Discharge measurement structures

Discharge measurement structures

Third revised edition

Edited by
M.G. Bos

Publication 20

International Institute for Land Reclamation and Improvement/ILRI

P.O.BOX45,6700 AA Wageningen, The Netherlands 1989.
Represented in the Working Group on Small Hydraulic Structures are the following
institutions:

International Institute for Land Reclamation and

Em Improvement/ILRI, Wageningen

H., Delft Hydraulics Laboratory, Delft

4 University of Agriculture, Departments of Hydraulics and

Irrigation, Wageningen

The first edition of this book appeared as

Publication no.20, ILRI, Wageningen
Publication no. 161, Delft Hydraulics Laboratory, Delft
Report no.4, Laboratory of Hydraulics and Catchment Hydrology, Wageningen

First edition 1976

Second edition 1978
Third revised edition 1989

0International Institute for Land Reclamation and Improvement/ILRI

Wageningen, The Netherlands 1989
This book or any part thereof must not be reproduced in any form without written permission of ILRI

ISBN 90 70754 15 O

Printed in the Netherlands

Preface to the first edition

The Working Group on Small Hydraulic Structures was formed in September 1971
and charged with the tasks of surveying current literature on small structures in open
channels and of conducting additional research as considered necessary.
The members of the Working Group are all engaged in irrigation engineering, hy-
drology, or hydraulics, and are employed by the Delft Hydraulics Laboratory (DHL),
the University of Agriculture (LU) at Wageningen, or the International Institute for
Land Reclamation and Improvement (ILRI) at Wageningen.

The names of those participating in the Group are:

Ing. W. Boiten (DHL)
Ir. M.G. Bos (ILRI)
Prof.Ir. D.A. Kraijenhoff van de Leur (LU)
Ir. H. Oostinga (DHL) during 1975
Ir. R.H. Pitlo (LU)
Ir. A.H. de Vries (DHL)
Ir. J. Wijdieks (DHL)

The Group lost one of its initiators and most expert members in the person of Professor
Ir. J. Nugteren (LU), who died on April 20, 1974.
The manuscripts for this publication were written by various group members. Ing.
W. Boiten prepared the Sections 4.3, 4.4, and 7.4; Ir. R.H. Pitlo prepared Section
7.5; Ir. A.H. de Vries prepared the Sections 7.2, 7.3, 9.2, and 9.7, and the Annexes
2 and 3. The remaining manuscripts were written by Ir. M.G. Bos. All sections were
critically reviewed by all working group members, after which Ir. M.G. Bos prepared
the manuscripts for publication.
Special thanks are due to Ir. E. Stamhuis and Ir. T. Meijer for their critical review
of Chapter 3, to Dr. P.T. Stol for his constructive comments on Annex 2 and to Dr.
M.J. Hall of the Imperial College of Science and Technology, London, for proof-
reading the entire manuscript.

This book presents instructions, standards, and procedures for the selection, design,
and use of structures, which measure or regulate the flow rate in open channels. It
is intended to serve as a guide to good practice for engineers concerned with the design
and operation of such structures. It is hoped that the book will serve this purpose
in three ways: (i) by giving the hydraulic theory related to discharge measurement
structures; (ii) by indicating the major demands made upon the structures; and (iii)
by providing specialized and technical knowledge on the more common types of struc-
tures now being used throughout the world.
The text is addressed to the designer and operator of the structure and gives the
hydraulic dimensions of the structure. Construction methods are only given if they
influence the hydraulic performance of the structure. Otherwise, no methods of con-
struction nor specifications of materials are given since they vary greatly from country
to country and their selection will be influenced by such factors as the availability
of materials, the quality of workmanship, and by the number of structures that need
to be built.

The efficient management of water supplies, particularly in the arid regions of the
world, is becoming more and more important as the demand for water grows even
greater with the worlds increasing population and as new sources of water become
harder to find. Water resources are one of our most vital commodities and they must
be conserved by reducing the amounts of water lost through inefficient management.
An essential part of water conservation is the accurate measurement and regulation
of discharges.
We hope that this book will find its way, not only to irrigation engineers and hydro-
logists, but also to all others who are actively engaged in the management of water
resources. Any comments which may lead to improved future editions of this book
will be welcomed.

Wageningen, October 1975 M.G.BOS

Editor

Preface to the second edition

The second edition of this book is essentially similar to the first edition in 1976, which
met with such success that all copies have been sold. The only new material in the
second edition is found in Chapter 7, Sections 1 and 5. Further all known errors have
been corrected, a number of graphs has been redrawn and, where possible, changes
in the lay-out have been made to improve the readability.
Remarks and criticism received from users and reviewers of the first edition have
been very helpful in the revision of this book.

Wageningen, July 1978 M.G.BOS

Editor
Preface to the third edition

This third edition retains the concept of the two previous editons, of which some 6700
copies have been sold. Nevertheless, major revisions have been made: in Sections 1.9,
4.1, 4.3, and 7.1 (which all deal with broad-crested weirs and long-throated flumes);
in Sections 1.5, 1.16, and 3.2.2; and in Annex 4.
Minor classifications have been added and errors corrected. Further, the typeface
and lay-out have been changed to improve the legibility of the text and accomodate
some additional information.

Wageningen, January 1989 Dr. M.G. Bos

Editor
Contents

Page

1 BASICPRINCIPLES OF FLUID FLOW AS APPLIED TO MEASURING STRUCTURES 17

1.1 General 17
1.2 Continuity 18
1.3 Equation of motion in the s-direction 19
1.4 Piezometric gradient in the n-direction 20
1.5 Hydrostatic pressure distribution in the m-direction 22
1.6 The total energy head of an open channel cross-section 23
1.7 Recapitulation 25
1.8 Specific energy 25
1.9 The broad-crested weir 28
1.9.1 Broad-crested weir with rectangular control section 29
1.9.2 Broad-crested weir with parabolic control section 31
1.9.3 Broad-crested weir with triangular control section 32
1.9.4 Broad-crested weir with truncated triangular control section 33
1.9.5 Broad-crested weir with trapezoidal control section 34
1.9.6 Broad-crested weir with circular control section 37
1.10 Short-crested weir 39
1.11 Critical depth flumes 41
1.12 Orifices 42
1.13 Sharpcrested weirs 45
I . 13.1 Sharp-crested weir with rectangular control section 47
1.13.2 Sharp-crested weir with parabolic control section 47
1.13.3 Sharp-crested weir with triangular control section 48
I . 13.4 Sharp-crested weir with truncated triangular control section 49
I . 13.5 Sharp-crested weir with trapezoidal control section 49
I . 13.6 Sharp-crested weir with circular control section 50
1.13.7 Sharp-crested proportional weir 52
1.14 The aeration demand of weirs 54
I . 15 Estimating the modular limit for long-throated flumes 58
I . 15.1 Theory 58
1.15.2 Energy losses upstream of the control section 58
I . 15.3 Friction losses downstream of the control section 60
1.15.4 Losses due to turbulence in the zone of deceleration 61
1.15.5 Total energy loss requirement 62
I . 15.6 Procedure to estimate the modular limit 64
I . 16 Modular limit of short-crested weirs 65
1.17 Selected list of literature 65
 i
2 AUXILIARY
EQUIPMENT FOR MEASURING STRUCTURES 67

2.1 Introduction 67
i 2.2 Head measurement station 68
2.3 The approach channel 69
2.4 Tailwater level 70
2.5 Staff gauge 70
2.6 Stilling well 72
2.7 Maximum stage gauge 76
2.8 Recording gauge 77
2.9 The float-tape and the diameter of the float 78
2.10 Instrument shelter 80
2.1 1 Protection against freezing 81
2.12 Differential head meters 81
2.13 Selected list of references 85

3 THE SELECTION OF STRUCTURES 87

3.1 Introduction 87
3.2 Demands made upon a structure 87
3.2.1 Function of the structure 87
3.2.2 Required fall of energy head to obtain modular flow 89
3.2.3 Range of discharges to be measured 92
3.2.4 Sensitivity 94
3.2.5 Flexibility 96
3.2.6 Sediment discharge capability 97
3.2.7 Passing of floating and suspended debris 1 o0
3.2.8 Undesirable change in discharge 101
3.2.9 Minimum water level in upstream channel 101
3.2.10 Required accuracy of measurement 102
3.2.11 Standardization of structures in an area 102
3.3 Properties and limits of application of structures 103
3.3.1 General 103
3.3.2 Tabulation of data 103
3.4 Selecting the structure 110
3.5 Selected list of references 119

4 BROAD-CRESTED
WEIRS 121

4.1 Horizontal broad-crested weir 121

4.1.1 Descrip tion 121
4.1.2 Evaluation of discharge 121
4.1.3 Modular limit 128
4.1.4 Limits of application 128
4.2 The Romijn movable measuring/regulating weir 129
 4.2.1 Description 129
Evaluation of discharge 131
4.2.2
4.2.3 Modular limit 132
4.2.4 Commonly used weir dimensions 133
4.2.5 Limits of application 137
4.3 Triangular broad-crested weir 137
4.3.1 Description 137
4.3.2 Evaluation of discharge 140
4.3.3 Modular limit 142
4.3.4 Limits of application 143
4.4 Broad-crested rectangular profile weir 143
4.4.1 Description 143
4.4.2 Evaluation of discharge 145
4.4.3 Limits of application 147
4.5 Faiyum weir 147
4.5.1 Description 147
4.5.2 Modular limit 148
4.5.3 Evaluation of discharge 150
4.5.4 Limits of application 151
4.6 Selected list of references 151

5 SHARP-CRESTED WEIRS 153

5.1 Rectangular sharp-crested weirs 153

5.1.1 Description 153
5.1.2 Evaluation of discharge 154
5.1.3 Limits of application 157
5.2 V-notch sharp-crested weirs 158
5.2.1 Description 158
5.2.2 Evaluation of discharge .160
5.2.3 Limits of application 164
- 5.2.4 Rating tables 164
5.3 Cipoletti weir 164
5.3.1 Description 164
5.3.2 Evaluation of discharge 165
5.3.3 Limits of application 166
5.4 Circular weir 167
- 5.4.1 Description 167
5.4.2 Determination of discharge 167
5.4.3 Limits of application 169
5.5 Proportional weir 169
5.5.1 Description 169
5.5.2 Evaluation of discharge 171
5.5.3 Limits of application 172
5.6 Selected list of references 173
6 SHORT-CRESTED WEIRS 175

6.1 Weir sill with rectangular control section 175

6.1.1 Description 175
6.1.2 Evaluation of discharge 176
6.1.3 Limits of application 176
6.2 V-notch weir sill 177
6.2.1 Description 177
6.2.2 Evaluation of discharge 178
6.2.3 Limits of application 180
6.3 Triangular profile two-dimensional weir 180
6.3.1 Description 180
6.3.2 Evaluation of discharge 182
6.3.3 Modular limit 183
6.3.4 Limits of application 184
6.4 Triangular profile flat-Vee weir 185
6.4.1 Description 185
6.4.2 Evaluation of discharge 186
6.4.3 Modular limit and non-modular discharge 188
6.4.4 Limits of application 191
6.5 Butchers movable standing wave weir 191
6.5.1 Description 191
6.5.2 Evaluation of discharge 194
6.5.3 Limits of application 195
6.6 WES-Standard spillway 195
6.6.1 Description 195
6.6.2 Evaluation of discharge 199
6.6.3 Limits of application 20 1
6.7 Cylindrical crested weir 202
6.7.1 Description 202
6.7.2 Evaluation of discharge 203
6.7.3 Limits of application 206
6.8 Selected list of references 206

7 FLUMES 209

7.1 Long-throated flumes 209

7.1. I Description 209
7.1.2 Evaluation of discharge 21 1
7.1.3 Modular limit 216
7.1.4 Limits of application 218
7.2 Throatless flumes with rounded transition 218
7.2.1 Description 218
7.2.2 Evaluation of discharge 220
7.2.3 Modular limit 22 1
7.2.4 Limits of application 222
7.3 Throatless flumes with broken plane transition 223
7.3. Description 223
7.4 Parshall flumes 224
7.4.1 Description 224
7.4.2 Evaluation of discharge 227
7.4.3 Submerged flow 24 1
7.4.4 Accuracy of discharge measurement 245
7.4.5 Loss of head through the flume 245
7.4.6 Limits of application 246
7.5 H-flumes 247
7.5.1 Description 247
7.5.2 Evaluation of discharge 252
7.5.3 Modular limit 252
7.5.4 Limits of application 253
7.6 Selected list of references 267

8 ORIFICES 269

8.1 Circular sharp-edged orifice 269

8.1.1 Description 269
8.1.2 Determination of discharge 269
8.1.3 Limits of application 27 1
8.2 Rectangular sharp-edged orifice 272
8.2.1 Description 272
8.2.2 Determination of' discharge 273
8.2.3 Modular limit 275
8.2.4 Limits of application 276
8.3 Constant-head-ori fice 277
8.3.1 Description 277
8.3.2 Determination of discharge 279
8.3.3 Limits of application 280
8.4 Radial or tainter gate 28 1
8.4.1 Description 28 1
8.4.2 Evaluation of discharge 282
8.4.3 Modular limit 284
8.4.4 Limits of application 286
8.5 Crump-De Gruyter adjustable orifice 286
8.5.1 Description 286
8.5.2 Evaluation of discharge 289
8.5.3 Limits of application 289
8.6 Metergate 29 1
8.6.1 Description 29 1
8.6.2 Evaluation of discharge 294
8.6.3 Metergate installation 295
8.6.4 Limits of application 298
8.7 Neyrpic module 299
8.7.1 Description 299
8.7.2 Discharge characteristics 299
8.7.3 Limits of application 305
8.8 Danadean tub 306
8.8.1 Description 306
8.8.2 Evaluation of discharge 306
8.8.3 Limits of application 308
8.9 Selected list of references 309

9 MISCELLANEOUS
STRUCTURES
311

9.1 Divisors 311

9.1.1 Description 31 1
9.1.2 Evaluation of discharge 3 12
9.1.3 Limits of application 313
9.2 Pipes and small syphons 314
9.2.1 Description 3 14
9.2.2 Evaluation of discharge 315
9.2.3 Limits of application 317
9.3 Fountain flow from a vertical pipe 318
9.3.1 Description 318
9.3.2 Evaluation of discharge 319
9.3.3 Limits of application 320
9.4 Flow from horizontal pipes 321
9.4.1 Description 321
9.4.2 Evaluation of discharge 322
9.4.3 Limits of application 326
9.5 Brink depth method for rectangular canals 326
9.5.1 Description 326
9.5.2 Evaluation of discharge 327
9.5.3 Limits of application 329
9.6 Dethridge meters 329
,9.6.1 Description 329
9.6.2 Evaluation of flow quantity 334
9.6.3 Regulation of discharge 336
9.6.4 Limits of application 336
9.7 Propeller meters 338
9.7.1 Description 338
9.7.2 Factors affecting propeller rotation 339
9.7.3 Head losses 342
9.7.4 Meter accuracy 343
9.7.5 Limits of application 343
9.8 Selected list of references 344
ANNEX1
Basic equations of motion in fluid mechanics 345

1. I Introduction 345
1.2 Equation of motion-Euler 345
1.3 Equation of motion in the s-direction ' 351
1.4 Piezometric gradient in the n-direction 353
1.5 Hydrostatic pressure distribution in the m-direction 354

ANNEX2
The overall accuracy of the measurement of flow 356

2.1 General principles 356

2.2 Nature of errors 356
2.3 Sources of errprs I 357
2.4 Propagation of errors 359
2.5 Errors in measurements of head 362
2.6 Coefficient errors 364
2.7 Example of error combination 365
2.8 Error in discharge volume over long period 367
2.9 Selected list of references 367

ANNEX3
Side weirs and oblique weirs 368

3.1 Introduction 368

3.2 Side weirs 368
3.2.1 General 368
3.2.2 Theory 369
3.2.3 Practical C,-values 372
3.2.4 Practical evaluation of side weir capacity 373
3.3 Oblique weirs 374
3.3.1 Weirs in trapezodal channels 374
3.4 Selected list of references 375

ANNEX4
Suitable stilling basins 377

4.1 Introduction 377

4.2 Straight drop structures 377
4.2. I Common drop 377
4.2.2 U.S. ARS basin 380
4.3 Inclined drops or chutes 38 1
4.3.1 Common chute 38 1
4.3.2 SAF Basin 383
4.4 Riprap protection 383
4.4.1 Determining maximum stone size in riprap mixture 386
4.4.2 Filter material placed beneath riprap 386
4.4.3 Permeability to water 386
4.4.4 Stability of each layer 388
4.4.5 Filter construction 389
4.5 Selected list of references 390

LISTOF PRINCIPAL SYMBOLS 392

SUBJECT INDEX 394
1 Basic principles of fluid flow as applied to
measuring structures

1.1 General
The purpose of this chapter is to explain the fundamental principles involved in eva-
luating the flow pattern in weirs, flumes, orifices and other measuring structures, since
it is the flow pattern that determines the head-discharge relationship in such structures.
Since the variation of density is negligible in the context of these studies, we shall
regard the mass density (p) of water as a constant. Nor shall we consider any flow
except time invariant or steady flow, so that a streamline indicates the path followed
by a fluid particle.

The co-ordinate system, used to describe the flow phenomena at a point P of a stream-
line in space, has the three directions as illustrated in Figure 1.1.
Before defining the co-ordinate system, we must first explain some mathematical
concepts. A tangent to a curve is a straight line that intersects the curve at two points
which are infinitely close to each other. An osculating plane intersects the curve at
three points which are infinitely close to each other. In other words, the curvature
at a point P exists in the local osculating plane only. Hence the tangent is a line in
the osculating plane. The normal plane to a curve at P is defined as the plane perpendi-
cular to the tangent of the curve at P. All lines through P in this normal plane are .
called normals, the normal in the osculating plane being called the principal normal,

J &O*

Figure 1. I The co-ordinate system

17
and the one perpendicular to the osculating plane being called the bi-normal.
The three co-ordinate directions are defined as follows:
s-direction: The direction of the velocity vector at point P. By definition, this vector
coincides with the tangent to the streamline at P (v, = v);
n-direction: The normal direction towards the centre of curvature of the streamline
at P.By definition, both the s- and n-direction are situated in the oscula-
ting plane;
m-direction: The direction perpendicular to the osculating plane at P as indicated
in Figure 1.1.
It should be noted that, in accordance with the definition of the osculating plane,
the acceleration of flow in the m-direction equals zero (a, = O).
Metric units (SI) will be used throughout this book, although sometimes for practical
purposes, the equivalent Imperial units will be used in addition.

1.2 Continuity
An elementary flow passage bounded by streamlines is known as a stream tube. Since
there is, per definition, no flow across these boundaries and since water is assumed
here to be incompressible, fluid must enter one cross-section of the tube at the same
volume per unit time as it leaves the other.

___)

Figure 1.2 The stream tube

From the assumption of steady flow, it follows that the shape and position of the
stream tube do not change with time. Thus the rate at which water is flowing across
a section equals the product of the velocity component perpendicular to the section
and the area of this section. If the subscripts 1 and 2 are applied to the two ends
of the elementary stream tube, we can write:
Discharge = dQ = v,dA, = v,dA, (1-1)
This continuity equation is valid for incompressible fluid flow through any stream
tube. If Equation 1-1 is applied to a stream tube with finite cross-sectional area, as
in an open channel with steady flow (the channel bottom, side slopes, and water surface
being the boundaries of the stream tube), the continuity equation reads:

18
 Q = JAvdA = SA = constant
or
-
v,A, = S,A2

where S is the average velocity component perpendicular to the cross-section of the

open channel.

1.3 Equation of motion in the s-direction

Since we do not regard heat and sound as being types of energy which influence the
liquid flow in open channels, an elementary fluid particle has the following three inter-
changeable types of energy per unit of volume:
1/zpv2 = kinetic energy per unit of volume
pgz = potential energy per unit of volume
P = pressure energy per unit of volume.

Consider a fluid particle moving in a time interval At from Point 1 to Point 2 along
a streamline, there being no loss of energy due to friction or increased turbulence.
(See Fig. 1.3.) Since, on the other hand, there is no gain of energy either, we can write
, (1/zpv2+ pgz + P), = (1/zpv2+ pgz + P)2 = constant (1 -3)
This equation is valid for points along a streamline only if the energy losses are negligi-
i ble and the mass density (p) is a constant. According to Equation 1-3
+
1/2pv2 pgz + P = constant
or
v2/2g + P/pg + z = H = constant

I where, as shown in Figure 1.3,

19
v2/2g = the velocity head
P/pg = the pressure head
Z = the elevation head
P/pg + z = the piezometric head
H = the total energy head.

The last three heads all refer to the same reference level. The reader should note that
each individual streamline may have its own energy head. Equations 1-3, 1-4, and
1-5 are alternative forms of the well-known Bernoulli equation, of which a detailed
derivation is presented in Annex 1.

1.4 Piezometric gradient in the n-direction

On a particle (ds, dn, dm) following a curved streamline, a force F is acting towards
the centre of curvature in order to accelerate the particle in a sense perpendicular
to its direction of motion. Since in Section 1.1 the direction of motion and the direction
towards the centre of curvature have been defined as the s- and n-direction respectively,
we consider here the movement of a particle along an elementary section of a streamline
in the osculating plane.

By Newton's second law of motion

F = ma
the centripetal acceleration (a) in consequence of the passage along a circle with a
radius (r) with a velocity (v), according to mechanics, equals:

a = -V2 (1-7)
r
Since the mass (m) of the particle equals p(ds dn dm), the force (F) can be expressed
as
V2
F = pdsdndm- (1-8)
r
This force (F) is due to fluid pressure and gravitation acting on the fluid particle.
It can be proved (see Annex 1) that the negative energy gradient in the n-direction
equals the centripetal force per unit of mass (equals centripetal acceleration). In other
words:

or

(1-10)

After integration of this equation from Point 1 to Point 2 in the n-direction we obtain
the following equation for the fall of piezometric head in the n-direction (see Fig. 1.4)

20
 Figure 1.4 Key to symbols

(1-1 I )

In this equation

= the piezometric head at Point 1

= the piezometric head at Point 2

+

I
v2
= the difference between the piezometric heads at Points 1 and 2
{ dn due to the curvature of the streamlines
From this equation it appears that, if the streamlines are straight (r = co), the integral
has zero value, and thus the piezometric head at Point 1 equals that at Point 2, so
that

[k+ = [s+ .I2 = constant (1-12)

k P-PgYo
4
Figure 1.5 Hydrostatic pressure distribution

21
At the water surface in an open channel, P, = O; hence
pz = yo-z
Pg
or
p2 = Pg(Y0-Z) (1-13)
Thus, if r = co there is what is known as a hydrostatic pressure distribution. If the
streamlines are curved, however, and there is a significant flow velocity, the integral
may reach a considerable value.

At a free overfall with a fully aerated air pocket underneath the nappe, there is atmos-
pheric pressure at both Points 1 and 2, while a certain distance upstream there is a
hydrostatic pressure distribution. The deviation from the hydrostatic pressure distri-
bution at the end of the weir.is mainly caused by the value of the integral (see Fig.l.6).
A decrease of piezometric head, which is due to the centripetal acceleration, necessarily
induces a corresponding increase of velocity head:

(1-14)

To illustrate the effect of streamline curvature on the velocity distribution in 'the n-

direction, Figure 1.6 shows the velocity distribution over a cross-section where a hy-
drostatic pressure distribution prevails and the velocity distribution at the free overfall.

2F velocity
distribution

Figure 1.6 Pressure and velocity distribution at a free overfall

1.5 Hydrostatic pressure distribution in the m-direction

As mentioned in Section 1.1, in the direction perpendicular to the osculating plane,
not only v, = O, but also

22
 Consequently, there is no net force acting in the m-direction, and therefore the pressure
distribution is hydrostatic. Hence
P +.pgz = constant (1-15)
or
P
-
Pg
+ z = constant (1-16)

1.6 The total energy head of an open channel cross-section

According to Equation 1-4, the total energy per unit of volume of a fluid particle
can be expressed as the sum of the three types of energy
'12 pv2 + pgz + P (1-17)
We now want to apply this expression to the total energy which passes through the
entire cross-section of a channel. We therefore need to express the total kinetic energy
of the discharge in terms of the average flow velocity of the cross-section.
In this context, the reader should note that this average flow velocity is not a directly
measurable quantity but a derived one, defined by
-v = Q
- (1-18)
A
Due to the presence of a free water surface and the friction along the solid channel
boundary, the velocities in the channel are not uniformly distributed over the channel
cross-section (Fig. 1.7). Owing to this non-uniform velocity distribution, the true ave-
rage kinetic energy per unit of volume across the section, ('h p~~),,,,,~, will not neces-
sarily be equal to ' / 2 pVz.
In other words:
('/z PV2)a"erage
=U '/z PV2 (1-19)
The velocity-distribution coefficient (U),always exceeds unity. It equals unity when
t m uniform-cross- c r o s s - s e c m b e c o m e s greater th-urther
flow
-
departs from uniform.
sle- For with steady turbulent flow, U-values range between 1 .O3
and I . IO. In many cases the velocity head makes up only a minor part of the total
-

Figure 1.7 Examples of velocity profiles in a channel section

23
 energy head and a = 1 can then be used for practicalpupmgs. Thus, the average
i t i c e n ~ o f v o l u m ofwater e passing a cross-sction equals a l/z
p~2.
The variation of the remaining two terms over the cross-section is characterized
by Equations 1-9 and 1-15. If we consider an open channel section with steady flow,
where the streamlines are straight and parallel, there is no centripetal acceleration
and, therefore, both in the n- and m-direction, the sum of the potential and pressure
energy at any point is constant. In other words;
pgz + P = constant (1-20)
for all points in the cross-section. Since at the water surface P = O, the piezometric
level of the cross-section coincides with the local water surface. For the considered
cross-section the expression for the average energy per unit of volume passing through
the cross-section reads:
E = h p i + P + pgz (1-21)
or if expressed in terms of head

(1 -22)

where H is the total energy head of a cross-sectional area of flow. We have now reached
the stage that we are able to express this total energy head in the elevation of the
+
water surface (P/pg z) plus the velocity head a3/2g.
In the following sections it will be assumed that over a short reach of accelerated
flow, bounded by channel cross-sections perpendicular to straight and parallel stream-
lines, the loss of energy head is negligible with regard to the interchangeable types
of energy (Figure 1.8). Hence:

ag+[&+z] I (1 -23)

Thus, if we may assume the energy head (H) in both cross-sections to be the same,
we have an expression that enables us to compare the interchange of velocity head
and piezometric head in a short zone of acceleration.

head
meosurement control
sectlon section
I

1 flow I
1- I
I

Figure 1.8 The channel transition

24
1.7 Recapitulation
For a short zone of acceleration bounded by cross-sections perpendicular to straight
and parallel streamlines, the following two equations are valid:
Continuity equation (1-2)
Q = VIAl = VZA2
Bernoullis equation (1 -23)
-2

In both cross-sections the piezometric level coincides with the water surface and the
latter determines the area A of the cross-section. We may therefore conclude that
if the shapes of the two cross-sections are known, the two unknowns i ,and i,can
be determined from the two corresponding water levels by means of the above equa-
tions.
It is evident, however, that collecting and handling two sets of data per measuring
structure is an expensive and time-consuming enterprise which should be avoided if
possible. It will be shown that under critical flow conditions one water level only is
sufficient to determine the discharge. In order to explain this critical condition, the
concept of specific energy will first be defined.

1.8 Specific energy

The concept of specific energy was first introduced by Bakhmeteff in 1912, and is
defined as the average energy per unit weight of water at a channel section as expressed
with respect to the channel bottom. Since the piezometric level coincides with the water
surface, the piezometric head with respect to the channel bottom is
P
-
Pg
+ z = y, the water depth (1 -24)

so that the specific energy head can be expressed as:

Ho = y + clv2/2g (1-25)
We find that the specific energy at a channel section equals the sum of the water depth
(y) and the velocity head, provided of course that the streamlines are straight and
parallel. Since 7 = Q/A, Equation 1-25 may be written

Ho= y Q
+ 2gA2
CL- (1 -26)

where A, the cross-sectional area of flow, can also be expressed as a function of the
water depth, y. From this equation it can be seen that for a given channel section
and a constant discharge (Q), the specific energy in an open channel section is a func-
tion of the water depth only. Plotting this water depth (y) against the specific energy
(Ho)gives a specific energy curve as shown in Figure 1.9.

25
Figure 1.9 The specific energy curve

The curve shows that, for a given discharge and specific energy there are two alter-
nate depths of flow. At Point C the specific energy is a minimum for a given discharge
and the two alternate depths coincide. This depth of flow is known as critical depth
(Y&
When the depth of flow is greater than the critical depth, the flow is called subcritical;
if it is less than the critical depth, the flow is called supercritical. The curve illustrates
how a given discharge can occur at two possible flow regimes; slow and deep on the
upper limb, fast and shallow on the lower limb, the limbs being separated by the critical
flow condition (Point C).
When there is a rapid change in depth of flow from a high to a low stage, a steep
depression will occur in the water surface; this is called a hydraulic drop. On the
other hand, when there is a rapid change from a low to a high stage, the water surface
will rise abruptly; this phenomenon is called a hydraulic jump or standing wave.
The standing wave shows itself by its turbulence (white water), whereas the hydraulic
drop is less apparent. However, if in a standing wave the change in depth is small,
the water surface will not rise abruptly but will pass from a low to a high level through
a series of undulations (undular jump), and detection becomes more difficult. The
normal procedure to ascertain whether critical flow occurs in a channel contraction
- there being subcritical flow upstream and downstream of the contraction - is to

look for a hydraulic jump immediately downstream of the contraction.

From Figure 1.9 it is possible to see that if the state of flow is critical, i.e. if the specific
energy is a minimum for a given discharge, there is one value for the depth of flow
only. The relationship between this minimum specific energy and the critical depth
is found by differentiating Equation 1-26 to y, while Q remains constant.

(1 -27)

Since dA = B dy, this equation becomes

(1 -28)

26
Photo 1 Hydraulic jumps

27
I 7

I If the specific energy is a minimum dH,/dy = O, we may write

(1 -29)

Equation 1-29 is valid only for steady flow with parallel streamlines in a channel of
small slope. If the velocity distribution coefficient, U, is assumed to be unity, the crite-
rion for critical flow becomes
-2
Vc--- AC or V = i,= (g A,/B,)O.SO (1-30)
2g 2Bc
Provided that the tailwater level of the measuring structure is low enough to enable
the depth of flow at the channel contraction to reach critical depth, Equations 1-2,
1-23, and 1-30 allow the development of a discharge equation for each measuring
device, in which the upstream total energy head (HI)is the only independent variable.
Equation 1-30 states that at critical flow the average flow velocity V, = (g A,/B,)n.5n
It can be proved that this flow velocity equals the velocity with which the smallest
disturbance moves in an open channel, as measured relative to the flow. Because of
this feature, a disturbance or change in a downstream level cannot influence an up-
stream water level if critical flow occurs in between the two cross-sections considered.
The 'control section' of a measuring structure is located where critical flow occurs
and subcritical, tranquil, or streaming flow passes into supercritical, rapid, or shooting
flow.
Thus, if critical flow occurs at the control section of a measuring structure, the
upstream water level is independent of the tailwater level; the flow over the structure
is then called 'modular'.

1.9 The broad-crested weir

A broad-crested weir is an overflow structure with a horizontal crest above which
the deviation from a hydrostatic pressure distribution because of centripetal acce-
leration may be neglected. In other words, the streamlines are practically straight
and parallel. To obtain this situation the length of the weir crest in the direction
of flow (L) should be related to the total energy head over the weir crest as
u 7 < H , / L < 0.50. H,/L I 0.07 because otherwise
- = c
the energy losses above the
weir crest c a p b e m a y occur on the-creqt; iI,/L_> 0.50,
so tha; onlylight curvature-lGs occurTabove the crest and a hydrostatic
pressure distribution may be assumed.
~

If the measuring structure is so designed that there are no significant energy losses
in the zone of acceleration upstream of the control section, according to Bernoulli's
equation (1-23):
H, = h, + ~V,'/2g= H = y + aV2/2g

(1-31)
where H, equals the total upstream energy head over the weir crest as shown in Figure

28
 c c

o s,
zs --
._
O
5k c

Figure 1.10 IhStrdtiOn of terminology

1.10. Substituting Q = VA and putting c1 = 1 .O gives

Q = A (2g(Hl - Y ) } ~ . ~ (1-32)
Provided that critical flow occurs at the control section (y = y,), a head-discharge
equation for various throat geometries can now be derived from
Q = Ac {2idHi -YJ)~.~ (1 -33)

I 1.9.1 Broad-crestedweir with rectangular control section

For a rectangular control section in which the flow is critical, we may write A, =
b,y, and A,/B, = y, so that Equation 1-30 may be written as Y2/2g = I2 y,. Hence:

(1 -34)

Substitution of this relation and A, = b, into Equation 1-33 gives, after simplification

(1-35)

This formula is based on idealized assumptions such as: absence of centripetal forces

+ bC

Figure I. I 1 Dimensions of a rectangular control section

29
 in the upstream and downstream cross-sections bounding the considered zone of acce-
leration, absence of viscous effects and increased turbulence, and finally a uniform
velocity distribution so that also the velocity distribution coefficient can be omitted.
In reality these effects do occur and they must therefore be accounted for by the intro-
duction of a discharge coefficient C,. The C,-value depends on the shape and type
of the measuring structure.
Q = C 2 (-g>o
2 b,
(1 -36)
d3 3
Naturally in a field installation it is not possible to measure the energy head H Idirectly
and it is therefore common practice to relate the discharge to the upstream water
level over the crest in the following way
Q = C C -
3 -3g
>,, b, h,1.50 (1-37)
w h e r e x s a correction coefficient for neglecting the velocity head in the approach
channel, c1,v,~/2g. Generally, the approach velocity coefficient

c, = [?T (1-38)

where u equals the power of h, in the head-discharge equation, being u 1.50 for
__I_

a rectangulauontro1,sec~oLT-
cc__

Thus C, is greater than unity and is .related to the shape of the .ap.pr.oach channel
section6?6o the pow& ofh, in the head-discharge equation.
Values of C, a s o n of the area r&o CdA*/AI are shown in Figure 1.12 for

coefficient of
approach velocity C,
1.20.
I I I
- - - - -- - - - triangular control u=2.5
_-_ parabolic control u=2.0
~ rectangular control u=l.5
1.15 -
control section

1.10 -

1.05

1.00 -
O o .1 0.2 0.3 0.L 0.5 0.6 O .7 0.8
area r a t i o v j c d A*/A,
A* = wetted area at control sectlon if waterdepth equals y = hl
Al = wetted are at head measurement station

Figure 1.12 C, values as a function of the area ratio &CdA*/A~ (from Bos 1977)

30
Photo 2 Flow over a round-nose broad-crested weir with rectangular control section

various control sections where A * equals the imaginary wetted area at the control
section if we assume that the water depth y = h,; A, equals the wetted area at the
head measurement station and Cd is the discharge coefficient. In Chapters 4 to 9, the
Cd-valueis usually given as some function of H l . Since it is common practice to measure
h, instead of H,. a positive correction equal to vI2/2gshould be made on h, to find
the true Cd-valuewhenever the change in c d as a function of H I is significant.
\

In the literature, Equation 1-37 is sometimes written as

Q = CiC,b,h,'50 (1 -39)

-
It should be noted that in this equation the coefficient Ci has the dimension
To avoid mistakes and to facilitate easy comparison of discharge
the metric and the

1.9.2 Broad-crested weir with parabolic control section

-
T-'1.
coefficients in both

For a parabolic control section, having a focal distance equal to f, (see Figure 1.13)
with A, = 2/3 B,y, and B, = 2&, we may write Equation 1-30 as

(1 -40)

Hence

(1-41)

Substituting those relations into Equation 1-33 gives

Q = HIz.O (1 -42)

31
Figure 1.13 Dimensions of a parabolic control section

As explained in Section 1.9.1, correction coefficients have to be introduced to obtain

a practical head-discharge equation. Thus

3
Q = CdC V J T fg hI2.O (1-43)

1.9.3 Broad-crested weir with triangular control section

0 0
For a triangular control section with A, = y: tan- and B, = 2yc tan- (see Figure
2 2
1.14), we may write Equation 1-30 as:

( 1-44)

Hence

(1 -45)

0
Substituting those relations and A, = y: tan- into Equation 1-33 gives
2

Figure 1.14 Dimensions of a triangular control section

32
 Q = '6 [? g]uIOtan!?2 H12.50 (1 -46)
25 5
For similar reasons as explained in Section 1.9.1, a Cd- and C,-coefficient have to
be introduced to obtain a practical head-discharge equation. Thus
16 2 0
Q = cdC, 25 [5 g] tan- h,*.50
2
(1 -47)

1.9.4. Broad-crested weir with truncated triangular control section

For weirs with a truncated triangular control section, two head-discharge equations
have to be ,used: one for the conditions where flow is confined within the triangular
section, and the other, at higher stages, where the presence of the vertical side walls
has to be taken into account. The first equation is analogous to Equation 1-47, being

Q c ~ 16c 2, ~ tan-
= [ ~
2 ~ ]
0 h,2.50 (1 -48)

which is valid if HI < 1.25 Hb.

The second equation will be derived below.

For a truncated triangular control section
0 1
A, = H: tan-
2
+ Bc(y,- Hb) = bcy, --
2
BcHb

According to Equation 1-30 we may write (see Figure .IS)

(1 -49)

Hence

(1 -50)

Figure I . 15 Dimension of a truncated triangular control section

33
Photo 3 Flow over a broad-crested weir with triangular control section

2 1
Substituting those relations and A, = - B,H, -- B,Hb into Equation
3 3 1-33 gives

(1-51)

For similar reasons as explained in Section 1.9.I , discharge and approach velocity
coefficients have to be introduced to obtain a practical head-discharge equation. Thus

(1-52)

which is valid provided H I 2 1.25 Hb.

1.9.5 Broad-crested weir with trapezodal control section

For weirs with a trapezodal control section with A, = b,y, + z,yf and B, = b, +
2zCy0we may write Equation 1-30 as (Figure 1.16)
5-4-- bCY, +
2
- ZCY:
(1-53)
2g 2B, 2b, + 4z,y,
+
Since H = H, = v,2/2g yo we may write the total energy head over the weir crest
as a function of the dimensions of the control section as
H I = 3 bcY, + 5 ZCY2 (1-54)
2 b, + 4 Z C Y ,
From this equation it appears that the critical depth in a trapezoidal control section
is a function of the total energy head H I ,of the bottom width b, and of the side slope
ratio z, of the control section.

34
It also shows that, if both b, and z, are known the ratio y,/Hl is a function of H I .
Values of yc/HIas a function of z, and the ratio Hl/bcare shown in Table I . I .
Substitution of A, = bcy, + zcy,2into Equation 1-33 and introduction of a discharge
coefficient gives as a head-discharge equation
Q = G {bCyc+ z,Y,~){%(Hi - YC))O.~O \
(1 -55)
Since for each combination of b,, zo and H,/b,, the ratio y,/H, is given in Table 1.1,
the discharge Q can be computed because the discharge coefficient has a predictable
value. In this way a Q-HI curve can be obtained. If the approach velocity head vI2/2g
is negligible, this curve may be used to measure the discharge. If the approach velocity
has a significant value, v12/2gshould be estimated and h, = H I-v12/2gmay be obtained
in one or two steps.
In the literature the trapezodal control section is sometimes described as the sum
of a rectangular and a triangular control section. Hence, along similar lines as will
be shown in Section I . 13 for sharp-crested weirs, a head-discharge equation is obtained
by superposing the head-discharge equations valid for a rectangular and a triangular
control section. For broad-crested weirs, however, this procedure results in a strongly
variable C,-value, since for a given H, the critical depth in the two superposed equa-
tions equals 2/3H, for a rectangular and 4/5Hcfor a triangular control section. This
difference of simultaneous y,-values is one of the reasons why superposition of various
head-discharge equations is not allowed. A second reason is the significant difference
in mean flow velocities through the rectangular and triangular portions of the control
section.

\ /

Figure I . 16 Dimensions of a trapezoidal control section

35
 Table 1.1 Values of the ratio yc/HI as a function of& and H,/b, for trapezodal control sections

Side slopes of channel, ratio of horizontal to vertical (zc: I )

HdbC Vertical 0.25:1 0.5O:l 0.75:l 1 : l 1.5:1 2.1 2.5:l 3.1 4:l
.o0 ,667 ,667 ,667 ,667 .667 ,667 ,667 .667 ,667 ,667
.o 1 ,667 ,667 ,667 ,668 ,668 ,669 ,670 ,670 .67I .672
.o2 ,667 ,667 ,668 ,669 ,670 ,671 ,672 .674 .675 ,678
.O3 ,667 ,668 ,669 ,670 .67I ,673 ,675 ,677 .679 ,683
.O4 ,667 ,668 ,670 .67I ,672 ,675 ,677 ,680 ,683 ,687
.O5 ,667 ,668 .670 ,672 .674 ,677 ,680 .683 ,686 ,692
.O6 ,667 ,669 ,671 .673 ,675 ,679 ,683 ,686 ,690 ,696
.O7 ,667 ,669 ,672 ,674 ,676 .681 ,685 ,689 ,693 ,699
.O8 ,667 ,670 ,672 .675 ,678 ,683 ,687 ,692 .696 ,703
.O9 ,667 .670 ,673 ,676 .679 ,684 ,690 ,695 ,698 .706
.IO ,667 ,670 ,674 .677 ,680 ,686 ,692 ,697 .701 ,709
.I2 ,667 .67I ,675 .679 ,684 .690 ,696 ,701 .706 ,715
.I4 .667 ,672 .676 .68I ,686 .693 ,699 ,705 .711 ,720
.I6 ,667 ,672 ,678 ,683 ,687 ,696 ,703 ,709 ,715 ,725
.18 .667 ,673 .679 ,684 ,690 ,698 ,706 ,713 ,719 .729
.20 ,667 ,674 .680 ,686 ,692 ,701 ,709 .717 ,723 ,733
.22 .667 ,674 .681 ,688 ,694 ,704 ,712 ,720 .726 ,736
.24 ,667 ,675 ,683 ,689 ,696 ,706 ,715 ,723 ,729 .739
.26 ,667 ,676 ,684 ,691 ,698 ,709 ,718 ,725 ,732 ,742
.28 ,667 ,676 ,685 ,693 .699 .7I 1 ,720 ,728 ,734 ,744
.30 ,667 ,677 .686 ,694 ,701 ,713 .723 ,730 ,737 .747
.32 ,667 ,678 ,687 ,696 ,703 ,715 ,725 ,733 ,739 ,749
.34 ,667 ,678 ,689 ,697 .705 ,717 ,727 ,735 ,741 .751
.36 ,667 ,679 ,690 ,699 ,706 ,719 ,729 ,737 ,743 ,752
.38 ,667 ,680 .691 ,700 ,708 ,721 .73I ,738 .745 ,754
.40 ,667 .680 ,692 ,701 .709 ,723 .733 ,740 .747 ,756
.42 ,667 ,681 ,693 ,703 ,711 ,725 ,734 ,742 ,748 .757
.44 .667 ,681 .694 ,704 ,712 ,727 ,736 ,744 ,750 .759
.46 ,667 ,682 ,695 .705 ,714 ,728 ,737 .745 ,751 ,760
.48 ,667 ,683 ,696 .706 .715 ,729 ,739 .747 ,752 .761
.so ,667 ,683 ,697 ,708 ,717 ,730 ,740 ,748 ,754 ,762
.60 .667 ,686 ,701 ,713 ,723 ,737 .741 ,754 ,759 .161
.70 ,667 .688 ,706 .718 ,728 .742 ,752 ,758 .764 .771
.80 ,667 ,692 ,709 ,723 ,732 ,746 ,756 ,762 ,767 .774
.90 ,667 ,694 ,713 ,727 ,737 ,750 ,759 ,766 .770 ,776
1.o ,667 .697 ,717 .730 ,740 .754 ,762 .768 ,773 .778
I .2 ,667 ,701 .723 ,737 .747 ,759 .767 ,772 .776 ,782
I .4 ,667 .706 ,729 ,742 ,752 ,764 .77I .776 ,779 ,784
I .6 ,667 ,709 ,733 ,747 ,756 ,767 ,774 ,778 ,781 ,786
1.8 ,667 ,713 ,737 .750 ,759 ,770 ,776 .78I ,783 ,787
2 .667 .717 .740 ,754 ,762 ,773 ,778 ,782 .785 ,788
3 ,667 ,730 ,753 ,766 ,773 .781 ,785 ,787 ,790 ,792
.4 ,667 ,740 ,762 ,773 ,778 ,785 ,788 ,790 ,792 ,794
5 ,667 ,748 ,768 ,777 .782 ,788 ,791 ,792 .794 ,795
IO ,667 ,768 ,782 ,788 ,791 ,794 .795 ,796 .797 .798
03 ,800 ,800 ,800 ,800 ,800 300 ,800 ,800 ,800

36
1.9.6 Broad-crested weir with circular control section

For a broad-crested weir with a circular control section we may write (see Figure
1.17) ..
1
A, = -d,Z(O-sin)) (1 -56)
8
B, = d,sin l/,O and (1 -57)
d (1-58)
y, = (1 -cos '/,O) = d, sin2 1/40
2
Substitution of values for A, and B, into Equation 1-30 gives
vc
- - A, - d, O-sin8 (1-59)
2g 2B, 16 sin '/,O
and because H = H, = y, + v,2/2g we may write the total energy head over the
weir crest as

(1-60)

For each value of yJd, = sin2 1/40a matching value of the ratios A,/d,Z and HJd,
can now be calculated with the above equations. These values, and the additional
values on the dimensionless ratios v,2/2gdCand yJH, are presented in Table 1.2.
For a circular control section we may use the general head-discharge relation 'given
earlier (Equation 1-33)
Q = C d Ac (2gWi - YJ}' 50 (1-61)
where the discharge coefficient Cd has been introduced for similar reasons to those
explained in Section 1.9.1. The latter equation may also be written in terms of dimen-
sionless ratios as

Figure 1.17 Dimensions of a circular control section

, 37
 Table 1.2 Ratios for determining the discharge Q of a broad-crested weir and long-throated flume with
circular control section (Bos 1985)

YJ~, v,2/2gdc ~ , / d , A& ycm1 fw yc/dC v2/2& HiPC Acid: yc/Hi f(e)
.O1 ,0033 ,0133 .O013 ,752 0.0001 .5 I .20 14 .7 I 14 ,4207 .7 I7 0.2556
.O2 ,0067 ,0267 ,0037 ,749 0.0004 .52 ,2065 .7265 ,4127 ,716 0.2652
.O3 .O101 ,0401 ,0069 ,749 0.0010 .53 .2 117 ,7417 ,4227 ,715 0.2750
.O4 .O1 34 ,0534 ,0105 ,749 0.0017 .54 .2 I70 ,7570 ,4327 .7 13 0.2851
.O5 ,0168 ,0668 ,0147 ,748 0.0027 .55 ,2224 ,7724 ,4426 ,712 0.2952

.O6 ..O203 .O803 ,0192 ,748 0.0039 .56 ,2279 .7879 ,4526 ,711 0.3056
.O7 ,0237 ,0937 ,0242 ,747 0.0053 .57 .2335 ,8035 ,4625 ,709 0.3161
.O8 . ,0271 ,1071 ,0294 ,747 0.0068 .58 ,2393 ,8193 ,4724 ,708 0.3268
.O9 ,0306 ,1206 .O350 ,746 0.0087 .59 ,2451 ,8351 ,4822 ,707 0.3376
.IO ,0341 ,1341 ,0409 ,746 0.0107 .60 ,2511 ,8511 ,4920 .705 0.3487

.I 1 ,0376 ,1476 .O470 ,745 0.0129 .61 ,2572 3672 ,5018 .703 0.3599
.I2 ,0411 ,1611 SO34 ,745 0.0153 .62 ,2635 ,8835 ,5115 ,702 0.3713
.I3 ,0446 ,1746 ,0600 ,745 0.0179 .63 ,2699 . ,8999 ,5212 .700 0.3829
.I4 ,0482 ,1882 ,0688 ,744 0.0214 .64 ,2765 .9 165 ,5308 ,698 0.3947
.15 ,0517 ,2017 .O739 ,744 0.0238 .65 ,2833 .9333 ,5404 ,696 ,0.4068

.16 ,0553 ,2153 ,0811 .743 0.0270 .66 ,2902 ,9502 .5499 ,695 0.4189
.I7 ,0589 ,2289 ,0885 ,743 0.0304 .67 ,2974 ,9674 ,5594 ,693 0.4314
.I8 ,0626 ,2426 .O96 1 ,742 0.0340 .68 .3048 ,9848 S687 .69 1 0.4440
.I9 ,0662 ,2562 ,1039 ,742 0.0378 .69 .3125 1.0025 S780 ,688 0.4569
.20 ,0699 ,2699 . I I I8 ,741 0.0418 .70 ,3204 1.0204 ,5872 ,686 0.4701

.21 .O736 ,2836 . I I99 ,740 0.0460 .71 .3286 1.0386 S964 ,684 0.4835
.22 ,0773 .2973 .I281 ,740 0.0504 .72 ,3371 1.O571 ,6054 .681 0.497 1
.23 .O8 1 I .3 11 1 .I365 ,739 0.0550 .73 ,3459 1.0759 ,6143 ,679 0.5109
.24 ,0848 ,3248 .I449 ,739 0.0597 .74 . ,3552 1.0952 ,6231 .676 0.5252
.25 .O887 ,3387 .I535 ,738 0.0647 .75 .3648 1.1148 .6319 ,673 0.5397

.26 ,0925 ,3525 ,1623 ,738 0.0698 .76 ,3749 1.1349 ,6405 ,670 0.5546
.27 ,0963, ,3663 ,1711 ,737 0.0751 .77 ,3855 1.1555 ,6489 ,666 0.5698
.28 ,1002 ,3802 ,1800 ,736 0.0806 .78 .3967 1.1767 ,6573 ,663 0.5855
.29 .IO42 ,3942 ,1890 ,736 0.0863 .79 ,4085 1.1985 ,6655 ,659 0.6015
.30 ,1081 ,4081 ,1982 ,735 0.0922 30 .4210 1.2210 ,6735 ,655 0.6180

.3 1 . I 12I ,4221 .2074 .734 0.0982 .8 I ,4343 1.2443 .68 I5 ,651 0.635 1
, .32 . I 161 ,4361 ,2167 ,.734 0.1044 32 ,4485 1.2685 .6893 ,646 0.6528
.33 ,1202 ,4502 ,2260 ,733 0.1 108 .83 ,4638 1.2938 ,6969 ,641 0.6712
.34 ,1243 ,4643 ,2355 ,732 0.1 I74 .84 ,4803 1.3203 ,7043 ,636 0.6903
.35 .1284 ,4784 ,2450 ,732 0.1289 .85 ,4982 1.3482 ,7115 ,630 0.7102

.36 ,1326 ,4926 ,2546 ,731 0.1311 36 ,5177 1.3777 ,7186 ,624 0.7312
.37 ,1368 ,5068 ,2642 ,730 O. 1382 .87 ,5392 1.4092 ,7254 ,617 0.7533
.38 .1411 ,5211 ,2739 ,729 0.1455 38 ,5632 1.4432 ,7320 ,610 0.7769
.39 ,1454 ,5354 .2836 ,728 0.1529 .89 S900 1.4800 .7384 ,601 0.8021
.40 ,1497 ,5497 .2934 ,728 0.1605 .90 ,6204 1.5204 .7445 ,592 0.8293

.41 .I541 ,5641 ,3032 ,727 0.1683 .91 .6555 1.5655 .7504 ,581 0.8592
.42 ,1586 .5786 ,3130 ,726 0.1763 .92 .6966 1.6166 ,7560 ,569 0.8923
.43 ,1631 ,5931 ' ,3229 ,725 0.1844 .93 ,7459 1.6759 ,7612 ,555 0.9297
,44 ,1676 ,6076 .3328 ,724 0.1927 .94 .8065 1.7465 .7662 ,538 0.9731
.45 ,1723 ,6223 .3428 ,723 0.2012 .95 ,8841 1.8341 ,7707 .5 18 I .O248

38
Substitution of Equations 1-56, 1-58, and 1-60 into Equation 1-62 yields
(1 -63)
(0 - sin
where f(e) = is a shape factor for the control section.
8(8 sin e
2
If d, is known and H I is set to a given value, the related value of f(0) can be read
from Table 1.2. Substitution of this value and the Cd value to Equation 1.62 yields
the discharge Q. The iterative procedure of Section 1.9.5 should be used to transform
this Hl-Q relationship int? an h,-Q relationship.
Table 1.2 also contains columns presenting dimensionless values for the velocity
head, water depth, and related area of flow.

1.10 ' Short-crested weir

The basic difference between a broad-crested weir and a short-crested weir is that
n o < h m T 6 ? 3 h e o r t crest can the curvatureTf the s t E l i G g e neglected; there
CL-

is t h G no h y d r o s t a f i m distribution. The two-dimensional flow pattern over

a short-crested weir can be described by the equations of motion in the s- and n-direc-
tions whereby the problem of determining the local values of v and r is introduced.
m, like those involved in three-dimensional flow, is not tractable by existing
theory-and thus r e c o u r s _ e t ~ b e made to hydraulic model tests. -
7

U.S. Soil Conservation Cylindrical crastad

Service Protile Weir weir

Figure 1.18 Various types of short-crested weirs

39
 Thus experimental data are made to fit a head-discharge equation which is structurally
-
similar to that of a broad-crested weir b.utjn_which the dischape coefficient expresses
in addition to these factorssplaine_d in S
the influence ,of streamline curvature ____. s
m
*act, the same measuring structure can act as a broad-crested weir for low heads
(H,/L < 0.50), while with an increase of head (H,/L > 0.50) the influence of the
streamline curvature becomes significant, and the structure acts as a short-crested weir.
For practical purposes, a short-crested weir with rectangular control section has a
head-discharge equation similar to Equation 1-37, i.e.
Q = ['
C ' C - - g ]05'b, h,l.SO
'3 3
(1-64)

The head-discharge equations of short-crested weirs with non-rectangular throats are

structurally similar to those presented in Section 1.9. An exception to this rule is pro-
vided by those short-crested weirs which have basic characteristics in common with
sharp-crested weirs. As an example we mention the WES-spillway which is shaped
according to the lower nappe surface of an aerated sharp-crested weir and the triangu-
lar profile weir whose control section is situated- above a separation bubble down-
stream of a sharp weir crest.
Owing to the-gressure and velocity distributions above the sw a- indicated
-P-
i i F i s r e 1.19, the discharge coefficient (-cd)
of a short-crested weir TsAgher than
T h a t of a broad-creste-d-xeir,The rate of departure from the hydrostatic pressure d i s t r
bution is defined by the local centripetal acceleration v2/r (see Equation 1-10).

(1 -65)

Depending on the degree of curvature in the overflowing nappe, an underpressure

may develop near the weir crest, while under certain circumstances even vapour pres-
sure can be reached (see also Annex 1). If the overfalling nappe is not in contact with
the body of the weir, the air pocket beneath the nappe should be aeratedto avoid
an underpressure,wJhich increases the streamline curvatureroF-a
mre details on this aeration demand the reader is referred to Section 1.14.

I I /
I v; 129 I p
qE

Figure 1.19 Velocity and pressure distribution above a short-crested weir

40
1.11 Critical depth-flumes
A free flowing critical depth or standing wave flume is essentially a streamlined con-
striction built in an open channel where a sufficient fall is available so that critical
flow occurs in the throat of the flume. The channel constriction may be formed by
side contractions only, by a bottom contraction or hump only, or by both side and
bottom contractions.
The hydraulic behaviour of a flume is essentially the same as that of a broad-crested
weir. Consequently, stage-discharge equations for critical depth flumes are derived
in exactly the same way as was illustrated in Section 1.9.
In this context it is noted that the stage-discharge relationships of several critical
depth flumes have the following empirical shape
Q = C'h" (1 -66)
where C' is a coefficient depending on the breadth (b,) of the throat, on the velocity
head v2/2g at the head measurement station, and on those factors which influence
the discharge coefficient; h is not the water level but the piezometric level over the -
---
flume crest at a specified point in the converg-gApproach channel, and u is a factor
usually varying betw"3Zd-2.5 depending on the geometry of the control section
-
(see also Section 1.15).
Examples of critical depth flumes that have such a head-discharge relationship are
the Parshall flume, Cut-throat flume, and H-flume.

..

Photo 4 If H , / L < about 0.07, undulations may occur in the flume throat

41
Empirical stage-discharge equations of this type (Equation 1-66) have always been
derived for one particular structure, and are valid for that structure only. If such a
structure is installed in the field, care should be taken to copy the dimensions of the
tested original as accurately as possible.

1.12 Orifices
The flow of water through an orifice is illustrated in Figure 1.20. Water approaches
the orifice with a relatively low velocity, passes through a zone of accelerated flow,
and issues from the orifice as a contracted jet. If the orifice discharges free into the
air, there is modular flow and the orifice is said to have free discharge; if the orifice
discharges under water it is known as a submerged orifice. If the orifice is not too
close to the bottom, sides, or water surface of the approach channel, the water particles
approach the orifice along uniformly converging streamlines from all directions. Since
these particles cannot abruptly change their direction of flow upon leaving the orifice,
they cause the jet to contract. The section where contraction of the jet is maximal
is known as the vena contracta. The vena contracta of a circular
-
orifice is about half
thedkmeter
--
of the orifice itself. I_c___

e-fIre aischarging orifice shown in Figure 1.20 discharges under

the average head H I (if H , >> w) and that the pressure in the jet is atmospheric,
we may apply Bernoullis theorem
HI = (h, + vI2/2g)= v2/2g (1 -67)
Hence
v = ,/2gHI (1 -68)
This relationship between v and f i was first established experimentally in 1643
by Torricelli.

Figure I .20 The free discharging jet

42
I I

I Figure I .21 Rectangular orifice

If we introduce a C,-value to correct for the velocity head and a C,-value to correct
for the assumptions made above, we may write
V = Cd c, J2ghI (1 -69)
According to Equation 1-2, the discharge through the orifice equals the product of
the velocity and the area at the vena contracta. This area is less thanthe orifice area,
the ratio between the two being called the contraction coefficient, 6. Therefore
Q = C d C v 6 A m (1 -70)

i The product of cd, C, and 6 is called the effective discharge coefficient Ce. Equation
1-70 may therefore be written as
Q=C A m (1-71)
~

~
-
Proximity of a boundinp surface of the approach channel on one side of the orifice
erevents the free approach of water and the contraction is partially suppressed on
I
e.- If the orifice edge is flush with the sides or bottom of the approach channel,
the contraction along this edge is fully suppressed. The contraction coefficient, how-
ever, does not vary greatly with the length of orifice perimeter that has suppressed
I contraction. If there is suppression of contraction on one or more edges of the orifice
and full contraction on at least one remaining edge, more water will approach the
orifice with a flow parallel to the face of the orifice plate on the remaining edge(s)
and cause an increased contraction, which will compensate for the effect of partially
or fully suppressed contraction.

dQ = Ce b, ,/- dm (I -72)

43
The total discharge through the orifice is obtained by integration between the limits
O and hb- h,:

Q = C,b, hbT'
O
d m d m (1 -73)

or
2
Q = C, b, 3 . A(hb1.50
- h, (1-74)

e across a rectangular sharp-crested

qation 1 -71-is used for all orXces, incl=
viations from the theoretical equation being
--
If the orifice discharges u d m t K i F i s k n o w n as a submerged orifice. Flow of
water through a'submerged orifice is illustrated in Figure 1.22.

If we assume that there is no energy loss over the reach of accelerated flow, that the
streamlines at the vena contracta are straight, and that the flow velocities in the eddy
above the jet are relatively low, we may apply Bernoulli' s theorem
H, = + 4 1 + VI2/% = (P/pg + 4,+ v,2/2g (1 -75)
and since (P/pg + z), = h, we may write Equation 1-75 as
v, = {2g(Hl-h2)}050 (1-76)
Using a similar argument to that applied in deriving Equation 1-71 we may obtain
a formula that gives the total discharge through a submerged orifice as
Q = C, A{2g(h, - h,)}' 50 (1-77)

Figure 1.22 Flow pattern through a submerged orifice

44.
1.13 Sharp-crestedweirs QI/L '5

If the crest length in the direction of flow of a weir is short enough not to influence
the head-discharge relationship of this weir (HJL greater than about 15) the weir
is called sharp-crested. In practice, the crest length in the direction of-flow is generally
equal to or less than 0.002 m so that even at a minimum head of3 m the nappe
is complete!y free -we& body after Dassing-gh weir and no adheredEDg
q a n z I f th3low springs clear from the downstream face of the weir, an air pocket
forms beneath the nappe from which a quantity of air is removed continuously by
the overfalling jet. Precautions are therefore required to ensure that the pressure in
the air pocket is not reduced, otherwise the performance of the weir will be subject
to the following undesirable effects:
a. Owing to the increase of underpressure, the curvature of the overfalling jet will
increase, causing an increase of the discharge coefficient (C,,).
b. An irregular supply of air to the air pocket will cause vibration of the jet resulting
in an unsteady flow.
If the frequency of the overfalling jet, air pocket, and weir approximate each other
there will be resonance, which may be disastrous for the structure as a whole. To
prevent these undesirable effects, a sufficient supply of air should be maintained to
the air pocket beneath the nappe. This supply of air is especially important for sharp-
crested weirs, since this type is used frequently for discharge measurements where a
high degree of accuracy is required (laboratory, etc.).
Figure 1.23 shows the profile of a fully aerated nappe over a rectangular sharp-
crested weir without side contractions as measured by Bazin and Scimeni. figure
shows that for-a sharp-crested weir the concept of critical flow is not applicable. _For
the derivation of the head-discharge equations it is assumed that sharp-crested weirs
L

behave like orifices with a free water surface,and the following assumptions are made:

OA = O 2 5 H1
OD = O B 5 H1
O8 =OB7 H1
OF =1.38 H l
AC =0.112 Hl
CE = O 6 5 H1
CI ===0.22 H l
3
IJ =0.18 H1
v1 =0.475
v2 =0.946-

Figure 1.23 Profile of nappe of a fully aerated two-dimensional weir (after Bazin 1896 and Scimeni 1930)

45
i. the height of the water level above the weir crest is h = h, and there is no contrac-
tion;
ii. velocities over the weir crest are almost horizontal; and
iii. the apmoach velocity head ~ ~ ~ 1is2neglected.
.g
The;elociG=&ry point oTth-ection is calculated with the equation
of Torricelli, which was derived in Section 1.12 (Figure 1.24).
v = J2g(h, + vI2/2g- m) (1-78)

The total flow over the weir may be obtained by integration between the limits m =O
and m = h,
h,
Q = (2g)0.50f x(h, -m)o.50dm (1 -79)
O

where x denotes the local width of the weir throat as a function of m. After the introduc-
tion of an effective discharge coefficient, Ce, to correct for the assumptions made,
the general head-discharge equation of a sharp-crested weir reads (see also Section
1.12)
hl
Q = C,(2g)0.50 x(h, - m)0.50dm (1-80)
O

The reader should note that the assumptions made above deviate somewhat from
reality as shown in Figure 1.23 and are even partly in contradiction with the velocity
distribution as calculated by Equation 1-79. In practice, however, Equation 1-80 has
proved to be satisfatory and is widely used throughout the world. Since, also, the
effective discharge coefficient is almost constant, a different set of head-discharge
ived below f o i V a r i G kinds of sharp-crested weirs.

Figure I .24 Parameters of a sharp-crested weir

46
1.13.1 Sharp-crested weir with rectangular control section

For a rectangular control section, (Figure 1.25) x = b, = constant, Equation 1-80

may be written as
hl
Q = Ce(2g)O so b,(h, - m)O 50 dm (1-81)
O

or
n
L
Q = Ce- (2g)O. b, hl.50 (1 -82)
3
So, apart from a constant factor, Equation 1-82 has the same structure as the head-
discharge relation for a broad-crested weir with rectangular control section (Equation
1-37).

I I . ..
Figure I .25 Dimensions of a rectangular control section

1.13.2 Sharp-crested weir with parabolic control section

For a parabolic control section (Figure 1.26) x = 2@, and Equation 1-80 may
be written as
hl
Q = Q,(2g)0.s0 1 2(2fm(h, - m)}o.sodm (1-83)
O

After substituting m = h(l -cos a)/2,Equation 1-83 is transformed into

Q = Ce(2g)0.s0
2(2f)0.50
h
[$] 2 n

O
(1 - sin a d a

or

(1 -84)

In the above a was introduced for.mathematical purposes only.

47
 Ih,=h

Figure 1.26 Dimensions of a parabolic control section

1.13.3 Sharp-crested weir with triangular control section

For a triangular control section, (Figure 1.27) x = 2m tan 0/2, and Equation 1-80
may be written as
hl
Q = Ce(2g)0.50J [2 tan;] m (h,-m)0.50dm (1-85)
O

or
8 0
Q = Ce-(2g)0.50
tan- h,2.50 (1 -86)
15 2
So, apart from a constant factor, Equation 1-86 has the same structure as the head-
discharge relation for a broad-crested weir with triangular control section (Equation
1-47).

I
4/
/
J(
r-

Figure 1.27'Dimensions of a triangular control section

1.13.4 Sharp-crested weir with truncated triangular control section

The head-discharge relation for a truncated triangular control section as shown in

Figure 1.28 is obtained by subtracting the head-discharge equation for a triangular
control section with a head (h, -Hb) from the head-discharge equation for a triangular
rs, superimposing%
tion is allowed, pro-

The head-discharge equation (h, > H,) reads

8 0
Q C, - (1 -87)
=
15 (2g)O tan-2 [hI2 - (h, - HJ2

or
4 B
Q = Ce-(2g)O [hI250 - (h, - Hb)' ' O ] (1 -88)
15 Hb

If the head over the weir crest is less than H,, Equation 1-86 should be used to calculate
the discharge.

Figure I .28 Dimensions of a truncated triangular control section

1.13.5 Sharp-crested weir with trapezodal control section

The head-discharge relation for a trapezoidal control section as shown in Figure 1.29
is obtained by superimposing the head-discharge equations for a rectangular and trian-
gular control section respectively, resulting in

(1 -89)
Figure 1.29 Dimensions of a trapezodal control section

1.13.6 Sharp-crested weir with circular control section

For a circular control section as shown in Figure 1.30, the values for x, m, and dm
can be written as x = 2 r sin c1 = d, sin 2p = 2 d, sin p cos p
m = r( 1 - cos a) = d, sin2 p
dm = 2 d, sin p cos p dp

Substitution of this information into Equation 1-80 gives

bh
Q = C,(2g)0.50f (2d, sin p cos 0) (h, - d, sin2p)0.5dp (1 -90)
O
h
After introduction of k2 = 2 (being < 1) and some further modifications Equation
1-90 reads d,

Substitution of sin p = k sin $ and introduction of A$ = (1 k2 ~in~$)O.~

~ leads to

Figure 1.30 Dimensions of a circular control section

50
 'I2 sin2+
Q = Ce4(2g)o.sd,"[ J -d+-(1 + k2) 'I2j wsin4+
d+ +k2 j
O A+ O
. -
O
- - ___ - -.

Now the complete elliptical integrals K and E of the first and second kind respectively,
are introduced. K and E are functions of k only and are available in tables.

Y - -.O61 + ,668~+ 1.465~2

O ,l -2 .3 .4 .S .6 .7 .8 .9 1
hl/D

I , .WCiLY .",Y, ' 32 6x25 XUbL 01 1.641U 1.9384

.I8 .O914 ,1080 .53 ,7064 ,8344 .88 1.6699 1.9725
.I9 ,1014 ,1198 .54 ,7306 ,8630 .89 1.6988 2.0066
.20 ,1119 ,1322 .55 ,7551 ,8920 .90 1.7276 2.0407

.21 ,1229 ,1452 .56 ,7799 ,9212 .91 1.7561 2.0743

.22 ,1344 ,1588 .57 ,8050 ,9509 .92 1.7844 2.1077
.23 ,1464 ,1729 .58 ,8304 ,9809 .93 1.8125 2.1409
.24 .I589 .1877 .59 ,8560 1.0111 .94 1.8403 2.1738
.25 ,1719 ,2030 .60 ,8818 1.0416 .95 1.8678 2.2063

.26 ,1854 .2190 .61 ,9079 1.0724 .96 1.8950 2.2384

.27 ,1994 ,2355 .62 ,9342 1.1035 .97 1.9219 2.2702
.28 .2139 .2527 .63 .9608 1.1349 .98 1.9484 2.3015
.29 ,2289 ,2704 .64 .9876 1.1666 .99 1.9144 2.3322
.30 ,2443 ,2886 .65 1.0147 1.1986 1.00 2.000 -

.31 .2601 ,3072 .66 1.0420 1.2308 4

.32 ,2763 ,3264 .67 1.0694 1.2632 Q = C , Is fi or
.33 ,2929 ,3460 .68 1.0969 1.2957
.34 ,3099 ,3660 ' .69 1.1246 1.3284 Q = Ce +i d'.:
.35 ,3273 ,3866 .70 1.1524 1.3612

Values of o from Stevens 1957

51
 'I2 d$
E= S -A$
O
(1 -93)

(1-94)

For the separate integrals of Equation 1-92 the following general reduction formula
can be derived (n being an arbitrary even number)

' I 2 sinn$ - n-2 1 + k2 ' I 2 -d$--- n-3 1 ' I 2 sinn4$

J ___ (1 -95)
n-l k2 A$ n-1 k2 A$ d$
O O O

Combinations of Equations 1-92, 1-93, 1-94, and 1-95 gives

4
Q = Ce-(2g)O.' dc2.5(2( 1 - k2 + k4)E - (2 - 3k2 + k4) K} ( I -96)
15
or
4
Q = Ce- o = Ce 4, dc2.5
(2g)0.5dc2.5 (1-97)
15
Equation 1-97 was first obtained by Staus and Von Sanden in 1926.
4
Values of o = {2(1 - k2 + k4) E - (2 - 3k2 + k4) K} and of +i = -
15
(2g)0.5o are
presented in Table 1.3.

1.13.7 Sharp-crested prbportional weir

A proportional weir is defined as a weir in which the discharge is linearly proportional

to the head over the weir crest. In other words, the control section over a proportional
weir is shaped in such a way that the sensitivity of the weir

;:; = 1.0 (1 -98)

In order to satisfy this identity the curved portion of. the weir profile must satisfy
. ~ a constant), so that the theoretical head-discharge equation,
the relation x = ~ n ' ) (cis
according to Equation 1-80, reads

Q = Ce(2g)0-5c J
h
-;[ I r 5 dn (1 -99)
O

Substitution of a new dummy variable

modification, to
p into tan p. = -:[ I]
0.5
leads, after some

7c
Q = C,(2g)0.5c - h, (1-100)
2
This mathematical solution, however, is physically unrealizable because of the infinite

52
wings of the weir throat at n = O. To overcome this practical limitation, Sutro (1908)
proposed that the weir profile should consist of a rectangular portion at the base of
the throat and a curved portion above it, which must have a different profile law
to maintain proportionality.

The discharge through the rectangular section under a head h, above the weir crest
equals, according to Equation 1-82
L
Q, = Ce3 (2g)O. b, [hl.5- ho.] (1-101)

where b, equals the width of the rectangular portion, ho = (h,-a) equals the head
over the boundary line CD, and a equals the height of the rectangular portion of
the control section as shown in Figure 1.3I . The discharge through the curved portion
of the weir equals according to Equation 1-80
h0
Q, = Ce(2g)0. 1 (ho - n)0.5xdn (1-102)
O

Thus the total discharge through the weir equals

ho
Q = 2 +
Q r + Q,= Ce(2g)0[~b,(h,~-h0~5) (h0-n)05xdn] (1-103)
O

The discharge through the weir must be proportional to the head above an arbitrarily
chosen reference level situated in the rectangular portion of the weir. The reference
level A B is selected at a distance of one-third of the rectangular portion above the
weir crest to facilitate further calculations. So the total discharge through the weir
also reads
Q = K(h, -a/3) (1-104)
where K is a weir constant. Since proportionality is valid for heads equal to or above
the boundary line CD, it must hold also if ho = O. Substitution of ho = O and h, = a
into Equations 1-103 and 1-104gives

Figure I .31 Dimensions of a proportional Sutro weir notch

53

A
 2
Q = Ce- (2g)0.5b, a'.5 and
3
2
Q=-Ka
3
Consequently the weir constant equals
K = Ceb,(2ga)'.' . ( 1- 1 05)
Substitution of the latter equation into Equation 1-104 gives
Q = C, (2ga)0.5b,(h, - a/3) (1-106)
as a head-discharge equation. The relationship between x and n' for the curved position
of the weir can be obtained from the condition that Equations 1-103 and 1-106 should
be equal to each other, thus
h0
2
-
3
b, [h,1.5
- +J (ho- n')0.5xdn' = b,a0.5(h, - a/3)
O

From this equation h, and hocan be eliminated and the following relationship between
x and n can be obtained (Pratt 1914).

X/b, = 1 ( 1 - 1 07)

1.14 The aeration demand of weirs

Under those circumstances where the overfalling jet is not in contact with the body
of the weir, an air pocket exists under the nappe from which a quantity of air is removed
continuously by the overfalling jet. If the air pocket is insufficiently aerated, an under-
pressure is created. This underpressure increases the curvature of the nappe. One of
the results of this feature is an increase of the discharge coefficient (CJ. For a given
head (h,) the discharge is increased, and if the discharge is fixed, the measured head
over the weir is reduced. Obviously, this phenomenon is not a desirable one as far
as discharge measuring weirs are concerned.
Based on data provided by Howe (1955) the writers have been able to find a relation-
ship that gives the maximum demand of air (qJ required for full aeration in m3/s
per metre breadth of weir crest as

(1-108)

where q, equals the unit discharge over the weir, h, is the head over the weir, and
yp equals the water depth in the pool beneath the nappe as shown in Figure 1.32.
The poolwater depth yp is either a function of the tailwater level or of the unit discharge
q, and the drop height AZ. If a free hydraulic jump is formed downstream of the
weir, ypmay be calculated with Equation 1- 109, which reads

54
 eny- level
y---

required air
SUPPIY:q,ir

Figure I .32 Definition sketch aeration demand

0.22
YP = AZ(&) (1-109)

The dimensionless ratio q2/gAz3is generally known as the drop number. If the jump
downstream of the weir is submerged, the poolwater depth may be expected to be
about equal to the tailwater depth; yp N y2.
I
As an example we consider a fully suppressed weir with a breadth b, = 6.50 m and
~

water discharging over it under a head hl = 0.60 m, giving a unit discharge of 0.86
I m3/s per metre, while the pool depth yp = 0.90 m. Equation 1-108 gives the maximum
air demand for full aeration under these conditions as

qair= 0.1 - 0.047 m3/s per metre

(0.90/0.60)1.5-
or 6.5 x 0.047 = 0.305 m3/s for the'full breadth of the weir. The diameter of the
air vent(s) to carry this air flow can be determined by use of the ordinary hydrodynami-
cal equations, provided the underpressure beneath the nappe is low so that the mass
density of air (pair)can be considered a constant. In calculating the air discharge, how-
ever, the effective head over the vent must be stated in metres air-column rather than
in metres water-column. For air at 20C, the ratio pair/pwater equals approximately
11830.
To facilitate the flow of air through the vent(s) a differential pressure is required
over the vent, resulting in an underpressure beneath the nappe. In this example we
suppose that the maximum permissible underpressure equals 0.04 m water column.

Suppose that the most convenient way of aeration is by means of one steel pipe 2.50
m long with one right-angle elbow and a sharp cornered entrance; the head-loss over
the vent due to the maximum air discharge then equals
fL + K, + K,, vfir
5 = P-.[K, + - (1-110)
Pg Pw IZ DP
where

55
P,/pg = permissible underpressure beneath the nappe in metres water-column
K, = entrance loss coefficient (use K, = 0.5)
f = friction coefficient in the Darcy-Weisbach equation, being:
h, = f(L/D) (v2/2g).Use f = 0.02
L = length of vent pipe
Dp = diameter of vent pipe
K, = bend loss coefficient (use K, = 1.1)
K,, = exit loss coefficient (use K,, = I .O)
v,, = average flow velocity of the air through the vent pipe.
According to continuity, the total flow of air through the vents
1
4
Qair = b cq air = - D,2 vair (1-1 11)

Substitution of the data of the example and the latter equation into Equation 1-110
gives
0.305*
0.04 =
830
~

'[2.6 + '.O2
DP
2.50] 12.14 D;

so that the internal diameter of the vent pipe should be about O. 16 m.

An underpressure beneath the nappe will deflect the nappe downwards and thus
give a smaller radius to the streamlines, which results in a higher discharge coefficient.
Consequently, for a measured head over the weir crest h,, the discharge will be greater
than the one calculated by the head-discharge equation (see Annex I).

Photo 5 Non-aerated air pocket

56
Photo 6 Fully aerated air pocket

positive percentage e r r o r
in the calqulated discharge
XQ
100
80
60

40

20

10
8
6

1
0.8
0.6

0.4

0.2

o 1.
0.001 0.032 OD04 0.01 OD2 OD4 OD6 0.1 0.2 0.4 0.6 0.8 1
ratio of underpressure beneath the nappe over upstream head p2/pgh,

Figure I .33 Increment of the discharge over a rectangular weir with no side contractions (after data from
Johnson, Hickox and present writers)
Based on experimental data provided by Johnson (1935), Hickox (1942) and our own
data a curve has been produced on double log paper (see Figure 1.33), resulting in
the following empirical formula for the positive percentage error in the discharge
X, = 20(P,/pghl)092 (1-112)
In our example, where P,/pg = 0.04, and h, = 0.60 m; the ratio P2/pghl = 0.067,
resulting in a positive error of 1.7% in the discharge. Figure 1.33 shows that if the
underpressure beneath the nappe increases, due to underdimensioning of the air-
vent(s), the percentage error in the discharge increases rapidly, and the weir becomes
of little use as a discharge measuring device.

1.15 Estimating the modular limit for long-throated flumes

1.15.1 Theory

The fundamental condition for flow at the modular limit is that the available loss
of head between the channel cross-sections where the upstream head, H l , and the
downstream head, H,, are to be determined, is just sufficient to satisfy the requirement
for critical flow to occur at the control section. This situation will be analyzed by
dividing this minimum loss of energy head, H I- H2,into three parts (Bos 1985):
1. The energy head loss, H I - H,, between the upstream head measurement section
(gauging station) and the control section in the flume throat (Section 1.15.2);
2. The energy losses, AHf, due to friction between the control section and the down-
stream head measurement section (Section 1.15.3);
3. The losses, AH,, due to turbulence in the diverging transition (Section 1.15.4).
Figure 1.34 indicates the lengths of those parts of the structure for which these three
energy losses are to be calculated.

t$----length tor H,-Hc -e

l ngth for
4 - tHA

I< L / 3 L length for A

H
4d
-
I 1

;'

o e
v-

Figure 1.34 Lengths of structure parts for which H l - H,, AHf, and AH, are to be calculated

1.15.2 Energy losses upstream of the control section

The head-discharge relationship for a rectangular, parabolic, or triangular control

section, and for parts of all other control section shapes, can be written in the exponen-
tional form

58
65

y 801 - y 801
(SI 1.1) =n
801 - 0801

(PI 1.1) _.-

YP b
OP
 a' Qb +flow rate

Figure 1.35 Illustration of terms in Equations 1. I 14 and I . 1 15

As stated when Cd was being introduced in Equation 1.36, its value follows from the
need to correct for:
i. Energy losses between the gauging station and the control section;
ii. The effect of curvature of streamlines in the control section;
iii. The non-uniformity of the velocity distribution in both sections.
For heads that are low with respect to the throat length, the influence of streamline
curvature and of the non-uniformity of the velocity distribution is negligible with res-
pect to the energy losses (Ackers and Harrison 1963; Replogle 1975; Bos 1985; Bos
and Reinink 1981). Consequently, it can be assumed that Cdonly expresses the energy
losses between the gauging station and the control section. Acting on this assumption
and replacing h, by H, in Equation I . 1 14 results in:
Q = KH," (1.116)
Combining Equations I . 1 13 and 1. I 16 gives
H: = CdHIU (1.117)
which can also be written as.(Bos 1985)
HI-H, = Hl (1 -Cd'") (1.118)
The right-hand member of this equation approximates the loss of hydraulic energy
between the gauging and control sections. This equation, however, is only valid if
the influence of streamline curvature at the control section on the Cd value is insignifi-
cant.

1.15.3 Friction losses downstream of the control section

Although flow is non-uniform in the diverging transition, the energy losses due to
friction are estimated by applying the Manning equation to the three reaches shown
in Figure 1.34.

60
I . Reach of the flume throat downstream of the control section; the length of this
reach is held at L/3;
2 . Length of the reach of the actual diverging transition of bottom and side walls,
Ld
3. Length of a canal reach from the end of the transition to the measurement section
of the downstream sill-referenced head (Le 2: 5 y2).
So that

AHthroat =
1
7 (&) 2
(1.119)

AH,ram = Ld
( nQ > ( I . 120)

(1.121)

In the calculation of AH,,,,, and average area of flow, A, = (Ac + A2)/2 can be used.
The n value in each of the equations depends on the construction material of the related
reach of the structure and canal. The total energy losses due to friction, A H , between
the control section and the section where h, is measured then equals the sum of the
losses over the three reaches
AHthroat -k AHtrans + AHcanal (1.122)
In contrast to the dimensions of the area of flow in the approach channel and the
control, the dimensions of the downstream area of flow depend on the unknown value
of H,. The calculation of the modular limit therefore requires the solution by iteration
of an implicit function of the downstream head (Section I. 15.6).

1.15.4 Losses due t o turbulence in the zone of deceleration

In the diverging transition, part of the kinetic energy is converted into potential energy.
The remainder is lost in turbulence. With flow at the modular limit, losses due to
turbulence in the hydraulic jump are low (Peterka 1958) so the simple classical expres-
sion of Borda for energy losses in an expansion of a closed conduit can be used

( 1.123)

in which
5 = the energy loss coefficient, being a function of the expansion ratio of the
diverging transition;
v,-v2 = decrease in average flow velocity between the control section and the
downstream head measurement section.
Here again, v2 depends on the unknown downstream head, H,, so that the solution
of Equation I . 123 is part of the iteration process (Section 1.15.6).
International literature contains few data that allow the measured total energy loss

61
 5 10 15
expansion ratio EM of bottom and/or SI

Figure I .36 Values of 5 as a function of the expansion ratio of downstream transition (Bos and Reinink
1981)

over flumes to be broken down into the above three parts and permit 5-values to
be calculated. Blau (1960), Engel (1934), Inglis (1929), and Fane (1927), however,
published sufficient data on the geometry of structures and channels to allow the total
head loss, AH, to be broken down into a friction part and a turbulence part. The
calculated 5 values that were obtained from this literature are shown in Figure 1.36.
They correspond with the 5 values for the B and C series of the experiments conducted
by Bos and Reinink (1981).

1.15.5 Total energy loss requirement

The total energy loss over a flume or weir at the modular limit can be estimated by
adding the three component parts as discussed in the preceding sections:
HI - H, = Hl (1 - Cd'/") + AHf + 5 (v, - ~,)~/2g (1.124)
For the considered rate of flow through the structure, this equation gives the minimum
loss of energy head required for modular flow. That part of the above equation which
expresses the sum of the energy losses due to friction, H, (1 - Cdl/")+ AHr, becomes
a large percentage of the total energy loss, H I - H,, when diverging transitions are

62
long (high AHf values). This is mainly because the relatively high flow velocities in
the downstream transition are maintained over a greater length. On the other hand,
very gradual downstream transitions have a favourable energy conversion (low 6
value). As a result, very gradual transitions may, as a whole, lose more energy than
more rapid but shorter transitions. Since, in addition, the construction cost of a very
gradual transition is higher than that of a shorter one, there are good arguments in
favour of limiting the ratio of expansion to about 6-to-l.
Rather sudden expansion ratios like I-to-I or 2-to-1 are not effective because the
high velocity jet leaving the throat cannot suddenly change direction to follow the
boundaries of the transition. In the resulting flow separation zones, turbulence con-
verts kinetic energy into heat and noise. If for any reason the channel cannot accommo-
date a fully developed gradual transition of 6-to- 1 it is recommended that the transition
be truncated to L d = Hlmaxrather than to use a more sudden expansion ratio (see
Figure 1.37 and Photo 7). The end of the truncated expansion should not be rounded,
since it guides the water into the channel boundary; a rounded end causes additional
energy losses and possible erosion.

The modular limit of a weir or flume can be found by dividing both sides of Equation
1.124 by H I ,giving
(H2/Hl)ath4L = Cdlu-AHf/HI-6 (vc-v2)2/2gHl ( 1.125)
Equation 1.125 is a general expression for the modular limit of any long-throated
flume, and is also valid for the hydraulically similar broad-crested weir.

flow
t not recommended

I
throat I I

recommended truncot ion

I I
I I
I I
i

Figure 1.37 Truncation of a gradual downstream transition (Bos, Replogle & Clemmens 1984)

63
1.15.6 Procedure t o estimate the modular limit

Modularity of a discharge measurement structure in a given channel implies that the

required head loss for modular flow must be less than the available head loss. The
required head loss, however, can only be estimated after the type and dimensions of
the structure have been chosen.
As will be explained in Chapter 3, this choice is also governed by other considera-
tions, such as range of flows to be measured, and accuracy. An optimal solution to
the design problem can only be found in an iterative design procedure. Part of this
procedure is estimating the modular limit for Q,, and Qm,, of the tentative design
in the relevant design loop.
In the following procedure to estimate the modular limit, it is assumed that the
relationship between Q, h,, and Cd is known. To estimate the modular limit of a weir
or flume in a channel of given cross-section, both sides of Equation 1.125 must be
equalized as follows:
1. Determine the cross-sectional area of flow at the station where h, is measured,
and calculate the average velocity, v,;
+
2. CalculateH, = h, vl2/2g;
3. For the given flow rate and related head, note down the C d value;
4. Determine the exponent u;
For a rectangular (u = 1.5), parabolic (u = 2.0), or triangular control section
(u = 2.5), the power u is known from the head-discharge equation. For all other
singular or composite control shapes, use Equation 1.114 or 1.115;
5. Calculate cd;
6. Use Section 1.9 to find yc at the.control section. Note that y, is a function of
H , and of the throat size and shape;
7. Determine the cross-sectional area of flow at the control section with the water
depth, y,, and calculate the average velocity, v,;
8. Use Figure 1.36 to find an 6 value as a function of the angle of expansion;
9. Estimate the value of h, that is expected to suit the modular limit and calculate
A, and the average velocity v,;
10. Calculate E,(v, -vJ2/2gHI;
+
1 1. Determine AHf = AHthroa, AH,,,, + AHcanal by applying the Manning equation
with the appropriate value of n to L/3 of the throat, to the transition length, and
to the canal up to the h2measurement section (see Section 1.15.3);
12. Calculate AHf/HI;
13. Calculate H, = h, + v,2/2g;
14. Calculate H,/H,;
15. Substitute the values (3,(lo), (12), and (14) into Equation 1.125;
16. If Equation 1.125 does not match, repeat steps (9) through (15).
Once some experience has been acquired, Equation 1.125 can be solved with two or
three iterations. Since the modular limit varies with the upstream head, it is advisable
to estimate the modular limit at both minimum and maximum anticipated flow rates
and to check if sufficient head loss (Hl - H,) is available in both cases.

64
1.16 Modular limit of short-crested weirs
As mentioned in Section 1.10 streamline curvature above a short weir crest causes
a non-hydrostatic pressure distribution. Because of the related velocity distribution
(see Figure 1.19) the discharge per unit width of a short-crested weir is more than
the discharge over a broad-crested weir operating under the same head, h,.
On the other hand, however, the degree of streamline curvature is influenced by
the elevation of the tailwater channel bottom and by the water level in this tailwater
channel. A high tailwater level will reduce streamline curvature and thus also reduce
the weir discharge. As a result the modular limit of a short-crested weir is less than
that of a broad-crested weir. As a general rule it may be said that there is a direct
relationship between the values of Cd and the modular limit. Figure 1.38 shows this
relationship for some common weir profiles.

1.5- -1.0

1.4 - - 0.9

1.3 - - 0.8

1.2 - - 0.7

1.1 - - 0.6

1.0- - 0.5

0.9- - 0.4

0.8 - - 0.3

Figure I .38 Influence of shape of weir crest and related streamline curvature on C , and modular limit (Bos
1978)

1.17 Selected list of literature

Ackers, P. and Harrison, A.J.M. 1963. Critical depth flumes for flow measurements in open channels.
Department of Industrial and Scientific Research, Hydraulic Research Station, Wallingford, U.K. Hyd-
raulic Research Paper 5. 50 pp.
Bazin, H.E. 1896. Expriments nouvelles sur Icoulement en dversoir. Annales des Ponts et Chausses.
V01.7, pp.249-357.

65
Blau, E. 1960. Die modelmisige Untersuchung von Venturikanlen verschiedener Grsse und Form. Verf-
fentlichungen der Forschungsanstalt fur Schiffahrt, Wasser und Grundbau 8. Akademie-Verlag, Berlin,
German Democratic Public.
Bos, M.G. 1977. The use of long-throated flumes to measure flowsin irrigation and drainage canals. Agricul-
tural Water Management, Elsevier, Amsterdam. Vol. I: 2: pp. 1 11-126.
Bos, M.G. 1978. De selectie van meet- en regelkunstwerken in waterlopen. (The selection of measurement
and control structures in channels) Cultuurtechnisch Tijdschrift nr. 4, LD, Utrecht, The Netherlands.
Bos, M.G. 1985. Long-throated flumes and broad-crested weirs, Nijhoff, Dordrecht, The Netherlands,
p. 141.
Bos, M.G. and Y. Reinink, 1981. Head loss over long-throated flumes. Journal of the Irrigation and Drain-
age Division, American Society ofcivil Engineers. Vol. 107: I R I . pp. 87-102.
Bos, M.G., J.A. Replogle and A.J. Clemmens 1984. Flow measuring flumes for open channel systems.
John Wiley, New York. 321 pp.
Carlier, M. 1972. Hydraulique gnrale et applique. Collection du Centre de recherches et d'essais de
Chatou. Eyrolles, Paris.
Clemmens, A.J., J.A. Replogle and M.G. Bos 1987. Flume: a computer model for estimating flow rates
through long-throated measuring flumes. U.S. Dept. of Agriculture, ARS-57. p. 64.
Engel, F.V.A.E. 1934. The Venturi flume. The Engineer. Vol. 158, August 3 . pp. 104-107. August IO, pp.
131-133.
Fane, A.B. 1927. Report on flume experiments on Shirhing Canal. Punjab Irrigation Branch. Paper 110:
Punjab Engineering Congress, Bombay. pp. 37-51, plate A G .
Formica, G. 1955. Esperienze preliminari sulk perdite di carico nei canali, dovute a cambiamenti di sezione.
L'Energia elettrica. Milano, Vo1.32, No.7. pp.554-568.
Henderson, F.M. 1966. Open channel flow. The Macmillan Company, New York.
Hickox, G.H. 1944. Aeration of spillways. Transactions of the American Society of Civil Engineers. V01.109,
pp.537-556.
Howe, J.W., G.C. Shieh and A.O. Obadia. 1955. Aeration demand of a weir calculated. Civil Engineering
Vo1.25, N O S , p.289. Easton, Pa. &Personal Communication, 1972 (Howe).
Idelcik, I.E. Memento des pertes de charge. Coll.du Centre de recherche et d'essais de Chatou. Eyrolles,
Paris, 1969. (Transl.from Russian.)
Inglis, C.C. 1928. Notes on standing wave flumes and flume meter falls. Technical Paper 15. Public Works
Department, Bombay. 35 pp.
King, H.W. and E.F. Brater. Handbook ofhydraulics. 5 " Edition. McGraw-Hill Book Comp.
Knapp, F.H. 1960. Ausfluss, Uberfall und Durchfluss im Wasserbau. Verl. G. Braun, Karlsruhe.
Pratt, E.A. 1914. Another proportional-flow weir; Sutro weir. Engineering News, Vo1.72, No.9, p.462.
Peterka, A.J. 1958 (revised 1964). Hydraulic design of stilling basins and energy dissipators. U.S. Depart-
ment of the Interior, Bureau of Reclamation, Washington D.C. 222 pp.
Replogle, J.A. 1975. Critical flow flumes with complex cross-section. American Society of Civil Engineer.
Specialty Conference Proceedings: Irrigation and Drainage in an Age of Competition for Resources.
Logan,Utah, U.S.A. Aug. 13-15,pp. 366-388.
Rouse, 1938. H. Fluid Mechanics for hydraulic engineers. Dover Publications, Inc. New York. Reprint
1961.
Rouse, H. 1950. Engineering Hydraulics. John Wiley & Sons, Inc., New York. Fourth printing 1964.
Scimeni, E. 1930. Sulla forme delle vene tracimani. L'Energia elettrica, Milano, Vo1.7, No.4, pp.293-305.
Stevens, J.C. 1957. Flow through circular weirs. J.of the Hydraulics Division of the American Society of
Civil Engineers, Vol. 83, No. H Y 6. Paper 1455.
Ven Te Chow. 1959.Open-channel hydraulics. McGraw-Hill Book Comp. New York.

66
2 Auxiliary equipment for measuring
structures

2.1 Introduction
Most structures built for the purpose of measuring or regulating discharges consist
of a converging section with accelerating subcritical flow, a control section with a
transition to supercritical flow, and a downstream transition where the flow velocity
is reduced to an acceptable value.
Upstream of the structure is an approach channel, which influences the velocity
distribution of the approaching flow. Downstream of the structure is a tailwater chan-
nel, which is of fundamental importance in the design of the structure because of the
range of tailwater levels that will result from varying discharges.
The difference in elevation between the crest of the control section and the piezo-
metric head in the approach channel is known as the upstream head over the crest
of the structure and is denoted by h,. If the structure is located in a channel where
the discharge is determined upstream, h, corresponds with the discharge and the struc-
ture serves as a measuring device only. If the structure is located at a canal bifurcation,
h, can be altered by moving the weir crest so that the structure can be used both
as a measuring and as a regulating device. The upstream head over the crest can be
determined by reading the water surface elevation in the approach channel on a staff

Figure 2. I General lay-out of a discharge measurement structure

67
gauge whose gauge datum elevation coincides with the crest of the structure. Determin-
ing the gauge datum elevation is generally known as zero-setting and this should be
repeated at regular intervals to avoid serious errors in the measurement of h,. That
part of the approach channel where the water surface elevation is measured is known
as the head measurement or gauging station.

2.2 Head measurement station

The head measurement station should be located sufficiently far upstream of the struc-
ture to avoid the area of surface draw-down, yet it should be close enough for the
energy loss between the head measurement station and the structure to be negligible.
This means that it will be located at a distance equal to between two and four times
h, max from the structure. For several standard measuring flumes, this general rule has
been disregarded and the piezometric head is measured at a well-prescribed point in
the converging section where there is a significant acceleration of flow. Thus the mea-
sured piezometric head is lower than the real upstream head over the crest, which
hampers the comparison of stage-discharge equations and the minimum required loss
of head (modular limit, see also Section 1.8). The stage-discharge relationship of such

Photo 1 The elevation of a movable weir can be read from a fixed gauge

68
Photo 2 Sharp-nosed intermediate piers tend to trap floating trash

flumes can only be obtained by laboratory calibration (tables and/or formulae). The
only advantage of this procedure is that an approach velocity coefficient is not needed.
The water level upstream of the structure may be measured by a vertical or an in-
clined gauge. A hook, point, or staff gauge can be used where incidental measurements
are required, or a float-operated recording gauge where a continuous record is needed.
Regardless of the type of gauge used, it should be located to one side of the approach
channel so that it will not interfere with the flow pattern over the structure.

2.3 The approach channel

All structures for measuring and regulating discharges require an approach channel
with a flow free from disturbance and with a regular velocity distribution. This can
be obtained by having a straight section free of projections at the sides and on the
bottom. The channel should have reasonably uniform cross-sections and be straight
for a length equal to approximately 10 times its average width,.provided that the
breadth of the control section is equal to or greater than half the width of the approach
channel. If the breadth of the control section is less than this, the length of the approach
channel should be at least 20 times the breadth of the control section. In canals that
carry no debris, the desired flow conditions can be provided by suitably placed baffles
formed by vertical vanes or laths. These baffles should not be located nearer to the
head measurement station than 10 times h,.
If super-critical flow occurs upstream of the structure, a hydraulic jump should
be introduced to ensure a regular velocity dist.ribution at the head measurement sta-

69
tion. This jump should be located at a distance of not less than 30 times h, from the
structure.
In cases where the entry to the converging section is through a bend, where the
approach channel is too short, or where a hydraulic jump occurs within the distance
mentioned above, either the approach channel must be modified or the structure must
be calibrated in situ, for example by use of the velocity-area method or salt dilution
method.

2.4 Tailwater level

The difference between the water level immediately below the downstream transition
(tailwater level) and the elevation of the crest of the structure is known as the down-
stream head over the crest and is denoted by h,. Tailwater level, and thus the submer-
gence ratio h,/h,, is affected by the hydraulic properties of the tailwater channel and
by the occurrence of transitions in that channel.
The measuring structure should be so designed that modular flow is maintained
under all operating conditions. If there is only a limited head loss available, both the
elevation of the crest in relation to the downstream energy level and the length and
shape of the downstream transition should be selected in such a way that modular
flow is ensured (Section 1.15).
If the tailwater channel is relatively wide or if the tailwater level is affected by a
downstream structure, it may occur that the measuring structure is modular at its
maximum design capacity, but non-modular with lesser discharges. Under such cir-
cumstances a decrease in the upstream head means an increase in the submergence
ratio h,/h,. The crest of the control section should then be raised so that h,, and thus
the ratio h,/h,, decreases to below the modular limit.
If the measuring structure is modular over its entire operating range, it is not necess-
ary to make tailwater measurements (see Section 1.8). If the flow conditions are non-
modular, however, both h, and h2must be recorded to allow the discharge to be caicu-
lated. The tailwater level should be measured immediately downstream of the decelera-
tion transition where normal channel velocities occur. The equipment to be used for
this purpose may be the same as that used for measuring the upstream water level
or it may be of a lower accuracy, and thus more simple, depending on the frequency
with which submerged flow occurs (see also Section 2.12).
It is evident that collecting and handling two sets of data per measuring structure
is an expensive and time-consuming enterprise, which should be avoided as much as
possible. Other even more important reasons for applying a modular structure are
that in an irrigation canal system a water user with his own canal inlet cannot increase
the discharge by lowering the tailwater level while, on the other hand, all persons
concerned have a simple way of checking whether they receive their proper share of
the available water.

2.5 Staff gauge

Where no detailed information on the discharge is needed or in stream channels where

70
the flow fluctuation is gradual, periodic readings on a calibrated staff gauge may pro-
vide adequate data. A staff gauge should also be provided if the head is registered
by a float-operated recorder as it will enable comparison of the outside water level
with the head in the float well.
Supports for the staff gauge should not interfere with the flow pattern in the struc-
ture, and should be independent of the stilling well. Most permanent gauges are plates
of enamelled steel, cast aluminium, or polyester, bolted or screwed in sections to a
timber or steel pole. A typical gauge is shown in Figure 2.2.
The gauge should be placed in such a manner that the water level can be read from
the canal bank. Care should be taken that the staff gauge is firmly secured. The follow-
ing type of support has proved satisfactory for permanent installations: a section of
180 mm channel iron is embedded about 0.50 m in a concrete block and extended

I n I

MO

5mrr

7
-74
Figure 2.2 Typical staff gauge

71
above the block to the maximum height required. The concrete block should extend
well below the maximum expected frost penetration and at least 0.60 m below the
minimum bed level of a natural stream. The top of the block should be O . 10 m below
the lowest head to be measured. A staff of durable hardwood, 0.05 x O. 15 m, is bolted
to the channel iron above the concrete block, and the enamelled gauge section is fas-
tened to this staff with brass screws. Staffgauges may be fastened to any other support-
ing structure, provided that its elevation is constant.

2.6 Stilling well

The primary purpose of a stilling well is to facilitate the accurate registration of a
piezometric or water level in open channels where the water surface is disturbed by
surges or wave action.
The stilling well should be vertical and of sufficient height and depth to cover the
entire range of expected water levels. In natural streams it should have a minimum
margin of 0.60 m above the estimated maximum level to be recorded. In canals the
minimum margin should be equal to the canal freeboard. Whenever the stilling well
is used in combination with a float-operated recorder, it is common practice to extend
the well to about 1.00 m above ground/platform level, so that the recorder can be
placed at a suitable working height.
Care should be taken to ensure that if the float is rising its counterweight does not
land on top of the float, but keeps well above it or passes the float. If a high degree

Photo 3 A stilling well made of steel pipes

72
of accuracy is required, the counterweight should not be permitted to become sub-
merged over part of the operating range since this will change the submergence rate
of the float and thus affect the recorded water level. This systematic error may be
prevented (i) by locating the counterweight inside a separate water-tight and water-free
pipe, (ii) by mounting two different-sized wheels on the axle of the recorder, the large-
diameter wheel serving to coil up the float wire and the small-diameter wheel coiling
up the counterweight wire, (iii) by extending the stilling well pipe to such a height
that the counterweight neither touches the float wheel at low stage nor the water sur-
face at maximum expected stage.

The cross-sectional dimensions of the well depend on a number of factors: (i) whether
a dip-stick, staff gauge, pressure logger, or a float-operated recorder is used, (ii) type
of construction material, (iii) height of the well, (iv) possible protection against free-
zing, (v) required stability, (vi) the necessity to have access to the inside.
If the well is used in combination with a dip-stick, a minimum diameter of 0.10 m
to 0.15 m is advised to give access to a hand. A reference point, on which the stick
will rest and whose elevation coincides with the exact crest elevation, is provided inside
the well. A dip-stick can supply very accurate information on head.
If the well is used in combination with a staff gauge, the length of the well, as mea-
sured from the face of the gauge, should not be less than twice the depth to minimum
water level in the well. The well width should not be less than 0.20 m to allow sufficient
room for the gauge to be fixed by screws to the side of the well.
If a pressure logger is used, the well should be about 1.5 times larger than the logger.
A minimum diameter of O . 10 m is recommended.

structure
- pref abr i c a t ed
metal structure

I
-=-+
/intake

t c o n c r e t e plug

Figure 2 3 Examples of a stilling well used in combination with a dip-stick

73
Figure 2.4 Stilling well used in combination with a staff gauge

If the well is to accommodate the float of an automatic water level recorder, it should
be of adequate size and depth to give clearance around the float at all stages. If the
well is a metal, PVC, or concrete pipe, its diameter should be 0.06 m larger than the
diameter of the float to avoid capillary effect; if the well is rectangular and constructed
of brickwork, concrete, wood, or similar materials, the float should not be nearer
than 0.08 m to the wall of the well. The bottom of the well should be some distance,
say 0.15 m, below the lowest intake, to avoid the danger of the float touching the
bottom or any silt that might have accumulated. This silt should be removed at regular
intervals. In general, an access door should be provided to allow the recorder setting
to be checked and to permit the removal of silt without the well having to be entered.
If the well is set back into the channel embankment, the access door should be
placed just above the embankment; if the well is installed in the channel, the door
should be placed just slightly above low water. A second access door will allow the
float tape length to be adjusted and gears to be changed without the recorder having
to be removed. To avoid corrosion problems, it is recommended that the hinges of
these access doors be of a rust-resistant metal such as stainless steel, brass, or bronze.
A more simple solution is to support the door by wing nuts on short bolts welded
to the well.
The foundation level of both the structure and the stilling well should be well below
the maximum expected frost penetration and sufficiently below minimum bed level
of canal or stream to provide stability and eliminate undercutting. To prevent the
stilling well plus intake from functioning as a short-cut for ground water flow, to
prevent siltation, and to facilitate zero-setting of a recorder, the well should be water-
tight. The inner base of a steel well should be sealed with bitumen where it meets
the concrete foundation.
Since the primary purpose of the stilling well is to eliminate or reduce the effects
of surging water and wave action in the open channel, the cross-sectional area of the
intake should be small. On the other hand, the loss of head in the intake during the
estimated maximum rate of change in stage should be limited to say 0.005 m. This
head loss causes a systematic error; a rising water level is always recorded too low
and a falling water level too high (Section 2.9). As a general guide to the size and
number of intakes, their total cross-sectional area should be approximately 1 per cent
of the inside horizontal cross-sectional area of the well.

74
 L 9Ox90xQmm
long-weld or braze
to top of well for rLeave down about
attaching shelter 3 mm from top

14 mm hole for
SECTION 8 - 8

Access doors
should have a lap

rfety
on a11 sides

. . 1 1; I
-
I
Boltorweld
brass or 1440' Mik

bronze -
L!

DETAIL OF ACCESS
DOOR.

access door

E l e v a t i o n w e s t int a ke

NOTE
ALL DIMENSIONS IN CM
UNLESS OTHERWISE INDICATED

SECTION A - A

Figure 2.5 Example of a steel stilling well for low head installations (after U.S. Dept. of Agriculture)

The intake pipe or slot should have its opening at least 0.05 m below the lowest level
t o be gauged, and it should terminate flush with and perpendicular to the boundary
of the approach channel. The area surrounding the intake pipe or slot should be care-
fully finished with concrete or equivalent material over a distance of 10 times the dia-
meter of the pipe or width of the slot. Although the minimum requirement is one
slot or pipe,-on field installations it is advisable to install at least two at different
levels to avoid the loss of valuable data if one intake should become clogged.
In most stilling wells, the intake pipes will require periodical cleaning, especially those
in rivers carrying sediments. Permanent installations can be equipped with a flushing
tank as shown in Figure 2.6. The tank is filled either by hand pump or with a bucket,
and a sudden release valve will flush water through the intake pipe, thereby removing
the sediment. For tightly clogged pipes and on temporary structures, a sewer rod or
'snake' will usually provide a satisfactory way of cleaning.

75
 h tank

Figure 2.6 Example of an intake pipe system with flush tank

A method that delays plugging involves the construction of a large cavity in the
floor of the approach channel at the head measurement station. Its size may be of
the order of 0.1 m3. The stilling well pipe then enters this cavity and is fitted with
a pipe elbow which is turned down so that sediment cannot fall directly into the pipe.
The cavity must fill with sediment before the stilling well pipe can be clogged. The
cavity must be covered with a steel plate coincident with the bottom of the approach
channel. Taking into consideration the probable increased bedload trapping of trans-
verse slots in this plate and the low quality pressure detection likely with parallel slots,
Replogle and Frazier (1973) advised the use of a battery of 12/3 mm holes drilled into
the 5 mm grating plate. They reported that laboratory use showed no pressure detec-
~

tion anomalies and that field use showed no sedimentation plugging problems, al-
though periodic grating and cavity cleaning is required.
1
1

2.7 Maximum stage gauge I

If records are kept to gain information on maximum flow and no continuously opera-
ting recorder is installed, a flood gauge may be used to protect and retain a high-water
mark for subsequent observations. The types recommended by the U.S. Department
of Agriculture all use powdered cork to mark the maximum water level. As an example,
Figure 2.7 shows a gauge that consists of a pipe containing a removable calibrated
stick, 2.5 cm square, from which the cork is wiped off after each observation. A small
metal or plastic cap, 4.0 cm in diameter and 1.5 cm deep, is attached to the bottom
end of the stick to hold a supply of powdered cork.
The 50 mm galvanized pipe is equipped with a perforated cap (4 perforations of
12/6 mm) at the bottom and another cap at the top. The top cap should be easily
removable to allow observations but should have provisions for a padlock to prevent
vandalism. The pipe should be securely anchored in an upright position as described
in Section 2.5 for a staff gauge. The top of the pipe should be accessible, also at flood
stages to facilitate observations. Since the flood gauge is intended to register high
water marks, the pipe should be long enough to extend from the moderate high water
mark, which is expected on an average of say twice per year, to a point above the
maximum stage expected.

76
 n FLOW

\i/

0 5 0 m m galva-
s t e e l pipe

DETAIL OF INTAKE CAP

Anchor t o supporting
structure with

32 x 3 s t r a p s w i t h

0 . 5 m pipe I e n a

40 mm d i a m e t e r
metal or plastic cap.
15 mm deep. attachQd

Reference level
18" steel stud welded
t o inside of cap. Top a t
( s e e detail) s a m e height as cap flange

Figure 2.7 Details of a maximum stage gauge (after U.S. Department of Agriculture 1962)

2.8 Recording gauge

Automatic water stage recorders are instruments that produce graphical, digital or
punched paper tape records of water surface elevation in relation to time. The usual
accessories to a recorder and its clock are a float, a counterweight, a calibrated float
tape, two tape clamps with rings, a box ofcharts or paper tapes, and the manufacturer's
instructions.
The use of such a recorder has the following advantages over an ordinary attendant-
read staff gauge: (i) in rivers with daily fluctuations, continuous records provide the
most accurate means of determining the daily average, (ii) the entire hydrograph is
recorded with the maximum and minimum stages as a function of time, (iii) observa-
tions can be made at remote places where observers are not available or in locations
that are not accessible under all weather conditions.
Various meteorological instrument manufacturers produce a variety of commercial-
ly available recorders. Most recorders permit the accurate registration of a wide range

77
 in stage on a scale which can be read easily. The majority also have several time and
stage-scale ratios available, and may run as long as 60 days before the clock has to
be rewound or the battery, chart or tape replaced.
Some recorders are driven by clocks operated either by spring or weight; the digital
recorder is an electrically operated device. No further details of recorders are given
here, since the manufacturers description and instructions are both detailed and com-
plete, while technical progress soon makes any description obsolete.

2.9 The float-tape and the diameter of the float

If a float-operated recorder is selected, it should be equipped with a calibrated float
tape that passes over the float wheel. The float and counterweight should be attached
to the ends of the tape by ring connectors. If the recorder is not equipped with a
tape index pointer, one should be attached either to the shelterhouse floor or to the
instrument case. The purpose of the calibrated tape and the index pointer is to enable
the observer to check the registered water level against the actual water level in the
float well and that shown on the independently placed staff gauge. As such, they pro-
vide an immediate check on whether recorder, float, and inlet pipe or slots are function-
ing properly.
All water level recorders operate only if a certain initial resistance is overcome. This
resistance, which is due to friction in the recorder and on the axle, can be expressed
as a resisting torque, Tf, on the shaft of the float wheel (Figure 2.8).
, If the counterweight exerts a tensile force, F, on the float-tape, this force must in-
crease or decrease by AF before the recorder will operate so that
A F r > Tf (2- 1)
where
AF = change in tensile force on float-tape between float and float wheel
r = radius of the float wheel
T, = resisting torque due to friction on the float wheel axle.
When we have, for example, a continuously rising water level in the well, a decrease
in the tensile force, AF, is required, which is possible only if the upward force acting
on the submerged part of the float increases. Consequently, the float has to lag behind
the rising water table by a distance Ah so that the volume of the submerged float
section will increase by
x
AV = - DAh (2-2)
4
where D equals the diameter of the float. According to Archimedes law, the upward
force will increase linearly with the weight of the displaced volume of water, hence
x
AF = TD2Ahpg (2-3)

Substitution of Equation 2-3 into Equation 2-1 shows that the friction in the recorder
and on the axle causes a registration error of the water level

78
 Ah > 4Tf (2-4)
pgnD2r
~

This lagging behind of the float causes a systematic error; a rising water level is always
registered too low and a falling water level too high. Accepting the recorders internal
friction moment, Tf, as a basic datum this systematic error can only be reduced by
enlarging either the float diameter, D, or the radius of the float-wheel, r.
Submergence of the counterweight and an increase of weight of the float tape or
cable on one side of the float wheel (and consequently a decreasing weight on the
other side) cause a known change in tape force at the float. This change in force,
AF, results in a systematic registration error, Ah, which can be calculated by Equation
2-3. These systematic errors can also 6e reduced by enlarging the float diameter.
The reader should note that the phenomenon just described produces a systematic
error that adds to the one mentioned in Section 2.6, i.e. an error due to the head
loss in the intakes.

- AF

1
I
U
zero-end Of
float tape

]counter weight

-e./ d
r::J
1):

Figure 2.8 Forces acting on a float tape

79
2.10 Instrument shelter
The housing of the recorder can vary from those used for permanent stations on large
streams, which allow the observer to enter, to very simple ones, just large enough
to cover the recorder and hinged to lift in the same direction as the instrument cover.
A major disadvantage of the latter type is that it is impossible to service the recorder
during bad weather, and further that the shelter provides no room for the storage
of charts and other supplies. For our purposes, the instrument shelter should meet
the following criteria: The shelter should be ventilated to prevent excessive humidity
from distorting the chart paper. All ventilation openings should be covered with a
fly screen (Figure 2.9). The shelter door should be hinged at the top so that when

xx) 0 hole to 14 0 holes for 12 mm Galv iron roof

SECTION

k==lgl
Shelter floor.

FRONT VIEW SIDE VIEW

ALL DIMENSIONS IN MM. UNLESS OTHERWISE INDICATED

Shelter to te painted inside and outside
with two coats of white paint

Figure 2.9 Example of an instrument shelter (after U.S. Dept. of Agriculture)

80
it is opened it will provide cover for the observer. An iron strip with a small notch
near one end should be attached to either side of the door and should run through
a staple on each side of the door opening, thus holding the opened door in position.
To prevent vandalism, all hinges and safety hasps should be placed so that they
cannot be removed while the door is locked. The flooring should be solid and of a
suitable hardwood which will not warp. The shelter floor should be anchored to the
well, for instance by bolting it at the four corners to small angle irons welded to the
top of the float well. Condensation can be reduced by glueing or spraying a 3 mm
layer of cork to the inside of both the metal shelter and the recorder cover. Silica
gel can be utilized as a desiccant, but the moisture should be removed from the gel
at regular intervals by heating it in an oven to about 150C (300F).

2.11 Protection against freezing

During winter it may be necessary to protect the stagnant water in the stilling well
against freezing. This can be done by employing one or more of the following methods,
depending on location and climate. If the well is set into the bank, an isolating subfloor
can be placed inside the well just below ground level. Care should be taken, however,
that both the float and counterweight can still move freely over the range of water
levels expected during winter. If the well is heated with an electric heater or cluster
of lights, or when a lantern or oil heater is suspended just above the water level, the
subfloor will reduce the loss of heat. A reflector to concentrate the light or heat energy
on the water surface will increase the heating efficiency.
A layer of low-freezing-point oil, such as fuel oil, in the well can be used as protec-
tion. The thickness of the oil layer required equals the greatest thickness of ice expected,
plus some allowance for water-stage fluctuations.
To prevent leakage of oil and erroneous records, a watertight stilling well will be
necessary. Since the mass density of oil is less than that of water, the oil will stand
higher in the well than the water surface in the open channel. Consequently, the re-
corder must be adjusted to give the true water stage.

2.12 Differential head meters

The differential head meter is an important device in structures where the difference
between two piezometric heads or water levels has to be known. Examples of such
structures are the constant head orifice and other submerged orifices.
The importance of the differential head meter is such that the success or failure
of the measuring and/or regulating structure often depends entirely upon the efficiency
of the particular meter used. Four types, all employing two adjacent stilling wells,
will be described here.

U-hook type

The U-hook type is the most simple and sturdy of differential head meters. It has

81
Figure 2. I O U-hook type differential head meter

no moving parts and consists merely of two scales fixed to one short beam (Figure
2.1O).
When the u-hook is placed over the divide wall between the two stilling wells, both
scales are hanging in the water. The differential head is obtained by reading both
scales independently and calculating the difference in immersion.

Hanging scale type

A differential head meter of the hanging scale type is a rather simple and inexpensive
device from which the full differential head can be read from a free hanging scale.
The meter consists of a float and an index which are hung over two disc wheels and
a second float plus scale which hang over a third disc wheel. The three disc wheels
are mounted on the same beam. Bicycle axles could be used for this purpose (see Figure
2.1 1).
The length of the scale should be about 0.10 m more than the maximum expected
differential head. The height of the beam above ground level should be such that the
scale stays clear of the steel stop-plate at low stage while the scale should remain hang-
ing free at high stages. Zero-setting of the index should preferably be done by turning
a swivel in the cable between the upstream float and scale. The device cannot be coup-
led to an automatic recorder.

Tube-float type

This robust differential head meter, which works without interruption under field con-
ditions, can be constructed by using two tube-floats. These floats can be made from
a section of 0 1 5 0 mm galvanized pipe, welded closed at the bottom and equipped

82
 scale

-swivel
DETAIL
INDEX

20
40

r III] stillingwell level

L 250 J

crest level

dimensions in mm

Figure 2. I 1 Differential head meter with hanging scale

with a screw cap plus hook at the top. Ballast is placed inside the watertight tube-floats
so that they are heavier than the water they replace. Two of these floats, hanging
over a bicycle wheel equipped with a zinc tyre, form a balance which, after immersion
of the floats, adjusts itself in such a way that the pipes have either the same draught
or a constant difference in draught, the latter occurring if the weights of the two tube-
floats are not exactly the same.

When the head between the two stilling wells is changing, each of the floats will move
over half the change in head. By transmitting the movement of the floats, as illustrated
in Figure 2.12, a differential head meter is obtained, which shows the difference in
head on a real or enlarged scale depending on the diameter of the disc-wheel and
the length of the balanced hand. The diameter of the disc-wheel should be such that
half its circumference is equal to or slightly larger than half the maximum difference

83
 disc wheel
(I balam

dimensions in mm

Figure 2.12 Tube-float differential head meter (after Romijn 1938)

in head to be measured. In this case the scale only fills half a circle, which facilitates
observations.
A change of head will cause a point on the circumference of the disc-wheel to move
half that dimension. Provided the hand is twice as long as the radius of the wheel,
its point moves over a distance twice as far as the movement of one float. Hence,
it shows the real change in head. The length of the tube-floats should be such that,
at both maximum and minimum stages, the floats are neither submerged nor hanging
free above the water surface.
Index-setting of the hand should preferably be done by turning a swivel in the cable
between the downstream float (IJ) and the disc-wheel. If required, the differential head
can be recorded by an automatic recorder.

Suction lift type

A portable differential head meter which facilitates accurate observation is the suction
lift type. This instrument consists of two glass tubes which are joined at the top by
a tee that is connected to a transparent conduit in which a partial vacuum can be created
by means of a simple hand-operated pump. The lower ends of the glass tubes are con-
nected with the stilling wells for the upstream and downstream heads. (Figure 2.13)

84
 r=====
stopcock valve -_ --
d

slidin indicator
and r8irror
attached to tape
I
tope

vernier- I1 IHI

t o upstr to downstream
stilling stilling well

Figure 2.13 Differential head meter of the suction lift type with direct reading scale

The meter is operated as follows. The stopcock valve is opened and a partial vacuum
is created by means of the hand pump so that water flows into the container and
all air is removed from the conduits. Then the stopcock valve is closed. Subsequently,
by operating the valve, some air is admitted so that the two liquid levels become visible
in the glass tubes. The difference in head can now be obtained by reading the elevation
of each liquid level independently on a scale placed behind the tubes.
A device developed by the Iowa Institute simplifies this process by the use of a
continuous tape over pulleys mounted at the top and bottom of the gauge.
The zero end of the tape is set at one liquid level and a sliding indicator moved
to the other level. Subsequently, the difference in head is given as a direct reading
on the tape.

To prevent the small diameter conduits from becoming clogged, they should be used
in combination with stilling wells and the conduit openings should be carefully
screened. A conduit diameter of 0.5 to 1 .O cm will usually be adequate.

2.13 Selected list of references

British Standards Institution. 1965, 1969. Methods of Measurement of Liquid Flow in Open Channels.
B.S.3680: Part 4A: Thin plate weirs and venturi flumes. Part 48: Long base weirs. British Standards
House, London W.I.
Replogle, J.A. and G. Frazier. 1973. Depth detection in critical depth flumes. Annual Report. U.S. Water
Conservation Laboratory, USDA, Phoenix, AZ.
Romijn, D.G. 1938. Meetsluizen ten behoeve van Irrigatiewerken. Vereniging van Waterstaats Ingenieurs
in Nederlandsch-Indi. Bandung.
Rouse, H. 1964. Engineering Hydraulics. John Wiley & Sons, Inc., New York, London. 4th printing.

85
Troskolanski, A.T. 1960. Hydrometry. Theory and practice of hydraulic measurements. Pergamon Press,
Oxford, London, New York, Paris.
U.S. Department of Agriculture 1962. Field manual for research in Agricultural Hydrology. Agriculture
Handbook No. 224. Washington, D.C.
U.S. Department of the Interior. 1967. Water Measurement Manual. U.S.Bureau of Reclamation. 2nd
Ed. Denver, Colorado. Repr. 1971.

86
 3 The selection of structures
3.1 Introduction
In selecting a suitable structure to measure or regulate the flow rate in open channels,
all demands that will be made upon the structure should be listed. For discharge mea-
suring and regulating structures, hydraulic performance is fundamental to the selec-
tion, although other criteria such as construction cost and standardization of struc-
tures may tip the balance in favour of another device.
The hydraulic dimensions of the discharge measuring or regulating structures des-
cribed in the following chapters are standardized. The material from which the device
is constructed, however, can vary from wood to brick-work, concrete, polyester, metal,
or any other suitable material. The selection of the material depends on such criteria
as the availability and cost of local material and labour, the life-time of the structure,
pre-fabrication etc. Constructional details are not given in this book except for those
steel parts whose construction can influence the hydraulic performance of the struc-
ture.
Although the cost of construction and maintenance is an important criterion in
the selection of structures, the ease with which a discharge can be measured or regu-
lated is frequently more important since this will reduce the cost of operation. This
factor can be of particular significance in irrigation schemes, where one ditchrider
or gatesman has to control and adjust I O to 20 or more structures daily. Here, ease
of operation is labour saving and ensures a more efficient distribution of water over
the irrigated area.
Although other criteria will come into play in the final selection of a discharge mea-
suring or regulating structure, the remarks in this chapter will be limited to a selection
based solely on hydraulic criteria.

I 3.2
3.2.1
Demands made upon a structure
Function of the structure

Broadly speaking, there are four different types of structures, each with its own particu-
lar function:
- discharge measuring structure;

- discharge regulating structure;

- flow divider;
- flow totalizer;

Discharge measuring structure

The function of such a structure is to enable the flow rate through the channel in
which it is placed to be determined. If the structure is not required to fulfil any other
function, such as water-level control, it will have no movable parts. Discharge
measurement structures can be found in natural streams and drainage canals, and

87
also in hydraulic laboratories or in industries where flow rates need to be measured.
All flumes and fixed weirs are typical examples of discharge measurement structures.

Discharge regulating structure

These structures are frequently found in irrigation canals where, as well as having
a discharge measuring function, they also serve to regulate the flow and so distribute
the water over the irrigated area. Discharge regulating structures can be used when
water is drawn from a reservoir or when a canal is to be split up into two or more
branches. A discharge regulating structure is equipped with movable parts. If the struc-
ture is a weir, its crest will be movable in a vertical direction; if an orifice (gate) is
utilized, the area of the opening will be variable. Almost all weirs and orifices can
be used as discharge regulating structures.
In this context it is curious to note that in many irrigation canal systems, the dis-
charge is regulated and measured by two structures placed in line in the same canal.
The first structure is usually a discharge regulating gate and the second, downstream
of the first, is a discharge measuring flume. It would seem to be a waste of money
to build two such structures, when one would suffice. Moreover, the use of two struc-
tures requires a larger loss of head to operate within the modular flow range than
if only one is used. Another even more serious disadvantage is that setting the required
discharge with two structures is a more time consuming and complicated procedure
than if a single regulating structure is used. Obviously, such procedures do not contri-
bute to the efficient management of the available water.

Flow divider

It may happen that in an irrigated area we are only interested in the percentage distribu-
tion of the incoming flow into two or more branch canals. This percentage distribution
can be achieved by constructing a group of weirs all having the same crest level but
with different control widths. If the percentage distribution has to vary with the flow
rate in the undivided canal, the crest level of the weirs may differ or the control sections
may have different shapes. Sometimes the required percentage distribution of flow
over two canals has to vary while the incoming flow remains constant. This problem
can be solved by using a movable partition (or divisor) board which is adjusted and
locked in place above a fixed weir crest (see Section 9.1).
Although a flow divider needs no head measurement device to fulfil its function,
a staff gage placed in the undivided canal can give additional information on the flow
rate, if this is required by the project management.

Flow totalizer

If we want to know the volume of water passing a particular section in a given period,
we can find this by using a flow totalizer. Such information will be required, for in-
stance, if a farmer is charged for the volume of water he diverts from the irrigation

88
canal system, or if an industry is charged for the volume of effluent it discharges into
a stream. The two flow totalizers treated in this book both have a rotating part and
a revolution counter which can be fitted with an additional counter or hand to indicate
the instantaneous flow rate.

3.2.2 Required fall of energy head t o obtain modular flow

Flumes and weirs

The available head and the required head at the discharge measuring site influence
both the type and the shape of the structure that will be selected. For weirs and flumes,
the minimum required head AH to operate in the modular flow range can be expressed
as a fraction of the upstream energy head H , or as (HI - H,)/H,. This ratio can also
be written as 1 - H,/H,, the last term of which describes the limit of the modular
flow range, i.e., the modular limit (see also Section 1.15).
The modular limit is defined as the value of submergence ratio H,/HI at which the
real discharge deviates by 1 % from the discharge calculated by the head-discharge
equation. We can compare the required fall over weirs of equal width by considering
their respective modular limits. The modular limit of weirs and flumes depends basi-
cally on the degree of streamline curvature at the control section and on the reduction
of losses of kinetic energy if any, in the downstream expansion. Broad-crested weirs
and long-throated flumes, which have straight and parallel streamlines at their control
section and where part of the kinetic energy is recovered, can obtain a modular limit
as high as H,/HI = 0.95. As mentioned in Chapter 1, the discharge coefficient of
a weir increases if the streamline curvature at the control section increases. At the
same time, however, a rising tailwater level tends to reduce the degree of streamline
curvature, and thus reduces the discharge.
Consequently we can state that the modular limit of a weir or flume will be lower
as the streamlines are more strongly curved under normal operation. The extreme
examples are the rectangular sharp-crested weir and the Cipoletti weir, where the tail-
water level must remain at least 0.05 m below crest level, so that streamline curvature
at the control section will not be affected. Modular limits are given for each structure
and are summarized in Section 3.3.
The available head and the required head over a structure are determining factors
for the crest elevation, width and shape of the control section, and for the shape of
the downstream expansion of a discharge measurement structure. This can be shown
by the following example.
Suppose a 0.457 m (1.5 ft) wide Parshall flume is to be placed in a trapezoidal
concrete-lined farm ditch with 1-to-1.5 side slopes, a bottom width of 0.50 m, and
its crest at ditch bottom level. In the ditch the depth-discharge relationship is con-
trolled by its roughness, geometry, and slope. If we use the Manning equation,
v = l/n R2I3so.s,with a value of n = 0.014 and s = 0.002, we obtain a satisfactory
idea of the tailwater depth in the ditch. Tailwater depth data are shown in Figure
3.1, together with the head-discharge curve of the Parshall flume and its 70% submer-
gence line (modular limit).
An examination of the 70% submergence curve and the stage-discharge curve shows

89
Figure 3. I Stage-discharge curves for 1.5 ft Parshall flume and for a concrete-lined ditch. Flume crest coin-
cides with ditch bottom

that submerged flow will occur at all discharges below 0.325 m3/s, when the flume
crest coincides with the ditch bottom. Figure 3.1 clearly shows that if a design engineer
only checks the modularity of a device at maximum stage, he may unknowingly intro-
duce submerged flow conditions at lower stages. The reason for this phenomenon
is to be found in the depth-discharge relationships of ditch and of control section.
In the given example, a measuring structure with a rectangular control section and
a discharge proportional to about the 1.5 power of upstream head is used in a trapezo-
dal channel which has a flow rate proportional to a greater power of water depth
than 1.5. The average ditch discharge is proportional to y2'.*.On log-log paper the
depth-discharge curve (ditch) has a flatter slope than the head-discharge curve of the
flume (see Figure 3.1). To avoid submerged flow conditions, the percentage submer-
gence line of the measuring device in this log-log presentation must be to the left of
the channel discharge curve throughout the anticipated range of discharges.
The coefficient of roughness, n, will depend on the nature of the surface of the
downstream channel. For conservative design the roughness coefficient should be
maximized when evaluating tailwater depths.
Various steps can be taken to avoid submergence of a discharge measuring device.
These are:
The 1.5 ft Parshall flume of Figure 3.1 can be raised 0.03 m above ditch bottom. The
stage-discharge curve of the flume in terms of ha + 0.03 m plots as a curve shown

90
in Figure 3.2. The corresponding 70% submergence curve plots to the left of the stage-
discharge curve of the ditch.
The 1.5 ft Parshall flume of Figure 3.1 can be replaced either by a flume which
requires more head for the same discharge, thus with a rating curve that plots more
to the left on log-log paper, or by a flume which has a higher modular limit than
70%. A flat-bottom long-throated flume with 0.45 m wide control and 1 to 6 down-
stream expansion will be suitable.
It must be recognized that the two previous solutions with a Parshall flume require
a loss of head of at least 0.31 m at the maximum discharge capacity of the flume,
being Q = 0.65 m3/s (see Figure 3.2). If this head loss exceeds the available head,
the design engineer must select a structure with a discharge proportional to an equal
or greater power of head than the power of the depth yz of the ditch. For example,
he may select a flat-bottom, long-throated flume with a trapezodal control section
and a gradual downstream expansion. Such a flume can be designed in such a way
that at Q = 0.65 m3/s an upstream head h, = 0.53 m and a modular limit of about
0.85 occur resulting in a required head loss of only 0.08 m. He could also use a long-
throated flume with a (truncated) triangular, parabolic, or semi-circular control sec-
tion (Bos 1985).

Figure 3.2 Stage-discharge curves for flume and ditch of Figure 3.1, but flume crest 0.03 m above ditch
bottom

91
 Orifices

At the upstream side of free flowing orifices or undershot gates, the upper edge of
the opening must be submerged to a depth which is at least equal to the height of
the opening. At the downstream side the water level should be sufficiently low so
as not to submerge the jet (see Chapter 8). For this reason free flowing orifices, especial-
ly at low flows, require high head losses and are less commonly used than submerged
orifices. The accuracy of a discharge measurement obtained with a submerged orifice
depends on the accuracy with which the differential head over the orifice can be mea-
sured. Depending on the method by which this is done and the required accuracy
of the discharge measurement, a minimum fall can be calculated with the aid of An-
nex 2. In general, we do not recommend the use of differential heads of less than
0.10 m.

3.2.3 Range of discharges to be measured

The flow rate in an open channel tends to vary with time. The range between Qmin
and Qmaxthrough which the flow should be measured strongly depends on the nature
of the channel in which the structure is placed. Irrigation canals, for example, have
a considerably narrower range of discharges than do natural streams. The anticipated
range of discharges to be measured may be classified by the ratio
Y = Qmax/Qmin (3-1)
From the limits of application of several weirs, a maximum attainable y-value can
be calculated. Taking the example of the round-nosed horizontal broad-crested weir
(Section 4. l), the limits of application indicate that HI/Lcan range between 0.05 and
0.50 m. As a result we obtain a maximum value of y which is

This illustrates that whenever the ratio y = Qmax/Qmin exceeds about 35 the horizontal
broad-crested weir described in Section 4. I cannot be used. Weirs or flumes that utilize
a larger range of head, or which have a head-discharge relationship proportional to
a power of head greater than 1.5, or both, can be used in channels where y = Qm,,/Qmin
exceeds 35. The following example shows how the y-value, in combination with the
available upstream channel water depth y,, influences the choice of a control section.
The process of selection is as follows:
Find a suitable flume and weir for
Qmin = 0.015 m3/s
-+ y = 200
Qma, = 3.00m3/s
Y1 = h, + pI Q 0.80m
The flume is to be placed in an existing trapezoidal channel with a 4 m wide bottom
and 1-to-2 side slopes. At maximum water depth y, = 0.80 m, the Froude number
in the approach channel is Fr = v,/(gA,/B,)~= 0.27. It is noted that for Fr < 0.50
the water. surface will be sufficiently stable.

92
From the relatively high y-value of 200 we can conclude that the control section of
the structure should be narrower at minimum stage than at maximum stage. Meeting
the requirements of this example are control sections with a narrow bottomed trape-
zium, or a triangular or truncated triangular shape. Because of the limited available
width we select a truncated triangular control section of which two solutions are illus-
tra ted below.

Triangular profile flat-V weir (Figure 3.3)

According to Section 6.4.2 the basic head-discharge equation of this weir reads
4 B
Q = C,C, - (2g)0.52 [h:. - (h, - Hb)2.] (3-3)
15 H b

in which the term (he - Hb)2.5should be deleted if he is less than H,. If we use the
l-to-2/1-to-5 weir profile and a 1-to-10 cross slope, the minimum channel discharge
can be measured at the minimum required head, since Q at 0.06 m head is
4 4.0
Q0,06= 0.66 x 1 x - (2g)0.5x -(0.06 - 0.0008)2.5
15 0.20
Q0.06 = 0.0133 m3/s
Another restriction for the application of this type is the ratio h,/p,, which should
not exceed 3.0. The required width of the weir can be found by trial and error:
Since y, = h, +
p, < 0.80 m, the maximum head over the weir crest h, max =
0.60 m when p, = 0.20 m. Using a width B, of 4 m, we find for the discharge capacity
at h, = 0.60 m (for C , see Fig. 6.10)
4 4.00
Q0.60 = 0.66 x 1.155 x E(2g)O. x __ x [(0.60-0.0008)2~5-(0.60-0.0008-0.20)2~5]
0.20
40.60 = 3.205 m3/s
This shows that the full discharge range can be measured with the selected weir.

Long-throated flume with truncated triangular control (see Fig. 3.3)

According to Section 7.1.2, the head-discharge relationships for this flume read

Triangular profile welr Long throated flume

8~:4.00 BE: 3.00
I I

Figure 3.3 Two examples of suitable control sections

93
 [z
Q = C C' 32 3 g]"" B,(h, - 1/2 Hb)2.5 (3-5)

if H l 2 1.25 Hb.
Using a flat-bottomed flume with a throat length of L = 0.80 m (HJL d l), we
can select a suitable control section. After some experience has been acquired two
trials will usually be sufficient to find a control section which will pass the maximum
discharge. For the control section shown in Figure 3.3 the c d - and C,-values can be
found as follows:
For h, = 0.80 m, H,/L N 1, Figure 7.3 shows that c d = 1.O25

The area ratio .

0.25 x 1.5 + 0.55 x 3.0 = o.46

C d 1F = 1.025 x
0.80 x 5.60
and we find in Figure 1.12 that the related C,-value is about 1.06. Substitution of
these values into Equation 3-5 yields a discharge capacity at h, = 0.80 m equal to

Q0.80 = 1.025 x 1.06 x '('9.81)'''

3 3 x 3 (0.80-0.125)3'2

Q0.80= 3.08 m3/s

At minimum applicable head of h, = 0.1L = 0.08 m (see Section 7.1.4) Cd = 0.93
and C , N 1.0.
Using Equation 3-4 we find that at h, = 0.08 m the discharge capacity is

Q0.08= 0.0128 m3/s

This shows that both minimum and maximum discharges can be measured with the
selected structure. These structures are only two of the many which meet 'the demands
set on the discharge range and upstream water depth.

3.2.4 Sensitivity

The accuracy to which a discharge can be measured will depend not only on the errors
in the c d - and C,-values but also on the variation of the discharge because of a unit
change of upstream head. Hence, on the power u of h, in the head-discharge equation.
In various countries, the accuracy of a discharge measuring structure is expressed in
the sensitivity, S, of the structure. This is defined as the fractional change of discharge
of the structure that is caused by the unit rise, usually Ah, = 0.01 m, of the upstream
water level. For modular flow

94
Using the relationship
Q = Constant x h," (3-7)
we can also write Equation 3-6 as
Const x uhl"-,Ah,
S = (3-8)
Const x h,"
U
S = -Ah, (3-9)
h,
The value of Ahl can refer to a change in waterlevel, head reading error, mislocation
of gauging station, etc. In Figure 3.4 values of S x 100 in per cent are shown as
a function of Ah,/h, and the U-value, the latter being indicative of the shape of the
control section.

Presented as an example is a 90-degree V-notch sharp-crested weir which discharges

at h, = 0.05 m. If the change in head (error) Ah, = 0.005 m, we find
2.5
sx 100 = 100-0.005
0.05
= 25%

This shows that especially at high U-values and low heads the utmost care must be
taken to obtain accurate h, values if an accurate discharge measurement is required.
In irrigated areas, where fluctuations of the head in the conveyance canals or errors
in head reading are common and the discharge through a turn-out structure has to
be near constant, a structure having a low sensitivity should be selected.

ratio Ahl/hj

Figure 3.4 Sensitivity as a function of relative change in head and shape ofcontrol section (modular flow)

95
3.2.5 Flexibility

Because of a changing flow rate, the head upstream of an (irrigation) canal bifurcation
usually changes. Depending on the characteristics of the structures in the supply canal
and that in the off-take canal, the relative distribution of water may change because
of the changing head. To describe this relative change of distribution the term flexibi-
lity is used, which has been defined as the ratio of the rate of change of discharge
of the off-take or outlet Qo to the rate of change of discharge of the continuing supply
canal Q, or

(3-10)

In general the discharge of a structure or channel can be expressed by the Equation

Q = Constant h," (3-1 1)
Hence we can write
dQ/dh, = Const uhlU-I (3-12)
Division by Q and by Const h," gives
dQlQ = udhilh, (3-13)
Substitution of Equation 3-13 into Equation 3-10 for both Q, and Q, results in

(3-14)

Since a change in water level in the upstream reach of the supply canal causes an
exactly equal change in h,,oand hl,s,the quotient dh,,,/dh,,, = I , and thus

(3- 15)

The proportional distribution of water over two or more canals may be classified ac-
cording to the flexibility as follows:

a. F = l
For F = 1 we may write

(3- 16)
off-take canal

-
incoming flow Q + 1-11
w
continuing
supply canal

Figure 3.5 Definition sketch

96
To meet this requirement for various heads, the structures on the off-take and supply
canal must be of the same type and their crest or sills must be at the same level.

b. F < I
If less variation is allowed in the off-take discharge than in the supply canal discharge,
the flexibility of the bifurcation has to be less than unity and is said to be sub-propor-
tional. The easiest way to obtain F < 1 is to select two different types of structures,
for example:
- an orifice as off-take; u = 0.5;

- a weir with rectangular (or other) control in the supply canal: u = 1.5 (or more).
We now find that

Usually h,,sis less than 3 hl,o,and then the flexibility of the bifurcation will be less
than unity. F < I can be an advantage in irrigation projects where, during the growing
season, canal water level rises due to silting and weed growth. A low flexibility here
helps to avoid a water shortage at the downstream end of the supply canal.

c. F > 1
If more variation is allowed in the off-take discharge than in the supply canal dis-
charge, the flexibility of the bifurcation has to be greater than unity and is said to
be hyper-proportional. Here again, the easiest way this can be obtained is by using
two different types of structures. Now, however, the structure with low U-value (ori-
fice) is placed in the supply canal while the off-take has a weir with a U-value of 1.5
or more. Thus

Since in this case hl,sis always greater than hl,o,the flexibility of the bifurcation will
be much more than unity. This is especially useful, for example, if the off-take canal
leads to a surface drain which can be used to evacuate excess water from the supply
canal system.

3.2.6 Sediment discharge capability

Besides transporting water, almost all open channels will transport sediments. The
transport of sediments is often classified according to the transport mechanism or
to the origin of the sediments, as follows from Figure 3.6. The expressions used in
this diagram are defined as follows:

Bed-load

Bed-load is the transport of sediment particles sliding, rolling, or jumping over and
near the channel-bed, generally in the form of moving bed forms such as dunes and

97
 o r i g i n of
transported suspended-
sediments
sediment
material transport
load process
m
y bed-load I
Figure 3.6 Terminology in sediment transport

ripples. Many formulae have been developed to describe the mechanism of the bed-
load, some being completely based upon experiment, while others are founded upon
a model of the transport mechanism. Most of these equations, however, have in com-
mon that they contain a number of constants which have to be modified according
to the field data collected for a certain river. In fact, all the deviations in bed-load
from the theoretical results are counteracted by selecting the right constants. Most
of the available bed-load functions can be written as a relation between the transport
parameter
X = T/,/&fY
and the flow parameter
Y = pys/AD
where
T = transport in solid volume per unit width [sometimes expressed in terms
of the transport including voids, S, according to T = S(l -E), where E
is the porosity];
y = depth of flow (often y is replaced by the hydraulic radius R);
D = graindiameter;
A = relativedensity = (p,-p)/p;
s = hydraulic gradient;
p = so-called ripple factor, in reality a factor of ignorance, used to obtain
agreement between measured and computed values of T.

As an example of such an X versus Y relation the well known Meyer-Peter & Mller
bed-load function may be given
X = A(Y - 0.047)32 (3-17)
with A = 8.

Typical bed-load equations like the Meyer-Peter & Mller equation d o not include
suspended-load. Equation (3- 17) differs from the total-load equation given below,
although the construction of both equations will appear to be similar.

98
 Suspended-load

Suspended-load is the transport of bed particles when the gravity force is counter-
balanced by upward forces due to the turbulence of the flowing water. This means
that the particles make larger or smaller jumps, but return eventually to the channel-
bed. By that time, however, o,ther particles from the bed will be in suspension and,
consequently, the concentration of particles transported as suspended-load does not
change rapidly in the various layers. A strict division between bed-load and suspended-
load is not possible; in fact, the mechanisms are related. It is therefore not surprising
that the so-called total-load (bed-load and suspended-load together) equations have
a similar construction to that of the bed-load equations. An example of a total-load
equation is the equation of Engelund & Hansen (1967), which reads
X = 0.05Y5'2 (3-18)

Wash-load

Wash-load is the transport of small particles finer (generally < 50 pm) than the bulk
of the bed material and rarely found in the bed. Transport quantities found from
bed-load, suspended-load, and total-load formulae do not include wash-load quanti-
ties.
Whereas for a certain cross-section quantities of suspended-load and bed-load can
be calculated with the use of the locally valid hydraulic conditions this is not the case
for wash-load. The rate of wash-load is mainly determined by climatological charac-
teristics and the erosion features of the whole catchment area.
Since there is normally no interchange with bed particles, wash-load is not important
for local scour or silting. Owing to the very low fall velocity of the wash-load particles,
wash-load only contributes to sedimentation in areas with low current velocities (reser-
voirs, dead river branches, on the fields). Owing to the small fall velocity, in turbulent
water the concentration of the particles over a vertical (generally expressed in parts
per million, p.p.m.) is rather uniform, so that even with one water sample a fairly
good impression can be obtained. However, the wash-load concentration over the
width of a channel may vary considerably.
The most appropriate method of avoiding sediment deposition in the channel 'reach
upstream from a structure is to avoid a change of the flow parameter Y = pRs/AD.
This can be done, for example, by avoiding a backwater effect in the channel. To
do so, a structure should be selected whose head-discharge curve coincides with the
stage-discharge curve of the upstream channel at uniform flow.
Since the U-value of most (trapezoidal) channels varies between u 1: 2.2 for narrow
bottomed channels and u 21 1.7 for wide channels, the most appropriate structures
are those with a trapezoidal, parabolic, semicircular, or (truncated) triangular control
section.
To avoid the accumulation of sediments between the head measurement station
and the control section, a structure that has either a flat bottom or a low bottom
hump with sloping upstream apron is recommended. Flat bottomed long-throated
flumes, which can be tailored to fit the channel stage-discharge curve, are very suitable
(Bos 1985).

99
Photo 1 Most weirs can be fitted with a movable gate

Many well-designed irrigation canal systems are equipped with a sand trap situated
immediately downstream of the head works or diversion dam. The diversion dam
will usually be dimensioned in such a way that a minimal volume of sediments is di-
verted from the river. Other systems draw their water from reservoirs or wells. As
a result such irrigation canals do not have bed-load transport but will have a certain
amount of suspended-load and wash-load. Because of the flow regulating function
of the structure, the deposition of silt immediately upstream of it cannot always be
avoided even if uniform flow is maintained upstream of the head measurement station.
If an adjustable orifice is used as a discharge regulating structure, it is recommended
that a bottom sill be avoided. If a movable weir is used, it should be fitted with a
movable bottom gate that can be lifted to wash out sediments. This gate arrangement
is described in Section 4.2. Its use is not restricted solely to the Romijn weir; it can
be used in combination with all weirs described in Chapters 4,5, and 6 .

3.2.7 Passing of floating and suspended debris

All open channels, and especially those which pass through forested or populated
areas, transport all kinds of floating and suspended debris. If this debris is trapped
by the discharge measuring structure, the approach channel and control section be-
come clogged and the structure ceases to act as a discharge measuring device.
In irrigation canals it may be practical to install a trash rack at strategic points
to alleviate the problem of frequently clogged structures. This applies especially if

1O0
narrow openings or orifices are used. In drainage channels, however, because of their
larger dimensions, the installation of trash-racks would not be practical. For drainage
canals therefore one should select structures that are not vulnerable to clogging. All
sharp-crested weirs and orifices are easily clogged and are thus not recommended if
floating debris has to be passed. Weirs with a sloping upstream face or weirs with
a rounded nose or crest and all flumes will pass debris relatively easily.
Piers which have no rounded nose or are less than 0.30 m wide, which thus includes
sharp-edged movable partition boards, tend to trap debris.

3.2.8 Undesirable change i n discharge

Structures may be damaged through vandalism or by persons who stand to benefit

from a faulty or non-operating structure. To prevent such damage, the design engineer
should keep structures as simple as possible and any movable parts should be as sturdy
as is economically justified. It may also happen that attempts will be made to alter
the discharge of a structure by changing the hydraulic conditions under which the
structure should operate.
Particularly vulnerable to damage are the sharp-crested weir and sharp-edged ori-
fices. It is possible to increase the discharge of these structures by rounding (i.e. dama-
ging) the sharp edge, roughening the upstream face, or by blocking the aeration vent
to the air pocket beneath a fully contracted nappe. Because of this and also because
of their vulnerability to clogging, sharp-crested weirs and sharp-edged orifices are only
recommended for use in laboratories, on experimental farms or at other places where
frequent inspection of the structures is common.
It is obvious that the discharge of structures which operate under submerged flow
can easily be influenced by altering the water level in the tailwater channel. It is there-
fore recommended that modular structures be used wherever off-takes, outlets, or
turn-outs are required.
Lack of maintenance will usually cause algal growth to occur on a structure. On
a sharp-crested weir, algal growth will lead to a roughening of the upstream weir face
and a rounding of the sharp edge. Both phenomena cause the contraction to decrease
and thus lead to an increase in the weir discharge at constant head.
On a broad-crested weir algal growth causes a roughening of the weir crest and
a rise in its height. This phenomenon, however, causes the weir discharge to decrease
at constant head.
The least influenced by algal growth is the short-crested weir. Its discharge will
scarcely be affected because of the strong influence of streamline curvature on the
discharge coefficient relative to the influence of a change of roughness of the weir
crest. In selecting a discharge measuring or regulating structure and organizing its
maintenance, this phenomenon should be taken into account.

3.2.9 Minimum water level in upstream channel

Several discharge measurement structures have a second function, which is to retain

water in the upstream channel reach, especially at low flows. In flat areas in moderate

101
climates, structures in drainage channels can be used to maintain a minimum water
level in the channels during the dry season, thus controlling the groundwater level
in the area. To perform this function, the weir crest elevation must be above the up-
stream channel bottom. If the variation between required minimum and required max-
imum water levels in the channel is small and the discharge varies considerably, a
movable weir may be the only possible solution.
On the other hand, in hot climates it may be desirable to design discharge measure-
ment structures so that the channels in which they are placed will go dry if no flow
occurs. This may be a necessary precaution to prevent the spread of serious diseases
like malaria and bilharzia. It may also be convenient to have irrigation canals go dry
by gravity flow so that maintenance work can be performed. This will require that
all structures in supply canals and drainage channels have zero crest elevation or a
drain pipe through the weir sill. If a raised weir crest is needed during other periods,
a movable weir will provide the answer.

3.2.10 Required accuracy of measurement

In the head-discharge equation of each structure there is a discharge coefficient and

an approach velocity coefficient, or a combination of these coefficients. The accuracy
with which a discharge can be measured with a particular structure depends to a great
extent on the variation of these coefficients determined under similar hydraulic condi-
tions.
For all of the structures described, an expected error in the product CdCVor in
the combined coefficient is given in the relevant section on the evaluation of discharge.
These errors are also listed in Section 3.3. Often, the error in C,C, is not constant
but decreases if the Cd-value increases, which usually occurs if the head over a crest
increases.
Besides the error in the coefficients, the most important error in a discharge measure-
ment is the error inherent to the determination of a head or head differential. The
error in head mainly depends on the method and accuracy of zero setting and the
method used to measure the head. It can be expressed in a unit of length independently
of the value of head to be measured. As a result enormous errors often occur in a
discharge measurement if the structure operates under minimal applicable head or
head differential (see also Sensitivity, Section 3.2.5).

3.2.11 Standardization of structures in an area

It may happen that in a certain area, several structures will be considered suitable
for use, each being able to meet all the demands made upon discharge measuring
or regulating structures. It may also happen that one of these suitable structures is
already in common use in the area. If so, we would recommend the continued use
of the familiar device, especially if one person or one organization is charged with
the operation and maintenance of the structures. Standardization of structures is a
great advantage, particularly for the many small structures in an irrigation canal sys-
tem.

102
3.3 Properties and limits of application of structures
3.3.1 General

In Section 3.2 the most common demands made upon discharge measuring or regulat-
ing structures are described. In Chapters 4 to 9, the properties and limits of application
of each separate structure are given in the sections entitled Description and Limits
of application. To aid the design engineer in selecting a suitable structure, we have
tabulated the most relevant data.

3.3.2 Tabulation of data

Table 3.1 consists of 18 columns giving data on the following subjects

Column 1 - Name of the standard discharge measuring or regulating device. In
brackets is the section number in which the device is discussed. Each
section generally consists of sub-sections entitled: Description, Evalua-
tion of discharge, Modular limit, Limits of application.
Column 2 - A three-dimensional sketch of the structure.
Column 3 - Shape of the control section perpendicular to the direction of flow and
the related power u to which the head or differential head appears in
the head-discharge equation.
Column 4 - Possible function of the structure. If the area of the control section
cannot be changed, the structure can only be used to measure dis-
charges; this is indicated by the letter M in the column. If the weir
crest can be made movable by use of a gate arrangement as shown
in Section 4.2, or if the area of an orifice is variable, the structure can
be used to measure and regulate discharges and has the letters MR
in the column. The Dethridge and propeller meters can measure a flow
rate in m3/s and totalize the volume in m3.The discharge can be regula-
ted by a separate gate, which is, however, incorporated in the standard
design. These two devices have the letters MRV in the column.
Column 5 - Minimum value of H , or Ah in metres or in terms of structural dimen-
sions.
Column 6 - As Column 5 , but giving maximum values.
Column 7 - Minimum height of weir crest or invert of orifice above approach chan-
nel bottom; in metres or in terms of structural dimensions.
Column 8 - Minimum dimensions of control section; b,, B,, w, and D,.
Column 9 - Range of notch angle 8 for triangular control sections.
Column 10 - Minimum discharge (Qm,Jin m3/s x or I/s of the smallest possible
structure of the relevant type, being determined by the minima given
in Columns 5,8,and 9.
Column 1 1 - Maximum discharge: q in m2/s, being the discharge per metre crest
width if this width is not limited to a maximum value, or Q in m3/s
if both the head (differential) and control section dimensions are limi-
ted to a maximum. No maximum discharge value is shown if neither
the head (differential) nor the control dimensions are limited by a
theoretical maximum. Obviously, in such cases, the discharge is limited
because of various practical and constructional reasons.

103
TABLE 3 . 1 . DATA ON VARIOUS STRUCTURES
1 2 3 4 5 6 7 8 9

Name o f s t r u c t u r e Sketch o f s t r u c t u r e Shape of M = Hl min H1 max minimum minimum range o f

and s e c t i o n number control section measuring crest size o f notch
i n which s t r u c t u r e perpendicular t o or or height control angle 8
i s described f l o w and U-value ::aiuring
Ah min Ah max above b o r B. degrees
approach w and D
L regulating rhnnnrl

Round-nose
horizontal
broad-crested
" -
rectangular
1.5
m 0.06 m
0.05 L 0.5 L
0.15 m
0.33 H I
0.30 m
0.30
H mx
0!2 L
veir (4.1)

Ramijn mvable
measuring1
regulating
u -
rectsngvlar
1.5
m 0.05 m
0.12 L 0.78 L
0.15 m
0.33 HI
0.30 m
H max

veir ( 4 . 2 )

Triangular (truncated) m 0.06 m 0.15 m 0.30 m 30 to

broad-crested triangular 0.05 L 0.5 L to 0.33 H I H max I80
weir (4.3) "-1.7 to 2.5 0.7 L 0!2 L

Broad-crested
rectangular
profile weir (4.4)
-
rectangular
" 1.5
m 0.06 m
0.08 L 0.85 L*-
1.50 L-
0.15 m
if 0.4 h
if 0.65 d,
0.30 m
h mx
0!2 L

Palp3
vel=(4.5) u -
rectangular
1.6
M 0.06 m
0.08 L 1.6 L
0.15 m 0.05 m
3hlIAl

Rectangular
sharp-crested
veira (5.1)
Y -
rectangular
1.5
H,
or
0.07 m 0.60 m
0.5 b
0.30 m
hl
0.30 m
B-b>4 hl

m- 0.03 m 2.4 p 0.10 m 0.15 m

0.5 h l

V-notch
sharp-crested
veirs ( 5 . 2 )
u -
triangular
2.5
H

M
0.05 m

0.05 m
0.60 m
1.2 p
0.38 m
0.10 m

0.45 m
B122.5 h l

B125.0 h l
90

25 to 100
0.4 p

Cipoletti
Yell (5.3) Y -
trapezoidal
1.5
m 0.06 m 0.60 m 0.30 m
hl
b20.30
0 . 5 hl
m

Circular circular M 0.03 m 0.10 m d20.20 m

veir ( 5 . 4 ) u is variable 0.1 d 0.9 d 0.5 d
but h 2 . 0

Proportional
veir ( 5 . 5 ) Y -
proportional
1.0
M 0.03 m
2 s
such that
~ ( 0 . 0 0 5m
e - 0
or
p20.15 m
0.15 m

Weir sill vith H 0.09 m 0.90 m O 0.30 m

rectangular 0.75 L 0.5 b b21.25 b,
control section
(6.1)

V-notch weir n h,-0.03m hl-l.83 m 0.15 m 12652'

ail1 (6.2) 143'08'
15722'

Triangular profile
m-dimeneional
veir (6.3)
u -
rectangular
1.5
M 0.03 m'
steel
3.00 m
3.0 p
0.06 m
0.33 HI
0.30 m
2 HI

0.06 m
Concrete
 Y ' modular l i m i t error i n sensitive- debris sediment
Qmi n Qmax in
CdCV or ness a t passing passing
m'ls or QmX "2'"l minimum capacity capacity
q mdx i n %K or 'e head t t very good; t good; Re*rkS
"IS m2/s head loss 6) % per 0 fair;
0.01 m - poor; - - very poor

b -
0.0066
0.30 m
4 -
H1-2.0 m
4.7 35 0.70 to'
0.95
2(21-20 Cd) 25 + 0 * value depends on
elope backface and on
ratio p IH
2 2

0.0057' W0.860* 30 0.30 3 30 + + values refer to

b-0.30 m b-1 .50 m
-
standard veir vith
L 0.60 m

0.0026 variable 830* 0.80 to 2(21-20 cd) 42 + t o o O triangular cantrol

at 8-30' 0.95 depending 0 . 0 5 L$H1\(0.7 L
on 0

0.0064 q-5.07 0.66 to IOF-8 25 0 0 depending on

Hl-2.0 m 35- 0.38 I<F$l.24 veir height p
81

0.0011 q-5.1
H1-2.0 m
90 0.66' 5 25 o *"su.lly lover

0.00997 q-o.813 24.5

if b2l.Z m
head loss- 25 __ __ Fully contracted veir
HI +O.OS m
0.00137 variable about 30 head loss- 25 __ Full vidth 6 partially
HI c0.05 m contracted veire

0.0008 W0.390 about head loss 50 -- __ partially contracted

about about 500 > H,
0.0002
if 0-28'4'
W0.145
if 0-100'
about
I50
head loss 50 -- __ fully contracted
> Hl

4'0.0082 q-O. 864 36.4 head loss 5 25 -- __

b-0.30 m HI+0.05 m

0.00091 variable 55.9 head loss 2 67

d-0.20 m if d>0.30 m ~ ~ + 0 . 0m5

0.0058 variable small.but head loss 2 33 -_ _- B > 0.005 m

a-0.006 m depends Hl+0.05 m good if
b-0.15 m on avalue P - 0

0.013 q-1.366 32 0.20 5 17 + +

b-0.30 m

0.0005 Q-25.4*' 50000 * 0.30 3 83 O - 'three notch angles only

0.0007 Q-30.6 43000 **depending on Alr a l u e s
0.0010 4'49.4 49000

0.0031 9-10,18 I O00 0.75 10 cv-9 50 + + *depends on crest ma-

h1-0.03m or or terial.
0.0088 350 25 Applies to I-to-5 back
b-0.30 m face
h l-0.06m
TABLE 3.1. DATA ON VARIOUS STRUCTURES (cont.)
1 2 3 4 5 6 7 8 9
Name of s t r u c t u r e Sketch of structure Shape o f M = H1 min H1 max minimum minimum range o f
and section number control section measuring or crest size o f notch
i n which s t r u c t u r e perpendicular t o or height control angle 8
i s described flow and U-value ::a;uring Ah min Ah max above b o r B. degrees
h regulating approach w and D
channel
bottom p
Triangular profile (truncated) M 0.03 m 3.00 m 0.06 m 0.30 m 168O34'
flat-vee vei? (6.4)
u - 1.7 to
stee1
0.06 m
concrete
3.0 p 0.33 H I 2 HI
174'16'

Butcher's -able
standing vave weir
(6.5)
-
rectangular
u 1.6
w 0.05 m 1.00 m 1.4 hl- 0.30 m
hl

WES-Standard
spillvay (6.6) " -
rectangular
1.5
M 0.06 m depends'
on hd
0.15 m
0.2 h l
0.30 m
Hl
5.0 p

Cylindrical
crested veir (6.7) U,*J
3%-
_I
u-
rectangular
1.5
m o.06 m
0.1 r
depends'
on r
3.0 D
0.15 m
0.33 h l
0.30 m
2 Hl

"-
Long-throated rectangular M 0.06 m O 0.30 m *
1.5 0.1 L 1.0 L but
(truncated) M 1.0 B ers0.5
B H . 1 0 u* 30 to I80
for for for
all a11 all
M f1-s flume* flumea
820.30 'm side slopt
Note: variable*'

u -
parabolic
2.0
(semi)-circulsr
M

M
i n geneml
Hlt3.0 m f >O.IO 'm

d 3 0.20 m*
u is variable
but 4 2.0

Throatless flumes
with rounded transition
(7.2)
u -
rectangular
1.5
M 0.06 m 2.00 m
1.5 R
o 0.20 m
H, max

Throatless flume* M 0.06 m 1.80 m o 0.305 m

vith broken plane only
transition (7.3)

Parshnll flumes
(22 types) (7.4) u -
rectangular
1.55
M 0.015 m
and
0.21 m
to
o 0.0254 m
to

--
0.03 m 0.33 m level 0.0762 m
u 1.522 to M 0.03 m, 0.45 m to floor
O. I524 m
u I .607 0.045 m. 0.76 m to
and 2.438 m

u - 1.60 M
0.076 m
0.09 m 1.07 m
to
3.048 m
to
1.83 m 15.24 m
H-flumes slopmg M 0.01 m 0.11 m o
-
see
( 3 type.) (7.5) traperilrm to to Figure
" 2.0 to 0.04 m 0.30 m 7.21
2.4
0.01 m 0.14 m
to to
0.03 m 1.36 m
0.03 m 1.06 m
and
1.21 m
Circular sharp-edged
orifice (8.1) -
Cll2"lP.r
u 0.5
M Ah20.03 m
hl a d
- 0.5 d A < 10 A l

d 3 0.02 m

Rectangular
sharp-edged rectangular M but M R AhF3.03 m - o b20.30 m
orifice (8.2) Y * 0.5 if
suppressed yI'o15 ~ 2 0 . 0 2m

constant head
orifice (8.3)
i -
rectangular
u 0.5
w Ahhr0.06 m
y, 3 2.5 w
Ah-0.06 m O usually'
b-0.60 m
and
b-0.75 m
 Y ' modular l i m i t error i n sensitive- debris sediment
%in %ax in cdcv or ness a t passing passing
m'/s or %ax " 2 4 minimum capaciti capacity
q max i n
%G or 'e head ++ very good; + good; Remarks
m'/s m21s head loss 6) % per 0 fair;
0.01 m - poor; - - very poor

0137 depends on 100.000* 0.67 10 cv-8 83 ++ + Applies to I-to-5 back

m0.03 m degree of h120.03 m if h1-0.03m face only.
0275 truncation
17,500 42 *y-values decrease if con-
-0.06 m h1>0.06 m i f h l -0.06 m trol ie more truncated

.
-0.30 m

0077 9-2.30 I20 0.70 3 32 + good if gate arrangement

~0.30m as in Section 4.2

025 variable' about 1000 0.30 5 25 ++ + * mini" pressure on

-1.0 m but depends crest limited to -4.0 m
on h -value water column (see
Pig.6.17)

,0064 variable' about 750. 0.33 5 25 ++ + 'mini- pressure on

-0.30 m but depends crest limited to -4.0 m
on ratio vater c d l " (see
Hl/= Pig.6.23)
.O066 variable 35 0.70 to 2(21-20cd) 25 ++ * for all f l m e ; at ma-
-0.30 m ..
vith 0.95 for xi" stage: B > 0.30 m
.O0898 throat depending ++ ++
length E315 on al1
28 to 42
-90 downstream
,0036 transition flmes "side slope ratio
mO.08 m ++ + + horr : vert. varies
E250 27 to 40
lope 1:2 betveen ! : I to 4:l

.O027 1 O0 33 ++ + +
-0.10 m
.O026 I O0 c 33 + + + +
-0.20 m if d20.60 m

.O050 q-4.82 I90 about' B 25 ++ if radii of rounding

=O.ZO m HI-2.00 m 0.50 and i f d o m e t r e m tran-
sition comply vith
Section 7.2.2

25 + + not rec-ended to be
constructed due to lack
af data

0.00009 0.0054 about 0.50 3 103 to 52 ++ + very small flumee; 1.2.

to to 55 and 3 inch vide
O. 00077 0.0321
0.0015 0.III about 0.60 3 53 to 21 ++ + small f l m e s ; 0.5.0.75.
to to 75 and I.O. 1.5, 2.3 to 8 feet
0.0972 3,949 0.70 vide
0.16 8.28 about 0.80 3 18 +* large flmes; IO 10
to to 93.04 50 feet vide
I05
0.75 m'ls m'ls

0.00001 2 0.0003 about 0.25 3 (240 HS-flmes, W O . 4 . 0.6,

to to I O0 0.8 and 1.0 ft
0.00034 0.0223
0.00003 I 0.009 about 0.25 3 (240 H-fl-s.WO.5, 0.75,
to to 750 1.0, 1.5. 2.0, 2.5, 3.0.
0.0014 2.336 and 4.5 ft
0.0018 2.369 about 0.25 3 .d 80 + HL-fl-s, D-3.5 and
and and 1500 4.0 ft
0.0020 3.326

0.0001 4 variable 5.8' submerged I 17 _- __ '0.03 m < Ah < 1.0 m

d-0.02 m

_-
0.0028 variable 5.8.
submerged 2 to 3 I7 O
if p-O
'0.03
and A -
m EA h < 1.0 m
constant

0.0086. Q-O. 140' 16** submerged. 2 7 8 _- W o sizes of orifice

but usually gates,0.60x0.45 m 6 0.75
0.0107 po. 280 26 AHt>0.30 m x0.60 m are ".used
** If A varies
TABLE 3 . 1 . DATA ON VARIOUS STRUCTURES (cont.)
1 2 3 4 5 6 7 8 9

Name of s t r u c t u r e Sketch of s t r u c t u r e Shape o f M = Hl min Hl max minimum minimum range o f

.nd s e c t i o n number control section measuring or crest size o f notch
i n which s t r u c t u r e perpendicular t o or height control angle 1
i s described flow and U-value :afuring Ah min Ah max above
approach
b o r E,
w and D
degree:
& regulating
channel
bottom p

Radial or
Tainter gete (8.4) u -
rectangvlar
0.5
NR ~ ~ 3 0 . 1m5 ~ ~ ( 1 .r2
yI>I.25 w
O bW.30 m

w>O.O2 m
YI>0.I r

Crump-de GNyter
adjuetable orifice
(8.5)
u -
rectangular
0.5
NR 0.03 m
1.58 v
0.60 m 0.20 m
p-b
b20.20 m
0.02 m < w
6 0.38 m
6 0 . 6 3 hl

Meter gate (8.6) Section of m hl>l.O D 0.17 D D Xl.30 m

circle
Ah>0.05 m Ahi0.45 m WI.75 O
"-0.5
w>O.O2 m

Neyrpic modules
(8.7) u -
rectangular
0.5
NR hd-0.17m
hd-0.28m
hd 6 P
and
hd<0.35 p2
0.16 m

0.26 m
0.05 m

0.05 m

circular or appro*. spprox. dN.02 m OoSBC1800

Danaidaan tub rectangular 0.10 m 5.0 m
(8.8) b>O.O2 m

0.06 m 1.0 p
Divisors
0.50 r 0.35 p2 0.33 H I 2.0 H l
4.0 r

-
Pipes and
small siphons
(9.2)
circular
" 0.5 - U 0.03 m 1.20 m 1.0 D D 20.015 m
D N.03 m

<D
Fountain flow
from vertical
pipe (9.3)
circular
u
"
--
1.35 or
0.53
U 0.03 m 4.00 m 0.025
D < 0.609

Flow from
circular ye-0.02m y, < D 83.05 m
horizontal pipes 0.1 D 0.56 D
(9.4)
" - 1.5
-
D,
r
Y < 0.15 'm - 0 . 0 5 6 Dn
(versus Y) Y20.025 6 0.15 m

Brruk depth
method for
rectangular canals
" -
rectangular
1.5
U ye>0.03m -* P - 0 0.30 m
y,
(9.5)

Dethridge rectangular NRV y1-0.30m y1-o.90m - 0.52 m

meterli (9.6) no "value yl=o.38m yI-O.90m - 0.78 m

Propeller v > 0.45 v < 5.0 0.05 m

meters (9.7) "BUSllY NRV
circular mle ds <Dp6
no u v a 1 v e
1.82 m
 10 11 12 13 14 15 16 17 18

y = modular l i m i t error i n sensitive- debris sediment

Qmi n amax in
m'/s o r Qmax HZlH1
CdCV or ;f;=f; passing
capacity
passing
capacity
q max i n G or 'e
head ++
0
very good; + good;
Remarks
m'ls m21s head loss (*) X per fair;
0.01 m -- poor; - - very poor

0.005 variable about' variable 5 8 + + * If A varies

yl-0.15 m 35 y is greater if gate is
lifted entirely

0.0088 q-O. 742 IO up to* 3 *If wlhl is small

0.25

0.0076 Q-2.10 7 to 65 h2 > 0.15 m 3 to 6 usually

0.20 m $ D d 1.22 m
D -0.30 m D -1.22 m AHt 2 0.30 m

0.0005 q-0.100 I* 0.60 5 3 __ 0 Type X I

0.0010 q-0.200 I 0.60 5 I .8 __ O Type KX 2
Discharge i8 regulated
by openinglclosing gates

1.00027
1-0.02 m
variable 7 hl+6d* 2 5 0 '6-
ficient
contraction eoef-

il-O.10 m

0.0075 q-5.69 30' 0.60 5 25 -- + Other weir profiles are

b-0.30 m Hl-2.00 m poesible

0.00006 variable 6 usually 10 17 __ __ L > 2 0 D

~ 0.00037 6 submerged 10 17 __ __ 6 0 ( L E 2 0 D

0.00068 Q-2.45 237 pipe must 15 to 20 50 __ -.

D -0.025 m D -0.609 m discharge

free into
the air

0.00062 variable 62 pipe must 3 I O0 _- __ Brink depth method

D -0.05 m discharge

0.0020 c=0.100 2.5 free into

the air
20 __ __ Trajectory method;
X-0.152, 0.305 and
D -0.05 m D -0.15 m . 0.457 m

0.0081 q-4.82 about head loss 3 25 + + yelYc.o.715

b-0.30 m Hl-2.0 m I75 2.1 H I
* Approach canal
length > I2 y

0.015 Q-0.070 4.6 head loss 5 + Small meter

2 0.08 m
0.040 rj=0.160 3.5 2 0.09 m 5 + + Large meter
at yl min.

0.00088 Q-13.0 IO "S"dlY 5' O + If propeller is

maintained
D -0.05 m D - I .82 m Ah > 0.50 m frequently
Column 12 - Value of y = Qmax/Qmin of the structure. If Qmaxcannot be calculated
directly, the y-value can usually be determined by substituting the limi-
tations on head (differential) in the head-discharge equation, as shown
in Section 3.2.3. I

Column 13 - Modular limit H,/H, or required total head loss over the structure.
The modular limit is defined as that submergence ratio H,/H, whereby
the modular discharge is reduced by 1OO/ due to an increasing tailwater
level.
Column 14 - Error in the product CdCVor in the coefficient Ce.
Column 15 - Maximum value of the sensitivity of the structure times 100, being

100 S = !!- Ah, 100

hi
where the minimum absolute value of h, is used with the assumption
Ah, = 0.01 m. The figures shown give a percentage error in the mini-
mum discharge if an error in the determination of h, equal to 0.01 m
is made. The actual error Ah obviously depends on the method by
which the head is determined.
Column 16- Classifies the structures as to the ease with which they pass floating
and suspended debris.
Column 17 - Classifies the structures as to the ease with which they pass bed-load
and suspended load.
Column 18 - Remarks.

3.4 Selecting the structure

Although it is possible to select a suitable structure by using Table 3.1, an engineer
may need some assistance in selecting the most appropriate one. To help him in this
task, we will try to illustrate the process of selection. To indicate the different stages
in this process we shall use differently shaped blocks, with connecting lines between
them. A set of blocks convenient for this purpose is defined in Figure 3.7.
All blocks except the terminal block, which has no exit, and logical decision blocks,
which have two or more exits, may have any number of entry paths but only one
exit path. A test for a logical decision is usually framed as a question to which the
answer is Yes or No, each exit from the Lozenge block being marked by the appro-
priate answer.

A block diagram showing the selection process is shown in Figure 3.8. The most impor-
tant parts of this process are:
- The weighing of the hydraulic properties of the structure against the actual situation-
or environment in which the structure should function (boundary conditions);
- The period of reflection, being the period during which the engineer tests the type

of structure and decides whether it is acceptable.

Both parts of the selection process should preferably bepassed through several times
to obtain a better understanding of the problem.
To assist the engineer to find the most appropriate type of structure, and thus the

1 IO
 stort or end
of progmm

(orithmotic)
operation

written
informotion

printing of
answer
(section No)

connector

E
a
-.
directions
flow Of

Figure 3.7 Legend of blocks diagram

relevant section number in the next chapters of this book, we have included Figure
3.9, which treats approximately that part of the selection process enclosed by the dotted
line in Figure 3.8. In constructing the diagram of Figure 3.9 we have only used the
most important criteria. The use of more criteria would make the diagram longer
and more complex.
After one or more suitable structures (sections) are found we recommend that Table
3.1 be consulted for a first comparative study, after which the appropriate section
should be studied. During the latter study one takes the secondary boundary condi-
tions into account and continues through the reflection branch of Figure 3.8 until
the proper structure has been selected.
It is stressed again that in this chapter the selection of structures is based purely
upon the best hydraulic performance. In reality it is not always desirable to alter the
existing situation so that all limits of application of a standard structure are fulfilled.
If, however, a structure is to be used to measure discharges and its head-discharge
relationship is not known accurately, the structure must either be calibrated in a hy-
draulic laboratory or calibrated insitu. Calibration in situ can be performed by using
the area-velocity method or the salt dilution method.

111
 0 ENGINEER

needs to
measure

see
and gives
boundary Fig.3.9
conditions

should

NO
I

of situation
using boundary

compares
boundary
conditions
and pro-
perties of

tests
structure (s)

__i""";;;!what is U
NO -
71
structure accept

r
unacceptable structure
able
and why

(K-)
Q
Figure 3.8 Selecting process of a discharge measuring or regulating structure

112
 q START

o
prime
function of
structure

Discharge function
measurement

Discharge
regulation

0 from pipe

OYES

/horik
LQ

Sect ions
I
o
decide if
information
on flow is
primarily
zontal or 9.2 or 9 . 4 required in
\tical pipe m3/period O K
v ms

4 VERT.
t I
o
Section 9.3 criteria

YES
o
cp> HAPPY
YES
fi device
ceptable
ac-

o
i
, discharRe
water into
open chan-
criteria
are met

0 YES
I

Figure 3.9a Finding the relevant structure (or section)

113
 NO

NO
calculate
y=@ax/Qmin

NO

sharp-crested
weir or sharp-

variation curve on
in head H,

U-value to slope
8.1,e.Z h next lower (U-value)
half point of curve

P No

NO Sect ion

Figure 3.9b Finding the relevant structure (or section)

114
 Y .6
calculate re-
quired modu-
lar limit
u
I

0Nno o
i+
p-value
Q>3.33 m/s U H1>1.36 m -

()YES
AYES VNO

Sections Approximate

H2/HI>O. 66

u 0 &/sa
6.3;6.4; U
YES H1>2.0 m e

/\Q
I
4.3 6 7.1

PNO
6.3;7.1

u .=>
4.3;4.5;
6.4;6.7;

Figure 3 . 9 ~
e
YES V
NO gular con- -
7Jtrol section

Finding the relevant structure (or section)

4.5;6.3;
6.7;7.1
 Q
max. head
l o s s AH
B max.

Q min.

Calculate
y-Qmax.lQ min

more paral-
l e l structu-

6.5 h 6.7

I I I

Q/B-value

I
approximate

apr.

- - - - 4.1 ;4.3;

4.1;43; -0- H2/Hl>O.33

4.1;4.2;

6.7:g.l

Figure 3.9d Finding the relevant structure (or section)

116
 Is regulator to be con-
structed on off-take
channel or on continu-

determine
required
flexibility

iNO QNO

NO but also an

0YES

orifice is
commonly
r'
--
OYES
YES

weir is
commonly
used

ONO
Sections
:iI dary crite-
ria to de-
cide whether
orifice or
weir is used
.%continue on
branch B or C

P
Sect ions
0.4;0.5;

Figure 3.9e Finding the relevant structure (or section)

117
Photo 2 The side walls of the channel in which the weir is placed are not parallel

Photo 3 If the limits of application of a measuring structure cannot be fulfilled, laboratory tests can provide
a head-discharge curve

118
3.5 Selected list of references
Berkhout, F.M.C. 1965. Lecture notes on irrigation engineering. F 18. University of Technology, Delft.
Bos, M.G. 1985. Long-throated flumes and broad-crested weirs, Nijhoff Publishers, Dordrecht, p. 141.
Engelund, F. and E. Hansen. 1967.A monograph on sediment transport in alluvial streams. Teknisk Forlag,
Copenhagen.
Mahbub, S.I. and N.D. Gulhati. 1951. Irrigation outlets. Atma Ram & Sons, India. 184 pp.
Meyer-Peter, E. and R. Miiller. Formulas for bed-load transport. Proc.Second meeting of the International
Association for Hydraulic Structures Res., Stockholm 1948.Vol. 2, Paper 2.
Netherlands Engineering Consultants (NEDECO) 1973. Rio Magdalena and Canal del Dique survey pro-
ject. The Hague.
Replogle, J.A. 1968. Discussion of rectangular cutthroat flow measuring flumes (Proc. Paper 5628). J. of
the Irrigation and Drainage Division of the ASCE. Vol. 94. No. IR3. pp.359-362.

119
4 Broad-crested weirs
Classified under the term 'broad-crested weirs' are those structures over which the
streamlines run parallel to each other at least for a short distance, so that a hydrostatic
pressure distribution may be assumed at the control section. To obtain this condition,
the length in the direction of flow of the weir crest (L) is restricted to the total upstream
energy head over the crest (HI). In the following sections the limitation on the ratio
H,/L will be specified for the following types of broad-crested weirs:
4.1 Horizontal broad-crested weir;
4.2 The Romijn movable measuring/regulating weir;
4.3 Triangular broad-crested weir;
4.4 Broad-crested rectangular profile weir;
4.5 Faiyum weir.
For details on other types of broad-crested weirs see Bos et al. (1984) and Bos (1985).

4.1 Horizontal broad-crested weir

4.1.1 Description

This weir is in use as a standard discharge measuring device and, as such, is described
in the British Standard 3680, 1969, which is partly quoted below. The weir comprises
a truly level and horizontal crest between vertical abutments. The upstream corner
is rounded in such a manner that flow separation does not occur. Flow separation
also can be avoided by using an upstream ramp which slopes between 2 - to - 1 and
3 - to - 1 (horz. to vert.). See Figure 1.34 for a longitudinal profile. This upstream
sloping face is a cost-effective solution if the weir is constructed in concrete. Down-
stream of the horizontal crest there may be a vertical face or a downward slope, de-
pending on the submergence ratio under which the weir should operate at modular
flow.
The weir structure should be rigid and watertight and be at right angles to the direc-
tion of flow.
The dimensions of the weir and its abutments should comply with the requirements
indicated in Figure.4.1. The minimum radius of the upstream rounded nose (r) is O. 1 1
HI,,,, although for the economic design of field structures a value r = 0.2 HI,,, is
recommended. The length of the horizontal portion of the weir crest should not be
less than 1.45 H,,,,. To obtain a favourable (high) discharge coefficient (cd) the crest
length (L) should be close to the permissible minimum. In accordance with Section
2.2 the head measurement section should be located a distance of between two and
three times HI,,, upstream of the weir block.

4.1.2 Evaluation of discharge

According to Equation 1-37 Section 1.9.1, the basic stage-discharge equation for a
broad-crested weir with a rectangular throat reads

121
Figure 4.1 Dimensions of round-nose broad-crested weir and its abutments (adapted from British Standards
Institution 1969)

For water of ordinary temperatures, the discharge coefficient (cd) is a function of

the upstream sill-referenced energy head (HI), and the length of the weir crest in the
direction of flow (L). It can be expressed by the equation (Bos 1985)
Cd = 0.93 + O.IOH,/L (4-2)
The appropriate value of the approach velocity coefficient (C,) can be read from Figure
1.12 (Chapter 1).
The error in Cdof a well maintained broad-crested weir, which has been constructed
with reasonable care and skill, can be deduced from the equation (Bos 1985).
X, = i-(3 I HJL- 0.55 1 + 4) per cent (4-3)
The method by which this error is to be combined with other sources of error is shown
in Annex 2.
Table 4.1 gives a series of rating tables for rectangular weirs. The groupings of weir
width were selected to keep the error due to the effects of the sidewalls to less than
1%. Ratings are given for a number of sill heights to aid in design. Discharges in
these tables are limited to keep the approach channel Froude number below 0.45.
Interpolation between sill heights will give reasonable results. If the approach area
is larger than that used to develop these rating tables, either because of a higher sill
or a wider approach, the ratings must be adjusted for C, (see Figure 1.12). To simplify
this process, the discharge over the weir for a C, value of 1 .O is given in the far right
column of each grouping. This discharge column is labeled as pI = co, since for C,
= 1.0 the velocity of approach is zero, as would be the case if the weir were the outlet

122
Photo 1 Downstream view of a broad-crested weir

of a deep reservoir or lake. Under this circumstance, the weir has the lowest discharge
for a given upstream head. Note that at the very low heads, the discharge for the
weirs with rectangular approach channels approaches pI = 00 because the approach
velocities are small.
The ratings given in Table 4.1 are for the throat lengths L given at the head of
each group columns. When the maximum design discharge of a structure is much
less than the maximum discharge shown in the rating table, the aforementioned throat
length may be longer than necessary. A value of L = 1.5 HI,,, is a reasonable com-
promise between providing a long enough throat to avoid the effects of streamline
curvature and minimizing the size of the structure. The throat length may be reduced
to this value provided that it does not become shorter than about two-thirds of the
L value in the table heading. Such a length reduction causes the weir discharge to
increase by less than 1%. The length of the converging transition Lb should be between
2 and 3 times pI. The distance between the gauging station and the start of the throat
+
(La Lb) should be between 2 and 3 times HI,,,, and the distance between the gauging
station and the start of the converging transition Lashould be greater than H I,ax.

123
Table 4.1 Rating Tables for rectangular Weirs in Metric Units with Discharge per Meter Width*

0.10 < b, 4 0 . 2 0 m 0.20 < b, <0.30m 0.30 4 b, < 0.50m 0.5 < b, < 1.0m
L = O.2m L = 0.35 m L = 0.5m L = 0.75m

hl q hl q hl q hl q
(m) (m3/spe* (m) (m3/sper (m) (m3/sper (m) (m3/sper
meter width) meter width) meter width) meter width)

PI = PI = PI = PI = PI = PI = PI = PI = PI = PI = PI =
0.05m m 0.1 m m O.lm 0.2m m 0.1 m 0.2m 0.3 m 00

,025 .O064 .O063 .O50 .O186 ,0183 ,0182 ,0181

,030 ,0085 .O084 .O55 .O2 16 ,0212 .o2 I o ,0209
.O14 ,0026 ,0026 .O35 .O108 ,0107 .O35 .o IO8 ,0106 .O106 ,060 .O248 ,0242 .O240 .0239
,016 ,0032 .O032 .O40 ,0133 .O131 .O40 .O133 ,0131 ,0130 .O65 .O28 I ,0274 ,0272 ,0270
.O18 ,0039 .O038 .O45 .O160 ,0157 .O45 .O160 ,0157 .O156 ,070 .O3 16 ,0308 .O305 ,0303
.O20 ,0046 ,0045 ,050 .O189 ,0184 .O50 ,0188 ,0185 .O183 .O75 ,0352 .O342 .O339 ,0336

.O22 ,0054 .O053 ,055 ,0220 .O213 .O55 .O2 19 ,0214 .o212 .O80 ,0390 ,0378 ,0374 ,0371
.O24 .O062 .O060 .O60 ,0252 .O244 .O60 ,0251 ,0245 .O242 .O85 ,0429 ,0416 .O41 1 ,0407
.O26 ,0070 .O068 .O65 .O285 .O275 .O65 .O285 ,0278 ,0274 .O90 ,0470 .O454 ,0449 ,0444
.O28 .O079 .O076 .O70 .O321 ,0308 .O70 .O320 .O3 I 2 ,0307 .O95 .o5 I2 .O494 ,0488 ,0482
.O30 .O088 .O085 .O75 ,0357 .O342 .O75 ,0357 ,0347 ,0341 ,100 ,0555 .O535 ,0528 ,0521

.O32 ,0097 ,0094 .O80 ,0396 .O377 .O80 ,0395 ,0383 .O376 .IO5 ,0600 ,0577 ,0570 .0561
.O34 .O107 .O103 .O85 .O435 .O414 .O85 .O435 ,0421 .O4 1 2 ,110 .O646 .O621 .O6 I2 .0602
.O36 ,0117 ,0112 ,090 ,0476 .O451 .O90 ,0476 .O460 .O450 .I15 ,0693 .O665 ,0656 .0644
.O38 .O128 .O122 .O95 ,0519 .O490 .O95 .O519 ,0500 ,0488 .I20 .O742 .O71 1 ,0700 ,0688
.O40 .O138 .O132 .IO0 ,0563 ,0529 .IO0 ,0561 ,0540 ,0528 .I25 ,0792 .O758 ,0746 ,0732

.O42 .O150 ,0142 .IO5 ,0608 .O570 .IO5 ,0606 ,0583 ,0567 .I30 .O843 .O806 ,0793 ,0776
.O44 .O161 .O153 .I10 ,0655 ,0611 .I10 ,0652 ,0626 ,0608 .I35 .O896 ,0855 ,0840 ,0822
.O46 ,0173 .O164 .I15 .O702 ,0654 .I15 .O700 ,0671 .O65 I ,140 ,0949 .O905 ,0889 ,0869
.O48 .O185 .O175 ,120 .O752 ,0697 .I20 .O748 ,0717 ,0694 .I45 .I004 .O956 ,0939 .O9 I6
.O50 .O197 .O186 ,125 .O802 ,0741 .I25 ,0798 ,0764 ,0738 .I50 ,1061 .I009 ,0989 ,0965

.O52 ,0210 ,0197 ,130 ,0854 .O787 ,130 ,0850 ,0812 .O783 ,155 .I I18 .I062 .IO41 ,1014
.O54 ,0223 ,0209 ,135 .O907 ,0833 .I35 .O902 ,0861 ,0828 .I60 ,1176 .I116 .I094 ,1064
.O56 .O236 ,0221 ,140 ,0961 .O880 ,140 ,0956 .O91 I .O875 .I65 ,1236 .I172 ,1147 .I I I5
,058 ,0250 ,0233 .I45 ,1017 ,0928 .I45 .IO11 ,0962 ,0923 .I70 ,1297 .I228 . I202 ,1166
.O60 .O264 .O245 .I50 ,1074 ,0977 .I50 . I067 ,1014 ,0971 .I75 ,1359 ,1285 . I257 .1219

.O62 ,0278 ,0257 ,155 ,1132 ,1026 ,155 , I 125 .I068 ,1020 .I80 ,1422 .I344 ,1314 . I272
.O64 .O293 ,0270 .I60 ,1191 ,1077 . I60 ,1183 .I122 . I070 .185 .I486 .I403 ,1371 ,1325
.O66 .O307 ,0283 .I65 ,1251 .I128 ,165 ,1243 ,1177 . I 121 .I90 .I552 ,1464 .I430 ,1380
.O68 .O322 ,0296 .I70 ,1312 ,1180 ,170 .I304 .I234 ,1173 ,195 ,1618 .I525 ,1489 ,1435
,070 ,0338 ,0309 ,175 ,1375 .I233 ,175 ,1366 ,1291 ,1225 ,200 .I686 . I587 .I549 ,1492

.O72 .O353 ,0323 .I80 ,1439 ,1286 .I80 .I429 .I349 .I278 .210** ,1824 ,1715 .I671 . I606
.O74 ,0369 ,0337 ,185 ,1504 ,1340 ,185 .I493 .I409 ,1332 220 .I967 . I846 ,1798 . I723
.O76 ,0385 ,0350 ,190 ,1567 .I396 .I90 .I559 .I469 ,1387 ,230 .21 I3 ,1981 .I927 . I843
.O78 .O402 ,0365 .I95 ,1633 ,1451 .I95 .I625 .I530 .I442 ,240 ,2264 .21 I9 ,2060 ,1965
.O80 ,0419 ,0379 ,200 ,1701 ,1508 ,200 ,1693 .I593 . I498 ,250 .2419 ,2262 2197 .2090

.O82 ,0436 ,0393 ,205 ,1770 .I565 .205 ,1762 . I656 ,1555 .260 .2578 ,2407 .2336 ,2217
.O84 ,0453 .O408 ,210 ,1840 .I623 .210 ,1831 . I720 .I612 .270 .2741 .2557 2479 ,2348
.O86 ,0470 .O423 ,215 .I911 ,1681 .215 .I902 . I786 .I671 ,280 ,2908 .2709 ,2625 ,2480
.O88 ,0488 .O438 ,220 ,1983 .I741 220 .I974 . I852 ,1730 ,290 ,3078 2866 ,2775 ,2610
.O90 ,0506 .O453 ,225 .2056 ,1801 .225 ,2047 ,1919 .I789 ,300 ,3253 ,3025 ,2927 ,2752

.O92 .O524 .O468 ,230 ,2130 .I861 230 ,2121 ,1987 .I849 .310 ,3431 .3 188 ,3083 .2892
.O94 ,0543 .O484 ,235 ,2205 .I923 235 ,2196 ,2056 ,1910 .320 ,3613 .3355 ,3242 .3034
.O96 ,0562 ,0499 ,240 ,2272 ,2125 .I972 .330 .3799 ,3524 ,3404 .3 178

124
Table 4.1 (continued)

0.10$bC~0.20m 0.20$ b C 4 0 . 3 0 m 0.304bC40.50m 0.5 4 b, 4 1.0m

L = 0.2 m L = 0.35 m L = 0.5m L = 0.75 m

PI = PI = PI = PI = PI = PI = PI = PI = PI = PI = PI =
0.05m m 0.1 m m 0.1 m 0.2m m O.lm 0.2m 0.3m m

.O98 ,0581 .o5 15 ,245 .2349 .2 I96 ,2034 .340 .3988 .3697 ,3568 ,3325
.IO0 ,0600 .O531 ,250 .2427 .2268 ,2097 ,350 .4181 ,3873 ,3736 ,3473

.105** ,0649 .O571 .260** ,2587 .2414 ,2225 .360 ,4378 .4053 ,3907 ,3624
.I10 ,0700 .O61 3 .270 ,2750 ,2563 ,2355 .370 .4235 ,4081 ,3777
.I15 .O753 ,0656 ,280 ,2917 ,2716 .2488 ,380 ,4421 ,4258 ,3932
,120 .O806 ,0699 .290' ,3088 ,2872 .2623 .390 ,4610 ,4438 ,4089
,125 .O861 .O744 ,300 ,3262 ,3032 .2760 .400 ,4802 ,4620 .4248

,130 ,0918 .O789 .310 ,3441 ,3195 .2900 ,410 ,4998 .4806 ,4409
,320 ,3623 .3361 ,3042 ,420 ,5196 ,4994 ,4573
,330 ,3808 .3531 ,3186 ,430 ,5397 ,5185 ,4738
,440 ,5601 S379 .4905
,450 SS09 376 SO74

,460 .6019 ,5776 ,5245

.470 .6232 ,5978 ,5418
,480 ,6448 .6183 ,5593
.490 .6667 ,6391 ,5769
SO0 .6888 ,6601 S948

AH = 0.01 2 m AH = 0.025m AH = 0.027 m 0.044 m AH = 0.028 m 0.048 m 0.063 m

or or or or or
O.lHI O.IHI . 0.1H1 0.1H1 0.1H1

* L~ = 2 or 3 times p I ; La > Hlmax; La + Lb > 2 to 3 times Hlmax. (continued)

** Change in head increment

125
 Table 4.1 (continued)

1.0 < b, < 2.0m b, 3 2.0 m

L = I.Om L = 1.0m

hl 4
(m)
hl (m3/s per meter width)
q (m) (m3/s per meter width)

PI = PI = PI = PI = PI = PI = PI = PI =
0.2m 0.3m 0.4m co 0.2m 0.4m 0.6m 00

,100 .O521 .o51 I ,0508 ,0506

.I20 .O695 ,0680 ,0675 ,0671

.O70 ,0304 .O301 ,0300 .O298 ,140 ,0889 .O866 ,0858 ,0852
.O80 ,0374 .O370 ,0369 .O298 .I60 .I099 ,1067 ,1056 . I046
.O90 ,0450 .O445 ,0442 .O439 ,180 .I326 .I283 . I268 .1253
,100 ,0531 ,0524 ,0521 .O516 .200 .I596 ,1513 ,1493 ,1473

.I10 .O6 I 6 ,0608 ,0604 .O597 ,220 ,1827 ,1756 ,1732 ,1704
.I20 .O706 ,0696 ,0691 .O683 .240 .2101 ,2013 ,1982 .1946
.I30 ,0801 .O788 .O782 ,0771 .260 ,2389 ,2283 ,2245 .2199
,140 .O900 .O885 .O877 .O864 .280 .2691 .2565 ,2519 ,2461
.I50 .I004 ,0985 .O976 .O960 .300 .3008 ,2859 ,2805 .2733

.I60 ,1112 .I090 .I079 ,1059 .320 ,3337 .3165 ,3101 ,3015
.I70 .I224 . I 198 , I 185 .I161 .340 .3681 ,3483 ,3409 ,3306
.I80 .I339 ,1319 .I295 .I267 .360 .4037 .3812 ,3727 .3606
. .I90 ,1459 ,1426 . I408 ,1375 ,380 ,4406 ,4153 ,4056 .3914
.200 ,1583 ,1545 ,1525 ,1487 .400 ,4788 ,4505 ,4395 ,4231

.210 ,1711 ,1668 . I646 ,1601 .420 ,5182 ,4868 ,4744 .4556
,220 .I842 ,1794 ,1769 .I718 ,440 ,5588 ,5241 ,5103 ,4889
.230 ,1977 ,1924 ,1896 ,1838 ,460 ,6007 ,5626 ,5472 5229
,240 .2116 ,2058 ,2027 .I961 ,480 ,6437 ,6020 ,5851 ,5577
,250 .2259 ,2194 ,2160 ,2086 ,500 ,6878 ,6425 ,6239 5932

.260 ,2405 ,2334 .2291 ,2214 ,520 ,7331 .6840 .6636 .6295
.270 ,2542 ,2477 ,2436 ,2344 ,540 ,7796 ,7265 ,7042 .6664
.280 ,2708 ,2624 ,2579 ,2477 ,560 ,8271 .7699 ,7458 .7041
.290 ,2864 ,2774 ,2725 .26 I2 ,580 ,8758 ,8144 ,7884 .7425
.300 ,3024 ,2927 ,2873 ,2749 ,600 ,9527 ,8600 ,8319 .7815

,310 ,3188 ,3083 ,3025 ,2889 ,620 ,9765 ,9063 3762 ,8212
.320 ,3355 .3242 ,3180 ,3032 ,640 1.O28 ,9537 .9214 ,8615
,330 ,3525 ,3404 ,3337 .3176 ,660 1.O8I 1.002 .9674 .9025
.340 ,3698 ,3569 ,3498 ,3323 ,680 1.135 1.051 1.014 .9441
,350 ,3875 ,3738 ,3661 ,3472 ,700 1.191 1.101 I .O62 .9864

,360 .4055 .3909 ,3828 ,3623 ,720 1.153 1.111 1 .O29

,370 ,4238 ,4083 ,3997 ,3776 ,740 1.205 1.160 I .O73
.380 ,4424 ,4261 .4 168 .3931 ,760 1.257 1.210 1.117
.390 ,4614 ,4441 ,4343 ,4088 ,780 1.311 1.262 1.161
.400 ,4806 ,4624 ,4520 .4248 ,800 1.366 1.314 1.207

,410 ,5002 .48 I O ,4701 ,4409 320 1.422 1.367 1.252

.420 ,5200 ,4999 ,4883 .4573 ,840 1.478 1.420 I .299
.430 ,5401 ,5190 ,5069 .4738 ,860 1.535 1.474 1.346
,440 ,5607 S385 S257 ,4905 380 1.593 1.530 1.393
.450 ,5815 ,5582 ,5447 ,5075 ,900 1.652 1.586 1.441

126

i
Table 4.1 (continued)

1.0 4 b, 4 2.0 m b, 2 2.0 m

L = I.Om L = 1.0m

hl q hl q
(m) (m3/s per meter width) (m) (m3/s per meter width)

p1 = PI = PI = PI = PI = PI = PI = PI =
0.2 m 0.3 m 0.4m O.2m 0.4m 0.6m

,460 .6025 ,5782 ,5641 ,5246 ,920 1.712 ,642 1.490

,470 ,6238 ,5984 ,5837 S419 .940 1.773 ,700 1.539
,480 ,6455 ,6189 ,6035 394 .960 1.834 ,758 1.588
,490 5674 ,6398 .6236 ,5771 .980 1.897 817 1.638
so0 ,6896 ,6608 .6440 ,5950 I .o00 1.960 ,877 1.689

SI0 ,7122 .6822 .6646 ,6130

.520 ,7350 .7038 ,6855 .63 I2
,530 .7580 .7257 ,7065 ,6496
.540 .7814 ,7478 ,7279 ,6682
.550 3050 ,7702 ,7495 ,6869

,560 3290 ,7929 ,7715 .7059

,570 .8532 ,8158 ,7936 .7249
,580 ,8776 ,8390 ,8159 ,7442
,590 ,9024 ,8624 ,8385 ,7636
,600 .9274 ,8861 ,8613 ,7832

,610 ,9527 ,9102 .8844 ,8029

,620 .9782 ,9343 ,9077 ,8228
,630 1 .O04 ,9588 .9312 ,8429
,640 1 .O30 .9835 .9550 ,8632
.650 1.056 1 .O08 ,9790 ,8836

,660 1 .O83 I .O34 I .O03 ,9041

,670 1.110 I .O59 I .O28 ,9249

AH = 0.046 m 0.066 m 0.086 m AH = 0.047 m 0.087 m 0.124 m

or or or or or
0.1H1 O.IH1 0.1H1 0.1H1 O.IH1

* L~ = 2 or 3 times pI; La La
2 HImax; + Lb > 2 to 3 times Himax
** Change in head increment

127
4.1.3 Modular limit

The flow over a weir is modular when it is independent of variations in tailwater level.
For this to occur, assuming subcritical conditions in the tailwater channel, the tailwater
energy level (H,) must not rise beyond a certain percentage of the upstream energy
head over the weir crest (HI). Hence, the height of the weir above the bottom of the
tailwater channel (p,) should be such that the weir operates at modular flow at all
discharges. The modular limit can be read from Figure 4.2 as a function of H,/p,
and the slope of the back face of the weir. A more accurate design value of p2 may
be established by the method persented in Section I . 15.

4.1.4 Limits of application

a. The practical lower limit of h, is related to the magnitude of the influence of fluid
properties, to the boundary roughness, and to the accuracy with which h, can be
determined. The recommended lower limit is 0.06 m or 0.05 times L, whichever
is greater.
b. The limitations on HJp, arise from difficulties experienced when the Froude
. ~ approach channel exceeds 0.45.
number Fr, = V , / ( ~ A , / B ,in) ~the
c. The limitations on H,/L arise from the necessity of ensuring a sensible hydrostatic
pressure distribution at the critical section of the crest and of preventing the forma-
tion of undulations above the weir crest. Values of the ratio H,/L should therefore
range between 0.08 and 0.7.
d. The breadth (b,) of the weir crest should not be less than L/5.

modular limit H2/H,

ratio H1/p2

Figure 4.2 The modular limit as a function of H,/pz

128
4.2 The Romijn movable measuring/regulating weir
4.2.1 Description

The Romijn weir was developed by the Department of Irrigation in Indonesia as a

regulating and measuring device for use in relatively flat irrigated regions where the
water demand is variable because of different requirements during the growing season
and because of crop rotation. A description of the weir was published in 1932 by
Romijn, after whom the structure is named.
The telescoping Romijn weir consists of two sliding blades and a movable weir which
are mounted in a steel guide frame:
a. the bottom slide is blocked in place under operational conditions and acts as a
bottom terminal for the movable weir
b. the upper slide is connected to the bottom slide by means of two steel strips placed
in the frame grooves and acts as a top terminal for the movable weir;
c. the movable weir is connected by two steel strips to a horizontal lifting beam. The
weir crest is horizontal perpendicular to the flow and slopes 1-to-25 upward in
the direction of flow. Its upstream nose is rounded off in such a way that flow
separation does not occur. The operating range of the weir equals the maximum
upstream head (h,) which has been selected for the dimensioning of the regulating
structure (see Figure 4.3).

2 p5m
STIFFENER
iR PLATE '
5 0.5 p2min
or
3 0.5 H, max
or
50.15 m

BOTTOM S I D E

Figure 4.3 The Romijn movable weir

129
 Although the Romijn weir has been included in this chapter on broad-crested weirs,
from a purely hydraulic point of view this is not quite correct. Above the I-to-25
sloping weir crest the streamlines are straight but converging so that the equipotential
lines are curved. At the same time, the control section is situated more towards the
end of the crest than if the crest were truly horizontal. Therefore, the degree of down-
ward curvature of the overflowing nappe has a significant influence on the C,-value.
To prevent the formation of a relatively strong eddy beneath the weir crest and
the overflowing nappe, the weir should have a vertical downstream face. The reason
for this is that especially under submerged flow conditions the nappe will deflect up-
wards due to the horizontal thrust of the eddy, resulting in up to 7% lower weir flows.
The downstream weir face, which breaks the force of the eddy should have a minimum
height of 0.5 pzmln or 0.5 Hlma,or O. 15 m, whichever is greater.
'As mentioned, the bottom slide, and thus the upper slide, is blocked in place during
normal flow conditions. However, to flush sediments that have collected upstream
of the weir, both slides can be unlocked and raised by moving the weir crest upward.
After flushing operations the slides are pushed in place again by lowering the weir
crest. To discourage misuse of the weir, the maximum flow capacity beneath the lifted
bottom gate must be less than the flow over the weir in its lowest position. For this
to occur, the travel of the upper gate is restricted so that the bottom gate cannot
be lifted higher than 0.5 Hlma,above the approach channel bottom.
The weir abutments are vertical and are rounded in such a way that flow separation
does not occur. A rectangular approach channel is formed to assure an even flow
distribution. The upstream head over the weir, h,, is measured in this approach channel
at a distance of between two and three times Hlmax upstream of the weir face. The
dimensions of the abutment should comply with the requirements indicated in Figure
4.4. The radius of the upstream rounding-off of the abutments may be reduced to
r k Hlma,if the centre line of the weir structure is parallel to or coincides with the
centre line of the undivided supply canal (in-line structure) or if the water is drawn
direct from a (storage) basin.
If several movable weirs are combined in a single structure, intermediate piers should
be provided so that two-dimensional flow is preserved over each weir unit, allowing
the upstream head over the weir to be measured independently per unit. The parallel
section of the pier should therefore commence at a point located at a distance of H,,,,
upstream of the head measurement station and extend to the downstream edge of
the weir crest. Piers should have streamlined noses, i.e. of semi-circular or tapered
semi-elliptical profile (i-to-3 axis). To avoid extreme velocity differences over short
distances, the thickness of the intermediate piers should be equal to or more than
0.65 Hlma,,with a minimum of 0.30 m.
Since the weir crest moves up and down, a fixed staff gauge at the head measurement
station does not provide a value for the upstream head over the crest unless the weir
crest elevation is registered separately in terms of gauged head. To avoid this proce-
dure, the weir is equipped with a gauge that moves up and down with the weir crest
(see Fig. 4.4). Zero level of this gauge coincides with the downstream edge of the weir
crest, so that the upstream head over the crest equals the immersed depth of the gauge
and can be read without time lag. The movable gauge is attached to the extended
lifting beam as shown in Figure 4.7.

130
 __
aeration groove
3 5H,max

-
-
DIVERTED FLOW

- .-
22H1max- - 2 H p u ~ - 2 t03H1max
TRANSITION

arrangement

3 2 times maximum depth to

I
water surface in well

Figure 4.4 Hydraulic dimensions of weir abutments

4.2.2 Evaluation of discharge

According to Equation 1-37, Section 1.9.1, the basic head discharge equation for a
broad-crested weir with a rectangular control section reads

Values of the discharge coefficient Cd may be read from Figure 4.5 as a function of
the ratio HJL.

Since the weir crest height above the approach channel bed (p,) is variable and to
a certain extent independent of the head over the weir crest h,, the approach velocity
cannot be predicted unless p, is known. Engineers therefore tend to use either a con-
stant C,-value of 1.055 for all values of H,/L or use Figure 4.5 to determine Cd by
assuming that h, N H I .
Values for the approach velocity coefficient C, may be read from Figure 1.12 as
+
a function of the dimensionless ratio Cdhl/(hl p,);where p, is the variable height
of the movable weir crest above the bottom of the rectangular approach channel. Over
the range of p,-values, an average C,-value may be used in Equation 4-4(see also
Figure 4.8).

131
DISCHARGE
COEFFICIENT C A

Figure 4.5 Values of Cd as a function of HI/L for the Romijn weir

If a movable Romijn weir has been constructed and installed with reasonable care
and skill, its discharge coefficient Cd may be expected to have an error of less than
3%. If an average value of Cd = 1.055 is used for all ratios of HJL, this Cd-values
may be expected to have an error of less than 4%. To obtain these accuracies the
weir should be properly maintained. The error in the C,-coefficient depends on the
minimum value of pI and the operating range of the movable weir. For the two most
common weir types the error in C, may be obtained from Section 4.2.4 and Figure
4.8. The method by which the coefficient errors have to be combined with other sources
of error is shown in Annex 2.

4.2.3 Modular limit

In order to obtain modular flow the submergence ratio H,/H, for which the modular
discharge is reduced by 1%O owing to the increasing tailwater level, should not exceed
0.30.
Results of laboratory tests have shown that the drowned flow reduction factor,
and thus the modular limit, depends on a number of factors, such as the value of
the ratio H,/L and the crest height above the tailwater channel bottom, pz. Since most
energy loss occurs in the bottom eddy immediately downstream of the weir crest, little
or no influence on the modular limit was observed if the side walls of the weir either
terminated abruptly or flared under 1-to-6. Values of the average drowned flow reduc-
tion factor, f, (i.e. the factor whereby the equivalent modular discharge is decreased
due to submergence) varies with H,/H, as shown in Figure 4.6.
To prevent underpressure beneath the nappe influencing the discharge, the air pock-
et beneath the nappe should be fully aerated, for example by means of the two aeration
grooves as shown in Figure 4.4.

132
4.2.4 Commonly used weir dimensions

The reader will have noted that all dimensions of both the weir and its abutments
are related to the maximum value selected for the total energy head over the weir
crest (HImax). The loss of head required for modular flow is also related to the total
energy head as Ah = h, - h, 2 0.70 HI,,,.
Since the limiting factor in most relatively flat irrigated areas is the available head
for open canal and weir flow, the maximum value of h, is limited to a certain practical
value which approximates 0.45 m. The length of the weir crest in the direction of
flow consequently equals L = 0.60 m, of which 0.50 m is straight and sloping 1 to
25 upward in the direction of flow and the remaining O. 10 m forms the rounded nose,
its radius also being O. 10 m.
Theoretically any weir breadth greater or equal to 0.30 m may be used but to obtain
a degree of standardization in the structures of an irrigation project a limited number
of breadths should be employed. It is often practicable to use a breadth not greater
than b, = 1.50 m, since a central handwheel can then be used to move the weir while
the groove arrangement can be a relatively simple one consisting of steel blades sliding
in narrow (0.0 1 m) grooves. If the breadth b, exceeds 1 S O m, a groove arrangement
as shown in Section 6.5.1 may be used.
Examples of constructional drawings are shown in Figure 4.7.

SUBMERGENCE RATIO
HzHi

. 0.01 I#

0 -
0.6 0.7 0.8 0.9 1.o
DROWNED FLOW REDUCTION FACTOR f

Figure 4.6 Drowned flow reduction factor for Romijn weir

133
i
W
A
P
P

HOLE 25x65: @
L. 50 "100.8
8-10

/?lpLr1" 1
8

W I R TABLE AND
HOISTING B E A M 8*1oo.c92.e
SECTION A _ B L
UPSTREAM VIEW
TERMINAL GATES
L50xlOOx8
____
strip 4 5 x 1 0
L5ox1ooxa s r i p 100 x 8
 DETAILS HANDWHEEL (housing and bronze n u t ) DETAILS CROSS SECTION A-B
I 600
1650,
I

__
TT
6 0 120

I I
660
BLOCKING WEDGE
thick 12
7
LOCKING HANDLE

121
'25 75
, I
2b

DETAILTOP CORNER FRAME

4
d

W
ul Figure 4.7 The Romijn movable measuring/regulating weir (dimensions in mm)
If the Romijn weir is installed in accordance with Figure 4.3, which is the normal
method of installation, the values for h, and p, vary in such a way that
0.05 m < h, & 0.45 m
0.55 m < p, < 0.95 m
0.60 m < h, + p, < 1.00 m
Due to the variation of both h, and p,, the approach velocity coefficient is not a func-
tion of h, alone, but ranges between the broken lines shown in Figure 4.8. In irrigation
practice it is confusing to work with several C,-values for the same upstream head.
Therefore the use of an average C,-value, as a function of the upstream head h, only,
is advised. It follows from Figure 4.8 that this average C,-value may be expected to
have an error of less than 1%. The discharge in m3/s per metre width of weir crest
can be calculated from Equation 4-4 and Figures 4.5 and 4.8. Values of q for each
0.01 m of head are presented in Table 4.2, Column 2.

An alternative method of installing the weir is to use no bottom slide. The movable
weir is then lowered behind a drop in the channel bottom, this drop acting as a bottom
terminal. With this method, the height of the weir crest above the bottom of approach
channel is less than with the normal method of installation. Consequently, the ap-
proach velocity and thus the C,-value is significantly higher. For a standard weir with

APPROACH VELOCITY COEFFICIENT Cv (dimensionless)

NOTE: The t o t a l upstream head over the weir ( h , )

should be measured between 0.90 m and 1.35 m up-
stream of t h e f a c e of the w e i r i n a rectangular
approach channel whose width equals the width of
t h e w e i r (b,)and the water depth equals (PI+ h , ) .
The flowwise length o f the weir c r e s t i s L-0.60 m. ,
/

1.15
I
I //

1.14 I

,/

Upper limit of h,

O 0.05 0.10 0.20 0.30 0.40 0.45 m e t r e

UPSTREAM HEAD OVER THE WEIR CREST(h1)

Figure 4.8 Approach velocity coefficient (C,) as a function of the head over the movable weir crest (h,)

136
 a length of the weir crest in the direction of flow of 0.60 m, values of p, and h,'range
in such a way that:
0.05 m d h, d 0.45 m
0.15 m 6 p, 6 0.55 m
0.20m d h, + p, 6 0.60m
+
Values of the ratio Cdhl/(hl p,) thus range more widely than before, as do C, values
as a function of h,. Minimum and maximum possible Cv-values are shown in Figure
I
4.8. Here, the average Cv-value to be used may be expected to have an error of less
than 4%. Values of q for each 0.01 m of head may be calculated from Equation 4-4
and from Figures 4.5 and 4.8, and are presented in Table 4.2, Column 3.

I 4.2.5 Limits of application

The limits of application of a movable Romijn weir for reasonable accuracy are:
a. The practical lower limit of h, is related to fluid properties and to the accuracy
with which gauge readings can be made. The recommended lower limit of h, is
0.05 m or 0.08 L, whichever is greater; I
b. To reduce the influence of boundary layer effects at the sides of the weir, the weir
breadth b, should not be less than 0.30 m nor less than the maximum value of
Hl;
c. The height of the weir crest above the bottom of the approach channel should
not be less than O. 15 m nor less than 0.33 Hlmax;
d. To obtain a sensibly constant discharge coefficient, the ratio H,/L should not exceed
0.75;
e. The submergence ratio H,/HI should not exceed 0.30 to obtain modular flow.

4.3 Triangular broad-crested weir

4.3.1 Description

On natural streams where it is necessary to measure a wide range of discharges, a

triangular control has several advantages. Firstly it provides a large breadth at high
flows so that the backwater effect is not excessive. Secondly, at low flows the breadth
is reduced so that the sensitivity of the weir remains acceptable. These advantages,
combined with the fact that a triangular control section has a critical depth equal
to 0.8 H l so that the weir can take a high submergence before its capacity is affected,
makes this weir type an interesting flow measuring device. A description of the weir,
although slightly different in shape, was published in 1964 by Bos.
The weir profile in the direction of flow shows an upstream rounded nose with
a minimum radius r equal to 0.1 1 Hlmax to prevent flow separation. For the economic
design of field structures, however, a value r = 0.20 Hlmax is recommended. To obtain
a sensibly hydrostatic pressure distribution above the weir crest, the length of the hori-
zontal portion of the crest should not be less than 1.75 Hlmax. To obtain a favourable
(high) discharge coefficient the crest length L should be close to the permissible mini-
mum. The weir should be placed between vertical abutments and be at right angles

137
Table 4.2 Discharge per metre width of weir crest for the movable Romijn measuring/regulatingweir

Head Discharge q in m3/s per metre width for two methods of installation
hl
metre normal alternative
0.55 m < pI < 0.95 m 0.15111 < p, < 0.55111
0.05 0.0195 0.0196
0.06 ,0258 ,0260
0.07 ,0327 ,0332
0.08 ,0402 ,0408
0.09 ,0483 .O491
0.10 ,0568 ,0579

0.11 ,0658 ,0672

0.12 ,0752 ,0770
0.13 ,0850 ,0873
O. 14 ,0952 ,0980
0.15 ,106 ,109

0.16 ,117 ,121

0.17 ,128 ,133
0.18 ,140 ,145
0.19 ,152 ,158
0.20 ,164 ,171

0.21 ,176 ,185

0.22 ,189 ,199
0.23 ,202 ,213
0.24 ,216 ,228
0.25 . ,230 .243

0.26 ,244 ,259

0.27 ,258 ,275
0.28 ,273 .292
0.29 ,288 ,310
0.30 ,304 ,327

0.31 ,319 ,345

0.32 .336 ,365
0.33 ,353 ,384
0.34 ,370 ,404
0.35 ,388 ,426

0.36 ,407 .448

0.37 .425 ,470
0.38 ,444 ,493
0.39 ,464 ,517
0.40 ,484 ,541

0.41 ,504 ,566

0.42 ,525 ,592
0.43 .547 ,619
0.44 ,569 ,646
0.45 .591 ,675

NOTE: The number of corresponding figures given in the columns for discharge should not be taken to
imply a corresponding accuracy of the values given, but only to assist in the interpolation and
rounding off for various values of head.

138
to the direction of flow. The upstream head over the weir crest should be measured
in the rectangular approach channel at a distance of between two and three times
Hlmax upstream from the weir face (see also Chapter 2).
Essentially, there are two types of triangular broad-crested weirs:
(i) if the maximum weir width is unrestricted (i.e. if the available weir width is such
that in combination with a selected weir notch angle 8, the water level in the control
section does not reach the intersection of side slopes and vertical abutments), the weir
type is referred to as less-than-full. For this type of weir, one head-discharge equation
applies for the entire operating range from Hlmin to Hlmax.
(ii) if the weir is installed in a channel with restricted width, the water level at the
control section may sometimes rise above the top of the side slopes. This weir type
is referred to as over-full, and somewhere in between Hlminand Himaxwe have to
change over from the head-discharge equation for a triangular control section to that
of a truncated triangular control section.
As shown in Sections 1.9.3 and 1.9.4, critical depth in a triangular control section

Photo 2 Triangular broad-crested weir

139
Figure 4.9 Definition sketch for triangular broad-crested weir

equals y, = 0.80 Hl, so that the weir is just full if H, = 0.80 HI or HI = 1.25 H,,
where H, denotes the difference in elevation between the top of the side slopes and
the vertex of the weir notch (see Figure 4.9) and equals H, = 1/2 B, cot 8/2.

4.3.2 Evaluation of discharge

As discussed already in Section 4.3.1 we can distinguish between two different

cases of head-discharge relationships, as follows
Less-Than-Full(H, 5 1.25 H,)
In this case the basic head-discharge equation for a triangular control section
is applicable, which, according to Section 1.9.3, reads

Q = [3 g] tanZe h12.50
16 2 0.50
CdCV25 (4-5)

where the discharge coefficient may be read as a function of the ratio H,/L from Figure
4.10. The approach velocity coefficient may be read from Figure 1.12 as a function
of the dimensionless ratio
A* h12tane/2
CdA = Cd x
I Bdhl + PI)

Over-Full (HI > 1.25 H,)

In this case the basic head-discharge equation for a truncated triangular control section

140
applies (see Section 1.9.4)

Q =CC 2
'3 3
[zg7-50B,(hl - 1/2Hb)1.50 (4-6)

where values of Cd again may be read from Figure 4.1O as a function of the ratio
HJL. It should be noted that if H,/L exceeds 0.50 the weir cannot be termed broad-
crested. If ratios H,/L 2 0.50 are used, the overfalling nappe should be fully aerated,
and it should be noted that the modular limits given in Section 4.3.2 will decrease
significantly with increasing H,/L-values. C,-values'may be obtained from Figure 1.12
as a function of the dimensionless ratio C,A*/A, = Cd(h,- l/ZHb)/(hI p,). +

4
L

Figure 4.10 Cd values as a function of H,/L of broad-crested weirs and long-throated flumes of all shapes
and sizes (Bos 1985)

141
 The error in the discharge coefficient (including C,) of a triangular broad-crested (trun-
cated) weir, which has been constructed with reasonable care and skill, may be deduced
form the equation

I X, = f (3 I H,/L - 0.55 I I + 4) per cent (4-7)

I
I

The method by which this error has to be combined with other sources of error is
I
shown in Annex 2.

4.3.3 Modular limit

a. Less-than-full case
The modular limit, or that submergence ratio HJH, which produces a 1 YOreduction
in the equivalent modular discharge, depends on a number of factors, such as the
value of the ratio H,/Hband the slope of the downstream weir face. Results of various
tests by the Hydraulic Laboratory, Agricultural University, Wageningen, 1964-1971,
and by Smith & Liang (1969), showed that for the less-than-full type weir (H,/Hb
d 1.25) the drowned flow reduction factor (0 (i.e. the factor whereby the equivalent
modular discharge is decreased due to submergence), varies with H,/H,, as shown
in Figure 4.11. The modular limit for weirs with a vertical back-face equals H,/H,
= 0.80. This modular limit may be improved by constructing the downstream weir
face under a slope of I-to-6 (see also Figure 4.2) or by decreasing p2.
b. Over-fullcase
No curve is available to evaluate the modular limit of over-full type weirs. It may
be expected, however, that the modular limit will change gradually to that of a broad-
crested weir as described in Section 4.1.1 if the ratio H,/Hbincreases significantly above
1.25. A more accurate estimate of the modular limit can be made by use of Section
1.15.

SUBMERGENCE R A T I O H s l HI

o 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.8 1.0

DROWNED FLOW REDUCTION FACTOR f

Figure 4. I 1 Drowned flow reduction factor as a function of H,/H,

142
4.3.4 Limits of application

The limits of application of the triangular broad-crested weir and truncated weir for
reasonable accuracy are:
- The practical lower limit of h, is related to the magnitude of the influence of fluid
properties, boundary roughness, and the accuracy with which h, can be determined.
The recommended lower limit is 0.06 m or 0.07 times L, whichever is greater;
- The weir notch angle 8 should not be less than 30";
- The recommended upper limit of the ratio H,/p, = 3.0, while p, should not be
less than O. 15 m.
- The limitation on H,/L arises from the necessity of ensuring a sensible hydrostatic
pressure distribution at the control section. Values of the ratio H,/L should therefore
not exceed 0.50 (0.70 if sufficient head is available);
- The breadth B, of a truncated triangular broad-crested weir should not be less than
L/5.

4.4 Broad-crested rectangular profile weir

4.4.1 Description

From a constructional point of view the broad-crested rectangular profile weir is a

rather simple measuring device. The weir block shown in Figure 4.12 has a truly flat
and horizontal crest. Both the upstream and downstream weir faces should be smooth
vertical planes. The weir block should be placed in a rectangular approach channel
perpendicular to the direction of flow. Special care should be taken that the crest

Figure 4.12 Broad-crested rectangular profile weir (after BSI 1969)

143
surface makes a straight and sharp 90-degree intersection with the upstream weir face.
The upstream head over the weir crest should be measured in a rectangular approach
channel as shown in Figure 4.12. The head measurement station should be located
at a distance of between two and three times Hlmaxupstream from the weir face.

Depending on the value of the ratio HJL, four different flow regimes over the weir
may be distinguished:
a) H,/L < 0.08
The depth of flow over the weir crest is such that sub-critical flow occurs above
the crest. The control section is situated near the downstream edge of the weir
crest and the discharge coefficient is determined by the resistance characteristics
of the crest surface. Over this range the weir cannot be used as a measuring device.
b) 0.08 < HJL < 0.33
At these values of HJL a region of parallel flow will occur somewhere midway
above the crest. The water surface slopes downward at the beginning of the crest
and again near the end of the crest. From a hydraulic point of view the weir may
be described as broad-crested over this range of HJL only. The control section
is located at the end of the section where parallel flow occurs. Provided that the
approach velocity has no significant influence on the shape of the separation bubble
(see Figure 4.13) the discharge coefficient has a constant value over this HJL-range.
c) 0.33 < H,/L < about 1.5 to 1.8
In this range of H,/L values the two downward slopes of the water surface will
merge and parallel flow will not occur above the crest. Streamline curvature at
the control has a significant positive effect on the discharge, resulting in higher
C,-values. In fact the weir cannot be termed broad-crested over this range but
should be classified as short-crested. The control section lies at station A above
the separation bubble shown in Figure 4.13.
d) H,/L > about 1.5
Here the ratio HI/L has such a high value that the nappe may separate completely

OH1
Re=-
-
1000 20 -
10000 2
50000 0.4
100000 0.2

b :0.109 Hl

Figure 4.13 Assumed structure of entry-edge separation bubble as a function of H I and the Reynolds
number (Hall 1962)

144
 from the crest and the weir in fact acts as a sharp-crested weir. If H l / L becomes
larger than about 1.5 the flow pattern becomes unstable and is very sensitive to
the sharpness of the upstream weir edge. For H , / L values greater than 3.0 the
flow pattern becomes stable again and similar to that over a sharp-crested measur-
ing weir (see Chapter 5).

To prevent underpressures beneath the overflowing nappe from influencing the head-
discharge relationship, the air pocket beneath the nappe should be fully aerated when-
ever H,/L exceeds 0.33. Dimensions of the aeration duct should be determined as
shown in Section 1.14.
The modular limit, or that submergence ratio H,/H, which produces a 1 YOreduction
from the equivalent modular discharge, depends on the ratio H,/L. If 0.08 Q HJL
Q 0.33, the modular limit may be expected to be 0.66. If H,/L = 1.5, however, the
modular limit is about 0.38 and over the range 0.33 < H,/L < 1.5 the modular limit
may be obtained by linear interpolation between the given values. Provided that the
ratio h,/(h, + p,) d 0.35, Figure 4.18, too, can be used to obtain information on
the reduction of modular flow due to submergence.

4.4.2 Evaluation of discharge

The basic head-discharge equation derived in Section 1.9.1 can be used to evaluate
the flow over the weir. This equation reads

(4-8)

where the approach velocity coefficient C, may be read from Figure 1.12 as a function
of the dimensionless ratio CdA*/Al= Cdhl/(hl+ p,). Experimental results have shown
that under normal field conditions the discharge coefficient is a function of the two
+
ratios h,/L and h,/(h, p,). As mentioned in the previous section, the discharge coeffi-
cient remains constant if there is parallel flow at the control section and if the approach
velocity does not influence the shape of the separation pocket. Hence Cd remains fairly
constant if both
0.08 < h,/L Q 0.33 and
+
h,/(h, PI) Q 0.35
The average value of Cd within these limits is 0.848 and is referred to as the basic
discharge coefficient. If one of the limits is not fulfilled the basic Coefficient should
be multiplied by a coefficient correction factor F which is always greater than unity
since both streamline curvature at the control section and. a depression of the separa-
tion bubble have a positive influence on weir flow. Values of F as a function of h,/L
and h,/(h, + pI)can be read from Figure 4.14.

There are not enough experimental data available to give the relation between c d
and the ratios h,/L and h,/(h, + p,) with satisfactory accuracy over the entire
range. If, however, the influence of the approach velocity on Cd is negligible, (i.e. if
h,/(h, + pI) < 0.35), C,-values can be read as a function of h,/L from Figure 4.15.

145
 1.15

1.14

1.13

1.12
LL
1.11

2U 1.10
LL
z 1.0s
o
1.OE
a
g 1.07
U
+
z 106
w
p
h.
1.05
W
o 1.04

1.03

1.02

1.01

1.o0

RATIO h l l L

Figure 4.14 Coefficient correction factor F as a function of h l / L and hl(hl + p i ) (adapted from Singer
1964)

1.06

1 .O4

1 .o2

u 1.00
U
I-
= 0.98
wU
0.96
IJ.
w
0.94

i
Q
0.92

5 0.90
e
0.88

0.86

0.84
0.1 0.2 0.3 0.4 05 0.6 0.7 0.8 0.Q 1.0 1.1 1.2 13 1.4 1.5

RATIO h q l L
_.
Figure 4.15 Cd-values and F-values as a function of hl/L, provided that hl!(hl + pI) < 0.35
146
The error in the discharge coefficient (including C,) of a rectangular profile weir, con-
structed with reasonable care and skill, may be obtained from the equation
X, = f (10F- 8) per cent (4-9)
To obtain this accuracy the structure should be properly maintained. The method by
which this error should be combined with other sources of error is shown in Annex2.

4.4.3 Limits of application

The limits of application of the rectangular profile weir essential for reasonable accura-
cy are:
- The practical lower limit of h, is related to the magnitude of the influence of fluid
properties, to boundary roughness, and to the accuracy with which h, can be deter-
mined. The recommended lower limit of h, is 0.06 m or 0.08 times L, whichever
is greater.
- The recommended upper limit of the ratio h,/(h, + p,) is 0.60, while p, should
not be less than O. I 5 m.
- The ratio h,/L should not be less than 0.08 and should not exceed 1.50. If, however,

the influence of the approach velocity on Cd is significant (i.e. if h,/(h, + p,) >
0.35), Cd-valuesare only available provided that the ratio h,/L d 0.85;
- The breadth b, of the weir should not be less than 0.30 m nor less than hlmax, nor
less than L/5;
- The air pocket beneath the nappe should be fully aerated whenever the ratio h,/L

exceeds 0.33.

4.5 Faiyum weir

4.5.1 Description

The Faiyum weir is essentially a rectangular profile weir with a crest shape identical
to that described in Section 4.4. The only significant difference is that with the latter
weir two-dimensional weir flow was assured by placing the weir block in a rectangular
approach channel. In contrast, the Faiyum weir consists of a rectangular control sec-
tion placed in a wall across an open channel of arbitrary cross-section (Figure 4.16).
The weir originates from the Faiyum Province in Egypt and a detailed description
of it was given in 1923 by Butcher.

Special care should be taken that the crest surface makes a sharp 90-degree intersection
with the upstream weir face. The crest may either be made of carefully aligned and
joined pre-cast granite concrete blocks with rubbed-in finish or it may have a metal
profile as upstream edge.
Although one is free to install the Faiyum weir across an approach channel of arbi-
trary cross-section, care should be taken that the approach velocity is sufficiently low
so that it does not influence the contraction at the upstream edge of the weir crest.
For this to occur, the area ratio b,h,/A, should not exceed 0.35for all values of h,.
A, denotes the cross-sectional area of the approach channel at the head measurement

147
 I I

DETAIL UPSTREAM EDGE

WEIR CREST (example)

Figure 4.16 Upstream view of Faiyum weir

station. This head measurement station should be located a distance of between two
and three times h,,,, upstream from the weir face.
The upstream corners of the vertical and parallel side walls are known to have a
significant influence on both contraction of the weir flow and the boundary layer dis-
placement thickness of the side walls.
Both effects make it impossible to apply the basic two-dimensional head-discharge
equation to the full width of the control section unless the upstream corners of the
side walls aredimensioned in such a way that the combined effects of lateral contrac-
tion and side-wall boundary layers are counterbalanced.
One way of ensuring that the weir discharge is proportional to the breadth b, of
the control section is to make the radius R of the upstream corners dependent on
the weir breadth b, and the crest length L. As a result of his experimental research
work on the Faiyum weir, Butcher produced a diagram giving the radius R as a func-
tion of the weir breadth b, for the most common crest length (L = 0.50 m) of the
weir. Figure 4.17 shows a dimensionless rendering of Butchers diagram. The two dot-
ted curves in the figure show the eventual limits of variation of the radius R, corres-
ponding to a maximum difference of & 1YOin the two-dimensional weir discharge.

To prevent underpressure beneath the overflowing nappe influencing the head-

discharge relationship, the air pocket beneath the nappe should be fully aerated.

4.5.2 Modular limit

The modular limit, or the submergence ratio H,/Hl that produces a 1% reduction
in the equivalent modular discharge of the Faiyum weir, is a function of the ratio
HJL. If 0.08 < H,/L < 0.33 the weir acts as a broad-crested weir and the modular
limit may be expected to be 0.66. If streamline curvature occurs at the control section,
however, the weir becomes more sensitive to submergence and consequently has a

148
 ratio bJL

Figure 4.17 Radius of upstream corner of side wall as a function of b, and L (adapted from Butcher 1923)

lower modular limit, which may be obtained from Figure 4.18.

We can read from Figure 4.18 that if, for example, HJL = 1.0 the modular limit
equals 0.40 and that if, for this H,/L value, the submergence ratio increases to HJH,
= 0.71, the modular discharge is reduced by 10%. Figure 4.18 can also be used for
the rectangular profile weirs described in Section 4.4, provided that the area ratio
b,h,/A, does not exceed 0.35.

Photo 3 A cluster of Faiyum weirs

149
 REDUCTION OF MODULAR DISCHARGE

RATIO H 1 I L 1% 5% 10
1.8

1.6

1.4

1.2

1.o

0.8

0.6

0.4

O 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6

RATIO H z I L

Figure 4.18 Diagram showing both reduction of modular discharge and variation of H,/L due to submer-
gence (adapted from Butcher 1923)

4.5.3 Evaluation of discharge

The basic head-discharge equation derived in Section 1.9.1 for modular flow through
a rectangular control section can be used to evaluate the weir flow. This equation
reads
2 2
Q = Cd c, T&g b, h,"'' (4- 1O)

where values of C d are similar to those shown in Figure 4.15 and where the approach
velocity coefficient C, can be obtained from Figure 1.12 as a function of the ratio
CdA*/A, = Cdb,h,/Al. The reader will note that due to the restriction on the area
ratio b,h,/A,, C, has a maximum value of 1.035.
The accuracy of the discharge coefficient of the Faiyum weir is unknown. A well

150
maintained structure, however, constructed with reasonable care and accuracy has
an acceptable accuracy for field conditions. The percentage error in the product CdCV
is expected to be less than 5% over the entire range of h,/L.
The method by which this percentage error should be combined with other sources
of error is shown in Annex 2.

4.5.4 Limits of application

The limits of application of the Faiyum weir for reasonable accuracy are:
a. The upstream corners of the parallel and vertical side walls should be selected in
accordance with Figure 4.17;
b. The practical lower limit of h, is related to the magnitude of the influence of fluid
properties, to boundary roughness, and to the accuracy with which h, can be deter-
mined. The recommended lower limit is 0.06 m;
c. The area ratio b,h,/A, should not exceed 0.35;
d. The breadth of the control section should not be less than 0.05 m;
e. The ratio h,/L should not be less than 0.08 and should not exceed 1.6;
f. The airpocket beneath the nappe should be fully aerated whenever hl/L exceeds
0.33.

4.6 Selected list of references

Agricultural University Wageningen. Various laboratory reports on the V-notch broad-crested weir.
Hydraulic Laboratory, Wageningen.
Bos, M.G. 1975 The Romijn movable measuring/regulatingweir. In Irrigation and Drainage Paper, No.26/
2, FAO, Rome. pp. 203-217.
Bos, M.G. 1985. Long-throated flumes and broad-crested weirs, Nijhoff, Dordrecht, The Netherlands.
Bos, M.G., J.A. Replogle and A.J. Clemens 1984. Flow measuring flumes for open channel systems.
321 pp. John Wiley, New York.
Bos, R.J. 1964. De lange meetoverlaat met V-vormige kruin. Polytechnisch Tijdschrift No.7B. No.8B,
pp. 301-310. pp. 254-259.
British Standards Institution 1969. Methods of measurement of liquid flow in open channels. British Stan-
dard 3680, Part 4B, Long base weirs. London.
Butcher, A.D. 1923.Submerged weirs and standing wave weirs. Government Press, Cairo. 17 pp.
Delft Hydraulics Laboratory 1953. Invloed van stromingsvorm van een Romijn meetoverlaat op de afvoer-
coefficient. Report Hydraulics Laboratory, Delft.
Hall, G.W. 1967. Analytic determination of the discharge characteristics of broad-crested weirs using
boundary layer theory. Proceedings of the Inst. of Civil Engineers. Vol. 22, Paper 6607. pp. 177-190.
Harrison, A.J.M. 1967. The streamlined broad-crested weir. Proceedings of the Inst. of Civil Engineers.
Vol. 38, pp. 657-678.
Horton, R.E. 1907. Weir experiments, Coeflicients and Formulas. Water Supply Paper 200, Geological
Survey, U.S.Department of the Interior, Washington D.C.
ISO, TC 113-WG 2, The Netherlands. 1967. A reference document on the round-nose broad-crested weir.
Wageningen, Agric. Univ. Hydraul. Laboratory. unpubl.
Keutner, C. 1934. Strmungsvorgnge an breitkronigen Wehrkrpern Einlaufbauwerken. Der Bauinge-
nieur, Vol. 15. p. 366.
Leliavsky, S. 1965.Irrigation Engineering: Siphons, Weirs and Locks. Design textbook in Civil Engineering,
Vol.11, Chapman and Hall Ltd., London.
Pitlo, R.H., and M. Smit 1970. Discussion to reference of Smith & Liang. Proc. of the Am. Soc. of Civil
Engineers, Irrigation and Drainage Division, R3, pp. 364-369.

151
Romijn, D.G. 1932. Een regelbare meetoverlaat als tertiaire aftapsluis. De Waterstaatsingenieur, Bandung.
Nr. 9.
Romijn, D.G. 1938. Meetsluizen ten behoeve van irrigatiewerken. Handleiding door De Vereniging van
Waterstaats Ingenieurs in Nederlandsch Indi. pp.58.
Singer, J. 1964. Square-edged broad-crested weir as a flow measuring device. Water and Water Engineering.
Vol. 68, Nr. 820, pp. 229-235.
Smith, C.D., and W.S. Liang. 1969. Triangular broad-crested weir. Proc. ofthe Am. Soc. of Civil Engineers,
Irrigation and Drainage Division, IR 4. Paper 6954. pp. 493-502 and closure in IR 4. 1971, pp. 637-640.
Vlugter, H. 1940. De regelbare meetoverlaat. De Waterstaatsingenieur, Bandung. Nr. 10.
Woodbum, J.G. 1930. Tests of broad-crested weirs. Proc. of the Am. Soc. of Civil Engineers, Vol. 56.
No. 7, pp. 1583-1612.Also: Trans. ASCE, Vol. 96,1932, Paper 1797, pp. 387-408.

152
5 Sharp-crested weirs

Classified under the term sharp-crested or thin-plate weirs are those overflow struc-
tures whose length of crest in the direction of flow is equal to or less than two milli-
metres. The weir plate should be smooth and plane, especially on the upstream face,
while the crest surface and the sides of the notch should have plane surfaces which
make sharp 90-degree intersections with the upstream weir face. The downstream edge
of the notch should be bevelled if the weir plate is thicker than two millimetres. The
bevelled surfaces should make an angle of not less than 45-degrees with the surface
of a rectangular notch and an angle of not less than 60 degrees if the throat section
is non-rectangular (see Figure 5. I).

-
FLOW
FLOW

min

rectangular notch, trapezoidal

and circular weirs V-notch and sutm weirs

Figure 5.1 Flow-wise cross-section over a sharp-crested (thin-plate) weir

In general sharp-crested weirs will be used where highly accurate discharge measure-
ments are required, for example in hydraulic laboratories and industry. To obtain
this high accuracy, provision should be made for ventilating the nappe to ensure that
the pressure on the sides and surfaces of the nappe is atmospheric (see Section 1.14).
The downstream water level should be low enough to ensure that it does not interfere
with the ventilation of the air pocket beneath the nappe. Consequently, the required
loss of head for modular flow will always exceed the upstream head over the weir
crest (h,) by about 0.05 m, which is one of the major disadvantages of a sharp-crested
weir.

5.1 Rectangular sharp-crestedweirs

5.1.1 Description

A rectangular notch, symmetrically located in a vertical thin (metal) plate which is

153
placed perpendicular to the sides and bottom of a straight channel, is defined as a
rectangular sharp-crested weir. Rectangular sharp-crested weirs comprise the follow-
ing three types:
a. Fully contracted weirs, i.e. a weir which has an approach channel whose bed and
walls are sufficiently remote from the weir crest and sides for the channel boundaries
to have no significant influence on the contraction of the nappe.
b. Full width weirs, i.e. a weir which extends across the full width of the rectangular
approach channel (B,/b, = 1.0). In literature this weir is frequently referred to
as a rectangular suppressed weir or Rehbock weir.
c. Partially contracted weir, i.e. a weir the contractions of which are not fully deve-
loped due to the proximity of the walls and/or the bottom of the approach channel.
In general, all three types of rectangular weirs should be located in a rectangular ap-
proach channel (See Figure 5.2 and 5.3). If, however, the approach channel is suffi-
ciently large {B,(h, + p,) 2 lOb,h,} to render the velocity of approach negligible,
and the weir is fully contracted, the shape of the approach channel is unimportant.
Consequently, the fully contracted weir may be used with non-rectangular approach
channels. The sides of the rectangular channel above the level of the crest of a full-width
weir should extend at least 0.3 hlmax downstream of the weir crest.

The fully contracted weir may be described by the limitations on BI-b,, b,/B,, h,/p,,
h,/b,, h,, b,, andp, asshowninTable5.1.

Table 5.1 Limitations of a rectangular sharp-crested fully contracted weir

BI-b, > 4h,

h,/p, < 0.5
h,/b, < 0.5
0.07 m < h, < 0.60 m
b, > 0.30111
p, > 0.30m

A comparison of these limitations with those given in Section 5. I .3 shows that the
fully contracted weir has limitations that are both more numerous and more stringent
than the partially contracted weir and full width weir.

5.1.2 Evaluation of discharge

As mentioned in Section 1.13.1, the basic head-discharge equation for a rectangular

sharp-crested weir is
2
Q = C C 3 J 2 gb, hl. (5-1)

To apply this equation to fully contracted, full-width, and partially contracted thin-
plate weirs, it is modified as proposed by Kindsvater and Carter (1957),

154
 2
Q = Ce3 f i be he'.5 (5-2)

where the effective breadth (be) equals b, + Kb and the effective head (he) equals h,

Figure 5.2 The rectangular sharp-crested weir (thin-plate weir)

1to2

Figure 5.3 Enlarged view ofcrest and side of rectangular sharp-crested weir

155
Photo 1 Rectangular sharp-crested weir

+ Kh.The quantities K, and Khrepresent the combined effects of the several pheno-
mena attributed to viscosity and surface tension. Empirically defined values for Kh
as a function of the ratio bc/Bl are given in Figure 5.4 and a constant positive value
for Kh = 0.001 m is recommended for all values of the ratios bc/Bland h,/p,.
Ce is an effective discharge coefficient which is a function of the ratios bc/Bl and
hl/p, and can be determined from Figure 5.5 or from Table 5.2.

Table 5.2 Values for Ce as a function of the ratios bc/Bl and hl/pl (from Georgia Institute of Technology)

bC/h Ce bc/B 1 Ce

1.o 0.602 + 0.075 hI/pI 0.5 +

0.592 0.01 1 hI/pl
0.9 +
0.599 0.064 hI/pl 0.4 0.591 + 0.0058 hl/pl
0.8 0.597 + 0.045 hl/pl 0.3 +
0.590 0.0020 h,/pI
O. 7 +
0.595 0.030 h,/pl 0.2 0.589 - 0.0018 h,/pl
0.6 0.593 + 0.018 h,/pI o. 1 0.588 - 0.0021 hi/p1
O 0.587 - 0.0023 hi/p1

For a rectangular sharp-crested weir which has been constructed with reasonable care
and skill, the error in the effective discharge coefficient (Ce)in the modified Kindsvater
and Carter equation is expected to be less than 1%. The tolerance on both K, and

156
K, is expected to be of the order of k 0.0003 m. The method by which these errors
are to be combined with other sources of error is shown in Annex 2.

5.1.3 Limits of application

a. The practical lower limit of h, is related to the magnitude of the influence

value of K, in metres
0.005

0.003

0.001

-0.001
O 0.2 0.4 0.6 0.8 1 .o 1.2
ratio b,/B1

Figure 5.4 Values of K, as a function of b,/B, (derived from tests a t the Georgia Institute of Techno-
logy by Kindsvater and Carter 1957)

value of Ca

O 0.4 0.8 1.2 1.6 2.0 2.4

value of hl/pl

Figure 5.5 Ceas a function of the ratios bJB, and h,/p, (after Georgia Institute of Technology)

157
 of fluid properties and the accuracy with which h, can be determined. The recom-
mended lower limit is 0.03 m;
b. Bss (1929) has shown that critical depth will occur in the approach channel up-
stream from a weir if the ratio h,/p, exceeds about 5. Thus, for values of h,/p,
greater than 5 the weir is not a control section as specified in Section 1.13. Further
limitations on hl/pl arise from observational difficulties and measurement errors.
For precise discharge measurements the recommended upper limit for h,/p, equals
2.0, while p, should be at least 0.10 m;
c. The breadth (bJ of the weir crest should not be less than O . 15 m;
d. To facilitate aeration of the nappe the tailwater level should remain at least 0.05
m below crest level.

5.2 V-notch sharp-crested weirs

5.2.1 Description

A V-shaped notch in a vertical thin plate which is placed perpendicular to the sides
and bottom of a straight channel is defined as a V-notch sharp-crested weir.
The line which bisects the angle of the notch should be vertical and at the same
distance from both sides of the channel (see Section 5). The V-notch sharp-crested
weir is one of the most precise discharge measuring devices suitable for a wide range
of flow. In international literature, the V-notch sharp-crested-weir is frequently re-
ferred to as the Thomson weir. The weir is shown diagrammatically in Figures 5.6
and 5.1.

The following flow regimes are encountered with V-notch sharp-crested or thin-plate
weirs:
a. Partially contracted weir, i.e. a weir the contractions of which along the sides
of the V-notch are not fully developed due to the proximity of the walls and/or
bed of the approach channel.
b. Fully contracted weir, i.e. a weir which has an approach channel whose bed and
sides are sufficiently remote from the edges of the V-notch to allow for a sufficiently
great approach velocity component parallel to the weir face so that the contraction
is fully developed.
These two types of V-notch sharp-crested weirs may be classified by the following
limitations on h,/p,, h,/B,, hl, pI and BI. It should be noted that in this classification
fully contracted flow is a subdivision of partially contracted flow.

Table 5.3 Classification and limits of application of V-notch sharp-crested (thin-plate) weirs (after I S 0
1971, France)

Partially contracted weir Fully contracted weir

hi/pi c 1.2 h,/p, f 0.4

h,/B, f 0.4 h,/B, C 0.2
0.05 m < h l < 0.6m 0.05 m < h, f 0.38 m
PI > 0.1 m p1 > 0.45m
BI > 0.6m BI > 0.90m

158
I

Figure 5.6 V-notch sharp-crested weir

159
From this table it appears that from a hydraulica1 point of view a weir may be fully
contracted at low heads while at increasing h, it becomes partially contracted.
The partially contracted weir should be located in a rectangular approach canal.
Owing to a lack of experimental data relating to the discharge coefficient over a suffi-
ciently wide range of the ratios hl/pl and p,/B,, only the 90-degree V-notch should
be used as a partially contracted V-notch weir. The fully contracted weir may be placed
in a non-rectangular approach channel provided that the cross-sectional area of the
selected approach channel is not less than that of the rectangular channel as prescribed
in Table 5.3.

5.2.2 Evaluation of discharge

As shown in Section I . 13.3, the basic head-discharge equation for a V-notch sharp-
crested weir is

Q =
8
Ce-&
e
tanZ h,2.5 (5-3)
15
To apply this equation to both fully and partially contracted sharp-crested weirs, it
is modified to a form proposed by Kindsvater and Carter (1957)

Q =
8
Cen&
e
tan2 heZ.5 (5-4)

where 8 equals the angle induced between the sides of the notch and heis the effective
head which equals h, + Kh.The quantity Kh represents the combined effects of fluid
properties. Empirically defined values for Kh as a function of the notch angle (O) are
shown in Figure 5.8.

For water at ordinary temperature, i.e. 5C to 30C (or 40F to 85F) the effective
coefficient of discharge (Ce) for a V-notch sharp-crested weir is a function of three
variables

value of Kh
in millimetres

value of notch angle 0 in degrees

Figure 5.8 Value of Kh as a function of the notch angle

160
value of Ce
0.61

0.60

0.59

0.58

0.57

0.56
O 20 40 60 80 1O0 120
value of notch angle 0 in degrees

Figure 5.9 Coefficient ofdischarge Ce as a function of notch angle for fully contracted V-notch weirs

(5-5)

If the ratios h,/p, < 0.4 and h,/B, < 0.2, the V-notch weir is fully contracted and
Cebecomes a function of only the notch angle 0, as illustrated in Figure 5.9.
If on the other hand the contraction of the nappe is not fully developed, the effective
discharge coefficient (Ce)can be read from Figure 5.10 for a 90-degree V-notch only.
Insufficient experimental data are available to produce Ce-values for non-90-degree
partially contracted V-notch weirs.
The coefficients given in Figures 5.9 and 5.10 for a V-notch sharp-crested weir can
be expected to have an accuracy of the order of 1.O% and of 1 .O% to 2.0% respectively,
provided that the notch is constructed and installed with reasonable care and skill
in accordance with the requirements of Sections 5 and 5.2.1. The tolerance on Kh
is expected to be of the order of 0.0003 m. The method by which these errors are
to be combined with other sources of error is shown in Annex 2.

effective discharge
coefficient Ce
/PI Ce

Figure 5.10 Ce as a function of hl/pl and pl/B, for 90-degree V-notch sharp-crested weir. (From British
Standard 3680: Part 4A and ISO/TC 113/GT 2 (France-IO) 1971)

161
Table 5.4 Discharges for V-notch sharp-crested weirs for heads in metres (adapted from ISO/TC 113/GT 2 (France-
IO) 1971)

Head Discharge ]/sec Head Discharge I/sec Head Discharge I/sec Head Discharge I/scc
~

metre 90" '/L900 '/,9O0 metre 90" '/,90 '/,90" metre 90" '/* 90" 90"
~ ~ ~

0.050 0.803 0.406 0.215 0.100 4.420 2.249 1.161 0.150 12.066 6. I30 3.140 0.200 24.719 12.506 6.379
0.051 0.843 0.427 0.225 0.101 4.530 2.305 1.190 0.151 12.267 6.231 3.192 0.201 25.208 12.662 6.458
0.052 0.884 0.448 0.236 o. I02 4.641 2.362 1.219 0.152 12.471 6.334 3.245 0.202 25.339 12.819 6.537
0.053 0.926 0.469 0.247 O. 103 4.754 2.420 1.249 0.153 12.676 6.437 3.297 0.203 25.652 12.977 6.617
0.054 0.970 0.491 0.259 O. I04 4.869 2.478 1.278 0.154 12.883 6.542 3.350 0.204 25.969 13.136 6.698
0.055 1.015 0.514 0.271 0.105 4.985 2.537 1.309 0.155 13.093 6.648 3.404 0.205 26.288 13.296 6.780
0.056 1.061 0.537 0.283 O. I06 5.103 2.598 1.339 0.156 13.304 6.755 3.458 0.206 26.610 13.457 6.862
0.057 1.108 0.561 0.295 0.107 5.222 2.659 1.371 0.157 13.517 6.863 3.513 0.207 26.934 13.620 6.944
0.058 1.156 0.586 0.308 0.108 5.344 2.720 1.402 0.158 13.732 6.971 3.568 0.208 27.261 13.784 7.028
0.059 1.206 0.611 0.321 0.109 5.467 2.783 1.434 0.159 13.950 7.081 3.624 0.209 27.590 13.949 7.111

0.060 1.257 0.637 0.334 0.110 5.592 2.847 1.466 0.160 14.169 7.192 3.680 0.210 27.921 14.115 7.196
0.061 1.309 0.663 0.348 0.111 5.719 2.911 1.499 0.161 14.391 7.304 3.737 0.21 I 28.254 14.282 7.281
0.062 1.362 0.691 0.362 0.112 5:847 2.976 1.533 0.162 14.614 7.417 3.794 0.212 28.588 14.450 7.366
0.063 1.417 0.7 I8 0.376 0.113 5.977 3.042 1.566 0.163 14.840 7.531 3.852 0.213 28.924 14.620 7.453
0.064 1.473 0.747 0.391 0.114 6.108 3.109 1.601 0.164 15.067 7.646 3.911 0.214 29.264 14.264 7.539
0.065 1.530 0.776 0.406 0.115 6.242 3.177 '1.635 0.165 15.297 7.762 3.969 0.215 29.607 14.964 7.627
0.066 1.588 0.806 0.421 0.116 6.377 3.246 1.670 0.166 15.529 7.879 4.029 0.216 29.953 15.138 7.715
0.067 1.648 0.836 0.437 0.117 6.514 3.315 1.706 0.167 15.763 7.998 4.089 0.217 30.301 15.313 7.803
0.068 1.710 0.867 0.453 0.1 18 6.653 3.386 1.742 0.168 15.999 8.117 4.149 0.218 30.651 15.489 7.893
0.069 1.772 0.899 0.470 0.119 6.793 3.457 1.778 0.169 16.237 8.237 4.210 0.219 31.004 15.666 7.982

0.070 1.836 0.932 0.486 0.120 6.935 3.529 1.815 0.170 16.477 8.358 4.272 0.220 31.359 15.844 8.073
0.071 1.901 0.965 0.503 0.121 7.079 3.602 1.853 0.171 16.719 8.481 4.334 0.221 21.717 16.024 8.164
0.072 1.967 0.999 0.521 0.122 7.224 3.667 1.891 0.172 16.964 8.604 4.397 0.222 32.077 16.204 8.255
0.073 2.035 1.033 0.539 0.123 7.372 3.751 1.929 0.173 17.210 8.728 4.460 0.223 32.439 16.386 8.347
0.074 2.105 1.069 0.557 0.124 7.522 3.827 1.968 0.174 17.459 8.854 4.524 0.224 32.803 16.570 8.441
0.075 2.176 1.105 0.575 0.125 7.673 3.904 2.007 0.175 17.709 8.980 4.588 0.225 33.168 16.754 8.535
0.076 2.248 1.141 0.594 0.126 7.827 3.982 2.046 0.176 17.963 9.108 4.653 0.226 33.535 16.940 8.629
0.077 2.322 1.179 0.613 0.127 7.982 4.060 2.086 0.177 18.219 9.237 4.718 0.227 33.907 17.727 8.724
0.078 2.397 1.217 0.633 0.128 8.139 4.140 2.127 0.178 18.478 9.367 4.784 0.228 34.282 17.315 8.819
0.079 2.473 1.256 0.653 0.129 8.298 4.220 2.168 0.179 18.378 9.497 4.851 0.229 34.659 17.504 8.915

0.080 2.551 1.296 0.673 0.130 8.458 4.302 2.209 0.180 19.001 9.629 4.918 0.230 35.039 17.695 9.01 I
0.081 2.630 1.336 0.694 0.131 8.621 4.384 2.251 0.181 19.265 9.762 4.986 0.231 35.421 17.886 9.108
0.082 2.710 1.377 0.715 0.132 8.785 4.467 2.294 0.182 19.531 9.896 5.054 0.232 35.806 18.079 9.207
0.083 2.792 1.419 0.737 0.133 8.951 4.551 2.337 0.183 19.800 10.032 5.122 0.233 36.139 18.274 9.306
0.084 2.876 1.462 0.759 0.134 9.119 4.636 2.380 0.184 20.071 10.168 5.192 0.234 36.582 18.469 9.405
0.085 2.961 1.505 0.781 0.135 9.289 4.722 2.424 0.185 20.345 10.305 5.261 0.235 36.974 18.666 9.504
0.086 3.048 1.549 0.803 0.136 9.461 4.809 2.468 0.186 20.621 10.444 5.332 0.236 31.369 18.864 9.605
0.087 3.136 L594 0.826 0.137 9.634 4.897 2.513 0.187 20.899 10.584 5.503 0.237 37.766 19.063 9.706
0.088 3.225 1.640 0.850 0.138 9.810 4.986 2.559 0.188 21.180 10.726 5.475 0.238 38.166 19.263 9.808
0.089 3.316 I .686 0.874 0.139 9.987 5.075 2.604 0.189 21.463 10.867 5.547 0.239 38.568 19.465 9.9 I O

0.090 3.409 1.734 0.898 0.140 10.167 5.166 2.651 0.190 21.748 11.010 5.620 0.240 38.973 19.668 10.013
0.091 3.503 1.782 0.922 0.141 10.348 5.258 2.697 0.191 22.034 11.155 5.693 0.241 39.380 19.872 10.116
0.092 3.598 1.830 0.947 0.142 10.532 5.351 2.744 0.192 22.322 11.300 5.766 0.242 39.790 20.079 10.220
0.093 3.696 I.880 0.973 0.143 10.717 5.444 2.792 0.193 22.612 11.447 5.481 0.243 40.202 20.287 10.325
0.094 3.795 1.930 0.998 0.144 10.904 5.539 2.840 0.194 22.906 11595 5.916 0.244 40.617 20.496 10.430
0.095 3.895 1.981 1.025 0.145 11.093 5.635 2.889 0.195 23.203 11.743 5.992 --.0.245 41.034 20.705 10.536
0.096 3.997 2.033 1.051 0.146 11.284 5.732 2.938 0.196 23.501 11.893 6.068 0.246 41.454 20.916 10.642
0.097 4.101 2.086 1 .O78 0.147 11.476 5.830 2.988 0.197 23.802 12.044 6.145 0.247 41.877 21.127 10.750
0.098 4.206 2.139 1.106 0.148 11.671 5.929 3.038 0.198 24.106 12.197 6.222 0.248 42.302 21.340 10.858
0.099 4.312 2.194 1.133 0.149 11.867 6.029 3.089 0.199 24.411 12.351 6.300 0.249 42.730 21.555 10.967

162
Head Discharge I/sec Head Discharge I/sec Head Discharge I/sec

metre 90" '/,90 'I490 metre 90" L/2900 'I490" metre 90" 'I290" '/4 90"

0.250 43.160 21.772 1 I .O77 0.300 68.106 34.268 17.410 0.350 100.19 50.313 25.512
0.251 43.593 21.990 11.187 0.301 68.675 34.552 17.555 0.351 100.91 50.672 25.693
0.252 44.028 22.209 1 1.299 0.302 69.246 34.837 17.700 0.352 101.63 51.033 25.875
0.253 44.466 22.429 11.410 0.303 69.821 35.124 17.845 0.353 102.36 51.397 26.057
0.254 44.907 22.649 11.523 0.304 70.398 35.412 17.992 0.354 103.08 51.758 26.240
0.255 45.350 22.873 I I .635 0.305 70.980 35.702 18.139 0.355 103.81 52.121 26.424
0.256 45.796 23.098 1 1.749 0.306 71.568 35.995 18.287 0.356 104.54 52.487 26.609
0.257 46.245 23.323 1 I .863 0.307 72.159 36.290 18.435 0.357 105.28 52.856 26.794
0.258 46.696 23.549 I I .978 0.308 72.750 36.585 18.585 0.358 106.02 53.227 26.981
0.259 47.150 23.777 12.094 0.309 73.341 36.880 18.735 0.359 106.77 53.596 27.168

0.260 47.606 24.005 12.210 0.310 73.936 37.177 18.885 0.360 107.52 53.967 27.355
0.261 48.065 24.235 12.326 0.311 74.534 37.477 19.037 0.361 108.27 54.340 27.544
0.262 48.527 24.466 12.443 0.312 75.135 37.779 19.189 0.362 109.02 54.717 27.733
0.263 48.991 24.699 12.561 0.313 75.738 38.081 19.342 0.363 109.78 55.096 27.923
0.264 49.458 24.933 12.680 0.314 76.344 38.384 19.495 0.364 110.54 55.473 28.114
0.265 49.928 25.168 12.799 0.315 76.954 38.687 19.650 0.365 111.30 55.582 28.306
0.266 50.400 25.404 12.920 0.316 77.566 38.995 19.805 0.366 112.06 56.231 28.498
0.267 50.876 25.642 13.041 0.317 78.181 39.304 19.960 0.367 112.84 56.616 28.691
0.268 51.353 25.881 13.162 0.318 78.802 39.615 20.117 0.368 113.62 57.003 28.885
0.269 51.834 26.121 13.284 0.319 79.428 39.927 20.274 I 0.369 114.39 57.391 29.080

0.270 52.317 26.363, 13.407 0.320 80.057 40.241 20.432 0.370 115.17 57.780 29.275
0.271 52.802 26.606 13.529 0.321 80.685 40.553 20.590 0.371 115.95 58.171 29.412
0.272 53.291 26.851 13.653 0.322 81.314 40.867 20.750 0.372 116.73 58.560 29.669
0.273 53.782 27.098 13.778 0.323 81.947 41.184 20.910 0.373 117.52 58.950 29.867
0.274 54.276 27.347 13.903 0.324 82.583 41.503 21.071 0.374 118.31 59.345 30.065
0.275 54.772 27.596 14.030 0.325 83.222 41.824 21.232 0.375 119.11 59.742 30.264
0.276 55.272 27.845 14.157 0.326 83.863 42.147 21.395 0.376 119.91 60.141 30.465
0.277 55.774 28.097 14.284 0.327 84.508 42.471 21.558 0.377 120.71 60.542 30.666
0.278 56.282 28.351 14.413 0.328 85.155 42.796 21.721 0.378 121.52 60.944 30.867
0.279 56.794 28.607 14.542 0.329 85.806 43.123 21.886 0.379 122.32 61.346 31 .O70

0.280 57.306 28.863 14.671 0.330 86.459 43.451 22.051 0.380 123.13 61.747 31.723
0.281 57.819 29.119 14.802 0.331 87.116 43.779 22.217 0.381 123.94 62.150 31.477
0.282 58.335 29.377 14.933 0.332 87.775 44.107 22.384
0.283
0.284
58.853
59.375
29.638
29.901
15.065
15.197
0.333
0.334
88.438
89.103
44.438
44.773
22.551
22.719
r 2

0.285
0.286
0.287
0.288
59.899
60.425
60.955
61.487
30.163
30.427
30.691
30.959
15.330
15.464
15.598
15.734
0.335
0.336
0.337
0.338
89.772
90.448
91.128
91.811
45.108
45.446
45.785
46.125
22.888
23.058
23.228
23.400
1, 90 degree v-mtch

0.289 62.023 31.229 15.870 0.339 92.491 46.467 23.572

0.290 62.560 31.499 16.006 0.340 93.175 46.810 23.744 I

0.291
0.292
0.293
63.101
63.645
64.195
31.769
32.040
32.315
16.143
16.281
16.420
0.341 93.862
0.342 94.551
0.343 95.244
47.153
47.497
47.842
23.910
24.092
24.627
TQl-7 # 93 degree Y-mtch

0.294 64.748 32.591 16.559 0.344.95.940 48.191 24.442

0.295 65.303 32.869 16.699 0.345 96.638 48.542 24.619
0.296 65.858 33.146 16.840 0.346 97.340 48.895 24.796
0.297 66.416 33.424 16.982 0.347 98.045 49.249 24.974
0.298
0.299
66.976
67.539
33.704
33.985
17.124
17.267
0.348 98.753
0.349 99.471
49.604
49.958
25.152
25.332
a 90 degree V-mtch

Note: The number of significant figures given for the discharge does not imply a corresponding accuracy
in the knowledge of the value given.

163
5.2.3 Limits of application

The limits of application of the Kindsvater and Carter equation for V-notch sharp-
crested weirs are:
a. The ratio h,/p, should be equal to or less than 1.2;
b. The ratio h,/B, should be equal to or less than 0.4;
c. The head over the vertex of the notch h, should not be less than 0.05 m nor more
than 0.60 m;
d. The height of the vertex of the notch above the bed of the approach channel (p,)
should not be less than O. 10 m;
e. The width of the rectangular approach channel should exceed 0.60m;
f. The notch angle of a fully contracted weir may range between 25 and 100 degrees.
Partially contracted weirs have a 90-degree notch only;
g. The tailwater level should remain below the vertex of the notch.

5.2.4 Rating tables

Commonly used sizes of V-notches for fully contracted thin-plate weirs are the 90-deg-
ree, 90-degree and 90-degree notches in which the dimensions across the top
are twice, equal to and half the vertical depth respectively. The related ratings are
given in Table 5.4.

5.3 Cipoletti weir

5.3.1 Description

A Cipoletti weir is a modification of a fully contracted rectangular sharp-crested weir

and has a trapezodal control section, the crest being horizontal and the sides sloping
outward with an inclination of 1 horizontal to 4 vertical (Figure 5.1 I). Cipoletti (1886)
assumed that, due to the increase of side-contraction with an increasing head, the
decrease of discharge over a fully contracted rectangular sharp-crested weir with
breadth b, would be compensated by the increase of discharge due to the inclination
of the sides of the control-section. This compensation thus allows the head-discharge
equation of a full width rectangular weir to be used. It should be noted, however,

2 to 3 h, max
n

---approach channel
upstream view

Figure S.1 I Definition sketch of a Cipoletti weir

164
Photo 2 Cipoletti weir

that experiments differ as to the degree to which this compensation occurs. Inherently,
the accuracy of measurements obtained with a Cipoletti weir is significantly less than
that obtainable with the rectangular or V-notch sharp-crested weirs described in Sec-
tion 5.1 and 5.2 respectively.

5.3.2 Evaluation of discharge

The basic head-discharge equation for the Cipoletti weir is the same as that of a rectan-
gular fully contracted weir. Hence
2
Q = CdCV3 J2g b, hl'.' (5-6)

where, within certain limits of application, the discharge coefficient Cd equals 0.63.
The approach velocity coefficient C, may be obtained from Figure 1.1 1. A rating table

165
for the discharge q in m3/s per metre width, with negligible approach velocity, is pre-
sented in Table 5.5.
The accuracy of the discharge coefficient for a well maintained Cipoletti weir is
reasonable for field conditions. The eiror in the product CdCVis expected to be less
than 5%. The method by which this coefficient error is to be combined with other
sources of error is shown in Annex 2.

5.3.3 Limits of application

The limits of application of the (fully contracted) Cipoletti weir are:

a. The height of the weir crest above the bottom of the approach channel should
be at least twice the head over the crest with a minimum of 0.30 m;
b. The distance from the sides of the trapezodal control section to the sides of the
approach channel should be at least twice the head over the crest with a minimum
of 0.30 m;
c. The upstream head over the weir crest h, should not be less than 0.06 m nor more
than 0.60 m;
d. The ratio h,/b, should be equal to or less than 0.50.
e. To enable aeration of the nappe, the tailwater level should be at least 0.05 m below
crest level.
Provided the Cipoletti weir conforms to the above limits of application, it may be
placed in a non-rectangular approach channel.
Table 5.5 Discharge of the standard Cipoletti weir in m3/s.m
~ ~____________

Head Discharge Head Discharge Head Discharge

metre m3/s.m metre m3/s.m metre m3/s.m
-
0.06 0.0273 0.26 0.247 0.46 0.580
0.07 0.0344 0.27 0.261 0.47 0.599
0.08 0.0421 0.28 0.275 0.48 0.618
0.09 0.0502 0.29 0.290 0.49 0.638
0.10 0.0588 0.30 0.306 0.50 0.657

0.11 0.0678 0.3 1 0.321 0.51 0.677

0.12 0.0773 0.32 0.337 0.52 0.697
0.13 0.0871 0.33 0.352 0.53 0.717
0.14 0.0974 0.34 0.369 0.54 0.738
0.15 O. 108 0.35 0.385 0.55 0.758

0.16 0.119 0.36 0.402 0.56 0.779

0.17 0.130 0.37 0.418 0.57 0.800
0.18 0.142 0.38 0.435 0.58 0.821
0.19 0.154 0.39 0.453 0.59 0.843
0.20 0.166 0.40 0.470 0.60 0.864

0.21 0.179 0.41 0.488

0.22 0.192 0.42 0.506
0.23 0.205 0.43 0.524
0.24 0.219 0.44 0.543 NOTE: The approach velocity has
0.25 0.232 0.45 0.561 been neglected (C, Y 1 .OO)

166
 5.4 Circular weir
5.4.1 Description

1 A circular control section located in a vertical thin (metal) plate, which is placed per-
~

pendicular to the sides and bottom of a straight approach channel, is defined as a

circular thin plate weir. These weirs have the advantage that the crest can be turned
and bevelled with precision in a lathe, and more particularly that they do not have
to be levelled. Circular sharp-crested weirs, in practice, are fully contracted so that
the bed and sides of the approach channel should be sufficiently remote from the
control section to have no influence on the development of the nappe (Figure 5.12).
The fully contracted weir may be placed in a non-rectangular approach channel pro-
vided that the general installation conditions comply with those laid down in Section
5.4.3.

5.4.2 Determination of discharge

According to Equation 1-93, the basic head-discharge equation for a circular sharp-
crested weir reads
4
Q = C, O -J2g dc2.5= Ce$i dc2. (5-7)
15
4
where o is a function of the filling ratio h,/dc = k2. Values of o and $i = - o
15
are shown in Table 5.6.
For water at ordinary temperatures, the discharge coefficient is a function of the
filling ratio h,/dc.Staus (1 93 1 ) determined experimental values of C, for various weir
diameters. Average values of Ce as a function of h,/d, are shown in Table 5.7.

~
- \
\

Figure 5.12 Circular weir dimensions

167
Table 5.6 Values of o and 4 as a function of the filling ratio hl/d, = k2 of a circular sharp-crested weir

hlldc o @i hi/dc o 4 h,/dc o +i

dimension- dimension- dimension-
less m'/l/s less m%/s less m'hls

0.01 0.0004 0.00047 0.36 ,345I ,4076 0.71 1.1804 1.3943

.o2 .O013 .O0I54 .37 ,3633 ,4291 2 .72 1.2085 1.4275
.O3 .O027 ,003 I9 .38 ,3819 ,451 1 .73 1,2368 1.4609
.O4 ,0046 .O0543 .39 .4009 ,4735 .74 1.2653 1.4946
.O5 .O07 1 ,00839 .40 .4203 .4965 .75 1.2939 1.5284

.O6 ,0102 ,0120 .4 I ,4401 ,5199 .76 1.3226 1.5623

.O7 ,0139 .O I64 .42 ,4603 ,5437 .77 1.3514 1.5963
.O8 .O182 .o2 15 .43 ,4809 .568 I .78 1.3802 1.6303
.O9 .O231 ,0273 .44 ,5019 ,5929 .79 1.409I 1.6644
.IO ,0286 ,0338 .45 ,5233 .6182 .80 1.4380 1.6986

.I1 ,0346 ,0409 .46 ,545 1 ,6439 .8 1 1.4670 1.7328

.I2 ,0412 ,0487 .47 ,5672 ,6700 .82 1.4960 1.7671
.13 ,0483 ,0571 .48 ,5896 .6965 .83 1.5250 1.8013
.I4 ,0560 ,0661 .49 ,6123 ,7233 34 1.5540 1.8356
.I5 ,0642 ,0758 so ,6354 ,7506 35 1.5830 1.8699

.I6 ,0728 ,0860 .51 ,6588 .7782 .86 1.6120 1.9041

.I7 ,0819 ,0967 .52 ,6825 ,8062 37 1.6410 1.9384
.I8 ,0914 ,1080 .53 ,7064 ,8344 .88 1.6699 1.9725
.I9 .IO14 ,1198 .54 ,7306 ,8630 39 1.6988 ' 2.0066
.20 .I119 .I322 .55 .7551 ,8920 .90 1.7276 2.0407

.21 ,1229 ,1452 .56 ,7799 .9212 .91 1.7561 2.0743

.22 ,1344 ,1588 .57 ,8050 ,9509 .92 1.7844 2.1077
.23 .1464 ,1729 .58 ,8304 .9809 .93 1.8125 2.1409
.24 ,1589 ,1877 .59 ,8560 1.0111 .94 1.8403 2.1738
.25 ,1719 ,2030 .60 ,8818 1.0416 .95 1.8678 ' 2.2063

.26 ,1854 ,2190 .61 .9079 1.0724 .96 1.8950 2.2384

.27 ,1994 ,2355 .62 ,9342 1.1035 .97 1.9219 2.2702
.28 .2139 ,2527 .63 ,9608 1.1349 .98 1.9484 2.3015
.29 ,2289 ,2704 .64 ,9876 1.1666 .99 1.9744 2.3322
.30 ,2443 ,2886 .65 1.0147 1.1986 I .o0 2.000 -

,2601 1.0420 4
.31
.32 ,2763
,3072
,3264
.66
.67 1.0694 1.2632
Q = Ce E f i o d2.5
.33 ,2929 ,3460 .68 1.0969 1.2957
or
.34 ,3099 ,3660 .69 1.1246 i.3254
.35 ,3273 ,3866 .70 1.1524 1.3612 Q = Ce +i dc2.5

Values o f o from Stevens 1957

So far as is practicable, circular weirs should be installed and maintained so as to

make the approach velocity negligible (HI 2: hl).
The error in the effective discharge coefficients for a well maintained circular sharp-
crested weir, as presented in Table 5.7, may be expected to be less than 2%. The method
by which this error is to be combined with other sources of error is shown in Annex 2.

168
Table 5.7 Average discharge coefficient for circular sharp-crested weirs

1.o0 0.606 0.65 0.595 0.30 0.600

0.95 0.604 0.60 0.594 0.2s 0.604
0.90 0.602 0.55 0.593 0.20 0.610
0.85 0.600 0.50 0.593 0.15 0.623
0.80 0.599 0.45 0.594 0.10 0.650
0.75 0.597 0.40 0.595 0.05 0.75
0.70 0.596 0.35 0.597 O -

The lower quarter of a circular weir is sometimes described as a parabola of which

the focal distance equals the radius of the circle. According to Equation 1-80, the
head-discharge relationship then reads

Q = C e t J m h,-O (5-8)

where the effective discharge coefficient differs less than 3% from those presented in
Table 5.7, provided that hl/dc < about 0.25.

5.4.3 Limits of application

The limits of application of the circular sharp-crested weir are:

a. The height of the crest above the bed of the approach channel should not be less
than the radius of the control section with a minimum of O. I O m;
b. The sides (boundary of the rectangular, trapezodal, or circular approach channel)
should not be nearer than the radius rc to the weir notch;
c. The ratio HJdCshould be equal to or more than O. 10;
d. The practical lower limit of H , is 0.03 m;
e. To enable aeration of the nappe the tailwater level should be at least 0.05 m below
crest level.
If only the lower half of the circular control section is used, the same limits of applica-
tion should be observed.

5.5 Proportional weir

5.5.1 Description

The proportional or Sutro weir is defined as a weir in which the discharge is linearly
proportional to the head over an arbitrary reference level which, for the Sutro weir,
has been selected at a distance of one-third of the height (a) of the rectangular section
above the weir crest. The Sutro weir consists of a rectangular portion joined to a curved
portion which, according to Equation 1-103, has as a profile law (see Section 1.13.7)

169
 x/bc = 1 --?tan-I
x fi (5-9)

to provide proportionality for all heads above the boundary line CD (Figure 5.13).
This somewhat complex equation of the curved profile may give the impression that
the weir is difficult to construct. In practice, however, it is quite easy to make from
sheet metal and, by using modern profile cutting machines, very fine tolerances can
be obtained. Table 5.8 gives values for z'/a and x/bc from which the coordinates of
the curved portion can be computed when the two controlling dimensions, a and b,,
are known. The values of z'/a and x/bc are related by Equation 5-9.
Several types of the Sutro proportional weirs have been tested, both symmetrical and
unsymmetrical forms being shown in Figure 5.13.
Both types are fully contracted along the sides and along the crest. Ventilation of
the nappe is essential for accurate measurements so that the tailwater level should
be at least 0.05 m below crest level. Of special interest are the so-called crestless weirs

Photo 3 Portable Sutro weir equipped with recorder

170
 c
c

bc weir discharge

symmetrical and unsymmetrical pmportional-flow sutm weir and a

typical head-discharge curve.

i
Figure 5.13 Sutro weir dimensions I
Table 5.8 Values of z/a and x/bc related by Equation 5-9

z/a x/b, z/a x/bc z/a xlbc

o. 1 0.805 I .o 0.500 IO 0.195

0.2 0.732 2.0 0.392 12 O. 179
0.3 0.68 I 3.0 0.333 14 O. 166
0.4 0.641 4.0 0.295 16 O. 156
0.5 0.608 5.0 0.268 18 0.147
0.6 0.580 6.0 0.247 20 0.140
0.7 0.556 7.0 0.230 25 0.126
0.8 0.536 8.0 0.216 30 0.115
0.9 0.517 9.0 0.205
1 .o 0.500 10.0 0.195

in which the symmetrical weir profile has been superimposed directly on the bottom
of the approach channel to prevent the accumulation of sediments upstream of the
weir plate. With all three types, the weir crest should be truly horizontal and perpendi-
cular to the flow. Weirs with a linear head-discharge relationship are particularly suit-
able for use as downstream control on rectangular canals with constant flow velocity,
as controls for float regulated chemical dosing or sampling devices, or as a flow meter
whereby the average discharge over any period is a direct function of the average
recorded head.

5.5.2 Evaluation of discharge

As shown in Section I . 13.7, the basic head-discharge equation for a linearly propor-
tional weir is
Q = c d b c J2ga(h, - a/3) (5-10)
where the discharge coefficient Cd is mainly determined by the geometrical proportions
of the control section, which, according to Equation 5-9, is governed by the values
of a and bc. The values of C, for symmetrical and unsymmetrical weirs are presented
in Tables 5.9 and 5.10 respectively.

171
Table 5.9 Discharge coefficients of symmetrical Sutro weirs as a function of a and b, (after Soucek, Howe
and Mavis 1936)

a b, (metres)
(metres)
0.15 0.23 0.30 0.38 0.46

0.006 0.608 0.613 0.617 0.6185 0.619

0.015 0.606 0.61 1 0.615 0.617 0.6175
0.030 0.603 0.608 0.612 0.6135 0.614
0.046 0.601 0.6055 0.610 0.6115 0.612
0.061 0.599 0.604 0.608 0.6095 0.610
0.076 0.598 0.6025 0.6065 0.608 0.6085
0.091 0.597 0.602 0.606 0.6075 0.608

Table 5.10 Discharge coefficients of unsymmetrical Sutro weirs as a function of a and b, (after Soucek,
Howe and Mavis 1936)

a b, (metres)
(metres)
0.15 0.23 0.30 0.38 0.46

0.006 0.614 0.619 0.623 0.6245 0.625

0.015 0.612 0.617 0.621 0.623 0.6235
0.030 0.609 0.614 0.618 0.6195 0.620
0.046 0.607 0.61 15 0.616 0.6175 0.618
0.061 0.605 0.610 0.614 0.6155 0.616
0.076 0.604 0.6085 0.6125 0.614 0.6145
0.091 0.603 0.608 0.612 0.6135 0.614

The coefficients given in Tables 5.9 and 5.10 can be expected to have an accuracy
of the order of 2%, provided the control is constructed and installed with reasonable
care and skill. To maintain this coefficient accuracy, the weir should be cleaned fre-
quently. The method by which this error is to be combined with other sources of error
is shown in Annex 2.
If contraction is fully suppressed along the weir crest, contraction along the curved
edges of the weir will increase to such an extent that the wetted area of the jet at
the vena contracta remains about constant (see orifices Section I . 12). Experimental
results obtained by Singer and Lewis (1966) showed that the coefficient values in
Tables 5.9 and 5.10 may be used for crestless weirs provided that the weir breadth
b, is not less than O. 15 m.

5.5.3 Limits of application

The weir discharge is linearly proportional to the head provided that the head is greater
than about 1.2a. However, to obtain a sensibly constant discharge coefficient, it is
advised to use h, = 2a as a lower limit. In addition, h, has a practical lower limit
which is related to the magnitude of the influence of fluid properties and the accuracy

172
with which h, can be determined. The recommended lower limit is 0.03 m.
The maximum value of h, is related to the magnitude of the influence of fluid proper-
ties. Further, h, - a = z' is restricted to a value whereby the value of x, as computed
by Equation 5-9, is not less than 0.005 m. For similar reasons, the height of the rectan-
gular portion (a) should not be less than 0.005 m.
The breadth (b,) of the weir crest should not be less than 0.15 m to allow the use
of the standard discharge coefficient.
To achieve a fully contracted weir, the ratio b,/p, should be equal to or greater
than 1.O and the ratio B,/b, not less than 3.0.
Linearly proportional weirs that do not comply with the limits on the breadth of
the crest can be employed satisfactorily provided that such weirs are first calibrated
to obtain the proper coefficient value. Due to lack of experimental data, no standard
C,-values are given for b, < O. 15 m.
To allow sufficient aeration of the nappe, tailwater-level should be at least 0.05
m below crest level.

5.6 ' Selected list of references

Banks, W.H.H., C.R. Burch, and T.L. Shaw 1968. The design of proportional and logarithmic thin-plate
weirs. J. of Hydraulic Research, Delft. Vol. 6. No. 2, pp. 75-106.
Bos, M.G. 1985. Long-throated flumes and broad-crested weirs. Nijhoff Publishers, Dordrecht. p. 141.
Bss, P. 1929. Berechnung der Abflussmengen und der Wasserspiegellage bei Abstiirzen und Schwellen
unter besonderer Bercksichtigung der dabei auftretenden Zusatzspannungen. Wasserkraft u. Wasser-
wirtschaft, Vol. 22. pp. 13-33.
British Standards Institution. 1965. Methods of measurement of liquid flow in open channels. BS 3680,
Part 4A: Thin-plate weirs and venturi flumes. London.
Cipoletti, C. 1886. Modulo per la dispensa delle acque atramazzo libero di forma trapezia e coeffciente
di contrazione constante. Esperimenti e formole per grandi stramazzi a soglia inclinata e orizontale.
Milano, Hocpli. 88 p.
Franke, P.G. 1962. Messiiberfalle. Das Gas- und Wasserflche, 103 Jahrg. Nr. 40, pp. 1072-1075. Nr. 42,
pp. 1137-1140 and Nr. 44, pp. 1178-1181.
Kindsvater, C.E. and R.W.C. Carter. 1957. Discharge characteristics of rectangular thin-plate weirs. Jour-
nal of the Hydraulics Division of the ASCE, Vol. 83, No. HY 6. Paper 1453.
L'Association Francaise de Normalisation. 1971. Mesure de dbit de l'eau dans les chenaux au moyen
dedvcrsoireenminceparoi. X 10-311.ISO/TC 113/GT2(France-10), 152.
Pratt, E.A. 1914,'Another proportional-flow weir. Sutro weir. Engineering News, Vol. 72, No. 9, p. 462.
Rehbock, Th. 1909. Die Ausbildung der Uberflle beim Abfluss von Wasser ber Wehre nebst Beschreibung
der Anlage zur Beobachtung von Uberfallen im Flusslaboratorium zu Karlsruhe, Karlsruhe. Festschrift
der Grossherzoglichen Technischen Hochschule Fridericjana.
Rehbock, Th. 1929. Wassermessung mit scharfkantigen Uberfallwehren. Z. des Vereines Deutscher In-
genieure. 73 No. 24, pp. 817-823, Berlin.
Singer, J. and D.C.G. Lewis 1966. Proportional-flow weirs for automatic sampling or dosing. Water &
Water Engineering. V. 70, No. 841, pp. 105-1I I .
Soucek, H.E. Howe and F.T. Mavis. 1936. Sutro weir investigations furnish discharge coefficients. Engi-
neering News-Record. New York. I17.,,no.20. pp. 679-680.
Staus, A. 1931. Der Beiwert kreisrunder Uberfalle. Wasserkraft u. Wasserwirtschaft. p. 42, No. 4.
Stevens, J.C. 1957. Flow through circular weirs. J. of the Hydraulics Div. of the ASCE, Vol. 83, No. HY
6. Paper 1455.
Thomson, J. 1859. On experiments on the measurement of water by triangular notches in weir-boards.
Report of the 28th meeting British Ass. for Advancement of Science, held at Leeds, London. Sept. 1858,
pp. 181-185.
United States Bureau of Reclamation 197I . Water measurement manual. Second edition, Denver, Col.

173
Wells, J.R.1954. Discharge characteristics of rectangular notch weirs in rectangular channels. MSc. Thesis
presented to the Georgia Institute of Technol. Atlanta, Ga.

174
6 Short-crested weirs
In general, short-crested weirs are those overflow structures, in which the streamline
curvature above the weir crest has a significant influence on the head-discharge rela-
tionship of the structure.

6.1 Weir sill with rectangular control section

6.1.1 Description

A common and simple structure used in open waterways as either a drop or a check
structure is the rectangular control shown in Figure 6.1. .
The control is placed in a trapezodal approach channel, the bottom of which has
the same elevation as the weir crest (p, = O). The upstream head over the weir crest
h, is measured a distance of 1.80 m from the downstream weir face in the trapezodal
approach channel. To prevent a significant change in the roughness or configuration
in the approach channel boundary from influencing the weir discharge, the approach
channel should be lined with concrete or equivalent material over the 2 metres imme-
diately upstream of the weir. The crest surface and sides of the notch should have
plane surfaces which make sharp 90-degree intersections with the upstream weir face.
These sharp edges may be reinforced by a non-corrodible angle iron. If a movable
gate is required on the (check) structure, the grooves should be located at the down-
stream side of the weir and should not interfere with the flow pattern through the
control section.
bC
U >I concrete lined
I I ,-approach channel
I I

1.25 bc I
K >I
SECTION 1-1

1.80m
K
II r"

I concrete Iinec
I ;

LONGITUDINAL SECTION

Figure 6.1 Weir sill with rectangular control section (after Ree 1938)

175
No specific data are available on the rate of change of the weir discharge if the tailwater
level rises above the weir crest. It may be expected, however, that no significant change
in the Q - h, relationship will occur provided that the submergence ratio h,/h, does
not exceed 0.20.

6.1.2 Evaluation of discharge

As stated in Section 1.10, the basic head-discharge equation for a short-crested weir
with rectangular control section is
2 2
Q = CdC, 3& b, h,'.5

where values of the discharge coefficient c d may be obtained from Figure 6.2 as a
function of the dimensionless ratios bJh, and L/h,. Values of the approach velocity
coefficient C, can, be read as a function of CdA*/A, from Figure 1.12, where
A* = b,h,.
For a weir which has been constructed and maintained with reasonable care and
skill the error in the product cdc, in Equation 6-1, may be expected to be less than
5%. The method by which the coefficient error is to be combined with other sources
of error is shown in Annex 2.

6.1.3 Limits of application

For reasonable accuracy, the limits of application of a weir sill with rectangular control
section are:
a. The practical lower limit of h, is related to the magnitude of the influence of fluid

RATIO Llh,
1.4

Figure 6.2 Values of Cd as a function of bJh, and L/h, (adapted from Ree 1938 and after own data points)

176
 properties, to the boundary roughness in the approach section, and to the accuracy
with which h, can be determined. The recommended lower limit is 0.09 m;
b. The crest surface and sides of the control section should have plane surfaces which
make sharp 90-degree intersections with the upstream weir face;
c. The bottom width of the trapezoidal approach channel should be 1.25 b,;
d. The upstream head h, should be measured 1.80 m upstream of the downstream
weir face. Consequently, h, should not exceed half of this distance, i.e. 0.90 m;
e. To obtain modular flow the submergence ratio h,/h, should not exceed 0.20.

6.2 V-notch weir sill

6.2.1 Description

In natural streams, where it is necessary to measure a wide range of discharge, a trian-

gular control section has the advantage of providing a wide opening at high flows
so that it causes no excessive backwater effects, whereas at low flows its opening is
reduced so that the sensitivity of the structure remains acceptable.
The U.S. Soil Conservation Service developed a V-notch weir sill with 2-to-l,3-to-l,
and 5-to-1 crest slopes to measure flows up to a maximum of 50 m3/s in small streams.
Dimensions of this standard structure are shown in Figure 6.3.
The upstream head over the weir crest h, should be measured a distance of 3.00
m upstream from the weir, which equals about 1.65 times the maximum v_alue of h,

Elevation
tg.8/2= z = 2 , 3 o r 5

. . .

i 0.40 ;
Section 1-1

Figure 6.3 Dimension sketch of a V-notch weir sill (after U.S. Dept. of Agriculture 1962)

177
of 1.83 m (6 ft). A reasonably straight and level approach channel of arbitrary shape
should be provided over a distance of 15 m upstream of the weir. The weir notch
should be at least O. 15 m from the bottom or the sides of the approach channel. To
prevent the structure from being undermined, a reinforced concrete apron is required.
This should extend for about 3.50 m downstream from the weir, 0.60 m below the
vertex of the notch, 6.0 m across the channel, and it should have a 1 .O m end cutoff
wall. The middle 3.0 m section of this apron should be level and the two 1S O m sides
should slope slightly more than the weir crest.
No specific data are available on the rate of change of the weir discharge if the
tailwater level rises above the weir crest. It may be expected, however, that there will
be no significant change in the Q-h, relationship provided that the submergence ratio
h2/h,does not exceed 0.30.

6.2.2 Evaluation of discharge

The basic head-discharge equation for a short-crested weir with a triangular control
section is as shown in Section I .9.3:
Q = C,Cvz[5g]
162 0.5 tan28 h,2.5

where tan@/2equals G.Based on hydraulic laboratory tests conducted by the U.S.

Soil Conservation Service at Cornel1 University, Ithaca, N.Y., rating tables have been
developed giving the discharge in m3/s at each 0.3048 m (1 foot) of head for a number
of wetted areas, A,, at the head measurement station. These are presented in Table
6.1. From this table, it is possible to read, for example, that the discharge over a 5-to-1
V-notch weir under a head h, = 0.915 m and a wetted area of the approach channel
of A, = 6.50 m2 equals 7.70 m3/s. For a wetted area of A, = 15.0 mz, and therefore
with a lower approach velocity, the weir discharge equals 6.56 m3/s under the same
head. The head-discharge relationship for these weirs can be obtained by plotting the
discharge for each 0.3048 m (1 foot) of head and the corresponding wetted area of
the approach channel.
Discharges for heads up to 0.20 m can be obtained from Table 6.2. A smooth line
is drawn through the plotted points and a rating table for each 0.01 m of head is
produced from this curve. It should be understood that any significant change in the
approach cross-section, due either to cutting or filling, requires a revision of the Q - h,
curve.
It can be expected that for a well-maintained V-notch weir which has been con-
structed with reasonable care and skill the error in the discharges shown in Tables
6.1 and 6.2 will be less than 3%. The method by which this error is to be combined
with other sources of error is shown in Annex 2.

178
Table 6.1 Rating table for V-notch weir sill (adapted from data of U.S. Soil Conservation Service at Cornell
University, Ithaca)

hl = 0.305111 h l = 0.610m h, = 0.915m h l = 1.219m h, = 1.524m hl = 1.829m

(1 ft) (2ft) (3 ft) (4ft) (5 ft) (6ft)
A l in Qin A l in Qin A l in Qin A l in Qin A l in Qin Al in Qin
m2 m3/s m2 m3/s m2 m3/s m2 m3/s m2 m3/s m2 m3/s

2-to-IV-notchweir

0.55 0.159 1.20 1.17 2.40 3.74 4.30 7.56 6.75 13.7 9.50 25.4
0.60 0.157 1.25 1.13 2.45 3.60 4.50 7.00 7.00 12.5 9.75 21.3
0.65 0.156 1.30 1.09 2.50 3.48 4.75 6.64 7.25 11.8 10.00 19.6
0.70 0.156 1.35 1.06 2.65 3.23 5.00 6.38 7.50 11.4 10.25 19.0
0.75 0.155 1.40 1.03 3.00 2.93 5.50 6.06 7.75 11.2 10.50 18.5
0.80 0.154 1.45 1.02 3.50 2.78 6.00 5.84 8.00 10.9 11.0 17.6
0.90 0.154 1.50 1.00 4.00 2.70 6.50 5.69 8.50 10.6 12.0 16.8
1.00 0.153 1.60 0.980 4.50 2.64 7.00 5.61 9.00 10.4 13.0 16.2
1.50 0.153 1.75 0.962 5.00 2.61 7.50 5.54 9.75 10.2 14.0 15.8
2.00 0.152 2.00 0.942 6.00 2.57 8.00 5.46 10.50 9.96 17.0 15.1
3.00 0.152 2.50 0.927 7.50 2.55 10.0 5.30 12.00 9.74 20.0 14.8
4.00 0.152 5.00 0.898 10.00 2.52 13.0 5.24 15.00 9.35 25.0 14.6
5.00 0.151 7.50 0.895 15.00 2.50 16.0 5.21 20.0 9.16 30.0 14.4
6.00 0.151 10.0 0.895 20.00 2.49 23.0 5.18 30.0 9.01 40.0 14.3
7.00 0.151 14.0 0.895 25.00 2.49 . 32.0 5.18 40.0 9.00 60.0 14.2

3-to-1V-notchwcir

0.75 0.237 1.85 1.62 3.75 5.24 6.50 12.0 10.5 19.2 15.0 30.6
0.80 0.234 2.00 1.56 4.00 4.83 6.75 10.8 11.0 18.0 15.5 28.8
0.90 0.232 2.50 1.44 4.50 4.43 7.00 10.3 12.0 16.8 16.0 27.6
1.00 0.230 2.75 1.42 5.00 4.25 7.50 9.69 13.0 16.0 16.5 27.0
1.20 0.228 3.00 1.40 5.50 4.14 8.00 9.31 14.0 15.5 17.0 26.5
1.50 0.226 3.50 1.38 6.00 4.06 8.50 9.02 15.0 15.1 18.0 25.9
2.00 0.225 4.00 1.37 6.50 4.02 9.00 8.81 16.0 14.9 19.0 24.9
3.00 0.224 4.50 1.36 7.00 3.98 9.50 8.67 17.0 14.7 20.0 24.4
5.00 0.224 5.00 1.35 8.00 3.91 10.0 8.55 18.0 14.5 22.5 23.6
7.50 0.224 5.50 1.35 9.00 3.88 11.0 8.37 20.0 14.4 25.0 23.1
6.00 1.34 10.0 3.85 12.0 8.25 22.5 14.2 27.5 22.8
7.00 1.34 12.5 3.79 15.0 8.03 25.0 14.0 30.0 22.5
8.00 1.33 15.0 3.77 20.0 7.90 30.0 13.9 40.0 22.0
10.00 1.33 20.0 3.75 30.0 7.80 40.0 13.8 50.0 21.8
14.00 1.33 25.0 3.74 45.0 7.79 55.0 13.7 60.0 21.7

5-to-IV-notchweir

1.50 0.386 2.80 2.77 5.60 8.76 11.0 16.4 16.0 30.0 22.0 49.4
2.00 0.382 3.00 2.67 5.75 8.49 11.25 16.1 17.0 27.8 23.0 46.3
3.00 0.378 3.25 2.58 6.00 8.17 11.5 15.8 18.5 26.9 25.0 43.4
6.00 0.376 3.50 2.52 6.25 7.87 12.0 15.5 20.0 26.0 27.5 41.1
10.00 0.376 3.75 2.49 6.50 7.70 12.5 15.2 22.5 25.0 30.0 39.8
4.00 2.45 7.00 7.42 13.0 15.0 25.0 24.5 32.5 39.1
4.50 2.42 7.50 7.26 14.0 14.6 27.5 24.1 35.0 38.5
5.00 2.39 8.00 7.10 15.0 14.4 30.0 23.8 37.5 38.0
6.00 2.35 9.00 6.99 16.0 14.2 32.5 23.7 40.0 37.6
7.00 2.33 10.0 6.84 17.5 14.0 35.0 23.6 45.0 37.1
8.50 2.32 12.0 6.67 20.0 13.8 40.0 23.4 50.0 36.6
10.00 2.30 15.0 6.56 25.0 13.5 50.0 23.3 55.0 36.4
15.00 2.23 20.0 6.48 30.0 13.4 60.0 23.2 60.0 36.2
20.00 2.28 30.0 6.42 40.0 13.3 75.0 23.0 75.0 36.0
25.00 2.28 45.0 6.40 65.0 13.2 90.0 22.8 90.0 36.0

179
Table 6.2 Discharge values for heads up to 0.20 m ofV-notch weirs m3/s x

Discharge in litres per second for V-notch weirs

Head 2-to-1 3-to-1 5-to-I
(metres) (a) (b) (c)

0.03 0.5 0.7 1.o

0.04 0.9 1.3 2.0
0.05 1.7 2.3 3.7
0.06 2.6 3.7 6.0
0.07 3.7 5.3 8.6
0.08 5.2 7.3 12.1
0.09 6.9 9.9 16.1
0.10 8.9 12.9 21.1
0.1 1 11.2 16.4 26.8
0.12 13.9 20.5 33
0.13 17.0 25.3 41
O. 14 21.8 31 52
0.15 . 24.5 37 59
0.16 28.3 43 68
0.17 34 51 80
0.18 39 59 93
0.19 45 67 107
0.20 51 71 . 123

NOTE:
Applicable to stations with cross-sectional areas at head measurement station equal to or greater than
(a) = 0.55 m2 for 0.30 m head
(b) = 0.75 m2 for 0.30 m head
(c) = 1.40 m2 for 0.30 m head

6.2.3 Limits of application

For reasonable accuracy, the limits of application of the V-notch weir sill are:
a. The head over the weir crest should be at least 0.03 m and should be measured
a distance of 3.00 m upstream from the weir.
b. The notch should be at least O. 15 m from the bottom or the sides of the approach
channel;
c. The approach channel should be reasonably straight and level for 15.0 m upstream
from the weir.
d. To obtain modular flow the submergence ratio h2/h, should not exceed 0.30.

6.3 Triangular profile two-dimensional weir

6.3.1 Description

The triangular profile two-dimensional weir is sometimes referred to in the literature

as the Crump weir, a name credited to E S . Crump, who described the device for
the first time in a paper in 1952. The profile of the weir in the direction of flow shows
an upstream slope of 1 (vertical) to 2 (horizontal) and a downstream slope of either
1-to-5 or I-to-2. The intersection of the two sloping surfaces forms a straight horizontal

180
 crest at right angles to the flow direction in the approach channel. Care should be
taken that the crest has a well-defined corner of durable construction. The crest may
either be made of carefully aligned and joined precast concrete sections or have a
cast-in non-corrodible metal profile (Figure 6.4).

Tests were carried out at the Hydraulics Research Station at Wallingford ( U K ) to

determine the maximum permissible truncation of the weir block in the direction of
flow whereby the discharge coefficient was to be within 0.5% of its constant value.
It was found that for a I-to-2 / 1-to-5 weir the minimum horizontal distance from
the weir crest to point of truncation of the weir block equals 1.0 HI,,, for the 1-to-2
slope and 2.0 Hlmax for the 1-to-5 slope. For a I-to-2 / I-to-2 weir, these minimum
distances equal 0.8 Hlmaxfor the upstream slope and 1.2 Hlma,for the downstream
slope.
The upstream head over the weir crest h, should be measured in a rectangular ap-
proach channel at a sufficient distance upstream from the crest to avoid the area of
surface draw-down, but close enough to the crest for the energy loss between the head
measurement station and the control section to be negligible. For this to occur, the
head measurement station should be at a distance LI = 6pI upstream from the weir
crest for a I-to-2 / I-to-5 weir and at L, = 4p, for a 1-to-2 / 1-to-2 weir. If no particular-
ly high degree of accuracy is required in the maximum discharges to be measured,
savings can be made in the construction cost of the structure by reducing the distance
+
from the crest to head measurement station to 2p, 0.5H,,,,. The additional error
introduced will be of the order of 0.25% at an H,/p, value of 1, of 0.5% at an H,/p,
value of 2, and of 1% at an H,/pl value of 3.
If the weir is to be used for discharge measuring beyond the modular range, crest
tappings should be provided to measure the piezometric level in the separation pocket
formed immediately downstream of the crest. The crest tapping should consist of a

Figure 6.4 Triangular profile two-dimensional weir

\ 181
sufficient number (usually 4 to 12) of @ 0.01 m holes drilled in the weir crest block
on 0.10 m centres 0.019 m downstream from the weir crest as shown in Figure 6.5.
The edges of the holes should not be rounded or burred.
Preferably, the crest tapping should be located at the centre of the weir, but may
be off-centre provided that the side walls do not interfere with the pressure distribution
in the separation pocket. A distance of about 1.20 m from the side walls should be
sufficient. Weirs with a breadth b, of less than 2.5 m should have the crest tapping
in the centre.

6.3.2 Evaluation of discharge

According to Sections I . 10 and 1.13.1, the basic head-discharge equation for a short-
crested weir with rectangular control section reads
2
Q = CdC, 3 f i b, (6-3)

where the effective head over the weir crest he = h, + K,, K, being an empirical
quantity representing the combined effects of several phenomena attributed to visco-
sity and surface tension. A constant value of K, = 0.0003 m for 1-to-2 / 1-to-5 weirs,
and of Kh = 0.00025 for 1-to-2 / 1-to-2 weirs is recommended. For field installations
where it is not practicable to determine h,-values accurate to the nearest 0.001 m the
use of Kh is inappropriate. Consequently values of he N h, may be used on these
installations.
Over the selected range of the ratio h,/p,, being h,/p, < 3, the discharge coefficient
is a function of the dimensionless ratio H,/p2as illustrated in Figure 6.6.
The curve for the l-to-2/1-to-2 weir shows that the discharge coefficient for low
values of p2 begins to fall at a value H,/p, = 1.0 and is 0.5% below the average deep
downstream value at H,/p2 = 1.25. The curve for the l-to-2/1-to-5 weir shows corres-
ponding values of HJp, = 2.0 and H,/p2 = 3.0, thereby indicating that the discharge
coefficient for a 1-to-5 downstream slope is considerably more constant in terms of
the proximity of the downstream bed. For high p2 values, the discharge coefficient
,- 0.019m

Gmnite concrete
precast wir
block with rubbed

Crest tapping c h o m k v
0.20 m

I I
I
0.25 m -4 i
P
Figure 6.5 Alternative solutions for crest tappings

182
DISCHAFGE COEFFICIENT Cd

RATIO H,b,

Figure 6.6 Two-dimensional triangular profile weirs, effect of downstream bed level on modular Cd-value
(after White 1971)

of the 1-to-2/1-to-2 weir has a higher value (C, = 0.723) than the 1-to-2/1-to-5 weir
(C, = 0.674) since the streamlines above the crest of the latter have a larger radius
of curvature (see also Section 1.IO):
The approach velocity coefficient C, = (H,/h,)3/2is related to the ratio
{cdh,/(h, + p,)} b,/B, and can be read from Figure 1.12.
The error in the product cdc, of a well-maintained triangular profile weir with
modular flow, constructed and installed with reasonable care may be deduced from
the equation
X, f (10CV-9)percent
= (6-4)
The method by which this error is to be combined with other sources of error is shown
in Annex 2.

6.3.3 Modular limit

The modular limit, or that submergence ratio H,/H, which produces a 1% reduction
from the equivalent modular discharge, depends on the height of the crest above the
average downstream bed level. The results of various tests are shown in Figure 6.7,
where the modular limit H,/H, is given as a function of the dimensionless ratio H,/p,.
For non-modular flow conditions, the discharge as calculated by Equation 6-3, i.e.

183
the discharge that would occur with low tailwater levels, has to be reduced by a factor
which is a function of the downstream head over the weir crest. For non-modular
flow, the discharge thus equals

Q = cdc, f 32 f i b, (6-5)

The drowned flow reduction factor f i s easier to define and evaluate for weirs which
have a constant discharge coefficient: Figure 6.7 shows that the 1-to-2/1-to-5 weir
has a more favourable modular limit, while Figure 6.6 shows that the C,-coefficient
is constant over a wider range of H,/p,. The Hydraulics Research Station, Wallingford
therefore concentrated its study on the drowned flow performance of the 1-to-2/1-to-5
weir. A graph has been produced giving values of the product C,f as a function of
+
the two-dimensionless ratios {Cdh,/(h, pl)}b,/B, and h,/h,, where h, equals the pie-
zometric pressure within the separation pocket. The product C,f can be extracted from
Figure 6.8 for values of the two ratios. Substitution of C,f into Equation 6-5 then
gives the weir discharge for its non-modular range.

6.3.4 Limits of application

For reasonable accuracy, the limits of application of the triangular profile weir are:
a. For a well-maintained weir with a non-corrodible metal insert at its crest, the recom-
mended lower limit of h, = 0.03 m. For a weir with a crest made of precast concrete
sections or similar materials, h, should not be less than 0.06 m;
b. The weir, in common with other weirs and flumes, becomes inaccurate when the
Froude number, Fr, = v,/(gA,/B,)/2,in the approach channel exceeds 0.5, due to
the effects of surface instability in the form of stationary waves. The limitation
Fr, < 0.5 may be stated in terms of h, and p,. The recommended upper limit of
hllp, is 3.0;
RATIO H./ H1
0.9

0.0

0.7

0.6

05

0.4

03

0.2

0.1

O
O 0.5 1.0 1.5 2.0
RATIO Hl/p2

Figure 6.7 Modular limit as a function of H,/p2 (after Crump 1952, and H.R.S. Wallingford, 1966 and
1971)

184
c. The height of the weir crest should not be less than 0.06 m above the approach
channel bottom (pI 2 0.06 m);
d. To reduce the influence of boundary layer effects at the sides of the weir, the breadth
of the weir b, should not be less than 0.30 m and the ratio b,/H, should not be
less than 2.0;
e. To obtain a sensibly constant discharge coefficient for l-to-2/1-to-2 profile weirs,
the ratio HJp, should not exceed 1.25. For l-to-2/1-to-5 profile weirs, this ratio
should be less than 3.0.

6.4 Triangular profile flat-V weir

6.4.1 Description

In natural streams where it is necessary to measure a wide range of discharges, a trian-

gular control has the advantage of providing a wide opening at high flows so that
it causes no excessive backwater effects, whereas at low flows its opening is reduced
so that the sensitivity of the structure remains acceptable. The Hydraulics Research
Station, Wallingford investigated the characteristics of a triangular profile flat-V weir
with cross-slopes of 1-to-10 and 1-to-20. (For the two-dimensional triangular profile
weir, see Section 6.3.) The profile in the direction of flow shows an upstream slope
of 1-to-2 and a downstream slope of either 1-to-5 or 1-to-2 (Figure 6.9). The intersec-

185
 Figure 6.9 Triangular profile flat-V weir

. tions of the upstream and downstream surfaces form a crest at right angles to the
flow direction in the approach channel. Care should be taken that the crest has a
well-defined corner made either of carefully aligned and joined precast concrete sec-
tions or of a cast-in non-corrodible metal profile.
The permissible truncation of the weir block is believed to be the same as that of
' the two-dimensional weir (see Section 6.3.1). Therefore the minimum horizontal dis-
tance from the weir crest to the point of truncation whereby the c,-value is within
0.5% of its constant value, equals 1.0 Hlmax for the upstream and 2.0 Hlmaxfor the
downstream slope of a 1-to-2/1-to-5 weir. For a l-to-2/1-to-2 weir these minimum
distances equal 0.8 Hlmax for the upstream slope and 1.2 Hlmaxfor the downstream
slope.
The upstream head over the weir crest h, should be measured in a rectangular ap-
proach channel at a distance of ten times the V-height upstream of the crest, i.e. L,
= 10 Hb. At this location, differential drawdown across the width of the approach
channel is negligible and a true upstream head can be measured accurately.
If a l-to-2/1-to-5 weir is to be used for discharge measuring beyond its modular
range, three crest tappings should be provided to measure the piezometric 'level in
the separation pocket, h,, immediately downstream (0.019 m) of the crest (see also
Figure 6.5). One crest tapping should be at the centre line, the other two at a distance
of O. 1 B, offset from the centre line.

6.4.2 Evaluation of discharge

According to Section 1.10, the basic head-discharge equation for a short-crested flat-V
weir with vertical side walls reads

186
Photo 6.1 An l-to-2/1-to-5 shaped weir on a natural stream

in which the term (he - H,)2.5should be deleted if he is less than H,. The effective
head over the weir crest he = h, - Kh, Kh being an empirical quantity representing
the combined effects of several phenomena attributed to viscosity and surface tension.
Values for Kh are presented in Table 6.3.

Table 6.3 Kh-values for triangular profile flat-V weirs (White 1971)

weir profile I-to-20 1-10-IO

cross slope cross slope

I -to-2/ 1-to-2 0.0004 m 0.0006 m

I-to-2/1-to-5 0.0005m 0.0008 m

For the 1-to-2/1-to-5 weir, an average C,-value of 0.66 may be used for both cross
slopes provided that the ratio h,/p, < 3.0. The C,-value of a 1-to-2/1-to-2 weir is
more sensitive to the bottom level of the tailwater channel with regard to crest level.
An average value of Cd = 0.71 may be used provided that he/p, does not exceed 1.25.

187
 C., values

he' Hb

Figure 6.10 C,-values as a function of hJpl and h,/Hb (after White 1971)

The approach velocity coefficient C, can be read as a function of the ratios hJp, and
h,/Hb in Figure 6.1O.
The error in the product CdCVof a well-maintained triangular profile weir with
modular flow, constructed with reasonable care and skill may be expected to be
X, = f (10CV-8)percent (6-7)
The method by which this error is to be combined with other sources of error is shown
in Annex 2.

6.4.3 Modular limit and non-modular discharge

The modular limit again is defined as that submergence ratio H,/H, which produces
a 1% reduction from the equivalent modular discharge as calculated by Equation 6-6.
Results of various tests have shown that for a 1-to-2/1-to-2 weir the drowned flow
reduction factor, f, and thus the modular limit, are functions of the dimensionless
ratios H,/H,, H,/Hb,H,/p,, H,/p,, and the cross slope of the weir crest.
Because of these variables, the modular limit characteristics of a l-to-2/1-to-2 weir
are rather complex and sufficient data are not available to predict the influence of
the variables. A limited series of tests in which only discharge, cross-slope, and down-
stream bed level (p,) were varied was undertaken at Wallingford. The results of these
tests, which are shown in Figure 6.1 1, are presented mainly to illustrate the difficulties.
For a l-to-2/1-to-5 profile weir, the drowned flow reduction factor is a less complex
phenomenon, and it appears that the f-value is a function of the ratios HJH, and
H,/Hbonly (Figure 6.12). Tests showed that there is no significant difference between
the modular flow characteristics of the weirs with either 1-to-10 or 1-to-20 cross slopes.
As illustrated in Figure 6.12, the drowned flow reduction factor f equals 0.99 for

188
 a4

Geometry
I Two-dimensional w e 7
1 : 2 0 Cross slcpe weir
0.2 1:lO Ct'usssbpe web-

O
0.1 E

-,0
O ( 1.5 21) 2.5 3.0
H1 'P2 d

Figure 6.11 Modular limit conditions, triangular profile I-to-2/1-to-2 flat-V (after White 1971)

modular limit values between 0.67 and 0.78, depending on the modular value of H,/H,.
For non-modular flow conditions, the discharge over the weir is reduced because of
high tailwater levels, and the weir discharge can be calculated from Equation 6-8,

O. 70

epz

DROWNED FLOW REDUCTION FACTOR t

Figure 6.12 Modular limit conditions, I-to-2/1-to-S flat-V weir (adapted from White 1971)

189
 4 B
Q = CdCVf -(2g)0.53[h,2.5- (he- Hb)2.5]
15 Hb

This equation is similar to Equation 6-6 except that a drowned flow reduction factor
f has been introduced. For 1-to-2/1-to-5 profile weirs, f-values have been determined
and, in order to eliminate an intermediate step in the computation of discharge, they
have also been combined with the approach velocity coefficient as a product C,f. This
product is a function of he/Hb,hp/h,, and Hb/p, and as such is presented in Figure
6.13. To find the proper C,f-value, one enters the figure by values of h,/Hb and hp/he
and by use of interpolation in terms of Hb/p, a value of the product C,f is obtained.
Substitution of all values into Equation 6-8 gives the non-modular discharge.

0 9 5 L '

- Hb/ P i < 0.3

Deep approach -
----Hb/p~=Io
Intermediate approach

0.5 1.0 1.5 2 .o

he/Hb

Figure 6.13 Values of C,f for a l-to-2/1-to-5

flat-V weIr as a functlon of h,/Hb, hp/h, and (after
White 1971)

190
6.4.4 Limits of application

For reasonable accuracy, the limits of application of a triangular profile flat-V weir
are:
a. For a well-maintained weir with a non-corrodible metal insert at its crest, the recom-
mended lower limit of h, = 0.03 m. For a crest made of pre-cast concrete sections
or similar materials, h, should not be less than 0.06 m;
b. To prevent water surface instability in the approach channel in the form of station-
ary waves, the ratio hJp, should not exceed 3.0;
C. The height of the vertex of the weir crest should not be less than 0.06 m above
the approach channel bottom;
d. To reduce the influence of boundary layer effects at the sides of the weir, the width
of the weir B, should not be less than 0.30 m and the ratio BJH, should not be
less than 2.0;
e. To obtain a sensibly constant discharge coefficient for 1-to-2/1-to-2 profile weirs,
the ratio H,/p, should not exceed 1.25. For I-to-2/1-to-5 profile weirs, this ratio
should be less than 3.0;
f. The upstream head over the weir crest should be measured a distance of 10 H,
upstream from the weir crest in a rectangular approach channel;
5 To obtain modular weir flow, the submergence ratio H,/H, should not exceed 0.30
for I-to-2/1-to-2 profile weirs and should be less than 0.67 for l-to-2/1-to-5 profile
weirs. For the latter weir profile, however, non-modular flows may be calculated
by using Equation 6-8 and Figure 6.13.

6.5 Butchers movable standing wave weir

6.5.1 Description

Butchers weir was developed to meet the particular irrigation requirements in the
Sudan, where the water supplied to the fields varies because of different requirements
during the growing season and because of crop rotation. A description of the weir
was published for the first time in 1922 by Butcher, after whom the structure has
been named.* The weir consists of a round-crested movable gate with guiding grooves
and a self-sustaining hand gear for raising and lowering it. The cylindrical crest is
horizontal perpendicular to the flow direction. The profile in the direction of flow
shows a vertical upstream face connected to a 1-to-5 downward sloping face by a
0.25 hlmaxradius circle, where hlmaxis the upper limit of the range of heads to be expected
at the gauge located at a distance 0.75 hlmax upstream from the weir face.

The side walls are vertical and are rounded at the upstream end in such a way that
flow separation does not occur. Thus a rectangular approach channel is formed to
assure two-dimensional weir flow. The upstream water depth over the weir crest h,
is measured in this approach channel by a movable gauge mounted on two supports.
The lower support is connected to the movable gate and the upper support is bolted
to the hoisting beam. The gauge must be adjusted so that its zero corresponds exactly
* Nowadays the structure is manufactured commercially by Boving Newton Chambers Ltd., Rotherham,
SGO 1TF. U.K.

191
 9
Lifting g e a r

K Bronze gauge

r-'
E /
Gauge /
/
/
/
/
/
1 /

2 L /

1 Fixed weir-
/
4iternative '
'ixed weir-

Figure 6.14 Butcher's movable gate

with the weir crest. Because of their liability to damage the supports have been kept
rather short; a disadvantage of this shortness is that the water surface elevation is
measured in the area of surface drawdown so that the hydraulic dimensions of both
the approach channel and weir cannot be altered without introducing an unknown
change in the product of CdCV.The centre line of the gauge should be 0.75 hlmax up-
stream from the weir face.
The weir can be raised high enough to cut off the flow at full supply level in the
feeder canal and, when raised, leakage is negligible. In practice it has been found ad-
vantageous to replace the lower fixed weir, behind which the weir moves, with a con-

192
 90 DIVERSION

W
U
2
a

c- 2 h 1 max
y
A WEIR LENGTH I
I

w
I AERATION

I I pTGROWE
I---,-

MOVABLE

FLOW ~

DETAILS O F WEIR ABUTMENT

STRIP 130x8

NCE

...
L..__^ L.^

/ MOVABLE WEIR FACE

DETAILS GROOVE ARRANGEMENT

Figure 6.14 (cont.)

Crete or masonry sill whose top width is about 0.10 m and whose upstream face is
not flatter than 2-to- I .
The maximum water depth over the weir crest, and thus the maximum permissible
discharge per metre weir crest, influences the weir dimensions. Used in the Sudan
are two standard types with maximum values of h, = 0.50 m and h, = 0.80 m respecti-
vely. It is recommended that 1.00 m be the upper limit for h,. The breadth of the
weir varies from 0.30 m to as much as 4.00 m, the larger breadths used in conjunction
with high hlmax-values. As shown in Figure 6.14, pI = 1.4 hlma,,which results in low
approach velocities.

193
The modular limit is defined as the submergence ratio h,/h, which produces a 1%
reduction from the equivalent modular discharge. Results of various tests showed that
the modular limit is h,/h, = 0.70. The average rate of reduction from the equivalent
modular discharge is shown in percentages in Figure 6.15.

6.5.2 Evaluation of discharge

Since the water depth over the weir crest is measured in the area of water surface
drawdown at a distance of 0.75 h,,,, upstream from the weir face, i.e. h,,,, upstream
from the weir crest, the stage-discharge relationship of the weir has the following em-
pirical shape
Q =c b, h,'.6 (6-9)
where h, equals the water depth at a well-prescribed distance L, = 0.75 hlmax
upstream
from the weir face. It should be noted that this water depth is somewhat lower than
the real head over the weir crest. For weirs that are constructed in accordance with
the dimensions shown in Figure 6.14, the effective discharge coefficient equals c =
2.30 m0.4s-I. The influence of the approach velocity on the weir flow is included in
this coefficient value and in the exponent value 1.6.
The error in the discharge coefficient c of a well-maintained Butcher movable weir
which has been constructed with reasonable care and skill may be expected to be less
than 3%. The method by which this error is to be combined with other sources of
error is shown in Annex 2.

SUBMERGENCE RATIO H,/H,

PERCENT-
./ REDUCTION FROM EQUIVALENT MODULAR FLCM

Figure 6.15 Modular flow condition

194
6.5.3 Limits of application

For reasonable accuracy, the limits of application of Equation 6-9 for Butchers mova-
ble weir are:
a. All dimensions of both the weir and the approach channel should be strictly in
accordance with the dimensions shown in Figure 6.14;
b. The width of the weir b, should not be less than 0.30 m and the ratio bJh, should
not be less than 2.0;
c. The upstream water depth should be measured with a movable gauge at a distance
of 0.75 h,,,, upstream from the weir face;
d. To obtain modular flow, the submergence ratio h,/h, should not exceed 0.70;
e. The recommended lower limit of h, = 0.05 m, while h, should preferably not exceed
1 .O0 m.

6.6 WES-Standard spillway

6.6.1 Description

From an economic point of view, spillways must safely discharge a peak flow under
the smallest possible head, while on the other hand the negative pressures on the crest
must be limited to avoid the danger of cavitation. Engineers therefore usually select
a spillway crest shape that approximates the lower nappe surface of an aerated sharp
crested weir as shown in Figure 6.16.

Theoretically, there should be atmospheric pressure on the crest. In practice, however,

friction between the surface of the spillway and the nappe will introduce some negative
pressures. If the spillway is operating under a head lower than its design head, the
nappe will normally have a lower trajectory so that positive pressures occur throughout
the crest region and the discharge coefficient is reduced. A greater head will cause
negative pressures at all points of the crest profile and will increase the discharge coeff-
cient.
The magnitude of the local minimum pressure at the crest (P/pg),in has been mea-
sured by various investigators. Figure 6. I7 shows this minimum pressure as a function
of the ratio of actual head over design head as given by Rouse & Reid (1935) and
Dillman (1933).

h, -equivalent head for

comparable sharp-crested weir

Figure 6.16 Spillway crest and equivalent sharp-crested weir

195
 A O W HEWDESIGN HEAD; h, /hd

Figure 6.17 Negative pressure on spillway crest (after Rouse & Reid I935 and Dillman 1933)

The avoidance of severe negative pressures on the crest, which may cause cavitation
on the crest or vibration of the structure, should be considered an important design
criterion on high-head spillways. In this context it is recommended that the minimum
pressure on the weir crest be - 4 m water column. This recommendation, used in
combination with Figure 6.17, gives an upper limit for the actual head over the crest
of a spillway.

On the basis of experiments by the U.S. Bureau of Reclamation the U.S. Army Corps
of Engineers conducted additional tests at their Waterways Experimental Station and
produced curves which can be described by the following equation
x"= K hdn-'Y (6-10)
which equation may also be written as

(6-1 1)

where X and Y are coordinates of the downstream crest slope as indicated in Figure
6.18 and hd is the design head over the spillway crest. K and n are parameters, the
values of which depend on the approach velocity and the inclination of the upstream
spillway face. For low approach velocities, K and n-values for various upstream slopes
are as follows:

196
Slope of upstream face K n

vertical 2,000 1,850

3to1 1,936 1,836
3 to2 1,939 1,810
1 to 1 1,873 1,776

The upstream surface of the crest profile varies with the slope of the upstream spillway
face as shown in Figure 6.18.

.Photo 6.2 WES-spillway operating under low head

197
 I .1.936 hA0.836y

C r r s t axis

I
I
0.01o y

Figure 6.18 WES-standard spillway shapes (U.S. Army Corps,of Engineers 1952)

198
6.6.2 Evaluation of discharge

The basic head discharge equation for a short-crested weir with a rectangular control
section reads
Q =
2F
Ce - - g b, HI1.
3 3
(6-12)

Since the WES-standard spillway evolved from the sharp-crested weir, we might also
use an equation similar to that derived in Section I . 13.I , being

(6-13)

A comparison of the two equations shows that Ce* = Ce/$, so that it is possible
to use whichever equation suits ones purpose best.
In these two equations the effective discharge coefficient Ce (or Ce*)equals the pro-
duct of Co(or Co*),C I and C , (Ce = CoCIC2).Co(or Co*)is a constant, CIis a function
of pl/hdand H,/h,, and C, is a function of pl/hl and the slope of the upstream weir
face.
As illustrated in Figure 6.16 the high point of the nappe, being the spillway crest,
is O. 11 h,, above the crest of the alternative sharp-crested weir (see also Figure 1.23).
As a result, the spillway discharge coefficient at design head, h, is about 1.2 times
that of a sharp-crested weir discharging under the same head, provided that the ap-
proach channel is sufficiently deep so as not to influence the nappe profile. Model
tests of spillways have shown that the effect of the approach velocity on Ceis negligible
when the height, p,, of the weir is equal to or greater than 1.33 h, , where h, is the
design head excluding the approach velocity head. Under this condition and with an
actual head, H I , over the spillway crest equal to design head h,, the basic discharge
coefficient equals C, = 1.30 in Equation 6-12 and Ce* = 0.75 in Equation 6-13.
C, can be determined from a dimensionless plot by Chow (1959), which is based
on data of the U.S. Bureau of Reclamation and of the Waterways Experimental Sta-
tion (1952), and is shown in Figure 6.19.
The values of C, in Figure 6.19 are valid for WES-spillways with a vertical upstream
face. If the upstream weir face is sloping, a second dimensionless correction coefficient
C2 on the basic coefficient should be introduced; this is a function of both the weir
face slope and the ratio pl/Hl.Values of C2can be obtained from Figure 6.20.
By use of the product Ce = C,CIC2 an energy head-discharge relationship can now
be determined provided that the weir flow is modular. After calculation of the approxi-
mate approach velocity, vI, this Q-HI relationship can be transformed to a Q-h, curve.

To allow the WES-spillway to function as a high capacity overflow weir, the height
pz of the weir crest above the downstream channel bed should be such that this channel
bed does not interfere with the formation of the overflowing jet. It is evident that
when p2 approaches zero the weir will act as a broad-crested weir, which results in
a reduction of the effective discharge coefficient by about 23 percent. This feature
is shown in Figure 6.21. This figure also shows that in order to obtain a high C,-value,
the ratio p,/H, should exceed 0.75.
Figure 6.21 also shows that, provided p,/H, 2 0.75, the modular discharge as calcu-

I99
lated by Equation 6-12 is decreased to about 99% of its theoretical value if the submer-
gence ratio H2/HIequals 0.3. Values of the drowned flow reduction factor f, by which
the theoretical discharge is reduced under the influence of both p2/Hland H,/H,, can
be read from Figure 6.2 1.
The accuracy of the discharge coefficient Ce = CoCICz of a WES-spillway which
has been constructed with care and skill and is regularly maintained will be sufficient

RATIO Hllhd
1.3

1.2

1.1

1.o

a9
oa
a7
0.6

0.5
0.4

a3

a2
0.1
o
0.70 0.75 OB0 0.85 0.90 0.95 1.00
CORRECTON FACTOR C1

Figure 6.19 Correction factor for other than design head on WES-spillway (after Chow 1959, based on
data of USBR and WES 1952)

CORRECTION COEFFICIENT C2

I I I I I I I I I I I I I I

t?ATIO P I / H ~

Figure 6.20 Correction factor for WES-spillway with sloping upstream face (after U.S. Bureau of Reclama-
tion 1960)

200
 - 02
- 0.1

O.

0.1

0.2

0.3

0.4

a5

i 0.6
P
Q 0.7
z

1.0 u
O 05 ID 1.5 20 25 3.0 3.5 40
RATIO p,/H,

Figure 6.21 Drowned flow reduction factor as a function of p2/Hl and H2/Hl (Adapted from U.S. Army
Corps of Engineers, Waterways Experimental Station 1952)

for field conditions. The error of Ce may be expected to be less than 5%. The method
by which this error is to be combined with other sources of error is shown in Annex 2 .

6.6.3 Limits of application

For reasonable accuracy, the limits of application of a weir with a WES-spillway crest
are:
a. The upstream head over the weir crest hi should be measured a distance of 2 to
3 times hlmax upstream from the weir face. The recommended lower limit of hi is
0.06 m;
b. To prevent water surface instability in the approach channel, the ratio pl/h, should
not be less than 0.20;
c. To reduce the influence of boundary layer effects at the side walls of the weir, the
ratio b,/H, should not be less than 2.0;
d. To obtain a high Ce-value, the ratio p2/Hlshould not be less than about 0.75;
e. The modular limit H2/H, = 0.3, provided that the tailwater channel bottom does
not interfere with the flow pattern over the weir (p2/H, >, 0.75);

20 1
f. The minimum allowable pressure at the weir crest equals - 4.0 m water column
(P/pg 2 - 4.0 m).

6.7 Cylindrical crested weir

6.7.1 Description

A cylindrical crested weir is an overflow structure with a rather high discharge coeffi-
cient and is, as such, very useful as a spillway. The weir consists of a vertical upstream
face, a cylindrical crest which is horizontal perpendicular to the direction of flow,
and a downstream face under a slope I-to-1 (CY= 45") as shown in Figure 6.22. The
abutments are vertical and should be rounded in such a manner that flow separation
does not occur.

If the energy head over the weir crest as a function of the radius of the crest is small
(HJr is small), the pressure on the weir crest is positive; if, however, the ratio Hl/r
becomes large, the position of the overfalling nappe is depressed below that of a free
falling nappe and the pressure of the crest becomes negative (sub-atmospheric) and
at the same time causes an increase of the discharge coefficient. The magnitude of
the local minimum pressure at the crest (P/pg)min was measured by Escande & Sana-
nes (1959), who established the following equation from which P/pg minimum can
be calculated
P/pg H, - (Hl -Y> {(r + ny)/rI2'"
= (6-14)
where n = 1.6 + 0.35 cot CY and y equals the water depth above the weir crest, which
approximates 0.7 H, provided that the approach velocity is negligible. For a weir with
a 1-to-1 sloping downstream face (cot CY = 1) the minimum pressure at the weir crest
in metres water column (P/pg),i, with regard to the energy head HI is given as a func-
tion of the ratio h,/r in Figure 6.23. To avoid the danger of local cavitation, the mini-
mum pressure at the weir crest should be limited to -4 m water column. This limitation,

2 to3 HI max
I
1
I

Figure 6.22 The cylindrical crested weir

202
 (PIPq I min.
H.
10

0.0

-10

-u)

-30

-4.0
O 1 2 3 4 5 6 7 8 9 10
RATIO Hl/'

Figure 6.23 Minimum pressure at cylindrical weir crest as a function of the ratio H ,/r

together with the maximum energy head over the weir crest, will give a limitation
on the ratio H,/r which can be obtained from Figure 6.23.
To allow the cylindrical-crested weir to function as a high capacity overflow weir,
the crest height above the downstream channel bed should be such that this channel
bed does not interfere with the formation of the overflowing nappe. Therefore the
ratio pJH, should not be less than unity.

6.7.2 Evaluation of discharge

The basic head-discharge equation for a short-crested weir with a rectangular control
section reads, according to Section 1.10
2 2
Q = Ce 3&g b, (6- 15 )

where the effective discharge coefficient Ceequals the product of Co(which is a function
of H,/r), of C, (which is a function of p,/H,) and of C2 (which is a function of p,/H,
and the slope of the upstream weir face) (Ce = C,C,C,). The basic discharge coefficient
is a function of the ratio H,/r and has a maximum value of Co = 1.49 if H,/r exceeds
5.0 as shown in Figure 6.24.

The Co-values in Figure 6.24 are valid if the weir crest is sufficiently high above the
average bed of the approach channel (p,/Hl > about 1.5). If, on the other hand, p,
approaches zero, the weir will perform as a broad-crested weir and have a Ce-value
of about 0.98, which corresponds with a discharge coefficient reduction factor, C,,
of 0.98/1.49 N 0.66. Values of the reduction factor as a function of the ratio p,/H,
can be read from Figure 6.25.
No results of laboratory tests on the influence of an upstream sloping weir face

203
are available. It may be expected, however, that the correction factor on the basic
discharge coefficient, C2, will be about equal to those given in Figure 6.20 for WES-
spillway shapes.

DISCHARGE COEFFICIENT Cd

l.!

1A

1.3

1.2

1.1

74
1.c

RATO Hllv
x r-0.025m , G.O. MATTHEW ,1963
0 r i ...... , AL. VERWOERD,1941
+ r :0.030m , W. J.v.d.OOR0 ,1941
0.7 * r=0.0375m., L.ESCANOEond
CI rz0.075 , F SANANES ,1959

Figure 6.24 Discharge coefficient for cylindrical crested weir as a function of the ratio H,/r

DISCHARGE COEFFICIENT
REWCTION FACTOR C 1
p l l H l r 1.5
1.c

O.!

aa

+DATA FROM W.J.ud.OORD.1941

a7

2.0 3.0
RATIO pl/Hl

Figure 6.25 Reduction factor C, as a function of the ratio pl/Hl

204
 -(H~+PI)/Yc
1.5 20 25 3.0 3.5 4.0 4.5
0.0

$05

1
aio

0.15

valid tor rectangular approach channel

O20

Q30

O45

Figure 6.26 Graph for the conversion of H I into h, (after Van der Oord 1941)

SUBMERGEKE RATO %Hl

DROWNED FLOW REDUCTION FACTOR f

Figure 6.27 Drowned flow reduction factor as a function of H2/Hl

205
 For each energy head over the weir crest, a matching discharge can be calculated
with the available data, resulting in a Q-Hl curve. With the aid of Figure 6.26, this
Q-H, relationship can be changed rather simply into a Q-h, relationship. For each
value of the ratio (Hl + pl)/yc a corresponding value of (v12/2g)/yccan be obtained,
where ycis the critical depth over the weir crest, so that h, = Hl - vI2/2gcan be calcula-
ted.
If we define the modular limit as that submergence ratio H2/H, which produces
a 1% reduction from the equivalent discharge (f = 0.99), we see in Figure 6.27 that
the modular limit equals about 0.33. Values of the drowned flow reduction factor
as a function of the submergence ratio can be obtained from Figure 6.27.

The accuracy of the effective discharge coefficient of a well-maintained cylindrical-

crested weir which has been constructed with reasonable care and skill will be sufficient
for field conditions. It can be expected that the error of C, = CoC,C2 will be less
than 5%. The method by which this error is to be combined with other sources of
error is shown in Annex 2.

6.7.3 Limits of application

For reasonable accuracy, the limits of application of a cylindrical-crested weir are:

a. The upstream head over the weir crest h, should be measured a distance of 2 to
3 times hlmaxupstream from the weir face. The recommended lower limit of
h, = 0.06m;
b. To prevent water surface instability in the approach channel, the ratio pl/hl should
not be less than 0.33;
c. To reduce the boundary layer effects of the vertical side walls, the ratio bJH, should
not be less than 2.0;
d: On high head installations, the ratio h,/r should be such that the local pressure
at the crest is not less than 4 m water column;
e. To prevent the tailwater channel bottom from influencing the flow pattern over
the weir, the ratio p2/HIshould not be less than unity;
f. The modular limit H2/H, = 0.33.

6.8 Selected list of references

Bazin, H.E. 1896. Expriences nouvelles sur Icoulement en dversoir. Annales des Ponts et Chausses.
Vol. 7. pp. 249-357.
British Standards Institution. 1969. Methods of measurement of Liquid flow in open channels. British
Standard 3680, Part 4B, Long base weirs. 39 pp.
Butcher, A.D. 1923. Submerged weirs and standing wave weirs. Min. of Public Works, Cairo. 17 pp.
Butcher, A.D. 1921/22. Clear overfall weirs. Res. Work Delta Barrage. Min. of Public Works, Cairo.
Chow, Ven Te 1959. Open-Channel Hydraulics. McGraw-Hill Book Company Inc., New York. 680 pp.
Crump, E.S. 1952. A new method of gauging stream flow with little afflux by means of a submerged weir
of triangular profile. Proc. Inst. Civil Engrs., Part I , Vol. 1. pp. 223-242.
Dillman, O. 1933. Untersuchungen an berfiillen. Mitt. des Hydr.lnst. der Tech. Hochschule Mnchen.
NO. 7. pp. 26-52.
Escande, L. and F. Sananes 1959. Etude des seuils dversants fente aspiratrice. La Houille Blanche, 14
No. B, Dec. Grenoble. pp. 892-902.

206
Leliavsky, S. 1965. Irrigation Engineering: Syphons, Weirs and Locks. Vol.11. Chapman & Hall Ltd., Lon-
don. 296 pp.
Matthew, G.D. 1962.On the influence ofcurvature, surface tension and viscosity on flow over round-crested
weirs. Proc. Inst. Civil Engrs., Vo1.25. pp. 51 1-524.
Oord, W.J. van der 1941. Stuw met cirkelvormige kruin. MSc Thesis. Techn. University Delft.
Rouse, H. and L. Reid. 1935. Model research on spillway crests. Civil Eng. Vol. 5 . January, p. IO.
U.S. Army Corps of Engineers 1952. Corps of Engineers Hydraulic Design Criteria. Prepared for Office
of the Chief of Engineers. Waterways Experimental Station Vicksburg, Miss. Revised in subsequent years.
U.S. Bureau of Reclamation 1960. Design of small dams. USBR Denver 61 1 pp.
U.S. Bureau of Reclamation 1948. Studies of crests for overfall dams. Boulder Canyon Project Final Re-
ports. Part VI. USBR Denver, Hydraulic Investigations, Bull. 3.
U.S. Department of Agriculture 1962. Field manual for research in agricultural hydrology. Agric. Hand-
book No. 224, Washington. 2 15 pp.
Verwoerd, A.L. 1941. Capaciteitsbepalingvan volkomen en onvolkomen overlaten met afgeronde kruinen.
Waterstaatsingenieur in Nederlandsch-Indi. No. 7. pp. 65-78 (11).
Vlugter, H. 1932. De volkomen overlaat. (Modular weirs) De Waterstaats Ingenieur, No.4, Bandung,
Wallingford Hydraulic Research Station 1970. The triangular profile Crump weir. Effects of a bend in
the approach channel. Report EX 518. W.HRS, England.
Water Resources Board. 1970. Crump Weir Design. Reading Bridge House, England. T N 8 (rev.), 92 pp.
White, W.R. and J.S. Burgess 1967. Triangular profile weir with 1.2 upstream and downstream slopes.
Hydr. Res. Sta. Wallingford, England. Rep. No. INT 64, pp. 41-47.
White, W.R. 1968. The flat vee weir. Water & Water Eng. V 72 No. 863, pp. 13-19.
White, W.R. 1971. The performance of two-dimensional and flat-V triangular profile weirs. Proc. Inst.
Civil Engrs. Suppl. (ii), Paper 7350 S. 48 pp.

207
7 Flumes
A critical depth-flume is essentially a geometrically specified constriction built in an
open channel where sufficient fall is available for critical flow to occur in the throat
of the flume. Flumes are in-line structures, i.e. their centre line coincides with the
centre line of the undivided channel in which the flow is to be measured. The flume
cannot be used in structures like turnouts, controls and other regulating devices.
In this chapter the following types of critical-depth flumes will be described: Long-
throated flumes (7. I), Throatless flumes with rounded transition (7.2), Throatless
flumes with broken plane transition (7.3), Parshall flumes (7.4), H-flumes (7.5). The
name Venturi flume is not used in this chapter, since this term is reserved for flumes
in which flow in the constriction is sub-critical. The discharge through such a constric-
tion can be calculated by use of the equations presented in Section 1.7.

7.1 Long-throated flumes

7.1.1 Description

Classified under the term long-throated flumes are those structures which have a
throat section in which the streamlines run parallel to each other at least over a short
distance. Because of this, hydrostatic pressure distribution can be assumed at the con-
trol section. This assumption allowed the various head-discharge equations to be de-
rived, but the reader should note that discharge coefficients are also presented for
high H,/L ratios when the streamlines at the control are curved.
The flume comprises a throat of which the bottom (invert) is truly horizontal in
the direction of flow. The crest level of the throat should not be lower than the dead
water level in the channel, i.e. the water level downstream at zero flow. The throat
section is prismatic but the shape of the flume cross-section is rather arbitrary, pro-
vided that no horizontal planes, or planes that are nearly so, occur in the throat above
crest (invert) level, since this will cause a discontinuity in the head-discharge relation-
ship. Treated in this section will be the most common flumes, i.e. those with a rectangu-
lar, V-shaped, trapezodal, truncated V, parabolic, or circular throat cross-section.
For other shapes see Bos (1985).
The entrance transition should be of sufficient length, so that no flow separation
can occur either at the bottom or at the sides of the transition. The transition can
be formed of elliptical, cylindrical, or plane surfaces. For easy construction, a transi-
tion formed of either cylindrical or plane surfaces, or a combination of both, is recom-
mended. If cylindrical surfaces are used, their axes should be parallel to the planes
of the throat and should lie in the cross-section through the entrance of the throat.
Their radii should preferably be about 2 Hlmax. With a plane surfaced transition, the
convergence of side walls and bottom should be about 1.3. According to Wells &
Gotaas (1956) and Bos & Reinink (1 98 l), minor changes in the slope of the entrance
transition will have no effect upon the accuracy of the flume. It is suggested that,
where the flume has a bottom contraction or hump, the transitions for the crest and
for the sides should be of equal lengths, i.e. the bottom and side contraction should
begin at the same point at the approach channel bottom as shown in Figure 7. I .

209
 "1 section A-A sectional view B-B

r B

RECTANGULAR FLUME

section A-A sectional view B-B

throat downsTream
section expansion
m

L B TRAPEZOIDAL FLUME

Figure 7. I Alternative examples of flume lay-out

With flat bottomed flumes, the floor of the entrance transition and of the approach
channel should be flat and level and at no point higher than the invert of the throat,
up to a distance 1 .O Hlmax
upstream of the head measurement station. This head mea-
surement station should be located upstream of the flume at a distance equal to be-

210
tween 2 and 3 times the maximum head to be measured.
Even if a flume is fitted with a curved entrance transition, it is recommended that
the downstream expansion beyond the throat be constructed of plane surfaces. The
degree of expansion influences the loss of energy head over the expansion and thus
the modular limit of the flume (Section 1.15).

7.1.2 Evaluation of discharge

The basic stage-discharge equations for long-throated flumes with various control
sections have been derived in Section 1.9 and are shown in Fig.7.2. As indicated, the
reader should use Table 7.1 to find y,-values for a trapezoidal flume, and Table 7.2
to find the ratios A,/d,2 and y,/dc as a function of HJd, for circular flumes.
For all control sections shown, the discharge coefficient Cd is a function of the ratio
H,/L and is presented in Figure 7.3. The approach velocity coefficient C, may be read
from Figure 1.12 as a function of the dimensionless ratio CdA*/AI.
The error in the product cdc, of a well maintained long-throated flume which has
been constructed with reasonable care and skill may be deduced from the equation
X, (3 I Hl/L-0.55 1 + 4)
= (7- 1)
The method by which this coefficient error is to be combined with other sources of
error is shown in Annex 2.

Photo 1 Long-throated flumes can be portable

21 1
 HEAD-DISCHARGE HOW TO FIND
EQ. TO BE USED THEY,-VALUE

Q = CdCV 3 (3g) b,h:/z

1/ 2

Use Table 3.1

Use Table 7.2

I f H1 < 0.70 d,
Q = CddFfi [f(S)]

use table 7.2 to find f(Bl Use Table 7.2

I f H1 b 0.70 d,
Q=C c 2 ( 2 g )1/2 dc(h~-0.1073d,) 3/2 y, = $ Hl + 0.0358dC
d v3 3

Figure 7.2 Head-discharge relationship for long-throated flumes (from Bos 1985)

212
 I
Cd value
1.16

21 3
Table 7.1Values of the ratio yc/HI as a function of zc and Hl/bcfor trapezodal control sections
~

Side slopes of channel, ratio of horizontal to vertical (zc:I )

Hl/b, Vertical 0.25:l 0.5O:l 0.75:l 1:l 1.5:1 2:1 2.5:l 3:1 4:l

.o0 ,667 ,667 ,667 ,667 ,667 ,667 ,667 ,667 ,667 ,667
.o 1 .667 ,667 ,667 .668 ,668 ,669 ,670 ,670 ,671 ,672
.o2 .667 ,667 .668 ,669 ,670 ,671 ,672 ,674 .675 .678
.O3 ,667 ,668 ,669 .670 .67I ,673 ,675 ,677 ,679 ,683
.O4 ,667 ,668 ,670 .67I .672 ,675 ,677 ,680 ,683 ,687
.o5 ,667 ,668 ,670 ,672 ,674 ,677 ,680 ,683 ,686 .692
.O6 ,667 .669 ,671 ,673 ,675 ,679 ,683 ,686 ,690 ,696
.O7 ,667 ,669 ,672 ,674 ,676 ,681 ,685 ,689 ,693 ,699
.O8 ,667 ,670 ,672 ,675 .678 ,683 ,687 ,692 ,696 ,703
.O9 ,667 ,670 ,673 ,676 ,679 ,684 ,690 ,695 ,698 ,706
.IO .661 ,670 .674 .677 ,680 ,686 ,692 ,697 .701 .709
.I2 .667 .671 ,675 ,679 ,684 ,690 ,696 .701 .706 .7I5
.I4 .667 .672 ,676 .681 ,686 ,693 ,699 ,705 ,711 ,720
.I6 ,667 ,672 ,678 ,683 .687 ,696 ,703 ,709 ,715 ,725
.I8 ,667 ,673 ,679 ,684 .690 .698 ,706 ,713 ,719 ,729
.20 ,667 ,674 ,680 ,686 692 .701 ,709 ,717 ,723 ,733
.22 ,661 ,674 .681 ,688 ,694 ,704 .7I2 ,720 ,726 ,736
.24 ,667 ,675 ,683 .689 ,696 ,706 .7I5 ,723 ,729 .739
.26 ,667 ,676 ,684 ,691 ,698 ,709 .7I8 ,725 ,732 ,742
.28 ,667 ,676 ,685 ,693 ,699 ,711 ,720 ,728 ,734 ,744
.30 ,667 ,677 .686 ,694 ,701 ,713 ,723 ,730 .737 .747
.32 ,667 ,678 .687 ,696 ,703 ,715 ,725 .733 .739 .749
.34 ,667 ,678 ,689 ,697 .705 ,717 ,727 ,735 ,741 ,751
.36 ,667 ,679 ,690 .699 ,706 ,719 ,729 ,737 ,743 .752
.38 .667 ,680 ,691 ,700 ,708 ,721 .73I ,738 ,745 .754
.40 ,667 ,680 ,692 .701 .709 ,723 ,733 ,740 ,747 ,756
.42 ,667 ,681 ,693 ,703 .711 .725 ,734 ,742 ,748 ,757
.44 ,667 ,681 ,694 ,704 ,712 ,727 .736 ,744 ,750 ,759
.46 .667 .682 .695 ,705 ,714 ,728 .737 ,745 ,751 ,760
.48 ,667 ,683 ,696 ,706 .715 .729 ,739 ,747 ,752 ,761
.5 ,667 ,683 ,697 ,708 ,717 ,730 ,740 .748 .754 .762
.6 ,667 ,686 ,701 ,713 ,723 ,737 ,747 ,754 ,759 ,767
.7 ,667 ,688 .706 .718 ,728 ,742 .752 ,758 ,764 .771
.8 ,667 ,692 ,709 .723 .732 ,746 ,756 ,762 ,767 ,774
.9 ,667 ,694 .7I3 ,727 ,737 ,750 ,759 ,766 ,770 ,776
I .o .667 ,697 .7I7 ,730 ,740 ,754 .762 .768 ,773 ,778
1.2 ,667 ,701 ,723 .737 ,747 ,759 ,767 ,772 ,776 ,782
1.4 ,667 ,706 .729 ,742 ,752 ,764 ,771 ,776 ,779 .784
I .6 ,667 ,709 ,733 ,747 .756 ,767 ,774 ,778 .781 ,786
1.8 ,667 ,713 ,737 ,750 ,759 ,770 .776 .781 ,783 .787
2 ,667 ,717 ,740 .754 ,762 ,773 ,778 ,782 ,785 ,788
3 ,667 ,730 ,753 ,766 .773 .%I ,785 ,787 ,790 .792
4 ,667 ,740 ,762 ,773 ,778 .785 .788 .790 ,792 ,794
5 ,667 ,748 ,768 ,777 ,782 .788 .79I ,792 ,794 ,795
10 ,667 ,768 ,782 .788 .791 .794 .795 .796 ,797 ,798

03 ,800 ,800 ,800 ,800 ,800 200 300 ,800 300

214
Table 7.2 Ratios for determining the discharge Q of a broad-crested weir and long-throated flume with
circular section (Bos 1985)

Ycldc v,2/2gdc Hlldc Ac/d,2 YclHI f(e) Yelde vc2/2gdc Hddc Ac/d,2 YJHI f(Q)

.o1 .O033 .O I33 .O0 13 ,752 0.0001 .5 I ,2014 .71 I4 .4027 ,717 0.2556
.o2 ,0067 ,0267 ,0037 ,749 0.0004 .52 .2065 ,7265 .4 I27 ,716 0.2652
.O3 .o101 .O40 I .O069 ,749 0.00 I o .53 .21 I7 ,7417 ,4227 .715 ,0.2750
.O4 .O I34 ,0534 ,0105 ,749 0.00 I7 .54 .2 I70 ,7570 .4327 ,713 0.2851
.O5 .O 168 ,0668 ,0147 ,748 0.0027 .55 .2224 ,7724 ,4426 ,712 0.2952
.O6 ,0203 ,0803 .O 192 ,748 0.0039 .56 ,2279 I ,7879 ,4526 .711 0.2952
.O7 ,0237 ,0937 .O242 ,747 0.0053 .57 ,2335 ,8035 .4625 ,709 0.3161
.O8 ,0271 ,1071 ,0294 ,747 0.0068 .58 ,2393 ,8193 ,4724 ,708 0.3268
.O9 ,0306 .I206 ,0350 ,746 0.0087 .59 ,245 1 ,8351 .4822 ,707 0.3376
.IO ,0341 ,1341 ,0409 ,746 0.0107 .60 ,251 1 ,851 1 ,4920 ,705 0.3487
.I I ,0376 .I476 ,0470 .745 0.0 I29 .6 1 ,2572' ,8672 SO18 ,703 0.3599
.I2 .O41 I ,161 I ,0534 ,745 0.0153 .62 ,2635 3835 ,5115 ,702 0.3713
.I3 ,0446 ,1746 ,0600 ,745 0.0 I79 .63 ,2699 ,8999 ,5212 ,700 0.3829
.I4 ,0482 .I882 ,0688 ,744 0.0214 .64 ,2765 .9 I65 ,5308 ,698 0.3947
.I5 .O5 17 ,2017 .O739 ,744 0.0238 .65 ,2833 ,9333 ,5404 ,696 0.4068
.16 ,0553 ,2153 .O81 I ,743 0.0270 .66 .2902 ,9502 ,5499 ,695 0.4189
.I7 ,0589 ,2289 .O885 ,743 0.0304 .67 ,2974 ,9674 ,5594 ,693 0.43 I4
.I8 ,0626 .2426 ,0961 .742 0.0340 .68 ,3048 ,9848 ,5687 .69 1 0.4440
.I9 ,0662 .2562 ,1039 .742 0.0378 .69 ,3125 1.0025 ,5780 ,688 0.4569
.20 ,0699 ,2699 . I I I8 ,741 0.0418 .70 .3204 1.0204 ,5872 .686 0.4701
.21 ,0736 ,2836 ,1199 ,740 0.0460 .7 1 ,3286 1.0386 ,5964 ,684 0.4835
.22 .O773 ,2973 .I281 ,740 0.0504 .72 .3371 1.0571 ,6054 .68 I 0.4971
.23 .O8 11 .31 I 1 ,1365 .739 0.0550 .73 ,3459 1.O759 .6 I43 ,679 0.5109
.24 ,0848 ,3248 ,1449 ,739 0.0597 .74 ,3552 1.0952 ,623 I ,676 0.5252
.25 ,0887 ,3387 ,1535 .738 0.0647 .75 ,3648 1.1148 ,6319 ,673 0.5397
.26 ,0925 ,3525 ,1623 ,738 0.0698 .76 ,3749 1.1349 ,6405 ,670 0.5546
.27 ,0963 ,3663 ,1711 ,737 0.075 I .77 ,3855 1.1555 .6489 ,666 0.5698
.28 ,1002 .3802 . I800 ,736 0.0806 .78 ,3967 1.1767 ,6573 ,663 0.5855
.29 ,1042 ,3942 ,1890 ,736 0.0863 .79 ,4085 1.1985 ,6655 .659 0.6015
.30 ,1081 ,408 1 ,1982 ,735 0.0922 .80 .42 1O 1.2210 ,6735 .655 0.6180
.3 I . I 121 .422 I ,2074 ,734 0.0982 .8 I ,4343 1.2443 .68 15 ,651 0.6351
.32 ,1161 ,4361 .2 I67 ,734 0.1044 32 ,4485 1.2685 ,6893 .646 0.6528
.33 ,1202 ,4502 ,2260 ,733 0.1 108 33 ,4638 1.2938 .6969 .64 1 0.6712
.34 . I243 ,4643 ,2355 ,732 0.1 174 .84 ,4803 1.3203 .7043 ,636 0.6903
.35 .I284 .4784 ,2450 ,732 0.1289 .85 .4982 1.3482 ,7115 ,630 0.7102
.36 .I326 ,4926 ,2546 .731 0.131 1 .86 ,5177 1.3777 .7 I86 ,624 0.73 12
.37 ,1368 ,5068 ,2642 ,730 O. 1382 .87 ,5392 1.4092 ,7254 ,617 0.7533
.38 ,1411 ,521 I ,2739 .729 O. I455 .88 ,5632 1.4432 .7320 .610 0.7769
.39 . I454 ,5354 ,2836 ,728 O. 1529 .89 ,5900 1.4800 .7384 .60 I 0.8021
.40 . I497 ,5497 .2934 ,728 O. 1605 .90 ,6204 1.5204 .7445 ,592 0.8293
.4 1 ,1541 ,5641 .3032 ,727 0.1683 .9 1 ,6555 1.5655 ,7504 ,581 0.8592
.42 . I586 ,5786 .3130 ,726 O. I763 .92 ,6966 1.6166 .7560 ,569 0.8923
.43 ,1631 ,5931 ,3229 .725 O. I844 .93 ,7459 1.6759 .7612 .555 0.9297
.44 . I676 ,6076 .3328 ,724 O. I927 .94 ,8065 1.7465 .7662 ,538 0.973 1
.45 . I723 ,6223 ,3428 ,723 0.2012 .95 ,8841 1.8341 ,7707 .5 18 1.0248
.46 .I769 ,6369 ,3527 ,722 0.2098
.47 ,1817 ,6517 .3627 ,721 0.2186
.48 ,1865 ,6665 ,3727 ,720 0.2276
.49 .I914 .68 14 ,3827 ,719 0.2368
.50 ,1964 .6964 ,3927 ,718 0.2461
7.1.3 Modular limit

The modular limit of flumes greatly depends on the shape of the downstream expan-
sion. The relation between the modular limit and the angle of expansion, can be ob-
tained from Section l . 15. Practice varies between very gentle and costly expansions
of about 1-to-15, to ensure a high modular limit, and short expansions of 1-to-6. It
is recommended that the divergences of each plane surface be not more abrupt than
I-to-6. If in some circumstances it is desirable to construct a short downstream expan-
sion, it is better to truncate the transition rather than to enlarge the angle of divergence
(see also Figure 1.35). At one extreme if no velocity head needs to be recovered, the
downstream transition can be fully truncated. It will be clear from Section 1.15 that
no expanding section will be needed if the tailwater level is always less than yc above
the invert of the flume throat.
At the other extreme, when almost all velocity head needs to be recovered, a transi-
tion with a gradual expansion of sides and bed is required. The modular limit of long-
throated flumes with various control cross sections and downstream expansions can
be estimated with the aid of Section 1.15.
As an example, we shall estimate the modular limit of the flume shown in Figure
7.4, flowing under an upstream head h, = 0.20 m at a flow rate of Q = 0.0443 m3/s.
The required head loss Ah over the flume, and the modular limit H,/H, are determined
as follows
a. Cross-sectional area of flow at station where h, is measured equals
A, = b,y, + z,y,, = 0.75 x 0.35 + 1.0 x 0.35, = 0.385m2
v, = Q/A, = 0.0443/0.385 = O. 115 m/s;
b. The upstream sill-referenced energy head equals
H, = h, + vI2/2g= 0.20 + 0.1 15,/(2 x 9.81) = 0.201 m;
c. The discharge coefficient C d = 0.964;
d. The exponent u = 1.50 (rectangular control section);
e. cd = 0.964 = 0.976;
f. For a rectangular control section yc = 2/3 Hl = O. 134 m;
g. The average velocity at the control section is

v =-=
Q 0.0443
= 1.110m/s
ycb, 0.134 x 0.30
h. With the 1-to-6 expansion ratio the value of 6 equals 0.66;
i. We tentatively estimate the modular limit at about 0.80. Hence, the related h,-value
is 0.80 x 0.20 = 0.16 m. Further
A, = b2y2+ z,y,2 = 0.3 13 m2
v2 = Q/A, = 0.141 m/s
j. ~ ( ~ , - v , ) ~ / 2 g H
= ,0.66(l.110-0.141)2/(2 x 9.81 x 0.201) = 0.157;
k. The energy losses due to friction downstream from the control section can be found
by applying the Manning equation with the appropriate n-value to L/3 = 0.20 m
of the throat, to the downstream transition length, Ld = 0.90 m, and to the canal

216
 I I

I - - I
L=0.60 & 0.90

I I I \ \
i
I
,rI I I I I I

L B C 4

sectional view B-B sectional view C-C

Figure 7.4 Long-throated flume dimensions (example)

up to the h, measurement section. The latter length equals (Bos 1984)

Le = IO (pl + L/2) - L, = 10 (O. 15 + 0.30) - 0.90 = 3.60 m
Using a Manning n-value of 0.016 for the concrete flume and canal the friction
losses are
L nv,
= 3(F)
AHthroa, = 0.00239 m

AH,,,,, = Ld p,,r+;?r = 0.00057 m

AHcana,
= = 0.00016m

217
 Hence AH, N 0.003 m. It should be noted that for low h,-values and relatively
long transitions, the value of AH, becomes significantly more important. The value
of AH, is relatively insensitive for minor changes of the tailwater depth y,. Hence,
for a subsequent pass through this step in the procedure the same AH,-value may
be used;
I. Calculate AHdH, = 0.003/0.201 = 0.015;
m. The downstream sill-referenced energy head at the tailwater depth used at Step
i equals
H, = h, + v,2/2g = 0.16 + 0.14,/(2 x 9.81) = 0.161 m
n. The ratio H,/H, equals then 0.801;
o. Substitution of the values of steps e, j, I, and n into Equation 1.125 gives at modular
limit H,/H,
0.801 = 0.976-0.015-0.157 = 0.804
which is almost true. Hence, h , - h, = 0.04 m for this flume if h, = 0.20 m.
Once some experience has been acquired a close match of Equation 1.125 can be ob-
tained in two to three iterations. Since the modular limit varies with the upstream
head, it is advisable to estimate the modular limit at both minimum and maximum
anticipated flow rates and to check if sufficient head loss is available.
The computer program FLUME (Clemmens et al. 1987) calculates the modular limit
and head loss requirement for broad-crested weirs and long-throated flumes.

7.1.4 Limits of application

The limits of application of a long-throated flume for reasonably accurate flow

measurements are:
a. The practical lower limit of h, is related to the magnitude of the influence of fluid
properties, boundary roughness, and the accuracy with which h, can be determined.
The recommended lower limit is 0.07 L;
b. To prevent water surface instability in the approach channel the Froude number
Fr = vI/(gAI/Bl)'/2should not exceed 0.5;
c . The upper limitation on the ratio H,/L arises from the necessity to prevent stream-
line curvature in the flume throat. Values of the ratio H,/L should be less than
1.o;
d. The width B, of the water surface in the throat at maximum stage should not be
less than L/5;
e. The width at the water surface in a triangular throat at minimum stage should
not be less than 0.20 m.

7.2 Throatless flumes with rounded transition

7.2.1 Description

Throatless flumes may be regarded as shorter, and thus cheaper, variants of the long-
throated flumes described in Section 7.1. Although their construction costs are lower,

218
Photo 2 Throatles flume with rounded transaction

throatless flumes have a number of disadvantages, compared with long-throated

flumes. These are:
- The discharge coefficient Cd is rather strongly influenced by H, and because of

streamline curvature at the control section also by the shape of the downstream
transition and by H,;
- The modular limit varies with H, and has a lower value;
- The control section can only be rectangular;

- In general, the Cd-value has a rather high error of about 8 percent.

Two basic types of throatless flumes exist, one having a rounded transition between
the converging section and the downstream expansion, and the other an abrupt (broken
plane) transition. The first type is described in this section, the second in Section 7.3.

A throatless flume with rounded transition is shown in Figure 7.5. In contradiction

to its shape, the flow pattern at the control section of such a flume is rather complicated
and cannot be handled by theory. Curvature of the streamlines is three-dimensional,
and a function of such variables as the contraction ratio and curvature of the side
walls, shape of any bottom hump if present, shape of the downstream expansion, and

219
 \ Ratio of side contraction: b,/b,

Figure 7.5 The throatless flume

the energy heads on both ends of the flume. Laboratory data on throatless flumes
are insufficient to determine the discharge coefficient as a function of any one ofthe
above parameters.
The Figure 7.6 illustrates the variations in Cd.Laboratory data from various investi-
gators are so divergent that the influence of parameters other than the ratio H,/R
is evident.

7.2.2 Evaluation of discharge

The basic head-discharge equation for flumes with a rectangular control section equals
2 2
Q = CdCV3&g b, h,3/2 (7-2)

From the previous section it will be clear that a Cd-valuecan only be given if we intro-
duce some standard flume design. We therefore propose the following:
- The radius of the upstream wing walls, R, and the radius, Rb, of the bottom hump,

if any, ranges between 1.5 Hlmax and 2.0 Hlmax;

- The angle of divergence of the side walls and the bed slope should range between
1-to-6 and I-to-10. Plane surface transitions only should be used;
- If the downstream expansion is to be truncated, its length should not be less than
1.5(B2-b,),where B, is the average width of the tailwater channel.

220
DISCHARGE COEFFICIENT Cd
I

I A
A
1 0 Khafagi flat bottom
Blau (Karlshorst) flat bottem
A Wou (Potsdam) rounded hump
( H ~ / R b ) ~ ~ ~ = 0also
. 2 5at
downstream transitlon
A Blou (Karlshorst) hump upstream
rounded (H1/Rb)max=+0.64
downstream 1 :10
I I

O 05 1.o 1.5 2.0

RATIO H1/R

Figure 7.6 Cd-values for various throatless flumes

If this standard design is used, the discharge coefficient Cd equals about unity. The
appropriate value of the approach velocity coefficient, C,, can be read from Figure
1.12 (Chapter 1).
Even for a well-maintained throatless flume which has been constructed with rea-
sonable care and skill, the error in the above indicated product cdc, is rather high,
and can be expected to be about 8 percent. The method by which this coefficient error
is to be combined with other sources of error is shown in Annex 2.

7.2.3 Modular limit

Investigating the modular limit characteristics of throatless flumes is a complex pro-

blem and our present knowledge is limited. Tests to date only scratch the surface of
the problem, and are presented here mainly to illustrate the difficulties. Even if we
take the simplest case of a flume with a flat bottom, the plot of H2/H, versus H,/b,,
presented in Figure 7.7 shows unpredictable variation of the modular limit for different
angles of divergence and expansion ratios b,/B2.
It may be noted that Khafagi (1942) measured a decrease of modular limit with
increasing expansion ratio b,/B, for 1-to-8 and 1-to-20 flare angles. For long-throated
flumes this tendency would be reversed and in fact Figure 7.7 shows this reversed

22 1
MODULAR LIMIT H2/H1
1.o

0.9

0.8

0.7

Figure 7.7 Modular limit conditions of flat bottomed throatless flumes (after Khafagi 1942)

trend for a 1-to-6 flare angle. The modular limits shown in Figure 7.7 are not very
favourable if we compare them with long-throated flumes having the same b,/B, ratio
and an abrupt (a = 180) downstream expansion. The modular limit of the latter
equal 0.70 if bc/B2.= 0.4 and 0.75 if b,/B, = 0.5.
The variation in modular limit mentioned by Khafagi is also present in data reported
by Blau (1960). Blau reports the lowest modular limit for throatless flumes, which
equals 0.5; for Hl/bc = 0.41, AJA, = 0.21, bJB, = 0.49, wingwall divergence and
bed slope both I-to- 1 O.
There seems little correlation between the available data, which would indicate that
the throatless flume is not a suitable modular discharge measurement structure if the
ratio H,/H, exceeds about 0.5.

7.2.4 Limits of application

The limits of application of a throatless flume with rounded transition for reasonably
accurate flow measurements are:
a. Flume design should be in accordance with the standards presented in Section 7.2.2;
b. The practical lower limit of h, depends on the influence of fluid properties, boun-
dary roughness, and the accuracy with which is h, can be determined. The recom-
mended lower limit is 0.06 m;
c. To prevent water surface instability in the approach channel the Froude number
Fr = v,/(gA,/B,)~ should not exceed 0.5;
d. The width b, of the flume throat should not be less than 0.20 m nor less than HI,,,.

222
7.3 Throatless flumes with broken plane transition
7.3.1 Description

The geometry of the throatless flume with broken plane transition was first developed
in irrigation practice in the Punjab and as such is described by Harvey (1912). Later,
Blau (1960) reports on two geometries of this flume type. Both sources relate discharge
and modular limit to heads upstream and downstream of the flume, h, and h, respecti-
vely. Available data are not sufficient to warrant inclusion in this manual.
Since 1967 Skogerboe et al. have published a number of papers on the same flume,
referring to it as the cutthroat flume. In the cutthroat flume, however, the flume
discharge and modular limit are related to the piezometric heads at two points, in
the converging section (ha)and in the downstream expansion (hb)as with the Parshall
flume. Cutthroat flumes have been tested with a flat bottom only. A dimension sketch
of this structure is shown in Figure 7.8.
Because of gaps in the research performed on cutthroat flumes, reliable head-
discharge data are only available for one of the tested geometries (b, = 0.305 m, overall
length is 2.743 m). Because of the non-availability of discharge data as a function
of hl and h, (or H l and H,) the required loss of head over the flume to maintain modu-
larity is difficult to determine.
In the original cutthroat flume design, various discharge capacities were obtained
by simply changing the throat width b,. Flumes with a throat width of I , 2, 3,4, 5,
and 6 feet (1 ft = 0.3048 m) were tested for heads ha ranging from 0.06 to 0.76 m.
All flumes were placed in a rectangular channel 2.44 m wide. The upstream wingwall
had an abrupt transition to this channel as shown in Figure 7.8.
Obviously, the flow pattern at the upstream piezometer tap is influenced by the
ratio b,/B,. Eggleston (1967) reports on this influence for a 0.3048 m wide flume. A
variation of discharge at constant ha up to 2 percent was found. We expect, however,
that this variation will increase with increasing width b, and upstream head. Owing
to the changing entrance conditions it even is possible that the piezometer tap for

2.743

Figure 7.8 Cutthroat flume dimensions (after Skogerboe et al. 1967)

223
measuring ha will be in a zone of flow separation. As already mentioned in Section
7.2.3, the ratios b,/B, and bJL, are also expected to influence the head-discharge rela-
tionship.
Bennett (1972) calibrated a number of cutthroat flumes having other overall lengths
than 2.743 m. He reported large scale effects between geometrically identicalcutthroat
flumes, each of them having sufficiently large dimensions (b, ranged from 0.05 to 0.305
m). Those scale effects were also mentioned by Eggleston (1967), Skogerboe and Hyatt
(1969), and Skogerboe, Bennett, and Walker (1972). In all cases, however, the reported
large scale effects are attributed to the improper procedure of comparing measure-
ments with extrapolated relations. As a consequence of the foregoing, no head-dis-
charge relations of cutthroat flumes are given here. Because of their complex hydraulic
behaviour, the use of cutthroat flumes is not recommended by the present writers.

7.4 Parshall flumes

7.4.1 Description

Parshall flumes are calibrated devices for the measurement of water in open channels.
They were developed by Parshall (1922) after whom the device was named. The flume
consists of a converging section with a level floor, a throat section with a downward
sloping floor, and a diverging section with an upward sloping floor. Because of this
unconventional design, the control section of the flume is not situated in the throat
but near the end of the level crest in the converging section. The modular limit of
the Parshall flume is lower than that of the other long-throated flumes described in
Section 7.1.
In deviation from the general rule for long-throated flumes where the upstream
head must be measured in the approach channel, Parshall flumes are calibrated against
a piezometric head, ha, measured at a prescribed location in the converging section.
The downstream piezometric head h, is measured in the throat. This typical American
practice is also used in the cutthroat and H-flumes.

Parshall flumes were developed in various sizes, the dimensions of which are given
in Table 7.3. Care must be taken to construct the flumes exactly in accordance with
the structural dimensions given for each of the 22 flumes, because the flumes are not
hydraulic scale models of each other. Since throat length and throat bottom slope
remain constant for series of flumes while other dimensions are varied, each of the
22 flumes is an entirely different device. For example, it cannot be assumed that a
dimension in the 12-ft flume will be three times the corresponding dimension in the
4-ft flume.
On the basis of throat width, Parshall flumes have been some what arbitrarily classi-
fied into three main groups for the convenience of discussing them, selecting sizes,
and determining discharges. These groups are very small for 1-, 2-, and 3-in flumes,
small for 6-in through 8-ft flumes and large for IO-ft up to 50-ft flumes (USBR
1971).

224
Table 7.3 Parshall flume dimensions (millimetres)

Dimensions as shown in Figure 7.9

b, A a B C D E L G H K M N P R X Y Z

1" 25.4 363 242 356 93 167 229 76 203 206 19 - 29 - - 8 13 3

2" 50.8 414 276 406 135 214 254 114 254 257 22 - 43 - - 16 25 6
3" 76.2 467 31 I 457 178 259 457 152 305 309 25 - 57 - - 25 38 13

6" 152.4 621 414 610 394 397 610 305 610 - 76 305 114 902 406 51 76 -
9" 228.6 879 587 864 381 575 762 305 457 - 76 305 114 1080 406 51 76 -
I' 304.8 1372 914 I343 610 845 914 610 914 - 76 381 229 1492 508 51 76 -
1'6" 457.2 1448 965 1419 762 1026 914 610 914 - 76 381 229 1676 508 51 76 -
2' 609.6 1524 1016 1495 914 1206 914 610 914 - 76 381 229 1854 508 51 76 -
3' 914.4 1676 1118 1645 1219 1572 914 610 914 - 76 381 229 2222 508 51 16 -
4 1219.2 1829 1219 I794 1524 1937 914 610 914 - 76 457 229 2711 610 51 76 -
5' 1524.0 1981 1321 1943 1829 2302 914 610 914 - 76 457 229 3080 610 51 76 -
6' 1828.0 2134 1422 2092 2134 2667 914 610 914 - 76 457 229 3442 610 51 76 -
7' 2133.6 2286 1524 2242 2438 3032 914 610 914 - 76 457 229 3810 610 51 76 -
8' 2438.4 2438 I626 2391 2743 3397 914 610 914 - 76 457 229 4172 610 51 76 -

10' 3048 - I829 4267 3658 4756 1219 914 1829 I52 343 305 229 -
12' 3658 - 2032 4877 4470 5607 1524 914 2438 152 343 305 229 -
15' 4572 - 2337 7620 5588 7620 1829 1219 3048 229 457 305 229 -
2 0 6096 - 2845 7620 7315 9144 2134 I829 3658 305 686 305 229 -
25' 7620 - 3353 7620 8941 10668 2134 1829 3962 305 686 305 229 -
3 0 9144 - 3861 7925 10566 12313 2134 1829 4267 305 686 305 229 -
4 0 12191 - 4877 8230 13818 15481 2134 1829 4877 305 686 305 229 -
50' 15240 - 5893 8230 17272 18529 2134 1829 6096 305 686 305 229 -
- - - -

....-__ - ~ - -- -_ A

Photo 3 Transparant model of a Parshall flume

225
 Very small flumes (1 , 2, and 3)

The discharge capacity of the very small flumes ranges from 0.09 I/s to 32 I/s. The
capacity of each flume overlaps that of the next size by about one-half the discharge
range (see Table 7.4). The flumes must be carefully constructed. The exact dimensions
of each flume are listed in Table 7.3. The maximum tolerance on the throat width
b, equals +0.0005 m.
The relatively deep and narrow throat section causes turbulence and makes the h,
gauge difficult to read in the very small flumes. Consequently, an h,-gauge, located
near the downstream end of the diverging section of the flume is added. Under sub-
merged flow conditions, this gauge may be read instead of the h,-gauge. The h, readings
are converted to h, readings by using a graph, as will be explained in Section 7.4.3,
and the converted h, readings are then used to determine the discharge.

Small flumes ( 6 , 9 , I, I, 2 up to 8)

The discharge capacity of the small flumes ranges from 0.0015 m3/s to 3.95 m3/s. The
capacity of each size of flume considerably overlaps that of the next size. The length
of the side wall of the converging section, A, of the flumes with 1 up to 8 throat
width is in metres:

A = -b,2+ 1.219 (7-3)

where b, is the throat width in metres. The piezometer tap forsthe upstream head,
h,, is located in one of the converging walls a distance of a = 3 A upstream from
the end of the horizontal crest (see Figure 7.9). The location of the piezometer tap for
the downstream head, h,, is the same in all the small flumes, being 51 mm
(X = 2 inch) upstream from the low point in the sloping throat floor and 76 mm (Y = 3
inch) above it. The exact dimensions of each size of flume are listed in Table 7.3.

Large flumes (1 O up to 50)

The discharge capacity of the large flumes ranges from O. 16 m3/s to 93.04 m3/s. The
capacity of each size of flume considerably overlaps that of the next size. The axial
length of the converging section is considerably longer than it is in the small flumes
to obtain an adequately smooth flow pattern in the upstream part of the structure.
The measuring station for the upstream head, ha, however, is maintained at a = b,/3
+ 0.813 m upstream from the end of the horizontal crest. The location of the piezo-
meter tap for the downstream head, h,, is the same in all the large flumes, being
305 mm (12 in) upstream from the floor at the downstream edge of the throat and
229 mm (9 in) above it. The exact dimensions of each size of flume are listed in Table
7.3.
All flumes must be carefully constructed to the dimensions listed, and careful level-
ling is necessary in both longitudinal and transverse directions if the standard discharge
table is to be used. When gauge zeros are established, they should be set so that the
ha-, hb-, and h,-gauges give the depth of water above the level crest - not the depths
above pressure taps.

226
 surface, S
surface F F

SECTION A-A

converging throat diverging

section section section

'I
-
I
__
III-
i-.
A

Figure 7.9 Parshall flume dimensions

If the Parshall flume is never to be operated above the 0.60 submergence limit, there
is no need to construct the portion downstream of the throat. The truncated Parshall
flume (without diverging section) has the same modular flow characteristics as the
standard flume. The truncated flume is sometimes referred to as the 'Montana flume'.

7.4.2 Evaluation of discharge

The upstream head-discharge (ha-Q) relationship of Parshall flume of various sizes,

as calibrated empirically, is represented by an equation, having the form
Q = Kh," (7-4)
where K denotes a dimensional factor which is a function of the throat width. The
power u varies between 1.522 and 1.60. Values of K and u for each size of flume
are given in Table 7.4. In the listed equations Q is the modular discharge in m3/s,
and ha is the upstream gauge reading in metres.
The flumes cover a range of discharges from 0.09 l/s to 93.04 m3/sand have overlap-

227
ping capacities to facilitate the selection of a suitable size. Each of the flumes listed
in Table 7.4 is a standard device and has been calibrated for the range of discharges
shown in the table. Detailed information on the modular discharge for each size-of
flume as a function of h, are presented in the Tables 7.5 to 7.1 I .

Table 7.4 Discharge characteristics of Parshall flumes

Throat Discharge range Equation Head range Modular

width b, in m3/s x IOT3 Q = K haU in metres limit
in feet (Q in m3/s) hb/ha
or inches minimum maximum minimum maximum

1" 0.09 5.4 0.0604 hal." 0.015 0.21 0.50

2" 0.18 13.2 0.1207 hal." 0.015 0.24 0.50
3" 0.77 32.1 0.1771 ha1." 0.03 0.33 0.50
6 1.50 111 0.3812 ha"" 0.03 0.45 0.60
9" 2.50 25 1 0.5354 0.03 0.61 0.60

I' 3.32 457 0.6909 0.03 0.76 0.70

1' 6 4.80 695 1.056 0.03 0.76 0.70
2' 12.1 937 1.428 0.046 0.76 0.70
3' 17.6 1427 2.184 0.046 0.76 0.70
4' 35.8 1923 2.953 0.06 0.76 0.70
5' 44. I 2424 3.732 0.06 0.76 0.70
6' 74.1 2929 4.519 0.076 0.76 0.70
7' 85.8 3438 5.312 hal.60' 0.076 0.76 0.70
8' 97.2 3949 6.1 12 0.076 0.76 0.70

in m3/s

10 0.16 8.28 7.463 hal.60 0.09 1.O7 0.80

12' 0.19 14.68 8.859 0.09 1.37 0.80
15' 0.23 25.04 10.96 0.09 1.67 0.80
20' 0.31 37.97 14.45 hal.60 0.09 1.83 0.80
25' 0.38 47.14 17.94 0.09 1,.83 0.80
30 0.46 56.33 21.44 ha'.60 0.09 1.83 0.80
40 0.60 74.70 28.43 0.09 1.83 0.80
50' 0.75 93.04 35.41 hal.60 0.09 1.83 0.80

228
Table 7.5 Free-flow discharge through 1 Parshall measuring flume in I/s computed from the formula
Q = 0.0604 hal.55

Head ha
(ml ,000 .O0I .O02 .O03 .O04 ,005 ,006 .O07 .O08 ,009

.o I 0.09 0.10 0.11 0.12 0.13

.o2 0.14 0.15 0.16 0.17 0.19 0.20 0.21 0.22 0.24 0.25
.O3 0.26 0.28 0.29 0.31 0.32 0.33 0.35 0.36 0.38 0.40
.O4 0.41 0.43 0.44 0.46 0.48 0.49 0.51 0.53 0.55 0.56
.o5 0.58 0.60 0.62 0.64 0.66 0.67 0.69 0.71 0.73 0.75
.O6 0.77 0.79 0.81 0.83 0.85 0.87 0.89 0.92 0.94 0.96
.O7 0.98 1.00 1.02 1.05 1.07 1.09 1.11 1.14 1.16 1.18
.os 1.20 1.23 1.25 1.28 1.30 1.32 1.35 1.37 1.40 1.42
.O9 1.45 1.47 1.50 1.52 1.55 1.57 1.60 1.62 1.65 1.68
.10 1.70 1.73 1.76 1.78 1.81 1.84 1.86 1.89 1.92 1.95
.I 1 1.97 2.00 2.03 2.06 2.09 2.11 2.14 2.17 2.20 2.23
.I2 2.26 2.29 2.32 2.35 2.38 2.41 2.44 2.47 2.50 2.53
.I3 2.56 2.59 2.62 2.65 2.68 2.71 2.74 2.77 2.80 2.84
.I4 2.87 2.90 2.93 2.96 3.00 3.03 3.06 3.09 3.13 3.16
.I5 3.19 3.22 3.26 3.29 3.32 3.36 3.39 3.43 3.46 3.49
.I6 3.53 3.56 3.60 3.63 3.66 3.70 3.73 3.77 3.80 3.84
.I7 3.87 3.91 3.95 3.98 4.02 4.05 4.09 4.12 4.16 4.20
.I8 4.23 4.27 4.31 4.34 4.38 4.42 4.45 4.49 4.53 4.57
.I9 4.60 4.64 4.68 4.72 4.75 4.79 4.83 4.87 4.91 4.95
.20 4.98 5.02 5.06 5.10 5.14 5.18 5.22 5.26 5.30 5.34
.2 1 5.38

Table 7.6 Free-flow discharge through 2 Parshall mesuring flume in l/s computed from the formula
Q = 0.1207

Head ha
(ml ,000 .OOI .O02 .O03 .O04 .O05 .O06 .O07 .O08 .O09

.o 1 0.18 0.20 0.22 0.24 0.26

.o2 0.28 0.30 0.33 0.35 0.37 0.40 0.42 0.45 0.47 0.50
.O3 0.53 0.55 0.58 0.61 0.64 0.67 0.70 0.73 0.76 0.79
.O4 0.82 0.85 0.89 0.92 0.95 0.99 1.o2 1.O6 1.o9 1.13
.o5 1.16 I .20 I .23 I .27 1.31 I .35 1.38 1.42 1.46 I .so
.O6 1.54 1.58 1.62 1.66 1.70 I .74 I .79 I .83 I .87 1.91
.O7 I .96 2.00 2.04 2.09 2.13 2.18 2.22 2.27 2.31 .2.36
.O8 2.41 2.45 2.50 2.55 2.60 2.64 2.69 2.74 2.79 2.84
.O9 2.89 2.94 2.99 3.04 3.09 3.14 3.19 3.24 3.30 3.35
.10 3.40 3.45 3.51 3.56 3.62 3.67 3.72 3.78 3.83 3.89
.11 3.94 4.00 4.06 4.1 1 4.17 4.22 4.28 4.34 4.40 4.45
.I2 4.51 4.57 4.63 4.69 4.75 4.81 4.87 4.93 4.99 5.05
.I3 5.11 5.17 5.23 5.29 5.35 5.42 5.48 5.54 5.60 5.67
.14 5.73 5.79 5.86 5.92 5.99 6.05 6.12 6.18 6.25 6.3 1
.I5 6.38 6.44 6.51 6.58 6.64 6.71 6.78 6.84 6.91 6.98
.I6 7.05 7.12 7.19 7.25 7.32 7.39 7.46 7.53 7.60 7.67
.I7 7.74 7.81 7.88 7.96 8.03 8.10 8.17 8.24 8.31 8.39
.I8 8.46 8.53 8.61 8.68 8.75 8.83 8.90 8.98 9.05 9.12
.I9 9.20 9.27 9.35 9.43 9.50 9.58 9.65 9.73 9.81 9.88
.20 9.96 10.04 10.12 10.19 10.27 10.35 10.43 10.51 10.59 10.66
.21 10.74 10.82 10.90 10.98 1 I .O6 11.14 1 I .22 1 I .30 11.38 1 I .47
.22 11.55 11.63 11.71 11.79 11.87 I I .96 12.04 12.12 12.20 12.29
.23 12.37 12.45 12.54 12.62 12.71 12.79 12.87 12.96 13.04 13.13
.24 13.21

229
Table 7.7 Free-flow discharge through 3" Parshall measuring flume in I/s computed from the formula
Q = 0.1771 ha'.550

Upper-
head ha
(d .O00 .O01 .O02 .O03 .O04 .O05 .O06 .O07 .O08 .O09

.O3 0.77 0.81 0.85 0.90 0.94 0.98 1 .o2 1.O7 1.1 1 1.16
.O4 1.21 1.25 1.30 1.35 1.40 1.45 1s o 1.55 1.60 1.65
.o5 1.70 1.76 1.81 1.87 1.92 1.98 2.03 2.09 2.15 2.20
.O6 2.26 2.32 2.38 2.44 2.50 2.56 2.62 2.68 2.75 2.81
.O7 2.87 2.94 3.00 3.06 3.13 3.20 3.26 3.33 3.40 3.46
.O8 3.53 3.60 3.67 3.74 3.81 3.88 3.95 4.02 4.09 4.17
.O9 4.24 4.31 4.39 4.46 4.53 4.61 4.69 4.76 4.84 4.91
.IO 4.99 5.07 5.15 5.23 5.30 5.38 5.46 5.54 5.62 5.70
.I I 5.79 5.87 5.95 6.03 6.12 6.20 6.28 6.37 6.45 6.54
.12 6.62 6.71 6.79 6.88 6.97 7.05 7. I4 7.23 7.32 7.41
.I3 7.50 7.59 7.68 7.77 7.80 7.95 8.04 8.13 8.22 8.32
.I4 8.41 8.50 8.60 8.69 8.78 8.88 8.97 9.07 9.16 .9.26
.15 9.36 9.45. 9.55 9.65 9.75 9.85 9,94 10.04 10.14 10.24
.16 10.34 10.44 10.54 10.64 10.75 10.85 10.95 1 1.o5 11.15 1 1.26
.I7 11.36 11.46 11.57 11.67 11.78 1 I .88 11.99 12.09 12.20 12.31
.I8 12.41 12.52 12.63 12.74 12.84 12.95 13.06 13.17 13.28 13.39
.19 13.50 13.61 13.72 13.83 13.94 14.05 14.16 14.28 14.39 14.50
.20 14.62 14.73 14.84 14.96 15.07 15.19 15.30 15.42 15.53 15.65
.2 I 15.76 15.88 16.00 16.1 I 16.23 16.35 16.47 16.59 16.70 16.82
.22 16.94 17.06 17.18 17.30 17.42 17.54 17.66 17.79 17.91 18.03
.23 18.15 18.27 18.40 18.52 18.64 18.77 18.89 19.01 19.14 19.26
.24 19.39 19.51 19.64 19.77 19.89 20.02 20.15 20.27 20.40 20.53
.25 20.66 20.78 20.9 I 21.04 21.17 21.30 21.43 21.56 21.69 21.82
.26 21.95 22.08 22.21 22.34 22.48 22.61 22.74 22.87 23.01 23.14
.27 23.27 23.41 23.54 23.67 23.81 23.94 24.08 24.21 24.35 24.49
.28 24.62 24.76 24.89 25.03 25.17 25.31 25.44 25.58 25.72 25.86
.29 26.00 26.14 26.28 26.42 26.56 26.70 26.84 26.98 27.12 27.26
.30 27.40 27.54 27.68 27.83 27.97 28.1 1 28.25 28.40 28.54 28.68
.3 I 28.83 28.97 29.12 29.26 29.41 29.55 29.70 29.84 29.99 30.14
.32 30.28 30.43 30.58 30.72 30.87 3 1.O2 31.17 31.32 3 1.46 31.61
.33 3 I .76 31.91 32.06

230
Table 7.8 Free-flow discharge through 6 Parshall measuring flume in I/s computed from the formula
Q = 0.3812 ha'.580

Upper-
head ha
(m) ,000 .O01 ,002 .O03 ,004 ,005 ,006 ,007 .O08 .O09

.O3 1.5 I .6 I .7 1.7 I .8 1.9 2.0 2. I 2.2 2.3

.O4 2.4 2.4 2.6 2.6 2.7 2.8 2.9 3.0 3.1 3.2
.o5 3.4 3.5 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.4
.O6 4.5 4.6 4.7 4.8 5.0 5.1 5.2 5.3 5.4 5.6
.O7 5.7 5.8 6.0 6.1 6.2 6.4 6.5 6.6 6.8 6.9
.O8 7.0 7.2 7.3 7.5 7.6 7.8 7.9 8.0 8.2 8.3
.O9 8.5 8.6 8.8 8.9 9. I 9.2 9.4 9.6 9.7 9.9
.10 10.0 10.2 10.4 10.5 10.7 10.8 11.0 11.2 11.3 11.5
.I 1 11.7 11.8 12.0 12.2 12.3 12.5 12.7 12.8 13.0 13.2
.I2 13.4 13.6 13.7 13.9 14.1 14.3 14.4 14.6 14.8 15.0
.I3 15.2 15.4 15.6 15.7 15.9 16.1 16.3 16.5 16.7 16.9
.14 17.1 17.3 17.4 17.6 17.8 18.0 18.2 18.4 18.6 18.8
.I5 19.0 19.2 19.4 19.6 19.8 20.0 20.2 20.4 20.7 20.9
.I6 21.1 21.3 21.5 21.7 21.9 22.1 22.3 22.5 22.8 23.0
.I7 23.2 23.4 23.6 23.8 24.1 24.3 24.5 24.7 24.9 25.2
.I8 25.4 25.6 25.8 26.0 26.3 26.5 26.7 27.0 27.2 27.4
.19 27.6 27.9 28. I 28.3 28.6 28.8 29.0 29.3 29.5 29.7
.20 30.0 30.2 30.4 30.7 30.9 31.2 31.4 31.6 31.9 32.1
.21 32.4 32.6 32.9 33.1 33.4 33.6 33.8 34.1 34.4 34.6
.22 34.8 35.1 35.4 35.6 35.8 36.1 36.4 36.6 36.9 37.1
.23 37.4 37.6 37.9 38.2 38.4 38.7 38.9 39.2 39.5 39.7
.24 40.0 40.2 40.5 40.8 41.0 41.3 41.6 41.8 42.1 42.4
.25 42.6 42.9 43.2 43.5 43.7 44.0 44.3 44.6 44.8 45.1
.26 45.4 45.6 45.9 46.2 46.5 46.8 47.0 47.3 47.6 47.9
.27 48.2 48.4 48.7 49.0 49.3 49.6 49.9 50.2 50.4 50.7
.28 51.0 51.3 51.6 51.9 52.2 52.5 52.8 53.0 53.3 53.6
.29 53.9 54.2 54.5 54.8 55. I 55.4 55.7 56.0 56.3 56.6
.30 56.9 57.2 57.5 57.8 58. I 58.4 SS.! 59.0 59.3 59.6
.3 1 59.9 60.2 60.5 60.8 61.1 61.4 61.8 62. I 62.4 62.7
.32 62.0 63.3 63.6 63.9 64.2 64.6 64.9 65.2 65.5 65.8
.33 66.1 66.4 66.8 67.1 67.4 67.7 68.0 68.4 68.7 69.0
.34 69.3 69.6 70.0 70.3 70.6 70.9 71.3 71..6 71.9 72.2
.35 72.6 72.9 73.2 73.6 73.9 74.2 74.6 74.9 75.2 75.5
.36 75.9 76.2 76.5 76.9 77.2 77.6 77.9 78.2 78.6 78.9
.37 79.2 79.6 19.9 80.2 80.6 80.9 81.3 81.6 82.0 82.3
.38 82.6 83.0 83.3 83.7 84.0 84.4 84.7 85.1 85.4 85.8
.39 86.1 86.5 86.8 87.2 87.5 81.9 88.2 88.6 88.9 89.3
.40 89.6 90.0 90.3 90.7 91.0 91.4 91.8 92.1 92.5 92.8
.4 I 93.2 93.6 93.9 94.3 94.6 95.0 95.4 95.1 96. I 96.4
.42 96.8 97.2 97.5 97.9 98.3 98.6 99.0 99.4 99.7 100.I
.43 100.5 100.8 101.2 101.6 102.0 102.3 102.7 103.1 103.4 103.8
.44 104.2 104.6 104.9 105.3 105.7 106.1 106.4 106.8 107.2 107.6
.45 108.0 108.3 108.7 109.1 109.5 109.8 110.2 110.6

23 1
Table 7.9 Free-flow discharge through 9 Parshall measuring flume in I/s computed from the formula
Q = 0.5354

Upper-
head ha
( 4 .O00 .O0 1 .O02 .O03 ,004 .O05 .O06 .O07 .O08 .O09

.O3 2.5 2.6 2.8 2.9 3.0 3.2 3.3 3.4 3.6 3.7
.O4 3.9 4.0 4.2 4.3 4.5 4.7 4.8 5.0 5. I 5.3
.o5 5.5 5.6 5.8 6.0 6.2 6.3 6.5 6.7 6.9 7.0
.O6 7.2 7.4 7.6 7.8 8.0 8.2 8.4 8.6 8.8 9.0
.O7 9.2 9.4 9.6 9.8 10.0 10.2 10.4 10.6 10.8 11.0
.O8 11.2 11.4 11.7 11.9 12.1 12.3 12.5 12.8 13.0 13.2
.O9 13.4 13.7 13.9 14.I 14.4 14.6 14.8 15.1 15.3 15.6
.IO 15.8 16.0 16.3 16.5 16.8 17.0 17.3 17.5 17.8 18.0
.II 18.3 18.5 18.8 19.0 19.3 19.6 19.8 20. I 20.4 20.6
.I2 20.9 21.2 21.4 21.7 22.0 22.2 22.5 22.8 23.0 23.3
.I3 23.6 23.9 24.2 24.4 24.7 25.0 25.3 25.6 25.9 26.2
.14 26.4 26.7 27.0 27.3 27.6 27.9 28.2 28.5 28.8 29.1
.I5 29.4 29.7 30.0 30.3 30.6 30.9 31.2 31.5 31.8 32.1
.I6 32.4 32.7 33.0 33.4 33.7 34.0 34.3 34.6 35.0 35.3
.17 35.6 35.9 36.2 36.6 36.9 37.2 37.5 37.8 38.2 38.5
.18 38.8 39.2 39.5 39.8 40.2 40.5 40.8 41.2 41.5 41.8
.I9 42.2 42.5 42.9 43.2 43.6 43.9 44.2 44.6 44.9 45.3
.20 45.6 46.0 46.3 46.7 47.0 47.4 47.7 48.1 48.4 48.8
.21 49.2 49.5 49.9 50.2 50.6 51.0 51.3 51.7 52.1 52.4
.22 52.8 53.2 53.5 53.9 54.3 54.6 55.0 55.4 55.8 56.1
.23 56.5 56.9 57.3 51.6 58.0 58.4 58.8 59.2 59.5 59.9
.24 60.3 60.7 61.1 61.5 61.9 62.2 62.6 63.0 63.4 63.8
.25 64.2 64.6 65.0 65.4 65.8 66.2 66.6 67.0 67.4 67.8
.26 68.2 68.6 69.0 69.4 69.8 70.2 70.6 71.0 71.4 71.8
.27 72.2 72.6 73.0 73.4 73.9 74.3 74.7 75. I 75.5 75.9
.28 76.4 76.8 77.2 77.6 78.0 78.4 78.9 79.3 79.7 80.1
.29 80.6 81.0 81.4 81.8 82.3 82.7 83.1 83.6 84.0 84.4
.30 84.8 85.3 85.7 86.2 86.6 87.0 87.5 87.9 88.3 88.8
.31 89.2 89.7 90.1 90.5 91.0 91.4 91.9 92.3 92.8 93.2
.32 93.7 94. I 94.6 95.0 95.5 95.9 96.4 96.8 97.3 97.7
.33 98.2 98.6 99.1 99.5 100.0 100.5 100.9 101.4 101.8 102.3
.34 102.8 103.2 103.7 104.2 104.6 105.1 105.6 106.0 106.5 107.0
.35 107.4 107.9 108.4 108.8 109.3 109.8 110.2 110.7 111.2 111.7
.36 112.2 112.6 113.1 113.6 114.1 114.6 115.0 115.5 116.0 116.5
.37 117.0 117.4 117.9 118.4 118.9 119.4 119.9 120.4 120.8 121.3
.38 121.8 122.3 122.8 123.3 123.8 124.3 124.8 125.3 125.8 126.3
.39 126.8 127.3 127.8 128.3 128.8 129.3 129.8 130.3 130.8 131.3
.40 131.8 132.3 132.8 133.3 133.8 134.3 134.8 135.3 135.8 136.3
.41 136.8 137.4 137.9 138.4 138.9 139.4 139.9 140.4 141.0 141.5
.42 142.0 142.5 143.0 143.5 144.1 144.6 145.1 145.6 146.2 146.7
.43 147.2 147.7 148.2 148.8 149.3 149.8 150.4 150.9 151.4 151.9
.44 152.5 153.0 153.5 154.1 154.6 155.1 155.6 156.2 156.7 157.3
.45 157.8 158.3 158.9 159.4 160.0 160.5 161.0 161.6 162.1 162.6
.46 163.2 163.7 164.3 164.8 165.4 165.9 166.5 167.0 167.6 168.1
.47 168.6 169.2 169.8 170.3 170.8 171.4 172.0 172.5 173.1 173.6
.48 174.2 174.7 175.3 175.8 176.4 177.0 177.5 178.1 178.6 179.2
.49 179.8 180.3 180.9 181.4 182.0 182.6 183.1 183.7 184.3 184.8
so 185.4 186.0 186.5 187.1 187.7 188.2 188.8 189.4 190.0 190.5
.SI 191.1 191.7 192.2 192.8 193.4 194.0 194.6 195.1 195.7 196.3
.52 196.9 197.4 198.0 198.6 199.2 199.8 200.4 200.9 201.5 202. I
.53 202.7 203.3 203.9 204.4 205.0 205.6 206.2 206.8 207.4 208.0
.54 208.6 209.2 209.8 210.3 210.9 211.5 212.1 212.7 213.3 213.0
.55 214.5 215.1 215.7 216.3 216.9 217.5 218.1 218.7 219.3 219.9
.56 220.5 221.1 221.7 222.3 222.9 223.5 224.1 224.7 225.3 225.9
.57 226.6 227.2 227.8 228.4 229.0 229.6 230.2 230.8 231.4 232.0
.58 232.7 233.3 233.9 234.5 235.1 235.7 236.4 237.0 237.6 238.2
.59 238.8 239.4 240.1 240.7 241.3 241.9 242.6 243.2 243.8 244.4
.60 245.0 245.7 246.3 246.0 247.6 248.2 248.8 249.4 250.1 250.7
.61 251.3

232
Table 7.10 Free-flow discharge through Parshall measuring flumes 1-to-8 foot size in I/s computed from
the formulae as shown in Table 7.4

Upper- Discharge in I/s for flumes of various throat widths

head ha
(") 1 1.5 2 3 4 5 6 7 8
feet feet feet feet feet feet feet feet feet

30 3.3 4.8
32 3.7 5.3
34 4.0 5.8
36 4.4 6.4
38 4.8 6.9
40 5.2 7.5
42 5.6 8.1
44 6.0 8.7
46 6.4 9.3 12.1 17.6
48 6.8 9.9 12.9 18.8
50 7.2 10.5 13.7 20.0
52 7.7 11.2 14.6 21.3
54 8.1 11.9 15.5 22.6
56 8.6 12.5 16.4 23.9
58 9.I 13.2 17.3 25.3
60 9.5 14.0 18.2 26.7
62 10.0 14.7 19.2 28.1 36.7 45.2
64 10.5 15.4 20.2 29.5 38.6 47.6
66 11.0 16.2 21.1 31.0 40.5 50.0
68 11.6 16.9 22.I 32.4 42.5 52.4
70 12.1 17.7 23.2 33.9 44.4 54.8
72 12.6 18.5 24.2 35.5 46.5 57.4
74 13.1 19.2 25.2 37.0 48.5 59.9
76 13.7 20.I 26.3 38.6 50.6 62.5 74.1 85.8 97.2
78 14.2 20.9 27.4 40.2 52.7 65.I 77.3 89.4 101.3
80 14.8 21.7 28.5 41.8 54.9 67.8 80.4 93.I 105.6
82 15.4 22.0 29.6 43.5 57.0 70.5 83.7 96.9 109.8
84 15.9 23.4 30.7 45.2 59.3 73.2 87.0 100.7 114.2
86 16.5 24.3 31.9 46.8 61.5 76.0 90.3 104.6 118.6
88 17.1 25.1 33.0 48.6 63.8 78.9 93.6 108.5 123.0
90 17.7 26.0 34.2 50.3 66.1 81.7 97.1 112.5 127.5
92 18.3 26.9 35.4 52.1 68.4 84.6 100.5 116.5 132.1
94 18.9 27.8 36.6 53.8 70.8 87.6 104.0 120.6 136.8
96 19.5 28.7 37.8 55.6 73.2 90.5 107.6 124.7 141.5
98 20.I 29.7 39.0 57.5 75.6 93.5 111.2 128.9 146.2
I O0 20.8 30.6 40.2 59.3 78.0 96.6 114.8 133.1 151.1
I02 21.4 31.5 41.5 61.2 80.5 99.7 118.5 137.4 156.0
I04 22.0 32.5 42.8 63.1 83.0 102.8 122.2 141.8 160.9
I06 22.7 33.5 44.0 65.0 85.6 106.0 126.0 146.1 165.9
IO8 23.4 34.4 45.4 66.9 88.1 109.1 129.8 150.6 171.0
I10 24.0 35.4 46.6 68.9 90.7 112.4 133.7 155.1 176.1
I12 24.7 36.4 48.0 70.8 93.3 1 15.6 137.6 159.6 181.2
I I4 25.4 37.4 49.3 72.8 96.0 118.9 141.5 164.2 186.5
I I6 26.0 38.4 50.7 74.8 98.6 122.2 145.5 168.8 191.8
118 26.7 39.5 52.0 76.9 101.3 125.6 149.5 173.5 197.1
120 27.4 40.5 53.4 78.9 104.0 129.0 153.6 178.2 202.5
I22 28.I 41.5 54.8 81.0 106.8 132.4 157.7 183.0 208.0
I24 28.8 42.6 56.2 83.1 109.6 135.9 161.8 187.9 213.5
I26 29.5 43.6 51.6 85.2 112.4 ,139.4 166.0 192.7 219.0
I28 30.2 44.7 59.0 87.3 1 15.2 142.9 170.2 197.6 224.6
130 31.0 45.8 60.4 89.5 118.0 146.5 174.5 202.6 230.3
132 31.7 46.9 61.9 91.6 120.9 150.1 178.8 207.6 236.0
I34 32.4 48.0 63.4 93.8 123.8 153.7 183.1 212.7 241.8
I36 33.2 49.1 64.8 96.0 126.8 157.4 187.5 217.8 247.6
138 33.9 50.2 66.3 98.2 129.7 161.0 191.9 223.0 253.5
I40 34.7 51.3 67.8 100.5 132.7 164.8 196.4 228.1 259.4
I42 35.4 52.5 69.3 102.7 135.7 168.5 200.9 233.4 265.4
144 36.2 53.6 70.8 105.0 138.7 172.3 205.4 238.7 271.4
I46 37.0 54.8 72.4 107.3 141.8 176.1 210.0 244.0 277.5
I48 37.7 55.9 73.9 109.6 144.9 180.0 214.6 249.4 283.7
150 38.5 57.1 75.4 112.0 148.0 183.8 219.2 254.8 289.8

233
Table 7.10 continued

Upper- Discharge in I/s for flumes of various throat widths

head ha
1"( 1 1.5 2 3 4 5 6 I 8
feet feet feet feet feet feet feet feet feet
152 39.3 58.3 77.0 114.3 151.1 187.7 223.9 260.2 296. I
154 40. I 59.4 78.6 116.7 154.2 191.7 228.6 265.8 302.4
156 40.9 60.6 80.2 119.0 157.4 195.6 233.4 271.3 308.7
158 41.7 61.8 81.8 121.4 160.6 199.6 238.2 276.9 315.1
160 42.5 63.0 83.4 123.8 163.8 203.6 243.0 282.5 321.5
162 43.3 64.2 85.0 126.3 167.1 207.7 247.9 288.2 328.0
164 44.1 65.5 86.6 128.7 170.3 211.8 252.8 293.9 334.5
I66 44.9 66.7 88.3 131.2 173.6 215.9 257.7 299.7 341.1
I68 45.7 68.0 89.9 133.7 176.9 220.0 262.7 305.5 347.7
I70 46.6 69.2 91.6 136.2 180.3 224.2 267.7 311.3 354.4
I72 47.4 70.4 93.3 138.7 183.6 228.4 272.7 3 17.2 361.1
174 48.2 71.7 95.0 141.2 187.0 232.6 277.8 223. I 367.9
176 49.1 73.0 96.7 143.8 190.4 236.9 282.9 329. I 374.7
178 50.0 74.3 98.4 146.4 193.8 241.2 288.0 335.1 381.6
I80 50.8 75.6 100.1 148.9 197.3 245.5 293.2 341.2 388.5
182 51.7 76.8 101.8 151.5 200.8 249.8 298.4 347.2 395.5
I84 52.5 78.2 103.6 154.2 204.2 254.2 303.7 353.4 402.5
I86 53.4 79.5 105.3 156.8 207.8 258.6 309.0 359.5 409.5
1 88 54.3 80.8 107.1 159.4 211.3 263.0 314.3 365.8 416.6
190 55.2 82. I 108.8 162.1 214.8 267.5 319.6 372.0 423.8
192 56.0 83.4 110.6 164.8 218.4 272.0 325.0 378.3 43 I .o
194 56.9 84.8 112.4 167.5 222.0 276.5 330.4 384.6 438.2
196 57.8 86. I 114.2 170.2 225.6 281.0 335.9 39 1 .O 445.5
198 58.7 87.5 116.0 172.9 229.3 285.6 34 I .4 397.4 452.8
200 59.6 88.8 117.8 175.7 233.0 290.2 346.9 403.8 460.2

202 60.6 90.2 119.7 178.4 236.6 294.8 352.4 410.3 467.6
204 61.5 91.6 121.5 181.2 240.4 299.4 358.0 416.8 475.1
206 62.4 93.0 123.4 184.0 244.1 304. I 363.6 423.4 482.6
208 63.3 94.4 125.2 186.8 247.8 308.8 369.3 430.0 490.1
210 64.2 95.8 127.1 189.6 25 1.6 313.6 375.0 436.6 497.7
212 65.2 97.2 129.0 192.4 255.4 318.3 380.7 443.3 505.4
214 66.1 98.6 130.9 195.3 259.2 323.1 386.4 450.0 513.0
216 67.1 100.0 132.8 198.2 263.0 327.9 392.2 456.8 520.8
218 68.0 101.4 134.7 201.0 266.9 332.7 398.0 463.6 528.6
220 69.0 102.9 136.6 203.9 270.8 337.6 403.8 470.4 536.4
222 69.9 104.3 138.5 206.8 274.7 342.5 409.7 477.3 544.2
224 70.9 105.8 140.5 209.8 278.6 347.4 415.6 484.2 552. I
226 71.8 107.2 142.4 212.7 282.3 352.3 42 I .6 491.1 560. I
228 72.8 108.7 144.4 215.7 286.5 357.3 427.5 498.1 568.0
230 73.8 110.2 146.4 218.6 290.4 362.2 433.5 505. I 576.1
232 74.8 111.6 148.3 22 I .6 294.4 367.3 439.5 512.2 584.2
234 75.8 113.1 150.3 224.6 298.5 372.3 445.6 519.2 592.3
236 76.7 114.6 152.3 227.6 302.5 377.4 451.7 526.4 600.4
238 77.7 116.1 154.3 230.7 306.6 382.4 447.8 533.5 608.6
240 78.7 117.6 156.3 233.7 310.6 387.6 464.0 540.7 616.8

245 81.2 121.4 161.4 241.4 320.9 400.4 479.5 558.9 637.6
250 83.8 125.2 166.6 249.1 33 I .3 413.5 495.2 577.2 658.7
255 86.3 129.1 171.7 257.0 341.8 426.7 511.1 595.8 680.0
260 88.9 133.0 177.0 264.9 352.4 440.0 527.1 614.6 701.5
265 91.5 137.0 182.3 272.9 363.2 453.6 543.4 633.7 723.3
270 94.2 141.0 187.6 281.0 374. I 467.2 559.8 652.9 745.4
275 96.8 145.0 193.1 289.2 385.1 481.1 576.5 672.4 767.7
280 99.5 149.1 198.5 297.5 396.2 495.0 593.3 692. I 790.2
285 102.3 153.2 204.0 305.9 407.4 509. I 610.3 712.0 813.0
290 105.0 157.3 209.6 314.3 418.7 523.3 627.4 732. I 836.1
295 107.8 161.5 215.2 322.8 430.2 537.7 644.8 752.4 859.4
300 110.6 165.8 220.9 331.4 441.7 552.2 662.3 772.9 882.9

234
Table 7.10 continued

Upper- Discharge in l/s for flumes of various throat widths

head h,
"1( 1 1.5 2 3 4 5 6 7 8
feet feet feet feet feet feet feet feet feet

305 113.4 170.0 226.7 340.2 453.4 566.9 680.0 793.6 906.7
310 116.2 174.3 232.4 348.9 465.2 58 I .7 697.8 814.6 930.7
315 119.1 178.7 238.3 357.8 477. I 596.7 715.9 835.7 954.9
320 122.0 183.1 244.2 366.7 489.1 61 1.8 734.1 857.0 979.4
325 124.9 187.5 250. I 375.7 501.2 627.0 752.5 878.6 1004
330 127.8 191.9 256. I 384.8 513.4 642.4 771.0 900.3 1029
335 130.8 196.4 262.2 394.0 525.8- 657.9 789.8 922.3 1054
340 133.8 201.0 268.2 403.2 538.2 673.6 808.6 944.4 1080
345 136.8 205.5 274.4 412.6 550.7 689.4 827.7 966.7 I I05
350 139.8 210.1 280.6 422.0 563.4 705.3 846.9 989.3 1131
355 142.8 214.7 286.8 431.3 576. I 721.4 866.3 1012 I I57
360 145.9 219.4 293.1 441.0 589.0 737.6 885.8 1035 I I83
365 149.0 224.1 299.4 450.6 602.0 753.9 905.5 1058 1210
370 152.1 228.8 305.8 460.3 615.0 770.3 925.4 1081 1237
375 155.3 233.6 312.2 470. I 628.2 786.9 945.4 I105 1264
380 158.4 238.4 318.7 480.0 641.4 803.6 965.6 1128 1291
385 161.6 243.3 325.2 489.9 654.8 820.5 985.9 I I52 1318
390 164.8 248.2 331.8 499.9 668.3 837.4 1006 I I76 1346
395 168.0 253. I 338.4 510.0 681.9 854.6 1027 1201 1374
400 171.3 258.0 345. I 520.1 695.5 87 1.8 I048 1225 1402

405 174.6 263.0 351.8 530.3 709.3 889.2 I069 1250 1430
410 177.9 268.0 358.5 540.6 723.2 906.6 I090 I274 1459
415 181.2 273.0 365.3 551.0 737.1 924.2 IIII I299 1487
420 184.5 278.1 372.2 561.4 75 I .2 942.0 1 I33 I325 1516
425 187.9 283.2 379.1 571.9 765.4 959.8 I I54 1350 1545
430 191.2 288.4 386.0 582.5 779.6 977.8 1176 1375 1575
435 194.6 293.5 393.0 593.1 794.0 995.9 I I98 1401 1604
440 198.0 298.7 400.0 603.8 808.4 1014 I220 1427 1634
445 201.5 304.0 407.1 614.6 823.0 1032 I242 1453 1664
450 204.9 309.2 414.2 625.4 837.6 1051 I264 1479 1694
455 208.4 314.6 42 I .4 636.4 852.3 I070 1287 I506 1724
460 211.9 319.9 428.6 647.3 867.2 I088 1310 I532 1755
465 215.4 325.2 435.8 658.4 882. I 1 IO7 1332 I559 1786
470 219.0 330.6 443.1 669.5 897. I 1 I26 1355 I586 1817
475 222.5 336. I 450.4 680.7 912.2 I145 I378 1613 1848
480 226. I 341.5 457.8 692.0 927.4 I I64 1402 1640 1879
48 5 229.7 347.0 465.2 703.3 942.7 I I84 1425 I668 191 I
490 233.3 352.5 472.6 714.7 958.1 I203 I448 1695 1942
495 236.9 358.1 480. I 726.1 973.5 1223 1472 I723 1974
500 240.6 363.6 487.7 737.6 989. I I242 1496 1751 2006

505 244.2 369.2 495.3 749.2 1005 1262 I520 I779 2039
510 247.9 374.9 502.9 760.9 1020 1282 I544 I808 2071
515 251.6 380.6 510.5 772.6 1036 I302 1568 1836 2104
520 255.4 386.3 518.2 784.4 1052 1322 I592 1865 21 37
525 259.1 392.0 526.0 796.2 I068 I342 1617 1893 2170
530 262.9 397.7 533.8 808. I I084 I363 I642 1922 2203
535 266.7 403.5 541.6 820. I I101 1383 1666 1951 2239
540 270.5 409.3 549.5 832. I 1117 I404 1691 1981 2271
545 274.3 415.2 557.4 844.2 1133 1424 1716 2010 2304
550 278. I 421.1 565.3 856.4 1150 1445 1741 2040 2339
555 282.0 427.0 573.3 868.6 1 I66 I466 I767 2070 2373
560 285.9 432.9 581.3 880.9 I I83 1487 I792 2099 2407
565 289.8 438.8 589.4 893.2 I199 I508 1818 2130 2442
570 293.7 444.8 597.5 905.6 1216 I529 I844 2160 2477
575 297.6 450.8 605.6 918.1 1233 1551 1869 2190 2512
580 301.6 456.9 613.8 930.6 I250 I572 1895 2221 2547
585 305.5 463.0 622.0 943.2 I267 I594 1922 2252 2582
590 309.5 469. I 630.3 955.9 I284 1615 I948 2282 2618
595 313.5 475.2 638.6 968.6 I302 1637 I974 2313 2654
600 317.5 481.4 646.9 98 I .4 1319 I659 200 1 2345 2690

235
Table 7.10 continued

Upper- Discharge in I/s for flumes of various throat widths

head ha
(") 1 1.5 2 3 4 5 6 7 8
feet feet feet feet feet feet feet feet feet

605 321.6 487.5 655.3 994.2 1336 1681 2027 2376 2726
610 325.6 493.7 663.7 I007 1354 I703 2054 2408 2762
615 329.7 500.0 672.2 I020 1371 I725 208 1 2439 2798
620 333.8 506.2 680.7 1033 1389 I748 2108 247 I 2835
625 337.9 512.5 689.2 I046 I407 I770 2135 2503 2872
630 342.0 518.9 697.8 1059 1424 1793 2163 2535 2909
635 346. I 525.2 706.4 1072 1442 1815 2190 2567 2946
640 350.3 531.6 715.0 1086 1460 1838 2218 2600 2983
645 354.5 538.0 723.7 1099 1478 1861 2245 2632 302 I
650 358.6 544.4 732.4 1112 1496 1884 2273 2665 3059
655 362.9 550.9 741.1 1 I26 1515 1907 2301 2698 3097
660 367. I 557.3 749.9 1 I39 1533 1930 2529 273 1 3135
665 371.3 563.8 758.8 1 I53 1551 1953 2357 2764 3173
670 375.6 570.4 761.6 1166 1570 1977 2386 2798 321 1
675 379.8 576.9 776.5 1180 1588 2000 2414 283 1 3250
680 384.1 583.5 785.4 I I94 I607 2024 2443 2865 3289
685 388.4 599.1 794.4 I208 1625 2047 2472 2899 3328
690 392.8 596.8 803.4 1221 I644 2071 2500 2933 3367
695 397.1 603.4 812.5 1235 1663 2095 2529 2967 3406
700 401.5 610.1 821.5 I249 1682 21 I9 2558 300 I 3446

705 405.8 616.8 839.7 I263 1701 2143 2588 3035 3485
710 410.2 623.6 839.8 1277 I720 2167 2617 3070 3525
715 414.6 630.4 849.0 1292 I739 2191 2646 3105 3565
720 419.1 637.2 858.2 1306 1758 2216 2676 3139 3605
725 423.5 644.0 867.5 I320 1778 2240 2706 3174 3645
730 428.0 650.8 876.8 I334 1797 2265 2736 3210 3686
735 432.4 657.7 886. I 1349 1817 2289 2765 3245 3727
740 436.9 664.6 895.4 1363 1836 2314 2796 3280 3767
745 441.4 671.5 904.8 1377 1856 2339 2826 3316 3808
750 445.9 678.4 914.3 1392 1875 2364 2856 3351 3850
755 450.4 685.4 923.7 1406 1895 2389 2886 3387 3891
760 455.0 692.4 933.2 1421 1915 2414 2917 3423 3932

236
Table 7.1 I Free-flow discharge through Parshall measuring flumes I O to 50 feet size in m3/s. Computed
from the formulae as shown in Table 7.4

Upper- Discharge per m3/s for flumes of various throat widths

head ha
"() 10 12 15 20 25 30 40 50
feet feet feet feet feet feet feet feet

90 o. I58 o. 188 0.233 0.307 0.381 0.455 0.603 0.751

95 O. 173 0.205 0.254 0.334 0.415 0.496 0.658 0.819
1O0 O. 187 0.223 0.275 0.363 0.451 0.539 0.714 0.889
105 0.203 0.241 0.298 0.392 0.487 0.582 0.772 0.962
I IO 0.218 0.259 0.321 0.423 0.525 0.627 0.832 1 .O4
115 0.234 0.278 0.344 0.454 0.564 0.674 0.893 1.11
120 0.251 0.298 0.369 0.486 0.603 0.721 0.956 1.19
I25 0.268 0.318 0.393 0.519 0.644 0.770 1 .o2 1.27
130 0.285 0.339 0.419 0.552 0.686 0.8 I9 1.O9 1.35
I35 0.303 0.360 0.445 0.587 0.728 0.870 1.15 1.44
140 0.321 0.381 0.472 0.622 0.772 0.923 1.22 1.52
145 0.340 0.403 0.499 0.658 0.8 17 0.976 1.29 1.61
I50 0.359 0.426 0.527 0.694 0.862 1 .O3 1.37 I .70
I55 0.378 0.449 0.555 0.732 0.909 I .O9 1.44 1.79
I60 0.398 0.472 0.584 0.770 0.956 1.14 1.51 1.89
165 0.418 0.496 0.613 0.809 1 .o0 1.20 1.59 1.98
170 0.438 0.520 0.643 0.848 I .o5 I .26 I .67 2.08
175 0.459 0.545 0.674 0.889 1.10 1.32 1.75 2.18
I80 0.480 0.570 0.705 0.930 1.15 I .38 1.83 2.28
185 0.502 0.595 0.737 0.971 1.21 1.44 1.91 2.38
190 0.524 0.621 0.769 1.01 1.26 1 so 1.99 2.48
195 0.546 0.648 0.801 I .O6 1.31 1.57 2.08 2.59
200 0.568 0.675 0.835 1.10 1.37 I .63 2.16 2.70
205 0.591 0.702 0.868 1.14 1.42 I .70 2.25 2.80
210 0.614 0.739 0.902 1.19 1.48 1.77 2.34 2.92
215 0.638 0.757 0.937 1.24 ' 1.53 1.83 2.43 3.03
220 0.662 0.786 0.972 1.28 1.59 I .90 2.52 3.14
225 0.686 0.814 1.01 1.33 I .65 I .97 2.61 3.26
230 0.71 I 0.844 1 .O4 1.38 1.71 2.04 2.71 3.37
235 0.736 0.873 1 .O8 1.42 1.77 2.11 2.80 3.49
240 0.761 0.903 1.12 1.47 1.83 2.19 2.90 3.61
245 0.786 0.933 1.15 1.52 1.89 2.26 3.00 3.73
250 0.8 I2 0.964 1.19 1.57 1.95 2.33 3.09 3.85
255 0.838 0.995 1.23 1.62 2.02 2.41 3.19 3.98
260 0.865 1 .O3 I .27 1.67 2.08 2.48 3.29 4.10
265 0.891 1 .O6 1.31 1.73 2.14 2.56 3.40 4.23
270 0.919 1 .O9 1.35 I .78 2.21 2.64 3.50 4.36
275 0.946 1.12 1.39 I .83 2.27 2.72 3.60 4.49
280 0.974 1.16 1.43 I .89 2.34 2.80 3.71 4.62
285 I .o02 1.19 1.47 1.94 2.41 2.88 3.82 4.75
290 1.030 1.22 1.51 1.99 2.48 2.96 3.92 4.89
295 I .O58 I .26 I .55 2.05 2.54 3.04 4.03 5.02
300 I .O87 I .29 1.60 2.1 I 2.61 3.12 4.14 5.16
305 1.1 16 I .33 1.64 2.16 2.68 3.21 4.25 5.30
310 1.146 I .36 1.68 2.22 2.75 3.29 4.36 5.44
315 1.175 I .40 1.73 2.28 2.83 3.38 4.48 5.58
320 1.205 1.43 1.77 2.33 2.90 3.46 4.59 5.72
325 1.236 I .47 1.81 2.39 2.97 3.55 4.71 5.86
330 I .266 1.50 1.86 2.45 3.04 3.64 4.82 6.01
335 I .297 1.54 I .90 2.51 3.12 3.73 4.94 6. I5
340 1.328 I .58 I .95 2.57 3.19 3.82 5.06 6.30
345 1.360 1.61 2.00 2.63 3.27 3.91 5.18 6.45
350 1.391 1.65 2.04 2.69 3.34 4.00 5.30 6.60
355 1.423 I .69 2.09 2.76 3.42 4.09 5.42 6.75
360 I .455 1.73 2.14 2.82 3.50 4.18 5.54 6.91
365 1.488 1.77 2.19 2.88 3.58 4.27 5.67 7.06
370 1.521 1.81 2.23 2.94 3.66 4.37 5.79 7.22
375 1.554 I .84 2.28 3.01 3.73 4.46 5.92 7.37
380 1.587 1.88 2.33 3.07 3.81 4.56 6.05 7.53
385 1.621 I .92 3.38 3.14 3.90 4.66 6.17 7.69
390 I .65 I .96 2.43 3.20 3.98 4.75 6.30 7.85
395 I .69 2.00 2.48 3.27 4.06 4.85 6.43 8.0 I

237
Table 7. I 1 continued

Upper- Discharge in m3/s for flumes of various throat widths

head h,
"1( 10 12 15 20 25 30 40 50
feet feet feet feet feet feet feet feet
400 1.72 2.04 2.53 3.34 4.14 4.95 6.56 8.17
405 1.76 2.09 2.58 3.40 4.22 5.05 6.69 8.34
410 I .79 2.13 2.63 3.47 4.31 5.15 6.83 8.50
41 5 1.83 2.17 2.68 3.54 4.39 5.25 6.96 8.67
420 1.86 2.21 2.74 3.61 4.48 5.35 7.10 8.84
425 1.90 2.25 2.79 3.68 4.56 5.45 7.23 9.01
430 1.93 2.30 2.84 3.74 4.65 5.56 7.37 9.18
435 I .97 2.34 2.89 3.81 4.74 5.66 7.51 9.35
440 2.01 2.38 2.95 3.89 4.82 5.76 7.64 9.52
445 2.04 2.43 3.00 3.96 4.91 5.87 7.78 9.69
450 2.08 2.47 3.05 4.03 5.00 5.98 7.92 9.87
455 2.12 2.51 3.1 I 4.10 5.09 6.08 8.06 10.0
460 2.15 2.56 3.16 4.17 5.18 6.19 8.21 10.2
465 2.19 2.60 3.22 4.24 5.27 6.30 8.35 10.4
470 2.23 2.65 3.27 4.32 5.36 6.41 8.49 10.6
475 2.27 2.69 3.33 4.39 5.45 6.52 8.64 10.8
480 2.31 2.74 3.39 4.47 5.54 6.63 8.79 10.9
485 2.34 2.78 3.44 4.54 5.64 6.74 8.93 11.1
490 2.38 2.83 3.50 4.62 5.73 6.85 9.08 11.3
495 2.42 2.88 3.56 4.69 5.82 6.96 9.23 11.5
500 2.46 2.92 3.62 4.77 5.92 7.07 9.38 11.7
505 2.50 2.97 3.67 4.84 6.01 7.19 9.53 11.9
510 2.54 3.02 3.73 4.92 6.1 1 7.30 9.68 12.1
515 2.58 3.06 3.79 5.00 6.20 7.42 9.83 12.2
520 2.62 3.1 I 3.85 5.08 6.30 7.53 9.99 12.4
525 2.66 3.16 3.91 5.15 6.40 7.65 10.1 12.6
530 2.70 3.21 3.97 5.23 6.50 7.76 10.3 12.8
535 2.74 3.26 4.03 5.31 6.59 7.88 10.5 13.6
540 2.78 3.31 4.09 5.39 6.69 8.00 10.6 13.0
550 2.87 3.40 4.21 5.55 6.89 8.24 10.9 13.6
560 2.95 3.50 4.33 5.71 7.09 8.48 11.2 14.0
570 3.04 3.60 4.46 5.88 7.30 8.72 11.6 14.4
580 3.12 3.71 4.58 6.04 7.50 8.97 11.9 14.8
590 3.21 3.81 4.71 6.21 7.71 9.22 12.2 15.2
600 3.30 3.91 4.84 6.38 7.92 9.47 12.6 15.6
610 3.38 4.02 4.97 6.55 8.13 9.72 12.9 16.1
620 3.47 4.12 5.10 6.73 8.35 9.98 13.2 16.5
630 3.56 4.23 5.23 6.90 8.57 10.2 13.6 16.9
640 3.65 4.34 5.37 7.08 8.78 10.5 13.9 17.3
650 3.75 4.45 5.50 7.25 9.00 10.8 14.3 17.8
660 3.84 4.56 5.64 7.43 9.23 11.0 14.6 18.2
670 3.93 4.67 5.77 7.61 9.45 11.3 15.0 18.7
680 4.03 4.78 5.91 7.80 9.68 11.6 15.3 19.1
690 4.12 4.89 6.05 7.98 9.91 11.8 15.7 19.6
700 4.22 5.01 6.19 8.17 10.1 12.1 16.1 20.0
710 4.3 I 5.12 6.34 8.35 10.4 12.4 16.4 20.5
720 4.41 5.24 6.48 8.54 10.6 12.7 16.8 20.9
730 4.51 5.35 6.62 8.73 10.8 13.0 17.2 21.4
740 4.61 5.47 6.77 8.93 11.1 13.2 17.6 21.9
750 4.71 5.59 6.92 9.12 11.3 13.5 17.9 22.3
760 4.8 I 5.71 7.06 9.3 I 11.6 13.8 18.3 22.8
770 4.91 5.83 7.21 9.51 11.8 14.1 18.7 23.3
780 5.01 5.95 7.36 9.71 12.1 14.4 19.1 23.8
790 5.12 6.08 7.52 9.91 12.3 14.7 19.5 24.3
800 5.22 6.20 7.67 10.1 12.6 15.0 19.9 24.8
810 5.33 6.32 7.82 10.3 12.8 15.3 20.3 25.3
820 5.43 6.45 7.98 10.5 13.1 15.6 20.7 25.8
830 5.54 6.58 8.13 10.7 13.3 15.9 21.1 26.3
840 5.65 6.70 8.29 10.9 13.6 16.2 21.5 26.8
850 5.75 6.83 8.45 11.1 13.8 16.5 21.9 27.3
860 5.86 6.96 8.61 11.4 14.1 16.8 22.3 27.8
870 5.97 7.09 8.77 11.6 14.4 17.2 22.8 . 28.3
880 6.08 7.22 8.93 11.8 14.6 17.5 23.2 28.9
890 6.19 7.35 9.10 12.0 14.9 17.8 23.6 29.4

238
Table 7.1 1 continued

Upper- Discharge in m3/s for flumes of various throat widths

head ha
"(1 10 12 15 - 20 25 30 40 50
feet feet feet feet feet feet feet feet

900 6.31 7.48 9.26 12.2 15.2 18.1 24.0 29.9

910 6.42 7.62 9.42 12.4 15.4 18.4 24.4 30.5
920 6.53 7.75 9.59 12.6 15.7 18.8 24.9 31.0
930 6.64 7.89 9.76 12.9 16.0 19.1 25.3 31.5
940 6.76 8.02 9.93 13.1 16.2 19.4 25.8 32.1
950 6.87 8.16 10.1 13.3 16.5 19.8 26.2 32.6
960 6.99 8.30 10.3 13.5 16.8 20. I 26.6 33.2
970 7.1 I 8.44 10.4 13.8 17.1 20.4 27. I 33.7
980 7.23 8.58 10.6 14.0 17.4 20.8 27.5 34.3
990 7.34 8.72 10.8 14.2 17.7 21.1 28.0 34.8
1O00 7.46 8.86 11.0 14.4 17.9 21.4 28.4 35.4
I010 7.58 9.00 11.1 14.7 18.2 21.8 28.9 36.0
I020 7.70 9.14 11.3 14.9 18.5 22.1 29.3 36.5
1030 7.82 9.29 11.5 15.1 18.8 22.5 29.8 37.1
I040 7.95 9.43 11.7 15.4 19.1 22.8 30.3 37.7
1050 8.07 9.58 11.8 15.6 19.4 23.2 30.7 38.3
I060 8.19 9.72 12.0 15.9 19.7 23.5 31.2 38.9
1070 9.87 12.2 16.1 20.0 23.9 31.7 39.5
1080 10.0 12.4 16.3 20.3 24.2 32.2 40.1
1090 10.2 12.6 16.6 20.6 24.6 32.6 40.6
1100 10.3 12.8 16.8 20.9 25.0 33.1 41.2
Ill0 10.5 13.0 17.1 21.2 25.3 33.6 41.8
1120 10.6 13.1 17.3 21.5 25.7 34.1 42.4
1130 10.8 13.3 17.6 21.8 26. I 34.6 43. I
I I40 10.9 13.5 17.8 22.1 26.4 35.1 43.7
I150 11.1 13.7 18.1 22.4 26.8 35.6 44.3
I I60 11.2 13.9 18.3 22.7 27.2 36.1 44.9
I I70 11.4 14.1 18.6 23. I 27.6 36.5 45.5
I180 11.5 14.3 18.8 23.4 27.9 37.0 46.1
I I90 11.7 14.5 19.1 23.7 28.3 37.6 46.8
I200 11.9 14.7 19.3 24.0 28.7 38. I 47.4
1210 12.0 14.9 19.6 24.3 29.1 38.6 48.0
1220 12.2 15.1 19.9 24.7 29.5 39. I 48.7
I230 12.3 15.3 20.1 25.0 29.9 39.6 49.3
I240 12.5 15.5 20.4 25.3 30.2 40.1 50.0
1250 12.7 15.7 20.7 25.6 30.6 40.6 50.6
1260 12.8 15.9 20.9 26.0 31.0 41.1 51.3
1270 13.0 16.1 21.2 26.3 31.4 41.7 51.9
I280 13.1 16.3 21.4 26.6 31.8 42.2 52.6
1290 13.3 16.5 21.7 27.0 32.2 42.7 53.2
1300 13.5 16.7 22.0 27.3 32.6 43.3 53.9
1310 13.6 16.9 22.3 27.6 33.0 43.8 54.5
1320 13.8 17.1 22.5 28.0 33.4 44.3 55.2'
1330 14.0 17.3 22.8 28.3 33.8 44.9 55.9
1340 14.1 17.5 23.1 28.7 34.2 45.4 56.6
I350 14.3 17.9 23.4 29.0 34.7 46.0 57.2
1360 14.5 17.9 23.6 29.3 35. I 46.5 57.9
I370 14.7 18.1 23.9 29.7 35.5 47.0 58.6
1380 18.3 24.2 30.0 35.9 47.6 59.3
I390 18.6 24.5 30.4 36.3 48.2 60.0
I400 18.8 24.8 30.7 36.7 48.7 60.7
1410 19.0 25.0 31.1 37.2 49.3 61.4
I420 19.2 25.3 31.4 37.6 49.8 62. I
1430 19.4 25.6 31.8 38.0 50.4 62.8
1440 19.6 25.9 32.2 38.4 51.0 63.5
1450 19.9 26.2 32.5 38.9 51.5 64.2
I460 20. I 26.5 32.9 39.3 52.1 64.9
1470 20.3 26.8 33.2 39.7 52.7 65.7
1480 20.5 27.1 33.6 40. I 53.2 66.3
I490 20.7 27.4 34.0 40.6 53.8 67.0
I500 21.0 27.6 34.3 41.0 54.4 67.7
1510 21.2 27.9 34.7 41.5 55.0 68.5
1420 21.4 28.2 35.1 41.9 55.6 69.2
1530 21.6 28.5 35.4 42.3 56. I 69.9

239
Table 7. I 1 continued

Upper- Discharge in m3/s for flumes of various throat widths

head ha
(") 10 12 15 20 25 30 40 50
feet feet feet feet feet feet feet feet

1540 21.9 28.8 35.8 42.8 56.7 70.7

1550 22. I 29. I 36.2 43.2 57.3 71.4
1560 22.3 29.4 36.5 43.7 57.9 72.1
1570 22.6 29.7 36.9 44. I 58.5 72.9
1580 22.8 30.0 37.3 44.6 59. I 73.6
1590 23.0 30.3 37.7 45.0 59.7 74.4
1600 23.2 30.7 38. I 45.5 60.3 75.1
1610 23.5 31.0 38.4 45.9 60.9 75.9
1620 23.7 31.3 38.8 46.4 61.5 76.6
1630 24.0 31.6 39.2 46.9 62. I 77.4
I640 24.2 31.9 39.6 47.3 62.7 78. I
1650 24.4 32.2 40.0 47.8 63.4 78.9
I660 24.7 32.5 40.4 48.2 64.0 79.7
I670 24.9 32.8 40.8 48.7 64.6 80.4
1680 33. I 41.1 49.2 65.2 81.2
1690 33.5 41.5 49.6 65.8 82.0
1700 33.8 41.9 50. I 66.5 82.8
1710 34.1 42.3 50.6 67. I 83.5
1720 34.4 42.7 51.1 67.7 84.3
1730 34.7 43. I 51.5 68.3 85.1
1740 35.1 43.5 52.0 69.0 85.9
1750 35.4 43.9 52.5 69.6 86.7
I760 35.7 44.3 53.0 70.2 87.5
1770 36.0 44.7 53.5 70.9 88.3
1780 36.4 45. I 53.9 71.5 89. I
1790 36.7 45.5 54.4 72.2 89.9
1800 37.0 45.9 54.9 12.8 90.7
1810 37.3 46.4 55.4 73.5 91.5
1820 37.7 46.8 55.9 74. I 92.3

240
7.4.3 Submerged flow

When the ratio of gauge reading h, to ha exceeds the limits of 0.60 for 3-, 6-, and
9-in flumes, 0.70 for 1- to 8-ft flumes and 0.80 for 10- to 50-ft flumes, the modular
flume discharge is reduced due to submergence. The non-modular discharge of Par-
shall flumes equals
Qs = Q-QE (7-5)
where Q equals the modular discharge (Tables 7.5 to 7.11) and QEis the reduction
on the modular discharge due to submergence.
The diagrams in Figures 7.10 to 7.16 give the corrections, QE,for submergence for
Parshall flumes of various sizes. The correction for the 1-ft flume is made applicable
to the 1.5-ft up to 8-ft flumes by multiplying the correction QEfor the I-ft flume by
the factor given below for the particular size of the flume in use.

Size of flume
correction
b, in ft b,inm factor

1 0.3048 1.o
1.5 0.4572 1.4
2 0.6096 I .8
3 0.9144 2.4
4 1.2191 3.1
5 I S240 3.7
6 1.8288 4.3
7 2. I336 4.9
8 2.4384 5.4

Similarly, the correction for the 10-ft flumes is made applicable to the larger flumes
by multiplying the correction for the 10-ft flume by the factor given below for the
particular flume in use.

Size of flume
correction
b, in ft b,in m factor

IO 3.048 I .o
12 3.658 1.2
15 4.572 1.5
20 6.096 2.0
25 7.620 2.5
30 9.144 3.0
40 12.192 4.0
50 15.240 5.0

If the size and elevation of the flume cannot be selected to permit modular-flow opera-
tion, the submergence ratio h,/h, should be kept below the practical limit of 0.90,

24 1
LPSTREAM HEAD
ha in metres

Figure 7.10 Discharge correction for submerged flow; 1 Parshall flume

CORRECTKM 0. in L h

Figure 7.1 I Discharge correction for submerged flow; 2 Parshall flume

242
UPSTREAM HEAD
ho in metres
100

.?O

.so

.30

.10

.10

.O7

.O3

.O1

DI
0.1 0.1 0.3 0.5 O.? 1 1 3 5 7 10 ao 30 50 70 700

CORRECTION OEin L/s

Figure 7.12 Discharge correction for submerged flow; 3" Parshall flume
UPSTREAM HEAD
h.in
<O0

.I O

.50

.30

.?O

.lo

.O7

0 5

.O3

O1

01
0.1 0.a 0.3 0.5 o7 7 1 3 5 7 10 10 30 5 0 -70 100

CORRECTION O, in L/s

Figure 7.13 Discharge correction for submerged flow; 6 Parshall flume

243
UPSTREAM HEAD
ha in metres
too

.70

.50

.io

.ao

.lO

07

.o3

Q. R R W L FLUME
.oa

.o1
0.1 0.a 0.3 0.5 0.7 ? a i s 7 10 ao io so 70 im
CORRECTIOC( 0. in L/S

Figure 7.14 Discharge correction for submerged flow; 9 Parshall flume

Figure 7.15 Discharge correction for submerged flow; I Parshall flume, correction QE(m3/s)

244
 CORRECTON in m h

Figure 7. I6 Diagram for determining correction to be subtracted from free-discharge flow to obtain sub-
merged flow discharge through 1 0 Parshall flumes

since the flume ceases to be a measuring device if submergence exceeds this limit. It
is recommended to use a long-throated flume (Section 7.1) instead of a non-modular
Parshall flume.
As mentioned, turbulence in the relatively deep and narrow throat of the very small
flumes makes the h,-gauge difficult to read. If an h,-gauge is used under submerged
flow conditions, the h,-readings should be converted to h,-readings with the aid of
Figure 7.17, and the converted h,-values are then used to determine the submerged
discharge with the aid of Figures 7.10 to 7.14.

7.4.4 Accuracy of discharge measurement

The error in the modular discharge read from the Tables 7.5 to 7.1 1 is expected to
be about 3%. Under submerged flow conditions the error in the discharge becomes
greater, until at 90% submergence the flume ceases to be a measuring device. The
method by which this discharge error is to be combined with errors in h,, h,, and
the flume dimensions are shown in Annex 2.

7.4.5 Loss of head through the flume

The size and elevation of the crest of the flume depend on the available loss of head
through the flume Ah( N AH). Since for the Parshall flume h, and h, are measured
at rather arbitrary locations, the loss of head through the flume Ah is not equal to

245
 . -
o. 3(

OX

o 10

O
O o 10 020 030
v HEAO (metres)

Figure 7.17 Relationship of h, and h b gauges for I", 2" and 3" Parshall flumes for submergences greater
than 50 percent (after Parshall 1953)

Figure 7.18 Section of Parshall flume

the difference between ha and hb but has a greater value (Figure 7.18). The head loss
Ah can be determined from the diagrams in Figures 7.19 and 7.20 for small and large
flumes. For very small flumes no data on Ah are available.

7.4.6 Limits of application

The limits of application of the Parshall measuring flumes essential for reasonable
accuracy are:
a. Each type of flume should be constructed exactly to the dimensions listed in Table
, 7.3;

246
b. The flume should be carefully levelled in both longitudinal and transverse direc-
tions;
c. The practical range of heads h, for each type of flume as listed in Table 7.4 is
recommended as a limit on h9;
d. The submergence ratio h,/h,should not exceed 0.90.

to 8' F SH4LL FL

30 85 80 C 2 0.05
L
0.10 o O
PERCENTAGE OF SUBMERGEN( 4EAD LOSS AH THROUGH FLUME in metres

Figure 7.19 Head-loss through Parshall flumes. I ' up to 8' Parshall flumes

247
 >to 5d PARSHALL FWMES

95 90 es eo 70 601 0.02 0.05 0.10 0.20 0.50 4

PERCENTAGE OF SUBMERGENC HEAD LOSS AH THROUGH FLUME in metres

Figure 7.20 Head-loss through Parshall flumes (10-50 feet wide)

7.5 H-flumes
7.5.1 Description

On natural streams where it is necessary to measure a wide range of discharges, a

structure with a V-type control has the advantage of providing a wide opening at
high flows so that it causes no excessive backwater effects, whereas at low flows its
opening is reduced so that the sensitivity of the structure remains acceptable. To serve
this purpose the U.S. Soil Conservation Service developed the H-type flume, of which
three geometrically different types are available. Their proportions are shown in Figure
7.21. They are:

HS-flumes

Of this small flume, the largest size has a depth D equal to 0.305 m (1 ft) and a
maximum capacity of 0.022 m3/s.

248
 H-flumes

Of this normal flume, the largest size has a depth D equal to 1.37 m (4.5 ft) and
a maximum capacity of 2.36 m3/s.

HL-flumes

The use of this large flume is only recommended if the anticipated discharge exceeds
the capacity of the normal H-flume. The largest HL-flume has a depth D equal to
1.37 m (4.5 ft) and a maximum capacity of 3.32 m3/s.

Since all three types are calibrated measuring devices, they should be constructed in
strict accordance with the drawings in Figure 7.21. It is especially important that the
slanting opening be bounded by straight sharp edges, that it has precisely the propor-
tional dimensions shown, and that it lies in a plane with an inclination of the exact
degree indicated in Figure 7.21. All cross sections of the flume should be symmetrical
about the longitudinal axis. The flume floor should be truly level. All plates should
be flat and should display no appreciable warp, dent, or other form of distortion.
All three types of flume should be located downstream of a rectangular approach
channel which has the same bottom width as the entrance of the flume, i.e., 1.05D
for the HS-flumes; 1.90D for the H-flumes; and 3.20D for the HL-flumes. The mini-
mum length of this approach channel is 2D. In practice, the flume sections are frequent-
ly constructed from sheet steel or other suitable material, while the approach section
is made of concrete, masonry, etc. The two parts should be given a watertight join
with the use of bolts and a gasket. The bolts should be suitable for both fastening
and levelling the flume. To prevent silting in the approach channel, its longitudinal
slope may vary from flat to about 0.02.
The upstream head ha is measured in the flume at a well defined location which
is shown separately for each flume in Figure 7.21. The piezometric head should be
measured in a separate well having a piezometer tap immediately above the flume
bottom. Since the head is measured at a location of accelerating flow and where stream-
lines are curved it is essential that the piezometer tap be located in its precise position
if accurate flow measurements are to be obtained.
To assure reliable head readings despite heavy sediment loads and the accompany-
ing sediment deposition in the flume, an 1-to-8 sloping floor was provided for H-
flumes. This false floor concentrates flows along the side wall having the stilling well
intake. Low flows can scour the sediment from the little channel formed along this
wall. The proportions of the sloping floor for the H-flume are given in Figure 7.22.
If the H-flume is equiped with a false floor the true flow rate differs slightly from
the figures given in Table 7.14. The percentage deviation in the free flow rate is shown
in Figure 7.23.

249
 k 1 . 0 5 D d k l . 5 D - 4

FRONT ELEVATION SIDE ELEVATION

F l . 1 0 4
H-FLUME
\
\
1.35D \
\
\
1 . 9 D 4 k 1 . 3 5 D d

w
0.3D
SIDE ELEVATION

0.50

FRONT ELEVATION w
0.25D
SIDE ELEVATION

Figure 7.21 Dimensions of the types HS-, H- and HL-flume (after Holtan, Minshall & Harrold 1962)

250
I sz

MOOU 3SlW 30 lN3Wd013h30

T7
Figure 7.23 Deviation in free flow rate through H-flumes with a sloping floor from rating tables 7.14 for
H-flumes with a flat floor (after Gwinn)

7.5.2 Evaluation of discharge

All three types of H-flumes have a rather arbitrary control while an upstream piezo-
metric head ha is measured at a station in the area of water surface drawdown. Under
these circumstances, the only accurate method of finding a head-discharge relationship
is by calibration in a hydraulic laboratory. Based on this calibration, an empirical
formula, expressing the discharge in m3/s as a function of the head ha in metres, could
be established of the general form
log Q = A + B log ha + C[log hal2 (7-6)
Values of the numbers A, B, and C appear in Table 7.12 for each flume type. Based
Table 7. I2 Data on three types of H-flumes

Flume Flume depth D Maximum Number in Equation 7-6 Rating

type capacity table
ft m m3/sx A B C
HS 0.4 ,122 2.27 -0.4361 +2.5151 +0.1379 7.13.a
HS 0.6 ,183 6.14 -0.4430 +2.4908 +0.1657 7.13.b
HS 0.8 ,244 12.7 -0.4410 +2.4571 +0.1762 7.13.c
HS 1.0 ,305 22.3 -0.4382 +2.4193 +0.1790 7.13.d

H 0.5 .I52 9.17 +0.0372 + 2.6629 +O. I954 7.14.a

H 0.75 ,229 26.9 f0.0351 +2.6434 +0.2243 7.14.b
H 1.o ,305 53.5 +0.0206 +2.5902 +0.2281 7.14.c
H 1.5 ,457 150 f0.0238 +2.5473 +0.2540 7.14.d
H 2.0 ,610 309 f0.0237 +2.49 18 +0.2605 7.14.e
H 2.5 ,762 542 +0.0268 + 2.4402 +0.2600 7.14.f
H 3.0 ,914 857 +0.0329 +2.3977 +0.2588 7.14.g
H 4.5 1.37 2366 +O.OS88 +2.3032 f0.2547 7.14.h

HL 3.5 1:07 2370 +0.3081 +2.3935 +0.2911 7.15.a

HL 4.0 1.22 3298 +0.3160 +2.3466 +0.2794 7.15.b

252
I

on Equation 7-6, calibration tables were prepared for each flume; see Tables 7.13 for
the HS-flumes, Table 7.14 for the H-flumes and Table 7.15 for the HL-flumes.
The error in the modular discharge given in Tables 7.13,7.14 and 7.15 may be expected
to be less than 3%. The method by which this error is to be combined with other
sources of error is shown in Annex 2.

7.5.3 Modular limit

The modular limit is defined as the submergence ratio h,/h, which produces a 1%
reduction from the equivalent modular discharge as calculated by Equation 7-6. Re-
sults of various tests showed that the modular limit for HS- and H-flumes is h,/h,
= 0.25, for HL-flumes this limit is 0.30. Rising tailwater levels cause an increase of
the equivalent upstream head ha at modular flow as shown in Fig.7.24. Because of
the complex method of calculating submerged flow, all HS- and H-flumes should be
installed with a submergence ratio of less than 0.25 (for HL-flumes 0.30).

e/. a

Submergence ratio hZ/h,

Submergence ratio b / h a

Figure 7.24a/b Influence of submergence on the modular head of HS-, H-, and HL-flumes. (Data on HL-
flumes based on personal communication, Gwinn 1977)

253
7.5.4 ' Limits of application

The limits of application of all H-flumes are:

a. The inside surface of the flume should be plane and smooth while the flume dimen-
sions should be in strict accordance with Figure 7.21.
b. The practical lower limit of h, is mainly related to the accuracy with which h, can
be determined. For heads less than 0.06 m, point gauge readings are required to
. obtain a reasonably accurate measurement. The lower limit of ha for each type
of flume can be read from Tables 7.13 to 7.15.
c. To obtain modular flow the submergence ratio h,/h, should not exceed 0.25.
d. To prevent water surface instability in the approach channel, the Froude number
Fr, = v,/(gA,/B>'/zshould not exceed 0.5.

Table 7.13a Free-flow discharge through 0.4 ft HS-flume in I/s

ha .O00 ,001. .O02 .O03 .O04 .O05 .O06 .O07 .O08 .O09
(ml

0.01 0.0 I2 0.015 0.017 0.020 0.024 0.027 0.031 0.035 0.039 0.044
0.02 0.049 0.054 0.059 0.065 0.071 0.077 0.084 0.091 0.098 o. 105
0.03 0.1 I3 0.121 0.130 O. 138 0.147 0.156 0.166 0.176 0.186 0.197
0.04 0.208 0.219 0.230 0.242 0.255 0.267 0.280 0.293 0.307 0.321
0.05 0.335 0.350 0.365 0.380 0.396 0.412 0.428 0.445 0.462 0.480
0.06 0.497 0.516 0.534 0.553 0.573 0.592 0.612 0.633 0.654 0.675
0.07 0.697 0.719 0.741 0.764 0.787 0.811 0.835 0.860 0.884 0.910
0.08 0.935 0.961 0.988 1.o1 1.04 1.07 1.10 1.13 1.16 1.19
0.09 1.21 1.25 1.28 1.31 1.34 1.37 I .40 1.44 1.47 1.50
0.10 1.54 I .57 1.61 I .64 1.68 1.71 1.75 1.79 1.83 1.87
0.11 1.90 1.94 1.98 2.02 2.06 2.10 2.15 2.19 2.23 2.27

Table 7.13b Free-flow discharge through 0.6 ft HS-flume in I/s

ha ,000 .O01 .O02 .O03 .O04 .O05 .O06 .O07 .O08 .O09
(m)

0.02 0.064 0.070 0.076 0.083 0.091 0.098 0.106 ' 0.114 O. 123 0.131
0.03 0.141 0.150 0.160 0.170 0.181 0.191 0.202 0.214 0.226 0.238
0.04 0.251 0.263 0.277 0.290 0.304 0.318 0.333 0.348 0.363 0.379
0.05 0.395 0.412 0.429 0.446 0.463 0.48 1 0.500 0.518 0.537 0.557
0.06 0.577 0.597 0.618 0.639 0.660 0.682 0.704 0.727 0.750 0.773
0.07 0.797 0.821 0.846 0.871 0.896 0.922 0.948 0.975 1.o0 I .O3
0.08 1.O6 I .O9 1.11 1.14 1.17 I .20 1.23 1.26 1.30 1.33
0.09 1.36 1.39 1.43 1.46 1.49 1.53 1.56 1.60 1.63 1.67
o. I O 1.71 1.74 1.78 1.82 1.86 1.90 1.93 1.97 2.01 2.06
0.11 2.10 2.14 2.18 2.22 2.27 2.31 2.35 2.40 2.44 2.49
0.12 2.53 2.58 2.63 2.68 2.72 2.77 2.82 2.87 2.92 2.97
0.13 3.02 3.07 3.12 3.18 3.23 3.28 3.34 3.39 3.45 3.50
O. I4 3.56 3.61 3.67 3.73 3.78 3.84 3.90 3.96 4.02 4.08
0.15 4.14 4.20 4.27 4.33 4.39 4.46 4.52 4.58 4.65 4.72
O. I6 4.78 4.85 4.92 4.98 5.05 5.12 5.19 5.26 5.33 5.40
0.17 5.47 5.55 5.62 5.69 5.77 5.84 5.92 5.99 6.07 6.14

254
OTable 7 . 1 3 ~Free-flow discharge through 0.8 ft HS-flume in I/s

ha .O00 .O01 .O02 .O03 .O04 .O05 ,006 ,007 ,008 ,009
( 4
0.03 0.227 0.239 0.252 0.266 0.280
0.04 0.294 0.308 0.324 0.339 0.355 0.371 0.388 0.404 0.422 0.440
0.05 0.458 0.476 0.495 0.514 0.534 0.554 0.574 0.595 0.617 0.638
0.06 0.660 0.683 0.706 0.729 0.753 0.777 0.802 0.827 0.852 0.878
0.07 0.904 0.931 0.958 0.986 1.01 1.04 1.07 1.10 1.13 1.16
0.08 1.19 1.22 1.25 1.29 1.32 1.35 1.38 1.42 1.45 1.49
0.09 1.52 1.56 1.59 1.63 1.67 1.70 1.74 1.78 1.82 1.86
0.10 1.90 1.94 1.98 2.02 2.06 2.10 2.15 2.19 2.23 2.28
0.11 2.32 2.37 2.41 2.46 2.50 2.55 2.60 2.65 2.69 2.74
0.12 2.79 2.84 2.89 2.94 2.99 3.05 3.10 3.15 3.20 3.26
0.13 3.31 3.37 3.42 3.48 3.54 3.59 3.65 3.71 3.77 3;83
0.14 3.89 3.95 4.01 4.07 4.13 4.19 4.25 4.32 4.38 4.45
0.15 4.51 4.58 4.64 4.71 4.77 4.84 4.91 4.98 5.05 5.12
0.16 5.19 5.26 5.33 5.40 5.48 5.55 5.62 5.70 5.77 5.85
0.17 5.92 6.00 6.08 6.15 6.23 6.31 6.39 6.47 6.55 6.63
0.18 6.71 6.79 6.88 6.96 7.04 7.13 7.21.... 7.30 7.39 7.47
0.19 7.56 7.65 7.74 7.82 7.91 8.00 8.10 8.19 8.28 8.37
0.20 8.47 8.56 8.65 8.75 8.84 8.94 9.04 9.14 9.23 9.33
0.21 9.43 9.53 9.63 9.73 9.83 9.94 10.0 10.1 10.2 10.4
0.22 10.5 10.6 10.7 10.8 10.9 11.0 11.1 11.2 11.3 11.4
0.23 11.5 11.7 11.8 11.9 12.0 12.1 12.2 12.3 12.5 12.6
0.24 12.7

Table 7.13d Free-flow discharge through 1 .O ft HS-flume in I/s

ha .O00 .OOI ,002 .O03 .O04 ,005 .O06 .O07 ,008 .O09
(m)
0.04 0.339 0.355 0.372 0.389 0.407 0.425 0.443 0.462 0.482 0.501
0.05 0.521 0.542 0.563 0.584 0.606 0.629 0.651 0.674 0.698 0.722
0.06 0.746 0.771 0.797 0.822 0.849 0.875 0.902 0.930 0.958 0.986
0.07 1.o2 1.04 1.07 1.10 1.14 1.17 1.20 1.23 1.26 1.30
0.08 1.33 1.36 1.40 1.43 1.47 1.50 1.54 1.58 1.61 1.65
0.09 I .69 1.73 1.77 ' 1.81 1.85 1.89 1.93 1.97 2.01 2.05
0.10 2.10 2.14 2.18 2.23 2.27 2.32 2.36 2.41 2.46 2.51
0.11 2.55 2.60 2.65 2.70 2.75 2.80 2.85 2.90 2.96 3.01
0.12 3.06 3.11 3.17 3.22 3.28 3.33 3.39 3.45 3.50 3.56
0.13 3.62 3.68 3.74 3.80 3.86 3.92 3.98 4.04 4.1 1 4.17
0.14 4.23 4.30 4.36 4.43 4.49 4.56 4.63 4.69 4.76 4.83
0.15 4.90 4.97 5.04 5.1 1 5.18 5.25 5.32 5.40 5.47 5.54
0.16 5.62 5.69 5.77 5.85 5.92 6.00 6.08 6.16 6.24 6.32
0.17 6.40 6.48 6.56 6.64 6.73 6.81 6.89 6.98 7.06 7.15
0.18 7.23 7.32 7.41 7.50 7.58 7.67 7.76 7.85 7.94 8.04
0.19 8.13 8.22 8.31 8.41 8.50 8.60 8.69 8.79 8.89 8.98
0.20 9.08 9.18 9.28 9.38 9.48 9.58 9.69 9.79 9.89 9.99
0.21 10.1 10.2 10.3 10.4 10.5 10.6 10.7 10.8 11.0 11.1
0.22 11.2 11.3 11.4 11.5 11.6 11.7 11.9 12.0 12.1 12.2
0.23 12.3 12.4 12.6 12.7 12.8 12.9 13.0 13.2 13.3 13.4
0.24 13.5 13.6 13.8 13.9 14.0 14.2 14.3 14.4 14.5 14.7
0.25 14.8 14.9 15.1 15.2 15.3 15.5 15.6 15.7 15.9 16.0
0.26 16.1 16.3 16.4 16.5 16.7 16.8 17.0 17.1 17.3 17.4
0.27 17.5 17.7 17.8 18.0 18.1 18.3 18.4 18.6 18.7 18.9
0.28 19.0 19.2 19.3 19.5 19.6 19.8 19.9 20.1 20.2 20.4
0.29 20.6 20.7 20.9 21.0 21.2 21.4 21.5 21.7 21.8 22.0
0.30 22.2 22.3

255
Table 7.14a Free-flow discharge through 0.5 ft H-flume in m3/s x (l/s)

.O00 .O02 .O04 .O06 .008

0.01 0.03 1 0.044 0.059 0.077 0.097

0.02 0.119 O. I45 0.172 0.203 0.236
0.03 0.272 0.311 0.353 0.398 0.446
0.04 0.497 0.551 0.609 0.669 0.733

0.05 0.801 0.871 0.946 1.o2 1.10

0.06 1.19 I .28 1.37 1.47 1.57
0.07 I .67 1.78 1.89 2.00 2.12
0.08 2.25 2.37 2.51 2.64 2.78
0.09 2.93 3.07 3.23 3.38 3.55

0.10 3.71 3.88 4.06 4.24 4.42

0.11 4.61 4.81 5.01 5.21 5.42
0.12 5.63 5.85 6.08 6.31 6.54
0.13 6.78 7.02 7.27 7.53 7.79
O. 14 8.05 8.32 8.60 8.88 9.17

Table 7.14b Free-flow discharge through 0.75 ft H-flume in (l/s)

.o00 .O02 .O04 .O06 .008

0.01 0.044 0.061 0.080 0.103 0.128

0.02 0.155 0.186 0.220 0.256 0.296
0.03 0.339 0.384 0.433 0.486 0.541
0.04 0.600 0.662 0.728 0.797 0.869
0.05 0.945 1.O3 1.11 I .20 I .29
0.06 1.38 1.48 1.58 1.69 1.80
0.07 1.91 2.03 2.15 2.28 2.41
0.08 2.54 2.68 2.83 2.97 3.13
0.09 3.28 3.44 3.61 3.78 3.95
0.10 4.13 4.31 4.50 4.70 4.89
0.11 5.10 5.30 5.52 5.73 5.96
0.12 6.18 6.42 6.65 6.90 7.14
0.13 7.40 7.65 7.92 8.19 8.46
0.14 8.74 9.03 9.32 9.61 9.91
0.15 10.2 10.5 10.9 11.2 11.5
0.16 11.8 12.2 12.5 12.9 13.2
0.17 13.6 14.0 14.4 14.7 15.1
0.18 15.5 15.9 16.3 16.7 17.2
0.19 17.6 18.0 18.5 18.9 19.4
0.20 19.8 20.3 20.8 21.2 21.7
0.21 22.2 22.7 23.2 23.7 24.2
0.22 24.8 25.3 25.8 26.4 26.9

256
Table 7 . 1 4 ~Free-flow discharge through 1 .O ft H-flume in m3/s x (I/s)

ha ha
(m) .O00 .O02 .O04 .O06 .O08 (m) ,000 .O02 .O04 .O06 .O08

0.00 0.15 11.0 11.3 11.7 12.0 12.3

0.01 0.127 0.157 0.16 12.7 13.1 13.4 13.8 14.2
0.02 0.190 0.226 0.265 0.308 0.236 0.17 14.5 14.9 15.3 15.7 16.1
0.03 0.403 0.455 0.511 0.571 0.634 0.18 16.5 16.9 17.4 17.8 18.2
0.04 0.701 0.771 0.845 0.922 1.00 ' 0.19 18.7 19.1 19.6 20.0 20.5

0.05 1.09 1.18 1.27 1.37 1.47 0.20 21.0 21.4 21.9 22.4 22.9
0.06 1.57 1.68 1.79 1.91 2.03 0.21 23.4 23.9 24.5 25.0 25.5
0.07 2.16 2.28 2.42 2.56 2.70 0.22 26.1 26.6 27.2 27.7 28.3
0.08 2.84 2.99 3.15 3.31 3.47 0.23 28.9 29.4 30.0 30.6 31.2
0.09 3.64 3.82 3.99 4.18 4.36 0.24 31.8 32.4 33.1 33.7 34.2

0.10 4.56 4.75 4.95 5.16 5.37 0.25 35.0 35.6 36.3 37.0 37.6
0.11 5.59 5.81 6.04 6.27 6.50 0.26 38.3 39.0 39.7 40.4 41.1
0.12 6.74 6.99 7.24 7.50 7.76 0.27 41.8 42.6 43.3 44.0 44.8
0.13 8.03 8.30 8.58 8.86 9.15 0.28 45.5 46.3 47.1 47.9 48.6
0.14 9.45 9.75 10.1 10.4 10.7 0.29 49.4 50.2 51.0 51.9 52.7
0.30 53.5

Table 7.14d Free-flow discharge through 1.5 ft H-flume in I/s

0.00 0.25 38.2 38.9 39.6 40.3 41.0

0.01 0.26 41.7 42.5 43.2 43.9 44.7
0.02 0.269 0.316 0.367 0.421 0.479 0.27 45.4 46.2 47.0 47.8 48.5
0.03 0.542 0.608 0.677 0.751 0.829 0.28 49.3 50.1 51.0 51.8 52.6
0.04 0.910 0.996 1.09 1.18 1.28 0.29 53.4 54.3 55.1 56.0 56.8

0.05 1.38 1.49 1.60 1.71 1.83 0.30 57.7 58.6 59.5 60.4 61.3
0.06 1.75 2.08 2.21 2.35 2.49 0.31 62.2 63.1 64.1 65.0 66.0
0.07 2.64 2.78 2.94 3.10 3.26 0.32 66.9 67.9 68.8 69.8 70.8
0.08 3.43 3.60 3.78 3.96 4.15 0.33 71.8 72.8 73.8 74.9 75.9
0.09 4.34 4.54 4.74 4.95 5.16 0.34 76.9 78.0 79.0 80.1 81.2

0.10 5.38 5.60 5.83 6.06 6.29 0.35 82.3 83.4 84.5 85.6 86.7
0.11. 6.54 6.78 7.04 7.30 7.56 0.36 87.8 89.0 90.1 91.3 92.4
0.12 7.83 8.10 8.38 8.67 8.96 0.37 93.6 94.8 96.0 97.2 98.4
0.13 9.25 9.55 9.86 10.2 10.5 0.38 99.6 101 102 103 105
0.1410.8 11.1 11.5 11.8 12.2 0.39 106 107 108 110 111

0.15 12.5 12.9 13.2 13.6 14.0 0.40 112 114 115 116 118
0.16 14.4 14.8 15.1 15.5 16.0 0.41 119 120 122 123 125
0.17 16.4 16.8 17.2 17.6 18.1 0.42 126 127 129 130 132
0.18 18.5 19.0 19.4 19.9 20.4 0.43 133 135 136 138 139
0.19 20.8 21.3 21.8 22.3 22.8 0.44 141 142 144 145 147

0.20 23.3 23.8 24.3 24.9 25.4 0.45 148 150

0.21 25.9 26.5 27.0 27.6 28.2
0.22 28.7 29.3 29.9 30.5 31.1
0.23 31.7 32.3 33.0 33.6 34.2
0.24 34.9 35.5 36.2 36.9 37.5

257
Table 7.14eFree-flow discharge through 2.0ft H-flumein m3/sx (I/s)

ha ha
(m) .O00 .O02 .O04 .O06 ,008 (m) .O00 .O02 .O04 ,006 ,008

0.00 0.30 61.9 62.9 63.8 64.7 65.7

0.01 0.31 66.6 67.6 68.6 69.5 70.5
0.02 0.469 0.535 0.606 0.32 71.5 72.5 73.5 74.6 75.6
0.03 0.681 0.760 0.844 0.932 1.02 0.33 76.6 77.7 78.7 79.7 80.8
0.04 1.12 1.22 1.33 1.44 1.55 0.34 81.9 83.0 84.1 85.2 86.3

0.05 1.67 1.79 1.92 2.05 2.19 0.35 87.5 88.6 89.7 90.9 92.0
0.06 2.33 2.48 2.63 2.79 2.95 0.36 93.2 94.4 95.6 96.7 97.9
0.07 3.11 3.29 3.46 3.64 3.83 0.37 99.2 1O0 I02 103 104
0.08 4.02 4.21 4.41 4.62 4.83 0.38 105 107 IO8 I09 1IO
0.09 5.04 5.27 5.49 5.72 5.96 0.39 112 I I3 1 I4 1 I6 1 I7

0.10 6.20 6.45 6.70 6.96 7.22 0.40 118 120 121 I23 124
0.1 I 7.49 7.76 8.04 8.33 8.62 0.41 125 127 128 130 131
0.12 8.91 9.22 9.52 9.84 10.2 0.42 132 I34 I35 I37 138
0.1310.5 10.8 11.1 11.5 11.8 0.43 140 141 I43 I44 146
0.14 12.2 12.5 12.9 13.3 13.7 0.44 147 148 150 152 154

0.15 14.0 14.4 14.8 15.2 15.6 0.45 155 157 I58 160 162
0.16 16.1 16.5 16.9 17.3 17.8 0.46 163 165 I67 I68 170
0.17 18.2 18.7 19.1 19.6 20.1 0.47 172 173 175 I77 179
0.18 20.5 21.0 21.5 22.0 22.5 0.48 180 182 I84 186 187
0.19 23.0 23.5 24.1 24.6 25.1 0.49 189 191 193 I95 196
0.20 25.7 26.2 26.8 27.3 27.9 0.50 198 200 202 204 206
0.21 28.5 29.1 29.7 30.2 30.9 0.51 208 210 21 1 213 215
0.22 31.5 32.1 32.7 33.3 34.0 0.52 217 219 221 223 225
0.23 34.6 35.3 35.9 36.6 37.3 0.53 227 229 231 233 235
0.24 38.0 38.7 39.4 40.1 40.8 0.54 237 240 242 244 246

0.25 41.5 42.2 42.9 43.7 44.4 0.55 248 250 252 254 256
0.26 45.2 46.0 46.7 47.5 48.3 0.56 259 261 263 265 267
0.27 49.1 50.0 50.7 51.5 52.3 0.57 270 272 274 276 279
0.28 53.2 54.0 54.9 55.7 56.6 0.58 281 283 286 288 290
0.29 57.5 58.3 59.2 60.1 61.0 0.59 293 295 297 300 302
0.60 305 307 309

258
Table 7.14f Free-flow discharge through 2.0 ft H-flume in m3/s x (l/s)

ha ha
(m) ,000 .O02 ,004 .O06 .O08 (m) .O00 .O02 .O04 .O06 .O08

0.00 0.40 125 126 128 129 131

0.01 0.41 132 134 I35 136 138
0.02 0.649 0.732 0.42 139 141 142 144 145
0.03 0.820 0.912 1.01 1.11 1.22 0.43 147 149 150 152 153
0.04 1.33 1.45 1.57 1.69 1.82 0.44 155 156 I58 160 161

0.05 1.96 2.10 2.25 2.40 2.55 0.45 163 165 166 168 169
0.06 2.71 2.88 3.05 3.23 3.41 0.46 171 173 175 176 178
0.07 3.59 3.78 3.98 4.18 4.39 0.47 180 181 183 185 187
0.08 4.60 4.82 5.04 5.27 5.51 0.48 189 I90 192 194 196
0.09 5.75 5.99 6.24 6.50 6.76 0.49 198 I99 20 1 203 205

0.10 7.02 7.30 7.58 7.86 8.15 0.50 207 209 21 1 213 215
0.11 8.44 8.75 9.05 9.36 9.68 0.51 216 218 220 222 224
0.12 10.0 10.3 10.7 11.0 11.4 0.52 226 228 230 232 234
0.13 11.7 12.1 12.4 12.8 13.2 0.53 236 239 24 1 243 245
0.14 13.6 14.0 14.4 14.8 15.2 0.54 247 249 25 1 253 255

0.15 15.6 16.0 16.4 16.9 17.3 0.55 257 260 262 264 266
0.16 17.6 18.2 18.7 19.1 19.6 0.56 268 27 1 273 275 277
0.17 20.1 20.6 21.1 21.6 22.1 0.57 280 282 284 286 289
0.18 22.6 23.1 23.6 24.2 24.7 0.58 291 293 296 298 30 I
0.19 25.2 25.8 26.4 26.9 27.5 0.59 303 305 308 310 313

0.20 28.1 28.7 29.2 29.8 30.5 0.60 315 317 320 322 325
0.21 31.1 31.7 32.3 33.0 33.6 0.61 327 330 332 335 337
0.22 34.2 34.9 35.6 36.2 36.9 0.62 340 343 345 348 350
0.23 37.6 38.4 39.0 39.7 40.4 0.63 353 355 358 361 363
0.24 41.1 41.9 42.6 43.4 44.1 0.64 366 369 371 374 377

0.25 44.9 45.6 46.4 47.2 48.0 0.65 380 382 385 388 39 I
0.26 48.8 49.6 50.4 51.2 52.0 0.66 393 396 399 402 405
0.27 52.9 53.7 54.6 55.4 56.3 0.67 408 410 413 416 419
0.28 57.2 58.1 59.0 59.9 60.8 0.68 422 425 428 43 I 434
0.29 61.7 62.7 63.5 64.5 65.4 0.69 437 440 443 446 449

0.30 66.4 67.3 68.3 69.3 70.3 0.70 452 455 458 46 1 464
0.31 71.3 72.3 73.3 74.3 75.3 0.71 467 470 4 74 477 480
0.32 76.4 77.4 78.5 79.5 80.6 0.72 483 486 489 493 496
0.33 81.7 82.8 83.9 85.0 86.1 0.73 499 502 506 509 512
0.34 87.2 88.3 89.5 90.6 91.8 0.74 515 519 522 525 529
0.75 532 535 539 542
0.35 93.0 94.1 95.3 96.5 97.7
0.36 98.9 100 101 102 104
0.37 105 106 108 109 110
0.38 112 113 114 115 117
0.39 118 119 121 122 .I24

259
Table 7.14g Free-flow discharge through 3.0 ft H-flume in m3/s x (I/s)

ha ha
(4 .O00 .O02 .O04 .O06 ,008 (m) .O00 .O02 .O04 .O06 .O08
-
0.00 0.50 216 218 220 222 224
0.01 0.51 226 228 230 232 234
0.03 0.959 1.06 1.18 1.29 1.41 0.53 246 248 25 1 253 255
0.04 1.54 1.67 1.81 1.95 2.09 0.54 257 259 26 1 263 266
0.05 2.25 2.40 2.57 2.74 2.91 0.55 268 270 272 274 277
0.06 3.09 3.27 3.46 3.66 3.86 0.56 279 28 1 283 286 288
0.07 4.06 4.28 4.49 4.72 4.95 0.57 290 293 295 297 300
0.08 5.18 5.42 5.66 5.92 6.17 0.58 302 304 307 309 312
0.09 6.43 6.70 6.98 7.26 7.54 0.59 314 317 319 32 1 324
0.10 7.83 8.13 8.44 8.75 9.06 0.60 326 329 33 1 334 336
0.1 1 9.38 9.71 10.0 10.4 10.7 0.61 339 34 I 344 347 349
0.12 11.1 11.4 11.8 12.2 12.5 0.62 352 354 357 360 362
0.13 12.9 13.3 13.7 14.1 14.5 0.63 365 368 370 373 376
O. 14 14.9 15.4 15.8 16.2 16.7 0.64 378 38 1 384 387 389
0.15 17.1 17.6 18.0 18.5 19.0 0.65 392 395 398 400 403
0.16 19.4 19.9 20.4 20.9 21.4 0.66 406 409 412 415 418
0.17 21.9 22.4 23.0 23.5 24.0 0.67 420 423 426 429 432
0.18 24.6 25.1 25.7 26.3 26.8 0.68 435 438 441 444 447
0.19 27.4 28.0 28.6 29.2 29.8 0.69 450 453 456 459 462
0.20 30.4 31.1 31.7 32.3 33.0 0.70 465 468 47 1 475 478
0.21 33.6 34.3 35.0 35.6 36.3 0.71 481 484 487 490 494
0.22 37.0 37.7 38.4 39.1 39.8 0.72 497 500 503 506 510
0.23 40.5 41.3 42.0 42.8 43.5 0.73 513 516 519 523 526
0.24 44.3 45.1 45.8 46.6 47.4 0.74 529 533 536 539 543
0.25 48.2 49.0 49.8 50.7 51.5 0.75 546 550 553 556 560
0.26 52.3 53.2 54.0 54.9 55.8 0.76 563 567 570 574 577
0.27 56.6 57.5 58.4 59.3 60.2 0.77 581 584 588 592 595
0.28 61.2 62.1 63.0 64.0 64.9 0.78 599 602 606 610 613
0.29 65.9 66.8 67.8 68.8 69.8 0.79 617 620 624 628 632
0.30 70.8 71.8 72.8 73.8 74.9 0.80 635 639 643 647 650
0.31 75.9 77.8 78.0 79.1 80.2 0.81 654 658 662 666 669
0.32 81.2 82.3 83.4 84.5 85.7 0.82 673 677 68 1 685 689
0.33 86.8 87.9 89.1 90.2 91.4 0.83 693 697 70 1 705 709
0.34 92.5 93.7 94.9 96.1 97.3 0:84 713 717 72 1 725 729
0.35 98.5 99.7 101 102 103 0.85 733 737 74 I 745 749
0.36 105 106 107 109 110 0.86 753 757 762 766 770
0.37 111 112 114 115 116 0.87 774 778 783 787 79 1
0.38 118 119 120 122 123 0.88 795 800 804 808 813
0.39 125 126 127 129 130 0.89 817 82 I 826 830 835
0.40 132 133 135 136 138 0.90 839 843 848 852 857
0.41 139 141 142 144 145
0.42 147 148 150 151 153
0.43 154 156 158 159 161
0.44 163 164 166 167 169
0.45 171 173 174 176 178
0.46 179 181 183 185 186
0.47 188 190 192 194 195
0.48 197 199 201 203 205
0.49 207 208 210 212 214

260
Table 7.14h Free-flow discharge through 4.5 ft H-flume in m3/s x IO (I/s)

ha ha
(m) ,000 ,002 .O04 .O06 .O08 (m) .O00 .O02 .O04 .O06 .O08

0.00 0.40 152 154 I 55 157 159

0.01 0.41 160 162 164 165 167
0.02 0.42 169 170 172 174 176
0.03 1.39 1.53 1.68 1.84 2.00 0.43 177 179 181 183 184
0.04 2.17 2.35 2.53 2.72 2.9 I 0.44 186 188 190 192 193

0.05 3.12 3.32 3.53 3.76 3.98 0.45 195 197 199 20 1 203
0.06 4.22 4.46 4.70 4.95 5.21 0.46 205 207 208 210 212
0.07 5.48 5.75 6.02 6.31 6.60 0.47 214 216 218 220 222
0.08 6.90 7.20 7.52 7.83 8.16 0.48 224 226 228 230 232
0.09 8.49 8.82 9.17 9.52 9.88 0.49 234 236 238 240 243

0.10 10.2 10.6 11.0 11.4 11.8 0.50 245 247 249 25 1 253
0.11 12.2 12.6 13.0 13.4 13.8 0.51 255 257 260 262 264
0.12 14.3 14.7 15.1 15.6 16.1 0.52 266 268 27 I 273 275
0.13 16.5 17.0 17.5 18.0 18.5 0.53 277 280 282 284 287
0.14 19.0 19.5 20.0 20.5 21.0 0.54 289 29 1 294 296 298

0.15 21.6 22.1 22.7 23.2 23.8 0.55 301 303 305 308 310
0.16 24.4 25.0 25.6 26.2 26.8 0.56 313 315 317 320 322
0.17 27.4 28.0 28.6 29.2 30.0 0.57 325 327 330 332 335
0.18 30.5 31.2 31.9 32.5 33.2 0.58 337 340 343 345 348
0.19 33.9 34.6 35.3 36.0 36.7 0.59 350 353 355 358 361

0.20 37.4 38.2 38.9 39.7 40.4 0.60 363 366 369 371 375
0.21 41.2 42.0 42.7 43.5 44.3 0.61 377 380 382 385 388
0.22 45.1 45.9 46.8 47.6 48.4 0.62 390 393 396 399 402
0.23 49.3 50.1 51.0 51.8 52.7 0.63 405 407 410 413 416
0.24 53.6 54.5 55.4 56.3 57.2 0.64 419 422 425 427 430

0.25 58.1 59.1 60.0 61.0 61.9 0.65 433 436 439 442 445
0.26 62.9 63.9 64.8 65.8 66.8 0.66 448 45 1 454 457 460
0.27 67.8 68.9 69.9 70.9 72.0 0.67 463 466 470 473 476
0.28 73.0 74.1 75.1 76.2 77.3 0.68 479 482 485 488 49 1
0.29 78.4 79.5 80.6 81.7 82.8 0.69 495 498 50 1 504 507

0.30 84.0 85.1 86.3 87.4 88.6 0.70 511 514 517 520 524
0.31 89.8 91.0 92.2 93.4 94.6 0.71 527 530 534 537 540
0.32 95.8 97.0 98.3 99.5 101 0.72 544 547 55 I 554 557
0.33 102 103 105 106 107 0.73 561 564 568 57 1 575
0.34 109 110 Ill 113 1 I4 0.74 578 582 585 589 592

0.35 115 117 118 119 121

0.36 122 124 125 126 128
0.37 129 131 132 134 135
0.38 137 138 140 141 143
0.39 144 146 148 149 151

26 1
Table 7.14hFree-flow discharge through 4.5ft H-flumein m3/s x (I/s) (cont.)

ha ha
(m) .O00 .O02 .O04 ,006 ,008 (m) .O00 .O02 .O04 .O06 .008

0.75 596 599 603 606 610 1.05 1281 1287 1292 1299 i304
0.76 614 617 62I 625 628 1.06 1310 1316 1321 1327 1333
0.77 632 636 639 643 647 1.07 1339 1345 I350 I356 1362
0.78 650 654 658 662 666 1.08 1368 1374 I380 1386 1392
0.79 669 673 677 681 685 1.09 1398 1403 1409 1415 1421

0.80 689 693 696 700 704 1.10 1427 1434 1440 1446 1452
0.81 708 712 716 720 724 1.11 1458 1464 1470 1476 1482
0.82 728 732 736 740 744 1.12 1489 1495 1501 1507 1513
0.83 748 752 757 761 765 1.13 1520 1526 1532 1539 1545
0.84 769 773 777 781 786 1.14 1551 1558 I564 I570 1577

0.85 790 794 798 802 807 1.15 1583 1590 1596 I603 1609
0.86 811 815 820 824 828 1.16 1616 1622 1629 1635 1642
0.87 833 837 84I 846 850 1.17 1648 1655 1661 1668 1675
0.88 855 859 863 868 872 1.18 1681 1688 1695 1701 1708
0.89 877 881 886 890 894 1.19 1715 1722 1728 1735 1742

0.90 899 904 909 913 918 1.20 1749 1756 1763 I769 1776
0.91 922 927 932 936 941 1.21 1783 1790 1797 I804 1811
0.92 946 950 955 960 965 1.22 1818 1825 I832 I839 1846
0.93 969 974 979 984 988 1.23 1853 1860 I867 I875 1882
0.94 993 998 1003 1 O08 1013 1.24 1889 1896 1903 1910 1918

0.95 1018 1023 1028 1032 1037 1.25 1925 1932 1939 1947 1954
0.96 1042 1047 1052 1057 I062 1.26 1961 1969 1976 1983 1991
0.97 1068 1073 1078 1083 1088 1.27 1998 2006 2013 2020 2028
0.98 1093 1098 1 IO3 1108 1114 1.28 2035 2043 2050 2058 2066
0.99 1119 1I24 1 I29 1 I34 1 I40 1.29 2073 2081 2088 2096 2104

1.00 1145 1150 1 I56 1161 1 I66 1.30 2111 2119 2127 2134 2142
1.01 1172 1 I77 1 I82 1188 1 I93 1.31 2150 2158 2165 2173 2181
1.02 1198 1204 1209 1215 1220 1.32 2189 2197 2205 2212 2220
1.03 1226 123I 1237 1242 1248 1.33 2228 2236 2244 2252 2260
1.04 1253 1259 1265 I270 I276 1.34 2268 2276 2284 2292 2300

1.35 2308 2317 2325 2333 2341

1.36 2349 2357 2366

262
Table 7.15a Free-flow discharge through 3.5ft HL-flume in I/s (m3/s x

ha
(m) .O00 .O02 ,004 .O06 .008

0.05 4.86 5.19 5.52 5.86 6.22

0.06 6.58 6.95 7.34 7.73 8.14
0.07 8.55 8.98 9.41 9.86 10.32
0.08 10.79 I 1.27 11.75 12.25 12.77
0.09 13.29 13.82 14.36 14.92 15.48

0.10 16.06 16.65 17.24 17.85 18.47

0.1 I 19.11 19.75 20.40 21.07 21.75
0.12 22.44 23.14 23.85 24.57 25.31
0.13 26.05 26.81 27.58 28.36 29.16
0.14 29.96 30.78 31.61 32.45 33.31
0.15 34.17 35.05 35.94 36.84 37.76
0.16 38.69 39.63 40.58 41.54 42.52
0.17 43.51 44.51 45.53 46.55 47.59
0.18 48.65 49.71 50.79 51.88 52.99
0.19 54.10 55.23 56.38 57.53 58.70

0.20 59.89 61.08 62.29 63.52 64.75

0.21 66.00 * 67.27 68.54 69.83 71.14
0.22 72.45 73.79 75.13 76.49 77.86
0.23 79.25 80.65 82.06 83.49 84.93
0.24 86.39 87.86 89.34 90.84 92.36
0.25 93.88 95.42 96.98 98.55 100.14
0.26 101.73 103.35 104.98 106.62 108.28
0.27 109.95 1 I I .64 113.34 115.06 116.79
0.28 118.53 120.30 122.07 123.86 125.67
0.29 127.49 129.33 131.18 133.05 134.93

0.30 136.83 138.74 140.67 142.61 144.57

0.31 146.55 148.54 150.55 152.57 154.61
0.32 156.66 158.73 160.82 162.92 165.03
0.33 167.17 169.32 171.48 173.66 175.86
0.34 178.07 180.30 182.55 184.81 187.09
0.35 189.38 191.69 194.02 196.36 198.73
0.36 201.10 203.50 205.91 208.33 210.78
0.37 213.24 215.72 218.21 220.72 223.25
0.38 225.80 228.36 230.94 233.53 236.15
0.39 238.78 241.43 244.09 246.77 249.47

0.40 252.19 254.92 257.68 260.45 263.23

0.41 266.04 268.86 271.70 274.56 277.43
0.42 280.33 283.24 286.17 289.12 292.08
0.43 295.06 298.06 301 .O8 304. I2 307.18
0.44 310.25 313.34 316.45 319.58 322.73
0.45 325.89 329.07 332.28 335.50 338.74
0.46 341.99 345.27 348.57 351.88 355.21
0.47 358.56 361.93 365.32 368.73 372.16
0.48 375.60 379.07 382.35 386.05 389.58
0.49 393.12 396.68 400.26 403.86 407.48

0.50 411.12 414.77 418.45 422.15 425.86

0.51 429.60 433.35 437. I3 440.92 444.74
0.52 448.57 452.43 456.30 460. I9 464.11
0.53 468.04 472.00 475.97 479.96 483.98
0.54 488.01 492.07 496.14 500.24 504.35
0.55 508.49 512.65 516.82 521.02 525.24
0.56 529.48 533.74 538.02 542.32 546.64
0.57 550.98 555.34 559.73 564.13 568.56
0.58 573.00 577.47 58 I .96 586.47 59 I .o0
0.59 595.55 600.13 604.72 609.34 613.97

263
Table 7.15a (cont.) Free-flow discharge through 3.5 ft HL-flume in I/s (m3/s x 10-7

ha
(4 .O00 .O02 .O04 .O06 .008

0.60 618.63 623.3 I 628.01 632.74 637.48

0.61 642.25 647.03 651.84 656.67 661.53
0.62 666.40 67 I .30 676.22 681.16 686.12
0.63 691.10 696.1 I 701.14 706.19 71 1.26
0.64 716.35 72 I .47 726.61 73 I .77 736.95
0.65 742.16 747.39 752.64 757.91 763.21
0.66 768.52 773.87 779.23 784.61 790.02
0.67 795.46 800.91 806.39 81 1.89 817.41
0.68 822.96 828.52 834.12 839.73 845.37
0.69 851.03 856.71 862.42 868. I5 873.91

0.70 879.68 885.49 891.31 897.16 903.03

0.71 908.92 914.84 920.78 926.75 932.74
0.72 938.75 944.79 950.85 956.93 963.04
0.73 969. I7 975.33 981.51 987.71 993.94
0.74 1000.I9 1006.47 1012.77 1019.10 1025.44
0.75 1031.82 1038.22 1044.64 1051.08 1057.56
0.76 1064.05 1070.57 1077.12 1083.69 1090.28
0.77 1096.90 1103.54 1 I10.21 I 1 16.90 1123.62
0.78 1130.36 1137.13 1143.92 1150.74 1157.58
0.79 I 164.45 1171.34 1 178.26 1185.20 1192.17

0.80 1199.17 1206.19 1213.23 1220.30 1227.40

0.81 1234.52 1241.66 1248.83 1256.03 1263.25
0.82 1270.50 1277.78 1285.08 1292.40 1299.75
0.83 1307.13 1314.54 1321.97 1329.42 1336.90
0.84 1344.41 I35 I .94 1359.50 1367.09 1374.70
0.85 1382.34 1390.00 1397.69 1405.41 1413.15
0.86 1420.92 1428.72 1436.54 1444.39 1452.27
0.87 1460.17 1468.1O 1476.06 1484.04 1492.05
0.88 1500.09 1508.I5 1516.24 1524.36 1532.50
0.89 1540.67 1548.87 1557.10 1565.35 1573.63

0.90 158I .94 1590.27 1598.63 1607.02 1615.44

0.9 I 1623.88 1632.35 1640.85 1649.38 1657.93
0.92 1666.51 1675.12 1683.75 1692.42 1701.11
0.93 1709.83 1718.58 1727.35 1736.I6 1744.99
0.94 1753.85 1762.73 1771.65 1780.59 1789.56
0.95 1798.56 1807.59 1816.65 1825.73 1834.85
0.96 1843.99 1853.16 1862.35 1871.58 1880.84
0.97 1890.12 1899.43 1908.77 1918.14 1927.54
0.98 1936.97 1946.42 1955.91 1965.42 1974.96
0.99 1984.53 1994.I3 2003.76 2013.42 2023.1 I

I .o0 2032.82 2042.57 2052.35 2062.15 2071.98

1.01 2081.85 2091.74 2101.66 2111.61 2121.59
I .o2 2131.60 2141.64 2151.71 2161.81 2171.94
I .O3 2182.10 2192.28 2202.50 2212.75 2223.03
I o4
... 22zz...
__. 3.
1 2243.67
~~ .. 2254.04 2264.44 2274.86
I .o5 2285.32 2295.81 2306.33 2316.87 2327.45
I .O6 2338.06 2348.70 2359.37 2370.07

264
 Table 7.15b Free-flow discharge through 4 ft HL-flume in I/s (m3/s x

ha
(m) .o00 .O02 .O04 .O06 .008

0.05 5.38 5.73 6.10 6.48 6.86

0.06 7.26 1.67 8.09 8.52 8.96
0.07 9.41 9.88 10.35 10.84 1 1.34
0.08 I I .84 12.36 12.90 13.44 13.99
0.09 14.56 15.13 15.72 16.32 16.93

0.10 17.55 18.19 18.84 19.49 20.16

0.11 20.84 21.54 22.24 22.96 23.69
0.12 24.43 25.18 25.95 26.73 27.51
0.13 28.32 29.13 29.96 30.79 3 I .65
0.14 32.51 33.38 34.27 35.17 36.09
0.15 37.01 37.95 38.90 39.86 40.84
0.16 41.83 42.83 43.85 44.88 45.92
0.17 46.97 48.04 49.12 50.21 51.32
0.18 52.44 53.57 54.72 55.88 57.05
' 0.19 58.24 59.44 60.65 61.88 63.12

0.20 64.37 65.64 66.92 68.21 69.52

0.21 70.85 72.18 73.53 74.90 76.27
0.22 77.67 79.07 80.49 81.93 83.38
0.23 84.84 86.32 87.81 89.3 I 90.83
0.24 92.37 93.92 95.48 97.06 98.65
0.25 100.26 101.88 103.52 105.17 106.83
0.26 108.51 110.21 I 1 I .92 113.64 115.38
0.27 117.14 118.91 120.69 122.50 124.31
0.28 126.14 127.99 129.85 131.73 133.62
0.29 135.52 137.45 139.38 141.34 143.31

0.30 145.29 147.29 149.31 151.34 153.39

0.3 I 155.45 157.53 159.62 161.73 163.86
0.32 166.00 168.16 170.34 172.53 174.73
0.33 176.96 179.20 181.45 183.72 186.01
0.34 188.31 190.63 192.97 195.32 197.69
0.35 200.08 202.48 204.90 207.34 209.79
0.36 212.26 214.75 217.25 219.77 222.3 I
0.37 224.86 227.43 230.02 232.62 235.24
0.38 237.88 240.54 243.21 245.90 248.61
0.39 251.33 254.08 256.84 259.61 262.41
0.40 265.22 268.05 270.89 273.76 276.64
0.41 279.54 282.46 285.39 288.34 291.32
0.42 294.30 297.31 300.33 303.38 306.44
0.43 309.51 312.61 315.72 318.86 322.01
0.44 325.18 328.36 331.57 334.79 338.03
0.45 341.29 344.57 347.87 351.18 354.52
0.46 357.87 361.24 364.63 368.04 311.47
0.47 374.92 379.39 38 I .86 385.37 388.89
0.48 392.43 395.99 399.51 403.16 406.78
0.49 410.42 414.07 417.75 421.44 425.15
0.50 428.88 432.63 436.41 440.20 440.00
0.51 447.83 45 I .68 455.55 459.44 463.35
0.52 467.27 471.22 475.19 479.17 483.18
0.53 487.20 491.25 495.31 499.40 503.51
0.54 507.63 511.78 5 15.94 520. I3 524.33
0.55 528.56 532.81 537.07 541.36 545.67
0.56 550.00 554.34 558.71 563.10 567.51
0.57 57 I .94 576.39 580.87 585.36 589.87
0.58 594.40 598.96 603.53 608.13 612.75
0.59 617.39 622.04 626.72 63 I .43 636.15

265
Table 7.15b (cont.) Free-flow discharge through 4 ft HL-flume in l/s (m3/s x

ha
(m) .O00 .O02 .O04 .O06 .008
0.60 640.89 645.66 650.44 655.25 660.08
0.61 664.92 669.80 674.69 679.60 684.54
0.62 689.49 694.47 699.47 704.49 709.53
0.63 714.60 719.68 724.79 729.92 735.07
0.64 740.24 745.44 750.65 755.89 761.15
0.65 766.44 771.74 777.07 782.41 787.79
0.66 793.18 798.59 804.03 809.49 814.97
0.67 820.48 826.00 831.55 837.13 842.72
0.68 848.34 853.98 859.64 865.32 871.03
0.69 876.76 882.51 888.29 894.09 899.91
0.70 905.75 91 1.62 917.51 923.42 929.36
0.71 935.31 941.30 947.30 953.33 959.38
0.72 965.46 971.55 977.67 983.82 989.99
0.73 996.18 1002.39 1008.63 1014.89 1021.18
0.74 1027.49 1033.82 1040.18 1046.56 1052.96
0.75 1059.39 1065.84 1072.31 1078.81 1085.34
0.76 1091.88 1098.45 I105.05 1 1 11.67 11 18.31
0.77 I 1 24.98 1131.67 1138.38 1145.12 1151.89
0.78 1158.68 1165.49 1172.32 1179.19 1186.07
0.79 1192.98 1199.92 1206.88 1213.86 1220.87
0.80 1227.90 1234.96 1242.04 1249.15 1256.28
0.81 1263.44 1270.62 1277.82 1285.05 1292.31
0.82 1299.59 1306.90 1314.23 1321.59 1328.97
0.83 1336.37 1343.81 1351.26 1358.74 1366.25
0.84 1373.79 1381.34 1388.93 1396.54 1404.17
0.85 141 1.83 1419.52 1427.23 1434.96 1442.73
0.86 1450.52 1458.33 1466.I7 1474.04 1481.93
0.87 1489.84 1497.79 1505.76 1513.75 1521.77
0.88 1529.82 1537.89 1545.99 1554.12 1562.27
0.89 1570.44 1578.65 1586.88 1595.I3 1603.42
0.90 161 1.73 1620.06 1628.42 1636.81 1645.23
0.91 1653.67 1662.14 1670.63 1679.15 1687.70
0.92 1696.28 1704.88 1713.51 1722.16 1730.84
0.93 1739.55 1748.29 1757.05 1765.84 1774.66
0.94 1783.50 1792.37 1801.27 1810.20 1819.15
0.95 1828.I3 1837.14 1846.17 .I85523 1864.32
0.96 1873.44 1882.58 1891.75 1900.75 1910.18
0.97 1919.43 1928.71 1938.02 1947.36 1956.73
0.98 1966.12 1975.54 1984.99 1994.46 2003.97
0.99 2013.50 2023.06 2032.65 2042.26 205 I .9 I
I .o0 2061.58 2071.28 208 1 .O1 2090.76 2100.55
1.01 2110.36 2120.20 2130.07 2139.97 2149.90
I .o2 2159.85 2 169.84 2179.85 2189.89 2199.96
1 .O3 2210.06 2220.18 2230.34 2240.52 2250.73
1 .O4 2260.97 227 I .24 228 I .54 229 1.87 2302.23
1 .o5 2312.61 2323.03 2333.47 2343.95 2354.45
1 .O6 2364.98 2375.54 2386. I3 2396.75 2407.40
1 .O7 2418.07 2428.78 2439.52 2450.28 246 I .O8
1 .O8 247 I .90 2482.76 2493.64 2504.55 25 15.49
I .O9 2526.47 2537.47 2548.50 2559.56 2570.65
1.10 2581.77 2592.93 2604.1 1 2615.32 2626.56
1.11 2637.83 2649.13 2660.46 2671.82 2683.21
1.12 2694.63 2706.09 2717.57 2729.08 2740.62
1.13 2752.19 2763.80 2775.43 2787.09 2798.79
1.14 2810.5 I 2822.27 2834.O 5 2845.87 2857.72
1.15 2869.60 2881.50 2893.44 2905.41 2917.41
1.16 2929.45 2941.51 2953.60 2965.73 2977.88
1.17 2990.07 3002.29 3014.54 3026.8 I 3039. I3
1.18 3051.47 3063.84 3076.25 3088.68 3101.15
1.19 3113.65 3126.18 3 138.74 3151.33 3 163.96
I .20 3176.61 3189.30 3202.02 3214.77 3227.55
1.21 3240.37 3253.21 3266.09 3279.00 3291.94

266
 7.6 Selected list of references
Ackers, P. and A.J.M. Harrison 1963. Critical-depth flumes for flow measurement in open channels. Dept.
of Sci. &Industrial Research. Hydraulic Res. Sta., Hydr. Research Paper No. 5. London.
Balloffet, A. 1951. Critical flow meters (Venturi flumes). Proc. ASCE. 81, Paper 743.
Bennett, R.S. 1972. Cutthroat flume discharge relations. Thesis presented to Colorado State University,
Fort Collins, Col., in partial fulfillment of requirements for the degree of Master of Science.
Bos, M.G. 1985. Long-throated flumes and broad-crested weirs. Martinus Nijhoff Publishers, Dordrecht.
p 141.
Bos, M.G. and Y, Reinink 198I.Head loss over long-throated flumes. Journal of the Irrigation and Drainage
I
Division, ASCE, Vol. 107, IRI, pp.87-102.
Blau, E. 1960. Die modelmssige Untersuchung von Venturikanlen verschiedener Grsse und Form. Ver-
ffentlichungen der Forschungsanstalt fiir Schiffahrt, Wasser- und Grundbau 8. Akademie Verlag,
Berlin.
British Standards institution 1965. Methods of measurement of liquid flow in open channels. BS3680. Part
4. Weirs and flumes. 4A: Thin plate weirs and venturi flumes.
Caplan, F. 1963. Nomograph for Free Flow Discharge through a Parshall Flume. Water & Sewage Works.
Clemmens, A.J., J.A. Replogle & M.G. Bos 1987. Flume: a computer model for estimating flow rates
through long-throated measuring flumes, U.S. Dept. of Agriculture, ARS-57, p.64.
Eggleston, K.O. 1967. Effects of boundary geometry on critical and subcritical flow through measuring
flumes. Utah Water Research Laboratory. College of Engineering, Utah State University. Logan, Utah.
Engel, F.V.A.E. 1934. The venturi flume. The Engineer. London. Vol. 158. Aug. 3, pp. 104-107. Aug.
10,pp. 131-133.
Fane, A.C. 1927. Report on flume experiments on Sirhind Canal. Punjab Irrigation Branch. Paper 1 IO,
Punjab Eng. Congress.
Gwinn, Wendell R. and Donald A. Parsons. 1976. Discharge equations for HS, H, and H L flumes. J.
of the Hydraulics Div. ASCE, Vol. 102. No. H Y I., pp. 73-88.
Harvey, W.B. 1912. Harveys Irrigation Outlet. Punjab Irrigation Branch.
Holtan, H.N., N.E. Minchal and L.L. Harrold. 1962. Field manual for research in agricultural hydrology.
Soils & Water Conservation Research Div., Agricultural Research Service, Agricultural Handbook
No.224. Washington D.C.
Idelcik, I.E. 1969. Memento des pertes de charge. Collection du Centre de recherche et dessais de Chatou.
Eyrolles. Paris.
Inglis, C.C. 1929. Notes on standing wave flumes and flume meter falls. Public Works Dept. Technical
Paper No. 15. Bombay.
Jameson, A.H. 1930. The development of the venturi flume. Water & Water Engineering. March 20, pp.
105-107.
Khafagi, A. 1942. Der Venturikanal. Theorie und Anwendung. Diss. Driickerei AG. Gebr. Leemann &
Co., Zurich.
Parshall, R.L. 1945. Improving the Distribution of Water to Farmers by Use of the Parshall Measuring
Flume. Soil Conservation Service. Bull. 488. U.S. Dept. of Agriculture and Colorado Agricultural Experi-
ment Station. Colorado A and M College. Fort Collins, Colorado.
Parshall, R.L. 1950. Measuring Water in Irrigation Channels with Parshall Flumes and Small Weirs. Soil
Conservation Circular No. 843. U.S. Dept. of Agriculture.
Parshall, R.L. 1953. Parshall Flumes of Large Size. Bull. 426-A (Reprint of Bull. 386). Colorado Agric.
Experiment Sta., Colorado State University, Fort Collins, Colorado.
Peterka, A.J. 1965. Water Measurement Procedures. Irrigation Operators Workshop. Hydraulics Labora-
tory Report No. Hyd.-552. Bureau of Reclamation, November I5 (unpublished).
Robinson, A.R. 1957. Parshall Measuring Flumes of Small Sizes. Technical Bulletin No. 61. Agricultural
Experiment Station. Colorado State University. Fort Collins, Colorado.
Robinson, A.R., and A.R. Chamberlain. 1960. Trapezoidal flumes for open channel flow measurement.
Trans. Amer. Soc. Agri. Engnrs. Nr. 2, pp. 120-128.
Robinson, A.R. 1965. Simplified Flow Corrections for Parshall Flumes under Submerged Conditions. Civil
Engineering ASCE.
Robinson, A.R. 1968. Trapezoidal flumes for measuring flow in irrigation channels. Agr. Res. Service ARS
41-140. U.S. Dept. ofAgriculture. 15 pp.

267
Schuster,J.C.(Ed.). 1970. Water measurement procedures. Irrigation operators workshop. Report on REC-
OCE-70 pp.49.
Skogerboe, G.V., M.L. Hyatt, R.K. Anderson, and K.O. Eggleston 1967. Design and calibration of sub-
merged open channel flow measurement structures. Part 3: Cutthroat flumes. Report WG31-4. Utah
Water Res. Laboratory. College of Engineering. Utah State University. Logan, Utah.
Skogerboe, G.V. and M.L. Hyatt 1967. Analysis of submergence in flow measuring flumes. J. of the Hyd-
raulics Div., ASCE . Vol. 93, No. HY4, Proc. Paper 5348. pp. 183-200.
Skogerboe G.V., M.L. Hyatt, J.D. England and J.R. Johnson 1967. Design and Calibration of submerged
open channel flow measurement structuies. Part 2. Parshall flumes. Report WG 3/-3. Utah Water Re-
search Laboratory. College of Eng. Utah St. University. Logan, Utah.
Skogerboe, G.V., and M.L. Hyatt 1967. Rectangular cutthroat flow measuring flumes. J. of the Irrigation
and Drainage Div. ASCE, Vol. 93, No. IR4. pp. 1-33.
Skogerboe, G.V., R.S. Bennett, and W.R. Walker 1972. Generalized discharge relations for cutthroat
flumes. J. of the Irrig. and Drainage Div. ASCE, Vol. 98. No. IR4. pp. 569-583.
U.S. Department of the Interior 1967. U.S. Bureau of Reclamation: Water measurement manual. Second
Ed. Denver, Colorado, 327 pp.
Ven Te Chow 1959. Open channel hydraulics. McGraw-Hill Book Comp., New York. pp. 680.
Wells, E.A., and H.B. Gotaas 1956. Design of Venturi flumes in circular conduits. J. Sanit. Engng. Div.,
ASCE. Paper No. 2937. pp.749-77 I .

268
 A well defined opening in a plate or bulkhead, the top of which is placed well below
the upstream water level, is classified here as an orifice.

8.1 Circular sharp-edged orifice

8.1.1 Description
b

8.1.2 Determination of discharge

The basic head-discharge equations for orifice flow, according to Section 1.12, are

269
 B--7 I

WITH CLEAR
VINYL PLASTIC

,PLASTIC CEMENTED
TO 2 m m IRON
WITH GLASS SEALER
WINDOW DETAIL

CLEAR VINYL PLASTK

8 0 - 2 0 5 2 m m THICK
TO COVER SLOT IN
PLATE CEMENTED
TO PLTE WITH
GLASS SEALER.

SECTION B-B
ORIFICE DETAIL
SECTION A-A

Figure 8.1 Portable orifice plate (adapted form U.S. Soil Conservation Service)

270
 Q = cdcv A J 2 g o
I
for submerged flow conditions, and

if the orifice discharges freely into the air. In these two equations h,-h, equals the
head differential across the orifice and Ah equals the upstream head above the centre
of the orifice (see Figures 1.8 and 1.19). A is the area of the orifice and equals
Ed2,where d is the orifice diameter.
i Orifices should be installed and maintained so that the approach velocity is negli-
gible, thus ensuring that C, approaches unity. Calibration studies performed by var-
ious research workers have produced the average Cd-valuesshown in Table 8.1.
The error in the discharge coefficient for a well-maintained circular sharp-crested
orifice, constructed with reasonable care and skill, is expected to be of the order of
1%. The method by which the coefficient error is to be combined with other sources
of error is shown in Annex 2 .

Table 8. I Average discharge coefficients for circular orifices (negligible approach velocity)

Orifice diameter Cd cd
d in metres free flow submerged flow

0.020 0.61 0.57

0.025 0.62 0.58
0.035 0.64 0.61
0.045 0.63 0.61
0.050 0.62 0.61
0.065 0.61 0.60
2 0.075 0.60 0.60

8.1.3 Limits of application

To ensure full contraction and accurate flow measurement, the limits of application
of the circular orifice are:
a. The edge of the orifice should be sharp and smooth and be in accordance with
the profile shown in Figure 5.1;
b. The distance from the edge of the orifice to the bed and side slopes of the approach
and tailwater channel should not be less than the radius of the orifice. To prevent
the entrainment of air, the upstream water level should be at a height above the
top of the orifice which is at least equal to the diameter of the orifice;
c. The upstream face of the orifice plate should be vertical and smooth; .
d. To make the approach velocity negligible, the wetted cross-sectional area at the
upstream head-measurement station should be at least 10 times the area of the
orifice;
e. The practical lower limit of the differential head, across the orifice is related to
fluid properties and to the accuracy with which gauge readings can be made. The
recommended lower limit is 0.03 m.

27 1
8.2 Rectangular sharp-edged orifice
8.2.1 Description

A rectangular sharp-edged orifice used as a discharge measuring device is a well-

defined opening in a thin (metal) plate or bulkhead, which is placed perpendicular
to the bounding surfaces of the approach channel. The top and the bottom edges
should be horizontal and the sides vertical.
Since the ratio of depth to width of (irrigation) canals is generally small and because
changes in depth of flow should not influence the discharge coefficient too rapidly,
most (submerged) rectangular orifices have a height, w, which is considerably less
than the breadth, b,. The principal type of orifice for which the discharge coefficient
has been carefully determined in laboratory tests is the submerged, fully contracted,
sharp-edged orifice. Since the discharge coefficient is not so well defined where the
contraction is partially suppressed, it is advisable to use a fully contracted orifice wher-
ever conditions permit. Where sediment is transported it may be necessary to place
the lower edge of the orifice at canal bed level to avoid the accumulation of sediments
on the upstream side. If the discharge must be regulated it may even be desirable
to suppress both bottom and side contractions so that the orifice becomes an opening
below a sluice gate.
A submerged orifice structure is shown in Figure 8.2. A box is provided downstream
from the orifice to protect unlined canals from erosion. Both the sides and the floor
of this box should be set outward from the orifice a distance of at least two times
the height of the orifice. To ensure that the orifice is submerged or to cut off the
flow, an adjustable gate may be provided at the downstream end of the orifice box.

Figure 8.2 Orifice box dimensions (adapted form U.S. Bureau of Reclamation)

272
This gate should be a sufficient distance downstream from the orifice so as not to
disturb the issuing jet.
The top of the vertical orifice wall should not be higher than the maximum expected
water level in the canal, so that the wall may act as an overflow weir if the orifice
should become blocked. Suitable submerged orifice-box dimensions for a concrete,
masonry, or wooden structure as shown in Figure 8.2 are listed in Table 8.2.

Table 8.2 Recommended box sizes and dimensions for a submerged orifice (after U.S. Bureau of Reclama-
tion 1967)

Orifice size Height of Width of Length Width Length of

structure ' head wall downstream
height breadth head wall
W bc D E L B T

0.08 0.30 1.20 3.00 0.90 0.75 0.60

0.08 0.60 1.20 3.60 0.90 1 .o5 0.60
0.15 0.30 1.50 3.60 I .o5 0.75 0.90
0.15 0.45 1.50 4.25 I .o5 0.90 0.90
0.15 0.60 1.50 4.25 1.o5 1.O5 0.90
0.23 0.40 1.80 4.25 1.o5 0.90 0.90
0.23 0.60 1.so 4.90 1.O5 1 .o5 0.90

8.2.2 Determination of discharge

The basic head-discharge equation for submerged orifice flow, according to Section
1.12 is
Q = c d c v AJ2g(h,-h,) (8-3)
where h, - h, equals the head differential across the orifice, and A is the area of the
orifice and equals the product wb,. In general, the submerged orifice should be de-
signed and maintained so that the approach velocity is negligible and the coefficient
C, approaches unity. Where this is impractical, the area ratio A*/A, may be calculated
and a value for C, obtained from Figure 1.12.
For a fully contracted, submerged, rectangular orifice, the discharge coefficient
Cd = 0.61. If the contraction is suppressed along part of the orifice perimeter, then
the following approximate discharge coefficient may be used in Equation 8-3, regard-
less ofwhether the orifice bottom only or both orifice bottom and sides are suppressed
Cd = 0.61 (1 + 0.15 r) (8-4)
where r equals the ratio of the suppressed portion of the orifice perimeter to the total
perimeter.

If water discharges freely through an orifice with both bottom and side contractions
suppressed, the flow pattern equals that of the free outflow below a vertical sluice
gate as shown in Figure 8.3. The free discharge below a sluice gate is a function of
the upstream water depth and the gate opening:

273
 u
Figure 8.3 Flow below a sluice gate

Q = CdC, b C w J Z 6 A (8-5)
If we introduce the ratios n = yI/w and 6 = y/w, where 6 is the contraction coefficient,
Equation 8-5 may be written as
Q = CdCvbcw1.J- (8-6)
which may be simplified to
Q = K bcw1.J2g = A wo. KJ& (8-7)
where the coefficient K is a function of the ratio n = yJw as shown in Table 8.3.
Table 8.3 Coefficients for free flow below a sluice gate

Ratio Contraction Discharge Coefficient

coefficient coefficient
Eq. 8-6 Eq. 8-7 Eq. 8-7

n=Yl/w 6 Cd K I2 s-

1.50 0.648 0.600 0.614 2.720

1.60 0.642 0.599 0.641 2.838
1.70 0.637 0.598 0.665 2.946
1.80 0.634 0.597 0.689 3.052
I .90 0.632 0.597 0.713 3.159
2.00 0.630 0.596 0.735 3.255
2.20 0.628 0.596 0.780 3.453
2.40 0.626 0.596 0.823 3.643
2.80 0.625 0.598 0.905 4.010
3.00 0.625 0.599 0.944 4.183
3.50 0.625 0.602 1 .O38 4.597
4.00 0.624 0.604 1.124 4.977
4.50 0.624 0.605 1.204 5.33 I
5.00 0.624 0.607 1.279 5.664

Adapted from Franke 1968

274
Some authors prefer to describe a sluice gate as a half-model of a two-dimensional
jet as shown in Figure 1.20, the bottom of the channel being the substitute for the
plane of symmetry of the jet. Hence a discharge equation similar to Equation 1-67
is used to determine the free flow below the gate. This is
Q=ceA& (8-8)
where Ce also expresses the influence of the approach velocity, since it is a function
of the ratio yI/w. The results of experiments by Henry (1950) are plotted in Figure
8.4, which show values of Ceas a function of yl/w and y;/w for both free and submerged
flow below the sluice gate. The C,-values read from Figure 8.4 will result in consider-
able errors if the difference between yl/w and y2/wbecomes small ( < 1 .O). This condi-
tion will generally be satisfied with small differential heads and thus we recommend
that the submerged discharge be evaluated by the use of Equations 8-3 and 8-4.
The results obtained from experiments by Henry, Franke and the U.S. Bureau of
Reclamation are in good agreement. In this context it should be noted that the velocity
does not occur anywhere in the flow system; it simply serves as a convenient
reference velocity for use in Equation 8-8.
The discharge coefficients given for the fully contracted submerged orifice (C, =
0.61) and for free flow below a sluice gate in Table 8.3 can be expected to have an
error of the order of 2%. The coefficient given in Equation 8-4 for flow through a
submerged partially suppressed orifice can be expected to have an error of about 3 % .
The method by which the coefficient error is to be combined with other sources
of error is shown in Annex 2.

8.2.3 Modular limit

Free flow below a sluice gate occurs as long as the roller of the hydraulic jump does
not submerge the section of minimum depth of the jet, which is located at a distance
of
discharge
coefficient

3
n =yl/w

Figure 8.4 Discharge coefficient for use in Equation 8-8 (after Henry 1950)

275
 e= w/F = yJn6 (8-10)
downstream of the face of the vertical gate. To ensure such free flow, the water depth,
y2, downstream of the hydraulic jump should not exceed the alternate depth to
y = Fw, or according to the equation

2 < [dl
W
+ 16 (g - 1) - 11 (8111)

Relative numbers y2/w worked .out with the theoretical minimum contraction coeffi-
cient 6 = 0.61 1, corresponding to high values of the ratio n, are given in Figure 8.5
as a function of yJw.

8.2.4 Limits of application

To ensure accurate flow measurements, the limits of application of the rectangular

sharp-edged orifice are:
a. The upstream edge of the orifice should be sharp and smooth and be in accordance
with the profile shown in Figure 5.1;
b. The upstream face of the orifice should be truly vertical;
c. The top and bottom edges of the orifice should be horizontal;
d. The sides of the orifice should be vertical and smooth;
e. The distance from the edge of the orifice to the bed and side slopes of the approach
and tailwater channel should be greater than twice the least dimension of the orifice
if full contraction is required;
f. The wetted cross-sectional area at the upstream head-measurement station should
be at least 10 times the area of the orifice so as to make the approach velocity
negligible; this is particularly recommended for fully contracted orifices;

RATIO y2/w

1:
3

O 1 1.5 2 3 4 5 6 7 8 9 10

RATIO n = y , / w

Figure 8.5 Limiting tail-water level for modular flow below a sluice gate

276
g. If the contraction is suppressed along the bottom or sides of the orifice, or along
both the bottom and sides, the edge of the orifice should coincide with the bounding
surface of the approach channel;
h. The practical lower limit of the differential head across the submerged orifice is
related to fluid properties and to the accuracy to which gauge readings can be made.
The recommended lower limit is 0.03 m;
i. If the contraction along bottom and sides is suppressed, the upstream head should
be measured in the rectangular approach channel;
j. The upper edge of the orifice should have an upstream submergence of 1 . 0 or ~
more to prevent the formation of air-entraining vortices;
k. A practical lower limit of w = 0.02 m and of y, = O. 15 m should be observedi

8.3 Constant-head-orifice
8.3.1 Description

The constant-head-orifice farm turnout (CHO) is a combination of a regulating and

measuring structure that uses an adjustable submerged orifice for measuring the flow
and a (downstream) adjustable turnout gate for regulation. The turnouts are used
to measure and regulate flows from main canals and laterals into smaller ditches and
are usually placed at 90" angle to the direction of flow in the main canal. The CHO
was developed by the United States Bureau of Reclamation and is so named because
its operation is based upon setting and maintaining a constant head differential, Ah,
across the orifice. Discharges are varied by changing the area of the orifice. A typical
constant head-orifice turnout installation is shown in Figure 8.6.
To set a given flow, the orifice opening A required to pass the given discharge is
determined from a graph or table, and the orifice gate is set at this opening. The down-
stream turnout gate is then adjusted until the head differential as measured over the
orifice gate equals the required constant-head, which usually equals 0.06 m. The dis-
charge will then be at the required value. The rather small differential head used is
one of the factors contributing to the inaccuracy of discharge measurements made
by the CHO. For instance, errors of the order of 0.005 m in reading each staff gauge
may cause a maximum cumulative error ofO.O1 m or about 16% in Ah, which is equiva-
lent to 8% error in the discharge. Introducing a larger differential head would reduce
this type of error, but larger flow disturbances would be created in the stilling basin
between the two gates. Furthermore, it is usually desirable to keep head losses in an
irrigation system as low as possible.
Since the downstream gate merely serves the purpose of setting a constant head
differential across the orifice gate, its shape is rather arbitrary. In fact, the turnout
gate shown in Figure 8.6 may be replaced by a movable weir or flap-gate if desired.
If the CHO is connected to a culvert pipe thht is flowing full, the air pocket immediately
downstream of the turnout gate should be aerated by means of a ventilation pipe.
The diameter of this pipe should be 1/6 of the culvert diameter to provide a stable
flow pattern below the turnout gate.
If the flow through the downstream gate is submerged, a change of tailwater level
of the order of a few centimetres will cause an equivalent change of water level in
the basin between the two gates. Under field conditions, the discharge in the main

277
 0.05
0.10

L( 00.08bad

DETAIL 0
into WUI1
Dcetocotbarr
lluh with mcpcte

wc ,c.i. slide gate with flat back mt and square

II bottom slide. lift d m r h w bdtr as
mim gate notshom.
s l e l slide gate with flat
back seat. lift and anchw bolts
as requted. pte nd shown. ----____ ',
reinf.

LONGITUDINAL SE(3TK)N

Figure 8.6 Example of a constant-head-orifice (adapted form USBR 1970)

canal is likely to be large compared with the discharge through the turnout. Hence
the head differential over the orifice gate will change with any change in tailwater
level, resulting in a considerable error in the diverted flow. The reader will note that
if reasonable'accuracy is required in discharge measurement, the flow below the turn-
out gate should be supercritical at all tailwater levels. For this to occur, the turnout
gate requires a minimum loss of head which may be calculated as explained in Section
8.2.2 and with the aid of Figure 8.5. The combined loss of head over the orifice gate
(usually 0.06 m) and over the turnout gate (variable) to produce modular flow is consi-
derable.
Usually the CHO is placed at an angle of 90" from the centre line of the main canal,
and no approach channel is provided to the orifice gate. As a result, the flow in the
main canal will cause an eddy and other flow disturbances immediately upstream of
the orifice gate opening, thus affecting the flow below the orifice gate. Such detrimental
effects increase as the flow velocity in the main canal increases and are greater if the
CHO is working at full capacity. Full-scale tests showed a deviation of the discharge
coefficient of as much as 12% about the mean C,-values with high flow velocities (1 .O
m/s) and with larger orifice gate openings. The approach flow conditions, and thus
the accuracy of the CHO can be improved significantly by introducing an approach
channel upstream of the orifice gate. For example, if the CHO is used in combination
with a culvert under an inspection road, the CHO could be placed at the downstream
end of the culvert, provided that the culvert has a free water surface.

278
Since the CHO is usually operated at a differential head of 0.06 m (0.20 foot) it is
clear that extreme care should be taken in reading heads. Fluctuations of the water
surfaces just upstream of the orifice gate and in the stilling basin downstream of the
orifice can easily result in head-reading errors of one or more centimetres if the heads
are read from staff gauges. This is particularly true if the CHO is working at full
capacity. Tests have revealed that, with larger orifice-gate openings, staff gauge read-
ings may show a negative differential head while piezometers show a real differential
head of 0.06 m. Head-reading errors can be significantly reduced if outside stilling
wells are connected to 0.01 m piezometers placed in the exact positions shown in Figure
8.6. Two staff gauges may be installed in the stilling wells, but more accurate readings
will be obtained by using a differential head meter as described in Section 2.12. Head-
reading errors on existing structures equipped with outside staff gauges can be reduced
by the use of a small wooden or metal baffle-type stilling basin and an anti-vortex
baffle. The dimensions and position of these stilling devices, which have been deve-
loped by the U.S. Agricultural Research Service, are shown in Figure 8.7.
Because of the above described error in discharge measurement, the construction
of a new CHO is not recommended.

8.3.2 Determination of discharge

The basic head-discharge equation for a submerged orifice, according to Section 1.13
reads
Q = CA- (8-12)
where, for the CHO, the differential head Ah usually equals 0.06 m. The discharge
coefficient C is a function of the upstream water depth, y,, and the height of the orifice

ORIFICE GATE

STILLING BASIN FOR

DOWNSTREAM STAFF GAUGE

NOTE: BOTH STILLING BASIN AND

ANTI VORTEX BAFFLE EXTEND
COMPLETELY ACROSS CHANNEL
AND FIT TIGHTLY AGAINST
SIDE WALLS.
DIMENSIONS IN MM.
Figure 8.7 Device to reduce water level fluctuations at CHO staff gauges (after U.S. Agricultural Research
Service, SCS 1962)

279
w. Experimental values of C as a function of the ratio y,/w are shown in Figure 8.8.
The reader should note that the coefficient C also expresses the influence of the ap-
proach velocity head on the flow.
From Figure 8.8 it appears that the discharge coefficient, C, is approximately 0.66
for normal operating conditions, i.e. where the water depth upstream from the orifice
gate is 2.5 or more times the maximum height of the gate opening, w. Substitution
of the values Cd = 0.66, Ah = 0.06 m, and g = 9.81 m/s2 into Equation 8.1 gives
the following simple area-discharge relationship for the CHO:
Q = 0.716 A = 0.716 b , ~ (8- 13)
If the breadth of the orifice is known, a straight-line relationship between the orifice
gate opening and the flow may be plotted for field use.
The error in the discharge coefficient given for the Constant-Head-Orifice
(C = 0.66) can be expected to be of the order of 7%. This coefficient error applies
for structures that have an even velocity distribution in the approach section. If an
eddy is formed upstream of the orifice gate, however, an additional error of up to
12% may occur (see also Section 8.3.1).
The method by which the coefficient error is to be combined with other sources
of error, which have a considerable effect on the accuracy with which flow can be
measured, is shown in Annex 2. In this context, the reader should note that if the
upstream gate is constructed with uninterrupted bottom and side walls and a sharp-
edged gate, Equations 8-3 and 8-4 can be used to determine the discharge through
the orifice with an error of about 3%.

8.3.3 Limits of application

The limits of application of the Constant-Head-Orifice turnout are:

a. The upstream edge of the orifice gate should be sharp and smooth and be in accor-
dance with the profile shown in Figure 5.1;
b. The sides of the orifice should have a groove arrangement as shown in Figure 8.6;

10

U
+ O9
w
u
t; O8
U
W
a
Q 07
xE
06
10 20 30 40 50 60 70 80 90 100 110 120

RATIO yllw- .

Figure 8.8 Variation of discharge coefficient, C , as a function of the ratio y l / w (indoor tests)

280
c. The bottom of the approach section upstream of the orifice gate should be horizon-
tal over a distance of at least four times the upstream water depth.
d. To obtain a somewhat constant value for the discharge coefficient, C, the ratio
yJw should be greater than 2.5;
e. The approach section should be such that no velocity concentrations are visible
upstream of the orifice gate.

8.4 Radial or tainter gate

8.4.1 Description

The radial or tainter gate is a movable control; it is commonly used in a rectangular

canal section. It has the structural advantage of not requiring a complicated groove
arrangement to transmit the hydraulic thrust to the side walls, because this thrust
is concentrated at the hinges. In fact, the radial gate does not require grooves at all,
but has rubber seals in direct contact with the undisturbed sides of the rectangular
canal section.
Figure 8.9 shows two methods by which the radial gate can be installed, either with
the gate sill at stream bed elevation or with its sill raised.

GATE WITH SILL AT STREAMBED ELEVATION

VERTICAL
L

CATE WITH RAISED SILL

Figure 8.9 Flow below a radial o r tainter gate

28 1
8.4.2 Evaluation of discharge

Free flow through a partially open radial gate is commonly computed with the follow-
ing equation:
Q = COC, wb,& (8-15)
The Coefficient, Co, depends on the contraction of the jet below the gate and may
be expressed as a function of the gate opening w, gate radius r, trunnion height a,
and upstream water depth yI, for a gate sill at streambed elevation. Figure 8.10 gives
C,-values for a/r ratios of 0.1, 0.5, and 0.9. Coefficient values for other a/r-values
may be obtained by linear interpolation between the values presented.

The coefficient C, is a correction to Co for gate sills above streambed elevation and
depends upon sill height pi and the distance between the step and the gate seat L,
as shown in Figure 8.11. Insufficient information is available to determine the effects,
if any, of the parameter pl/r.
It should be noted that the velocity &in Equation 8- 15 does not occur anywhere
in the flow system, but simply serves as a convenient reference velocity.
The experiments on which Figure 8.10 is based showed that the contraction coeffi-
cient, 6, of the jet below the gate is mainly determined by the angle 8 and to a much
lesser extent by the ratio yI/w. For preliminary design purposes, Henderson (1966)
proposed Equation 8-16 to evaluate F-values.
F = I - 0.75 ( e p o o ) + 0.36 (e/900)2 (8-16)
where 8 equals the angle of inclination in degrees.

Equation 8-16 was obtained by fitting a parabola as closely as possible to Toch's results
(1952, 1955) and data obtained by Von Mises (1917) for non-gravity, two-dimensional
flow through an orifice with inclined side walls. Values of 6 given by Equation 8-16
and shown in Figure 8.12 can be expected to have an error of less than 5%, provided
that 8 < 90".
If the discharge coefficient Coin Equation 8-15 is to be evaluated from the contrac-
tion coefficient, we may write, according to continuity and Bernoulli:
6
(8-17)
co = Jm
The discharge coefficient, Co, given in Figure 8.10 and Equation 8-17 for free flow
below a radial gate can be expected to have errors of less than 5% and between 5
and 10% respectively. The error in the correction coefficient CI,given in Figure 8.1 1
can be expected to have an error of less than 5%. The method by which these errors
have to be combined with other sources of error is shown in Annex 2.

282
 O
"O6

PY
LL
8 al r -01
2 OS

I
E

04

RATIO y,/r

O
U
07
z
w
u
LL

i 06

4 alrz0.5

05

0 4
O0 O2 04 06 O8 10 12 14
RATIO Ir
Y.

alr-O. 9

RATIO yl/r

Figure 8.10 Co-values as a function of a/r, yl/r and w/r (from U . S . Army Engineer Waterways Experiment
Station 1960)

283
 correction
coefficient

ratio LIPI

Figure 8.1 1 C,-values for radial gates with raised sill (from U.S. Army Engineer Waterways Experiment
Station)

lip angleein degrees

Figure 8.12 Effect of lip angle on contraction coefficient

8.4.3 Modular limit

Modular flow below a radial gate occurs as long as the roller of the hydraulic jump
does not submerge the section of minimum depth of the jet (vena contracta). To pre-

284

i
vent such submergence, the water depth, y2, downstream of the hydraulic jump should
not exceed the alternate depth to y = 6w or according to the equation

W < i[/l + 16(E-l)- 11 (8-18)

For each radial gate the modular limit may be obtained by combining Equation 8-16
(or Figure 8.12) and Equation 8-18.
If flow below the gate is submerged, Equation 1-73 as derived in Section 1.12 may
be used as a head-discharge relationship. It reads
Q = C,~,wJ2g(y,-y,) (8-19)

Photo 1 Radial gates are suitable flow control structures

285
Insufficient experimental data are available to present reasonably accurate Ce-values.
For design purposes, however, the coefficient Ce may be evaluated from the contrac-
tion coefficient 6 for free flow conditions (Figure 8.12).
A combination of the Bernoulli and the continuity equations gives for Ce

(8-20)

It should be noted that the assumption that the contraction coefficient is the same
for free flow as for submerged flow is not completely correct.

8.4.4 Limits of application

The limits of application of the radial or tainter gate are:

a. The bottom edge of the gate should be sharp and horizontal from end to end;
b. The upstream head should be measured in a rectangular approach channel that
has the same width as the gate;
c. The gate opening over water depth ratio should not exceed 0.8 (w/y, 6 0.8);
d. The downstream water level should be such that modular flow occurs (see Equation
8-18).

8.5 Crump-De Gruyter adjustable orifice

8.5.1 Description

The Crump-De Gruyter adjustable orifice is a short-throated flume fitted with a verti-
cally movable streamlined gate. It is a modification of the adjustable proportional
module, introduced by Crump in 1922. De Gruyter (1926) modified the flume align-
ment and replaced the fixed roof-block with an adjustable sliding gate and so ob-
tained an adjustable flume that can be used for both the measurement and regulation
of irrigation water (see Figure 8.1 3).
Usually the orifice is placed at an angle of 90 from the centre line of the main
canal which may cause eddies upstream of the orifice gate if canal.velocities are high.
For normal flow velocities in earthen canals, the approach section shown in Figure
8.13 is adequate. If canal velocities are high, of the order of those that may occur
in lined canals, the approach section should have a greater length so that no velocity
concentrations are visible upstream of the orifice gate. The structural dimensions in
Figure 8.13 are shown as a function of the throat width b, and head h,.
Provided that the gate opening (w) is less than about z/3 Hl - in practice one takes
w 6 0.63 h, - and the downstream water level is sufficiently low, supercritical flow
will occur in the throat of the structure so that the discharge depends on the upstream
water level (h,) and the gate opening (w) only.
With the use of Equation 1-33, the discharge through the non-submerged (modular)
structure can be expressed by
Q =C d C v b c ~ J w (8-21)

286
where b, equals the breadth of the flume throat and w is the gate opening which equals
the 'water depth' at the control section of the flume. To obtain modular flow, a minimal
loss of head over the structure is required. This fall, Ah, is a function of both h, and
w, and may be read from Figure 8.14, provided that the downstream transition is
in accordance with Figure 8.13.

From Figure 8.14 we may read that for a gate opening w = 0.2 h, the minimal fall
required for modular flow is 0.41 h,, and that if w = 0.4 h, the minimal fall equals
0.23 h,. This shows that, if h, remains about constant, the adjustable orifice requires
a maximum loss of head to remain modular when the discharge is minimal. Therefore,
the required value of the ratio y = Qmax/Qmin is an important design criterion for the

ADJUSTABLE GATE

4P2
BROKEN BACK TRANSITION

NOTE '
O N STANDARD STRUCTURES pl=bc and L=bc

Figure 8.13 The Crump-De Gruyter adjustabel orifice dimensions as a function of h, and b,

287
 SECTION OVER ADJUSTABLE GATE
I

FLOW

DETAIL OF GROOVE ARRANGEMENT

Figure 8.13 cont.

elevation of the flume crest. If, for example, both y and h, are known, the minimum
loss of head, Ah, required to pass the range of discharges can be calculated from Figure
8.14. On the other hand, if both y and Ah are known, the minimum h,-value, and
thus the flume elevation with regard to the upstream (design) water level, is known.

When a design value for h, has been selected, the minimum throat width, b,, required
to pass the required range of discharges under modular conditions can be calculated
from the head-discharge equation and the limitation on the gate opening, which is
w < 0.63 h,. Anticipating Section 8.5.2 we can write
Qm,, < 0.94 bc(0.63hl) J2g(hI - 0.63 hl) (8-22)
which results in a minimum value of b,, being
~

(8-23)

With the use of Figures 8.13 and 8.14 and Equation 8-23, all hydraulic dimensions
of the adjustable orifice can be determined.

I 288
RATIO Ah/h,

1.0

o. 9
o. 8
O. 7

O.6

0.5

o. 4

o.3
o. 2
o. 1
O

RATIO W/h,

Figure 8.14 Characteristics of the Crump-De Cruyter adjustable orifice (after De Cruyter 1926)

8.5.2 Evaluation of discharge

As mentioned in Section 8.5.1, the basic head discharge equation for a Crump-De
Gruyter adjustable orifice reads
Q=CdCcbcWJ2g0 (8-24)
where the discharge coefficient C, equals 0.94 and the approach velocity coefficient
can be obtained from Figure 1.12. Table 8.4 shows the unit discharge q in m3/s per
metre flume breadth as a function of h, and w, for negligible approach velocity
(C, N 1.0).
If reasonable care and skill has been applied in the construction and installation
of a Crump-De Gruyter adjustable orifice, the discharge coefficient may be expected
to have an error of about 3%. The method by which the error in the coefficient is
to be combined with other sources of error is shown in Annex 2.

8.5.3 Limits of application

The limits of application of the Crump-De Gruyter adjustable orifice are:

a. To obtain modular flow the gate opening (w) should not exceed 0.63 h,, and the
minimum fall over the structure, Ah, should be in accordance with Figure 8.14;
b. The practical lower limit of w is 0.02 m;
c. The bottom of the flume control section should be horizontal and its sides vertical;
d. The thickness of the adjustable gate in the direction of flow should be 0.5 Hlmax
and the upstream curvature of the gate should equal 0.375 Hlmax
leaving a horizontal
lip with a length of0.125 HlmU(see Figure 8.13);

289
 Table 8.4 Rating table for the Crump-De Gruijter adjustable flume

h) Upstream head over flume crest H Iin metres

W
O 0.30 0.32 0.34 0.36 0.38 0.40 0.42 0.44 0.46 0.48 0.50 0.52 0.54 0.56 0.58 0.60
win win
metres unit discharge q in m3/s perm metres,
0.02 0.044 0.045 0.046 0.048 0.049 0.050 0.052 0.054 0.055 0.056 0.057 0.058 0.059 0.060 0.061 0.062 0.02
0.03 0.064 0.068 0.070 0.072 0.074 0.076 0.078 0.080 0.082 0.084 0.085 0.087 0.088 0.090 0.092 0.094 0.03
0.04 0.084 0.088 0.090 0.094 0.097 0.100 0.102 0.105 0.108 0.110 0.113 0.116 0.118 o. 120 o. 122 O. 124 0.04
0.05 0.104 0.108 0.112 0.116 0.119 0.122 0.126 O. 130 0.133 0.136 0.140 0.143 0.146 O. I49 0.152 O. 154 0.05
0.06 0.122 0.127 0.132 0.137 0.142 0.146 0.150 O. 154 0.158 0.162 0.165 0.168 0.171 O. I74 0.178 O. 182 0.06
0.07 0.140 0.145 0.150 0.156 0.162 0.167 0.172 O. I77 0.182 0.186 0.190 0.195 0.200 0.204 0.208 0.212 0.07
0.08 0.156 0.163 0.170 0.176 0.182 0.188 0.194 0.200 0.206 0.211 0.216 0.221 0.226 0.231 0.236 0.241 0.08
0.09 O. 172 0.180 0.187 0.194 0.201 0.208 0.215 0.222 0.228 0.234 0.240 0.246 0.252 0.258 0.263 0.268 0.09
0.10 0.186 0.195 0.204 0.212 0.220 0.228 0.235 0.242 0.249 0.256 0.263 0.270 0.276 0.282 0.288 0.294 0.10
0.11 0.200 0.210 0.219 0.228 0.237 0.246 0.253 0.262 0.270 0.278 0.285 0.292 0.300 0.308 0.314 0.320 0.11
0.12 0.212 0.223 0.234 0.244 0.254 0.264 0.274 0.283 0.292 0.300 0.308 0.316 0.323 0.330 0.338 0.346 0.12
0.13 0.224 0.236 0.248 0.259 0.270 0.280 0.290 0.300 0.310 0.319 0.328 0.337 0.346 0.354 0.362 0.370 0.13
0.14 0.234 0.247 0.260 0.273 0.286 0.297 0.308 0.319 0.330 0.340 0.350 0.359 0.368 0.377 0.386 0.395 0.14
0.15 0.242 0.257 0.272 0.286 0.299 0.312 0.324 0.335 0.346 0.358 0.370 0.380 0.390 0.400 0.410 0.420 0.15
0.16 0.250 0.266 0.282 0.298 0.312 0.326 0.339 0.352 0.364 0.376 0.388 0.399 0.410 0.420 0.430 0.440 0.16
0.17 0.256 0.274 0.292 0.308 0.324 0.339 0.354 0.368 0.381 0.394 0.406 0.418 0.430 0.442 0.453 0.464 0.17
0.18 0.260 0.280 0.299 0.318 0.334 0.350 0.366 0.381 0.396 0.410 0.424 0.437 0.450 0.462 0.474 0.486 0.18
0.19 0.262 0.284 0.305 0.325 0.344 0.362 0.380 0.396 0.410 0.425 0.440 0.454 0.468 0.480 0.492 0.504 0.19
0.20 0.288 0.310 0.331 0.352 0.372 0.390 0.408 0.424 0.440 0.456 0.472 0.486 0.498 0.512 0.525 0.20
0.21 0.316 0.338 0.360 0.380 0.400 0.419 0.438 0.454 0.472 0.488 0.502 0.518 0.532 0.546 0.21
0.22 0.342 0.366 0.388 0.408 0.428 0.448 0.466 0.484 0.502 0.518 0.532 0.548 0.564 0.22
0.23 0.344 0.370 0.394 0.417 0.438 0.458 0.478 0.496 0.514 0.532 0.550 0.566 0.582 0.23
0.24 0.374 0.400 0.424 0.446 0.468 0.488 0.508 0.528 0.548 0.566 0.584 0.600 0.24
0.25 0.404 0.427 0.452 0.476 0.498 0.519 0.540 0.560 0.578 0.596 0.615 0.25
0.26 0.432 0.458 0.482 0.506 0.528 0.549 0.572 0.592 0.612 0.631 0.26
0.27 0.462 0.489 0.514 0.538 0.562 0.583 0.604 0.624 0.646 0.27
0.28 0.464 0.493 0.520 0.546 0.570 0.594 0.616 0.638 0.659 0.28
0.29 0.496 0.525 0.552 0.578 0.604 0.628 0.650 0.672 0.29
0.30 0.528 0.558 0.586 0.612 0.636 0.660 0.684 0.30
0.31 0.562 0.590 0.618 0.644 0.669 0.694 0.3 I
0.32 0.594 0.624 0.651 0.679 0.704 0.32
0.33 0.600 0.628 0.658 0.687 0.714 0.33
0.34 0.632 0.662 0.694 0.720 0.34
0.35 0.666 0.698 0.728 0.35
0.36 0.700 0.732 0.36
0.37 0.738 0.37
0.38 0.742 0.38
Note: Valid for negligible approach velocity (hl !z Hl)
e. The minimum breadth of the flume should be in accordance with Equation 8-23,
but b, should not be less than 0.20.m;
f. For standard flumes pI equals b,; PI may be changed, however, provided it remains
equal to or greater than 0.10 m.

8.6 Metergate
8.6.1 Description

A metergate is rather commonly used in the U.S.A. for measuring and regulating
flow at irrigation water off-takes. Basically, it is a submerged orifice arranged so that
its area is adjustable by a vertical screw lift. It may also be regarded as a submerged
calibrated valve gate at the upstream end of a pipe section. A typical metergate installa-
tion is shown in Figure 8.15. Constructional details of the gate with a rectangular
gate leaf are shown in Figure 8.16.
Usually the metergate is placed at right angles to the center line of the main canal

amount of opening is shown by

distance between notch on lift
rod and top of lift nut

a minimum of 0.30 m or other minimum

reater than 0.45 m

Figure 8.15 Metergate installation (courtesy of ARMCO)

29 1
292
 or lateral from which it diverts flow. If the flow velocity in the main canal becomes
significant, it will cause eddies and other flow disturbances along the upstream
wingwalls that form the approach to the gate. To prevent such disturbances from
reducing the flow through the metergate, the approach to the gate should be shaped
so that no velocity concentrations are visible on the water surface upstream of the
orifice. To achieve this the approach section should have a minimum length of about
5 D,, where D, equals the diameter of the pipe and also the diameter of the gate ope-
,
I ning.
As explained in Section 1.12, the flow through a submerged orifice is directly related
I to the differential head over the opening. It is essential that the stilling well intakes
(piezometers) be located exactly as they were in the original calibrated metergate. The
upstream piezometer should be placed in the vertical headwall, at least 0.05 m from
the gate frame and also 0.05 m from any change in headwall alignment if viewed from
the top. The intake should be flush with the headwall surface and at least 0.05 m
below minimum water level during operation. For the downstream piezometer, two
locations are possible, depending on the method of discharge evaluation:
- on the centre line of the top of the pipe, at exactly 0.3048 m (1 foot) downstream
from the downstream face of the gate. This location is used on most commercially-
manufactured* gates. The discharge is read from tables which are supplied with
each gate;
- on the centre line of the top of the pipe at D,/3 downstream from the downstream
gate face. This location is recommended by the U.S. Bureau of Reclamation and
is supported by the present writers. The discharge can be evaluated by using Equa-
tion 8-25 and Figure 8.18 (see Section 8.6.2).
If corrugated pipe is used, the downstream piezometer should always be at the top
of a corrugation.
The piezometer location at exactly 0.3048 m downstream from the downstream gate
face means that the various metergates are not hydraulic scale models of each other.
Another disadvantage is that for small pipe diameters the downstream piezometer

PIEZOMETRIC PRESSURE LINE-,

1 (Not to scale 1

LL ,PIU.AT TOP OF

0.305m

Figure 8.17 Effect of piezometer location on measured head

* The metergate is commercially manufactured by ARMCO Steel Corporation, P.O.BOX700 Middletown,

Ohio 45042, USA. Our listing of this supplier should not be construed as an endorsement of this company
or their product by the present writers.

293
is situated in a region with a rapid change of pressure, as illustrated in Figure 8.17.
As a result any minor displacement of the piezometer from the tested location will
cause large errors in the determination of the differential pressure.
Flow through the metergate is proportional to the square root of the head difference,
Ah, between the two stilling wells, which may be measured by one of the differential
head meters described in Section 2.12. The practical lower limit of Ah is related to
the accuracy with which piezometer readings can be made. The recommended lower
limit is 0.05 m. If practicable, the upstream water level should be kept at a height
which ensures that the metergate operates under large differential heads.
To ensure that the downstream stilling well contains sufficient water for a reading
of head to be taken, the pipe outlet must have sufficient submergence. This submer-
gence depends, among other things, on the friction losses in the downstream pipe
and the maximum head differential over the stilling wells. On field installations the
head differential is usually limited to 0.45 m while the meter pipe must be longer than
6 Dp or 7 D, so that a submergence of 0.30 m will usually be sufficient. A method
by which the required submergence can be calculated is shown in Section 8.6.3.

8.6.2 Evaluation of discharge

Flow through a metergate may be evaluated by the following formula

Q = CeApJ- (8-25)
where A, = /,nD,2 is the nominal area of the pipe. It should be noted that the coeffi-
cient Ce is not the same as the discharge coefficient introduced in the orifice equation
derived in Section 1.12, where the orifice area (A) appears in the discharge equation.
Figure 8.18 gives C,-values as a function of the gate opening for gates with either
a rectangular or a circular gate leaf, and with their downstream pressure tap at D,/3
downstream from the downstream face of the gate. The curve for circular leaves was
derived from tables published by ARMCO; that for rectangular leaves was taken from
the U.S. Bureau of Reclamation, 1961.
Although the curves in Figure 8.18 were obtained for particular approach condi-
tions, all approach sections that comply with the conditions outlined in Section 8.6.4
may be used in combination with the Ce-curves shown. This was demonstrated by
tests, conducted by the U.S. Bureau of Reclamation (1961), which showed that Ce-
values are not influenced by approach conditions if the gate opening remains less than
50%; in the range from 50% to 75% the Ce-valuemay increase slightly. Gate openings
greater than 75% are not recommended for discharge regulation since, in this range,
the Ce-value shows considerable variation (see also Figure 8.20).
The discharge coefficient shown in Figure 8.18 may be expected to have an error
of less than 3% for gate openings up to 50%, and an error of less than 6% for gate
openings up to 75%. The method by which this error is to be combined with other
sources of error is shown in Annex 2.
Each commercially-manufactured meter gate is accompanied by a discharge table
(Imperial units). Generally, these tables are sufficiently accurate, but the U.S. Bureau
of Reclamation in some instances found errors of 18% or more. Discharge tables are
available for gates ranging from 0.20 m (8) to 1.22 m (48).

294
coefficient Ce

O 20 40 60 80 1O0
percent gate opening

Figure 8.18 Ce-values for pressure tap located at D,/3

Provided that water rises sufficiently high in the downstream stilling well, the degree
of submergence does not affect the accuracy of the meter.

8.6.3 Metergate installation

For a metergate to function properly it must be installed at the proper elevation and
be of the proper size. To aid in the selection of gate size and elevation we give the
following suggestions in the form of an example:

Given:
- Upstream water surface elevation 100.00 m;
- Downstream water surface elevation 99.70 m (thus Ah,,, = 0.30 m);
- Turnout discharge 0.140 m3/s;
- Depth of water in downstream measuring well, h,, should be O. 15 m above crown
of metergate;
- Length of metergate pipe, L,= 8.50 m;

- Submergence of metergate inlet, h,, should not be less than D, above the crown

of the pipe;
- A metergate with rectangular leaf is used.

Find:
1.. Metergate size;
2. Elevation at which metergate should be placed.

295
 LP
y
I
downstream
oressure tao well
1
I

SECTION THROUGH INSTALLATION

Figure 8.19 Example of metergate installation (USBR 1961)

Metergate size

a. When downstream scour is a problem, an exit velocity has to be selected that will
not cause objectionable erosion, say v 6 0.90 m/s. From A, = Q/v we find
A, 2 O. 140/0.90 = O. 156 m2 or Dp 2 0.445 m;
An 18-inch (D, = 0.457 m) metergate is required.
b. When downstream scour is not a problem, we select a metergate that operates at
gate openings not exceeding75% (see Section 8.6.2). For 75% gate opening the coeffi-
cient Ce N 0.51 (Figure 8.18) and the maximum differential head Ah N 1.8 Ah,
(Figure 8.20). Taking into account some losses due to friction in the pipe, we assume
Ah N 1.60 Ah,,, = 1.60 x 0.30 = 0.48 m.From Equation 8-25: Q = CeAp(2gAh)0.5
we obtain the minimum area of the pipe: A, 2 0.0895 m2 and thus D, 2 0.34
m. Our initial estimate is a 14-inch metergate (D, = 0.356 m);
c. Check capacity of selected gate. It is common practice to express the loss of hydrau-
lic head as a function of the velocity head, v2/2g. For a metergate the velocity head
in the pipe can be found by substituting the continuity equation Q/Ap = v into
Equation 8-25, which leads to
v = CJ2gAh (8-26)
or
V2
- = C,2 Ah (8-27)
2g
The total (available) loss of head over the structure, Ah,,,, equals the sum of the energy
loss over the gate, the friction losses in the meterpipe, and the exit losses, so that
Ah,,, = Ahgate+ 5fv2/2g + 5 e x v2/2g (8-28)
If we assume that no recovery of kinetic energy occurs at the pipe exit (tex= 1.0)
we can write
Ah,,, = Ahr + &fv2/2g (8-29)

296
where Ahr denotes the drop of piezometric head to a recovery point downstream of
the downstream pressure tap which equals the energy losses over the gate plus the
velocity head in the meterpipe.
The substitution of Equation 8-27 into Equation 8-29 and division by Ah leads to
AhJAh = Ahr/Ah i- Slcz (8-30)
where the friction loss coefficient Sf equals fL,,/D, (assume f = 0.025 for concrete
and steel pipes) and values of Ce and Ahr/Ah can be obtained from Figures 8.18 and
8.20 respectively as a function of the gate opening.
cr
In our example Ah = 0.30 m and = fL,/D, = 0.025 x 8.50/0.356 = 0.60. For
75% gate opening Ce v 0.51 and the ratio Ah/Ahr N 1.80, so that according to Equa-
tion 8-30 the maximum value of Ah = 0.42 m. Using this adjusted value of Ah, the
turnout capacity at 75% gate opening equals

~ ~~~ ~

AP~R~ACH~ EC
TION
1

i
FLOOR SIDE WALLS
SYMBOL
DISTANCE M W IN DISTANCE

I
SLOPE BELCW EMRT PLAN FROM E D E
- OFElmwCE OF ENTRANCE

(ULLEL 3.0 D,

+
24

24
;CE; 1.0 D .
0.63Dp PARALLEL 2.25 Dp

1:2DOWNw\AD 0.17 Dp 8:l FLARING 0 . 2 5 Dp

2. o

1.9

1.8

1.7

1.6

g 1.2
d
1.1

1.0
O 20 40 60 80 100

PERCENTAGE GATE OPENING

Figure 8.20 Gate opening versus Ah/Ah,

29 7
 Q N 0.51 x n x 0.3562(2g x 0.42)''. N 0.146 m3/s
A 14-in metergate is adequate.

Elevation at which metergate should be placed

If the differential head over the metergate structure is a constant, in our example
Ah = 0.30 m, the head difference Ah measured between the two wells is at its maximum
with gate openings of around 50%. Using Equation 8-30 the following Ah-values can
be computed:
gate opening 35% 40 yo 50% 55% 60% 75%
Ah (m) 0.399 0.437 0.452 0.458 0.454 0.422
- To meet the requirement of water surface O. 15 m above the crown of the pipe (0.05 m
above bottom of well) in the downstream well, elevation of crown entrance would
be set at EL = 100.00-Ah,,, - h, = 100.00 - 0.46 -0.15 = 99.39 m
- To meet the upstream submergence requirement, h,, of 1.0 D,, the crown of the

pipe entrance should be set at EL = 100.00 - D, N 99.64 m.

The depth requirement for a measurable water surface in the downstream well is the
governing factor and the metergate should be set with its crown of entrance not higher
than EL = 99.39 m.

8.6.4 Limits of applications

The limits of application of the metergate are:

a. The crown of the pipe entrance should have an upstream submergence of 1.0 D,
or more;
b. Submergence of the pipe outlet should be such that the water surface in the down-
stream well is not less than O. 15 m above the crown of the pipe;
c. The approach channel should be such that no velocity concentrations are visible
upstream of the gate (see Figure 8.15);
d. The length of the gate pipe should be 6 D, or more;
e. The head differential over the stilling wells should not be less than 0.05 m. Its practi-
cal upper
_ _ limit is about 0.45 m;
f. During operation (flow measurement), gate openings should not be greater than
75%;
g. If Figure 8.18 is used to obtain Ce-values, the downstream pressure tap should
be located at exactly D,/3 downstream from the downstream face of the gate;
h. The downstream pressure tap should be located on the centre line of the top of
the pipe. The intake pipe should be flush with the inside surface of the pipe and
absolutely vertical. If corrugated pipe is used the intake should be at the top of
a corrugation;
i. The bottom of the approach section should be at least 0.17 D, below the invert
of the gate opening.

298
 8.7 Neyrpic module
8.7.1 Description

The Neyrpic module* was designed to allow the passage of an almost constant flow
from an irrigation canal in which the variation of the water level is restricted. The
structure consists of a fixed weir sill with a 60-degree sloping upstream face and a
12-degree sloping downstream face. The weir crest is rounded, its radius equal to
0.2 hd, where hd is the design head. Above the weir either one or two steel plates are
fixed in a well defined position. These sloping (35-degree) sharp-edged plates cause
an increase of contraction of the outflowing jet when the upstream head increases.
The near constant orifice discharge per unit width is a function of the height of the ,
inclined blade above the weir. Since this height cannot be altered the only way to
regulate flow is to combine several orifices of different widths into one structure. The
minimum width of an orifice is 0.05 m which coincides with 0.005 m3/s for the XI-type
module shown in Figure 8.2 I .
Flow through the structure is regulated by opening or closing sliding gates. These
gates are locked in place either fully opened or fully closed since partially opened
gates would disturb the contraction of the jet. The gates slide in narrow grooves in
the 0.01 m thick vertical steel divide plates. The position of the gates should be such
that in an opened position the orifice flow pattern is not disturbed. Possible gate posi-
tions are shown in Figures 8.21 and 8.22.
Essentially two types of modules are available:
- Type XI**: This single baffle module is shown in Figure 8.21 and has a unit discharge

of O. 1O0 m2/s;
- Type XX2**: This double baffle module has two inclined orifice blades, the up-
stream one having the dual function of contracting the jet at low heads and of acting
as a weir at high heads. Water passing over the upstream blade is deflected in
an upstream direction and causes additional contraction of the jet through the
downstream orifice. As a result the discharge through the structure remains within
narrow limits over a considerable range of upstream head. The type XX2 has a
unit discharge of 0.200 m2/s.Details of the module are shown in Figure 8.22.
If unit discharges other than those given in the examples are required, the module
may be scaled up according to Froude scale law.

8.7.2 Discharge characteristics

At low heads the upper nappe surface is not in contact with the inclined baffle plate

* T h e module was developed and is commercially manufactured by Alsthom Fluides, 93121 La Courneuve,
France. Our listing of this supplier should not be construed as an endorsement of this company or their
product by the present writers.

,1 ** The Roman numeral stands for the discharge in I/s per 0.10 m width and the Arab numeral 1 of 2
stands for the number of baffles.

299
 LOCK C O N T g
DEMONTABLE BLOCKING PIN

rT

Figure 8.21 Module type XI dimensions (after Neyrpic)

300
 SLIDING GATE

DIMENSDNS IN M M

Figure 8.22 Module type XX2 dimensions (after Neyrpic)

30 1
Photo 2 Neyrpic module type X60

and the structure acts as a short-crested weir with rectangular control section. Accord-
ing to Section 1.10, the head-discharge equation for such a weir reads:

(8-31)

The discharge coefficient c d is shown in Figure 8.23 as a function of the dimensionless

ratio HJr. Since for practical reasons h, is used instead of HI, the approach velocity
coefficient C, was introduced. The value of C, = (H,/hJ3/*is related to the ratio
+
Cdh, b,/(h, pl)B, and can be read from Figure 1.12.
If the weir discharge approximates the design discharge plus 5%, the upper nappe
surface touches the inclined baffle plate and orifice flow commences (Figure 8.24).
With rising head, flow passes through a transitional zone to stable orifice flow. As
shown in Section 1.12 the modular discharge through an orifice equals

302
1.3 -

1.2

1.1

1.o
/

303
 N
S

E'
s5 i
mi
E
Y
%
I
U

Pt
.FE
'C E
4 .

V
E U
C
m m
0 .

Q
01. OL uv WUI av3~

304
 UPSTREAM HEAD
in metres

-10% -5% Q +5% .IO%

DISCHARGE ,
m3is per m

Figure 8.25 Discharge characteristics of Neyrpic module type XX2 (rising stage)

8.7.3 Limits of application

The limits of application of the Neyrpic module are:

a. The upstream water level should be kept between the limits shown in Figures 8.21
and 8.22;
b. To reduce the influence of the approach velocity on the flow pattern through the
module, the ratio hd/p,should not exceed unity;
c. To prevent the tailwater channel bottom from influencing the flow pattern through
the orifice, the ratio pz/hdshould not be less than 0.35;
d. To obtain modular flow, the ratio h,/hd should not exceed 0.60.

305
8.8 Danadean tub
8.8.1 Description

The Danadean tub is a vessel which receives a flow of water from above and discharges
it through a (circular) orifice or a (rectangular) slot in its bottom. After some time
the water surface in the Danadean tub stabilizes to a head h,, being the head that
makes the orifice discharge at the same rate as water flows into the tub (Qin = Q,,,).
The head h, can be read by means of a piezometer as shown in Figure 8.26. If-the
area A of the orifice is known, the discharge can be calculated (see Section 8.8.2).
If the head-discharge equations are to be applicable, however, the contraction of the
jet must not be hindered. Therefore, the bottom of the tub must have a minimum
clearance of d/6 to the free water surface below the tub. Here 6 denotes the ratio
of the cross-sectional area of the fully contracted jet to that of the efflux section. The
ratio 6 is known as the contraction coefficient.
The bottom of the tub must be smooth and plane so that the velocity component
along the bottom (upstream face of orifice plate) is not retarded. Provided that the
tub bottom has a perfectly plane surface, it may be horizontal or sloping under an
angle p as shown in Figure 8.27.

8.8.2 Evaluation of discharge

To determine the discharge through the opening in the Danadean tub, we 'use an
equation similar to Equation 1-67. This reads
Q=G A $ & (8-33)

e
arant

fQ0"t

Figure 8.26 Danadedn tub (circular)

306
Figure 8.27 Definition sketch for orifice (circular) and slot (rectangular)

The discharge coefficient, c d , depends on the contraction of the jet, which, obviously,
is a function of the boundary geometry of the tub. Sufficient values of the contraction
coefficient are given in Table 8.5 to permit interpolation for any boundary condition.

Table 8.5 Coefficients ofjet contraction

b d p = 45" = 90" p = 135" p = 180"

i Or D 6 6 6 6

0.0 0.746 0.61 1 0.537 0.500

o. 1 0.747 0.612 0.546 0.513
0.2 0.747 0.616 0.555 0.528
0.3 0.748 0.622 0.566 0.544
0.4 0.749 0.63 1 0.580 0.564
0.5 0.752 0.644 0.599 0.586
0.6 0.758 0.662 0.620 0.613
0.7 0.768 0.687 0.652 0.646
0.8 0.789 0.722 0.698 0.691
0.9 0.829 0.781 0.761 0.760
I .o I .O00 I .O00 1.O00 1.o00

(after Von Mises 19 17)

By using the contraction coefficient in the continuity and pressure-velocity equations

(Bernoulli), Rouse (1 948) gives the following relationships for the discharge coefficient
of water flowing through a slot
6
c - (8-34)
d-Jm
The corresponding expression for Cdfor discharge from an orifice reads
6
c, = (8-35)
-J
307
 and
B D

Figure 8.28 Variation of efflux coefficients with boundary proportions. Valid if = 90" (after Rouse 1949)

Since the right-hand term of each equation is a function of quantities depending on

boundary geometry, the discharge coefficient Cdcan be evaluated. A typical plot of
Cdversus boundary geometry is shown in Figure 8.28 to indicate its trend in compa-
rison with that of 6.

If reasonable care and skill has been applied in the construction and installation of
a Danadean tub, the discharge coefficient may be expected to have an error of about
2%. The method by which this error is to be combined with other sources of error
is shown in Annex 2.
The reader may be interested to note that the discharge equation and related coeffi-
cient values given also apply if the orifice is placed at the end of a straight vertical
pipe which discharges its jet free into the air.

8.8.3 Limits of application

The limits of application of the Danaidean tub are:

a. The edge of the opening should be sharp and be in accordance with the profile
shown in Figure 5.1 ;
b. The ratios b/B and d/D should not exceed 0.8;
c. The contraction of the jet must not be hampered. To ensure this, the bottom of
the tub must have a minimum clearance of d/6 (or b/6) above the downstream
water level.

308
1 8.9 Selected list of references
Al-Khoudari, N . 1973. The module masque type-X. Agric. Univ. Wageningen. MSc. Thesis. 39 pp.
Armco Drainage & Metal Products Inc. 195I . Water measurement tables. Portland, Oregon.
Franke, P.G. and F. Valentin 1969. The determination of discharge below gates in case of variable tailwater
conditions. J. of Hydraulics Res. 7. No. 4, pp. 433-447.
Gentilini, B. 1947. Flow under inclined or radial sluice gate. Technical and experimental results. La Houille
Blanche 1947, Vol. 2.
Gruyter, P. de 1926. Een nieuw type aftap - tevens meetsluis. De Waterstaats-ingenieur. (No.12) and 1927
(No. I). Batavia, Java.
Henderson, F.M. 1966. Open channel flow. The MacMillan Comp., New York 522 pp.
Henry, H.R. 1950. Discussion to Diffusion of submerged jets. Transactions of the American Society of
Civil Engineers. 115. pp.687-697.
Kruse, E.A. 1965. The constant-head orifice farm turnout. U.S. Dept. of Agric. Report ARS 41-93, Fort
Collins, Col. 24 pp.
Metzler, D.E. 1948. A Model Study of Tainter Gate Operation. State Univ. of Iowa MSc Thesis.
Mises, R. von 1917. Berechnung von Ausfluss und Uberfallzahlen. Z. des Vereines Deutscher lngenieure
61. No.21, pp. 447-452; No.22, pp. 469-474; No.23, pp. 493-498. Berlin.
Neyrpic. 1955. Irrigation canal equipment. Ets Neyrpic, Grenoble, France. 32 pp.
Romijn, D.G. 1938. Meetsluizen ten behoeve van irrigatiewerken. Handleiding uitgegeven door de Vereni-
ging van Waterstaat-ingenieur in Nederlandsch-Indi. Bandung 58 pp.
Schuster, J.C. 1970. Water measurement procedures. Irrigation operators workshop. U.S. Bureau of Recla-
mation, Denver, Col. Report REC-OCE-70 -38.49 pp.
Stokmans, J.A. 1970. Metingen aan de constant-head-orifice-turnout. Agric. Univ. Wageningen. MSc. The-
sis not published.
Toch, A. 1952. The Effect of a Lip Angle Upon Flow Under a Tainter Gate. State Univ. of Iowa MSc
Thesis.
U.S. Army Engineer Waterways Experiment Station 1939. Model Study of the spillway for New Lock
and Dam No. I . St.Lucie Canal, Florida. Technical Memorandum No.153-1, Vicksburg, Miss.
U.S. Army Engineer Watenvays Experiment Station 1954. Spillway for New Cumberland Dam, Ohio River,
West Virginia. Technical Memorandum No. 2-386, Vicksburg, Miss.
U.S. Army Engineer Waterways Experiment Station 1958. Stilling Basin for Warrior Dam, Ohio River,
West Virginia. Technical Report No. 2-485, Vicksburg, Miss.
U.S. Army Engineer Waterways Experiment Station 1960. Spillways and Stilling Basins, Jackson Dam,
Tombigbee River, Alabama. Technical Report No.2-53 I , Vicksburg, Miss.
U.S. Army Engineer Waterways Experiment Station 1960. Hydraulic design criteria. Sheets 320-4 to 320-7.
Vicksburg, Miss.
U.S. Bureau of Reclamation. Water measurement manual 1967. U.S. Gov. printing office, Washington.
Second Ed. 326 pp.
U.S. Soil Conservation Service 1962. National Engineering Handbook. Chap. 9, Section 15: Measurement
of irrigation water. U.S. Government printing office, Washington. 70 pp.

309
9 Miscellaneous structures
9.1 Divisors
9.1.1 Description

Many of the worlds older irrigation systems are co-operative stock companies in
which the individual water users have rights to proportional parts of the supply of
water furnished by their canal system, the divisions being in the ratio of the stock
owned in the canal company. Under this system it was often considered unnecessary
to measure the water so long as each user got his proportionate part of it. This led
to the use of divisors or division boxes as have been described by Cone (1917). These
divisors, however, are not recommended for use as measuring devices where any consi-
derable reliability is required, and will not be described here. Our attention will be
confined to divisors which can be used both for measuring and for making a fair divi-
sion of the water.
Most divisors are built to divide the flow in a ditch into two ditches, but they are
sometimes made to divide the flow into three parts or more. The divisor consists essen-
tially of a weir and a movable partition board. The partition board is hinged as shown
in Figure 9.1. Provision is usually made for locking the board to a timber or steel
profile across the weir crest when the desired set has been made.

Figure 9.1 Divisor (adapted from Neyrpic)

31 1
 Photo I Proportional divisor with fixed pier in between the two weirs

The structure shown in Figure 9.1 was designed by Neyrpic and consists of a slightly
curved weir sill with a 60-degree sloping upstream face and a 12-degree sloping down-
stream face. The weir crest is rounded, its radius being equal to r = 0.2 h,,,,, where
hlmaxis the maximum upstream head. Viewed from above, the weir crest is curved
with a minimum radius of 1.75 b,; the crest width b, should not be less than 2 Hlmax.
The upstream head, h,, is to be measured in a rectangular approach channel at a dis-
tance of between 2 hlmax and 3 hlmax upstream from the weir crest.
The upstream edge of the partition board should be sharp (< 0.005 m thick) and
should be located immediately downstream of the weir crest, in the area where flow
is super-critical. A disadvantage of sharp-edged partition boards is that trash and
floating debris are caught, so that frequent maintenance is required to obtain a propor-
tional division of water.
The flow-wise weir profile is not a determining factor in the proportional division
of water. In principle, any crest profile is suitable, especially the broad-crested weir
(Section 4.1). An example of such an application is shown in Photo 2.

9.1.2 Evaluation of discharge

I .
According to Section 1.lo, the basic head-discharge equation for a short-crested weir
with a rectangular control section reads

312
where the approach velocity coefficient C, may be read from Figure 1.12 as a function
of the area ratio CdA*/A,.The discharge coefficient of the Neyrpic weir profile is
a function of the ratio H,/r as shown in Figure 9.2.'
The modular C,-values shown in Figure 9.2 are valid if the weir crest is sufficiently
high above the average bed of both the approach and tailwater channel so as not
to influence the streamline curvature above the weir crest. To ensure this, the ratio
pl/HI should not be less than 0.33 and the ratio p2/HIshould not be less than 0.35.
To obtain modular flow, the ratio H,/H, should not exceed 0.60. It should be noted
that the weir width b, is measured along the curved weir crest.
The accuracy of the discharge coefficient of a well maintained divisor which has
been constructed with reasonable care and skill will be sufficient for field conditions.
The error in the product cdc, may be expected to be less than 5 per cent. The method
by which this error is to be combined with other sources of error is shown in Annex 2.

9.1.3 Limits of application

The limits of application of a divisor equipped with a Neyrpic weir crest are:
a. The upstream head over the weir crest h, should be measured at a distance of 2
to 3 times hlmaxupstream from the weir crest. The recommended lower limit of
h, = 0.06m;
b. To prevent water surface instability in the approach channel, the ratio pl/HIshould
not be less than 0.33;

Photo 2 A broad-crested weir with a movable partition board

313
DISCHARGE COEFFICIENT
cd

RATIO H,/r

Figure 9.2 Cd-valuesas a function of the ratio H,/r

c. To prevent the tailwater channel bottom from influencing the flow pattern over
the weir crest, the ratio p2/H,should not be less than 0.35;
d. To reduce boundary layer effects of the vertical side walls, the ratio b,/H, should
not be less than 2.0;
e. To obtain sensibly two-dimensional flow over the weir crest, the horizontal radius
of curvature of the weir crest should not be less than 1.75 b,;
f. The ratio H,/r should not be less than 0.20;
g. To obtain modular flow, the ratio H2/H,should not exceed 0.60.
If these limits cannot be met, the use of a broad-crested weir is recommended.

9.2 Pipes and small syphons

9.2.1 Description

On irrigated farms, short sections of pipe are frequently used to distribute water over
the fields. Commonly used for this purpose are plastic, aluminium, or galvanizkd steel
pipes and siphons. Some examples are shown in Figure 9.3.
If such pipes are to be used to estimate discharges, the hydraulic losses at the en-
trance and exit of the pipe have to be known. To prevent these losses from varying
too greatly, we have drawn up instructions for use which are listed under the limits
of application (Section 9.2.3).
The effective (differential) head, Ah, over the pipe or siphon has to be measured
as accurately as possible, but the installation also has to be practical. For field measure-
ments a transparent hose acting as a siphon, as illustrated in Figure 9.4, will be found
useful. By keeping the hose in a vertical prosition Ah can be read from a scale. Since
tailwater level will drop as soon as the device is installed, the meter has to be placed
and read quickly to obtain a reasonably accurate Ah-value.

314 I
9.2.2 Evaluation of discharge

From a hydraulica1 viewpoint, two types of pipes (or siphons) can be distinguished:
- small diameter pipe, being a pipe with a .length L considerably more than D,
(L > 20D,);
- large diameter pipe, which has a relatively short length of 6 D, < L d 20 D,.
For either pipe the discharge can be evaluated with the equation

-
-
FIELD LATERAL DITCH

PIPE
A. SUBMERGED TAKE OUT

WATER SURFICE

0. FREE-DISCUARGING PIPE

EFFECTIVE HEAD, ah

C. SUBMERGED PIPE

,-EFFECTIVE HEnD,ah

D. SlPM

Figure 9.3 Discharge through ditch-furrow pipes and siphons

315
 FLEXIBLE
r:teFSPARENT

SUPPLY DITCH

Figure 9.4 Method of head measurements

where v is average flow velocity in pipe and 5 denotes the head loss coefficient.
Substituting the continuity equation into Equation 9-2 yields
Q ED;(^)
2gAh OS
(9-3)
4
For small diameter pipes friction losses in the pipe play a significant role and the
head loss coefficient is estimated to equal
L
k= 1.9+f- (9-4)
D P

or for pipes with a length between 1 .O0 and 1.50 m, i.e. average L = 1.25 m
1.25 f
5= 1.9+- (9-5)
DP
where f i s the friction loss coefficient of Darcy-Weissbach. For an equivalent sand
roughness k = 5 x m, f i s a function of the Reynolds number Re and the ratio
D,/k. If Re > lo5, k = 5 x 10-5m, and 300 < D,/k < 1200, it follows that
0.028 > f > 0.019.
For the large diameter pipes entrance and exit losses are the most significant
sources of hydraulic losses and the head loss coefficient is estimated to equal
6 = 2.1 (9-6)
A combination of Equations 9-3 and 9-5 results in Figure 9.5 from which the pipe
discharge can be read as a function of Ah and D, for small diameter pipes. A combina-
tion of Equations 9-3 and 9-6 produces Figure 9.6, from which similar information
about large diameter pipes can be obtained.
The error in the discharge read from Figures 9.5 and 9.6 is expected to be about
10%. The method by which this discharge error is to be combined with errors in Ah
and D, is shown in Annex 2.

316
 1.20

1.00
O90
0.00
OM

it
.-c
,035
Q
0.45
0.40

a30
a=
am
I 0.18
OS6
I 0.14
a12

O10
M9
am
0.07
Oab
0.05

0.04

om
2 3 4 5 6 7 0 9 1 0
-0 in m%x103
Figure 9.5 Rates of flow through smooth pipes or siphons

9.2.3 Limits of application

To produce a reasonably accurate estimate of the discharge through a pipe or siphon,

the following limits of application are considered essential.
a. Pipes should have clear cut edges (no rounding-off) and a constant diameter from
entrance to end. The pipe entrance should protrude from the ditch embankment
and the flow velocity in the ditch should be less than one third of the average velocity
in the pipe;
b. The pipe should be made of technically smooth material. For D, < 0.05 m, PVC
or aluminium are suitable, while if D, > 0.05 m galvanized steel is also suitable;
c. To prevent air-bubbles from collecting at the top of a siphon, it is recommended
that v 2 1.3 (g D, sin E ) O . ~ , where c1 denotes the angle of the downstream siphon
limb from the horizontal;
d. To eliminate bend-losses, the radius of bends should not be less than 8 D,;
e. No air-entraining vortex should be visible at the pipe entrance;
f. The exit cross-section of the pipe has to flow entirely full. For a free discharging
horizontal pipe, this occurs if Q > 1.18 D,2.5m3/s. (See also Sections 9.4 and
9.5);
g. The recommended lower limit of Ah is 0.03 m. The recommended lower limit of
D, is 0.015 m for small diameter pipes and 0.03 m for large diameter pipes.

317
 1 .zo
1 .o0
am
0.70
0.60
0.50
0.45
0.40
0.35
am
0.25

0.20
0.18
0.16
0.14
012

0.10
0.09
0.08
0.07
0.06
0.05

0.04

a03
I / I / I /I I l I l / l I 1 1 / 1 1 1 /I I I I J I I I I I I
1 2 3 4 5 6 7 8 9 x) 12 14 16 1820 25 30 35 4 0 50 60 7080 100
-GI in m3s x 1 ~ - 3
Figure 9 .6 Rates of flow through smooth pipes or siphons

9.3 Fountain flow from a vertical pipe

9.3.1 , Description

Fountain flow from a vertical pipe into the air can occur during pumping tests, or
when there is flow from pressure conduits or from artesian wells. Such flow can occur
either as weir flow or as jet flow.

Weirf l o w
Water discharges from the pipe with sub-critical flow and is similar to flow over a
curved sharp-crested weir. Weir flow occurs if the height to which the water rises above
the pipe is equal to or less than 0.37 D,.

Jetflow
Water discharges from the pipe with supercritical flow. Jet flow occurs if the height
of the jet exceeds 1.4 D,, as determined by sighting over the jet to obtain the average
rise.

The principal difficulty of measuring the discharge from a vertical pipe is to get an

318
accurate measurement of the height to which the water rises above the end of the
pipe. This is usually done with a sighting rod. As shown in Figure 9.7, the sighting
rod is attached to the pipe from which the jet is to come. To obtain proper head read-
ings, we have to set the movable arm at the head at which the water stays the longest
time. Thus we measure its average head, not the maximum head.

9.3.2 Evaluation of discharge

The discharge from a vertical pipe can be estimated by using the equations given by
Lawrence and Braunworth (1 906), which for sighting rod readings in the metric system
are:
Q = 5.47 D,' *'h,' 35 (9-7)
and
Q = 3.15D,'99h,053 (9-8)
Equation 9-7 is valid for weir flow (h, < 0.37 D,) and Equation 9-8 is valid for jet
flow (h, 2 1.4 D,). For jet heights between 0.37 D, and 1.4 D,, the flow is somewhat

WVABLE ARM OR WINTER

TO SET TO mp OF JET

GRADUATED SCALE

ZERO OF Y A L E LEVEL
WITH EDGE OF PIPE

CLAMP m HOLD GRADUATED YALE

VERTICAL PIPE FROM WHICH JET IS TO COME

Figure 9 7 Sketch showing application of movable pointer and scale in measuringjets

319
less than given by either of these equations. Figure 9.8, prepared from Lawrence and
Braunworth data, shows flow rates in m3/s for standard pipes and for jet heights up
to 4.0 m.
The accuracy with which the jet flow can be evaluated may be expected to be about
15% for sighting rod readings. For weir flow these accuracies are about 20%.

9.3.3 Limits of application

The limits of application that enable a reasonable estimate of the discharge from a
vertical pipe are:
a. Pipes should have clear cut edges and a constant diameter over at least a length
of 6 D,;
b. Pipes should be vertical for at least a length of 6 D, from the top of the pipe;
c. The practical range of pipe diameters is 0.025 m < D, < 0.609 m;
d. The practical range of heads is 0.03 m < h, < 4.0 m.

DISCHARGE Q (m3/sec)

Figure 9.8 Discharge from vertical pipes

320
I
9.4 Flow from horizontal pipes
9.4.1 Description

Flow from a horizontal pipe can be estimated by using either the California pipe
method* developed by Van Leer (1922) or the trajectory method developed at Purdue
University by Greeve (1 928). The California pipe method applies only to pipes flowing
less than half full, whereas the more general trajectory method applies equally well
to both partially and completely filled pipes. The California pipe method consists of
measuring the end depth at the pipe outlet and is valid if ye = D, - Y < 0.56 D,
(see Figure 9.9).
The Purdue trajectory method consists of measuring two coordinates of the upper
surface of the jet as shown in Figure 9.10. If the pipe is flowing with a depth of less

PARTIALLY FILLED P I E
( ~ ~ $ 0 Dpfcr
5 6 XzO)

Figure 9.9 Dimension sketch partially filled pipe

SCALE ADJUSTED AND X

-
I
I

-
FIXED TO DESIRED X-VALUE
c I
>

-
SET SCREW

PIPE FLOWING FULL

Figure 9.10 Purdue coordinate method

\ \
\

* The California pipe method is identical to the brink depth method for circular canals.

32 1
Photo 3 Flow from a horizontal pipe

than 0.56 D, at the outlet, the vertical distance from the upper inside surface of the
pipe to the surface of the flowing water, Y, can be measured at the outlet of the pipe
where X = O. For higher discharges, Y can be measured at horizontal distances X
from the pipe outlet of O. 15,0.305 or 0.46 metre.

9.4.2 Evaluation of discharge

California pipe method (X = O)

The California pipe method is based on the unique relationship between the depth,
ye, of flow at the pipe outlet and the pipe discharge, Q. A dimensionless plot of this
relationship is shown in Figure 9.1 1.
Provided that ye d 0.56 D, the pipe discharge can be calculated from this figure
for any diameter D,. Discharge values in m3/s x for 2- to 6-inch diameter (0.05
to O. 15 m) standard pipes are shown in Figure 9.13A as a function of Y = D, - ye.
The user will experience difficulty in making the measurement Y exactly at the brink.
Since the upper nappe surface is curved, any small error in the location of the gauge
will cause large errors in Y . Actually, the only method by which Y can be measured
accurately is by installing a point gauge at the center line of the pipe exactly above
the brink (see also Figure 9.10). Since the upper nappe surface at the brink is instable,
the accuracy of the Y-value can be greatly improved by repeating its measurement

322
and taking the average value.
The error in the discharge value as derived from Figure 9.1 1 for partially filled pipes
may be expected to be less than 3 per cent. The method by which the various errors
have to be combined with other sources of error is shown in Annex 2.

Purdue trajectory method

The shape of the jet from a horizontal pipe can be interpreted by the principle of
a projectile (Figure 9.12). According to this principle, it is assumed that the horizontal
velocity component of the flow is constant and that the only force acting on the jet
is gravity. In time t, a particle on the upper surface of the jet will travel a horizontal
distance X from the outlet of the pipe equal to
x = v,t (9-9)
where v, is the velocity at the point where X = O. In the same time t, the particle
will fall a vertical distance Y equal to
Y = gt2 (9- 1O)

R A T I O YelDp
0.8

0.7

O .6

0.5

O .4

o .3

o .2

o .1

O
O 1 .2 .3 .4 .5 .6 .7 .e

Figure 9. I 1 Flow from horizontal pipes by California pipe method or brink depth method

323
 \
\
\
\
\
Y
\
b
Figure 9.12 Derivation ofjet profile by the principle of projectile

Eliminating t from the above two equations and multiplying each term by the inside
pipe area I4
?T D: and a discharge coefficient ( c d N 1.10) leads to

X2
Q = Cd 14 nD,2 J g 2y (9-1 1)

Discharge values in m3/s x IO (I/s) for 2- to 6-inch diameter (0.05 to O. 15 m) standard

pipes are shown in graphs in Figure 9.13B to D.
Due to the difficulty of making the vertical measurement Y in the Purdue trajectory
method (ye > 0.56 D, or pipe flowing full), the error in flow measurement found
COORDINATE Y
In metres

1 .2 .3 .4 .5 6 .7 .e 9 1 2 3 4 5 6 78910 20 30

MSCHARGE I m3/. l3 o/5)

Figure 9.13A Flow from horizontal pipes by either Purdue trajectory method or by California pipe method

324
by using Figure 9.13 may be expected to be about 10 to 15 per cent. If this error
is not to be exceeded, the pipe should be truly horizontal and straight for at least
6 times D, from the outlet. If it slopes downward, the discharge taken from Figure
9.13 will be too low. If it slopes upward, the discharge will be too high.

COORDINATE Y
in metres

.om .O04 .O06 D10 ,015 .O2 .O3 .O4 05 .O7 .lo0
DISCHARGE in m3/5

Figure 9.13B (cont.)

COORDINATE Y
in metres

.o02 .o04 .o07 .O1 .O15 .O2 .O3 .O4 .O6 .O8 .10
DISCHARGE in 177315
Figure 9.13C (cont.)

325
 E in metres
C O ~ R D ~ N A TY

.O02 .O04 .O07 .a .O2 .O3 .O4 .O6 .O8 10

DISCHARGE in m3/s
Figure 9.13D(cont.)

9.4.3 Limits of application

The limits of application that enable a reasonably accurate estimate of the discharge
from a horizontal pipe are:
a. Pipes should have clear cut edges and a constant diameter over at least a length
of 6 D, from the outlet;
b. Pipes should be straight and truly horizontal over at least a length of 6 D, from
the outlet;
c. Pipes must discharge freely into the air.

9.5 Brink depth method for rectangular canals

9.5.1 Description

When the bottom of a low gradient canal drops suddenly, a free overfall is formed
which, since flow changes to supercritical, may be used as a discharge measurement
device. In principle, any canal cross section can be used for flow measurement provided
that the free overfall is calibrated.
Sufficiently accurate experimental data, however, are only available for rectangular
and circular cross sections. Since the circular section was treated in Section 9.4, we
will confine our remarks here to the brink depth method for rectangular canals.

The simplest case of a free overfall is that of a rectangular canal with sidewalls continu-
ing downstream on either side of the free nappe over a distance of at least 0.3 H,,,,,
so that at the brink the atmosphere has access only to the upper and lower side of
the nappe. This is a two-dimensional case with a confined nappe, and is the only
form of the problem for which serious attempts have been made to find a solution.

326
Some experiments, however, have been made on a free overfall with unconfined
nappe, i.e. where the side walls end at the sudden drop.
In the situation shown in Figure 9.14, flow takes place over a confined drop which
is sharp enough (usually 90 degrees) to guarantee complete separation of the nappe.
The bottom of the tailwater channel should be sufficiently remote so as not to influence
the streamline curvature at the brink section. To ensure that this does not happen,
the drop distance should be greater than 0.6 yc.
The user will experience difficulty in making the measurement ye exactly at the brink.
Since the upper nappe surface is curved, any small error in the location of the gauge
will cause large errors in ye. Actually, the only method by which ye can be measured
accurately is by installing a point gauge in the middle of the canal exactly above the
brink. Since a point gauge is vulnerable to damage, however, a staff gauge, with its
face flush with the side wall, will be found more practical. The location of the brink
should be marked on the gauge face to enable ye readings to be made. The brink depth
as measured at the side wall will be higher than that in the middle of the canal, because
of side wall effects. To limit the effect of roughness on the brink depth as measured
with a staff gauge, the side walls as well as the bottom of the canal should be smooth.
If the brink depth is measured with a point gauge, no significant influence of roughness
is found, as is illustrated for three values of the equivalent sand roughness, k, in Figure
9.15.

9.5.2 Evaluation of discharge

If we assume that the streamlines in the rectangular canal are straight and parallel,
we may, according to Equation 1-26, write the specific energy in the canal as

(9-12)

Differentiation of Ho to y, while q remains constant leads to

(9- 13)

.l

M -.

MINIMUM DROP DISTANCE -- - -+q

Figure 9.14 Flow profile at the free overfall

327
 y, in m.

.O7

smooth channel
o so=o
.O6 A so~o.oo(
+ s0~0.002
x S0=00025

.O5
kZ0.08m x109
a so=o
.O4
* so~o.ool
o so=oocQ
4 S0:Q0025

k:0.30m x10-3
o so=o
o so~o.ool
.o2
+ s0=aoo2
e S0z0.0Q25

.o1

O
.o1 02 .O4 .05
-
t .07 .O8 .(
yein m

Figure 9. I5 Relation between ye and yc (after Kraijenhoff van de Leur and Dommerholt 1972)
9

If the depth of flow is critical (y = yc), dH,/dy equals zero, and we may write

(9- 14)

Assuming CL = 1 and substituting Q = bcq leads to

Q = bc& Ycy2 (9- 15)

The experiments of Rouse (1936), and further experiments by various authors, showed
that for a confined nappe the brink section has a flow depth equal to

328
 ye = 0.715 yc (9- 16)
resulting in the discharge equation

= 5.18bcye32 (9- 17)

Q = ( A 5y
b c & 0.71
As shown in Figure 9.15, slight variations in the roughness of the canal boundaries
and in the canal bottom slope are of little significance on the ratio ye/yc. If the free
overfall has an unconfined nappe, however, the ratio ye/yc is somewhat less than in
the two-dimensional case, being equal to 0.705.
For a free overfall which is constructed and maintained with reasonable care and
skill, the coefficients 0.715 and 0.705 can be expected to have an error of the order
of 2% and 3% respectively, provided ye is measured in the middle of the channel. If
ye is measured at the side walls an additional error in ye occurs due to boundary rough-
ness (see Section 9.4.2 for other possible errors). The method by which these errors
are to be combined with other sources of error is shown in Annex 2.

9.5.3 Limits of application

The limits of application of the brink depth method for rectangular canals are:
a. Perpendicular to the flow, the brink should be truly horizontal and the side walls
of the rectangular approach canal should be parallel from end to end;
b. To obtain a uniform velocity distribution, the length of the approach channel
should not be less than 12 ye;
c. The longitudinal slope of this approach channel should preferably be zero but not
more than s = 0.0025;
d. The practical lower limit of ye is related to the magnit.ude of the influence of fluid
properties and the accuracy with. which ye can be measured. The recommended
lower limit is 0.03 m;
e. The y,-value should be measured in the middle of the canal, preferably by means
of a point gauge;
f. The width of the canal should not be less than 3 yemaxnor less than 0.30 m;
g. To obtain free flow, the drop height should not be less than 0.6 yc,,,.

9.6 Dethridge meter

9.6.1 Description

The Dethridge meter is a rather commonly used device for measuring the volume
of irrigation water supplied to farms from main and lateral canals in Australia. The
meter was designed by J.S. Dethridge of the State Rivers and Water Supply Com-
mission, Victoria, in 1910. This Commission provided the present information on the
standard device, of which today about 40000 are in operation in irrigation areas
throughout Australia. The meter consists of an undershot water wheel turned by the
discharging water passing through its emplacement, which is a short concrete outlet
specially formed to provide only the minimum practicable clearance of the lower half
of the wheel at its sides and round the lowest 70 degrees of its circumference. Two

329
 standard sizes of the meter are used: the 1.524 m (5 ft) diameter large meter which
is suitable for discharges from 0.040 m3/s to 0.140 m3/s, and the small, 1.219 m (4 ft)
diameter meter for discharges from 0.015 m3/s to 0.070 m3/s. The main dimensions
of both meters, which are similar in general form, are shown in Figure 9.16.
The wheel is made up of a cylinder of 2 mm thick mild steel sheet, bearing eight
external vanes of the same material, each welded against the surface of the cylinder
on a widely distended V, with the root of the V leading in the direction of the wheels
rotation. At the root of each vane is a small air vent so that compartments between
the vanes can fill completely with water while being submexed by rotation of the
wheel. The outer corners of the vanes are chamferred.
The internal bracing used to consist of three crossed pairs of timber spokes ( f O. 10
x 0.05 m) placed at the middle and both ends of the cylinder. Today they have given
way to 16 mm steel rods in parallel pairs, welded on either side of the 25 mm internal
diameter pipe-axle of the wheel (see Figure 9.17).

The concrete structure in which the wheel has been placed has upstream of the wheel
a simple rectangular section, with level floor in the vicinity of the wheel. At the wheel
the walls remain plane and parallel but the floor is intended to accomodate an arc
of about 70 degrees of the wheels circumference. Immediately downstream of the
. wheel the walls are flared outward and the floor is sloped up to a lip of sufficient

a t e seefig.Q.17

Diameter of cylinder .
1.016m (a=)
Outside diamter
of wheel to tip of vanes
1524 m (1.219 )

\
Figure 9.16 Dethridge meter
Photo 4 One of the vanes was painted red to check the revolutions counter and required gate openings

height (see Figure 9.17) to ensure submergence of the passage swept by the vanes
under the wheel.
Most Dethridge meters are equipped with cheap wooden bearing blocks, usually
seasoned Red Gum or other durable hardwood, dressed to dimensions shown in Figure
9.18. A disadvantage of these blocks is that they wear and are not always replaced
in time so that the wheel may scrape on the concrete. A variety of more permanent
type bearings was tested under the supervision of the above mentioned Commission
and it appeared that the best installation would be a non-corrosive ball bearing which
does not require any maintenance. Details of the type adapted as standard by the
State Rivers and Water Supply Commission, Victoria, are shown in Figure 9.18.
The operational life of revolution counters mounted to the wheel axle is quite irregu-
lar due to their fragile construction, the wire connection to the axle, and the jerky
motion of the wheel. None of the counters in use can be considered satisfactory but
(since 1966) tests showed that a pendulum actuated revolution counter fitted in a sealed
casing inside the drum of the wheel may be satisfactory (see Figure 9.17 and Photo 4).
It is important that the Dethridge meter be installed at the correct level in relation
to full supply level in the undivided irrigation canal, so as to make the best use of
the generally limited head available. The standard setting of the large meter is to have
the floor of the concrete structure, at entry, 0.38 m below design supply level to the
meter, being full supply level at the next check downstream of the meter. For the
small meter this depth is 0.30 m. If excess head over the meter is available the depth
may be increased up to 0.90 m, with the necessity ofcourse, ofcorrespondingly increas-
ing the height of the sluice gate and head wall (see also Figure 9.19).
I-

331 ..
W +16mngalvanized b d l s R15m long
W forAnthony bearings 025m long for
h) timber blocks.

-5

Floor& side fillet io be given

Floor reinforcement firdl rendering to requirrd
continued into walls. (M06mclearance from vane
Of wheel.

Pendant m i g h t
rewlulion rmnier
-
spokes both ends.
Ihickvelded lo

(0.572)
DEVELOPMENT OF VANE s!andard m e d h

mlded t o d m

A L
SECTION A.A
SIDE ELEVATION

Figure 9.17 Dethridge meter dimensions (small meter dimensions shown between brackets, if different from large meter)
 0.006~0.W6xoJ2recess for cwu Plok #17mm holes for *16mm bolls
o f countrr housing in crntrr of blocks
( o n counter side o n l y )

Hole f o r a x l e
r l l m m holes f a fuing Diameter to sui1 25. int.diam.
countrr housing

TIMBER AXLE BLOCK (dimensions In mm)

135

ANTHONY BEARING DETAILS (dimensions in mm)

W
W
W Figure 9.18 Alternative wheel bearing arrangements
 TOP VIEW
upstream downstream
check check

flow
t
L 2

upstream SIDE VIEW

k-

Figure 9.19 Setting of meter in relation to supply canal

Supply level should not exceed 0.90 m above the meter sill at entry to avoid the jet
below the sluice gate from driving the wheel. This Pelton wheel effect reduces the
volume of water supplied per revolution. Discharge regulations are usually effectuated
by adjusting a sluice gate immediately upstream of the wheel. Provided that supply
level does not exceed 0.90 m above the meter sill at entry, the gate may be hand-
operated. Gates may be locked in place as shown in Figure 9.20.
The main advantage of the Dethridge meter is that it registers a volume of supplied
water; it is simple and robust in construction, operates with small headloss, and it
will pass ordinary floating debris without damage to or stoppage of the wheel.

9.6.2 Evaluation of flow quantity

If there were no clearances between the wheel and the concrete structure, the meter
would give an exact measurement of the water passing through it, as each revolution
of the wheel would pass an invariable quantity. With the provision for the necessary
clearances, however, leakage occurs through the clearance space at a rate dependent
not only on the rotation of the wheel, but dependent also on other factors such as
the difference in water levels immediately upstream and downstream of the wheel,
and the depth of submergence. For free flow over the end sill, rating curves for both
wheels are given in Figure 9.21.

As shown, the quantity of water passed per revolution of the wheel varies to some
extent with the running speed of the wheel. For the conversion of revolutions to water
quantity supplied, constant ratios are assumed, being 0.82 m3/rev for the large wheel

334
and 0.35 m3/rev for the small wheel. Leakage around the wheel increases, and thus
more water is supplied than registered, if there are large bottom clearances, large side
clearances, high tailwater levels, and if the wheel is rotating at less than about three
revolutions per minute.
The positive error resulting from excessive side clearances is smaller than that from
bottom clearances. Increase in supply level has only a small effect on the rating.

A Dethridge meter which has been constructed and installed with reasonable care

Hondles flottened ot ends

fllmm holes to be
drilled in left
side angle -
(see dejail )

1
Chain 8 Locking
pin(Sedctd1) -

meter downstream view

# These dimensions to be used for positioning h g l e s XWB

0.038 . -

GATE I N LOCKED POSTION

ANGLE 6 ANGLE A
SIDE ANSLE 8
~

OF LARGE METER SIDE ANGLES OF S M A U METER

NOTE : Angle A 0 f large meter

has no holes

Figure 9.20 Gate dimensions

335
O 0.050 ai00
I I I I I I I I I I I
O 1 2 3 4 5 6 7 8 9 10 m3/min
DISCHARGE

Figure 9.21 Rating curves for free flow over end sill for large and small meter

and skill may be expected to measure the total quantity of water passsing through
it with an error of less than 5%. It is obvious that this quite reasonable degree of
accuracy for the measurement of irrigation deliveries can only be achieved if adequate
and regular maintenance is provided.

9.6.3 Regulation of discharge

As mentioned in Section 9.6.1, the discharge through the Dethridge meter is regulated
by a sluice gate. Provided that flow over the end sill is modular, meter discharge can
be set by adjusting the gate opening according to Figure 9.22.
If the meter is submerged, the most convenient method of setting a flow rate is
to adjust the sliding gate so that the wheel makes the required revolutions per minute
to pass this flow. Figure 9.21 may be used for this purpose, provided that tailwater
levels remain less than 0.17 m over the end sill to avoid excessive leakage through
the clearances of the large wheel. For the small wheel this value is 0.13 m. Approximate
limits of tailwater level to obtain modular flow through the Dethridge meter are shown
in Figure 9.23 for both meters.

9.6.4 Limits of application

The limits of application of the Dethridge meter are:

a. The practical lower limit for the supply level over the entry sill is 0.38 m for the
large meter and 0.30 m for the small meter. The upper limit for this supply level
is 0.90 m for both meters;

336
G A T E OPENING in metres
0.33

O .!o

0.10

O a m 0.080

TURNOUT DISCHARGE m3~scc ' TURNOUT DISCHARGE m3~5eC

Figure 9.22 Gate calibration curves for Dethridge meters

DOWNSTREAM WATERDEPTH
OVER END SILL. m.

O 0.050 o .loo 0.150

&/sec

TURNHOUT DISCHARGE

Figure 9.23 Approximate limits of tailwater for modular flow over downstream lip

337
b. Tailwater level should not be more than 0.17 m over the end sill of the large meter.
This value is O. 13 m for the small meter;
c. The wheel should neither make less than about 3 r.p.m. nor more than about 10
to 12 r.p.m. Consequently, the discharge capacity ranges between 0.040 m3/s and
0.140 m3/s for the large meter and between 0.015 m3/sand 0.070 m3/s for the small
meter (see also Figure 9.21);
d. Clearance between the floor and side fillets of the structure and the wheel should
not exceed 0.006 m for both meters. Clearance between the side walls and the wheel
should not exceed 0.009 m for the large meter and 0.006 m for the small meter.

9.7 Propeller meters

9.7.1 Description*

Propeller meters are commercial flow measuring devices used near the end of pipes
or conduits flowing full, or as in-line meters in pressurized pipe systems. The meters
have been in use since about 1 9 13 and are of many shapes, kinds, and sizes. The mate-
rial presented in this section applies to all makes and models of meters, in general,
and serves to provide a better understanding of propeller operation.
Propeller meters utilize a multibladed propeller (two to six blades) made of metal,
plastic, or rubber, rotating in a vertical plane and geared to a totalizer in such a manner
that a numerical counter can totalize the flow in cubic feet, cubic metres, or any other
desired volumetric unit. A separate indicator can show the instantaneous discharge
in any desired unit. The propellers are designed and calibrated for operation in pipes
and closed conduits and should always be fully submerged. The propeller diameter
is always a fraction of the pipe diameter, usually varying between 0.5 to 0.8 D,. The
AIR E N T 6fDp
/
, O10 m min

-
-

I IVANES I
I
I
I
I 2 7 OD I
I

Figure 9.24 Typical propeller meter installation

* The information presented in this section is for the major part an abstract from an excellent review on
propeller meters by Schuster and USBR (1970 and 1967)

338
measurement range of the meter is usually about 1 to IO; that is the ratio y = Qmax/Qmin
N IO. The meter is ordinarily designed for use in water flowing at 0.15 to 5.0 m/s
although inaccurate registration may occur for the lower velocities in the O. 15 to 0.45
m/s range. Meters are available for a range of pipe sizes from 0.05 to 1.82 m in diameter.
The principle involved in measuring discharges is not a displacement principle as
in the Dethridge meter described earlier, but a simple counting of the revolutions of
the propeller as the water passes it and causes it to rotate. Anything that changes
the pattern of flow approaching the meter, or changes the frictional resistance of the
propeller and drive gears and shafts, affects the accuracy of the meter registration.

9.7.2 Factors affecting propeller rotation

Spiral flow

Spiral flow caused by poor entrance conditions from the canal to the measuring culvert
is a primary cause of discharge determination errors. Depending on the direction in
which the propeller rotates, the meter will over or under register. Flow straightening
vanes inserted in the pipe upstream from the propeller will help to eliminate errors
resulting from this cause. Meter manufacturers usually specify that vanes be several
pipe diameters in length and that they be located in a straight, horizontal piece of
pipe just upstream from the propeller. The horizontal pipe length should not be less

Photo 5 To avoid such a vortex the gate opening must be sufficiently deep below the upstream water level

339
than 7 D,. Vanes are usually made in the shape of a plus sign to divide the pipe into
equal quarters. Because the area taken up by the vanes near the centre of the pipe
tends to reduce the velocity at the centre of the propeller such a vane type has a negative
influence on the registered discharge (about 2%) and some manufacturers suggest us-
ing vanes that do not meet in the middle. One or two diameters of clear pipe, however,
between the downstream end of the vanes and the propeller will nullify any adverse
effects caused by either type of vane.
If straightening vanes are not used, a long length of straight horizontal pipe (30
or more diameters long) may be required to reduce registration errors.

Velocity profiles

Changes in velocity distribution, or velocity profile, also influence registration. If the

distance between the intake and the propeller is only 7 or so diameters long, the flow
does not have time to reach its normal velocity distribution, and a blunt, rather evenly
distributed velocity pattern results as shown in Figure 9.25, Case A. On the other
hand, if the conduit length is 20 to 30 diameters or longer, the typical fully developed
velocity profile as shown in Figure 9.25, Case B, occurs.
Here, the velocity of flow near the centre of the pipe is high compared with the
velocity near the walls. A meter whose propeller diameter is only one-half the pipe
diameter would read 3 to 4 per cent higher than it would in the flat velocity profile.
A larger propeller could therefore be expected to produce a more accurate meter be-
cause it is driven by more of the total flow in the line. Laboratory tests show this
to be true. When the propeller diameter exceeds 75 per cent of the pipe diameter,
the changes in registration due to variations of the velocity profile are minor.

W a i t y profiles

_
_- --
Short pipe Long pipe RopelIer
CASE A , . CASE B

Figure 9.25 Velocity profiles (after Schuster 1970)

Propeller motion

Since the meter, in effect, counts the number of revolutions of the propeller to indicate
the discharge, any factor that influences the rate of propeller turning affects the meter
registration. Practically all propeller effects reduce the number of propeller revolution
which would otherwise occur, and thus result in under-registration. Propeller shafts
are usually designed to rotate in one or more bearings. The bearing is contained in
a hub and is protected from direct contact with objects in the flow. However, water

340
often can and does enter the bearing. Some hubs trap sediment, silt, or other foreign
particles, and after these work into the bearing a definite added resistance to turning
becomes apparent. S?me propellers are therefore designed for flow through cleaning
action so that particles do not permanently lodge in the bearings. Care should be
taken in lubricating meter bearings. Use of the wrong lubricant (perhaps none should
be used) can increase the resistance to propeller motion, particularly in cold water.
It should also be established that the lubricant is reaching the desired bearing or other
surfaces after it is injected. For some meters, the manufacturers do not recommend
lubrication of the bearings.
Floating moss or weeds can,foul a propeller unless it is protected by screens. Heavy
objects can break the propeller. With larger amounts, or certain kinds, of foreign ma-
terial in the water, even screens may not solve the problem.
The propeller meter will require continuous maintenance. Experience has shown
that maintenance costs can be reduced by establishing a regular maintenance pro-
gramme, which includes lubrication and repair of meters, screen cleaning, replacement
about every 2 years, and general maintenance of the turnout and its approaches. In
a regular programme many low-cost preventive measures can be made routine and
thereby reduce the number of higher cost curative measures to be faced at a later
time. Maintenance costs may be excessive if meters are used in sediment-laden water.

Effect of meter setting

Unless the meter is carefully positioned in the turnout, sizeable errors may result.
For example, a meter with an 0.30 m propeller in an 0.60 m diameter-pipe discharging
0.22 m3/s, set with the hub centre 0.025 m off the centre of the pipe, showed an error
of 1.2 percent. When the meter was rotated 1 1.5O in a horizontal plane (8 mm measured
on the surface of the 76 mm-diameter vertical meter shaft housing), the error was
4 per cent; for 23 O, the error was 16 per cent (under-registration).

Effect of outlet bdx design

The geometry of the outlet box downstream from the flow meter may also affect meter
accuracy. If the outlet is so narrow as to cause turbulence, boils, and/or white water,
the meter registration may be affected.
Figure 9.26 shows two designs of outlet boxes (to scale). Design B is believed to
be the smallest outlet box that can be built without significantly affecting the meter
calibration. The vertical step is as close to the meter as is desirable. Larger outlet
structures - those providing more clearance between the meter and vertical step -
would probably have less effect on the registration. More rapidly diverging walls than
shown in Figure 9.26 should be avoided since they tend to produce eddies over the
meter and/or surging flow through the meter and/or surging flow through the turnout.
This has been observed as a continuously swinging indicator which follows the chang-
ing discharge through the meter. The surging may often be heard as well as seen.
Large registration errors can occur when rapidly or continually changing discharges
are being measured.

341
 n
Worp from vertical
to 1.5 : 1 - - --.
A

U
PLAN

A SECTION A-A

Figure 9.26 Outlet box design (after Schuster 1970)

9.7.3 Head losses

Head losses across a propeller meter are usually regarded as being negligible, although -
there is evidence that losses may run as high as two velocity heads. In many cases
turnout losses including losses through the pipe entrance, screens, sand trap, pipe,
etc., are large enough to make the losses at the meter seem negligible. Some allowance
for meter losses should be made during turnout design, however, and the meter manu-
facturer can usually supply the necessary information. Table 9.1 may serve to give
an impression of the head losses that occur over a typical propeller meter installation
as shown in Figure 9.24, and in which the horizontal pipe length is 7 D,.

Table 9.1 Head losses over propeller meter installation (after USBR 1967)

Q, m3/s D,, m Ah, m

0.085 0.30 0.50

O. 140 0.36 0.54
0.280 0.46 0.66

342
 Warp lrom vertical-,
to 2 5 : l

i 1

U
PLAN

..

B SECTION A - A

Figure 9.26 (cont.)

9.7.4 Meter accuracy

The accuracy of most propeller meters, stated in broad terms, is within 5 per cent
of the actual flow. Greater accuracy is sometimes claimed for certain meters and this
may at times be justified, although it is difficult to repeat calibration tests, even under
controlled conditions in a laboratory, to within 2 per cent. A change in lubricating
practice or lubricant, along with a change in water temperature, can cause errors of
this magnitude. A change in line pressure (the head on the turnout entrance) can cause
errors of from 1 to 2 per cent.

9.7.5 Limits of application

The limits of application of the propeller meter for reasonable accuracy are:
a. The propeller should be installed under the conditions it was calibrated for;
b. To reduce errors due to always existing differences in velocity profiles between
calibration and field structure, the propeller diameter should be as large as practic-
able. For a circular pipe a propeller diameter of 0.75 D, or more is recommended;

343
c. The minimum length of the straight and horizontal conduit upstream from the
propeller is 7 D,, provided flow straightening vanes are used;
d. If no flow straightening vanes are used, a straight horizontal pipe without any flow
disturbances and with a minimum length of 30 D, should be used upstream from
the propeller;
e. The flow velocity in the pipe should be above 0.45 m/s for best performance. In
sediment-laden water the velocity should be even higher to minimize the added
friction effect produced by worn bearings.

9.8 Selected list of references

Cone, V.M. 1917. Divisors. The Agric. Exp. Sta. Fort Collins, Col., Bulletin 228. 52 pp.
Delleur, J.W., J.C.I. Dooge, and K.W. Gent 1956. Influence of slope and roughness on the free overfall.
Proc. of the American Society of Civil Engineers. Vol. 82. No. HY 4, Paper No. 1038.
Dethridge, J.S. 1913. An Australian water meter for irrigation supplies. Engineering News, New York
7. No. 26. p. 1283.
Diskin, M.H. 1961. End depth at a drop in trapezodal channels. Proc. of the American Society of Civil
Engineers, Vol. 87. No. H Y 4, Paper No. 2851.
Greeve, F.W. 1928. Measurement of pipe flow by the coordinate method. Purdue Engineering Exp. Sta.,
Bulletin 32.
ISO. 1967. Second Draft proposal for an I S 0 Recommendation on liquid flow measurement in open chan-
nels by weirs and flumes. Brink depth method for measurement of flow at a free overfall. ISO/TC 113/WG
2 (secretariat 26) 95.
Kraijenhoff van de Leur, D.A., and Dommerholt, A. 1972. Brink depth method for flow measurements.
Supplementary data from The Netherlands to ISO/TC 1 13/WG2 (secretariat-26)95.
Lawrence, F.E., and P.L. Braunworth 1906. Fountain flow of water in vertical pipes. American Society
of Civil Engineers. Transactions Vol .57, pp. 256-306.
Leer, B.R. van 1922. The California pipe method of water measurement. Engineering News-Record. 1924.
Meacham, I. 1956. Water measurement in Victorian irrigation districts. The Dethridge meter. Ann. Bulletin,
International Commission on Irrigation and Drainage, New Delhi. pp. 14-16.
Murky, K. 1966. Dethridge meter investigations. Aqua. Vo1.17. No.!?, pp. 202-211. Off. J. State Rivers
and Water Supply Commission. Victoria, Australia.
Neogy, B.N. 1972. Brink depth for trapezodal broad-crested weir. Proc. of the Am. Society of Civil Engin-
eers. December 1972, No.HY 12, Paper No. 9427.
Neyrpic 1955. Irrigation canal equipment. Neyrpic, Grenoble. 32 pp.
Rohwer, C. 1943. Discharge of pipes flowing partly full. Civil Engineering. ASCE.
Rouse, H. 1936. Discharge Characteristics of the free overfall. Civil Engineering. Vol. 6, No. 4.
Schuster, J.C. 1970. Water measurement procedures. Irrigation operators workshop. Div. of general re-
search, Engineering and Research Centre, Bureau of Reclamation, Denver Col. Report REC-OCE-70-38.
49 PP.
Scott, V.E., and C.E. Houston 1959. Measuring irrigation water. Circular 473. California Agricultural
Exp. Sta., Univ. of California.
Smith, C.D. 1962. Brink Depth foracircularchannel. Proc. Am. Soc. Civ. Engnrs. No. HY6, Paper No.3327.
United States Bureau of Reclamation 1967. Water Measurement Manual. Denver, Colorado. 327 pp.

344
1 Annex 1
Basic equations of motion in fluid mechanics

1.1 Introduction
It is assumed that the reader of this book is familiar with the basic laws of fluid mecha-
nics. Nevertheless some of these laws will be discussed in this annex to summarise
material and to emphasize certain subjects which are important in the context of dis-
charge measurement structures in open channels.

1.2 Equation of motion- Euler

In fluid mechanics we consider the motion of a fluid under the influence of forces
acting upon it. Since these forces produce an unsteady motion, their study is essentially
one of dynamics and must be based on Newtons second law of motion
F = ma (Al.l)
where F is the force required to accelerate a certain mass (m) at a certain rate (a).
If we consider the motion of an elementary fluid particle (dx dy dz) with a constant
mass-density (p), its mass (m) equals
m = pdxdydz (Al .2)
The following forces may act on this particle:
a. The normal pressures (P) exerted on the lateral faces of the elementary volume
by the bordering fluid particles;
b. The mass forces, which include in the first place the gravitational force and in the
second the power of attraction of the moon and the sun and the Coriolis force.
These forces, acting on the mass (p dx dy dz) of the fluid particle, are represented
together by their components in the X-, Y - ,and Z-direction. It is common practice
to express these components per unit of mass, and therefore as accelerations; for
example, the gravitational force is expressed as the downward acceleration g;
c. Friction. There are forces in a fluid which, due to friction, act as shear forces o n
the lateral faces of the elementary particle (dx dy dz). To prevent complications
unnecessary in this context, the shear force is regarded as a mass force.
Gravitation and friction are the only mass forces we shall consider. If the fluid
is in motion, these two forces acting on the particle (dx dy dz) do not have to
be in equilibrium, but may result in an accelerating or decelerating force (pos. or
neg.). This net force is named:
d. Net impressed force. This force equals the product of the mass of the particle and
the acceleration due to the forces of pressure and mass not being in equilibrium.
The net impressed force may be resolved in the X-, Y - ,and Z-direction.
If we assume that the pressure at a point is the same in all directions even when the
fluid is in motion, and that the change of pressure intensity from point to point is

345
continuous over the elementary lengths dx, dy, and dz, we may define the normal
pressures acting, at time t, on the elementary particle as indicated in Figure A I . I .
Acting on the left-hand lateral face (X-direction) is a force

ap
-k (P-'!,xdx 1 dydz

while on the right-hand face is a force

-'(P + ax
ap
dx) dy dz
'/2

The resulting normal pressure on the elementary fluid particle in the X-direction equals

(A 1.3)

The resultant of the combined mass forces in the X-direction equals

p dx dydz k,
where k, is the acceleration due to gravitation and friction in the X-direction. Hence
in the X-direction, normal pressure and the combined mass forces on the elementary
particle result in a total force
ap
F, = --dXdydz
ax + k,pdxdydz (A 1.4)

1 b P
P'T --dY
bY
I
Figure A l . I Pressure distribution on an elementary fluid particle

346
*
p--
1 LP
-dZ
bz
 t+dt)

Figure A 1.2 The velocity as a function of time and position

Similarly, for the forces acting on the mass (p dx dy dz) in the Y - and Z-direction,
we may write

(A1.5)

and

(A 1.6)

The reader should note that in the above equations k,, k,, and k, have the dimension
of an acceleration.
In a moving liquid the velocity varies with both position and time (Figure AI .2).
Hence:
v = f(x, Y, 2, t) (A 1.7)
and as such
vx = fx(x, Y, z, t)
vy = f,(x, Y, 2, t)
and
vz = fz (x, Y, z, t>
If we consider the X-direction first, we may write that at the time (t + dt) and at
the point (x +
dx, y dy, z + +
dz) there is a velocity component in the X-direction
which equals v, +
dv,.

347
The total differential of v, is equal to

dv, av,
= -dt + Lav av av
d x + L d y + --"-dz (A 1.8)
at ax aY aZ
In Figure A1.3 we follow a moving fluid particle over a time dt, and see it moving
along a pathline from point (x, y, z) towards point (x + dx, y + dy, z + dz) where
it arrives with another velocity component (v, + dv,). The acceleration of the fluid
particle in the X-direction consequently equals

a = -dv, (A 1.9)
dt
while the elementary variations in time and space equal
dx = v,dt (A1.lO)
dy = v,dt (Al.l I )
dz = V, dt (A 1 . I 2)
Equation A1.8, which is valid for a general flow pattern, also applies to a moving
fluid particle as shown in Figure A1.3, so that Equations A1.10 to A1.12 may be
substituted into Equation A1.8, giving

dv, av,
= -dt+
at
av,
-V,dt
ax
+ Lav
aY
av
v,dt +Lv,dt
aZ
(A1.13)

and after substitution of Equation AI .9

(A 1 . I 4)

Figure A I .3 The flow path of a fluid particle

348
and similarly

(A I . 15)

(A1.16)

Substitution of Equations A 1.2, A 1.4, and A 1.14 into Equation A I . 1 gives

+ k,pdxdydz + av + >v,
-Edxdydz
ax
= pdxdydz 2 ~ y

aY av
aZ 1
or

(A1.17)

In the same manner we find for the Y- and Z-direction

av
+ -av + Yav av = I ap + k
$ $vx
aYv y + Yv,
az Pay y
(A 1.1 8)

1 aP
av,
at
av,
+ -vx
ax
av,
+ -v
ay y
av,
+ -v,
aZ =
P
+ k, (A 1 . I 9)

These are the Euler equations of motion, which have been derived for the general
case of unsteady non-uniform flow and for an arbitrary Cartesian coordinate system.
An important simplification of these equations may be obtained by selecting a coordi-
nate system whose origin coincides with the observed moving fluid particle (point
P). The directions of the three axes are chosen as follows:
- s-direction: the direction of the velocity vector at point P, at time t. As defined,
this vector coincides with the tangent to the streamline at P at time t (vs = v).
- n-direction: the principal normal direction towards the centre of curvature of the
streamline at point P at time t. As defined, both the s- and n-direction lie in the
osculating plane.
- m-direction: the binormal direction perpendicular to the osculating plane at P at
time t (see also Chapter 1).
If we assume that a fluid particle is passing through point P at time t with a velocity
v, the Eulerian equations of motion can be written as:
av,
- avs av av
+ -v, I ap
at +-vs
as +-v,
an am
=
P as
+ ks ( A l .20)

avn +-vsavn
-
at as
av
+-1v,
an
av
+-!v,
am
=---'
P an
+ k, (A1.21)

av, av,
av,
at + -vS
av,
as
+ -v,
an
+ -v,
am
=
I ap
P am
+ k, (A 1.22)

Due to the selection of the coordinate system, there is no velocity perpendicular to

the s-direction; thus
v, = O and v, = O (A 1.23)

349
Therefore the equations of motion may be simplified to

(A 1.24)

(Al .25)

( A I .26)

Since the streamline at both sides of P is situated over an elementary length in the
osculating plane, the variation of v, in the s-direction equals zero. Hence, in Equation
A1.26

(A 1.27)

In Figure A I .4 an elementary section of the streamline at point P at time t is shown

in the osculating plane. It can be seen that

% ds
as ds
-- _ (A 1.28)
tandp =
av r
v, + A d s
as
or

( A I .29)

av
In the latter equation, however, S d s is infinitely small compared with v,; thus we
may rewrite Equation A l .29 as

- n -3
a v- (A 1.30)
as r
or
av, v2 ( A l .31)
= 2-
r
Substitution of Equations A 1.27 and A 1.3 1 into Equation A I .26 and A 1.25 respecti-
vely gives Eders equations of motion as follows
av, +aL
v V= - - -i +ap
k (A 1.32)
at as pas .

av,
x++
v2
-
- P as
+ k, ( A I .33)

350
 M o. & A d s /
elementary xciion (1-2)
of a streamline a t time i
in the osculating plane
(osculating plane)
--__
J---
\
\
. ,/A
I

vs + r ds

Figure A I .4Elementary section of a streamline

av, (Al .34)

at
These equations of motion are valid for both unsteady and non-uniform flow. Here-
after, we shall confine our attention to steady flow, in which case all terms 8. ./at
equal zero.
Equations A 1.32, A 1.33, and AI .34 are of little use in direct applications, and they
tend to repel engineers by the presence of partial derivative signs; however, they help
ones understanding of certain basic equations, which will be dealt with below.

1.3 Equation of motion in the s-direction

If we follow a streamline (in the s-direction)?we may write v, = v, and the partial
derivatives can be replaced by normal derivatives because s is the only dependent vari-
able. (Thus a changes into d). Accordingly, Equation Al .32 reads for steady flow
dv 1 dP
=---+ k
-V (Al .35)
ds p ds
where k, is the acceleration due to gravity and friction. We now define the negative
Z-direction as the direction of gravity, The weight of the fluid particle is - p g ds dn
dm of which the component in the s-direction is
dz
- p g ds dn dm-
ds
and per unit of mass
dz
-p g ds dn dm -
ds - dz
- - g z (Al .36)
p ds dn dm

351
 I-dtrectlon

S-direction

--;--I
d

2
/ W (due t o friction)

pgdsdndm
d zs

\
\
\
\

\
\
\ pgdsdndm

V
Figure A1.5 Forces due to gravitation and friction acting on an elementary fluid particle

The force due to friction acting on the fluid particle in the negative s-direction equals
per unit of mass

-w =
-W (Al .37)
p ds dn dm
;The acceleration due to the combined mass-forces (k,) acting in the s-direction accor-
dingly equals
k = - w - g - dz (A1.38)
ds
Substitution of this equation into Equation A1.35 gives
dv
-v = 1 dP dz (Al .39)
g--w
ds p ds ds
or
dv dP dz
pv-+
ds
-ds+ pg- = - p w
ds
(A 1.40)

or
d (I/* p v2 + P + pgz) = -p w ds (A1.41)
The latter equation indicates the dissipation of energy per unit of volume due to local

352
friction. If, however, the decelerating effect of friction is neglected, Equation Al .41
becomes
d
-(y2
ds
pv2 + P + pgz) = o (A 1.42)

Hence
'/2 pv2 + P + p g z = constant (A 1.43)
where
p v2 = kinetic energy per unit of volume
p g z = potential energy per unit of volume
P = pressure energy per unit of volume
If Equation A1.43 is divided by pg, an equation in terms of head is obtained, which
reads
v2
2g
- + PgP + z = constant = H
- (A 1.44)

where
v2/2g = the velocity head
P/pg = the pressure head
Z = the elevation head
P/pg+z = the piezometric head
H = the total energy head
The last three heads all refer to the same reference level (see Figure 1.3, Chapter I).
The Equations A1.43 and A1.44 are alternative forms of the well-known Bernoulli
equation, and are valid only if we consider the movement of an elementary fluid parti-
cle along a streamline under steady flow conditions (pathline) with the mass-density
(p) constant, and that energy losses can be neglected.

1.4 Piezometric gradient in the n-direction

The equation of motion in the n-direction reads for steady flow (see Equation Al .33)
v2 - 1 dP
-
r pdn + kn (Al .45)

Above, the a has been replaced by d since n is the only independent variable. The
term v2/r equals the force per unit of mass acting on a fluid particle which follows
a curved path with radius! at a velocity! (centripetal acceleration). In Equation A 1.45,
k, is the acceleration due to gravity and friction in the n-direction. Since v, = O, there
is no friction component. Analogous to its component in the direction of flow here
the component due to gravitation can be shown to be

k =-g- dz (AI .46)

dn

353
Substitution into Equation A1.45 yields
v2 - 1 dP
-
dz
r p d n gdn (A 1.47)

which, after division by g, may be written as

(Al .48)

After integration of this equation from point 1 to point 2 in the n-direction we obtain
the following equation for the change of piezometric head in the n-direction
(
+ i(
+i J z), - z ) ~= -1 Tdn
g ,
v2
(Al .49)

where (P/pg + z) equals the piezometric head at point 1 and 2 respectively and
1 2v*
-j-dn
g, r
is the loss of piezometric head due to curvature of the streamlines.

1.5 Hydrostatic pressure distribution in the m-direction

Perpendicular to the osculating plane, the equation of motion, according to Euler,
reads for steady flow

(A1.50)

I
._O
U
0
.-L
?

Figure A1.6 The principal normal direction

354
4Since there is no velocity component perpendicular to the osculating plane (v, = O),
there is no friction either. The component of the acceleration due to gravity in the
m-direction is obtained as before, so that

(A1.51)

Substitution of this acceleration in the equation of motion (Equation A 1.50) gives

1 dP dz
g-=o (A1.52)
pdm dm
which may be written as

-&(5+ z) = o (A1.53)

It follows from this equation that the piezometric head in the m-direction is
P
-
Pg
+ z = constant (A1.54)

irrespective of the curvature of the streamlines. In other words, perpendicular to the

osculating plane, there is a hydrostatic pressure distribution.

355
Annex 2
The overall accuracy of the measurement of flow
2.1 General principles
Whenever a flow rate or discharge is measured, the value obtained is simply the best
estimate of the true flow rate which can be obtained from the data collected; the true
flow rate may be slightly greater or less than this value. This annex describes the calcu-
lations required to arrive at a statistical estimate of the range which is expected to
cover the true flow rate.
The usefulness of the flow rate measurement is greatly enhanced if a statement of
possible error accompanies the result. The error may be defined as the difference be-
tween the true flow rate and the flow rate which is calculated from the measured water
level (upstream head) with the aid of the appropriate head-discharge equations.
It is not relevant to give an absolute upper bound to the value of error. Due to
chance, such bounds can be exceeded. Taking this into account, it is better to give
a range which is expected to cover the true value of the measured quantity with a
high degree of probability. This range is termed the uncertainty of measurement, and
the confidence level associated with it indicates the probability that the range quoted
will include the true value of the quantity being measured. In this annex a probability
of 95% is adopted as the confidence level for all errors.

2.2 Nature of errors

Basically there are three types of error which must be considered (see Figure A2.1):
a. Spurious errors (human mistakes and instrument malfunctions);
b. Random errors (experimental and reading errors);
c. Systematic errors (which may be either constant or variable).

Spurious errors are errors which invalidate a measurement. Such errors cannot be
incorporated into a statistical analysis with the object of estimating the overall accu-
racy of a measurement and the measurement must be discarded. Steps should be taken
to avoid such errors or to recognize them and discard the results. Alternatively, correc-
tions may be applied.
Random errors are errors that affect the reproducibility of measurement. It is as-
sumed that data points deviate from the mean in accordance with the laws of chance
as a result of random errors. The mean random error of a summarized discharge over
a period is expected to decrease when the number of discharge measurements during
the period increases. As a result, the integrated flow over a long period of observation

Note: Sections 1 and 2 of this annex are based on a draft proposal of an I S 0 standard prepared by Kinghorn,
1975.

356
mean measured
v a l u e o f quan-l
t i t y t o be ; -ed
termined
I
I
.L
r.7
i.
tandomf.
* *
apparent
i 1l e g iti-
mate e r r o r

systematic
I error assessed w i t h spe-
t r u e value of1 1
w c i f i c confidence
quantity t o I
be determined: level
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
f r

It1
I t i m e
I
4
p e r i o d d u r i n g which
a s i n g l e value o f
f l o w r a t e i s measured
Figure A2.1 Illustration of terms

will have a mean random error that approaches zero. It is emphasized that this refers
to time-dependent errors only, and that the length of time over which observations
should be made has to be several times the period of fluctuations of flow.
Systematic errors are errors which cannot be reduced by increasing the number
of measurements so long as equipment and conditions remain unchanged. Whenever
there is evidence of a systematic error of a known sign, the mean error should be
added to (or subtracted from) the measurement results. A residual systematic error
should be assessed as half the range of possible variation that is due to this systematic
error.
A strict separation of random and systematic errors has to be made because of
their different sources and the different influence they have on the total error. This
influence will depend on whether the error in a single measurement is concerned, or
that in the sum of a series of measurements.

2.3 Sources of errors

For discharge measurement structures, the sources of error may be identified by consi-
dering a generalized form of head-discharge equation:
Q = wC,C,f&bh, (A2.1)
where w and u are numerical constants which are not subject to error. The acceleration

357
due to gravity, g, varies from place to place, but the variation is small enough to be
neglected in flow measurement. So the following errors remain to be considered:
6C = error in product Cd C,
6f = error in drowned flow _reductionfactor f
6b = error in dimensional measurement of weir; e.g. the width of the weir b,
or the weir notch angle 8
6h = error in h, and/or Ah
The error 6C of each of the standard structures described in Chapters 4 to 9 is given
in the relevant sections on evaluation of discharge. These errors are considered to
be constant and systematic. This classification is not entirely correct because Cd and
C, are functions of h,. However, the variations of the errors in Cd and C, as a function
of h, usually are sufficiently small to be neglected.
When flow is modular, the drowned flow reduction factor f i s constant (f = 1.0)
and is not subject to error. As a result, for modular flow 6f = O. When flow is non-
modular the error 6f consists of a systematic error, Sf,, being the error in the numerical
value off, and of systematic and random errors caused by the fact that f i s a function
of the submergence ratio Sh = H2/HIN h,/h,.
The error 6b depends on the accuracy with which the structure as constructed can
be measured, and is also a systematic error. In practice this error may prove to be
insignificant in comparison with other errors.
The error Fh, has to be split into a random error 6h, and a systematic error 6h,.
Those errors may contain many contributory errors. Possible sources of contributory
errors are:
1. Internal friction of the recording system;
2. Inertia of the indication mechanism;
3. Instrument errors;
4. Zero setting;
5 . Settling or tilting sideways of the structure with time;
6. The crest not being level, or other construction faults not included in 6b;
7. Improper maintenance of the structure (this also may cause an extra error 6C);
8. Reading errors.
We have to be careful in recognizing whether an error is random or systematic. Some
sources can cause either systematic or random errors, depending on circumstances.
Internal friction of the recorder, for example, causes a systematic error of a single
measurement or a number of measurements in a period when either rising or falling
stage is being considered, but a random error if the total discharge through an irriga-
tion canal per season is being considered. On natural streams, however, falling stage
may occur over a much longer period than rising stage and here the internal friction
of the recorder once again results in a systematic error. Also zero setting may cause
either a systematic or a random error. If a single measurement or measurements within
the period between two zero settings are considered, the error will be systematic; it
will be random if one is considering the total discharge over a period which is long
in comparison with the interval between zero settings. The errors due to (3), (9,and
(6) are considered to be systematic, that due to (8) being random.
In the following sections the term relative error will frequently be found. By this
we mean the error in a quantity divided by this quantity. For example, the relative
error in h, equals xh] = 6h,/h,.
358
 2.4 Propagation of errors
The overall error in the flow Q is the resultant of various contributory errors, which
themselves can be composite errors. The propagation of errors is to be based upon
the standard deviation of the errors. The standard deviation o out of a set of measure-
ments on Y may be estimated by the equation
n

2 -
c (Y,-Y)Z
1=1
( 3 - (A2.2)
n- 1
where
-
Y = the arithmetic mean of the n-measurements of the variable Y
Y, = the value obtained by the ithmeasurement of the variable Y
n = the total number of measurements of Y
The relative standard deviation o equals o divided by the observed mean. Hence

0= Y1
- [
, i 1n-1
(Y1-P] (A2.3)

The relative standard deviation of the mean o; of n-measurements is given by

o
(3; = ~
(A2.4)
fi
If Equations A2.2 to A2.4 cannot be used to estimate the relative standard deviation,
it may be estimated by using the relative error of the mean for a 95% confidence level,
X,. The value of X, is either given (X,), or must be estimated.

To estimate oit is necessary to know the distribution of the various errors. In this
context we distinguish three types of distribution (see Figure A2.2).
- normal distribution: For practical purposes it is assumed that the distribution of
the errors in a set of measurements under steady conditions can be sufficiently close-
ly approximated by a normal distribution. If ois based on a large number of obser-
vations, the error of the mean for a 95% confidence level equals approximately two
times o (o= 0.5 X). This factor of two assumes that n is large. For n = 6 the
factor should be 2.6; n = 10 requires 2.3 and n = 15 requires 2.1;
- uniform distribution: For errors X having their extreme values at either +X,,, or
-X,,, with an equal probability for every error size in this range, o equals
0.58 X,,, (0 = 0.58 X,,,);
- point binomial distribution: For errors X which always have an extreme value of
either +X,,, or -X,,,, with an equal probability for each of these values, o equals
1 .o x,,, ((3 = X,,,).

To determine the magnitude of composite errors the standard deviation has to be

used. The composite standard deviation can be calculated with the following equation

I 359
 95V. confidence band

I. 2 I x 1
STANDARD NORMAL DISTRIBUTION

p,50x-.,I
UNlFO R M DI ST R I BUTION

-',ax
I- G
I
I
I
G
L
+ ~ m x

POINT BINOMIAL DISTRIBUTION

Figure A2.2 Possible variation of measured values about the average (actual) value

,o;=
in which
J C Gi di
i=l
(A2.5)

(A2.6)

where
o; = relative standard deviation of the composite factor T;
or = relative standard deviation of the factor Fi;
Fi = relevant factor influencing Q; the error of this factor is uncorrelated with

360
 the errors in other contributory factors of Equations A2.5 and A2.6; Fi may
itself be a composite factor.
It is emphasized that only factors with uncorrelated errors can be introduced in Equa-
tion A2.5. This means that it is incorrect to determine oh by substituting o:, o;,ob,
and oh into Equation A2.5 because the errors in f and h, are correlated. One must
start from relevant (= contributing to 6C, 6b, 6f and 6h,) errors or standard deviations
which are mutually independent. For weirs and flumes, those independent errors are
generally SC, 6b, Sf,,*, 6h, (containing 6hIRand 6h,,) and 6Hz (containing 6H2, and
6HZs).The,first three errors are systematic errors. The last two errors are often compo-
site errors themselves, and their magnitude has to be determined with the use of Equa-
tions A2.5 and A2.6. Substitution into Equation A2.6 of the independent factors con-
tributing to the overall error in Q and their relative standard deviations yields the
first two terms of the following equations.

* df,, is the error in the numerical value o f f and has no relation to 6hl. Systematic and random errors
in f caused by its relation to h, and H2 are not independent and cannot be substituted into Equation A2.5.

The right-hand side of these equations is found by partial differentiation of Equation

A2.1 to-C, b, f,,, h, and H, respectively. In doing so we have to take into account
that fis a function of Sh 1: H,/h,. Putting

(A2.7)

and substituting the above information into Equation A2.5 gives

06 = [O: + ob + o;: + (u-G)~o,,+ G2 (~;122]/~ (A2.8)
As has been mentioned in the section on sources of error, we have to distinguish be-
tween systematic and random errors because of their different influences on the accur-
acy of measured volumes over long periods. Using the given information on the char-
acter of various errors, we can divide Equation A2.8 into two equations; one for ran-
dom errors and the other for systematic errors, as follows:
0 6 =~ [(U-G) ohlR2 + G2 oh2RZ] I2 (A2.9)

361
 06s = ';.[ + ob2+ CT;: + (u-G)~oh,? + G2 (~;1~?] '12 (A2.10)

For most discharge measuring structures, the error Sf, is unknown. We know, how-
ever, that i f f does not deviate much from unity (near modular flow), the error Sf,
is negligible. For low values o f f (f < appr. 0 . Q the error in the numerical value
o f f , Sf,, becomes large, but then the absolute value of G becomes so large that the
structure ceases to be an accurate measuring device. As mentioned, Sf, is usually un-
known and therefore it is often assumed that Sf, N O and thus also o;,E O.
To determine G we need a relationship between the drowned flow reduction factor
and the submergence ratio. If we have, for example, a triangular broad-cTested weir
operating at a submergence ratio H,/H, = 0.925, we can determine G (being a measure
for the 'slope' of the S,-f-curve) from Figure 4.1 1 as

G = - Af/f -- (0.775 - 0.825)/0.80 - -4.

AshlSi, (0.932 - 0.9 18y0.925 -
It should be noted that G always has a negative value.
From Equations A2.9 and A2.10, it may be noted that 06 increases sharply if I GI
increases, i.e. if the slope of the H2/H,-f-curve in Figure 4.1 1 becomes flat. If flow
is modular, the drowned flow reduction factor fis constant and is not subject to error
(f = 1.O). Thus, o;,, = O and G = O , and as a consequence Equations A2.9 and A2.1 O
reduce to
O6R = OhlR (A2.11)
and
obs= [0:2 + ob2 + u2 O;lS2] '12 (A2.12)
It is noted again that Equations A2.11 and A2.12 are only valid if flow is modular.
It can be proved that the combination of a sufficiently large number of errors not
having a normal distribution tends to a composite error having a normal distribution.
So we may assume that the overall error of the flow rate measurement has a normal
distribution even if the overall error is the result of the combination of a few errors
not having a normal distribution. Thus, the overall relative error of the flow rate for
a single discharge measurement approximates
XQ = 2 [ob: + 06:]1/2 (A2.13)
It should be realized that the relative error XQis not a single value for a given device,
but will vary with discharge. It is therefore necessary to consider the error at several
discharges covering the required range of measurement. In error analysis, estimates
of certain errors (or standard deviations) will often be used. There is a general tendency
to underestimate errors. In some cases they may even be overlooked.

2.5 Errors in measurements of head

When errors are quoted, the reader should be aware that the general tendency is for
them to be underestimated. He should also realize that errors having a 95 per cent
confidence level must be estimated by the user.

362
Chapter 2.2 indicates that the head measurement station should be located sufficiently
far upstream of the structure to avoid the area of surface drawdown, yet it should
be close enough for the energy loss between the head measurement station and struc-
ture to be negligible. For each of the standard structures described in Chapters 4 to
9, the location of the head measurement station has been prescribed. In practice, how-
ever, it very often happens that this station is located incorrectly, resulting in very
serious errors in head.
Insufficient depth of the foundation of the structure or the head measurement de-
vice, or both, can cause errors in the zero-setting since ground-frost and changes in
soil-moisture may move the structure and device. To limit errors in zero setting it
is recommended that the setting be repeated at least twice a year; for example, after
a period of frost, after a rainy season, or during summer or a dry season. The reading
error of a staff-gauge is strongly influenced by the reading angle and the distance
between the gauge and the observer, the turbulence of the water, and the graduation
unit of the gauge.
For example, a staff gauge with centimeter graduation placed in standing water
can be read with a negligible systematic error and a random reading error of
0.003 m. If the same gauge is placed in an approach channel with a smooth water
surface, the gauge becomes more difficult to read and a systematic reading error of
0.005 m and a random reading error of 0.005 m may be expected. Little research has
been done on this subject, although Robertson (1966) reports on the reading error
of a gauge with graduation in feet and tenths of a foot located in reasonably still
water in a river. He recorded a systematic reading error of 0.007 m and a random
reading error of 0.007 m. The graduation unit of the reported gauge equaled 0.03
m. If the water surface is not smooth or the position of the observer is not optimal,
or both, reading errors exceeding one or more graduation units must be expected.
It is obvious that a dirty gauge face hinders readings and will cause serious reading
errors. Staff gauges should therefore be installed in locations where it is possible for
the observer to clean them.
Since reading a gauge in standing water causes a smaller reading error than one
read in streaming water, the use of a stilling well must be considered whenever the
accuracy of head readings has to be improved. The stilling well should be designed
according to the instructions given in Chapter 2.6.
When a float-operated automatic water level recorder is used great care should be
given to the selection of the cable, although it is recommended that a calibrated float
tape be used instead. The cable or tape should not stretch and should be made of
corrosion-resistant material.
Several errors are introduced when a float-operated recorder is used in combination
with a stilling well. These are:
- Lag error due to imperfections in the stilling well. This error, caused by head losses

in the pipe connecting the stilling well with the approach channel during rising or
falling discharges or head losses caused by a leaking stilling well, has also been
considered in Chapter 2.6;
- Instrument errors, due to imperfections in the recorder. This error depends on con-

tributory errors due to internal friction of the recorder, faulty zero setting, and back-
lash in the mechanism, etc.
The magnitude of internal friction should be given by the manufacturer of the recorder.

363
The reader should realize, however, that manufacturers are sometimes rather optimis-
tic and that their data are valid for factory-new recorders only. Regular maintenance
will be required to minimize internal friction. The errors due to internal friction and
those caused by a change in cable weight hanging on one side of the float wheel or
submergence of the counter weight are considered in Chapter 2.9. The magnitude of
all these errors is inversely proportional to the square of the float diameter (d*). To
give an idea of the order of magnitude of errors that may occur in automatic recorders
we cite three examples:
- Stevens (1919) reports on a recorder equipped with a 00.25 m float, a steel cable,
and a 4 kg counter weight. The following errors were observed: Error due to submer-
gence of counterweight 0.0015 m. Difference in readings between falling and rising
stage due to internal friction 0.002 m. An increasing total weight of cable plus
counter weight hanging on one side of the cable wheel caused a registration error
of 0.06%;
- Robertson (1966) reports on the reading error of recorder charts. When a writing
mechanism with 1:1 reduction (full scale) was used, the systematic reading error
was negligible and the random error was 0.010 m. For a writing mechanism with
1O:l reduction, however, a systematic error of 0.010 m and a random error of
0.016 m was reported. No float diameter was mentioned;
- Agricultural University (1966) at Wageningen reports on laboratory tests conducted
under ideal conditions with a digital recorder giving a signal for a 0.003 m head inter-
val. Equipped with a @ 0.20 m float the digital reading showed a negligible systematic
error and a random error of 0.002 m. In addition, a difference of 0.002 m was found
between readings for falling and rising stage. The errors found in the Wageningen
tests must be regarded an absolute minimum.
It should be noted that if waves are dampened in the approach channel by means
of a stilling well a systematic error may be introduced. This is a result of the non-linear
relationship between the head and the discharge.

2.6 Coefficient errors

The coefficient errors presented in Chapters 4 to 9 are valid for well-maintained clean
structures. To obtain the accuracies listed, sediment, debris, and algal growth must
be removed regularly. To keep the structure free of weed, fungicides can be used.
The best method is probably to add, say, 0.5 per cent by weight of cement copper
oxide to facing concrete during mixing. Copper sulphate or another appropriate fungi-
cide can be applied to existing concrete but frequent treatment will be required. Algal
growth on non-concrete structural parts can be prevented by regular treatment with
an anti-fouling paint such as that used on yachts.
It must be realized that algal growth on broad-crested weirs and flumes increases
friction and raises the crest. Consequently algal growth has a negative influence on
Cd-values. On sharp-crested weirs or sharp-edged orifices, algal growth reduces the
velocity component along the weir face, causing an increase of Cd-values.
Nagler (1929) investigated this type of influence on a sharp-crested weir whose up-
stream weir face was roughened with coarse sand. He found that, compared with the
coefficient value of a smooth-faced weir the discharge coefficient increased by as much

364
as 5 per cent when h, = 0.15 m and by 7 per cent when h, = 0.06 m. Algal growth
on the upstream face of sharp-crested weirs may cause a 'rounding-off' of the edge
which, in addition to reducing the velocity component along the weir face, causes
a decrease of contraction and consequently results in an increase of the discharge coef-
ficient. For a head of 0.15 m, Thomas (1957) reported an increase of some 2, 3, 5.5,
11, and 13.5 per cent due solely to the effect of rounding-off by radii of a mere 1,
3,6,12, and 19 mm respectively. Another factor that will cause the discharge coefficient
to increase is insufficient aeration of the air pocket beneath the overfalling nappe of
a sharp- or short-crested weir (see also Chapter 1.14).

2.7 Example of error combination

In this example all errors mentioned are expected to have a 95 per cent confidence
level. We shall consider a triangular broad-crested weir as described in Chapter 4.3,
flowing less-than-full, with a vertical back face, a crest length L = 0.60 m, a weir
notch angle 0 = 120, and a crest height pi = 0.30 m. According to Chapter 4.3,
the following head-discharge equation applies

(A2.14)

Both upstream and downstream heads were measured by identical digital recorders
giving a signal for every 0.003 m head difference (thus maximum reading error is
0.0015 m). The random error due to internal friction of the recorder was 0.002 m.
The systematic error in zero setting was estimated to be 0.002 m due to internal friction
of the recorder and 0.001 m due to the procedure used. The latter error is due to
the difficulty of determining the exact elevation of the crest.
In addition to these errors, it was found that over the period between two successive
zero settings the stilling well plus recorder had subsided 0.005 m more than the structure.
To correct for this subsidence, all relevant head readings were increased by 0.0025 m,
leaving a systematic error of 0.0025 m. The frequency distribution of the error due
to subsidence is unknown, but is likely to be more irregular than a normal distribution.
If subsidence occurs over a period which is short compared with the interval between
two zero settings,the ratio o:/X, approaches unity. In our example we assume o;/Xi
to equal 0.75.
The error in the discharge coefficient (including C,) is given by the equation
X, = f (3 I H,/L - 0.55 I + 4 per cent (A2.15)
The overall error in a single discharge measurement for three different states of flow
has been calculated in Table A2.1. From this example it appears that even if accurate
head registration equipment is used, the accuracy of a single measurement at low heads
and at small differential heads Hl - H, is low. For an arbitrary hydrograph, the random
error in the total discharge over a long period equals zero. If, however, the hydrograph
shows a considerably shorter period of rising stage than of falling stage, as in most
streams and sometimes in irrigation canals, the internal friction of an automatic re-
corder (if used) causes a systematic error which cannot be neglected.
The factor that has the greatest influence on the accuracy of discharge measurements

365
is the accuracy with which the head h, or the differential head Ah can be measured.
This warrants a careful choice of the equipment used to make such head measurements.
This holds especially true for structures where the discharge is a function of the head
differential, h, - h,, across the structure, as it is for instance for submerged orifices.
If h, and h, are measured independently by two separate gauging systems, the errors
of both measurements have to be combined by using Equation A2.5. In doing so,
the errors have to be expressed as percentage errors of the differential head (h, - h,),
thus not of h, and h, separately. If a differential head meter as described in Chapter
2.12 is used to measure (h, - h,), errors due to zero-settings and in some cases due
to reading of one head are avoided, thereby providing more accurate measurements.

Table A2.1 Examples of accuracy computation

State of flow

Source of error Type of Ratio h, = 0.06111 h, = 0.40m h, = 0.40m

error oi/Xi H2 = nil H, = 0.30111 H, = 0.37m
Cd 0.920 Cd = 0.996 Cd = 0.996
f = 1.0 f = 1.0 f Y 0.80

cd cv S 0.50 O; = 2.6% O', = 1.1% O', = 1.1%

procedure of zero
setting S 0.50 o;, = 0.8% O&, = 0.1% O;,( N 4,)= 0.1%
internal friction-zero
setting S 1.o o;, = 3.3% oh, = 0.5% O i l (Y Oh2) = 0.5%

internal friction R 1.0 CT;, = 3.3% = 0.5% O i l (Y Oh,) = 0.5%

subsidence S 0.75 O;, 3.1% O&, = 0.45% O;,( N O;,) = 0.45%

digital reading R 0.58 O;, 2 1.5% = 0.23% O;,( N O&) = 0.23%

crest level S 0.50 O;, = 0.8% O;, = 0.13% (N 4 2 ) = 0.13%

Calculated Equation
value used

OhlR A2.5 o~,R= 3.6% o L I R = 0.55% o;,R Y oI;*R= 0.55%

oh,s A2.5 o&,R = 4.7% O;,R = 0.70% O;,R N O;,R = 0.70%

G A2.7 G = -4.1

ObR A2.9 or 11 o ~ R= 9% o ~ R = 1.40% ObR Y 3.9%

OQS A2.10 or 12 Obs = 12% Obs = 2.05% 06s > 5.6%*

XQ A2.13 XQ = 30% XQ = 4.95% XQ > 13.6%*

* obs and XQ are greater than values shown because the systematic error of the f-value is unknown and
not included in this computation

366
2.8 Error in discharge volume over long period
If during a 'long' period a great number of single discharge measurements (n > 15)
are made and these measurements are used in combination with head readings, to
calculate the discharge volume over an irrigation season or hydrological year, the per-
centage random error Xvol.R tends to zero and can be neglected.
The systematic error Xvo,.sof a volume of water measured at a particular station
is a function of the systematic percentage error of the discharge (head) at which the
volume was measured. Since the systematic percentage error of a single measurement
decreases if the head increases, a volume measured over a long period of low discharges
will be less accurate than the same volume measured over a (shorter) period of higher
discharge. As a consequence we have to calculate Xvol.sas a weighted error by use
of the equation
JQObSdt
Xv0l.S = 2 ~ (A2.16)
JQdt
which may also be written as
k
C Qi06At
i= 1
X"0l.S= 2 (A2.17)
C Qi At
i= I

where At = period between two successive discharge measurements. By using Equa-

tions A2.16 and A2.17 the reader will note that the value of Xvol.s= Xvo,will be signifi-
cantly lower than the single value X, and will be reasonably small, provided that a
sufficient number of measurements are made over the period considered.

2.9 Selected list of references

Agricultural University, Wageningen 1966. Voortgezet onderzoek van registrerende waterstands meters.
Hydraulica Laboratorium. Nota No.4, 15 pp.
British Standard Institution 1969. Measurement of liquid flow in open channels. Part 4: Weirs and flumes.
4B: Long base weirs. BS 3680. London. BSI. 39 pp.
Kinghorn, F.C. 1975. Draft proposal for an IS0 standard on the calculation of the uncertainty of a measure-
ment of flowrate. Doc. No. ISO/TC 30/WG 14:24 E.
Nagler, F.A. 1929. Discussion of precise weir measurements. Transaction ASCE. Vol. 93. p. 115.
Robertson, A.I.G.S. 1966. The magnitude of probable errors in water level determination at a gauging
station. Water Resources Board. T N 7, Reading, England. Reprinted 1970.
Stevens, J.C. 1919. The accuracy of water-level recorders and indicators of the float type. Transactions
ASCE. Vol. 83.
Thomas, C.W. 1957. Common errors in measurement of irrigation water. Journal Irrigation and Drainage
Div. Proc. Am. Soc. ofCiv. Eng. Vol. 83, No. IR 2. Paper 1361, pp.1-24.

367
Annex 3
Side weirs and oblique weirs

3.1 Introduction
Most of the weirs described in this book serve mainly to measure discharges. Some,
however, such as those described in Chapters 4 and 6 can also be used to control
upstream water levels. To perform this dual function, the weirs have to be installed
according to the requirements given in the relevant chapters. Since these weirs are
usually relatively wide with respect to the upstream head, the accuracy of their flow
measurements is not very high. Sometimes the discharge measuring function of the
weir is entirely superseded by its water level control function, resulting in a contraven-
tion in their installation rules. The following weirs are typical examples of water level
control structures.
Side weir: This weir is part of the channel embankment, its crest being parallel to
the flow direction in the channel. Its function is to drain water from the channel when-
ever the water surface rises above a predetermined level so that the channel water
surface downstream of the weir remains below a maximum permissible level.
Oblique weir: The most striking difference between an oblique weir and other weirs
is that the crest of the oblique weir makes an angle with the flow direction in the
channel. The crest must be greater than the width of the channel so that with a change
in discharge the water surface upstream of the weir remains between narrow limits.
Some other weir types which can maintain such an almost constant upstream water
level will also be described.

3.2 Side weirs

3.2.1 General

In practice, sub-critical flow will occur in almost all rivers and irrigation or drainage
canals in which side weirs are constructed. Therefore, we shall restrict our attention
to side weirs in canals where the flow remains subcritical. The flow profile parallel
to the weir, as illustrated in Figure A3.1, shows an increasing depth of flow.
The side weir shown in Figure A3.1 is broad-crested and its crest is parallel to the
channel bottom. It should be noted, however, that a side weir need not necessarily
be broad-crested. The water depth downstream of the weir y2 and also the specific
energy head are determined by the flow rate remaining in the channel (Q2) and
the hydraulic characteristics of the downstream channel. This water depth is either
controlled by some downstream construction or, in the case of a long channel, it will
equal the normal depth in the downstream channel. Normal depth being the only
water depth which remains constant in the flow direction at a given discharge (Q2),
hydraulic radius, bottom slope, and friction coefficient of the downstream channel.

368
 I I
B
I
I

I I
I b I
k

CROSS SECTION

section
.... . 1
section 2

I
I- I
x=x1
x=x2

WATER SURFACE PROFILE

Figure A3.1 Dimension sketch of side weir.

3.2.2 Theory

The theory on flow over side weirs given below is only applicable if the area of water
surface drawdown perpendicular to the centre line of the canal is small in comparison
with the water surface width of this canal. In other words, if y - p I < O. 1 B.
For the analysis of spatially varied flow with decreasing discharge, we may apply
the energy principle as introduced in Chapter I , Sections 1.6 and 1.8. When water
is being drawn from a channel as in Figure A3.1, energy losses in the overflow process
are assumed to be small, and if we assume in addition that losses in specific energy
head due to friction along the side weir equal the fall of the channel bottom, the energy
line is parallel to this bottom. We should therefore be able to write

(A3. I )

If the specific energy head of the water remaining in the channel is (almost) constant

369
 n e n
0.l 0.2

Figure A3.2 Ho-y diagram for the on-going channel

while at the same time the discharge decreases, the water depth y along the side weir
should increase in downstream direction as indicated in Figures A3.1 and A3.2, which
is the case if the depth of flow along the side weir is subcritical (see also Chapter
1, Figure 1.9).
Far upstream of the side weir, the channel water depth y equals the normal depth
related to the discharge QI and the water has a specific energy Ho,o,which is greater
than Ho,.Over a channel reach upstream of the weir, the water surface is drawn down
in the direction of the weir. This causes the flow velocity to increase and results in
an additional loss of energy due to friction expressed in the loss of specific energy
head Ho,o- H o,2. Writing Equation A3.1 as a differential equation we get

(A3.2)

or

(A3.3)

The continuity equation for this channel reach reads dQ/dx = - q, and the flow rate
per unit of channel length across the side weir equals

(A3.4)

The flow rate in the channel at any section is

Q = A J Z g o
and finally

370
 dA
-= B-dy
dx dx
so that Equation A3.3 can be written as follows
dy 4Cs (Ho- y). (y - p).
(A3.5)
dx - 3IB A/B 2y - 2H0+
where C, denotes the effective discharge coefficient of the side weir. Equation A3.4
differs from Equation 1-36 (Chapter 1) in that, since there is no approach velocity
towards the weir crest, y has been substituted for Ho. Equation A3.5, which describes
the shape of the water surface along the side weir, can be further simplified by assuming
a rectangular channel where B is constant and A/B = y, resulting in

(A3.6)

For this differential equation De Marchi (1934) found a solution which was confirmed

x =- [ (-)
3I.B 2Ho-3p H o - y 0.5 - 3 arcsin
2cs Ho-P Y-P
(zr]
experimentally by Gentilini (1938) and Collinge (1957) and reads
K + (A3.7)

where K is an integration constant. The term in between the square brackets may!
be denoted as $(y/Ho) and is a function of the dimensionless ratios y/H0,, and p/H,,,
as shown in Figure A3.3. If pI, y2, and Ho,2are known, the water surface elevation
at any cross section at a distance (x - x2)along the side weir can be determined from
the equation*

If the simplifying assumptions made to write Equation A3.1 cannot be retained or

in other words, if the statement
V2
j-C2R - S t a n i << y2 - yI (A3.9)

is not correct, the water surface elevation parallel to the weir can only be obtained
by making a numerical calculation starting at the downstream end of the side weir
(at x = x2). This calculation also has to be made if the cross section of the channel
is not rectangular.
For this procedure the following two equations can be used

(A3.10)

* If the flow along the weir is supercritical and no hydraulic jump occurs along the weir and the same
simplifying assumptions are retained, Equations A3.1 to A3.8 are also valid. Greater discrepanties, however,
occur between theory and experimental results. Also, the water surface profile along the weir has a shape
different form that shown in Figure A3.1.

371
 Figure A3.3 Values of +(y/H,,*) for use in Equation A3.8

(A3. I I )

where; Ax = length of the considered channel section, u = subscript denoting up-

stream end of section, d = subscript denoting downstream end of section, C = coeffi-
cient of Chzy, R = hydraulic radius of channel.
It should be noted that before one can use Equations A3.10 and A3.11 sufficient
information must be available on both A and R along the weir. The accuracy of the
water surface elevation computation will depend on the length and the chosen number
of elementary reaches Ax.

I 3.2.3 Practical C,-value

I
The reader will have noted that in Equations A3.3 to A3.9 an effective discharge coeffi-
I cient Cs is used. For practical purposes, a value
c s = 0.95 Cd (A3.12)
may be used, where Cd equals the discharge coefficient of a standard weir of similar
crest shape to those described in Chapters 4 and 6 .
If Equations A3.4 to A3.11 are used for a sharp-crested side weir, the reader should
be aware of a difference of fi
in the numerical constant between the head-discharge
equations of broad-crested and sharp-crested weirs with rectangular control section.
In addition it is proposed that the discharge coefficient (C,) of a sharp-crested weir
be reduced by about 10% if it is used as a side weir. This leads to the following C,-value

312
to be used in the equations for sharp-crested side weirs
C, 0.90 J";Ce N 1.55 Ce (A3.13)

3.2.4 Practical evaluation of side weir capacity

Various authors proposed simplified equations describing the behaviour of sharp-

crested side weirs along rectangular channels. However, discrepancies exist between
the experimental results and the equations proposed, and it follows that each equation
has only a restricted validity. In this Annex we shall only give the equations as proposed
by Forchheimer (1 930), which give an approximate solution to the Equations A3.3
and A3.4 assuming that the water surface profile along the side weir is a straight line.
The Forchheimer equations read
AQ = Y1 +
C,-2 J z-g- S[------P~,]'.~
Y2 (A3.14)
3 3 2
and
Y2 - YI
(A3.15)
= v12/2g- v,2/2g - AH,
where AH, is the loss of specific energy head along the side weir due to friction. AH,
can be estimated from

(A3.16)

The most common problem is how to calculate the side weir length S, if AQ =
QI - Q2, y2 and pI are known. To find S an initial value of yI has to be estimated,
which is then substituted into the Equations A3.14 and A3.15. By trial and error yI
(and thus S) should be determined in such a way that = 1.O.
The Equations A3.14 and A3.15 are applicable if

and
Fr, = -
kl (A3.17)

Y1-P 2 0 (A3.18)
If the above limits do not apply, the water depth y, at the entrance of the side weir
and the side weir length S required to discharge a flow Q, - Q2 should be calculated
by the use of Equation A3.1, which reads

(A3.19)

In combination with the equation

(A3.20)

373
The latter equation is a result of substituting Equation A3.12 into Equation A3.8.
In using Equation A3.20 the reader should be aware that the term xI - x2 is negative
since x, < x2. As mentioned before, values of +(y/H,J can be read from Figure A3.3
as a function of the ratios p,/H,,, and Y/H,,~.

3.3 Oblique weirs

3.3.1 Weirs in rectangular channels

According to Aichel (1953), the discharge q per unit width of crest across oblique
weirs placed in a rectangular canal as shown in Figure A3.4 can be calculated by the
equation

(
q = 1- '
PI
P) qn (A3.21)

where qnis the discharge over a weir per unit width if the same type of weir had been
placed perpendicular to the canal axis ( E = 90") and p is a dimensionless empirical
function of the angle of the weir crest (in degrees) with the canal axis.
Equation A3.21 is valid provided that the length of the weir crest L is small with
respect to the weir width b and the upstream weir face is vertical. Values of the
coefficient are available (see Figure A3.5) for
hllp, < 0.62 and E > 30" . (A3.22)
or
h,/p, < 0.46 and E < 30" (A3.23)

3.3.2 Weirs in trapezoidal channels

Three weir types, which can be used to suppress water level variations upstream of
the weir are shown in Figure A3.6. Provided that the upstream head over the weir
crest does not exceed 0.20 m (h, < 0.20 m) the unit weir discharge can be estimated
by the equation
.

//
\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\~\'
Figure A3.4 Oblique weir in channel having rectangular cross section

314
Figure A3.5 p-values as a function of E

9 = rq, (A3.24)
where qn is the discharge across a weir per unit width if the weir had been placed
perpendicular to the canal axis (see Chapters 4 and 6) and r is a reduction factor
as shown in Figure A3.6.

3.4 Selected list of references

Aichel, O.G. 1953. Abflusszahlen fr schiefe Wehre. (Discharge ratios for oblique weirs.) Z.VDI 95. No.
1, Jan. 1, pp. 26-27.
Collinge, V.K. 1957. The discharge capacjty of side weirs. Proc. of the Inst. of Civil Engineers, Vol. 6,
Febr., pp. 288-304.
Engels, H. 1917. Versuche ber Streichwehre. Mitt. aus dem Dresdener Flussbau-Laboratorium.
Forschungsarbeiten auf dem Gebiete des Ingenieurwesens No. 200. Berlin.
Engels, H. 1917. Weitere Versuche ber Streichwehre. Mitt. aus dem Dresdener Flussbau-Laboratorium.
Forschungsarbeiten auf dem Gebiete des Ingenieurwesens no. 201. Berlin. 55 pp.
Forchheimer, Ph. 1930, Hydraulik. 3. Aufl., pp. 406-408.
Frazer, W.: 1957. The behaviour of side weirs in prismatic rectangular channels. Proc. of the Inst. of Civil
Eng., Vol. 6 , Febr., pp. 305-328.

375
Gentilini, B. 1938. Richerche sperimentali sugli sfioratori longitudinali (prima serie di prove). LEnergie
Elettrica, Milano. 15, Sept No. 9, pp. 583-595.
Henderson, F.M. 1966. Open Channel Flow. MacMillan Comp. New York. 521 pp.
De Marchi, G. 1934. Saggio di teoria de funzionamente degli stramazzi laterali. LEnergie Elettrica, 1 I ,
Nov., pp. 849-860. Milano.
Schaffemak, F. 1918. Streichwehrberechnung. sterreichische Wochenschrift f.d.ffentl. Baudienst. Heft
36.
Schmidt, M. 1954. Zur Frage des Abflusses iiber Streichwehre. Mitt. Nr. 41, Inst. fiir Wasserbau der Tech.
Universitt Berlin-Charlottenburg.
Schmidt, M. 1954-1955. Die Berechnung von Streichwehren. Die Wasserwirtschaft, pp. 96-100.
Ven Te Chow, 1959. Open channel hydraulics. McGraw-Hill, New York, 680 pp.

Drop in canal
bottom 2 h, max
crest length 1 0.15 m
L-J-
Upstream view for all three types

..................................I.............................,
A-T-1 o( 1M Oblique weir
tor a <45O r = 0.95

Giraudet weir
tor 45O<a< 70 LYY 0 . 4 0 m
b =2A+L
r E 0.90

-
Z weir
b = Ll+L2
I- 0.85

Figure A3.6 Weirs in trapezodal channels

376
Suitable stilling basins
4.1 Introduction
Unless a weir or flume is founded on rock, a downstream stilling basin will be neces-
sary. The floor of the stilling basin should be set at such a level that the hydraulic
jump, if formed, occurs on the sloping downstream weir face or at the upstream end
of the basin floor so that the turbulence in the jump will abate to a level which will
not damage the unprotected downstream channel bed. Calculations for the floor level
should be made for several discharges throughout the anticipated range of modular
flow. To aid the engineer in designing a suitable stilling basin, hydraukdesign criteria
of a number of devices are given below.

4.2 Straight drop structures

4.2.1 Common drop

Illustrated in Figure A4.1 is a drop structure that will dissipate energy if installed
downstream of a weir with a vertical back face. The aerated free falling nappe will
strike the basin floor and turn downstream at Section U. Beneath the nappe a pool
is formed which supplies the horizontal thrust required to turn the nappe downstream.
Because of the impact of the nappe on the basin floor and the turbulent circulation
in the pool beneath the nappe, some energy is lost.
Further energy will be dissipated in the hydraulic jump downstream of section U.
The remaining energy head downstream from the basin, H,, does not vary greatly

flow control conveyance J

, energy dissipation J
, protected transition
/p. /T' T
control
section

-- 7- T
drop of
- --

energy
level:AH
I
forinfluence of

-1 basin length 4L
g
-

Figure A4. I Straight drop structures

311
with the ratio AZ/H, and is equal to about 1.67 H , (adapted from Henderson 1966).
This value of 1.67 H I provides a satisfactory estimate for the basin floor level below
the energy level of the downstream canal. The hydraulic dimensions of a straight drop
can be related to the following variables (see Figure A4.1):
H , = upstream sill-referenced energy head n = step height
AH = change in energy head across structure yu = flow depth at section U
Hd = downstream energy head yd = downstream flow depth
q = discharge per unit width of sill relative to basin floor
g = accelaration due to gravity y2 = flow depth in downstream
channel
These variables can be combined to make a first estimate of the drop height
AZ = (AH + Hd)-HI (A4. I )
Subsequently, the flow velocity and depth at section U may be estimated by
vu = &E (A4.2)
and by the continuity equation

y = -9 (A4.3)
u vu
The flow at section U can best be characterized by the dimensionless Froude number
Fr, = ~ V (A4.4)
Jsyu
This Froude number can be related directly to the straight drop geometry through
the length ratios yd/AZ and LJAH, values of which can be read from Figure A4.2
(see also Figure A4.1).
The length of the hydraulic jump L,, downstream from section U in Figure A4.1,
can be calculated by (Henderson 1966),
Lj = 6.9 (Yd - Yu) (A4.5)
It is important to realize that the downstream water depths (yd and y2) are caused
not by the drop structure, but by the flow characteristics of the downstream canal.
If these characteristics are such that the required depth y,, is produced, a jump will
form; otherwise it will not form and not enough energy will be dissipated within the
basin. Additional steps, such as lowering the basin floor and adding an end sill, must
be taken to assure adequate energy dissipation.

Because of seasonal changes of the hydraulic resistance of thecanal, however, the

flow velocity as calculated by Mannings equation changes together with the water
depth Yd. The jump thus tends to drift up and down the canal. This unstable behavior
is often undesirable, and is then suppressed by increasing the flow resistance by means
of an abrupt step at the end of the basin. Usually, this step is constructed at a distance
Lj = 5(n + y2) (A4.6)
downstream of section U. For design purposes, Figure A4.3 can be used to determine
the largest required value of n, if Fr, = vul&, yu,and y2are known.
378
Figure A4.2 Dimensionless plot of straight drop geometry (Bos e.a. 1984)

Figure A4.3 Experimental relationships between Fr,, y2/yu,and n/y, for an abrupt step (after Forster and
Skrinde 1950)

379
4.2.2 U.S. ARS basin

The U.S. Agricultural Research Service has developed an alternative basin which is
especially suitable if tailwater level is greater than the sequent depth and varies inde-
pendently of the flow rate. This impact block type basin was developed for low heads
and gives a good energy dissipation over a wide range of tailwater levels. The energy
dissipation is principally by turbulence induced by the impingement of the incoming
jet upon the impact blocks. The required downstream water depth, therefore, can be
slightly less than with the previous basin but can vary independently of the drop height
AZ. To function properly, the downstream water depth yd must not be less than
I .45 HI,while at Q,,, the Froude number Fr, should not exceed 4.5.
Upstream from section U, the length L, may be determined by use of Figure A4.2.
The linear dimensions of the basin downstream from section U are shown in Figure
A4.4 as a function of H I .

-Lp+ 1.7H1 d-
Figure A4.4 Impact block type basin .

380
4.3 Inclined drops or chutes
4.3.1 Common chute

Downstream from the control section of either a weir or flume, a sloping downstream
face or expansion is a common design feature. The slope of the downstream face usual-
ly varies between 1 to 4 and 1 to 6. By approximation we may write that the flow
velocity over the downstream face equals
vu = SlYu (A4.7)
where q is the unit discharge on the downstream face and y, is the water depth at
a particular point on the downstream apron.
Values of y, may be determined by the use of Table A4.1. The symbols used in
Table A4.1 are defined in Figure A4.5.
A hydraulic jump will form in the horizontal (rectangular) basin provided that the
tailwater depth is greater than the sequent depth y, to y, and vu. Minimum values
of y, may be read from Figure A4.3 for rectangular basins. The length of such a hori-
zontal basin equals that part of the basin which is situated downstream of Section
U in Figure A4.1, and equals L, = 5(n + y,).
It is recommended that a tabulation be made of the Froude number Fr, near the
toe of the downstream face, and of the depth of flow y" throughout the anticipated

I ---_

i'
11 k-slope
I
length
section u

I rounded, r E1/2
Hi

o reduced
p s l o p e length-

Figure A4.5 Definition sketch for Table A4. I

38 1
Table A4.1 Dimensionless Ratios for Hydraulic Jumps

0.2446 3.00 0.3669 1.1006 1.4675 1.1006 O. 1223 1.2229

0.2688 3.10 0.3599 1.1436 1.5035 1.1157 0.1 190 1.2347
0.2939 3.20 0.3533 1.1870 1.5403 1.1305 0.1159 1.2464
0.3198 3.30 0.3469 1.2308 1.5777 1.1449 0.1130 1.2579
0.3465 3.40 0.3409 I .2749 1.6158 1.1590 0.1 103 1.2693
0.3740 3.50 0.3351 1.3194 1.6545 1.1728 0.1077 1.2805
0.4022 3.60 0.3295 1.3643 1.6938 1.1863 0.1053 1.2916
0.43 12 3.70 0.3242 1.4095 1.7337 1.1995 0.1030 1.3025
0.4609 3.80 0.3191 1.4551 1.7742 1.2125 0.1008 1.3133
0.4912 3.90 0.3142 I . 5009 1.8151 1.2253 0.0987 1.3239
0.5222 4.00 0.3094 1.5472 1.8566 1.2378 0.0967 1.3345
0.5861 4.20 0.3005 1.6407 1.9412 1.2621 0.0930 I .355l
0.6525 4.40 0.2922 1.7355 2.0276 1.2855 0.0896 1.3752
0.721 1 4.60 0.2844 1.8315 2.1159 1.3083 0.0866 1.3948
0.7920 4.80 0.2771 1.9289 2.2060 1.3303 0.0837 1.4140
0.8651 5.00 0.2703 2.0274 2.2977 1.3516 0.081 1 1.4327
0.9400 5.20 0.2639 2.1271 2.3910 1.3723 0.0787 1.4510
1.0169 5.40 0.2579 2.2279 2.4858 1.3925 0.0764 1.4689
1.0957 5.60 0.2521 2.3299 2.5821 1.4121 0.0743 1.4864
1.1763 5.80 0.2467 2.4331 2.6798 1.4312 0.0723 1.5035
1.2585 6.00 0.2417 2.5372 2.7789 I .4499 0.0705 1.5203
1.3429 6.20 0.2367 2.6429 2.8796 1.4679 0.0687 1.5367
1.4280 6.40 0.2321 2.7488 2.9809 1.4858 0.0671 1.5529
1.5150 6.60 0.2277 2.8560 3.0837 1.5032 0.0655 1.5687
1.6035 6.80 0.2235 2.9643 3.1878 1.5202 O O641 1.5843
1.6937 7.00 0.2195 3.0737 3.2932 1.5368 0.0627 1.5995
1.7851 7.20 0.2157 3.1839 3.3996 1.5531 0.0614 1.6145
1.8778 7.40 0.2121 3.2950 3.5071 1.5691 0.0602 1.6293
1.9720 7.60 0.2085 3.4072 3.6157 1.5847 0.0590 1.6437
2.0674 7.80 0.2051 3.4723 3.7254 1.6001 0.0579 1.6580
2.1641 8.00 0.2019 3.6343 3.8361 1.6152 0.0568 1.6720
2.2620 8.20 O. I988 3.7490 3.9478 1.6301 0.0557 1.6858
2.3613 8.40 0.1958 3.8649 4.0607 1.6446 0.0548 . 1.6994
2.4615 8.60 0.1929 3.9814 4. I743 1.6589 0.0538 1.7127
2.5630 8.80 0.1901 4.0988 4.2889 1.6730 0.0529 1.7259
2.6656 9.00 O. 1874 4.2171 4.4045 1.6869 0.0521 1.7389
2.7694 9.20 O. 1849 4.3363 4.521 1 1.7005 0.0512 1.7517
2.8741 9.40 O. 1823 4.4561 4.6385 1.7139 0.0504 1.7643
2.9801 9.60 0. 1799 4.5770 4.7569 1.7271 0.0497 1.7768
3.0869 9.80 o. 1775 4.6985 4.8760 1.7402 0.0489 1.7891
3.1949 10.00 0.1753 4.8208 4.9961 1.7530 0.0482 1.8012
3.469 I 10.50 0.1699 5.1300 5.2999 1.7843 0.0465 1.8309
3.7491 11.00 0.1649 5.4437 5.6087 1.8146 0.0450 1.8594
4.0351 11.50 0.1603 5.7623 5.9221 1.8439 0.0436 1.8875
4.3267 12.00 O. I560 6.0853 6.2413 1.8723 0.0423 1.9146
4.6233 12.50 O. I520 6.4124 6.5644 1.9000 0.041 1 1.9411
4.9252 13.00 O. 1482 6.7437 6.8919 1.9268 0.0399 1.9667
5.2323 13.50 0.1447 7.0794 7.2241 1.9529 0.0389 1.9917
5.5424 14.00 0.1413 7.4189 7.5602 1.9799 0.0379 2.0178
5.8605 14.50 0.1381 7.7625 7.9006 2.0032 0.0369 2.0401
6.1813 15.00 0.1351 8.1096 8.2447 2.0274 0.0361 2.0635
6.5066 15.50 O. 1323 8.4605 8.5929 2.051 1 0.0352 2.0863
6.8363 16.00 O. 1297 8.8153 8.9450 2.0742 0.0345 2.1087
7.1702 16.50 0.1271 9.1736 9.3007 2.0968 0.0337 2.1305
7.5081 17.00 O. 1247 9.5354 9.6601 2.1 190 0.0330 2.1520
7.8498 17.50 0. 1223 9.9005 10.0229 2.1407 0.0323 2.1731
8.1958 18.00 0.1201 10.2693 10.3894 2.1619 0.0317 2.1936
8.5438 18.50 0.1180 10.6395 10.7575 2.1830 0.031 1 2.2141
8.8985 19.00 0.1159 11.0164 11.1290 2.2033 0.0305 2.2339
9.2557 19.50 0.1140 11.3951 11.5091 2.2234 0.0300 2.2534
9.6160 20.00 0.1122 11.7765 11.8887 2.2432 0.0295 2.2727
discharge range. The sequent depth rating should be plotted with the stage-discharge
curve of the tailwater channel to ensure that the jump forms on the basin floor.

4.3.2 SAF Basin

An alternative stilling basin suitable for use on low-head structures was developed
at the St. Anthony Falls Hydraulic Laboratory (SAF-basin) of the University of Min-
nesota. The basin is used as a standard by the U.S. Soil Conservation Service, and
has been reported by Blaisdell(l943, 1959). The general dimensions of the SAF-basin
are shown in Figure A4.6.
The design parameters for the SAF-basin are given in Table A4.2.

Table A4.2 Design parameters of the SAF-basin

Fr, =vu JgA,IB, LB/Y2 TW/Y2

1.7 to 5.5 4.5/~r,0.~~ 1.1-Fr,2/120

5.5 to 1 1 4.5/Fr:.76 0.85

1 1 to17 4.5 Fr,0.76 1 .O - Fr,2/800

In Table A4.2 yz is the theoretical sequent depth of the jump corresponding to yu

as shown in Figure A4.3. The height of the end sill is given by C = 0.07 y2 and the
freeboard of the sidewall above the maximum tailwater depth to be expected during
the life of the basin is given by z = y2/3.
The sidewalls of the basin may be parallel or they may diverge. Care should be
taken that the floor blocks occupy between 40 and 55% of the stilling basin width,
so that their width and spacing must be increased with the amount of divergence of
the sidewalls. The effect of air entrainment should not be taken into account in the
design of the basin; however, its existence within the stilling basin calls for a generous
freeboard (y,/3).

4.4 Riprap protection

To prevent bank damage by erosive currents passing over the end sill of a basin or
leaving the tail of a structure, riprap is usually placed on the downstream channel
bottom and banks. Several factors affect the stone size required to resist the forces
which tend to move riprap. In terms of flow leaving a structure, these factors are
velocity, flow direction, turbulence and waves. The purpose of this section is to give
the design engineer a tool to determine the size of riprap to be used downstream from
discharge measurement devices or stilling basins and to determine the type of filter
or bedding material placed below the riprap.

383
 RECTANGULAR STILLING BASIN
HALF PLAN

HALF PLAN
TRAPEZOIDAL STILLING BASIN

CENTERLINE SECTION

A_--; h ----------

f - - t
TRAPEZOIDAL STILLING BASIN RECTANGULAR STILLING BASIN

DOWNSTREAM ELEVATION

Figure A4.6 SAF-basin dimensions

384
Design A Tailwater depth calculated byTW/y2 = 1.1 - Fr,2/120

Design B Tailwater depth is 15% greater than in Design A

PHOTOS:
I :20 scale model of SAF stilling basin discharging 1200 m3/s in prototype
b, = 40.0 m, AH = 3.50 m

385
4.4.1 Determining maximum stone size i n riprap mixture

From published data, a tentative curve was selected showing the minimum stone dia-
meter as a function of the bottom velocity. This curve is shown in Figure A4.7. Down-
stream of stilling basins, the conception bottom velocity is difficult to define because
of the highly turbulent flow pattern. The velocity at which the water strikes the riprap
is rather unpredictable unless the basin is tested.
For practical purposes, however, Peterka (1964) recommends that, to find the stone
diameter in Figure A4.7, use be made of the average velocity based on discharge di-
vided by cross-sectional area at the end sill of the stilling basin. If no stilling basin
is needed because Fr, < 1.7, Figure A4.7 should be entered with the impact velocity,
being
v, = JW (A4.8)
More than 60% of the riprap mixture should consist of stones which have length,
width, and thickness dimensions as nearly alike as is practicable, and be of curve size
or larger; or the stones should be of curve weight or heavier and should not be flat
slabs.

4.4.2 Filter material placed beneath riprap

If riprap stones of a protective lining were to be installed directly on top of the fine
material in which the canal is excavated, grains of this subgrade would be washed
through the openings in between the riprap stones. This process is partly due to the
turbulent flow of canal water in and out of the voids between the stones and partly
due to the inflow of water that leaks around the structure or flows into the drain.
To avoid damage to a riprap protection because of the washing of subgrade, a filter
must be placed between the riprap and the subgrade (see Figure A4.8). The protective
construction as a whole and each separate layer must be sufficiently permeable to
water entering the canal through its bed or banks. Further, fine material from an
underlying filter layer or the subgrade must not be washed into the voids of a covering
layer.

4.4.3 Permeability to water

To maintain a sufficient permeability to water of the protective construction of Figure

A4.8, the following d,,/dlsratios should have a value between 5 and 40 (USBR 1973):
d,, layer and d,, layer and d,, layer 1 = 5 to40 (A4.9)
d,, layer 2 d,, layer 1 d,, subgrade
where dI5equals the diameter of the sieve opening whereby 15% of the total weight
of the sample passes the sieve. Depending on the shape and gradation of the grains
in each layer, the above-mentioned 5 to 40 range of the ratios can be narrowed as
follows (Van Bendegom 1969):

386
 1D
9 mo

-
Bw
t; Mo
d
. 4002
Ec 5S 300
.E
u
L
$ 4 E;
!.nE
Ea-
3 40

c :5
0
L
o 2 10
E
0
6
4
3
2
al
o 1
0
0 O0
O6
7
04
6 a3
5

3 004
003
002
2

O004
ow3 LEGEND

0.m
o
Wo2
.-
.p
DUBUT 1786
W N K E I W 1845
0
OOOOB
a001 .a BLACKWELL 1857
O0006 .L SAINJON 1871
7
am4 .t SUCHIER 1074
6 00003
.$ GILBERT 1814
5 wo02

4 om1
00oooB
OXC.36
3 woo04
woo03
OXC.32
2
o m 1
woW0
000000 FOREIGN.PROT0 11)
BONNEVILLE FLUME I 2 1
oooooo BONNEVILLE CLOSUREIZ)
MOOOO
oDxoo + COLUMBIA RIVER.PROT0 12)
4001 LOS ANGELES DIST.PROT0 I 2 1
8 O LUIDER ZEE.PROT0 12)
aMMo0
0 PASSAMA~UODY.MODELI21
B STILLING BASIN.MODELI3)
7
4 CHANNEL.MODELI3)
6 ME NARY DAM.MODEL(4)
5
11) E H HOOKER.ASCE.VOL 36
4 V ) C I GRlMM hN LEUPOLD. 1939
I31 WATERWAVS EXP STATION
~~

14) BONNEVILLE LABORATORY

3

1
1 2 4 5 0 .o 1 2
bottom velocit n mlsec

Figure A4.7 Curve to determine maximum stone size

387
 erosion
protect ion

construct ion
filter

original
moterlol

Figure A4.8 Example of filter between riprap and original material (subgrade) in which canal is excavated

1. Homogeneous round grains (gravel) 5 to 10

2. Homogeneous angular grains (broken gravel, rubble) 6 to 20
3. Well-graded grains 12 to 40
To prevent the filter from clogging it is, in addition, advisable that for each layer
d5 2 0.75 mm (A4.10)

4.4.4 Stability of each layer

To prevent the loss of fine material from an underlying filter layer or the subgrade
through the openings in a covering layer, two requirements must be met:

The following d,,/d,, ratios should not exceed 5 (Bertram 1940)

dl5 layer 3
d,, layer 2
and
dl5 layer 2
dB5layer 1
and
dl5 layer 1
d,, subgrade ' (A4.11)

while the dso/d,o should range between 5 and 60 (U.S. Army Corps of Engineers 1955).
d50 layer 3 d50 layer 2 dsolayer 1
and = 5 to 60 (A4.12)
d5o layer 2 and d50layer 1 d50 subgrade
As before, the ratio in Equation A4.12 depends on the shape and graduation of the
grains as follows:
I . Homogeneous round grains (gravel) 5 to 10
2. Homogeneous angular grains (broken gravel, rubble) I O to 30
3. Well-graded grains '12 to 60
The requirements in this section describe the sieve curves of the successive filter layers.
Provided that the sieve curve of the riprap layer and the subgrade are known, other
layers can be plotted. An example of plotting sieve curves of a construction consisting
of one riprap and two filter layers is shown in Figure A4.9. In practice one should
use materials that have a grain size distribution which is locally available, since it
is uneconomic to compose a special mixture. To provide a stable and effectively func-
tioning filter, the sieve curves for subgrade and filter layers should run about parallel
for the small-diameter grains.

.:
388

i'
i
 Figure A4.9 Sieve curves of a filter construction

4.4.5 Filter construction

To obtain a fair grain size distribution throughout a filter layer, each layer should
be sufficiently thick. The following thicknesses must be regarded as a minimum for
a filter construction made in the dry
- sand, fine gravel O.5 to O. 10 m
- gravel O. 10 to 0.20 m
- stones 1.5 to 2 times the largest stone diameter.
With filters constructed under water, these thicknesses have to be increased considera-
bly to compensate for irregularities in the subgrade and because it is more difficult
to apply an even layer under water.

Many variations can be made on the basic filter construction. One or more of the
layers can be replaced with other materials. With some protective linings, only the
riprap layer is maintained, while the underlying layers are replaced by one single layer.
For example
- concrete blocks on a nylon filter
- stones on braided azobe slabs on plastic filter

- gabions on fine gravel

- nylon-sand mattresses
The usual difficulty with these variants is their perviousness to underlying sand. The
.
openings in each layer should not be greater than 0.5 x d,, of the underlying material.
If openings are greater, one should not replace all underlying layers but maintain as
many layers (usually one) as are needed to prevent the subgrade from being washed
through the combined layer.
At structure-to-filter and filter-to-unprotected channel joints, the protective con-
struction is most subject to damage. This is because the filter layer is subject to subsi-

389
 we,ep hole

1 . -
(wood)

Figure A4.10 Examples of filter construction details (after van Bendegom 1969)

dence while the (concrete) structure itself is well founded. Underlying material (sub-
grade) may be washed out at these joints if no special measures are taken. It is recom-
mended that the thickness of the filter construction be increased at these places. Some
examples of common constructional details are shown in Figure A4.10.

As a rule of thumb we may suggest a length of riprap protection which is neither

less than 4 times the (maximum) normal depth in the tailwater channel, nor less than
the length of the earth transition, nor less than 1.50 m.

4.5 Selected list of references

Van Bendegom, L. et al. 1969. Principles governing the design and construction of economic revetments
for protecting the banks of rivers and canals for ocean and inland navigation. 20th Intern. Navigation
Congr. Paris, 43 pp.
Berry, N.K. 1948. The start of bed load movement. Thesis. Univ. of Colorado, USA.
Bertram, G.E. 1940. An experimental investigation of protective filters. Publications of the Graduate School
of Engineering. Harvard University. No. 2657.
Blaisdell, F.W. 1943. The SAF stilling basin. U.S. Dept.of Agric. Soil Conservation Service. St. Anthony
Falls Hydraulic Laboratory, Minneapolis, Minn.

390
Blaisdell, F.W. 1959. The SAF stilling basin. A structure to dissipate the destructive energy in _high-velocity
flow from spillways. U.S. Dept.of Agric. Service in cooperation with the Minnesota Agric. Exp. Sta.
& St. Anthony Falls Hydraulic Laboratory. Agric. Handbook 156. Washington D.C. U.S. Gov. Printing
Office.
Bos! M.G., J.A. Replogle and A.J. Clemmens 1984. Flow measuring flumes for open channel systems.
John Wiley, New York. 321 pp.
Canals and Related Structures 1961. Commissioners office, Denver (Col.). U.S. Dept.of the Interior,
Bureau of Recl. Design Standards 3.
Design of Small Dams. 1973. 2nd edition, U.S. Dept.of the Interior. Bureau of Recl., Washington D.C.
U.S. Gov. Printing Office. 816 pp.
Dort, J.A. van and M.G. Bos. 1974. Drainage principles and applications. Chapt.29:Main drainage system
Publication 16. Vol. IV. Wageningen. pp. 124-224.
Forster, J.W., and R.A. Skrinde. 1950. Control of the hydraulic jump by sills. Transactions, American
Society of Civil Engineers. Vol. 115, pp. 973-987.
Henderson, F.M. 1966. Open Channel Flow. MacMillan Co., New York 522 pp.
Mavis, F.T., and L.M. Laushey 1948. A reappraisal of the beginnings of bed movement-competent velocity.
Proc. of the Int. Assoc. for Hydraulic Research. Stockholm. pp. 213-218.
Peterka, A.J. 1964. Hydraulic design of stilling basins and energy dissipators. U.S. Dept. of the Interior.
Bureau of Recl. Water Resources Techn. Publ. Engineering Monograph No. 25.223 pp.
U.S. Army Corps of Engineers 1955. Drainage and erosion control-subsurface drainage facilities for air-
fields. Part XIII, Chapter 2, Engineering Manual, Militairy Construction, Washington, DC. 15 pp.
Vlugter, H. 1941. 121/2Jaar hydrodynamische research aan waterloopkundige modellen in Nederlandsch
Indi. De Ingenieur in Ned. Indi. No. 9.

39 1
List of principal symbols
A cross-sectional area L2
a height of rectangular weir section (Sutro) L
a acceleration LT-2
B channel surface width L.
bC breadth at bottom of control section L
be +
effective breadth of weir crest (bc Kh) L
Cd discharge coefficient dimensionless
C" approach velocity Coefficient dimensionless
Ce effective discharge coefficient ( c d c v ) dimensionless
C subscript for critical flow condition dimensionless
D diameter of float L
DP diameter of pipe L
dC diameter of circular weir L
E energy ML2T-2
E complete elliptical integral of the first kind dimensionless
e exponential number, 2.71828 dimensionless
F force MLT-2
F coefficient correction factor dimensionless
Fr Froude number, Q(B/gA3)''2 dimensionless
f friction coefficient in the Darcy-Weisbach equation dimensionless
f drowned flow reduction factor dimensionless
G weight MLT-2
G relative slope factor dimensionless
g gravitational acceleration LT-2
H total energy head over crest L
Ho specific energy L
Hl total upstream energy head over crest L
H2 total downstream energy head over crest L
h, upstream head over crest L
h2 tailwater head over crest L
he +
effective upstream head over crest (h, Kh) L
Ah head loss over structure (h, - h2) L
K weir constant dimensionless
K head loss coefficient dimensionless
K complete elliptical integral of the second kind dimensionless
k filling ratio circular weir (h,/d3.5 dimensionless
k acceleration due to mass forces LT-2
L flowwise length of crest L
L length of channel reach L
1 length of pipe L
m mass M
m coordinate direction (binormal) dimensionless
n coordinate direction (principal normal) dimensionless
n number of data dimensionless

392
P wetted perimeter of flow cross-section dimensionless
P pressure M L-l T-2
PI height of crest above approach channel bed L
P2 height of crest above tailwater channel bed L
Q discharge rate L3T-I
Q, discharge rate through rectangular section L3T-'
QC discharge rate through curved section L3T-I
Qair volumetric air discharge rate L3T-I
9 discharge per unit width L2T-'
R hydraulic radius (A/P) L
Rb radius of embankment L
r radius of circular weir L
r radius of curved streamline L
r radius of float-wheel L
r radius of round-nose weir crest L
S length of side weir L
ski submergence ratio (H2/H1) dimensionless
Sh submergence ratio (h2/hl) dimensionless
S m modular'limit dimensionless
S coordinate direction (velocity direction) dimensionless
Tf resisting torque due to friction ML2T-2
TW tailwater level L
t time T
U power of head or of differential head dimensionless
V volume of fluid L3
V
-
fluid velocity LT-I
V average fluid velocity (Q/A) LT-I
W friction force MLT-2
W acceleration due to friction LT-2
W underflow gate opening L
X relative error dimensionless
X horizontal distance L
X breadth of weir throat at height y above crest L
X factor due to boundary roughness dimensionless
X Cartesian coordinate direction dimensionless
Y vertical distance L
Y vertical depth of flow L
Y coordinate direction dimensionless
Z coordinate direction dimensionless
Z side slope ratio horz/vert dimensionless
AZ drop height L
CY velocity distribution coefficient dimensionless
CY diversion angle degrees
P half angle of circular section ('h CY) degrees
Y QmaJQmin dimensionless
6 error dimensionless
6 contraction coefficient dimensionless

393
A small increment of dimensionless
A (p, - p)/p: relative density dimensionless
0 weir notch angle degrees
0 angle of circular section degrees
7.t circular circumference-diameter ratio: 3.14 16 dimensionless
P mass density of water M L-3
Pair mass density of air M L-3
PS mass density of bed material ML-3
o circular section factor dimensionless
5 friction loss coefficient dimensionless
o standard deviation dimensionless
o relative standard deviation dimensionless

Subject index

A
Access door 2.6 Baffle-type stilling basin 8.3.1
Accuracy of measurement 3.2.10;A 2.1 Bank damage, prevention of erosion
Accuracy of propeller meters 9.7.4 A 4.4
Actual head 6.6.1; 6.6.2 Basic discharge coefficient 4.4.2
Adjustable orifice 8.5.1 Bedding material beneath riprap
Adjustable proportional module 8.5.1 A 4.4.2
Adjustable sliding gate 8.5.1 Bed-load, see also Total-load equation
Aeration demand 1.10 Bed-load 3.2.6
Aeration demand of weirs 1.14 Bernoullis equation 1.7;8.4.3;A1.3
Air bubbles, prevention of 2.12 Bifurcation 3.2.5
Air pocket 1.13; 1.14; 8.3.1 Bi-normal 1.1
Air pocket underneath the nappe 1.4 Bottom velocity A4.4.1
Algal growth 1.12 Boundary layer, influence of 4.1.2
Algal growth, prevention of A 2.6 Boundary layer, displacement
Algal growth on weirs 3.2.8 thickness of sidewalls 4.5.1
Alternate depth 1.8; 8.2.3; 8.4.3 Brink depth method 9.5
Angle of divergence 1.15.2 Broad-crested weir 1.9; 4
Anti-vortex baffle 8.3.1 Butchers weir 6.5
Approach channel 2.1; 2.3
Archimedes law 2.9
Average flow velocity 1.6
C
California pipe method 9.4.2
B Canal bifurcation
Cartesian coordinate system
3.2.5
A 1.2
Backwater effect, avoiding of 3.2.6 Cavitation, danger of 6.6.1; 6.7.1
Baffle module 8.7.1 Centripetal acceleration 1.4; A1.4
Baffles 2.3 Channel expansions 1.15

394
Check structure 6.1.1 Discharge determination errors, causes
Chezy coefficient 3.2.6 9.7.2
Chutes, inclined A 4.3 Discharge equations, see Head
Cipoletti weir 5.3 discharge, Stage discharge
Circular sharp-edged orifice 8.1 Discharge measuring, displacement
Circular weir 5.4 principle 9.6.1
Cleaning of the intake pipes 2.6 Discharge measuring structure, see also
Coefficient, see also Contraction Measuring structures
Coefficient ,Discharge coefficient Discharge measuring structure,
Coefficient correction factor 4.4.2 function of 3.2.1
Coefficient errors A 2.6 Discharge measuring structure, errors
Confidence level for errors A 2.1 A2.4
Constant-head-orifice 8.3 Discharge volume measurement, error
Continuity equation 1.2;1.7 in A2.8
Contraction coefficient Diseases, prevention of 3.2.9
1.12; 8.4.2; 8.8.2 Distribution of errors A2.4
Control section 1.8 Division boxes 9.1.1
Converging section 2.1 Divisors 9.1
Co-ordinate directions 1.1 Downstream expansion 1.15.1
Coriolis force A1.2 Downstream expansion, truncation of
Corrugated pipes 8.6.1 1.15; 7.2.2
Counterweight 2.6; 2.9 Downstream head over the crest 2.4
Crest elevation 3.2.2 Drop 6.1.1
Critical depth of flow 1.8 Drop, inclined A4.3
Critical depth flumes 1.11;7 Drop height 1.14; A4.2.1
Crump weir 6.3.1 Drop structure A4.2
Crump-De Gruyter adjustable orifice
8.5 E
Curvature of streamlines A1.4
Cutthroat flume 1.11; 7.3.1 Eddy 8.3.1
Cylindrical crested weir 6.7 Effective discharge coefficient
1.12; 1.13
D Effective discharge coefficient of side
weir A3.2.2; A3.2.3
Damage to measuring structures 3.2.8 Elevation head 1.3; A1.3
Danadean tub 8.8 Energy, see also Kinetic energy,
Darcy-Weissbach equation 1.14; 9.2.2 Potential energy
Debris, passing of 3.2.7 Energy dissipation A4.2.2
Design head 6.6.1; 6.6.2 Energy losses at base of straight drop
Dethridge meter 9.6 A4.2.1
Differential head meter 2.12 Energy losses beneath the nappe A4.2.1
Dip-stick 2.6 Energy losses over hydraulic jump
Discharge, percentage error in 1.14 A4.2.1
Discharge, undesirable change of 3.2.8 Energy losses over metergate 8.6.3
Discharge coefficient 1.9.1 Entrainment of air, prevention 8.1.3
Discharge coefficient of side weir Entrance transition 7.1.1
A3.2.2 Equation ofmotion 1.3; A1.2; A1.3

395
Erosion, bank damage by A4.4 Gate opening 8.6.2
Error, see Measurement error, Registra- Gauge, see Flood gauge, Recording
tion error, Systematic error gauge
Euler, equation of motion A l .2;A1.5 Gauging station 2.1
Expansions in closed conduits 1.15.2 Gradual expansion 1.15.2
Gravitational force A l .2
F Guiding grooves 6.5.1

Faiyum weir 4.5

Filter construction ' A4.4.3
H
Filter material beneath riprap A4.4.2 Head-discharge equation 1.9; 1.13
Float, diameter of 2.9 Head losses across a propeller meter
Float tape 2.6; 2.8 9.7.3
Float wheel 2.8 Head losses in the intakes 2.9
Float-operated recorder 2.5; 2.6 Head measurement 2.1
Flood gauge 2.7 Head measurement, errors in A2.5
Flow disturbances 8.3.1 Head measurement station 2.2
Flow divisor, see also Divisors Head measurement station, location of
Flow divisor, function of 3.2.1 2.2; 6.3.1
Flow geometry at straight drops Head meters 2.12
A4.2.1 Head-reading errors, reduction of
Flow parameter 3.2.6 8.3.1
Flow straightening vanes 9.7.2 H-flume 1.11; 7.5
Flow totalizer 3.2.1 High-water mark 2.7
Fluid mechanics, basic equations A l Horizontal crest 1.9
Flume, see also Cutthroat flume, Horizontal pipe, flow from 9.4
H-flume, Parshall flume, Critical- Hydraulic drop 1.8
depth flume Hydraulicjump 1.8; 2.3; 8.2.3; 8.4.3
Flume, required fall of energy head Hydraulic jump, sequent depth A4.2.1
3.2.2 Hydrostatic pressure distribution
Forchheimer equation A3.2.4 1.4; A I S
Foundation level of a stilling well 2.6 Hydrostatic pressure in the m-direction
Fountain flow from a pipe 9.3.1 1.5
Free discharge 1.12
Free discharging orifice 1.12
Free overfall 9.5.1
I
Freezing, protection of recorders 2.1 1 Impact block type basin A4.2.2
Friction A l .2 Inclined drops A4.3
Friction moment, see Internal moment Instrument errors in measurement
Froude number A4.2.1 A2.5
Fully aerated nappe 1.13 Instrument shelter 2.10
Fully contracted orifice 8.1.1 Intake pipes 2.6
Fully suppressed contraction 1.12 Intermediate piers 4.2.1
Internal friction moment (recorders)
G 2.9
Irrigation water, measuring of volume
Gate leaf, shape 8.6.2 9.6.1

396
 J Modular flow
errors in measurement of
1.8
A2.4
i Jet below the gate 8.4.2 Modular limit 1.15.2; 3.2.2
Jet flow from the pipe 9.3.1 calculation of 1.15.3
Modular limit of flume, estimate of
K Montana flume
7.1.3
7.4 1
Kinetic energy 1.3; A1.3 Motion, equation of 1.3
Movable gate 6.5.1
L Movable gauge
Movable partition board
6.5.1
9.1.1
Lag error in measurement A 2.5
Level recorders 2.8; 2.9
Level recorders, housing of 2.10
N
Level recorders, protection against Nappe, curvature of 1.14
freezing 2.11 underpressure beneath 1.14
Long-throated flumes 1.15.3; 7.1 Near constant orifice discharge 8.7.1
Long-throated flumes, application of Negative pressure 6.6.1
7.1.4 of the crest 6.7.1
Net impressed force A1.2
M Newtons law of motion
Neyrpic module
1.4;A1.2
8.7
Manning equation 3.2.2 Neyrpic weir profile 9.1.2
Mass forces A1.2 Normal depth A3.2.1
Maximum stage gauge 2.7 Normal plane 1.1
Measurement error 3.2.10; A2.2 Normal pressure A1.2
Measurement of flow, accuracy of
A2.1
Measurement of flow, propagation of
O
errors A2.4 Oblique weirs A3.3
Measuring device 2.1 Orifice . 1.12; 8
Measuring structures, required fall of energy head 3.2.2
accuracy of 3.2.4 Orifice box 8.2.1
application of 3.3 Orifice flow, head discharge equations
calibration of 3.4.3 8.1.2
damage to 3.2.8 Orifice plates 8.1.1
function of 3.2.1 Osculating plane 1.1;Al.2
selection of 3.4 Outlet box, influence on meter
Meter registration, factors affecting accuracy 9.7.2
9.7.2
Metergate 8.6
flow through 8.6.1
P
installation of 8.6.3 Parshall flume 1.11; 7.4
Meyer-Peter/Mller bed-load function modular flow in 3.2.2
3.2.6 Pathline under steady flow conditions
Minimum pressure at the weir crest A1.3
6.7.1

397
Pendulum actuated revolution counter
9.6.1
S
Permeability of filter material A4.4.2 SAF basin A4.3.2
Piezometer tap 8.6.1 Sediment discharge capability 3.2.6
Piezometric gradient 1.4; A1.4 Selection of structures 3.1
Piezometric head 1.3; A1.3 Sensitivity of structure 3.2.4
measurement 2.2 Sensitivity of weir 1.13.7
Pipes, see also Horizontal pipe, Vertical Separation pocket 6.3.1; 6.4.1
Pipe Separation bubble 1.10; 4.4.1
Pipes, use of 9.2.1 Sequent depth rating A4.3.1
spiral flow in 9.7.2 Sharp-crested side weirs A3.2.4
Point binomial distribution of errors Sharp-crested weirs 1.10; 1.13; 5
A2.4 Short-crested weirs 1.10; 6
Potential energy, conversion into Side weir A3.2
kinetic energy 1.15.3 Side weir capacity, evaluation A3.2.4
Pressure detection 2.6 Sighting rod readings 9.3.2
Pressure energy 1.3; A1.3 Siphons 9.2.1
Pressure head 1.3; A1.3 Sluice gate 8.2.1; 8.2.2; 9.6.1
Propeller meters 9.7 Specific energy 1.8
Propeller rotation 9.7.2 Spillway face, inclination of 6.6.1
Proportional divisor 9.1.1 Spillways 6.6.1; 6.7.1
Proportional weir 5.5 Spiral flow in pipes 9.7.2
Purdue trajectory method 9.4.2 Spurious errors in measurement A2.2
Staff gauge 2.5; 2.6
R Stage-discharge equations (empirical)
1.11
Radial gate 8.4 Stage-discharge relationship 6.5.2
Random errors in measurement A2.2 Standard deviation A2.4
Recorders, see Level recorders Standing wave 1.8
Recording gauge 2.8 Standing wave flume 1.1 1
Rectangular gate leaf 8.6.2 Standing wave weir, movable 6.5
Rectangular sharp-edged orifice 8.2 Stilling basin A4
Registration error, see also Accuracy, Stilling well ' 2.6
Measurement error, Meter errors in measurement A2.5
registration Stone size in riprap mixture A4.4.1
Registration error 2.9; 9.7.2 Straight drop structures A4.2
Regulating device 2.1 Stream tube 1.2
Regulating structures, see also Streamline curvature, influence of 1.1O
Measuring structures Streamlines 1.2
Regulating structures, selection 3.4 Structures see Measuring structures
Relative error in measurement A2.3 Subcritical flow 1.8
Relative standard deviation A2.4 Submerged calibrated valve gate 8.6.1
Resonance of the overfalling jet 1.13 Submerged orifice 1.12; 8.1.1
Revolution counters 9.6.1 Submergence ratio 1.15.1; 1.15.2; 2.4
Ripple factor 3.2.6 Suction lift head meter 2.12
Riprap protection A4.4 Supercritical flow 1.8
Romijn movable weir 4.2

398
Superposition of head-discharge Upstream spillway face, inclination of
equations 1.9.5; 1.13.4 6.6.1
Suspended-load 3.2.6 Upstream water level (over the crest)
Sutro weir 5.5.1 1.9.1
Systematic error 2.6; 2.9; A2.2 U.S. ARS basin A4.2.2
Systematic percentage error in
measurement A 2.8
V
T Valve gate
Velocity above the weir crest
8.6.1
1.10
Tailwater channel 2.1; 2.4 Velocity coefficient 1.6; 1.9.1
Tailwater level 2.1; 2.4 Velocity distribution 1.4; 1.6
Tailwater measurement 2.4 Velocity head 1.3;A1.3
Tainter gate 8.4 Velocity profile, influence on
Tape index pointer 2.8 registration 9.7.2
Thomson weir 5.2.1 Vena contracta 1.12; 5.5.2
Throat of the flume 1.11 Ventilation pipe 8.3.1
Throatless flumes 7.2; 7.3 Venturi flume 7
Torricellis equation 1.13 Vertical pipe, flow from 9.3
Total energy head 1.3; A1.3 V-notch sharp-crested weir 5.2
Total energy head of open channel 1.6 V-notch weir sill 6.2
Total-load equation 3.2.6 Volume of irrigation water, measuring
Trajectory method 9.4.2 9.6.1
Transition reach 1.15.2
Transport parameter
Trash rack
3.2.6
3.2.7
w
Triangular broad-crested weir 4.3 Wash-load 3.2.6
Triangular profile flat-V weir 6.4 Water level, registration error 2.9
Triangular profile weir 6.3 Water surface along side weir A3.2.2
Truncated Parshall flume 7.4.1 Water wheel 9.6.1
Truncated transition 1.15.3 Weir, see also Broad-crested weir,
Truncation, point of 6.3.1 Short-crested weir, Sharp-crested
Truncation of downstream expansion weir
. 7.2.2 Weir, aeration demand 1.14
Tube-float differential head meter 2.12 required fall of energy head 3.2.2
Turnout gate 8.3.1 sensitivity 1.13.7
Weir block, truncation of 6.3.1
U permissible truncation 6.4.1
Weir constant 1.13.7
Undershot gates 3.2.2 Weir face, inclination of 6.6.1
Undershot water wheel 9.6.1 Weir notch angle 4.3.1; 4.3.4
Undular jump 1.8 WES-spillway 1.10; 6.6
Uniform distribution of errors A2.4
Upstream channel, minimum water
level in 3.2.9
Z
Upstream energy head over the crest Zero-setting 2.1
1.9; 2.1

399
Currently available ILRI publications
No. Publications Author ISBN No.

11 Reclamation of salt-affected soils in Iraq. P. J. Dieleman (ed.) -

14 Irrigation requirements for double cropping of low- G. A. W. van de Goor 90 70260 840
land rice in Malaya. and G. Zijlstra
15 Planning of service centres in rural areas of developing D. B. W. M. van Dusseldorp -
countries
16 Drainage principes and applications (in 4 volumes) - 90 70260 123,
-13 1, -62 X
I
and -63 8
16 Principes y aplicationes del drenaje (en 4 volmenes). - -
17 Land evaluation for rural purposes. R. Brinkman and A. J. Smyth 90 70260 859
19 On irrigation efficiencies. M. G. Bos and J. Nugteren 90 70260 875
20 Discharge measurements structures. (3rd edition) M G Bos (ed.) 9070754150
21 Optimum use of water resources. N. .A. de Ridder and A. Erez -
23 Land evaluation for agricultural development. K. J. Beek -
24 Drainage and reclamation of salt-affected soils. J. Martinez Beltrn -
25 Proceedings of the International Drainage Work- J. Wesseling (ed.) 9070260549
shop
26 Framework for regional planning in developing coun- J. M. van Staveren and 90 70260 832
tries D. B. W. M. van Dusseldorp.
27 Land reclamation and water management. - 90 70260 689
28 Proceedings of the Workshop on Land Evaluation for P. Laban (ed ) 9070260689
Forestry
29 Numerical modelling of groundwater basins: A user- J. Boonstra and 90 70260 697
oriented manual N. A. de Ridder
30 Proceedings of the Symposium on Peat Lands Below H. de Bakker and 90 70260 700
Sea Level. M. W. van den Berg
31 Proceedings of the Bangkok Symposium an Acid Sul- H.Dost and 9070260719
phate Soils. N. Breeman (eds.)
32 Monitoring and evaluation of agricultural change. Josette Murphy and 90 70260 743
Leendert H. Sprey
33 Introduction to farm surveys. Josette Murphy and 90 70260 735
Leendert H. Sprey
34 Evaluation permanente du dveloppement agricole. Josette Murphy and 90 70260 89 1
Leendert H.Sprey
35 Introduction aux enqutes agricoles en Afrique. Josette Murphy and 90 70260 956
Leendert H. Sprey
36 Proceedings of the International Workshop on Land W. Siderius (ed.) 9070260948
Evaluation for Extensive Grazing (LEEG).
37 Proceedings of the ISSS Symposium on Water and J. Bouma, P. A. C. Raats 90 72060 972
solute movement in heavy clay soils. (ed.)
38 Aforadores de caudal para canales abiertos. M. G. Bos, J. A. Replogle 9070260921
and A. J. Clemmens
39 Acid Sulphate Soils: A baseline for research and devel- D. Dent 90 70260 980
opment.
40 Land evaluation for land-use planning and conserva- W. Siderius (ed.) 90 70260 999
tion in sloping areas.
41 Research on water management of rice fields in the S. EL. Guindy & 9070754088
Nile Delta, Egypt. I. A. Risseeuw;

400
~

42 Proceedings, Symposium 25th International J. Vos (ed.) 90 70754 I 1 8

Course on Land Drainage
43 BASCAD: A Mathematical Model for Level Basin J. Boonstra & 90 70754 12 6
Irrigation M. Jurrins
44 Selected Papers of the Dakar Symposium on Acid H. Dost (ed.) 9070754134
Sulphate Soils
45 Health and Irrigation (vol. 11) J. M. V. Oomen, J. de Wolf 90 70754 17 7
and W. R. Jobin

No. Bulletins Author ISBN No.

. ~ .-

1 The auger hole method. W. F. van Beers 90 70260 816

4 On the calcium carbonate content of young marine B. Verhoeven -
sediments.
6 Mud transport studiesin coastal watcr from the A. J . de Groot
Western Scheldt to the Danish frontier.
Some nomographs for the calculation of drain spac- W. F. J. van Beers -
8
ings.
9 The Managil South-Western Extension: An extension D. J. Shaw -

to the Gezira Scheme.

IO A viscous fluid model for demonstration of ground- F. Homma 90 70260 824
water flow to parallel drains.
I1 Analysis and evaluation of pumping test data. G. P. Kruseman and 90 70260 808
N. A. de Ridder
1Is Anlisis y evaluacin de los datos de ensayos por G . P. Kruseman and -

bombeo. N. A. de Ridder
I IF Interprktation et discussion des pompages dessai. G. P.Kruseman and -
N.A. de Ridder
I2 Gypsifereous Soils. J . G. van Alphen and -
F. de los Rios Romero
13 Groundwater hydraulics of extensive aquifers. J. H. Edelman 90 70260 794

No. Bibliographies

7 Agricultural extension in developing countries. C. A. de Vries -

8 Bibliography on cotton irrigation. C. J. Brouwer and -
L.F. Abell
9 Annotated bibliography on surface irrigation meth- S. Raadsma, G. Schrale -
ods.
IO Soil Survey interpretation. R. H. Brook -
13 Abstract journals on irrigation, drainage and water L.F. Abell -
resources engineering.
18 Drainage: An annotated guide to books and journals. G. Naber 90 70260 93 X
. .

Other publications

Papers International Symposium. 9070260751

Polders of the World (3 volumes). 76X and 77 8
Final Report Symposium Polders of the World. -
Proceedings Symposium Lowland Development in Indonesia. 90 70754 07 X

401