^L.

FOREST RESEARCH CP-JTER

^^ß</4k ''^ t'entrai Eorerence File

l_l^3i^S '

the Sllf stilling basin

A STRUCTURE TO DISSIPATE THE

DESTRUCTIVE ENERGY

IN HIGH-VELOCITY FLOW

FROM SPILLWAYS

Agriculture Handbook No. 156

Agricultural Research Service

In Cooperation with the

Minnesota Agricultural Experiment Station

and the Saint Anthony Falls Hydraulic Laboratory

UNITED STATES DEPARTMENT OF AGRICULTURE

 PREFACE

This publication is a condensed report of the research that led to the

development of the St. Anthony Falls (SAF) stilling basin. It is prepared
especially for the use of those who have occasion to design this efficient and
economical outlet structure for dissipating the destructive energy in the
high-velocity flow at the exit end of chutes, dams, closed conduit spillways,
and similar structures.
The experimental work begun in January 1941 was completed in De-
cember 1943. The results of the tests were first reported in a processed
publication in December 1943 that was revised in May 1949 (SCS-TP-79).
A detailed report on the research has been published in the Transactions
of the American Society of Civil Engineers, volume 113, 1948, '^Develop-
ment and Hydraulic Design, Saint Anthony Falls Stilling Basin.''
This cooperative V study in the solution of problems concerning the
hydraulics of soil and water conservation structures was made by the staff
of the Soil and Water Conservation Research Division, Agricultural Re-
search Service, U. S. Department of Agriculture, in cooperation with the
Saint Anthony Falls Hydraulic Laboratory and the University of Min-
nesota Agricultural Experiment Station.

CONTENTS

Page Page
Preface 2 Test Results—Continued
The Problem 3 Sidewall Height 10
Previous Work 3 Wingwalls 10
The Test Program 3 Shape of Basin 10
Laboratory Facilities and Test Methods. _ 4 Cutoff Wall 10
Hydraulic Jump 4 Effect of Entrained Air 11
Test Results 5 Conclusions 11
Length of Basin_ 5 Application of Results 11
Chute Blocks 6 Design Chart 11
Floor Blocks 6 Solution of a Typical Problem 12
Force on Floor Blocks 6 Field Experience 12
EndSiU 7 Summary 14
Tailwater Depth 7 Literature Cited 14

Washington, D. C. Issued April 1959

For sale by the Superintendent of Documents, U. S. Government Printing Office

Washington 25, D. C. - Price 10 cents
the SflF stilling basin
a structure to dissipate
the destructive energy in high-velocity
flow from spillways
By FRED W. BLAISDELL, hydraulic engineer, Soil and Water Conservation Research Division, Agricultural Research Service

THE PROBLEM jet enters the pool above its bottom. This type
is, therefore, somewhat more efficient than the
The research summarized in this pubhcation is simple stilling basin. Although both the Stanley
a direct result of the need for a stiUing basin to and the Schoklitsch stilling basins are undoubtedly
dissipate the energy in the high-velocity discharge satisfactory in dissipating energy, a smaller and
from culverts, chutes, and other types of spillway. more economical stilling basin was needed,
^ Engineering literature abounds with descrip-
tions of stilling basins located at dams throughout THE TEST PROGRAM
the world. Each structure, however, had been
individually designed for a specific location. Ad- Exploratory tests were made on the hydraulic
ditional studies were required to adapt a design jump, the Schoklitsch, and other published de-
to other locations. signs of stilling basins. Analytical studies were
When the SAF stilling basin study was initiated also made of several other stilling basin designs.
in 1941, at the request of the United States Soil As a result of these preliminary studies, some
Conservation Service, little had been accomplished stilling basins were eliminated from further con-
toward the development of a universal design. sideration because of their inferior performance
In general, the structures built by the Soil Con- in dissipating energy while others were eliminated
servation Service are of such size that few of because their size and cost for equivalent per-
them can economically justify the individual formance were greater than for the more efficient
model studies that proved so profitable in the stilling basin.
development of stilling basins for large dams. On the basis of the exploratory tests, the
It was essential, therefore, that an eflScient and rectangular stilling basin, developed by the United
economical stilling basin be developed and that States Bureau of Reclamation and described by
design rules be formulated so future stilling Warnock {19), was selected for further study.
basins could be designed without recourse to This stilling basin had chute and floor blocks to
further model studies. dissipate the energy and an end sill to deflect
the stream away from the bed. The length of
PREVIOUS WORK the stilling basin was 75 percent and the depth
Sb percent of the hydraulic jump length and
A study of the literature on stilling basins depth, respectively, but the indications were
carried out in 1941 revealed only two investiga- that the size could be reduced still further.
tions leading to the development of generalized Using this basic form of stilling basin, studies
stilling basin designs. To the writer's knowledge, were directed toward determining the minimum
the only additional generalized studies that have dimensions for efficient energy dissipation and
been published to July 1958 are those by Bradley the laws governing the design of the various
and Peterka {5-11)} In the simple stilling basin elements making up the basin.
studied by Stanley {18), the energy in the high- The test program was divided into three parts:
velocity flow is absorbed in a pool formed by a (1) The culvert-outlet series in which the basin
sill or low dam. The Schoklitsch energy dis- proportions were determined for a narrow range
sipator {17) is similar to the simple stilling basin of the Froude number; (2) the flume-outlet series
in that an end. sill is used to form a pool but the of check tests, which covered a large range of
the Froude number; and (3) the turbine-room
^ Italic numbers in parentheses refer to Literature series of large-scale check tests. The results of
Cited, p. 14. these tests are discussed in this publication.
LABORATORY FACILITIES AND TEST elevations were measured with point gages located
on traveling carriages. Centerline profiles of the
METHODS water surface and eroded sand bed were sketched
All experiments on the SAF stilling basin were on data sheets with the aid of a grid of uniformly
made at the St. Anthony Falls Hydraulic Labora- spaced strings placed against the glass observation
tory of the University of Minnesota, Minneapolis. panels.
The laboratory is located on Hennepin Island at The stream bed downstream from the stilling
St. Anthony Falls in the Mississippi Kiver. Up basin was formed in concrete sand. The effec-
to 300 c. f. s. can be diverted from the river above tiveness of each arrangem^ent of stilling basin in
the falls and returned to the river below the falls removing the destructive energy in the water was
after dropping about 50 feet through the labora- measured by the erosion of this sand bed. The
tory. Water for the various experiments is ob- loose sand was scoured to its approximate ulti-
tained through pipes connected to a supply canal mate depth in 30 minutes, and this length of run
running the full length of the laboratory. was used for the culvert and flume series of
The culvert and flume series of tests were con- tests. Two-hour runs were found to be more
ducted in a channel 24 inches deep, 18 inches satisfactory for the large-scale turbine-room series.
wide, and 8 feet long; the turbine-room series in Before beginning each experiment, the stream
a channel 6 feet wide and 24 feet long. Since it bed was filled in with sand and a small stream of
was impossible to observe from above the opera- water was used to fill the channel without eroding
tion of the stilling basin under the jet and 'Vhite the stream bed. At the beginning of the run
water/' all tests were conducted on half-models; the valve was opened quickly to give the desired
that is, the models were split along their centerline discharge and at the end of the run the valve was
and one-half of the model was pressed against a closed quickly. Data on the discharge, tailwater
glass plate through which the action of the stilling level, and water-surface profile were obtained
basin could be observed. Check tests showed that during the run. Photographs were made during
identical results can be obtained from either full many runs. After the run the sand bed was
or half models. A glass observation panel 8 feet drained and the erosion was recorded.
long was located on one side of the 18-inch channel The procedure used in determining the best
and a panel 2 feet high and 12 feet long on the proportions of the various elements making up the
side of the 6-foot channel; the model centerlines stilling basin was to run a group of tests and make
were located along the face of the panels. changes on only one element between each test,
Water for the culvert and flume series was so tnat any differences observed in tue perform-
obtained through a 4-inch pipe and the discharge ance of tlie stilling basin could be attributed to
controlled by a 3-inch gate valve. The rate of tJie change in the pertinent element. This single
flow was measured by a calibrated 1-foot type element was varied in shape, size, or position
IIS flume. Water for the turbine-room series until the best proportions and location had been
was obtained through an 18-inch pipeline tnat ascertained. Other elements were then studied in
was reduced to 12 inches before reaching the model. like manner until the best proportions of all parts
The rate of flow was controlled by a 12-inch gate of the stilling basin had been tentatively deter-
valve and measured by the calibrated pressure mined. Because of the interdependency of the
difference across the 18-inch by 12-inch reduction, various elements, it was necessary to repeat
by a calibrated 1.5-foot type H flume, or by a certain steps to ascertain the effect of subsequent
Pitot tube located at the stilling basin entrance. changes on elements studied previously.
For the culvert series of tests, the approach to After ascertaining the most satisfactory stilling
the models was a 3-inch square pipe. The depth basin proportions for a single rate of flow, the
of flow at the entrance to the stilling basin was dimensions were varied for other rates of flow to
determined by the discharge and the width of the determine the laws governing the proportions of
open channel transition used between the pipe the stilling basin and its elements. All these tests
and the stilling basin. The depth at the stilling were part of the culvert-outlet series. Both the
basin entrance for the flume and turbine-room flume-outlet and turbine-room series were check
series was set by means of adjustable gates located tests. The only important revisions in the design
just upstream from the models. For these series resulting from the tests with other rates of flow
it was possible to regulate both the depth and were in the end sill height and the wingwall
velocity at the stilling basin entrance. The shape and position. All changes indicated as a
approach channel width was 3 inches for the result of the check tests were made and verified.
flume series and 1 foot for the turbine-room series.
All models were made of wood—waterproof HYDRAULIC JUMP
plywood for the smaller models and pine for the
larger models. All the dimensions of the SAF stilling basin
The tailwater depth was controlled by stop logs are related, either directly or indirectly, to the
located in both channels at the downstream end hydraulic jump. The theoretical equation for the
of the test section. Depths and sand-contour hydraulic jump is
 dl Length of Basin
4 The full length of the rectangular stilling basin
described by Warnock (19) was not utilized in
where ¿2 is the depth after the jump, di the depth dissipating the energy in the water. (The length
preceding the jump, Vi the velocity preceding the of this basin, LB, was 75 percent of the hydraulic
jump, and g the acceleration due to gravity (32.2 jump length, or, LB=0.75 X 5^2=3.75^2-) The
feet per second per second). The derivation of stilling basin was shortened in steps until a
this equation can be found in most books on minimum length equal to 0.70¿2 was reached.
hydraulics; for example, the '^Handbook of Surprisingly, the depth of the scour hole was not
Hydrauhcs'' (15, pp. 8-23 to 28). Numerous increased by this shortening until a length of
experiments by others have proved the validity stilling basin less than I.25S2 was tested. The
of this equation, which can be simplified to channel erosion was markedly but not dangerously
increased when i^—1.00¿2. When the basin
dr length was 0.70^2? the scour at the end of the
d2=~ (-l+VSf+l) (2)^
stilling basin as well as in the downstream channel
was considered excessive and the energy dissipa-
where the Froude number F is given by the equa- tion in the stilling basin was poor. A stilling
tion basin length of 1.25¿2 was used in subsequent
tests in which the positions and sizes of the other
F=IJ (1) elements making up the basin were investigated.
The Froude number was about 30 for tests up
This dimensionless number (F), a constant to this point.
for similar flow conditions in the model and the Further study of tlie stilling basin length was
prototype, is also used in the determination of initiated after tentatively determining the best
the size of the stilling basin. sizes arid locations of the chute and floor blocks
The length of the hydraulic jump is assumed to and trie end sill. Triese tests covered a range of
be 5¿2j after Bakhmeteff and Matzke (1).^ trie Froude number from 3 to 150 and were part
of both trie culvert and flume-outlet series. As a
result of these tests, it was discovered that the
TEST RESULTS stilling basin was too short for Froude numbers
less than 30 and longer than necessary for larger
The SAF stilling basin design was developed Froude numbers. The stilling basin length,
and verified as a result of 271 tests. The number therefore, was varied for each of 12 values of F
of tests in each series and the range of the vari- until the best length was determined. The
ables are given in table 1, where Qis the discharge, performance of each length of basin was ^^rated"
R=Vidilv is the Reynolds number, and v is and plotted, with ÍB/¿2 and F as coordinates, and
the kinematic viscosity. The tests made on a curve drawn through the plotted points. Both
each element comprising the SAF stilling basin the experience obtained during the experiments
will be discussed separately. and the plotted data were used in locating this
curve. The equation of this curve,
2 Numbers in parentheses opposite the equations refer
to the equations listed on the design chart, pp. 8 and 9.
^ Bradley and Peterka (5, 6) show that the hydraulic ¿B_4.5
jump length varies with the Froude number, reaching a d2 "F^-^s
(4)
maximum length, of about Q.ld2. However, this difference,
in the length has no effect on the SAF stilling basin is suggested as giving a minimum safe length of
because the hydraulic jump length does not enter directly
into the SAF design, and the tests on the SAF stilling stilling basin; it is conservative, but not to the
basin covered the practical range of Froude numbers. point where the material in the outlet is wasted.

TABLE 1.—Tests of SAF stilling basin and range of test variables

Series Tests Q Fi d. C?2 F R X 10-3

Number C.f.s. F. p. s. Ft. Ft.

Culvert outlet 100 0. 09 to 0. 4 4. 3 to 12 0. 04 to 0. 17 a 17 to 0. 8 3 to 57 12. 7 to 45
Flume outlet 108 . 04 to .8 2. 8 to 22 . 05 to . 15 . 13 to 1. 8 5 to 200 14. 2 to 237
Turbine room _ 66 . 40 to 21. 9. 7 to 44 . 03 to 1. 27 . 49 to 5. 5 7 to 288 40. 6 to 2, 100
Total tests and 274 . 04 to 21. 2. 8 to 44 . 03 to 1. 27 . 13 to 5. 5 3 to 288 12, 7 to 2, 100
total range in
variabhîs.
 Equation 4 was developed for a range of the The width and spacing of the floor blocks should
Froude number from 3 to 150, but it was later be the same as for the chute blocks. However,
used to design experimental stilling basins having for those stilling basins in which the sidewalls
values of F as high as 300. The results of all diverge in plan, the width and spacing of the floor
subsequent tests show that stilling basin lengths blocks should be increased over the chute block
determined from equation 4 are satisfactory. width and spacing to compensate for the greater
stilling basin width at the floor block location.
Chute Blocks No floor block should be located closer to the
stilling basin sidewall than 3(¿i/8. Floor blocks
The chute blocks, located at the entrance to the located closer cause a high boil that might overtop
stilling basin, serve to increase the effective depth of the sidewall.
the entering stream, break the stream up into a num- Insufficient water can pass between the floor
ber of small jets, and help create the turbulence blocks if they occupy too much of the stilling
required for effective energy dissipation. basin width; they then act more like a sill than
The original height of the chute blocks was ¿i, like individual blocks. The test results show that
and the width and spacing, 0.75ái. A test on a satisfactory conditions exist when the floor blocks
solid chute block, such as is used in the Schoklitsch occupy between 40 and 55 percent of the stilling
energy dissipator, showed that less energy was basin width. The aggregate width of all floor
dissipated in the stilling basin and that flow condi- blocks, therefore, should be held within these
tions in the channel downstream from tiie stilling limits, even if it is necessary to reduce the width
basin were not so good. A second test was made of the floor blocks to do so.
in which the tops of the chute blocks were sloped The floor blocks always should be placed down-
to direct the jets at the floor blocks. The result stream from the openings in the chute blocks to
of this change was to increase the depth of erosion break up the jets issuing from between the chute
near the end of the stilling basin. blocks and passing along the stilling basin floor.
The chute blocks used for all subsequent tests A single test made with the floor blocks in line
had a height of di and a width and spacing of with the chute blocks was sufficient to show the
0.75¿i. These proportions proved to be entirely inferiority of this arrangement.
satisfactory. It makes no difference in the per- The floor blocks may be piers square in plan
formance of the stilling basin whether a chute with vertical faces, or their downstream faces may
block or a space is next to the sidewall as long as the slope as shown on the design chart.
blocks are symmetrical about the centerline of the
outlet. Force on Floor Blocks
Floor Blocks A knowledge of the forces exerted on the floor
blocks is necessary for their structural design.
Energy is removed from the water by impact No tests were made to determine these forces.
against the floor blocks and considerable turbu- It is possible, however, to compute the maximum
lence is created by them. probable forces, and the results of experiments by
The first tests on the floor blocks were made to others are available to modify these computed
determine their best longitudinal position. These values.
tests show that it is equally as bad to have the The impact force on the floor blocks required
distance between the chute and floor blocks too to turn the flow 90"^ is given by the equation
short as it is to have tJae distance between the floor F=AVi'w¡gj where F is the total force, A is the
blocks and end sill too short. If the distance area of the face of the blocks, and w is the unit
between the chute and floor blocks is too short, the weight of water (62.5 pounds per cubic foot). It is
blocks act like a solid chute block. If the distance convenient to write the impact force in terms of
between the floor blocks and end sill is too short, (¿1 and F, since both of these values are required
the blocks and sill act as a unit in deflecting the in the design of the stilling basin. The force
jet upward. per unit width of the floor block,/, is
Nearly identical results were obtained when the
floor blocks were located X^/S and is/2 from the f=wdi^F
upstream end of the stilling basin. The results
for the Í5/3 spacing were slightly better, but the This equation gives the maximum impact force
difference probably is insignificant. The floor on the floor blocks per unit width of block. The
blocks were located is/3 from the upstream end equation assumes that all the water approaching
of the basin for all subsequent tests. No reason the block is turned at right angles to its original
was discovered for changing their longitudinal direction. Much of the water changes direction
location as a result of these tests. only slightly, so the actual force must be con-
Floor blocks were tried with heights both siderably less than the computed maximum.
greater and less than di. This height of floor Other factors that influence the force on the floor
block was either as good as or better than greater blocks are the shape, width, and spacing of the
and lesser heights. Accordingly, a floor block blocks, the effect of the chute blocks, and the fact
height equal to di was used for subsequent tests. that the mean velocity at the floor blocks is re-
duced by the chute blocks and the roller. Such nolds number ß, a well-defined curve was obtained,
information as is available regarding the effect although there is no reason to believe that such a
of these factors will be presented. relationship should exist. A study of tñe equation
The forces on stepped blocks and streamlined for this curve showed that the height of the end
blocks measured experimentally at the Massa- sill was unbelievably low for values of R within
chusetts Institute of Technology have been re- the practical range. This naturally cast suspicion
ported by Harleman {llf). The stepped blocks on the form of the end-sill height equation and led
approximate the shape of the vertical-faced blocks to the turbine-room tests, which were made at
used in the SAF stilling basin. Harleman states: higher Reynolds numbers.
'The maximum force exerted by the baffle piers The turbine-room tests confirmed the opinion
is of the order of 20 percent of the pressure force that the end-sill height was too low for the higher
due to the downstream depth.'' Since the down- values of R and indicated that c\d<¿ was independ-
stream force is applied across the full width of ent of both F and R. A tentative equation (c=
the stilling basin and the stepped blocks occupied 0.07(¿2) was derived early in the test program.
50 percent of the basin width, the maximum This equation was checked by other tests until it
measured force per unit width of blocks is 40 became apparent that this end-sill height would be
percent of the pressure force per unit width due to satisfactory. Subsequent tests were made to
the downstream depth. Using this latter figure, verify this equation.
it is found that the maximum force per unit width A review of the data obtained during the flume-
of block varies from 27 percent of the theoretical outlet series of tests shows that satisfactory erosion
value for F=3 to 38 percent of the theoretical conditions were obtained when c=0.07¿2. The
force for F-=300. higher end sills given by the equation containing
Unpublished results of tests made at the St. R produced slightly better erosion patterns for
Anthony Falls Hydraulic Laboratory in connec- the lower values of ß, but the difference is so small
tion with a model study of the Chippewa River as not to warrant the use of separate equations.
Reservoir Dam of the Northern States Power The recommended equation for the height of the
Co. substantiate the MIT values. Piezometric end sill is
pressures were measured on the face of a baffle c=0.07¿2 (6)
pier located below a Tainter gate, and the pres-
sures were integrated to determine the total force
on the baffle. Forces determined for three dif- Taîlwatcr Depth
ferent rates of flow amounted to 43 percent, 24 Use of the blocks and end sill in the stilling basin
percent, and 27 percent of the impact forces permits the depth of the tailwater above the stilling
computed as outlined above. basin floor level to be decreased over that for the
In view of these data and until better informa- theoretical tailwater depth of the hydraulic jump.
tion is available, it is suggested that the force Warnock {19) recommended a 15-percent reduc-
on the floor blocks exerted by the approaching tion, so that the actual tailwater depth, ¿'2, would
stream be taken as 40 percent of the computed be 0.85(¿2. Tests were made to check this figure.
impact force, ov If the tailwater depth is too low, the roller on
the hydraulic jump will be washed out of the still-
f=OA wdi^F=2b d^F ing basin and the floor blocks and end sill will
simply deflect the stream and break it up. The
End Sill broken-up stream will land on the water surface
The end sill, located at the downstream end of some distance downstream from the outlet and
the stilling basin, deflects the bottom currents erode the bed only slightly, with the depth of the
upward and away from the stream bed. In scour hole near the end of the stilling basin remain-
addition, a ground roller is created under the ing unchanged. The structure will not be endan-
deflected stream, which brings bed material from gered tJaereby, but the energy dissipation in the
downstream and deposits it at the end of the stilling basin is poor and the spray may prove
stilling basin. objectionable.
The height of the end sill, c, for the culvert- In determining the minimum permissible value
outlet series was made ¿2/7. This end-sill height of (¿'27 the tailwater depth was decreased until the
proved to be satisfactory for the narrow range of roller was washed out of the stilling basin. The
the Froude number for which it was derived, but relative tailwater depth at which this occurred
the equation was inadequate for a larger range was plotted against F. Originally the relative
of the Froude number. tailwater depth was assumed to be dWd<i, When a
A thorough study of the end-sill height for paper {ß) describing development and hydraulic
values of F from 5 to 200 was made as part of the design of the SAF stilling basin was published,
flume-outlet series of tests. The best height of however, one of the discussers showed that plotting
end sill for each set of otherwise constant condi- the ratio d^d^ permitted the derivation of a rela-
tions was selected, and the selected values were tionship that is continuous over the range of
found to be independent of the Froude number. Froude numbers covered by the tests. As a
However, when ¿¡d^ was plotted against the Rey- result of this discovery, dWdx was plotted against
 A. SOLUTION OF EQUATIONS (l),(3) AND (4) B. SOLUTION OF EQUATION (2)
Procedure. Equation (I) df to Vf to F Procedure iFtodf to dg
Equation (3) df to F to d2
Equation (4) F to d2to Lg

4\ï
sil

dg in feet I 2 3 4 5 is 8 to 20 30
\iW^ Vri'iT|lM^avi''.'/V|1ihi|iiliii,'|.Vi|i'i'i''i'i'l'i'i'.'i''i'i''
c in feet .07 OJ 0.2 0.3 0.4 06 0.8 I 2

C. SOLUTION OF EQUATIONS (5) AND (6)

Procedure : Equation (5) d2 to z
Equation (6)d2 to c

D. EXAMPLE E.DESIGN EQUATIONS

GiVEN: d, = 0.6 ft and v, = 40 ft. per sec. d,
(I) F.- -^ (2) dg -
SOLUTIONS OE EQUATIONS gd, 2 ("H-^eFi-1)
(using curves and nomograptis)
4.5dg
(I) F = 63 (2) dg =7.4 ff (3) d¡= 6.1 ft. (3)d'^=: l.4d,F' (4) Lc

(4) LB=6.3ft (5)z= 2.48 ft. (6) 0.52 ft. (5) z= dg/3 (6) c -= 0.07 dg

FIGURE 1 ^ DESIGN CHART for SAF STILLING BASIN

Agricultural Research Service, U. S« Department of Agriculture, in cooperation with the Minnesota
Agricultural Exipefiment Station atttl the St* Anthony Falls Hydraulic Laboratory, University of Minnesota.
8
RECTANGULAR STILLING BASIN
HALF - PLAN

O» to 90*
45*Prtftrrtd

TRAPEZOIDAL STILLING BASIN

HALF- PLAN

CENTERLINE SECTION

F. PROPORTIONS OF THE SAF STILLING BASIN

6. DEFINITIONS OF SYMBOLS
Bi " width of stilling basin at upstream end, in feet above stilling basin floor, in feet

Bg " width of stilling basin at floor blocks, in feet D' - side wall divergence, D' longitudinal to I transverse

Bj - width of stilling basin at downstream end, in feet F - the Froude number

c - height of end sill, in feet g - acceleration due to gravity, in feet per second per second
df - depth of flow at entrance to stilling basin, in feet Lß - length of stilling basin, in feet

d¿ - downstream depth computed by momentum equation ^/ " velocity at entrance to stilling basin, in feet per second
for the hydraulic Jump, in feet ' " ^^'9^* of stilling basin side walls above maximum

dg - water surface elevation in downstream channel tailwater level, in feet

F and an enveloping curve drawn above the rela- sidewall height); or (2) a wingwall of triangular
tive depths at which the roller is washed out of shape in downstream elevation, the top having a
the stilling basin. The equation of this curve is slope of 1:1. The triangular shape of wall is
recommended, because it is equally as satisfactory
in preventing scour as is the extended wingwall
(3) and, in addition, requires less material.
Wingwalls have been customarily located per-
Sidewall Hcighf pendicular to the centerline of the outlet structure.
Tests have shown, however, that the wingwalls
The flow in the stilling basin is very turbulent, may be extended parallel to the basin centerline
and, as a result, the water surface is so rough that if field conditions make it necessary to do so,
some freeboard above tailwater level is necessary although the boil height is considerably higher.
if overtopping of the sidewalls is to be prevented. Nevertheless, the best overall conditions are
In addition to the surface roughness, a standing obtained if the triangular wingwalls are located
wave, or boil, is caused by the floor blocks and at an angle of about 45° to the outlet centerline.
end sill, which in itself requires freeboard above Subsequent tests of other types of stilling basins
the tailwater level. For Froude numbers less have confirmed the superior performance of the
than about 20, the crest of the boil is in the wingwall having a 1:1 top slope located at an
stilling basin, whereas, for higher Froude numbers, angle of 45° to the outlet centerline (4, 12, IS).
the boil crest occurs downstream from the end of
the basin and its full height need not be considered Shope of Basin
in designing the sidewall height. The size of the stilling basin varies with the
Average profiles of the water surface in the initial flow depth if Vi does not change; any reduc-
stilling basin were obtained for all series, but the tion in dx will reduce (¿2, the length of the basin,
maximum height of splash was obtained for only the height of the sidewalls, and the depth of
the turbine-room series. It is from these latter excavation. In addition, a larger percentage of
tests that the height of the sidewall is determined. the energy in the water entering the stilling basin
The maximum height of the splash Zraax in the will be dissipated. A saving in overall cost of the
stilling basin was divided by ¿2. There is con- outlet will ordinarily be possible if a flaring-
siderable scatter to the data, but Zmaxld2 is appar- sidewall transition is placed between the culvert
ently independent of F. The range of 2m,axld2 is or chute and the stilling basin to accomplish this
from —0.02 to 0.31. A study of the data shows reduction in di (3). In those cases where a
that if the height of the stilling basin sidewall, 0, transition is used, the diverging transition side-
above the maximum tailwater level is given by walls should be extended to form the stilling basin
the equation walls. The resulting stilling basin is trapezoidal
z=d2l^, (5) in plan, as is shown on the design chart, page 8.
A few tests were made on a trapezoidal-shaped
the freeboard will be sufficient to keep the splash stilling basin in the culvert-outlet series. The
in the stilling basin. Because of the scatter in stilling basin was designed for flow conditions at
the data, the freeboard provided by this equation its entrance. The width and spacing of the floor
will, in some cases, be greater than is necessary to blocks were multiplied by the ratio B2/B1 to com-
protect fully the structure, but the safety factor pensate for the increase in the width of the stilling
is not excessive for the average case. basin at their location. All blocks had their axes
parallel to the centerline of the basin. Flow
Wingwalls conditions in the downstream channel were some-
Wingwalls at the end of the stilling basin are what improved through the use of the trapezoidal
used as retaining walls to hold back the earth stilling basin. This is because the velocity of the
fill. The ordinarily used wingwall is rectangular flow was lower at the exit from the basin, and the
in downstream elevation. Since the scour around widening of the stream to fill the downstream
the end of this wall is severe, other wingwall shapes channel reduced the size of the eddies along the
were investigated. channel sides near the stilling basin.
The principal cause of the scour around the end
of the wingwall is an eddy along each side of the Cutoff Wall
downstream channel that is driven by the stream A cutoff wall is used at the end of the stilling
leaving the stilling basin. It is imperative that basin to prevent scour from undermining the
the concentration of the flow from this eddy be basin. Obviously, the depth of the cutoff wall
kept off the stream bed. Two methods can be must be greater than the maximum depth of ero-
used to prevent the eddy from attacking the sion at the end of the stilling basin.
stream bed: (1) A submerged extension of the Serious erosion near the end of the stilling basin
wingwall, having a height equal to half the tail- is prevented by the end sill, which deflects upward
water depth and a length equal to 0.6 of the side- the stream leaving the basin. A ground roller
wall height (the minimum length of rectangular under the deflected stream brings material up-
wingwall used in the experiments was 0.4 of the stream and further aids in preventing erosion.
10
In the laboratory, the scour sometimes reached an e. The widths and spacings of the floor
elevation slightly below the floor of the stilling blocks for diverging stilling basins
basin, but the scour never reached a depth at the should be increased in proportion to
end of the basin greater than the thickness of a the increase in stilling basin width
floor slab that might be used. Therefore, a cut- at the floor block location.
off wall of only nominal depth need be used at the f. The floor blocks may be piers square
end of the stilling basin. in plan with vertical faces, or their
downstream faces may slope as shown
Effect of Entrained Air on the design chart.
g. The force per foot width exerted on
Air is ordinarily entrained by the water flowing the floor blocks by the approaching
in chutes laid on a steep slope. This results in a stream may be taken as
greatly increased depth of flow of the mixture.
However, no air was naturally entrained by the
water during the model tests because of the low f=25di'F
velocities or the short length of channel. Because
entrained air may afi^ect the performance of the 4. The height of end sill is
stilling basin, a few tests were made in which from
10 to 117 percent of air was mixed with the water. c=0.07d2 (6)
The stilling basins were designed as if the water 5. The depth of the tailwater above the stilling
were free of air, and duplicate tests were run both
with and without air entrainment. Identical basin floor is
results, within the limits of experimental preci-
sion, were obtained from the duplicate tests. d'2==lAF'-'% (3)
Although di is greater when air entrainment
occurs, ¿2 remains unchanged, since the air sepa- 6. The height of the sidewall above the maxi-
rates from the water, owing to the lower velocities mum tailwater depth to be expected during
in the downstream channel. No increase in side- the life of the structure is
wall height is required as a result of air entrain-
ment. z=d2l^ (5)
The results of these tests show that the effect of
air entrainment can be neglected in the design 7. Wingwalls should be equal in height and
of the SAF stilling basin. The resulting stnicture length to the stilling basin sidewalls. The
will safely handle any flows in which air is en- top of the wingwall should have a 1:1 slope.
Wingwalls flaring at 45° with the outlet
trained. centerline are preferred to wingwalls that
are perpendicular or parallel to the center-
CONCLUSIONS line.
8. The stilling basin sidewalls may be parallel
The following conclusions are reached as a (rectangular stilling basin) or diverge as an
result of the tests made to develop and verify the extension of the transition sidewalls (trape-
SAF stilling basin design : zoidal stilling basin).
1. The length of the stilling basin for Froude 9. A cutoff wall of nominal depth should be
numbers between 3 and 300 is used at the end of the stilling basin.
10. The effect of entrained air should be neg-
LB = 4:M2IF' (4) lected in the design of the stilling basin.
During the tests it was noticed that the per-
2. The height of the chute blocks and the floor formance of the SAF stilling basin was excellent
blocks is ¿i; their width and spacing are at discharges less than the design discharge. At
approximately 3di/4; either a chute block the design flow the SAF stilling basin provides an
or a space may be located next to the side- economical method of dissipating energy and
wall if the blocks and spaces are sym- preventing dangerous stream bed erosion.
metrical about the outlet centerline.
3. The floor block criteria are as follows;
a. The distance from the upstream end APPLICATION OF RESULTS
of the stilling basin to the floor blocks
is LB/S, Design Chart
b. No floor block should be placed closer
to the sidewall than 3i/i/8. The results of all the tests on the SAF stilling
c. The floor blocks should be placed basin are summarized on the design chart for the
downstream from the openings be- SAF stilling basin (fig. 1).
tween the chute blocks. The proportions of a SAF stilling basin can be
d. The floor blocks should occupy be- determined from the chart without the aid of
tween 40 percent and 55 percent of any instrument or any other design chart or table.
the stilling basin width. The use of the design charts is explained thereon,
11
a typical problem is solved, and the principal difficulty can be overcome by making the block
dimensions determined. width and spacing 8 inches. The proportion then
The stilling basin dimensions obtained from the becomes 2x8/36=0.44. The total force on each
design charts will result in a good design. Slight block is 745x8/12 = 500 pounds.
variations in the dimensions, however, will have Another arrangement of the blocks is to make
little or no effect on the performance of the basin. them 6 inches wide as before, but to place half a
To simplify the construction, all odd dimensions chute block at each side of the chute and the
should be changed to even dimensions. two other equally spaced blocks between them.
Three equally spaced floor blocks can then be
used in the basin, one straddling the centerline
Solution of a Typical Problem and the others placed 6 inches on either side of
the center block. The block nearest to the sidewall
A rectangular SAF stilling basin is to be con- is therefore 3 inches from the sidewall. This is
structed at the end of a 3-foot wide chute. The greater than the allowable minimum of 0.6x%=
depth and velocity at the end of the chute are 0.225 foot=2% inches. The proportion of the
0.6 foot and 40 f. p. s., respectively, the design basin width occupied by the floor blocks is 3x%6 =
tailwater elevation is 377.0, and the maximum 0.50, a satisfactory figure. The total force on
tailwater elevation in the downstream channel each floor block is 745x%2===372 pounds.
anticipated during the lif-e of the structure is Arrangement of the blocks is up to the designer.
378.5 for the design discharge of 72 c. f. s. Either arrangement given above would be sat-
Reading the principal dimensions from the isfactory.
design charts it is found that: F=82.8, ¿2=7.43
feet, (Z'2=6.13 feet, iß=6.28 feet, 0=2.48 feet,
and c=0.520 foot. In order to simplify the con- Field Experience
struction, LB=Q feet 3 inches, and c=ñ or 7 The first SAF stilling basin was built in western
inches can be used without affecting the operation Iowa in 1944. Since that time a considerable
of the structure. The elevation of the top of the number of SAF stilling basins have been built.
sidewalk, which is determined from the maximum The exact number is unknown to the writer;
tailwater elevation, is 378.5+2.48 = 380.98; use publications describing the design of the SAF
381.00. The force on the floor blocks is 25x0.62x stilling basin are readily available for use by
82.8 = 745 pounds per foot of width. anyone without restriction, and there is no way
The elevation of the basin floor is 377.0—6.13 = to determine how many stilling basins have been
370.87. The tailwater level and required tafl- built according to the SAF design.
water depth also should be checked at discharges The writer has seen a number of SAF stilling
less than the design value to insure proper stilling basins and has had reports on the performance
action at all flows. Finally, consideration of the of other stilling basins. All reports received by
possibility that the channel bed elevation—and, the writer and all SAF stilling basins observed by
as a result, the tailwater level—^may become lower him have shown satisfactory performance. The
in time, suggests that the stilling basin floor be following field structures are known to have
set below the calculated elevation. The amount handled flows that approach the capacity for which
will depend upon local conditions and the judg- they were designed, so their performance will be
ment of the designer. The wingwall will have a described.
length of about 9 feet, depending on the sidewall The most thorough and complete test of the SAF
height, and its top a slope of 1 : 1. A cutoff wall stilling basin was that performed by William O.
under the stilling basin having a depth of 2 feet Ree (16) at the Stillwater (Okla.) Outdoor
or more should be used. Hydraulic Laboratory of the Agricultural Re-
Several arrangements of the 6- or 7-inch high search Service. Mr. Ree concluded (p, 13) :
chute and floor blocks are possible, the floor
Tests of the St. Anthony Falls Stilling Basin during
Till- 1 1 6 feet 3 inches ^ « ... . a 2-year period at the Stillwater Outdoor Hydraulic
blocks bemg placed =2 leet 1 mch,
Laboratory showed that the stilling basin was very
say 2 feet, downstream from the upper end of effective and completely satisfactory. Very little
scour of the channel bed occurred. It should be
the basin. noted, however, that the bed material at the point
The chute and floor blocks and the spaces of discharge was a rather firm clay. A sandy material
between them can be made 0.6x%=0.45 foot= might have shown a little different result.
5% inches, say 6 inches. This gives 36/6 = 6
spaces across the stilling basin. Now, locate chute Splash was not an important problem.
blocks 3 inches from either side of the chute and Figure 2 shows this stilling basin.
one straddling the centerline. Two floor blocks If someone is unduly concerned regarding
can now be located in the basin downstream from erosion in sandy material, figure 3 shows a SAF
spaces between the chute blocks. No floor blocks stilling basin at the exit of a 48-inch diameter
should be located next to the basin walls. The closed conduit spillway located at an airfield in
proportion of the basin width occupied by the northwestern Florida. The soil at the site of
floor blocks is 2x6/36=0.33. This proportion for this structure is a clayey sand. No scour of this
floor blocks is lower than is recommended. This readily erodible bed material is evident.

12
 In a monthly report, Glenn H. Baker, Soil
Conservation Service engineering specialist, corn-
men ted on a visit to Spruce Knob Lake, in W. Va.,
made in 1953, as follows:
At this visit I had an opportunity to observe the
performance of the SAF type stilling basin in opera-
tion. During the inspection the gate was completely
removed from the opening to the 26-inch diameter
drain which caused the maximum planned discharge.
The basin performed according to expectations almost
exactly as indicated by the model test shown in
Ohio. There was a minimum of erosion in the
channel below the dam, and the other engineers were
impressed with the operation of this type of structure.
The model referred to was one developed for
demonstration purposes in which the pipe had a
diameter of 1% inches—one-seventeenth the size
of the Spruce Knob Lake pipe.
The third structure that will be mentioned is
a SAF stilling basin at the end of a chute. This
spillway is located in Crawford County, Iowa.
On June 22, 1947, the storm runoff rate exceeded
the design capacity of the spillway by 50 percent,
this figure being based on information made
available to the writer by Floyd Nimmo, con-
struction engineer, through M. M. Gulp, Chief,
Design and Construction Branch, Engineering
Division, U. S. Soil Conservation Service. Figure
4 shows views of this structure taken before and
after the excessive storm of June 22, 1947. It is
readily apparent from these photographs that the FIGURE 2.—SAF stilling basin at Stillwater (Okla.)
SAF stilling basin gave excellent protection to Outdoor Hydraulic Laboratory: A, Drawing of basin;
the downstream channel despite the excessive B, with full-capacity flow.
flow that passed through it.

FIGURE 3.—SAF stilling basin at exit of 48-inch closed conduit spillway in northwestern Florida.

13
 SUMMARY
The stilling basin developed as a result of the
model studies at the St. Anthony Falls Hydraulic
Laboratory of the University of Minnesota has
become known as the "SAF stilling basin." It
has five distinct advantages :
1. The characteristics and proportions of the
stilling basin have been determined over a
wide range of conditions to be expected in the
field; the performance can be predicted
without making additional model studies.
2. The design procedure has been generalized.
3. The size of the stilling basin has been reduced
to the minimum that will assure protection
to the structure and prevent excessive erosion
in the downstream channel.
4. The SAF stilling basin is very economical to
construct.
5. Use of the SAF stilling basin under actual
field conditions has demonstrated its effec-
tiveness and has verified the predictions
based on the laboratory tests.
A design chart, giving the proportions of the
SAF stilling basin and the design equations and
graphical solution of these equations, is presented
FIGURE 4.—Box inlet chute spillway and SAF stilling
basin in Crawford County, Iowa: A, Before storm of on the center fold, pages 8 and 9.
June 22, 1947; B, after storm of June 22, when storm
runoff rate exceeded design capacity by 50 percent.

E. I. Eowland, Arizona State Supervisor for the LITERATURE CITED

Bureau of Land Management, U. S. Department
of the Interior, has furnished a number of interest- (1) BAKHMETEFP, B. A., and MATZKE, A. E.
ing photographs of a SAF stilhng basin (figs. 5 1936. THE HYDRAULIC JUMP IN TERMS OF DYNAMIC
and 6). In a letter to the writer, dated October SIMILARITY. Amer. Soc. Civ. Engin. Trans.
101: 630-680, illus. (Paper No. 1935.)
31, 1955, Mr. Rowland writes: (2) BLAISDELL, F. W.
Enclosed are a few photographs of a drop structure 1948. DEVELOPMENT AND HYDRAULIC DESIGN,
placed across the San Simon Wash in southeastern SAINT ANTHONY PALLS STILLING BASIN.
Arizona. This is a replacement for an earlier design Amer. Soc. Civ. Engin. Trans. 113: 483-
structure which was not adequate and washed out in 561, illus. (Paper No. 2342.)
1954. This structure, known as the San Simon drop (3)
structure, has worked very successfully this year. 1949. FLOW THROUGH DIVERGING OPEN CHANNEL
The peak flow water stood at 6%' depth in the im- TRANSITIONS AT SUPERCRITICAL VELOCITIES.
pounded area above the drop structure. The spillway Soil Conserv. Serv. SCS-TP-76, 21 pp., iUus.
lip is at elevation 88 feet (assumed) and the peak (4) and DONNELLY, C. A.
water within the reservoir was at 94}i foot stage. It 1951. HYDRAULIC DESIGN OP THE BOX INLET DROP
was calculated that the maximum flow through the SPILLWAY. Minn. St. Anthony Falls Hy-
structure at this elevation was approximately 2,200 cfs. draul. Lab. Tech. Paper 8, Ser. B, 63 pp.,
At this peak flow, the hydraulic jump in the box illus. (Rep. by Soil Conserv. Serv. as SCS-
outlet, as indicated by the flow line of water through TP-106. 1951.)
^ the lower structure was 13 feet above the floor of the (5) BRADLEY, J. N., and PETEBKA, A. J.
structure at this point. The heavy splash line as 1955. RESEARCH STUDY ON STILLING BASINS,
indicated by the mud deposits on the side walls [fig. 6] ENERGY DISSIPATORS, AND ASSOCIATED AP-
reached 16 feet in height above the floor. The walls PURTENANCES. U. S. Bur. Reclam. Lab.
are 18 feet high at this point. Rpt. Hyd-399, 125 pp., illus.
It is extremely gratifying to note that the drop
structure operated very effectively for volume flow (6) and PETERKA, A. J.
reduction and reduced channel cutting. You will note 1957. THE HYDRAULIC DESIGN OF STILLING BASINS:
in the picture [fig. 5,B], taken after all flow through HYDRAULIC JUMPS ON A HORIZONTAL APRON
the spillway had stopped on September 26, that there (BASIN I). Amer. Soc. Civ. Engin. Proc,
was no channel cutting, and practically no cutting Hvd. Div. Jour., 83 (HY 5): 1401-1 to
around the lower wing walls except that which was 1401-24, iflus.
caused by foreign drainage on the east wing, which (7) and PETERKA, A. J.
will be corrected. 1957. HYDRAULIC DESIGN OF STILLING BASINS:
Of most interest to us was the small sand fan which HIGH DAMS, EARTH DAMS, AND LARGE CANAL
developed immediately below the lip of the structure STRUCTURES (BASIN II). Amer. Soc. Civ.
and can be noted in the picture taken September 26 Engin. Proc, Hyd. Div. Jour., 83 (HY 5) :
[fig. 5,JS]. 1402-1 to 1402-14, iUus.

14
 /^--íífL

■^^Hg'-:'

FiGUKE 5.—San Simon drop structure—Spillway crest elevation is 88 feet; dam crest elevation, 103; end sill elevation,
68; chute, 40 feet wide; and stilling basin sidewalls, 18 feet high: A, After completion of structure, July 10, 1955;
and B, after flow of 2,200 c. f. s., September 26, 1955.

(8) Bradley, J. N., and PETERKA, A. J. (11) and PETERKA, A. J.

1957. HYDRAULIC DESIGN OF STILLING BASINS: 1957. HYDRAULIC DESIGN OP STILLING BASINS:
SHORT STILLING BASIN FOR CANAL STRUC- SMALL BASINS FOR PIPE OR OPEN CHANNEL
TURES, SMALL OUTLET WORKS, AND SMALL OUTLETS NO TAIL WATER REQUIRED (BASIN
SPILLWAYS (BASIN III). Amer. Soc. Civ. vi). Amer. Soc Civ. Engin. Proc, Hyd.
Engin. Proc, Hyd. Div. Jour., 83 (HY 5) : Div. Jour., 83 (HY 5) : 1406-1 to 1406-17,
- 1403-1 to 1403-22, illus. illus.
(9) and PETERKA, A. J. (12) DONNELLY, C. A.
1957. HYDRAULIC DESIGN OF STILLING BASINS: 1947. DESIGN OF AN OUTLET FOR BOX INLET DROP
STILLING BASIN AND WAVE SUPPRESSORS FOR SPILLWAY. Soil Conserv. Serv. SCS-TP-63,
CANAL STRUCTURES, OUTLET WORKS, AND 31 pp., illus.
DIVERSION DAMS (BASIN IV) . Amer. Soc.
Civ. Engin. Proc, Hyd. Div. Jour., 83 (13) and BLAISDELL, F. W.
(HY 5): 1404-1 to 1404-20, illus. 1954. STRAIGHT DROP SPILLWAY STILLING BASIN.
(10) and PETERKA, A. J. Minn. St. Anthony Falls Hydraul. Lab.
1957. HYDRAULIC DESIGN OF STILLING BASINS: Tech. Paper 15, Ser. B, 35 pp., illus.
STILLING BASIN WITH SLOPING APRON (BASIN (14) HARLEMAN, D. R. F.
v). Amer. Soc. Civ. Engin. Proc, Hyd. 1955. EFFECT OF BAFFLE PIERS- ON STILLING BASIN
Div. Jour., 83 (HY 5) : 1405-1 to 1405-32, PERFORMANCE. Boston Soc. Clv. Engin.
illus. Jour. 42 (2): 84-99, illus.

15
 f«
\

i.'^.^.» '^

FIGURE 6.—Mud deposit on sidewalls of San Simon drop structure. The maximum flow line is 13 feet above the stilling
basin floor, and the maximum splash line is about 16 feet above the basin floor. Photographed August 10, 1955.

(15) KING, H. W. (18) STANLEY, C. M.

1954. HANDBOOK OP HYDRAUUCS. ED. 4. 556 1934. STUDY OF STILLING-BASIN DESIGN. Amer.
pp., illus. New York. Soc. Civ. Engin. Trans. 99: 490-523, illus.
(16) RBE, W. O. (Paper No. 1867.)
1951. RESULTS OF TESTS ON A CHUTE WITH A ST. (19) WARNOCK, J. E.
ANTHONY FALLS STILLING BASIN. Soil Con- 1940. SPILLWAYS AND ENERGY DISSIPATORS. Iowa
serv. Serv. SCS-TP-107, 13 pp., illus. Univ. Studies in Engin. Bul. 20 (ProC
(17) ScHOKLiTSCH, A. (Translated by Samuel Schulitz.) Hydraulics Conf.), pp. 142-159, illus.
1937. HYDRAULIC STRUCTURES. 2 V., illus. Amer-
ican Society of Mechanical Engineers, New
York.

16
U. S. GOVERNMENT PRINTING OFFICE: 1959