0% found this document useful (0 votes)

17 views

Ur Process Pocket

Uploaded by

Anand Agrawal

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

17 views

Ur Process Pocket

Uploaded by

Anand Agrawal

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 356

Line sizing calculations

Link: Line Sizing calculations

This tool was developed for process engineers to quickly estimate line pressure drops (single-phase gas
or liquid) estimate equivalent length for pipelines, and access a list of standards checklists on site. As a
part of a larger project to develop what is similar to Carl Branan’s book “process engineers Pocket
Handbook” these tools would allow a process engineer to quickly calculate/estimate equipment
efficiencies or sizing using standardized calculations

The aim here is to take little-known data from the field (flow, pressures, temperatures,
compositions..etc.) and use it as input for a rough estimation without having to return to the office to
use commercial software or calculations Excel sheets to validate or to calculate. Additionally, These tools
may also serve as a gathered data validation tool.

Nomenclature

T Absolute Temperature (K) P Absolute Pressure (kg/cm.2a)

K (Cp/Cv) Ratio of specific heats, Cp/Cv Pr Reduced pressure (psia)
Z Compressibility factor Tr Reduced temperature (R)
Q Volume Flow rate (std.m3/hr) Pc Critical pressure (psia)
Ꝭ Density (kg/m3) Tc Critical temperature (R)
Re Reynold’s number M.wt Molecular weight (g / g. mol)
D Pipe diameter (in) y Component composition mol fraction
Pv Vapor pressure (kg/cm2.a) H Elevation (m)

Input Required
Table 1: Note 1: Input required depending on the case

Gas pressure drop Calculations Liquid pressure drop calculations

Flow rate [note 1] STD.m3/hr P1 Kg/cm2.g
Upstream pressure [note 1] Kg/cm2.a T o
C
o 3
average temperature C Q m /hr
[note 1]
Downstream pressure Kg/cm2.a Density Kg/m3
M.Wt / K (Cp/Cv) / Z Cp
- /-/-/ kgm-3 / cP
factor/Density and Viscosity
Viscosity
Alternative option:
Vol% or mol%
Composition
L m
D In
Head [note 1] m
Vapor pressure [note 1] Kg/cm2.a

Output Obtained

Gas pressure drop Calculations Liquid pressure drop calculations

Upstream/Downstream Kg/cm2.a Downstream Kg/cm2.g
pressure pressure
Flow rate STDm3/hr – Nm3/hr – Kg/cm2
pressure drop
m3/hr
Dp% to Upstream pressure % velocity m/s
dp/100 m Kg/cm2/100 m Reynlods number
friction factor/E - friction factor
Reynolds number - assumed roughness mm
Out velocity / Sonic velocity m/s NPSHa[note 1] m
Out Mach number -
ꝬV^2 Kg/m.s2
1/sqrt(k) -
viscosity cP
M.Wt / K (Cp/Cv) / Z factor -

Pressure drop calculations

1.1 Liquid pressure drop calculations
1.1.1 Nelson (Using fanning friction factor) [6]
Fanning’s equation (Originally derived for turbulent flow) for flow in circular conduit is:
0.323 ∗ 𝑓 ∗ 𝑠 ∗ 𝑢2 ∗ 𝐿
∆𝑃𝑓 =
𝑑
7738 ∗ 𝑑 ∗ 𝑢 ∗ 𝑠
𝑅𝑒 =
𝑍
Where:
∆𝑃𝑓 : pressure drop (psi)

s: specific gravity
u: average velocity (ft/s)
L: Equivalent length (ft)
d: pipe diameter (in)
f: fannings friction factor (equals ¼ moody’s friction factor)
𝑅𝑒 : Reynold’s number
Z: viscosity (cP)
Note: fannings equations isn’t valid for gases when pressure drop is larger than 10% of
downstream pressure
Total equivalent length = actual length + fittings estimated length
Nelson’s fittings equivalent length[6]
Equivalent length table P.396

Calculations of equivalent length through nomographs and K p. 397

𝒖𝟐
𝒉=𝑲∗ (ft of fluid pressure drop)
𝟔𝟒.𝟒

Fanning’s friction factor chart P.398[6]

1.1.2 Darcy equation [2]
𝑓 ∗ 𝜌 ∗ 𝑢2 ∗ 𝐿
∆𝑃𝑓 = 1.01972 ∗ 10−5 ∗
2∗𝑑
Where: ∆𝑃𝑓 : pressure drop (kg/cm2)
F: moody’s friction factor
𝜌: Liquid density (kg/m3)
L: equivalent Length (m)
D: pipe inside diameter (m)

Reynold’s number
𝝆𝒗𝒅
𝑹𝒆 =
𝝁
μ: Viscosity (kg/m.s) or 0.001*cP
v: fluid velocity (m/s)
Darcy friction factor
You either opt to use the famous moody’s chart or use the Colebrook-white
equation
Laminar flow
𝟔𝟒
𝒇=
𝑹

Turbulent flow
Colebrook-White equation
𝟏 𝒆 𝟐. 𝟓𝟏
= −𝟐 𝐥𝐨𝐠 𝟏𝟎 ( + )
√𝒇 𝟑. 𝟕𝑫 𝑹√𝒇

Where: f: Darcy friction factor

D: pipe inside diameter, in (or mm)
e: absolute pipe roughness, in (or mm)[4]
Pipe material Roughness (in) Roughness (mm)
Riveted steel 0.035-0.35 0.9-9
Commercial steel/welded steel 0.0018 0.045-0.05
Cast iron 0.01 0.26
Galvanized iron 0.006 0.15
Asphalted cast iron 0.0047 0.12
Wrought iron 0.0018 0.045
PVC, drawn tubing, glass 0.000059 0.0015
Concrete 0.0118-0.118 0.3-3.0
Moody’s chart
Estimation of equivalent length (Other than nelson’s)
1. JGC Quick Estimation

When the piping layout is not available, the equivalent length (Le) of the piping will be
estimated based on the straight length (Ls) as follows:

1. Process area: 3.0 times Ls

2. Common area: 1.5 times Ls
3. Offsite area: 1.3 times Ls

Note that Ls is the sum of XYZ coordinate length. For large size or high pressure piping, it is
recommended to estimate the number of elbows tees and valves and evaluate the equivalent
length, assuming piping layout. [5]

2. GPSA table[1]:
This table is the table used to estimate your equivalent length in the tool developed

Figure 1: Source: GPSA databook P. 17-6

1.1.3 Hazen-Williams equation [3]
the hazen-williams equation take a different approach. It’s based on the head loss
in 10-year-old pipe and on actual testing of arious sie pipes of different degrees of
roughness. These test were conducted in the first part as the last century and
were limited to a relatively narrow range of pipe diamteres. The tables should not
be used for pipe diameters less than 200 mm (8 in) nor for diameters greater
thatn 915 mm (36 in). their values mus be adjusted for pipe age and materials,
through the use of correction factors. The hazen-williams calculations are strictly
designed for water at ambient temperatures and are not applicable to other
liquids. These tables are base on the following formula[7]:

16 𝑄 1.852 𝑆. 𝑔
∆𝑃𝑓 = 0.98066 ∗ 1.1101 ∗ 10 ∗( )
𝐶 𝐷4.87
Where: ∆𝑃𝑓 : frictional pressure drop (kg/cm2)

Q: flow rate (m3/hr)

D: pipe inside diameter (mm)

S.gr: Liquid specific gravity

C: Hazen-williams C factor

Hazen-williams C factor
Pipe material Hazen-williams C factor
Smooth pipes (all metals) 130-140
Cast iron (old) 100
Iron (worn/pitted) 60-80
Polyvinyl chloride (PVC) 150
Brick 100
Smooth wood 120
Smooth masonry 120
Vitrified clay 110
Note: This equation is yet to be utilized in our tool
1.2 Gases pressure drop calculations
Several equations are available that relate the gas flow rate with gas properties, pipe diameter and
length, and upstream and downstream pressures. These equations are listed as follows:
1. General Flow equation
2. Weymouth equation
3. Panhandle A equation
4. Panhandle B equation

Other equations which aren’t discussed here: Oliphant, Fritzsche, Muller, IGT, Spitzglass high, Spitzglass
low [4]

General flow equation [3]

The steady-state, isothermal flow behavior of gas in pipelines is defined by a general energy equation of
the form:
0.5
11.4946 ∗ 10−4 𝑇𝑏 𝑃12 − 𝑒 𝑠 𝑃22
𝑄= ∗ ( ) ∗ 𝐷2.5
24√𝑓 𝑃𝑏 𝐺 𝑇𝑓 𝐿𝑒 𝑍
And the elevation adjustment term s is given by:
0.0684𝐺(𝐻2 − 𝐻1)
𝑠=
𝑇𝑓 𝑍
𝐿(𝑒 𝑠 − 1)
𝐿𝑒 =
𝑠

Where:
Q: gas flow rate (stdm3/hr)
Tb: base temperature K (15.5556 oC +273.15 in tool)
Pb: Base pressure kPa (101.325 Kpa in tool)
Tf: average flow temperature K
P1: Upstream pressure kPa
P2: Downstream pressure kPa
L: Pipe length (km)
Le: Equivalent length of pipe (km)
H1: Upstream elevation (m)
H2: Downstream elevation (m)
G: gas gravity (air =1)
f: Darcy’s friction factor
D: pipe inside diameter, mm.
Z: Gas compressibility

The equivalent length, Le, and the term 𝑒 𝑠 take into account the elevation difference
between the upstream and downstream ends of the pipe segment. The parameter s depends upon the
gas gravity, gas compressibility factor, the flowing temperature, and the elevation difference.
𝑷𝒃 𝑮𝑸
𝑹𝒆 = 𝟎. 𝟓𝟏𝟑𝟒 ∗ 𝟐𝟒 ∗ 𝟏𝟎𝟎 ∗ ∗
𝑻𝒃 𝝁𝑫
𝝁: viscosity in cP
Z is determined based on average P where:
𝟐 𝑷𝟏𝑷𝟐
𝑷𝒂𝒗𝒈. = (𝑷𝟏 + 𝑷𝟐 − )
𝟑 𝑷𝟏 + 𝑷𝟐

1.2.2 Weymouth equation [3]

This formula is generally used for short pipelines and gathering systems. Like the Panhandle equations,
this equation also uses an efficiency factor. The equation cannot be generally applied to any variety of
diameters and roughness, and in the flow region of partially developed turbulence, it is not valid. The
Weymouth Equation may be used to approximate fully turbulent flow by applying correction factors
determined from the system to which it is to be applied. The Weymouth equation is used for high
pressure, high flow rate, and large diameter gas gathering systems. [4]

Figure 2: API 14E

0.5
−3
𝑇𝑏 𝑃12 − 𝑃22
𝑄 = 3.7435 ∗ 10 ∗ 24 ∗ 𝐸 ∗ ( ) ∗ 𝐷2.667
𝑃𝑏 𝐺 𝑇𝑓 𝐿𝑒 𝑍

Where: E: Pipeline efficiency, a decimal value less than 1.0

remaining units are the same for the general flow equation

Pipleline efficiency E
Note: API RP 14E doesn't use E for the Weymouth equation.

1.2.3 Panhandle A equation [3]

The Panhandle A Equation was developed for use in natural gas pipelines, incorporating an efficiency
factor for Reynolds numbers in the range of 5 to 11 million. In this equation, the pipe roughness is not
used. [4] This equation was intended to reflect the flow of gas through smooth pipes. When “adjusted”
with an efficiency factor, E, of about 0.90, the equation is a reasonable approximation of the partially
turbulent flow equation. The equation becomes less accurate as flow rate increases. Many users of the
Panhandle A Equation assume an efficiency factor of 0.92. [1]

Figure 3: source: API 14E

0.5394
𝑇𝑏 1.0788
𝑃12 − 𝑃22
𝑄 = 4.5965 ∗ 10−3 ∗ 24 ∗ 𝐸 ∗ ( ) ( ) ∗ 𝐷2.6182
𝑃𝑏 𝐺0.8539 𝑇𝑓 𝐿𝑒 𝑍

remaining units are the same for the general flow equation

1.2.4 Panhandle B equation [3]

The Panhandle B equation, also known as the revised Panhandle equation, is used in large diameter,
high pressure transmission lines. In fully turbulent flow, it is found to be accurate for values of Reynolds
number in the range of 4 to 40 million. [4] this revised equation is only slightly Reynolds number
dependent. Therefore, it more nearly approximates fully turbulent flow behavior. The equation can be
adjusted through the use of an efficiency term that makes it applicable across a relatively limited range
of Reynolds numbers. Other than this, however, there are no means for adjustment of the equation to
correct it for variations in pipe surface. Adjusted to an average flowing Reynolds number, the equation
will predict low flow rates at low Reynolds numbers, and high flow rates at high Reynolds numbers, as
compared to a fully turbulent flow equation. Efficiencies based on the Panhandle B equation decrease
with increasing flow rate for fully turbulent flow. The efficiency factor, E, used in the Panhandle B
equation generally varies between about 0.88 and 0.94.[1]
0.51
−2
𝑇𝑏 1.02
𝑃12 − 𝑃22
𝑄 = 1.002 ∗ 10 ∗ 24 ∗ 𝐸 ∗ ( ) ( 0.961
) ∗ 𝐷2.53
𝑃𝑏 𝐺 𝑇𝑓 𝐿𝑒 𝑍

The successful application of these transmission line flow equations in the past has largely involved
compensation for discrepancies through the use of adjustment factors, usually termed “efficiencies.”
These efficiencies are frequently found in practice by determining the constant required
to cause predicted gas equation behavior to agree with flow data. As a result, the values of these factors
are specific to particular gas flow equations and field conditions and, under many circumstances, vary
with flow rate in a fashion that obscures the real nature of flow behavior in the pipe. The Reynolds
number dependent equations, such as the Panhandle equations, utilize a friction factor expression
which yields an approximation to partially turbulent flow behavior in the case of the Panhandle A
equation, and an approximation to fully turbulent behavior in the case of the Panhandle B.

These equations suffer from the substitution of a fixed gas viscosity value into the Reynolds number
expression, which, in turn, substituted into the flow equation, results in an expression

with a preconditioned bias. Regardless of the merits of various gas flow equations, past

practices may dictate the use of a particular equation to maintain continuity of comparative capacities
through application of consistent operating policy. A summary of comparisons between transmission
factors used in the above gas equations are shown in Fig. 17-7. [4]

Example from Gas Hydraulics book, sashi menon. [4]

In Figure 2.5, we consider a pipeline 100 mi long, NPS 16 with 0.250 in. wall thickness, operating at a
flow rate of 100 MMSCFD. The gas flowing temperature is 80°F. With the upstream pressure fixed at
1400 psig, the downstream pressure was calculated using the different flow equations. By examining
Figure 2.5, it is clear that the highest pressure drop is predicted by the Weymouth equation and the
lowest pressure drop is predicted by the Panhandle B equation. It must be noted that we used a pipe
roughness of 700 μin. for both the AGA and Colebrook equations, whereas a pipeline efficiency of 0.95
was used in the Panhandle and Weymouth equations. Figure 2.6 shows a comparison of the flow
equations from a different perspective. In this case, we calculated the upstream pressure required for
an NPS 30 pipeline, 100 miles long, holding the delivery pressure constant at 800 psig. The upstream
pressure required for various flow rates, ranging from 200 to 600 MMSCFD, was calculated using the five
flow equations. Again it can be seen that the Weymouth equation predicts the highest upstream
pressure at any flow rate, whereas the Panhandle A equation calculates the least pressure. We
therefore conclude that the most conservative flow equation that predicts the highest pressure drop is
the Weymouth equation and the least conservative flow equation is Panhandle A.
Remaining calculations
Liquid Calculations
NPSHa = Psuction – Pv = P1 - ∆𝑃𝑓 + H – Pv
(P1 and Pv variables in kg/cm2.a and P1, Pv & - ∆𝑃𝑓 converted to m.liq by *10/S.Gr)

Gas Calculations
𝑷𝒃 (𝒌𝒈/𝒄𝒎𝟐 ) (𝟏. 𝟎𝟑𝟑) 𝑻 (°𝑲)
𝑸𝒂𝒄𝒕𝒖𝒂𝒍 = 𝑸𝒔𝒕𝒅𝒎𝟑 /𝒉𝒓 × ×
𝑷 (𝒌𝒈/𝒄𝒎𝟐 ) 𝑻𝒃 (°𝑲)(𝟏𝟓. 𝟓𝟔 °𝑪)
Similarly, 𝑸𝒏𝒐𝒓𝒎𝒂𝒍 (𝒏𝒎𝟑/𝒉𝒓) can be calculated at T = 0 +273.15 K and P=Pb
𝑸𝒂𝒄𝒕𝒖𝒂𝒍
𝑽𝒆𝒍𝒐𝒄𝒊𝒕𝒚 =
𝑨𝒄𝒓𝒐𝒔𝒔 𝒔𝒆𝒄𝒕𝒊𝒐𝒏𝒂𝒍

𝟗. 𝟖𝟏 ∗ 𝒌 ∗ 𝑹 ∗ 𝑻
𝑺𝒐𝒏𝒊𝒄 𝒗𝒆𝒍𝒐𝒄𝒊𝒕𝒚 𝒗𝒔 = √
𝑴𝒘𝒕
𝒗
𝒎𝒂𝒄𝒉 𝒏𝒖𝒎𝒃𝒆𝒓 =
𝒗𝒔
𝑪𝒆
𝑬𝒓𝒓𝒐𝒔𝒊𝒐𝒏𝒂𝒍 𝒗𝒆𝒍𝒐𝒄𝒊𝒕𝒚 𝑽𝒆 =
√𝝆
Table 2: Source: JGC

Ce Services
180-240 Solid-free fluids where corrosion is not anticipated or when corrosion is controlled by
inhibition or by employing corrosion resistant alloys, continuous.
300 Same as above but intermittent
120 Solids and/or corrosives are expected
Table 3: C = 100 for continous and C=125 for itermittent services are conservative [8]

Where:
R: gas constant 847.9 (kgf/m2)(m3)/(kg-mole)(°K)

K: Cp/Cv

Ve: erosional velocity (m/s)

Pm: Homogeneous density (kg/m3)

Ce: Empirical constant

If [5]:
1 The flow is subsonic
Mach number <
√𝑘
1 The flow is sonic
Mach number =
√𝑘
1 The flow is supersonice
Mach number >
√𝑘

Compressibility factor Z [3]

First, Tr and Pr are calculated as follows (T and P are absolute values):

𝑛

𝑇𝑐 = ∑ 𝑦𝑖 ∗ 𝑇𝑐𝑖
𝑖=1
𝑛

𝑃𝑐 = ∑ 𝑦𝑖 ∗ 𝑃𝑐𝑖
𝑖=1

𝑇 𝑃
𝑇𝑟 = , Pr =
𝑇𝑐 𝑃𝑐
Where Tci and Pci are each component’s critical temperature and pressure

Naturally, The Standing and Katz method is applied where Z is obtained as a function of Tr and Pr
through the famous chart (figure no. ). Alternatively, Dranchuk, Purvis, and Robinson method is used. In
this method the Benedict-Webb-Rubin equation of state is used to correlate the Standing-Katz Z
factor chart. Eight coefficients A1, A2, etc., are used in this equation as shown [1]:
𝑨𝟐 𝑨𝟑 𝑨𝟓 𝑨𝟓 𝑨𝟔 𝝆𝒓 𝟓 𝑨𝟕 𝝆𝒓 𝟑
𝒁 = 𝟏 + (𝑨𝟏 + + ) 𝝆𝒓 + (𝑨𝟒 + ) 𝝆𝒓 𝟐 + +
𝑻𝒑𝒓 𝑻𝟑 𝑻𝒑𝒓 𝑻𝒑𝒓 𝑻𝒑𝒓 𝟑 (𝟏 + 𝑨𝟖 𝝆𝒓 𝟐 )𝒆𝒙𝒑(−𝑨𝟖 𝝆𝒓 𝟐 )

Where:
𝟎. 𝟐𝟕 ∗ 𝑷𝝆𝒓
𝝆𝒓 =
𝒁𝑻𝝆𝒓

𝑨𝟏 = 𝟎. 𝟑𝟏𝟓𝟎𝟔𝟐𝟑𝟕 𝑨𝟐 = −𝟏. 𝟎𝟒𝟔𝟕𝟎𝟗𝟗𝟎

𝑨𝟑 = −𝟎. 𝟓𝟕𝟖𝟑𝟐𝟕𝟐𝟗 𝑨𝟒 = 𝟎. 𝟓𝟑𝟓𝟑𝟎𝟕𝟕𝟏
𝑨𝟓 = −𝟎. 𝟔𝟏𝟐𝟑𝟐𝟎𝟑𝟐 𝑨𝟔 = −𝟎. 𝟏𝟎𝟒𝟖𝟖𝟖𝟏𝟑
𝑨𝟕 = 𝟎. 𝟔𝟖𝟏𝟓𝟕𝟎𝟎𝟏 𝑨𝟖 = 𝟎. 𝟔𝟖𝟒𝟒𝟔𝟓𝟒𝟗
NOTE: Through trial and errors Z is assumed and then calculated until error is within tolerance.

K (Cp/Cv) [1]
𝐶𝑝 𝑀𝐶𝑝
𝑘= = , 𝑅 = 1.986 𝐵𝑡𝑢/(𝑙𝑏𝑚𝑜𝑙 ˚𝐹)
𝐶𝑣 𝑀𝐶𝑝 − 𝑅
∑𝑛𝑖=1 𝑦𝑖 ∗ 𝑀𝐶𝑝
𝑘=
(∑𝑛𝑖=1 𝑦𝑖 ∗ 𝑀𝐶𝑝) − 1.986

MCp values at different temperatures for a variety of compounds are found in figure 1 [1]
Figure 4: Mcp, Tc and Pc values [1]

Figure 5: Calculations of Average M.wt and k [1]

Figure 6: Standing-Katz Chart [1]

Gas Viscosity [1]

Figures 23- 22, 23-23 and 23-24 present data on hydrocarbon gases. To correct for pressure, adjust the
gas viscosity from Fig. 23-22 from atmospheric pressure using Fig. 23-24, or read directly from Fig. 23-
23. Figure 23-24 is better when the reduced temperature is greater than 1.0.
The method of Dean and Stiel41 is best for calculating the viscosity of a gaseous mixture with large
amounts of non-hydrocarbons. This method is particularly useful for handling natural gas with high CO2
content. Compared to 30 CO2-N2 mixtures, this method has an average deviation of 1.21% at pressures
up to 3525 psia. The method uses a factor, ξ, defined as [1]:
Figure 7source 1. GPSA Engineering Data Book Section 23 2nd edition p. 23-31
Figure 8: source 1. GPSA Engineering Data Book Section 23 2nd edition p. 23-32
Figure 9: source 1. GPSA Engineering Data Book Section 23 2nd edition p. 23-33
Calculations Examples using commercial software, hand written calculations and the tool developed:

Example No.1: Liquid pressure drop calculations (Units converted to match table units)

Computer Methods for chemical engineers (2nd edition Page no. 37) (input highlighted)
Table 4: Note: Software pgk utilized is PR, using ASME stem is recommended though for this particular case as fluid is Water!

Hand Calculations Tool Developed

Property Unit Commercial software[Note]
in book (Darcy’s)
P1 (Kg/cm2.g) 1.033 1.033 1.033
T (oC) 25 25 25
Q (m3/hr) 8 - 8
Density Kg/m3 1007 1000 1007
Viscosity (Cp) 0.8904 1 0.8904
L (m) 10 10 10
D (in) 1 1 1
P2 Kg/cm2.g 0.303 - 0.282
pressure drop 0.73 0.54 0.751
velocity m/s 3.948 4 3.98
Reynlods number 119062 1.06 * 10^5 120117
friction factor - 0.0176 0.0245
assumed m 0.0000457 0.0
0.00005
roughness

Manual calculations:
𝑄 8∗4
𝑣= = = 4 𝑚/𝑠
𝐴 3600 ∗ 3.14 ∗ 0.02662
𝜌𝑣𝑑 1007 ∗ 4 ∗ 0.0266
𝑅= = = 120333.33
𝜇 0.8904 ∗ 0.001
𝜀 0.00005
The relative roughness of the smooth pipe is =
𝑑 0.0266

Turbulent flow, Iterate through the following function to obtain f using the goal
seek function in excel or use moody’s chart
Colebrook-White equation
1 𝑒 2.51
= −2 log10 ( + )
√𝑓 3.7𝐷 𝑅√𝑓

f = 0.0245
𝑓 ∗ 𝜌 ∗ 𝑢2 ∗ 𝐿 1.01972 ∗ 10−5 ∗ 0.0245 ∗ 1007 ∗ 42 ∗ 10
−5
∆𝑃𝑓 = 1.01972 ∗ 10 ∗ =
𝑑2 0.0266 ∗ 2
= 0.751 𝑘𝑔/𝑐𝑚2
P2 = P1 - ∆𝑃𝑓 = 0.282 kg/cm2
Calculations Examples using commercial software, hand written calculations and the tool developed:

Example No.2: Gas pressure drop calculations (Units converted to match table units)

Computer Methods for chemical engineers (2nd edition Page no. 43) (input highlighted)

Hand Tool
Commercial
Property Unit Calculations Developed
software[Note]
in book (NiqSim)
P1 Kg/cm2 3.518 3.518 3.518
P2 Kg/cm2 3.384 3.334 3.376
o
T C 25 25 25
L m 20000 20000 20000
D m 0.154 0.154 0.154
flow_rate Std_m3/hr 424.1 363 kg/hr 424.1
Composition Vol% 85 CH4 15 CO2 85 CH4 15 CO2 85 CH4 15 CO2
flow_rate Nm3/hr - - 402.02
Actual_flow_rate m3/hr - - 133.99
Dp% to % 3.8
Upstream 4.022
pressure
dp/100 m Kg/cm2/100m 0.00067 - 0.0007
friction factor/E - 0.024 -
Reynolds 70480 71400
71670
number
Out velocity m/s 1.98 1.98 1.99
Sonic Velocity m/s 398.8 - 399.24
Out Mach 0.005 -
0.005
number
ꝬV^2 - - 11.527
mol.weight 20.24 - 20.238
z 0.9931 - 0.993
k 1.308 - 1.30
1/root(k) - - 0.876
viscosity cP 0.0118 0.012 0.0116
Table 5: Note: Gas pipe for compressible gases were used, adiabatic pipe could also be used with similar results
Manual Calculations
General gas equation
K, Pr and Tr calculations
Carbon
Gas Methane dioxide total
Mol wt 16.043 44.01 Tc 207.59 K
Pc 667 1070 Pc 51.144952 kg/cm2.a
Tc -116.66 87.76 M.wt 20.23805 g/gmol
0F 8.23 8.38
50 F 8.42 8.7 T1 298.15 K
60 F 8.46 8.76 76.37 F
100 F 8.65 9 P1 3.518 kg/cm2.a
150 F 8.95 9.29
200 F 9.28 9.56
250 F 9.64 9.81
300 F 10.01 10.05
mol fraction 0.85 0.15 1
Tc -85.997 F P1
Pc 727.45 Psia Tr 1.436
M.wt 20.23805 g/gmol Pr 0.068
P1 MCp 8.505

K1 1.304648

Viscosity calculations
Viscosity at 1 atm µa (cP) Fig. 32.-22 0.0104
Viscosity addition due to CO2 15 mol% Fig. 32.-22 0.0008
Total µa (cP) 0.0112
µ/µa at Pr = 0.068 and Tr = 1.54 Fig. 32.-24 1.05
µ (cP) 1.05*0.0112 0.01176
Z calculations

Iterate through the following equation or use katz chart at Tr and Pr

𝑨𝟐 𝑨𝟑 𝑨𝟓 𝟐
𝑨𝟓 𝑨𝟔 𝝆𝒓 𝟓 𝑨𝟕 𝝆𝒓 𝟑
𝒁 = 𝟏 + (𝑨𝟏 + + ) 𝝆 + (𝑨𝟒 + )𝝆 + +
𝑻𝒑𝒓 𝑻𝟑 𝒓 𝑻𝒑𝒓 𝒓 𝑻𝒑𝒓 𝑻𝒑𝒓 𝟑 (𝟏 + 𝑨𝟖 𝝆𝒓 𝟐 )𝒆𝒙𝒑(−𝑨𝟖 𝝆𝒓 𝟐 )

Where:
𝟎. 𝟐𝟕 ∗ 𝑷𝝆𝒓
𝝆𝒓 =
𝒁𝑻𝝆𝒓

Till Error < 0.001 for instance.

For both P1 & T1: Tr and Pr (obtained above)

(Use the Goal seek function in an Excel Sheet, Change your assumed Z till ABS(Z_assumed –
Z_calculated)*1000 = 0 for instance)

Z = 0.992

𝑤𝑡 20.24
𝐺 = 𝑀. = = 0.698
29 29
𝑷𝒃 𝑮𝑸 𝟏𝟎𝟏.𝟑𝟐𝟓 𝟎.𝟔𝟗𝟖∗𝟒𝟐𝟒.𝟏
𝑹𝒆 = 𝟎. 𝟓𝟏𝟑𝟒 ∗ 𝟐𝟒 ∗ 𝟏𝟎𝟎 ∗ ∗ = 𝟎. 𝟏𝟓𝟑𝟒 ∗ 𝟐𝟒 ∗ 𝟏𝟎𝟎 ∗ ∗ =70714
𝑻𝒃 𝝁𝑫 𝟐𝟕𝟑+𝟏𝟓.𝟓𝟓𝟔 𝟎.𝟎𝟏𝟏𝟕𝟔∗𝟏𝟓𝟒

Turbulent flow, Iterate through the following function to obtain f using the goal seek function in excel or
use moody’s chart
Colebrook-White equation
1 𝑒 2.51
= −2 log10 ( + )
√𝑓 3.7𝐷 𝑅√𝑓

f = 0.0206
0.5
11.4946∗ 10−4 𝑇𝑏 𝑃12 −𝑒 𝑠 𝑃22
𝑄= ∗ ( ) ∗ 𝐷2.5 , where s = 0 as no elevation change assumed
24√𝑓 𝑃𝑏 𝐺 𝑇 𝑓 𝐿𝑒 𝑍

0.5
11.4946 ∗ 10−4 273 + 15.56 3.5182 − 𝑃22
424.1 = ∗ ( ) ∗ 1542.5
24√0.0206 101.325 0.698 ∗ (273.15) ∗ 20000 ∗ 0.992

Solving for P2, P2 would equal ≈ 3.374 kg/cm2.a

A Quick take on Line sizing

Line sizing is an essential aspect of designing chemical plants. It involves determining the appropriate
diameter of pipes to transport fluids (gases, liquids, or multiphase mixtures) within a chemical plant.
Proper line sizing ensures that the plant operates efficiently, safely, and with minimal maintenance
while meeting the desired process parameters.

Main concerns in line sizing:

1. Pressure drop: A key consideration is minimizing pressure drop across the pipeline. High
pressure drops can lead to insufficient flow rates, increased energy consumption, and reduced
plant efficiency. Engineers use various correlations and models to calculate pressure drops and
select appropriate pipe diameters.
2. Flow regime: The flow characteristics (laminar, turbulent, or transitional) depend on the fluid's
properties, flow rate, and pipe diameter. Engineers should ensure that the desired flow regime
is maintained to avoid issues such as poor heat and mass transfer, increased pressure drop, or
flow instabilities.
3. Velocity limits: Fluid velocities in pipes should be maintained within specific limits to avoid
issues such as erosion, noise, and vibration. Higher velocities can cause erosion and damage to
pipes, while low velocities can lead to sedimentation or other issues.
4. Economic factors: The cost of piping and installation should also be considered when sizing
pipelines. Larger diameter pipes are more expensive and may require more support structures,
while smaller pipes may be cheaper but can result in higher pressure drops and energy
consumption.

Limitations of line sizing:

1. Accuracy of models and correlations: The models and correlations used for pressure drop
calculations and flow regime predictions are often based on empirical data and may not
accurately represent the specific conditions of a plant. Engineers must exercise judgment when
applying these tools and may need to consult experimental data or conduct simulations to
validate their assumptions.
2. Complexity of multiphase flows: In cases where multiple phases (gas, liquid, solid) are
present, predicting flow behavior and pressure drops can be more challenging. Engineers may
need to rely on experimental data, simulations, or specialized models for multiphase flow
systems.
3. Changes in operating conditions: Plant operating conditions can change over time due to
factors such as equipment wear, fouling, or changes in feedstock properties. These changes can
affect the flow behavior and pressure drops in pipelines, potentially requiring adjustments to
pipe diameters or other design parameters.

In summary, line sizing in chemical plants involves balancing various factors such as pressure drop, flow
regime, velocity limits, and economic considerations. Engineers must also consider the limitations of the
models and correlations used for line sizing, and be prepared to adapt to changing operating conditions.
Proper line sizing ensures that chemical plants operate safely, efficiently, and with minimal
maintenance. The basic principle to determine the line sizes shall be based on economical aspects, i.e.,
minimizing the sum of operational costs and investment costs. However, line sizes should not
exceed the limitations. In some instances, the process requirements will take precedence over
the economical aspects; for example, in the case of pump suction lines where the NPSH is the
main concern. [5]

References
1. GPSA Engineering Data Book Section 23 (Physical properties) and section 17 (fluid flow and
piping).
2. Computer methods for chemical engineering, Second Edition, Chapter 2: Fluid flow in pipes.
3. Piping Calculations manual, E. Shashi Menon. Chapter 7 Gas piping system (equations Q units
were converted to stdm3/hr instead of stdm3/day).
4. Gas Hydraulics, E. Shashi Menon. Chapter 2 pressure drop due to friction.
5. JGC Hydraulics standard JGS 210-120-1-37E
6. Petroleum refinery Engineering, W. L. Nelson.
7. American national standard for rotodynamic pumps for pump piping ANSI/HI 9.6.6 -2009
8. API RP 14E fifth edition, recommended practice for design and installation of offshore
production platform piping systems

T Temperature (oC) P Pressure (kg/cm2)

ṁ Mass Flow rate (kg/hr) µ Viscosity (Pa.s)

ρ Density (kg/m3) A Cross sectional Area (m2)

Di Tube inside diameter (mm) SG Specific gravity

Do Tube outside diameter (mm) h Head loss (m)

Re Reynold’s number u, v Velocity (m/s)

Q Volume flow rate (m3/hr) f Friction factor

L Length (m) Cv Control valve Cv (gpm/psi)

K Resistance coefficient NPSH Net positive suction Head (m)

SS Pump specific speed SSS Pump suction-specific speed

Input Required

Hydraulic system Calculations Pumps Specs Checking

Flow rates (main or branches) m3/hr Pump Type List of HI pumps

Density kg/ m3 Suction / Discharge Lines D In

Viscosity cP Pump speed rpm

Vapor Pressure Kg/cm2.a Rated Capacity m3/hr

Operating conditions Suction/Discharge Pressures Kg/cm2.g

Suction/ Discharge Vessel Pressures Kg/cm2.g Vapor pressure Kg/cm2.a

Suction/ Discharge Vessel Elevations m Specific Gravity -

Fittings Viscosity c.St

List input of fittings - NPSHa m

Diameters/ d/D for reducers In/ In./In. MCSF m3/hr

Pump service List of applications

Output Obtained

Summary Table Summary table

Differential pressure kg/cm2 NPSHa – NPSH3 m

Differential Head m Calculated NPSHa m

Hydraulic power Kw Common NPSHr – JGC Charts m

NPSHa m Specific speed -

Suction/Discharge velocities m/s US Specific speed -

Suction/Discharge Reynold’s number US Suction Specific speed -

Suction/Discharge pipe DP kg/cm2 Suction Specific speed -

Suction/Discharge pipe DH m HI calculated Eff. %

Suction/Discharge pressure kg/cm2.g Eff. Reduction due to SS %

Pump calculations datasheet Eff. Normal Deviation %

Hydraulic system analysis HI Min. – Max. Calculated power

JGC charts Eff %

Branan Calculated Eff. %

Min. – Max. Calculated power kW

Corrected BHP kW

Max. Rated Output kW

Selected motor/motor η kW / %

Min. – Max Accepted BEP m3/hr

POR for BEPmin – BEP max m3/hr

MCSF at max. impeller diameter m3/hr

Pump System Calculations

i. Suction, Discharge pressures and NPSHa calculations

ii. Pipes pressure drop calculations
iii. Fittings pressure drop calculations
iv. Control valves pressure dop calculations
v. How and Why to draw the hydraulic gradient
vi. How and Why to draw the system resistance curve
vii. Key points to take into consideration when specifying a centrifugal pump
1. Suction below CL (Center Line)

Static Suction Lift = Ls Pump Suction Diam = Ds

Static Discharge Head = Ld Pump Discharge Diam = Dd
Total Static Head Ltot = Ls + Ld
Friction Head = Hfs and Hfd
Inlet and Discharge Nozzle Losses = Hi and Ho
Velocity Head = V2/2g
Pump Suction and Discharge Nozzle Elevations = Lns and Lnd
Suction Pressure Head = Ps [if open tank = 0]
Discharge Pressure Head = Pd [if open tank = 0]
Pin = Ps - Ls - Hfs - Hi + Lns = Hs - Vs2/2g + Lns
Pout = Pd + Ld + Hfd + Ho - Lnd = Hd - Vd2/2g - Lnd
Total Dynamic Suction Lift Hs = Ps - Ls - Hfs - Hi - Vs2/2g
Total Dynamic Discharge Head Hd = Ld + Pd + Hfd + Ho + Vd2/2g
Total Dynamic Head Htot = Hd - Hs
NPSHa(abs) = Ps + Patm - Ls - Hfs - Hi + Vs2/2g - Pvap [where
Closed Tank: Pvap = Vapor Pressure(abs)]
or = Pamb - Ls - Hfs - Hi + Vs2/2g - Pvap
Open Tank: [NPSHa should exceed NPSHr by at least 0.5 m]
Dynamic Suction Lift (req'd) Hs(r) = Ps - Ls - Hfs - Hi
Dynamic Discharge Head (req'd) Hd(r) = Ld + Pd + Hfd + Ho
Dynamic Head (req'd) Htot(r) = Hd(r) - Hs(r)
NPSHa(req'd) NPSHa(r) = NPSHa - Vs2/2g

2. Suction Above CL (Center Line)

Static Suction Head = Ls Pump Suction Diam = Ds

Static Discharge Head = Ld Pump Discharge Diam = Dd
Total Static Head Ltot = Ls + Ld
Friction Head = Hfs and Hfd
Inlet and Discharge Nozzle Losses = Hi and Ho
Velocity Head = V2/2g
Pump Suction and Discharge Nozzle Elevations = Lns and Lnd
Suction Pressure Head = Ps [if open tank = 0]
Discharge Pressure Head = Pd [if open tank = 0]
Pin = Ps + Ls - Hfs - Hi - Lns = Hs - Vs2/2g + Lns
Pout = Pd + Ld + Hfd + Ho - Lnd = Hd - Vd2/2g - Lnd
Total Dynamic Suction Head Hs = Ls + Ps - Hfs - Hi - Vs2/2g
Total Dynamic Discharge Head Hd = Ld + Pd + Hfd + Ho + Vd2/2g
Total Dynamic Head Htot = Hd - Hs
NPSHa(abs) = Ps + Patm - Ls - Hfs - Hi + Vs2/2g - Pvap
Closed Tank:
[where Pvap = Vapour Pressure(abs)]
or = Pamb - Ls - Hfs - Hi + Vs2/2g - Pvap
Open Tank:
[NPSHa should exceed NPSHr by at least 0.5 m]
Dynamic Suction Head (req'd) Hs(r) = Ls + Ps - Hfs - Hi
Dynamic Discharge Head (req'd) Hd(r) = Ld + Pd + Hfd + Ho
Dynamic Head (req'd) Htot(r) = Hd(r) - Hs(r)
NPSHa(req'd) NPSHa(r) = NPSHa - Vs2/2g
Liquid pressure drop calculations (Hfs and Hfd) [1]

For pipe segments

• Use Darcy equation (check previous calculations documentation for liquid pressure
drop)

𝑓 ∗ 𝜌 ∗ 𝑢2 ∗ 𝐿
∆𝑃𝑓 = 1.01972 ∗ 10−5 ∗
2∗𝑑
Where: ∆Pf : pressure drop (kg/cm2)

f: moody’s friction factor

ρ: Liquid density (kg/m3)

L: equivalent Length (m)

D: pipe inside diameter (m)

Reynold’s number
ρvd
Re =
μ
μ: Viscosity (kg/m.s) or 0.001*cP

v: fluid velocity (m/s)

Laminar flow
64
f=
R
Turbulent flow
Colebrook-White equation
1 e 2.51
= −2 log10 ( + )
√f 3.7D Re√f
Where: f: Darcy friction factor
D: pipe inside diameter, in (or mm)
e: absolute pipe roughness, in (or mm) (0.045-0.05 for commercial steel)
• Convert ∆P to head loss
𝑛 10
ℎ𝑡𝑜𝑡𝑎𝑙 (𝑚) = ∑ ∆P𝑖 ∗
𝑖=0 𝑆. 𝐺

For numerous fittings [2]

• Use Equivalent lengths mentioned in previous documentation or

• Use crane K factor or
• Use Hooper 2K factors equation (this is the used method in our tool)
𝐾1 1 𝑖𝑛𝑐ℎ
𝐾= + 𝐾∞ (1 + )
𝑅𝑒 𝐷𝑖𝑛

Fitting K1 Kꚙ
Elbow, 90°, Standard (R/D = 1), Screwed 800.0 0.40
Elbow, 90°, Standard (R/D = 1), Flanged/welded 800.0 0.25
Elbow, 90°, Long-radius (R/D = 1.5), All types 800.0 0.20
Elbow, 90°, Mitered (R/D = 1.5), 1 weld (90° angle) 1000.0 1.15
Elbow, 90°, Mitered (R/D = 1.5), 2 weld (45° angle) 800.0 0.35
Elbow, 90°, Mitered (R/D = 1.5), 3 weld (30° angle) 800.0 0.30
Elbow, 90°, Mitered (R/D = 1.5), 4 weld (22.5° angle) 800.0 0.27
Elbow, 90°, Mitered (R/D = 1.5), 5 weld (18° angle) 800.0 0.25
Elbow, 45°, Standard (R/D = 1), All types 500.0 0.20
Elbow, 45°, Long-radius (R/D 1.5), All types 500.0 0.15
Elbow, 45°, Mitered (R/D=1.5), 1 weld (45° angle) 500.0 0.25
Elbow, 45°, Mitered (R/D=1.5), 2 weld (22.5° angle) 500.0 0.15
Elbow, 45°, Standard (R/D = 1), Screwed 1000.0 0.70
Elbow, 180°, Standard (R/D = 1), Flanged/welded 1000.0 0.35
Elbow, 180°, Long-radius (R/D = 1.5), All types 1000.0 0.30
Elbow, Used as, Standard, Screwed 500.0 0.70
Elbow, Elbow, Long-radius, Screwed 800.0 0.40
Elbow, Elbow, Standard, Flanged/welded 800.0 0.80
Elbow, Elbow, Stub-in type branch 1000.0 1.00
Tee, Run, Screwed 200.0 0.10
Tee, Through, Flanged or welded 150.0 0.05
Tee, Tee, Stub-in type branch 100.0 0.00
Valve, Gate, Full line size, Beta = 1 300.0 0.10
Valve, Ball, Reduced trim, Beta = 0.9 500.0 0.15
Valve, Plug, Reduced trim, Beta = 0.8 1000.0 0.25
Valve, Globe, Standard 1500.0 4.00
Valve, Globe, Angle or Y-type 1000.0 2.00
Valve, Diaphragm, Dam type 1000.0 2.00
Valve, Butterfly, 800.0 0.25
Valve, Check, Lift 2000.0 10.00
Valve, Check, Swing 1500.0 1.50
Valve, Check, Tilting-disc 1000.0 0.50
• Convert K to head loss
𝑛 𝑣𝑖2
ℎ𝑡𝑜𝑡𝑎𝑙 (𝑚) = ∑ 𝐾𝑖 ∗
𝑖=0 2𝑔
For control valves

Given control valve diameter (Used in our tool)

• Given the control valve diameter and if the control valve Cv isn’t available, use Fisher’s
table for approximate Cv estimation[3]:

Valve Valve Plug Flow Port Dia. Rated Cv Fl Xt Fd

Size Style Characteristic (in.) Travel
(NPS) (in.)
0.5 Post-Gulded Equal-Percentage 0.38 0.5 2.41 0.9 0.54 0.61
0.75 Post-Gulded Equal-Percentage 0.56 0.5 5.92 0.84 0.61 0.61
1 Cage-Guided Linear 1.125 0.75 20.6 0.84 0.64 0.34
1 Cage-Guided Equal-Percentage 1.125 0.75 17.2 0.88 0.67 0.38
1.5 Cage-Guided Linear 1.875 0.75 39.2 0.82 0.66 0.34
1.5 Cage-Guided Equal-Percentage 1.875 0.75 35.8 0.84 0.68 0.38
2 Cage-Guided Linear 2.3125 1.125 72.9 0.77 0.64 0.33
2 Cage-Guided Equal-Percentage 2.3125 1.125 59.7 0.85 0.69 0.31
3 Cage-Guided Linear 3.4375 1.5 148 0.82 0.62 0.3
3 Cage-Guided Equal-Percentage 3.4375 1.5 136 0.82 0.68 0.32
4 Cage-Guided Linear 4.375 2 236 0.82 0.69 0.28
4 Cage-Guided Equal-Percentage 4.375 2 224 0.82 0.72 0.28
6 Cage-Guided Linear 7 2 433 0.84 0.74 0.28
6 Cage-Guided Equal-Percentage 7 2 394 0.85 0.78 0.26
8 Cage-Guided Linear 8 3 846 0.87 0.81 0.31
8 Cage-Guided Equal-Percentage 8 3 818 0.86 0.81 0.26

• Calculate Ktotal and estimate △Psizing

∑ 𝐾 = 𝐾1 + 𝐾2 + 𝐾𝐵1 + 𝐾𝐵2 For an inlet reducer
2
𝑑2
Where, 𝐾1 = 0.5 ( )
𝐷1 2
K1 = Resistance coefficient of upstream
For an outlet reducer
K2 = Resistance coefficient of downstream
2
fittings 𝑑2
𝐾2 = 1.0 ∗ ( )
𝐷2 2
KB1 = Inlet Bernoulli coefficient
For a valve installed between identical reducers
KB2 = outlet Bernoulli coefficient
(assumed in tool)
𝑑 4 2
𝐾𝐵1 = 1 − ( ) 𝑑2
𝐷1 𝐾1 + 𝐾2 = 1.0 ∗ ( 2 )
𝐷
𝑑 4
𝐾𝐵2 = 1 − ( )
𝐷2

−1/2 𝑞
∑ 𝐾 𝐶𝑣 2 𝐶𝑣 =
𝐹𝑃 = [1 + ( ) ]
𝑁2 𝑑2 ∆𝑃𝑠𝑖𝑧𝑖𝑛𝑔
𝑁1 𝐹𝑃 √
𝜌1 /𝜌𝑜

Where N1 = 1.0 and N2 = 890, ρ0 is 1000 kg/m3 and Q in m3/hr

d is control valve size in in and D is line size in inch

From the Cv equation estimate △Psizing in bar

• Convert △Psizing to hloss in m

hloss = △Psizing 1.0197210/SG

Control valve diameter not given (not used in our tool)

• Estimate △Psizing using the following guideline [4]:

“the sum of the following pressure drops at maximum flow may be used for this purpose:

1. 20% of the friction drop in the circuit (excluding the valve). (A circuit generally includes
all equipment between the discharge of the pump, compressor or vessel and the next
point downstream of which pressure is controlled. In most cases this latter point is a
vessel)
2. 10% of the static pressure of the vessel into which the circuit discharges up to the
pressure of 200 psig, 20 psig from 200 psig to 400 psig, and 5% above 400 psig.”

𝑆. 𝐺
𝐶𝑣 = 𝑄𝐿 √
△𝑃
Where:
QL : liquid flow in gpm
△ 𝑃 : pressure drop across the valve in psi
“In order to ensure that the valve is in a controlling position at maximum flow rate, the valve Cv
is the maximum process value determined above divided by 0.8. The reasons for using this
factor are that:

1. It is not desirable to have the valve fully open at maximum flow since it’s not then in a
controlling position.
2. The valves supplied by a single manufacturer often vary as much as 10-20% in Cv.
3. Allowance must be made for pressure drop, flow rate, etc, values which differ from
design.”

Control valve sizing[4]

After estimating valve Cv, valve sizing can be estimated using the following equations:
𝑣𝑎𝑙𝑣𝑒 𝐶𝑣 1 Single-seated valves
𝑆 (𝑖𝑛𝑐ℎ𝑒𝑠) = ( )2
9
𝑣𝑎𝑙𝑣𝑒 𝐶𝑣 1 Double-seated valves
𝑆 (𝑖𝑛𝑐ℎ𝑒𝑠) = ( )2
12
𝑣𝑎𝑙𝑣𝑒 𝐶𝑣 1 Butterfly valve sizes
𝑆 (𝑖𝑛𝑐ℎ𝑒𝑠) = ( )2
20
Opening percentage = Cv desired / Cv max. for the valve

Estimating Pump Hydraulic power and Efficiency [5]

𝑄∗𝐻∗𝑆𝐺
𝐻𝑦𝑑𝑟𝑎𝑢𝑙𝑖𝑐 𝑝𝑜𝑤𝑒𝑟 = 367.46
, Q in m3/hr and H in m

Branan efficiency equation[5]

η= 80 – 0.2855F+3.78*10-4-FG-2.38*10-7FG2 + 5.39*10-4*F2-6.39*10-7*F2G+4*10-10*F2G2

Where:
η: pump percentage efficiency
F: developed head (ft)
G: flow (GPM)
Range of applicability:
F: 50 – 300 ft
G: 100 – 1000 GPM

For G: 25 – 99:
η= η (at G=100 GPM) – 0.35% * G * (100-G)
BHP = hydraulic power / ηestimated
Drawing the Hydraulic Gradient [6]

Effect of elevation

There are situations where the ground elevation may have drastic peaks and valleys that require
careful consideration of the pipeline topography. In some instances, the total pressure required
to transport a given volume of liquid through a long pipeline may depend more on the ground
elevation profile than on the actual frictional pressure drop. If we calculated the total pressure
required for a flat pipeline as 825 psi and an uphill pipeline to be 1002 psi. In the uphill case the
static elevation difference contributed to 17 percent of the total pressure required. Thus the
frictional component was much higher than the elevation component.

Tight line operation

When there are drastic elevation differences in a long pipeline, sometimes the last section of
the pipeline toward the delivery terminus may operate in an open-channel flow. This means that
the pipeline section will not be full of liquid and there will be a vapor space above the liquid.

Such situations are acceptable in ordinary petroleum liquid (gasoline, diesel, and crude oil)
pipelines compared to high vapor pressure liquids such as liquefied petroleum gas (LPG) and
propane. To prevent such open-channel flow or slack line conditions, we must pack the line by
providing adequate back pressure at the delivery terminus as illustrated in Fig. 6.15.

Hydraulic Gradient

The graphical representation of the pressures along the pipeline, as shown in Fig. 6.16, is the
hydraulic gradient. Since elevation is measured in feet, the pipeline pressures are converted to
feet of head of liquid and plotted against the distance along the pipeline superimposed on the
elevation profile. If we assume a beginning elevation of 100 ft, a delivery terminus elevation of
500 ft, a total pressure of 1000 psi required at the beginning, and a delivery pressure of 25 psi at
the terminus, we can plot the hydraulic pressure gradient graphically by the following method.

Drawing the required system curve

Single pipe System

The system curve is a function of elevation – or static head

and the major and minor losses in the system and can be
expressed by:
hsys = h + hL = dh + kQ2
where:
hsys: system head (m)
h = h2-h1 = elevation (static head) difference between the inlet
and outlet in the system (m)
hL: major and minor head loss in the system (m)
a generic expression of major and minor head loss is:
hL = kQ2 where:
Q: flow rate
k: constant describing the total system characteristics –
including all major and minor losses
Branch systems [7]

The first example of a branched pipe system is a circulation system. In such a case Hstat = 0. The
volume flow Qp, which
passes through the pump, is divided at the branch point. Continuity dictates that

Qp=Qa+Qb

Both branches A and B have their own system curves, the sum of which determines the
resultant system curve. The intersection point of the resultant system curve and the pump curve
determines the pump’s optimum operating point.

Figure 9.2a also shows how large a portion of the pump flow

flows through each respective branch. The division of flow

depending upon the magnitude of the losses in each respective

branch. In this example it is assumed that the branch point is

near to the pump i.e. that flow losses between the pump and

the branch point can be neglected.

The next example, figure 9.2b, also takes into consideration

the losses in the main pipeline up to the branch point. This is Figure 1: h is the difference between Pb and Pa in meters

done by reducing the pump curve by the magnitude of the losses up to the branch point. The
“reduced” curve is then matched to the remaining system curve as before.

In the third example the pump curve is first reduced to the branch point. Thereafter the system
curves A+B are added to the resultant system curve Hsyst A+B. The intersection point between
the reduced pump curve and the resultant system curve determines the pump operational
point, as shown in figure 9.2c.

In the fourth example there is a level difference (in this example a supply head) between the
supply vessel on the suction side and the branch point. The system curve for the supply pipeline
exhibits therefore, a static head (Hstat < 0).

The method is the same as in the previous examples. First reduce the pump curve with the
supply line curve up to the branch point. Then determine the system curves for the branches
A+B from the branch point. The resultant branch curve is then matched to the pump's reduced
curve as shown in figure 9.2d.

This method can be applied in principle to calculate any pipework system, however complicated
and containing any number of branch points. The first step is to determine the resultant system
curve for the branch furthest from the pump in relation to the furthest branch point. The next
step is to deal with the next branch point and so on until the branch point nearest to the pump
is reached.

A common problem is to dimension a piping network with many branches and many tapping
points in such a way as to guarantee a predetermined capacity at each tapping point.
Dimensioning is carried out so that the pipeline requiring the most pressure, usually the pipeline
to the furthest tapping point, is dimensioned first. The branch lines are then calculated on the
basis of the excess pressure available at the branch points. The excess pressure at the branch
points can be such that it is necessary to equip a branch line with extra throttling to prevent it
from "stealing" capacity from the other tapping points.

Figures 9.2a and 9.2b

Figures 9.2c and 9.2d

1. Pump handbook: Sectio 8.2 BRANCH-LINE PUMPING SYSTEMS

2. Pump and System Curves for Multi-branched Systems , modern pumping today
3. Reference 7: Pumpbook, Process Contact Scandinavia Chapter 9: Pump system

System Analysis of a Centrifugal pump[8]

The first step toward the goal of selecting the proper pump for a service involves four areas of
analysis:

1. Pump flow boundary conditions

2. Flow requirements
3. Fluid specifications
4. Criticality of a system
Pump flow boundary conditions

Pump boundary conditions must be defined clearly and with as much certainty as is possible. A
comprehensive knowledge of all boundary parameters is essential to correct user specifications
and vendor bid preparation.

1.1 Suction conditions

NPSHa should be determined as well as the margin between NPSHa and NPSHr. The following
questions to be considered:

• Does the calculated value for NPSHa allow for increases in system resistance in the
suction piping due to fouling?
• Where suction is from a vessel, was the minimum possible operating level used in
calculating NPSHa? The minimum level at which maximum flow is expected should
provide the limiting guidelines here.
• What margin between NPSHa and NPSHr is considered as acceptable?
• Has the maximum expected flow been used to set NPSHa?

Where NPSHa is minimal, and may increase in value will incur

a major additional cost. Pump which has an unrealistically low
NPSHr curve will result in a heavy maintenance burden and
Unreliable operation. Bearing and seal life will be reduced.
Impeller erosion will lead to lower head/flow relationship
And fall-off in efficiency
Suction specific speed should be estimated afterwards:
𝑁∗√𝑄𝐵𝐸𝑃
𝑆𝑠 = (𝑁𝑃𝑆𝐻𝑟 0.75
𝐵𝐸𝑃 )
As a general rule, Ss values above 11,000 are to be avoided
Pumps which develop high stage heads of greater than 650 ft and high Ss require greater than
normal NPSH margins. Where very high Ss pumps are are necessary to the fit the applications, a
suction backflow recirculation insert for the suction nozzle may be reviewed with the vendor.
This can avoid cavitation surge which results when high Ss pumps are operated at flows much
below BEP (normally less than 70% BEPm but dependent on new NPSH (NPSHa – NPSHr) and Ss
combinations.
The system resistance curve must be clearly defined. How much static head is built into the
pump discharge in terms of downstream pressure in a receiving vessel, or height which is
considerably in excess of normal flow? Control valve sizing of pump. A larger than normal
control valve may be be required to provide the artificial head loss at rated and minimum flow,
while still accommodating the low loss it must provide at maximum expected flow. The effect if
system resistance on maximum possible flow and required control valve head loss is shown in
figure 5. In new installations, pipe size may be increased to flatten a steeply rising system
resistance curve to accommodate greater flow flexibility.
The differential pressure which the pump will see be derived
From the system resistance curve. This must be converted to
Differential or total head (H). in arriving at H, the range of SG
Expected must be reviewed, as any lowering in SG will require
Additional pump head to meet required discharge pressure
Conditions. Accordingly, Max. Head requirement should be set
To the lowest expected SG.
While defining system resistance to accommodate normal and 1
Changing conditions, it’s also important not to be overly
conservative. Imposing too high head values on a pump
specification
for given flows will result in the pump operating much below its
BEP point, and in the lower efficiency region when these heads
prove to be lower than expected. Reliability and maintenance
costs will suffer likewise.
The static component of the discharge system resistance can also
limit maximum capacity. Where the possibility exists of an
increase in the pressure of the receiving downstream vessel, these
must be looked at in determining rated conditions.

1.2 Flow requirements

Min./Normal/rated flow should be determined. The rated flow should reflect the maximum flow
the system can envisage under current consideration, but must also consider the long range
outlook. Minimum flow requirements can conflict with rated requirements and recirculation
facilities may be required.

When determining these design flows, care must be taken to avoid an extremely conservative
approach. Higher than expected flow will result in a larger than required pump as in head
consideration. A more simple single stage, overhung pump application may require a double-
suction between bearings design under increased flow requirements.

While it’s of prime importance to define maximum and minimum flow requirements properly,
it’s also important to clarify the percentage of time at which the pump will operate at minimum,
normal and rated (or maximum) flows. Where a pump is used for two different services, the
lower flow may require excellent turndown while the higher flow will impose more stringent
NPSHa restrictions. Long term operation at lower flow can mean higher maintenance costs due
to higher bearing loads and shaft deflections.

Where complete shutoff (isolation) of a pump is an expected occasional occurrence, provision

must be made to recycle flow to prevent vapor locking , due to overheating of the trapped
liquid. Where shutoff will be for mins rather than hours, a recycle flow of 10% of BEP will suffice.
For MCF, a total flow of 30-40% of BEP is more realistic , although this can be much higher for
high Ss and/or high head (H)

The percentage of BEP flow at which discharge and suction

recirculation within the impeller geometry.

the points of low flow cavitation and high temperature rise are
only valid considerations where extremely low flows are
considered probable for short periods which may cause severe
cavitation and eventual vapor locking of the pump. Such events
will lead to to mechanical seal failure and require protection
against even short duration of one or two minutes, where
volatile
liquids close to their vapor points being pumped.
1.3 Fluid specifications
System analysis must include a clear definition of the fluid to be
pumped and show all variations expected in fluid quality.
Fluid temperature variations must be defined. This will assist in
Specifying the NPSHa and must include heating and cooling
which may occur between the vessel or drum being drawn from
the
pump suction, in the case of vapor-liquid interface.
Higher temperatures will necessitate bearing housing cooling
and
require that mechanical seal arrangements be suitably designed.
Corrosion due to chemical due to chemical attack or oxidation
Must be considered. Compatibility of materials to resist
electrolytic reaction is important.
Erosion due to high percentage of particulate matter may
cause a premature performance decline. Large particles may
necessitate an open faced impeller. It may prove necessary to
specify wear plates or a hard coating to prolong life. Wear ring
flushing from an external source may be required.
Fluid toxicity may dictate the use of dual seals as may very high
temperatures, high flammability, and/or high vapor pressures.
Carcinogenic, strongly acid or strongly alkaline fluid impose
similar needs for more elaborate sealing and often required an
external clean fluid supply.
Entrained gases may cause cavitation and may have a very
negative effect on pump’s ability to produce the required
differential pressure.
1.4 Criticality of the service
Check if pump is spared or un-spared?
Loss of flow process debits (where a specific time can be tolerated for pump outage for
maintainability needs and reparability)
Safety Consideration (where fluid is toxic, highly flammable or carcinogenic special care may be
required in materials, mechanical seals and structural areas)
Continuous operation is normally viewed as the prime criteria for sparing, but intermittent
operation may also demand a spare pump. The definition of intermittent is important. A pump
may be required to operate one week in four, but 24-hout service during that may be crucial,
requiring a spare for high reliability of service.
Liquid specification In addition to the parameters outlined in API 610 datasheet,
comment must be made on solids content, toxicity & setup
temperature
Operating and site conditions Define min., normal and maximum flows. Remember to add a
table of time of which these flows are expected to operate in
Suction pressure min., normal and maximum

The maximum discharge pressure The max. pressure that will be encountered under conditions of
max. flow and min. suction pressure
Site conditions These would influence items such as electrical or steam tracing
requirements, lubricant quality, type of lubrication, motor
protection, etc..
Performance Important performance variables such as specific speed, suction
specific speed, NPSHr, MCF, POR, AOR.

The minimum flow Whether its stable of thermal is to be specified

2. Specifications Checking

1. Desired BEP Range

2. Predicted Operating Regions (POR & AOR)
3. Predicted Efficiency
4. Predicted Motor
5. Predicted MCSF
6. Predicted NPSHr
7. Predicted/required Head rise to shutoff

1. Desired BEP range

API 610

“Rated flow shall be within the region of 80% to 110% of the best efficiency flowrate of the
pump as furnished.”

“the BEP flowrate for the pump as furnished should preferably be between the rated point and
the normal point”

Expected BEP range accordingly is from 0.91* Qrated to 1.25 * Qrated

2.1 Predicted operating regions

HI preferred operating region POR (ANSI_HI 9.6.3-2012)

Specific Speed Preferred Operating Region

Metric US customary (POR)

< 87 < 4500 Between 70% & 120% of BEP

> 87 > 4500 Between 80% & 120% of BEP

For slurry pumps handling high concentrations of abrasive solids, the generally acceptable range
of rates of flow is primarily determined by wear considerations of the wear components. Refer
to ANSI/HI 12.1-12.6 Rotodynamic (Centrifugal) slurry pumps for nomenclature, definitions,
Applications, and Operation for information.

API 610

“pumps shall have a preferred operating region of 70% to 120% of the best efficiency flowrate
of the pump as furnished.” Rated flow shall be within the region of 80% to 110% of the best
efficiency flowrate of the pump as furnished.

HI Allowable operating region AOR (ANSI_HI 9.6.3-2012)

AOR depends on a large number of factors, some of which are application-specific. This
discussion is limited to thos factors related to operating rates of flow and pump design. For the
purpose of this guideline it’s assumed that the pumped liquid is a Newtonian pure liquid with no
vapor, gas, suspended solids or abrasives.
Certain fluid mixtures may have other characteristics that affect the AOR. for example, the
minimum rate of flow when pumping a fluid that contains entrained gasses may be limited bt
gas accumulation at the eye of the impeller. These considerations are outside the scope of this
guideline;

Factors such as proper equipment selection, installation, maintenance and operation which also
greatly affect pump reliability are not factors in defining the AOR.

Bearing life will be reduced and noise, vibration and components stresses will be increased
when a pump is operated outside its POR. As a result, service life of a pump operated within the
AOR may be lower than within the POR. While the predicted bearing life will vary significantly
over the AOR. It’s recommended that the calculated bearing life be a minimum of two years of
operation in this range (basic rating life L10 equal to or greater than 17500 hours)

End suction Cent. Pumps (ASME B31.1-2012)

AOR: MCF in ANSI table below () to 120% of the flow at the best efficiency point (BEP)

Vertical in-line centrifugal pumps (ASME B73.2-2016)

AOR: MCF in ANSI table below () to 120% of the flow at the best efficiency point (BEP)

2.3 Predicted Efficiency

2.3.1 HI Predicted Efficiency (ANSI/ HI 20.3)

Steps
1. Select your pump type from the following table
2. Calculate Q (m3/s) and obtain optimum attainable efficiency
3. Calculate Specific speed and obtain Efficiency reduction due to specific speed
𝒑𝒖𝒎𝒑 𝒔𝒑𝒆𝒆𝒅 𝒙 𝑸𝟎.𝟓 (𝒎𝟑 /𝒔)
Specific speed ns =
𝑯𝟎.𝟕𝟓
4. Obtain Deviation from normally attainable efficiency
Example

Determine the normally attainable efficiency of an API-type, single-stage, end suction process
pump driven by a 50 Hz motor when pumping clear water at 30 oC given Q equals 700 m3/hr or
0.19 m3/s. H equals 130 m, NPSHa equals 7.5 m, and the customer has specified a maximum
pump rotational speed of 1500 rpm.

Assuming a speed of 1500 rpm, calculate ns:

1470 ∗ 0.190.5
𝑛𝑠 = = 17
1300.75
Because this calculated value is already below the specific speed for optimum efficiency, no
consideration will be given to selections at speeds below 1500 rpm.

* Selected pump type: F

From chart, at 0.19 m3/s and curve F, efficiency corresponding to the optimum specific speed for
API end suction process type pumps = 85.1%

Efficiency correction (reduction due to specific speed) from figure 20.3b = 3%

Predicted efficiency = optimum efficiency – efficiency reduction

Predicted efficiency = 85.1 – 3 = 82.1%

From chart (fig. 20.e) at 0.19 m3/s, the normal deviation is ±3%

Predicted Efficiency = [82.1-3 , 82.1 +3] = [79.1-85.1]

Other methods to quickly estimate efficiency

JGC Charts for Centrifugal pumps

𝑟𝑝𝑚 ∗ 𝑄𝑔𝑝𝑚 0.5

𝑆𝑆 =
𝐻𝑓𝑡 0.75
4. Predicted motor

After calculating BHP estimated for your pump at max Q multiply it by one of the following
factors mentioned in API 610

Nameplate Rating Percentage of rated pump power

Kw HP

< 18.5 < 25 125

22-55 30-75 115

> 75 > 100 110

Estimate motor efficiency from the following table and Calculate rated motor output

Motor output = BHPmax * API factor/motor efficiency

Select the nearest motor to you calculated motor output

5. Predicted MSCF

5.1 Minimum flow rate at max. impeller to avoid suction circulation (ANSI/HI 9.6.3-2012)

After calculating suction specific speed & estimating BEP obtain result from the following curve

𝒑𝒖𝒎𝒑 𝒔𝒑𝒆𝒆𝒅 𝒙 𝑸𝟎.𝟓 (𝒎𝟑 /𝒔)

𝑆𝑆𝑆 =
𝑵𝑷𝑺𝑯𝒓𝟎.𝟕𝟓

Other Method to estimate MCSF at rated impeller

Horizontal End Suction Centrifugal pumps (ASME B73.1-2012)

1. Check Approximate Hydraulic coverage table from the ASME standard and select the
appropriate pump
2. Check MCSF percentage for your selected pump in MCF table using affinity laws
𝑄𝐵𝐸𝑃_𝑛𝑒𝑤
𝑀𝐶𝐹% 𝑎𝑡 𝑟𝑎𝑡𝑒𝑑 ℎ𝑒𝑎𝑑 = 𝑀𝐶𝐹% ∗
𝑄𝐵𝐸𝑃 𝑏𝑒𝑓𝑜𝑟𝑒 𝑡𝑟𝑖𝑚𝑚𝑖𝑛𝑔

Note: This tool trims the impeller diameter to achieve the rated head, QBEP here is QBEP
for the pump after adjusting pump head for rated head not Q rated in specs input
Vertical in-line centrifugal pumps (ASME B73.2-2016)
6. Predicted/required NPSHr

• Obtain NPSH margin recommended

NPSHr Desired is either NPSHa/NPSH margin ratio or NPSHa – NSPH margin (or whichever
greater)

7. Predicted/required Head rise to shutoff

As per API 610, if the parallel operation is specified and the pumps are not individually flow
controlled, Head rise from rated point to shutoff shall be at least 10%.

Generally if specified, pump head curve shall be continuously rising from rated point to shutoff.
HI (ANSI/HI 9.6.1-2012)

Application Margins

Petroleum (Hydrocarbon) process pumps Min. margin ratio of 1.1 or a margin of 1 m

within AOR

Chemical process pumps SSS < 210 metric or 11000 US units SSS >= 210 metric

Min. margin ratio of 1.1 or a margin of 0.6 m Pumps operating within POR

Min. margin ratio of 1.1 or a

margin of 0.6 m

Pumps operating Outside POR

Min. margin ratio of 1.2 or a

margin of a m

Electric power plant pumps Power (kW/Stage) POR NPSH Ratio AOR NPSH ratio/Margin

Boiler feed < 225 1.1 (ratio) 1.3 (ratio)

Boiler feed > 225 and < 500 1.2 (ratio) 1.5 (ratio)

Condensate All 1 (ratio) 1 (ratio)

Cooling Tower All 1.1 (ratio) 1.3 (ratio)

Circulating cooling water All 1.05 (ratio) 1.0 m

Nuclear power plant pumps Power (kW/Stage) POR NPSH Ratio AOR NPSH ratio/Margin
(1.0 m minimum ) (1.5 m minimum )

Waste water (cast iron impeller) <45 1.1 (ratio) 1.2 (ratio)

Waste water (stainless steel impeller) <45 1.05 (ratio) 1.1 (ratio)

Waste water (cast iron impeller) > 45 1.2 (ratio) 1.3 (ratio)

Waste water (stainless steel impeller) > 45 1.1 (ratio) 1.2 (ratio)

Water (stainless or Al-Bronze impeller) < 75 1.05 (ratio) 1.1 (ratio)

Water (stainless or Al-Bronze impeller) > 75 1.1 (ratio) 1.2 (ratio)

Pulp and paper stock pumps SSS < 145 metric or 7500 US units SSS > 145 metric

Min. margin ratio of 1.1 throughout AOR Min. margin ratio of 1.2 or a
margin of 1.0 m

Building Services SSS < 145 metric or 7500 US units SSS > 145 metric

Min. margin ratio of 1.0 throughout AOR Min. margin ratio of 1.1 or a
margin of 0.6 m

Slurry pumps Pumps operating within AOR

Min. margin ratio of 1.1 or a margin of 0.6 m

Oil and Gas industry pumping applications

Pipeline pumps Application specific!

water injection pumps Application specific! Typical margins are set

based on :

a) Pump NPSH3 at max. expected flow

b) Min. NPSHa expected at max.
expected flow

General industrial pumping applications Pumps operating within POR Pumps operating within AOR

Min. margin ratio of 1.05 or a margin of 0.6 Min. margin ratio of 1.1 or a
m margin of 1.0 m

ASME B73.1-2012 / ASME B73.2-2016

An operating NPSH margin is necessary to ensure satisfactory operation. A minimum margin of 3 ft (0.9m) or a margin ratio of 1.2 (whichever
yields a higher NPSH requirement) should be made available. This margin should be increased if variables exist that will increase the NPSHR of
the pump
JGC NPSHr Charts[9]

NPSHr for Centrifugal pumps

NPSHr for Sundyne pumps
Hydraulic analysis for a pump system Example (2 branches) [4]
Elements of chemical process engineering (Page 49)
Calculate pump discharge pressure for the following case:
Parameter Units (converted)

Rated Capacity m3/hr 262.027

Density Kg/ m3 900
Viscosity cP 1.08
Vapor pressure Kg/cm2.a 0
Flow Rate (branch 1) m3/hr 96.3234
Supply vessel pressure Kg/cm2.g -0.98
Supply vessel static head m 13.716
Delivery pressure (branch 1) Kg/cm2.g 3.45
Delivery vessel static head(branch 1) m 0
Delivery pressure (branch 2) Kg/cm2.g -0.98
Delivery vessel static head(branch 2) m 13.716
Fittings:
Line Fitting Diameter Length Estimated Elevation Count d/D
(in) (m) Dp (m) of
Fittings
Suction Pipe Segment 10 13.715 0 0 1 0
Suction Pipe Segment 10 12.2 0 0 1 0
Discharge Pipe Segment 8 5.5 0 5.5 1 0
Discharge Pipe Segment 8 19.5 0 0 1 0
Discharge Misc./ Equipment DP 8 0 0.55 0 1 0
Discharge 1 Misc./ Equipment DP 6 0 0.41 0 1 0
Discharge 1 Pipe Segment 6 5.4 0 -5.4 1 0
Discharge 1 Pipe Segment 6 33 0 0 1 0
Discharge 1 Control Valve 4 0 0 0 1 0
Discharge 2 Pipe Segment 6 8.22 0 22 1 0
Discharge 2 Pipe Segment 6 28.65 0 0 1 0
Discharge 2 Elbow, 90°, Long-radius (R/D = 1.5), All types 6 0 0 0 4 0
Discharge 2 Valve, Gate, Full line size, Beta = 1 6 0 0 0 2 0
Discharge 2 Control Valve 3 0 0 0 1 0
Discharge 1 Elbow, 90°, Long-radius (R/D = 1.5), All types 6 0 0 0 10 0
Discharge 1 Elbow, 90°, Long-radius (R/D = 1.5), All types 6 0 0 0 8 0
Discharge 1 Valve, Gate, Full line size, Beta = 1 6 0 0 0 6 0
Discharge 1 Tee, Through, Flanged or welded 6 0 0 0 1 0
Output
Summary Table Units Book Tool
3
Flow rate m /hr 262.03 262.03
2
Differential pressure Kg/cm 6 5.64
Differential head m 66.67 62.71
Hydraulic power kW - 40.3
NPSHa m - 14.14
Suction velocity m/s -
1.43
Suction Re - 303499.42
2
Suction pipe DP Kg/cm 0.058 0.02
Suction pipe DH m 0.64 0.17
Suction pressure Kg/cm2.g 0 0.24
Pump discharge pressure 6 5.88
Density Kg/ m3 900.0 900.0
Viscosity cP 1.08 1.08
Vapor pressure Kg/cm2.a - 0.0
Flow rate (branch 1) m3/hr 96.32 96.32
Flow rate (branch 2) m3/hr 165.7 165.7
Control Valve Cv 1 176.8 92.3
Control Valve Cv 2 91.2 118.4
Solution
Suction pressure calculations
Pressure drop calculations
𝐷2 (25.4∗0.001∗10)2
𝐴=𝜋 4
= 3.14 ∗ 4
=0.05064

v = 𝑄⁄𝐴 = (262/3600)/5.064 =1.431

ρvd 900∗1.43∗0.254
Re = μ
= 1.08∗0.001
=304168.5

Using moody’s Chart, At Re =304168.5 , f = 0.01607

𝑓 ∗ 𝜌 ∗ 𝑢2 ∗ 𝐿
−5
∆𝑃𝑓 = 1.01972 ∗ 10 ∗
2∗𝑑
0.01607 ∗ 900 ∗ 1.432 ∗ 25.908
= 1.01972 ∗ 10−5 ∗ = 0.0153
2 ∗ 0.254
Psuction = Pvessel - ∆𝑃𝑓 +Helevation = 0.0532 – 0.0153 + 1.2375 = 1.2754 kg/cm2.a = 0.24 kg/cm2.g
NPSHa = Psuction – Pvapor = 14.14 – 0 = 14.14 m

Discharge pressure (branch 1) calculations

Control valve
For a control valve 4” Cv = 236
42
∑ 𝐾 = 1.5 (1 − 2 ) = 0.47
6
𝐶𝑣 = 0.8 ∗ 236 = 188.8
−1/2
∑ 𝐾 𝐶𝑣 2 0.47 18808 2 −1/2
𝐹𝑝 = [1 − ( 2) ] = [1 − ( 2 ) ] = 0.996
𝑁2 𝑑 890 4
𝑣2
h=𝑘∗ = 3.57 m
2𝑔
0.9
∆𝑃𝑐𝑜𝑛𝑡𝑟𝑜𝑙 𝑣𝑎𝑙𝑣𝑒 = 3.57 ∗ 10 = 0.321 Kg/cm2
Fittings pressure drop
Line Fitting Diameter Count K1 𝑲∞ K
(in) of
Fittings
Discharge 1 Elbow, 90°, Long-radius (R/D = 1.5), All types 6 10 800 0.2 2.376
Discharge 1 Elbow, 90°, Long-radius (R/D = 1.5), All types 6 8 800 0.2 1.901
Discharge 1 Valve, Gate, Full line size, Beta = 1 6 6 300 0.1 0.71
Discharge 1 Tee, Through, Flanged or welded 6 1 150 0.05 0.06
∑ 𝑘 = 2.376 + 1.901 + 0.71 + 0.06 = 5.047
𝑣2 1.4312
∑ℎ = 𝑘 ∗ = 5.047 ∗ = 0.527
2𝑔 2 ∗ 9.81
0.9
∆𝑃𝑓𝑖𝑡𝑡𝑖𝑛𝑔𝑠 = 0.527 ∗ = 0.047
10
Pressure drop calculations

𝐷2 (25.4∗0.001∗6)2
𝐴=𝜋 = 3.14 ∗ = 0.018232
4 4

v = 𝑄⁄𝐴 = (96.32/3600)/ 0.1524 = 1.47

ρvd 900∗1.47∗0.0254∗6
Re = μ
= 1.08∗0.001
=310618.2
Using moody’s Chart, At Re = 310618.2, f = 0.0178
𝑓 ∗ 𝜌 ∗ 𝑢2 ∗ 𝐿
∆𝑃𝑓 = 1.01972 ∗ 10−5 ∗ =
2∗𝑑
−5
0.0178 ∗ 900 ∗ 1.472 ∗ 43.8
= 1.01972 ∗ 10 ∗ = 0.0505
2 ∗ 0.1524
Pressure at tee point
△P total= Pfittings + ∆𝑃𝑓 +∆𝑃𝑐𝑜𝑛𝑡𝑟𝑜𝑙 𝑣𝑎𝑙𝑣𝑒 + Helevation + ∆𝑃𝑚𝑖𝑠𝑐 =0.047+0.0505+ 0.321 -
(5.4*0.9/10)+0.41=0.3425

Discharge pressure (branch 2) calculations

Control valve
For a control valve 4” Cv = 148
42
∑ 𝐾 = 1.5 (1 − 2 ) = 0.263
6
𝐶𝑣 = 0.8 ∗ 148 = 118.4
−1/2
∑ 𝐾 𝐶𝑣 2 0.47 118.4 2 −1/2
𝐹𝑝 = [1 − ( 2) ] = [1 − ( ) ] = 0.998
𝑁2 𝑑 890 42
𝑣2
h=𝑘∗ = 26.8 m
2𝑔
0.9
∆𝑃𝑐𝑜𝑛𝑡𝑟𝑜𝑙 𝑣𝑎𝑙𝑣𝑒 = 26.8 ∗ 10 = 2.412 Kg/cm2
Fittings pressure drop
Line Fitting Diameter Count K1 𝑲∞ Ktotal
(in) of
Fittings
Discharge 2 Elbow, 90°, Long-radius (R/D = 1.5), All types 6 4 800 0.2 0.94
Discharge 2 Valve, Gate, Full line size, Beta = 1 6 2 300 0.1 0.235
∑ 𝑘 = 1.175
𝑣2 1.472
∑ℎ = 𝑘 ∗ = 1.175 ∗ ( ) = 0.37 𝑚
2𝑔 2 ∗ 9.8

∆𝑃𝑓𝑖𝑡𝑡𝑖𝑛𝑔𝑠 = 0.033 Kg/cm2

Pressure drop calculations
𝐷2 (25.4∗0.001∗6)2
𝐴=𝜋 4
= 3.14 ∗ 4
=0.018232 m2

v = 𝑄⁄𝐴 = ((262 − 96.32)/3600)/0.008103 = 2.524225 m/s

ρvd 900∗5.68∗0.0254∗4
Re = μ
= 1.08∗0.001
=534294.2
Using moody’s Chart, At Re = 534294.2, f = 0.0168

𝑓 ∗ 𝜌 ∗ 𝑢2 ∗ 𝐿
−5
∆𝑃𝑓 = 1.01972 ∗ 10 ∗ =
2∗𝑑
0.0168 ∗ 900 ∗ 2.5242 ∗ 36.82
= 1.01972 ∗ 10−5 ∗ = 0.125
2 ∗ 0.1524
Pressure at tee point
△P total= Pfittings + ∆𝑃𝑓 +∆𝑃𝑐𝑜𝑛𝑡𝑟𝑜𝑙 𝑣𝑎𝑙𝑣𝑒 + Helevation + ∆𝑃𝑚𝑖𝑠𝑐 =
0.033+0.125+2.412+(13.72*0.9/10) + (8.22*0.9/10)= 4.55 Kg/cm2

Check the highest calculated junction point pressure

The highest pressure drop is for branch 2 and so is the calculated junction point, meaning that
branch 1 should have a higher than calculated pressure drop to equalize the estimated pressure
at junction point due to branch 2
The control valve of branch 1 should decrease its opening
Estimated pressure drop that would be added to the control valve
∆𝑃𝑎𝑑𝑑𝑖𝑡𝑜𝑛𝑎𝑙 = 𝐶𝑎𝑙𝑐𝑢𝑙𝑎𝑡𝑒𝑑 𝑃𝑗𝑢𝑛𝑐𝑡𝑖𝑜𝑛− 𝑏𝑟𝑎𝑛𝑐ℎ 2 − 𝐶𝑎𝑙𝑐𝑢𝑙𝑎𝑡𝑒𝑑 𝑃𝑗𝑢𝑛𝑐𝑡𝑖𝑜𝑛− 𝑏𝑟𝑎𝑛𝑐ℎ 1 = (4.55-
0.98+1.235)-(3.45-0.3425+0) = 4.805 – 3.11=1.695 Kg/cm2
∆𝑃𝑠𝑖𝑧𝑖𝑛𝑔 = ∆𝑃𝑠𝑖𝑧𝑖𝑛𝑔 +1.695= 2.016 Kg/cm2
𝑄
𝑁𝑒𝑤 𝐶𝑣 = = 92.3 Notes: the book approach for
∆𝑃𝑠𝑖𝑧𝑖𝑛𝑔 calculating fittings pressure
𝑁1 𝐹𝑝 √ drops is using “the equivalent
𝜌1 /𝜌𝑜
length” for every fitting
Opening percentage = New Cv / valve 100% Cv = 92.3/188.4*1.25=39.3%
Also, the book approach for
Discharge pressure (at pump discharge flange) calculations
Sizing control valves is the
Pressure drop calculations
𝐷2 (25.4∗0.001∗8)2
second method we
𝐴 = 𝜋 4 = 3.14 ∗ 4
= 0.032413 m2 mentioned earlier
v = 𝑄⁄𝐴 = (262/3600)/0.032413= 2.245338 m/s
ρvd 900∗∗0.0254∗8
Re = μ = 1.08∗0.001 =475263.2
Using moody’s Chart, At Re =475263.2 , f = 0.017
𝑓∗𝜌∗𝑢2 ∗𝐿 0.017∗900∗2.245∗25
∆𝑃𝑓 = 1.01972 ∗ 10−5 ∗ 2∗𝑑 = 1.01972 ∗ 10−5 ∗ =0.048386
2∗0.0254
Pdischarge = Pmeeting point + ∆𝑃𝑓 + ∆𝑃𝑚𝑖𝑠𝑐 + Helevation = 4.805+0.05+0+0.55+(5.5*0.9/10)=5.9
Kg/cm2.g
Hpump_head = (5.9- 0.24)*10/0.9 = 62.88 m
Pump efficiency estimation using branan equation
η= 80 – 0.2855F+3.78*10-4-FG-2.38*10-7FG2 + 5.39*10-4*F2-6.39*10-7*F2G+4*10-10*F2G2

substituting for
F = 206 ft
G = 1153.7 gpm
Gives, η = 59.95 %
Power Calculations
𝑄∗𝐻∗𝑆𝐺 262∗62.88∗0.9
Hydraulic power = 367.46 = 367.46 = 40.35 Kw
Rated power = Hydraulic power/pump efficiency = 40.35/0.5995 = 67.306 Kw

How the Tool works

1. Select number of Discharges and Complete the tables below
2. Add at least one pipe segment to the suction line and each discharge
3. If 2 or 3 discharges are selected add at least one pipe segment for the main discharge
and one segment for each branch
4. add a Control valve for each branch (or a Control valve for the main discharge in case of
1 discharge) add fittings in their order!

How

A. Using the template Excel sheet

B.
Keep in mind the requirements/ possible inputs for each fitting:

1. Pipe segments: [Diameter, Length, Elevation Change, Count of fitting]

2. Misc. / Equipment DP: [Estimated DP, Count of fittings]
3. Control Valve: [Diameter]
4. Entrance/Exit Nozzle: [Diameter, Count of fittings]
5. Reducers/Enlargements: [Diameter, D1/D2,Count of fittings]
6. Any other fittings: [Diameter, Count of fitting]

B. Using input form

NOTE:

You can remove any entry simply by selecting it and clicking “remove selected entry”
5. Check your calculations output or download the Calculations sheet
E
6. Go to the “Pump specs Checking” Tab and fill out the table below or click on “Use
pump system data” to auto fill
7. Check your output table

Note 1: you can skip the first tab and directly use the pump specs checking tab if you already have your pump
specifications
References

1. Piping calculations manual, E. Shashi Menon. Chapter 6, Oil piping systems

2. Hooper, William B. “Calculate Heads loss cause by change in pipe size” Chemical
engineering 95, no. 16,”The 2K method predicts Head losses in pipe fittings” Chem. Eng.,
p. 97.
3. Control valve Handbook fifth edition, Fisher, p. 109
4. Elements of chemical process engineering chapter 1, Basic Process Engineering
principles
5. Process Engineers Pocket Handbook, chapter 2: pumps and motors
6. Piping calculations manual, E. Shashi Menon. Chapter 6, Oil piping systems, Effect of
elevation
7. Pumpbook, Process Contact Scandinavia Chapter 9: Pump system
8. Centrifugal pump specification and selection – a system approach by stan T. Shiels
9. JGC standard practice (JGS 210-120-1-27E) : Pumps
Compressors efficiency calculations

Link: Compressor calculations

This tool was developed for the APRCO process team to quickly estimate compressor efficiency and its
anticipated power consumption. As a part of a larger project to develop what is similar to Carl Branan’s
book “process engineers Pocket handbook” these tools would allow a process engineer to quickly
calculate/estimate equipment efficiencies or sizing using standardized calculations

The aim here is to take little-known data from the field (flow, pressures, temperatures,
compositions..etc.) and use it as input for a rough estimation without having to return to the office to
use commercial software or calculations excel sheets to validate or to calculate. Additionally, These
tools may also serve as a gathered data validation tool.

Link to Centrifugal pump evaluation tool

Link to Heater Efficiency calculations tool

Nomenclature

T Absolute Temperature (K) P Absolute Pressure (kg/cm.2a)

K (Cp/Cv) isentropic exponent, Cp/Cv Pr Reduced pressure (psia)
n polytropic exponent Tr Reduced temperature (R)
ƞ𝑎𝑑 Adiabatic efficiency in decimals Pc Critical pressure (psia)
ƞ𝑝𝑜𝑙𝑦 Polytropic efficiency in decimals Tc Critical temperature (R)
BHP Brake horse power (Kw) M.wt Molecular weight (g / g. mol)
Z Compressibility factor y Component composition mol fraction
G Flow rate (kg/hr)

Input Required

Flow rate Nm3/hr

Suction pressure Kg/cm2.g
o
Suction temperature C
Discharge pressure Kg/cm2.g
o
Discharge temperature C
M.Wt / K (Cp/Cv) / Z factor -
Alternative option: Composition

Output Obtained

M.Wt / K (Cp/Cv) / Z factor -

Polytropic exponent -
Adiabatic Efficiency %
Polytropic Efficiency %
Brake horsepower Kw
Calculations

Polytropic exponent
1
𝑛=
ln 𝑇𝑑/𝑇𝑠
1−
ln 𝑃𝑑/𝑃𝑠
Adiabatic Temperature [1]
𝑘
𝑃𝑑 𝑘−1
𝑇𝑎𝑑𝑖𝑎𝑏𝑎𝑡𝑖𝑐 = 𝑇𝑠𝑢𝑐𝑡𝑖𝑜𝑛 𝑥 ( )
𝑃𝑠
Adiabatic Efficiency [1]
𝑘
𝑃𝑑 𝑘−1
𝑇𝑠𝑢𝑐𝑡𝑖𝑜𝑛 𝑥 (( 𝑃𝑠 ) − 1)
ƞ𝑎𝑑 =
𝑇𝐷𝑖𝑠𝑐ℎ𝑎𝑟𝑔𝑒 − 𝑇𝑠𝑢𝑐𝑡𝑖𝑜𝑛

Polytropic Efficiency [derived from ref. 1]

𝑃𝑑
𝑘−1 ln 𝑃𝑠
ƞ𝑝𝑜𝑙𝑦 = 𝑥
𝑘 𝑇𝑑
ln 𝑇𝑠

Brake horsepower [2]

𝑘
1 𝑘 𝑍𝑅𝑇𝑠𝑢𝑐𝑡𝑖𝑜𝑛 𝑃𝑑 𝑘−1
𝐵𝐻𝑃 = ∗𝐺∗ ∗( ) ∗ (( ) − 1)
(ƞ𝑎𝑑 ∗ 36) 𝑘−1 𝑀. 𝑤𝑡 𝑃𝑠

Physical properties calculations

Average Molecular weight [1]

𝑛

𝑀. 𝑤𝑡 = ∑ 𝑦𝑖 ∗ 𝑀. 𝑤𝑡𝑖
𝑖=1

MCp values at different temperatures for a variety of compounds are found in figure 1 [1]
Compressibility factor Z [3]

First, Tr and Pr are calculated as follows (T and P are absolute values):

𝑛

𝑇𝑐 = ∑ 𝑦𝑖 ∗ 𝑇𝑐𝑖
𝑖=1
𝑛

𝑃𝑐 = ∑ 𝑦𝑖 ∗ 𝑃𝑐𝑖
𝑖=1

𝑇 𝑃
𝑇𝑟 = , Pr =
𝑇𝑐 𝑃𝑐
Where Tci and Pci are each component’s critical temperature and pressure

𝑨𝟐 𝑨 𝟑 𝑨𝟓 𝑨𝟓 𝑨𝟔 𝝆𝒓 𝟓 𝑨𝟕 𝝆𝒓 𝟑
𝒁 = 𝟏 + (𝑨𝟏 + + ) 𝝆𝒓 + (𝑨𝟒 + ) 𝝆𝒓 𝟐 + +
𝑻𝒑𝒓 𝑻𝟑 𝑻𝒑𝒓 𝑻𝒑𝒓 𝑻𝒑𝒓 𝟑 (𝟏 + 𝑨𝟖 𝝆𝒓 𝟐 )𝒆𝒙𝒑(−𝑨𝟖 𝝆𝒓 𝟐 )

Where:
𝟎. 𝟐𝟕 ∗ 𝑷𝝆𝒓
𝝆𝒓 =
𝒁𝑻𝝆𝒓

𝑨𝟏 = 𝟎. 𝟑𝟏𝟓𝟎𝟔𝟐𝟑𝟕 𝑨𝟐 = −𝟏. 𝟎𝟒𝟔𝟕𝟎𝟗𝟗𝟎

Figure 2: Calculations of Average M.wt and k [1]

Figure 3: Standing-Katz Chart [1]

Calculations Examples using commercial software, hand written calculations and the tool developed:

Example No.1: Unknown composition, Known physical properties and operating conditions

AICHE standard [4] example (Page no. 43) (input highlighted)

Property Unit AICHE Tool

(converted Units)
Capacity Nm³ /hr 9610.1 9610.1
Mass Capacity kg/hr 18867.31 18867.315
Suction pressure kg / cm²_g 3.164 4.1976
Suction temperature °C 23.05 23.05
Discharge Pressure kg / cm²_g 11.71 12.774
Discharge temperature °C 110.2 110.2
Discharge temperature (adiabatic) °C - 91.132
Power consumption Kw 390 395.05
Gas Molecular weight g / g. mol 44.01 44.01
Z - 0.9771 0.9759
K (Cp/Cv) - 1.2709 1.2709
Compressor Pressure ratio - 2.633 2.6395
Compressor Temperature Ratio - 1.28 1.2942
The Polytropic constant - 1.3422 1.3618
The Polytropic efficiency % 83.43 80.21
Isentropic (adiabatic) Efficiency % - 78.12

Hand Calculations

𝑻𝒔𝒖𝒄𝒕𝒊𝒐𝒏 = 𝟐𝟑. 𝟎𝟓 + 𝟐𝟕𝟑. 𝟏𝟓 = 𝟐𝟗𝟔. 𝟐 𝑲

𝑻𝒅𝒊𝒔𝒉𝒂𝒓𝒈𝒆 = 𝟏𝟏𝟎. 𝟐 + 𝟐𝟕𝟑 = 𝟑𝟖𝟑. 𝟑𝟓 𝑲
𝑲𝒈
𝑷𝒔𝒖𝒄𝒕𝒊𝒐𝒏 = 𝟒. 𝟏𝟗𝟕𝟔 + 𝟏. 𝟎𝟑𝟑𝟐𝟑 = 𝟓. 𝟐𝟑𝟎𝟖𝟑
𝒄𝒎𝟐𝒂
𝑲𝒈
𝑷𝒅𝒊𝒔𝒉𝒂𝒓𝒈𝒆 = 𝟏𝟐. 𝟕𝟕𝟔 + 𝟏. 𝟎𝟑𝟑𝟐𝟑 = 𝟏𝟑. 𝟖𝟎𝟗𝟐𝟑
𝒄𝒎𝟐𝒂
𝑷𝒅 𝟏𝟑. 𝟖𝟎𝟗𝟐𝟑
𝑪𝒐𝒎𝒑𝒓𝒆𝒔𝒔𝒊𝒐𝒏 𝒓𝒂𝒕𝒊𝒐 = = = 𝟐. 𝟔𝟑𝟗
𝑷𝒔 𝟓. 𝟐𝟑𝟎𝟖𝟑
𝑻𝒅 𝟑𝟖𝟑. 𝟑𝟓
𝑻𝒆𝒎𝒑𝒆𝒓𝒂𝒕𝒖𝒓𝒆 𝒓𝒂𝒕𝒊𝒐 = = = 𝟏. 𝟐𝟗𝟒𝟐
𝑻𝒔 𝟐𝟗𝟔. 𝟐
1 1
𝑛= = = 1.3619
ln 𝑇𝑑/𝑇𝑠 ln 1.2942
1− 1−
ln 𝑃𝑑/𝑃𝑠 ln 2.639
𝑘
𝑃𝑑 𝑘−1 1.2709
𝑇𝑎𝑑𝑖𝑎𝑏𝑎𝑡𝑖𝑐 = 𝑇𝑠𝑢𝑐𝑡𝑖𝑜𝑛 𝑥 ( ) = 296.2 𝑥(2.639)0.2709 − 273.15 = 91.1322 𝐶
𝑃𝑠
𝑘
𝑃𝑑 𝑘−1 1.2709
𝑇𝑠𝑢𝑐𝑡𝑖𝑜𝑛 𝑥 (( 𝑃𝑠 ) − 1) 296.2𝑥 ((2.639)0.2709 − 1)
ƞ𝑎𝑑 = = 𝑥 100% = 78.12%
𝑇𝐷𝑖𝑠𝑐ℎ𝑎𝑟𝑔𝑒 − 𝑇𝑠𝑢𝑐𝑡𝑖𝑜𝑛 383.35 − 296.2
𝑃𝑑
𝑘−1 ln 𝑃𝑠 0.2709 0.9704
ƞ𝑝𝑜𝑙𝑦 = 𝑥 = 𝑥 𝑥 100% = 80.21%
𝑘 𝑇𝑑 1.2709 0.2578
ln 𝑇𝑠
𝑘
1 𝑘 𝑍𝑅𝑇𝑠𝑢𝑐𝑡𝑖𝑜𝑛 𝑃𝑑 𝑘−1
𝐵𝐻𝑃 = ∗𝐺∗ ∗( ) ∗ (( ) − 1)
(ƞ𝑎𝑑 ∗ 36) 𝑘−1 𝑀. 𝑤𝑡 𝑃𝑠
1 1.2709 0.9759𝑥8.3143𝑥296.2 1.2709
= 𝑥 100 𝑥 𝑥 𝑥 ((2.639)0.2709 − 1)
78.12 𝑥 36 0.2709 44.01
= 395.05 𝐾𝑤
Tool input and Output
Example No. 2 : known composition, unknown physical properties and operating conditions

Computer Methods for chemical engineers (2nd edition Page no. 78) (input highlighted)

Composition Mol. (Volume) %

Methane 80
Ethane 10
Carbon dioxide 5
Nitrogen 5

Property Unit Commercial Hand Calculations Tool

software in book
Capacity Nm³ /hr - - 116.89
Mass Capacity kg/hr 100 100 101.38223731228456
Suction pressure kg / cm²_g 2.026 300 Kpa 2.0259186389337844
Suction temperature °C 30 30 30
Discharge Pressure kg / cm²_g 9.164 1000 Kpa 9.163932129779282
Discharge temperature °C 144.5 144.4 144.4
Discharge temperature (adiabatic) °C 118.1 - 115.64785528190123
Power consumption Kw 6.54 6.69 6.619462274189613
Gas Molecular weight g / g. mol 19.44 19.4 19.44257
Z - 0.9926 1 (assumed) 0.9930121050499339
K (Cp/Cv) - 1.2662 (Isen. Exponent)* 1.31 1.26051842489981
Compressor Pressure ratio - 3.33 3.33 3.333333333333333
Compressor Temperature Ratio - 1.37 - 1.3773709384793007
The Polytropic constant - 1.3632 - 1.362274188171721
The Polytropic efficiency % 77.707 - 77.71706331465492
Isentropic (adiabatic) Efficiency % 75 75 (assumed) 74.8670063652983
NOTE: Isentropic mentioned in commercial software in performance tab
Hand Calculations

𝑻𝒔𝒖𝒄𝒕𝒊𝒐𝒏 = 𝟑𝟎 + 𝟐𝟕𝟑. 𝟏𝟓 = 𝟑𝟎𝟑. 𝟏𝟓 𝑲

𝑻𝒅𝒊𝒔𝒉𝒂𝒓𝒈𝒆 = 𝟏𝟒𝟒. 𝟒 + 𝟐𝟕𝟑 = 𝟒𝟏𝟕. 𝟓𝟓 𝑲
𝑲𝒈
𝑷𝒔𝒖𝒄𝒕𝒊𝒐𝒏 = 𝟐. 𝟎𝟓𝟗 + 𝟏. 𝟎𝟑𝟑𝟐𝟑 = 𝟑. 𝟎𝟗𝟐𝟐𝟑
𝒄𝒎𝟐𝒂
𝑲𝒈
𝑷𝒅𝒊𝒔𝒉𝒂𝒓𝒈𝒆 = 𝟗. 𝟏𝟔𝟑𝟗 + 𝟏. 𝟎𝟑𝟑𝟐𝟑 = 𝟏𝟎. 𝟏𝟗𝟕𝟏𝟑
𝒄𝒎𝟐𝒂
𝑷𝒅 𝟏𝟎. 𝟏𝟗𝟕𝟏𝟑
𝑪𝒐𝒎𝒑𝒓𝒆𝒔𝒔𝒊𝒐𝒏 𝒓𝒂𝒕𝒊𝒐 = = ≈ 𝟑. 𝟑𝟑
𝑷𝒔 𝟑. 𝟎𝟗𝟐𝟐𝟑
𝑻𝒅 𝟑𝟖𝟑. 𝟑𝟓
𝑻𝒆𝒎𝒑𝒆𝒓𝒂𝒕𝒖𝒓𝒆 𝒓𝒂𝒕𝒊𝒐 = = = 𝟏. 𝟑𝟕𝟕
𝑻𝒔 𝟐𝟗𝟔. 𝟐
1 1
𝑛= = = 1.362
ln 𝑇𝑑/𝑇𝑠 ln 3.33
1− 1−
ln 𝑃𝑑/𝑃𝑠 ln 1.377
K, Pr and Tr calculations

Gas Methane Ethane Nitrogen Carbon dioxide total

Mol wt 16.043 30.07 28.0134 44.01 Tc 190.27528 K
Pc 667 706.6 492.5 1070 Pc 47.97656 kg/cm2.a
Tc -116.66 89.92 -232.53 87.76 M.wt 19.44257 g/gmol
0 8.23 11.44 6.95 8.38
50 8.42 12.17 6.95 8.7 Tsuction 303.15 K
60 8.46 12.32 6.95 8.76 Tdischarge 417.55 K
100 8.65 12.95 6.96 9 Psuction 3.09223 kg/cm2.a
150 8.95 13.78 6.96 9.29 Pdischarge 10.19713 kg/cm2.a
200 9.28 14.63 6.97 9.56
250 9.64 15.49 6.98 9.81
300 10.01 16.34 7 10.05
mol fraction 0.8 0.1 0.05 0.05
Tc -91.5745 F Suction
Pc 682.385 Psia Tr 1.593218013
M.wt 19.44257 g/gmol Pr 0.064452933
Suction MCp 9.3505 Discharge
Ksuction 1.269672 Tr 2.19445219
Discharge
MCp 9.7135 Pr 0.212544002
Kdischarge 1.257004
Average K 1.263338
𝑘
𝑃𝑑 𝑘−1 1.26
𝑇𝑎𝑑𝑖𝑎𝑏𝑎𝑡𝑖𝑐 = 𝑇𝑠𝑢𝑐𝑡𝑖𝑜𝑛 𝑥 ( ) = 303.15 𝑥(3.33)0.26 − 273.15 = 115.6 𝐶
𝑃𝑠
𝑘
𝑃𝑑 𝑘−1 1.26
𝑇𝑠𝑢𝑐𝑡𝑖𝑜𝑛 𝑥 (( 𝑃𝑠 ) − 1) 303.15𝑥 ((303.15)0.26 − 1)
ƞ𝑎𝑑 = = 𝑥 100% = 74.8%
𝑇𝐷𝑖𝑠𝑐ℎ𝑎𝑟𝑔𝑒 − 𝑇𝑠𝑢𝑐𝑡𝑖𝑜𝑛 417.55 − 303.15
𝑃𝑑
𝑘−1 ln
ƞ𝑝𝑜𝑙𝑦 = 𝑥 𝑃𝑠 = 0.26 𝑥 1.2039 𝑥 100% = 77.71%
𝑘 𝑇𝑑 1.26 0.3199
ln
𝑇𝑠
Z calculations

Iterate through the following equation

Where:
𝟎. 𝟐𝟕 ∗ 𝑷𝝆𝒓
𝝆𝒓 =
𝒁𝑻𝝆𝒓

Till Error < 0.001 for instance.

For both suction and discharge Tr and Pr (obtained above) then average your result

(Use the Goal seek function in an Excel Sheet, Change your assumed Z till ABS(Z_assumed –
Z_calculated)*1000 = 0 for instance)

Z = 0.993
𝑘
1 𝑘 𝑍𝑅𝑇𝑠𝑢𝑐𝑡𝑖𝑜𝑛 𝑃𝑑 𝑘−1
𝐵𝐻𝑃 = ∗𝐺∗ ∗( ) ∗ (( ) − 1)
(ƞ𝑎𝑑 ∗ 36) 𝑘−1 𝑀. 𝑤𝑡 𝑃𝑠
1 1.26 0.993𝑥8.3143𝑥303.15 1.26
= 𝑥 100 𝑥 𝑥 𝑥 ((3.33)0.26 − 1) = 6.62 𝐾𝑤
74.8 𝑥 36 0.26 19.44
Tool input and output
References:

1. GPSA Engineering Data Book Section 13 (Compressors).

2. Computer methods for chemical engineering, Second Edition, Chapter 2: Fluid flow in pipes.
3. Piping Calculations manual, E. Shashi Menon. Chapter 7 Gas piping system.
4. AICHE equipment testing procedure: Centrifugal compressors: A guide to performance
Evaluation and Site testing

Link: Heat Exchanger Calculations

This tool was developed for process engineers to quickly estimate heat exchangers (single-phase gas or
liquid) efficiencies and create preliminary designs using kern, bell Delaware and NTU methods, and
access a list of useful tables on site. As a part of a larger project to develop what is similar to Carl
Branan’s book “process engineers Pocket Handbook” these tools would allow a process engineer to
quickly calculate/estimate equipment efficiencies or sizing using standardized calculations

The aim here is to take little-known data from the field (flow, pressures, temperatures,
compositions..etc.) and use it as input for a rough estimation without having to return to the office to
use commercial software or calculations Excel sheets to validate or to calculate. Additionally, these tools
may also serve as a gathered data validation tool.

Nomenclature

T Temperature (oC) P Pressure (kg/cm2.a)

ṁ Mass Flow rate (kg/hr) R Heat capacity ratio
ρ Density (kg/m3) S Temperature efficiency
Re Reynold’s number Ft Correction factor
Di Tube inside diameter (mm) µ Viscosity (Cp)
Do Tube outside diameter (mm) K Thermal conductivity (W/m.K)
L Tube Length (m) Cp Heat Capacity (Kg/Kcal.hr. oC)
U Overall Heat transfer coefficient A Exchanger Area (m2)
W/m2.K
Q Heat Duty (W and Kcal/hr) Ds Shell Diameter
Leff Effective length (m) Nt Number of tubes
De Equivalent Diameter (mm) h Heat transfer coefficient
W/m2.K
PT pitch size (mm) LMTD Logarithmic mean temperature difference

B baffle spacing (mm) Gs cross flow mass velocity m/m2.s

Np Number of tube passes C Clearance (mm)
Nb Number of baffles
Input Required
Table 1: Note 1: Input required depending on the case

Kern & Bell Delaware Rating Calculations Kern & Bell Delaware Preliminary Calculations
Flow rates kg/hr Flow rates kg/hr
o o
Inlet/outlet temperatures C Inlet/outlet temperatures C
Inlet/outlet pressures Kg/cm2.a Inlet/outlet pressures Kg/cm2.a
Average Densities [1] Average Densities [1]
Average Heat Capacities [1] Kcal/kg. oC Average Heat Capacities [1] Kcal/kg. oC
Average Heat Thermal Conductivities[1] W/m. °C Average Heat Thermal W/m. °C
Conductivities[1]
Average Viscosities [1] cP Average Viscosities [1] cP
Alternative option: Alternative option:
Liquid/gas: Composition Liquid/gas: Composition Vol% or
Vol% or mol%
Liquid oil fraction: SG. & temperatures Liquid oil fraction: SG. & mol%
- /°C
temperatures - /°C
Fouling factors W/m. K Fouling factors W/m. K
Exchanger Geometry Desired features in the design
Tube Diameter & thickness mm Bell Delaware
Pitch type and pitch -/mm Pitch type and pitch ratio -/-
Shell diameter mm Kern
Number of tubes / number of passes Assumed Uservice W/m2. °C
Tube Length mm Length mm
Baffle spacing mm Pitch type and pitch ratio -/-
Baffle cut %
Output Obtained

Kern and Bell Delaware Calculations NTU calculations

Duty KCal/hr Q W
∆TLMTD UA W/K
Ft Cr W/K
Corrected LMTD Cmin W/K
Surface Area m² Cmax W/K
Tube Heat transfer Coef. W/m²-°C effectiveness
Shell Heat transfer Coef. W/m²-°C NTU
Uclean W/m²-°C Thi °C
Udirty W/m²-°C Tho °C
Uservice W/m²-°C Tci °C
Over Design % Tco °C
Over Surface %
Shell Reynolds Number
Tube Reynolds Number
Tube Velocity m/s
Shell Velocity m/s
Shell D mm
Baffle Spacing mm
Number of baffles
Do mm
Di mm
Length mm
Number of tubes
Number of passes
Tube pitch mm
pitch type
baffle cut %
Heat Exchangers calculations

1. Shell and tube heat exchangers’ rating

Shell and tube heat exchanger rating is an essential process in evaluating the performance and
efficiency of a shell and tube heat exchanger. Heat exchanger rating involves determining the actual
heat transfer rate, pressure drop, and overall thermal performance of the heat exchanger based on
its operating conditions and design specifications.

During the rating process, several key factors are taken into account. These include the inlet and
outlet temperatures of both the hot and cold fluids, the flow rates of the fluids, the physical
properties of the fluids (such as density, specific heat capacity, thermal conductivity, and viscosity),
and the heat exchanger's geometric and design parameters (such as tube diameter, tube length,
shell diameter, and number of tube passes).

The rating process typically involves two main calculations:

1. Heat Transfer Calculation: This calculation estimates the actual heat transfer rate occurring in
the heat exchanger.
2. Pressure Drop Calculation: This calculation determines the pressure drop experienced by the
fluids as they flow through the heat exchanger.

By performing heat transfer and pressure drop calculations, the rating process allows engineers to
evaluate the actual performance of the heat exchanger and compare it against the design
specifications.

Additionally, heat exchanger rating also provides valuable insights into the heat exchanger's thermal
performance over time, considering factors like fouling, scaling, or degradation. This information
aids in determining maintenance schedules, cleaning intervals, and potential efficiency
improvements.

Rating Procedures: [4]

1. Calculate the required overall coefficient

2. Calculate the clean overall coefficient
3. Check if Uc > Ureq (If not then the exchanger isn’t suitable)
4. Calculate the dirty overall coefficient by adding shell and tube fouling factors
5. Check if Ud > Ureq (If not then the exchanger isn’t suitable)
1.1 Heat exchangers rating (Kern’s Method)[1]

Where

T1, T2: inlet/outlet of hot stream

T1, t2: inlet/outlet of cold stream

Duty (for single-phase heat exchangers)
Hot stream = ṁ.Cp(T1-T2)

Cold stream = ṁ.Cp(t2-t1)

The logarithmic mean temperature difference

𝑇 −𝑇
R=
𝑡 −𝑡
𝑡 −𝑡
S=
𝑇 −𝑡
R: Heat capacity ratio

S: Or P, Temperature efficiency

Using Fig. 18 in reference [1]. We get the correction factor Ft or from [2]

√𝑅 + 1 ∗ ln( )
𝐹𝑡 =
( √ )
(𝑅 − 1) ∗ ln
( √ )

(𝑇1 − 𝑡2) − (𝑇2 − 𝑡1)

𝐿𝑀𝑇𝐷 = ( )
ln
( )
𝐶𝑜𝑟𝑟𝑒𝑐𝑡𝑒𝑑 𝐿𝑀𝑇𝐷 = 𝐹𝑡 ∗ 𝐿𝑀𝑇𝐷
Heat Exchanger area (m²)
Area = N ∗ πD L ∗ n
Nt: no. of tubes

Do: tube outer diameter (m)

L: tube Length (m)

n: no. of units

Overall Heat transfer coefficient (required or service)

Q calculated (Kcal / ℎ𝑟)
U (kcal / m². hr. °C) =
A (m ) ∗ corrected LMTD

Overall Heat transfer coefficient

Calculated overall heat transfer coefficient
1
𝑈=
+ 𝑟𝑜 + 𝑟𝑤 + 𝑟𝑖 ∗ +( ∗ )

U = Calculated overall heat transfer coefficient (fouled) (Kcal/h-m¹C)

ho = Shellside film coefficient (Kcal/h-m¹C)
hi = Tubeside film coefficient of fluid inside tubes (Kcal/h-m¹C)
ro= Shellside fouling resistance on outside of tubes (h-m². C/Kcal)
ri = Tubeside fouling resistance on inside of tubes (h-m². C/Kcal)
rw = Resistance of tube wall referred to outside surface of tube (h-m². C/Kcal)
Ao/Ai = Ratio of outside to inside surface tubing

Shell side heat transfer coefficient & pressure drop

Equivalent diameter
4 × 𝑓𝑟𝑒𝑒 𝑓𝑙𝑜𝑤 𝑎𝑟𝑒𝑎
𝐷 =
𝑤𝑒𝑡𝑡𝑒𝑑 𝑑𝑖𝑎𝑚𝑒𝑡𝑒𝑟
/
𝐷 = , for square pitch

𝐷 = , for triangular pitch

Bundle cross flow area

𝐷 𝐶𝐵
𝐴 =
𝑃
Ds: Shell inside diameter m C: Clearance m

PT: pitch size m B: baffle spacing m

Shell Reynold’s number & friction factor

𝑓 = exp(0.576 − 0.19 ln 𝑅𝑒)
Where

400 < 𝑅𝑒 < 1,000,000

Shell side heat transfer coefficient [1]
.
𝐷𝑒 𝐷𝑒. 𝐺𝑠 𝐶𝑝. µ µ .
ℎ ∗ = 0.36 . .( )
𝑘 µ 𝑘 µ
for

2,000 < 𝑅𝑒 < 1,000,000

ho: shell side heat transfer coefficient (W/m2.K)
De: equivalent diameter (m)
k: thermal conductivity (W/m.K)
µ: Viscosity (N.s/m2)
Gs: shell side mass velocity (kg/s.m2) 𝐺 =
̇

Cp: heat capacity (j/kg.K)

µ : viscosity at wall temperature (N.s/m2)
Shell side pressure drop
𝑓𝐺 (𝑁 + 1)𝐷
∆𝑝 = 𝑛.
2⍴𝐷 𝜙
µ𝑠 0.14
Where 𝜙 = ( ) ,𝑁 = − 1 (Number of baffles) and (𝑁 + 1) is the number
µ𝑠,𝑤

of times the shell fluid passes the tube bundle. n number of units in series.
Tube side
𝜋𝑑 𝑁
𝐴 = ×
4 𝑁
Nt: no. of tubes
Np: no. of passes
Fluid mean velocity
𝑚̇
𝑈 =
𝜌 ×𝐴
ρ: fluid density kg/m3

𝑅 = , µ in N.s/m2
µ

For laminar flow

ṁ
𝐺 = (dimensionless)

Where graetz number between 10 and 10,000

µ .
𝑁𝑢 = 2 ∗ (𝐺 ) .( )
µ
For graetz number less than 10 assume N u=4.36 for uniform surface heat flux or
3.66 for constant surface temperature [1]
For Re > 2300
(𝑓/2)(𝑅 − 1000)𝑃
𝑁𝑢 =
1 + 12.7 (𝑃 − 1)

𝑓 = (1.58 ln 𝑅 − 3.28)
Pr: Prandtl number Cpµ/k
𝑁𝑢 . 𝑘
ℎ =
𝑑
Tube side pressure drop [3]

4𝑓𝐿𝑁 ⍴𝑢
∆𝑝 = 𝑛. + 4𝑁
𝑑 2
n: number of units in series
1.2 Heat exchangers rating (Bell-Delaware Method) [note]
In Bell Delaware method, the fluid flow in the shell is divided into a number of individual
streams. Each of these streams introduces a correction factor which is used to correct heat
transfer coefficient and pressure drop across the shell. This article gives step-by-step guidance on
doing heat exchanger rating analysis based on Bell-Delware method.
Note:
This section is mostly the Che Guide article with minor clarifications Link

a. Shell Side Heat Transfer Coefficient, hs

Cross flow area, Sm in m2 is the minimum flow area in one baffle space at the center of the tube
bundle. It is calculated by following equation:
DotL = Ds - (12.5+( Ds /200))
Sm = B[(Ds - DOTL) + (DOTL - Do)(PT - Do)/PT,eff ]*10-6

where, PT is tube pitch, B is central baffle spacing, DOTL is outer tube limit diameter, Ds is shell
diameter and Do is tube outside diameter. (all units in mm)

PT,eff = PT for 30° and 90° layouts

PT,eff = 0.707*PT for 45° layout

Shell side cross flow mass velocity, GS (kg/m².s) is defined as:

GS = 𝑚 /Sm ∗ 3600

where, mS is shell side mass flow rate. Shell side Reynolds number Re S is then calculated from
ReS = Do.GS / μS *10

where, μS is the shell side fluid dynamic viscosity at average bulk temperature in cP.

Shell side Prandtl number PrS is calculated as following:

PrS = CP,S . μS*1.163 / kS ∗ 3.6 ∗ 10

where, CP,S is the shell side fluid specific heat (kcal/kg.oC) and kS is the shell side fluid thermal
conductivity (W.m/k).

Colburn j-factor for an ideal tube bank is defined as:

1.33
𝑗=𝑎 . (𝑅𝑒 )

𝑎
𝑎=
1 + 0.14(𝑅𝑒 )

where a1, a2, a3 and a4 are the correlation constants listed below.
The ideal tube bank based coefficient (W/m2.k) is calculated from –

µ .
ℎ = 1.163 ∗ 𝑗. 𝐶 , . 𝐺 . (𝑃𝑟 ) . ( )
µ,

where, μS,W is shell side fluid viscosity at wall temperature.

a.1 Correction factor for Baffle Window Flow, JC

The factor JC accounts for heat transfer in the baffle windows. It has a value of 1.0 for exchanger
with no tubes in the windows.

JC = 0.55 + 0.72FC
FC = 1 - 2FW
FW = (θCTL - Sin(θCTL))/2π
θCTL = 2cos-1(Ds(1 - 2*Bc/100)/DCTL)
DCTL = DOTL - Do
where, Bc is segemental baffle cut in %. Fc is Fraction of tubes in cross flow. Fw is Fraction of tubes in
windows flow. Dctl is central tube limit diameter in mm (the diameter of the circle that passes through the
centers of the outermost tubes in the bundle)
Figure 1: Baffle geometry used for calculating the fraction, Fw, of tubes
in one baffle window
a.2 Correction factor for Baffle Leakage, JL

The correction factor JL considers the effects of the tube-to-baffle and shell-to-baffle leakage
streams on heat transfer.
DSB = 3.1+0.004* Ds
JL = 0.44(1-rS) + (1-0.44(1-rS))exp(-2.2rL) LTB = 0.4 mm
𝑟 = Ssb /(Ssb + Stb) If the longest unsupported
𝑟 = (Ssb + Stb)/ Sm
Ssb = Ds*DSB(π - 0.5θDS)*10 tube length is < 3 ft,
Stb = (π/4)((Do+LTB)2 - Do2)Nt(1-FW)*10 otherwise: 0.8 mm
θDS = 2cos-1(1 - 2Bc/100)
Figure 2: Ssb Shell-to-baffle leakage
area.
Where, Nt is number of tubes, DSB is diametral clearance between shell & baffle and LTB is
diametral clearance between tube and baffle (mm). rS is ratio of shell-baffle to total area and rL is
ratio of leakage to cross flow. θDS is Baffle window angle, Ssb is Shell to baffle leakage area and
Stb is Tube to baffle leakage area. (m2)

a.3 Correction factor for Bundle Bypass effects, J B

Bundle bypass correction factor JB accounts for the bundle bypass stream flowing in the gap
between the outermost tubes and the shell. The number of effective rows crossed in one cross
flow section, Ntcc between the baffle tips is provided by following equation.

Ntcc = (Ds/Pp)(1 - 2Bc/100)

Pp = PT 30.5/2 for 30° layout
Pp = PT / 20.5 for 45° layout
Pp = PT for 90° layout

Ratio of sealing strips to tube rows rss is provided by 𝑁𝑡𝑐𝑐 [11]

Nss = 𝑖𝑛𝑡𝑒𝑔𝑒𝑟
6
rss = Nss/ Ntcc

where Nss is number of sealing strips (pairs) in one baffle.

The bundle bypass flow area, Sb (m2) is defined as the area between the outermost tubes and the shell
at the shell centerline in one central baffle space. It is part of the cross-flow area , Sm.
Sb = B(Ds - DOTL - Do/2)*10

where, B is central baffle spacing. Correction factor J B is then calculated as following -

JB = exp(-Cj(Sb / Sm)(1 - (2rss)1/3)) for rss < 0.5

JB = 1 for rss >= 0.5
Cj = 1.35 for ReS < 100
Cj = 1.25 for ReS >= 100

a.4 Correction factor for adverse temperature gradient, J R

The factor JR accounts for the decrease in the heat transfer coefficient with downstream distance
in laminar flow.
Ntcw = (0.8/Pp)(Ds(Bc/100) - (Ds-(DOTL-Do))/2 )
NB = 1 + int((L - 2Ls - LBIn - LBOut)/B)
NC = (Ntcw + Ntcc)(1 + NB)
JRL = (10/NC)0.18
JR = 1, ReS > 100
JR = JRL + (20-ReS)(JRL - 1)/80, ReS <= 100, ReS > 20
JR = JRL, ReS <= 20
where, L is tube length, Ls is tubesheet thickness, LBIn is inlet baffle spacing and LBOut is outlet baffle
spacing. All units in mm. Ntcw is the number of tube rows crosses, NB is number of baffles and Nc is Tube
rows crossed in entire exchanger.
Ls ~ DS √P/C (mm)
a.5 Correction factor for unequal baffle spacing, J S P : Design pressure (kPa or
Kg/cm2G) of shell or tube
n1 = 0.6, ReS >= 100 side, whichever is greater.
n1 = 1/3, ReS < 100 C : Constant (578 for kPaG,
JS = ((NB-1)+(LBIn/B)1-n1 + (LBOut/B)1-n1)/((NB-1)+(LBIn/B) + (LBOut/B)) 58.3 for Kg/cm2G)
Or whichever greater than:
Shell side heat transfer coefficient is calculated as Ls = 50 mm Ds ≥ 500mm
Ls = 0.1* Ds for Ds < 500mm
hs = hIdeal(JC.JL.JB.JS.JR)
LBIn , LBOut is assumed = B
b. Shell Side Pressure Drop, ΔPs

Friction factor for ideal tube bank is calculated as following:

1.33
𝑓=𝑏 . (𝑅𝑒 )

𝑏
𝑏=
1 + 0.14(𝑅𝑒 )

where b1, b2, b3 and b4 are correlational constants listed below.

Pressure drop for an ideal tube bank is calculated from

ΔPIdeal = 2f(GS²/ρS)(μS/μS,W)0.14 Ntcc

b.1 Correction factor for Baffle Leakage, RL

RL = exp(-1.33(1+rS)rLp)
p = 0.8 - 0.15(1+rS)

b.2 Correction factor for Bundle Bypass effect, R B

RB = exp(-Cr(Sb / Sm)(1 - (2rss)1/3)) for rss < 0.5

RB = 1 for rss >= 0.5
Cr = 4.5 for ReS < 100
Cr = 3.7 for ReS >= 100

b.3 Correction factor for unequal baffle spacing, RS

RS = 0.5((B/LBIn)2-n + (B/LBOut)2-n)
n = 0.2, ReS >= 100
n = 1.0, ReS < 100

1. Pressure drop for window section, ΔPW

Following terms are calculated as -

SWG = (Ds²/8)(θDS - Sin(θDS))*10
SWT = Nt.FW(πDo²/4)*10
SW = SWG - SWT
GW = mS/3600*(Sm.SW)0.5
DW = 4.SW *10 /(π.Do.Nt.FW + θDS.Ds)
SWG Gross window area m2, SWT Window area occupied with tubes m2, SW Net Cross flow area m2, Gw Mass
velocity baffle window Kg/m².s, Dw Hydraulic diameter baffle window in m.

Pressure drop for laminar and turbulent flow is calculated.

ΔPW, Turb = NB.RL(2+0.6*Ntcw).GW²/(2.ρS)

26. 𝐺W. 𝐶 Ntcw 𝐵 𝐺W²
ΔPW, Laminar = NB.RL( + + )
ρS PT − D o D w ² ρS
ΔPW = ΔPW, Turb , ReS ≥ 100
ΔPW = ΔPW,Laminar , ReS < 100

2. Pressure drop in Central Baffle spaces

ΔPC = (NB - 1)ΔPIdeal.RL.RB

3. Pressure drop in entrance & exit baffle spaces

ΔPE = ΔPIdeal(1 + Ntcw/Ntcc).RB.RS

Shell side pressure drop in bar is calculated as following -

ΔPS = ΔPW + ΔPC + ΔPE

c. Tube Side Heat Transfer Coefficient, ht

vt =mt/(⍴t*at*3600)
Reynold's number and Prandtl number are calculated as following -
at = 10-6*Nt* π. ∗
ReT = 𝐷 .v.ρt/μt ∗ 10 Pn: number of passes
PrT =1.163* Cp,t.μt/kt

where, Di is tube inside diameter mm, v is velocity m/s, ρ t is density, μt is viscosity in cP, kt is
thermal conductivity W/m.K and Cp,t is specific heat for fluid on tube side kcal/kg.oC.

For laminar flow, ReT < 2300, Sieder and Tate correlation is used for Nusselt's nubmer.

Nu = 1.86(ReT.PrT.Di/Leff)1/3
Leff = L - 2*Ls , where Leff is effective tube length in mm for fixed tube sheet, see [2] for others

For turbulent flow, ReT > 10,000, following equation developed by Petukhov-Kirillov can be
used.

f = (1.58 ln(ReT) - 3.28)-2

Nu = (f/2)ReT.PrT/(1.07+12.7(f/2)0.5(PrT2/3-1))
For transient flow, Nusselt number can be interpolated from Nu Laminar & Nu Turbulent.

Heat transfer coefficient ht in (W/m2.k) is calculated as following - For floating head

Le = L – 2xTS- XLZ [1]
ht = Nu.(kt/Di*1000)(μt/μt, w)0.14 For U-Tube
Le = L –TS+XU[2] [3] [nozzle at/after U-bend]
d. Tube Side Pressure Drop, ΔPt Le = L –TS-50 mm [4] [nozzle before U-bend]
[1] XLZ: floating head dead space (for S or T
Tube side pressure drop in bar is calculated by following equation - type)
[2] XU: U-bend length
ΔPt = (4.f.Leff.Np/Di + 4.Np)ρt.v²/2 [3] the U-bend is included in the effective
length.
where, Np is number of tube passes. [4] 50 mm: distance between TL and the
last baffle
e. Overall Heat Transfer Coefficient, U UX = 1/3 Dotl

XLZ: obtained from table in ref [3] P. 15

Resistance due to tube wall m2.k/W is calculated as following

Rtube = Do*0.001/(2.ln(Do/Di).ktube)

where, ktube is thermal conductivity of tube material W/m.k. Overall clean heat transfer
coefficient, UClean is calculated as per below equation

UClean = 1/hS + Do/(Di.ht) + Rtube)

Overall dirty heat transfer coefficient, UDirty is calculated as per below expression

UDirty = 1/(1/UClean + fshell + ftube)

where, fshell & ftube are fouling factors for shell and tube side in m2.k/W.

Heat transfer coefficient required, URequired is calculated as following

URequired = Q /(A x LMTDcorrected)

where, Q is heat duty in W, A is heat transfer area and LMTD corrected is corrected logarithmic
mean temperature difference.

Over Surface, % = (UClean/URequired - 1)*100

Over Design, % = (UDirty/URequired - 1)*100

Accuracy of Correlations Between Kern’s Method and the Bell–Delaware’s Method [4]
Heat exchangers designs have been arrived at using the Kern’s, Tinker’s, or the Bell’s method. The
suitability and accuracy of these have been reviewed in the literature. Kern’s method cannot be applied
to a TEMA type T floating head heat exchanger without sealing strips or with unsealed pass partition
lanes. Whitley presented a study of the errors found in heat transfer coefficient and pressure drop
predictions obtained with the Kern and Bell methods. Palen and Taborek show that the Bell–Delaware
method allows the prediction of shell-side film coefficients in the range from 50% lower to 100% higher
than the real values. Table below shows a comparison of the Kern, Bell–Delaware, and the Tinker’s
methods.
Kern Bell-Delaware Tinker
1. Ease of use Simple More involved More involved
2. Resultant design Very conservative Not conservative Not conservative
3. Cost of equipment Very high Relatively low Relatively low
4. Does it account for
a. By pass and leakage streams No Yes Yes
b. Inlet and outlet baffle
spacing being different than
the central one No Yes Yes
c. Number of tube rows being
different in inlet and outlet zones
than in the center. No Yes No
d. Seal strips No Yes Yes
e. Different tube layouts and
baffle cuts No Yes Yes
f. Effectiveness of tube rows in
window by a separate
calculation. No Yes Yes
Original plots
extended.
Method remains Δp calculated lam Method remains
g. Laminar flow unchanged. differently. unchanged.
Accounts for it by
Assumes shell full number of tubes in Accounts for it by mean
h. Size of tube bundle of tubes cross flow bundle width.
i. Pressure drop in the nozzle. Yes No No
Yes (takes an
j. Δp due to gradual fouling of average
heat exchanger working unit) No Yes
5. Basis of Reynolds number Equivalent
calculation diameter Tube O.D. Tube O.D.
Cross-flow mass velocity
Geometric mean of multiplied by a factor
Cross flow over the cross-flow and for tubes in baffle
6. Which mass velocity used? tube bundle window flow. window.
Source: Petroleum Refining Design and applications, vol. 4, A. Kayode coker

1.3 NTU Method

The NTU (Number of Transfer Units) method is a widely used approach for rating shell and tube heat
exchangers. It is a simplified yet effective method for evaluating the heat transfer performance of such
heat exchangers.

The NTU method is based on the concept of heat transfer effectiveness, which represents the ratio of
the actual heat transfer in the heat exchanger to the maximum possible heat transfer. It takes into
account the temperature difference between the hot and cold fluids, the heat exchanger configuration,
and the fluid flow rates.

In the NTU method, the heat exchanger is typically represented as a series of heat transfer units.
Each heat transfer unit consists of a hot fluid stream and a cold fluid stream. The NTU value
represents the number of these heat transfer units in the heat exchanger

The NTU method involves the following steps:

1. Determining the NTU Value: The NTU value is calculated based on the heat exchanger geometry,
flow rates, and heat transfer characteristics. It can be determined using equations or graphical
methods.
2. Calculating the Heat Transfer Effectiveness: The effectiveness ( ε) of the heat exchanger is
determined using the NTU value and the heat exchanger configuration. The effectiveness
represents the ratio of the actual heat transfer to the maximum possible heat transfer. It is a
measure of how efficiently heat is transferred between the hot and cold fluids.
3. Estimating the Heat Transfer Rate: Once the effectiveness is known, the heat transfer rate in the
heat exchanger can be calculated. This involves multiplying the maximum possible heat transfer
rate (which is determined by the temperature difference between the hot and cold fluids) by the
effectiveness.
4. Estimating outlet temperatures by simple heat balance in shell and tube

It should be noted that the NTU method assumes certain simplifications and may not account for all
complex factors, such as fouling or non-uniform fluid flow distribution. In such cases, more detailed
analysis and modeling may be required.

Capacity ratio
ṁ
Capacity ratio=
ṁ

Number of transfer units

𝑈𝐴
NTU =
𝐶
Cmin: stream with the lowest mCp
Overall temperature range & Effectiveness
Overall temperature range = T −t
QMax = CMin.(THot In - TCold In)
Q=ε*Cmin *(Th1 -t c1 )
THot Out = THot In - Q /( ṁH.CpH)
TCold Out = TCold In + Q /( ṁC.CpC)
Heat Exchanger Rating Assessment [8][9]

Applications of rating can be for operational performance, for

changes in process conditions, or in process design. There are three fundamental
points in determining if a heat exchanger performs well for given operating
conditions or for a new service:
(1) What actual coefficient UA value can be “performed” by the two fluids as the
result of their flow rates, individual film coefficients ht and hs, and fouling
resistance?
(2) From the heat balance: Q =( ṁ .𝐶𝑝 ) (T1-T2)= ( ṁC .𝐶𝑝 )(t2-t1), for given area A,
and actual temperatures, required U value (UR) can be calculated.
(3) The operating pressure drops for the two streams passing through the existing
heat exchanger.
The criteria can be established for the suitability of an existing exchanger for
given or new services as two necessary and sufficient conditions:
(a) UA must exceed UR to give desired overdesign (%OD) so that the heat
exchanger can meet changing process conditions for a reasonable period of
service continuously. While UA denotes the capability of the exchanger based
on its dimension and geometry, UR indicates the requirement from the process heat
transfer.
(b) Operating pressure drops on both sides must be less than allowable pressure
drops.
When these two conditions are fulfilled, an existing exchanger is suitable for the
process conditions for which it was rated. When the process conditions undergo
significant changes, a rating should be performed to make sure the exchanger can perform the task
satisfactorily under the new conditions. You could also check design R f against Rf calculated to check
your current fouling resistance.
the simple U value analysis is backed up with pressure drop information to confirm that the
performance loss is fouling related.
Actual heat exchanger Pressure Drop[8]
As the tube wall thickness increases with fouling deposits, pressure drop measurement
must be conducted and used as the basis for pressure drop rating calculations. In
doing so, the tube wall thickness including fouling deposits are assumed and iterated
until the calculated pressure drops from the rating software converges with the
measured ones.
Typical fouled exchanger pressure drops are 1.3–2 times that of clean exchangers
(Barletta, 1998). For extreme cases, fouled exchanger pressure drops are much
higher than that of clean exchangers.
It is recommended that hydraulic calculations should be conducted in an
exchanger rating software (e.g., HTRI) as the rating software is more rigorous in
pressure drop calculations than flowsheet simulation software.
Instrumentation and Data Access[9]
Most refineries have sophisticated plant data historians that record detailed plant
operating conditions with high resolution. Electronically accessible temperature and
flow data are crucial for effective exchanger performance monitoring. Manual temperature
data entry for monitoring most exchangers is feasible in principle. However, a great
deal of effort is required to collect field temperature readings from local temperature
indicators or “temperature guns.” Manual monitoring is very difficult to maintain at any
meaningful frequency level. The most frequent surveys of this type that the authors have
seen were completed every 3 months, and then only when a team from a corporate
engineering support group provided the staff.
Ideally, data should be available for each exchanger’s shell-side and tube-side outlet
temperature and each fluid’s temperature as it first enters the heat exchanger network.
Inlet temperatures for all other exchangers can be set equal to the outlet of an upstream
connected exchanger without appreciable loss in accuracy. The exchanger outlet
temperatures must be adjusted for any bypasses. Exchangers can also be modeled by
grouping identical exchangers and identifying common configurations (two series/one
parallel, two series/two parallel, etc.). Note, however, that in such groupings some
resolution is lost as to the exact location of any fouling.
Exchanger grouping for monitoring may also be determined by the availability of
bypass and isolation valves to allow online exchanger isolation and removal from
service. In general, though, monitoring resolution should be determined by available
temperature instrumentation.
Crude feed, product rundown, and tower pump-around flow rates are generally well
measured. Ideally, flow rates should be corrected for flowing density versus design for
each meter and an overall mass balance check should be performed across the unit to
ensure flow meter consistency.
Flow splits are usually less well measured unless the splits are actively manipulated
by operators. Temperatures around an associated mixer can be used to estimate an
unmetered flow split. This is especially useful around a controlled bypass. Failing this,
valve position in conjunction with valve flow characteristics can sometimes be used to
estimate flow distribution between branches. However, this is complicated by the
unequal pressure resistance on different branches, particularly when fouling is involved.
Finally, in the absence of any of these, bypasses should be assumed closed and other flow
splits to have an equal distribution.
Alternatively, data reconciliation can be used to find best-fit flow split data, though
data reconciliation then becomes a more difficult nonlinear optimization exercise. This
can also introduce more uncertainty as the model now has extra optimization variables.
Rigorous calculation of film heat transfer coefficients requires thermal and physical
data about each stream. Distillation data and liquid specific gravities are sufficient to
synthesize the required liquid properties of density, heat capacity, thermal conductivity,
and viscosity at flowing conditions in each heat exchanger. The best approach is to use
rigorous physical property simulation to calculate these properties from composition or
crude assay information. However, standard shortcut methods can also be used in
spreadsheets.
If available, electronically sourced laboratory data should be used to characterize
crude and oil product properties. It is important to include all components when
characterizing the whole crude oil stream. Different distillation types and light ends
analysis can be combined in the process simulation and the result checked against the
crude unit material balance.

Data Quality and Data Reconciliation

Exchanger tube-side heat duties (Qt) and shell-side heat duties (Qs)
are calculated independently and compared for consistency. Shell- and tube-side heat
imbalances are not uncommon. It is rare that raw plant data give
reliable (±5%) heat balance closures for all exchangers, especially with larger preheat
train systems that may have up to 60 exchangers excluding product and pump-around
coolers.

With the right tools it may be possible to manually adjust exchanger outlet temperatures to come up
with acceptable heat balances across the preheat train system; however, this is not a trivial undertaking
due to the complexity of interactions between exchangers. The final solution is also subjective as it
involves a set of sequential decisions as to which exchanger to start with and whether to adjust the
tube- or shell-side temperature in order to achieve a heat balance. There are many possible heat
balance solutions. The question is: Which is the most representative solution?
Mathematical data reconciliation can effectively answer this question. It is a vital
component of any successful fouling monitoring application. Reconciliation is a least squares
optimization process where the objective function minimizes the deviation between reconciled, heat
balanced data and the raw plant data. Temperature, fluid flow, and flow split data can be included in the
data reconciliation model. The relative weight the reconciliation model places on different data types is
governed by assigning trust factors or typical errors to each plant measurement. Occasionally, though
not often, it may be necessary to include local exchanger bypasses in the reconciliation problem.
Given that local bypass flows are never measured, bypasses must be estimated and thus very low trust
factors (or high typical errors) should be assigned to these variables.
A measure of “acceptability” should be established for any optimized solution, to determine whether to
accept or reject the current data set. If reconciled plant data deviations on a particular plant
measurement are consistently higher than expected instrumentation errors, the flow meter or
temperature indicator should be checked.

Heat Exchanger fouling monitoring [10], read more on [8] [9]

To track the amount of fouling in an exchanger one can calculate Uo frequently (weekly) and plot Uo versus
time. The pattern of the fouling curve will usually repeat itself after each cleaning. 1/Uo = R= RCLEAN +RDIRT
If conditions have changed significantly from design condition, h i and ho can be adjusted to current
conditions using the equations from the Natural Gasoline Processing Handbook summarized in
Key Formulas. RCLEAN can then be recalculated. The current RDIRT can be compared to the design RDIRT.
Many heat exchange services are regulated by instrumentation to provide a stream to a downstream
process at a consistent temperature. This temperature is accomplished by bypassing part of the
exchanger feed around the exchanger and mixing this bypassed material with the exchanger effluent.
The degree of bypassing is regulated by a temperature controller which senses the downstream mixture
temperature. When the exchanger is clean, the maximum amount of feed is bypassed; and when the
exchanger is fouled to the maximum, no feed is bypassing the exchanger. By recording the clean
bypassing condition (percent of feed bypassed), you can easily estimate the amount of exchanger
fouling by observing the amount of feed bypassed with the partially fouled exchanger.
The following equations which can be readily derived from the
above and are often quoted in the literature:
Fraction dirty = rn/rd = (Ud / UN)[UC – Un ] / (UC – Ud )
where:
rd = RDIRT design
rn = RDIRT current (now)
UC = Clean U with no fouling factors
UD = Design U which includes fouling factors
Un = Current U (now)
3. Shell &Tube Preliminary sizing

The preliminary sizing stage focuses on establishing initial estimates and configurations before
proceeding to detailed design calculations. During preliminary sizing, several key parameters need to be
considered. These include the heat duty or the amount of heat to be transferred, the temperature
difference between the two fluids, the flow rates of the fluids, and the physical properties of the fluids,
such as density and specific heat capacity.

The goal of preliminary sizing is to determine the approximate size and layout of the shell and tube heat
exchanger, including the number of tubes, tube diameter, tube length, and shell diameter. This is
achieved by employing empirical correlations, design guidelines, and engineering experience to estimate
the required surface area for heat transfer.

Additionally, factors like pressure drop, fouling, material selection, and construction constraints are also
taken into account during the preliminary sizing phase.

The results obtained from preliminary sizing serve as a foundation for detailed design and analysis,
where more precise calculations and considerations are made to optimize the heat exchanger's
performance, efficiency, and cost.

In preliminary sizing of shell and tube heat exchangers, there are certain aspects and considerations that
are generally not accounted for. These may include:

1. Vibration Analysis
2. Detailed Pressure Drop Calculations.
3. Thermal Stress Analysis

While preliminary sizing may not incorporate these detailed analyses and considerations, it is important
to note that they are crucial in the subsequent stages of design and engineering to ensure the heat
exchanger's reliability, performance, and longevity.

3.1 Bell-Delaware Preliminary Design: Polley et al.[4][6]

G. T. Polley developed rapid design algorithms for the design of both shell and tube heat exchangers and
compact heat exchangers. These algorithms are based on the full use of allowable pressure drops of
both of the streams being contacted as the design objective and a set of simultaneous equations. In the
case of a shell and tube heat exchanger, they assumed that the best shell-side performance can be
gained by making baffle window flow velocities and bundle cross-flow velocities equal. This in turn
results in a “similarity concept” that can be used for the derivation of simple performance equations
from shell-side models. They determined the exchanger geometry
from values as follows:
1. The tube-side film heat coefficient can be directly related to the tube-side velocity and thus to
the exchanger tube count.
2. From the tube count and total surface area, the tube length can be determined.
3. The shell diameter can be calculated from the tube count.
4. Finally, with the shell diameter known, and the shell-side velocity being determined from the
shell-side film coefficient, the number of baffles and baffle spacing required within the
exchanger can be determined. They inferred that the rapid design algorithm avoids the need to
evaluate much potential geometry while ensuring the full use of the allowable pressure drop.
The only restrictions are:
The pressure drop (ΔP) referred to is that associated with flow through the exchanger bundle, as no
account is taken of any nozzle or header pressure drops. Allowance for these drops must be made ahead
of design and checked after design. However, this restriction is not considered to adversely affect the
design. Kern correlations are generally considered too inaccurate for use in modern exchanger design.
Their methodology started with a consideration of the Bell-Delaware Method as described earlier and
proceeded to consider current state-of-the-art commercial methodologies.
The algorithms used in the design have been tested with data from literature, which show that in the
case of the shell and tube heat exchanger algorithm, there appears to be the first one which makes full
use of both allowable pressure drops and thereby identifies the smallest exchanger for a given duty.
Alternatively, in the case of compact heat exchangers, a major use of such an algorithm would be the
identification of the best surface combination for a specific duty. The research showed that the basic
algorithm can be applied using the Bell-Delaware method for shell and tube
exchangers, as the approach can be further extended to even more sophisticated methods through the
use of geometrical similarity (e.g., 25% baffle cut, baffle spacing equal to shell diameter).

A = area (m2)
d = tube diameter (m)
Ds = shell diameter (m)
Ft = ΔTLMTD correction factor (non-counter-current flow)
h = film heat transfer coefficient (W/m2 oK)
K = dimensional constant, solely dependent on physical properties, volumetric flow rate, and a
single
characteristic dimension.
R = fouling resistance (m2 K/W)
Δp = pressure drop (kPa)
Subscripts
i = inside surface
o = outside surface
s = shell side
t = tube side
Source 7: Rapid Design Algorithms for Shell-and-tube and Compact Heat Exchangers, G. T. Polley
Walkthrough

a. Initialization
a. Assume an exchanger with shell diameter Ds = 387 mm baffle cut = 25%, 6 m length, 2 passes, same
tube diameter and same pitch ratio and type as specified.
b. Estimate the number of tubes that could fit in your shell diameter
c. Estimate shell Reynold’s number

θDS = 2cos-1(1 - 2Bc/100)

θCTL = 2cos-1(Ds(1 - 2*Bc/100)/DCTL)
FW = (θCTL - Sin(θCTL))/2π
SWG = (Ds²/8)(θDS - Sin(θDS)) *10
SWT = Nt.FW(πDo²/4)*10
SW = SWG - SWT
Based on the assumption that baffle spacing is the same as baffle cut area
Sm = SW
GW = mS/3600*(Sm.SW)0.5
GS = mS/Sm
ReS = Do.GS / μS*10
d. Calculate shell heat transfer coefficient and pressure drop
Sm = B[(Ds - DOTL) + (DOTL - Do)(PT - Do)/PT,eff ]*10
PT,eff = PT for 30° and 90° layouts Di = Do – 2*thickness
PT,eff = 0.707*PT for 45° layout Ltb = 0.4
GS = mS/3600*Sm
Ls= 0.1* Ds
ReS = Do.GS / μS*10
PrS = 1.163*CP,S.μS / kS Dsb = 3.1+0.004* Ds
Leff = L- 2* Ls
Dotl= Ds - (12.5+( Ds/200))
1.33
𝑗=𝑎 . (𝑅𝑒 )
𝑃
𝐷𝑜
𝑎
𝑎=
1 + 0.14(𝑅𝑒 )

µ .
ℎ = 1.163 ∗ 𝑗. 𝐶 , . 𝐺 . (𝑃𝑟 ) . ( )
µ,

where, μS,W is shell side fluid viscosity at wall temperature.

a.1 Correction factor for Baffle Window Flow, JC

JC = 0.55 + 0.72FC
FC = 1 - 2FW
FW = (θCTL - Sin(θCTL))/2π
θCTL = 2cos-1(Ds(1 - 2*Bc/100)/DCTL)
DCTL = DOTL – Do
a.2 Correction factor for Baffle Leakage, JL
JL = 0.44(1-rS) + (1-0.44(1-rS))exp(-2.2rL)
rS = Ssb /(Ssb + Stb)
rL = (Ssb + Stb)/ Sm
Ssb = Ds*DSB(π - 0.5θDS)*10
Stb = (π/4)((Do+LTB)2 - Do2)Nt(1-FW)*10
θDS = 2cos-1(1 - 2Bc/100)

a.3 Correction factor for Bundle Bypass effects, J B

Ntcc = (Ds/Pp)(1 - 2Bc/100)
Pp = PT 30.5/2 for 30° layout
Pp = PT / 20.5 for 45° layout
Pp = PT for 90° layout
rss = Nss/ Ntcc
Sb = B(Ds - DOTL - Do/2)*10
JB = exp(-Cj(Sb / Sm)(1 - (2rss)1/3)) for rss < 0.5
JB = 1 for rss >= 0.5
Cj = 1.35 for ReS < 100
Cj = 1.25 for ReS >= 100

a.4 Correction factor for adverse temperature gradient, J R

Ntcw = (0.8/Pp)(Ds(Bc/100) - (Ds-(DOTL-Do))/2 )
NB = 1 + (int)(L - 2Ls - LBIn - LBOut)/B
NC = (Ntcw + Ntcc)(1 + NB)
JRL = (10/NC)0.18
JR = 1, ReS > 100
JR = JRL + (20-ReS)(JRL - 1)/80, ReS <= 100, ReS > 20
JR = JRL, ReS <= 20

a.5 Correction factor for unequal baffle spacing, J S

n1 = 0.6, ReS >= 100
n1 = 1/3, ReS < 100
JS = ((NB-1)+(LBIn/B)1-n1 + (LBOut/B)1-n1)/((NB-1)+(LBIn/B) + (LBOut/B))

Shell side heat transfer coefficient is calculated as

hs = hIdeal(JC.JL.JB.JS.JR)

1.33
𝑓=𝑏 . (𝑅𝑒 )
𝑃
𝐷𝑜

𝑏
𝑏=
1 + 0.14(𝑅𝑒 )

ΔPIdeal = 2f(GS²/ρS)(μS/μS,W)0.14 Ntcc

b.1 Correction factor for Baffle Leakage, RL

RL = exp(-1.33(1+rS)rLp)
p = 0.8 - 0.15(1+rS)
b.2 Correction factor for Bundle Bypass effect, R B
RB = exp(-Cr(Sb / Sm)(1 - (2rss)1/3)) for rss < 0.5
RB = 1 for rss >= 0.5
Cr = 4.5 for ReS < 100
Cr = 3.7 for ReS >= 100

b.3 Correction factor for unequal baffle spacing, RS

n = 0.2, ReS >= 100
n = 1.0, ReS < 100
RS = 0.5((B/LBIn)2-n + (B/LBOut)2-n)

1. Pressure drop for window section, ΔPW

Following terms are calculated as -

SWG = (Ds²/8)(θDS - Sin(θDS)) *10

SWT = Nt.FW(πDo²/4)*10
SW = SWG - SWT
GW = mS/3600*(Sm.SW)0.5
DW = 4.SW /(π.Do.Nt.FW + θDS.Ds)
SWG Gross window area, SWT Window area occupied with tubes, SW Net Cross flow area, Gw Mass velocity baffle
window, Dw Hydraulic diameter baffle window in m.

Pressure drop for laminar and turbulent flow is calculated.

ΔPW, Turb = NB.RL(2+0.6*Ntcw).GW²/(2.ρS)
26. 𝐺W. 𝐶 Ntcw 𝐵 𝐺W²
ΔPW, Laminar = NB.RL( + + )
ρS PT − D o D w ² ρS
ΔPW = ΔPW, Turb , ReS ≥ 100
ΔPW = ΔPW,Laminar , ReS < 100

2. Pressure drop in Central Baffle spaces

ΔPC = (NB - 1)ΔPIdeal.RL.RB

3. Pressure drop in entrance & exit baffle spaces

ΔPE = ΔPIdeal(1 + Ntcw/Ntcc).RB.RS

Shell side pressure drop in bar is calculated as following -

ΔPS = ΔPW + ΔPC + ΔPE

e. Calculate tube heat transfer coefficient and pressure drop

ReT = Di.v.ρt/μt*10
PrT = 1.163*Cp,t.μt/kt

For laminar flow, ReT < 2300, Sieder and Tate correlation is used for Nusselt's nubmer.

Leff = L - 2*Ls , where Leff is effective tube length 𝑖𝑛 𝑚𝑚

Nu = 1.86(ReT.PrT.Di/Leff)1/3
For turbulent flow, ReT > 10,000, following equation developed by Petukhov-Kirillov can be
used.

f = (1.58 ln(ReT) - 3.28)-2

Nu = (f/2)ReT.PrT/(1.07+12.7(f/2)0.5(PrT2/3-1))

For transient flow, Nusselt number can be interpolated from Nu Laminar & Nu Turbulent.

ht = Nu.(kt/Di)(μt/μt, w)0.14

a. Tube Side Pressure Drop, ΔPt

ΔPt = (4.f.Leff.Np/Di + 4.Np)ρt.v²/2

f. Calculate shell and tube constants from the following equations

.
∆𝑃 = 𝑘 ∗ 𝐴 ∗ ℎ
.
∆𝑃 = 𝑘 ∗ 𝐴 ∗ ℎ

b. Solving for the allowable pressure drops

1. Solve the following non-linear equations to obtain the desired heat transfer coefficients for both
shell and tube and the required area (hs,ht and A) that can achieve your allowable pressure drops for
both shell and tubes

𝐴= Q /( LMTDcorrected)*(1/(1/hS + Do/(Di.ht) + Rtube)) + fshell + ftube)

.
∆𝑃 = 𝑘 ∗ 𝐴 ∗ ℎ
.
∆𝑃 = 𝑘 ∗ 𝐴 ∗ ℎ
Where ∆𝑃 and ∆𝑃 are shell and tube allowable pressure drops

Solve the following equations to obtain tube Reynold’s number and friction factor

f = (1.58 ln(ReT) - 3.28)-2

Nu = (f/2)ReT.PrT/(1.07+12.7(f/2)0.5(PrT2/3-1))
where Nu = ℎ /.(kt/Di)(μt/μt, w)0.14
after obtaining Re iterate this Re till ht calculated = ht solved from the previous step.

𝑣 = Di.Re.ρt/μt
at = mt/(⍴t*vt*3600)
*
Nt = at/ π. ∗

Leff =A/(tn. π.Do*10 )

For laminar flow, ReT < 2300, Sieder and Tate correlation is used for Nusselt's nubmer.
Leff = L - 2*Ls , where Leff is effective tube length
Nu = 1.86(ReT.PrT.Di/Leff)1/3

For turbulent flow, ReT > 10,000, following equation developed by Petukhov-Kirillov can be
used.

Nu = (f/2)ReT.PrT/(1.07+12.7(f/2)0.5(PrT2/3-1))
f = (1.58 ln(ReT) - 3.28)-2

For transient flow, Nusselt number can be interpolated from Nu Laminar & Nu Turbulent.

ht = Nu.(kt/Di)(μt/μt, w)0.14
if ht = ht.solved stop iteration, else change Re accordingly and start over.

Now you’ve calculated Tube length, Number of baffles, Ret and Vt and tube count.

c. Iterate!
1. Estimate Shell diameter from the tube count. If it didn’t change proceed, if changed repeat from
initialization with a new shell diameter accordingly.
2. Calculate ∆𝑃 If it didn’t change from your allowable shell pressure drop proceed, if it changed repeat
from initialization with a new baffle cut accordingly.
3. If no baffle cut % achieved ∆𝑃 start from initialization with a new number of passes.

If A, ht and ∆𝑃 as assumed/calculated then hs and ∆𝑃 should match your assumptions/requirement.

3.2 Kern Preliminary Design[4]:

1. Establish physical properties of fluids at the caloric or arithmetic mean temperature, depending
upon the temperature range and order of magnitude of the properties.
2. Establish the heat duty of the exchanger.
3. Estimate or assume a specific unit and define its size and characteristics, based upon reasonable
values of overall U and ΔTLMTD.
4. Determine the ΔTLMTD with correction if needed.
5. Assume Uservice
6. Calculate the area required.
Q calculated Kcal ℎ𝑟
A (m ) =
U kcal m . hr. °C ∗ corrected LMTD
7. Calculate number of tubes from
N = Area / πDoL ∗ n
8. Calculate the tube-side velocity based upon the calculated number of tubes per pass and the
heat balance.
9. Determine the tube-side film coefficient.
𝐷
ℎ = ℎ ∗ ( )
𝐷

10. Determine the shell-side film coefficient for an assumed baffle spacing.
a. Calculate De and Gs.
b. Calculate the Reynolds number, Re, expressed as
𝑅𝑒 = 𝐷 .𝐺 S / 𝜇

c. Calculate ho from
.
𝐷𝑒 𝐷𝑒. 𝐺𝑠 𝐶𝑝. µ µ .
ℎ ∗ = 0.36 . .( )
𝑘 µ 𝑘 µ

d. If ho appears too low, assume closer baffle spacing, up to 1/5 of the shell diameter and
recalculate Gs and ho. If this second trial is obviously too low, then a larger shell size
may be indicated; therefore, return to step 3, re-evaluate the assumed U to be certain
that is attainable.
11. If the ho appears to have possibilities of satisfying the design, continue to a conclusion by
assuming the tube-side and shell-side fouling and calculate U overall.
12. Compare values calculated in steps 10 and 5. If the calculated Uoverall is too small, re-assume a
new lower U for step 5 or try closer baffle spacing in step 9 but do not get baffles closer than 1/5
the shell I.D.
13. Calculate the percent of over Design. A reasonable figure is 10–20%.
14. Calculate the shell-side pressure drop. If p is too high, reassume U (step 3).
15. Calculate the tube-side pressure drop.

If the tube-side pressure drop exceeds a critical allowable value for the process system, then recheck by
either lowering the flow rate and changing the temperature levels or reassuming a unit with fewer
passes on tube-side or more tubes per pass. The unit must then be rechecked for the effect of changes
on heat transfer performance. The following figure illustrates the influence of various geometrical
parameters on heat exchanger heat transfer and pressure drop.
Need to Increase U

Increase hi and ho Increase A

Decrease Increase Ds with Employ multiple

Increase number Increase L appropriate shells in series or
of tubes. baffle spacing
number of tubes parallel
Decrease Do or baffle cut

Increase F

Use counter flow configuration

Use multiple shell configuration

Need to reduce pressure drop

Tube side Shell side

Decrease Increase Do Decrease L Use

and Increase Increase Increase
number of Increase double or
Ds and baffle tube
tube passes
baffle cut triple
number of spacing pitch
segmental
tubes
baffles

Source 4: Petroleum refining Design and applications handbook, vol 4

Source 4: Petroleum refining Design and applications handbook, vol 4
Quick notes on Heat exchangers’ Sizing [7] Tube Design
Construction  A good design will make full use of
1. Fixed tubesheet allowable pressure drop as this
 Low cost will yield the highest heat transfer
coefficient.
 Leakage of shell-side is minimum
 h = 0.027(DGs/μ)0.8(Cp.μ)0.33(k/D)
 Bundle is fixed into shell and can’t be
removed; the outside of the tubes can’t  h α μ-0.47
be cleaned mechanically.  h α k0.67
 Limited to clean services  h α Gs0.8
 Not suitable for large temperature  ∆P α Gs2
differentials between the tubes and  Increasing mass velocity increases
shell. h but rapidly also increases ∆P
2. U-Tube  Very high velocity may lead to
 The lower cost for the single tube-sheet erosion, liquid min velocity 1 m/s
is offset by the bending of the tubes and while max. is 2.5-3 m/s.
larger shell diameter.  If number of passes is increased
 Insides of tubes can’t be cleaned from 2 to 4 (doubled), velocity
effectively. also doubles ∆P2 = ∆P1*23
 Should not be used for dirty tube  With an increase in the number of
services inside tubes tube passes, h increases but tube-
3. Floating Head side pressure drop rises steeply.
 most versatile and costliest  Other options is to increase ht is to
 permits free expansion of tubes increase number of tubes,
 can be used for both shell and tube dirty decrease tubes diameter decrease
services tube length while increasing shell
diameter and number of tubes.
 prone to leakage, limited where shell-
side is non-hazardous and non-toxic with  Tubes smaller than ¾“ shouldn’t
moderate pressure and temperatures be used for fouling services. The
use of smaller diameter ½” only
warranted for small shell areas 20-
30 m2 .
Step-wise calculations for viscous liquids arrangement where a large
temperature cross can be achieved.
 When the variation in tube-side
inlet/outlet viscosity is pronounced, a  TEMA J: a divided-flow shell
single point calculation for ht and ∆Pt wherein the shell-side fluid enters
gives unrealistic results. the shell at the center and divides
into two halves, one flowing to the
 Particularly true when variation in cases
left and the other to the right and
where a combination of turbulent and
leaving separately, they’re then
laminar flow (Re variation hence ht
combined into a single stream( J 1-2
variation).
shell). Alternatively, the stream
 In such cases it’s advised to perform
may split into two halves that enter
step-wise calculations.
the shell at the two ends, flow
Shell-side Design toward the center and leave as a
Shell selection single stream ( J 2-1 shell)
 TEMA G: shell is a split-flow shell.
 TEMA E: most common, shell-side enters
This construction is usually
from one end and leaves from the other.
employed for horizontal thermo-
 TEMA F: two-pass shell that has a syphon reboilers. There is only a
longitudinal baffle that divides shell into central support plate and no
two passes, shell-side fluid enters at one baffles.
end, traverses the entire length of the
 Can’t be used with heat exchangers
exchanger through one-half the shell
with tube lengths higher than 3 m,
cross-sectional area, turns around and
since this would exceed the
flows through the second pass, then
maximum unsupported tube length
finally leaves at the end of the second
limit specified by TEMA.
pass.
 TEMA H: When larger tube length is
 Used for temperature-cross
needed.
situations – that is, the cold stream
 Basically two G shells placed side by
leaves at a temperature higher than
side, so that there are two full
the outlet temperature. If two-pass F
support plates. Described as double
shell has only two tube passes, this
split configuration as the flow is
becomes a true countercurrent
split twice and recombined twice.
 Invariably employed for horizontal  Rotated triangular pattern offer
thermosyphon reboilers. no valuable advantage over a
 Pressure drop is drastically less and there triangular pattern
are no cross baffles for both H and G.  Square pattern is typically
 TEMA X: pure cross-flow shell where the employed for dirty services.
shell side fluid enters, flows across the  Produces less turbulence hence
tubes and exits from the opposite side of lower hs. when Re is less than 2,000
the shell. rotated square is employed as it
 Pressure drop is extremely low. offers higher turbulence
 Employed for cooling or condensing vapors
Tube pitch
at low pressure, particularly vacuum. Full
support plates can be located if needed for  For triangular pattern, minimum
structural integrity. pitch ratio recommended by
 TEMA K: employed for kettle reboilers. TEMA is 1.25.
 Has an integral vapor-disengagement  For square pattern, minimum
space embodied in an enlarged shell. pitch ratio recommended by
 Full support plates can be located if TEMA is 1.25 or pitch = OD + 6
needed for structural integrity. mm whichever larger.
 Minimizing tube pitch minimizes
Tube layout patterns shell diameter and higher tube
 Triangular (or rotated triangular) will pitch reduce shell side pressure
accommodate more tubes than square (or drop.
rotated square.
Baffling
 Triangular pattern produces more
turbulence hence higher hs. Baffle spacing
 However, at typical tube pitch ratio of 1.25  Minimum baffle spacing specified
it doesn’t permit mechanical cleaning of by TEMA is 1/5 shell diameter or
tubes, since access lanes are available. 2” (50.8 mm) whichever greater.
 Triangular pattern is limited to clean shell-  Closer baffling causes poor bundle
side services but may used for dirty shell- penetration by the shell-side fluid,
side services provided that chemical difficulty in mechanical cleaning
cleaning is suitable and effective. and poor stream distribution.
Baffle cut
 Maximum baffle spacing is shell  Baffle cut can vary between 15 -
inside diameter. Higher baffle 45 % of the shell inside diameter.
space leads to predominantly  It’s recommended to employ
longitudinal flow, which is less baffle cut between 20 and 35%.
efficient than cross flow, and  Horizontal baffle cut is
larger unsupported tube spans, recommended for single phase
which makes the exchanger prone fluids on shell-side as this
to tube failure due to flow- minimizes accumulation of
induced vibration. deposits at the bottom of the shell
 hs α v0.6-0.7 for Re > 1,000 and also prevents stratification.
 ∆P α v1.7-2.0  In case of two pass shell (TEMA F),
 hs α v0.33for Re < 1,00 a vertical cut is preferred for ease
 ∆P α v1.0 of fabrication and bundle
 When a baffle space in reduced, assembly.
higher velocity resulted leads to
Equalize cross-flow and window
reduction in pressure drop higher
velocities
than the ratio of heat transfer
increase.  The window velocity and cross-
 Optimum ration of baffle cut to velocity should be as close as
shell inside diameter is between possible – preferably within 20%
0.3 and 0.6 of each other.
 If they differ more than that ,
repeated acceleration and
deceleration take place along the
length of the tube bundle,
resulting in inefficient conversion following parameters will affect
of pressure drop to heat transfer stream analysis and therby the
shell-side performance:
Shell-side stream analysis
1. Baffle spacing and baffle cut
2. Tube layout angle and tube
pitch
3. Number of lanes in the flow
direction and lane width
4. Clearance between the tube
and the baffle hole.
5. Clearance between the shell ID
and the baffle
 Stream A: fairly efficient, because 6. Location of sealing strips and
the shell side is in contact with the sealing rods
tubes.  Very low baffle spacing (or low
 Stream B: highly effective (main baffle cut%) tends to increase the
stream) leakage and by-pass streams as it
 Stream C: Fairly efficient, as increases resistance to main
stream C is in contact with the stream B.
peripheral tubes around the
Reducing ∆Ps by modifying baffle
bundle.
design
 Stream F: Fairly efficient, as
stream F is in contact with the Single pass shell and single
tubes along the pass-partition segmental baffles
lanes.  If ∆Ps is too high even after
increasing baffle space and baffle
 Stream E: Almost encounters no
cut% to highest values
heat transfer at all since stream
recommended, double segmental
flows along the shell wall where
baffles should be considered.
there are no tubes.
 Since the flow fraction depend Single pass shell and double
strongely upon the path segmental baffles
resistances, varying any of the
 Shifting from single segmental  Offers 4% reduction of an identical
baffle to double segmental baffle single pass exchanger but also
directly reduce cross flow velocity 40% reduction in hs.
to half, because shell flow is
No-tube-in-window segmental
divided into two parallel flows,
baffle
this greatly reduces ∆Pc.
 However, ∆Pw can’t be reduced  As baffle spacing increases to
appreciably (assuming that reduce ∆Ps, the heat exchanger
maximum recommended baffle becomes more prone to tube
cut is already reached for the failure due to flow-induced
single segmental baffle cut. But vibration.
overall there is a significant  Double segmental baffles are less
reduction in ∆Ps. likely to experience such
 There will be also a reduction in hs problems.
as cross velocity decreases but a  If these problems persist, No-
reduction lower than ∆Ps. tube-in-window segmental baffle
could be used so that the
Divided-flow shell and single- unsupported span is the baffle
segmental baffles
spacing.
 If allowable ∆Ps can’t be reached  The tube-in-window requires a
even with double segmental larger shell diameter for a given
baffles. A divided flow shell (TEMA number of tubes, escalating the
J) with single segmental baffles cost by 10%.
should be considered next, it  The higher cost is offset by the
should have approx. 1/8 ∆Ps of an higher hs since pure cross flow (no
identical single pass exchanger parallel co-current) is more
(1/23). efficient.
 Offers larger reduction in ∆Ps than
double segmental baffle as
velocity is reduced both in
window and in cross flow.
Divided-flow shell and double-
segmental baffles
Cross flow shell since there are no other
parameters that can be modified.
 When ∆Ps is very low TEMA X is
used where pure cross flow is Mean Temperature difference
employed with very low velocities.
 Ft for pure countercurrent flow =1,
 Support plates will be needed to
1-2 shell and tube heat exchanger
meet TEMA requirements and
is normally lower as some of flow
prevent any induced tube
is co-current.
vibration and since they’re parallel
 Ft is recommended to be not less
to shell-side flow ∆Ps isn’t increase
than 0.8.
Increasing Tube pitch  Increasing number of shells
 As far as thermal hydraulics are increase Ft, multipule shells must
concerned, The optimum tube- be used if temperature cross
pitch-to-tube-diameter ratio for exists (cold steam outlet is higher
conversion of pressure drop to than hot stream outlet).
heat transfer is 1.25-1.35 for  It’s important to realize that LMTD
turbulent flow and 1.4 for laminar and Ft factor concept assumes
flow. that there is no significant
 Increasing tube pitch to reduce variation in the overall heat-
∆Ps isn’t generally recommended transfer coefficient along the
as: length of the shell. However,
1. It increases shell diameter there are some services where
(additional cost) this is not true (e.g.: cooling of a
2. Reducing ∆Ps by increasing viscous liquid, phase change
baffle spacing, baffle cut or applications…etc.) in this case, the
changing shell type will result simplistic overall MTD approach
in a cheaper design. “End points” will be inaccurate
and the exchanger must be
 In case of X shells, it may be
broken into several sections and
necessary to increase the tube
the calculations performed zone-
pitch above the TEMA minimum
wise “weighted average”.
to meet pressure drop limitations,
Temperature profile distortion  Since each of the shell-side
streams has a different
 effectiveness for heat transfer, an empirical equation was
and since mixing developed for δ of the form:
of these streams in the shell is
incomplete, each stream exhibits a
different temperature profile
along the exchanger.
Where,
 The cross-flow stream (B) is the
Ta = inlet temperature of shell-
most effective for heat transfer,
side fluid
and therefore its temperature
profile lies closest to that of the Tb = outlet temperature
tube-side fluid, resulting in the shellside fluid
lowest driving force Ta = inlet temperature of tube-
for heat transfer. The least side fluid
effective stream for heat transfer mE/mo = fraction of stream E
is the shell-to-baffle leakage ReB = Reynold’s number for the
stream (E), which also experiences B stream
the least amount of mixing with  Conditions favoring high
the other streams. Hence, it leakage flow rates and low
experiences the largest driving Reynolds numbers (hence,
force for heat transfer. low δ) are high shell-side fluid
 The net result of these different viscosity and small baffle
temperature profiles is that the
spacing.
mean temperature difference in
the exchanger can be less than  The effect of temperature
that calculated using the profile distortion can be
logarithmic mean temperature mitigated by using multiple
difference (LMTD) correction shells in series because the
factor, which does not take shell-side fluid becomes well
account of these effects. This mixed as it exits one shell and
phenomenon is referred to as enters the next.
temperature profile distortion.  The minimum recommended
△Tmean = δF(△Tln)cf δ is 0.75. Below this, two or
δ ranges from 0 to 1.0. more shell in series must be
Q = UAδF(△Tln)cf employed
Other factors
“β” is the baffle cut factor, which accounts for non-crossflow patterns
caused by too small or too large a baffle cut. If the baffle cut is not
adequacy, the fraction of crossflow is decreased. It is recommended that β
is more than 0.9, and it should be checked by the HTRI program.
Another two corrections hereunder are considered. Refer to HTRI Design
Manual.
(*1) tube row effect, accounts for number of tube rows in crossflow.
(*2) boundary layer overlap correction in single-phase laminar flow for
finned tubes.
Useful tables for your trials[5]
r2 = (f). r1[2] ∆P2 = (f). ∆P2[2]
Variable[1] Flow Regime
Shell (f) Tube (f) Shell (f) Tube (f)
0.27
Viscosity Turbulent (μ2 /μ1) (μ2 /μ1)0.47 (μ2 /μ1) 0.15
(μ2 /μ1)0.2
Viscosity bulk to 𝜇 𝜇 .
Streamline . (μ2 /μ1)
wall correction 𝜇 𝜇
Thermal
Turb. Or Streamline (k1 /k2)0.67 (k1 /k2)0.67
conductivity
Sp. heat Capacity Turb. Or Streamline (Cp1 /Cp2)0.33 (Cp1 /Cp2)0.33
Mass velocity or ṁ Turbulent (G1 /G2)0.6 (G1 /G2)0.8 (G2 /G1)1.85 (G2 /G1)1.8
Streamline (G1 /G2)0.6 (G1 /G2)0.33 (G2 /G1)
Density Turb. Or Streamline (ρ1/ ρ2) (ρ1/ ρ2)
Tube diameter Turbulent (Do2 / Do1)0.4 (Di2 / Di1)0.2 (Do1 / Do2)0.15 (Di1 / Di2)1.2
Tube diameter Streamline (Di2 / Di1)0.33 (Di1 / Di2)2
Tube length Streamline (L2 /L1)0.33
Tube passes Turb. Or Streamline (Np2/ Np1)

No. baffle spaces Turb. Or Streamline SP2/SP1

No. tube rows
RC2/RC1
crossed[3]
Source 9: GPSA section 9: heat exchangers

[1] Use consistent units for any one variable in both cases
[2] F is the ratio of the new value to the old value for a given variable. The overall f is the
product of the individual fs.
[3] Number of rows of tubes exposed to cross flow (as opposed to parallel flow). This
Number is determined by baffle and bundle geometry.
[4] (hi/h2) = r2/ r1
.
ℎ 𝑟 𝜇 . 𝑘 . 𝐶 𝐺 . 𝐷 .
= = . . . .
ℎ 𝑟 𝜇 𝑘 𝐶 𝐺 𝐷
. . .
∆𝑃 𝜇 𝜌 𝑁 𝐺 𝐷
== . . . .
∆𝑃 𝜇 𝜌 𝑁 𝐺 𝐷

Note: Check for the lowest h in your calculations it should be the dominating factor
and the most factor that would influence your overall U.
Methods to improve the flow fraction
To increase “B” fraction, “A”, “C”, “E” and “F” fractions should be decreased by the following methods:
Method ＼ Fraction A C E F
Enlargement of baffle spacing Effective Effective
Increasing baffle cut Effective Effective
Decreasing tube-to-baffle clearance(*1) Effective
Installation of sealing strips Effective
Installation of seal rods Effective
Change of tube passes arrangement Effective
Change of baffle cut orientation Effective
NOTE(*1): Refer to TEMA RCB-4.2 with regard to standard its clearance
The recommended limitation about flow fractions are shown hereunder:
B stream : For turbulent flow; Min. 50%, For laminar flow; Min. 40%
C stream : Max. 10%
E stream : Max. 15%
F stream : Max 20%, preferably 15% or below
Guidelines to be kept in mind
The dynamic head, ρV2(*1), is a limitation to prevent or minimize erosion of the tube bundle,
and is used as a guideline for the necessity for impingement protection devices. Impingement
protection devices should be provided when ρV2 through the inlet nozzle exceeds the
following:
Non corrosive, non-abrasive, single-phase fluid: 2230 (kg/m-sec 2)
All other liquids, including a liquid at its bubble point: 740 (kg/m-sec 2)
Note(*1): ρ is the fluid density, and V is the linear velocity of fluid. For two-phase fluid, the
mean density should be calculated assuming a homogeneous vapor-liquid mixture.
The allowable baffle cut range for each baffle style is as follows:
Segmental: 10～49% (single phase: recommended 25%, 10-15 accepted)
Double segmental: 10～30%
Triple segmental: 10～40%
Segmental NTIW: 15～30%
Minimum spacing[2]
Segmental baffles normally should not be spaced closer than 1/5 of the shell inside diameter, or
50 mm, whichever is greater. This is the design practice specified by TEMA.
Max. spacing[2]
Baffle or support plates should be spaced such that the unsupported tube span does not exceed
the values
Table: TEMA Maximum Unsupported Tube Spans

Tube materials and temperature limits (°C)

Carbon Steel / Other alloy (400) Aluminum, Aluminum alloys
Tube O.D. Low alloy (455) Copper, Copper alloys
Copper-Nickel (316) Titanium, Titanium alloy
Nickel (454), Ni-Cr-Fe Alloy (538) (at code max. allow temp.)
19.0 or 19.05mm 1,525 mm 1,320 mm
25.0 or 25.4 mm 1,880 mm 1,625 mm
The effective temperature difference △TLMTD[2]
The effective temperature difference should be greater than 5.6 "C (10°F)
Minimum temperature approach
Recommended minimum temperature approach for shell and tube exchangers is 10 oC
Cooling water (Cooling tower water and sea water) [2]
(a) Cooling water is usually routed through the tube side of an exchanger
(b) The velocity of the cooling water in the shell side should be 0.6m/sec or more. A
higher velocity is desirable, but exchangers should be designed without causing flow-
induced vibration problems.
Tube Material Cooling Tower Water (CTW) Sea Water (SW)
Carbon steel Stainless steel 0.9~ Not used
Admiralty brass Aluminum brass 0.9∼2.4 0.9∼2.4 (*1)
Aluminum bronze Not used 0.9∼3.0
Copper-nickel 70 - 30 0.9∼4.6 0.9∼2.9
Copper-nickel 90 - 10 0.9∼3.7 0.9∼2.0
Monel 1.8∼ 1.8∼
Titanium 1.2∼ 1.2∼
Table 1: Recommended Cooling Water Velocity in Tubes (m/sec)

NOTE(*1): If sea water is not treated with an effective iron ion, the maximum water
velocity should be less than the following conditions: 1.2m/sec for admiralty brass,
1.8m/sec for Aluminum brass. Engineers should confirm the sea water supply
conditions.
(c) To reduce the tendency to foul, the maximum allowable outlet temperature for
cooling water is limited as follows:
SW: 50°C
CTW: 60°C
(d) Another important criterion is the film temperature on the tube surface including a
fouling layer. If the film temperature exceeds the following conditions, severe fouling
could result.
SW: 65°C
CTW: 70°C
Slurry handling[3]
25.0 mm O.D. or larger tubes should be used to prevent plugging of the tubes with
slurry “chunks”, and not less than 12 BWG thickness because of the possibility of
erosion that may occur at the higher velocities.

Velocity limitations (m/sec)

Max. Min
2.44 1.37
Tube side velocities[3]
 Tube side velocity for most materials and services should be held between 1.3 – 2.5
m/s
 Belo 1 to 1.2 m/s fouling may be excessive, much above 2.5 m/s erosion can become
a problem
Design margin[3]
10% design margin is recommended

4. Tool assumptions
1. Flow is steady and isothermal, and fluid properties are independent of time.
2. Fluid density is dependent on the local temperature only or is treated as constant
3. The pressure at a point in the fluid is independent of direction.
4. There are no energy sinks or sources along the streamline; flow stream mechanical
energy dissipation is idealized as zero.
5. The friction factor is considered constant with passage flow length.
6. Tube wall temperature viscosity effect is negligible.
7. Single phases for both shell and tube (two phases to be added soon)

For Bell-Delaware method:

1. Leff = L - 2*Ls , where Leff is effective tube length in mm for fixed tube sheet
2. Ls = 0.1* Ds
𝑁𝑡𝑐𝑐
3. Nss = 𝑖𝑛𝑡𝑒𝑔𝑒𝑟
6
4. DSB = 3.1+0.004* Ds
5. LTB = 0.4 mm
6. DotL = Ds - (12.5+( Ds /200))
References
1. Process Heat transfer, Kern.
2. JGC standard practice “Shell & tube Exchangers” JGS 210-120-1-24E
3. TOTAL process engineering design manual
4. Petroleum Refining Design and applications, vol. 4, A. Kayode Coker
5. GPSA Engineering Data Book Section 9 (Heat exchangers)
6. Rapid Design Algorithms for Shell-and-tube and Compact Heat Exchangers, Graham
Polley.
7. Effectively design shell-and-tube heat exchanges, Rajiv Mukherjee.
8. Energy and Process Optimization for the Process Industries, Frank Zhu.
9. Energy Management and Efficiency for the Process Industries, Alan P. Rossiter, Beth P.
Jones.
10. Applied Heat Transfer and Heat Exchangers, Aramco Engineering Encyclopedia
11. Process Heat Transfer Principles and Applications, Robert W. serth.
Other Recommended Readings and resources

1. Chemical process engineering by A. Kayode Coker vol 2 (present the manual calculations
of heat exchangers in different cases and compares results with commercial software.
Overall, an excellent book, worth checking out and should be added to your library)
2. Monitoring the Thermal Efficiency of Fouled Heat Exchangers: A Simplified Method, M.
A. S. Jerónimo et al
3. Data Reconciliation and Fouling Analysis in Heat Exchanger Network, Ahmad Nuruddin
bin Abdul Aziz.
4. Optimization of operating conditions for mitigating fouling in heat exchanger networks,
C. Rodriguez and R. Smith.
Resources
1. Bell-Delaware Calculations excel sheet
2. Kern Calculations excel sheet
3. NTU calculations excel sheet
Preliminary Design Example
Kern Method [4] page: 291
Use Kern’s Method to obtain a preliminary design for the following conditions
Kerosene API 42, 25,000 kg/hr (Shell side)
Crude oil API 34, 85,000 kg/hr (Tube side)
Kerosene Inlet Average Outlet
o
Temperature C 200 145 90
o
Specific heat kJ/kg. C 2.72 2.47 2.26
Thermal conductivity W/m. oC 0.13 0.132 0.135
Density Kg/m3 690 730 770
-2
Viscosity mNsm 0.22 0.43 0.8
Crude Oil
o
Temperature C 79 59.5 40
Specific heat kJ/kg. oC 2.09 2.05 2.01
o
Thermal conductivity W/m. C 0.133 0.134 0.135
Density Kg/m3 800 820 840
-2
Viscosity mNsm 2.4 3.2 4.3
Assumptions:
Uservice = 350 W/m2. oC
Length = 5000 mm, Do = 19.05 mm, Di = 14.83 mm, triangular pitch 30o, pitch ratio = 1.25
Allowable ∆Ps = 0.9 bar, ∆Pt = 0.9 bar,
Output
Table: input for tool was only S.G and temperatures.

Parameter Units Coker’s [1] Commercial Tool Tool

software kern Polley
Surface Area m2 77.64 50.2 82.85 54.84
hs coefficient 1325.69 1510.3 1237.35 1857.4
ht coefficient 1555 1543.8 901.1 2489.5
Uservice W/m2. oC 350 736.2 315.87 561.760
Uclean W/m2. oC 508.9 472.27 413.26
Udirty W/m2. oC 440 508.2 365 420.88
Over Design % 25.7 1 13.53 -1.8
Over surface % 45 49.51 33.4
∆Ps Kg/cm2 0.7 bar 0.74 0.4 0.86
∆Pt Kg/cm2 0.96 bar 0.329 0.41 0.93
Length mm 5000 4876.8 5000 7159
Number of tubes - 260 181 213 128
Tube diameter mm 19.05 19.05 25.4 19.05
Tube thickness mm 2.11 2.11 2.11 2.11
Tube pitch mm 23.81 23.81 31.75 23.81
Ds mm 522 438.15 627 348.3
Number of passes - 4 6 4 2
Baffle spacing mm 104.32 215.9 125.4 79.67
Baffle cut % 25 38.04 25 14
Q = ṁCp∆T =(25,000/3600)*247(200-90) = 1886.8 kw
∆TLMTD = = 80.34 ˚𝐶
( )

P= = 0.2437, R = = 2.82
F = 0.8673 , ∆TCMTD= F * ∆TLMTD = 69.43
Q=UA∆TCMTD , A = Q/U∆TCMTD = 1886.8 x 103/(350x69.43) = 77.64 m2
Nt = A/πDoL = 77.64/(3.14x0.01905x5) = 259.46 (Say 260)
Tube cross sectional area = Ac = πDi2/4 = π(0.01483)2/4 = 0.0001727 m2
Area per pass = Nt/pnAc =(260/2x0.0001727)= 0.02246 m2
Gs = Wh/ρ = 85,000/3600*820= 0.02879 m3/s
Tube velocity = ut = 0.02879/0.02246 = 1.28 m/s
Estimate Shell diameter
From table 8.46 for two passes K1 = 0.249 , n1 = 2.207
.
𝐷 =𝐷 ∗ = 19.05 ∗ = 444 mm
.
Clearance is between 50 to 80 mm for split-ring floating head
Choose C= 56 mm, Ds = 444 +56 = 500 mm
∗ . ∗ .
𝑅 = = = 4864
. ∗
. ∗ ∗ . ∗
Pr= Cpμ/k = =48.96
.
Re= 4864, L/di =5000/14.84 = 337, from figure 8.55 the friction factor Jh
Jh = 3.5 x 10
Nu = Jh Re Pr 0.33(μ/μw)0.14
.
= 𝐽 𝑅𝑒𝑃𝑟 .
, assume negligible μ/μw
ℎ = 3.5 x 10 (0.134/0.01483)(4864)(48.96)0.33=555.48 W/m2. oC
In order to increase hi , tube passes is doubled, this will halve cross sectional area and double
the tube velocity.
ut = 2 x 1.28 = 2.56 m/s
Re = 2 x 4864 = 9728, Jh = 4.9 x 10
hi= 4.9 x 10 (0.134/0.01483)(9728)(48.96)0.33= 1555 W/m2. oC √√
From table 8.46 for two passes K1 = 0.175 , n1 = 2.285
.
𝐷 =𝐷 ∗ = 19.05 ∗ = 465.6 mm
.
Ds = 465.6 +56 = 521.6 mm
Baffle spacing trials
B = Ds/5 = 521.6/5 = 104.32
( ) . .
Shell flow area = = 𝑥522𝑥144 = 10853.04 mm2 =0.01853 m2
.
.
Equivalent shell diameter = De = (𝑝 − 0.913𝑑 ) =13.53 mm
Gs = Wh/ρ = 25,000/3600*730= 0.00951 m3/s
Shell velocity = us = 0.00951/0.01853 = 0.51 m/s
∗ . ∗ .
𝑅 = = = 11,714
. ∗
. ∗ ∗ . ∗
Pr= Cpμ/k = = 8.17
.
-3
jh= 5.8 x 10
.
Nu = = 𝐽 𝑅𝑒𝑃𝑟 .

ho= 5.8 x 10 (0.132/0.01353)(11,714)(8.17)0.33= 1325.69 W/m2. oC

Uclean and Udirty
𝑑
1 1 𝑑 𝑑 ln 1
𝑑
= + +
𝑈 ℎ 𝑑 2𝑘 ℎ
19.05
1 1 19.05 1 19.05 0.001905 ln 1
=( + 0.00035) + + 14.83 +
𝑈 1555 14.83 ℎ 14.83 2 ∗ 55 1325.69

𝑑
1 1 𝑑 1 𝑑 𝑑 ln 1
𝑑
= + + + +𝑅
𝑈 ℎ 𝑑 ℎ 𝑑 2𝑘 ℎ
19.05
1 1 19.05 1 19.05 0.001905 ln 1
= + 0.00035 + + 14.83 + + 0.0002
𝑈 1555 14.83 ℎ 14.83 2 ∗ 55 1325.69
= 440 W/m2. oC
Over Design = ∗ 100% = 25.7%
Pressure drops
Tube pressure drop
𝐿 μ ρv
∆Pt = 𝑁 8𝐽 + 2.5
𝐷 μ 2
=4[8x4.8x10 x337.15+2.5]820x2.56 /2 =96443.9 N/m2 = 0.96 bar
-3 2

Shell pressure drop

𝐿 𝐿 ρv μ .
∆Ps = 8𝑗 8𝐽
𝐷 𝐵 2 μ
=8[4.8x10 x(522/13.75)(5000/104)(730x0.512/2)] =66537.87 N/m2 = 0.7 bar
-2
Bell Delaware (Polley’s Method) [6]
Use Bell Delaware’s Method to obtain a preliminary design for the following conditions

Parameter Unit Shellside Tubeside

Flow rate Kg/hr 22.4*3600 77.96*3600
Fluid density Kg/m3 740 1000
Heat Capacity J/kg.K 2407 4187
Viscosity cP 0.494 1
Thermal conductivity W/m.K 0.105 0.61
o
Inlet temperature C 100 7.1
o
Outlet temperature C 42.5 16.6
Allowable ∆P kPa 13.7 11.66
Dirt factor K.m2/W 0.0 0.0
Wall resistance K.m2/W 0.00003
Heat duty kW 3100

Geometry Unit Polley’s Tool

Do mm 16 15.88
Di mm 13.5 13.4
Tube layout Degree 30 30
Tube pitch mm 20.8 20.64
Baffle/shell clearance mm 5.5 5.366
Tube/baffle clearance mm 0.5 0.4 “assumed”
Bundle/shell clearance mm 10 15.33
Output
Parameter Units Polley’s Tool
Geometry
Ds mm 563 566.6067
Tube length mm 1815 1721.38016
Baffle cut % 29.3 26
Baffle spacing mm 253 190.57
No. of baffles - 6 8
No. of tubes - 574 677
No. of tube passes - 2 2
Installed area m2 52.3 58.14
Required area m2 52.3 50.55
Performance
Shellside Re - 21,398 27,211.4
Shellside ∆P kPa 13.7 15.95
Tubeside ∆P kPa 11.69 19.6
Shellside Coeff. W/m2. oC 1406 2124.5
Tubeside Coeff. W/m2. oC 6641 6380.67
Udirty W/m2. oC 1088 982
1. Initialization
Assume exchanger with shell diameter Ds = 387 m, baffle cut=25%, 6 m length, 2 passes for the
following heat exchanger:
Geometry Unit Value
Do mm 16
Di mm 13.5
Tube layout Degree 30
Tube pitch mm 20.8
Baffle/shell clearance mm 5.5
Tube/baffle clearance mm 0.5
Bundle/shell clearance mm 10

a. Estimate the number of tubes that could fit in your shell diameter
b. Estimate shell Reynold’s number
θDS = 2cos-1(1 - 2Bc/100)
θCTL = 2cos-1(Ds(1 - 2*Bc/100)/DCTL)
FW = (θCTL - Sin(θCTL))/2π
SWG = (Ds²/8)(θDS - Sin(θDS))
SWT = Nt.FW(πDo²/4)
SW = SWG - SWT
Based on the assumption that baffle spacing is same as baffle cut area
Sm = SW
GW = mS/(Sm.SW)0.5
GS = mS/Sm
ReS = Do.GS / μS
c. Calculate shell heat transfer coefficient and pressure drop
Sm = B[(Ds - DOTL) + (DOTL - Do)(PT - Do)/PT,eff ] Di = Do – 2*thickness
PT,eff = PT for 30° and 90° layouts Ltb = 0.4
PT,eff = 0.707*PT for 45° layout
Ls= 0.1* Ds
GS = mS/Sm
ReS = Do.GS / μS Dsb = 3.1+0.004* Ds
PrS = CP,S.μS / kS Leff = L- 2* Ls
Dotl= Ds - (12.5+( Ds/200))
1.33
𝑗=𝑎 . (𝑅𝑒 )
𝑃
𝐷𝑜
𝑎
𝑎=
1 + 0.14(𝑅𝑒 )

µ .
ℎ = 𝑗. 𝐶 , . 𝐺 . (𝑃𝑟 ) . ( )
µ,

where, μS,W is shell side fluid viscosity at wall temperature.

a.1 Correction factor for Baffle Window Flow, JC
JC = 0.55 + 0.72FC
FC = 1 - 2FW
FW = (θCTL - Sin(θCTL))/2π
θCTL = 2cos-1(Ds(1 - 2*Bc/100)/DCTL)
DCTL = DOTL - Do

a.2 Correction factor for Baffle Leakage, JL

JL = 0.44(1-rS) + (1-0.44(1-rS))exp(-2.2rL)
rS = Ssb /(Ssb + Stb)
rL = (Ssb + Stb)/ Sm
Ssb = Ds*DSB(π - 0.5θDS)
Stb = (π/4)((Do+LTB)2 - Do2)Nt(1-FW)
θDS = 2cos-1(1 - 2Bc/100)

a.3 Correction factor for Bundle Bypass effects, J B

Ntcc = (Ds/Pp)(1 - 2Bc/100)
Pp = PT 30.5/2 for 30° layout
Pp = PT / 20.5 for 45° layout
Pp = PT for 90° layout
rss = Nss/ Ntcc
Sb = B(Ds - DOTL - Do/2)
JB = exp(-Cj(Sb / Sm)(1 - (2rss)1/3)) for rss < 0.5
JB = 1 for rss >= 0.5
Cj = 1.35 for ReS < 100
Cj = 1.25 for ReS >= 100

a.4 Correction factor for adverse temperature gradient, J R

a.5 Correction factor for unequal baffle spacing, J S

n1 = 0.6, ReS >= 100
n1 = 1/3, ReS < 100
JS = ((NB-1)+(LBIn/B)1-n1 + (LBOut/B)1-n1)/((NB-1)+(LBIn/B) + (LBOut/B))

Shell side heat transfer coefficient is calculated as

hs = hIdeal(JC.JL.JB.JS.JR)

1.33
𝑓=𝑏 . (𝑅𝑒 )
𝑃
𝐷𝑜

𝑏
𝑏=
1 + 0.14(𝑅𝑒 )
ΔPIdeal = 2f(GS²/ρS)(μS/μS,W)0.14 Ntcc

b.1 Correction factor for Baffle Leakage, RL

RL = exp(-1.33(1+rS)rLp)
p = 0.8 - 0.15(1+rS)

b.2 Correction factor for Bundle Bypass effect, R B

RB = exp(-Cr(Sb / Sm)(1 - (2rss)1/3)) for rss < 0.5
RB = 1 for rss >= 0.5
Cr = 4.5 for ReS < 100
Cr = 3.7 for ReS >= 100

b.3 Correction factor for unequal baffle spacing, RS

n = 0.2, ReS >= 100

n = 1.0, ReS < 100
RS = 0.5((B/LBIn)2-n + (B/LBOut)2-n)

1. Pressure drop for window section, ΔPW

Following terms are calculated as -

SWG = (Ds²/8)(θDS - Sin(θDS))

SWT = Nt.FW(πDo²/4)
SW = SWG - SWT
GW = mS/(Sm.SW)0.5
DW = 4.SW /(π.Do.Nt.FW + θDS.Ds)
SWG Gross window area, SWT Window area occupied with tubes, SW Net Cross flow area, Gw Mass velocity
baffle window, Dw Hydraulic diameter baffle window.

Pressure drop for laminar and turbulent flow is calculated.

ΔPW, Turb = NB.RL(2+0.6*Ntcw).GW²/(2.ρS)
26. 𝐺W. 𝐶 Ntcw 𝐵 𝐺W²
ΔPW, Laminar = NB.RL( + + )
ρS PT − D o D w ² ρS
ΔPW = ΔPW, Turb , ReS ≥ 100
ΔPW = ΔPW,Laminar , ReS < 100

2. Pressure drop in Central Baffle spaces

ΔPC = (NB - 1)ΔPIdeal.RL.RB

3. Pressure drop in entrance & exit baffle spaces

ΔPE = ΔPIdeal(1 + Ntcw/Ntcc).RB.RS

Shell side pressure drop is calculated as following

ΔPS = ΔPW + ΔPC + ΔPE
d. Calculate tube heat transfer coefficient and pressure drop
ReT = Di.v.ρt/μt
PrT = Cp,t.μt/kt
For laminar flow, ReT < 2300, Sieder and Tate correlation is used for Nusselt's nubmer.
Nu = 1.86(ReT.PrT.Di/Leff)1/3
Leff = L - 2*Ls , where Leff is effective tube length

For turbulent flow, ReT > 10,000, following equation developed by Petukhov-Kirillov
can be used.

Nu = (f/2)ReT.PrT/(1.07+12.7(f/2)0.5(PrT2/3-1))
f = (1.58 ln(ReT) - 3.28)-2

For transient flow, Nusselt number can be interpolated from Nu Laminar & Nu Turbulent.
ht = Nu.(kt/Di)(μt/μt, w)0.14

a. Tube Side Pressure Drop, ΔPt

ΔPt = (4.f.Leff.Np/Di + 4.Np)ρt.v²/2

ΔPs = 0.47 Kg/cm2, ΔPt =0.5159 Kg/cm2, hs =1061 W/m².°C, ht =7071

W/m².°C, A=48.147 m2

e. Calculate shell and tube constants from the following equations

.
∆𝑃 = 𝑘 ∗ 𝐴 ∗ ℎ
.
∆𝑃 = 𝑘 ∗ 𝐴 ∗ ℎ

ks = 1.912 , kt= 0.906

1. Solving for the allowable pressure drops

1. Solve the following non-linear equations to obtain the desired heat
transfer coefficients for both shell and tube and the required area (h s,ht
and A) that can achieve your allowable pressure drops for both shell and
tubes

𝐴= Q /( LMTDcorrected)*(1/(1/hS + Do/(Di.ht) + Rtube)) + fshell + ftube)

∆𝑃 = 𝑘 ∗ ℎ .
→ 11.66 = kt*(ht )3.5→ ht = 908 W/m².°C

∆𝑃 = 𝑘 ∗ ℎ .
→ 13.7 = ks*(hs )4.412→ hs = 6335 W/m².°C

A = 64.74 m2

Where ∆𝑃 and ∆𝑃 are shell and tube allowable pressure drops

Solve the following equations to obtain tube Reynold’s number and friction
factor
Nu = (f/2)ReT.PrT/(1.07+12.7(f/2)0.5(PrT2/3-1))
f = (1.58 ln(ReT) - 3.28)-2
where Nu = ℎ /.(kt/Di)(μt/μt, w)0.14
after obtaining Re iterate this Re till ht calculated = ht solved from the
previous step.
𝑣 = Di.Re.ρt/μt
at = mt/(⍴t*vt*3600)
Nt = at/ π. ∗

Leff =A/(tn. π.Do)

For laminar flow, ReT < 2300, Sieder and Tate correlation is used for Nusselt's nubmer.

Nu = 1.86(ReT.PrT.Di/Leff)1/3
Leff = L - 2*Ls , where Leff is effective tube length

For turbulent flow, ReT > 10,000, following equation developed by Petukhov-Kirillov
can be used.

Nu = (f/2)ReT.PrT/(1.07+12.7(f/2)0.5(PrT2/3-1))
f = (1.58 ln(ReT) - 3.28)-2

For transient flow, Nusselt number can be interpolated from Nu Laminar & Nu Turbulent.

ht = Nu.(kt/Di)(μt/μt, w)0.14
if ht = ht.solved stop iteration, else change Re accordingly and start over.

Re = 21866.32, ht =6376.52 W/m².°C , vt =1.631 m/s, at=0.047 m2/s,Leff =

1.72 m, Nt = 677
2. Iterate!
1. Estimate Shell diameter from the tube count. If it didn’t change proceed,
if changed repeat from initialization with a new shell diameter
accordingly.
Ds = 566.60
2. Calculate ∆𝑃 If it didn’t change from your allowable shell pressure drop
proceed, if it changed repeat from initialization with a new baffle cut
accordingly.
3. If no baffle cut % achieved ∆𝑃 start from initialization with a new
number of passes.

Baffle cut% = 26%

∆𝑃 = 0.196 Kg/cm2

NTU Method
For the previous example:
Fluid Data Hot Cold
Flowrate kg/s 22.4 78.0
Inlet Temperature °C 100.0 7.1
Specific Heat kJ/kg.°K 2.407 4.19

Exchanger Data
Heat Exchanger Type Shell & Tube - 1 Shell - 2,4, .. Tube Passes
Heat Transfer Coefficient kW/m².°K 1.09
Area m² 52.30
Solution
UA = 52.3*1.088 = 56.9 kW/°C
Cmin = 22.4*3600*2.407* 0.239006*1.163 = 53952.94 W/oC
Cmax = 78*3600*4.19*0.239006 *1.163 = 327039.1 W/ oC
Cr = Cmin/Cmax = 0.1652
NTU = UA/Cmin = 56.9/53.95 = 1.06
Qmax = Cmin(T1-t1) = (22.4*3600*2.407* 0.239006) * (100-7.1) = 4309741 kCal

2
𝜀= = 0.6177
1 + exp (−𝑁𝑇𝑈 1 + 𝐶 )
1+𝐶 + 1+𝐶
1 − exp (−𝑁𝑇𝑈 1 + 𝐶 )

Q = ε Qmax = 0.6177 * 4309741= 2662323 Kcal/hr

T2 = T1 -Q/m.Cp = 100 – 2662323/ 46391.18 = 42.6 oC

t2 = t1 +Q/m.Cp = 7.1 + 2662323/281203 = 16.6 oC

hence, exchanger would achieve the desired outlet temperatures.

Oil fractions’ properties [Notes] Valid in the range of 50-300 F, otherwise
1. Density you may take viscosity at point 37.38 C and
Liquid [4] use the following one-point approximation
ρ = sg*1000*(1-α*((t*1.8+32)-60)) 2. From 1 viscosity point [7]
α: Thermal expansion coefficient c = -0.8696
API α ( oC-1) b= log10(µ37.38 ˚C; 1 atm )-c
14.9 0.00035 s = 0.2008*b+1.6180
34.9 0.0004 Log(µ) = (b/(1+((T-Tanalysis)/310.93))**s)+c
50.9 0.0005 µ: Viscosity in c.St
63.9 0.0006
3. From 2 viscosity points
78.9 0.0007
Solve the two points using the logarithmic
88.9 0.0008
93.9 0.00085 function
100 0.0009 A = (z2 – z1) / (x2 – x1)
Vapor B = z1 – a * x1
Molecular weight [1] x1, x2: temperature points
1⁄𝑀 = 0.0001644 ∗ 𝐴𝑃𝐼 − 0.000972 z1, z2: log of viscosity points
16.018463 ∗ 𝑚_𝑤𝑡 ∗ 𝑝 ∗ 14.7 µ = 10a (t+273.15) +b
⍴=
10.732 ∗ 𝑧 ∗ (𝑡 + 459.67) Vapor [7]
P: atm. ,t in o F ⍴: kg/m3 µOg=-0.0092696 +T0.5(0.001383-5.9712x10-5M0.5)+1.1249x10-5M
2. Heat Capacity, Latent Heat & Thermal T: temperature in oK
conductivity [Note 1] Note: The reliability of this equation is about 6%

Cp = (1/ 𝑠𝑔))(0.388+0.00045(t*1.8+32)) µg=10-4A[exp(B x ρC)]

Cv = Cp-(0.09/sg) A = [(12.6 + 0.021M) T 1.5] / (116 + 10.6M + T)
HLatent = (1/sg)*(110.9-0.09*(t*1.8+32)) * B = 3.45 + 0.01M+ ((548/T)
0.555927 C = 2.4 -0.2 B
k = (0.813/sg)*(1-0.0003*((t*1.8+32)-32)) ρ: Density in kg/m3
*0.1442279 Hydrocarbon mixtures
-
kgas =( -0.000500777+1.0906*10 1. Density
5
*(t*1.8+32)+0.061137256/√M.wt + liquid
0.000158966*(t*1.8+32)/√M.wt)*1.730735 GPSA (page 23-24) [2]
Cp and Cv in Kcal/kg.oC 1. Calculate C3+ Density using table
HLatent in Kcal/Kg
23-16
k and kgas in W/m.oC
2. Calculate wt% C2 in mixture in C2+
3. Viscosity
3. Calculate Density of CO2+
Liquid
4. Calculate wt% in total
1. From no viscosity points[5]
5. Use fig 23.14 for pseudo-density
API Equation
mixture at 60F and 14.7 psia
34 3.518-0.01591*t+1.734*10-5t2
30 5.804-0.029831*t+1.2485*10-5*t2 6. Use fig. 23-15 for pressure
20 9.21-0.0469*t+3.167*10-5*t2 correction
10 18.919-0.1322*t+2.431*10-5*t2 7. Use fig 23-17 for temperature
correction
8. Calculate density at operating 2. Soave-Redlich-Kwong
pressure and temperature 3. Peng-Robinson
⍴= , [6] where: 2. Viscosity of liquids/gases
∑
⍴ Liquid
⍴: Density of the mixture [kg/m3]  The Orrick-Erbar and Letsou-Stiel methods
⍴ : Density of component i are recommended for estimating the liquid
𝑤 : Weight fraction of component i viscosity. [4] [Notes]
Elbro et al [6] [Notes] Ludwig’s [3] [Notes]
µG = A+ BT + CT2
𝑉 = 𝑛 .△𝑉
Simple Methods [6]
△ 𝑉 = 𝐴 + 𝐵 ∗ 𝑇 + 𝐶 ∗ 𝑇 , Where: Method 1: If only one viscosity datum is
Vm is the molar volume of liquid, ni is the available, viscosity at other temperatures
number of group i, and △Vi is the molar can be approximately extrapolated by the
group volume equation:
Example: µµ0tt233
Estimate the liquid density of isopropyl Method 2: If two pairs of temperature-
viscosity data are available, it is often a
acetate at 20C.
good approximation to assume
µ1 is linear, reciprocal to absolute
temperature:
ln µL = ln A + B/T
Liquid mixtures[6]
Vm = V (CH3COO) + V (CH) + 2V (CH3)
At 20C, or 293.15 oK µ = ∑𝑥 .µ , for liquid hydrocarbons mixtures
V (CH3COO) = 42.820 - 20.50×10-3 (293.15) +
16.42×10-5 (293.15)2 = 50.92 ln µ = ∑ 𝑥 . ln µ , for liquid non-hydrocarbons
V (CH) = 6.297 - 21.92×10-3 (293.15) = -0.13 mixtures , xi is weight fraction
V (CH3) = 18.960 + 45.58×10-3 (293.15) = 32.32 Vapor viscosity
Vm = 50.92 - 0.13 + 2 (32.32) = 115.44 cm3/mol  The Reichenberg method is recommended.[4][Notes]
The molecular weight is 102.133, then
Ludwig’s [3]
⍴= 102.133/115.44 = 0.885 g/cm3
µG = (A + BT + CT2) x 10 -4
Ludwig’s [3]
Gas mixtures[6] (Herning Zipperer)
⍴ = 𝐴𝐵 ( )
∑ .µ .
A, B & n: constants µ = ∑
, yi is mole fraction
.
Hydrocarbon gas density Saturated Liquid viscosity
Ludwig’s [3] Ludwig’s [3]
ρideal = Mwt *Vm=Mwt.P/RT Log µL = A + (B/T) + CT + DT2
ρ = Mwt *Vm=Mwt.P/ZRT µL : cP, T: oK
where Vm is the molar volume L/g.mol GPSA
R: Universal gas const. 8.314 Use previous documentation (Line sizing)
kPa.dm3/mol.k Use fig. 23-21 to 23-29
P: Gas pressure in kPa 3. Thermal conductivity
T: Gas temperature in oK Gas Thermal conductivity
Other methods (EOS) [2] : Ludwig’s [3] [Notes]
1. Van der Waals Kgas = A+ BT + CT2
Kgas: W/m.K ∑ . .( )
𝑘 = , for gas mixtures
Stiel-Thodos Method[Notes] ∑ .( )

𝜔 = 𝑦𝜔 GPSA[2]
Using Charts
𝑍 = 0.291 − 0.08𝜔
1. Use fig 23-31 at operating
𝑀 = 𝑦𝑀 temperature to find KA
𝑇 =𝑇 2. Calculate Tr and Pr
3. Use fig. 23-32 to find thermal
𝑇 = (𝑇 𝑇 )
conductivity ratio
𝑉 =𝑉
4. K = ratio * KA
(𝑉+𝑉 )  Stiel and thodos Method is recommended
𝑉 =
8 For mixtures of simple gases, it doesn’t
∑ ∑ 𝑦𝑦𝑉 𝑇 apply to mixtures containing high CO2[2]
𝑇 =
𝑉  Miscellaneous gases can be obtained from
𝑉 = 𝑦𝑦𝑉 fig 23-33 and hydrocarbon gases from fig
23-34
𝑃 =𝑍 𝑅𝑇 /𝑉
/ Liquid Thermal conductivity
𝑇 𝑀 Ludwig’s [3] [Notes]
𝛤 = 210
𝑃 KL = A+ BT + CT2
𝑃 = 𝑉 /𝑉 KL: µcal/s.cm.oC
For organic compounds, the correlation for thermal
if ρrm<0.5: conductivity of
liquid as a function of temperature is [2]:
(λ−λ∘)ΓZcm5=1.22×10−2[exp(0.535ρr)−1] log10 kliq
if 0.5< ρrm <2.0: Thermal conductivity for liquid mixtures
(λ−λ∘)ΓZcm5=1.14×10−2[exp(0.67ρrm)−1.069] ∑ . .
if 2.0< ρrm <2.8: = ∑
or = ∑𝑤 , where
.
(λ−λ∘)ΓZcm5=2.6×10−3[exp(1.155ρrm)−2.016] Xi: Mole fraction of component i
Wi: weight fraction of component i
Tcm, Pcm, Zcm, Vcm are critical temperature, Mi: molecular weight of component i
pressure, compressibility factor and specific From graph 23-35 liquid paraffin
volume of mixture in oK, bar, -, cm3/mol hydrocarbons
Vm, Mm, λ, λ∘ are specific volume at 1 atm,  The Sato-Riedel and Missenard methods
molecular weight of mixture g/mol, are-recommended.[6] [Notes]
thermal conductivity and thermal Missenard method
conductivity at 1 atm in W/m.k k/k∗=1+QPr0.7
yi, yj are mol. Fractions of mixture k*: k at low pressure (W/m.K)
components
𝛤 is the reduced, inverse thermal
conductivity in [W/m.K]-1

Thermal conductivity for gas mixtures

Sato-Riedel Method [Notes] R: gas constant 8.3145 J mol-1 K-1
1.1053 3 + 20(1 − 𝑇𝑟) / CPL: J mol-1 K-1
𝑘= k is total the number of different kinds of
√𝑀𝑊 3 + 20(1 − 𝑇𝑏𝑟) / groups
Tr: reduced temperature oK
Tbr: reduced boiling temperature oK at 1 Vapor Heat capacity Cpv
atm Ludwig’s [3]
Kgas: W/m.K Cp = A + BT + CT2 + DT3+ET4
4. Heat capacity Cp Cp: joule/mol. K
Liquid Heat Capacity Benson method is recommended for ideal
Ludwig’s [3] [Notes] gas Cp [6]
Cp = A + BT + CT2 + DT3 Harrison-Seaton method, is much simpler
Cp: joule/mol. K than Benson method and gives only slightly
GPSA inferior results. [6]
Use table (fig 23-2) in P. 23-2 for specific Heat Capacity for gas or liquid mixtures
heats at 60 oF and 14.696 psia for gases and Cp = 𝑥 . Cp
liquid
xi is components weight fraction
 The Chueh-Swanson method is
References
recommended for estimation of liquid heat
[1] Bureau of standards report 1929: Thermal
capacity at 20C, and for temperature
dependency the Missenard method is properties of petroleum products
recommended. [6] [Notes] [2] GPSA Engineering Databook, Section 23.
The assumption is made that various groups [3] Ludwig’s applied process design for
in a molecule contribute a definite value to chemical and petrochemical plants, vol. 1
the total molar heat capacity. (chapter 3 and Appendix C)
Example:
[4] Petroleum Refinery Engineering, W. L.
Estimate the liquid heat capacity of 1,4-
pentadiene at 20C. Nelson
[5] Petroleum Refining Design and applications,
vol. 1, A. Kayode coker
[6] JGS 210-120-1-11E “Estimation of physical
properties”
[7] Prediction of the viscosity of crude oil
CpL = 2 (CH2=) + 2 (-CH=) + (-CH2-) +
fractions from a single measurement, A.
corrections noted in Table 4.2.3
= 2 (5.20) + 2 (5.10) + 7.26 + 4.5 + 2.5 Miadonye and V.R. Puttagunta
= 34.9 cal/mol･K Notes
The observed value is 35.1cal/mol･K. 1. It’s worth mentioning that the most
accurate method to obtain oil fractions
Method of Ruzicka and Domalski[Notes] properties is oil characterization
𝑇 𝑇 calculations if ASTM assay or Tb are
𝐶 =𝑅 𝐴+𝐵 +𝐷
100 100 available. For more on that topic the
following references are recommended:
𝐴= 𝑛 .𝑎 ,𝐵 = 𝑛 .𝑏 ,𝐷 = 𝑛 .𝑑
A. Characterization and Properties of
T: temperature in oK Petroleum Fractions, M. R. Riazi
B. Petroleum Refining Design and liquids, Bruce E. Poling et al, fifth edition P.
Applications Handbook, Volume 1, 6.19 and its constants in P. 6.20
chapter 3 10.The Chueh-Swanson method and Missenard
C. Petroleum Refinery Engineering, method are described in The properties of
Nelson, chapter 5 gases and liquids, Bruce E. Poling et al,
D. Petroleum Refinery Process fourth edition P. 137 and its constants in P.
Modeling: Integrated Optimization 138-13
Tools and Applications, Y A Liu, Ai- 11.Reichenberg method is described in The
Fu Chang, Chapter 1. properties of gases and liquids, Bruce E.
2. kgas equation was obtained through regressing Poling et al, fifth edition P. 9.12
Maxwell databook chart “Thermal conductivity
of hydrocarbons” P. 215 Section 12 Flow of Heat
Or from fig 23-36
3. Ludwig’s equation constants can be found
in reference [3] Appendix C pages 827-862
4. For Elbro Method constants A, B and C,
refer to The properties of gases and liquids,
Bruce E. Poling et al, fifth edition P. 4.41
5. Orrick and Erbar Method is described in
Viscosity of Liquids, Theory, Estimation,
Experiment and Data P. 315
𝜂 𝐵
ln =𝐴+
𝜌𝑀 𝑇
η: Liquid viscosity in cP
ρ: density at 20 oC
M: molecular weight
T: absolute temperature in oK
A and B are constants in table 4.30 P. 315 As
mentioned above.
6. Missenard Method is described in The
properties of gases and liquids, Bruce E.
Poling et al, fifth edition P. 10.54
7. Sato- Riedel Method is described in The
properties of gases and liquids, Bruce E.
Poling et al, fourth edition P. 550
8. Stiel and thodos method is described in The
properties of gases and liquids, Bruce E.
Poling et al, fifth edition P. 10.35 and its
constants in Apenddix A
9. Method of Ruzicka and Domalski is
described in The properties of gases and
Examples
Calculate water properties (Density, Viscosity, Thermal conductivity and heat capacity) at 50 oC
Ludwig’s
Substituting the coefficients (As, Bs, Cs, Ds, and Es) from Appendix C
𝒏
⍴𝑳 = 𝑨𝑩 (𝟏 𝑻𝒓 )
A = 0.347, B = 0.274, n = 0.286, Tc = 647.4 oK (from table C-1 in appendix C)
Tr = (273+50)/647.4 = 0.498
ρ = 0.347*0.286-(1-0.498)^0.286=1004 kg/m3
Log µL = A + (B/T) + CT + DT2
A = -10.73 , B =18.28, C=0.01966, D= -0.00001466 (from table C-3 in appendix C)
Log µL = -10.73 +(18.28/323) +0.01966*323 -0.00001466*3232
KL = A+ BT + CT2
A = -916.62, B = 12.547 and C = -0.015212
KL = -916.62 + 12.547*(50+273) -0.015212*3232 =1549.54 µcal/s.cm.oC
Cp = A + BT + CT2 + DT3
A = 0.6741, B=0.002825, C=-0.000008371, D =8.601*10-9 (from table C-5 in appendix C)
Cp = 0.6741+0.002825*323 -0.000008371*323 2 +8.601*10-9*3233 = 1.003 Kcal/kg.oC
heat capacity of 1-butene at a temperature of 200 oC (473.15 K).
using coefficients from table C-4 in appendix C
Cp = 24.915+ (0.20348*473.15)+(5.9828*10-5*473.152)+(-1.4166*10-7*473.153)+(
4.7053*10-11*473.154)=123.36 j/mol. K
Example 23-2 - Calculate the density of the following gas mixture at:
Temperature=100o F
Pressure =1000 psia
Component yi [%] Tci [R] Tpc [R] Pci [psia] Pci [psia] MW
CH4 83.19% 343.0 285.3 666.4 554.4 16.043
C2H6 8.48% 549.6 46.6 706.5 59.9 30.070
C3H8 4.37% 665.7 29.1 616.0 26.9 44.097
i-C4H10 0.76% 734.1 5.6 527.9 4.0 58.123
n-C4H10 1.68% 765.3 12.9 550.6 9.3 58.123
i-C5H12 0.57% 828.8 4.7 490.4 2.8 72.150
n-C5H12 0.32% 845.5 2.7 488.6 1.6 72.150
C6H14 0.63% 913.3 5.8 436.9 2.8 86.177
100.00% 392.7 661.6 20.426
Tr (100+459.67)/392.7=1.425
Pr 1000/661.6=1.512
Z-factor 0.86 (from Fig.23-4 with Pr and Tr)
Density 20.426*1000/10.732/0.86/559.67=3.95 lbm/ft 3
Example 23-2 - Calculate the density of the following liquid mixture at:
Temperature=120o F
Pressure =1760 psia

Density
Volume
Component Mole fraction Mol wt Weight, lb (60F),
, cu ft
lb/cf
Methane 0.20896 16.043 3.352 - -
Carbon dioxide 0.39730 44.010 17.485 51.016 0.3427
Ethane 0.01886 30.070 0.567 - -
Propane 0.02387 44.097 1.053 31.619 0.0333
n-Butane 0.03586 58.123 2.084 36.423 0.0572
n-Pentane 0.02447 72.150 1.766 39.36 0.0449
n-Hexane 0.01844 86.177 1.589 41.4 0.0384
n-Heptane 0.02983 100.204 2.989 42.92 0.0696
n-Octane 0.02995 114.231 3.421 44.09 0.0776
n-Decane 0.18208 142.285 25.907 45.79 0.5658
n-Tetradecane 0.03038 198.394 6.027 47.85 0.1260
Total 1.00000 66.241

Calculate density of C3+ = 44.836/1.0128=44.275 lbm/ft3

Weight % C2 in C2+ = 100*0.567/(0.567+44.836)=1.25% Wt %
Density of CO2+ = (45.403+17.485)/(45.403/44+0.3427)=45.75 lbm/ft3
Wt % CH4 in Total = 100*3.352/66.241=5.1% Wt %
(at 60F and 14.7
Pseudo-density of mixture = 42.9 lbm/ft3
psia, Fig 23-14)
Pressure correction = 0.7 lbm/ft3 (from Fig 23-15)
Density at 60F and 1,760
= 42.9+0.7=43.6 lbm/ft3
psia
Temp correction to 120F
= -1.8 lbm/ft3
from Fig. 23-17
Density at 60F and 1,760
= 43.6-1.8=41.8 lbm/ft3
psia
Viscosity
Example 23-9/10 Calculation of gas mixture viscosity.
Given:
Pressure = 1,000 psia
Temperature = 100 degF
MW = 22.0 lb/lb-mole
Tc = 409 deg R
Pc = 665 psia
Gid = 22/28.9625=0.760 dimensionless
μA = 0.0105 cP (from Fig. 23-22 at 100
deg F)
Tr = (100+459.7)/409=1.37 dimensionless
Pr = 1000/665=1.50 dimensionless
μ/μA = 1.21 dimensionless (from Fig. 23-24)
μ = 1.21*0.0105=0.0127 cP

Given:
Pressure = 300 psia
Temperature = 50 deg F
Methane = 80 mol %
Nitrogen = 15 mol %
Carbon dioxide = 5 mol %

Component yi MW Pci, Tci, deg R Vci, ft3/lb Vci • MWi • yi Zci • yi

psia
CH4 0.80 16.043 667.0 343.0 0.0988 1.27 0.2306
N2 0.15 28.013 492.5 227.2 0.0510 0.21 0.0434
CO2 0.05 44.010 1,069. 547.4 0.0342 0.08 0.0137
5
Mixture 1.00 19.237 335.9 1.558 0.2877

Solution
Steps
Tcm = 335.9 deg R (from Table above)
3
Vcm = 1.558 ft /lb (from Table above)
Pcm = 0.2877*10.73*335.9/1.558=665.6 psia
Zcm = 0.2877 dimensionless
5.4402*(335.9)0.1666/(665.60.6666)/(19.23)0.5
ξ = dimensionless (from Eq. 23-20)
=0.043
Tr = 1.518 dimensionless
ξμA = 0.000489 dimensionless (from Eq. 23-21)
μA = 0.000489/0.043=0.014 cP (from Eq. 23-22)

Thermal Conductivity
Example 23-12 Find the thermal conductivity of a natural gas.
Given:
Pressure = 700 psia
Temperature = 300 deg F
Tc = 440 deg R
Pc = 660 psia
Solution Steps
kA = 0.0248 Btu/[(hr-sq.ft.-deg (from Fig.23-35)
F)/ft]
Tr = (300+459.67)/440=1.73 dimensionless
Pr = 700/660=1.06 dimensionless
k/kA = 1.15 dimensionless (from Fig.23-36)
k = 1.15*0.0248=0.0285 Btu/[(hr-sq.ft.-deg
F)/ft]

Example 23-13 -- Find the thermal conductivity of the gaseous mixture shown in Fig. 23-41 at
200°F and one atmosphere.

Given:
Component Mole Thermal conductivity, MW MW-3 (yi)MW-3 (yi)(ki)MW-3
Fraction Btu/(hr-ft-degF)
CO2 0.10 0.0127 44.01 3.530 0.3530 0.00448
H2S 0.20 0.0136 34.076 3.242 0.6483 0.00882
N2 0.05 0.0175 28.013 3.037 0.1518 0.00266
CH4 0.60 0.0258 16.043 2.522 1.5131 0.03904
C2H6 0.05 0.0176 30.07 3.109 0.1555 0.00274
Total 1.00 2.8218 0.05773
Solution
km = 0.05773/2.8218=0.0205 Btu/[(hr-sq.ft.-degF)/ft]

Calculate Naphtha properties (S.G =0.727) at 25oC and vapor naphtha at 2 kg/cm2.a & 150o C
density 0.727*1000*(1-0.0005*((25*1.8+32)-60)) 720.8 Kg/m3
Cp = (1/ 𝑠𝑔))*(0.388+0.00045*(t*1.8+32)) 0.49569 Kcal/kg.C
Cv Cp-(0.09/sg) 0.3719 Kcal/kg.C
thermal conductivity = (0.813/sg)*(1-0.0003*((t*1.8+32)-32)) *0.1442279 0.15911 W/m.k
viscosity 3.518-0.01591*t+1.734*10-5t2 2.39574 cP
Mol. Weight 1/(0.0001644 ∗ 𝐴𝑃𝐼 − 0.000972) 106.29846 -
16.018463 ∗ 𝑚_𝑤𝑡 ∗ 𝑝 ∗ 14.7 5.92723 Kg/m3
Gas Density =
10.732 ∗ 𝑧 ∗ (𝑡 + 459.67)
Gas thermal ( -0.000500777+1.0906*10-5*(t*1.8+32)+0.061137256/√M.wt + 0.02316 W/m.k
conductivity 0.000158966*(t*1.8+32)/√M.wt)*1.730735
=-0.0092696 +T0.5(0.001383-5.9712x10-5M0.5) 0.00771 cP
Gas Viscosity +1.1249x10-5M
Source 1: Bureau of standards report

Report Conclusions
Specific volume of vapor: The data given in Table 9 for products of gravity 50°
to 150° A. P. I. appear to be sufficiently reliable for most industrial purposes. The
data given for products of gravity 20° to 50° A. P. I. are admittedly approximate and
are included mainly for the benefit of those users of petroleum products who have no
information at hand as to the source or the volatility of these products. Additional
experimental data on molecular weights of products of gravity 10° to 40° A. P. I.
appear desirable.
Thermal conductivity: The data given in Table 10 appear to be sufficient for all
practical uses of such data.
Specific heat: The data given in Tables 12 and 13 appear to be sufficiently
reliable for all practical applications wherein only moderate pressures (less than 50
lbs. /in.2) are involved. Additional experimental data on petroleum vapors and also
on petroleum liquids at high temperatures and pressures are desirable.
Heat content: The data given in Tables 16 to 19 should provide simplicity and
convenience in calculations of the quantities of heat involved in the heating and
cooling of petroleum products.
Figure 23-15: Density
correction for
compressibility of
hydrocarbon liquids
Figure 23-14: Pseudo Density of mixtures
Figure 23-17: Density Correction for Thermal Expansion of Hydrocarbon Liquids
FIG. 23-33
Thermal Conductivity
of Miscellaneous
Gases
at One Atmosphere

FIG. 23-32 Thermal Conductivity Ratio for Gases

FIG. 23-34
Thermal Conductivity
of H.Cs Gases
at One Atmosphere
Figure 23-4: Standing-Katz Chart
Figure 23-4: Standing-Katz Chart
Supplementary Material

Metal Composition(%) Properties sat 293 K(20◦C)

Cp k α × 105
(J/kg·K) (W/m·K) (m2 /s)
Aluminum 94-96A1,3-
Duralumin 5Cu,traceMg 833 164 6.676
Silumin
Copper 87A1,13Si 871 164 7.099
Aluminum 95Cu,5A1 410 83 2.330
bronze
Bronze 75Cu,25Sn 343 26 0.859
Red brass 85Cu,9Sn,5Zn 385 61 1.804
Brass 70Cu,30Zn 385 111 3.412
German silver 62Cu,15Ni,22Zn 394 24.9 0.733
Constantan
Iron 60Cu,40Ni 410 22.7 0.612
Castiron ∼4C 420 52 1.702
Wrought iron
Steel 460 59 1.626
Carbon steel 1C 473 43 1.172
1.5C 486 36 0.970
Chromesteel 1Cr 460 61 1.665
5Cr 460 40 1.110
10Cr 460 31 0.867
Chrome-nickel 15Cr,10Ni 460 19 0.526
steel 20Cr,15Ni 460 15.1 0.415
Nickel steel 10Ni 460 26 0.720
20Ni 460 19 0.526
40Ni 460 10 0.279
60Ni 460 19 0.493
Nickel-chrome 80Ni,15Cr 460 17 0.444
steel 60Ni,15Cr 460 12.8 0.333
40Ni,15Cr 460 11.6 0.305
20Ni,15Cr 460 14.0 0.390
Manganese 1Mn 460 50 1.388
steel 5Mn 460 22 0.637
Silicon steel 1Si 460 42 1.164
5Si 460 19 0.555
Stainless steel Type304 461 14.4 0.387
Type347 461 14.3 0.387
Tungsten steel 1W 448 66 1.858
5W 435 54 1.525
Source: Process Heat Transfer Principles and Applica ons
Fluids U (W/m2 K)
Water to water 1300-2500
Ammonia to water 1000-2500
Gases to water 10-250
Water to compressed air 50-170
Water to lubricating oil 110-340
Light organics (µ < 5 X 10^-4 Ns/m2) to water 370-750
Medium organics (5 x 10^-4 < µ < 10x10^-4 Ns/m2 to
240-650
water
Heavy organics (µ > 10 X 10^-4 Ns/m2 to lubricating oil 25-400
Steam to water 2200-3500
Steam to ammonia 1000-3400
Water to condensing ammonia 850-1500
Water to boiling Freon-12 280-1000
Steam to gases 25-240
Steam to light organics 490-1000
Steam to medium organics 250-500
Steam to heavy organics 30-300
Light organics to light organics 200-350
Medium organics to medium organics 100-300
Heavy organics to heavy organics 50-200
Light organics to heavy organics 50-200
Heavy organics to light organics 150-300
Crude oil to gas oil 130-320
Plate heat exchangers: water to water 3000-4000
Evaporators: steam/water 1500-6000
Evaporators: steam/other fluids 300-2000
Evaporators of refrigeration 300-1000
Condensers: steam/water 1000-4000
Condensers: steam/other fluids 300-1000
Gas boiler 10-50
Oil bath for heating 30-550
Source: Petroleum Reﬁning Design and applica ons, vol. 4, A. Kayode coker
Typical film heat
Fluid Fluid condition transfer coefficients
(W/m2.K)

Water Liquid 5,000-7,500

Ammonia Liquid 6,000-8,000

Light organics Liquid 1,500-2,000

Medium organics Liquid 750-1,500

Heavy organics Liquid

Heating 250-750
Cooling 150-400
Very heavy
Liquid
organics
Heating 100-300
Cooling 60-150

Gas 1-2 bar abs 80-125

Gas 10 bar abs 250-400

Gas 100 bar abs 500-800

Condensing Heat
Transfer
Steam, ammonia No noncondensable 8,000-12,000
Pure component, 0.1 bar abs, no
Light organics 2,000-5,000
noncondensable

Light organics 0.1 bar, 4% noncontensable 750-1,000

Medium organics Pure or narrow condensing range, 1 bar abs 1,500-4,000

Heavy organics Narrow condensing range, 1 bar abs 600-2,000 .

Light
multicomponent
Medium condensing range, 1 bar abs 1,000-2,500
mixture, all
condensable
Medium
multicomponent
Medium condensing range, 1 bar abs 600-1,500
mixture, all
condensable
Heavy
multicomponent
Medium condensing range, 1 bar abs 300-600
mixture, all
condensable
Vaporizing Heat
Transfer

Water Pressure <5 bar abs, dT = 25 K 5,000-10,000

Water Pressure 5-100 bar abs, dT = 20 K 4,000-15,000

Ammonia Pressure <30 bar abs, dT = 20 K 3,000-5,000

Pure component, pressure < 30 bar abs, dT = 20

Light organics 2,000-4,000
K

Narrow bolling range, pressure 20-150 bar abs,

Light organics 750-3,000
dT at 15-20

Narrow boiling range, pressure < 20 bar

Medium organics 600-2,500
abs,dTmax = 15 K

Narrrow boiling range, pressure <20 bar abs,

Heavy organics 400-1,500
dTmax = 15 K
Source: Petroleum Reﬁning Design and applica ons, vol. 4, A. Kayode coker
Stream fouling factors
Natural gas 0.0002–0.0004
Column verhead products 0.0002
Lean oil 0.0004
Rich oil 0.0002
Natural gasoline and liquefied petroleum gases 0.0002
Atmospheric tower overhead vapors 0.0002
Light Naphtha 0.0004
Vacuum overhead vapors 0.0008
SALT Crude oil 0 to 120°C velocity m/s <0.6 0.0005
SALT Crude oil 0 to 120°C velocity m/s 0.6-1.2 0.0004
SALT Crude oil 0 to 120°C velocity m/s <1.2 0.0004
SALT Crude oil 120°C to 180°C velocity m/s <0.6 0.0009
SALT Crude oil 120°C to 180°C velocity m/s 0.6-1.2 0.0007
SALT Crude oil 120°C to 180°C velocity m/s <1.2 0.0007
DRY Crude oil 0 to 120°C velocity m/s <0.6 0.0005
DRY Crude oil 0 to 120°C velocity m/s 0.6-1.2 0.0004
DRY Crude oil 0 to 120°C velocity m/s <1.2 0.0004
DRY Crude oil 120°C to 180°C velocity m/s <0.6 0.0005
DRY Crude oil 120°C to 180°C velocity m/s 0.6-1.2 0.0004
DRY Crude oil 120°C to 180°C velocity m/s <1.2 0.0004
SALT Crude oil 180°C to 230°C velocity m/s <0.6 0.001
SALT Crude oil 180°C to 230°C velocity m/s 0.6-1.2 0.0009
SALT Crude oil 180°C to 230°C velocity m/s <1.2 0.0007
SALT Crude oil 230°C and more velocity m/s <0.6 0.0012
SALT Crude oil 230°C and more velocity m/s 0.6-1.2 0.001
SALT Crude oil 230°C and more velocity m/s <1.2 0.001
DRY Crude oil 180°C to 230°C velocity m/s <0.6 0.0007
DRY Crude oil 180°C to 230°C velocity m/s 0.6-1.2 0.0005
DRY Crude oil 180°C to 230°C velocity m/s <1.2 0.0004
DRY Crude oil 230°C and more velocity m/s <0.6 0.0009
DRY Crude oil 230°C and more velocity m/s 0.6-1.2 0.0007
DRY Crude oil 230°C and more velocity m/s <1.2 0.0007
Gasoline 0.0004
Naphtha and light distillates 0.0005
Kerosene 0.0005
Light gas oil 0.0005
Heavy gas oil 0.0005–0.0008
Heavy fuel oils 0.0008–0.0012
Vacuum tower bottoms 0.002
Atmosphere tower bottoms 0.0014
Cracking and coking units - Overhead vapors 0.0004
Cracking and coking units - Light cycle oil 0.0004
Cracking and coking units - Heavy cycle oil 0.0005–0.0007
Fuel oil no. 2 0.0004
Fuel oil no. 6 0.0009
Transformer oil 0.0002
Engine lube oil 0.0002
Quench oil 0.0007
Refrigerant liquids 0.0002
Hydraulic fluids 0.0002
Industrial organic heat transfer media 0.0002
Molten salts 0.00009
Ammonia, liquid 0.0002
Ammonia, liquid (or bearing) 0.0005
Calcium chloride solutions 0.0005
Manufactured gas 0.002
Diesel engine exhaust gas 0.002
Steam (nonoil bearing) 0.00009
Exhaust team (oil bearing) 0.004
Compressed air 0.002
Refrigerant vapors in refrigerant cycle condensers 0.004
Ammonia vapor 0.0002
CO vapor 2 0.0002
Chlorine vapor 0.0004
Coal fuel gas 0.002
Natural gas flue gas 0.0009
MEA and DEA solutions 0.0004
DEG and TEG solutions 0.0009
Stable side draw and bottom products 0.0002
Caustic solutions 0.0004
Vegetable oils 0.0005
Acid gases 0.0004
Solvent vapors 0.0002
Stable overhead products 0.0002
Crude and vapor unit gases and vapors 0.0002
Atmospheric tower overhead vapors 0.0002
Overhead vapors - Cracking and coking unit streams 0.0004
Light-cycle oil - Cracking and coking unit streams 0.0004
Heavy-cycle oil - Cracking and coking unit streams 0.0005–0.0007
Light coker gas oil - Cracking and coking unit streams 0.0005–0.0007
Heavy coker gas oil - Cracking and coking unit streams 0.0007–0.0009
Bottom slurry oil (minimum 4.5 ft/s) - Cracking and coking unit
streams 0.0005
Light liquid products - Cracking and coking unit streams 0.0005
Feed - Naphtha hydrotreater 0.0005
Effluent - Naphtha hydrotreater 0.0004
Naphtha - Naphtha hydrotreater 0.0004
Overhead vapors - Naphtha hydrotreater 0.0003
Overhead vapor- Visbreaker 0.0005
Visbreaker bottoms- Visbreaker 0.0020
Source: Petroleum Reﬁning Design and applica ons, vol. 4, A. Kayode coker & TEMA 9th edi on
Tubesheet type and no. of
passes tube size and pitch 8 10 12 13¼ 15¼ 17¼ 19¼ 21¼ 23¼ 25 27 29 31 33 35 37

Fixed Tubes One-Pass ¾ in. on 15/ 16 in ∆ 33 69 105 135 193 247 307 391 481 553 663 763 881 1019 1143 1,269

Fixed Tubes One-Pass ¾ in. on 1in. ∆ 33 57 91 117 157 217 277 343 423 493 577 667 765 889 1007 1,127

Fixed Tubes One-Pass ¾ in. on 1in ◽ 33 53 85 101 139 183 235 287 355 419 495 587 665 765 865 965

Fixed Tubes One-Pass 1 in. on ¼ in. ∆ 15 33 57 73 103 133 163 205 247 307 361 427 481 551 633 699

Fixed Tubes One-Pass 1 in. on 1 ¼ in ◽ 17 33 45 65 83 111 139 179 215 255 303 359 413 477 545 595

Fixed Tubes Two-Pass ¾ in. on 15/ 16 in ∆ 32 58 94 124 166 228 300 370 452 528 626 734 846 964 1088 1,242

Fixed Tubes Two-Pass ¾ in. on 1 in. ∆ 28 56 90 110 154 208 264 326 398 468 556 646 746 858 972 1,088

Fixed Tubes Two-Pass ¾ in. on 1 in ◽ 26 48 78 94 126 172 222 280 346 408 486 560 644 746 840 946

Fixed Tubes Two-Pass 1 in. on 1 ¼ in ∆ 16 32 52 62 92 126 162 204 244 292 346 410 462 530 608 688

Fixed Tubes Two-Pass 1 in. on 1 ¼ in ◽ 12 26 40 56 76 106 136 172 218 248 298 348 402 460 522 584

U Tubes Two-Pass ¾ in. on 15/ 16 in ∆ 8 34 64 94 134 180 234 304 398 460 558 648 768 882 1008 1,126

U Tubes Two-Pass ¾ in. on 1in. ∆ 8 26 60 72 108 158 212 270 336 406 484 566 674 772 882 1,000

U Tubes Two-Pass ¾ in. on 1in ◽ 12 30 52 72 100 142 188 242 304 362 436 506 586 688 778 884

U Tubes Two-Pass 1 in. on 1 ¼ in ∆ XX 8 26 42 58 84 120 154 192 234 284 340 396 466 532 610

U Tubes Two-Pass 1 in. on 1 ¼ in ◽ XX 12 22 38 58 76 100 134 180 214 256 304 356 406 464 526

Fixed Tubes Four-Pass ¾ in. on 15/ 16 in ∆ XX 48 84 108 154 196 266 332 412 484 576 680 788 904 1024 1,072

Fixed Tubes Four-Pass ¾ in. on 1in. ∆ XX 44 72 96 134 180 232 292 360 424 508 596 692 802 912 1,024

Fixed Tubes Four-Pass ¾ in. on 1in ◽ XX 48 72 88 126 142 192 242 308 366 440 510 590 688 778 880
Tubesheet type and no. of
passes tube size and pitch 8 10 12 13¼ 15¼ 17¼ 19¼ 21¼ 23¼ 25 27 29 31 33 35 37

Fixed Tubes Four-Pass 1 in. on 1 ¼ in ∆ XX 24 44 60 78 104 138 176 212 258 308 368 422 486 560 638

Fixed Tubes Four-Pass 1 in. on 1 ¼ in ◽ XX 24 40 48 74 84 110 142 188 214 260 310 360 414 476 534

U Tubes Four-Pass ¾ in. on 15/ 16 in ∆ XX 28 56 84 122 166 218 286 378 438 534 622 740 852 976 1,092

U Tubes Four-Pass ¾ in. on 1in. ∆ XX 20 52 64 98 146 198 254 318 386 462 542 648 744 852 968

U Tubes Four-Pass ¾ in. on 1in ◽ XX 24 44 64 90 130 174 226 286 342 414 482 560 660 748 852

U Tubes Four-Pass 1 in. on 1 ¼ in ∆ XX XX 20 36 50 74 110 142 178 218 266 322 376 444 508 584

U Tubes Four-Pass 1 in. on 1 ¼ in ◽ XX XX 16 32 50 66 90 122 166 198 238 286 336 384 440 500

Fixed Tubes Six-Pass ¾ in. on 15/ 16 in ∆ XX XX XX 80 116 174 230 294 372 440 532 632 732 844 964 1,106

Fixed Tubes Six-Pass ¾ in. on 1in. ∆ XX XX XX 66 104 156 202 258 322 388 464 548 640 744 852 964

Fixed Tubes Six-Pass ¾ in. on 1in ◽ XX XX XX 54 78 116 158 212 266 324 394 460 536 634 224 818

Fixed Tubes Six-Pass 1 in. on 1 ¼ in ∆ XX XX XX 34 56 82 112 150 182 226 274 338 382 442 514 586

Fixed Tubes Six-Pass 1 in. on 1 ¼ in ◽ XX XX XX XX 44 66 88 116 154 184 226 268 318 368 430 484

U Tubes Six-Pass ¾ in. on 15/ 16 in ∆ XX XX XX 74 110 156 206 272 358 416 510 596 716 826 944 1,058

U Tubes Six-Pass ¾ in. on 1in. ∆ XX XX XX 56 88 134 184 238 300 366 440 518 626 720 826 940

U Tubes Six-Pass ¾ in. on 1in ◽ XX XX XX 56 80 118 160 210 268 322 392 458 534 632 718 820

U Tubes Six-Pass 1 in. on 1 ¼ in ∆ XX XX XX 30 42 68 100 130 168 206 252 304 356 426 488 562

U Tubes Six-Pass 1 in. on 1 ¼ in ◽ XX XX XX XX 42 60 80 110 152 182 224 268 316 362 420 478
Tubesheet type and no. of
passes tube size and pitch 8 10 12 13¼ 15¼ 17¼ 19¼ 21¼ 23¼ 25 27 29 31 33 35 37

Fixed Tubes eight-Pass ¾ in. on 15/ 16 in ∆ XX XX XX XX 94 140 198 258 332 398 484 576 682 790 902 1,040

Fixed Tubes eight-Pass ¾ in. on 1in. ∆ XX XX XX XX 82 124 170 224 286 344 422 496 588 694 798 902

Fixed Tubes eight-Pass ¾ in. on 1in ◽ XX XX XX XX XX 94 132 174 228 286 352 414 490 576 662 760

Fixed Tubes eight-Pass 1 in. on 1 ¼ in ∆ XX XX XX XX XX 66 90 120 154 190 240 298 342 400 466 542

Fixed Tubes eight-Pass 1 in. on 1 ¼ in ◽ XX XX XX XX XX XX 74 94 128 150 192 230 280 334 388 438

U Tubes eight-Pass ¾ in. on 15/ 16 in ∆ XX XX XX 68 102 142 190 254 342 398 490 578 688 796 916 1,032

U Tubes eight-Pass ¾ in. on 1in. ∆ XX XX XX 52 82 122 170 226 286 350 422 498 600 692 796 908

U Tubes eight-Pass ¾ in. on 1in ◽ XX XX XX 48 70 106 146 194 254 306 374 438 512 608 692 792

U Tubes eight-Pass 1 in. on 1 ¼ in ∆ XX XX XX 24 38 58 90 118 154 190 238 290 340 404 464 540

U Tubes eight-Pass 1 in. on 1 ¼ in ◽ XX XX XX XX 34 50 70 98 142 170 206 254 300 344 396 456

TEMA P or S One-Pass 1.25 in. on 1 9/16 in ◽ 12 21 29 38 52 70 85 108 136 154 184 217 252 289 329 372

TEMA P or S Two-Pass 1.25 in. on 1 9/16 in ◽ 12 12 28 34 48 66 84 108 128 154 180 212 248 276 316 368

TEMA P or S Four-Pass 1.25 in. on 1 9/16 in ◽ 4 8 16 34 44 56 70 100 128 142 158 204 234 270 310 354
Tubesheet type and no.
of passes tube size and pitch 8 10 12 13¼ 15¼ 17¼ 19¼ 21¼ 23¼ 25 27 29 31 33 35 37

TEMA P or S Six-Pass 1.25 in. on 1 9/16 in ◽ 0 12 18 24 48 50 80 96 114 136 172 198 236 264 304 340

U Tubes Two-Pass 1.25 in. on 1 9/16 in ◽ 4 12 26 36 44 60 82 100 128 154 176 212 242 280 324 358

U Tubes Four-Pass 1.25 in. on 1 9/16 in ◽ 4 8 20 28 44 60 76 100 120 148 172 204 240 280 312 352

U Tubes Six-Pass 1.25 in. on 1 9/16 in ◽ 6 12 12 15 32 56 79 100 120 130 160 198 234 274 308 350

TEMA L or M One-
Pass 1.25 in. on 1 9/16 in ∆ 15 27 38 55 66 88 117 136 170 198 237 268 312 357 417 446

TEMA L or M Two-
Pass 1.25 in. on 1 9/16 in ∆ 10 22 36 44 64 82 106 134 164 188 228 266 304 346 396 446

TEMA L or M Four-
Pass 1.25 in. on 1 9/16 in ∆ 8 16 26 42 52 78 98 124 146 166 208 242 284 322 372 422

TEMA L or M Six-Pass 1.25 in. on 1 9/16 in ∆ 12 12 24 40 50 68 96 108 148 168 192 236 276 324 364 408

TEMA P or S One-
Pass 1.25 in. on 1 9/16 in ∆ 13 18 33 38 57 81 100 126 159 183 208 249 291 333 372 425

TEMA P or S Two-
Pass 1.25 in. on 1 9/16 in ∆ 10 20 26 44 58 72 94 120 146 172 206 238 282 326 368 412

TEMA P or S Four-
Pass 1.25 in. on 1 9/16 in ∆ 4 8 26 34 48 62 86 116 132 150 190 224 262 298 344 394

TEMA P or S Six-Pass 1.25 in. on 1 9/16 in ∆ 0 12 18 24 44 68 80 102 132 148 180 220 256 296 336 384

U Tubes Two-Pass 1.25 in. on 1 9/16 in ∆ 6 14 28 34 52 72 90 118 148 172 200 242 282 326 362 416
Tubesheet type and no.
of passes tube size and pitch 8 10 12 13¼ 15¼ 17¼ 19¼ 21¼ 23¼ 25 27 29 31 33 35 37

U Tubes Four-Pass 1.25 in. on 1 9/16 in ∆ 4 12 20 28 48 68 84 112 132 160 188 228 264 308 344 396

U Tubes Six-Pass 1.25 in. on 1 9/16 in ∆ 6 12 18 30 40 64 78 102 120 152 180 216 250 292 336 384

TEMA P or S One-
Pass 5/8 in. on 13/16 in ◽ 55 88 140 178 245 320 405 502 610 700 843 970 1127 1288 1479 1647

TEMA P or S Two-
Pass 5/8 in. on 13/16 in ◽ 48 78 138 172 232 308 392 484 584 676 812 942 1096 1250 1438 1604

TEMA P or S Four-
Pass 5/8 in. on 13/16 in ◽ 34 62 112 146 208 274 352 442 536 618 742 868 1014 1172 1330 1520

TEMA P or S Six-Pass 5/8 in. on 13/16 in ◽ 24 56 100 136 192 260 336 424 508 600 716 840 984 1148 1308 1480

U Tubes Two-Pass 5/8 in. on 13/16 in ◽ 52 90 140 180 246 330 420 510 626 728 856 998 1148 1318 1492 1684

U Tubes Four-Pass 5/8 in. on 13/16 in ◽ 40 80 128 164 232 312 388 488 596 692 816 956 1108 1268 1436 1620

U Tubes Six-Pass 5/8 in. on 13/16 in ◽ 32 74 108 148 216 292 368 460 562 644 780 920 1060 1222 1388 1568

TEMA L or M One-
Pass 3/4 in. on 15/16 in ∆ 64 85 122 151 204 264 332 417 495 579 676 785 909 1035 1164 1304

TEMA L or M Two-
Pass 3/4 in. on 15/16 in ∆ 48 72 114 142 192 254 326 396 478 554 648 762 878 1002 1132 1270

TEMA L or M Four-
Pass 3/4 in. on 15/16 in ∆ 34 52 94 124 166 228 290 364 430 512 602 704 814 944 1062 1200

TEMA L or M Six-Pass 3/4 in. on 15/16 in ∆ 24 50 96 112 168 220 280 348 420 488 584 688 792 920 1036 1168

TEMA P or S One-
Pass 3/4 in. on 15/16 in ∆ 34 60 109 126 183 237 297 372 450 518 618 729 843 962 1090 1233
Tubesheet type and no. of
passes tube size and pitch 8 10 12 13¼ 15¼ 17¼ 19¼ 21¼ 23¼ 25 27 29 31 33 35 37

TEMA P or S Two-Pass 3/4 in. on 15/16 in ∆ 32 62 98 120 168 228 286 356 430 498 602 708 812 934 1064 1196

TEMA P or S Four-Pass 3/4 in. on 15/16 in ∆ 16 52 78 106 146 202 258 324 392 456 548 650 744 868 990 1132

TEMA P or S Six-Pass 3/4 in. on 15/16 in ∆ 18 44 68 100 136 192 248 316 376 444 532 624 732 840 972 1100

U Tubes Two-Pass 3/4 in. on 15/16 in ∆ 32 64 98 126 180 238 298 370 456 534 628 736 846 978 1100 1238

U Tubes Four-Pass 3/4 in. on 15/16 in ∆ 24 52 88 116 160 224 280 352 428 500 600 696 812 928 1060 1200

U Tubes Six-Pass 3/4 in. on 15/16 in ∆ 24 52 78 108 148 204 262 334 408 474 570 668 780 904 1008 1152

TEMA P or S One-Pass 3/4 in. on 1 in ◽ 28 52 80 104 136 181 222 289 345 398 477 554 637 730 828 937

TEMA P or S Two-Pass 3/4 in. on 1 in ◽ 26 48 76 90 128 174 220 272 332 386 456 532 624 712 812 918

TEMA P or S Four-Pass 3/4 in. on 1 in ◽ 16 44 66 70 128 154 204 262 310 366 432 510 588 682 780 882

TEMA P or S Six-Pass 3/4 in. on 1 in ◽ 12 24 56 80 114 160 198 260 308 344 424 496 576 668 760 872

U Tubes Two-Pass 3/4 in. on 1 in ◽ 28 52 78 96 136 176 224 284 348 408 480 562 648 748 848 952

U Tubes Four-Pass 3/4 in. on 1 in ◽ 24 44 72 92 132 176 224 280 336 392 468 548 636 728 820 932

U Tubes Six-Pass 3/4 in. on 1 in ◽ 12 32 70 90 120 160 224 274 328 378 460 530 620 718 816 918

TEMA L or M One-Pass 3/4 in. on 1 in ∆ 42 73 109 136 183 237 295 361 438 507 592 692 796 909 1023 1155
Tubesheet type and no.
of passes tube size and pitch 8 10 12 13¼ 15¼ 17¼ 19¼ 21¼ 23¼ 25 27 29 31 33 35 37

TEMA L or M Two-
Pass 3/4 in. on 1 in ∆ 40 66 102 128 172 228 282 346 416 486 574 668 774 886 1002 1124

TEMA L or M Four-
Pass 3/4 in. on 1 in ∆ 26 52 88 112 146 208 258 318 382 448 536 632 732 836 942 1058

TEMA L or M Six-Pass 3/4 in. on 1 in ∆ 24 44 80 102 148 192 248 320 372 440 516 604 708 812 920 1032

TEMA P or S One-
Pass 3/4 in. on 1 in ∆ 31 56 88 121 159 208 258 320 400 450 543 645 741 843 950 1070

TEMA P or S Two-
Pass 3/4 in. on 1 in ∆ 26 48 78 106 148 198 250 314 384 442 530 618 716 826 930 1052

TEMA P or S Four-
Pass 3/4 in. on 1 in ∆ 16 42 62 94 132 182 228 290 352 400 488 574 666 760 878 992

TEMA P or S Six-Pass 3/4 in. on 1 in ∆ 12 40 68 88 132 180 220 276 336 392 468 556 648 740 856 968

U Tubes Two-Pass 3/4 in. on 1 in ∆ 32 52 84 110 152 206 266 330 400 472 554 648 744 852 974 1092

U Tubes Four-Pass 3/4 in. on 1 in ∆ 24 48 76 100 140 188 248 316 384 440 528 616 716 816 932 1056

U Tubes Six-Pass 3/4 in. on 1 in ∆ 24 40 74 98 136 182 234 296 356 424 502 588 688 788 908 1008

TEMA P or S One-
Pass 1 in. on 1.25 in ◽ 17 30 52 61 85 108 144 173 217 252 296 345 402 461 520 588

TEMA P or S Two-
Pass 1 in. on 1.25 in ◽ 12 30 48 56 78 108 136 166 208 240 280 336 390 452 514 572
Tubesheet type and no. of
passes tube size and pitch 8 10 12 13¼ 15¼ 17¼ 19¼ 21¼ 23¼ 25 27 29 31 33 35 37

TEMA P or S Four-Pass 1 in. on 1.25 in ◽ 8 16 42 52 62 104 130 154 194 230 270 310 366 432 494 562

TEMA P or S Six-Pass 1 in. on 1.25 in ◽ 12 18 24 50 64 96 114 156 192 212 260 314 368 420 484 548

U Tubes Two-Pass 1 in. on 1.25 in ◽ 14 30 44 60 80 104 132 172 212 244 290 340 400 456 518 584

U Tubes Four-Pass 1 in. on 1.25 in ◽ 8 24 40 48 72 100 132 168 204 240 284 336 384 444 504 576

U Tubes Six-Pass 1 in. on 1.25 in ◽ 6 12 32 44 74 100 120 148 198 230 274 328 372 440 502 566

TEMA L or M One-Pass 1 in. on 1.25 in ∆ 27 42 64 81 106 147 183 226 268 316 375 430 495 579 645 729

TEMA L or M Two-Pass 1 in. on 1.25 in ∆ 26 40 66 74 106 134 176 220 262 302 360 416 482 554 622 712

TEMA L or M Four-Pass 1 in. on 1.25 in ∆ 8 34 52 62 88 124 150 204 236 274 336 390 452 520 586 662

TEMA L or M Six-Pass 1 in. on 1.25 in ∆ 12 24 44 56 92 114 152 186 228 272 324 380 448 504 576 648

TEMA P or S One-Pass 1 in. on 1.25 in ∆ 18 33 51 73 93 126 159 202 249 291 345 400 459 526 596 672

TEMA P or S Two-Pass 1 in. on 1.25 in ∆ 14 28 48 68 90 122 152 192 238 278 330 388 450 514 584 668

TEMA P or S Four-Pass 1 in. on 1.25 in ∆ 8 16 42 52 78 112 132 182 216 250 298 356 414 484 548 626

TEMA P or S Six-Pass 1 in. on 1.25 in ∆ 12 18 44 44 76 102 136 172 212 240 288 348 400 464 536 608

U Tubes Two-Pass 1 in. on 1.25 in ∆ 14 28 52 64 90 122 152 196 242 286 340 400 456 526 596 668

U Tubes Four-Pass 1 in. on 1.25 in ∆ 12 24 40 56 80 112 140 180 224 264 320 380 436 504 572 636

U Tubes Six-Pass 1 in. on 1.25 in ∆ 6 24 40 52 78 102 136 176 216 246 300 352 414 486 548 614

Source : Source: Process Heat Transfer Principles and Applica ons & Petroleum Reﬁning Design and applica ons, vol. 4, A. Kayode coker
Birmingham Wire Gage (inches) (mm)
BWG
7 0.180 4.572
8 0.165 4.191
9 0.148 3.759
10 0.134 3.404
11 0.120 3.048
12 0.109 2.769
13 0.095 2.413
14 0.083 2.108
15 0.072 1.829
16 0.065 1.651
17 0.058 1.473
18 0.049 1.245
19 0.042 1.067
20 0.035 0.889
21 0.032 0.813
22 0.028 0.711
23 0.025 0.635
24 0.022 0.559
25 0.020 0.508
26 0.018
27 0.016
28 0.014
29 0.013
30 0.012
31 0.010
32 0.009
33 0.008
34 0.007
35 0.005
36 0.004
Source: Heat Exchangers: Selec on, Ra ng, and Thermal Design
Source: Heat Exchangers: Selec on, Ra ng, and Thermal Design
Fired Heater Efficiency Calculations

Link: Fired Heater Efficiency Calculations

This tool was developed for process engineers to quickly estimate heaters efficiency using heat loss
simplified method and the infamous API 560 also indirect heat loss method and access a list of useful
tables on site. As a part of a larger project to develop what is similar to Carl Branan’s book “process
engineers Pocket Handbook” these tools would allow a process engineer to quickly calculate/estimate
equipment efficiencies or sizing using standardized calculations

The aim here is to take little-known data from the field (flow, pressures, temperatures, compositions..etc.)
and use it as input for a rough estimation without having to return to the office to use commercial
software or calculations Excel sheets to validate or to calculate. Additionally, these tools may also serve as
a gathered data validation tool.

Nomenclature

T Temperature (oF) P Pressure (psia)

ρ Density (kg/m3) Cp Heat Capacity (BTU/lb.oF)
Pv Vapor pressure (psia) ha Air Heat Correction (BTU/lb)
hl Fuel LHV (BTU/lb) hf Fuel Heat Correction (BTU/lb)
hs Stack heat correction(BTU/lb) hm atomizing medium heat correction
η Fired Heater Efficiency (%) hr radiant section Heat Loss(BTU/lb)
Required Input
Table 1: Note 1: Input required depending on the case

Simple Equation API Gas API fuel oil

O2 percent in Flue gas O2 percent in Flue gas O2 percent in Flue gas
Stack temperature oC Stack temperature oC Stack temperature oC
Gas composition in vol% Fuel oil specific gravity
Atomizing steam & Ts Fuel oil LHV [optional]
Datum design T/ Humidity % Fuel oil CH ratio [optional]
ambient T [optional] Atomizing steam & Ts/ ambient T [optional]
Fuel Temp [optional] Datum design T/ Humidity %/ Fuel Temp [optional]
Obtained Output

Simple Equation API Gas API fuel oil

Simpilified Efficiency % Simpilified Efficiency % Simpilified Efficiency %
Thermal efficiency Thermal Efficiency
Fuel Efficiency Fuel Efficiency
API spread sheet API spread sheet
Flue gases heat content table Flue gases heat content table
Gas LHV Fuel oil LHV/ CH

1. Simplified Equation[2]

𝑓𝑙𝑢𝑒 𝑔𝑎𝑠 𝑂2 %∗91.2

Excess Air (EA)% = , when O2 analysis is Dry
20.95−𝑓𝑙𝑢𝑒 𝑔𝑎𝑠 𝑂2 %

𝑓𝑙𝑢𝑒 𝑔𝑎𝑠 𝑂2 %∗111.4

Excess Air (EA)% = , when O2 analysis is Wet
20.95−𝑓𝑙𝑢𝑒 𝑔𝑎𝑠 𝑂2 %
(0.0237+0.000189∗𝐸𝐴%)(𝑇𝑠𝑡𝑎𝑐𝑘 −𝑇𝑎𝑚𝑏. )∗100
Efficiency η = 100 −
100+𝑄𝐿

Where ambient is assumed 80 oF, QL is Casing heat loss% and assumed 2%

2. API 560 Annex G Method[1]

2.1 Fuel Gas

1. Collect data

Fluids properties

• Fuel composition (S.G., CH ratio & LHV if fuel, Complete composition & LHV if gas)
• Air & Fuel inlet conditions (humidity% and T)
• Calorific values (preferred for fuel oil but can be calculated for gas)

Furnace process data

• Key temperatures (ambient, fuel, air and stack temperatures)

• Flows (not a necessity)
• Flue gas composition (% O2, %CO2)

2. Calculate The following Using API spreadsheet:

a. Dry Air formed (lbs)/lb of fuel

b. CO2 formed (lbs)/lb of fuel
c. H2O formed (lbs)/lb of fuel
d. SO2 formed (lbs)/lb of fuel
e. N2 formed (lbs)/lb of fuel

𝑚 = ∑𝑛𝑖=0 𝑥𝑖 . 𝑚𝑖 where:

m: amount formed or required of a flue gas (Air, N2, H2O, CO2 or SO2) per lb fuel.
xi : mass fraction of fuel gas component.
mi: amount formed by the fuel gas component per lb fuel.
n: number of fuel gas components.

f. Gas mixture LHV BTU/lb (hl)

𝑛

𝐿𝐻𝑉 𝑝𝑒𝑟 𝑙𝑏 𝑓𝑢𝑒𝑙 = ∑ 𝑥𝑖 . 𝐿𝐻𝑉𝑖

𝑖=0

3. Calculate pounds of moisture per pound of fuel = Pounds of wet air per pound of fuel
- Air required:

a. Moisture in Air

𝑃𝑣𝑎𝑝 𝑅𝑒𝑙𝑎𝑡𝑖𝑣𝑒 ℎ𝑖𝑚𝑖𝑑𝑖𝑡𝑦% 18

Moisture in air = × × (Pounds of moisture per pound of wet air)
14.696 100 28.85
Where:
Pvapor = vapor pressure of water at ambient temperature in psia from steam tables or from
the following equation:
Note: Water vapor pressure calculated per Wagner and Pruss (1993)
T in °K and
ln(Psat/Pc) = (a1τ + a2τ1.5 + a3τ3 + a4τ3.5 + a5τ4 + a6τ7.5)Tc/Ta P in kPa
Tc = 647.1 pc = 22,064.0 a1 = -7.8595 a2 = 1.8441
a3 = -11.7866 a4 = 22.6807411 a5 = -15.962 a6 = 1.8012
τ =1−T/Tc = 0.5813 ln(Psat/Pc) = -10.657

b. Calculate Pounds of wet air per pound of fuel required

𝐴𝑖𝑟 𝑟𝑒𝑞𝑢𝑖𝑟𝑒𝑑
Pounds of wet air per pound of fuel required=
1−𝑀𝑜𝑖𝑠𝑡𝑢𝑟𝑒 𝑖𝑛 𝑎𝑖𝑟

c. Calculate Pounds of moisture per pound of fuel

= Pounds of wet air per pound of fuel - Air required per pound of fuel

d. Calculate Corrected Pounds of H2O per pound of fuel

Pounds of H2O per pound of fuel

= H2O formed + Pounds of moisture per pound of fuel + atomizing steam

4. Calculate Excess Air percent

a. Calculate Pounds of dry excess air per pound of fuel

𝑁2 𝑓𝑜𝑟𝑚𝑒𝑑 𝐶𝑂2 𝑓𝑜𝑟𝑚𝑒𝑑 𝐻2𝑂 𝑓𝑜𝑟𝑚𝑒𝑑
(28.85 ×𝑃𝑒𝑟𝑐𝑒𝑛𝑡 𝑂2)( 28
+ 44
+ 18
)
Pounds of dry excess air per pound of fuel = 𝑃𝑜𝑢𝑛𝑑𝑠 𝑜𝑓 𝐻2𝑂
20.95−𝑃𝑒𝑟𝑐𝑒𝑛𝑡 𝑂2 ((1.6028×𝑃𝑜𝑢𝑛𝑑𝑠 𝑜𝑓 𝑎𝑖𝑟 𝑟𝑒𝑞𝑢𝑖𝑟𝑒𝑑)+1)

𝑃𝑜𝑢𝑛𝑑𝑠 𝑜𝑓 𝑒𝑥𝑐𝑒𝑠𝑠 𝑎𝑖𝑟 𝑝𝑒𝑟 𝑝𝑜𝑢𝑛𝑑 𝑜𝑓 𝑓𝑢𝑒𝑙

Percent excess air % = × 100
𝐴𝑖𝑟 𝑟𝑒𝑞𝑢𝑖𝑟𝑒𝑑

5. Calculate total pounds of H2O per pound of fuel (corrected for excess air)
𝑝𝑒𝑟𝑐𝑒𝑛𝑡 𝑒𝑥𝑐𝑒𝑠𝑠 𝑎𝑖𝑟
Total lbs of H2O /lb of fuel (corrected for excess air) = 100
× 𝑝𝑜𝑢𝑛𝑑𝑠 𝑜𝑓 𝑚𝑜𝑖𝑠𝑡𝑢𝑟𝑒 𝑝𝑒𝑟 𝑝𝑜𝑢𝑛𝑑 𝑜𝑓 𝑓𝑢𝑒𝑙

6. Calculate Stack Heat loss

a. From formed flues gases per lb fuel calculate enthalpies then heat content for each flue
gas enthalpies can be estimated using Figure G.6 and Figure G.7 or by the following
approximation N2.
Component Composition Enthalpy Heat
Wet Dry
lb/lb mol/lb mol% mol% Btu/lb Btu/lb
Nitrogen, N2 12.95 0.46 59.03 71.34 56.98 738
Carbon dioxide, CO2 2.42 0.05 7.01 8.47 52.33 126
Water, H2O 2.43 0.14 17.26 100.77 245
Excess O2 0.88 0.03 3.50 4.23 53.17 47
Excess N2 2.90 0.10 13.21 15.96 56.98 165
Total per 21.58 0.78 100.00 100.00 1,322

HN2 =-19.929903 + 0.25445255 Ts+ 5.572710310-6* Ts2+ 3.157217210-9 Ts 3 -1.429813710-12 Ts4

HH2O =-18.059441 + 0.22117674* Ts + 5.638387*10-5* Ts2 -4.3167144*10-8* Ts 3+ 1.5150613* 10-11*Ts4
HH2O =-11.245005 + 0.30633377* Ts + 2.6291631*10-4* Ts2 -1.4045071*10-7* Ts 3+ 3.0355352*10-11* Ts4
HO2 =-19.19339 + 0.25924657* Ts -8.4198438*10-5* Ts2+ 8.7932994*10-8* Ts 3 -2.5240497*10-11* Ts4
HSO2 = -4*10-6 *Ts 2 + 0.208 *Ts - 23.292
Heat Content = Enthalpy * mflue gas/mfuel
Where Enthalpy in Btu/lb and Ts is T stack in oF (approximated for figures G.6 and G.7 in Ref. 1)
b. Sum of heat content represents hs (Stack heat loss)

7. Calculate radiant section Heat Loss (Assume 2% heat loss of total LHV)
hr = 2*0.01*hl
8. Calculate hf Fuel sensible Heat correction BTU/lb of fuel
hf = Cpf.(Tf – Td)
where Cpf is fuel heat capacity, Tf: fuel temperature and Td design datum temperature
9. Calculate ha air sensible heat correction BTU/lb of fuel
ha = Cpair (Tambient – Td).ma/mf
ma/mf is total air per lb fuel
10. Calculate atomizing medium heat correction hm BTU/lb of fuel
hm = Cpm(Tm – Td).mm/mf or mm/mf . ∆Hm
where Cpm is steam heat capacity, Tf: steam temperature and Td design datum temperature
11. Calculate Efficiencies
a. Net Thermal Efficiency
𝐻𝑒𝑎𝑡 𝑈𝑠𝑒𝑑
𝜂=
𝐻𝑒𝑎𝑡 𝑐𝑜𝑛𝑡𝑒𝑛𝑡 𝑓𝑜𝑟 𝑎𝑙𝑙 𝑖𝑛𝑝𝑢𝑡
((𝐻𝑒𝑎𝑡 𝐹𝑟𝑜𝑚 𝐹𝑢𝑒𝑙 + 𝐴𝑖𝑟 𝐻𝑒𝑎𝑡 + 𝐹𝑢𝑒𝑙 𝐻𝑒𝑎𝑡 + 𝐴𝑡𝑜𝑚𝑖𝑧𝑖𝑛𝑔 𝑠𝑡𝑒𝑎𝑚 ℎ𝑒𝑎𝑡)‐ (𝑟𝑎𝑑𝑖𝑎𝑛𝑡 𝐿𝑜𝑠𝑠 + 𝑆𝑡𝑎𝑐𝑘 𝐿𝑜𝑠𝑠))
=
(𝐻𝑒𝑎𝑡 𝐹𝑟𝑜𝑚 𝐹𝑢𝑒𝑙 + 𝐴𝑖𝑟 𝐻𝑒𝑎𝑡 + 𝐹𝑢𝑒𝑙 𝐻𝑒𝑎𝑡 + 𝐴𝑡𝑜𝑚𝑖𝑧𝑖𝑛𝑔 𝑠𝑡𝑒𝑎𝑚 ℎ𝑒𝑎𝑡)
(ℎ𝐿 + ℎ𝑎 + ℎ𝑓 + ℎ𝑚 ) − (ℎ𝑟 − ℎ𝑠 )
=
ℎ𝐿 + ℎ𝑎 + ℎ𝑓 + ℎ𝑚

b. Fuel Efficiency

𝐻𝑒𝑎𝑡 𝐴𝑏𝑠𝑜𝑟𝑏𝑒𝑑 𝐻𝑒𝑎𝑡 𝐹𝑟𝑜𝑚 𝐹𝑢𝑒𝑙 (ℎ𝐿 + ℎ𝑎 + ℎ𝑓 + ℎ𝑚 ) − (ℎ𝑓 − ℎ𝑠 )

𝜂𝑓 = ∗ 100 =
𝑇𝑜𝑡𝑎𝑙 𝐻𝑒𝑎𝑡 𝑎𝑣𝑎𝑖𝑙𝑎𝑏𝑙𝑒 ℎ𝐿
2.2 Fuel Oil
1. Obtain CH ratio and LHV analysis of fuel oil or estimate them

API = 141.5/SG -131.5

CH =0.0022 API2 – 0.1296 API + 9.1398, approximated from Maxwell Databook P. 186-188
LHV = -0.3726 API2 – 57.429 API + 16797 (btu/lb), approximated from Maxwell Databook P. 180

2. Use Carbon ratio and other Oil content (H2O, ash, sulfur, Sodium and others) to estimate C% and
H% and other contents

Component Mol Mass Air CO2 H2O N2 SO2

Mass % lb/lb
Carbon, C 12.0 88.50 11.51 3.66 0.00 8.85 0
Hydrogen, H2 2.016 11.50 34.29 0.00 8.94 26.36 0
Sulfur, S 32.1 0.00 4.31 2.00 0.00 3.31 2
Inerts 0.00 0.00 0.00 0.00 0.00 0
Total per lb of fuel 100.00 14.13 3.24 1.03 10.86 0
3. Input calculated volume percent in API spreadsheet and Repeat Gas calculations steps from 1 to 10

References
1. API 560 ANNEX G.
2. Aramco stds: calculating process heater thermal efficiency

Recommended Readings
1. Aramco stds: calculating process heater thermal efficiency (extended document on the
various methods of estimating thermal efficiencies (direct & indirect methods)
2. Correctly Modeling and Calculating Combustion Efficiencies In Fired Equipment,
David Schmitt
3. Get the most from your fired heater, ashutosh Garg and H. Ghosh chemical engineering.
4. Optimize fired heater operations to save money , A. Garg, Hydrocarbon processing.
Column Column Column Column
Column 1 Column 2 3 4 Column 5 Column 6 Column 7 Column 8 Column 9 Column 10 Column 11 12 Column 13 14

Net SO2
Heating Dry Air H2O N2
Heating CO2 formed formed
Fuel Component total Value required Dry Air CO2 formed H2O formed
Volume value (pounds of N2 formed (pounds
M.Wt weight (british (pounds of required formed (pounds of formed (pounds
fraction (british CO2 per (lbs) of SO2
(lbs) thermal air per (lbs) (lbs) H2O per (lbs) of N2 per
thermal pound) per
units) pound) pound) pound)
units) pound)

Carbon 12 0.00 0 0.00 11.51 0.00 3.66 0 8.85 0 0

Hydrogen 73.98 2.016 1.49 51600 76958.14 34.29 51.14 0 0 8.94 13.33344 26.36 39.314274 0

Oxygen 32 0.00 0 0.00 -4.32 0.00 0 0 0 0 -3.32 0 0

Nitrogen 2.79 28 0.78 0 0.00 0 0.00 0 0 0 0 1 0.7812 0

Carbon monoxide 28 0.00 4345 0.00 2.47 0.00 1.57 0 0 0 1.9 0 0

Carbon dioxide 44 0.00 0 0.00 0 0.00 1 0 0 0 0 0 0

Methane 5.42 16 0.87 21500 18644.8 17.24 14.95 2.74 2.376128 2.25 1.9512 13.25 11.4904 0

Ethane 3.72 30.1 1.12 20420 22864.68 16.09 18.02 2.93 3.28078 1.8 2.015496 12.37 13.8509364 0

Ethylene 28.1 0.00 20290 0.00 14.79 0.00 3.14 0 1.28 0 11.36 0 0

Propane 4.36 44.1 1.92 19930 38320.61 15.68 30.15 2.99 5.749052 1.63 3.134099 12.05 23.169258 0

Propylene 42.1 0.00 19690 0.00 14.79 0.00 3.14 0 1.28 0 11.36 0 0

Butane 4.89 58.1 2.84 19670 55884.24 15.46 43.92 3.03 8.608503 1.55 4.40369 11.88 33.7521492 0

Butylene 56.1 0.00 19420 0.00 14.79 0.00 3.14 0 1.28 0 11.36 0 0

Pentane 3.93 72.1 2.83 19500 55253.84 15.33 43.44 3.05 8.642267 1.5 4.250295 11.78 33.3789834 0

Hexane 0.23 86.2 0.20 19390 3844.26 15.24 3.02 3.06 0.606676 1.46 0.28946 11.71 2.3216246 0

Benzene 78.1 0.00 17270 13.27 3.38 0.69 10.2 0 0

Methanol 32 0.00 8580 6.48 1.38 1.13 4.98 0 0

Ammonia 17 0.00 8000 6.1 0 1.59 5.51 0 0

Sulfur 32.1 0.00 0 0.00 4.31 0.00 0 0 0 3.31 0 2

Hydrogen Sulfide 0.68 34.1 0.23 6550 1518.81 6.08 1.41 0 0 0.53 0.122896 4.68 1.0851984 1.88

Water 18 0 0 0 1 0 0

Total 100.00 12.28708 273289.38 206.05 29.26 29.50 159.14

Total per pound of fuel 100.00 1.00 22242.01763 16.7696231 2.381641 2.40094 12.9521469
EXCESS AIR AND RELATIVE HUMIDITY WORKSHEET

Atomizing steam:
0 Pounds per pound of fuel (assumed or measured)

CORRECTION FOR RELATIVE HUMIDITY

𝑃𝑣𝑎𝑝 𝑅𝑒𝑙𝑎𝑡𝑖𝑣𝑒 ℎ𝑖𝑚𝑖𝑑𝑖𝑡𝑦 18

Moisture in air = × ×
14.696 100 28.85

= 0.076328 50% 18
14.696 100 28.85

= -0.000016 Pounds of moisture per pound of wet air

where:
Pvapor = vapor pressure of water at ambient temperature in psia (from steam tables)

Pounds of wet air per pound of fuel required 𝐴𝑖𝑟 𝑟𝑒𝑞𝑢𝑖𝑟𝑒𝑑

=
1−𝑀𝑜𝑖𝑠𝑡𝑢𝑟𝑒 𝑖𝑛 𝑎𝑖𝑟

16.76962
1 -0.000016

16.770

Pounds of moisture per pound of fuel = Pounds of wet air per pound of fuel - Air required
= 16.770 -16.7696
= 0.00027

Pounds of H2O per pound of fuel = H2O formed + Pounds of moisture per pound of fuel +
atomizing steam

= 2.400944 +0.00027 +0
= 2.401

CORRECTION FOR EXCESS AIR

Pounds of dry excess air per pound of fuel

𝑁2 𝑓𝑜𝑟𝑚𝑒𝑑 𝐶𝑂2 𝑓𝑜𝑟𝑚𝑒𝑑 𝐻2𝑂 𝑓𝑜𝑟𝑚𝑒𝑑
(28.85 ×𝑃𝑒𝑟𝑐𝑒𝑛𝑡 𝑂2)( + + )
28 44 18
= 𝑃𝑜𝑢𝑛𝑑𝑠 𝑜𝑓 𝐻2𝑂
20.95−𝑃𝑒𝑟𝑐𝑒𝑛𝑡 𝑂2 ((1.6028× )+1)
𝑃𝑜𝑢𝑛𝑑𝑠 𝑜𝑓 𝑎𝑖𝑟 𝑟𝑒𝑞𝑢𝑖𝑟𝑒𝑑 Percent
O2 Wet%
3.5
= 100.975 (0.462577 +0.054128 +0.133401)
20.95 -3.50009
= 3.761878

𝑃𝑜𝑢𝑛𝑑𝑠 𝑜𝑓 𝑒𝑥𝑐𝑒𝑠𝑠 𝑎𝑖𝑟 𝑝𝑒𝑟 𝑝𝑜𝑢𝑛𝑑 𝑜𝑓 𝑓𝑢𝑒𝑙

Percent excess air = × 100
𝐴𝑖𝑟 𝑟𝑒𝑞𝑢𝑖𝑟𝑒𝑑

= 3.761878
16.76962

= 22.43%
EXCESS AIR AND RELATIVE HUMIDITY WORKSHEET

total pounds of H2O per pound of fuel (corrected for excess air)

𝑝𝑒𝑟𝑐𝑒𝑛𝑡 𝑒𝑥𝑐𝑒𝑠𝑠 𝑎𝑖𝑟

= × 𝑝𝑜𝑢𝑛𝑑𝑠 𝑜𝑓 𝑚𝑜𝑖𝑠𝑡𝑢𝑟𝑒 𝑝𝑒𝑟 𝑝𝑜𝑢𝑛𝑑 𝑜𝑓 𝑓𝑢𝑒𝑙
100

= 22.43% 0.000272 2.40

= 2.401

NOTE: All values used in the calculations above shall be on per-pound-of-fuel basis. Numbers in parentheses indicate values to be
taken from the "total per pound fuel" line of the combustion work sheet, and letters in parentheses indicate values to be taken
from the corresponding line of this work sheet.
FLUE GAS WORK SHEET
150 C
302 F

Exit flue gas temperature, Tc:

Column 1 Column 2 Column 3 Column 4 Column 5 Column 6 Column 7

Enthalpy at
Component pounds of
Tc (Btu per Heat Content (BTU per pound of
Wet Mol. % Wet Wt% component T/year
pound fuel)
formed/hr
pounds of component formed per formed)
pound of fuel
Carbon
dioxide 2.38 6.93% 11.1% 17482.30 62936.29 52.81542 125.7874
water vapor 2.40 17.07% 11.2% 17626.44 63455.17 101.6308 244.0436
Nitrogen 15.92 72.77% 74.0% 116889.47 420802.09 57.49809 915.6013
Oxygen 0.79 3.16% 3.7% 53.63188 42.36888
Air 3.76 27613.86 99409.89
SO2 0.0355 0.07% 0.2% 39.15918 1.389333
total 21.53 100.00% 100.00% 179612.07 - 1327.801
Design Basis
Design datum temperature, Td = 60 oF
Ambient air temperature, Ta,a = 28 oF
Air temperature, Ta = 28 oF
Relative Humidity = 50% %

Stack gas Oxygen content vol % = 3.5

Radiation massic heat loss, lower massic heat value of fuel, hr =
2.5 %

Fuel:

Lower massic heat value of the fuel burned, hL = 22,242 Btu/lb

Fuel temperature, Tf = 100 o
F
Atomizing-medium temperature, Tm = 0 o
F

Heat loss calculations

Radiation massic heat loss, lower massic heat value, hr = 556 Btu/lb

Exit flue-gas temperature, Te = 150 oC

Stack massic heat loss, see next sheet 'Combustion', hs = 1,328

Av specific heat capacity of the air, over dataum cpa = 0.24 Btu/lb.oF
Air sensible massic heat correction, Δha = -158 Btu/lb
Av specific heat capacity of the fuel, over datum cpf = 0.48 Btu/lb.oF
Fuel sensible massic heat correction, Δhf = 19.2 Btu/lb
Steam temperature 0.0
Atomizing medium sensible massic heat correction, Δhm = 0.00 Btu/lb

Efficiency calculations
Net thermal efficiency, expressed as a %, e
% = 91.5%
Gross thermal efficiency, expressed as a %, eg

Fuel efficiency, expressed as a %, ef % = 90.9%

=[(hL+Δha+Δhf+Δhm)-
(hf+hs)]/hL
Fuel efficiency, industry practice, expressed as a %, e % = 91.5%
=(hL-hf-hs)/hL
Heater Maximum skin temperature

Link: https://processpocket.streamlit.app

This tool was developed for process engineers to quickly estimate Heaters’ maximum skin temperature
using the method described in API 530 and ISO 13704 on site. As a part of a larger project to develop
what is similar to Carl Branan’s book “process engineers Pocket Handbook” these tools would allow a
process engineer to quickly calculate/estimate equipment efficiencies or sizing using standardized
calculations.
The aim here is to take little-known data from the field (flow, pressures, temperatures,
compositions..etc.) and use it as input for a rough estimation without having to return to the office to
use commercial software or calculations Excel sheets to validate or to calculate. Additionally, these tools
may also serve as a gathered data validation tool.

Nomenclature

T Temperature (oC) P Pressure (kg/cm2.g)

ṁ Mass Flow rate (kg/hr) µ Viscosity (Pa.s)
ρ Density (kg/m3) λ Thermal conductivity (W/m.K)
Di Tube inside diameter (mm) Cp Heat Capacity (Kg/Kcal.hr. oC)
Do Tube outside diameter (mm) A Total Exposed surface Area (m2)
Q Heat Duty (W and Kcal/hr) h Heat transfer coefficient
W/m2.K
Np Number of tube passes 𝑞𝑚𝐴 flow mass velocity m/m2.s
FL FT
Re Reynold’s number Fcir
Q Volume flow rate (m3/hr) or (Nm3/hr) Pr Prandtl’s number
𝛿 Thickness (mm)
Input Required
Table 1:
Max. Tube wall Calculations
Oil Specific gravity -
Flow rate (Light ends) Nm3/hr
Flow rate (Oil) m3/hr
o
Inlet/outlet temperatures C
Average pressure Kg/cm2.a
Heater Geometry Data
Tube Spacing mm
Outside Diameter mm
Tube Thickness mm
Number of passes -
Exposed Surface Area m2
Coke decomposition mm
Coil Geometry -
Light Ends option:
Light ends: Composition Vol% or mol%
Output Obtained

Summary Table
o
Max. Skin temperature C
o
Mean bulk temperature C
Total mass flow rate Kg/hr
Heat Duty Kcal/hr
Heat Transfer coefficient W/m.k
Max heat flux W/m2
Vapor fraction -
Average Heat Capacities Kcal/kg. oC
Average Heat Thermal Conductivities W/m. °C
Average Viscosities cP
Reynold’s number -
Prandtl Number -

Heater Maximum skin temperature

1. Calculating max. heat flux density

Radiant heat flux

̂ = ṁ𝐶𝑝 ∆𝑡 , 2nd term if no phase change present in heater inlet/outlet
𝑄𝑅,𝑎𝑣𝑔 = ṁ. ∆𝐻

The maximum heat flux density at any point in a coil can be estimated as follows:

𝑄𝑚𝑎𝑥 = 𝐹𝑐𝑖𝑟 𝐹𝐿 𝐹𝑇 𝑄𝑅,𝑎𝑣𝑔 + 𝑄𝑐𝑜𝑛𝑣

Where

Qmax is the maximum radiant heat flux density in W/m2

Fcir is the factor accounting for circumferential heat-flux-density variations

FL is the factor accounting for the effect of tube metal temperature on the radiant heat flux density

FT is the factor accounting for the effect of tube metal temperature on the radiant heat flux density

QR, avg is the average heat flux density in W/m2

Qconv is the average convective heat flux density in W/m2

The circumferential variation factor, Fcir, is given as a function of tube spacing and coil geometry in
Figure 2. The factor given by this figure is the ratio of the maximum local heat flux density at the fully
exposed face of a tube to the average heat flux density around the tube. This figure was developed from
considerations of radiant heat transfer only.

As mentioned above, influences such as conduction around the tube and flue gas convection act to
reduce this factor. Since these influences are not included in this calculation, the calculated value will be
somewhat higher than the actual maximum heat flux density.
The longitudinal variation factor, FL, is not easy to quantify. Values between 1 .0 and 1.5 are most often
used. In a firebox that has a very uniform distribution of heat flux density, a value of 1.0 can be
appropriate. Values greater than 1.5 can be appropriate in a firebox that has an extremely uneven
distribution of heat flux density (for example, a long or a tall, narrow firebox with burners in one end
only).

FL = c10 x (1 + H / c11)

Where H = Inside firebox height, feet (mm) c10 = 1.00 and 1.10 for fuel gas and fuel oil/combination
firing, respectively c11 = 100 and 30500 for H in feet and mm, respectively

The tube metal temperature factor, FT, will be less than 1.0 near the coil outlet or in areas of maximum
tube metal temperature. It will be greater than 1.0 in areas of lower tube metal temperatures. For most
applications, the factor can be approximated as follows:
4 4
𝑇𝑔,𝑎𝑣𝑔 − 𝑇𝑡𝑚
𝐹𝑇 = ( 4 4 )
𝑇𝑔,𝑎𝑣𝑔 − 𝑇𝑡𝑚,𝑎𝑣𝑔
∗
𝑇𝑔,𝑎𝑣𝑔 : is the average flue-gas temperature, expressed in kelvins (degrees Rankine), in the radiant
section
∗
𝑇𝑡𝑚 : is the tube metal temperature, expressed in kelvins (degrees Rankine), at the point under
consideration
∗
𝑇𝑡𝑚,𝑎𝑣𝑔 : is the average tube metal temperature, expressed in kelvins (degrees Rankine), in the radiant
section.

The convective heat flux density in most parts of a radiant section is usually small compared with the
radiant heat flux density. In the shock section, however, the convective heat flux density can be
significant; it should therefore be added to the radiant heat flux density when the maximum heat flux
density in the shock section is estimated.

Calculating maximum tube temperature

In addition to the heat-transfer coefficient and the maximum heat flux density, the temperature profile
of the fluid in the coil is necessary for calculating the maximum tube metal temperature in the radiant
section of the heater. This profile, which is often calculated by the heater supplier, defines the variation
of the bulk fluid temperature through the heater coil. For operation at or near design, the design profile
can be used. For operation significantly different from design, a bulk temperature profile shall be
developed. Once the bulk fluid temperature is known at any point in the coil, the maximum tube metal
temperature can be calculated as follows:

𝑇𝑚𝑎𝑥 = 𝑇𝑏𝑓 + ∆𝑇𝑓𝑓 + ∆𝑇𝑐𝑜𝑘𝑒 + ∆𝑇𝑡𝑤

𝑄𝑚𝑎𝑥 𝐷𝑜
∆𝑇𝑓𝑓 = ( )
ℎ𝑓 𝐷𝑖 − 2𝛿𝑐𝑜𝑘𝑒
𝑄𝑚𝑎𝑥 𝛿𝑐𝑜𝑘𝑒 𝐷𝑜
∆𝑇𝑐𝑜𝑘𝑒 = ( )
𝜆𝑐𝑜𝑘𝑒 𝐷𝑖 − 𝛿𝑐𝑜𝑘𝑒
𝑄𝑚𝑎𝑥 𝛿𝑡,𝑎𝑣𝑔 𝐷𝑜
∆𝑇𝑡𝑤 = ( )
𝜆𝑡𝑚 𝐷𝑜 − 𝛿𝑡,𝑎𝑣𝑔

𝑇𝑚𝑎𝑥 is the maximum tube metal temperature, expressed in degrees Celsius

Tbf is the bulk fluid temperature, expressed in degrees Celsius

∆Tff is the temperature difference across the fluid film, expressed in degrees Celsius

∆Tcoke is the temperature difference across coke or scale, expressed in degrees Celsius

∆Ttw is the temperature difference across the tube wall, expressed in degrees Celsius

QR,max is the maximum radiant heat flux density, in W/(m2-K) for the outside surface

hf is the fluid-film heat-transfer coefficient, in W/(m2-K)

Do is the outside diameter, expressed in meters, of the tube;

Di is the inside diameter, expressed in meters , of the tube;

𝛿𝑐𝑜𝑘𝑒 is the coke and/or scale thickness, expressed in metres

𝜆𝑐𝑜𝑘𝑒 is the thermal conductivity of coke or scale, in W/(m2-K)

𝛿𝑡,𝑎𝑣𝑔 is the average tube thickness, expressed in metres

𝜆𝑡𝑚 is the thermal conductivity, in W/(m-K) of the tube metal. (from figure 1)

Heat transfer Coefficient iteration

A value necessary for calculating the maximum tube metal temperature is the fluid heat-transfer
coefficient at the inside wall of the tube. Although the following correlations are extensively used and
accepted in heater design, they have inherent inaccuracies associated with all simplified correlations
that are used to describe complex relationships.

For single-phase fluids, the heat-transfer coefficient is calculated by one of the two equations below, in
which Re is the Reynolds number and Pr is the Prandtl number. No correlation is included for the heat-
transfer coefficient in laminar flow, since this flow regime is rare in process heaters. There is inadequate
information for reliably determining the inside coefficient in laminar flow for oil in tube sizes that are
normally used in process heaters.

for liquid flow with Re > 10000:

0.14
𝜆𝑓,𝑇𝑏 𝜇𝑓,𝑇𝑏
ℎ𝑙 = 0.023 ∗ ( ) ∗ 𝑅𝑒0.8 ∗ 𝑃𝑟0.33 ∗ ( )
𝐷𝑖 𝜇𝑓,𝑇𝑤

for vapour flow with Re >15000:

𝜆𝑓,𝑇𝑏 𝑇𝑏 0.5
ℎ𝑣 = 0.021 ∗ ( ) ∗ 𝑅𝑒0.8 ∗ 𝑃𝑟0.4 ∗ ( )
𝐷𝑖 𝑇𝑤
𝐷𝑖 ∗ 𝑞𝑚𝐴
𝑅𝑒 =
𝜇𝑓,𝑇𝑏
𝐶𝑝 𝜇𝑓,𝑇𝑏
𝑃𝑟 =
𝜆𝑓,𝑇𝑏

hL is the heat-transfer coefficient, in W/(m2-K) for the liquid phase

hv is the heat-transfer coefficient, in W/(m2-K) for the vapor phase

𝜆𝑓,𝑇𝑏 is the thermal conductivity, in W/(m-K) of fluid at bulk temperature

Di is the inside diameter, expressed in meters of the tube

𝜇𝑓,𝑇𝑏 is the absolute viscosity, in Pa-s of fluid at bulk temperature

𝜇𝑓,𝑇𝑤 is the absolute viscosity, in Pa-s of fluid at wall temperature;

Tb is the absolute bulk temperature, expressed in kelvins of vapor

Tw is the absolute wall temperature, expressed in kelvins (degrees Rankine), of vapor

𝑞𝑚𝐴 is the mass flow flux, in kg/(m2-s) of the fluid = ṁ/(A*Np)

Cp is the specific heat capacity, in J/(kg-K) of the fluid at bulk temperature.

For two-phase flows, the heat-transfer coefficient can be approximated using the following equation:

hf = hv * wv + hL * wL

hf is the heat-transfer coefficient, in W/(m2-K), for two phases

wL is the mass fraction of the liquid

wv is the mass fraction of the vapor.

The liquid and vapor heat-transfer coefficients, hL and hv, should be calculated using the mixed-phase
area mass flow rate but using the liquid and vapor material properties, respectively.

Iteration part

Assume for liquid

0.14
𝜇𝑓,𝑇𝑏
( ) = 1.1
𝜇𝑓,𝑇𝑤

Assume for vapor

𝑇𝑏 0.5
( ) = 0.91
𝑇𝑤
1. Solve for hL and hv
2. Calculate ∆𝑇𝑓𝑓 , ∆𝑇𝑐𝑜𝑘𝑒 and ∆𝑇𝑡𝑤
3. Calculate mean tube wall temperature & oil film temperature & maximum tube temperature
Tfilm =Tb + ∆𝑇𝑓𝑓 (estimate μfilm at Tfilm)
∆𝑇𝑡𝑤
Tm = Tb + ∆𝑇𝑓𝑓 +
2
Tmax = Tb + ∆𝑇𝑓𝑓 + ∆𝑇𝑡𝑤 + ∆𝑇𝑐𝑜𝑘𝑒
0.14
𝜇𝑓,𝑇𝑏 𝑇 0.5
4. Calculate ( ) , ( 𝑏) and calculate your error percentage in assumptions
𝜇𝑓,𝑇𝑤 𝑇𝑤
0.14 0.5
𝜇 𝑇
5. Re-assume (𝜇 𝑓,𝑇𝑏 ) , (𝑇𝑏 ) accordingly until your assumption matches your calculated
𝑓,𝑇𝑤 𝑤
values!
6. Check mean tube temperature calculated against temperature assumed for tube thermal
conductivity
Calculating total mass flow rate of feed and light gases

Light gases
𝑚

𝑀𝑤𝑡 = ∑ 𝑀𝑤𝑡𝑖 ∗ 𝑦𝑖
1

ṁ𝑙𝑖𝑔ℎ𝑡 𝑔𝑎𝑠𝑒𝑠 = 0.044 ∗ 𝑀𝑤𝑡 ∗ 𝑄𝑙𝑖𝑔ℎ𝑡 𝑔𝑎𝑠𝑒𝑠

𝑛𝑙𝑖𝑔ℎ𝑡 𝑔𝑎𝑠𝑒𝑠 = ṁ𝑙𝑖𝑔ℎ𝑡 𝑔𝑎𝑠𝑒𝑠 /𝑀𝑤𝑡

𝑙𝑖𝑔ℎ𝑡 𝑒𝑛𝑑𝑠 𝑚𝑜𝑙𝑠 = 𝑛 ∗ 𝑚𝑜𝑙𝑠 𝑓𝑟𝑎𝑐𝑡𝑖𝑜𝑛𝑠

Oil

ṁ𝑜𝑖𝑙 = 𝜌𝑜𝑖𝑙 ∗ 𝑄𝑜𝑖𝑙

𝑚

𝑀𝑤𝑡𝑜𝑖𝑙 = ∑ 𝑀𝑤𝑡𝑖 ∗ 𝑦𝑖
1

𝑛𝑂𝑖𝑙 = ṁ𝑜𝑖𝑙 /𝑀𝑤𝑡𝑜𝑖𝑙

𝑛𝑡𝑜𝑡𝑎𝑙 = 𝑛𝑂𝑖𝑙 + 𝑛𝑙𝑖𝑔ℎ𝑡 𝑔𝑎𝑠𝑒𝑠

𝑐𝑜𝑚𝑝𝑜𝑛𝑒𝑛𝑡 𝑚𝑜𝑙𝑠
𝑡𝑜𝑡𝑎𝑙 𝑓𝑒𝑒𝑑 𝑚𝑜𝑙 𝑓𝑟𝑎𝑐𝑡𝑖𝑜𝑛𝑠 =
𝑛𝑡𝑜𝑡𝑎𝑙
NOTE: from composition calculated physical properties could be calculated for both light ends and oil
fractions. Oil characterization is beyond the sope of this document thought (sorry)
Figure 1: GPSA metal thermal conductivity
Figure 2: Fcir as function of tube spacing/D

1. Curve 1 = double row against wall, triangular spacing

2. Curve 2 = double row with equal radiation from both sides and two diameters between rows,
equilateral spacing
3. Curve 3 = single row against wall
4. Curve 4 = single row with equal radiation from both sides
These curves are valid when used with a tube-centre-to-refractory-wall spacing of 1,5 times the nominal
tube diameter. Any appreciable variation from this spacing should be given special consideration.
ISO 13704/API 530 -2008 Example

In the heater under consideration, the medium-carbon-steel tubes are in a single row against the wall. Other
aspects of the heater configuration are as follows:

Tube spacing is 203,2 mm ( 0,667 ft 8,0 in).

Do = 114,3 mm ( 0,375 ft 4,5 in);
𝛿 t,ave = 6.4 mm ( 0.020 8 ft 0.25 in);
Di = 101.6 mm ( 0.333 ft 4.0 in);
𝛿 f = 0 mm (0 in);
𝜆 tm = 42.2 W/(m K) [24.4 Btu/(h ft °F)] at an assumed tube metal temperature of 380 °C (720 °F).

The flow in the tubes is two-phase with 10 % mass vapour. Other operating conditions are as follows:
Flow rate (total liquid plus vapour) is 6.3 kg/s
Tb = 271 °C
qR.ave = 31 546 W/m2

The properties of the liquid at the bulk temperature are as follows:

µf.T b = 2 * 10-3 Pa s
𝜆 f,T b = 0.1163 W/(m K)
Cp,f 2 847 J/(kg K)

The properties of the vapour at the bulk temperature are as follows:

µv,T b 7.0 10-6 Pa.s
𝜆 v,T b 0.034 6 W/(m K)
Cp,v 2 394 J/(kg K)

From the inside diameter, the flow area is equal to 8.107 10 3 m2 (0.087 3 ft2). Using the total flow rate:
qmA = 6.3/(8.107 x 10-3),
qmA 777.1 kg/(m2 s).

The Reynolds number [Equation (B.2)] is calculated as follows:

For liquid:
(0,1016) (777.1)
𝑅𝑒 = = 3.95 𝑥 104
0.002
For vapor
(0.1016) (777.1)
𝑅𝑒 = = 1.13 𝑥 107
7 𝑥 10−6
The Prandtl number [Equation (B.3)] is calculated as follows:
For liquid:
(2 847) (0.002)
𝑃𝑟 = = 49
0.1163
For vapor:
(2 395) (7 𝑥 10−6 )
𝑃𝑟 = = 0.485
0.0346

Assume that for the liquid:

0.14
𝜇𝑓,𝑇𝑏
( ) = 1.1
𝜇𝑓,𝑇𝑤
Assume that for the vapour:
𝑇𝑏 0.5
( ) = 0.91
𝑇𝑤
These assumptions will be checked later. Calculating hl and hv:
0.1163
ℎ𝑙 = 0.023 ∗ (0.1016) ∗ (3.94 ∗ 104 )0.8 ∗ (49)0.33 ∗ 1.1 = 497 W/m2.K

0.0346
ℎ𝑣 = 0.021 ∗ (0.1016) ∗ (1.12 ∗ 107 )0.8 ∗ (0.486)0.33 ∗ 0.91 = 2126 W/m2.K

The two-phase heat-transfer coefficient can then be calculated:

hf = 0.9*497 +0.1*2126 = 659.9 W/m2.K

The ratio of tube spacing to tube diameter is as follows:

203.2 / 114.3 = 1.78

From Figure 2, Fcir 1.91.

Assume that for this heater, FL 1.1
FT 1.0, and qconv 0 (that is, there is no convective heat-flux density at this point).

Calculating max. heat flux:

qR,max = (1.91)(1.1)(1.0)(31 546) = 66278 W/m2

The temperature difference through each part of the system can now be calculated for the
fluid film:
𝑄𝑚𝑎𝑥 𝐷𝑜 66278 114.3
∆𝑇𝑓𝑓 = ( )= ( ) = 113 𝐾
ℎ𝑓 𝐷𝑖 − 2𝛿𝑐𝑜𝑘𝑒 659.9 101.6

for the tube wall:

𝑄𝑚𝑎𝑥 𝛿𝑡,𝑎𝑣𝑔 𝐷𝑜 66278 ∗ 6.4 114.3
∆𝑇𝑡𝑤 = ( )= ( ) = 11 𝐾
𝜆𝑡𝑚 𝐷𝑜 − 𝛿𝑡,𝑎𝑣𝑔 42.2 114.3 − 6.4
the maximum tube metal temperature is as follows:
𝑇𝑚𝑎𝑥 = 𝑇𝑏𝑓 + ∆𝑇𝑓𝑓 + ∆𝑇𝑐𝑜𝑘𝑒 + ∆𝑇𝑡𝑤 = 271 + 113 + 11 = 395 o C

Checking the assumed viscosity ratio, at the oil-film temperature calculated above, 271+ 113 = 384 °C, the viscosity is 1.1
mPa.s. So, for the liquid:
0.14
𝜇𝑓,𝑇𝑏 0.002 0.14
( ) =( ) = 1.820.14 = 1.09
𝜇𝑓,𝑇𝑤 0.0011
For the vapour:
𝑇𝑏 0.5 270 + 273 0.5
( ) = ( ) = 0.91
𝑇𝑤 384 + 273
Both values are close to the values assumed for the calculation of Kl and Kv. so no additional work is needed.

The mean tube wall temperature is as follows:

270 + 113 + 11/2 = 388 oC

This is close to the temperature assumed for the tube conductivity, so no additional work is required.
Fabricated Example to explore differences between a commercial software and tool

parameters Units Input for both

Feed S.G - 0.733
Feed flow m3/hr 23.487
hydrogen flow Nm3/hr 30980
temperature in oC 250
temperature out oC 290
Pressure Kg/cm2.g 4
Tube spacing mm 450
Outside Diameter mm 203.2
Thickness mm 6.02
Number of passes - 1
Exposed Surface Area m2 8.9
Coke decomposition mm 0
Fc 1.686 -
Fl 0.955 -
Ft 1.0 -

ASTM D86 Assay Temperature oC

0 201.920182
10 227.714783
30 229.657141
50 232.126014
70 234.027408
90 238.276357
100 243.0963
Hydrogen comp% 100

Commercial
Parameter Tool Units
software
o
Max Skin Temperature 408.557 380 C
o
Mean Bulk temperature 399.507 363 C
Total Mass flow rate 19964.63 20,000 kg/hr
Heat Duty 819933 824150 Kcal/hr
Heat transfer coefficient 1339.157 1404.1 W/m oC
Max Heat Flux 126555.7 122200 W/m oC
o
Inlet Temperature 250 250 C
o
Outlet Temperature 290 290 C
Inlet Pressure 4 4 Kg/cm2.g
Vapor Fraction 1 1 -
Reynold's Number V/L [3124642.0, 'NA'] -
Prandtl Number V/L [0.273, 'NA'] 0.46 -
Viscosity V/L [0.012, 'NA'] 0.0221 cP
Cp 1.04 1.0479 Kcal/kg. oC
Thermal Conductivity V/L [0.188, 'NA'] 0.2239 W/m.K
Metal Thermal Conductivity 43.375 W/m.C
Fc 1.686 1.678 -
Fl 0.955 0.9551 -
Ft 1 1 -

References

1. API 530, 6th edition 2008

2. ISO 13704: 2001
3. GPSA Engineering Databook Section 8: Fired Heaters
Safety Relief Valves Sizing
Link: https://processpocket.streamlit.app/Safety%20Valves%20Sizing

This tool was developed for process engineers to quickly estimate the outcome of the safety relief valve sizing
equations on site. As a part of a larger project to develop what is similar to Carl Branan’s book “process
engineers Pocket Handbook” these tools would allow a process engineer to quickly calculate/estimate
equipment efficiencies or sizing using standardized calculations

The aim here is to take little-known data from the field (flow, pressures, temperatures, compositions..etc.) and
use it as input for a rough estimation without having to return to the office to use commercial software or
calculations Excel sheets to validate or to calculate. Additionally, these tools may also serve as a gathered data
validation tool.

Nomenclature

Tn Temperature (oK) P Pressure (kPa)

T Temperature at relieving conditions (oK) P1/s/c/back Relief/Set/critical/back pressures (kPa)
a Temperature of insulation outer surface (1177 oK) t Thickness of insulation (m)
𝜆 Latent heat of vaporization (kcal/kg) M Molecular weight M.wt
H Enthalpy (kcal/kg) µ Viscosity (cP)
Re Reynold’s number k Gas specific heat ratio
ρ Density (kg/m3) αv cubical expansion coefficient (1/°K)
D Diameter (m) Cp Heat Capacity (Kg/Kcal.hr. oC)
L Length (m) Aw Wetted area (m2)
W W Required relieving capacity (kg/hr) A/Acalc Minimum required effective discharge area (mm2)
Q Required relieving capacity (Nm3 /min or Ae Effective area (mm2)
L/min for liquids)
Q Relief Load (kW or kcal/hr) G Specific gravity
h Depth of wetted portion (m) Z Compressibility factor
L1 NLL in vessel/column (m) F Environmental factor
C Coefficient determined from an expression of F2 Coefficient for subcritical gas flow sizing
the ratio of specific heats of the gas or vapor equation
U Heat transfer coefficient (kcal/(m2.h°C)) L2 Elevation above grade (m)
Kd The effective coefficient of discharge v Velocity (m/s)
KSH superheated correction factor (steam) Kb Capacity correction factor due to backpressure
Kv Viscosity correction factor (liquid) KN Correction factor for Napier equation (steam)
Kc Combination correction factor Kw Backpressure correction factor (liquid)
Input Required
Table 1: Note 1: Input required depending on the case

1. Calculating Relieving/Set Pressure 2. Relief Loads Estimation (Cont.)

Case (non-fire/fire) - Closed Outlet (Heat Exchangers)
o
MAWP kg/cm².g Hot side inlet temperature C
Design Pressure (if Set pressure kg/cm².g Normal Cold Side rate Kg/hr
doesn’t equal MAWP)
Disc Installation Type (If multiple One/additional/ Normal Cold Side composition %vol
device installation is chosen) Supplemental
o
2. Relief Loads Estimation Normal Cold Side conditions (T&P) C/ kg/cm².a
External Fire (Vaporization) Inadvertent Control Valve opening
normal stream composition %vol Control valve size inch
o
normal stream conditions (T&P) C/ kg/cm².a Control Valve opening %
Aw calculations m2 Control valve inlet and outlet %vol
(If the stream is liquid, it appears) stream composition
o
Type of Equipment Control valve inlet and outlet C/ kg/cm².a
normal stream conditions (T&P)
Geometry inputs m Thermal Expansion
Heat Absorbed kW Relieved stream composition %vol
o
External Fire (Gas expansion) Relieved stream conditions (T&P) C/ kg/cm².a
Same input + Max. Heat load Kcal/hr
o
vessel maximum wall temperature C 3. Gas or Vapor/ Steam
Closed Outlet (Vessel) Mass flow rate Kg/hr
Vessel Feed rate Kg/hr inlet/outlet stream composition %vol
o
feed composition and %vol Inlet (relieved) stream conditions (T&P) C/ kg/cm².a
o o
feed stream conditions (T&P) C/ kg/cm².a Outlet (relief header) stream conditions (T&P) C/ kg/cm².a
Feed vapor mass fraction (if not 0-1 range Kb, kd, kc input (optional) -
calculated)
Closed Outlet (Vessel with reboiler) 3. Liquid
Same + Same +
Kw, kd, kc input (optional) -
Output Obtained
1. Calculating Relieving/Set Pressure 3. Liquid
Max. accumulated pressure kg/cm².g Max. Flow Capacity Kg/hr
Relief device set pressure kg/cm².g Physical properties used in Kg/m3/cP
calculations: Density/ µ
Allowable overpressure kg/cm² Kw,kd,kc -
Overpressure % Minimum area cm2/in2
Relieving pressure kg/cm².a Selected Area cm2/in2
2. Relief Loads Estimation Orifice Designation
Calculated Relief Load kW Permissible inlet/outlet nozzles inches
Calculated Relieving flow rate Kg/hr Table of Calculations -
3. Gas or Vapor/ Steam Specifications sheet -
Flow Critical/Subcritical
Max. Flow Capacity Kg/hr
Physical properties used in -/cP/-/-
calculations: average cp/cv, µ, M.wt, Z
Kb,kd,kc + (kSH, kN) for steam -
Minimum area cm2/in2
Selected Area cm2/in2
Orifice Designation
Permissible inlet/outlet nozzles inches
Table of Calculations -
Specifications sheet -

1. Introduction
Sizing pressure relief valves (PSV) requires a vivid understanding of the process where the safety valves are
installed, the sizing equations, and the numerous relevant standards. In this document, we’ll be exploring the
hand calculations required to preliminary size different safety valves. First, we’ll explore how to estimate the
relieving and set pressures according to standards. Secondly, we’ll explore the basic safety orifices sizing
equations (for gases, steam, or liquids) and learn how to estimate the required coefficients. Finally, we’ll take a
glance at how to estimate the required relief loads for different processes and contingencies as an introduction
to this complex matter. However, the back pressure calculations were not discussed in this document. The
backpressure used in safety calculations is the sum of superimposed backpressure and built-up back pressure.
The superimposed backpressure is the static pressure and the relief header before the valve opening, while the
built-up back pressure is the pressure increase due to the valve opening building up extra pressure to overcome
the outlet friction losses. The estimation of the built-up back pressure could be discussed later in the context of
flare systems calculations.

This document did not discuss the various safety valve types' working mechanisms, the rupture disc types and
sizing procedures, or guidelines for sizing the inlet/outlet piping of PSV. However, you may find the mentioned
topics in references and recommended reads.
2. Set/relief Pressure Calculations
Table 2.1: Relief Pressure estimation table (percentages are applied to MAWP’s gauge pressures) [1]
Contingency Installation Single-Valve Installations Multiple-Valve Installations
Set Pressure Max. Accumulated Set Pressure Max. Accumulated
(%) pressure (%) (%) pressure (%)

Non-fire only First valve 100 110 [1] 100 116 [2]

Additional valve(s) --- --- 105 116 [2]

Fire only First valve 100 121 100 121

Additional valve(s) --- --- 105 121

Supplemental valve --- --- 110 121
Note : All values are percentages of the maximum allowable working pressure.
(1) 10% or 3 psi (0.21 kglcm2), whichever is greater.
(2) 16% or 4 psi (0.28 kg/cm2), whichever is greater.
(3) A supplemental valve installation provides relieving capacity for an additional hazard created by exposure to
fire or other unexpected sources of external heat. Supplemental valves are used only in addition to valves sized
for operating (non-fire) contingencies. [3]
(4) In the case of ASME-application liquid service valves (that is, for protection of a liquid-full vessel), maximum
accumulated pressure is limited to 110% of the maximum allowable working pressure for operating
contingencies. In the case of non-ASME-application liquid service valves (for protection of piping without vessels
included), 25 % overpressure is generally specified. [3]
(5) The minimum pressure differentials between the set pressure of the valve and the operation pressure of the
vessel are recommended as follows:
Table 2.2: minimum differential pressure limitations [3]
Set Pressure Ps Minimum recommended
Pressure differential
Ps ≤ 70 psi (4.9 kg/cm2) 5 psi (0.35 kg/cm2)
Ps ≤ 1000 psi (70 kg/cm2) 10% of Ps
Ps > 1000 psi (70 kg/cm2) 7% of Ps
Figure 2.1: Pressure Level Relationships for Pressure-relief Valves [1]
3. Safety Relief Valves Sizing

Gases Steam Liquids

Subcritical Flow Critical Flow

17.9 ∗ 𝑊 ∗ √𝑇𝑍 𝑊 ∗ √𝑇𝑍

𝐴= 𝐴=
𝐹2 𝐾𝑑 𝐾𝑐 √𝑀𝑃1 (𝑃1 − 𝑃2 ) 𝐶𝐾𝑑 𝑃1 𝐾𝑏 𝐾𝑐 √𝑀
47.95 ∗ 𝑄 ∗ √𝑇𝑀𝑍 2.676 ∗ 𝑄 ∗ √𝑇𝑀𝑍
𝐴= 𝐴=
𝐹2 𝐾𝑑 𝐾𝑐 √𝑃1 (𝑃1 − 𝑃2 ) 𝐶𝐾𝑑 𝑃1 𝐾𝑏 𝐾𝑐
258 ∗ 𝑄 ∗ √𝑇𝐺𝑍 14.41 ∗ 𝑄 ∗ √𝑇𝐺𝑍
𝐴= 𝐴=
𝐹2 𝐾𝑑 𝐾𝑐 √𝑃1 (𝑃1 − 𝑃2 ) 𝐶𝐾𝑑 𝑃1 𝐾𝑏 𝐾𝑐

𝑃1 190.5 ∗ 𝑊 11.78𝑄√𝐺
Acalc 𝑇= ∗𝑇 𝐴= 𝐴=
𝑃𝑛 𝑛 𝑃1 𝐾𝑑 𝐾𝑠ℎ 𝐾𝑁 𝐾𝑏 𝐾𝑐 𝐾𝑑 𝐾𝑤 𝐾𝑣 𝐾𝑐 √𝑃1 − 𝑃2
𝑃𝐶𝑓 2 𝑘⁄(𝑘−1)
=[ ]
𝑃1 𝑘+1
2 𝑘+1/𝑘−1
𝐶 = 0.03948√𝑘 [ ]
𝑘+1
𝑘 1 − 𝑟 (𝑘−1)/𝑘
𝐹2 = √[ ] 𝑟 2/𝑘 [ ]
𝑘−1 1−𝑟

𝑊𝑟𝑒𝑞 ∗ 𝐴𝑒 𝑊𝑟𝑒𝑞 ∗ 𝐴𝑒 𝑊𝑟𝑒𝑞 ∗ 𝐴𝑒

Wmax[3]
𝐴𝑐𝑎𝑙𝑐 𝐴𝑐𝑎𝑙𝑐 𝐴𝑐𝑎𝑙𝑐

A: mm2
A: mm2 A: mm2
W: kg/h
W: kg/h Q: L/min
Units Q: Nm3/min
P1,P2: kPa P1,P2: kPag
P1,P2: kPa
T: Kelvin T: Kelvin
T: Kelvin
W (kg/h ) = 0.044 * M.wt * Q (Nm3/hr)
Q (Nm3/min) = Q (Nm3/hr)/60
Conversion Q (L/min) = Q (m3/hr) *16.67
Notes P (kPa) = P (kg/cm2) * 98.0665
A (cm2) = A (mm2)*0.01
A (in2) = A (cm2) * 0.155
Table 3.1: Table of single-phase relief valve sizing equations [1]
NOTE 1: The Napier coefficient KN must be considered when P1 ≥ 106.5 kg/cm2a.
NOTE 2: for gas equations
Kd = effective coefficient of discharge. For preliminary sizing, use the following values:
= 0.975 when a pressure relief valve is installed with or without a rupture disk in combination.
= 0.62 when a pressure relief valve is not installed and sizing is for rapture disk
Wrated = maximum rated flow rate through the valve
3.1 Gas PSV Sizing[1]
1. Estimate if flow is critical or sub-critical from the following equation

2 𝑘/𝑘+1
𝑃𝑐 = 𝑃1 ∗ [ ] , where
𝑘+1

Pc is the critical pressure, P1 is the upstream pressure (relieving), both in absolute units
k is the specific heats ratio for any ideal gas

2. Based on the flow condition, you can proceed with one of the following equations:

Sub-critical Flow Critical Flow

Pc < Pback Pc > Pback
17.9 ∗ 𝑊 ∗ √𝑇𝑍 𝑊 ∗ √𝑇𝑍
𝐴= 𝐴=
𝐹2 𝐾𝑑 𝐾𝑐 √𝑀𝑃1 (𝑃1 − 𝑃2 ) 𝐶𝐾𝑑 𝑃1 𝐾𝑏 𝐾𝑐 √𝑀
47.95 ∗ 𝑄 ∗ √𝑇𝑀𝑍 2.676 ∗ 𝑄 ∗ √𝑇𝑀𝑍
𝐴= 𝐴=
𝐹2 𝐾𝑑 𝐾𝑐 √𝑃1 (𝑃1 − 𝑃2 ) 𝐶𝐾𝑑 𝑃1 𝐾𝑏 𝐾𝑐
258 ∗ 𝑄 ∗ √𝑇𝐺𝑍 14.41 ∗ 𝑄 ∗ √𝑇𝐺𝑍
𝐴= 𝐴=
𝐹2 𝐾𝑑 𝐾𝑐 √𝑃1 (𝑃1 − 𝑃2 ) 𝐶𝐾𝑑 𝑃1 𝐾𝑏 𝐾𝑐
Table 3.2: Gas PSV sizing equations
Where,
2 𝑘+1/𝑘−1
𝐶 = 0.03948√𝑘 [ ] , refer to Figure 3.7
𝑘+1
𝑘 1−𝑟 (𝑘−1)/𝑘
𝐹2 = 𝑐𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝑜𝑓 𝑠𝑢𝑏𝑐𝑟𝑖𝑡𝑖𝑐𝑎𝑙 𝑓𝑙𝑜𝑤 = √[ ] 𝑟 2/𝑘 [ ], refer to Figure 3.6
𝑘−1 1−𝑟
r: ratio of backpressure to upstream relieving pressure P2/P1
Balanced Pressure Relief Valves
Balanced pressure relief valves should be sized using critical-flow equations. The back pressure correction factor
in this application accounts for subcritical flow velocities and the tendency for the disc to drop below full lift (the
use of subcritical flow equations is appropriate only where full lift is maintained). For this application, the back
pressure correction factor, Kb, should be obtained from the manufacturer.
Figure 3.1: Kb for balanced bellows relief valves (Gas) [1]
[1]
3.2 Steam PSV Sizing
Pressure relief devices in steam service that operate at critical
low conditions may be sized using:
190.5∗𝑊
𝐴= , where
𝑃1 𝐾𝑑 𝐾𝑠ℎ 𝐾𝑁 𝐾𝑏 𝐾𝑐

Kb: The capacity correction factor is due to backpressure.

Applied for balanced bellows valves only. Refer to Figure 1
For conventional valves, use a value for Kb equals 1.0.
KN: correction factor for Napier equation
= (0.1906 * P1- 1000)/(0.2292 *P1 - 1061)
where P1 in kPa and 10339 kPa < P1 ≤ 22057 kPa
KSH: Correction factor for superheated steam.
For saturated steam use KSH =1.0
3.3 Liquid PSV Sizing [1]
Pressure relief devices for liquid services may be sized using:
11.78 ∗ 𝑄 ∗ √𝐺
𝐴=
𝐾𝑑 𝐾𝑤 𝐾𝑣 𝐾𝑐 𝐾𝑝 √𝑃1 − 𝑃2

Kd: effective coefficient of discharge that should be obtained from the valve manufacturer. For preliminary sizing
estimation, a discharge coefficient of 0.65 can be used.
Kw: correction factor for back pressure. If the backpressure is atmospheric, Kw =1.0.
Balanced-bellow valves in backpressure service will require the correction factor obtained from
Figure 2 Conventional valves require no special correction.
Kv: correction factor for viscosity as determined from the following equation:
2.878 342.75 −1 170 −0.5
𝑘𝑣 = (0.9935 + + ) or 𝑘𝑣 = ( + 1)
𝑅𝑒0.5 𝑅𝑒1.5 𝑅𝑒
When a pressure relief valve is sized for viscous liquid service, it should first be sized as if it were for non-viscous
type application (i.e., Kv = 1.0) so that a preliminary required discharge area, A, can be obtained from the previous
equation. From API standard orifice sizes (Table 3.3 may be used for preliminary estimation), the next larger
orifice size should be used in determining the Reynolds number, Re, the following equation:

𝑄∗(18800∗𝐺)
𝑅𝑒 = ,
𝜇√𝐴

µ: Liquid viscosity is cP Q: flow rate in L/min

G: specific Gravity A: selected area mm2
After Reynold’s number, Re, is determined, the factor Kv is obtained, and Kv is then applied in the liquid sizing
equation to correct the preliminary required discharge area. If the corrected area exceeds the chosen standard
orifice area, the above calculations should be repeated using the next larger standard orifice size.

Table 3.3: Conventional and balanced-bellows valve selection table

Figure 3.2: Kw for kb for relief valves due to back pressure (liquid) [1] (API RP 526 and GPSA Section 3) [3]
Figure 3.3: Kv for relief valves due to viscosity (liquid) [1] Figure 3.4: Kb for conventional/pilot operated relief valves (Gas) [1]

Figure 3.5: KSH for steam at T&P[3]

Figure 3.6: F2 Values for Gas PSV Sizing (subcritical Flow)

Figure 3.7: C Values for Gas PSV Sizing (Critical Flow)

3.4 Calculating Equation Coefficients 2. Steam
1. Gas or Vapors KN Correction factor for Napier equation
Kb Capacity correction factor = 1.0 where P1≤ 106.5 kg/cm2a or 1339 kPa
Kb = a + b(Pb /(Pset+POP)) 3 (abs. pressures) = (0.1906 * P1- 1000)/(0.2292 *P1 - 1061)
(Conventional and pilot operated) [6] where P1 in kPa and
γ Range % a b 10339 kPa (106.5 kg/cm2a) < P1 ≤ 22057 kPa (226.1
1.1 66-90 1.3026 -1.137*10-6 kg/cm2A)
1.3 63-90 1.294 -1.1703*10-6
1.5 56-90 1.203 -1.143*10-6 KSH Superheat steam correction factor
1.7 51-90 1.148 -1.109*10-6 For saturated steam at any pressure, KSH = 1.0
Kb =1/( a + b(PB/Ps)3) (gauge pressures) KSH can be obtained from figure 3.5 or KSH tables published
in ref [1] P.86-90.
(Balanced bellows) [6]
Overpressure Range a b
3. Liquid
10 30-50 0.8707 4.724*10-6
Kw Correction factor due to back pressure
20 30-50 0.976 8.36*10-7
If the back pressure is atmospheric, Kw = 1.0. for
conventional and pilot-operated PSVs, kw =1. Otherwise.
Use Figure 3.2.

1. The curves above represent a compromise of the values

recommended [1]
Figure 3.6: C for by several
Critical relief
flow gas valve
Sizing manufacturers
equation and
used when the make of the valve or the critical flow
Kw = 1.1165-0.01*(PB/Ps) for (PB/Ps) > 17 [6]
pressure point for the vapor or gas is unknown. The curves
Kv viscosity correction factor
are for set pressures of 50 psig and above. They are limited
2.878 342.75 −1
to back pressure below critical flow pressure for a given set 𝑘𝑣 = (0.9935 + 0.5 + )
pressure. For set pressure below 50 psig or subcritical flow, 𝑅𝑒 𝑅𝑒1.5
170 −0.5
the manufacturer must be consulted for values of Kb. [1] 𝑘𝑣 = (1 + )
2. For 21% overpressure, Kb equals 1.0 up to PB/PS = 50%.[1] 𝑅𝑒
3. For pilot-operated PRVs, the valve lift is not affected by Or use Figure 3.3
backpressure. For compressible fluids at critical 1.Use the next larger orifice area to calculate Re
flow conditions, a backpressure correction factor of 1.0 2.Correct calculated A without Kv by dividing it with new Kv
should be used for pilot-operated PRVs.[1] Kd Effective coefficient of discharge for PSV
Alternatively, kb can be estimated for conventional and 0.65, when a PRV is installed with or without a rupture disk
pilot-operated Relief valves using Figure 3.4 (Subcritical in combination;
flow) [1] 0.62, when a PRV is not installed, and sizing is for a rupture
Kd Effective coefficient of discharge for PSV disk
0.975 when a PRV is installed with or without a rupture disk
in combination;
0.62 when a PRV is not installed and sizing is for a rupture
disk
Kc combination correction factor
1.0 when a rupture disk is not installed.
Figure 3.7: F2 for Subcritical gas equation [1]
0.9 when a rupture disk is installed with a pressure relief
valve and the combination does not have a published value.
Notes on PSV sizing
PSV sizing could be an iterative procedure or a two-step process
After estimating the relieving capacity, you may want to check its impact on the relief header's total back
pressure. The increased back pressure due to the calculated relieved capacity could impact your
calculated/initially estimated Kb, resulting in another iteration/step to calculate the corrected relieved capacity.
Refer to section 5.3.4.2 on ref [1]

Inlet piping
inlet piping pressure losses due to friction should not be higher than 3% of set pressure and calculated using the
maximum rated capacity of the pressure relief device [1][3]. Exceptions were mentioned in recommended reads[II]
When two or more pressure relief devices are placed on one connection, the inlet piping internal cross-sectional
area shall be at least equal to the combined inlet areas of the pressure relief devices connected to it
Selection Criteria of PSV
1. Conventional PSVs
2. The sum of the maximum variable superimposed back pressure plus the built-up back pressure is less than
10% of the set pressure.
3. Fouling or corrosive conditions are not expected.
2. Bellows Type PSVs
1. The sum of the variable superimposed back pressure plus built-up back pressure exceeds 10% of set
pressure.
2. Fouling or corrosive conditions are expected and protection cannot be afforded by using alternative
materials or devices.

Backpressure Limitations on Bellows Type PSVs

Total backpressure shall not reduce differential pressure across the PSV to a value limiting PSV relieving capacity
to less than design capacity. This corresponds to a total backpressure of approximately 50% of set pressure for
gas services.

3. Pilot Operated PSVs

1. Pilot-operated PSVs are recommended when maximum set point accuracy is required
2. Pilot-operated PSVs shall generally be limited to clean gas service.
3. Tanks designed to API STD 650, which may also require pressure relief devices, may be protected by pilot-
operated relief valves. Pilot-operated relief valves shall protect tanks designed to API STD 620.

More on the advantages and disadvantages of different PSV types can be found at recommended reads [II] [III]
4. Examining possible cases (Contingencies)
Estimating the required relief loads could be a grueling task that requires extensive heat and mass balance
calculations at relief conditions. It could be best to use simulation tools for this one. However, this section
aims to familiarize readers with the basis and variables needed for different scenarios. Moreover, manual
calculations could be an alternative option where minimum data are available for simulation in some cases.
All causes of overpressure, or contingencies, must be evaluated for each PSV installation in terms of the
pressures generated and the rates at which fluids must be relieved. Causes of overpressure in process
equipment can range from a single event to a complex combination of events. The basis for calculating a
valve size follows calculations of valid contingencies. The contingency that requires the largest effective area
dictates the size of the PSV.
More on “guidelines on estimating relief loads for different contingencies” can be found on recommended
reads [V][I][III]
Table 4.1: Guidelines on estimating relief loads for different contingencies [2]
Item Pressure Relief Device Pressure Relief Device
No. Condition (Liquid Relief) (Vapor Relief)*
1 Closed outlets on vessels Maximum liquid pump- Total incoming steam and vapor plus that
in rate generated therein at relieving conditions
2 Cooling water failure to condenser — Total vapor to condenser at relieving conditions
3 Top-tower reflux failure — Total incoming steam and vapor plus that
generated therein at relieving conditions less
vapor condensed by side-stream reflux
4 Side-stream reflux failure — Difference between vapor entering and leaving
section at relieving conditions
5 Lean oil failure to absorber — None, normally
6 Accumulation of noncondensables — Same effect in towers as found for Item 2; in
other vessels, same effect as found for Item 1
7 Entrance of highly volatile — For towers, usually not predictable
material: Water in hot oil or Light
hydrocarbons in hot oil
8 Overfilling storage or surge vessel Maximum liquid —
Pump-in rate
9 Failure of automatic controls — Must be analyzed on a case-by-case basis
10 Abnormal heat or vapor input — Estimated maximum vapor generation including
non-condensables from overheating
11 Split exchanger tube — Steam or vapor entering from twice the cross-
sectional area of one tube; also same effects
found in Item 7 for exchangers
12 Internal Explosions — Not controlled by conventional relief devices
but by avoidance of circumstances
13 Chemical reaction — Estimated vapor generation from both normal
and uncontrolled conditions
14 Hydraulic expansion Use Liquid expansion —
1. Cold fluid shut in equation (table 4.7)
2. Line outside process area
3. Shut in
15 Exterior fire — Estimate by external fire equations (table 4.2)
16 Power failure (steam, electric, or — Study the installation to determine the effect of
other) power failure; size relief valve for the worst
condition that can occur
1. Fractionators — All pumps could be down, with the result that
reflux and cooling water would fail
2. Reactors — Consider failure of agitation or stirring, quench
or retarding steam; size valves for vapor
generation from a runaway reaction
3. Air-cooled exchangers — Fans would fail; size valves for the difference
between normal and emergency duty
4. Surge vessels Maximum liquid inlet —
rate
* Considerations may be given to the suppression of vapor production as the result of the device’s relieving
pressure being above operating pressure, assuming constant heat input. (Procedures for sizing pressure relief devices
are presented in Section 4 of API-RP-520.)
After the specifications are determined, the next activity is to calculate a preliminary effective discharge area.
Figure 4.1 shows the inputs used to determine the basis for calculating the size of a PSV.

Figure 4.1: Sizing input flowchart [2]

Relief Loads Estimation for individual contingencies

External Fire External Fire

Overpressure due to vaporization [4] Overpressure due to Gas Expansion [4]
𝑄 When the vessels that contain no liquid, such
𝑊 =
∆𝐻 as gas holders, are exposed to the fire
W: relieving load kg/hr contingency, overpressure will occur due to the
△H: Enthalpy difference between normal and relieving gas expansion. The following equation can
conditions Kcal/kg obtain the relieving loads for this case:
Q: total heat absorbed kcal/hr
NOTE: assuming to inlet or outlet flow
𝑄 = 𝑞𝑜 . 𝐹. 𝐴0.82 𝐴 (𝑇𝑤 − 𝑇1 )1.25
Q: total heat absorbed kcal/hr 𝑊 = 0.1406√𝑀𝑃1 ( )
𝑇11.1506
F: Environmental factor W: relieving load lb/hr
A: wetted surface m2 M: molecular weight
qo: 37130, P1: relieving pressure psia
if drainage isn’t provided qo: 61000 A: exposed surface area ft2
Bare Vessel Tω: Vessel max. wall temperature, 593 for CS
F=1 (oR)
Insulated Vessel T1: relieving temperature (oR)
𝑘 ∗ (𝑎 − 𝑇) T ∗P
𝐹= T1 = n 1⁄P
𝐶. 𝑡 n
k: thermal conductivity of insulation ( W/m-K) at average of Pn: normal operating pressure (psia)
904 oC and relieving temperature. Tn: normal operating temperature (oR)
Tf: temperature of vessel contents at relieving condition ( oC)
a: temperature of insulation outer surface (904 oC)
C: constant that equals 66570
t: thickness of insulation (m)
Type of equipment Factor F Factor F
API RP521 NFPA 30
Bare Vessel 1.0 1.0

Water application 1.0 0.3

facilities on bare vessel
Water application 1.0 0.15
facilities on insulated
vessel
Depressurizing and 1.0 -
emptying facilities
Underground storage 0.0 0.3
Earth covered storage 0.03 -
above grade

Table 4.2: Relief loads guidelines for External Fire

Closed Outlet
Drums / flashing drum with reboiler [4] Columns [4]
The vapor rate in the feed gas is treated as the relief When all outlets from the equipment are blocked, the
valve's relieving rate. The relief rate is determined by relieving load should be at least as great as the capacity
estimating the emergency reboiler heat duty. of the sources of pressure. If all outlets are not blocked,
See Vaporizer notes on table 4.5 the capacity of the unblocked outlets may properly be
considered

Reboiler Duty (steam) (1) Valve block of reflux line

1
𝑊= [(ℎ𝐹′ − ℎ𝐿 )𝑊𝐹′ − (ℎ𝐷
′
− ℎ𝐿 )𝑊𝐷′ − (ℎ𝐵′ − ℎ𝐿 )𝑊𝐵′
𝜆
+ ∆𝑄𝑅′ − ∆𝑄𝐶′ ]
λ /hL: latent heat of vaporization/Liquid enthalpy at the
top section of the tower (2 or 3 stages below the top tray)
h’F: feed stream enthalpy (relieving)
W’F: Feed stream flow rate (relieving)
h’D/ W’D: Top distillate enthalpy and flow rate (relieving)
h’B/ W’B: Bottom product enthalpy and flow rate
(relieving)
𝑡2′ − 𝑡1′
∆𝑇′ = △Q’R: Reboiler duty (relieving)
𝑇 ′ − 𝑡1′ △Q’C: Condenser duty (relieving)
ln( ′ )
𝑇 − 𝑡2′
∆𝑇′ 𝑈
∆𝑄′ = ( )( 𝑐𝑎𝑙𝑐 )∆𝑄 (kcal/hr)
∆𝑇 𝑈𝐷
∆𝑄′
𝑊= (kg/hr)
𝜆
Reboiler Duty (Oil) (2) Valve block of fractionation column overhead vapor
product
1
𝑊= [(ℎ𝐹′ − ℎ𝐿 )𝑊𝐹′ − (ℎ𝐵′ − ℎ𝐿 )𝑊𝐵′ + ∆𝑄𝑅′ ]
𝜆
Steam stripping is used:
1
𝑊𝐻𝐶 = [(ℎ𝐹′ − ℎ𝐿 )𝑊𝐹′ − (ℎ𝑠′ − ℎ𝑠′′ )𝑊𝑠′
𝜆
− (ℎ𝐵′ − ℎ𝐿 )𝑊𝐵′ ]
𝑊 = 𝑊𝐻𝐶 + 𝑊𝑠′
WHC: hydrocarbon relieved
W’s: steam relieved
(𝑇1 ′ − 𝑡2′ ) − (𝑇2 ′ − 𝑡1′ ) h’s: steam inlet enthalpy (relieving)
∆𝑇′ =
𝑇 ′ − 𝑡2′ h’’s: steam relieving enthalpy
ln( 1 ′ )
𝑇2 − 𝑡1′ estimate relieved steam based on a full open steam
∆𝑇′ 𝑈
∆𝑄′ = ( )( 𝑐𝑎𝑙𝑐 )∆𝑄 (kcal/hr) control valve and pressure difference between the steam
∆𝑇 𝑈𝐷
𝑇1 −𝑇2′ header and relieving pressure
′
∆𝑄 = ( )∆𝑄 (kcal/hr)
𝑇1 −𝑇2
∆𝑄′
𝑊= (kg/hr)
𝜆

Table 4.3: Relief loads guidelines for Drums and Columns

Discharge line of rotary machinery
Centrifugal pumps[4] Centrifugal compressors[4]

An overpressure problem does not occur on If the design pressure for the compressor discharge system is
the discharge line of a centrifugal pump higher than the pressure of surge point at maximum speed,
because the centrifugal pump discharge overpressure does not occur. If the design pressure is lower
system normally has a design pressure equal than that, overpressure protection should be considered. In
to or higher than the pump shut-off this case, the relieving load should be the flow rate (FD) at the
pressure. head equivalent to the design pressure (PD) at maximum
However, when the design pressure is set speed or should be the anti-surge flow (FS) at maximum
lower than the pump shut-off pressure, an speed, whichever is greater. That value is usually obtained
overpressure problem will occur, and the from such compressor performance curve as shown below:
flow rate at the head, equivalent to the set
pressure of the PSV minus the maximum
suction pressure, should be read from the
pump performance curve for the relieving
load.
The relieving load should be determined
based on the pump’s maximum speed when
the steam turbine driver is provided.

Reciprocating pumps[4] The calculation procedure is as follows.

The relieving load shall be equal to the 1. Assume the discharge pressure is 1.1 times the design
pump’s rated capacity. pressure. Based on the calculated discharge pressure,
estimate the suction pressure of the compressor assuming
the compressor is running on the surge control line. In
case of variable speed compressor, whole range of the
operating speed to be investigated.
2. Calculate weight flow through the compressor and
required power of the compressor based on the suction
and discharge pressures calculated in Step 1.
3. Check the calculated weight flow if it is available from
upstream side of the compressor.
4. Check the required power if it is available from the driver.
Reciprocating compressors[4] Relieving load is the maximum weight flow rate calculated in
The relieving load should be equal to the Step 2 considering the limitation of available flow rate and
compressor’s rated capacity. power checked in Step 3 and 4. In case the calculated suction
pressure is equal to or higher than the design pressure of the
suction side in Step 1, re-calculate the suction pressure
assuming that the compressor is still running on the surge
control line while the 1.1 times of the design
pressure is replaced with the design pressure of discharge
side.

Table 4.5: Relief loads guidelines for Rotary Machinery

Heat Exchangers
Vaporization in Hex [4] Vaporizer [4]

If the vapor pressure of the cold medium at the When the outlet of the cold side is blocked off, and the hot
inlet temperature of hot side is more than 1.3 medium continues to flow, overpressure will
times the design pressure of the cold side, occur, and the relieving load should be calculated based on the
overpressure protection due to vaporization is vaporizer heat duty at the relieving condition obtained by the
needed. If it is assumed that the outlet of the cold procedure shown in Table 4.3
side is blocked off, and the hot medium continues
to flow, the relieving load on the cold side should
be determined as follows :

𝑄 (𝑇1 − 𝑡𝑏𝑝 )
𝑊 = .
𝜆 (𝑇1 − 𝑡𝑎𝑣 ) Steam Vaporizer
𝑡1 + 𝑡2 (1) Inlet and outlet temperatures of process fluid to/from
𝑡𝑎𝑣 =
2 reboiler at relieving pressure will increase, because
W : relieving load (kg/hr) the boiling temperature of process fluid increases, when the
Q : normal heat exchanger duty (kcal/hr) operating pressure reaches the relieving pressure.
λ : latent heat of vaporization at tbp (kcal/kg) (2) When heating steam is supplied under steam flow control,
T1: hot side inlet temperature (oC) the saturated temperature of steam at the
Tbp: cold side inlet temperature (oC) pressure of supply header should be applied to ΔT calculation,
tav: average cold side temperature (normal because the flow control valve tend to open to maintain the
operation) (oC) steam flow at a constant. If the other control system is applied
t1/t2: cold side inlet/outlet temperature (oC) to the steam supply, the normal operating temperature of
steam may be used.
(3) Ucalc (calculated overall heat transfer coefficient ) should be
used for recalculation.

Hot Oil Vaporizer

When the hot oil supply stops due to failures such as a hot oil
pump stop, overpressure does not occur because
of no heat input to the reboiler. If the hot oil supply continues,
the heat duty should be calculated based on the temperature
profile under the relieving conditions.
(1) Inlet and outlet temperatures of process fluid to/from the
reboiler will increase because the boiling temperature of
process fluid increases when the operating pressure reaches
the relieving pressure.
(2) The flow rate of hot oil is maintained the same as in normal
operation.
(3) UCALC (calculated overall heat transfer coefficient) should be
used for recalculation.
Based on the above conditions, assume the outlet temperature
of hot oil (T2’) and obtain the reboiler duty
(ΔQ’) by trial and error calculation using the following
equations :

Table 4.6: Relief loads guidelines for Heat Exchangers

Liquid Thermal Expansion [4] Solar Radiation [4]

A 3/4-inch × 1-inch nominal pipe size (NPS) relief valve 𝑄 = 𝑞𝑠 𝜀𝐴

is commonly used, even though it will be oversized, Q: heat absorbed by solar radiation (kcal/hr)
since relieving load for thermal expansion will usually qs : heat flux by solar radiation (kcal/hr.m2)
be small. ԑ : emissivity
If there is reason to believe that this size is not A: projected heat transfer area (m2)
adequate, the relieving load should be obtained
according to the following equation
𝛼𝑄 solar radiation
𝑉=
𝐶 ∗ 𝑆. 𝐺 ∗ 𝐶𝑝 When the solar radiation data is not available, the noon values in
V: volumetric flow rate at flowing temperature (m3lhr) the following table can be applied to obtain the maximum heat
𝛼:cubical expansion coefficient (1/°K) absorbed from solar radiation.
Q: total heat transfer rate. ( kW) Maximum Expected Solar Radiation of Various North Latitudes
specific gravity, water = 1.0 (-) Latitude (kcal/h.m2) (kW/m2)
Cp specific heat of the fluid 24-hr N30° 353 0.41
C: factor (1000) average
N40° 353 0.41
N45° 353 0.41
Noon N30° 990 1.15
value
N35° 976 1.14
N40° 950 1.1
N45° 922 1.07
Above table presents typical highest values of monthly average
solar radiation on a horizontal surface throughout the year,
based on analysis of Weather Bureau records for a number of
stations through the United States. Refer to Section 12 Table 12-5
of Perry’s Chemical Engineer’’ Handbook 7th Edition.
Gravity of Oil [Deg API] αv 1/°C or 1/K (1/°F) typical value of emissivity
3 – 34.9 0.00072 (0.0004) Material Emissivity, ε
35 – 50.9 0.00090 (0.0005) Clean and polished metal 0.1
51 – 63.9 0.00108 (0.0006) Metals, general 0.3
64 – 78.9 0.00126 (0.0007) Rusty metal 0.8
79 – 88.9 0.00144 (0.0008) Painted surface 0.95
89 – 93.9 0.00153 (0.00085)
94 and lighter 0.00162 (0.0009)
Temperature °C αv 1/°C or 1/K For horizontal piping or vessels, vertical projected area on ground
(water) is used. For vertical piping or vessels, horizontal projected area
15.6 0.00018 may be used as conservative side.
20 0.00021
30 0.00030
40 0.00040
50 0.00047

Table 4.7: Relief loads guidelines for Liquid Thermal Expansion and Solar radiation
Inadvertent Control Valve Opening
Flashing Liquid [4]

(a) Calculate ΔP of the control valve at the relieving condition

(b) Breakthrough flow rate should be calculated on a liquid phase basis by using the selected CV value,
since LPG is in liquid phase at the inlet of the control valve.
(c) Calculate the flashed vapor flow rate (VF) by the flash calculation at the relieving pressure.
(d) If the vapor space is enough to accommodate the let down liquid for the operator’s response time, the
relieving load (VR) = VF – VN ( VN = vapor flow rate at normal operation).
(e) If the vapor space is not enough, consider the relieving of vapor-liquid mixture.
(f) In this case, pay attention to an occurrence of slug flow in two phase lines.
Gas Breakthrough [4]

(a) Calculate ΔP of the control valve at the relieving condition.

(b) Gas breakthrough flow rate (VB) should be calculated by using the selected CV value.
(c) Relieving load (VR) = VB – VN , ( VN = vapor flow rate at normal operation).
(d) In this case, pay attention to an occurrence of slug flow in two phase lines.
Liquid [Note 1] [7]
Estimation of control valve Cv[Note 2][7]
△𝑃
𝑄𝐿 = 0.86 ∗ 𝐶𝑣 . √ 𝑣𝑎𝑙𝑣𝑒 𝐶𝑣 1 Single-seated
𝐺 𝑑 (𝑖𝑛𝑐ℎ𝑒𝑠) = ( )2
9 valves
QL in m3 /hr
△P in kg/cm2
G: Specific gravity (-)
Steam [7]
△ 𝑃. 𝑃2 𝑣𝑎𝑙𝑣𝑒 𝐶𝑣 1 Double-seated
𝑄𝑠 = 394.155 ∗ 𝐶𝑣 . √ 𝑑 (𝑖𝑛𝑐ℎ𝑒𝑠) = ( )2
𝑇 12 valves
Qs in kg/hr
△P and P2 in kg/cm2
T in oK
Gas [Note 1][7]
△ 𝑃. 𝑃2
𝑄𝐺 = 386.67 ∗ 𝐶𝑣 √ 𝑣𝑎𝑙𝑣𝑒 𝐶𝑣 1
𝑍2 . 𝐺. 𝑇 Butterfly valve
𝑑 (𝑖𝑛𝑐ℎ𝑒𝑠) = ( )2
3 20 sizes
Qs in Nm /hr
△P and P2 in kg/cm2
T in oK
G is the molecular weight divided by 29
Table 4.8: Relief loads guidelines for Inadvertent open control valve

NOTE 1: Assuming a subcritical and turbulent flow

NOTE 2: Shortcut mentioned in ref [8] were used in the tool (Cv =10*d2), d in inches
Wetted Area Calculations [5]
To determine vaporization rate, the surface
area wetted by a vessel's internal liquid and is Semi-Ellipsoidal/Elliptical Heads
up to 7.6 m (25 feet) above grade, denoted as Or Torispherical/Dished Heads
"fire elevation", needs to be considered. The
term "grade" usually refers to ground level,
but may be at any level at which a sizable fire
could be sustained.
z
D
Vessels
Cylindrical part
h
Vertical
𝐴𝑤
𝐷2 ℎ
= 𝜋 (( − 0.5) 𝐵 + 1
8 𝐷
ℎ
L 1 4𝜀 ( − 0.5) + 𝐵
+ ln( 𝐷 ))
D 4𝜀 2 − √3

2
ℎ
√
𝐵 = 1 + 12 ( − 0.5)
𝐷

Aw = 2πD*L For 2:1 Elliptical head ε = 0.866,

Horizontal elsewise:

4𝑧 2
𝜀 = √1 −
𝐷2
D
z: inside dish depth
h
𝜀 : Eccentricity of elliptical heads
L
𝐴𝑤 = 𝐿𝐷 cos −1 (1 − (2ℎ/𝐷))
Heads
Hemispherical Heads

𝐷
𝐴𝑤 = 𝜋ℎ
2
Shortcut Calculations for various wetted Areas [4]

Figure 4.2: Heat exchanger U tube side Aw

Figure 4.3: Fixed Heat exchanger tube side Aw

Figure 4.4: Kettle reboiler Aw

Figure 4.5: Floating head Hex. Aw

Figure 4.6: Heat exchanger U shell-side Aw

Figure 4.7: Fixed heat exchanger Shell-side Aw

Figure 4.8: Spherical Tank Aw

Figure 4.9: Vertical vessel fixed head Aw

Figure 4.10: Vertical Vessel Elliptical Head Aw

Figure 4.11: Horizontal Vessel Aw

Figure 4.12: Trayed Column Aw

Examples
API RP 520 Example No. 1
Input Units API RP 520 Part I Tool
Mass flow Kg/hr 24270 24271
Molecular weight - 51 51 (Using composition)
o
Relieving Temperature K 348 74.85 oC
Design pressure of equipment kPa / kg/cm2 517 5.272
Backpressure kPa / kg/cm2 101.325 1.033
Overpressure % 10 10 (Calculated)

Output
Relieving pressure kPa / kg/cm2 670
Z - 0.9 0.98
Critical Pressure kPa / kg/cm2 392 -
Cp/Cv - 1.11 1.05
Kb - 1 1
Kc - 1 1
kd - 0.975 0.975
Acalculated cm2/ in2 36.98 / 5.73 39.5 / 6.12
Aselected cm2/ in2 41.16 / 6.38 41.16 / 6.38
Orifice Designation P P
1. Estimating Relief Pressure and constants

Relief pressure for one relief valve, no rupture disc, and non-fire case will equal 1.1 the equipment
design pressure
Pr = 1.1 * 517 = 670 kPa (6.8321 kg/cm2)
Kb = 1 as backpressure is atmospheric
Kc = 1, as there’s no rupture disc installed
Kd = 0.975
1. Checking whether the flow is critical

2 𝑘⁄(𝑘−1) 2 1.11⁄(1.11−1)
𝑃𝐶𝑓 = 𝑃1 ∗ [ ] = 670 ∗ [ ] = 670 ∗ 0.5825 = 390.33
𝑘+1 1.11 + 1
The PRV sizing is based on the critical flow equation since the backpressure (0 kPag) is less than the
critical flow pressure (291 kPag).

2. Calculating Orifice Area

2 (𝑘+1)/(𝑘−1) 2 (1.11+1)/(1.11−1)
𝐶 = 0.03948√𝑘 [ ] =0.03948*√1.11 [ ] = 0.03948 ∗ 0.63045 =
𝑘+1 1.11+1
0.02489

Using The C graph provided earlier (Figure 2.7), C = 0.0249 for Cp/Cv of 1.11
𝑊 ∗ √𝑇𝑍 24270 348 ∗ 0.9
𝐴= = √ = 3698 𝑚𝑚2
𝐶𝐾𝑑 𝑃1 𝐾𝑏 𝐾𝑐 √𝑀 0.0249 ∗ 0.975 ∗ 1 ∗ 1 ∗ 670 51

Step 3: Input relief/back conditions (T&P) Step 4: Download your calculations table!
API RP 520 Example No. 4
Input Units API RP 520 Part I Tool
Mass flow Kg/hr 69615 69615
Relieving Temperature C 427 427
Design pressure of equipment kPag / kg/cm2g 11032 112.5
Backpressure kPag / kg/cm2g 101.325 1.033
Overpressure % 10 10

Output
Relieving pressure kPa / kg/cm2 12236 (124.77) 124.77
Kb - 1 1
Kc - 1 1
kd - 0.975 0.975
KN - 1.01 1.01
KSH - 0.855 0.862
Acalculated cm2/ in2 12.87 / 1.995 12.75 / 1.97
Aselected cm2/ in2 18.41 / 2.853 18.41 / 2.853
Orifice Designation L L
1. Estimating Relief Pressure and constants

Relief pressure for one relief valve, no rupture disc, and non-fire case will equal 1.1 the equipment
design pressure
Pr = 1.1 * 11032 + 101.325 = 12236.5 kPa (124.77 kg/cm2)
Kb = 1 for conventional valve discharging to atmosphere
Kc = 1, as there’s no rupture disc installed
Kd = 0.975
KSH : 0.855 from tables
KN : P1 (124.77 kg/cm2.a) is > 106.5 kg/cm2.a
KN = (0.1906 *1774.7 - 1000)/(0.2292 *1774.7- 1061) = 1.01

2. Calculating Orifice Area

190.5 ∗ 𝑊 190.5 ∗ 69615
𝐴= = = 1287 𝑚𝑚2 (12.87 𝑐𝑚2 )
𝑃1 𝐾𝑑 𝐾𝑠ℎ 𝐾𝑁 𝐾𝑏 𝐾𝑐 12236.5 ∗ 0.975 ∗ 0.855 ∗ 1.01 ∗ 1 ∗ 1
See API 526 for the selection of the proper orifice size. API 526 provides standard effective

orifice areas in terms of letter designations. For this example, a “L” size orifice should be selected since it

has an effective orifice area of 2.853 in.2 (1841 mm2)

Step 1: Calculate Relief Pressure Step 2: Input mass flow rate and composition

Step 3: Input relief/back conditions (T&P) Step 4: Download your calculations table!
API RP 520 Example No. 5
Input Units API RP 520 Part I Tool
Mass flow L/min 6814 368010 (kg/hr)
Specific Gravity 0.9 0.9
Viscosity Saybolt / cP 2000 / 388.5 388.5
Design pressure of equipment kPag / kg/cm2g 1724 17.58
Backpressure kPag / kg/cm2g 345 3.52 (4.55 abs)
Overpressure % 10 10
Output
Relieving pressure kPag / kg/cm2 1896 (19.33 kg/cm2.g) 20.37 (19.337 gauge)
Kc - 1.0 1
Kd - 0.65 0.65
Kw - 1 0.97
Kv - 1 (initially) / 0.982 0.99
Acalculated cm2/ in2 31.22 / 4.84 30.71 / 4.83
Aselected cm2/ in2 41.16 / 6.38 41.16 / 6.38
Orifice Designation P P
1. Estimating Relief Pressure and constants

Relief pressure for one relief valve, no rupture disc, and non-fire case will equal 1.1 the equipment
design pressure
Pr = 1.1 * 1724 = 1896.4 kPa (kg/cm2)
Kw = 1 as backpressure is atmospheric
Kc = 1 as there’s no rupture disc installed
Kd= 0.975
Kv: initially assumed 1
2. Calculating Orifice Area

11.78𝑄√𝐺 11.78 ∗ 6814 0.9

𝐴= = √ = 3066 𝑚𝑚2 (30.66 𝑐𝑚2 )
𝐾𝑑 𝐾𝑤 𝐾𝑣 𝐾𝑐 √𝑃1 − 𝑃2 0.975 ∗ 1 ∗ 1 ∗ 1 1896 − 345

See API 526 for the selection of the proper orifice size. API 526 provides standard effective
orifice areas in terms of letter designations. For this example, a “P” size orifice should be selected since it
has an effective orifice area of 6.38 in.2 (4116 mm2)
3. Check Kv and recalculate Acalculated
𝑄 ∗ (18800 ∗ 𝐺) 6814*18800*0.9
𝑅𝑒 = = = 4625
𝜇√𝐴 388.5√4116
−0.5
170
𝑘𝑣 = ( + 1) = 0.982
𝑅𝑒
𝐴𝑅 3066
𝐴= = = 3122 𝑚𝑚2
𝐾𝑣 0.982
Step 1: Calculate Relief Pressure Step 2: Input mass flow rate and composition

Step 3: Input relief/back conditions (T&P) Step 4: Download your calculations table!
Fire Case Example: Benzene in a Horizontal Vessel
Input Units Reference [1] Tool
Units Converted
Fluid Benzene Benzene
Molecular weight - 78.11 78.11
Relieving Temperature K 37.78 + 273.15 [Note] 250 + 273.15 [Note 2]
Design pressure of equipment kPag / kg/cm2.g 1379 14.06 kg/cm2.g (1379 kPa)
Backpressure kPa / kg/cm2 Atmospheric Atmospheric
Overpressure % 21 21
Vessel Geometry data Bare Vessel Bare Vessel
Horizontal + Horizontal +
spherical heads Cylindrical heads
Diameter m 4.572 4.572
T-T Length m 9.144 9.144
Level (Normal liquid level) m 3.734 3.734
Height above grade m 4.572 4.572

Output
Relieving pressure kPa / kg/cm2 1769.88 18.04 kg/cm2.g (1769.7 kPa)
ΔH kCal/kg 95.62(input) 146 [Note 1]
Wetted Area (Aw) m2 83.7 121
Mass flow Kg/hr 14665 13020.15
Z - 1 (input) 0.81
Critical Pressure kPa / kg/cm2
Flow condition - Critical Critical
Cp/Cv - 1.12 (input) 0.97
Kb - 1 1
Kc - 1 1
kd - 0.975 0.975
C
Acalculated cm2/ in2 6.787 / 1.052 in2 8.20495 / 1.27177 in2
Aselected cm2/ in2 8.303 / 1.287 in2 8.303 / 1.287 in2
Orifice Designation J (nozzles 2”/3”) J (nozzles 2”/3”)
Notes:

1. Heat of vaporization is higher in the tool as a result of adding the heat required to bring the fluid
from the subcooled phase to saturation temperature
2. A temperature of 250 was assumed to obtain benzene at the vapor phase; assuming relived and
vaporized benzene at 37.78 C & 18 kg/cm2 is illogical. This, however, will increase Acalculaed.
3. Estimate Vessel wetted area

Level of the wetted area = 7.6 – 4.57 = 3.03 m

ratio of level = 3.03/4.57 = 0.663

Total Area of the horizontal vessel (spherical heads)

= (L+D)πD = (9.1 + 4.57)π*4.57 = 196.26 m2 9.1 m

Fraction of area (from Figure 4.11) = 0.6

4.57 m
Wetted area = Area fraction * total surface area
7.6 – 4.57

= 0.6*196.26 = 117.75 m2 Wetted Area

4.57 m 3.73+4.57 m 7.6 m

2ℎ
𝐴𝑤 (𝐶𝑦𝑙𝑖𝑛𝑑𝑟𝑖𝑐𝑎𝑙 𝑝𝑎𝑟𝑡) = 𝐿𝐷 cos −1 (1 − ( ))
𝐷
3.03
= 9.1 ∗ 4.57 𝑐𝑜𝑠 −1 (1 − (2 ∗ )) = 587 ∗ cos −1 (−0.326) = 79.14 𝑚2
4.57
𝐴𝑤 (2 ℎ𝑒𝑎𝑑𝑠) = 𝜋ℎ𝐷 = 3.14 ∗ 3.03 ∗ 4.057 = 38.6 𝑚2
𝐴𝑤 𝑡𝑜𝑡𝑎𝑙 = 79.147 + 38.6 = 117.74 𝑚2
4. Estimate Relief Load

Q = qoFA0.82

F = 1 for Bare vessel

qo = 37130 (adequate drainage is provided)

Q = 371301117.750.82 = 1853159.92 kCal/hr

𝑄 1853159.92
𝑊= = = 12693 𝑘𝑔/ℎ𝑟
∆𝐻 146
3. Estimating Relief Pressure and constants

Relief pressure for one relief valve, no rupture disc, and fire case will equal 1.21 the equipment design
pressure
Pr = 1.21 * 1379 + 101.325 = 1770 kPa (18.05 kg/cm2)
Kb = 1 as backpressure is atmospheric
Kc = 1, as there’s no rupture disc installed
Kd = 0.975
4. Checking whether the flow is critical

2 𝑘⁄(𝑘−1) 2 1.12⁄(1.12−1)
𝑃𝐶𝑓 = 𝑃1 ∗ [ ] = 1770 ∗ [ ] = 1770 ∗ 0.5805 = 1027.5 𝑘𝑃𝑎
𝑘+1 1.12 + 1
The PRV sizing is based on the critical flow equation since the backpressure (0 kPag) is less than the
critical flow pressure (1027 kPag).
5. Calculating Orifice Area

2 (𝑘+1)/(𝑘−1) 2 (1.12+1)/(1.12−1)
𝐶 = 0.03948√𝑘 [ ] =0.03948*√1.12 [ ] = 0.03948 ∗ 0.4 =
𝑘+1 1.12+1
0.0249

Using The C graph provided earlier (Figure 2.7), C = 0.025 for Cp/Cv of 1.12

𝑊 ∗ √𝑇𝑍 24270 348 ∗ 1

𝐴= = √ = 621 𝑚𝑚2
𝐶𝐾𝑑 𝑃1 𝐾𝑏 𝐾𝑐 √𝑀 0.025 ∗ 0.975 ∗ 1 ∗ 1 ∗ 1770 78.11

See API 526 for the selection of the proper orifice size. API 526 provides standard effective orifice areas
in terms of letter designations. For this example, a “J” size orifice should be selected since it has an
effective orifice area of 1.287 in.2 (830 mm2)
Step 1: Calculate Relief Pressure Step 2: Select estimate relief loads and choose “External Fire
due to vaporization”

Step 3: input relieved stream composition and conditions Step 4: Input wetted Area required inputs
Step 5: check your results and calculations’ matrix Step 6: go to “3. Gas or vapor” and use Calculated Relief Load

Step 7: Input mass flow rate and composition Step 8: Download your calculations table!

Note: 250 oC was used to obtain the vapor phase

Definitions
Maximum operating pressure is the expected maximum pressure during operation of the system
upstream of the pressure relief valve.

Maximum allowable working pressure (MAWP) is the maximum gauge pressure permissible at the top of
a completed vessel in its operating position for a designated temperature. The pressure is based on
calculations for each element in a vessel using nominal thickness, exclusive of additional metal thickness
allowed for corrosion and loading other than pressure. The maximum allowable working pressure is the
basis for the pressure setting of the pressure relief devices that protect the vessel.

Design gauge pressure refers to at least the most severe conditions of coincident temperature and
pressure expected during operation. This pressure may be used in place of the maximum allowable
working pressure in all cases where the MAWP has not been established. The design pressure is equal to
or less than the MAWP.

Accumulation is the pressure increase over the maximum allowable working pressure of the vessel
during discharge through the pressure relief device, expressed in pressure units or as a percent.
Maximum allowable accumulations are established by applicable codes for operating and fire
contingencies.

Overpressure is the pressure increase over the set pressure of the pressure relief device, expressed in
pressure units or as a percent. It is the same as accumulation when the relieving device is set at the
maximum allowable working pressure of the vessel.

Rated relieving capacity is the portion of the measured relieving capacity permitted by the applicable
code regulation to be used as a basis for the application of a pressure relief device. Stamped capacity is
the rated relieving capacity that appears on the device nameplate.

The stamped capacity is based on the set pressure plus the allowable overpressure for compressible
fluids and the differential pressure for incompressible fluids. The stamped capacity shall not exceed 90%
of the average capacity of the valves tested.[ASME VIII Div. 1 UG-13 1 (d)(l)]

Set pressure is the inlet gauge pressure at which the pressure relief device is set to open under service
conditions.

Pressure differential is the difference between the set pressure of the pressure relief device and the
operating pressure of the protected vessel.

Cold differential test pressure is the pressure at which the pressure relief valve is adjusted to open on the
test stand. The cold differential test pressure includes corrections for the service conditions of back
pressure or temperature or both.

Back pressure is the pressure that exists at the outlet of pressure relief devices as a result of the pressure
in the discharge system. It is the sum of the superimposed and built-up back pressure.

Superimposed back pressure: The static back pressure that exists at the outlet of a pressure relief device
at the time the device is required to operate. It is result of pressure in the discharge system corning from
other sources and may be constant or variable. Built-up back pressure: The built-up back pressure is the
increase in pressure in the discharge header that develops as a result of flow after the pressure relief
device or devices open.

The built-up back pressure is caused by flow from the particular device and others, if any, which
simultaneously discharge into the disposal system. This type of back pressure is variable. The built-up
back pressure shall be less than the allowable back pressure, 10% of the set pressure for the
conventional type or 50% of the set pressure for the balanced-bellows type.

Constant Back Pressure: The static back pressure that exists under normal operation where no relief
device is operated; i.e. constant back pressure of the superimposed back pressure. The constant back
pressure is used to determine the spring set pressure of conventional type pressure relief valve as a
difference between the set pressure and the constant back pressure. Therefore, the opening pressure
will vary depend on the built-up back pressure. The opening pressure of balanced-bellows type is the set
pressure of the valve and is independent of any back pressure.

Blowdown is the difference between the set pressure and the closing pressure of a pressure relief valve,
expressed as a percent of the set pressure or in pressure units.

Opening pressures the value of increasing inlet static pressure at which there is a measurable lift of the
disc or at which discharge of the fluid becomes continuous.

Closing pressure is the value of decreasing inlet static pressure at which the valve disc reestablishes
contact with the seat or at which lift becomes zero.

Simmer is the audible or visible escape of compressible fluid between the seat and disc at an inlet static
pressure below the set pressure and at no measurable capacity

Leak-test pressure is the specified inlet static pressure at which a seat leak test is performed. The term
relieving conditions is used to indicate the inlet pressure and temperature on a pressure relief device at a
specific overpressure. The relieving pressure is equal to the valve set pressure plus the overpressure. The
temperature of the flowing fluid at relieving conditions may be higher or lower than the operating
temperature.

Chatter refers to the motion that causes the disc to contact the seat and damage the valve and
associated piping. Chattering may result in lowered capacity and damage to the seating surfaces.

Flutter refers to the abnormally rapid reciprocating motion of the movable parts of a pressure relief valve
in which the disc does not contact the seat

Atmospheric discharge is the release of vapors and gases from pressure relief or depressing devices to
the atmosphere.

Flare system is a means for the safe disposal of waste gasses by closed pipeline and combustion system.
With an elevated mre the combustion is carried out at the top of a pipe or stack where the burner and
igniter are located. A ground flare is similarly equipped except that combustion is carried out at or near
ground level. A burn pit differs from a flare in that it is normally designed to handle both liquids and
vapors.

Vent stack is the elevated vertical termination of a disposal system that discharges vapors into the
atmosphere without combustion or chemical conversion of the relieved fluid.
References
[1] API RP 520 Part I, 10th edition.
[2] Engineering Encyclopedia: Aramco desktop standards: sizing and selecting Pressure relief valves
[3] Pressure relief devices rev. 5, JGC standard practice 2002 JGS 210-120-1-61E
[4] Pressure Relieving design rev. 5, JGC standard practice 2008 JGS 210-120-1-40E
[5] Volume and Wetted Area of Partially Filled Horizontal Vessels, Link:
https://neutrium.net/equipment/volume-and-wetted-area-of-partially-filled-horizontal-vessels/
[6] Sizing Pressure-Relief Devices, Daniel A. Crowl and Scott A. Tipler. (AIChE)
[7] Elements of chemical process engineering chapter 1, Basic Process Engineering principles
[8] Chemical process engineering Vol. 1, Chapter 5 P. 246 By Kayode Coker

T Temperature (oC or oK) P Pressure (kPa or bar)

Gg gas Specific gravity µ Viscosity (mPa.s/cP)
ρ Density (kg/m3) SG Specific gravity

Di Tube inside diameter (mm) V Velocity (m/s)

Do Tube outside diameter (mm) Z Gas compressibility factor. dimensionless
Re Reynold’s number P1 Upstream or inlet pressure (kPa or bar)
Q Volume flow rate (m3/hr) or (Nm3/hr) P2 Downstream or outlet pressure (kPa or bar)
Cv Control Valve flow coefficient (gpm/psi) Pc Absolute critical pressure (kPa or bar)
Kv Control Valve flow coefficient (m3.hr-1/bar) Pv Absolute vapor pressure at T1
D1 Inlet pipe diameter (mm) △P Pressure drop across the valve (△P =P1-
P2) (kPa or bar)
d Normal valve size (mm or in.) △Pmax(L) Maximum allowable liquid sizing
pressure drop (kPa or bar)
D2 outlet pipe diameter (mm) △Pmax(LP) Maximum allowable sizing pressure drop
with attached fittings (kPa or bar)
Fd Valve style modifier, dimensionless q Volume flow rate (m3/hr)
FF Liquid critical pressure ratio factor, Qmax Maximum flow rate at given valve
dimensionless upstream condition (m3/hr)
Fk Ratio of specific heat factor, T1 Absolute upstream temperature (oK)
dimensionless
FL Rated liquid pressure recovery factor. W Mass rate of flow (kg/h)
dimensionless
FLP Combined liquid pressure recovery x Ratio of pressure drop to the upstream
factor and piping geometry factor of a or inlet pressure (△P/P1)
valve with attached fittings,
dimensionless (Note: if no fitting
attached, FLP=FL)
Fp Piping geometry factor. dimensionless xT Rated pressure drop ratio factor,
dimensionless
k Ratio of specific heats, dimensionless Y Expansion factor (ratio of flow coefficient
for a gas to that for a liquid at the same
Reynolds number). dimensionless
K Head loss coefficient of a valve due to N Numerical constant
installed fittings attached to the valve,
dimensionless
M Molecular weight, dimensionless  Specific weight at the valve inlet
condition (Note: The standard condition
is defined as 15.56 oC and 1 atm)
Input Required

Calculate Cv (D,Q,△P Known) Calculate Cv,D (Q,△P Known)

D1,d,D2 [Note 1] in
Flow Rate [Note 1] Flow Rate [Note 1]
3
(Gas) Nm /hr (Gas) Nm3/hr
(Liquid) m3/hr (Liquid) m3/hr
(steam) kg/hr
Inlet pressure kg/cm².a Inlet pressure kg/cm².a
Outlet pressure kg/cm².a Outlet pressure kg/cm².a
Inlet temperature ºC Inlet temperature ºC
[Note 2]
Molecular weight - Molecular weight[Note 2] -
[Note 2]
Compressibility factor - Compressibility factor[Note 2] -
Avergage k (Cp/Cv) - Avergage k (Cp/Cv) [Note 2] -
Viscosity[Note 2] cP Viscosity[Note 2] cP
[Note 2]
Critical Pressure kg/cm².a Critical Pressure[Note 2] kg/cm².a
[Note 1]
xT - xT[Note 1] [Note 3] -
[Note 3]
FL - FL[Note 3] -
Fd[Note 3] - Fd[Note 3] -
Note 1: parameter required depending on the case selected (Liquid, Gas or Steam)
Note 2: Alterna ve op on: Use the tool proper es es mator
Note 3: Alterna ve op on: Use ANSI/ISA table for these factors using “Modify assump ons)
Output Obtained

Summary Table
Calculation - D inlet assumed in.
Inlet pressure kg/cm².a D outlet assumed in.
Flow Rate Nm³/hr FL -
Outlet pressure kg/cm².a Fd -
Inlet temperature ºC xT -
Molecular weight - FP -
Compressibility factor - xTP -
Avergage k (Cp/Cv) - FR -
Viscosity cP choked -
Shortcut Cv gpm/psi Y -
Kv m³.hr-1/bar Rev -
Cv gpm/psi laminar -
Control Valve size in.
Control Valve Specifications sheet Table of matching valves
Control valve sizing

1. Rules of thumb for Estimating △Psizing and QSizing & Other Guidelines
2. Different options for Calculating Valve Cv
3. Selecting Valve Size and calculating the opening percentage (or lift %).
4. General guidelines when selecting a control valve
1. Rules of thumb for Estimating △Psizing and QSizing[3][4]

Es mate △Psizing using the following guidelines:

“The sum of the following pressure drops at maximum ﬂow may be used for this purpose:

1. 20% of the friction drop in the circuit (excluding the valve). (A circuit generally
includes all equipment between the discharge of the pump, compressor, or vessel
and the next point downstream of which pressure is controlled. In most cases this
latter point is a vessel)
2. 10% of the static pressure of the vessel into which the circuit discharges up to the
pressure of 200 psig (14.06 kg/cm2.g),
20 psig from 200 psig to 400 psig (1.406 from 14.06 to 28.12 kg/cm 2.g)
and 5% above 400 psig.” [Note]

Control valve at Pump discharges[4]

System friction losses ΔPcv or sizing
∑ ∆𝑃 < 5.0 kg/cm2 ΔPcv = 0.5*∑ ∆𝑃
5.0 kg/cm2 < ∑ ∆𝑃 < 6.25 kg/cm2 ΔPcv = 2.5 kg/cm2
6.25 kg/cm2 < ∑ ∆𝑃 < 10 kg/cm2 ΔPcv = 0.4*∑ ∆𝑃
2
10 kg/cm < ∑ ∆𝑃 13.4 kg/cm2 ΔPcv = 4*∑ ∆𝑃
13.4 kg/cm2 < ∑ ∆𝑃 ΔPcv = 0.3*∑ ∆𝑃
NOTE: Another rule of thumb suggests that for good control where possible, make the
control valve take 50-60% of the system ﬂowing head loss [2]

Minimum pressure drop[4]

1) Liquid service ΔPcv = 0.7 kg/cm2
2) Vapor service ΔPcv = 0.2 kg/cm2

Es mate QSizing using the following guidelines:

“In order to ensure that the valve is in a controlling posi on at the maximum ﬂow rate, the
valve Cv is the maximum process value determined above divided by 0.8. The reasons for
using this factor are that:
1. It is not desirable to have the valve fully open at maximum flow since it’s not then in
a controlling position.
2. The valves supplied by a single manufacturer often vary as much as 10-20% in Cv.
3. Allowance must be made for pressure drop, flow rate, etc. values which differ from
design.” [1]
Another Rule of thumb is to use the greater of the following [2]
1. QSizing = 1.3* QNormal or 2*QNormal [4]
2. QSizing = 1.1* QMaximum

Velocity & Mach number [8] [2]

Keep the valve outlet velocity below the following guidelines [8]
Liquids 50 feet per second (15.24 m/s)
Gases Approaching Mach 1.0
Mixed Gases and Liquids 500 feet per second (152.4 m/s)

For saturated steam keep the control valve outlet velocity below 0.25 mach. Branan Also suggests
keeping the inlet velocity below 300 /s for 2” pipes and below 200 /s for larger sizes. [2]
The above figures are guidelines for typical applica ons.
In general, smaller-sized valves handle slightly higher veloci es and large valves handle lower
veloci es. Special applica ons have par cular velocity requirements;
a few of which are provided below.
Liquid applica ons – where the fluid temperature is close to the satura on point – should be
limited to 30 feet per second (9.1 m/s) to avoid reducing the fluid pressure
below the vapor pressure. This is also an appropriate limit for applica ons designed to pass the
full flow rate with a minimum pressure drop across the valve.
Valves in cavita ng service should also be limited to 30 feet per second (9.1 m/s) to minimize
damage to the downstream piping. This will also localize the pressure recovery
which causes cavita on immediately downstream from the vena contracta.
In flashing services, veloci es become much higher due to the increase in volume resul ng from
vapor forma on.
For most applica ons, it is important to keep veloci es below 500 feet per second. (152.4 m/s)
Expanded outlet-style valves – such as the Mark One-X – help to control outlet veloci es on such
applica ons. Erosion damage can be limited by using chrome-moly body material and hardened
trim. On smaller valve applica ons that remain closed for most of the me – such as heater drain
valves – higher veloci es of 800 to 1500 feet per second (243.84 - 457.2 m/s) may be acceptable
with appropriate materials.
Gas applica ons where special noise a enua on trim is used should be limited to approximately
0.33 Mach. In addi on, pipe veloci es downstream from the valve are cri cal to the overall noise
level. Experimenta on has shown that veloci es around 0.5 Mach can create substan al noise
even in a straight pipe. The addi on of a control valve to the line will increase the turbulence
downstream, resul ng in even higher noise levels.

Noise
Typical Noise limita on: 85-90 dB(A) [4][5]
Cau on: Noise levels over 110 dBA may cause vibra on in valves/piping resul ng in equipment
damage. [8]

2. Diﬀerent op ons for Calcula ng Valve Cv

Shortcut Cv
1. General shortcut[1]
𝐶 = 10𝑑 d: Control valve diameter (in.)

2. A more specific shortcut[2]

𝐶 = 𝐶 .𝑑 d: Control valve diameter (in.)
3. Simplified Cv Calculations
Where:
. QL = liquid flow in gpm at condi ons
𝐶 =𝑄 for liquids
△ △P = pressure drop across valve (psi)
SG = specific gravity of liquid at
𝐶 = Condi ons
△ . for Steam
Qs = steam rate (lb/hr)
P2 = pressure downstream of valve (psia)
. .
𝐶 = for Gases Qg= gas flow in sc (60 °F, 14.7 psia)
△ .
T = temperature of gas in °R (°F + 460)
4. ANSI/ISA 75.01.01 Equations for Control Valve Sizing[6] Gg = mol weight of gas divided by 29
Incompressible fluids Z = compressibility factor at downstream
condi ons
There are six steps for a control valve size selec on for liquids:
1) Specify the variables required to size the valve.
2) Determine the equa on constant (N).
3) Determine the piping geometry factor (FP).
4) Determine the maximum flow rate upstream of the valve (qmax) and the allowable
pressure drop (ΔPmax).
5) Solve for the required flow coefficient (Cv) value using an appropriate equa on. (Use Annex
B.1 flowcharts)
6) Select the appropriate flow coefficient table and the calculated flow coefficient
valve to select the suitable valve size.
Turbulent flow
Non-choked flow (without fi ngs) (ΔP < FL2(P1-FFPv) )
N1: 0.865
𝑸 𝝆 𝟏 ⁄𝝆 𝒐
𝐶𝑣 = Eq.1
𝑵𝟏 ∆𝑷 Q in m3/hr

P in bar
Non-choked ﬂow (with ﬁ ngs) (ΔP < (FLP/Fp) 2(P1-FFPv) )
ρ in kg/m3
𝑸 𝝆𝟏 ⁄𝝆𝒐
𝐶𝑣 = Eq.2 ρ0 : 1000 kg/m3
𝑵𝟏 𝑭 𝑷 ∆𝑷

Choked ﬂow (without ﬁ ngs) (ΔP > FL2(P1-FFPv) )

𝑸 𝝆𝟏 ⁄𝝆𝒐
𝐶𝑣 = Eq.3
𝑵𝟏 𝑭𝑳 𝑷𝟏 − 𝑭𝑭 𝑷𝒗

Choked ﬂow (with ﬁ ngs) (ΔP > (FLP/Fp) 2(P1-FFPv) )

𝑸 𝝆 𝟏 ⁄𝝆 𝒐
𝐶𝑣 = Eq.4
𝑵𝟏 𝑭𝑳𝑷 𝑷𝟏 − 𝑭𝑭 𝑷𝒗

Non-Turbulent ﬂow (without ﬁ ngs)

𝑸 𝝆𝟏 ⁄𝝆𝒐
𝐶𝑣 = Eq.5
𝑵𝟏 𝑭 𝑹 ∆𝑷
Specify the Variables Required to Size the Valve[5] The values of FL for the globe, v-notch ball, and bu erfly valves can
• Process fluid (e.g. oil and water); be extracted from Table 2. Ball valves with standard design are not
• Service condi ons and fluid physical proper es such as volume rate suitable for flow control; however, a v-notch ball valve with a unique
of flow (q), mass rate of flow (w), Upstream and downstream v-shaped hole in the ball allows precise flow control. Furthermore,
pressure values (P1 and P2), pressure drop across the valve (ΔP = P1 − bu erfly valves can be used for fluid control or thro ling in addi on
P2), upstream temperature (T1), Specific gravity (SG), Viscosity (µ), to fluid isola on.
Vapor pressure (Pv), Cri cal pressure (Pc), and solid%. If the valve is installed with fi ngs such as reducers a ached to it,
• Constants FL & Fd: A er selec ng required Valve type, trim type and the combined liquid pressure recovery factor and piping geometry
flow direc on, these valves can be es mated using Table 2. factor of a valve with a ached fi ngs known as parameter FLP is
applied instead of FL in Eq. (35) for maximum flow rate (qmax)
Determine Piping Geometry Factor (FP)
calcula on. FLP is calculated as per Eq. (30).
FP is a correc on factor that accounts for the pressure losses due to
piping fi ngs that are used to change the direc on or size of the Maximum Pressure Drop Calcula on for Valves Without Any
piping or take branches from the headers such as tees, elbows, and A ached Fi ngs
reducers. If fi ngs are used before or a er the valve, the effect of ΔPmax = FL2(P1-FFPv)
piping geometry must be considered for the valve sizing. However, if Maximum Pressure Drop Calcula on for Valves With A ached
no fi ng is a ached to the valve, FP is equal to one and does not Fi ngs
impact the valve sizing process. Equa on (36) shows how to calculate ΔPmax = (FLP/Fp) 2(P1-FFPv)
the piping geometry.
a) Solve for Flow Coefficient (valve diameter known)
Determine the Maximum Flow Rate (qmax) and Maximum Pressure Drop
Equa on (1 & 2) is used to calculate the modified flow efficient (Cv)
(ΔPmax)
value if ΔPmax > ΔP meaning that the flow will not be choked.
As its name implies, the maximum flow rate, also called limi ng flow
Equa on (3 & 4) applies to adjusted flow coefficient calcula on if
rate (qmax) or choked flow calculated from Eq. (35), cannot be
ΔPmax < ΔP when the flow will be choked.
increased even by increasing the pressure drop across the valve. For
1. Select FL from Table 2
example, flashing the vapor from liquids when the sta c valve
2. Calculate FF from Eq. 31
pressure drops below the liquid vapor pressure can cause a
3. Calculate ∆Pmax and determine if flow is choked or not.
maximum flow rate. It is important to calculate the maximum
4. Calculate Ci by Eq. 1 (non-choked) or Eq. 3 (Choked)
allowable pressure drop (ΔPmax) for selec ng the correct size of the
5. Calculate Rev using Eq. 28, Obtain Fd from Table 2.
valve if it is possible for the choked flow to develop. The calculated
6. If Rev > 10000, flow is turbulent and If valve size = pipe size
ΔPmax value is compared with the actual pressure drop across the
Use Ci previously Calculated. If not, Calculate Fp and
valve specified by the process engineer, and the lesser of these two is
subsequently its Cv using the iteration method described in
applied to the valve sizing process. If ΔPmax > ΔP, the flow will not be
flowchart.
choked and there is no need to determine the maximum flow rate as
7. If Rev < 10000, flow is non-turbulent. Calculate FR and its
per step 4. However,
subsequently its Cv using the iteration method described in
if a choked flow condi on exists (ΔPmax > ΔP), then step 5 for sizing
flowchart.
valves for liquids shall be changed by replacing the actual service
pressure differen al (ΔP = P1 − P2) with ΔPmax.
Select the Correct Valve Size
Maximum Flow Rate (qmax)
As a general rule, the smallest valve with a flow coefficient larger
𝑃 −𝐹 𝑃 Eq.35 than that calculated based on either Eq. (1 & 2) or Eq.(3 & 4),
𝑄 =𝑁𝐹𝐶 according to the informa on provided in Fisher representa ve Table
𝑆𝐺
shall be selected. Select the desired flow Characteris c and calculate
the es mated opening for the chosen valve.
The liquid cri cal pressure ra o factor (FF) can be calculated by using
Eq. (31), or Figure 5 for water. When a liquid pressure drops below its
b) Alterna ve op on for valve sizing (valve diameter isn’t known):
vapor pressure, the fluid state alters from a liquid to gas, that is called
Use the simplified equa ons described earlier then use the sizing
choked flow or vena contracta pressure condi on. FF is the ra o of
equa on in the following sec on (Another method for es ma ng
the vena contracta pressure at choked flow condi on to the vapor
control valve size (if tables not available))
pressure of the liquid at inlet temperature.
Check Noise levels, velocity and Cavita on index
𝑃 −𝑃
𝐶𝑎𝑣𝑖𝑡𝑎𝑡𝑖𝑜𝑛 𝑖𝑛𝑑𝑒𝑥 (𝛿) =
𝑃 −𝑃
Cavita on index: Ra o of pressure drop across the valve divided by
the difference between the inlet pressure and the vapor pressure of
the liquid. Valve selec on to ensure opera on above the cavita on
point is the primary use.

Criteria Cavita on consequence[5]

𝛿 > 2.0 No risk of cavita on
1.7 < 𝛿 < 2.0 There is no need for cavita on control
The trim is hardened for protec on
1.5 < 𝛿 < 1.7 There is some cavita on control required
1.0 < 𝛿 <1.5 Poten al for severe cavita on
𝛿 < 1.0 The flashing is occurring
Compressible fluids
There are six steps for a control valve size selec on for steam or gases:
1) Specify the variables required to size the valve.
2) Determine the equa on constant (N).
3) Determine the piping geometry factor (FP).
4) Determine expansion factor (ϒ).
5) Solve for required flow coefficient (Cv) value using an appropriate equa on. (Use
Annex B.2 flowcharts)
6) Select the appropriate flow coefficient table and the calculated flow coefficient
valve to select the suitable valve size.

Turbulent flow
Non-choked flow (without fi ngs) (ΔP/P1 < Fϒ.xT )

𝑾 N6 : 27.3
𝐶𝑣 = Eq.6 N7 : 417
𝑵𝟔 𝒀 𝒙𝑷𝟏 𝝆𝟏
N8 : 94.8
N9 : 2120
𝑾 𝑻𝟏 𝒁 W in kg/hr
𝐶𝑣 = Eq.7
𝑵𝟖 𝑷𝟏 𝒀 𝒙𝑴 Q in Nm3/hr
P in bar
𝑸 𝑴𝑻𝟏 𝒁 Density in kg/m3
𝐶𝑣 = Eq.8a T in kelvin
𝑵𝟗 𝑷𝟏 𝒀 𝒙

𝑸 𝑮𝒈𝑻𝟏 𝒁
𝐶𝑣 = Eq.8b
𝑵𝟕 𝑷𝟏 𝒀 𝒙

Non-choked ﬂow (with ﬁ ngs) (ΔP/P1 < Fϒ.xTP )

𝑾
𝐶𝑣 = Eq.9
𝑵𝟔 𝑭𝑷 𝒀 𝒙𝑷𝟏 𝝆𝟏

𝑾 𝑻𝟏 𝒁
𝐶𝑣 = Eq.10
𝑵𝟖 𝑭𝑷 𝑷𝟏 𝒀 𝒙𝑴

𝑸 𝑴𝑻𝟏 𝒁
𝐶𝑣 = Eq.11a
𝑵𝟗 𝑭𝑷 𝑷𝟏 𝒀 𝒙

𝑸 𝑮𝒈𝑻𝟏 𝒁
𝐶𝑣 = Eq.11b
𝑵𝟕 𝑭𝑷 𝑷𝟏 𝒀 𝒙
Choked ﬂow (without ﬁ ngs) (ΔP/P1> Fϒ.xT )

𝑾
𝐶𝑣 = Eq.12
𝟎. 𝟔𝟔𝟕 ∗ 𝑵𝟔 𝐅ϒ 𝒙𝑻 𝑷𝟏 𝝆𝟏

𝑾 𝑻𝟏 𝒁
𝐶𝑣 = Eq.13
𝟎. 𝟔𝟔𝟕 ∗ 𝑵𝟖 𝑷𝟏 𝐅ϒ 𝒙𝑻 𝑴

𝑸 𝑴𝑻𝟏 𝒁
𝐶𝑣 = Eq.14a
𝟎. 𝟔𝟔𝟕 ∗ 𝑵𝟗 𝑷𝟏 𝐅ϒ 𝒙𝑻

𝑸 𝑮𝒈𝑻𝟏 𝒁
𝐶𝑣 = Eq.14b
𝟎. 𝟔𝟔𝟕 ∗ 𝑵𝟕 𝑷𝟏 𝐅ϒ 𝒙𝑻

Choked ﬂow (with ﬁ ngs) (ΔP/P1> Fϒ.xTP)

𝑾
𝐶𝑣 = Eq.15
𝟎. 𝟔𝟔𝟕 ∗ 𝑭𝑷 𝑵𝟔 𝐅ϒ 𝒙𝑻 𝑷𝟏 𝝆𝟏

𝑾 𝑻𝟏 𝒁
𝐶𝑣 = Eq.16
𝟎. 𝟔𝟔𝟕 ∗ 𝑭𝑷 𝑵𝟖 𝑷𝟏 𝐅ϒ 𝒙𝑻 𝑴

𝑸 𝑴𝑻𝟏 𝒁
𝐶𝑣 = Eq.17a
𝟎. 𝟔𝟔𝟕 ∗ 𝑭𝑷 𝑵𝟗 𝑷𝟏 𝐅ϒ 𝒙𝑻

𝑸 𝑮𝒈𝑻𝟏 𝒁
𝐶𝑣 = Eq.17b
𝟎. 𝟔𝟔𝟕 ∗ 𝑭𝑷 𝑵𝟕 𝑷𝟏 𝐅ϒ 𝒙𝑻

Non-Turbulent ﬂow (without ﬁ ngs)

𝑾 𝑻𝟏 Eq.18
𝐶𝑣 = N27: 0.67
𝑵𝟐𝟕 𝑭𝑹 𝚫𝐏(𝑷𝟏 + 𝑷𝟐 )𝑴
N22: 1500
𝑸 𝑴𝑻𝟏
𝐶𝑣 = Eq.19
𝑵𝟐𝟐 𝑭𝑹 𝚫𝐏(𝑷𝟏 + 𝑷𝟐 )
Specify the Variables Required to Size the Valve [5] a) Solve for the Required Flow Coefficient (Cv) (valve diameter known)
• Process fluid (e.g. oil and water); Flow coefficient value is different from volumetric flow rate than
• Service condi ons and fluid physical proper es such as volume rate mass flow rate units. When the gas-specific gravity, Gg, is known for
of flow (q), mass rate of flow (w), Upstream and downstream volumetric flow rate units,
pressure values (P1 and P2), pressure drop across 1. the flow coefficient is calculated as per Eq. (15 & 16).
the valve (ΔP = P1 − P2), upstream temperature (T1), gas-specific 2. If the molecular weight, M, is known for volumetric flow rate
gravity that is defined as a dimensionless number equal to the ra o units, the flow coefficient is calculated as per Eq. (17a).
of the flowing gas density to the density of air with both at standard 3. If the specific weight at the valve inlet, γ, is known for mass flow
condi ons (Gg), molecular weight (M), the ra o of specific heats (k), rate units, the flow coefficient is calculated as per Eq. (17b).
gas compressibility factor (Z), viscosity (µ) and solid% . 4. If the molecular weight, M, is known for mass flow rate units, the
• Constants FL, Fd and xT: A er selec ng required Valve type, trim flow coefficient is calculated as per Eq. (16).
type and flow direc on, these valves can be es mated using Table 2. 1. Select xT from Table 2
2. Calculate Fγ from Eq. 34
3. Calculate FγxT and determine if flow is choked or not. (x <
Determine Piping Geometry Factor (FP) FγxT)
FP is a correc on factor that accounts for the pressure losses due to 4. If flow is choked Y = 0.667 if not Calculate Y.
piping fi ngs that are used to change the direc on or size of the 5. Calculate Ci by Eq. 6,7,8 (non-choked) or Eq. 12,13,14
piping or take branches from the headers such as tees, elbows, and (Choked)
reducers. If fi ngs are used before or a er the valve, the effect of 6. Calculate Rev using Eq. 28, Obtain Fd from Table 2.
piping geometry must be considered for the valve sizing. However, if 7. If Rev > 10000, flow is turbulent and If valve size = pipe size
no fi ng is a ached to the valve, FP is equal to one and does not Use Ci previously Calculated. If not, Calculate Fp and
impact the valve sizing process. Equa on (36) shows how to calculate subsequently its Cv using the iteration method described in
the piping geometry. flowchart.
8. If Rev < 10000, flow is non-turbulent. Calculate FR and its
Determine the Expansion Factor (Y) subsequently its Cv using the iteration method described in
When sizing components such as valves in the piping system handling flowchart.
compressible gas, it is essen al to account for all factors, including
the expansion factor that would affect the flow rate and pressure Select the appropriate flow coefficient table and the calculated flow
drop in the piping system. As compressible gas flows through each coefficient valve to select the suitable valve size.
component in the piping system, the head loss causes expansion as Use fisher tables (or any representa ve tables) to select a valve with
well as change in density, temperature, and velocity. The gas Cv larger than your calculated Cv. Select the desired flow
expansion process adds resistance to the flow rate, resul ng in a flow Characteris c and calculate the es mated opening for the chosen
rate reduc on for a given pressure drop. Thus, the gas expansion valve.
factor, Y, shall be taken into account.
𝑥 b) An alterna ve op on for valve sizing (valve diameter isn’t known):
𝑌 = 1−
3Fϒ 𝑥 Use the simplified equa ons described earlier then use the sizing
x: The pressure drop ra o= ΔP/P1; If x < 0.2, the change in gas density equa on in the following sec on (Another method for es ma ng
and expansion is small, the fluid can be assumed incompressible, and control valve size (if tables not available))
the parameter Y can be neglected. For x > 0.4, the expansion factor Check Noise levels, velocity and Mach Number
shall be considered as the fluid can be assumed compressible.
xT: The pressure drop ra o factor for valves installed without
a ached fi ngs can be extracted from Table 2. More precisely, xT is
the pressure drop ra o required to produce cri cal or maximum flow
through the valve when Fγ = 1. It should be noted that the value of Y
cannot be less than 0.667. When the calculated Y value is less than
0.667, the Y value shall be assumed to be 0.667.
Suppose the control valve is installed with fi ngs such as reducers or
elbows a ached to it. In that case, the effect of fi ngs shall be taken
into account by replacing XT by a factor of XTP that can be calculated
from Eq. (33) as follows.
𝑥
𝐹
𝑥 = Eq.33
𝑥 ∑𝑘 𝐶
1+ 𝑁 ( )
𝑑

N5: 0.00241, d in mm
Determina on of correc on factors[6]

Fp The piping geometry factor

∑𝐾 = 𝐾 + 𝐾 +𝐾 −𝐾 Eq. 21 For an inlet reducer

Where, 𝐾 = 0.5(1 − ) Eq. 23
K1 = Resistance coefficient of upstream
K2 = Resistance coefficient of downstream For an outlet reducer
fittings 𝐾 = 1.0 ∗ (1 − ) Eq. 24
KB1 = Inlet Bernoulli coefficient For a valve installed between identical reducers
KB2 = outlet Bernoulli coefficient (assumed in tool)
𝐾 =1− Eq. 22 𝐾 + 𝐾 = 1.5 ∗ (1 − ) Eq. 25

𝐾 =1− Eq. 22
N1 : 0.865
N2: 0.00214
Q in m3/hr
P in bar
∑ / Density in kg/m3
𝐹 = 1+ ( ) Eq. 36
ρ in kg/m3
ρ0 : 1000 kg/m3
D and D in mm

If Rev > 10,000 and Valve Size doesn’t equal pipe size:
Assume Fp =1 and Cv = Ci (your calculated Cv previously) and re-calculate Fp, check whether
flow is choked or not and re-calculate Cv accordingly, if Ci/C > 0.99 you can use your
calculated Cv. If not, re-assume Ci to be your calculated Cv and Fp to be your re-calculated Fp.
Reynolds number factor FR 1. For Transi onal flow
The Reynolds number factor FR is required when .
0.33𝐹 𝑅𝑒
non-turbulent flow condi ons are established 𝐹 =1+( ) log ( )
through a control valve because of a low 𝑛 . 10000
pressure differen al, a high viscosity, a very Where:
small flow coefficient, or a combina on thereof.
The FR factor is determined by dividing the flow 𝑁
𝑛 =
rate when non-turbulent flow condi ons exist by 𝐶
( )
the flow rate measured in the same installa on 𝑑
under turbulent condi ons. 2. For laminar flow
Tests show that FR can be determined from the
.
curves given in Figure 3a or 3b using a valve 𝐹 = 𝑛 𝑅𝑒 (not to exceed FR =1)
Reynolds number calculated from the following
equa on: For reduced trim valves where Ci /d2 at rated
travel is less than 0.016 N18 and Rev ≥ 10,
𝑁𝐹 𝑄 𝐹 𝐶 calculate F from the following equa ons:
𝑅𝑒 = ( + 1) .
Eq.28
𝑣 𝐶𝐹 𝑁 𝐷 1. For Transi onal flow
.
0.33𝐹 𝑅𝑒
This calcula on will require itera on. Proceed by 𝐹 =1+( ) log ( )
𝑛 . 10000
calcula ng the flow coefficient C for turbulent
flow. The valve style modifier Fd converts the n2= 1 + N32 ( ) /

geometry of the oriﬁce(s) to an equivalent

circular single ﬂow passage. See Table 2 for 2. For laminar ﬂow
typical values and Annex A in ANSI/ISA 75.01.01 .
for details. 𝐹 = 𝑛 𝑅𝑒 (not to exceed FR =1)

Next, establish Ci as per Equa on 26. Liquid pressure recovery factor without
Ci = 1.3C (Equa on 26) a ached fi ngs FL
Apply Ci as per Equa on 26 and determine FR. FR FL is the liquid pressure recovery factor of the
is determined from Figure 3a for full-size trim valve without a ached fi ngs. This factor
valves. accounts for the influence of the valve internal
FR is determined from Figure 3b for reduced trim geometry on the valve capacity at choked flow. It
valves where Ci /d 2 at rated travel is less than is defined as the ra o of the actual maximum
0.016 N18. flow rate under choked flow condi ons to a
theore cal, non-choked flow rate which would
C/FR ≤ Ci Eq. 29 be calculated if the pressure differen al used
If the condi on of Equa on 29 is sa sfied, then was the difference between the valve inlet
use the Ci established from Equa on 26. If the pressure and the apparent vena contracta
condi on of Equa on 29 is not met, then repeat pressure at choked flow condi ons. The factor FL
the above equa on by again increasing Ci by 30 may be determined from tests in accordance
percent. This may require several itera ons un l with ANSI/ISA-75.02-1996. Typical values of FL
the condi ons required in Equa on 29 are met. versus percent of rated flow coefficient are
The equa ons defining non-turbulent flow for shown in Figure 4a/b and Table 2.
full size and reduced trim valves are stated in
Combined liquid pressure recovery factor
Annex G in ANSI/ISA 75.01.01. (Summarized
below, Also check flowchart below) and piping geometry factor with a ached
fi ngs FLP
For full size trim where Ci /d2 ≥ 0.016 N18
FLP is the combined liquid pressure recovery
and Rev ≥ 10, calculate FR from the following
factor and piping geometry factor for a control
equa ons:
valve with a ached fi ngs. It is obtained in the d) Reynolds number; and
same manner as FL. e) specific heat ra o .
To meet a devia on of ±5% for FLP, FLP shall be The influence of items a), b), c), and e) is
determined by tes ng. When es mated values accounted for by the pressure differen al ra o
are permissible, the following equa on shall be factor xT.
used: The Reynolds number is the ra o of iner al to
𝐹 viscous forces at the control valve orifice. In the
𝐹 = case of compressible flow, its value is generally
𝐹 𝐶 Eq.30 beyond the range of influence, except where the
1+ 𝐾( ) flow rate or the Cv is very low or a combina on
𝑁 𝑑
of both exist .
The pressure differen al ra o xT is influenced by
the specific heat ra o of the fluid.
Liquid cri cal pressure ra o factor FF
FF is the liquid cri cal pressure ra o factor. This 𝑥
factor is the ra o of the apparent vena contracta
𝑌 =1−
3F 𝑥 Eq.32
pressure at choked flow condi ons to the vapor
pressure of the liquid at inlet temperature. At
vapor pressures near zero, this factor is 0.96. The value of x for calcula on purposes shall not
Values of FF may be determined from the curve exceed Fϒ.xT. If x > Fϒ.xT, then the flow becomes
in Figure 5 or approximated from the following choked and Y = 0.667.
equa on:
Pressure differen al ra o factor without fi ngs
xT
𝑃
𝐹 = 0.96 − 0.28 Eq.31 xT is the pressure differen al ra o factor of a
𝑃
control valve installed without reducers or other
fi ngs. If the inlet pressure P1 is held constant
and the outlet pressure P2 is progressively
Specific heat ra o factor Fϒ lowered, the mass flow rate through a valve will
The factor xT is based on air near atmospheric increase to a maximum limit, a condi on
pressure as the flowing fluid with a specific heat referred to as choked flow. Further reduc ons in
ra o of 1.40. If the specific heat ra o for the P2 will produce no further increase in flow rate.
flowing fluid is not 1.40, the factor Fis used to This limit is reached when the pressure
adjust xT. Use the following equa on to calculate differen al x reaches a value of Fϒ.xT. The
the specific heat ra o factor: limi ng value of x is defined as the cri cal
𝛾 differen al pressure ra o. The value of x used in
Fϒ = Eq.34 any of the sizing equa ons and in the
1.4
rela onship for Y (Equa on 32) shall be held to
this limit even though the actual pressure
Expansion factor Y differen al ra o is greater. Thus, the numerical
value of Y may range from 0.667, when x = Fϒ.xT,
The expansion factor Y accounts for the change
to 1.0 for very low differen al pressures.
in density as the fluid passes from the valve inlet
to the vena contracta (the loca on just
Representa ve values of xT for several types of
downstream of the orifice where the jet stream
control valves with full size trim and at full rated
area is a minimum). It also accounts for the openings are given in Table 2. Cau on should be
change in the vena contracta area as the exercised in the use of this informa on. When
pressure differen al is varied. precise values are required, they should be obtained
Theore cally, Y is affected by all of the following: by test.
a) ra o of port area to body inlet area;
b) shape of the flow path;
c) pressure differen al ra o x ;
Pressure differen al ra o factor with
a ached fi ngs xTP

If a control valve is installed with a ached

fi ngs, the value of xT will be affected.
To meet a devia on of ±5% for xTP, the valve and
a ached fi ngs shall be tested as a unit. When
es mated values are permissible, the following
equa on shall be used:
𝑥
𝐹
𝑥 = Eq.33
𝑥 ∑𝑘 𝐶
1+ ( )
𝑁 𝑑

In the above rela onship, xT is the pressure

differen al ra o factor for a control valve
installed without reducers or other fi ngs. K is
the sum of the inlet velocity head loss
coefficients (K1 + KB1) of the reducer or other
fi ng a ached to the inlet face of the valve.
If the inlet fi ng is a short-length, commercially
available reducer, the value of K1 may be
es mated using Equa on 23.

Figure: FR Iteration Calculations

N4 : 0.076
N32 : 127
N5 : 0.00241
D or d in mm
Q in m3/hr
Kinema c viscosity
(𝑣) in m2/s
Control Valve Sizing flow charts

Figure: Incompressible ﬂuids control Valve sizing ﬂowchart

Figure: Incompressible fluids control Valve sizing flowchart
Figure: Compressible fluids control Valve sizing flowchart
Figure: compressible fluids control Valve sizing flowchart
3. Selecting Valve Size and calculating the opening percentage
Another method for es ma ng control valve size (if tables not available) [2]
A er es ma ng valve Cv, valve sizing can be es mated using the following equa ons:
𝑣𝑎𝑙𝑣𝑒 𝐶 Single-seated valves
𝑆 (𝑖𝑛𝑐ℎ𝑒𝑠) = ( )
9
𝑣𝑎𝑙𝑣𝑒 𝐶 Double-seated valves
𝑆 (𝑖𝑛𝑐ℎ𝑒𝑠) = ( )
12
𝑣𝑎𝑙𝑣𝑒 𝐶 Butterfly valve sizes
𝑆 (𝑖𝑛𝑐ℎ𝑒𝑠) = ( )
20
[4]
Li Ra o (Opening percentage)

Li ra o of the control valve is preferred to be between 50% - 70% for normal ﬂow condi ons.
(50% ≤ Li % ≤ 70%)

Flow Characteris c Cv ra o () Li Ra o ()

Linear 1 1 1 1
 = 1− +  = ( − )/(1 − )
   
Equal percentage  =  log 
= +1
log 
Modiﬁed parabolic  = 𝑐.  log /𝑐
=
log 
Quick opening 1 1 1 1
 = 1− √ +  = ( − )/(1 − )
   
Bu erﬂy 1 1 1 1 /
 = 1−  +  = ( − )/(1 − )
   
: Cv Ra o ( Cv/Cmax)
: Li Ra o (L/Lmax)
: Rangeability ( Cmax/Cmin)
Typical Vendor Recommenda ons for Percent Travel Versus Flow [9]
Flow Condi on Percent of Rated Travel
Minimum 10
Normal 20-80
Maximum 90

Guidelines for Percent Travel at Various Flow Condi ons[9]

Flow Characteris c Percent Travel at Normal Flow Percent Travel at Maximum Flow
Equal Percentage 85 93
Linear 75 90
Modified Parabolic 80 90
4. General guidelines when selecting a flow characteristic and control valve type

Selec ng The required ﬂow characteris c[4]

(1) The flow characteristic of the control valve will be “Linear”, except for the following
items (2) or (3).
Also ‘Linear’ could be selected when the change in valve opening causes a maximum of
10% change in ΔP
(2) Equal percentage will be applied for the following services:
- For temperature control
- For the services where the pressure drop across control valve or flow rate varies
significantly
- For service with relatively small ϒo
(3) Quick opening (denoted on-off) will be applied for the following services:
- Self-actuated pressure regulator.
- On-off control service.
- For service with relatively large friction, such as product rundown line.

Type of control Valves[4]

(1) General features and typical application for control valves are summarized as follows:

Type General Application

Globe body valve Flow characteristic: Optional Most common, provided in the middle of
Rangeability (effective): 10 straight line.

- 1 ½ “ and smaller: single seat valve

- 2” and larger: Double seat valve

Single seat valve can be used for tight shut-off

services

Angle body valve Flow characteristic: Optional Application to flashing, erosive, high AP, or
Rangeability (effective): 10 slurry services
Available single seat only
Solid contained less than 0.05 kg/cm3
Butterfly valve Flow characteristic: Optional Applicable to high flow rate and low pressure
Rangeability (effective): 15 drop services
Sampled and economical, use for 4” and larger
Gas service with low pressure drop
High viscous, slurry service
Max. opening is usually limited to 60 degrees
for throttling
Don’t use with opening of 10% or less as a
control valve.
Ball valve Flow characteristic: Essentially EQ% Applicable to high flow rate and high shut-off
Rangeability (effective): 50 pressure services
Low resistance at full-open
Suitable as a shut-off valve
Solid contained service
High rangeability
Eccentric plug- Flow characteristic: Globe/Ball Advantages over butterfly valves
valve (Camflex) Rangeability (effective): 100

Three-way valve Flow characteristic: - Splitting into lines or mixing of streams.

Rangeability (effective): 10 For splitting flow: double seat valve
For mixing flows: single seat valve
Economical rather than two 2-way valves
Applicable for 4” and smaller
Saunders valve Flow characteristic: Special Corrosive, Slurry, high viscous service
Rangeability (effective): 10 Bodies can be fully lines with corrosion
resistant materials
Self-cleaning ability
Suitable as a shut-off valve
Examples from ANSI/ISA 75.01.01
Example no. 1 : Incompressible flow (water) non-choked, turbulent without fi ngs
Process data:
Fluid: water
Inlet temperature: T1 = 363 oK
Density: ρ1 = 965.4 kg/m3
Vapor pressure: Pv = 70.1 kPa
Thermodynamic cri cal pressure: Pc = 22,120 kPa
Kinema c viscosity: ν = 3.26 × 10–7 m2/s
Inlet absolute pressure: P1 = 680 kPa (6.93407 kg/cm2.a)
Outlet absolute pressure: P2 = 220 kPa (2.24338 kg/cm2.a)
Flow rate: Q = 360 m3/h
Pipe size: D1 = D2 = 150 mm
Valve data:
Valve style: globe
Trim: parabolic plug
Flow direc on: flow-to-open
Valve size: d = 150 mm
Liquid pressure recovery factor: FL = 0.90 (from Table 2)
Valve style modifier: Fd = 0.46 (from Table 2)

Calcula ons

𝑃 70.1
𝐹 = 0.96 − 0.28 = 0.96 − 0.28 = 0.944
𝑃 22120
∆Pmax = 𝐹𝐿 (𝑃 − 𝐹 𝑃 ) = 497.2 𝑘𝑃𝑎
Which is more than the differen al pressure (∆P = 460 kPa); therefore, the flow is non-choked,
and the flow coefficient C is calculated using Equa on 1.

𝑄 𝜌 ⁄𝜌
𝐶𝑣 = = 165
𝑁 ∆𝑃
Where Q = 360 m3 / h; N1 = 1 × 10–1 from Table 1; ρ1/ρo = 0.965; and ∆P = 460 kPa. Cv is in Kv’s
units.
Calculate Rev
𝑁𝐹 𝑄 𝐹 𝐶
𝑅𝑒 = ( + 1) . = 2.967 ∗ 10
𝑣 𝐶𝐹 𝑁 𝐷
N2 = 1.60 × 10–3 from Table 1; N4 = 7.07 × 10–2 from Table 1;
Fd = 0.46; Q = 360 m3 /h; ν = 3.26 × 10–7 m 2 /s; Ci = C = Kv = 165 m3/ h;
FL = 0.90; and D = 150 mm.
Since the valve Reynolds number is greater than 10,000, the flow is turbulent, and the flow
coefficient C as calculated above is correct.
Example no. 2 : Compressible flow – non-choked flow with a ached fi ngs
Process data:
Fluid: carbon dioxide
Inlet temperature: T1 = 433 K
Molecular mass: M = 44.01 kg/kmol
Kinema c viscosity: ν = 1.743 × 10–5 m 2 /s
Specific heat ra o: γ = 1.30
Compressibility factor: Z = 0.988
Inlet absolute pressure: P1 = 680 kPa
Outlet absolute pressure: P2 = 310 kPa
Flow rate: Q = 3,800 standard m3 /h at 101.325 kPa and 0°C
Inlet pipe size: D1 = 80 mm
Outlet pipe size: D2 = 100 mm
Reducers: short length, concentric
Valve data:
Valve style: rotary
Trim: eccentric rotary plug
Flow direc on: flow-to-open Valve size: d = 50 mm
Pressure differen al ra o factor: xT = 0.60 (from Table 2)
Liquid pressure recovery factor: FL = 0.85 (from Table 2)
Valve style modifier: Fd = 0.42 (from Table 2)

Calcula ons:
ϒ 1.3
Fϒ = = = 0.929
1.4 1.4
𝛥𝑃 680 − 310
𝑥= = = 0.544
𝑃 680
Which is less than Fϒ.xT = 0.557; therefore, the flow is non-choked and the flow coefficient is
calculated from Equa on 11. Next, Y is calculated from Equa on 32;
.
𝑌 =1− =1− = 0.674
ϒ ∗ . ∗ .
. ∗ ∗ .
𝐶𝑣 = = = 62.7 𝑓𝑜𝑟 𝑘
. ∗ ∗ ∗ . .
Calculate Rev
𝑁𝐹 𝑄 𝐹 𝐶 0.0707 ∗ 0.42 ∗ 3800 0.85 62.7
𝑅𝑒 = ( + 1) . = ( + 1) . = 8.96 ∗ 10
𝑣 𝐶𝐹 𝑁 𝐷 1.743 ∗ 10 √62.7 ∗ 0.85 80
Since the valve Reynolds number is greater than 10,000, the flow is turbulent. Now, calculate the
effect of the inlet and outlet reducers on C.
Since both reducers are concentric, short length, the velocity head loss coefficients can be
calculated as follows:
𝑑
𝐾 = 0.5(1 − ) = 0.186
𝐷
𝐾 =1− = 0.847

𝐾 =1− = 0.938
∑ 𝐾 = 𝐾 + 𝐾 + 𝐾 − 𝐾 =0.658
Now, the eﬀect of the reducers is calculated by itera on, star ng with Ci = C and FP(1) = 1.

/ /
∑𝐾 𝐶 0.658 62.7
𝐹 = 1+ ( ) = 1+ ( ) = 0.891
𝑁 𝑑 1.6 ∗ 10 50
Since FP(2) /FP(1) = 0.891/1 < 0.99, one more itera on step shall be done
𝐶 62.7 m
𝐶 = = = 70.4
𝐹 ( ) 0.891 h
/ /
∑𝐾 𝐶 0.658 70.4
𝐹 ( ) = 1+ ( ) = 1+ ( ) = 0.868
𝑁 𝑑 1.6 ∗ 10 50
Now, FP(3)/FP(2) = 0.868/0.891 > 0.99 so FP(3) will be used as FP for the ﬁnal calcula on.
𝑥 0.6
𝐹 0.868
𝑥 = = = 0.626
𝑥 𝐶 0.6 70.4
1+ ( ) 1+ ( )
𝑁 𝑑 1.8 ∗ 10 50
and with this Fϒ xTP = 0.582, which is greater than x = 0.544.

Finally, C results from Equa on 11 as follows:

𝑄 𝑀𝑇 𝑍 3800 44.01 ∗ 433 ∗ 0.988 m3

𝐶𝑣 = = = 72.2 𝑓𝑜𝑟 𝑘
𝑁𝐹 𝑃𝑌 𝑥 24.6 ∗ .868 ∗ 680 ∗ 0.674 0.544 h

Solving the case sing the tool (Check screenshots below)

Input steps

1. Select “Calculate Cv (D,Q,△P Known)” Case

2. Select “Gas or Vapor”
3. Select D1,d and D2 in inches
4. Input Normal flow rate (Nm3/hr) (flow at 1 atm and 0 oC)
5. Either input your process data manually or obtain it by composition and flow
conditions as indicated below (You can still edit the obtained physical result)
6. You can Select “Modify assumptions” to modify the assumed Valve type/trim type and
flow direction as indicated below
7. Click on “Calculate”
NOTE: for this example, physical proper es were modified in the input table to the physical
proper es men oned in the example, actual physical proper es obtained by tool yielded a Kv
of 72.6 but with choked flow (try it yourself!)
Supplementary material (fisher 2nd and 4th edi on represesnta ve tables)

VALVE_SIZE(IN.) Valve_plug_type FLOW_CHARACTERISTIC PORT_DIA.(IN.) RATED_TRAVEL(IN.) Cv Cg Cs Km C1

0.5 1-Flute Micro -Flute ™ Equal Percentage 0.25 0.75 0.31 10 0.5 0.7 32.3
0.5 2-Flute Micro -Flute ™ Equal Percentage 0.25 0.75 0.63 20 1 0.7 31.7
0.5 3-Flute Micro -Flute ™ Equal Percentage 0.25 0.75 0.95 32 1.6 0.75 33.7
0.5 Micro -Form ™ Equal Percentage 0.25 0.75 1.4 48 2.4 0.75 34.3
0.5 Micro -Form Equal Percentage 0.375 0.75 2.5 87 4.4 0.8 34.8
0.5 Micro -Form Equal Percentage 0.5 0.75 2.89 99.5 4.98 0.89 34.4
0.5 Top -Guided Equal Percentage 0.5 0.75 4.2 145 7.25 0.8 34.5
0.5 Port -Guided Quick Opening 0.75 0.4375 4.06 147 7.35 0.85 36.2
0.5 Top -Guided Quick Opening 0.75 0.75 6.27 227 11.4 0.8 36.2
0.75 Micro -Flute Equal Percentage 0.75 0.75 6.26 216 10.8 0.87 34.5
0.75 Top -Guided Equal Percentage 0.75 0.75 8.65 304 15.2 0.8 35.1
0.75 Top -Guided Quick Opening 0.75 0.75 9.08 322 16.1 0.8 35.5
1 Micro -Flute Equal Percentage 1 0.75 9.39 355 17.8 0.83 37.8
1 Micro -Form Equal Percentage 1 0.75 12.2 467 23.3 0.91 38.2
1 Cage -Guided Equal Percentage 1.3125 0.75 17.4 591 29.6 0.9 34
1 Cage -Guided Linear 1.3125 0.75 20.1 668 33.4 0.8 33.2
1 Cage -Guided Quick Opening 1.3125 0.75 21.4 690 34.5 0.79 32.2
1.25 Top -Guided Equal Percentage 1.25 0.75 22.2 785 39.3 0.8 35.4
1.25 Top -Guided Quick Opening 1.25 0.75 23 805 40.2 0.8 35
1.5 Micro -Form Equal Percentage 1 0.75 17.4 623 31.2 0.85 35.8
1.5 Top -Guided Equal Percentage 1.5 0.75 29.1 1000 50.2 0.75 34.5
1.5 Top -Guided Quick Opening 1.5 0.75 30 1070 53.4 0.75 35.6
1.5 Cage -Guided Equal Percentage 1.875 0.75 33.4 1190 59.5 0.89 35.6
1.5 Cage -Guided Linear 1.875 0.75 34.9 1220 61 0.85 35
1.5 Cage -Guided Quick Opening 1.875 0.75 38 1350 67.5 0.88 35.6
2 Cage -Guided Equal Percentage 1.3125 1.125 56.2 2070 104 0.85 34.9
2 Cage -Guided Linear 1.3125 1.125 65.3 2280 114 0.83 36.8
2 Cage -Guided Quick Opening 1.3125 1.125 67.2 2420 121 0.86 36
2.5 Cage-Guided Equal Percentage 2.875 1.5 87.2 3100 155 0.87 37.4
2.5 Cage-Guided Linear 2.875 1.5 86.5 3220 161 0.86 37.2
2.5 Cage-Guided Quick Opening 2.875 1.5 93.1 3470 174 0.83 37.2
3 Cage-Guided Equal Percentage 3.1875 1.5 121 4210 211 0.79 34.8
3 Cage-Guided Linear 3.1875 1.5 135 4680 234 0.8 34.7
3 Cage-Guided Quick Opening 3.1875 1.5 150 5090 255 0.76 33.9
4 Cage-Guided Equal Percentage 4.375 2 203 7360 368 0.82 36.2
4 Cage-Guided Linear 4.375 2 212 7540 377 0.79 35.6
4 Cage-Guided Quick Opening 4.375 2 235 8300 415 0.79 35.4
6 Cage-Guided Equal Percentage 7 2 357 12900 645 0.74 36.1
6 Cage-Guided Linear 7 2 417 14400 720 0.66 34.5
6 Cage-Guided Quick Opening 7 2 469 15700 785 0.67 33.5
8 Port-Guided V-Pup Equal Percentage 8 2 587 20400 1020 0.8 34.8
8 Top & Bottom Guided Throttle Modified Parabolic 8 2 602 20100 1010 0.75 33.4
8 Cage-Guided Equal Percentage 8 2 570 19000 950 0.72 33.3
8 Cage-Guided Linear 8 2 701 24400 1220 0.7 34.8
8 Cage-Guided Quick Opening 8 2 787 28400 1420 0.73 36.1
8 Cage-Guided Equal Percentage 8 3 808 29400 1470 0.72 36.4
8 Cage-Guided Linear 8 3 836 29900 1500 0.72 35.8
8 Cage-Guided Quick Opening 8 3 875 30800 1540 0.72 35.2
10 Port-Guided V-Port Modified Parabolic 10 3 987 33400 1670 0.75 33.8
10 Port-Guided Quick Opening 10 3 1280 43800 2190 0.75 34.2
10 Top & Bottom-Guided Quick Opening 10 3 1360 46500 2330 0.75 34.2
12 Port-Guided V-Port Modified Parabolic 12 3 1180 40900 2050 0.8 34.7
12 Top & Bottom-Guided Quick Opening 12 3 1660 55300 2770 0.75 33.3
12 Port-Guided Quick Opening 12 3 1700 56600 2830 0.75 33.3
VALVE_SIZE(IN.) Valve_plug_type RATED_VALVE_OPENING Cv Cg Cs Km C1 Kc
1 Rotary -V -Notch Ball 90 22.3 684 34.2 0.74 30.7 0.26
1.5 Butterfly Conventional Disc 60 26.3 662 33.1 0.55 25.2 0.35
1.5 Butterfly Conventional Disc 90 52 907 45.3 0.3 17.2 0.25
1.5 Rotary -V -Notch Ball 90 88.9 2050 103 0.55 23.1 0.26
2 Rotary -Eccentric Disc 90 75 1810 90.6 0.48 24.1 0.26
2 Butterfly -Conventional Disc 60 55.1 1390 69.4 0.55 25.2 0.35
2 Butterfly -Conventional Disc 90 110 1900 95 0.3 17.2 0.25
2 Butterfly -Fishtail Disc 90 91 1650 82.3 0.43 18.1 0.31
2 Rotary -V -Notch Ball 90 158 3530 177 0.46 22.3 0.26
2.5 Butterfly -Conventional Disc 60 91.4 2300 115 .55 25.2 0.35
2.5 Butterfly -Conventional Disc 90 182 3150 157 .30 17.2 0.25
2.5 Butterfly -Fishtail Disc 90 154 2790 139 .43 18.1 0.31
3 Rotary -Eccentric Disc 90 193 4310 216 .42 22.3 0.26
3 Butterfly -Conventional Disc 60 136 3440 172 .55 25.2 0.35
3 Butterfly -Conventional Disc 90 273 4710 235 .30 17.2 0.25
3 Butterfly -Fishtail Disc 90 234 4230 211 .43 18.1 0.31
3 Rotary -V -Notch Ball 90 372 7430 372 .46 20 0.26
4 Rotary -Eccentric Disc 90 418 8640 432 .36 20.7 0.26
4 Butterfly -Conventional Disc 60 271 6820 341 .55 25.2 0.35
4 Butterfly -Conventional Disc 90 542 9340 467 .30 17.2 0.25
4 Butterfly Fishtail Disc 90 490 8870 443 .43 18.1 0.31
4 Rotary -V -Notch Ball 90 575 10800 540 .37 18.8 0.26
6 Rotary Eccentric Disc 90 900 17100 855 .33 19 0.26
6 Butterfly Conventional Disc 60 768 18800 941 .55 24.5 0.35
6 Butterfly Conventional Disc 90 1750 25000 1250 0.3 14.3 0.25
6 Butterfly -Fishtail Disc 90 1410 22500 1120 0.43 16 0.31
6 Rotary -V -Notch Ball 90 944 18100 905 0.31 19.2 0.26
8 Rotary -Eccentric Disc 90 1920 35300 1770 .31 18.4 26
8 Butterfly Conventional Disc 60 1340 32900 1640 .55 24.5 35
8 Butterfly Conventional Disc 90 3050 43600 2180 .30 14.3 25
8 Butterfly -Fishtail Disc 90 2440 39100 1950 .43 16 0.31
8 Rotary -V -Notch Ball 90 1770 28900 1450 .28 16.3 26
10 Rotary -Eccentric Disc 90 3390 60000 3000 .28 17.7 26
10 Butterfly Conventional Disc 60 2170 53100 2650 .55 24.5 0.35
10 Butterfly Conventional Disc 90 4920 70400 3520 .30 14.3 0.25
10 Butterfly -Fishtail Disc 90 4010 64200 3210 .43 16 31
10 Rotary -V -Notch Ball 90 3160 56900 2850 .33 18 26
12 Rotary -Eccentric Disc 90 4890 84100 4200 .27 17.2 0.26
12 Butterfly Conventional Disc 60 3180 78000 3900 .55 24.5 0.35
12 Butterfly -Conventional Disc 90 7240 103000 5180 .30 14.3 0.25
12 Butterfly Fishtail Disc 90 5990 95800 4790 .43 16 0.31
12 Rotary -V -Notch Ball 90 4620 83200 4160 .33 18 0.26
14 Rotary -Eccentric Disc 90 6800 117000 5860 .27 17.2 0.26
14 Butterfly Conventional Disc 60 3880 95000 4750 0.55 24.5 0.35
14 Butterfly -Conventional Disc 90 8810 126000 6300 0.3 14.3 0.25
14 Butterfly -Fishtail Disc 90 7280 116000 5820 .43 16 0.31
16 Rotary -Eccentric Disc 90 9470 163000 8150 .27 17.2 0.26
16 Butterfly -Conventional Disc 60 5210 127000 6380 .55 24.5 0.35
16 Butterfly -Conventional Disc 90 11800 169000 8460 .30 14.3 0.25
16 Butterfly -Fishtail Disc 90 9830 157000 7860 .43 16 0.31
16 Rotary -V -Notch Ball 90 67 125000 6250 .33 18.5 0.26
18 Rotary Eccentric Disc 90 12000 206000 10300 .27 17.2 0.26
18 Butterfly Conventional Disc 60 6510 159000 7980 :55 24.5 0.35
18 Butterfly -Conventional Disc 90° 14800 212000 10600 .30 14.3 0.25
18 Butterfly -Fishtail Disc 90 12300 197000 9850 .43 16 0.31
20 Rotary -Eccentric Disc 90 14400 248000 12400 0.27 17.2 0.26
20 Butterfly Conventional Disc 60 8210 201000 10000 0.55 24.5 0.35
20 Butterfly Conventional Disc 90 18600 267000 13300 0.3 14.3 0.25
20 Butterfly -Fishtail Disc 90 15500 248000 12400 0.43 16 0.31
24 Rotary -Eccentric Disc 90 21600 372000 18600 0.27 17.2 0.26
24 Butterfly Conventional Disc 60 12200 298000 14900 0.55 24.5 0.35
24 Butterfly -Conventional Disc 90 27700 396000 19800 0.3 14.3 0.25
24 Butterfly Fishtail 90 23100 370000 18500 0.43 16 0.31
24 Rotary - V Notch Ball 90 13700 248000 12400 0.33 18.1 0.26
30 Butterfly -Conventional Disc 60 19900 487000 24300 0.55 24.5 0.35
30 Butterfly -Conventional Disc 90 45200 646000 32300 0.3 14.3 0.25
30 Butterfly -Fishtail Disc 90 40300 646000 32300 0.43 16 0.31
36 Butterfly -Conventional Disc 60 29400 721000 36100 0.55 24.5 0.35
36 Butterfly -Conventional Disc 90 66900 957000 47800 0.3 14.3 0.25
36 Butterfly -Fishtail Disc 90 61600 985000 49200 0.43 16 0.31
36 Butterfly Conventional Disc 60 54200 1330000 66400 .55 24.5 0.35
48 Butterfly Conventional Disc 90 123000 1760000 88100 .30 14.3 0.25
48 Butterfly Fishtail Disc 90 116000 1850000 92400 .43 16 0.31
48 Butterfly -Conventional Disc 60 85300 2090000 104000 .55 24.5 0.35
60 60Butterfly -Conventional Disc 90 194000 2770000 139000 .30 14.3 0.25
60 Butterfly -Fishtail Disc 90 182000 2910000 145000 .43 16 0.31
60 Butterfly Conventional Disc 60 123000 3030000 151000 .55 24.5 0.35
72 Butterfly Conventional Disc 90 281000 4010000 201000 .30 14.3 0.25
72 Butterfly -Fishtail Disc 90 262000 4200000 210000.00 .43 16 0.31
Valve Valve Plug Flow Port Dia. Rated Cv Fl Xt Fd
Size Style Characteristic (in.) Travel
(NPS) (in.)
0.5 Post-Gulded Equal-Percentage 0.38 0.5 2.41 0.9 0.54 0.61
0.75 Post-Gulded Equal-Percentage 0.56 0.5 5.92 0.84 0.61 0.61
1 Cage-Guided Linear 1.125 0.75 20.6 0.84 0.64 0.34
1 Cage-Guided Equal-Percentage 1.125 0.75 17.2 0.88 0.67 0.38
1.5 Cage-Guided Linear 1.875 0.75 39.2 0.82 0.66 0.34
1.5 Cage-Guided Equal-Percentage 1.875 0.75 35.8 0.84 0.68 0.38
2 Cage-Guided Linear 2.3125 1.125 72.9 0.77 0.64 0.33
2 Cage-Guided Equal-Percentage 2.3125 1.125 59.7 0.85 0.69 0.31
3 Cage-Guided Linear 3.4375 1.5 148 0.82 0.62 0.3
3 Cage-Guided Equal-Percentage 3.4375 1.5 136 0.82 0.68 0.32
4 Cage-Guided Linear 4.375 2 236 0.82 0.69 0.28
4 Cage-Guided Equal-Percentage 4.375 2 224 0.82 0.72 0.28
6 Cage-Guided Linear 7 2 433 0.84 0.74 0.28
6 Cage-Guided Equal-Percentage 7 2 394 0.85 0.78 0.26
8 Cage-Guided Linear 8 3 846 0.87 0.81 0.31
8 Cage-Guided Equal-Percentage 8 3 818 0.86 0.81 0.26
References

1. Chemical process engineering Vol. 1, Chapter 5 P. 246 By Kayode Coker

2. Process Engineers Pocket Handbook, chapter 1: Fluid Flow
3. Elements of chemical process engineering chapter 1, Basic Process Engineering
principles
4. JGC Standard Practice JGS 210-120-1-60E: Control Valves
5. Industrial Valves: Calculations for design, manufacturing, operation and safety decisions,
Karan Sotoodeh, Chapter 2: Valve Sizing
6. ANSI/ISA standard ISA-75.01.01-2007
7. Control valve Handbook 2nd edition, Fisher, p. 109
8. Valtek Control Valve sizing, Chapter 3.
9. Sizing control valves, Aramco standards, Process instrumentation.

1. Noise estimation in JGC Standard Practice JGS 210-120-1-60E: Control Valves

Appendix 4
2. Control Valve Selection, Aramco standards, Process instrumentation.
3. Sizing control valves, Aramco standards, Process instrumentation.
Vessel Sizing
Link: https://processpocket.streamlit.app/

This tool was developed for process engineers to estimate a vessel Size on site quickly and compare
various methods from the literature. As a part of a larger project to develop what is similar to Carl
Branan’s book “process engineers’ Pocket Handbook” these tools would allow a process engineer to
quickly calculate/estimate equipment efficiencies or sizing using standardized calculations.

The aim here is to take little-known data from the field (flow, pressures, temperatures,
compositions..etc.) and use it as input for a rough estimation without having to return to the office to
use commercial software or calculations Excel sheets to validate or to calculate. Additionally, these tools
may also serve as a gathered data validation tool.

Nomenclature*

T Temperature (oC or oR) P Pressure (kPa, bar or Psia)

ρ Density (kg/m3 or lb/ft3) µ Viscosity (cP)
D Vessel diameter (mm, in or ft) SG Specific gravity [-]
u or v Velocity (m/s or ft/s) Z Gas compressibility (dimensionless)
Re Reynold’s number (dimensionless)  Specific weight (dimensionless)
Q Volume flow rate M.wt Molecular weight (dimensionless)
(m3/hr or Nm3/hr) or (bbl/d and MMSCFD)
V Vessel Capacity (bbls) H or h Vessel Height(s) (Vertical) (ft or mm)
W Vessel Liquid Capacity (bpd) L Vessel Length(s)(Horizontal) (ft or mm)
k K factor (m/s or ft/s) A Area (m2 or ft2)
Dp Droplet Size (ft or microns) d Nozzle diameter (in. or mm)
t Time in mins E Joint Efficiency
C Corrosion Allowance
Note: Units will differ from one method to another; the TOTAL method uses SI units, while other methods use
US units
Input Required
User Defined Properties Using App properties
3
Gas Flow rate (kg/hr or Nm /hr) Total Flow rate (kg/hr)
Liquid (Oil Flow rate) (kg/hr or m3/hr) Water Flow rate (3 phase) (kg/hr)
Water Flow rate (3 phase) (kg/hr) Pressure (kg/cm2)
Pressure (kg/cm2) Temperature (ᵒC)
Temperature (ᵒC) Composition (% mol)
Gas Viscosity (cP)
Liquid Viscosity (cP)
Gas Density [option 1] (kg/m3)
Liquid Density [option 1] (kg/m3)
Oil API [option 2] [-]
Gas Molecular Weight [option 2] [-]
Gas Compressibility [option 2] [-]
Output Obtained (2-Phase Horizontal Separator)

Property Table
Gas Mass Fraction -
Gas Mass Flow rate kg/hr
Liquid Mass Flow rate kg/hr
Gas Flow rate Nm³/hr
Liquid Flow rate m³/hr
Stream Pressure kg/cm²
Stream Temperature C
Gas Density kg/m³
Liquid Density kg/m³
Gas Viscosity cP
Liquid Viscosity cP
Gas Molecular weight -
Gas Compressibility -

Total Hussien Method

Diameter D mm Diameter D mm
Tan to Tan Length T-TL mm Selected Length L mm
Length L mm Total Capacity (approx.) V m³
Total Capacity V m³ Selected Length to D ratio SR or -
(approx.) LD
Length to D ratio T_TLD - Head Type - -
assumed L/D LD - C cofficient C -
High Liquid Level HLL mm Reynlods number Re -
Low Liquid Level LLL mm Settling Velocity h3 m/s
K Value K m/s LD Constraint- Gas Capacity LD ft.in
Droplet Size Dp micron D2L Constraint - Liquid DL in2.ft
Capacity
Head Type - - Residence Time t mins

Svrcek Method API 12J Method

Total Flow Rate m kg/hr Total Flow Rate kg/hr
Gas Flow Rate mg kg/hr Gas Flow Rate Nm³/hr
Liquid Flow Rate moil kg/hr Liquid Flow Rate kg/hr
Diameter of Drum D mm Feed pressure kg/cm².a
Vessel Tan-Tan Length L mm Feed Temperature ºC
Total Capacity (approx.) V m³ Liquid Density kg/m³
L/D - - Gas Density kg/m³
Shell Thickness ts mm Molecular weight -
Head Thickness tH mm Compressibility factor -
Total Weight W kg Gas Viscosity cP
Inlet Nozzle size dF inch Gas specific gravity -
Liquid Outlet Nozzle size dL inch Retention Time minutes
Vapor Outlet Nozzle size dv inch Head Type -
High Liquid Level HHLL mm Normal liquid level %
Normal Liquid Level HNLL mm K factor m/s
Low Liquid Level HLL mm Gas Actual Velocity m/s
Holdup Time TH min Gas Actual Volume flow m³/s
Surge Time TS min Minimum gas flow area m²
MeshPad - - Minimum ID of separator in
Head Type - - Selected Diameter in
K factor K ft/s L/D -
Droplet Size Dp micron Selected Diameter (mm) mm
Material/Alloy grade and type: - - Vessel Length mm
Max. Operating Pressure MOP Barg Liquid volume (excluding bottom head) m³
Design Pressure DP Barg Liquid capacity of separator m³/hr
Design Temperature o Total Capacity (approx.) m³
DT C
Corrosion Allowance C mm
Joint Efficiency E -

Additions/modifications to output table for other cases

Vertical Separator (2 Phase)

Total Hussien Method

h1 mm
h2 mm
h3 mm
h4 mm
h5 mm
h6 mm
h7 mm
h8 mm

Svrcek Method API 12J Method

Surge Liquid Level hs mm H/D -

The intuition behind Vessel Sizing Calculations: minimum length for vapor-liquid
Quick takes disengagement should be satisfied.
2. Liquid capacity constraint: you must
For both Vertical and Horizontal separators,
ensure enough space for liquid to cover
there are general themes you may notice
throughout the following document. In general, surge and holdup times (or retention
the bigger the vessel, the better, but this would times). After the holdup and surge times
also result in a higher capital cost. The 1st are established and given your feed flow
question to be answered is: What are the rate, you can calculate the surge and
constraints for sizing a vessel while the second holdup volumes that your separator must
should be: How could we minimize cost while satisfy. A gas–oil separator has to have a
maintaining these constraints? sufficient volume to retain the liquid for
the specified retention time before it
Constraints: leaves the separator.
For two-phase separation, 2 constraints are at For three-phase separation two additional
play here: constraints are Also to be accounted for:
1. Gas Constraints: you must ensure enough 1. Water Droplets Settling Constraint: the
vapor velocity to separate entrained liquid water settling time from the oil phase
from vapor: into the water phase. This could be
• Vertical Separators: we’ll be calculating calculated by calculating the water
the velocity limit where separation is settling velocity using Stoke’s law and
ensured, Higher velocities will result in given the available height to separate
lower separation efficiency. This principle the water of oil required, the time limit
is directly tied to the vessel’s diameter for could be calculated by dividing both.
vertical separators. The lower the The residence time of the oil phase in
diameter the higher the vapor velocity. your sized separator must cover that
The velocity limit would result in a time of course.
minimum diameter which will be used for 2. Oil Droplets Settling Constraint: The
initial sizing and built on to calculate the same principle applies to separating the
vessel length. water phase's oil droplets!
• Horizontal Separators: A slightly different
approach since the geometry of Cost
horizontal separation results in an easier The initial estimation of a vessel was directly
gravity separation. The effective length, linked to its weight via its dimensions. The
along with the diameter, will also be vessel length height, and head type, along
accounted for. A larger diameter will not with the design pressure, design
ensure separation if the vapor’s traveling temperature, and corrosion allowance, will
length is well below the required length play a vital role in estimating the vessel
to allow for gravity separation. weight. Consequently, After the initial sizing
Together, diameter and effective vessel a weight minimization algorithm/iteration is
length will be used to satisfy the gas required.
constraint. Given a diameter, The actual
These are the general themes you may notice
vapor velocity will be calculated and the
while you’re reading the different approaches
presented here. Refer to them for a detailed
explanation of the equations involved or to K = A constant depending upon design and
chapters 6 and 7 of Petroleum and Gas Field operating conditions ft/s
Processing, Second Edition by Hussein et al. K factors for determining maximum allowable
superficial velocity
K factor estimation Table 1: Typical K factor Ranges
For all the methods mentioned below, the K Separator L (ft) K (ft/s)
factor is the constant for the vapor terminal
velocity equation which is a modification of 5 0.12 to 0.24
Stokes low. This constant depends on the design Vertical
and operating conditions and generally regarded 10 0.18 to 0.35
as measure for separation efficiency. Generally,
the higher K factor the higher the vapor velocity Horizontal 10 0.40 to 0.50
for separation meaning that separation could be
done are higher velocity and consequently a
Other Lengths 0.40 to 0.50 x (L/10)0.56
lower vessel diameter. K factor is affected by the
presence of mist eliminator, operating
pressures, droplet size…etc. Similar to K, Ks is Spherical All 0.2 to 0.35
the constant for liquid settling velocity out of • The maximum allowable superficial velocity
another liquid. K factor is usually estimated in
calculated from the above factors is for
the preliminary phase as it’s mainly a factor
separators normally having a wire mesh mist
determined by the designer that’s why you may
extractor. This rate should allow all liquid
encounter numerous estimations.
droplets larger than 10 microns to settle out
API RP 12J of the gas. The maximum allowable
Sizing of Two-Phase Oil-Gas Separators.[1] superficial velocity or other design criteria
The following calculations are presented as a should be considered for other types of mist
guide to the design and sizing of two-phase and extractors. The mist extractor
three-phase separators. Sizing should be based manufacturer's recommended minimum
on the maximum expected instantaneous rate.
distances upstream and downstream of the
Theory and Equations
wire mesh between the gas inlet and outlet
Gas capacities of separators may be determined
nozzles should be provided for full utilization
by a modification of Stokes' Law. When using
Stokes' Law, the capacity is based on the of the mist extractor
principle of the minimum droplet size that will • The oil capacity of a separator is a function
settle out of a moving gas stream at a given of retention time and gas-oil interface area.
velocity. The maximum allowable superficial The basic requirement is to retain the oil
velocity of the gas at operating conditions is long enough and provide sufficient interface
calculated by the following formula, Where: area for entrained gas to break out of the oil.
Va = K * ((ρL – ρG)/ ρG)0.5, Where Separator liquid capacity is normally based
Va = Maximum allowable superficial velocity in on a minute retention time for non-foaming
ft/s through the secondary separation section oils having a gravity of 35 O API and above. A
ρL = Density of the liquid in lb/ft3 at operating gravity lower than 35O API may require a
conditions
greater retention time.
ρG = Density of the gas in lb/ft3 at operating
conditions
• Foaming crudes offer a special problem in Table 2: retention time for Oil
sizing separators. Foam is a mixture of gas Oil Gravities Minutes
(Typical)
dispersed in a liquid and having a density less
Above 35" API 1
than the liquid but greater than the gas. A
20 - 30" API 1 to2
greater interface area and longer retention
10 - 20" API 2 to 4
time are needed to remove the gas from the
• Note: Assumed Retention time for the tool
liquid. Horizontal separators normally give
is the sum of surge + holdup time from
the largest interface area. Retention times of
Svrcek’s papers (Table no. 7)
as high as 15 minutes may be necessary.
• The settling volumes may be used in the V &
However, a retention time of 2 'to 5 minutes
W equations to determine the liquid capacity
is sufficient in most cases for the separators
of a particular vessel. For proper sizing, both
to handle foaming crudes. Where the well
the liquid capacity and gas capacity should
can be sampled in a test unit, a more
be determined. It may be noted that on most
accurate estimate of the required retention
high-pressure gas distillate wells, the gas-oil
time- can be determined. De-foaming
ratio is high and the gas capacity of a
separator designs often include a variety of
separator is usually the controlling factor.
proprietary internal configurations to
However, the reverse may be true for low-
improve capacity. These are beyond the
pressure separators used on well streams
scope of this specification (API RP 12J).
with low gas-oil ratios. The liquid discharge
• In addition to the well stream properties, the
or dump valve on the separator should be
gas capacity is influenced by the following:
sized based upon the pressure drop
a. Operating temperature is above the cloud
available, the liquid flow rate, and the liquid
point of oil
viscosity.
b. Operating temperature being above the
Sizing of Three-phase Gas-Oil-Water Separators
hydrate point of gas
The basic principles of oil and gas separation
c. Foaming tendency of liquid
have been covered under the Sizing of Two-
d. Uniformity of flow Phase Oil-Gas Separators. The following portion
e. De-foaming chemicals; if used will cover the separation of free water and oil:
The liquid capacity of a separator is primarily • All of the basic separators (vertical,
dependent upon the retention time of the liquid horizontal, spherical) may be used for
in the vessel. Good separation requires three-phase separation. Regardless of
sufficient time to obtain an equilibrium
shape, all three-phase vessels must meet
condition between the liquid and gas phase at
the following requirements:
the temperature and pressure of separation.
a. Liquid must be separated from gas in a
The liquid capacity of a separator or the settling
volume required based on retention can be primary separating section.
determined from the following equation: b. Gas velocity must be lowered to allow
Basic design criteria for liquid retention time in liquids to drop out.
two-phase separators are generally as follows: c. Gas must be scrubbed through an
efficient mist extractor.
d. Water and oil must be diverted to a
turbulence-free section of the vessel.
e. Liquids must be retained in the vessel • Determine which shape fits the particular
long enough to allow separation. installation best considering space, mounting,
f. The water-oil interface must be and ease of access for maintenance. Both
maintained. present and future operating conditions
g. Water and oil must be removed from should be considered.
the vessel at their respective outlets. • Determine whether unusual well stream
• Sizing a three-phase separator for water conditions (foam, sand, etc.) would make the
removal is mainly a function of retention vessel selected difficult to operate or
time. Required retention time is related to maintain.
the volume of the vessel, the amount of • Determine whether over-all economics is
liquid to be handled, and the relative affected by the installation or portability of
specific gravities of the water and oil. The the shape selected.
effective retention volume in a vessel is • Make certain that all design requirements
that portion of the vessel in which the oil such as heating coils for paraffin or hydrates
and water remain in contact with one and three-phasing for water removal have
another. As far as oil-water separation is been considered and are compatible with the
concerned, once either substance leaves shape selected.
the primary liquid section, although it may • Consider possible liquid slugging of the
remain in the vessel in a separate separator
compartment, it cannot be considered as a Vertical Separator (2 phase) Sizing Procedure
part of the retention volume. There are according to API method
two primary considerations in specifying 1. Assume Vessel height and NLL
retention time: 2. Select retention time from Table 2 based on
a. Oil settling time to allow adequate water oil API.
3. Using Table 1, Assume K value
removal from oil
4. Calculate The maximum allowable superficial
b. Water settling time to allow adequate oil velocity for gas
removal from water
𝑑𝐿 − 𝑑𝐺
The usual approach in design is to allow equal 𝑉𝑎 = 𝐾√
retention times for oil and water. This is 𝑑𝐺
accomplished with a wide range interface level 5. Step 1: Determine gas and oil properties
controller or variable water weir. Basic design
2.7𝛾𝑃(𝑃𝑠𝑖𝑎) 𝑙𝑏
criteria for liquid retention time in three-phase 𝜌𝑔 = ( 3)
separators are generally as follows: 𝑇(ᵒ𝑅)𝑍 𝑓𝑡
Table 3: Retention time for oil and water 𝑙𝑏
Oil Gravities Minutes (Typical) 𝜌𝐿 = 62.4 ∗ 𝑆. 𝐺 ( )
O
𝑓𝑡 3
Above 35 API 3 to 5
Below 35 O API 6. Calculate Actual Volume flow rate of gas
+100 O F 5 to 10 𝑄 𝑖𝑛 𝑆𝐶𝑓𝐷 ∗ 𝑀. 𝑤𝑡 𝑙𝑏𝑠/𝑚𝑜𝑙
O 𝑆𝐶𝐹 𝑠𝑒𝑐
+80 F 10 to 20
379.5 ∗ 86400 ∗ 𝑔𝑎𝑠 𝐷𝑒𝑛𝑠𝑖𝑡𝑦 𝑙𝑏/𝑓𝑡 3
+60 O F 20 to 30 𝑚𝑜𝑙 𝑑𝑎𝑦
Separator Selection - The following procedure 7. Calculate Minimum gas flow area
may be used when selecting a separator for a 𝑄𝑎𝑐𝑡𝑢𝑎𝑙
𝐴𝑚𝑖𝑛 = (𝑓𝑒𝑒𝑡 2 )
particular application: 𝑉𝑎
8. Determine minimum required ID for the 𝑄 𝑖𝑛 𝑆𝐶𝑓𝐷 ∗ 𝑀. 𝑤𝑡 𝑙𝑏𝑠/𝑚𝑜𝑙
separator 𝑆𝐶𝐹 𝑠𝑒𝑐
379.5 ∗ 86400 ∗ 𝑔𝑎𝑠 𝐷𝑒𝑛𝑠𝑖𝑡𝑦 𝑙𝑏/𝑓𝑡 3
𝑚𝑜𝑙 𝑑𝑎𝑦
𝐴𝑚𝑖𝑛 (𝑓𝑒𝑒𝑡 2 ) ∗ 4 7. Calculate Minimum gas flow area
𝐷𝑚𝑖𝑛 = √ ∗ 12 (𝑓𝑒𝑒𝑡)
𝜋 𝑄𝑎𝑐𝑡𝑢𝑎𝑙
𝐴𝑚𝑖𝑛 = (𝑓𝑒𝑒𝑡 2 )
9. Nearest Std. Diameter is selected (Table 13) 𝑢𝑚𝑖𝑛
10.Calculate H/D 8. Assume minimum Diameter in inches
Note: As per GPSA, typical horizontal H/D ratios 9. Calculate Minimum gas flow area and Dmin
are normally in the 2 to 4 range. 𝑟 − 𝑁𝐿𝐿
𝐴𝑠𝑒𝑔𝑚𝑒𝑛𝑡 = 𝑟 2 cos −1 ( ) − [(𝑟
11.Based on Assumed Height and NLL Calculate 𝑟
Liquid Volume, V (Excluding bottom head) − 𝑁𝐿𝐿√2𝑟 ∗ 𝑁𝐿𝐿 − 𝑁𝐿𝐿2 ] (𝑓𝑒𝑒𝑡 2 )
𝐷
𝑆𝑒𝑙𝑒𝑐𝑡𝑒𝑑 𝐼𝐷 (𝑖𝑛𝑐ℎ𝑒𝑠)2 ∗ 𝜋 𝑟 = ∗ 12 (𝑓𝑒𝑒𝑡)
𝑉= (𝐵𝑏𝑙𝑠) 2
4 ∗ 144 ∗ 5.615 ∗ 𝐻 ∗ 𝑁𝐿𝐿 10.Iterate through Dmin until Asegment= Amin
11.Select Nearest Std. Diameter (Table 13)
12.Calculate the Liquid Capacity of the
12.Calculate L/D
separator W
13.Based on Assumed Height and NLL Calculate
1440 ∗ 𝑉
𝑊= (𝑏𝑝𝑑) Liquid Volume, V (Excluding bottom head)
𝑡 ( 𝑟𝑒𝑡𝑒𝑛𝑡𝑖𝑜𝑛 𝑡𝑖𝑚𝑒 𝑖𝑛 𝑚𝑖𝑛𝑠)
𝑉 = 𝐴𝑠𝑒𝑔𝑚𝑒𝑛𝑡 (𝑓𝑒𝑒𝑡 2 ) ∗ 𝑆𝑒𝑙𝑒𝑐𝑡𝑒𝑑 𝐿 𝑖𝑛 𝑓𝑒𝑒𝑡
Verify that W > Liquid flow rate otherwise re-
∗ 0.17811 (𝐵𝑏𝑙𝑠)
assume H and NLL 14.Calculate Liquid Capacity of the separator W
Ensure L/D is in the range of 2-4 as per GPSA. 1440 ∗ 𝑉
Vertical Separator (3 phase) Sizing Procedure 𝑊= (𝑏𝑝𝑑)
𝑡 ( 𝑟𝑒𝑡𝑒𝑛𝑡𝑖𝑜𝑛 𝑡𝑖𝑚𝑒 𝑖𝑛 𝑚𝑖𝑛𝑠)
according to API method
Verify that W > Liquid flow rate otherwise re-
Same procedure except that W > (Heavy and
assume L, D and NLL
Light Liquid Flow rates)
Ensure L/D is in the range of 2.5-5 as per GPSA.
Ensure L/D is in the range of 2-4 as per GPSA.
Horizontal Separator (3 phase) Sizing
Horizontal Separator (2 phase) Sizing
Procedure according to API method
Procedure according to API method
Same procedure except that W > (Heavy and
1. Assume Vessel length and NLL
Light Liquid Flow rates)
2. Select retention time from table 2 based on
Ensure L/D is in the range of 2.5-5 as per GPSA.
oil API.
3. Using Table 1, Assume K value based on
selected length
4. Calculate The maximum allowable superficial
velocity for gas
𝑑𝐿 − 𝑑𝐺
𝑉𝑎 = 𝐾√
𝑑𝐺
5. Determine gas and oil properties

2.7𝛾𝑃(𝑃𝑠𝑖𝑎) 𝑙𝑏
𝜌𝑔 = ( 3)
𝑇(ᵒ𝑅)𝑍 𝑓𝑡
𝑙𝑏
𝜌𝐿 = 62.4 ∗ 𝑆. 𝐺( )
𝑓𝑡 3
6. Calculate Actual Volume flow rate of gas
TOTAL • Try to avoid vessels with wall thickness
Horizontal or vertical design (notes) [2] greater than 100 mm as these require
• Provided a sufficient L/D ratio is selected, a special fabrication and can prove expensive.
horizontal separator is more efficient than a Calculation theory and equations
vertical for the same flow area.
• Vapor velocity in a horizontal drum can • Valid only for pure gravity settlers with no
exceed the liquid settling velocity provided internals to enhance separation)
L/D > 1. For vertical drums, the velocity
Liquid-vapour settling velocity
cannot.
• Horizontal drums are more effective and 1. Vs = K* ((ρL- ρv)/ (ρv))1/2 m/s
0.5
geometrically more practical for heavy 2. K = 0.003615 (D/C) m/s
liquid phase removal than vertical drums. 3. UT = 0.003615 (D/C) * ((ρL- ρv)/ (ρv))1/2
0.5

• A rising liquid level in a vertical drum does K in m/s, Densities in kg/m3, D is droplet size in
not alter the vapor flow area. Consequently, microns and C is drag coefficient.
vertical drums are preferred for compressor • For medium and low pressure’ with gases of
and fuel gas KO drums. Vertical drums viscosity less than 0.01 cP Figure 2 can be
utilize a smaller plot area and are easier to used to estimate Vs.
instrument with alarms and shutdown • For higher pressures (> 50 bar) or viscosities
controls. For floating installations are over 0.01 cP it is necessary to calculate Vs.
preferred as less “sloshing” occurs. The drag coefficient is calculated using
• For high volume flow rates a split flow Figure 3 (curve 2) where:
horizontal drum is preferable as smaller CRe = (0.3072*10-11*ρv*D3*(ρL - ρv) / µ2
drum diameters can be used. The preferred
split flow arrangement is a single centre Equation in step 3 is then used to calculate Vs.
entry nozzle with two end exists. Head exits Liquid-liquid settling velocity (based on Stoke's
can be used where plot space is limited. law of terminal settling)
• Each design case must be evaluated
The following equation can be used for
separately but in general the following can
calculating the settling velocity of water in oil or
be used as a guideline:
the upward “settling” of oil in water. The
Vertical drums important fact is to use the viscosity of the
Compressor KO drums continuous phase i.e : for oil settling upwards
De-gassing boots through water use the water viscosity, for water
Fuel gas KO drums settling in oil use the oil viscosity.
Absorber feed KO drums Ut = (g*D2*(ρH – ρL)) / 18*µc m/s
Floating installations
Horizontal drums Ut, = terminal velocity m/s
Production separators HP g: gravitation acceleration m/s2
Reflux drums ρH = density heavy fluid kg/m3
3-phase separation ρL = density light fluid kg/m3
Flare KO drums µc = viscosity (continuous) kg/m.s
Setting the particle size to 125 microns and eliminator. It is recommended to install
using more useful units gives: a mist eliminator for most
applications. if not calculate Vs using
Ut = 0.513*(ρH – ρL) / µc
Ut mm/min µc in cP the equation in step 3 earlier.
3. De-rate the calculated settling velocity
• The above equation is valid for Reynold’s no by 85 % design margin to give a
of 0.1 – 0.3 If the calculated settling velocity maximum allowable vapor velocity.
is > 250 mm/min use 250 max 4. Calculate the drum internal diameter
and round to the nearest 50 mm. (note
Vessel volumes
further adjustment of ID: OD can be
Partial volumes of a horizontal cylinder can be made to suit standard head
calculated using the partial volume charts in dimensions).
Figure 2 or estimated using the following 5. Check if wall thickness is less than 100
equations : (for vessels with a diameter < 1.2 m mm.
ignore head volumes)
𝐷2 𝐷−2ℎ 𝐷
𝐴𝐿 = cos−1 [ ] − ( − ℎ)(𝐷ℎ − ℎ2 )0.5 m2
4 𝐷 2

Horizontal cylinder H1
VC = AL* L m in radians H2

2 dished head H3
Vdh = 0.21543 h2 (1.5 D – h) m3
ɸ
2 elliptical heads H4
Vel = 0.52 194 h² (1.5 D – h) m³ (most common)
2 hemispherical heads
HLA
Vhh = 1.047 h2 (1.5 D -h) m³ (gives extra vol) H5 HLL
Volume up to baffle
H6 NLL
For depth h
(0.52194 h2 (1.5D-h))/2+ AL*B (Elliptical heads) H7
LLL

• These formulas are accurate enough for LLA

general design and are easily programmed H8

onto a calculator for time-saving.
• For greater accuracy, the length L should be
the tan-tan length and not the flow path
length between nozzles. This is especially
true with large vessels and tight design. Calculate vessel height based on the following
criteria
Vertical separator
1. Decide if Figure 2 can be applied i.e. P < H1 max(15 % of ɸ or 400 mm)
50 bara, µ < 0.01 cp.
H2 100 mm if mesh selected
2. If applicable, use the 500 micron curve
150 mm for compressor KO
to evaluate settling velocity (this
H3 max (50 % of ɸ or 600 mm) If no mesh, use
assumes A mist eliminator will be
H1 + H2 + H3 = 60 % ɸ or 800 mm
installed) or 150 micron with no mist
H4 400 mm + d/2: d= inlet nozzle ɸ 3. Evaluate required vapor cross-sectional area,
Av.
H5 calculate based on
4. Assume the drum is 70 % full, i.e., h/D = 0.7,
1-2 minutes residence time
at maximum liquid inflow –min 200 mm and evaluate drum D to give the required Av,
H6 base on the following hold up times: (min using Figure 4 (to the nearest 50 mm). For
350) “dry” vessels, use h/D = .35
5. For required liquid surge volume, calculate
Reflux drums 4 min
vol at HLL, if insufficient adjust D or L (note if
Product drums 5 min with pump
L/D changes significantly recheck Av using
3 min no pump
new Vm).
Heater feed 8 min
HP sep. to LP sep. 4 min 6. Set position of LLL in drum and confirm
H7 1-2 min residence time required surge vol between HLL-LLL. If
– minimum 150 mm volume is insufficient increase ø, L or h.
Include volumes in heads.
H8 - 150 mm for bottom
7. When setting the LLL height, take into
connected LC
account any LSLL, LSL alarms, and vortex
300 mm for side
breakers that may set a minimum usable
connected LC
Note: For compressor suction drums that are value. This value is usually 300-350 mm.
normally dry set HLL at 450 mm Above tan line 8. Rationalize all heights and dimensions to the
and use bottom connected LC. This will reduce nearest 10 mm.
vessel height if required. No specific HLL-LLL
NOTES:
hold-up time is required.

Calculation procedure horizontal vessel (Vapor- • For high volumetric flows of gas with small
liquid) liquid volumes consider using a split flow
L’ arrangement. Design is as above but with
half vapor volume flow.
L • Normal design is with top entry and exit
nozzles. However, if space is limiting
(primarily offshore) head mounted nozzles
can be used to increase flow path.
ɸ HLL • L is designated as the flow path length i.e
NLL distance between inlet and outlet nozzle. L’
LLL or TTL is the tangent-tangent length. For 1st
estimates L’ = L + 1.5 ɸi + 1.5 ɸ2

1. Calculate settling velocity Vs for particle size ɸi = inlet nozzle diameter

150 micron , use Figure 2 or Vs equation. ɸ2 = outlet nozzle diameter
2. De-rate this by F = 0.85 and calculate • “Normal” liquid levels are taken as midway
required vapor velocity V m/s between the high and low levels
Vm=F*Vs* (L/D) m/s

Use L/D of 3 to 4 max (3 initial estimate)

Calculation procedure horizontal vessel 3phase drum. Baffle height and position can now
be set noting:
L’ • The baffle should be at least 75 mm
below the LLL
L • The baffle should be at least 2/3 down
the length of the drum from the inlet in
some cases the water droplets will
settle to the floor in a short distance.
ɸ HIL
HLL The baffle should still be set at a
NLL minimum of 2/3 along the vessel.
LIL
LLL 5. Set the HIL at baffle height 75 mm. The LIL
according to height determined vortex
B breaker + LSLL use a minimum of 300-350
mm.
6. Check if an oil droplet will rise through the
Sufficient residence time to allow separation of
the oil-water mixture as well as the oil surge water layer (from drum floor) to LIL before
and vapor flow areas must be provided. reaching water outlet. Use area at LIL with
normal oil + water flow rates. (This criteria
1. Proceed with steps 1 to 4 as for a two is very rarely governing but must be
phase separation. Use L/D = 3 (1st checked).
estimate) and evaluate L. 7. Calculate water surge time Vol@HIL -
2. Provision now has to be made to Vol@LIL, and residence time Vol@NIL-
accommodate both oil and water surge Vol@Outlet. Remember to use only one
volumes. Use Tan-Tan length L’ and not head volume, and length of drum up to
nozzle-nozzle distance L. baffle. Minimum acceptable times are 4-5
3. Calculate LLL required to give approx. mins. If calculated times are very long
4 mins oil Surge capacity (minimum). consider using a water boot arrangement.
Inspection will reveal whether sufficient 8. Rationalize all dimensions and “tidy” levels
height exists below LLL to include the to standard values if possible i.e: 150 mm,
interface levels. If not, adjust the vessel ɸ 200, 250, 300 etc. This allows use of
or L to give sufficient room. standard displacers.
• Note: If the water cut is very small, 9. Recalculate all residence times based on
consideration may be given to using a “tidied” levels (if required).
water boot instead of a baffle • Note: When calculating the final residence
arrangement see step 10. times, use the vessel tan-tan length, not
4. Having determined HLL and LLL now set the nozzle-to-nozzle distance L.
both position and height of baffle. 10. Boot calculation (See Figure 5)
Calculate terminal settling velocity of water If the water volumetric flow is so small as not to
droplet (Ut equation) at both HLL and LLL. warrant a separate baffled settling
Volumetric flow of liquid is in both cases compartment, as detailed above, a water boot
the oil plus the water. Calculate fall should be used instead. To design proceed as
distance of a droplet across length of the follows:
2. Proceed as previous up to step 3. Calculate vessel wall thickness using the ASME
3. Calculate settling distance of water droplet VII div. I formula. The wall thickness should be
when vessel is operating at LLL. Water calculated immediately after D is known to
droplet should reach floor of drum before confirm if t < 100 mm.
oil outlet. Remember that the oil exit 𝑃𝐷
nozzle will be raised above the floor as a 𝑡= +𝐶
2𝑆𝐸 − 1.2𝑃
standpipe. Adjust drum Ø or L to achieve
D: Diameter mm
settling. t: wall thickness mm
4. Check that settling is also possible when P: Design Pressure barg
operating at HLL, droplet to fall below draw E: Joint Efficiency
off nozzle level. Use 1 for seamless shells 0.85 otherwise
5. Size water draw-off boot Ø (try to use S: Max. Allowable Stress bar
standard pipe diameters). Calculate rising 1220 bar for Carbon Steel
velocity of the oil in water, set downward 1000 bar for SS Plate
velocity of water in boot at 90 % of this and For t < 100 No fabrication problem
evaluate boot Ø. Boot length by inspection 100 < t < 150 Vendor advice may be needed
(use standard displacers). t > 150 mm Major fabrication problem
• Note : Boot Ø must be less than 35% of
To meet standard vessel head sizes and wall
vessel v
thicknesses the following ranges should be
When heavy walled vessels are used a remote observed:
boot may be more economical to prevent large Vessel diameter:
cuts in the main vessel. 250 1250 mm in increments of 50 mm
1300-4000 mm in increments of 100 mm
NOZZLE SIZING Standard wall thicknesses:
Inlet nozzle 1 - 30 mm in increments of 1 mm
Size based on normal volumetric flow + 10 % 30 - 60 mm in increments of 2 mm
(liquid + vapor flow) Limit inlet velocity to 7 - 13 60 - 140 mm in increments of 5 mm
m/s

Round nozzle diameter up or down to the

nearest standard size.

Gas outlet Liquid outlet

Size on normal flow Normal flow + 10 %

Velocity limit 15-30 m/s Velocity limit
1-3 m/s HC
2-4 m/s water
Manholes: 450 mm or 600 Dmin = 2” (avoid
plugging)

Vessel wall thickness

Hussein Method Settling of Oil Droplets
Basic Assumptions [3]
In separating oil droplets from the gas in the
• No oil foaming takes place during the gas–oil gravity settling section of a separator, a relative
separation (otherwise retention time has to motion exists between the particle, which is the
be drastically increased. oil droplet, and the surrounding fluid, which is
• The cloud point of the oil and the hydrate the gas. An oil droplet, being much greater in
point of the gas are below the operating density than the gas, tends to move vertically
temperature. downward under the gravitational or buoyant
force, Fg. The fluid (gas), on the other hand,
• The smallest separable liquid drops are
exerts a drag force, Fd, on the oil droplet in the
spherical ones having a diameter of 100 μm.
opposite direction. The oil droplet will
• Liquid carryover with the separated gas does
accelerate until the frictional resistance of the
not exceed 0.10 gallon/ MMSCF. fluid drag force approaches and balances the
gravitational force; and, thereafter, the oil
Fundamentals
droplet continues to fall at a constant velocity
• The difference in densities between liquid known as the settling or terminal velocity. The
and gas is taken as a basis for sizing the gas drag force is proportional to the droplet surface
capacity of the separator (ρo – ρg). area perpendicular to the direction of gas flow
• A normal liquid (oil) retention time for gas to and its kinetic energy per unit volume. Hence,
separate from oil is between 30 s and 3 min. 𝜌𝑔 𝑢2
Under foaming conditions, more time is 𝐹𝑑 = 𝐶𝑑 ∗ 0.25 ∗ 𝜋 ∗ 𝑑 2 ∗
2
considered (5–20 min). Retention time is
Where Fg is given by
known also as the residence time, V/Q,
where V is the volume of vessel occupied by 𝐹𝑔 = 0.25 ∗ 𝜋 ∗ 𝑑 3 ∗ (𝜌𝑜 − 𝜌𝑔 )𝑔
oil and Q is the liquid flow rate.
where Cd is the drag coefficient, d is the
• In the gravity settling section, liquid drops
diameter of the oil droplet (ft), u is the settling
will settle at a terminal velocity that is
velocity of the oil droplet (ft/s), ρo and ρg are the
reached when the gravity force, Fg, acting on oil and gas densities (lb/ft3), respectively, and g
the oil drop balances the drag force, Fd, is gravitational acceleration (ft/s2). The settling
exerted by the surrounding fluid or gas. terminal velocity, u, is reached when Fd = Fg. By
• For vertical separators, liquid droplets (oil) equating Equations and solving for u, the
separate by settling downward against an droplet settling velocity, we obtain
up-flowing gas stream; for horizontal ones,
8 (𝜌𝑜 − 𝜌𝑔 ) 𝑑
liquid droplets assume a trajectory-like path 𝑢2 = 𝑔 .
6 𝜌𝑔 𝐶𝑑
while it flows through the vessel (the
trajectory of a bullet fired from a gun). The droplet diameter, d, is normally expressed
• For vertical separators, the gas capacity is in microns, where 1 μm is equal to 3.2808 × 10–6
proportional to the cross-sectional area of the ft. Let dm be the droplet diameter (in microns)
separator, whereas for horizontal separators, and substitute 32.17 for g in the preceding
gas capacity is proportional to the area of equation to obtain the following expression for
disengagement (LD; i.e., length × diameter) the settling velocity:
0.5 Sizing Vertical Gas–Oil Separators
(𝜌𝑜 −𝜌𝑔 ) 𝑑𝑚
𝑢 = 0.01186. [ . ] (ft/s)
𝜌𝑔 𝐶𝑑

In designing gas–oil separators, the smallest oil

droplet to be removed from the gas in the
gravity settling section is normally taken as 100
μm. Under such a condition, the mist extractor
will be capable of removing oil droplets smaller
than 100 μm without getting flooded. There are
other special separators known as gas scrubbers
that are normally used to remove liquid from
gas streams, which have been through regular
gas-liquid separators. Such gas streams are
mostly gas with little liquid that may have
formed due to condensation. Examples of such
separators are the gas scrubbers used at the
inlet to gas compressors and gas dehydration
facilities. Scrubbers are also used on vents and
flares. Since the amount of liquid is very small, When sizing a separator, two constraints should
be accounted for:
the design of such gas scrubbers could be based
1. Gas Capacity Constraint: For vertical
on separation of liquid droplets up to 500 μm in
the gravity settling section with no danger of separators, the upward average gas velocity
flooding the mist extractor. should not exceed the downward terminal
velocity of the smallest oil droplet to be
Gas Capacity of Separators separated. This condition is expressed
Qg = Agu (ft3/s) mathematically by equating the following
Equations:
Qg = 3.056 (P/TZ)*u*Ag MMSCFD 0.5
(ρo -ρg ) dm
u=0.01186. [ . ] (ft/s)
Solving for the gas velocity, u, gives ρg Cd
u = 0.327*Qg*(TZ/P)*(1/Ag) (ft/s)
u = 0.327*Qg*(TZ/P)*(1/Ag) (ft/s)

Where Z is the gas compressibility at the 0.327Qg(TZ/P)*(1/Ag) =

0.5
operating pressure (P in psia) and temperature (ρo -ρg ) dm
0.01186. [ . ]
ρg Cd
(T in oR) and Ag is the available area for gas flow
(ft2). Ag = 0.25*π*(D/12)2 (ft2)
Liquid Capacity of Separators 𝑇𝑍 (ρo -ρg ) dm
0.5
The basic relationship that combines the oil flow D2 = 5508 ∗ 𝑄𝑔 ∗ ∗ [ . ] in2
𝑃 ρg Cd
rate or oil capacity of a separator, Qo, the
volume of separator occupied by oil, Vo, and the D is the minimum acceptable diameter of the
retention time or residence time, t, is separator. Larger diameters yield lower gas
velocities and, thus, better separation of the oil
Qo (ft3/min) = Vo / t droplets from the gas. Smaller diameters, on the
Qo (bbl/day) =257* Vo / t other hand, result in higher gas velocities and,
therefore, the liquid droplets will be carried
over with the gas.
A summary of the steps involved in the 2. Oil Capacity Constraint: The oil has to be
calculation of the gas capacity of a gas–oil retained within the separator for a specific
separator is outlined in Figure below. In solving retention time, t. The volume of separator
0.5
𝑇𝑍 (ρo -ρg ) dm occupied by oil, Vo, is obtained by multiplying
D2 = 5.508 ∗ 𝑄𝑔 ∗ ∗ [ . ] , the
𝑃 ρg Cd
the cross-sectional area by the height of the
value of the drag coefficient, Cd, must be first
oil column, H (in). equation Qo (bbl/day)
determined. Cd is related to the Reynolds
=257* Vo / t could, therefore, be rewritten
number, Re, according to the following formula:
as:
Cd = 0.34 + 3/Re0.5 +24/Re
𝐷2 𝐻
where the Reynolds number is given by 𝑄𝑜 = 257 ∗ 0.25 ∗ 𝜋 ∗ ∗ ∗ 𝑡 −1 , (bbl/day)
12 12
Re = 0.0049 ρg.dm.u/µg Or
D2H=8.565*Qo*t in3

2 conditions
Sizing Procedure
Find V* qd and qf
Fd = F g
Vf.Af = qf qs. qs =
m = qs.pf
In summary, the size (diameter and seam-to-
seam length or height) of a vertical separator is
determined as follows:
Find V*
Vf.Af = qf Vmax = Vf
qs = qs.pf/ps 1. Determine the minimum allowable vessel
c(po – pg/pi)**0.5 diameter.
0.5
𝑇𝑍 (ρo -ρg ) dm
D2 = 5.508 ∗ 𝑄𝑔 ∗ ∗ [ . ] in2
𝑃 ρg Cd
Find V* qs = qf By ideal gas law change
(ps/ps) 2. For diameters larger than the minimum, the
Vf.Af = qf (pf/ps)(Ts/Tf)z As f(T,P)
following is used to determine combinations
of D and H.
D2H=8.565*Qo*t in3
Cd could only be determined by an iterative 3. The seam-to-seam length, Ls, for each
procedure as follows: combination of D and H is determined using
1. Assume a value for Cd (a value of 0.34 one of the following expressions as
could be used as a first assumption). appropriate:
2. Calculate the velocity, u, from: If D < 36 in, then Ls =(H+76)/12 ft
0.5 If D> 36 in, then Ls =(H+D+40)/12 ft
(ρo -ρg ) dm
u=0.01186. [ . ] 4. For each combination of D and Ls, the
ρg Cd
slenderness ratio, SR, defined as the ratio of
3. Calculate Re from
length to diameter, is determined.
Re = 0.0049 ρg dm u/µg
Separators with SR between 3 and 4 are
4. Calculate Cd and compare to the assumed
commonly selected.
value from:
Cd = 0.34 + 3/Re0.5 +24/Re Notes: dm in micron, densities in lb/ft3, Qg in
5. If no match is obtained, use the calculated MMSCFD and Qo in bbl/d, t in mins, H and D in
value of Cd and repeat steps 2–4 until inches, T in oR and P in Psia.
convergence is obtained.
Sizing Horizontal Gas–Oil Separators volumetric flow rate, Qg, by one-half of the
separator cross-sectional area, A, that is,
ug = Qg /(0.5[0.25*π*D2]) ft/s
Qg is usually reported in units of MMSCFD
and should, therefore, be converted into
actual cubic feet per second (ft3 /s); also D,
which is usually given in inches, should be
converted into feet in order to obtain the
velocity in units of feet per second (ft/s). The
preceding equation, therefore, becomes
ug = 120*Qg*TZ/(D2*P) ft/s
The gas travels horizontally along the
effective length of the separator, L (ft), in a
The horizontal separator is determined by its time, tg, that is given by:
required capacity for gas and oil. It has been tg = L/ug s
shown that the gas capacity constraint for This time must, at least, be equal to the time
vertical separators determines the minimum it takes the smallest oil droplet to be
allowable vessel diameter. For horizontal removed from the gas to travel a distance of
separators, however, the gas capacity constraint D/2 to reach the gas–oil interface. This
yields, as shown in the following section, a
settling time, ts, is obtained by dividing the
relationship between the diameter and effective
distance (D/2) by the settling velocity from u
length of the separator. This along with a similar
equation; therefore,
relationship derived from the liquid capacity
0.5
constraint are used in determining the size of 𝐷 (ρo -ρg ) dm
𝑡𝑠 = . 0.01186. [ . ]
the separator. In reality, either the gas capacity 2𝑥12 ρg Cd
constraint or the liquid capacity constraint Equating both above equation, substituting for
governs the design and only one of the two ug from ug = 120*Qg*TZ/(D2*P), and solving for
constraints equations is used in determining the
the product LD, we obtain
size. 0.5
𝑄𝑔 𝑇𝑍 (ρo -ρg ) dm
𝐿𝐷 = 422. 𝑃
.[ ρg
.C ] ft /in
d
In the following discussion, it is assumed that
This equation provides a relationship between
each of the gas and oil phases occupies 50% of
the effective separator volume. Similar the vessel diameter and effective length that
equations as those derived in the following satisfies the gas capacity constraint. Any
could be obtained for other situations where combination of D and L satisfying the above
either of the two phases occupies more or less equation ensures that all oil droplets having
than 50% of the separator effective volume. diameter dm and larger will settle out of the gas
flowing at a rate of Qg MMSCFD into the
When Sizing a separator, two constraints should
separator that is operating at P psia and T°R.
be accounted for:
3. Liquid Capacity Constraint: A gas–oil separator
2. Gas Capacity Constraint: Since the gas has to have a sufficient volume to retain the
occupies the top half of the separator, its liquid for the specified retention time before it
average flowing velocity within the leaves the separator. For a horizontal separator
separator, ug, is obtained by dividing the
that is half full of liquid, the volume occupied by The preceding equations and sizing procedures
the liquid is given by are very sufficient for the determination of
Vo = 0.5*0.25* π*(D/12)2.L ft3 separator diameter and length as well as for the
performance evaluation of existing separators.
Substituting in Qo (bbl/day) =257* Vo / t, the Students and practicing engineers should be
following equation is obtained: familiar with such design equations and
D2L = 1.428Qot ft3 procedures before attempting the use of
commercially available software. The retention
The equation provides another relationship time is an important parameter in designing
between D and L that satisfies the liquid gas–oil separators. It is best obtained from
capacity (retention) time constraint. laboratory tests that simulate the field
Sizing Procedure operating conditions. This, however, may not
always be available. In such cases, experience
For a given set of operating conditions and data from offset fields, if available, will be
(pressure, temperature, gas and oil flow rates, very valuable.
gas and oil properties, and oil retention time),
the size (diameter and seam-to-seam length) of
a horizontal separator is determined as follows:

1. Assume various values for the separator

diameter, D.
2. For each assumed value of D, determine the
effective length, Lg, that satisfies the gas
capacity constraint from LD equation and
calculate the seam-to-seam length, Ls, from:

Ls = Lg + D/12 ft

3. For each assumed value of D, determine the

effective length, Lo, that satisfies the liquid
capacity constraint from D2L equation and
calculate the seam-to-seam length, Ls, from:
Ls = (4/3)*Lo ft

4. For each value of D used, compare the

values of Lg and Lo to determine whether
the gas capacity constraint or the oil
capacity constraint governs the design of
the separator. Of course, the larger
required length governs the design.
5. Select reasonable combinations of D and Ls
such that the slenderness ratio (SR) is in the
range of 3 to 5. The cost and availability
would then determine the final selection.
Three-Phase Vertical Separators by dividing the oil flow rate by the cross-
sectional area of flow. If Qo is the oil rate (in
BPD), then
𝑄𝑜 𝑥 5.61 𝑥4 𝑥 144
𝑢𝑜 = ft/s
24 𝑥 3600𝑥𝜋𝐷2
𝑄𝑜
𝑢𝑜 = 0.0119 ft/s
𝐷2

The water droplet settling velocity:

∆ϒ∗(𝑑𝑚 )2
𝑢𝑤 = 1.787 ∗ 10−6 ∗ ft/s
𝜇𝑜

2 6686.𝑄𝑜 .𝜇𝑜
𝐷𝑚𝑖𝑛 = (𝛾 2 in2
0 −𝛾𝑊 ).𝑑𝑚

3. Gas Capacity Constraint:

0.5
2 𝑇𝑍 (ρo -ρg ) dm
𝐷𝑚𝑖𝑛 = 5508 ∗ 𝑄𝑔 ∗ ∗ [ . ]
𝑃 ρg Cd

𝜌𝑐 𝑢2 4. Liquid Retention Time (Capacity) Constraint:

𝐹𝑑 = 𝐶𝑑 ∗ 0.25 ∗ 𝜋 ∗ 𝑑 2 ∗ Vo = (1/12)3*0.25*π* D2Ho
2
Vw = (1/12)3*0.25*π* D2Hw
where Fg is given by
Vo + Vw = 4.543 × 10–4 D2(Ho + Hw) ft3
𝜋 5.61
𝐹𝑔 = ∗ 𝑑 3 ∗ (∆𝜌)𝑔 Vo = Q o . .t
6 24*60 0
5.61
𝑉𝑤 = 𝑄𝑤 . .𝑡
24 24𝑔 24∗60 𝑤
𝐶𝑑 = = 𝑉𝑜 + 𝑉𝐻 = 3.986 ∗ 10−13 (𝑄𝑜 𝑡0 + 𝑄𝑤 𝑡𝑤 )
𝑅𝑒 𝜌𝑑𝑢
(H𝑜 + H𝑤 ). 𝐷 2 = 8.576(𝑄𝑜 𝑡0 + 𝑄𝑤 𝑡𝑤 ) in3
𝐹𝑑 = 3𝜋𝜇′𝑑𝑢
In summary then, the diameter and seam-to-
∆𝜌𝑑 2 seam length of vertical three phase separators
𝑢=
18 are determined as follows:
∆𝜌∗(3.281∗10−6 ∗𝑑𝑚 )2
𝑢= ft/s 1. Determine the minimum diameter that
18
satisfies the water droplets settling
𝑢 = 2.684 ∗ 10−6 ∗ ∆𝜌 ∗ (𝑑𝑚 )2 ft/s constraint from
2 6686.𝑄𝑜 .𝜇𝑜
𝑢 = 1.787 ∗ 10−6 ∗ ∆ϒ ∗ (𝑑𝑚 )2 ft/s 𝐷𝑚𝑖𝑛 = (𝛾 2 in2
0 −𝛾𝑊 ).𝑑𝑚

Where Δγ = γw – γo. γo and γw are the specific 2. Determine the minimum diameter that
gravity of oil and water, respectively. satisfies the gas capacity constraint from
0.5
2 𝑇𝑍 (ρo -ρg ) dm
2. Water Droplets Settling Constraint: The 𝐷𝑚𝑖𝑛 = 5508 ∗ 𝑄𝑔 ∗ ∗ [ . ]
𝑃 ρg Cd
condition for the settling and separation of
3. The larger of the two minimum diameters
water droplets from the oil is established by
determined in steps 1 and 2 is then
equating the average upward velocity of the
considered as the minimum allowable vessel
oil phase, uo, to the downward settling
diameter.
velocity of the water droplets of a given size,
uw. The average velocity of the oil is obtained
4. For various values of diameter larger than can, therefore, be summarized in the following
the minimum allowable vessel diameter, steps:
Determine combinations of diameters and
1. Determine the value of Aw/A from:
liquid heights. 𝐴𝑤 𝑄𝑤 𝑡𝑤
= 0.5
(H𝑜 + H𝑤 ). 𝐷 2 = 8.576 ∗ (𝑄𝑜 𝑡0 + 𝑄𝑤 𝑡𝑤 ) in3 𝐴 𝑄𝑜 𝑡𝑜 +𝑄𝑤 𝑡𝑤
5. For each combination, determine the seam- 2. Use Figure 1 to determine the value of Ho/D
to-seam length from the following: for the calculated value of Aw/A.
If D < 36 in,
Figure 1: Ho/D ratio as a function Aw/Atotal
then Ls =(1/12)*(Ho+ Hw+76) ft
If D > 36 in,
then Ls =(1/12)*(Ho+ Hw+D+40) ft

Three-Phase Horizontal Separators

3. Determine the maximum oil pad

thickness, Ho,max from the following
equation with dm equal to 500 μm.
1.28∗10−3 𝑡𝑜 (𝛾𝑜 −𝛾𝑤 )𝑑𝑚 2
𝐻𝑜,𝑚𝑎𝑥 = in.
𝜇𝑜
4. Determine Dmax from:
1. Water Droplet Settling Constraint: 𝐻𝑜,𝑚𝑎𝑥
𝐷𝑚𝑎𝑥 = in.
1 (𝐻𝑜 /12) 𝐻𝑜 /𝐷
𝑡𝑤𝑑 = . min 5. For diameters smaller than Dmax,
60 1.787∗10−6 ∆𝛾dm
𝜇𝑜
1.28∗10−3 𝑡𝑜 (𝛾𝑜 −𝛾𝑤 )𝑑𝑚 2 determine the combinations of D and L
𝐻𝑜,𝑚𝑎𝑥 =
𝜇𝑜 that satisfy the gas capacity constraint,
𝐴𝑤 1 2𝐻𝑜 2𝐻𝑜 4𝐻𝑜 2 substituting 100 μm for dm.
= [cos −1 ( ) −( )(1 − 2 )−0.5 ] 𝑄𝑔 𝑇𝑍 ρg 𝐶𝑑 0.5
𝐴 𝜋 𝐷 𝐷 𝐷 𝐿𝐷 = 422 .( . ) in.ft
𝑃 ρo -ρg 𝑑𝑚
𝐴𝑤 𝐴𝑤
= 0.5 6. For diameters smaller than Dmax,
𝐴 𝐴𝑜 + 𝐴𝑤
𝐴𝑤 𝑄𝑤 𝑡𝑤 determine the combinations of D and L
= 0.5 that satisfy the retention time
𝐴 𝑄𝑜 𝑡𝑜 + 𝑄𝑤 𝑡𝑤
𝐷𝑚𝑎𝑥 =
𝐻𝑜,𝑚𝑎𝑥
in. constraint from:
𝐻𝑜 /𝐷
D2L = 1.429(Qoto+ Qwtw) in2.ft
2. Gas Capacity Constraint
7. Compare the results obtained in steps 5
𝑄𝑔 𝑇𝑍 ρg 𝐶𝑑 0.5
𝐿𝐷 = 422 .( . ) in.ft and 6 and determine whether the gas
𝑃 ρo -ρg 𝑑𝑚
3. Retention Time Constraint capacity or retention time (liquid
D2L = 1.429(Qoto+ Qwtw) in2.ft capacity) governs the separator design.
8. If the gas capacity governs the design,
The procedure for determining the diameter determine the seam-to-seam length of
and length of a three-phase horizontal separator
the separator, Ls, from Ls =L + D/12 If
the liquid retention time (liquid
capacity) governs the design, determine
Ls from Ls = (4/3)*L
9. Recommend a reasonable diameter and
length with a slenderness ratio in the
range of 3–5. In making the final
selection, considerations such as cost
and availability will be important. It
should be mentioned that, in some
cases, the slenderness ratio might be
different from the range of 3–5. In such
cases, especially when the slenderness
ratio is larger than 5, internal baffles
should be installed to act as wave
breakers in order to stabilize the gas–
liquid interface.

Notes:

T in absolute R
P in Psia
Densities in lb/ft3
Specific gravities  are dimensionless
Qo and Qw in bbl/day
Qg in MMSCFD
Droplet sizes dm,w and dm,o in microns
Diameter (D) in inches
LD in in.ft and D2L in in2.ft
Ho in inches
Areas in ft2
L in ft

Check resource no. 2 for further information.

Svrcek Method 7. Calculate the height from low liquid level to
Vertical 2 phase separator[4] normal liquid level, HH (1 ft minimum)
HH = VH / (π*0.25)*Dv2 ft
8. Calculate the height from normal liquid
level to high liquid level, Hs (or high level
alarm)
Hs = Vs / (π*0.25)*Dv2 ft
9. Calculate the height from high liquid level to
the centerline of the inlet nozzle, HLIN:
HLIN = 12 + dN , in (with inlet divertor)
HLIN = 12 + 0.5* dN, in (without inlet divertor)
10. Calculate the disengagement height, from
the centerline of the inlet nozzle to, HD:
a. The vessel top tangent line if there is
no mist eliminator or
b. The bottom of the demister
1. Calculate vertical terminal vapor velocity: HD = 0.5*dv or minimum of:
UT = K* ((ρL- ρv)/ (ρv))1/2 HD = 36 + 0.5*dv (without mist
Uv =0.75* UT eliminator)
HD = 24 + 0.5*dv (with mist eliminator)
(Table 5 for K values)
11. If there is a mist eliminator, take 6 in. for
2. Calculate the vapor volumetric flow rate the mist eliminator pad and take 1 ft from
Qv= Wv / (3600* ρv) ft3/s the top of the mist eliminator to the top
tangent line of the vessel.
3. Calculate the vessel (inside) diameter, if
12. Calculate the total height, HT, of the vessel
there’s a mist eliminator and 3-6 inches to
HT = HLLL + HH + Hs + HLIN + HD + HML , ft
accommodate a support ring and round it up
to the nearest 6 in.
Dvd = (4*Qv / π*Uv) ft

4. Calculate the liquid volumetric flow rate

Select holdup time from (Table 7) and
calculate the holdup volume

VH = TH * QL ft3

5. If surge volume isn’t specified. Select a

surge time from (Table 7) and calculate
surge volume
VS = TS * QL ft3

6. Obtain low liquid level height, HLLL from

(Table 8)
Horizontal two phase separator [4] Using Hv/D, obtain Av/At using table 11 and
calculate Av.
10. Calculate the minimum length to
accommodate the liquid holdup/surge:
L = (VH + VS) / (AT – Av –ALLL) , ft
11. Calculate the liquid dropout time,
Φ = Hv / Uv
12. Calculate the actual vapor velocity UVA:
UVA = Qv/Av ft/s
13. Calculate the minimum length required for
vapor-liquid disengagement, Lmin:
Lmin = UVA Φ , ft
1. Calculate the vapor volumetric flow rate 14. Ensure that 1.2 > L/Lmin > 0.8 by
2. Calculate the liquid volumetric flow rate manipulating Hv (increase if L/Lmin >1.2 and
3. Calculate vertical terminal vapor velocity: vice versa), make sure Hv doesn’t reach Hv
UT = K* ((ρL- ρv)/ (ρv))1/2 ft/s minimum specified in step 9. Manipulate D
Uv =0.75* UT ft/s if Hv minimum is reached.
15. Calculate the thickness of the shell and
(Table 5 for K values)
heads according to (Table 9).
4. Select holdup time from (Table 7) and 16. Calculate the surface area of the shell and
calculate the holdup volume heads according to (Table 9).
VH = TH * QL ft3 17. Calculate the approximate vessel weight
5. If surge volume isn’t specified. Select a according to (Table 9).
surge time from (Table 7) and calculate 18. Increase/Decrease D until L/D is within [1.5
surge volume – 6]
VS = TS * QL ft3 19. With the optimum vessel size (minimum
weight), calculate normal and high liquid
6. Obtain an estimate L/D from table 10 and levels.
initially calculate the diameter according to: ANLL = ALLL + VH /L
Calculate the total cross section area With ANLL/AT obtain HNLL from (Table 11)
7. Calculate the low liquid level height HLLL, HHLL = D - Hv
using (Table 8) or
HLLL = 0.5D+ 7 in.
Where D in ft and round up to the nearest in, if
D < 4 ft, HLLL = 9 in.
8. Using HLLL/D, obtain ALLL/AT using table 11
and calculate the low liquid area, ALLL.
9. If there’s no mist eliminator pad, the
minimum height of the vapor
disengagement area (Av) is the larger of
0.2D or 1 ft. if there’s a mist eliminator, the
minimum height is the larger of 0.2D or 2 ft.
Vertical three phase separator [5] QLL = WLL / 60 ρL
QHL = WHL / 60 ρH
7. Assume HL = 1 ft (minimum) nad calculate
settling time for the heavy liquid droplets to
settle throught the distance
tHL = 12HL / UHL
8. Assume HL = 1 ft (minimum) and calculate
settling time for the heavy liquid droplets to
settle through the distance
tLH = 12HH / ULH
9. If there is a baffle plate, calculate the area:
a. Calculate ρL- ρv
b. Assume HT (9 in minimum) and
Calculate HL + HR
c. Use (Figure 7) to obtain G
1. Calculate vertical terminal vapor velocity: d. Calculate AD
UT = K* ((ρL- ρv)/ (ρv))1/2 e. Assume WD = 4 in.
Uv =0.75* UT f. Calculate WD/D
g. Use table 11 to determine AD/A
(Table 5 for K values)
h. Calculate A = 0.25*π*D2
2. Calculate the vapor volumetric flow rate i. Calculate AD
Qv= Wv / (3600* ρv) ,ft3/s j. Select the larger value of AD
3. Calculate the vessel (inside) diameter, if k. Calculate the area of the baffle plate =
there’s a mist eliminator and 3-6 inches to setting area for the light liquid; AL = A –
accommodate a support ring and round it AD
up to the nearest 6 in. 10. Calculate the residence time of each phase
Dvd = (4*Qv / π*Uv) ,ft based on the volumes occupied by the light
and heavy phases:
4. Calculate the settling velocity of the heavy ϴLL = HLAL/QLL
liquid out of the light liquid using Stoke’s ϴHL = HHAH/QHL
law (the maximum is 10 in./min)
UHL = Ks (ρH- ρL)/μL Make sure that ϴLL > tHL and ϴHL > tLH.
Where Ks is obtained from (Table 12) or is Otherwise, repeat steps from 7 to 10.
calculated 11. Calculate the height of the light liquid above
5. Calculate the settling velocity of the heavy the outlet (holdup height) based on the
liquid out of the light liquid using Stoke’s required holdup time:
law (the maximum is 10 in./min) HR = QLLTH/AL and check the value with the
ULH = Ks (ρH- ρL)/μH previously assumed.
Where Ks is obtained from (Table 12) or is 12. Calculate the vessel height using these
calculated guidelines:
6. Calculate the light and heavy liquid HA = 6 in. minimum
volumetric flow rates, QLL and QHL HBN = 0.5dN + greater of (2 ft or Hs +0.5 ft)
HD = 0.5D or a minimum of eliminator. Using Hv/D in (Table 11), obtain
HD = 36 + 0.5*dN (without mist eliminator) Av/AT and calculate Av.
HD = 24 + 0.5*dN (with mist eliminator) 7. Set heights of the heavy and light liquids,
HHL and HLL
HT = HH + HL + HR + HA + HBN + HD
8. Find (AHL + ALL) /AT, Using (HHL + HLL) /D in
If a mist eliminator is used, additional
(Table 11), and calculate AHL+ALL
height is added, as shown in the section
9. Calculate the minimum length to
figure.
accommodate the liquid holdup/surge:
L = (VH + VS) / (AT – Av –(AHL+ALL) ), ft
Horizontal 3 phase (No weir, No boot) [5]
10. Calculate the liquid dropout time:
Φ = Hv/Uv
11. Calculate the actual vapor velocity:
UVA = Qv/Av
12. Calculate the minimum length required for
the vapor/liquid separation:
Lmin = UVA φ
13. Ensure that 1.2 > L/Lmin > 0.8 by
manipulating Hv (increase if L/Lmin >1.2
and vice versa), make sure Hv doesn’t reach
Hv minimum specified in step 9. Manipulate
D if Hv minimum is reached.
14. Calculate the setting velocities of the heavy
1. Calculate the vapor volumetric flow rate
liquid out of the light liquid phase and the
Qv= Wv / (3600* ρv) ,ft3/s
light liquid out of the heavy phase:
2. Calculate the light and the heavy liquid UHL = Ks (ρH- ρL)/μL
volumetric flow rates, QLL and QHL ULH = Ks (ρH- ρL)/μH
3. Calculate vertical terminal vapor velocity: 15. Calculate the settling times of the heavy
UT = K* ((ρL- ρv)/ (ρv))1/2 liquid out of the light phase and the light
Uv =0.75* UT liquid out of the heavy phase:
(Table 5 for K values) tHL = 12 (D-Hv – HHL)/UHL
tLH = 12 HHL / ULH
4. Select holdup time from table 7 and
16. Calculate the residence times of the light
calculate the holdup volume
and heavy liquids
VH = TH * QL ,ft3 , If surge volume isn’t specified.
ϴHL = AHLL/QHL
Select a surge time from table 7 and calculate
ϴLL = (AT – Av – AHL)L/QLL
surge volume
17. Ensure that ϴHL > tLH and ϴLL > tLH.
VS = TS * QL ,ft3
Otherwise, Increase L.
5. Obtain L/D from (Table 10) and initially L = max(tLHQHL/AHL, tHLQLL/(AT – Av – AHL))
calculate the diameter according to: 18. Calculate L/D, ensure L/D is between [1.5 –
D = (4(VH + VS)/(0.5π(L/D)))1/3 6] by manipulating D.
6. Set the vapor space height, Hv, the larger of 19. Calculate the thickness of the shell and
0.2D or 2 ft; 1ft if there is no mist heads according to table 9.
20. Calculate the surface area of the shell and 6. Set the vapor space height, Hv, the larger of
heads according to (Table 9). 0.2D or 2 ft; 1ft if there is no mist
21. Calculate the approximate vessel weight eliminator. Using Hv/D in (Table 11), obtain
according to (Table 9). Av/AT and calculate Av.
22. Using optimum vessel size (minimum 7. Set the light liquid heights in the vessel and
weight), calculate the normal and high the boot HLLV, HLLB. (1 ft each)
liquid levels: 8. Calculate the cross-sectional area of the
HHLL = D - Hv light liquid above the bottom of the vessel,
ANLL = (AHL + ALL) + VH ALLV, using HLLV/D in (Table 11).
Obtain HNLL using (Table 11) with the value 9. Calculate the minimum length to
of ANLL/AT. accommodate the liquid holdup/surge:
Horizontal 3 phase (boot) [5] L = (VH + VS) / (AT – Av –ALLV) ft
10. Calculate the liquid dropout time, φ, φ =
Hv/Uv
11. Calculate the actual vapor velocity:
UVA = Qv/Av
12. Calculate the minimum length required for
the vapor/liquid separation:
Lmin = UVA φ
13. Ensure that 1.2 > L/Lmin > 0.8 by
manipulating Hv (increase if L/Lmin >1.2 and
vice versa), make sure Hv doesn’t reach Hv
minimum specified in step 6. Manipulate D
1. Calculate the vapor volumetric flow rate if Hv minimum is reached.
Qv= Wv / (3600* ρv) ,ft3/s 14. Calculate the setting velocities of the heavy
liquid out of the light liquid phase:
2. Calculate the light and the heavy liquid
UHL = Ks (ρH- ρL)/μL
volumetric flow rates, QLL and QHL
15. Calculate the settling times of the heavy
3. Calculate vertical terminal vapor velocity:
liquid out og the light phase and the light
UT = K* ((ρL- ρv)/ (ρv))1/2
liquid out of the heavy phase:
Uv =0.75* UT
tHL = 12 (D-Hv – HHL)/UHL
(Table 5 for K values) 16. Calculate the residence times of the light
4. Select holdup time from table 7 and and heavy liquids
calculate the holdup volume ϴLL = (AT – Av)L/QLL
VH = TH * QL ,ft3 , If surge volume isn’t specified. 17. Ensure ϴLL > tHL, Otherwise, Increase vessel
Select a surge time from table 2 and calculate length.
surge volume L = tHLQLL/ (AT – Av)
VS = TS * QL ,ft3 18. Calculate L/D, ensure L/D is between [1.5 –
6] by manipulating D.
5. Obtain L/D from (Table 10) and initially
19. Calculate the thickness of the shell and
calculate the diameter according to:
heads according to (Table 9).
D = (4(VH + VS)/(0.6π(L/D)))1/3
20. Calculate the surface area of the shell and 6. Set the vapor space height, Hv, the larger of
heads according to (Table 9). 0.2D or 2 ft; 1ft if there is no mist
21. Calculate the approximate vessel weight eliminator. Using Hv/D in table 11, obtain
according to (Table 9). Av/AT and calculate Av.
22. Using optimum vessel size (minimum 7. Calculate the low liquid level in the light
wright), calculate the normal and high liquid liquid compartment using:
levels: HLLL = 0.5D + 7 (D in feet, HLLL is inches), If D
HHLL = D - Hv < 4.0, HLLL = 9 inches
ANLL = (AHL + ALL) + VH/: Calculate ALLL form HLLL/D (both in feet or
Obtain HNLL using (Table 11) with the value inches)
of ANLL/AT. 8. Calculate the weir height:
Horizontal 3 phase (Weir) [5] Hw = D - Hv
If Hw < 2 ft increase D and repeat
calculations from Step 6.
9. Calculate the minimum length of the light
liquid compartment to accommodate the
liquid holdup/surge:
L2 = (VH + VS) / (AT – Av –ALLL) , ft
Round to the nearest 0.5 ft. With L2
minimum as dN + 12
10. Set the interface height as Hw/2, obtaining
the heights of heavy and light liquids HHL,
HLL.
1. Calculate the vapor volumetric flow rate 11. For the liquid settling compartment,
Qv= Wv / (3600* ρv) ,ft3/s calculate the cross sectional area if the
heavy liquid by HHL/D and calculate the
2. Calculate the light and the heavy liquid
cross sectional area of the liquid liquid
volumetric flow rates, QLL and QHL
from:
3. Calculate vertical terminal vapor velocity:
ALL = AT – Av – AHL
UT = K* ((ρL- ρv)/ (ρv))1/2
12. Calculate the setting velocities of the heavy
Uv =0.75* UT
liquid out of the light liquid phase and the
4. Select holdup time from table 7 and light liquid out of the heavy phase:
calculate the holdup volume UHL = Ks (ρH- ρL)/μL
VH = TH * QL ,ft3 , If surge volume isn’t specified. ULH = Ks (ρH- ρL)/μH
Select a surge time from table 7 and calculate 13. Calculate the settling times of the heavy
surge volume liquid out of the light phase and the light
VS = TS * QL ,ft3 liquid out of the heavy phase:
5. Obtain L/D from table 10 and initially tHL = 12 (D-Hv – HHL)/UHL
calculate the diameter according to: tLH = 12 HHL / ULH
D = (16(VH + VS)/(0.6π(L/D)))1/3 Note: Check resource no. 1 and Ref. 4 and 5 for
further information.
TOTAL Supplementary Tables and Charts

Figure 2: Vs (Liquid Settling Velocity ms-1)

Figure 3: Drag Coefficient (C) Vs Re or C(Re^2)
Figure 4: Relationship Between Chordal Height and Circular segmental Area
Figure 5: Vessel Weight Estimation
Table 4: For gauging horizontal cylinderical tanks - Flat Ends
Figure 6: Relationship between chordal height and circular segment area
Svrcek Supplementary Tables and Charts
Table 5: York Demister Equation Table 7: Holdup and Surge Times
Mist Eliminator P in psia Services Hold-up times Surge Time
(NLL-HLL) (NLL-LLL)
1≤P <15 K = 0.1821+0.0029P+0.046ln(P) min min
15 ≤P < 40 K = 0.35 Unit Feed Drum 10 5
40 ≤ P < 5500 K = 0.43 – 0.023ln(P) Separators
GPSA Feed to column 5 3
Feed to other drum or 5 2
0 ≤ P < 1500 K = 0.35 – 0.01(P-100/100) P in psig
tankage
• Most vapors under vacuum K =0.2 with pump or through 5 2
• For glycol and amine solutions multiply K by 0.6- exchanger
0.8 without pump 2 1
• For vertical vessels without mist eliminator. Feed to fired heater 10 3
divide K by 2 Reflux or product accumulator
• For compressor suction scrubber, mole sieve Reflux only 3 2
scrubbers, mole sieve scrubbers and expander Reflux and product +3 +2
inlet separator multiply K by 0.7-0.8 based on reflux (3 min) + appropriate holdup time of
overhead product
Theoretical (no mist eliminator) Columns bottoms
Feed to another column 5 2
4𝑔𝐷𝑝 Feed to other drum or tankage
𝐾= √
3𝐶𝐷 with pump or through 5 2
exchanger
𝐶𝐷 = exp (𝑌) without pump 2 1
𝑌 = 8.411 − 2.243𝑋 + 0.273𝑋 2 − 1.865𝐸 −2 𝑋 3 feed to fired boiler 5-8 2-4
+ 5.201𝐸 −4 𝑋 4 based on reboiler vapor expressed as liquid (3 min) +
appropriate holdup time for the bottom product
0.95 + 8𝜌𝑣 𝐷𝑝3 (𝜌𝑙 − 𝜌𝑣 ) Compressor suction/interstage scrubber
𝑋 = ln(
𝜇𝑣2 3 mins between HLL (HLA) and HLSD
Notes: 10 min from bottom tangent line to HLA
Dp , ft Fuel gas KO drum
ρ , lb/ft3 20 ft slug in the incoming fuel gas line between NLL and HLSD
μ , cP Flare gas KO drum
1 micron = 3.28084E-6 ft 20 to 30 min. to HLL

Table 6: Factor to be multiplied in Holdup and Surge times

Personnel Factor Instrumentation Factor
Experienced 1 Well instrumented 1
Trained 1.2 Standard instrumented 1.2
Inexperienced 1.5 Poorly instrumented 1.5
Table 8: Low Liquid Level Height Table 9: Wall thickness, Surface Area and Approx. vessel Height
Vessel vertical LLL (in) Horizontal Wall Thickness Surface Area ft2
Diameter (ft) LLL (in) Shell 𝑃𝐷 πDL
+ 𝑡𝑐
< 300 psia > 300 psia 2𝑆𝐸 − 1.2𝑃
≤4 15 6 9 2:1 Elliptical 𝑃𝐷 1.09D2
+ 𝑡𝑐
6 15 6 10 Head 2𝑆𝐸 − 0.2𝑃
8 15 6 11 Hemispherical 𝑃𝐷 1.571D2
+ 𝑡𝑐
10 6 6 12 Head 4𝑆𝐸 − 0.4𝑃
Dished Head 𝑃𝐷 0.842D2
12 6 6 13 + 𝑡𝑐
16 6 6 15 𝑆𝐸 − 0.1𝑃
Appropriate 𝑙𝑏 𝑡
Inlet Nozzle Sizing 𝑊 = (490 3 )( )(𝐴𝑠 + 2𝐴𝐻 )
vessel height 𝑓𝑡 12
4𝑄𝑚 o
𝑑𝑁 ≥ ( )0.5 , ft P, design pressure, F (typically, operating pressure + (15-
𝜋60/√𝜌𝑚
30) psi or 10-15%, whichever greater
Qm = QL + QV , ft3/s T, design pressure, oF (typically, operating pressure + 25-50
o
ρm= ρL λ + ρm(1-λ), lb/ft3 F if Top> 200 oF, 250 oF

λ = QL/( QL+ QV) 1. Under 650 oF don’t reduce wall thickness

2. if overpressure caused by boiling, should be TBP
Table 10: L/D ratio guidelines
Vessel Operating Pressure (psig) L/D
D, diameter, in.
0 < P ≤ 250 1.5-3
S, allowable stress, psi
250 < P ≤ 500 3-4
E, joint efficiency, (0.6-1.0), 0.85 for spot examined joints,
500 < P 4-6
1.0 for 100% x-ray joints
tc, corrosion allowance
t, larger of ts and tH
Table 11: Cylindrical Height and area conversion
Selection on Head Types
2 3 4 2 3 4
Y = (a+cX+eX +gX +iX )/ (1.0+bX+dX +fX +X )
1. 2:1 elliptical heads are typically used when D < 15
H/D to A/AT A/AT to H/D ft and P > 100 psig.
Y = A/AT Y = H/D 2. Hemispherical heads are typically used when D>
X = H/D X = A/AT 15 ft regardless of P.
a= -4.75593E-5 a= 0.00153756 3. Dished Heads with knuckle radius = 0.6 D are
b= 3.924091 b= 26.787101 typically used when D < 15 ft and P < 100 psig
c= 0.174875 c= 3.299201
d= -6.358805 d= -22.923932 Notes:
e= 5.668973 e= 24.353518 P: Design Pressure
f= 4.018448 f= -14.844824 D: Drum diameter
g= -4.916411 g= -36.999376
h= -1.801705 h= 10.529572
i= -0.145348 i= 9.892851
Table 13: Standard Sizes As per API 12J
Table 12: Typical Ks values for liquid-liquid separations
STANDARD SEPARATOR SIZES AS PER API
Light phase Heavy Phase Min. Ks D [in] x H or L [ft]
Droplet (ft/s) 12¾ in x 5 ft
μm 12¾ in x 7½ ft
SG at 60 oF < 0.85 Water or caustic 127 0.333 12¾ in x 10 ft
SG at 60 oF < 0.85 Water or caustic 89 0.163
Water Furfural 89 0.163 16 in x 5 ft
MEK Water 89 0.163 16 in x 7½ ft
sec-Butyl alcohol Water 89 0.163 16 in x 10 ft
Methyl isobutyl Water 89 0.163
ketone 20 in x 5 ft
Nonyl alcohol Water 89 0.163 20 in x 7½ ft
20 in x 10 ft

24 in x 5 ft
24 in x 7½ ft
24 in x 10 ft

30 in x 5 ft
30 in x 7½ ft
30 in x 10 ft

36 in x 5 ft
36 in x 7½ ft
36 in x 10 ft
36 in x 15 ft

42 in x 7½ ft
42 in x 10 ft
42 in x 15 ft
Figure 7: G values as a function of density difference and the downcomer
allowable flow
48 in x 7½ ft
48 in x 10 ft
48 in x 15 ft

54 in x 7½ ft
54 in x 10 ft
54 in x 15 ft

60 in x 7½ ft
60 in x 10 ft
60 in x 15 ft
Example No. 1: Two-Phase Vertical Separator[1]

Parameter Value Unit Value Unit

Gas flow rate 27895 kg/hr 25 MMSCFD
Oil flow rate 19.8734 kg/hr 3000 Bbl/d
Gas Density 54.46 kg/m3 3.4 lbs/ft3
Oil Density 824.95 kg/m3 51.5 lbs/ft3
Gas specific gravity 0.7
Oil API 40
2
Operating pressure 57.27 kg/cm 800 Psig
o
Operating temperature 26.667 C 80 F
mol. Wt 20.3 20.3 lbs/mol
Z 0.9
Gas viscosity 0.013 cP assumed
Oil viscosity 0.45 cP assumed
surge 0 mins
holdup 1 mins
Input

API 12J Solution

Assume 10 ft Vessel height 30% Liquid Full
Using the K Table, K value is assumed 0.3 ft/s
Calculate The maximum allowable superficial velocity for
gas
𝑑𝐿 −𝑑𝐺 51.5−3.4
𝑉𝑎 = 𝐾 √ = 0.3 ∗ √ = 1.128 𝑓𝑡/𝑠
𝑑𝐺 3.4

Actual Volume flow rate of gas =

250000 𝑆𝐹𝐶𝐷 ∗ 20.3 𝑙𝑏𝑠/𝑚𝑜𝑙

= 4.552 𝑓𝑡 3 /𝑠𝑒𝑐
𝑆𝐶𝐹 𝑠𝑒𝑐 3
379.5 ∗ 86400 ∗ 3.4 𝑙𝑏/𝑓𝑡
𝑚𝑜𝑙 𝑑𝑎𝑦
4.552 𝑓𝑡/𝑠
Minimum gas flow area = = 4.035 𝑓𝑡 2
1.128 𝑓𝑡/𝑠
4.035∗144
Minimum ID of Separator = √ = 27.2 𝑖𝑛𝑐ℎ𝑒𝑠
0.7854
Nearest Std. Diameter is selected (30 inches)
Liquid Volume, V (Excluding bottom head) =
302 ∗ 0.7854 𝑖𝑛2 ∗ 3 𝑓𝑒𝑒𝑡
= 2.62 𝐵𝑏𝑙𝑠
𝑖𝑛2
144 2 ∗ 5.615 𝑓𝑡 3 /𝑏𝑏𝑙
𝑓𝑡
The Liquid Capacity of the separator:
1440 𝑉 1440 ∗ 2.62
W= = = 3772 𝐵𝑃𝐷
𝑡 1
W is higher that Oil rate (3000 BPD) thereby Liquid Capacity is satisfactory for design based on 30 in. ID x
10 ft Vertical Separator Size.
Hussien Method
Step 1: Determine gas and oil properties

2.7𝛾𝑃 2.7 ∗ 0.7 ∗ 814.5

𝜌𝑔 = = = 3.4 𝑙𝑏/𝑓𝑡 3
𝑇𝑍 540 ∗ 0.84

𝜌𝐿 = 62.4 ∗ 0.83 = 51.5 𝑙𝑏/𝑓𝑡 3

Step 2: Determine Cd
assumed drag coefficient of Cd = 1.21
0.5
𝜌𝐿 − 𝜌𝑔 𝑑𝑚
𝑢 = 0.01186 [( ) ] = 0.41 𝑓𝑡/𝑠
𝜌𝑔 𝐶𝑑
𝜌𝑔 𝑑𝑚 𝑢
𝑅𝑒 = 0.0049 = 52
𝜇𝑔
3 24
𝐶𝑑 = 0.34 + + = 1.22, Acceptable Cd assumption
𝑅𝑒 0.5 𝑅𝑒
Step 4: Check for Gas Capacity Constraint
𝑇𝑍 𝜌𝑔 𝐶𝑑 0.5
𝐷𝑚𝑖𝑛 2 = 5058𝑄𝑔 ( ) ( ) = 2059
𝑃 𝜌𝐿 − 𝜌𝑔 𝑑𝑚
∴ 𝐷𝑚𝑖𝑛 = 45.4 𝑖𝑛
Step 5: Check for liquid capacity constraint
D2H = 8.565*Qot = 25695 in2

Try different combinations of D.

Table 1
𝐷2 𝐻
D, in H, in Ls, ft SR 𝐻=
𝐷
30 28.6 8.7 3.5
36 19.8 8.0 2.7 𝐿𝑠
42 14.6 8.0 2.3
48 11.2 8.3 2.1
54 8.8 8.6 1.9
60 7.1 8.9 1.8
66 5.9 9.3 1.7
72 5.0 9.7 1.6
78 4.2 10.2 1.6
𝑆𝑅 = 𝐷 ∗ 12⁄𝐿
84 3.6 10.6 1.5 𝑠

90 3.2 11.1 1.5

Ls 8.7 ft Seam to seam length (select from Table 1)

SR 3.5 Slenderness ratio (typical value 3 to 4) (select from Table 1)
D 30.0 in Separator diameter (select from Table 1)

The solution selected has the highest SR ratio in the range of 2-4 as per GPSA
TOTAL method
Liquid-vapour settling velocity and Dmin

1. K = 0.3/3.28 m/s (GPSA’s value used earlier)

2. Vs = K* ((ρL- ρv)/ (ρv))1/2 = K* ((824.95- 54.46)/ (54.46))1/2=K*3.76 =0.091*3.76 = 0.34 m/s
3. Vm = 0.85*0.34=0.289 m/s (assuming De-rating factor of 85%)
4. Q = 27895/(54.46*3600) = 0.1423 m3/s
5. Dmin = (4*(Q/ Vm)/(π))0.5=(4*(Q/ Vm)/(π))0.5 =(4*(0.1423/ 0.289)/(π))0.5=0.79 m = 790 mm
6. Selected Diameter 800 mm
Required liquid Hold-up times
Liquid flow rate

h5: HLA – HLL 0.5 min 0.72 m3 360

h6: HLL – LLL 0.5 min 0.72 m3 360
h7: LL – LLA 0 min 0.00 m3 0.00
Nozzle Sizing
inlet flow = 0.142 m3/s Nozzle ID = 150 mm Actual velocity = 8.05 m/s
Gas flow = 0.142 m3/s Nozzle ID = 80 mm Actual velocity = 28 m/s
Liquid flow =6.69E-06m3/s Nozzle ID = 50 mm Actual velocity =0.003 m/s
Note 1: +10% increase in flows were neglected
Height Calculations φ = mm
h1 max(15 % of φ (120) or 400 mm) 400
h2 100 mm (mesh selected) 100
h3 max (50 % of ɸ (400) or 600 mm) 600
h4 400 mm + d/2 = 400 + 150/2 550
h5 min 200 mm 360
h6 360 mm 360
h7 150 mm 150
h8 150 mm 150
Total Height - 2670 (2700 mm selected)
Wall Thickness
Design Pressure P = 55.16 barg
Corrosion Allowance C = 3 mm
Diameter D = mm
Max Stress:
S= 1220 bar CS
1000 bar CS S = 1220
Joint efficiency E = 0.85
𝑃∗𝐷 55.16 ∗ 800
t= +𝐶 = + 3 = 25 𝑚𝑚
2 ∗ 𝑆 ∗ 𝐸 − 1.2𝑃 2 ∗ 1220 ∗ 0.85 − 1.2 ∗ 55.16
𝑡 𝐷
𝑚𝑖𝑛= +𝐶 = 1+3 = 4 𝑚𝑚
800
Vessel weight
t (thickness) = 25 mm shell weight = 2600*0.8 =2040 kg 2600/D from Figure 5
L = 2.7 m Heads weight = φ (m) * thickness (mm) * 20 kg= 0.8*25* 20 = 400 kg
D = 0.8 mm
Total weight = 2440 kg
Step 1: Select Desired Calculations and modify assumptions Step 2: Input operating conditions and physical props

Step 3: Check Operating conditions and physical props output Step 4: Check TOTAL output image and table
Step 5: Check Hussien’s output table Step 6: Check Svrcek output image and table

Step 7: Check API RP 12J output table

Example No. 2: Two-Phase Horizontal Separator (No Mist Eliminator)[4]

Parameter Value Unit Value Unit

Gas flow rate 66043.0 kg/hr 145600 (27.59) lb/hr (MMSCFD)
Oil flow rate 20910.6 kg/hr 46100 (5072) lb/hr (bbl/d)
Gas Density 64.23 kg/m3 4.01 lbs/ft3
Oil Density 622 kg/m3 38.83 lbs/ft3
Gas specific gravity 1.7
Oil API 95.99
2
Operating pressure 68.24 kg/cm 975 psig
o
Operating temperature 343.33 C 650 F
mol. Wt 48.27
Z 1
Gas viscosity 0.013 cP
Oil viscosity 0.24 cP
surge 5 mins
holdup 10 mins
Input

Svrcek Paper Solution

1. Calculate The vapor volumetric flow rate Assumptions for the Surge and Holdup
145600 𝑙𝑏/ℎ𝑟 times (Tables 6 and 7):
𝑄𝑣 = 𝑠 = 10.09 𝑓𝑡 3 /𝑠
3600 ∗ 4.01 𝑙𝑏/𝑓𝑡 3
ℎ Service: Unit Feed Drum
2. Calculate The liquid volumetric flow rate
Surge 10 mins, Holdup 5 mins
46100 𝑙𝑏/ℎ𝑟
𝑄𝐿 = = 19.79 𝑓𝑡 3 /𝑚𝑖𝑛
𝑚𝑖𝑛 3 Personal, Factor = 1
60 ∗ 38.83 𝑙𝑏/𝑓𝑡
ℎ
Well-Instrumented, Factor = 1
3. Calculate the vertical terminal velocity
For other examples the factors are
38.83 − 4.01 assumed both 1 and 1 too.
𝑈𝑡 = 0.13√ = 0.38 𝑓𝑡/𝑠
4.01

K = 0.13 (GPSA Value divided by two since “no mist eliminator is used”)
Uv = 0.75 Ut = 0.29 ft/s

4. Calculate the holdup volume

VH =tH*QL = 10 mins * 19.79 ft3/min = 197.9 ft3
5. Calculate the surge volume
Vs = ts*QL = 5 mins * 19.79 ft3/min = 98.95 ft3
6. Assume L/D = 5.0. initially set the diameter AT = π/4 *(5 ft)2 = 19.63 ft3
7. Calculate low liquid height:
HLLL=0.5D + 7 =0.5*5+7=9.5 in (use 10 in.)
8. Calculate the low liquid level area:
HLLL/D = 0.167
Using table 6, ALLL/AT = 0.11
ALLL = 0.11*19.63=2.16 ft2
9. Set Hv = 2 ft, Hv/D = 2/5 = 0.4
From Table 6, Av/AT = 0.374
Av = 0.373*19.63 = 7.34 ft2
10.Calculate the length to accommodate holdup/surge

𝑉𝐻 + 𝑉𝑠 197.9 + 98.95
𝐿= = = 29.3, 𝑠𝑎𝑦 29.5 𝑓𝑡
𝐴 𝑇 − 𝐴𝑣 − 𝐴𝐿𝐿𝐿 19.63 − 7.34 − 2.16

11.Calculate the liquid dropout time

2 𝑓𝑡
∅= = 6.9 𝑠
0.29 𝑓𝑡/𝑠

12.Calculate the actual vapor velocity

10.09 𝑓𝑡 3 /𝑠
𝑈𝑉𝐴 = = 1.37 𝑓𝑡/𝑠
7.34 𝑓𝑡 2

13.Calculate Lmin = 1.37 ft/s * 6.9 s = 9.45 ft

14.L>>Lmin but Hv is minimum and cannot be reduced so L cannot be reduced.
L/D = 29.5/5 = 5.9
15. Calculate the thickness of the shell and heads according to table 9:
• Table 9. use 2:1 elliptical
• Assume E = 0.85
• Assume SA 516 70 Carbon steel
Design Temp. = 650 oF
• From 9, S = 17500 psi
• CA = 1/16 in (Assumed)
• P = 975 * 1.1 = 1072 psig
• ts = 2.375 in

1072 ∗ 60 1
𝑡𝑠 = + = 2.322 𝑖𝑛
2 ∗ 17500 ∗ 0.85 − 1.2 ∗ 1072 16

• tH = 2.25 in

1072 ∗ 60 1
𝑡𝐻 = + = 2.240 𝑖𝑛
2 ∗ 17500 ∗ 0.85 − 0.2 ∗ 1072 16

16.Calculate the surface area of the shell and heads according to table 9:
As = π*5 ft*29.5 ft = 463.38 ft2 and
AH = 1.09 * (5 ft)2 = 27.25 ft2
17.Calculate the approximate vessel weight = 50,224 lb

𝑙𝑏 2.375 𝑖𝑛
𝑊 = 490 (463.38 𝑓𝑡 2 + 2 ∗ 27.25)
𝑓𝑡 3 12 𝑖𝑛
𝑓𝑡

18.Try D = 5.5 ft and repeat calculations until minimum weight of shell and head is obtained

API Method

Step 1: Determine gas and oil properties

2.7𝛾𝑃 2.7 ∗ 1.68 ∗ 1000.5

𝜌𝑔 = = = 4.1 𝑙𝑏/𝑓𝑡 3
𝑇𝑍 649.4 ∗ 1

𝜌𝐿 = 62.4 ∗ 0.62 = 38.8 𝑙𝑏/𝑓𝑡 3

Step 2: Calculate H/D
𝜌𝐿 − 𝜌𝑔 38.8 − 4.1
𝑉𝑎 = 𝐾 √ = 0.5 ∗ √ = 1.456 𝑓𝑡/𝑠
𝜌𝑔 4.1
Actual volume flow of gas
𝑄𝑔,𝑎𝑐𝑡𝑢𝑎𝑙 = 𝑄𝑔 ∗ 1000000 ∗ 𝑀. 𝑤𝑡/379.5/86400/𝜌𝑔 = 10.02
Minimum gas flow area
𝑄𝑔,𝑎𝑐𝑡𝑢𝑎𝑙
𝐴𝑔,𝑚𝑖𝑛 = = 6.88 𝑓𝑡 2
𝑉𝑎
4
𝐷𝑚𝑖𝑛 = √ ∗ 12 = 9 𝑖𝑛,
𝜋∗𝐴𝑔,𝑚𝑖𝑛
You’ll Iterate D till D = 66 in where W > Qo and L/D in 2.5-5 range.
L = 27 ft (Assumed Shell Length)
L/D = 12*66/120 = 4.9
Step 2: Calculate V and W
Liquid volume (excluding bottom head)
𝑉 = ((𝐷/12/2)^2 ∗ 𝐴𝐶𝑂𝑆((𝐷/12/2 − 𝐷/12 ∗ 𝑁𝐿𝐿)/(𝐷/12/2)) − ((𝐷/12/2) − (𝐷/12 ∗ 𝑁𝐿𝐿)
∗ 𝑆𝑄𝑅𝑇((2 ∗ 𝐷/12/2 ∗ 𝐷/12 ∗ 𝑁𝐿𝐿 − (𝐷/12 ∗ 𝑁𝐿𝐿)^2)))) ∗ (𝐿) ∗ 0.17811 = 80 𝑏𝑏𝑙
𝑉 ∗ 1440
𝑊= = 7706 𝑏𝑝𝑑
𝑡
Hussien Method

Step 1: Determine gas and oil properties

2.7𝛾𝑃 2.7 ∗ 1.68 ∗ 1000.5

𝜌𝑔 = = = 4.1 𝑙𝑏/𝑓𝑡 3
𝑇𝑍 649.4 ∗ 1

𝜌𝐿 = 62.4 ∗ 0.62 = 38.8 𝑙𝑏/𝑓𝑡 3

Step 2: Determine Cd
assumed drag coefficient of Cd = 1.33
0.5
𝜌𝐿 − 𝜌𝑔 𝑑𝑚
𝑢 = 0.01186 [( ) ] = 0.31
𝜌𝑔 𝐶𝑑
𝜌𝑔 𝑑𝑚 𝑢
𝑅𝑒 = 0.0049 = 48 𝐿𝑔 = 𝐿𝐷/𝐷
𝜇𝑔
3 24
𝐶𝑑 = 0.34 + + = 1.28, Acceptable Cd assumption 𝐿𝑠 (𝐺𝑎𝑠) = 𝐿𝑔 + 𝐷/12
𝑅𝑒 0.5 𝑅𝑒
Step 3: Check for Gas Capacity Constraint
0.5 𝐿𝑜 = 𝐷2 𝐿/𝐷2
𝑄𝑔 𝑇𝑍 𝜌𝐿 − 𝜌𝑔 𝐶𝑑
𝐿𝐷 = 422 ∗ ( ) ∗ [( ) ( )] = 501 𝑖𝑛. 𝑓𝑡
𝑃 𝜌𝑔 𝑑𝑚 𝐿𝑠 (𝑂𝑖𝑙) = 2.5 + 𝐿𝑜 , If D > 30 in
Step 4: Check for Liquid Capacity Constraint
4
𝐿𝑠 (𝑂𝑖𝑙 ) = ∗ 𝐿𝑜 , If D ≤ 30 in
3
1.428
𝐷2 𝐿 = 1.428𝑄𝑜 𝑡 = ∗ 50% ∗ 5072 ∗ 15 = 108642.2 𝑖𝑛2 𝑓𝑡
0.5
Try different combinations of D.
TABLE 1
SR = 12 ls
D, in Lg, ft Ls, ft (gas) Lo, ft Ls, ft (oil) Governing case (oil or gas)/D

6 83.57 84.07 3017.84 4023.8 Liquid Capacity 8047.57

12 41.79 42.79 754.46 1005.9 Liquid Capacity 1005.95

24 20.89 22.89 188.62 251.5 Liquid Capacity 125.74

30 16.71 19.21 120.71 161.0 Liquid Capacity 64.38

36 13.93 16.93 83.83 86.3 Liquid Capacity 28.78

42 11.94 15.44 61.59 64.1 Liquid Capacity 18.31

48 10.45 14.45 47.15 49.7 Liquid Capacity 12.41

54 9.29 13.79 37.26 39.8 Liquid Capacity 8.83

60 8.36 13.36 30.18 32.7 Liquid Capacity 6.54

66 7.60 13.10 24.94 27.4 Liquid Capacity 4.99

72 6.96 12.96 20.96 23.5 Liquid Capacity 3.91

Ls 27.4 ft Seam to seam length (select from Table 1)

SR 5.0 Slenderness ratio (typical value 3 to 5) (select from Table 1)
D 66 in Separator diameter (select from Table 1)
The solution selected has the highest SR ratio in the range of 2.5-5 as per GPSA
TOTAL
Liquid-vapour settling velocity and Dmin

1. K = 0.13/3.28 m/s (GPSA’s value used earlier)

2. Vs = K* ((ρL- ρv)/ (ρv))1/2 = K* ((622- 64.23)/ (64.23))1/2=K*2.95 =0.04*2.95 = 0.118 m/s
3. Vm = Vs * F *( L/D) = 0.118* 0.85*4 =0.4 m/s
(assuming a De-rating factor of 85% and (assumed L/D = 4)
4. Q =66043.0/(64.23*3600) = 0.285 m3/s
5. Av = Qg/Vm = 0.285/0.4 = 0.71 m2
Nozzle Sizing
inlet flow = 0.295 m3/s Nozzle ID = 200 mm Actual velocity = 9.4 m/s
Gas flow = 0.285 m3/s Nozzle ID = 150 mm Actual velocity =16.1 m/s
Liquid flow =20910.6/3600*622= 0.01 m3/s Nozzle ID = 80 mm Actual velocity =1.98 m/s

Drum Sizing
For first trial tres = 15 mins volume required =15 * QL =15*0.01*60= 9 m3

Trial 1 2
Selected h/D 0.7 assumed 0.6
modified
2 0.71
Vapor Area Av m Step 1 1.57
Step 6
% Total Area 70.5 Step 2 61.83
Step 5
2
Total Area Ar m 1.017 Step 3 2.54
Step 4
Liquid Area Al m2 0.305 Step 4 0.97
Step 7
Calculated drum φ mm 1130 Step 5
Selected Drum D mm 1200 Step 6 1800 ↗ Step 1
L/D (3 – 4 ) 4 assumed 4
Flowpath length L mm 4800 Step 7 7200 Step 2
T-T length L’ mm 5325 Step 8 7725 Step 3
HLL height mm 840 Step 9 1080 Step 8
3
Vol @ HLL m 4.8 Step 10 12.86 Step 9
LLL height mm 300 Step 11 300 Step 10
3
Vol @ LLL m 1.24 Step 12 2.22 Step 11
∆Vol m3 3.56 Step 13 10.64 Step 12
Calculated tres mins 5.93 Step 14 17.73 Step 13
Notes Small vessel Good enough!
tan-tan length L’ = L + 1.5φ1 + 1.5φ2 = 7200 + 1.5 * (150+200) = 7725 mm
Selected Diameter: 1650 mm x 7725 mm tan-tan length
Step 1: Select Desired Calculations and modify assumptions Step 2: Input operating conditions and physical props

Step 3: Check Operating conditions and physical props output Step 4: Check TOTAL output image and table
Step 5: Check Hussien’s output table Step 6: Check Svrcek output image and table

Step 7: Check API RP 12J output table

Example No. 3: Three-Phase Vertical Separator[5]

Parameter Value Unit Value Unit

Gas flow rate 188240.8 kg/hr 415000 (155) lb/hr (MMSCFD)
Oil flow rate 7484.3 kg/hr 16500 (1307.27) lb/hr (bbl/day)
Gas Density 11.17 kg/m3 0.6973 lbs/ft3
Oil Density 864.2 kg/m3 53.95 lbs/ft3
Water Density 994.9 kg/m3
Gas specific gravity 0.8
Oil API 32.24
Operating pressure 11.6 Kg/cm2 165 psig
o
Operating temperature 26.67 C 60 F
mol. Wt 24.488
Z 1
Gas viscosity 0.013
Oil viscosity 0.63
water flow rate 589.67 kg/hr 1300 (89) lb/hr (bbl/day)
surge 5 tw 5 Water retention time (assumed)
holdup 25 to 25 Oil retention time (assumed)
Input

Svrcek Paper Solution

1. Calculate the vertical terminal velocity. Using table 5, calculate K using the york demister
equations

53.95 − 0.6973
𝑈𝑡 = 0.313√ = 2.74 𝑓𝑡/𝑠
0.6973

𝑈𝑣 = 0.75 ∗ 2.74 = 2.05 𝑓𝑡/𝑠

2. Calculate The vapor volumetric flow rate

415000 𝑙𝑏/ℎ𝑟
𝑄𝑣 = 𝑠 = 165.32 𝑓𝑡 3 /𝑠
3600 ∗ 0.6973 𝑙𝑏/𝑓𝑡 3
ℎ
3. Calculate the vessel inner diameter

4∗165.32 𝑓𝑡 3 /𝑠
𝐷𝑉𝐷 = √ = 10.13 𝑓𝑡 , Use DVD = 10.5 ft
𝜋∗2.05 𝑓𝑡/𝑠

4. Calculate the settling velocity of the heavy liquid out of the light liquid. Using Table 12 ks = 0.163
Then:
UHL = 0.163*(62.11-53.95)/0.63 = 2.11 in/min
5. Calculate the settling velocity of the light liquid out of the heavy liquid.
ULH = 0.163*(62.11-53.95)/0.764 = 1.74 in/min
6. Calculate the light and heavy liquid volumetric flow rates
165000 𝑙𝑏/ℎ𝑟
𝑄𝐿𝐿 = = 5.10 𝑓𝑡 3 /𝑚𝑖𝑛
𝑚𝑖𝑛
60 ∗ 53.95 𝑙𝑏/𝑓𝑡 3
ℎ
1300 𝑙𝑏/ℎ𝑟
𝑄𝐻𝐿 = = 0.35 𝑓𝑡 3 /𝑚𝑖𝑛
𝑚𝑖𝑛 3
60 ∗ 62.11 𝑙𝑏/𝑓𝑡
ℎ
7. Assume HL =1 ft and calculate the time for heavy liquid to settle out of the light liquid phase:
tHL = 12*1.0/2.11 = 5.7 min
8. Assume HH = 1 ft and Calculate the time for the light liquid to rise out of the heavy liquid phase:
tLH = 12 * 1.0 /1.74 = 6.9 min
9. Calculate the baffle plate area
ρL- ρv = 53.95-0.6973 = 53.25 lb/ft3
Assume HR = 12 in, HL + HR = 24 in
Using Figure 3. G = 9800 gph/ft3
AD=7.48 gal/ft3*60 min/h*(5.10+0.35) (ft3)/9800 gph/ft3
Assume WD = 4 in
WD/D = 4/(12*10.5)=0.0317
Using table 3, AD/A = 0.0095
A = 0.25π*10.5 ft2=86.59 ft2
AD=0.0095*86.59=0.82 ft2
10. Calculate the residence time of each phase
ϴLL=1 ft * 85.77 ft2/5.10 ft3/min = 16.8 min
ϴHL=1 ft * 85.77 ft2/0.35 ft3/min = 247.4 min
11. Calculate the height of the light liquid above the outlet, based on holdup:
HR=5.10 ft3/min*25 min / 85.77 ft2= 1.5 ft
HS = (5.10+0.35) (ft3/min) (5 min)/86.59 ft2 = 0.31
Use HS=0.5 ft
12. Calculate dN according to table 5:
λ = QL/ (QL+ Qv) = (5.10 + 0.35)/(5.10+0.35+165.32*60)=0.0006
16500 1300
𝜌𝐿 = 𝑥53.95 + ∗ 62.11 = 54.55 𝑙𝑏/𝑓𝑡 3
17800 17800
ρM= ρL λ+ρv(1-λ)=54.55*0.0006+0.6973*(1-0.0006)=0.73
QM=165.32+(5.10+0.35)/60 = 165.41 ft3/s
4 𝑥 165.41 0.5
dN > ( ) = 1.73𝑓𝑡 𝑜𝑟 21 𝑖𝑛, 𝑈𝑠𝑒 24 𝑖𝑛
60𝜋/√0.73
Calculate HD = 24 + 24/2 = 36 in = 3.0 ft (min.)
Use HD=5.5 ft from figure 1, HT = 1 ft and s = 0.5 ft. Calculate HBN
HBN=0.5* 2 ft+ 2 ft= 3 ft
Set HA=0.5 ft
Final Dimensions
D = 10.5 ft, HH = 1.0 ft, HL = 1.0 ft, HR = 1.5 ft, HA=0.5 ft, HBN=3.0 ft, and HD=5.5 ft. Add 1.5 ft for
the mist eliminator.
HT = 14 ft
HT/D = 14/10.5 = 1.3
Add 2 ft to HT to be 16 ft(HR=2 ft, HD=7.0 ft) so that HT/D = 1.52 (HT/D should be in the range of 1.5 to 6.0)

API Method (3 phase)

Step 1: Determine gas and oil properties

2.7𝛾𝑃 2.7 ∗ 0.85 ∗ 165

𝜌𝑔 = = = 0.7 𝑙𝑏/𝑓𝑡 3
𝑇𝑍 520 ∗ 1

𝜌𝐿 = 62.4 ∗ 0.86 = 53.9 𝑙𝑏/𝑓𝑡 3

Step 2: Calculate H/D
𝜌𝐿 − 𝜌𝑔 53.9 − 0.7
𝑉𝑎 = 𝐾 √ = 0.35 ∗ √ = 3.0 𝑓𝑡/𝑠
𝜌𝑔 0.7
Actual volume flow of gas
𝑄𝑔,𝑎𝑐𝑡𝑢𝑎𝑙 = 𝑄𝑔 ∗ 1000000 ∗ 𝑀. 𝑤𝑡/379.5/86400/𝜌𝑔 = 159.9 𝑓𝑡 3 /𝑠
Minimum gas flow area
𝑄𝑔,𝑎𝑐𝑡𝑢𝑎𝑙
𝐴𝑔,𝑚𝑖𝑛 = = 53.57 𝑓𝑡 2
𝑉𝑎
4
𝐷𝑚𝑖𝑛 = √ ∗ 12 = 99 𝑖𝑛, Use D = 102 in
𝜋∗𝐴𝑔,𝑚𝑖𝑛
H = 7.5 ft (Assumed Shell Height)
H/D = 12*7.5/102 = 0.9
Step 2: Calculate V and W
Liquid volume (excluding bottom head)
𝜋 𝑁𝐿𝐿
𝑉 = ∗ 𝐷2 ∗ 𝐻 ∗ = 38 𝑏𝑏𝑙
4 144 ∗ 5.615
𝑉 ∗ 1440
𝑊= = 1819 𝑏𝑝𝑑
𝑡
The liquid capacity of the separator should be more than 1396.27 bpd (Oil + water flow rates)
You can increase the separator height to 17 ft to achieve L/D =2 and fall into the range of 2-4 as per
GPSA. V = 86 bbl and W = 4123 bpd.
Hussien Method (3 phase)

Step 1: Determine gas and oil properties

2.7𝛾𝑃 2.7 ∗ 0.85 ∗ 165

𝜌𝑔 = = = 0.7 𝑙𝑏/𝑓𝑡 3
𝑇𝑍 520 ∗ 1

𝜌𝐿 = 62.4 ∗ 0.86 = 53.9 𝑙𝑏/𝑓𝑡 3

Step 2: Determine Cd
assumed drag coefficient of Cd = 2.33
0.5
𝜌𝐿 − 𝜌𝑔 𝑑𝑚
𝑢 = 0.01186 [( ) ]
𝜌𝑔 𝐶𝑑
𝜌𝑔 𝑑𝑚 𝑢
𝑅𝑒 = 0.0049 = 18.29
𝜇𝑔
3 24
𝐶𝑑 = 0.34 + + = 2.35, Acceptable Cd assumption
𝑅𝑒 0.5 𝑅𝑒
Step 3: Determine minimum diameter for water droplet settling
6686𝑄𝑜 𝜇𝑜
𝐷𝑚𝑖𝑛 2 = = 164 𝑖𝑛2 , where dm for Water assumed 500 micron
(𝛾𝑜 −𝛾𝑚 )𝑑𝑚 2
∴ 𝐷𝑚𝑖𝑛 = 12.4 𝑖𝑛
Step 4: Check for Gas Capacity Constraint
𝑇𝑍 𝜌𝑔 𝐶𝑑 0.5
𝐷𝑚𝑖𝑛 2 = 5058𝑄𝑔 ( ) ( ) = 9551, where dm for Oil assumed 500 micron
𝑃 𝜌𝐿 −𝜌𝑔 𝑑𝑚
∴ 𝐷𝑚𝑖𝑛 = 97.7 𝑖𝑛
Step 5: Check for liquid capacity constraint
D2(Ho + Hw) = 8.576(Qoto+ Qwtw) = 284095.01 in2

Try different combinations of D.

TABLE 1
D, in Ho + Hw, in Ls, ft SR = 12 Ls / D Ho + Hw = D2(Ho + Hw) / D2
16 1110 98.8 74.1 𝐿𝑠 = (H𝑜 + H𝑤 + D + 40) /12, If D > 36 in
20 710 65.5 39.3
24 493 47.4 23.7 𝐿𝑠 = (H𝑜 + H𝑤 + 76) /12, If D ≤ 36 in

30 316 32.6 13.1

36 219 24.6 8.2
42 161 20.3 5.8
48 123 17.6 4.4
54 97 16.0 3.5
60 79 14.9 3.0
120 20 15.0 1.5
132 16 15.7 1.4

Ls 14.9 ft Seam to seam length (select from table 1)

SR 1.5 Slenderness ratio (typical value 1.5 to 3) (select from Table 1)
D 120.0 in Separator diameter (select from Table 1)
Step 1: Select Desired Calculations and modify assumptions Step 2: Input operating conditions and physical props

Step 3: Check Operating conditions and physical props output Step 4: Check Hussien’s output tables
Step 5: Check Svrcek output image and table Step 6: Check Svrcek output image and table (Cont.)

Step 7: Check API RP 12J output table

Example No. 4: Three-Phase Horizontal Separator (With Weir, No Mist Eliminator)[5]

Parameter Value Unit Value Unit

Gas flow rate 106594.2 kg/hr 235,000 lb/hr
Oil flow rate 20411.7 kg/hr 45,000 lb/hr
Gas Density 3.043496 kg/m3 0.19 lbs/ft3
Oil Density 648.7452 kg/m3 40.5 lbs/ft3
Water Density 993 kg/m3 62 lbs/ft3
Gas specific gravity 0.99
Oil API 86.61
2
Operating pressure 2.8 Kg/cm 25 psig
o
Operating temperature 37.78 C 100 F
mol. Wt 28.76
Z 1
Gas viscosity 0.013 cP assumed
Oil viscosity 0.682 cP
water flow rate 3401.94 kg/hr 7,500 Lb/hr
surge 10 mins
holdup 5 mins
Input

1. Calculate The vapor volumetric flow rate

235000 𝑙𝑏/ℎ𝑟
𝑄𝑣 = 𝑠 = 343.57 𝑓𝑡 3 /𝑠
3600 ∗ 0.19 𝑙𝑏/𝑓𝑡 3
ℎ
2. Calculate the light and heavy liquid volumetric flow rates
45000 𝑙𝑏/ℎ𝑟
𝑄𝐿𝐿 = = 18.52 𝑓𝑡 3 /𝑚𝑖𝑛
𝑚𝑖𝑛 3
60 ∗ 40.5 𝑙𝑏/𝑓𝑡
ℎ
7500 𝑙𝑏/ℎ𝑟
𝑄𝐻𝐿 = = 2.02 𝑓𝑡 3 /𝑚𝑖𝑛
𝑚𝑖𝑛 3
60 ∗ 62 𝑙𝑏/𝑓𝑡
ℎ
3. Calculate the vertical terminal velocity. Using table 2, calculate K = 0.175 from GPSA values in table
2 divided by two since there’s no mist eliminator

40.5 − 0.19
𝑈𝑡 = 0.175√ = 2.55 𝑓𝑡/𝑠
0.19

𝑈𝑣 = 0.75 ∗ 2.55 = 1.91 𝑓𝑡/𝑠

4. Holdup + surge as specified = 15 min.

VH+ VS, = (15 min) (18.52 ft3/min) = 211.80 ft3
Assume 10 min holdup, 5 min surge.
5. Assume L/D = 1.7
4 𝑥 277.8 1
𝐷=( )3 = 11.15 𝑓𝑡, 𝑈𝑠𝑒 11 𝑓𝑡
0.6𝜋 𝑥 1.7 𝑥 0.25

AT = 0.25π(11.0 ft)2 = 95.03 ft2

6. Since the mass rate of vapor is about 82% of the loading, set Hv to be much greater than the
minimum. Assume Hv = 0.70D = (0.70)*(11.0 ft) = 7.70 ft. Using Table 3, Av/AT = 0.748, Av= 71.08
ft2
7. HLLL = 0.5 * 11.0 + 7 = 12.5 in, use 13 in.
HLLL/D = 13/(11*12) = 0.098, Using Table 3, ALLL/AT = 0.051
ALLL=0.051*95.03 ft2 = 4.85 ft2
8. Hw = 11.0 - 7.70 = 3.30 ft
9. Calculate L2
277.8
𝐿2 = = 14.54 𝑓𝑡, Use L2 = 15 ft
95.03−71.08−4.85
10. HHL = HLL= 3.30/2 = 1.65 ft
11. HHL/D= 1.65/11= 0.15, From Table 11, AHL/AT = 0.094
AHL=0.094*95.03 ft2 = 8.93 ft2
ALL=95.03 – 71.08 – 8.93 = 15.02 ft2
12. From Table 12, ks=0.333 as the light phase is the oil with SG < 0.85 and the heavy phase is water
UHL= (0.333)(62 - 40.5)/0.24= 29.83 in./min
Use 10 in./min (maximum)
ULH= (0.333)(62 - 40.5)/0.682 = 10.50 in./min, Use 10 in./min (maximum)
13. tHL = 12 in./ft)(1.65 ft)/10 in./min = 1.98 min, use 2.0 mins
tLH = tHL = 2.0 min
14. Calculate L1
𝑡𝐿𝐻 𝑄𝐻𝐿 2 𝑚𝑖𝑛𝑠 ∗ 2.02 𝑓𝑡 3 /𝑚𝑖𝑛
= = 0.45 𝑓𝑡
𝐴𝐻𝐿 8.93 𝑓𝑡 2
𝑡𝐻𝐿 𝑄𝐿𝐿 2 𝑚𝑖𝑛𝑠 ∗ 18.52 𝑓𝑡 3 /𝑚𝑖𝑛
= = 2.47 𝑓𝑡
𝐴𝐿𝐿 15.02 𝑓𝑡 2
Use L1 = 3 ft
15. L = 3.0 + 15.0 = 18.0 ft
16. Ф = 7.7 ft/1.91 ft/s = 4.03 s
17. UVA = Qv /Av = (343.57 ft3/s )/ (71.08 ft2) = 4.83 ft/s
18. Lmin = (4.83 ft/s) * (4.03 s) = 19.5 ft
19. Since L < Lmin set L = 19.5 ft (set L2 =16.0 ft, L1=3.5 ft)
20. L/D = 19.5/11.0 = 1.78
21. Assume dished heads per Table 9.
Calculate the thickness of the shell and heads according to table 9:
• Assume E = 0.85
• Assume SA 516 70 Carbon steel
Design Temp. = 650 oF
• S = 17500 psi
• CA = 1/16 in (Assumed)
• P = 25 + 30 = 55psig
• tS = 3/8 in
55 ∗ 132 1
𝑡𝑠 = + = 0.307 𝑖𝑛
2 ∗ 17500 ∗ 0.85 − 1.2 ∗ 55 16

• tH = 0.5 in
0.885 ∗ 55 ∗ 132 1
𝑡𝐻 = + = 0.495 𝑖𝑛
2 ∗ 17500 ∗ 0.85 − 0.1 ∗ 55 16

22. Calculate the surface area of the shell and heads according to table 7:
As = π*11 ft*19.5 ft = 673.87 ft2
and
AH = 0.842 * (11 ft)2 = 101.88 ft2
23.Calculate the approximate vessel weight = 17920 lb

𝑙𝑏 0.5 𝑖𝑛
𝑊 = 490 (673.87 𝑓𝑡 2 + 2 ∗ 101.88 𝑓𝑡 2 ) = 17920 𝑙𝑏𝑠
𝑓𝑡 3 12 𝑖𝑛
𝑓𝑡

24. In this example, calculations were performed for only one diameter. However, nearly the minimum
L/D corresponded to a diameter of 11.0 ft, therefore. the next diameter should be smaller, resulting
in a larger L/D. Also, calculations should be performed using a diameter of 11.5 ft.
25. For the light liquid compartment
HHLL = Hw = 3.3 ft ~ 3 ft, 4 in.
ANLL = 4.85 + 185.200/16 = 16.43 ft2
ANLL/AT = 16.43/95.03 = 0.173
Using Table 11, HNLL/D = 0.229
HNLL = 0.229*11 = 2.52 ft ~ 2 ft, 6 in.
HLLL = 13 in.
Comment: Due to the small amount of heavy liquid and a large amount of vapor, a better design
would have used a boot. A vertical vessel should be compared, As well.
Step 1: Select Desired Calculations and modify assumptions Step 2: Input operating conditions and physical props

Step 3: Check Operating conditions and physical props output Step 4: Check Svrceks’s output image and table
References

[1] API SPEC 12J, 7th edition.

[2] TOTAL, Process Engineering Design Manual, Section 2: Vessels
[3] Petroleum and Gas Field Processing - Hussein K. Abdel-Aal, Mohamed Aggour, M. A. Fahim
[4] Design of Two-Phase Separators within the right limits, W.Y. Svrcek and W.D Monnery
[5] Successfully specify three-phase separators, W.Y. Svrcek and W.D Monnery

Resources

[1] Vessel Sizing Wizard Tool by Eng. Ali Farrokhzad

[2] Vessel Sizing (two and three-phase separators) Excel Sheet by Eng. Ajay S. Satpute
Feature Report
Engineering Practice

Updating the Rules

For Pipe Sizing Total costs

$/time
Fixed

Cost
The most economical velocity in piping costs

Pumping
continues to shift downwards power
cost
over the last 40 years Optimum

Alejandro Anaya Durand, Francisco Javier Pacheco Román, Pipe diameter, D

Maria Jose de Villafranca Casas, Ricardo Gabriel Suárez Suárez,
FIGURE 1. Fixed costs rise as the pipe
Adriana Shunashi García Cornejo, Juan Sampieri Espinoza, diameter increases. Power costs fall as the
Daniela Jara Carranza, Luis Francisco Villalobos pressure drop falls. The sum of these two
Vazquez de la Parra has a minimum
Faculty of Chemistry, UNAM
TABLE 1. VALUES USED IN THE GENERAUX EQUATION

P
ipe sizing calculations include Carbon Stainless Carbon Stainless
Alumi- Brass
certain values that are time Terms Steel Steel Steel Steel
num 2008 2008
dependent, such as costs. While 1998 1998 2008 2008
some costs rise, others decrease n 1.35 0.7793 1.472 0.924 0.769 0.907
over time. However, there will always x 29.52 130 6.607 30.7 22.26 32.3
be a minimum total. For flow of flu- Le’ 2.74 2.74 2.74 2.74 2.74 2.74
ids calculations, heuristic criteria are M 0.102 0.102 0.064 0.064 0.064 0.064
often used, which is the main reason E 0.5 0.5 0.5 0.5 0.5 0.5
why these values must be updated P 150 150 150 150 150 150
from time to time. In this paper rec- K 0.04 0.04 0.07 0.07 0.07 0.07
ommended velocities are updated for Y 365 365 365 365 365 365
several materials of construction, and
& 0.55 0.55 0.55 0.55 0.55 0.55
other velocities for new materials are
Z 0.1 0.1 0.1 0.1 0.1 0.1
calculated, using the Generaux Equa-
F 6.7 7.5 6.5 7.4 7.1 7.2
tion [1].
a+b 0.2 0.2 0.2 0.2 0.2 0.2
Background a’ + b’ 0.4 0.4 0.4 0.4 0.4 0.4
Process design is one of the most im-
portant tasks for engineering. In the An example of this phenomena is costs of 1998 derived by the Marshall
development, optimal use of resources the determination of the recommended and Swift Index [2]. Similarly recom-
must be accomplished in order to velocity for a fluid in pipe sizing. The mended velocities are updated in this
reach equilibrium between minimum velocity is intrinsically related to fluid paper with costs for the current year.
cost and maximum efficiency. properties and a host of economic pa- We compare 1998 values with the ac-
Engineers tend to use heuristic for- rameters, including energy and pipe tual calculated ones (2008) and calcu-
mulas to make quick estimates. Such material costs (It’s important to relate new values for other pipe materi-
calculated estimations are accurate member that pipe diameter is directly als such as brass and aluminum.
enough for grasping a quick overview proportional to its cost).
of the desired process. Since 1968 prices have changed Defining optimum
Some of these calculations involve due to inflation. In 1998 it was dem- For pipe design, several criteria have
certain factors that changing over the onstrated that the most economical to be taken into account. Common cri-
years. Nevertheless, the values fre- velocity in piping had shifted down- teria used are recommended velocity,
quently used as inputs in these equa- wards over the preceding 30 years [6]. economical criteria and other opera-
tions are over thirty years old, driving In that paper, recommended velocities tive criterion.
estimations even further away from were recalculated using the Gener- However, an economical criterion
reality [6]. aux equation [1] with parameters and is the decisive factor for determin-
2 CHEMICAL ENGINEERING WWW.CHE.COM JANUARY 2010
Recommended Fluid Velocities, 1998 to 2008 Costs

Recommended velocity, ft/s

20.00
Carbon 1998
15.00 Carbon 2008
10.00 Stainless
steel 2008
5.00 Stainless Examples of these are rent, taxes and
steel 1998
0.00 depreciation. Operational costs on
0 20 40 60 80 100 the other hand are the variable costs,
Density, lb/ft3 which typically include raw material,
FIGURE 2. Comparison between the recommended fluid velocities from 1998 & 2008 man power and energy services, and
fixed operational costs.
Recommended Velocities for Liquids, 2008 Costs A practical way to find the optimal
11.00 cost is to plot capital and operational
Recommended

Stainless
velocity, ft/s

9.00 steel costs (with the same scale) and then

Aluminium adding these values to form a new
7.00
5.00 Brass function that will have a minimal
Carbon value (the optimal value) as shown in
3.00 Figure 1.
steel
20 40 60 80 100
Several other methods can be used
Density, lb/ft3 @ µ = 1 cp; D = 1 ft
to obtain the minimum value in a
more exact way. Although more rigor-
Recommended Velocities for Gases with viscosity close to 0,02 Cp, 2008 Costs
ous procedures are slower to perform,
63
Recommended

Stainless they are used to optimize plant costs.

velocity, ft/s

steel
43 Aluminium
Optimum equation
23 Brass Although there are several rules for
Carbon calculating the optimum pipe diame-
3 steel
0 0.2 0.4 0.6 0.8 1 1.2 ter, this paper only considers the Gen-
Density, lb/ft3 @ µ = 0.2 cp; D = 1 ft eraux equation [1], shown in the box
on p. TK. This equation determines
FIGURE 3. Recommended velocities for gases (a) and liquids (b) according to 1998 the fluid velovcities required to obtain
and 2008 prices. Recommended velocities tend to fall down as years pass by
the most economic pipe diameter.
Recommended Fluid Velocities at different Diameters for Carbon steel, 2008 Costs Ten years ago this equation was
8 used to obtain updated recommended
7 D = 6 in velocities because the 1968 values
were still used in that time [6]. For
Recommended

6 D = 8 in
velocity, ft/s

D = 10 in this purpose an analysis was made to

5
D = 12 in demonstrate how recommended veloc-
4
D = 4 in ities were changing over the years.
3 D = 2 in
2 Updated recommended velocities
1 Price changes have a great impact on
0 the parameters used in the Generaux
0 20 40 60 80 100 120 equation. Table 1 shows the influence
Density, lb/ft3 of the change in energy and material
Recommended Fluid Velocities at different Viscosities, 2008 Costs costs as measured by the Marshall
8.00 and Swift Index [3]. Some of the val-
7.50 ues of these parameters were derived
7.00 Viscosity
Recommended

in order to obtain updated ones that

velocity, ft/s

6.50 = 0.2 cp˝

6.00 Viscosity could be more in line with the 2008
5.50 = 0.5 cp˝
5.00 costs.
4.50 Viscosity Figure 2 shows a comparison of the
4.00 = 1 cp
3.50 velocities calculated. Updated fluid ve-
Viscosity
3.00 = 2 cp locities of carbon steel pipe are about
0 20 40 60 80 100 120 48% lower than the 1998 recommen-
Density, lb/ft3 dations; the same goes to stainless
FIGURE 4. Plots of recommended velocities versus different viscosities and diam- steel pipe with a 45% decrease.
eters Viscosity has no significant influence in recommended velocity as diameter
does At higher diameter, economic velocities are also higher.
Cost evolution
ing a pipeline design. The analysis is as well minimal. In the midst of comparison between en-
meant to find the negligible total cost, Total costs are the sum of energy ergy and capital costs, energy cost has
which is the sum of capital and op- and capital costs. Fixed costs are prac- proved to have a greater impact. This
erational costs. By determining this tically independent of time, produc- cost can be defined by the variable K,
minimum a number of variables are tion amount and production volume. which surprisingly has increased by a
CHEMICAL ENGINEERING WWW.CHE.COM JANUARY 2010 3
TABLE 2. UPDATED RECOMMENDED FLUID VELOCITIES (2008 COSTS)
Carbon Steel 2008
Density, lb/ft3 100 62.4 50 1 0.1 0.075 0.01
Engineering Practice Viscosity, cP 1 1 1 0.02 0.02 0.02 0.02
Recommended velocity, ft/s 3.57 4.11 4.38 17.38 34.34 37.39 67.85
Stainless Steel 2008
factor of 1.75 since 1998. While energy Density, lb/ft3 100 62.4 50 1 0.1 0.075 0.01
costs increase in an inversely propor- Viscosity, cP 1 1 1 0.02 0.02 0.02 0.02
tional way, recommended velocities Recommended velocity, ft/s 5.42 6.23 6.65 26.37 52.10 56.73 102.95
tend to shift down over time despite of
Aluminum 2008
involving a higher initial investment
Density, lb/ft3 100 62.4 50 1 0.1 0.075 0.01
in piping design (material costs).
Viscosity, cP 1 1 1 0.02 0.02 0.02 0.02
Sample calculation No. 1 Recommended velocity, ft/s 4.48 5.15 5.50 21.79 43.06 46.88 85.08
Let us use a pipeline for the following Brass 2008
conditions. Density, lb/ft3 100 62.4 50 1 0.1 0.075 0.01
t1JQFTUBJOMFTTTUFFM Viscosity, cP 1 1 1 0.02 0.02 0.02 0.02
t'MPXQ = 250 gal/min = 0.557 ft3/s Recommended velocity, ft/s 5.43 6.25 6.67 26.44 52.25 56.89 103.24
t-JRVJEXBUFSBU¡'
t%FOTJUZR = 62.4 lb/ft3
Using the set of fluid velocities for GENERAUX EQUATION
1998 shown in Figure 3a; the recom-
mended velocity is V = 11.32 ft/s. The
cross sectional area of the pipe, S, is
calculated as
S = Q/V = 0.557/11.32 = 0.0492 ft2
This cross-section is reasonably close Nomenclature
to that of a 3-in dia., stainless-steel a Fractional annual depreciation on M Factor to
pipeline, dimensionless express cost of
Schedule 40 pipe.
b Fractional annual maintenance on piping installation, in terms of yearly
However, using the set of fluid veloc- cost of power delivered to the fluid,
pipeline, dimensionless
ities for 2008 shown in Figure 3b; the dimensionless
a’ Fractional annual depreciation on
recommended velocity value is signifi- pumping installation, dimensionless n Exponent in pipe–cost equation
cantly lower V = 6.23 ft/s , calculating b’ Fractional annual maintenance on (C = XD n ) , dimensionless
the cross-sectional area of the pipe S installation, dimensionless P Installation cost of pump and motor,
= Q/V = 0.0894 ft2. This cross-section C Installed cost of pipeline, including $/h.p.
is closer to a 4-in. dia. stainless-steel fittings, $/ft Q Fluid flow, ft3/s
Schedule 40 pipe. A larger pipe diame- D Inside pipe diameter, ft S Cross sectional area, ft2
ter is justified with the updated lower E Combined fractional efficiency of€ V Velocity, ft/s
recommended velocities. pump and motor, dimensionless X Cost of 1 ft of 1-ft-dia. pipe, $
F Factor for installation and fitting, Y Days of operation per year (at 24
Sample calculation No. 2 dimensionless h/d)
For this case let us use a pipeline for K Energy cost delivered to the motor, Z Fractional rate of return of incre-
the following conditions. $/kWh mental investment, dimensionless
t1JQFDBSCPOTUFFM Le’ Factor for friction in fitting, equivalent Φ Factor for taxes and other expenses,
t'MPXQ = 250 gal/min = 0.557 ft3/s length in pipe diameter per length of dimensionless
t-JRVJEXBUFSBU¡' pipe, 1/ft ρ Flow density, lb/ft3
t%FOTJUZR = 62.4 lb/ft3 μ Fluid viscosity, cP
Once again using the graph of fluid €
velocities for 1998 (Figure 3a), the rec- with the updated lower recom- decreased between 40 and 50% since
ommended velocity is V = 8 ft/s. The mended fluid velocities. 1998 being greater the decrease for
cross-sectional area of the pipe is S = € cheaper materials.
Q/V = 0.1358 ft2. This cross-section Conclusions €
is reasonably close to that of a 3.5-in. Revised values for the recommended References
dia., carbon-steel Schedule 40 pipe. fluid velocities on this paper have 1. Perry R. H. and Chilton. C.H. “Chemical
Engineer´s Handbook,” 5th Edition. pp 5–30
However, using the set of fluid proved to be highly sensitive to energy McGraw-Hill, New York, 1973.
velocities for 2008 shown in Fig- and material costs. However between 2. Marshall and Swift Equipment Cost Index.
ure 3b, the recommended velocity both factors, energy cost has more sig- In Chem. Eng. p. 230, September 2008, and
similarly in all issues back to 1998.
value is lower, V = 4.1 ft/s, and the nificance in the total cost than mate- 64%FQUPG&OFSHZ 8BTIJOHUPO %$
cross sectionas area of the pipe is S rial cost. It is of great importance to 4. Crane Corp., “Flow of Fluids,” Chicago, 1963.
= Q/V = 0.1358 ft2. This cross-sec- remember that these costs vary de- "OBZB %VSBOE " FU BM 6QEBUFE 3VMFT PG
tion is closer to a 5-in. dia., carbon- pending on the inflationary surge at Thumb for Pipe Sizing, Presented at October
1998 meeting of IMIQ. Oaxaca, México.
steel Schedule 40 pipe. As found in the current time, and so do the eco- "OBZB%VSBOE" BOEPUIFST 6QEBUFE3VMFT
the first example calculation above, nomic fluid velocities. In general terms for Pipe Sizing, Chem. Eng. May 1999, pp.
153–156,.
a larger pipe diameter is justified recommended velocities in fluids have
4 CHEMICAL ENGINEERING WWW.CHE.COM JANUARY 2010
Authors
Alejandro Anaya Durand Adriana Shunashi García Ricardo Gabriel Suárez
(Parque España No. 15b Col. Cornejo (1era Cda Ezequiel Suárez (Industria 6, Col.
Condesa, C.P. 06140, Mexico, Ordoñez No. 14 Col. Copilco Florida c.p.01030, Alvaro
%' 1IPOF &NBJM BBOBZB- El Alto, C.P. 06340, Coyoacán, 0CSFHØO .ÏYJDP %'
[email protected]) is a .ÏYJDP %' .ÏYJDP 1IPOF .ÏYJDP 1IPOF
professor of chemical engineer &NBJM &NBJMHBCPTVBSF[!
at the National University of [email protected]) is gmail.com) is a fifth-semester
Mexico (UNAM), and has over a fifth-semester chemical chemical engineering honor
48 years experience in project engineering honor student student at UNAM, he is vice
and process engineering. He at UNAM, she is an active president of the UNAM stu-
retired from Instituto Mexi- member of IMIQ. dent section of IMIQ.
cano del Petroleo in 1998 after
holding top positions. For 43 years he has been
an educator in chemical engineering in several Daniela Jara Carranza ( Juan Sampieri Espinoza
universities in Mexico, and presently he is also Escultores No. 42 Col. Cd (Avenida Universidad 2014,
consultant at several engineering companies. He Satélite, C.P 01021, Nau- Col. Copilco, c.p. 04350,
has published over 250 papers related to engi- calpan de Juarez , Estado $PZPBDÈO .ÏYJDP %'
neering and education; is a Fellow of the AIChE; EF .ÏYJDP .ÏYJDP 1IPOF .ÏYJDP 1IPOF
a member of National Academy of Engineering; &NBJM &NBJM HJPWTF@!IPU-
and has received the main chemical engineering [email protected]) mail.com) is a fifth-semester
awards Mexico. He holds a M.S. in project engi- is a fifth-semester chemical chemical engineering honor
neering from UNAM. engineering honor student student at UNAM, he is an
at UNAM, she is an active active member of IMIQ.
member of IMIQ.
Maria Jose de Villafranca
Casas (2da Cda de Tantoco #4 Luis Francisco Villalo-
Frac. La Presilla Col. M. Con- Francisco Javier Pacheco bos Vazquez de la Parra
USFSBT .ÏYJDP %'1IPOF Román (Avenida Universi- (Rincon de los Arcos 95,
&NBJMNKEFW- dad 2014, Col. Copilco, c.p. Col. Bosque Residencial del
[email protected]) 04350, Coyoacán, México, Sur, c.p. 16010,Xochimilco,
is a fifth-semester chemical %' .ÏYJDP 1IPOF .FYJDP %' .FYJDP 1IPOF
engineering honor student at &NBJM &NBJMMV-
UNAM; she is an active mem- [email protected]) is a isfranciscovillalobos@gmail.
ber of IMIQ and works as an fifth-semester chemical en- com) is a fifth-semester
intern at CMM (Mario Molina gineering honor student at chemical engineering honor
Center for Strategic Studies UNAM, he is an active mem- student at UNAM. He is an
in Energy and the Environment) ber of IMIQ. active member of IMIQ.

Circle XX on p. 62 or go to adlinks.che.com/23022-XX
CHEMICAL ENGINEERING WWW.CHE.COM JANUARY 2010 5
View publication stats

Heat Transfer
From Everand
Heat Transfer
Dimitri Gidaspow
No ratings yet
Pipe Sizing and Pressure Drop Calculation
100% (1)
Pipe Sizing and Pressure Drop Calculation
7 pages
Pipe Sizing and Pressure Drop Calculations
100% (2)
Pipe Sizing and Pressure Drop Calculations
14 pages
CE 331 Notes
No ratings yet
CE 331 Notes
95 pages
Basic Calculation For Process Engineer
100% (1)
Basic Calculation For Process Engineer
10 pages
Permeability Report
100% (3)
Permeability Report
21 pages
Process Engineering Calculations (Part 1)
100% (3)
Process Engineering Calculations (Part 1)
210 pages
Pressure Drop Calculations
100% (3)
Pressure Drop Calculations
28 pages
Basic Calculations For Line Pressure Drop
No ratings yet
Basic Calculations For Line Pressure Drop
28 pages
Line Sizing
No ratings yet
Line Sizing
20 pages
Pressure Drop Calculation Equations For Piping and Pipeline
No ratings yet
Pressure Drop Calculation Equations For Piping and Pipeline
6 pages
Pressure Drop Calculation Equations For PIPING and PIPELINE
100% (1)
Pressure Drop Calculation Equations For PIPING and PIPELINE
6 pages
Pressure Drop
No ratings yet
Pressure Drop
5 pages
Pressure Drop Calculation Equations For PIPING and PIPELINE
100% (6)
Pressure Drop Calculation Equations For PIPING and PIPELINE
3 pages
P - P D 10 4.367 Q: Metric: Fluid Flow Design Calculation
No ratings yet
P - P D 10 4.367 Q: Metric: Fluid Flow Design Calculation
22 pages
Pipeline Losess
100% (1)
Pipeline Losess
23 pages
Single Phase Flow Line Sizing
No ratings yet
Single Phase Flow Line Sizing
21 pages
Chapter 3 GTSa New
No ratings yet
Chapter 3 GTSa New
63 pages
Fluid Flow - Rules of Thumb For Chemical PDF
No ratings yet
Fluid Flow - Rules of Thumb For Chemical PDF
27 pages
RMS Inlet Pressure Calculation Theory
No ratings yet
RMS Inlet Pressure Calculation Theory
22 pages
Problem Piping
No ratings yet
Problem Piping
79 pages
Pressure Drop Evaluation Along Pipelines
No ratings yet
Pressure Drop Evaluation Along Pipelines
11 pages
Pipe Systems Design
No ratings yet
Pipe Systems Design
31 pages
Line Sizing Manual
No ratings yet
Line Sizing Manual
17 pages
Gas Pipeline Hydraulics
No ratings yet
Gas Pipeline Hydraulics
23 pages
Pressure Drop in Pipes Calculation
No ratings yet
Pressure Drop in Pipes Calculation
34 pages
Pressure Drop Along Pipe Length - Fluid Flow Hydraulic and Pneumatic
No ratings yet
Pressure Drop Along Pipe Length - Fluid Flow Hydraulic and Pneumatic
1 page
Pipe Sizing: Pressure Drop Equations Darcy-Weisbach Equation
No ratings yet
Pipe Sizing: Pressure Drop Equations Darcy-Weisbach Equation
21 pages
ASHRAE Pipe Sizing
No ratings yet
ASHRAE Pipe Sizing
21 pages
8.pressure Drop in Piping PDF
100% (1)
8.pressure Drop in Piping PDF
41 pages
Pressure Drop in Piping
80% (5)
Pressure Drop in Piping
41 pages
Fluid Engineering Flow in P.
No ratings yet
Fluid Engineering Flow in P.
6 pages
Teori Head Pompa
No ratings yet
Teori Head Pompa
6 pages
Single Phase W Orly Po
No ratings yet
Single Phase W Orly Po
105 pages
Chapter 2
No ratings yet
Chapter 2
27 pages
Basic Calculation For Process Engineer
No ratings yet
Basic Calculation For Process Engineer
10 pages
Garr
No ratings yet
Garr
7 pages
Formulario Transporte de Hidrocarburos
No ratings yet
Formulario Transporte de Hidrocarburos
7 pages
Sazeh Procedure For Line Sizing
No ratings yet
Sazeh Procedure For Line Sizing
28 pages
Pressure Drop Calculation
No ratings yet
Pressure Drop Calculation
6 pages
Pressure Drop Calculation
100% (2)
Pressure Drop Calculation
6 pages
Petrowiki Pressure Drop Equations
No ratings yet
Petrowiki Pressure Drop Equations
14 pages
FF-perdidas de Carga en Tuberias-01
No ratings yet
FF-perdidas de Carga en Tuberias-01
8 pages
Gas Hydraulic Pipeline
No ratings yet
Gas Hydraulic Pipeline
5 pages
Pressure Drop in Piping
No ratings yet
Pressure Drop in Piping
140 pages
Flow Characteristics:: B. Pressure Loss (Head Loss, Pressure
No ratings yet
Flow Characteristics:: B. Pressure Loss (Head Loss, Pressure
4 pages
Fluid Engineering - Flow in Pipes
No ratings yet
Fluid Engineering - Flow in Pipes
6 pages
TableBookErrata PDF
No ratings yet
TableBookErrata PDF
4 pages
Hazen-Williams Formula PDF
100% (1)
Hazen-Williams Formula PDF
4 pages
TableBookErrata PDF
No ratings yet
TableBookErrata PDF
4 pages
Hazen William Formula PDF
No ratings yet
Hazen William Formula PDF
4 pages
Friction Factor
No ratings yet
Friction Factor
10 pages
Pumps & Piping Hydraulics
No ratings yet
Pumps & Piping Hydraulics
120 pages
Pipeline Design: Pages
No ratings yet
Pipeline Design: Pages
32 pages
Mixed Lubrication in Hydrodynamic Bearings
From Everand
Mixed Lubrication in Hydrodynamic Bearings
Dominique Bonneau
No ratings yet
Carbon Dioxide Thermodynamic Properties Handbook: Covering Temperatures from -20° to 250°C and Pressures up to 1000 Bar
From Everand
Carbon Dioxide Thermodynamic Properties Handbook: Covering Temperatures from -20° to 250°C and Pressures up to 1000 Bar
Sara Anwar
No ratings yet
Principles of Hydraulics
From Everand
Principles of Hydraulics
Horst Walter Grollius
4.5/5 (5)
Thermo-hydrodynamic Lubrication in Hydrodynamic Bearings
From Everand
Thermo-hydrodynamic Lubrication in Hydrodynamic Bearings
Dominique Bonneau
No ratings yet
Southern Marine Engineering Desk Reference: Second Edition Volume I
From Everand
Southern Marine Engineering Desk Reference: Second Edition Volume I
Rolf N. Ekenes
No ratings yet
A Modern Course in Statistical Physics
From Everand
A Modern Course in Statistical Physics
Linda E. Reichl
3.5/5 (2)
Transport Phenomena II Essentials
From Everand
Transport Phenomena II Essentials
The Editors of REA
4/5 (1)
CFD Model of A Hydrocyclone
No ratings yet
CFD Model of A Hydrocyclone
14 pages
Pipe Flow Handout-Lecture1-7
No ratings yet
Pipe Flow Handout-Lecture1-7
23 pages
Ec 0411 PDF
No ratings yet
Ec 0411 PDF
7 pages
Lecture 08 1D Convection Diffusion
No ratings yet
Lecture 08 1D Convection Diffusion
25 pages
Chapter 2 - Fluid Statics
No ratings yet
Chapter 2 - Fluid Statics
30 pages
Bearing Details For Pump and Motor Data
No ratings yet
Bearing Details For Pump and Motor Data
1 page
Stack Downwashing
No ratings yet
Stack Downwashing
1 page
Fluid Mechanics: 6 Session
No ratings yet
Fluid Mechanics: 6 Session
9 pages
Norsok Standard P-001: Edition 5, Sep. 2006
100% (1)
Norsok Standard P-001: Edition 5, Sep. 2006
26 pages
Large Eddy Simulation Using An Unstrcutured Mesh Based Finite-Volume Solver
No ratings yet
Large Eddy Simulation Using An Unstrcutured Mesh Based Finite-Volume Solver
7 pages
Lecture19 Two Independents Variables 2
No ratings yet
Lecture19 Two Independents Variables 2
14 pages
B+V ELEVATOR SIDE DOOR Collar Type VS09 A4
100% (1)
B+V ELEVATOR SIDE DOOR Collar Type VS09 A4
19 pages
Curva de Operación de Bomba K6TS Austral
No ratings yet
Curva de Operación de Bomba K6TS Austral
1 page
Week7 ME130-2
No ratings yet
Week7 ME130-2
11 pages
Lecture2 - CFD - Course - Governing Equations (Compatibility Mode) PDF
No ratings yet
Lecture2 - CFD - Course - Governing Equations (Compatibility Mode) PDF
132 pages
Aerodynamic Tunnel, 300 X 300 MM: Engineering and Technical Teaching Equipment
No ratings yet
Aerodynamic Tunnel, 300 X 300 MM: Engineering and Technical Teaching Equipment
8 pages
WAS Pumps XFP100E CB1 50HZ PDF
No ratings yet
WAS Pumps XFP100E CB1 50HZ PDF
6 pages
Topic 6 Buoyancy Notes
No ratings yet
Topic 6 Buoyancy Notes
1 page
Experiment 8 The Hydraulic Jump
No ratings yet
Experiment 8 The Hydraulic Jump
11 pages
Test Data API 610
No ratings yet
Test Data API 610
4 pages
Air Vessel Sizing For Water Hammer Prevention in Pipelines
No ratings yet
Air Vessel Sizing For Water Hammer Prevention in Pipelines
4 pages
Fluid Power With Applications 7th Edition - Chapter 1
No ratings yet
Fluid Power With Applications 7th Edition - Chapter 1
53 pages
Fluid Mechanics Assignment
No ratings yet
Fluid Mechanics Assignment
11 pages
Civflum Quiz 1
No ratings yet
Civflum Quiz 1
4 pages
Kill Sheet Deviated Well
No ratings yet
Kill Sheet Deviated Well
2 pages
Conservation Equations For Mass Momentum and Energy
No ratings yet
Conservation Equations For Mass Momentum and Energy
12 pages
Fluid Systems/Bansal
No ratings yet
Fluid Systems/Bansal
17 pages
THE ENERGY EQUATION
No ratings yet
THE ENERGY EQUATION
5 pages
Hydraulic Jack
No ratings yet
Hydraulic Jack
9 pages