Chapter 3

System Hydraulics
and Design
This chapter discusses the fundamentals of liquid pipeline hydraulics and the design
and operation (Chapter 5) of hydrocarbon liquid pipeline systems from a hydraulics
point of view. Pipeline system design is mainly concerned with line sizing, equipment
sizing and location, and flow capacity; while system operation is concerned with pipeline system or facility start-up and shut-down, product receipt and delivery, flow rate
changes, emergency shut-down, equipment failure, etc.
A proper pipeline system design requires a system approach taking into account
the following design disciplines:

Hydraulics
Mechanical design
Geo-technical design
Operations and maintenance design

These disciplines are closely interrelated because any decisions or changes in one
area of design directly affect or limit the options in another area. Through the hydraulic design, the pipeline route, pipe size, operating pressure and temperature and the
number of pump stations are determined. From a hydraulic design, mechanical designs
can be developed to meet the criteria of the design basis. The mechanical design is
dictated by the relevant codes and standards, resulting in pipe material selection and
specifications as well as burial depth requirements. Geo-technical design addresses
surface loads, water crossings, buoyancy control and geo-hazard management, which
can significantly affect the cost and safety, if the pipeline route traverses challenging
environments. The operation and maintenance consideration includes the necessary
control systems to operate the system within its design parameters, taking account of
the operating tasks while maintaining the functional integrity of the system.
The scope of this chapter includes the governing principles and equations of liquid pipeline hydraulics and their solutions in steady states. The design of any pipeline
system is based on various design factors such as flow profile over time and operating
pressures.

3.1 FUNDAMENTALS OF LIQUID PIPELINE HYDRAULICS

3.1.1 Pipeline Flow Equations
Pipe flow is dictated by three conservation laws: mass, momentum, and energy conservation. The mass conservation law states that the net change rate of the fluid flow
in a segment of pipe is equal to the net packing rate of the fluid in the segment of pipe,
while the momentum conservation law states that the momentum applied to a fluid
element is conserved, equating the rate of change of momentum to the sum of the applied forces. The energy conservation law holds for fluid flow, so the net rate of energy
63

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

64 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

t ransport across a pipeline segment is the same as the rate of energy accumulation
within the pipeline section. Such energy includes the internal energy, compression or
expansion energy (work), and kinetic energy.
The mathematical models used for pipelines are based on equations derived from
the fundamental principles of fluid flow and thermodynamics. The hydraulic states of
a pipeline can be defined by four independent variables; pressure, temperature, flow
rate, and density, and thus four equations are required to relate these four independent
variables. These are momentum, mass, and energy conservation equations together
with the equations of state appropriate to the fluids in the pipeline. The three conservation laws can be expressed in the form of partial differential equations describing the
momentum equation, continuity equation, and energy equation. The one-dimensional
form of the conservation equation is adequate to describe the pipeline flow.
3.1.1.1 Continuity or Mass Conservation Equation
The mass conservation equation accounts for mass being conserved in the pipeline.
It requires knowledge of the density and compressibility of the fluid in the pipeline
together with flows, pressures, and temperatures.
(rA) (rvA)
+
= 0
t
x

(3 1)

where
A = Cross sectional area of the pipe
The cross sectional area can change due to the changes in pressure and temperature:

A = A0 1 + cP ( P - P0 ) + cT (T - T0 )

(3 2)

where the subscript zero refers to base or standard conditions. cT is the coefficient for
thermal expansion of the pipe material and its effect on transients is negligibly small.
CP has a large effect on the acoustic speed of a pressure wave and is defined as:

cP =

1 D
1 - m2
E w

(3 3)

where
E = Youngs modulus of elasticity of the pipe
w = Pipe wall thickness
m = Poisson's ratio
The first term in the continuity equation represents the change of mass in a pipe
segment. It is often called line pack change. The line pack can be increased or decreased due to pressure and temperature changes. The line pack change is useful for
gas pipeline operation. It should be distinguished from the line fill volume, which is the
quantity of fluid contained in a pipeline. It is also a useful quantity for batch pipeline
operation. The second term represents the difference between mass flow into and out
of the pipe segment.
3.1.1.2 Momentum Equation
The momentum equation describes the motion of the fluid in the pipeline. It requires
fluid density and viscosity in addition to the pressures and flows. Applying Newtons
second law of motion to a fluid element together with the Darcy-Weisbach frictional

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 65

force, the momentum conservation equation, in one dimensional form, is expressed
as

V
V P
h f rV | V |
+ rV
+
+ rg
+
= 0
t
x x
x
2D

(3 4)

where
r = Density of the fluid
V = Velocity of the fluid
P = Pressure on the fluid
h = Elevation of the pipe
g = Gravitational constant
f = Darcy-Weisbach friction factor
D = Inside diameter of the pipe
x = Distance along the pipe
t = Time
The first term is a force due to acceleration, and the second term a force due to
kinetic energy. These two terms are related to inertial force. The third term is a force
due to pressure difference between two points in a pipe segment. The fourth term is
a gravitational force, and the last term is a frictional force on the pipe wall, opposing
the flow.
The Darcy-Weisbach equation is used to calculate the pressure drop due to the
friction of fluid flow against the pipe wall. The friction pressure drop is linearly proportional to the fluid density and the friction factor, squarely proportional to fluid velocity, and inversely proportional to the pipe diameter. The friction pressure drop can be
expressed as follows:

f rV | V | 8 f rQ 2
= 2 5
2D
p D

(3 5)

In terms of flow rate, the frictional pressure drop is proportional to the square of
the flow rate and inversely proportional to the fifth power of the pipe diameter. Since
the frictional pressure drop and thus pipeline flow capacity depends highly on pipe
diameter, it is the most significant design parameter. The friction factor is related to
the energy losses resulting from fluid flow. It is a function of the Reynolds number and
pipe roughness. Depending on the Reynolds number, the type of pipe flow is classified
into three flow regimes: laminar flow, critical flow, and turbulent flow. Turbulent flow
can be further divided into partially turbulent, where the smooth pipe law applies, and
fully turbulent, where the rough pipe law applies.
The Reynolds number is dimensionless and the ratio of inertial forces to viscous
forces. It is defined by

Re =

| V | rD | V | D

=
m

(3 6)

where
m = dynamic viscosity (kg/m s)
n = m/r = kinematic viscosity (m2/s)
r = fluid density (kg/m3)
V = flow velocity, m/s
D = pipe inside diameter, m

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

66 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

The Reynolds number increases as flow rate or flow velocity increases, and is
always positive. The kinematic viscosity is frequently used for liquid pipeline design
because it is more readily available and is independent of density. A common kinematic
viscosity unit is stokes, but centistokes is a practical unit because the viscosities of
most hydrocarbon liquids are in centistokes range.
The friction factor is determined empirically and analytically represented by the
Colebrook-White correlation for turbulent flow regimes:

k
1
2.51
= - 2 log
+
for Re 4,000
f
3.7D Re f

(3 7)

where k is the pipe roughness, D the pipe inside diameter, and Re is the Reynolds
number. For laminar flow, the friction factor is:

f =

64
for Re 2400
Re

(3 8)

The critical flow regime is defined between 2,400 < Re < 4,000, in which the flow
is unstable. Laminar flow is independent of pipe roughness, while partially turbulent
flow is dependent on Reynolds number and pipe roughness, and fully turbulent flow is
dependent only on relative roughness being independent of Reynolds number.
The Moody diagram, shown in Figure 3-1, relates the friction factor in terms of
Reynolds number and relative roughness.
The Colebrook-White equation is not easily solvable without a computer because
the friction factor appears on both right and left sides of the correlation. To facilitate an

Figure 3-1. Moody diagram for friction factor

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 67

explicit calculation, several alternative forms of the correlation have been developed
and a few examples are given next:
Jains Approximation

f = [1.14 2 Log(k/D + 21.25Re0.9)]2

(3 9)

for 106 < k/D < 102 and 5000 < Re < 108
Churchills formula
f = 8[(8/Re)12 + (A + B) 1.5]1/12

(3 10)

where A = {2.456 Ln[(7/Re)0.9 + 0.27(k/D)]}16 and B = (37,530/Re)16

These equations correlate closely with friction factors on the Moody diagram. The
Fanning friction factor ff is occasionally used and related to the Darcy friction factor
as follows:
f = 4 ff

(3 11)

Other pressure drop equations, such as the Shell-MIT equations and Hazen illiams, are sometimes found in the literature. Since the Darcy-Weisbach equation
W
with the associated Darcy friction factor is most widely used in the petroleum pipeline
industry, it will be used throughout this book.
Most liquid hydrocarbon pipelines are operated in partially turbulent flow regimes,
with the exception of ethylene and ethane flow which may be in a fully turbulent regime and heavy crude which may be in a laminar flow regime.
3.1.1.3 Energy Equation
The energy equation accounts for the total energy of the fluid in and around the pipeline, requiring information regarding the flows, pressures, and fluid temperatures together with fluid properties and environmental variables, such as conductivity and
ground temperature.
4wrpCp T

T
r v v A
+ rvCv
+T
+
+
rCv +

D t
x
T x A x

f rv 2 | v | 4k dT
+
=0
2D
D dz

(3 12)

where
Cv = Specific heat of the fluid at constant volume
T = Temperature of the fluid
rp = Density of the pipe material
Cp = Heat capacity of the pipe material
k = Heat transfer coefficient
z = Distance from the pipe to its surroundings
The first term is the temperature change over time, the second is the rate of temperature change due to the net convection of fluid energy into the fluid element. The
third term describes the change rate due to expansion/compression of the fluid including the Joule-Thomson effect. The fourth term represents the heat flow to, or from,

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 75

Figure 3-4. Pressure profile with elevation profile

represented in head (m or ft), it is sometimes more convenient to graphically display

the pipeline pressure profile in head.
3.1.3.3 Solution of Energy Equation and Temperature Profile Calculation
In the previous examples, an isothermal assumption was made to calculate the pressure profile. The isothermal flow assumption can be justified for fluid which is transported near ground temperature. It is especially valid for a long transmission pipeline
with multiple pump stations, because the temperature approaches close to the ground
temperature within the first section and the temperature increases at the subsequent
pump stations are in the order of a few degrees. However, large changes in liquid
temperature can affect liquid density and/or viscosity, which will subsequently affect
pressure drop. Therefore, the following hydraulic problems should be treated as temperature dependent flow:
Heavy hydrocarbon liquids or waxy crudes whose viscosity changes significantly with temperature
Light hydrocarbon liquids whose density changes significantly with temperature
Injection temperature is significantly higher than the soil temperature
Pipelines with a large pipe size running in a hot ambient temperature condition
The liquid temperature rises or falls along a pipeline and rises through a pump
station. Temperature profile along the pipeline is influenced by external factors such as
ground temperature and soil conductivity as well as heat generated by friction. Fluid
temperature rises through a pump station mainly because of the inefficiency of the
pump and the small temperature drop through station piping. The temperature change
along a liquid pipeline consists mainly of the following components:
Temperature rise due to volume expansion in an isenthalpic process, raising
liquid temperature as the pressure drops;
Temperature change due to heat conduction with the surrounding ground and
ambient temperatures.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 69

state or by the previous pipeline state if it is available. At the end of a time interval, the
current pipeline state is calculated from the four equations using the initial state conditions and by applying the boundary values. Boundary conditions required to solve for
realistic operation analysis are: upstream pressure downstream pressure boundary,
upstream flow downstream pressure boundary, and upstream pressure downstream flow boundary.
There are many different ways to solve the difference equations representing the
partial differential equations. Three popular solution techniques for pipeline flow simulation are briefly described below. For more details refer to specialized books for solving partial differential equations [3].
3.1.2.1 Method of Characteristics
Streeter and Wylie [4] applied the method of characteristics extensively in solving various pipeline-related problems. The method of characteristics changes pipe length and
time coordinates to a new coordinate system in which the partial differential equation
becomes an ordinary differential equation along certain curves. Such curves are called
characteristic curves or simply the characteristics.
This method is elegant and produces an accurate solution if the solution stability condition is satisfied. This stability condition, called the Courant-Levy condition,
requires that the ratio of the discretized pipe length to time increment must be smaller
than the acoustic speed of the fluid in the pipeline. In other words, the time increment is
limited by the discretized pipe length and the fluid acoustic speed. This is not necessarily a limitation for real-time applications where the time increment is short. However,
it can be a severe limitation if applications such as a training simulator require flexible
time steps.
The method of characteristics is easy to program and can produce a very accurate
solution, it also does not require large computer computational capability.
3.1.2.2 Explicit Methods
In explicit methods, the finite difference equations are formulated in such a way that
the values at the current time step can be solved explicitly in terms of the known values
at the previous time step [5]. There are several different ways of formulating the equations, depending on the discretization schemes used and which variables are explicitly
expressed.
The explicit methods are restricted to a small time step in relation to pipe length in
order to keep the solution stable. Just like the method of characteristics, this is not an
issue for real-time applications but a severe limitation for applications requiring flexible time steps. For applications extending over a long time, an explicit method could
result in excessive amounts of computation.
Explicit methods are very simple for computer programming and can produce an
accurate solution. The computer computational capability requirements are relatively
light.
3.1.2.3 Implicit Methods
In implicit solution methods [6], the partial differentials with respect to pipe length
are linearized and then expressed by finite difference form at the current time step,
instead of the previous time as in the explicit method. The values at the current time
step are arranged in a matrix, so the solution requires the use of matrix inversion
techniques. There are several ways to arrange the numerical expressions, depending
on the discretization schemes and whether values are expressed during or at the end of
the time interval. Initially, a trial solution is n guessed and then successive changes to
the approximated solution are made iteratively until convergence is achieved within a
specified tolerance.
The implicit methods produce unconditionally stable solutions no matter what size
the time step or pipe length is. Unconditional stability does not mean the solution is

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

70 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

accurate. Other errors may make the solution inaccurate or useless. The methods can
generate accurate results if the pipe length and time step are short and the specified
tolerance is tight. Therefore, they can be used not only for real-time model but also for
applications requiring flexible time steps.
The disadvantages are that the methods require matrix inversion software, the
computer programming is complex, and the computer computational capability requirement is comparatively high, especially for a simple pipeline system. However, the
absence of a restriction on the size of time step generally outweighs the increase in the
extra requirements, particularly for large pipeline systems.
There are other solution techniques such as variational methods [7], a hybrid
explicit-implicit scheme, and succession of steady states. These are not discussed here.

3.1.3 Steady-State Solutions and Design Equations

A steady state is a condition of a pipeline system that does not change much over
time. Under a steady state, pressure and flow remain constant from one instant to
another, being considered independent of time. A pipeline system design can be
based on a steady-state assumption. In general, the assumption is valid when the
system is not subject to sudden changes in flow rates or other operating conditions
over a short period of time. However, a steady-state assumption is invalid for shortterm operation analysis and even for designing control systems, testing the level of
safety under abnormal operating conditions, etc, because these behaviors are timedependent.
Steady-state equations are good approximations of fluid behaviors for pipeline
design. Steady-state solutions can address design issues because a system design is
concerned with long time horizons. They are simpler and thus faster to get a solution
for each design case. In addition, time-dependent data may not be fully available during a design phase, so transient equations may not be usable. A steady-state solution
can generate pressure, flow, temperature, and density profiles along with a list of station suction and discharge pressures. Such a solution is generally adequate for pipeline
system design, excluding a control system design, because it can:
Determine liquid pipeline capacity,
Determine an efficient operating mode by selecting appropriate units if the line
pack changes or transients in the pipeline network are relatively small compared to the system line pack,
Calculate power or fuel usage and pump or compressor efficiency,
Identify pipeline operations and an alternate configuration.
In this section, the concept of hydraulics is summarized and a calculation method
is presented for design and operation analysis. For detailed hydraulic analysis and calculation, the readers may refer to other books on hydraulics or computer software. In
general, the following parameters are required to calculate pipeline hydraulics:

Pipe grade, size, wall thickness, and pipe roughness,

Pipe length,
Elevation profile,
Fluid properties such as density and viscosity,
Number of products for batched pipelines,
Discharge pressure and temperature,
Delivery or suction pressure,
Ground temperature and thermal conductivity

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 71

3.1.3.1 Solution of Continuity Equation and Volume Correction
Under a steady-state condition, the continuity and momentum equations can be easily solved. The continuity equation is reduced to a total differential equation under a
steady state as
d (rV )
=0
dx

From this steady-state form of continuity equation, we get

r(P,T)V(P,T) = r(P0,T0)V(P0,T0)

or

r(P,T)Q(P,T) = r(P0,T0)Q(P0,T0)

(3 14)

This relationship is the basis of converting volume or flow rate from one pressure
and temperature condition to another including volume correction to base conditions.
Its application is illustrated with the following base design example (this example will
be extended further to a realistic design case):
Example: Base Case
A crude oil pipeline from CE to QU is 200 km long and is 20 in nominal diameter, with a
0.281 wall thickness. It is constructed of 5LX-56 electric resistance welded steel pipe. At
the injection point, crude oil of 32API gravity and ambient pressure enters the pipeline at
an initial flow rate of 18,000 m3/d at 15C. The average operating pressure and temperature
are 4000 kPag and 4C. Calculate the flow rate at the operating conditions.
Figure 3-2 illustrates this pipeline configuration, which will be used for the subsequent example problems in this chapter. CE is the initial injection station, QU is the
final delivery station, and TO a side stream delivery point.
Solution:
It is assumed that the API correction equation or equation of state (Refer to Chapter 2)
is applicable to convert the density at the base condition to the density at the operating
pressure and temperature.
Step 1. To determine the flow rate at the operating conditions, the crude density at
the same conditions should be determined.
The density equivalent to 32API gravity is calculated by applying the API
gravity and the specific gravity relationship, thus the specific gravity is g = 141.5/
(131.5 + API) = 0.8654, and the density is

r = g 1000 = 865.4 kg/ m3 at 15C

Figure 3-2. Pipeline configuration

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

72 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

Step 2. Since the operating conditions are different from the base conditions of
the fluid, it is necessary to convert the density in order to determine the flow rate at the
operating condition, by applying the API equation of state:
Apply the API pressure correction at 4000 kPag: Cf = 0.6476 106 and CP =
1.0026
Apply the API temperature correction at 4C: CT = 1.0090,
Therefore, the density at 4000 kPag and 4C = 865.4 1.0026 1.0090 =
865.4 1.0116 = 875.4 kg/m3
Step 3: Calculate the flow rate at the operating conditions by applying the steadystate mass balance equation.
Pressure and temperature adjusted flow rate = 18,000 /1.0116 = 17,794 m3/d
This volume flow rate is lower because the density at the operating conditions
is higher. This is the consequence of mass conservation.
3.1.3.2 Solution of Momentum Equation and Pressure Profile Calculation
The momentum equation can also be simplified under a steady state. Since the kinetic
energy or velocity head term for long pipeline systems is negligibly small compared
to the total pressure requirement, the momentum equation can be simplified to a total
differential equation as shown below.

dP
dh f rV | V |
+ rg
+
=0
dx
dx
2D

(3 15)

It can be assumed that the liquid density and velocity are constant between two
points along the pipeline. This assumption is valid as long as the distance between two
points is not long. Therefore, the pressure-flow equation can be obtained by integrating
the above steady-state momentum equation:

Px = P0 rg(hx h0) frV2(X X0)/2D

(3 16)

The left hand side is the pressure at the downstream point. The first term on the
right hand side is the pressure at the upstream point, the second the static pressure
or elevation head, and the third the friction head. The total pressure requirement in a
pipeline system consists of the following components:
Pressure changes due to elevation changes, depending only on the product density and difference between the elevations between two points on the pipeline;
Friction pressure drop due to flow rate or velocity, fluid density and viscosity,
and pipe diameter;
Pressure changes due to changes in pipe diameter and subsequent changes in
flow velocity.
For a given flow rate, the above pressure-flow equation allows us to calculate the
downstream pressure if the upstream pressure is known, and the upstream pressure if
the downstream pressure is known. Also, the flow rate can be calculated if the upstream
and downstream pressures are known.
If the static pressure term is set aside, the above equation can be arranged as

(Px P0)/(X X0) = frV2/2D

(3 17)

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

82 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

Specific heat:
Viscosities:
Pour point:

1880 J/kgC
9.5 cSt at 35C and 43.5 cSt and 4C
0C

Solution:
It is assumed that the viscosity of this product is Newtonian and that the density and
viscosity depend on temperature. The fluid density and viscosity are calculated at the
starting point temperature in the segment between two profile points. Let the inlet pressure be 8605 kPag, the same as for the isothermal case.
Step 1. Since the density and viscosity change with temperature, the temperature
relationships of density and viscosity need to be established to calculate these quantities as the temperature profile is calculated.

Applying the API temperature correction term, we get

r(T) = r(15) Exp[ 0.00082 (T 15) (1 + 0.000656 (T 15))]
Applying the ASTM viscosity correlation, we get
Log (v + 0.7) = 11.4667 4.6062 Log(T + 273)

Step 2. Calculate the density and viscosity at the inlet conditions; r(35) = 851.0 kg/m3
and (16) = 9.5 cSt.
Step 3. Use the inlet temperature of the first segment to calculate the friction factor
of 0.0201 and the frictional pressure drop of 508 kPa.
Step 4. Calculate the temperature at the downstream point of the first segment.
The temperature increase due to the frictional pressure drop is 0.32C
To calculate the temperature drop due to conduction, the following values are
calculated iteratively:
the heat transfer coefficient, 0.324 W/m2C;
the log mean temperature, 34.1C;
the temperature drop at downstream temperature 2.1C;
hence the downstream temperature is 35 + 0.32 2.10 = 33.2C.
Step 5. Calculate the pressure and temperature at the other profile points by
repeating the above steps.
KMP
(km)
0
20
30
60
80
90

Elevation
(m)

Pressure
(kPag)

Temp (C)

KMP
(km)

Elevation
(m)

Pressure
(kPag)

Temp
(C)

30
55
45
30
70
100

8605
7889
7714
7060
6195
5674

35.0
33.2
32.4
30.0
28.6
27.9

100
130
150
160
180
200

130
100
60
110
150
130

5152
4586
4368
3666
2765
2364

27.3
25.4
24.3
23.9
22.8
21.9

It is expected that the total pressure requirement is lower than the pressure requirement under the isothermal assumption, because the operating temperature would
be higher and thus the values of density and viscosity are lower. Indeed, the delivery
pressure turns out to be much higher than the delivery pressure for the isothermal
case, and so the total pressure requirement is less by 2014 kPa. It is concluded that the
temperature effects have to be included in hydraulic calculations if the liquid injection
temperature is much higher than the ground temperature.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

74 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

Figure 3-3. Pressure profile and gradient

Note that the frictional pressure drop remains the same even though the elevation
changes.
Step 1. Use the same pressure gradient as obtained in the previous example.
Step 2. Calculate the pressures at the above profile points by adding the static
pressure difference to the frictional pressure drop.
KMP (km)

Elevation (m)

Pressure (kPag)

KMP (km)

30
55
45
30
70
100

8605
7650
7366
6384
5302
4676

100
130
150
160
180
200

0
20
30
60
80
90

Elevation (m)

Pressure (kPag)

130
100
60
110
150
130

4049
3196
2799
2001
919
350

Step 3. Assess the pressure profile.

The elevation difference between point KMP = 0 km and point KMP = 200 km
is 100 m. The static pressure difference is Ph = 873.1 9.8 100/1000 = 855 kPa or
8605 7750 = 855 kPa. Since the elevation at the delivery point is 100 m higher than
the elevation at the inlet point, the total pressure required at the inlet point is 8605kPag,
which is 855 kPa higher than the previous case for flat elevation.
As shown in Figure 3-4, the pressure profile is shifted by the elevation difference
from a reference point, which is in this case the delivery point. Note that the left y-axis
is represented in pressure and the right y-axis in head. Since the elevation profile is
Table 3-1. Elevation profile
KMP (km)
0
20
30
60
80
90

Elevation (m)

KMP (km)

Elevation (m)

30
55
45
30
70
100

100
130
150
160
180
200

130
100
60
110
150
130

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

88 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

and method of installation, and the type of pipe material selected. The operating pressure of a pipeline must be maintained within minimum and maximum pressures. These
pressure limits are critical for safe and efficient operation. The maximum operating
pressure in a liquids pipeline is constrained by the yield strength of the pipe material,
pipe diameter and wall thickness, the fluid density and the elevation of the lowest
point of the pipe, while the minimum pressures are determined by vapor pressures of
the liquids along the pipeline. The elevation affects the operating pressure due to high
static head for liquid pipelines.
The delivery pressure is generally defined in the contract between the pipeline
company and the shippers or third party pipeline to which the fluid is delivered. The
determination of the delivery pressures is influenced by the terminal equipment such
as tank and control valves as well as the elevation profile upstream of the terminal. A
peak elevation can dictate the pressure required, which can result in higher delivery
pressure at the terminal. The delivery pressure is determined by the fluid vapor pressure, pressure rating of the equipment at the delivery site, and pressure requirements
imposed by the delivery facilities such as a tank or connecting pipeline. Therefore,
the delivery pressure requirement dictates the operating pressure for a given flow
rate.
As noted earlier, temperature affects viscosity, density, and specific heat in liquid
lines. A temperature rise is beneficial in liquid pipelines as it lowers the viscosity and
density, thereby lowering the pressure drop. The cooling effect on non-Newtonian or
viscous fluids can be significant because their viscosity can increase significantly and
subsequently the pressure drop can be very high. To reduce the effect of temperature
cooling, the pipeline can be insulated and/or operated at high temperature. The viscous
fluids can be blended with light hydrocarbon liquids such as condensate. The temperature along the pipeline is least controllable due to its dependency on variable soil
thermal conductivity and ambient temperature.
The maximum temperature limit for buried pipe is determined by a combination
of the following three factors:
Ground conditions
Stress level the pipe material can withstand without buckling
Economics of pipeline flow (the liquid flows most efficiently at high
temperature)
The minimum temperature limit is normally determined by the metallurgical (fracture toughness) properties of the pipe material or by the ground conditions.
Fluid properties were fully discussed in the previous chapter. Summarized below
are fluid properties that directly and indirectly affect the design and operation of liquid
pipeline systems.
Density or specific gravity the higher the fluid density, the higher the pressure drop. The pressure drop due to friction is directly proportional to the fluid
density.
Compressibility or bulk modulus is not important for liquid pipeline capacity
calculation, but important for controlling pressure surges and determining line
pack changes.
Viscosity is important in calculating line size, hydraulics, and pumping requirements for liquid pipelines.
Vapor pressure determines the minimum pressure in the pipeline. It must be
high enough to maintain the fluid in a liquid state and to avoid cavitation at
inlet to a pump.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

76 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

Ambient
Ground
Heat out
Pipe
Heat in

Liquid
Heat generation

Heat out

Insulation
Heat out

Figure 3-5. Heat balancing mechanism

Some pipelines may be partially or wholly installed aboveground to save construction or maintenance cost. However, transmission pipelines are generally buried
in order to:
Minimize land use disturbance,
Provide longitudinal restraint along pipeline length,
Protect pipe from possible pipe material fatigue due to stress changes caused by
fluctuations in ambient temperature,
Minimize effects of changes in ambient temperature on fluid viscosity and
density,
Protect pipe from intentional or accidental damage, and
Use the pipeline right of way.
The temperature calculation from the energy equation is not simple even under a
steady-state condition. The steady-state energy equation can be derived by balancing
heat entering and leaving a pipe section, heat transferred from/to the pipe section, to/
from surrounding soil or ambient, and heat from friction. The heat balancing mechanism can be shown in Figure 3-5, and the heat balance is expressed as:

Hin Hout Hcon + Hw = 0

(3 18)

where
Hin = Heat entering a pipe section (w)
Hout = Heat leaving a pipe section (w)
Hcon = Heat transferred from/to the pipe section to/from surrounding soil or
ambient (w)
Hf = Heat from friction (w).
Described below is a temperature calculation procedure. Another method for calculating temperature profile is presented in Addendum 3.1, which includes a temperature calculation method for above-ground pipelines.
1. Assuming that the specific heat of the fluid remains constant at the entering
and leaving conditions, the heat entering and leaving a pipe section can be
expressed in terms of temperatures and engineering quantities as follows:

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 77

Hin Hout = rQCp(Ti T0)/3600

(3 19)

where
r = liquid density (kg/m3)
Q = flow rate (m3/hr)
Cp = specific heat of liquid, kJ/kg/C
Ti = temperature of liquid entering the pipe section, C
To = temperature of liquid leaving the pipe section, C
2. As the liquids flow through the pipe, the pipe pressure drops by friction, liquid
flows undergo an isenthalpic process, and as a result the pressure dissipated by friction becomes heat in the flowing fluid. The temperature of liquids rises in frictional
heating due to their volumetric properties as they are expanded in an isenthalpic
process. The effect of friction heating generally increases with flow rate, viscosity,
insulation, and line length. For large diameter pipelines and high flow rates, heat
generated by friction loss should be included in the temperature profile.
Heat of friction should be considered at high flow rates in large pipelines
to ensure that overheating does not occur. Pump stations operating on flow
control may experience increasing or decreasing discharge pressures as the
temperature of the fluid in the pipeline rises or falls after leaving the pump station. As the temperature increases, the fluid expands. As expansion continues
in the pipeline, the local pressure and volumetric flow rate increases. The heat
generated by frictional pressure drop is expressed as
Hw = q DPf = 0.278Q (DPf/Dx) L

(3 20)

where
Hw
= frictional heating, w
q
= liquid flow rate, m3/sec
DPf = frictional pressure drop, Pa
Q
= liquid flow rate, m3/hr
DPf/Dx = frictional pressure gradient, kPa/km
L
= length of the pipe section, km
3. Even though ground temperature along the pipeline is not normally measured
on a daily basis, it is an important parameter for designing a pipeline system.
Significant temperature changes can occur due to heat transfer through conduction between the liquid and surrounding soil. In describing the flow of heat
from pipeline to ground, Fouriers law of heat conduction is applied by taking
into account the heat transfer through pipe, insulation, and soil. The conduction
heat transfer can be expressed as:
Hcon = U A DTm = 2p DT L U DTm

where
U
A
DT
L
DTm
Tg

(3 21)

= overall heat transfer coefficient (w/m2/C)

= surface area of the outside of the pipe (m2)
= outside pipe diameter or insulated pipe diameter (m)
= length of the pipe section (m)
= Tm Tg = log mean temperature difference between the liquid in a
pipe section and its surrounding soil (C)
= ground temperature (C)

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

78 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

In heat transfer calculations, the log mean temperature can be used, because theoretically it produces a more accurate result in temperature calculation. In practice, there
are many factors that prevent the calculation of temperature accurately; these factors
include ground temperature, soil conductivity, etc.
In the above heat transfer equation, the overall heat transfer coefficient and log
mean temperature difference need to be determined. As shown in the figure below, the
overall heat transfer for pipe flow includes the heat transfer effects due to the boundary
layer, pipe wall, surrounding soil, and insulation if the pipe is insulated. Therefore, the
overall heat transfer coefficient is defined as
U = 1/(Rif + Rp + Rins + Rs)

(3 22)

where
Rif = thermal resistance due to the boundary layer that builds up on the inside of
the pipe wall (m2C/w)
Rp = thermal resistance of the pipe wall (m2C/w)
Rins = thermal resistance of insulation (m2C/w)
Rs = thermal resistance due to the surrounding medium (m2C/w)
However, the heat transfer effects due to the boundary layer and pipe wall are
much smaller than those due to surrounding soil or insulation. Therefore, these two
terms are usually ignored, and only the last two terms are considered in the overall heat
transfer calculation.
Pipelines are not frequently insulated unless the fluid viscosity is so high that it can
be significantly reduced by heating the fluid. If the fluid such as heavy crude is heated,
certain parts of the pipeline are insulated. For an insulated pipe, the heat resistance can
be determined by,
Rins = (DT/kins) Ln(DT/D)

(3 23)

Ground

Insulation

Liquid
film

Steel
Pipe

Corrosion
coating
Outer
Jacket

Figure 3-6. Cross section of insulated pipe

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 79

where
Ln = natural log
DT = the outside diameter of the insulated pipe in m (DT = D + 2 T),
kins = thermal conductivity of the insulation,
T = the insulation thickness in m.
In general, the thicker the better; however, insulation efficiency is not proportional
to the thickness. Although greater thickness reduces conductive heat transfer, it may
not offset the cost of the extra insulation nor reduce the overall heat transfer. The outer
jacket is intended to prevent water from making direct contact with insulation material,
thereby limiting or even destroying the insulation properties of the insulation. It should
be noted that pipeline insulation to reduce heat loss during cold weather may contribute
to overheating in summer, particularly for large diameter pipelines. Normally, pipes are
coated under the insulation layer.
As discussed earlier, most pipelines are buried along their entire length or at least
almost all of their length. The thermal conductivity of insulation can be ten or more
times lower than that of soil, but the depth of burial is much deeper than the insulation
thickness. In general, heat resistance of a buried pipe is greater than that of insulation,
and thus most heat transfer is concerned with heat conduction through the surrounding
soil. The heat resistance can be determined by:

where
DT
Xc

ks

Rs = (DT/ks) Ln{[2Xc + (4Xc2 - DT2)0.5]/DT}

(3 24)

= D if the pipe is not insulated (m)

= burial depth to the center line of the pipe (m)
= burial depth to the top of the insulation = DT/2
= thermal conductivity of the soil (w/m C)

The thermal conductivity is a measure of how easily heat conducts through the
material. It appears in Fourier's law of heat conduction. Generally, the thermal conductivity can be nearly constant over the temperature range normally encountered in
pipelines. Thermal conductivity is measured in units of W/(mC) (Table 3-2).
Certain portions of a pipeline may run above-ground, even for heated liquids, in
order to reduce the construction and other costs. Above-ground pipelines are usually
insulated. If the above-ground pipe length is long enough to affect the temperature profile, the heat transfer between the liquid and ambient air needs to be calculated. Since
the ambient air conditions can change significantly in a short time, their effects need
to be evaluated for design based on the average and worst conditions but are difficult
to assess for operation.
In heat transfer calculations, the log mean temperature difference between the
liquid in a pipe section and the surrounding soil is often used. This is because the fluid
Table 3-2. Thermal conductivity
Substance

Thermal conductivity (W/mC)

Sandy soil, dry

Sandy soil, moist
Sandy soil, wet
Clay soil, dry
Clay soil, moist
Clay soil, wet
Insulation

0.450.70
0.851.05
1.902.25
0.350.50
0.700.85
1.051.55
0.020.05

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

80 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

temperature drop in the pipe section shows an exponential behavior (Figure 3-7). The
log mean temperature is defined as:
Tm = Tg + (Ti T0)/Ln[(Ti Tg)/(T0 Tg)]

(3 25)

where
Ti = temperature of liquid entering the pipe section (C)
T0 = temperature of liquid leaving the pipe section (C)
Tg = ground or surrounding medium temperature (C)
Therefore, the log mean temperature difference is determined by the equation:
DTm = Tm Tg = (Ti T0)/Ln[(Ti Tg)/(T0 Tg)]

(3 26)

Note that a log mean temperature is similar to a simple arithmetic average temperature for short pipe lengths over which the temperature is calculated, and that both
the log mean temperature and arithmetic average temperature contain the downstream
temperature that has to be calculated in the temperature profile calculation. Therefore,
an iterative technique is used to calculate either the log mean or arithmetic average
temperature and this can be easily implemented in software. A manual calculation
can also generate a reasonable temperature profile to the known upstream temperature
instead of using the log mean temperature.
Combining the above equations for temperature, we have
T0 = Ti + Pf /(rCp) Hcon/(rQCp)

(3 27)

Temperature

where
T0 = Outlet temperature (C)
Ti = Inlet temperature (C)
DPf = frictional pressure drop, Pa
r = density (kg/m3)
Q = flow rate (m3/sec)
Cp = specific heat (J/kg C)

T0

Temperature Profile

Ground Temperature

TG

Pipe Length

Figure 3-7. Temperature profile

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 81

The heat conduction term, Hcon, includes T0. In other words, the above temperature equation contains the term T0 on both sides of the equation. Therefore, it requires
an iterative process to calculate T0 accurately. Except for heavy crudes, the friction
heating term is small compared to the heat conduction term, so the above temperature
equation can be simplified to:
rQCpdT = UADTmdx

(3 28)

where
A = pipe surface area
dx = differential in distance
This equation can be integrated to obtain
Tx = Tg + [T0 Tg] exp[ (2p UDTX)/(rQCp)]

(3 29)

This equation shows that the temperature profile decays exponentially and that the
delivery temperature drops closer to the ground temperature. If the frictional heating
term is included, the overall temperature profile is elevated. The temperature equation
indicates that, assuming the ambient temperature is lower than the liquid temperature,
the liquid cools faster and its viscosity increases as flow rate decreases.
Note that the effect of friction heating increases with flow rates and viscosity because the frictional pressure drop is high. Therefore, a frictional heating term should be
included for the case of high flow rates and/or high viscosity liquid. Also, the calculation of a temperature profile is so complex and prone to error that it is beneficial to use
a computer software package to obtain quick and accurate results. Temperature-related
problems are more severe for larger pipelines because the conduction heat loss is proportional to pipeline surface area.
The surrounding environment is the key factor in the overall heat transfer coefficient, which is most critical in calculating the temperature profile along the pipeline.
Table 3-3 shows the range of overall heat transfer coefficients for an on-shore pipelines surrounding environment [14].
Example: Base Case Extension 3
The previous base case is extended to include the temperature profile by removing the
isothermal assumption. Assuming that the pipeline is not insulated, calculate the pressure and temperature profiles using the following data:

Oil inlet temperature:

Average soil temperature:
Depth of cover:
Soil thermal conductivity:

35C
4C
1.2 m
0.5 W/mC

Table 3-3. Environment vs. overall heat transfer coefficients

Environment
Buried, dry soil (uninsulated)
Buried, dry soil (2 thick insulation)
Buried, wet soil (uninsulated)
Buried, wet soil (2 thick insulation)
Above-ground, exposed to atmosphere (uninsulated)
Above-ground, exposed to atmosphere
(2 thick insulation)

U Value (W/m2 C)
0.853.69
0.280.85
1.704.54
0.571.14
3.978.52
0.571.15

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

82 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

Specific heat:
Viscosities:
Pour point:

1880 J/kgC
9.5 cSt at 35C and 43.5 cSt and 4C
0C

Solution:
It is assumed that the viscosity of this product is Newtonian and that the density and
viscosity depend on temperature. The fluid density and viscosity are calculated at the
starting point temperature in the segment between two profile points. Let the inlet pressure be 8605 kPag, the same as for the isothermal case.
Step 1. Since the density and viscosity change with temperature, the temperature
relationships of density and viscosity need to be established to calculate these quantities as the temperature profile is calculated.

Applying the API temperature correction term, we get

r(T) = r(15) Exp[ 0.00082 (T 15) (1 + 0.000656 (T 15))]
Applying the ASTM viscosity correlation, we get
Log (v + 0.7) = 11.4667 4.6062 Log(T + 273)

Step 2. Calculate the density and viscosity at the inlet conditions; r(35) = 851.0 kg/m3
and (16) = 9.5 cSt.
Step 3. Use the inlet temperature of the first segment to calculate the friction factor
of 0.0201 and the frictional pressure drop of 508 kPa.
Step 4. Calculate the temperature at the downstream point of the first segment.
The temperature increase due to the frictional pressure drop is 0.32C
To calculate the temperature drop due to conduction, the following values are
calculated iteratively:
the heat transfer coefficient, 0.324 W/m2C;
the log mean temperature, 34.1C;
the temperature drop at downstream temperature 2.1C;
hence the downstream temperature is 35 + 0.32 2.10 = 33.2C.
Step 5. Calculate the pressure and temperature at the other profile points by
repeating the above steps.
KMP
(km)
0
20
30
60
80
90

Elevation
(m)

Pressure
(kPag)

Temp (C)

KMP
(km)

Elevation
(m)

Pressure
(kPag)

Temp
(C)

30
55
45
30
70
100

8605
7889
7714
7060
6195
5674

35.0
33.2
32.4
30.0
28.6
27.9

100
130
150
160
180
200

130
100
60
110
150
130

5152
4586
4368
3666
2765
2364

27.3
25.4
24.3
23.9
22.8
21.9

It is expected that the total pressure requirement is lower than the pressure requirement under the isothermal assumption, because the operating temperature would
be higher and thus the values of density and viscosity are lower. Indeed, the delivery
pressure turns out to be much higher than the delivery pressure for the isothermal
case, and so the total pressure requirement is less by 2014 kPa. It is concluded that the
temperature effects have to be included in hydraulic calculations if the liquid injection
temperature is much higher than the ground temperature.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

102 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

The total pressure requirements for all three combinations are higher than
their respective design pressure. Therefore, they require an intermediate pump
station to satisfy the total pressure requirement.
Step 3. Determine the number of intermediate pump stations and their power
requirements.
Only one intermediate pump station is required for all three cases because
the design pressures for all cases are less than half of the total pressure drops.
Assuming the suction pressure of the intermediate station is the same as the
delivery pressure, the discharge pressure at the inlet and intermediate stations
are as follows:

Pipe grade

Pipe size
(in/mm)

Design pressure
(kPag)

Discharge
pressure at inlet
point (kPag)

Discharge pressure
at intermediate
station (kPag)

X65
X70
X70

22/558.8
20/508.0
22/558.8

8246
9765
8880

5066
7870
5066

5066
7870
5066

The capital cost due to the extra pumping power requirement for the 20
pipe is higher than the cost for the 22 pipe size, while the pipe cost for X70
with 20 diameter may cost less than the other two options. The extra capital
cost for the 22 line is more than 20% and is incurred by the extra pipe material
and construction expenses. However, the extra capital cost of the 22 diameter
pipe might be partly compensated by lower unit pumping cost. Assuming that
the annualized cost for the 20 pipe case is lowest, it is selected as the base
design.
The facilities such as the initiating pump station for the selected base design
would be designed to accommodate the capacity until the capacity increases
in the 10th year. In the 10th year, the additional facility increases include the
pumping capacity at the inlet point for the additional flow and an intermediate
pump station with the pumping capacity of 30,000 m3/d.
4. Develop alternative design cases and perform comparative studies against the
base design
Alternative 1: This alternative design is to use a pipe wall thickness larger
than 0.281 in order to increase the design pressure slightly higher than the
total pressure requirement. No intermediate pump station is required if the design pressure is slightly higher than the total pressure requirement. Note that
the required total pressure will be increased due to slightly smaller inside pipe
diameter.
The design pressure for the X70 22 pipe is lower than the total pressure requirement, which in turn is lower than the maximum operating pressure range.
The next largest nominal wall thickness is 0.312 or 7.92 mm, and its design
pressure is 9857 kPag or 1430 psig, but the required total pressure is 9914 kPag
for a flow rate of 30,000 m3/d. Therefore, the wall thickness is not sufficient to
meet the total pressure requirement without an intermediate pump station.
The next largest nominal wall thickness is 0.344 or 8.74 mm, which
can allow the design pressure to increase up to 10,870 kPag. For this wall
thickness, the required total pressure turns out to be 10,051 kPag. Since this
design pressure is higher than the required total pressure, no intermediate
pump station is required for the flow rate expected beyond the 10th year,

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

84 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

alcohols, carbon dioxide, etc. The design section includes design criteria, design and
selection of piping components, piping joints, supports and restraints, and auxiliary
and other specific piping. The standard also specifies the following subjects:

Acceptable materials and limitations;

Dimensional requirements for piping components and threads;
Construction, welding, and assembly of components, equipment and facilities;
Inspection and testing, including repair of defects and test pressure;
Operation and maintenance procedures of pipeline, equipment and facilities,
right of way, communications, etc.
Internal and external corrosion control and monitoring.
The CSA pipeline standard Z662 is more comprehensive than B31.4 in its scope
and covers the following:
Petroleum liquids and gases including sour gas and oil field steam;
Onshore and offshore liquid and gas pipelines;
Steel pipe, reinforced composite and polyethylene pipes, and aluminum pipe.
There are many differences between Z662 and B31.4 in design specifications, materials, welding and in other areas. However, the discussion of the differences is beyond the
scope of this book. A summary of the differences can be found in [8]. In this book, ASME
B31.4 and if necessary, the Canadian standard, CSA Z662, are referenced whenever they
are used. Other standards referenced include ASME B16.5 for pipe flanges and flanged
fittings, ASME B16.34 for valves, and API 5XL for specifications for line pipe.

3.2.2 Design Factors [9]

3.2.2.1 Supply and Demand
The need for a pipeline system has to be identified before the pipeline system is built. This
need results from actual or anticipated requests for transportation of petroleum products.

Figure 3-8. Supply profile

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 85

The need can be a new pipeline or an increase in the capacity of an existing line, depending on the supply and/or demand locations and volumes specified in the requests.
As shown in Figure 3-8, the flow rates are initially low and increase to a future
flow rate. The flow rates can be decreased during the life of the project, and the supply and demand locations may also change. Therefore, an optimum design includes
pipeline system growth in terms of pipe and facilities requirements, taking into account
future incremental flow rate increase and eventual decrease.
The first step in identifying the need is to determine the supply and demand as well
as their respective locations. In general, the demand profile drives the pipeline capacity for petroleum products in consuming areas or oil importing countries, while the
supply profile drives the pipeline capacity for producing areas. However, the supply
and demand change over time, and their build-up patterns in terms of volume and time
greatly influence the determination of the economically optimum size of the pipeline
and facilities required for the entire range of flow rates. In other words, the supply/demand projection into the future is required to determine the optimum pipe size, facilities, timing of system expansion, and other requirements. The locations of supply and
delivery points strongly influence the selection of the pipeline route and subsequently
the locations of facilities and control points.
The supply information includes the oil reserves or production capacities (refinery
capacities) estimated at a given time as well as the locations where these volumes will
grow or shrink over time. Depending on the particular pipeline system under consideration, supply may or may not be a major factor. If the pipeline system is to be supplied
by a large supply source, it may be assumed that the supply will satisfy the demand
over the life of the project. On the other hand, if the pipeline system transports fluid
from many supply sources, demand may dictate the pipeline system design instead.
Therefore, transportation facilities should be designed and built to accommodate these
volume forecasts and the accuracy of the supply and demand forecasts reduces the risk
of over or under design of the system. Figure 3-8 shows an example of a supply profile
over time.
The demand is forecasted on the basis of average annual flows over the period of
the project; the yearly volume increases or variations are important for system design.
Seasonal variations in the demand also need to be taken into account in design. If the
pipeline system transports petroleum products such as gasoline to a large consuming area, seasonal variations in the demand can be more important than the annual
increase. In addition, the storage capacities around the consuming areas are also important not only to offset some of the peak requirements but also to avoid over design. If the pipeline system has no storage facilities available, the peak requirements
must be transported and the facilities must be sized accordingly to accommodate these
requirements.
3.2.2.2 Pipeline Route and Environmental Issues
The routing of the pipeline system is directly related to supply and demand locations.
The routing selection is important especially for new pipeline systems. A preliminary
route is selected using a combination of immersive video, aerial photography, LIDAR
(laser interferometry and distance ranging), and geographical information system
(GIS). The latter provides detailed geographical information such as major locations,
roads, rivers and lakes, mountains, and even existing pipelines [10]. If major obstacles
are located along the preliminary route, the route may be modified before hydraulic
studies are performed. In later phases of design, the preliminary route can be modified
as more detailed information is made available. For existing systems, the routing considerations may be as simple as paralleling the existing system. However, a new routing may offer significant benefits such as cost savings or additional volume pickups or
deliveries over the paralleling option.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

86 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

The routing selection factors may include terrain, supply sources, population cen
ters, environmental constraints, and other impediments. The weighting of these factors
can vary from location to location, but cost and timing are the major considerations
along with environmental impacts. The following factors should be taken into consideration in selecting the pipeline route because of the significant impact they may have
on the pipeline economics and permitting requirements:
Pipeline right of way affects construction and land acquisition costs
Compliance with environmental regulations affect construction timing/methods and hence costs
Elevation profile directly affects hydraulics and pumping requirements as well
as construction cost
Depth of cover or burial affects hydraulics due to heat conduction and the integrity of pipe as well as construction cost
Soil types along the route affects construction cost and heat conduction
Water crossing including rivers affects construction cost, requiring extra valves
and overcoming other environmental restrictions
Geotechnical considerations such as slope stability, earthquake, permafrost,
muskeg, etc.
Environmental assessments help pipeline operators develop the guidelines for
the pipeline system during the design, construction, and operation phases. They are
intended to protect the possible varied environments along the pipeline route. The following environmental issues may arise along the proposed pipeline route:

Soil resources/farm land

Protected areas
Areas of potential archaeological value
Wildlife, endangered species, etc.

3.2.2.3 Operating Parameters

Since the final purpose of the design is satisfactory system operation, the operating
parameters have to be defined in an early phase of the design. They may include operating flow range, operating pressures and temperatures, fluid properties, and ambient
conditions.
For optimum design and operation, required factors are not only the future growth
of the system throughputs, as discussed in Section 3.2.2.1, but also maximum and
minimum daily or annual throughputs. The pressure drop is almost proportional to the
square of flow rate or flow velocity. Liquid velocity in a pipeline is the velocity averaged across the cross section of the pipe and is calculated as follows:

V = Q/A

(3 30)

where:
V = Liquid velocity
Q = Flow rate
A = pipe cross sectional area
It may be noted that there are a number of situations where selecting a pipe size based
on the optimum fluid velocity is not appropriate and a detailed analysis will be required
The pressure gradient or pressure drop per unit length of pipe is an important
measure for designing a safe and economic pipeline system. Since the liquid velocity

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 87

is directly related to the frictional pressure drop, the maximum velocity is used as a
guideline for an optimum system design. In other words, the required facilities such as
pipeline and pump station and operating costs can be minimized by keeping the velocity around an optimum velocity. The maximum velocity can be different for fluids with
different density and viscosity. It also depends on surge conditions, potential erosion,
facility limits, and economics. Refer to Addendum 3.2 for the discussion of erosional
velocity.
Pipeline and piping a major proportion of a pipeline and facilities costs (for example petrochemical plants, piping makes up 20% to 30% of the total capital costs).
Therefore, optimizing the pipe size is a key to reducing capital costs.
The optimum pipe diameter is a balance between two opposing factors: material
costs and pumping (energy) costs. To obtain an exact optimum size would require a
rigorous analysis taking into account: energy costs and capital costs of pumps/piping.
These factors will change over time and several of them may be difficult to determine
accurately [9]. The following provide fluid velocity ranges that typically provide optimum velocity and hence pipeline diameter operation:
3.2.2.3.1 Low-Viscosity Liquids For low-viscosity liquids, (i.e., with a viscosity of
less than 10 cSt e.g., water, light oils, caustic solutions),
Pipe diameter
Below 75 mm NB (Nominal Bore)
75 mm NB to 150 mm NB
100 mm NB to 200 mm NB
Above 200 mm NB

Suggested velocity
0.9 m/s to 2.0 m/s
1.5 m/s to 3.5 m/s
1.8 m/s to 4.0 m/s
2.4 m/s to 4.5 m/s

These figures approximate only but generally provide an economic pipeline and
piping design.
3.2.2.3.2 High Viscosity Fluids As the liquid viscosity increases above 10 cSt, the
suggested velocities are lower than those listed above. However, for high viscosity
liquids (i.e., these with viscosities approaching 1000 cSt and higher), pipeline and piping design would not be based purely on economic factors. For high viscosity liquids,
keeping the pressure drop to within acceptable limits is likely to be the key. It may be
noted that there are a number of situations where selecting a pipe size based on the
optimum fluid velocity is not appropriate and a detailed analysis will be required.
No pipeline systems can operate continuously for a full calendar year due to operational restrictions such as system maintenance or other reduced capacity operations.
The average daily flow is obtained by dividing the annual throughput by 365 (yearly
calendar days), and the actual maximum daily flow by the actual number of operating days. The ratio of operating days to calendar days is called load factor, so the load
factor can be defined as the average daily flow divided by the actual maximum daily
flow. Normally, the maximum daily flow is used for design in order to compensate for
the downtime. In the design procedure, a load factor of up to 95% is used for a simple
pipeline, while it may be as low as 85% for more complex systems or pipelines operated with expected large flow variations.
The minimum flow rate has to be defined for system design and operation, because
all equipment has maximum and minimum operational limits in capacity and efficiency.
For example, a pump can only operate within a flow bound between the maximum and
minimum capacity. In a highly mountainous terrain, slack flow conditions may occur
at low flows so that extra equipment specifications are required to operate the pipeline
safely. Refer to Section 3.3.3 for a detailed discussion of slack flow conditions.
Choice of operating pressures directly affects pipeline safety and operating requirements. The requirements include shipping capacity and volume demands, location

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

88 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

and method of installation, and the type of pipe material selected. The operating pressure of a pipeline must be maintained within minimum and maximum pressures. These
pressure limits are critical for safe and efficient operation. The maximum operating
pressure in a liquids pipeline is constrained by the yield strength of the pipe material,
pipe diameter and wall thickness, the fluid density and the elevation of the lowest
point of the pipe, while the minimum pressures are determined by vapor pressures of
the liquids along the pipeline. The elevation affects the operating pressure due to high
static head for liquid pipelines.
The delivery pressure is generally defined in the contract between the pipeline
company and the shippers or third party pipeline to which the fluid is delivered. The
determination of the delivery pressures is influenced by the terminal equipment such
as tank and control valves as well as the elevation profile upstream of the terminal. A
peak elevation can dictate the pressure required, which can result in higher delivery
pressure at the terminal. The delivery pressure is determined by the fluid vapor pressure, pressure rating of the equipment at the delivery site, and pressure requirements
imposed by the delivery facilities such as a tank or connecting pipeline. Therefore,
the delivery pressure requirement dictates the operating pressure for a given flow
rate.
As noted earlier, temperature affects viscosity, density, and specific heat in liquid
lines. A temperature rise is beneficial in liquid pipelines as it lowers the viscosity and
density, thereby lowering the pressure drop. The cooling effect on non-Newtonian or
viscous fluids can be significant because their viscosity can increase significantly and
subsequently the pressure drop can be very high. To reduce the effect of temperature
cooling, the pipeline can be insulated and/or operated at high temperature. The viscous
fluids can be blended with light hydrocarbon liquids such as condensate. The temperature along the pipeline is least controllable due to its dependency on variable soil
thermal conductivity and ambient temperature.
The maximum temperature limit for buried pipe is determined by a combination
of the following three factors:
Ground conditions
Stress level the pipe material can withstand without buckling
Economics of pipeline flow (the liquid flows most efficiently at high
temperature)
The minimum temperature limit is normally determined by the metallurgical (fracture toughness) properties of the pipe material or by the ground conditions.
Fluid properties were fully discussed in the previous chapter. Summarized below
are fluid properties that directly and indirectly affect the design and operation of liquid
pipeline systems.
Density or specific gravity the higher the fluid density, the higher the pressure drop. The pressure drop due to friction is directly proportional to the fluid
density.
Compressibility or bulk modulus is not important for liquid pipeline capacity
calculation, but important for controlling pressure surges and determining line
pack changes.
Viscosity is important in calculating line size, hydraulics, and pumping requirements for liquid pipelines.
Vapor pressure determines the minimum pressure in the pipeline. It must be
high enough to maintain the fluid in a liquid state and to avoid cavitation at
inlet to a pump.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 89

Pour point is the lowest temperature at which oil flows and around which it
starts behaving more like a non-Newtonian fluid. Oil can be pumped below the
pour point, but here the design and operation require special consideration and
pumping equipment. It should be noted that the change in fluid characteristics
occurs gradually at a higher temperature than the pour point.
Specific heat affects heat transfer rate through conduction processes between
fluid and surrounding soil.
The ambient parameters include ambient air temperature and ground conditions.
These parameters play a critical role in design and operation, particularly for long pipelines or for pipelines in extreme environments such as a desert or the Arctic. In permafrost areas, for example, the fluid has to be chilled to a few degrees below 0C to avoid
melting the surrounding frozen soil. The ambient air temperature affects turbine driver
thermodynamic performance as well as the fluid properties due to conduction. Since
ambient conditions change daily and seasonally, these variations have to be taken into
consideration in design and operation.
Most pipelines are buried for various reasons. Even though it is costly to bury
pipe, buried pipelines offer significant advantages over aboveground pipelines:
Limited changes due to ambient temperature and minimum effects on fluid
properties such as viscosity
Pipeline is restrained by the soil along its length
Protection from intentional or accidental damage as well as against expansion
and contraction from ambient temperature changes
Allows surface use of pipeline right of way.
The greater the depth of burial, the lower the rate of heat transfer. The effect of soil
thermal conductivity on the fluid depends on the differential temperature between the
fluid and the surrounding soil. If the soil temperature is colder than the fluid temperature, the fluid temperature drops. This results in higher viscosity and a higher pressure
drop. If the soil temperature is hotter than the fluid temperature, the opposite results
occur. The following parameters are required to determine the temperature profile due
to heat transfer along the pipeline:

Receipt temperature determine the temperature profile along the pipeline

Soil temperature
Thermal conductivity
Depth of cover
Thermal insulation properties
Ambient temperature has a direct impact on soil temperature and turbine
performance

3.2.2.4 Pipe Parameters

Most liquid transmission lines are constructed of steel pipes. Steel pipes are structurally
strong and ductile; they do not fracture easily. Steel pipes are made of various grades
of steel with yield strength in the range of 30,000 to 120,000 psi. In the hydraulic design, line size is initially based on a preliminary choice of pipe grade, diameter, and
wall thickness from experience. Further calculations are needed to finalize the system
design based on the code requirements, project cost, and material availability.
The profitability of a pipeline operation is directly related to how much volume
is delivered from sources to destinations, and the maximum throughput is mostly determined by pipe size and pressure. Pipe grade, diameter, and wall thickness are the

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

90 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

largest factors in determining the throughput capacity. They also affect pipeline operating pressure and thus overall economics:
Pipe size the larger the inside diameter of the pipeline, the more fluid can be
moved through it and the smaller the pressure drop per unit length.
Pipe wall thickness determines the steel and construction cost and operating
pressure.
Pipe grade determines the steel strength and the operating pressure affecting
pipe construction and operating costs.
Pipe roughness affects pressure drop and cleaning pig run frequency.
Pipe coating protects against corrosion and other damage by inhibiting the
flow of electric current from the pipe to the surrounding soil.
API 5LX specifications are often applied to the acquisition of high pressure
steel pipeline in Grades X42 through X80.
Pipe size is the largest factor in determining the throughput and one of the most
important parameters in the design and operation of pipeline system to meet a set of
projected flow profiles. The minimum size may be selected based on the maximum
input pressure and the minimum output pressure for short pipelines, while the pipe
size together with other factors including pumping facilities have to be optimized for
longer systems.
Pipes are designated in pipe size, pipe wall thickness, and weight. A common
designation of pipe size is the nominal pipe size (NPS), which indicates the outside
diameter of a pipe. The internal diameter of the pipe defines the cross-sectional area
available for the flow of fluids. It is obtained by subtracting twice the pipe wall thickness from the outside diameter. For a given pipe diameter, several different wall thicknesses are available to satisfy different levels of the maximum design pressure. For a
specified pipe design pressure, the pipe wall thickness varies with the pipe grade and its
elevation changes. Nominated pipe sizes and wall thicknesses are intended to standardize pipelines and associated facilities.
The nominal pipe size outside diameter is expressed in millimeters or inches. The
weight of a unit pipe length is determined by the actual pipe size and wall thickness.
Table 3-4 lists pipe sizes and wall thicknesses with their corresponding weights.
3.2.2.4.1 Design Pressure The three pipe parameters determine the level of internal pressure that a pipe can withstand. Associated with these parameters together with
a design or safety factor is the maximum design pressure. The design pressure sets
the maximum limit that the pipeline is allowed to be pressurized safely. The design
Table 3-4. Pipe Size, standard wall thickness, and weight
Nominal pipe
size (NPS)
6
8
10
12
14
18
20
24
32
36
42
48

Pipe OD (mm/in)

Standard wall
thickness (mm/in.)

Weight (tons/km)

168.3 / 6.625
219.1 / 8.625
273.1 / 10.752
323.9 / 12.752
355.6 / 14
457.2 / 18
508.0 / 20
609.6 / 24
812.8 / 32
914.4 / 36
1066.8 / 42
1219.2 / 48

7.11 / 0.280
8.19 / 0.322
9.27 / 0.365
9.53 / 0.375
9.53 / 0.375
9.53 / 0.375
9.53 / 0.375
9.53 / 0.375
9.53 / 0.375
9.53 / 0.375
9.53 / 0.375
9.53 / 0.375

28.27
42.54
60.32
73.84
81.31
105.18
117.11
140.98
188.72
212.59
248.40
284.20

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 91

pressure is determined by modifying Barlows formula for a given pipe grade, pipe
size, and wall thickness to include a design safety factor:

Pdesign = (2S t/D0) F L J T

(3 31)

where
S = specified minimum yield strength (SMYS) of pipe, kPag, or psig
t = pipe wall thickness, mm or in.
D0 = outside pipe diameter, mm or in.
F = design factor or safety factor
L = location factor (L = 1 for liquid pipelines)
J = joint factor (to reflect the method of pipe joining generally taken to have a
value of 1)
T = temperature derating factor, to account for the effect of higher temperatures
on yield stress
The SMYS is a standard measure of the specified minimum yield strength for steel
pipe. Standards that are frequently used by the pipeline industry are API 5L: Specifications for Line Pipe, which includes API 5LX and 5LS. API 5LX specifies various
strength grades, ranging from Grade B, rated at 42,000 psig (289 MPag) to Grade
X120, rated at 120,000 psig (827 MPag), where the Grade X120 refers to the SMYS in
1000 psi. Pipes are manufactured to these specifications.
ASME B31.4 does not define the location factor. The design factor, F, specified
in ASME B31.4 is 0.72 for liquid pipelines regardless of the location of the pipeline,
while other codes such as CSA Z662 define the design factor differently depending on
the locations. The joint factor is 1 for all types of pipe manufactured to 5LX and 5LS
specifications. The temperature derating factor is generally taken as 1 for transmission
pipelines, because transmission lines are seldom operated beyond the temperature derating range. Several mechanical design aspects are discussed in the next chapter.
Effective pipe roughness is a pipe parameter that affects frictional pressure drop
and pipeline efficiency. It includes pipe roughness as well as other pressure loss terms
such as bends, welds, fittings, etc. It directly influences the friction factor of the fluid
flow; the larger the pipe roughness, the higher the frictional resistance. To reduce
roughness, pipes are internally coated or cleaned by pigging. Several examples of pipe
conditions and their corresponding roughness are listed in Table 3-5, showing also that
pipe roughness varies with pipe conditions.
3.2.2.4.2 Maximum Allowable Operating Pressure (MAOP) In the design of
pipelines and their components, the design engineer must ensure that the design pressure at any point along the pipeline is lower than or equal to the maximum design
pressure or maximum allowable operating pressure (MAOP). As discussed earlier,
the design pressure is proportional to pipe strength and the MAOP defines the maximum pressure permitted for steady-state pipeline operations which relates to the pipes
ability to withstand internal pressure. The MAOP is the sum of the pressure required
to overcome friction losses, static head pressure, and any required back pressure or
Table 3-5. Pipe roughness
Pipe conditions

Roughness (in.)

Roughness (mm)

New clean bare pipe

Scraper burnished pipe
Internally coated pipe
Pipe after two years of
atmospheric exposure

0.00050.0008
0.00030.0005
0.00020.0003
0.00180.0020

0.01270.0200
0.00760.0127
0.00510.0076
0.04450.0508

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

92 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

d elivery pressure. Therefore, the values of a point specific MAOP along the pipeline
vary with elevation changes.
In many jurisdictions, MAOP is obtained by choosing the lowest of the following
four values in a pipeline section:
Design pressure determined by Barlows formula,
Pressure established during hydrostatic testing of pipe with hydrostatic pressure limit equal to 80% of hydrostatic test pressure (hydrostatic test pressure
results in 90% of SMYS for new pipe), which is illustrated in Figure 3-9. Note
that in Canada and a number of other jurisdictions test pressures causing the
pipe to reach or go slightly above yield are permitted,
Flange rating: B16.5 based on grade, material and operating temperature,
Documented historical operating pressure.
Figure 3-9 shows the MAOPs determined after a hydrostatic test is performed,
assuming that the pipe grade, diameter, and wall thickness are uniform. Hydrostatic
testing must be performed on new pipelines, as specified in ASME B31.4 and other
standards, prior to in-service use. Hydrostatic testing is also used on operating pipelines to assess their structural integrity. For testing a new pipeline, the pipeline is divided into multiple pipe segments, which are tested individually. The length of each
segment and hence the overall number of test sections is determined on the basis of
acceptable elevation changes within the segment.
After a certain period of operation, some segments of pipe may have corroded
internally or externally, and thus effective pipe wall thickness is reduced. In such cases,
the new pipe wall thicknesses have to be determined and the pipe repaired or else the
MAOP of the segments must be lowered.
3.2.2.4.3 Pipe Wall Thickness Pipe wall thickness seldom remains uniform along
the pipeline. ASME B31.4 requires that an allowance of 10% over the internal design
pressure or 80% of specified minimum yield strength (SMYS) of the pipe is made to
take into account surges and other operational changes in pressure. After an optimum
pipe wall thickness is determined, a thorough transient analysis is performed using
potentially worst case operation scenarios. Based on this analysis, the pipe wall thicknesses need to be increased to satisfy local transient pressure requirements or can be

Figure 3-9. Hydrostatic test and MAOP

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 93

decreased not only to satisfy safe pressure requirement but also reduce pipe cost. As a
rule of thumb, pipe wall thickness tends to be larger than the optimum thickness around
river crossings or in deep valleys, while it is smaller at the highest elevations.
Section 401.2.3 of B31.4 specifies that a component of the pipeline system shall
be designed to withstand the maximum differential pressure between external and internal design pressures. External surface loading on the buried pipe at road and railroad crossings, or caused by heavy agricultural equipment may require extra pipe wall
thickness.
3.2.2.5 Pumping Parameters
All liquid pipeline systems have one or more pump stations in order to boost the pressure level of the liquid. In the early phase of the pipeline system, the number of pump
stations may be small due to low flow rate. As the flow rate requirements increase, one
way of addressing the system growth is to add more pump stations.
Pump characteristics and station design are detailed in the next chapter. Summarized below are the pumping parameters required for the selection of pumps and the
design of pump stations:

Pump Capacity
Performance curves
Operating ranges (flow, pressure and temperature)
Pump efficiency
Cooler/heater requirement parameters
Station auxiliary equipment requirements and specifications
Energy/Power requirements and specifications
Driver requirements and specifications
Piping requirements and specifications

3.2.2.6 Economic Factors

Several stakeholders are involved in building and operating a pipeline including both users and non-users of the pipeline system. Either directly or indirectly, these stakeholders
have an interest in the pipeline system. The users may include the shippers on the pipeline system as well as the owner and operating company. Non-users of the system are
land owners, the general public, environmentalists and multiple levels of governments.
Other non-users may include users of other transportation modes, such as trucking and
railroad companies, whose business could be directly affected by the pipeline system.
Some of the non-users such as land owners have an economic interest, but others such
as the general public may not be directly involved in the development of the system.
However, labor unions and/or environmentalists might show opposing interests to the
project; citing economic impacts vs. potential adverse consequences due to changes in
the socio-economic and natural environment. Governments, through their regulatory
agencies, make a decision by balancing the views of all of the stakeholders based on
sound engineering and economic merits. Therefore, an unbiased economic study including an environmental assessment is necessary to satisfy all the stakeholders.
For a new pipeline system, an economic study is necessary to provide a measure
of economic benefits for not only shippers and pipeline companies but also other key
stakeholders. The study must justify the need of a new pipeline system to satisfy the energy requirements in new markets. The study assesses the project feasibility, financing
requirements, and optimum system design and operation. If the pipeline is of strategic
importance for a country or a certain region, the assessment of the project feasibility
may not be critical. However, the need for a new pipeline or a major expansion of an
existing system can be justified through an economic analysis. The economic study
covers the financing requirements that may include the project profitability, amounts of

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

94 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

financing and their payment schedule. It also includes preliminary design and operation,
all costs, and comparative analysis of the capital costs along with the operating costs as
well as the proposed tariff structure in the case of a cost recovered public utility.
A pipeline economic analysis includes a process of optimizing the pipeline system, determining an optimum pipe size and pumping requirements over the life of the
project life. The economic study may include key, not necessarily all, design factors
discussed above. The optimizing process involves achieving a desired level of profitability, balancing the capital costs including material and construction against the operating costs. During the process, due considerations should be given to design factors
that are suitable for operating the pipeline system safely and economically.
The performance of an economic study is beyond the scope of this book, so no
attempt is made to discuss an economic analysis and tariff structures. However, some
of the major cost factors are discussed in this section because they influence pipeline
system design greatly and will be referred to again in the subsequent chapters:

Mechanical factors
1. Pipe grade, pipe size or diameter, and wall thickness
2. Pipeline route and depth of cover
Capacity factors
1. Operating parameters
2. Station spacing and pumping costs
Reliability and safety factors
1. Valve spacing
2. Other valve-related costs

3.2.2.6.1 Pipe Grade, Size, and Wall Thickness It is critical to optimize the pipe
grade, diameter, and wall thickness to minimize the project cost. The pipe cost is based
on the grade, diameter, and wall thickness. For most pipeline systems, the pipe cost
is the highest material cost. In addition, these three factors have a direct effect on the
cost of installation. Pipeline economics begins with the selection of the pipe material.
Since pipe material for transmission lines is steel, it boils down to the selection of pipe
grade. Higher grade steels are more costly to produce and because of their chemical
composition require specific welding procedures. Nevertheless they do result in thinner pipe wall hence less steel tonnage, lower transportation costs, and reduced amounts
of welding. A case specific study is needed to determine if such steels are the optimal
solution to a given project.
One common economic decision is whether to construct a large line initially, or
put in a smaller line first and parallel it or add pumps at a later time. Once the need for
a pipeline system is recognized, the maximum pipe size is determined such that it can
be economically optimized. The larger the pipe size, the larger the carrying capacity
and the lower unit shipping costs. The pipeline capacity increases approximately by
5/2 power for a fixed pressure drop, but the pipe material cost increases significantly
and construction costs increase almost linearly as the size is increased.
The design pressure is directly proportional to pipe wall thickness for the same
grade and size. The larger the wall thickness for a given pipe size, the higher the design
pressure. The larger the wall thickness, the higher the pipe and construction costs.
Higher grade pipe requires thinner pipe wall for the same design pressure, resulting in
lower steel weight and reduced cost even though higher pipe grade costs more per ton.
Cost savings can also result from reduced construction costs.
3.2.2.6.2 Pipeline Route Both direct and indirect costs due to time delays have to
be taken into account in selecting a pipeline route. As noted earlier the costs of selecting a pipeline route are related to pipeline length, terrain features, intermediate supply

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 95

and delivery locations, cost and restrictions on facilities and land, and permitting requirements. If possible, a straight line is selected to minimize the pipe cost, and severe
mountainous terrains are avoided because of high construction, pumping and maintenance cost requirements. Obtaining right-of-ways for certain portions of the route can
be difficult or even impossible due to environmental restrictions or land claims.
The determination of pipeline location must take account of population density, as
well as the proximity of features such as roads, railways, rivers, lakes, unusually sensitive areas, etc. The route should be evaluated in terms of the safety and environmental
issues, accessibility, extra material requirements, land claims, etc. Also, the locations
of facilities have a direct influence on construction cost.
The minimum depth of cover from a safety standpoint is specified in the applicable
codes and standards. However, the operational requirement depends on the temperature condition and thus varies along the pipeline route, particularly for long pipelines.
The effect of depth on the installation and labour cost component is largely dependent
upon the burial depth, soil conditions and location. Extra labour, material and/or equipment costs are incurred for conditions such as rocky ground, soft ground, e.g., muskeg,
river beds, roadbeds, railway crossings, etc.
3.2.2.6.3 Operating Parameters No extra cost is associated with the flow rate
because the design is based on it. Since operating pressures are based on maximum
allowable operating stress levels of pipe grade, pipe size and wall thickness, and class
location factors, a range of design pressures is available in the design phases. If higher
operating pressure is selected, the station spacing is increased, resulting in lower material and energy costs.
If the fluid viscosity is sensitive to temperature, the major cost items could be the
provision of heaters and heating, pipe insulation, and/or a blending operation.
3.2.2.6.4 Station Spacing and Pumping Costs Station spacing is determined by
factors such as pipe size, flow profiles, hydraulics and elevation profile, and capital and
operating costs. In an environment of high energy cost or rapid increase in flow, the
option with a larger pipe size is preferred, even though its capital cost is higher than
that of a smaller size. For a given flow profile, the larger the pipe size, the longer the
station spacing. The longer the station spacing, the lower the capital costs associated
with station construction and the pumping cost associated with power and energy.
3.2.2.6.5 Valve Spacing Valves are significant cost items. Placement of valves
provides for effective control of pressure or flow; sectionalizing the system in case of
emergency, isolation of components of the system, etc. The minimum valve spacing
and operation requirements are specified in the applicable codes and standards. The
number and locations are determined by such factors as system layout, product, adjacent population density, proximity to river crossings, etc.
3.2.2.6.6 Other Valve-Related Costs Other valve related costs have to be considered for safety in certain designs: namely, the need for and location of pressure-reducing
valves and pressure relief valves. The latter is discussed in Section 5.1.3 Surge Control.
3.2.2.6.7 Pressure-Reducing Station (PRS) A pressure-reducing station (PRS) is
usually installed to reduce the back pressure of a pipeline if the pipeline is sloping
down severely. This is due to the static pressure increase beyond the MAOP caused by
the elevation gain on the downstream side (refer to Section 3.3.3). A PRS is installed
to maintain the downstream pressure below the MAOP, independent of the upstream
pressure, unless the upstream pressure becomes less than the downstream pressure set
point. Through the downstream pressure controlling process, the upstream pressure
can be increased. The installation of a PRS has both cost and operation implications;
a PRS requires not only various types of valves including a relief valve and relief tank
but also a pig trap and launcher pair. An example of a PRS operation is discussed in
Section 5.1.4.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

96 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

Major capital costs are 30% to 40% of the total capital cost in material, 35% to
50% in labor and construction, 5% to 10% in right of way, and 12% to 15% in miscellaneous items. The materials include pipe, pump stations, valves and fittings, meter
stations, SCADA and telecommunication equipment, and tanks and manifold piping,
while the miscellaneous items include engineering, surveying, administration, regulatory filing, freight, taxes, etc. Among the major operating costs, general and administration costs such as payroll is the largest, and power and energy cost the next largest.
The rest are SCADA and telecommunication costs, utility costs, lease costs such as
ROW easements, office buildings, etc.

3.2.3 Hydraulic Design Procedure

A pipeline design process describes a way of combining the design considerations with
appropriate codes and standards. Of course, all of the design factors discussed in the
previous section are not always required; a certain type of pipeline design requires a
certain set of factors and another type requires a different set of factors. For example,
consideration of the pour point of light crude may not be required in a warm temperature environment, but may be required for a viscous heavy crude in a cold temperature
environment.
The system design is done in several phases; conceptual design, system planning,
and detailed engineering design. In the conceptual design phase, the following data is
available with which a preliminary hydraulic study is performed:

Product properties such as gravity and viscosity

Flow profile over the life of the project
Pipeline length and preliminary route with the points of injection and delivery
Macro-economic data such as trends of economic growth, demographic
changes, etc.

The conceptual design may include hydraulic and economic studies, which result
in overall system and financial requirements.
There are several types of pipeline system design; a new pipeline system, increasing the design capacity of an existing pipeline, and delivery from and/or injection to
other points outside the existing system. The increases in the capacity of the existing
system may require additional pump stations, a parallel line, or replacement of the
existing pipeline with a larger pipe size. Also, the route of the existing pipeline can be
moved due to significant supply/demand changes, or some existing pump stations may
be relocated to other sites to improve the operational efficiency and subsequently to
increase capacity.
After the conceptual design is approved, the pipeline system design is done to
achieve the minimum combined capital and operation costs. In the system planning
phase, the hydraulic and economic evaluation studies are performed in relative detail,
by taking into account the product properties and volumes to be transported, pipeline
route and terrain data, operating temperature ranges and possibly preliminary pressure
ranges, economic and financial data, and other factors such as environmental conditions and restrictions. Described below is a process for performing hydraulic and
economic studies:
1. Gather data
Receive the commitment from shippers for the proposed pipeline
Forecast the supply/demand volumes
Select a preliminary route for the pipeline

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 125

d ifference is large, the effects on the pressure and temperature profiles will also
be large.
Higher flow rates result in greater friction losses and thus lower pressures,
causing lower density and higher velocities. At the higher flow rates, the temperature is high and this, together with the low pressures, results in a further
pressure drop with flow rate.
The elevation effects on the pressure drop in the uphill segments are different
from that in the downhill segment. In the uphill segment, the total pressure
gradient remains the same, because the decrease in the frictional pressure gradient is compensated by the increase in the static pressure increase rate due
to the elevation gain. In the downhill segment, however, the magnitude of the
hydrostatic term exceeds the magnitude of the friction term, resulting in less
pressure drop.
The magnitude of these effects depends on the rate of change of the fluid properties with pressure and temperature under the particular flowing conditions. In many
cases, accurate values for these design parameters are unknown. For example, soil
temperature will vary considerably from place to place, adding another uncertainty. In
such situations, one should perform calculations for a range of values to examine the
overall uncertainty in the calculated pressures and temperatures.
Normally, when designing a liquid pipeline, consideration is given to use the
maximum flow that is required at a specific time and a larger pipe size taking into account future volume increase. However, a HVP and particularly dense phase pipeline
is designed with the following criteria in mind, , high operating pressure, because of
uncertainties in defining the product properties in their operating ranges, and limited
accuracy in determining temperature profile:
Low flow velocity resulting in low pressure drops to operate at high pressures,
requiring a larger size pipe,
A high pressure required at the storage facility of the HVP products, also requiring high delivery pressure,
Overpressure problem if the pipeline is shut down for a prolonged period, requiring blowdown valves,
Frequent block valve spacing to reduce spillage and increase safety,
Installation of blowdown valves on either side of each block valve to relieve
an overpressure condition or deal with other emergency conditions. For added
safety purposes, the blowdown valves need to be automated and the isolated
segment has to be blown down as quickly as possible. Normally, a flaring system may be provided to flare the spillage.
Hydrate problems in HVP pipelines have been reported in the presence of free
water. Since it is impractical to control the pressure and temperature conditions for
forming hydrates, it may be simpler to reduce the contents of free water.
Example: Ethane Pipeline
A pipeline company plans to build an NPS 12 pipeline, transporting ethane, with wall
thickness of 0.219. The total length of the pipeline is 200 km and its elevation profile
is assumed to be flat. Initially, no intermediate pump station is planned. The yearly
throughput is expected to grow to 1,500,000 tons. The inlet pressure is planned to be
600 kPa less than the maximum design pressure and the maximum inlet temperature
is 30C. Refer to the pressure-enthalpy diagram shown in Figure 3-18. Determine the
minimum operating pressure, and pressure and temperature profiles using the following data:

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

126 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

Pipe grade: X56

Maximum inlet temperature: 30C
Ground temperature: 4C
Heat capacity: 4.76 kJ/kgC
Ethane viscosity: 0.14 cSt
Soil conductivity: 0.5 W/mC
Depth of cover: 1.2 m

Solution:
Refer to the pressure-enthalpy diagram, Figure 3-18, which shows the phase behavior of ethane. The Pressure-Enthalpy diagrams show pressure on the vertical axis and
enthalpy on the horizontal axis. The diagrams are used in locating pipeline operating
points in terms of pressure and temperature and for designing control valves. Pipe
flow is almost an isenthalpic process, so the diagram shows a graph of the enthalpy
during various pressures and physical states. The critical point is defined at the critical pressure and critical temperature (point C in the figure), where the liquid phase
and vapor phase meet, and either phase cannot be distinguished. The rectangular
box in the diagram shows the operating pressure range of an ethane pipeline for an
operating temperature range (assuming that the operating temperature ranges from
0C to 30C (solid lines in the figure) and the pressure from 4500 kPa to 10,000kPa).
Since the operating temperature range is lower than the critical temperature, the
ethane in this operating condition remains in liquid phase. For different operating
temperatures, the operating pressure range should be different to avoid vaporization.

Figure 3-18. Ethane pressure-enthalpy diagram [2]

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 99

Section 3.3.6 describes a HVP pipeline design process with an ethane pipeline
as an example.
Although the heating effect on viscosity is inherent to all real fluid flow situations, the temperature effect on viscosity of heavy and waxy crudes is significant. Temperature of the highly viscous fluids at the entrance to the pipe can be
quite different from the temperature of the soil surrounding the pipeline system.
Viscous liquids such as heavy oil or waxy crude may be heated or blended with
lighter hydrocarbon liquids to reduce the viscosity for pumping. Section 3.3.7
describes a heavy oil pipeline design process as an example.
Intermediate hydrocarbon liquids such as light or medium crude and refined products such as diesel or gasoline are not as sensitive to temperature in terms of density and viscosity. Also, frictional heating is negligibly small for these products.
Therefore, the assumption of isothermal flow is reasonable if an adequate average
temperature is used for the operating temperature. However, the design consideration should include the vapor pressure because it depends on temperature. Section
3.3.1 begins with an isothermal pipeline system design example, demonstrating
the hydraulic design process. An average flow profile is added to the base design
problem in order to demonstrate the above design process for a realistic design
problem. The last three steps are not included in these examples because as mentioned quantitative economic analysis is beyond the scope of this book.

3.3.1 Crude Oil Pipeline System Isothermal Flow

Example: A crude oil pipeline from CE to QU is 200 km long. Refer to Figure 3-2 for
the pipeline configuration. At the injection point, crude oil of 32API (specific gravity
of 0.8654) and ambient pressure enters the pipeline at an initial flow rate of 18,000
m3/d at 15C. The operating temperature in winter and summer is 4C and 14C, respectively. Design a crude oil pipeline to transport the amounts defined in the flow
profile, using the data listed below:

Density: 865.4 kg/m3 at 15C and 875.4 kg/m3 at the operating temperature
Viscosities at 4C: 43.5 cSt
Pipe roughness: 0.0457 mm
Delivery pressure: 350 kPag
Load factor: 90%

The average flow profile is as follows:

Year 1 18,000 m3/d
Year 4 20,000 m3/d
Year 10 27,000 m3/d
Design an optimum pipeline system. Assume that the design factor of 0.72 as
specified in ASME B31.4 Codes is applicable and that the elevation profile is flat and
flow is isothermal.
Solution: The design considerations for this type of design problem are:
Satisfy the delivery pressure requirement that must be greater than the vapor
pressures of the delivered products,
Find an optimum solution in terms of not only capital and operating costs but
also hydraulics for flexible operation.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

100 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

Discussed below is the solution procedure described in Section 3.2.3, except an
economic analysis.
1. Gather data
It is assumed that the shipper commitments have been received, the approximate
volume forecasts are made, and a preliminary route of the pipeline is selected.
2. Prepare a set of design criteria.
Range of maximum operating pressure: from 8100 kPag to 9500 kPag based
on common practice for liquid pipelines.
Operating temperature: winter operating temperature of 4C is used.
Minimum operating pressure: 250 kPag
Pipe grade: X70 (483 MPag) and X65 (448 MPag)
Pipe sizes: 18 (457.2 mm), 20 (508.0 mm), and 22 (558.8 mm)
Pipe wall thickness: 0.25 (6.35 mm) and 0.281 (7.14 mm)
Maximum liquid velocity: 2 m/s
3. Develop a base design.
Step 1. Calculate the flow velocity and design pressure for each pipe grade,
size and wall thickness.
In the flow profile, the largest flow is scheduled from the 10th year on, and
the design flow rate is obtained by dividing the flow rate by the load factor:
27,000/0.9 = 30,000 m3/d. For the design flow rate, the table below gives the
flow velocity for each combination of the pipe size and wall thickness:
Pipe size
(in/mm)

Wall thickness
(in/mm)

Velocity (m/s)

Wall thickness
(in/mm)

Velocity (m/s)

18/457.2
20/508.0
22/558.8

0.250/6.35
0.250/6.35
0.250/6.35

2.24
1.80
1.48

0.281/7.14
0.281/7.14
0.281/7.14

2.26
1.81
1.49

The 18 pipe is excluded from further consideration because the velocity

exceeds the velocity limit by more than 10%, so the pipe sizes to be considered
further are 20 and 22. Next, calculate the design pressure for X65 and X60
grade pipes, respectively.
X65

X70

Pipe size
(in/mm)

Wall thickness
(in/mm)

Design pressure
(psig/kPag)

Wall thickness
(in/mm)

Design pressure
(psig/kPag)

20/508.0
20/508.0
22/558.8
22/558.8

0.250/6.35
0.281/7.14
0.250/6.35
0.281/7.14

1180/8132
1315/9067
1064/7334
1196/8246

0.250/6.35
0.281/7.14
0.250/6.35
0.281/7.14

1260/8688
1417/9765
1145/7897
1288/8880

Since the operating pressure range is between 8100 kPag and 9500 kPag,
the design pressures far outside of the range are removed from further consideration. Therefore, the selected combinations for X65 pipe are 20 with wall
thicknesses of 0.25 and 0.281, 22 with wall thickness of 0.281, and those
for X70 are 20 with wall thicknesses of 0.250 and 0.281, and 22 with wall
thickness of 0.281. The allowable design pressure for the 20 with 0.281 wall
thickness exceeds the maximum operating pressure, but the combination is
selected for further consideration because it is within a tolerance level.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 101

Step 2. Calculate the required total pressure drop and total pressure requirement or inlet pressure for the design flow rate of 20,000m3/d during the first
three years.
The design flow rate is obtained by dividing the given flow rate, 18,000m3/d,
by the load factor, 0.9. The total pressure drop is added to the delivery pressure
to get the total pressure requirement. The pressure calculation is based on the
worst condition, which is the winter temperature.

Pipe grade

Pipe size
(in/mm)

Wall thickness
(in/mm)

Design
pressure
(kPag)

Total
pressure
drop (kPag)

Total
pressure
req. (kPag)

X65
X65
X65
X70
X70
X70

20/508.0
20/508.0
22/558.8
20/508.0
20/508.0
22/558.8

0.250/6.35
0.281/6.35
0.281/7.14
0.250/7.14
0.281/7.14
0.281/7.14

8132
9067
8246
8688
9765
8880

7288
7288
4649
7400
7400
4649

7638
7638
4999
7750
7750
4999

All six combinations satisfy the total pressure requirements for 20,000 m3/d
flow. For the same combinations as above, calculate the required total pressure
drop and inlet pressure for 22,300 m3/d from the fourth year to the tenth year.

Pipe grade

Pipe size
(in/mm)

Wall thickness
(in/mm)

Design
pressure
(kPag)

Total
pressure
drop (kPag)

Total
pressure
req. (kPag)

X65
X65
X65
X70
X70
X70

20/508.0
20/508.0
22/558.8
20/508.0
20/508.0
22/558.8

0.250/6.35
0.281/6.35
0.281/7.14
0.250/7.14
0.281/7.14
0.281/7.14

8132
9067
8246
8688
9765
8880

8811
8946
5618
8811
8946
5618

9161
9296
5968
9161
9296
5968

Only the combinations of pipe size 22 with the wall thickness of 0.281 for
X65, and of the pipe size 20 with the wall thickness of 0.281 and the pipe size
22 with the wall thickness of 0.281 for X70 pipe, satisfy the pressure requirement with no intermediate pump station. It may not be cost-effective to install
and operate an intermediate pump station to accommodate a small amount of
the flow increase from the fourth year.
For the above three combinations, calculate the required total pressure and
inlet pressure for the flow rate of 30,000 m3/d from the tenth year on. It should
be noted that the pumping power requirement for the 20 pipe at the inlet point
is higher by 59% (8946/5618 = 1.59) than the power requirement for the 22
pipe size. Therefore, the pump units for the 20 pipe have to produce higher
head than those for the 22 pipe and thus their capital and operating costs are
higher. On the other hand, the required pressure for the 22 pipe is low for the
first 10 years, and so the facility usage would be limited unless further flow
increase is expected in earlier years.

Pipe grade

Pipe size
(in/mm)

Wall thickness
(in/mm)

Design
pressure
(kPag)

Total
pressure
drop (kPag)

Total
pressure
req. (kPag)

X65
X70
X70

22/558.8
20/508.0
22/558.8

0.281/7.14
0.281/7.14
0.281/7.14

8246
9765
8880

9432
15,039
9432

9782
15,389
9782

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

102 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

The total pressure requirements for all three combinations are higher than
their respective design pressure. Therefore, they require an intermediate pump
station to satisfy the total pressure requirement.
Step 3. Determine the number of intermediate pump stations and their power
requirements.
Only one intermediate pump station is required for all three cases because
the design pressures for all cases are less than half of the total pressure drops.
Assuming the suction pressure of the intermediate station is the same as the
delivery pressure, the discharge pressure at the inlet and intermediate stations
are as follows:

Pipe grade

Pipe size
(in/mm)

Design pressure
(kPag)

Discharge
pressure at inlet
point (kPag)

Discharge pressure
at intermediate
station (kPag)

X65
X70
X70

22/558.8
20/508.0
22/558.8

8246
9765
8880

5066
7870
5066

5066
7870
5066

The capital cost due to the extra pumping power requirement for the 20
pipe is higher than the cost for the 22 pipe size, while the pipe cost for X70
with 20 diameter may cost less than the other two options. The extra capital
cost for the 22 line is more than 20% and is incurred by the extra pipe material
and construction expenses. However, the extra capital cost of the 22 diameter
pipe might be partly compensated by lower unit pumping cost. Assuming that
the annualized cost for the 20 pipe case is lowest, it is selected as the base
design.
The facilities such as the initiating pump station for the selected base design
would be designed to accommodate the capacity until the capacity increases
in the 10th year. In the 10th year, the additional facility increases include the
pumping capacity at the inlet point for the additional flow and an intermediate
pump station with the pumping capacity of 30,000 m3/d.
4. Develop alternative design cases and perform comparative studies against the
base design
Alternative 1: This alternative design is to use a pipe wall thickness larger
than 0.281 in order to increase the design pressure slightly higher than the
total pressure requirement. No intermediate pump station is required if the design pressure is slightly higher than the total pressure requirement. Note that
the required total pressure will be increased due to slightly smaller inside pipe
diameter.
The design pressure for the X70 22 pipe is lower than the total pressure requirement, which in turn is lower than the maximum operating pressure range.
The next largest nominal wall thickness is 0.312 or 7.92 mm, and its design
pressure is 9857 kPag or 1430 psig, but the required total pressure is 9914 kPag
for a flow rate of 30,000 m3/d. Therefore, the wall thickness is not sufficient to
meet the total pressure requirement without an intermediate pump station.
The next largest nominal wall thickness is 0.344 or 8.74 mm, which
can allow the design pressure to increase up to 10,870 kPag. For this wall
thickness, the required total pressure turns out to be 10,051 kPag. Since this
design pressure is higher than the required total pressure, no intermediate
pump station is required for the flow rate expected beyond the 10th year,

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 103

and thus the capital and operating costs due to an extra pump station can be
saved. However, two points should be evaluated; the required pressure is
very high for a liquid pipeline and the extra capital cost. The required pressure in this case is much higher than the maximum operating pressure, and
normally crude oil pipelines are not operated at such a high pressure. Extra
pipe and construction costs will be incurred due to the extra pipe material
needed.
Therefore, these extra capital cost should be compared against the costs
of the base design in terms of annualized cost. The base design may a better
choice in terms of the overall cost and its pipeline system operation due to its
lower operating pressure.
Alternative 2: This alternative design is to use X80 grade pipe to increase the
design pressure, also allowing the operating pressure limit to be raised. If the
design pressure for this pipe grade is higher than the required total pressure of
9782 kPag, then no intermediate pump station is required even for the maximum flow rate.
For this grade, the design pressure is 10,149 kPag, which is higher than the
total pressure required from the 10th year on. Therefore, an intermediate pump
station is not needed. Also, the pipe material cost for X80 pipe is only slightly
higher than X70 pipe cost.
When compared against the base design, both designs are comparable, because this alternative design offers the lower cost solution even though its operating pressure range for a crude line seems to be high. To finalize the design, it
is necessary to perform sensitivity studies for these two designs.
Alternative 3: The base design is modified by adding storage tank capacity to
allow the system to transport more in summer during which time the transportation capacity is higher than during the winter when capacity decreases due
to lower viscosity and density. It costs much less to add the extra tank capacity
than to increase pipe diameter or wall thickness, but the tank operation does
add costs.
The summer capacity listed in the table below can be found by setting the
operating temperature at 14C and the discharge pressure at the maximum
operating pressure of 9500 kPag with and without an intermediate pump station; 24,580 m3/d without and 36,300 m3/d with an intermediate pumpstation.
For the same inlet pressure as for the base design, the capacity increasesto
23,280 m3/d without an intermediate pump station. Therefore,assuming that
the pipeline operates in summer condition for the first half of a year and in
winter condition for the rest of the year, this alternativedesign allows an
increase in the transportation capacity on a yearly basis.
This alternative design offers a more flexible solution than the base design,
even though it costs more. Also, its transportation capacity is 10% higher that
the base design capacity. To finalize the design, it is necessary to perform sensitivity studies for these three designs.
5. Perform a sensitivity study with respect to the flow profile.
Sensitivity on the modified flow profile: The flow rate is expected to gradually increase by approximately 1% yearly over the initially projected flow
profile; 21,000 m3/d or design flow rate of 23,300 m3/d for year 9 and more
than 27,000 m3/d or design flow rate of 30,000 m3/d beyond year 10. For
these flow rate changes, the required total pressures are calculated for the
three cases:

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

104 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

Design

Design
pressure

Base
Design 2
Design 3

9765
10,149
9765

Inlet pressure
(9th year flow)

Inlet pressure
(beyond 10th year)

9720
6234
9300

Discharge pressure
at intermediate
station

18,133
11,499
9300

9242
6275
9300

The base design barely satisfies up to the seventh year transportation requirement without an intermediate pump station, while Alternative 3 provides more than the ninth year requirement within the design pressure. Also,
Alternative 3 fully utilizes the facility, but Alternative 2 does not. Still, both
alternatives need an intermediate pump station from the 10th year on, and
the pumping capacity at the inlet station has to be increased at the same time.
The intermediate pump station will be located at 100 km from the inlet station, because the pumping head at both stations is the same (the criteria for
locating pump stations are discussed in the next chapter). With the discharge
pressure of 9300 kPag, Alternative 3 has a higher flow capacity than needed.
In summary, Alternative 3 is selected as the best design under this flow condition, because:
With slightly more capital and operating costs than the base design, Alternative 3 offers more flexible operation,
If needed, the flow capacity can be increased significantly.
Sensitivity on a fast flow growth: The flow rate is expected to grow at the
yearly rate of 1000 m3/d from the first year on and to level off at 32,000 m3/d;
18,000m3/d or design flow rate of 20,000 m3/d in the first year, 19,000 m3/d or
design rate of 21,100 m3/d in the second year, etc. For these flow rate changes,
the required total pressures are calculated for these three cases:

Design

Design pressure

Inlet pressure
(kPag)

Year of pump
installation

Base
Design 2
Design 3

9765
10,149
9765

9436
9782
9440

3rd year
10th year
4th year

Alternative 2 does not require an intermediate pump station until the 10th
year, while the other two require it in 3rd and 4th year, respectively. Alternative
2 needs higher initial capital cost due to the higher pipe grade and larger pipe
size. However, Alternative 2, using a large pipe size, offers a better option in
terms of the operating cost for such a high flow growth rate.

3.3.2 Pipeline Configurations

This section describes the key design and operation issues on different pipeline configurations. In addition to pipe, a pipeline network is composed of the following
facilities:
Injection points, also known as receipt or inlet stations, these are where the
products are lifted or injected into the line. Storage facilities such as tanks and
booster pumps are usually located at these locations.
Delivery point, also known as terminal, is where the product will be delivered
to the final consumer or to another pipeline.
An intermediate station can provide a side stream injection or delivery point.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 105

These stations allow the pipeline operator to inject or deliver part or all of the
product being transported.
Pump stations are located along the line to move the liquid through the
pipeline.
Block Valve Stations are the first line of consequence mitigation for pipelines.
With these valves the operator can isolate any segment of the line to perform
some specific maintenance work or isolate a rupture or leak. Block valve stations are usually located every 20 to 30 km, depending on the type of pipeline
and applicable standards.
Regulating station is a special type of valve station, where either pressure or
flow is controlled. Pressure regulators are usually located on the downhill side
of a peak, while flow regulators are installed at delivery stations.
Depending on the requirements and arrangements of these facilities, liquid pipeline networks can be diverse; some are short and straight, some are long with multiple
pump stations, or some are complex with multiple injection and delivery points. The
pipeline system design and operation has to comply with the required system network.
A simple pipeline consists of one inlet with a pump station and one delivery. In addition to simple networks, the following types of pipeline networks can be built and are
frequently encountered:
Pipelines including one injection and one delivery with multiple pump
stations
Pipelines including multiple injection and multiple delivery points with multiple pump stations
Pipelines with branch or lateral lines that connect to/from other pipelines or
facilities from/to the main line
Series pipelines of partial or entire length, referring to the connection of pipes
of the same or different diameters in series.
Parallel Pipelines of partial or entire length to increase throughput by reducing
pressure drop.
Note that the pressure gradients for these networks, except the first type, vary
because the flow rate of each segment is different, and so is the pumping requirements.
If such a network is anticipated in the initial design phase, the pump station locations
are determined accordingly. If the existing network has to be modified to meet the new
requirements, additional pump stations are added and/or certain stations need to be
modified.
3.3.2.1 Side Stream Delivery
Liquid may be delivered off the pipeline (stripping) at intermediate locations, thus
reducing the main line flow rate while the remainder of the product continues to the
main line terminal. The final delivery location is right on the main line or connected
through a branch line. Since the downstream flow is lower, the frictional pressure
drop is lower. Normally, a holding pressure control valve is installed at the delivery
point to maintain the delivery pressure level or a pressure regulator is placed on the
branch line. A block valve is installed downstream of the take-off point on the main
line and branch line to block the flow when a full stream delivery takes place on
either line.
The modes of side stream delivery operation can vary depending on the delivery
flow requirements or availability, nomination status, and pipeline operational status.
For example, the main line downstream of the take-off point cannot be operational
if a line break occurs there, or the branch line should be shut down if no volume is

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

106 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

n ominated to the branch line delivery site. Therefore, the following modes of side
stream delivery operation are possible, and thus have to be included in the design:
Strip delivery through the branch line or at the delivery point as originally
designed,
Full stream delivery through the branch line due to the main line problem in the
segment downstream of the take-off point,
Full stream delivery through the main line due to a problem in the branch line.
The design considerations for this type of design problem are:
Satisfying the delivery pressure requirements at both delivery locations, while
maintaining sufficiently high pressure at the take-off point. Note that the delivery pressures at both locations can be different because the delivery conditions
can be different.
Using a pipe with a smaller diameter downstream of the delivery point if the
side stream delivery volume is large.
Installing an extra facility such as a pressure regulator or pump at the take-off
point on the branch line in order to satisfy the branch line delivery pressure
requirement.
Selecting pumps to meet the maximum and minimum flow requirements. When
the main line is shut down downstream of the take-off point, the minimum flow
rate along the main line can be as low as or even lower than the design flow
rate of the branch line.
Example: A crude oil pipeline from CE to QU is 200 km long. It is constructed of
5LX-70 steel pipe with NPS = 20 and a 0.281 wall thickness. At the CE terminal,
the crude oil of 32API (specific gravity of 0.8654) enters the pipeline at the design
flow rate of 30,000 m3/d. Crude oil is taken off at TO, 136 km downstream of CE,
where up to 7200 m3/d is stripped off the pipeline, and the rest is delivered to the final
destination, QU. Occasionally, the full flow has to be delivered to QU. At TO, a 50-km
branch line is connected to a third party pipeline, which requires a delivery pressure of
3000kPag. This branch pipeline is constructed with X52 grade pipe, and the pipe size
is NPS= 12 (actual pipe diameter = 12.75) with a 0.219 wall thickness.
Determine the pressure requirements of the pipeline system, using the following data:

Average operating temperature: 4C

Density: 865.4 kg/m3 at 15C and 875.4 kg/m3 at the operating temperature
Viscosity at 4C: 43.5 cSt
Pipe roughness: 0.0457 mm
Delivery pressure at QU: 350 kPag

Assume that the design factor of 0.72 is applicable and that the elevation profile is
flat and flow is isothermal. Figure 3-10 shows the configuration of this pipeline system.
Solution:
It is assumed that the Alternative 3 design has been used for the main line in anticipation of flow increase and the intermediate pump station has been operating.
Step 1. Determine the design pressure of the main and branch lines using the Barlow formula with the hoop stress limited to 72% of the SMYS
Pmain = 2 70,000 0.281 0.72/20 = 1416 psig = 9765 kPag
Pbranch = 2 52,000 0.25 0.72/12.75 = 1286 psig = 8868 kPag

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 107

Figure 3-10. Side stream take-off

Step 2. Calculate the required pressures at TO and the discharge pressure at the
intermediate pump station.
1. First calculate the pressure required at TO for the design flow rate that can meet
the branch line delivery pressure requirement within the design pressure limit
of 8868 kPag.

Flow velocity = 1.086 m/s

Reynolds number = 7820
Relative roughness = 0.000146
Friction factor = 0.0334
Friction pressure drop = 2745 kPa

Therefore, the pressure required at TO is 3000 kPag + 2745 kPa = 5745

kPag. Assuming that no pump station is installed at TO, the actual pressure required at TO may be around 5900 kPag when minor pressure losses at TO and
the delivery site are taken into account (refer to Addendum 3.3 for the discussion of minor pressure losses).
2. Next, determine the discharge pressure required at the intermediate pump station.
Distance from the pump station to TO = 136 km 100 km = 36 km
Pressure gradient of the main line = 75.2 kPa/km
Total pressure drop between the pump station and TO = 75.2 kPa/km
36km = 2707 kPa
Discharge pressure required at the intermediate station = 5900 kPag + 2707
kPa = 8607 kPag
Discharge pressure difference with and without the branch line = 8607
7870 = 737 kPa
Step 3. Determine the total pressure requirement when the branch line is shut down.
When the branch line is shut down, the maximum flow rate along the entire main
line reaches the design flow rate of 30,000 m3/d, and thus the main line can transport
the design flow rate, as demonstrated in the previous example.
Step 4. Evaluate this design
This pressure is lower than the design pressure, but 737 kPa higher than the discharge pressure required for the main line pressure. Since the discharge pressure difference is large, there are several options to correct this large pressure difference problem;
No modification to the existing pump units at the intermediate station,
Add a pump at the intermediate station, locate TO closer to the intermediate
station,
The installation of a pump station at TO on the branch line.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

108 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

Discussing these options further,
Alternative 3 has been selected in anticipation of future flow increases. Therefore, the pump units would have been chosen so as to accommodate such flow
increases and thus pump head. If the pump driver has extra power, the pump
units may not need to be modified by increasing the pump impeller size. (Refer
to the next chapter on pumps.)
A pump is added to the existing pumps at the intermediate station to provide
the extra pumping head. If the extra head required is large, this option may be
viable but the 737 kPa head is too small to warrant another pump.
If there is no restriction in locating the take-off point, it can be the best option
to locate TO at 126 km:
Distance from the pump station to TO = 126 km 100 km = 26 km
Pressure gradient of the main line = 75.2 kPa/km
Total pressure drop between the pump station and TO = 75.2 kPa/km
26km = 1955 kPa
Discharge pressure required at the intermediate station = 5900 kPag + 1955
kPa = 7855 kPag
Discharge pressure difference with and without the branch line = 7855
7870 = 15 kPa
Even if the branch line gets slightly longer than the original distance, the
discharge pressure difference is small enough so as not to require any changes
to the existing pump station.
It is a costly option to install a small pump station on the branch line, because
extra capital and operating costs are required.
It may not be a viable option to use an 18 pipe downstream of the side stream
delivery point, because the pressure drop for the pipe size is so high that the
maximum design pressure limit will be violated.
If it is known that the branch line will be added at the time of the main line
design, other considerations need to be included in order to optimize the system
design:

Location of the intermediate pump station

Location of the take-off point
Pressure requirements
Selection of pumping units

3.3.2.2 Side Stream Injection

Instead of flow take-off, liquid may be injected from branch lines into the main pipeline, entering the main pipeline at these intermediate locations, adding flow rate to
the main line flow downstream of the injection point. Since the flow is lower in the
upstream segment of the injection point, the frictional pressure drop there is lower. A
block valve is installed upstream of the injection point on the main line side and closed
when a full stream injection takes place or a new batch is created at the injection point
(refer to Chapter 5).
The modes of side stream injection can vary depending on the injection flow requirements or availability, nomination status, and pipeline operational status. For example, the branch line cannot be operational if a line break occurs downstream of the
injection point. Therefore, the following modes of side stream injection are possible,
and thus have to be included in the design:

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 109

Side stream injection through the branch line or at the injection point as originally designed,
The main line upstream of the injection point is shut down due to a problem in
the segment, so the operational segment of the pipeline is the branch line connected to the main line at the injection point,
The branch line is shut down if no volume is available to be injected into the
branch line or other operational problems occur.
Unlike the side stream delivery problem, other operational scenarios are available;
the same or different product injection and batching or blending operation for different
product. If the same product is injected, the product is mixed with that in the main line
and there is no operational issue. If a different product is injected into the main line, the
following operational issues need to be addressed:
Two products are blended if a partial injection takes place and the properties of the blended product will be different from the liquid in the main line
before they are blended. Then, a new batch is created at the injection point
and its volume grows until the side stream injection is finished. The movement of the new batch has to be tracked until it is fully delivered to the
shipper.
If the two products are not allowed to be blended, then the injection should be a
full stream injection and a new batch is created at the injection point. The main
line flow is stopped upstream of the injection point.
This type of design problem requires the following design considerations:
The injection pressure on the branch line should be higher than the main
line pressure at the injection point. The branch line design is similar to that
of the delivery to the third party pipeline discussed in the previous design
problem.
A pipe with a larger diameter can be used downstream of the injection point if
the side stream injection volume is large.
For partial side stream injection, the pumps upstream of the injection point
should be designed to accommodate the reduced flow. If the side stream
injection rate is high, the upstream flow rate can be lower than the minimum mainline flow. If the viscosity of the injection fluid is much higher
than the viscosity of the main line liquid, then the pumping units at the
downstreamof the injection point have to be selected to accommodate high
viscosity.
The side stream injection flow rate can be much lower than the minimum main line flow. The pump stations downstream of the injection point
have to be designed to meet the low flow rate during a full stream injection, particularly if the injection flow is lower than the minimum main line
flow.
A block valve is installed on the upstream side of the main line to take into
account full stream injections into the main line.
Injection of the same product: no batch is created. If the injection is a
partial injection, the upstream flow is reduced and thus the pump stations
have to be designed to accommodate the maximum and minimum flow
requirements.
Partial injection of a different product: blending of two different products occurs and a new blended batch has different density and viscosity. The property

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

110 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

differences have to be taken into account in the design of the pipeline system
including pumps and tanks.
Full stream injection of a different product: a new batch retains the product
properties of the injection fluid. For this operation, the effects of the injection
fluid have to be taken into account in the selection of the pump units in the
downstream segment of the injection point, particularly if its viscosity is much
higher than the viscosity of the main line liquid.
Example: Product Blending
A crude oil pipeline from CE to QU is 200 km long. It is constructed of 5LX-70
steel pipe with NPS = 20 and a 0.281 wall thickness. At the CE terminal, the crude
oil of 32API enters the pipeline at the design flow rate of 30,000 m3/d. A 60-km
branch line is planned to transport a crude oil of 35 API (specific gravity of 0.850)
from a tank to SI, a side stream injection point where the crude oil enters the main
line at the design flow rate of 7200 m3/d. SI is initially located at 78 km downstreamofCE, because it is closest to the flow lifting point, LP. Considered initially
are X52 grade pipe and the pipe size is NPS = 10 (actual pipe diameter = 10.75)
with a 0.219 wall thickness. Figure 3-11 shows the configuration of this pipeline
system.
The product density and viscosity of 32API gravity are 875.4 kg/m3 and 43.5cSt,
and the density and viscosity of 35API gravity are 857.6 kg/m3 and 21.0 cSt at the
operating temperature, respectively. Determine the pressure requirements of the pipeline system, using the following data:
Average operating temperature: 4C
Pipe roughness: 0.0457 mm
Delivery pressure at QU: 350 kPag
Assume that the design factor of 0.72 is applicable and that the elevation profile is
flat and flow is isothermal.
Solution:
It is assumed that the Alternative 3 design has been used as before and the intermediate
pump station has been operating.
Step 1. Determine the design pressure of the main and branch lines using the Barlow formula with the design factor of 0.72.
Pmain = 2 70,000 0.281 0.72/20 = 1416 psig = 9765 kPag
Pbranch = 2 52,000 0.25 0.72/10.75 = 1525 psig = 10,518 kPag

Figure 3-11. Side stream injection

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 111

Step 2. Calculate the required pressures at SI on the main line.
1. Calculate the pressure profile of the main line and the pressures at SI and LP for
the design flow rate.
From the base design example, the pressure gradient is 75.2 kPa/km, the
pressure at CE is 7870 kPag, and the suction pressure at the intermediate
pump station and the delivery pressure at QU are 350 kPag. Therefore, the
minimum pressure required at SI becomes:
Pressure at SI = 7870 75.2 78 = 2005 kPag
Minor losses in pressure are expected due to facilities like a pressure regulator and block valves installed on the branch line. Taking into account various
minor losses, the actual pressure required at SI is assumed to be approximately
2100 kPag. The discharge pressure at CE has to be increased by 95 kPa and
the pressure at LP should be determined to satisfy this pressure requirement.
2. Calculate the discharge pressure at the branch line lifting point, LP.
Flow velocity = 1.547 m/s
Reynolds number = 19,280
Relative roughness = 0.000174
Friction factor = 0.0265
Friction pressure drop = 6222 kPa
Therefore, the discharge pressure required at LP is 2100 kPag + 6222 kPa=
8322 kPag. This pressure requirement is lower than the design ressure of 10,518
kPag, and thus the side stream injection is appropriate.
3. Check if the suction pressure at the intermediate pump station is adequate.
Since the main line pressure is increased by 95 kPa, the suction pressure will
be higher than the minimum suction pressure by that amount. This pressure
increase is well within the tolerance. Therefore, this design including the injection location is an adequate solution.
Step 3. Calculate the pressure requirement of the segments upstream and downstream of the injection point when the two products are blended.
1. Calculate the pressure profile upstream of the injection point.
If the design flow rate is injected, the upstream flow rate is 32,000 7200 =
24,800 m3/d, for which the pressure gradient 53.9 kPa/km. Then the discharge
pressure at CE is 2100 + 53.9 78 = 6304 kPag.
2. Calculate the density and viscosity of the blended crude. These quantities are calculated for the design flow rates of the two lines; approximately 80% of the main
line and 20% of the injection flow rate. Actually, they change depending on the
percentages of blending of these two products. However, it is assumed here that
they remain constant to simplify calculations for other blending percentages.
Density of the blended liquid at 4C = 871.8 kg/m3
Viscosity at 4C using the ASTM method = 38.2 cSt
3. Calculate the suction and discharge pressures at the intermediate station.
Flow velocity = 1.815 m/s
Reynolds number = 23,450
Friction factor = 0.0250
Pressure gradient = 72.9 kPa/km
Suction pressure at the intermediate station = 2100 72.9 (100 78) =
496 kPag
Discharge pressure required at the intermediate station = 72.9 100 + 350=
7640 kPag

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

112 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

Since the pressure difference at the station is slightly reduced due to lower
density and viscosity, the pumping requirement is reduced and no modification
to the pump units is needed. If the density and viscosity are higher than those of
the main line liquid, the pumping requirement will have to be increased.
Step 4. Determine the total pressure requirement when the branch line is shut in.
When the branch line is shut down, the maximum flow rate along the entire main
line reaches the design flow rate of 30,000 m3/d, and thus the main line can transport
the design flow rate without any changes to the main line.
Step 5. Evaluate this design
If the injection point is located closer upstream of the main line injection point,
the injection pressure has to be high, requiring high discharge pressure at the branch
line injection point. If it is higher than the design pressure, potential options include an
increase in branch line pipe size, increase in the branch line discharge pressure, and/or
locating the injection point further to a downstream point along the main line.
If it is known that the branch line will be added at the time of the main line design,
other considerations need to be included in order to optimize the system design:

Location of the intermediate pump station

Location of the injection point
Pressure requirements
Selection of pumping units

3.3.2.3 Pipeline in Series

Pipelines may include different pipes connected in a series. Such situations occur when
different flow rates are transported due to intermediate take-off or injection or different
pressures are required along certain pipe segments. Depending on the purpose of arranging pipes in series, there are three types of series arrangement; different pipe sizes,
different pipe wall thickness, and different pipe grade. Except for the flow change due
to side stream injection or delivery, the same flow rate goes through the pipes connected in series but the flow velocity of each segment is different.
The pressure requirement in a series pipeline for the entire pipeline network is
determined by applying the appropriate flow equation for each segment and combining
all the segment pressure drops. The total pressure requirement can also be determined
by calculating the pressure required for each segment and then adding all the pressures
over the entire length.
3.3.2.3.1 Different Pipe Sizes Connected in Series Different pipe sizes are connected in series in two cases: significant change in flow or in elevation. The larger the
pipe diameter, the slower the velocity, the smaller the friction factor, and the lower the
friction pressure loss. A larger pipe is required as the throughput along a pipeline increases significantly, or vice versa. Therefore, a pipe is connected in series at a junction

Figure 3-12. Series pipes

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 113

where there is a large flow increase or decrease due to side-stream injection or delivery.
Figure 3-12 shows a pipeline with different lengths and diameters connected in series;
flow is taken off at the end of L1, requiring a smaller pipe size downstream of the side
stream delivery, and flow is added at the end of L2, requiring a larger pipe size downstream of the side stream injection point.
If the future throughput may not be known beforehand, it is not easy to determine
the different pipe sizes for each segment. Therefore, it may be more economical to use
the same size pipe over the entire length of the pipeline in case future flow requirements are not well known or show an increasing trend, even if there is intermediate
take-off or injection.
Where the pipeline is sloping down significantly, the pipe pressure can be increased due to elevation gain on the downstream side beyond the pressure loss due
to friction. As a result, it may be more advantageous to use a smaller pipe size (refer
to Section 5.1.4) to increase the frictional pressure drop so that overall pressure gain
can be reduced. For the opposite case, it may be safer to use a larger pipe size where
the elevation gains significantly, particularly if it is difficult to maintain the peak point
pressure above the minimum required level.
The pressure gradients change with pipe sizes, and the total pressure requirement
is the sum of the total pressure required in each pipeline segment, including the static
pressure due to elevation changes. One method of calculating the pressure drop in a
series pipe is to use the equivalent length technique, in which the first pipe is hydraulically equivalent to another pipe if the frictional pressure drop in the first pipe is the
same as that in the other pipe with a different length. Refer to hydraulic books that
detail this method.
At the connection point, either a reducer or expander may be used to provide
smoother transition from one size of pipe to another size. Minor pressure losses occur
at each junction, and a pigging station with a pig trap and launch facility has to be installed. Dual diameter cleaning pigs may be required on this type of pipeline.
3.3.2.3.2 Different Pipe Wall Thickness The main reason for connecting pipes
with different wall thickness is to reduce the pipe material and construction costs while
at the same time maintaining the same level of safety. Unless the pipe pressure increases due to significant elevation gain, the pressure tends to decrease continuously
from upstream to downstream and so does the pressure requirement. In other words,
the discharge pressure of an upstream pump station is much higher than the delivery
or suction pressure of the downstream pump station. The design pressure of a pipe is
proportional to the pipe wall thickness. Therefore, a pipe with thinner wall can be used
on the delivery or suction side, while a thicker pipe wall on the discharge side. Since
the pipe costs less for thinner wall pipe, the overall material and construction costs can
be reduced. In practice, different pipe wall thicknesses are used to compensate for different pressure requirements locally in the pipeline system.
When a pump station is shut down, the suction pressure increases greatly due to
potential surge and subsequently the surge pressure moves towards the upstream. In
addition, when the pipeline reaches a new steady state, the pressure level of the suction
side increases substantially in order to maintain the delivery pressure or the suction
pressure set point at the next pump station. Therefore, the pipe wall thickness on the
suction side of an intermediate pump station has to be high enough to withstand the
higher pressures that result from shut-down of the station.
Note it is not uncommon to reverse the flow direction on some pipelines for operational reasons and this capability can be denied if different wall thicknesses are used.
3.3.2.3.3 Different Pipe Grade A different pipe grade may be used instead of using a pipe with different wall thickness to satisfy the different design pressure requirement, not only for reducing the cost but also maintaining the same level of safely. In

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

114 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

other words, the high pressure sections are constructed of a high grade pipe, while the
lower pressure sections are constructed of somewhat lower grade steel. The same precaution as mentioned for different pipe wall thickness has to be exercised.
The design strategy of using different pipe wall thicknesses and/or pipe grade may
not be a good option if more pump stations are added, or flow is reversed, at a later
time. When the pipe flow increases, it is an option to add a pump station between two
existing pump stations. If the pressure rating on the discharge side of the new pump
station is low due to low pipe grade and/or thinner pipe wall, the new pump has to discharge at a low pressure unless the pipe sections with low pressure rating are replaced
with thicker wall pipe or higher grade pipe.
3.3.2.4 Pipelines in Parallel
Excessive pressure drop can occur in certain sections of a pipeline system where a bottleneck is formed. As a result, the throughput can be severely limited throughout the
pipeline. Pipes are arranged in parallel to reduce the excessive pressure drop in a certain section of the pipeline, and as a result to increase the throughput in the bottleneck
and relieve the throughput limitation in the pipeline system.
Two or more pipes are connected at the upstream and downstream points, so that
the flow splits among individual pipes at the upstream point and combines into a single
pipe at the downstream point as illustrated in Figure 3-13. Such a piping system is
referred to as parallel piping or looped piping system. The liquid flowing through AB
splits into Pipe 1 and Pipe 2, through which the liquid flows separately into point C.
The liquid flows recombine at point C and move to point D. An example is given in
the Addendum 3.3.
The flow rate splits in such a way that there is a common pressure across each
parallel pipe and the total flow is the sum of the flows across all parallel pipes at the
splitting point and at the combining point. The pipe sizes of the parallel piping sections
can be determined to meet the overall pressure requirements for the required throughput. The sizes of parallel pipes can be different. If the pipe sizes of the parallel pipes
are different, so is the flow velocity through each pipe.
If the pipe sizes are different between the parallel pipes, the flow rate through each
parallel pipe is initially unknown. Two principles are used to calculate the flow split
and pressure across the parallel pipes:
Conservation of mass or total flow at the junction
Common pressure at the end of or pressure loss across each parallel pipe.
Applying the flow conservation principle at B or C,

Q = Q1 + Q2

(3 32)

where Q represents the flow rate in the base conditions. Applying the common pressure
principle, we have

PB PC = P1 = P2

Figure 3-13. Parallel pipes

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 115

where P1 and P2 are pressure drops between B and C along the parallel pipes 1 and
2, respectively.
A pipeline is looped to increase throughput. Since the frictional pressure drop is
lower with a parallel pipe, so is the pumping requirement. However, if the pipe sizes
in parallel are different, caution must be exercised for batch pipeline design and operation. Since the flow velocity through each pipe is different, the batch front through
a smaller pipe arrives at the other end earlier than the other batch front, allowing the
early arriving batch to be blended with the leading batch. This blending increases the
mixing volume, thereby increasing slop. A batch controller is installed at the other end
of a parallel pipe in order to avoid this blending problem.

3.3.3 Severe Elevation Change Slack Flow

It is a challenge not only to construct a pipeline in mountainous areas with severe
elevation changes but also to operate the pipeline. The difficulties result because the
total pressure required to transport in such an area may depend more on the elevation
change than on the frictional pressure drop. When the pressure of the liquid drops
below vapor pressure, the liquid evaporates or boils forming vapor pockets inside the
pipe as shown in the diagram below. This condition is called slack flow and shown in
Figure 3-14. Note that there is a free surface between the liquid and vapor, at which
significant turbulent mixing can take place. Therefore, batch interface mixing can be
significant under a slack flow condition.
A vapor pocket is formed where the elevation drops severely, because the pressure downstream of a peak point must be increased due to the elevation gain but the
required pressure there is brought down by the low back pressure setting. Refer to the
elevation profile-pressure gradient diagram shown in Figure 3-16. With such severe
elevation drops, the slack flow condition can occur downstream of the high points in
the profile, if the back pressure is set low. The vapor pockets tend to stay on the downstream side of the high point, and the liquid flow is restricted due to the vapor pockets,
resulting in high pressure drop.
The slack flow problem may not occur at a high flow because the frictional pressure drop can overcome the pressure increase due to elevation gain. However, the
problem becomes more pronounced at a lower flow rate because the frictional pressure
drop at a low flow is so small that the downstream pressure becomes much higher than
the pressure at higher flow rate. These points are demonstrated in Figure 3-16 (refer to
Slack Flow Design Problem), showing the two pressure profiles.
A slack flow condition disrupts the pipe flow, reducing pipeline transmission efficiency and increasing batch interface mixing sizes. Damage to the interior of the pipe
can result if the vapor pocket suddenly collapses. Slack flow operation is difficult to
avoid for liquid pipelines, if the elevation drops severely and the back pressure has

Figure 3-14. Vapor pocket in a slack flow condition

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

116 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

been set low due to pressure limitations on equipment. Even though slack flow is not
desirable, pipeline systems transporting low vapor products such as crude oils can be
successfully operated in a slack flow condition. However, slack flow operations need to
be avoided for batch lines in order to limit the growth of batch interface mixing.
The design considerations for this type of design problem are:
A minimum pressure, which is sufficiently higher than the vapor pressure, has
to be maintained at the peak point to prevent vaporization.
Since a slack flow condition occurs more frequently at a low flow rate, a thorough hydraulic analysis has to be performed, particularly at low flow rates, in
order to fully understand the consequences of the slack flow on the design and
to determine the extra facility requirement and pressure rating on the equipment. Valves and flanges in the downstream segment of the peak point should
have a high pressure rating if the back pressure is not reduced using the following methods; installation of smaller pipe size and/or pressure-reducing station
(PRS).
Smaller pipe sizes can be used downstream of the peak point to increase the
frictional pressure drop and at the same time reduce the pipe pressure. Pipe and
construction costs can be reduced significantly. However, separate pig launchers and receivers have to be installed at either end of the pipe segment with
smaller pipe size because the pipe size is changed.
A PRS may be installed to operate the pipeline in a full flow condition by keeping the back pressure low and at the same time maintaining the downstream
pressure lower than the MAOP. As an additional benefit, the PRS can help to
keep the peak point pressure above the vapor pressure of the liquid. A PRS
is needed on batch pipelines to be operated in a full flow. Occasionally, two
PRSs may be installed if the elevations change several times and the drops are
extremely severe, or a combination of a smaller pipe size and PRS is adopted
in the design (refer to OCP pipeline in Section 5.1.4).
A typical PRS station is shown in Figure 3-15. It is noted from the figure that
the number of the pressure control valves is selected depending on the flow rate
and their positions are adjusted depending on the downstream pressure. A pig
receiver/launcher is not necessarily required if a pig can bypass the pressurereducing station.

Figure 3-15. Pressure-reducing station

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 117

Example: Slack Flow Line
A crude oil pipeline from CE to QU is 200 km long, crossing a mountainous area. The
table below shows an elevation profile. At the CE terminal, the crude oil of 32API
gravity enters the pipeline at the design flow rate of 30,000 m3/d. The minimum flow
rate is 9000 m3/d. Determine the pressure requirements of the pipeline system, using
the following data:

Average operating temperature: 4C

Minimum delivery pressure at QU: 350 kPag
Pipe grade: 5LX-70
Pipe size: NPS = 20 and a 0.281 wall thickness.
Density at the operating temperature: 875.4 kg/m3
Viscosity at the operating temperature: 43.5 cSt
Kilometer
post (km)
0
20
30
60
80
90

Elevation (m)

Kilometer
post (km)

Elevation (m)

30
55
45
30
70
100

110
130
150
160
180
200

100
300
770
425
150
130

Assume that the design factor of 0.72 is applicable and that the flow is isothermal.
Solution:
It is assumed that the elevation changes are gradual between two profile points, the peak
point pressure is kept at 350 kPag, and the minimum suction pressures are the same as
the delivery pressure. An intermediate pump station is located at KMP = 110km. Note
that the elevation changes in the first section between CE and the intermediate station
are mild, but the changes in the second section are significant.
Step 1. Calculate the pressures for the design flow rate of 30,000 m3/d at the above
profile points. The discharge pressure at CE is 9220 kPag so as to satisfy the minimum
suction pressure requirement, and the discharge pressure at the intermediate station is
9091 kPag so as to keep the peak point pressure at 350 kPag. The delivery pressure of
2067 kPag is obtained in order to keep the pipeline flow in a full flow condition. As a
result, the pressure profile is determined as shown in the table below.
KMP (km)
0
20
30
60
80
90

Elevation (m)

Pressure (kPag)

KMP (km)

Elevation (m)

Pressure (kPag)

30
55
45
30
70
100

9220
7502
6836
4708
2862
1854

110
130
150
160
180
200

100
300
770
425
150
130

350/9091
5876
350
2550
3400
2067

The discharge pressure at the intermediate station has to be sufficient to overcome

the static pressure loss due to the elevation increase and the friction pressure drop and
at the same time to keep the peak pressure higher than the vapor pressure of the liquid.
Note that the delivery pressure is higher than the minimum required delivery pressure
of 350 kPag, because the pressure downstream of the peak point is gained due to the
elevation drop while maintaining the required peak point pressure in a full flow condition. If the delivery pressure is set at 350 kPag, then the peak pressure drops below the
vapor pressure and vapor pockets are formed to meet the set point pressure.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

118 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

Step 2. Calculate the pressures for the minimum flow rate of 9000 m3/d at the
above profile points. The table below shows the pressure profile for the minimum flow
rate.
KMP (km)
0
20
30
60
80
90

Elevation (m)

Pressure (kPag)

KMP (km)

Elevation (m)

Pressure (kPag)

30
55
45
30
70
100

1975
1575
1567
1415
886
536

110
130
150
160
180
200

100
300
770
425
150
130

350/6456
4558
350
3209
5375
5360

The discharge pressure at CE is 1975 kPag so as to satisfy the minimum suction

pressure requirement, and the discharge pressure at the intermediate station is 6456
kPag so as to keep the peak point pressure at 350 kPag. The difference in the two discharge pressures are significantly large because the friction pressure drop is small at a
low flow rate but the intermediate station has to pump the liquids at a higher pressure
to compensate for the large elevation gain up to the peak point.
Therefore, the discharge pressure or head at the intermediate station must be high
to satisfy the pressure requirement at the peak point, which is, as shown in Figure 3-16,
the control point that dictates the discharge pressure and the downstream pressure too.
Note that the delivery pressure is much higher than the minimum required delivery
pressure of 350 kPag as well as the delivery pressure for the design flow rate. This is
caused by the low frictional pressure drop at the low flow rate while requiring the same
pressure gain due to the elevation drop.
Figure 3-16 graphically shows three pressure profiles; pressure profile for the design flow, pressure profile for the minimum flow, and pressure profile for the design
flow with the delivery pressure set at 350 kPag. The pressure gradients AB and CD represent a full flow condition for the maximum design flow, while AB and CD are the
profiles for another full flow condition, and for the minimum design flow, respectively.
The lines EF and EF show the pressure gradients where the delivery pressure at QU is
set at the minimum delivery pressure of 350 kPag. The liquid in the segment between

Figure 3-16. Slack flow conditions and pressure gradients

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 119

the peak point and E or E flows in a slack flow condition for the maximum or minimum flow rates. In other words, the pipeline segment is at zero gauge or atmospheric
pressure. The segment between EF or EF remains in full flow. If the back pressure is
kept constant at the minimum delivery pressure, the slack flow segment grows larger
as the flow rate is reduced. A PRS may be needed at, or preferably upstream of, QU to
bring these slack flow lines to the full flow condition.
Step 3. Two alternative designs are available; reduce the pipe size from 20 to a
smaller pipe size and/or install a pressure-reducing station (PRS). If the pipe size is
reduced to 14, the delivery pressure at the minimum flow rate drops to 3218 kPag. A
PRS can be installed downstream of the peak point for keeping the peak point pressure
high enough while reducing the downstream pressure.

3.3.4 Severe Weather Conditions

Severe weather conditions significantly influence the pipeline design and operation. A
severe weather condition can result in extremely hot or cold ambient temperature and
have a similar effect on soil temperatures. If a pipeline operates in hot weather conditions, the pipeline system can pick up ambient heat. On the other hand, if a pipeline
operates in an extremely cold area, the ground remains frozen and the fluid has to be
transported at lower than the freezing temperature in order to avoid melting the ice in
the surrounding soil.
3.3.4.1 Pipeline in a Hot Environment
As discussed in Section 3.1.3, the liquid temperature can increase mainly due to pump
inefficiency, heat gain through the frictional heating, as well as from the surrounding
soil. The temperature increase due to pump inefficiency will be high if the station spacing is short, because the next pump will add more heat before the liquid temperature
drops sufficiently to the ground level temperature. The temperature increase due to
frictional heating is higher as the flow rate increases. Normally, the temperature rise
due to conduction is largest. If the surrounding soil temperature is high due to prolonged high ambient temperature, the liquid in the pipeline absorbs the heat from the
soil, raising its temperature. The temperature increase will be greater for larger diameter pipelines, because the larger the pipe surface area the larger the heat conduction.
The temperature increase results in a decrease in liquid viscosity and density as well as
a decrease in vapor pressure. The decrease in viscosity and density will help to reduce
the friction loss. However, the decrease in vapor pressure has the following negative
consequences:
The pipeline pressure drops below the vapor pressure unless the pumps discharge at higher pressure.
Evaporation of the liquid in the pipeline and storage tanks would increase.
If the temperature increase is high, then cooling facilities need to be installed
along the pipeline in order to cool the temperature of the liquid. The best locations for
any cooling facilities would be near rivers or other water crossing areas where line
temperatures are low.
3.3.4.2 Pipeline in a Cold Environment
In the Arctic, the temperature in winter is very low, but can be hot in summer. However,
the ground is permanently frozen in most areas. This condition is called permafrost.
It is expensive to construct and operate a pipeline in a permafrost zone. The operating
temperature is one of the most critical design parameters in an Arctic pipeline. Therefore, the following considerations must be given when designing a pipeline for a cold
climate:

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

120 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

Selection of pipe low temperature steel pipes are required to control fracture.
The pipeline is buried or installed aboveground the line is installed aboveground in areas where the ground is permanently frozen, to avoid the need to
chill the oil.
If buried, the fluid is chilled. If the liquid temperature enters a pipeline close to
or greater than the freezing point of water, the flowing temperature increases,
as discussed above, and becomes higher than the freezing point. The liquid will
warm the pipeline and eventually the surrounding soil, which will be softened
around the pipeline. This may lead to thaw settlement resulting in the pipeline being bent and eventual damage to the pipeline. Therefore, for pipelines
in permafrost zones, the operating temperature must be lower than the freezing
temperature for the soil.
If a crude oil pipeline is shut down for a prolonged period, the crude oil may
congeal in the pipeline. Therefore, the relationship between the crude viscosity
and temperature has to be determined and temperature cooling behavior evaluated while the pipeline is shut down. If there is a possibility of congealing during shut-down, special facilities may be needed to restart the pipeline.
A chiller is installed at the liquid injection point in order to reduce the liquid temperature below the freezing temperature. The liquid temperature is reduced to at least
5C in consideration of temperature increases due to pump inefficiency and heat conduction in summer. In addition to a chiller, a wax removing facility may be required at
the injection location, because wax can build up on the pipe wall at low temperatures.
The requirement of the wax removing facility is determined after analyzing the viscous
behavior of the liquid with respect to the temperature.

3.3.5 Batch Pipeline Hydraulics Design

Since the densities and viscosities of the batching products can be different, the
pressure gradients are different and the pipeline capacities vary too. This is the result of
the dependence of capacity on pressure drops of the products in the pipeline, the order
of the products along the pipeline, and the position of the products with respect to pipeline and pump stations. In addition, the vapor pressure of each batch differs, requiring
a different minimum pressure along the pipeline. If the differences of the vapor pressures are significant, the pipeline may be operated with different minimum pressures in
order to reduce pumping costs. An example is an ethane-propane batch pipeline, where
the ethane batch requires a minimum pressure of around 4500 kPag but the propane
batch a minimum pressure of about 1700 kPag. Elevation changes, particularly severe
changes, have to be included in the analysis of pressure drops and batch movements
to analyze the minimum pressure requirements. The design considerations for a batch
pipeline design problem are as follows:
1. Select the fluid that produces the biggest friction loss, i.e., with the largest
viscosity and/or the highest density within the operating temperature range.
The minimum operating pressure is determined for the fluid with the highest
vapor pressure in order to maintain all the batching products in a full flow condition. Selecting these two products ensures that all pumping stations provide
adequate pressure and power to sustain the design flow rates and pressure for
all the batching products, while keeping the operating pressure within the maximum and minimum pressure limits. If future growth in the pipeline capacity
is expected, this design approach is preferred because it provides enough room
for future growth in the throughput capacity.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 121

2. Pressure drop averaged over the batching products can be used for a hydraulic
design if one of the following conditions is met:
Batch sizes are smaller than the volumes of the pipe section between two
pump stations,
The system load factor is low, or
All batching products have similar viscosities and densities.
This approach can result in a tight system design in terms of future system growth,
but can be acceptable if the pump station spacing is very long, the future growth in the
capacity is limited, or all future batching products have properties similar to the existing products.
Example: Batch Pipeline
A petroleum product pipeline from CE to QU is 200 km long and is 20 in nominal
diameter, with a 0.281 wall thickness. It is constructed of 5LX-65 electric resistance
welded steel pipe. At the injection point, the following three products enter the pipeline
at the design flow rate of 30,000 m3/d in a batch mode:

Product

Density at 4C
(kg/m3)

Viscosityat 4C
(cSt)

Batch size at 4C
(m3)

Vapor pressure
(kPa)

32API
35API
Condensate

875.4
857.6
705.0

43.5
21.0
0.7

20,000
15,000
25,000

10
15
95

It is assumed that these values are measured at the average operating temperature
of 4C. Design the batch pipeline including the delivery pressure.
Solution:
It is assumed that the base design is used; an intermediate pump station is located
100km downstream of CE and the design pressure is 9765 kPag.
Step 1. Determine the line fill volumes of the two sections of the pipeline. The
line fill volume is the volume of liquid contained in a segment of pipe, and is the pipe
volume in the ambient conditions, even though actual volume of liquid shrinks under
pressure. Addendum 3.4 discusses the effect of pressure and temperature on line fill
volume. A section is defined as the pipeline between two pump stations or between a
pump station and the delivery point. Therefore, the first section is defined from CE to
the intermediate station, where the second section starts, ending at QU.
Since the length of each section is the same, so is the line fill volume of each
section. Assuming that the pipe volume does not change in a pressurized condition, the line fill volume of each section becomes 19,150 m3.
Since this volume is smaller than the size of a 32API batch, the batch covers
thewhole section when fully lifted at CE or has passed the intermediate station.
Step 2. Select the product with the largest viscosity and the product with the highest vapor pressure among the three batch products.
The 32API batch has the highest viscosity among the three products, with a
viscosity of 43.5 cSt.
The condensate batch has the highest vapor pressure of 95 kPa or 6 kPag. Taking into account the minor pressure losses and transient effect, extra pressure of
400 kPa is added to the highest vapor pressure to get the minimum pressure of
400 kPag at the delivery and pump station.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

122 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

Figure 3-17. Line fill of batches

Step 3. Determine the batch sequence. Here, the batch sequence is given below
without describing the sequencing method which is detailed in Chapter 5.
The batch sequence for minimizing the interfacial mixing is 32API 35API
Condensate, and the same sequence is repeated in the next batch cycle.
When these batches are placed in the pipeline, the batch line fill profile can be
shown in Figure 3-17.
Step 4. Calculate the pressure profile using the 32API properties.
Discharge pressure at CE = 7920 and suction pressure at the intermediate station = 400 kPag.
Discharge pressure at the intermediate station = 7920 and delivery pressure at
QU = 400 kPag.
Step 5. Determine the average pressure profile.
Calculate the pressure drops P32, P35 and Pcon for 32API, 35API and condensate, respectively, using the common minimum pressure of 400 kPag.
Calculate the total required pressure averaged over the weight of each batch
size:
Pavg = (20,000 P32 + 15,000 P35 + 25,000 Pcon)/(20,000 + 15,000 + 25,000)
This approach is acceptable if the load factor is low.

3.3.6 High Vapor Pressure (HVP) Pipeline Design

HVP products are defined as the liquids whose vapor pressure at 38 C exceeds
110kPa. High vapor pressure (HVP) pipelines are characterized by low density, low
viscosity, and the requirement to operate the system at high pressure to maintain the
fluid in a single phase in the pipeline. HVP products are highly flammable and heavier
than air even when they evaporate into a gaseous form. They expand greatly as the
temperature increases, and their vapors are not easily visible. If the HVP liquids leak
out of a pipeline, the vapors may creep along the ground or gather in low places, and
can explode if they encounter an ignition source. Therefore, extra precautions are necessary to transport and store the products.
The temperature effects on HVP and dense phase fluids (refer to Chapter 2 for the
definition of dense phase) are so sensitive that the temperature behaviors in the pipeline
should be taken into consideration to determine the pressure profile accurately.
The density of light hydrocarbon such as ethane or propane changes significantly with temperature. The viscosity of lighter hydrocarbon liquids is small
and does not vary with temperature significantly. Therefore, hydraulic design

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 123

for such fluids is relatively independent of viscosity, because the Reynolds
number is so high that the fluids flow in or close to a fully turbulent flow regime. However, the design consideration should include the dependence of
their high vapor pressures and phase changes on the operating temperature.
This is the subject addressed in this section.
Viscous liquids such as heavy oil or waxy crude need to be heated or blended
with diluent to reduce the viscosity for pumping. Although the viscous heating
effect is inherent to all real fluid flow situations, its relative influence on heavy
and waxy crudes is very high. The temperature of the highly viscous fluids at
the entrance to the pipe can be significantly different from the temperature of
the medium surrounding the pipeline system. This subject will be discussed
inthe next section.
The hydraulic design for fluids such as NGLs or LPGs is relatively independent of
viscosity, but more dependent on consideration of vapor pressure. High vapor pressure
(HVP) pipelines are characterized by low density, low viscosity, and the requirement
to operate the system at high pressure to maintain the fluid as a single phase liquid in
the pipeline. Single phase should be maintained throughout the pipeline by keeping the
local pressure above the vapor pressure.
The governing design parameters for HVP pipelines are thus the vapor pressure
and maximum temperature.
The vapor pressure is directly related to fluid temperature in the pipeline.
The maximum vapor pressure occurs at maximum temperature in the pipeline.
The delivery points for HVP liquids require much higher minimum pressures over
the vapor pressure of the liquid. Because of the complex dependence of fluid properties
on pressure and temperature in the dense phase, pressure and temperature calculations
should be performed simultaneously to maintain high accuracy. The delivery point for
HVP liquids may be equipped with a pressurized sphere and thus require much higher
minimum pressures over the vapor pressure of the liquid.
HVP products can be economically transported in liquid phase, except ethane and
ethylene which may be transported in dense phase. In order to avoid vaporization of
the HVP liquids, HVP pipelines have to be operated at high pressure, above a minimum pressure greater than the vapor pressures throughout the pipeline. Normally, the
minimum pressure is determined by adding extra pressure to the vapor pressure. The
extra pressure takes into account the transient effect and elevation difference along the
pipeline as well as piping losses through manifold and other equipment at pump stations. If the liquids are delivered to a tank, the delivery pressure should be much higher
because the tank is often pressurized at a very high pressure level.
Dependent upon the pressure and temperature conditions, the fluid in a pipeline
can exist as a liquid, gas or a mixture of both (two-phase flow). The phase behavior
does not play a critical role in designing and operating heavier hydrocarbon liquid
pipelines, because their operating ranges are far away from the phase change zone.
However, the phase behavior of the HVP liquids has to be taken into account in pipeline design and operation, because their pipelines operate closer to the zone where a
phase change occurs.
Some examples of HVP products include pentane, butane, propane, ethane and
ethylene. Any pipeline transporting these products in liquid phase is called an HVP
pipeline. The vapor pressures of these products are listed in Table 3-6 (these vapor
pressures are obtained from GPSA Handbook [11], measured at 40C instead of
38C).

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

124 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

Table 3-6. High vapor pressure product parameters
Products

Vapor pressure
(kPa)

Thermal
expansion (/C)

Critical
pressure (kPa)

Critical
temperature (K)

I-Pentane
N-Pentane
I-Butane
N-Butane
Propane
Ethane

151
116
530
379
1370
6000 (*)

0.00160
0.00154
0.00216
0.00194
0.00280
0.015 (+)

3381
3370
3640
3798
4244
4872

460.4
469.7
407.8
425.1
369.8
305.3

Ethylene

9700 (*)

0.025 (+)

5040

282.3

(*) The vapor pressures and thermal expansions of these liquids are highly dependent on the pressure and temperature conditions. Therefore, a representative value does not have a definite meaning
for these products. These values are estimated by extrapolating measured values and are presented
for an illustrative purpose only.
(+) These values are estimated about 40C at 9000 kPa and presented for an illustrative purpose
only.

As shown in the table, the vapor pressures and thermal expansions of ethane and
ethylene are significantly higher than the other HVP liquids. Normally, these two products are transported in dense phase. For a hydrocarbon mixture, there is no clear line
dividing dense phase from the liquid phase or other single line dividing the dense
phase from the gas phase, but the dense phase lies between critical temperature and
cricondentherm if the pressure is above the cricondenbar. Phase change from denseto-liquid or vice versa is gradual. Ethane (C2H6), ethylene (C2H4), and carbon dioxide
(CO2), can be liquefied in pipelines at temperature and pressures even below the
critical point, and treated as liquids in transportation. Dense phase liquid is a highly
compressible liquid that shows properties of both liquid and gas; a density similar to
that of a liquid, but a viscosity similar to that of a gas. For liquid pipeline design and
operation, it is considered that the fluids are in dense phase if the pressure and temperature are around the critical pressure and critical temperature but above the vapor
pressure.
Because of the complex dependence of fluid properties on pressure and temperature in or near to the dense phase, pressure, and temperature should be determined
as accurate as possible and thus their calculations must be performed simultaneously
to achieve the desired accuracy. Reference [12] details the method of calculating
pressure and temperature in dense phase and identifies the following key design
parameters:
The critical point is not well defined nor are the properties near the critical
point. Therefore, one should try to avoid approaching the critical points too
closely.
Since most ethane or ethylene pipelines are operated in a fully turbulent flow
regime, the friction factor is independent of the Reynolds number and depends
only on the relative roughness of the pipe. Therefore, the accuracy of the pressure profile is sensitive to the values of the relative roughness. Sometimes,
other HVP products flow in a similar fully turbulent regime.
Both pressure and temperature profiles are relatively sensitive to the specified
value of the overall heat transfer coefficient, which in turn depends on soil
conductivity. The soil conductivity not only varies along the pipeline but also
changes frequently with moisture content.
The effect of the seasonal variation in the average soil temperature depends
on the difference between the fluid temperature and soil temperature. If the

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 125

d ifference is large, the effects on the pressure and temperature profiles will also
be large.
Higher flow rates result in greater friction losses and thus lower pressures,
causing lower density and higher velocities. At the higher flow rates, the temperature is high and this, together with the low pressures, results in a further
pressure drop with flow rate.
The elevation effects on the pressure drop in the uphill segments are different
from that in the downhill segment. In the uphill segment, the total pressure
gradient remains the same, because the decrease in the frictional pressure gradient is compensated by the increase in the static pressure increase rate due
to the elevation gain. In the downhill segment, however, the magnitude of the
hydrostatic term exceeds the magnitude of the friction term, resulting in less
pressure drop.
The magnitude of these effects depends on the rate of change of the fluid properties with pressure and temperature under the particular flowing conditions. In many
cases, accurate values for these design parameters are unknown. For example, soil
temperature will vary considerably from place to place, adding another uncertainty. In
such situations, one should perform calculations for a range of values to examine the
overall uncertainty in the calculated pressures and temperatures.
Normally, when designing a liquid pipeline, consideration is given to use the
maximum flow that is required at a specific time and a larger pipe size taking into account future volume increase. However, a HVP and particularly dense phase pipeline
is designed with the following criteria in mind, , high operating pressure, because of
uncertainties in defining the product properties in their operating ranges, and limited
accuracy in determining temperature profile:
Low flow velocity resulting in low pressure drops to operate at high pressures,
requiring a larger size pipe,
A high pressure required at the storage facility of the HVP products, also requiring high delivery pressure,
Overpressure problem if the pipeline is shut down for a prolonged period, requiring blowdown valves,
Frequent block valve spacing to reduce spillage and increase safety,
Installation of blowdown valves on either side of each block valve to relieve
an overpressure condition or deal with other emergency conditions. For added
safety purposes, the blowdown valves need to be automated and the isolated
segment has to be blown down as quickly as possible. Normally, a flaring system may be provided to flare the spillage.
Hydrate problems in HVP pipelines have been reported in the presence of free
water. Since it is impractical to control the pressure and temperature conditions for
forming hydrates, it may be simpler to reduce the contents of free water.
Example: Ethane Pipeline
A pipeline company plans to build an NPS 12 pipeline, transporting ethane, with wall
thickness of 0.219. The total length of the pipeline is 200 km and its elevation profile
is assumed to be flat. Initially, no intermediate pump station is planned. The yearly
throughput is expected to grow to 1,500,000 tons. The inlet pressure is planned to be
600 kPa less than the maximum design pressure and the maximum inlet temperature
is 30C. Refer to the pressure-enthalpy diagram shown in Figure 3-18. Determine the
minimum operating pressure, and pressure and temperature profiles using the following data:

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

126 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

Pipe grade: X56

Maximum inlet temperature: 30C
Ground temperature: 4C
Heat capacity: 4.76 kJ/kgC
Ethane viscosity: 0.14 cSt
Soil conductivity: 0.5 W/mC
Depth of cover: 1.2 m

Solution:
Refer to the pressure-enthalpy diagram, Figure 3-18, which shows the phase behavior of ethane. The Pressure-Enthalpy diagrams show pressure on the vertical axis and
enthalpy on the horizontal axis. The diagrams are used in locating pipeline operating
points in terms of pressure and temperature and for designing control valves. Pipe
flow is almost an isenthalpic process, so the diagram shows a graph of the enthalpy
during various pressures and physical states. The critical point is defined at the critical pressure and critical temperature (point C in the figure), where the liquid phase
and vapor phase meet, and either phase cannot be distinguished. The rectangular
box in the diagram shows the operating pressure range of an ethane pipeline for an
operating temperature range (assuming that the operating temperature ranges from
0C to 30C (solid lines in the figure) and the pressure from 4500 kPa to 10,000kPa).
Since the operating temperature range is lower than the critical temperature, the
ethane in this operating condition remains in liquid phase. For different operating
temperatures, the operating pressure range should be different to avoid vaporization.

Figure 3-18. Ethane pressure-enthalpy diagram [2]

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 127

As shown in broken lines, the ethane will be in dense phase and the minimum pressure has to be increased if the operating temperature is increased to 37C at the
maximum pressure.
It is assumed that the pipe design factor is 0.72 and pipe roughness is 0.0018 or
0.0457 mm.
Step 1. Determine the maximum design pressure for the X60 grade pipe.
Applying Barlow formula with the design factor of 0.72, design pressure =
2 S t/D F = 2 56,000 psig 0.219/12.75 0.72 = 1385 psig = 9550
kPag
Step 2. Determine the density of ethane at the maximum inlet conditions using the
Pressure-Enthalpy diagram.
The maximum operating inlet pressure is 9550 600 = 8950 kPag, and the inlet
temperature is 30C.
From the ethane pressure-enthalpy diagram in Figure 3-18, the ethane specific
volume at the inlet conditions is about 0.00267 m3/kg, or the density is about
375 kg/m3. However, the pressure and temperature change as the ethane flows
along the pipeline, and so does the density.
Step 3. Determine the vapor pressure and minimum delivery pressure for this pipeline design, using the ethane Pressure-Enthalpy diagram.
Since the pipeline flow is an almost isenthalpic process, the ethane vapor pressure is determined close to 4000 kPa by following down the isenthalpic line to
the phase envelope.
The minimum delivery pressure is obtained by adding to the vapor pressure a
safety pressure of 600 kPa: 4000 + 600 = 4600 kPa or about 4500 kPag. The
safety pressure includes minor pressure losses due to valves, pump station piping loss, meter station piping loss, and transient effects.
Step 4. Calculate the volume flow rate and velocity at the design flow rate.
At the inlet conditions, the density is 375 kg/m3 and thus the volume flow rate
1,500,000/(0.375 365 24) = 457 m3/hr. The flow velocity at the inlet conditions is about 1.65 m/s. Note that the local velocity varies somewhat because
of mass conservation.
If the number of yearly operating days is less than 365 days, then the actual
number of operating days should be used. Then, the flow velocity is larger
because the total amount of yearly shipment is divided by a smaller number of
days than 365.
Step 5. Determine the pressure, temperature and density profiles (Figure 3-19).
The viscosity effect on the friction factor may be negligibly small. However,
the density and heat capacity density change with temperature, and the density
is a complex function of pressure and temperature. The pressure and temperature profiles may not be determined reliably without their accurate behaviors
with respect to pressure and temperature.
It is time-consuming to manually calculate the pressure and temperature profiles in detail. Therefore, it is suggested to use a pipeline simulator for hydraulic

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

128 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

design work within the boundaries that have been established in the above
steps.
To calculate the pressure and temperature profiles accurately, the total pipeline
length is broken down into multiple short pipe lengths, say 5 km spacing for
elevation changes or 10 km spacing for flat elevation. The profiles are plotted
in the figure below.
Step 6. Analyze the ethane pipeline design.
1. The profiles show the following behaviors:
The delivery pressure is set at 4500 kPag and discharge pressure is calculated at 8550 kPag for the flow rate of 457 m3/hr. The pressure gradient is
almost linear at high operating pressures, where the ethane remains in dense
phase within the operating range. Since the mass rate has to be conserved,
the flow velocities at high temperatures, where the densities are lower, are
faster than those at low temperatures. Therefore, the frictional pressure drop
is somewhat higher in the upstream segment where the operating temperature is high than that in the downstream segment. This hydraulic behavior
is similar to that of a gas pipeline.
The temperature drops to the ground temperature about 80 km from the injection point. For a lower flow rate, the temperature drops faster and reaches
the ground temperature nearer to the injection point, because the heat conduction is faster at a low flow velocity.
Within this operating range, the density varies from 365 kg/m3 to 415 kg/m3,
about 11% change. Note that the density profile does not necessarily keep

Figure 3-19. Pressure and temperature profiles of an ethane pipeline.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 129

increasing as the pressure and temperature drop, because the density has a
non-linear relationship with pressure and temperature.
2. Determine the pipeline capacity.
The maximum throughput can be determined by setting the injection pressure at the maximum pressure.
The initially planned maximum pressure is 8950 kPag. At the injection pressure, the capacity is 480 m3/hr, which is about 5% higher than the design
flow rate. If the expected flow increases beyond this capacity, the maximum
operating pressure can be increased to the MAOP.
If the operating pressure is allowed to go up to the MAOP, the capacity increases to 517 m3/hr, which is about 13% higher than the design flow rate.
This capacity is equivalent to the annual rate of 17.0 million tons.
If the throughput needs to be increased beyond this limit, an intermediate
pump station has to be installed.
3. Calculate the pressure and temperature of the liquid. Since the pipeline pressure
of HVP products is very sensitive to temperature changes, it is necessary to
understand the temperature and pressure behaviors in order to avoid potential
overpressure problems. If the ambient pressure is high while the pipeline is shut
in for a prolonged period of time, an overpressure problem can occur because
the ethane temperature can increase in the pipeline. It is necessary to install
automatic blowdown valves to relieve the pipeline pressure.

3.3.7 Heavy Crude Pipeline Hydraulic Design

Heavy or waxy crudes do not flow easily in normal operating temperature ranges
mainly because of high viscosity and their high pour points. The pour points of these
viscous crudes are higher than normal operating temperature ranges. Even though
light and medium crudes are easy to pump above their pour points, they can exhibit
similar behaviors if the crude temperature drops below their pour points, as is possible in cold climates. Hydraulic design for heavy crudes or for hydrocarbon liquids
transported below their pour points is influenced largely by the effect of temperature
on viscosity and related friction losses. Therefore, the design aspects discussed in
this section are equally applicable to not only heavy crudes transportation in normal
operating temperature ranges but also light and medium crudes transportation in very
cold areas.
Transportation of such crudes through pipelines requires much higher flowing
temperature than their pour points or else a reduction of the pour points by blending
with diluent. Reference [15], in the five part series, describes various issues of pumping heavy crudes and lists the following methods of transporting high pour point
crudes:
Blending with a hydrocarbon diluent to keep the fluid behavior as Newtonian.
The diluents frequently used for bitumen transportation are natural gas condensate and synthetic crudes.
Heating the crude to a higher inlet temperature to allow it to reach the delivery
or intermediate station before cooling to below its pour point.
Combination of the above two methods
Mixing hot water with the highly viscous crude to form an emulsion, primarily
being used to transport bitumen to its processing plant
Processing the crude before pipelining to remove the wax and bring down the
pour point and viscosity

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

130 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

Injecting paraffin inhibitors, primarily being used in crude oil production systems to reduce pour point by preventing paraffin deposition and wax crystallization on the pipe wall
Heating both the crude and the pipeline by steam tracing or electrical heating,
which is only applicable to short pipelines due to the poor economics of applying it to long transmission lines
However, before deciding which method is selected, it is necessary to evaluate the
physical properties of crude, the temperature behavior in the pipeline, restarting after
shutdown, and facilities design.
3.3.7.1 Determine the Physical Properties under Pipeline Conditions
The critical design parameters for heavy oil pipelines are the viscosity and pour point,
because the viscosity is directly related to fluid temperature in the pipeline and the
non-Newtonian viscosity behaviors appear near the pour point. The following physical
properties are important for designing a heavy or waxy oil pipeline system including
pipeline hydraulics, pump station, and terminals [14]:

Density or specific gravity

Wax content
Shear stress vs. shear rate for non-Newtonian region
Yield stress for non-Newtonian region
Bulk modulus
Heat capacity

Heavy crude is characterized by high density, high viscosity and high pour point,
and may contain a significant amount of wax and/or sulphur. Heavy crude may exhibit
non-Newtonian viscosity behavior at normal operating temperature ranges because its
pour point can be higher. It is known that the apparent viscosities of non-Newtonian
liquids are sensitive not only to temperature changes but also to the shear rate and cooling rates. Laboratory tests should be performed at the pipeline operating conditions to
determine the crudes viscosity types and behaviors in terms of the shear stress vs. shear
rate and yield stress over the operating temperature ranges including the pour points. The
types include Newtonian, dilatant, Bingham plastic, pseudoplastic, and thixotropic (timedependent) fluid, because heavy crudes show different fluid characteristics.
Another potential engineering problem in dealing with heavy crudes, and sometimes with light and intermediate crudes, is the significant presence of wax. A waxy
crude may exhibit Bingham plastic characteristics after gelling, requiring a finite shear
stress to initiate flow. Heavy and/or waxy crudes start developing a yield stress near
their pour point, which may require additional pressure to restart flow. It is known that
wax does not deposit in turbulent flow at high temperatures, certain parts of a pipeline
may have wax deposits, and wax deposits could have an insulating effect.
Table 3-7. Viscosity, temperature, and pour point [14]
Product

Specific
Gravity

Temperature
(C)

Viscosity
(cSt)

Temperature
(C)

Bitumen
Residuals
Crude
High wax
Diesel
Jet fuel
Gasoline
NGL

1.02
0.96
0.84
0.81
0.84
0.78
0.73
0.50

65
65
20
50
-1
-1
-1
-1

50,000
1000
11
7.4
2.8
2.2
0.8
0.23

120
120
50
60
27
27
27
38

Viscosity Pour Point

(cSt)
(C)
330
46
4
3.3
1.4
1.3

0.2

55
32
13
35

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 131

The common characteristic of heavy and waxy crudes is their high pour point. Due
to non-Newtonian behavior near the pour point, more pressure is required to pump in
the non-Newtonian range. No problem may arise in pumping heavy crude below its
pour point, if the fluid is kept in motion. However, when the crude temperature is below its pour point, a few unique behaviors are observed:
If a crude pipeline being pumped below its pour point is shut down, the resulting gelled state will require substantially more pressure to put it into motion.
This additional restart pressure is substantially less than if a crude pipeline being pumped above its pour point is shut down and allowed to cool down.
Density and bulk modulus of heavy oil are very high compared to other types
of crude. The high bulk modulus can result in a large potential surge during pump
shut-down or valve closure. The frictional pressure drop of a heavy crude pipeline is
significantly high due to the high density and viscosity, and so is the surge pressure
due to high bulk modulus. As a result, a heavy oil pipeline tends to be operated at
low flow velocity for economic and safety reasons. As usual, heat capacity is used
for calculating temperature profile. The yield stress is a parameter used for determining the pumping requirement upon restart. Therefore, yield stresses should be
measured over the range of temperatures and shutdown times which are expected in
the pipeline.
For extra heavy oil transportation through a pipeline, blending with diluent is
most effective. It lowers the pour point of the blended heavy crude and viscosity significantly. The level of blending diluent changes with the temperature and viscosity
behaviors of the crudes, and the diluent requirement varies within individual pipeline systems to meet their specifications. When delivered to a third-party pipeline,
the pipeline specifications for density and viscosity are 940 kg/m3 (19API) and up
to 350 cSt in Alberta, Canada [16]. This provides for lower diluent requirements in
summer months than in the winter. Typically, summer requirements are about 20%
less than maximum requirements in mid-winter. Normally, crudes with an API gravity
of 18 or higher may not require any diluent, unless the operating temperature is very
low. Even in winter months, diluent requirements may be less than 5%. Crudes lower
than 18API need to be blended to a level that will provide for optimum pumpability
and protection from congealing in case of line shutdown [13]. Regardless of density
or viscosity before blending, all blended crudes should have a common consistency
so that all heavy crude moving in a common carrier pipeline has the same hydraulic
characteristics.
It is too expensive and risky to transport bitumen a long distance by means of heating only. Therefore, in long lines it is necessary to blend it with a diluent. The key issue
is the availability of diluent at the injection location. If it is not available, then it has to
be shipped from other sources through another pipeline or by train. If the availability
of diluent is limited, the diluent can be separated from the blended bitumen after it is
delivered. It requires a separation plant at the delivery location and then the separated
diluent is shipped back to the injection location.
3.3.7.2 Determine the Pressure and Temperature throughout the Pipeline for
the Anticipated Flow Rates
Section 3.1.3 discusses the equations, surrounding environment, and procedures for
calculating pressure and temperature profiles. The surrounding environments can
vary significantly, resulting in different overall heat transfer coefficients. As noted
previously, the viscosity of extra heavy crude such as bitumen is very sensitive to
temperature change. Therefore, an accurate temperature calculation is necessary. In
order to improve calculation accuracy, temperature and pressure profile calculations

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

132 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

are performed by dividing the entire pipeline into many short pipe segments and
analyzing each segment separately.
As discussed earlier, hydraulic design for heavy crudes is influenced largely by the
effect of temperature on viscosity and related friction losses. If the crudes are heated,
a thermal analysis is required to predict the performance of the system over its design
temperature range and subsequently to determine the pumping requirements in a pipeline system. If the injection temperature is much higher than the ground temperature,
the frictional pressure drop accelerates as a result of cooling. As the fluid is cooled,
both the density and the viscosity increase, and the frictional pressure drop increases.
Due to the high viscosity of heavy crudes, the frictional pressure drop per unit
distance is very high even for low flow velocity, and thus the friction heating and
fluid temperature increase. Normally, liquid pipelines operate in the turbulent flow
regime and the boundary layer is thin. Therefore, the thermal resistance due to the
boundary layer that builds up on the inside of the pipe wall is negligibly small.
The contribution of the thermal resistance for extra heavy crudes to the overall heat
transfer coefficient turns out to be small, even if the crudes flow in the laminar flow
regime. However, the actual temperature drop reduces slightly as a result of the
added thermal resistance.
Pipelines for heavy crudes may be insulated to reduce heat loss, if the economics
is justified. Insulation thickness is important as the design and operation of a hot oil
pipeline depend on the amount of heat lost by the heavy crudes. If the temperature difference between the pipeline and the ground is greater, the insulation thickness can be
increased to a certain extent. However, it should not be thicker than the economic and
physical optimum insulation thickness. Insulation applied to large diameter pipelines
to maintain temperature at low flow rates and low ambient temperature may cause
overheating of the line for high flow rates at high ambient temperatures. For any given
insulation thickness, the heat loss is greater if it takes longer for the crude to travel
between the initial pumping station and terminal or between reheat stations. Therefore,
the velocity and viscosity of the oil determine the distance between stations and the
number of pump stations and/or reheat stations.
Design considerations for heavy crude pipelines with thermal effects should include the following:
The temperature behaviors of the environment and its effect on the physical
properties of the fluid over the range of operations.
The temperature effects on shutdown and restart.
In addition to hydraulic design, the operating temperature range affects the mechanical design and design for operations including shutdown and restart. The following types of hydraulic design and operation problems arise for heavy crude pipelines:

Pressure and temperature profile calculations,

Pipe line sizing,
Maximum throughput determination,
Pump and heater station spacing,
Heat retention for a certain period, thereby determining insulation thickness
and other facility requirements such as extra pumps.

3.3.7.3 Review the Restart after Shutdown

As pointed out above, heavy crudes show unique behaviors near or below their pour
points; the gelled state will require more pressure to put it into motion due to the

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 133

high yield stress below the pour point. Therefore, it is necessary to determine crude
temperature throughout a pipeline whose temperature may cool down during the expected shutdown period. During shutdown periods, fluid in the pipeline cools without
the heat of friction until flow resumes or the pipeline temperature reaches the ground
temperature.
During cooling, the temperature at a certain location may be calculated by analyzing the rate of heat loss of the crude in the pipeline:

T(t) = Tg + [T(0) Tg] exp{ (4UDTt)/[rCpd2)]}

(3 33)

where
T(t) = temperature at time t, (C)
T(0) = temperature at the time when the pipeline is shut down (C)
Tg = ground temperature (C)
DT = pipe outside diameter including insulation thickness for insulated pipe
t = time from start of static cooling, in second
d = inside pipe diameter, m
r = liquid density, kg/m3
Cp = heat capacity of the liquid averaged at the inlet and discharge temperatures,
kJ/kgC
If the heat conduction through the boundary layer and pipe wall is excluded
from the overall heat transfer coefficient, the calculated temperature would be lower
than the actual temperature, requiring somewhat lower restart pressure. If possible,
pipeline start-up or restart can be scheduled during periods of warmest ambient temperatures in order to avoid the difficult problems that may be encountered during
start-up.
3.3.7.4 Design Facilities
The effect of the yield stress of heavy crudes is non-trivial below pour point. When a
pipeline is shut in and thus the heavy crude cools down below the pour point, it requires
an extra pressure to put the crude in motion. This extra pressure requirement has to be
provided by a pump to initiate flow. The pressure required to initiate flow is sum of the
pressure differentials required to break the gel in each section of the pipeline. Since
yield strength is sensitive to temperature, the required pressure has to be determined on
each segment to reduce potential calculation error.
When starting up the pipeline after shut-in, the flow rate should be very low
to push the gelled crude gently. It is essential to establish the minimum flow rate
needed to be maintained during initial start-up, and it may be necessary to include
redundant provisions for emergency and planned shutdowns. In selecting a mainline pump, the maximum operating point should be satisfied as usual. If the minimum flow cannot be met by the mainline pump during an initiating period after
shut-in, special startup/restart pumps with the capability of high pressure and low
flow should be considered. Note that the performance of a centrifugal pump deteriorates for pumping high viscosity fluids, thus requiring a rerate of the pump perfor
mance. Refer to the next chapter for pump performance rerating for high viscosity
conditions.
Systems to consider would include standby pumps for displacing the crude oil
in the pipeline with water, and adding pour point depressant injection facilities. If
bitumen is blended with a diluent, a diluent blending and storage facility is required
at the lifting point. If the crude is heated for pumping, heaters have to be installed not
only at the initiating station but also intermediate pump stations. Depending on the

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

134 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

t emperature requirements along the pipeline, the number of heaters required at each
station can vary. The heater duty can be calculated from:

qh = rQCp (Td Ti)/hh

(3 34)

where

qh = heater duty required to heat the liquid to the discharge temperature, kJ/hr

r = liquid density, kg/m3

Q = liquid flow rate, m3/hr
Cp = heat capacity of the liquid averaged at the inlet and discharge temperatures,
kJ/kg C
Td = discharge temperature of the heater, C
Ti = inlet temperature of the heater, C
hh = efficiency of the heater
Example: Extra Heavy Crude Oil Pipeline
A pipeline company has decided to build a pipeline transporting bitumen to a common
carrier pipeline system. The company receives bitumen at 65C at the initiating station
with an expected maximum throughput of 151,000 bbls/day or 24,000 m3/day. The
delivery point of the common carrier pipeline is located 200 km from the lifting point.
The pipeline system crosses an unpopulated area, and the elevation profile is almost
flat. An X56 grade pipe with 28 pipe diameter and 0.350 wall thickness is considered, and the pipeline has to be buried 1.2 m below the ground surface. The maximum
shutdown period expected for scheduled maintenance or emergency repair is estimated
at 120 hours.
Soil temperatures are 4C in summer and 4C in winter and the soil conduc
tivity is 1.0 W/mC. Assume that the soil temperatures are uniform throughout
the pipeline. The bitumen gravity is 8.6API and its viscosities are 10,000 cSt and
100cSt at 45C and 115C. A series of laboratory tests has found that its pour point is
50C, below which the yield stress grows significantly exhibiting a Bingham plastic
behavior. The common carrier pipeline requires that the viscosity of the delivered
product must be maintained at 350 cSt or lower. The heat capacity of the blended
bitumen ranges from 2.01 kJ/kgC to 2.40 kJ/kgC depending on the temperature
and density.
Considering the pipeline length, high viscosity and pour point, and low operating
temperature, the company has decided to blend bitumen with condensate as a diluent
in order to facilitate easy transportation of bitumen. The bitumen and condensate are
blended at the production area before the blended bitumen (dilbit) is lifted. The API
gravity of the condensate is 76 and the viscosities of the condensate at 5C and 45C
are 0.7 cSt and 0.4 cSt. Determine the following:

diluent requirements in both summer and winter conditions,

pressure and temperature requirements,
a heater requirement,
temperature profile after the maximum shutdown period
pipe insulation requirement

Solution:
It is assumed that the pour point is low enough to transport the blended bitumen as a
Newtonian fluid and that the possible contents of diluent for winter condition are 45%,
40% and 35% and the contents for summer condition are 30%, 25% and 20%.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 135

Step 1. Calculate the base densities of the blended bitumen, and viscosities of and
volume requirements for the blended bitumen at the minimum temperatures.
The minimum temperature is the temperature required to satisfy the maximum viscosity requirement of 350 cSt, and the flow requirement is the total volume or daily flow rate
of the blended product. Assuming the daily bitumen production remains the same, the flow
requirement for thewinter blended bitumen is larger because the diluent requirement is
higher in winter.
Density of the bitumen at 15C and atmospheric pressure: r = 1000 141.5/
(131.5 + 8.6) = 1010 kg/m3
Density of the condensate at 15C and atmospheric pressure: r = 1000 141.5/
(131.5 + 76) = 682 kg/m3
Contents of
Diluent (%)

Density
(kg/m3)

Thermal
Expansion (C)

Minimum
Temperature (C)

862.4

8.27 104

321

34,800

40

878.8

7.95 10

15

330

33,600

35

895.2

7.66 104

23

342

32,400

30

911.6

7.37 104

32

330

31,200

25
20

928.0
944.4

7.12 104

41
50

324
339

30,000
28,800

45

6.89 104

Viscosity Flow Requirement

(cSt)
(m3/day)

Step 2. Calculate the pressure and temperature profiles. The maximum design
pressure for the selected pipe is 6950 kPag, so the blended bitumen is discharged at
6900 kPag.
The table below summarizes the injection and delivery temperatures and the delivery pressures for different amounts of diluent. The injection delivery temperatures are
determined in such a way that the viscosity at the delivery point is kept below 350 cSt.
The table shows the temperatures and their corresponding viscosities after the pipeline
is shut down for 120 hours.
Contents of
Diluent (%)
45
40
35
30
25
20

Injection
Temperature
(C)

Delivery
Temperature
(C)

Delivery
Pressure
(kPag)

Temperature
after 120 hours
(C)

Viscosity after
120 hours
(cSt)

15
34
52
62
83
104

7
15
23
32
41
50

532
1458
2019
2236
2774
3203

0.2
3.3
6.4
14.7
18.2
21.8

750
1264
2138
1917
2958
4359

The required injection temperature decreases as the amount of the diluent increases. As expected, the pressure requirements for the winter condition are higher
than those for the summer condition as a result of the operating temperatures in summer condition being much higher than in winter and the flow rates are lower for the
summer condition. Also, as the amount of diluent gets smaller, a heating facility has to
be installed to raise the injection temperature.
If this bitumen starts showing its non-Newtonian behavior about 2000 cSt, the
yield stress has to be measured in order to assess the requirement for extra pumping
facilities to dislodge the blended bitumen that was congealed during the 120 hours of
the shut-in period.
Step 3. Repeat the same calculations for the case where the pipeline is insulated
with 2 polyurethane insulation material. The insulation conductivity is 0.035W/mC.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

136 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

Contents
of Diluent
(%)
45
40
35
30
25
20

Injection
Temperature
(C)

Delivery
Temperature
(C)

Delivery
Pressure
(kPag)

Temperature
after 120 hours
(C)

Viscosity after
120 hours
(cSt)

10
18
29
38
51
63

9.7
16
23
31
40
48

390
381
841
1139
1706
2109

5.6
9.8
15.4
23.8
30.4
36.5

410
610
774
742
853
997

Compared to the results of the un-insulated case, the insulated pipeline has the
following advantages over the un-insulated pipeline:
No extra pumping facilities are required even after shutting down for 120
hours,
There is less need for a heater because the required injection temperature is low,
or the heater duty is lower than the duty for the un-insulated pipeline even if a
heater is installed, assuming that the same amount of the diluent is mixed,
The diluent requirement is much smaller than the requirement for the uninsulated pipeline,
Restarting after the shut-down is much easier due to low viscosity.
Step 4. Finalize the pipeline system design
The 28 pipe with wall thickness of 0.35 and pipe grade X56 satisfies the pressure requirements for both winter and summer conditions.
The insulated pipeline can save both capital and operating costs by reducing
the requirements for extra facilities such as a heater and an extra pump to deal
with the congealed non-Newtonian crude.
The selection of the diluent requirement vs. heater installation is based on the
cost comparison of the diluent costs against the heater costs. If the pipeline is
insulated, 35% of diluent and 65% of bitumen blending can be sufficient in
winter. If the temperature of the blended bitumen is higher than 63C, a heater
is not required and thus 20% of diluent may satisfy the summer transportation
requirement.

3.4 LOCATING PUMP STATIONS

The initiating point of the pipeline system, into which petroleum products are
lifted,must have a pump station. Also, a long pipeline may require multiple pump
stations along the mainline. The pumping requirements should be considered in
terms of the number and locations of the stations. The number of pump stations
is dictated by the installation and operating costs as well as the flow velocity and
controllability of the pipeline system and pump station. If the number of stations
increases, the costs and flow velocity increase while making the system control
difficult due to large surge pressure and its fast response. Refer to Section 5.1.3 for
controlling surge.
The key criteria of initially locating mainline pump stations are that the MAOP
should not be violated downstream of each pump station and each station has the same
differential pressure or head. Here, the differential pressure includes all minor pressure
losses due to station piping, bends, fittings, and various valves including control valve.
The second criterion offers the following advantages:

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 137

The total energy or power consumption is reduced by adding the same amount
of energy to the liquid at each pump station.
The pump maintenance and spare part inventory costs can be minimized, because the equipment can be identical.
However, these advantages should be compared against potential extra costs to
design a pipeline system as such. For example, the power line may be too far from an
optimum location to satisfy the above criterion.
This criterion is applicable to the design of all new pipelines in locating pump stations. However, the procedure of locating stations can be different for different terrains
or pipeline configuration:

Relatively flat terrain,

Complicated terrain in terms of elevation profile,
Simple pipeline system with one injection and one delivery,
Complex pipeline system with multiple injection and delivery points.

For a simple pipeline system with relatively flat terrain, the criteria for locating
stations results in almost equal station spacing along the pipeline, and the number of
pump stations can be determined by dividing the total required pressure by the difference between the MAOP and the minimum pressure;
No. of stations = Total required pressure / (MAOP minimum suction pressure)
The procedure of locating pump stations is to start from the delivery pressure,
drawing the pressure gradient upstream to the intersection of the maximum design
pressure, which is superimposed on the elevation profile. If the discharge pressure of
the initiating station is smaller than the design pressure, then reduce the design pressure
and move the initial locations to further downstream locations. The same differential
pressure can be calculated by dividing the total pressure requirements by the number
of pump stations.
Example 1: Simple Pump Station Location
Refer to the design example described in Section 3.3.1. The total required pressure
is 15,389 kPag, maximum design pressure 9765 kPag, and minimum delivery pressure 350 kPag. It is assumed that the minimum suction pressure is 350 kPag. Since
the elevation profile is flat, the number of pump stations is obtained from the above
formula:

15,389 / (9765 350) = 1.6

Therefore, the total number of pump stations required is 2; one at the initiating station and the other at an intermediate location. Applying the station location criteria, the
intermediate station is located at the mid-point of the pipeline as shown in Figure 3-20.
For a simple pipeline system with severe elevation changes, the station locations
can be determined by applying these criteria through trial and error on a graph. The
procedure of locating intermediate pump stations is as follows:
Step 1. Using the maximum design pressure as the discharge pressure at the initiating station, the first intermediate station is found at a location where the pressure
reaches the minimum suction pressure by drawing the pressure gradient on the elevation profile. In practice, a pressure allowance of 200 kPa to 300 kPa at the intake of the
pump station is required to account for the losses due to station piping, valves, fittings,
and other equipments.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

138 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

Pressure
(kPag)
Design Pressure = 9,765 kPag

PD = 7,780

0
Booster Pump

PS = 350

100 km
Main Line Pump

Distance

200 km

Figure 3-20. Locating intermediate pump stations for flat elevation

Step 2. Progressing downstream from the maximum design pressure at the intermediate station, the next intermediate station is located in the same way as above.
Repeat these steps until the minimum suction pressure of the last section is greater than
or equal to the delivery pressure.
Step 3. If the suction pressure is much greater than the delivery pressure, reduce
the discharge pressure equally at each pump station and then repeat the second step to
move the initial locations to upstream locations.
Step 4. If the discharge pressure has to be reduced significantly, the maximum design
pressure can be lowered by selecting lower grade pipe or thinner pipe wall thickness.
Figure 3-21 shows that the total pressure requirement is greater than the design
pressure. This pressure requirement can be met by installing an intermediate pump station or choosing a thicker pipe in the upstream segment where the required pressure is
Pressure
( kPag)

Head
(m)
Design Pressure

8,600

1,000

PD

PD

4,300

500
Ps

PB

PS
0

Distance (km)

200

Figure 3-21. Locating intermediate pump stations

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 139

violated. Assuming an intermediate pump station is installed, it can be located in such
a way that the differential pressure, PD PB, at the initiating station is the same as the
differential pressure, PD PS, of the intermediate station. In this case, the station location is shifted toward the high elevation side. The shift depends on the elevation profile
and site conditions.
Example 2: Pump Station Location in Changing Elevation Profile
A pipeline company plans to build and operate a crude oil pipeline, delivering to a tank
farm. The pressure rating of the tank equipment is designed at 700 kPag. The average
flow rate is 1175 m3/hr and the pipeline system is expected to operate at least 345 days
a year. The average operating temperature is 4C. The density and the viscosity of the
crude at the operating temperature is 870 kg/m3 and 40 cSt, respectively. The vapor
pressure is 80 kPa or 21 kPag, and a slack flow condition has to be avoided. Analyze
the pressure profile for the minimum design flow rate of 500 m3/hr. Assume the suction
pressure at each pump station is 350 kPag and the pump pressure differential should be
less than 8000 kPa. The pipe specifications are as follows:

Pipe sizes NPS = 20

Wall thickness 0.281
Pipe roughness 0.0018
Pipe grade API X60
Design factor 0.72

The pipeline length is 350 km and the elevation profile is given below.
KMP
Elevation

0
10 m

50
250 m

150
250 m

190
250 m

230
250 m

290
310 m

320
460 m

350
10 m

Solution:
Step 1. Determine the design flow rate and the maximum design pressure.
Since the number of yearly operating days is 345 days, the load factor is
345/365 = 94.5%, and thus the design flow rate is 1175/0.945 = 1243 m3/hr, or
rounding up to 1250 m3/hr.
The design pressure is obtained by applying the Barlow formula and the design
factor for the X60 pipe grade; 8370 kPag.
Step 2. Calculate the pressure gradient.

Flow velocity = 1.82 m/s

Reynolds number = 1.82 0.494/0.00004 = 22,500
Relative roughness = 0.00009
Friction factor = 0.0377
Pressure gradient = 0.0377 870 1.822/(2 0.494) = 110.0 Pa/m =
110.0 kPa/km

Step 3. Determine initial station locations and calculate the pressures at the
locations.
Assuming that the discharge pressure at the initiating station is 8230 kPag, the
first pump station will be located approximately 53 km downstream with a suction pressure of 354 kPag.
Since the elevation difference is zero for about 180 km downstream of the
first intermediate station, the next pump station can be located with a similar

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

140 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

differential pressure to the initiating station; 7920/110 = 72 km. Therefore, the
first and second pump stations are located at 125 km and 197 km, where the
discharge pressures are 8274 kPag and 8270 kPag, respectively.
The fourth station is located at 269 km if the elevation were 250 m. Since it
is higher than 250 m at 269 km point, the spacing will be shorter than 72 km
and the elevation is determined by prorating elevations between two adjacent
known locations. At the 266 km post, the elevation is prorated at 286 m and the
suction pressure becomes 374 kPag.
If the discharge pressure of the fourth station is set at 8270 kPag, then the pressure at KMP = 290 is 5596 kPag and the pressure at KMP = 320 is 1021 kPag.
Note that the KMP = 320 is the peak point in this pipeline system. Since the peak
point pressure is higher than the vapor pressure by 1042 kPa (1021(21)=
1042), theoretically the discharge pressure can be reduced by about 1000 kPa.
However, considering the transient effect on the pressure, an extra allowance
of about 300 kPa has to be added to the vapor pressure.
Step 4. Since the discharge pressure of the last station can be reduced by (1000
300) kPa, that is to 700 kPa, the initial station locations can be adjusted.
First, locate the first intermediate station at 52 km, where the suction pressure is
set at 350 kPag, the discharge pressure and differential pressure at the initiating
station are 8116 kPag and 7766 kPa.
Using the similar differential pressure, the station spacing of the next two stations is 70.5 km and the second and third station locations are 122.5 km and
193km, respectively. Then, setting the suction pressures of the second andthird
stations at 350 kPag, the discharge pressures of the first and second stations are
8105 kPag and the differential pressures are 7755 Pa. This differential pressure
is very close to the differential pressure at the initiating station.
The fourth station was located initially at 266 km. By taking into account the
elevation difference and locations due to the location shift, the new location is
determined at KMP = 263 km. Then, the discharge pressure at the third station
is 8160 kPag. If the discharge pressure at the fourth station is set at 8079 kPag,
the peak point pressure is calculated at 300 kPag.
If a surge analysis shows that the peak point pressure is too low, the pump stations need to be moved slightly toward downstream locations.
Step 5. Determine the delivery pressure when the station locations are finalized.
The hydraulic pressure gain from the peak point to the delivery point is 0.87
9.8 450 = 3837 kPa, but the friction pressure loss is 3300 kPa. Thus the delivery pressure is 887 kPag, which is greater than the tank equipment pressure
rating. Therefore, a pressure control valve (PCV) is needed upstream of the
tank farm.
Since the MAOP is much greater than the delivery pressure, a pressure-reducing
station (PRS) is not required as long as the peak point pressure is maintained
above the vapor pressure.
Step 6. Analyze the pressure profile for the minimum design flow rate.
Flow velocity = 0.727 m/s
Reynolds number = 0.727 0.494/0.00004 = 8980
Relative roughness = 0.00009

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 141

Head
(m)

Pressure
(kPag)
8,600
8,116

8.105

PD

8,105

8,160

1,000

8,079

4,300

300
350

350

500

350

350

887kPag
0

50

100

150

200

250

300

350

PS

Distance (km)

Figure 3-22. Station locations with elevation and pressure profiles

Friction factor = 0.0505

Pressure gradient = 0.0505 870 0.7272/(2 0.494) = 23.5 Pa/m = 23.5 kPa/
km
Since the pressure gradient is low, the first and second pump stations can be
bypassed. Assuming the suction pressure is set at 350 kPag at the third pump
station, the discharge pressure required at the initiating station is 6952 kPag. If
the discharge pressure at the third station is 4844 kPag, then the pressure at the
peak is 350 kPag and the delivery pressure becomes 3482 kPag.
Since this pipe pressure is much higher than the tank equipment pressure rating, the PCV must have the capacity to reduce pressure by 3482 700 kPa =
2782 kPa.
Figure 3-22 shows the pump station locations with elevation and pressure profiles
for the design and minimum flows. Note that the pressures at the delivery gate for
low flows are higher than those for high flows in order to keep the minimum pressure
required at the peak point.
In general, the same criteria are applied to more complex pipeline systems for locating intermediate pump stations by a trial and error method. Through this hydraulic
analysis, the approximate pump station locations are determined that would meet the
design and operating parameters. However, the same differential pressure at all pump
stations cannot always be achieved.
Example 3: Pump Station Location with a Branch Line
The pipeline from CE to QU is 214 km long and is 20 in nominal diameter, with a
0.281 wall thickness. It is constructed of API X-60 grade steel. At CE, diesel enters
the pipeline at the design flow rate of 1800 m3/hr. The booster pumps at CE discharge
into the main line pump at 350 kPag, and the minimum delivery pressure required at
QU is 350 kPag.
The diesel is taken off at TO, 176 km downstream of CE, where up to 600 m3/
hr is stripped off the pipeline, and the rest is delivered to the final destination, QU.
Occasionally, the full flow has to be delivered to QU. At TO, a 50-km branch line is
connected to a third party pipeline, which requires the delivery pressure of 3000 kPag.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

142 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

The pipeline is constructed with X52 grade pipe, and the pipe diameter is 16 with a
0.25 wall thickness.
Locate the pump stations along the main pipeline, using the following data:

Average operating temperature: 15C

Density: 850.0 kg/m3 at 15C at the operating temperature
Viscosities at 15C: 10.0 cSt
Pipe roughness: 0.0018
Delivery pressure at QU: 350 kPag

Assume that the design factor of 0.72 is applicable and that the elevation profile is
flat and flow is isothermal.
Solution:
Step 1. Calculate the design pressure (MAOP) of the main and branch lines.
MAOP of the main line = (2 60,000 0.281 0.72/20) 6.895 = 8470 kPag
MAOP of the branch line = (2 52,000 0.250 0.72/16) 6.895 = 8067 kPag
Step 2. Calculate the pressure required at TO on the branch line side.

Flow velocity = 1.37 m/s

Reynolds number = 1.37 0.394/0.00001 = 54,000
Relative roughness = 0.0001125
Friction factor = 0.0208
Pressure gradient = 0.0208 850 1.372/(2 0.394) = 42.3 Pa/m = 42.3 kPa/km
The pressure required at TO = 3000 + 42.3 50 = 5115 kPag, which is the
minimum pressure required at the take-off point.

Step 3. Calculate the pressure at CE.

Flow velocity upstream of TO = 2.62 m/s

Reynolds number = 2.62 0.494/0.00001 = 129,400
Relative roughness = 0.00009
Friction factor = 0.0175
Pressure gradient = 0.0175 850 2.622/(2 0.494) = 103.3 Pa/m = 103.3kPa/
km
The pressure required at CE = 5115 + 103.3 176 = 23,296 kPag
Since this pressure is much higher than the main line design pressure, pump
stations should be installed along the main line.
Step 4. Find the minimum number of pump stations and locate the required pump
stations along the main line.
Making a small allowance of 370 kPa in the discharge pressure, the discharge
pressure is set at 8100 kPag.
Since equal pumping head reduces overall cost, the equal spacing in flat terrain can achieve the equal pumping head. Also, too short a spacing should be
avoided to minimize capital and operating costs. It can then be safely assumed
that the suction and discharge pressures at each pump station are 350 kPag and
8100 kPag, respectively.
The station spacing is determined by (8100 350)/103.3 = 75 km, which is the
maximum station spacing. Therefore, 214/75 = 2.85 or three pump stations are

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 143

required along the main line. In order to maintain equal pump head for each
station, the spacing is 214 km/3 = 71.3 km or 72 km, if the pressure requirement of 5115 kPag at TO is satisfied. Therefore, the mainline pump stations are
temporarily located at 72 km and 144 km.
Step 5. In order to justify the selection, we need to prove that the pressure requirements at TO and QU are satisfied with the pump stations. Since the full flow can be
delivered to QU, we need to study the hydraulic behaviors of both operations.
TO is located at 32 km downstream of the third mainline pump station, and
the pressure required at TO is 5115 kPag. When the pump station discharges at
8100 kPag, the pressure at TO is 8100 103.3 32 = 4794 kPag. This pressure
does not satisfy the pressure required on the mainline side of TO. The upstream
pump has to be located at (8100 5115)/103.3 = 28.9 km from the TO or
176 28 = 148 kmp.
Dividing this distance in two stations, the station spacing is 148 km/2 = 74km,
which is still less than 75 km. Therefore, the new station locations become
KMP = 74 and KMP = 148. When the second pump is located at 148 km and
discharges at 8100 kPag, the pressure at TO is 5208 kPag, which is higher than
the required pressure there.
If the pressure is maintained and other pressure losses are less than 93 kPa
(5208 5115), no pumping is required along the branch line. Instead, a pressure
control valve is required at TO on the branch line side to regulate the delivery
pressure for low flow rate.
The delivery pressure at QU for full flow delivery The distance between the
third station and QU is 66 km. When the pump station discharges at 8100 kPag,
the full flow delivery pressure is 8100 103.3 66 = 1282 kPag. This pressure
falls outside the acceptable delivery pressure range, and thus a pressure regulator is required at the delivery point.
The delivery pressure at QU for partial flow delivery The partial flow rate is
1800 m3/hr 600 m3/hr = 1200 m3/hr.
Flow velocity between TO and QU = 1.75 m/s
Reynolds number = 1.75 0.494/0.00001 = 86,450
Relative roughness = 0.00009
Friction factor = 0.0189
Pressure gradient = 0.0189 850 1.752/(2 0.494) = 49.8 Pa/m = 49.8kPa/
km
The delivery pressure at QU = 8100 103.3 28 49.8 38 = 3315 kPag.
This pressure is much higher than the required delivery pressure range, and
thus a pressure control valve has to be installed at or upstream of the delivery location.
It should be noted that the differential pressure at the third pump station is different
from the pressure at the other stations.
As a final step of locating the pump stations, the best pump station locations are adjusted on the basis of the following criteria at the time of detail design and construction:

Site terrain conditions

Availability of power infrastructure
Availability of access roads
Potential impact to environment and habitat
Potential impact to the local land owners due to noise, etc.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

144 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

ADDENDA TO CHAPTER 3
A3.1 Temperature Calculation
Temperature has considerable influence on the design of pipelines and related facilities, including the establishment of facilities sizing and optimization, economic and
technical evaluation, etc.
Temperature and pressure influence all fluid properties. In fluid transmission pipelines, both pressure and temperature vary along the pipeline length. In long-distancetransmission pipelines traversing varied terrain, from permafrost regions to moderate
climate conditions, pipelines experience significant temperature changes. Temperature
change affects viscosity, density, and specific heat in liquid lines, particularly in crude
oil pipelines.
In any pipeline segment, the significant overall temperature change (DT) is due to
conduction and convection (DTc). However, there are other factors that affect the overall temperature change. These are (DTe) due to isentropic expansion caused by elevation change and due to isenthalpic expansion caused by friction (DTf) [17]. Therefore,
the overall temperature change in a pipeline segment is:

DT = DTc + DTe + DTf

(A3 1)

The following illustrates a method of overall temperature change due to conduction and convection, DTc. For a pipeline (Figure A3-1) buried at a finite depth (ho)
with insulation, the following expression for computing fluid flow temperature To is
applicable, Holman [18].
Nomenclature:
Cp
= Fluid isobaric specific heat
Dp or D = Pipe outside diameter

Figure A3-1. Heat transfer from a buried pipeline

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 145

Di
= Pipe diameter with insulation
d
= Pipe inside diameter
ha
= Air film coefficient
hf
= Fluid film coefficient
Dh
= Elevation change
Kg
= Soil/surrounding ground thermal conductivity
Ki
= Insulation conductivity
Kp
= Pipe thermal conductivity
DL
= Pipe segment length
Q
= Fluid flow rate
Ti
= Inlet fluid temperature
To or Tf = Outlet/Fluid temperature
Va
= Ambient air/surrounding fluid velocity
rQCp ( Ti - To ) =

2p DLki To - Tg

ln Di / Dp

(A32)

By introducing the shape factor, S, and rearranging the above equation, we have

To =

kg S
Tg
1+ a
kg
rQCp +
1+ a

rQCpTi +

(A33)

where
kg Di
ln
ki Dp

(A34)

2p DL
2

2h
2h
ln o + o - 1
Dp

Dp

(A35)

a=

and the shape factor is defined as
S=

For un-insulated pipeline, a = 1

For an above-ground or offshore pipeline (Figure A3-2) the corresponding fluid
flow temperature is:
-DL
To = Ta + (Ti - Ta ) Exp

rQCpU

(A36)

Where U = overall heat transfer coefficient and is given by:

U=

1 1
1
D 1
D
+
ln
+
ln i

p hf d 2kp d 2ki d

+ h D
a i

(A37)

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

146 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

Figure A3-2. Heat transfer from an above-ground or offshore pipeline

In Eq. (A37), radiation heat losses are ignored as they are small at most normal pipeline operating temperatures. When the pipe is not insulated, the third term in
Eq.(A37) is reduced to zero and Di in the fourth term is set equal to D (i.e., outside
diameter of the pipe).
For above-ground pipeline, the film coefficient (ha) for air can be calculated from
the following equation recommended by Dittus and Boelter, Holman [18].
N u = 0.023 ( Re )

0.8

( Pr )n

(A38)

Where the Nusselt number, Nu, Reynolds number, Re, and Prantl number, Pr, are defined as follows:

h D
Nu = a
ka
rV D
Re = a a
ma

-m aCpa
Pr =
ka

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 147

Fluid film coefficient (hf) for fluid flowing at velocity Vf through the pipe segment
is given by:
C
hf = 0.023k f0.6 pf
f

0.4

(Vf )0.8
d 0.2

(A39)

In the above equations, Cp, , r, and k, respectively, refer to isobaric (constant

pressure) specific heat, viscosity, density, and conductivity of the flowing medium in
the either denoted as suffix (a) for air and (f) for fluid.
For wind blowing over a pipe segment at a velocity of (Va), the film coefficient (ha)
can be calculated from the following equation:
n

k r V D
ha = C a a a i
Di m a

(A310)

Properties of air are provided elsewhere [18]. Values of constant C and exponent n
are dependent on the Reynolds number and are also given elsewhere [18].
For an offshore pipeline, ha can be calculated from Eq. (A3-10) with appropriate
values of Cp, r, k, and m for sea water, and knowing the current velocity.
The following expressions summarize the computation of DTe and DTf.
T V
T
DTe = - Es Ph =
=
Ph
P s Cp T p

(A311)

And

DTf = - JPf =

H
P
T

H
T Pf
P

=-

1 H
Pf
Cp P T

(A312)

Where Es and J are elevation sensitivity and Joule Thompson coefficients, respectively, Ph is pressure loss due to elevation change, and Pf is pressure loss in overcoming
friction. Es can be computed from graphs of pressure (P) and temperature (T) at constant entropy (s), and Pk can be calculated from graphs of enthalpy (H) versus pressure
(P) at constant temperature (T).
The sign of Joule Thompson coefficient J indicates whether fluid expansion
or compression will cause an increase or decrease in the temperature. As an example, in an expanding gas if J is positive, the gas will cool. A negative J in an
expanding gas indicates temperature rise, and is observed in expansion of some
specialgases,e.g., hydrogen. Methods for calculating Es and J are given elsewhere
[19].
The above procedure outlined above provides an accurate prediction of fluid flow
temperature under steady-state condition for buried and exposed pipelines. Sample
plots of temperature profiles for a liquid pipeline (carrying bitumen/condensate mixture) is shown in Figure A3-3.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

148 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

Figure A3-3. T
 emperature profile in a buried 12 pipeline transporting non-Newtonian
fluid[17]

A3.2 Erosional Velocity of Fluid

Liquid pipeline operations are limited by the following factors that impact the fluid
flow velocity:
Static electricity (affecting both low and high vapor pressure (HVP) products
pipeline)
Erosional velocity (affecting oil and low vapor pressure (LVP) pipeline)
Generation of static electricity is of a concern for pipeline transporting high vapor
pressure products such as LPG (propane, butane etc.). Industrys practice is to limit
pipeline fluid velocity to 3 to 5 m/s (< 16 ft/s), depending on liquids.
Erosion occurs due to high velocity, especially in the presence of sand or bubbles of
particles (Figure A3-4). Erosion is particularly severe when corrosive agents also exist
in the fluid. Erosion can be best controlled by proper design and operational limits.
Erosional velocity limits in liquid pipelines are based on gas/liquid density at the
operating pressure and temperature and the likely entrainment of particulates such as
sand in the pipeline in gathering and injection lines.
Erosional velocity can be calculated from equation, Ve = C/r0.5. In this case density r is replaced by rm (density mean value), representing the density at initial and
final flowing conditions:

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 149

Figure A3-4. Erosion in a pipeline

rm =

12, 409Si P + 2.7 RSg P

192.7P + RaTZ

(A313)

where
Ve = Erosional velocity, ft/sec
C = Constant defined as 75< C <150
P = Operating pressure, psia
Si = Liquid specific gravity (water = 1; use average specific gravity for hydrocarbon mixes such as interface) at standard conditions.
Ra = gas/liquid ratio, ft3/bbl at standard conditions
T = Operating Temperature oR
Sg = Gas specific gravity (air = 1) at standard conditions
Z = compressibility factor at the specified pressure and temperature,
Depending on gas/liquid ratio and the amount of entrainment of particulates, it is
a standard practice to limit the fluid velocity about 5 m/s. However, in two-phase flow
lines minimization of slugging in pipeline and separation equipment set the maximum
velocity. In this case the velocity is set to about 10 ft/sec (3 m/sec), ANSI/API RP-14E
[20]. This is particularly important in long lines with elevation changes.

A3.3 Minor Pressure Losses

A liquid pipeline system is composed of pipe, valves, flanges and fittings, and includes
facilities such as pump stations, meter stations, tanks, and pig launchers and traps. A
heavy crude pipeline may require heaters, while a pipeline in a permafrost zone needs
chillers.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

150 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

All pipeline components and facilities cause pressure losses as the liquid passes
through them. The facilities include valves, joints and fittings, pipe bends and elbows, pipe enlargers or contractors in addition to pipes. In most long-distance pipelines, the pressure losses due to valves and joints are comparatively small, so such
pressure losses are called minor pressure losses. Since the number of components
and facilities is small, the overall minor pressure losses are so small that they can
be approximated without significant errors. On the other hand, the pressure loss due
to these components can be substantial within facilities such as stations and tank
farms.
This addendum discusses the calculation of minor pressure losses that would occur when liquid flows through some valves, pipe and fittings. Depending on the components, minor losses can be calculated in two ways:
Minor losses are expressed as a kinetic energy term in terms of the pressure
loss coefficient, K:
DPm = KDV2/2

(A314)

Minor losses in terms of an equivalent length of straight pipe, Le:

Pm = frV2Le/(2D)

(A315)

where
DPm = Minor pressure loss
K = pressure loss coefficient (dimensionless).
V = velocity of liquid through valve, joint or fitting
r = liquid density
f = friction factor
D = pipe inside diameter
Le = equivalent length of straight pipe
The value of K is determined mainly by the flow geometry or the shape of the device. K is analogous to the term fL/D for a straight length in the Darcy equation.
It is more convenient to use K, because K is less dependent on the Reynolds
number and relative roughness. K and Le values are obtained from Reference [21].
Some representative values of K and Le are shown in Table A3-1. Since actual values
depend on the design and fabrication of the components, actual manufacturers data
shall be consulted in the final design.
Example of Equivalent Length Method: A piping system of a pump station is
140 m of NPS 24 pipe that has two 24 gate valves, three 24 ball valves, on swing
check valve, and two 90 standard elbows. Using the equivalent length concept, calculate the total pipe length of the station.
Solution:
Convert all valves and fittings in terms of 24 pipe as follows:

Two 24 gate valves = 2 24 8 0.0254 = 9.75 m

Three 24 ball valves = 3 24 3 0.0254 = 5.49 m
One 24 swing check valve = 1 24 100 0.0254 = 60.96 m
Two 90 elbows = 2 24 30 0.0254 = 36.58 m

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 151

Table A3-1. Loss coefficients and equivalent lengths
Loss coefficient
(K)

Types
Pipe entrance:
Sharp edged
Well-rounded
Pipe exit:
Sharp edged
Well-rounded
90 Bends:
Smooth threaded
Smooth flanged
Miter bend
Miter bend with vanes
Tee Branch flow:
Flanged
Threaded
Tee line flow:
Flanged
Threaded

0.5
0.03
1.0
1.0
0.9
0.3
1.1
0.2
1.0
2.0

Types
Valves (fully open):
Gate valve
Globe valve
Plug valve
Angle valve
Ball valve
Globe lift check valve
Angle lift check valve
Standard elbow:
45
90
Standard tee:
Flow through run
Flow through branch

Equivalent length
(Le/D)
(*)
8
340
18
150
3
600
55
16
30
20
60

0.2
0.9

(*) K values for valves vary with their type and size.

Total equivalent length of straight pipe and all fittings = 140 m + 112.78 m =
252.78 m
The pressure drop due to friction is calculated based on 252.78 m of pipe.
Gradual Enlargement and Reduction
Consider liquid flowing through a pipe of diameter D1. If the diameter enlarges or
reduces to D2, the pressure loss can be calculated as follows:

DPm = Kr(v1 v2)2/2 = Kr(A2/A1 1) 2 v22/2

(A316)

where v1 and v2 are the velocity of the liquid in D1 and D2 pipes and A1 and A2 the areas.
The value of K depends on the diameter ratio and the different angle due to the enlargement or reduction. The figure below shows a gradual pipe enlarger and a reducer.

Figure A3-5. Gradual pipe enlargement and reduction

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

152 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

The loss coefficients for an enlarger are given in Tables A3-2 and A3-3 below:
Table A3-2. Loss coefficients for enlarger
Angle

20

45

60

90

180

A2/A1 = 2.25

0.45

1.06

A2/A1 = 9

0.43

0.88

1.19

1.11

1.00

1.02

1.04

1.02

The loss coefficients for a reducer are given in the table below:
Table A3-3. Loss coefficients for reducer
Angle

20

45

60

90

180

A2/A1 = 0.5

0.05

0.06

0.06

0.12

0.26

A2/A1 = 0.25

0.03

0.07

0.07

0.17

0.41

Enlarger Example: Calculate the pressure loss due to a gradual enlargement in a

pipe that flows 1800 m3/hr of diesel from 8 diameter to 12 with an angle of 60. Both
sizes are internal diameters.
Solution:
Solution: The liquid velocity in the 12 pipe size is v2 = 1.714 m/sec, and diameter
ratio= 12/8 = 1.5. From the diagram, the value of K is 1.19 for area ratio = 2.25 and
angle = 60. Therefore, pressure loss due to gradual enlargement is

DPm = 1.19 850 (2.25 1)2 1.7142/2 = 6037 Pa = 6.037 kPa

Sudden Expansion: Minor loss for the sudden expansion of a pipe can be expressed as
DPm = r(1 D12/D22) 2 V12/2

(A317)

Sudden Contraction: Minor loss for the sudden contraction of a pipe can be
expressed as
DPm = r(1/Cc 1) 2 V 22/2

(A318)

where Cc is given in Table A3-4 below.

Table A3-4. Sudden contraction parameters
A2/A1
Cc

0.0
0.585

0.2
0.632

0.4
0.659

0.6
0.712

0.8
0.813

1.0
1.0

Example: A tank open to the atmosphere is filled with 40API (specific gravity
of 0.825) oil to a height of 10 m from the bottom. A tap at the bottom of the tank is
opened, and oil flows out from the outlet. Assuming that the flow is steady and incompressible, determine the tank discharge pressure at point 2.
Solution:
Assume that as the oil is discharged out of the tank, the tank level drops so slowly that
the flow velocity at point 1 is negligibly small. The data for this example are:
Pipe = 16 OD 0.25 wt: inside diameter = 0.394 m
Pipe roughness = 0.0018 = 0.0457 mm
Crude viscosity = 3 cSt of 40API crude

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 153

10m
Oil
200m
16 OD

Figure A3-6. Example of Oil Discharge from a Tank

Density = 825 kg/m3

Flow rate = 1000 m3/hour or velocity = 2.28 m/sec
Step 1. From the momentum conservation in steady state, we have
(P1 + rv12/2 + rgz1) - (P2 + rv22/2 + rgz2) = frv22/2 L + Krv22/2

where f is the Darcy friction factor and L the pipe length, and K = Kentrance + Kexit are
the pressure loss coefficients for pipe entrance and exit, respectively.
Step 2. P1 is atmospheric pressures, v1 = 0, v2 = v, z1 = 10 m, and z2 = 0. Therefore,
the equation becomes
P2 = rgz1 rv2/2 frv2/2 L (Kentrance + Kexit)rv2/2

where Kentrance = 0.5 and Kexit = 1.0.

Step 3. Calculate friction factor and pressure drop for the 200 m long 16 pipe

Re = 2.28 0.394/0.000003 = 299,000: partially turbulent

Relative roughness = 0.0000457 m/0.394 m = 0.000116
f = 0.0156
Pf/L = 0.0156 825 kg/m3 (2.28 m/sec)2/(2 0.394 m) = 84.9 Pa/m =
0.0849kPa/m
Total pressure drop for 200 m long pipe = 0.0849 200 = 17.0 kPa
Step 4. Determine the tank discharge pressure

Static pressure at the tank outlet, rgz1 = 825 9.8 10/1000 = 80.9 kPa
Velocity pressure, rv2/2 = 825 (2.28) 2/2 = 2144 pa = 2.14 kPa
Pressure losses due to entrance and exit, (Kentrance + Kexit)rv2/2 = 1.5 2.14 =
3.21
Pressure at the pipe discharge point = 80.9 2.14 - 17.0 3.21 = 58.6 kPa
Step 5. Discuss the result of this example
The pressure losses due to pipe entrance and exit can be relatively significant for
low pressure system, whereas they are negligibly small for large pressure system such
as a transmission pipeline.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

154 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

A.3.4 Effect of Pressure and Temperature on Pipe Volume

When a pipeline transporting hydrocarbon liquids experiences a change in pressure
and temperature, both the pipe and the liquid contained therein are affected. For a
given segment or cross section of a pipe (Figure A3-7), the change in pressure can be
due operational variables and the change in temperature due to environmental/ambient
conditions (surrounding the pipe) at a given time.
Considering a uniform stress in the pipe wall thickness and assuming that variations in pipe cross sectional diameter are negligible and that there is no variation in
temperature across any cross section and as well pipe ends are considered as fully
closed, so the internal volume of the pipe is equal to the volume of contained fluid,
then the following illustrate the relationship between pressure, temperature and volume in a section of a pipeline filled with a hydrocarbon liquid [9].
As indicated previously both the pipe and fluid are affected by changes in pressure (dP) and temperature (dT). Therefore, any incremental increase in volume of the
pipe (dVpipe) must be related to change in liquid volume (dVfluid) caused by change in
temperature or pressure. Thus,

dVpipe = dVfluid

(A319)

By rearranging the expressions for both the bulk modulus and the coefficient of
volume thermal expansion of the fluid, it can be written that

dVfluid = BV dT - V dP / K

(A320)

where
B = coefficient of volumetric thermal expansion of the fluid.
K = bulk modulus of fluid
The determination of the incremental increase in volume of the pipe (dVpipe) is
dependent upon the degree to which the pipeline is free to move. Consider first the
case where a welded pipeline has unrestrained movement such that it is free to expand
in both the radial and longitudinal directions.

Figure A3-7. Generalized view of a liquid pipeline under internal pressure

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 155

The incremental volume increase in the pipe will be due to the effects of both
pressure and temperature. In the case of pressure change, the pipe wall experiences a
state of biaxial stress, a circumferential or hoop stress Sq, and an axial stress Sx. The
respective hoop and axial strains are:

S - nSx DdP
eq = q
= 4tE (2 - n)

(A321)

S - nSq DdP
ex = x
= 4tE (1 - 2n)

(A322)

and

where

n = Poissons ratio
eq = Hoop strain (i.e., strain in radial direction)
ex = Axial strain (i.e., strain along the length of the pipeline)

Sx = Axial stress = DdP/4t

Sq = Hoop stress = DdP/2t
E = Modulus of elasticity
t = Pipe wall thickness

and the new pipe length L* and diameter D* become

L* = L(1 + ex)

(A323)

D* = D(1 + eq)

(A324)

The new volume of the pipe section is given by pL*(D*)2/4. Upon substitution
of the expressions for D* and L*, and ignoring the cross products of strain as being
negligible, the new pipe volume V* is given by:

V* =

pD 2 L
(1 + 2e q + e x)
4

(A325)

The incremental increase in volume, dV, due to incremental change in pressure,

dP, can be found by substituting the expressions for eq and ex into Eq. (A325), and
noting that dV = V* - V, where V is the original volume of the pipe. Hence, for fluid
and pipe:

dVpipe =

VDdP
(5 - 4 n)
4tE

(A326)

The incremental volume change due to a temperature change dT is calculated

using

dD = DadT

(A327)

dL = LadT

(A328)

and

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

156 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

where a is the coefficient of linear expansion of pipe material, and dD and dL are changes
in diameter D and length L, respectively. By neglecting the cross products of the coefficient of thermal expansion when evaluating the new volume, the incremental volume is:

dV = 3aV dT

(A329)

The combined effect of incremental pressure dP and temperature dT change can be

found by substituting Eqs. (A320), (A326) and (A329) into Eq. (A319):

dP
DdP

V
(5 - 4 n) + 3adT = V BdT 4tE

(A330)

Rearranging the terms and eliminating V:

dP =

( B - 3a ) dT
1 D (5 - 4 n)
+
4tE

(A31)

Equation (A331) provides the change in pressure due to a temperature change

for an unconstrained pipe with closed ends. It may be noted that, in this circumstance,
it is independent of pipe length. Similar expression can be derived for buried pipeline
which are considered as restrained. Refer to ref. [9].
For a buried pipeline change in pressure dP caused by change in temperature dT
and thence the change in volume dV due to the change in pressure can be computed
from the following expressions:

B - 2a (1 + v ) dT
dP =
D 1 - v 2 / tE + 1/ K

(A332)

The effect of axial thermal stress on restrained pipelines capped at the ends can be
neglected [22]. Ignoring the temperature increase effects simplifies Equation (A3.4.14)
such that, for a restrained pipe, the theoretical volume required to pressure (squeeze)
the pipeline section is given by:

1
D
dV = V dP
1 - v2 +
tE
K

(A333)

Example:
To illustrate the effect of a change of temperature on pressure consider a pipe of outside
diameter 114.3 mm and wall thickness 3.18 mm transporting water under pressure.
Neglect the effect of axial thermal stress on a restrained pipeline Eq. (A332) can be
simplified to become [22]:
dP =

B - 2a
D
1 - v2 + C
Et

where a = 1.17 105

B = (-64.268 + 17.105 T- 0.20396 T2 + 0.016048 T3) 106

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

System Hydraulics and Design n 157

Figure A3-8. P
 ressure change due to temperature change (medium water in 48 pipeline, no
flow condition)

E = 2.0 108
n = 0.3
C = 1/K = 4.86 107 (K= 2.057 106 kPa assumed for water at 10C)
Note that C (the reciprocal of the bulk modulus for liquid, in this case water) is
a function of the pressure and temperature. Assuming the temperature T is 10C then
upon substituting, B has the value 81.0728 106 and the pressure increase due to a
1C temperature change is:
dp =

81.0728 x10 - 6 - 2 1.17 10 -5

114.3
(1 - 0.32 ) + 4.86 10 -7
200 106 12.7

That is a pressure increase of 98.9 kPa.

In a similar manner, the change in pressure due to a temperature increase over
a range of temperatures for a large restrained transmission pipeline whose outside
diameter is 1219 mm (48), having a wall thickness of 12.7 mm (1/2) and containing
water, can be calculated. The resulting pressure change due to temperature change is
shown in Figure A3-8.
If it is assumed that water temperature increases from 1 to 3C during a period of
time then the end result is a pressure decrease of 100 kPa would be obtained by summing the pressure decreases in each time interval, that is 59 kPa between 1 and 2C,
and 41 kPa between 2 and 3C. It may be noted that had the temperature decreased
from 3 to 1C the pressure would have increased by 100 kPa.

REFERENCES

[1] American Petroleum Institute, 1984, Petroleum Liquid Volume Correction, API Publication
1101.
[2] NIST Standard Reference Database 4, Supertrap version 3, National Institute of Standards and
Technology, Gaithersburg, MD 20899, U.S.A.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use

158 n Hydrocarbon Liquid Transmission Pipeline and Storage Systems

[3] Finlayson, B. A., 1980, Nonlinear Analysis in Chemical Engineering, McGraw-Hill, New York,
N.Y.
[4] Wylie, E. B., and Streeter, V. L., 1983, Fluid Transients, FEB Press, Ann Arbor, MI.
[5] Carnahan, B., Luther, H. A., and Wilkes, J. O., 1969, Applied Numerical Methods, John Wiley &
Sons, Inc. New York, N.Y.
[6] Wylie, E. B., Stoner, M. A., Streeter, V. L., 1971. Network System Transient Calculations by
Implicit Model, Soc. Pet. Eng. J.
[7] Rachford, H. H., and Dupont, T., 1974, A Fast, Highly Accurate Means of Modeling Transient
Flow in Gas Pipeline Systems by Variational Methods, Soc. Pet. Eng. J.
[8] Meyer, K. J., Smith, R. W., Murray, A., 2008, Comparison of US and Canadian Transmission
Pipeline Consensus Standards, Proc. of IPC 2008, Sep. 29Oct. 3, pp. 17.
[9] Mohitpour, M., Golshan, H., Murray, A., 2007, Pipeline Design and Construction A Practical
Approach, 3rd Ed. ASME Press, New York, NY, USA.
[10] Adam, S., and Davis, K., 2009, Pipeline Geomatics, ASME, New York, NY.
[11] 1998, Engineering Data Book (SI Version), 11th Edition, Gas Processors Suppliers Association
(GPSA), Tulsa, OK, USA.
[12] Gregory, G. A., Aziz, K., and Moore, R. G., 1979, Computer Design of Dense-Phase Pipelines,
J. Petroleum Technology.
[13] Urquhart, R. D., 1986, Heavy Oil Transportation Present and Future, J. Can. Pet. Eng.,
pp.6871.
[14] Arnold, C. L., 1981, How Temperature Affects Pipeline Hydraulics Design for Heavy Crudes,
Resid, Oil & Gas J., pp. 104120.
[15] Smith, B., 1979, Pumping Heavy Crudes 1: Guidelines set out for pumping heavy crudes,
Oil & Gas J., pp 111 114, May 28, 1979, Pumping Heavy Crudes 2: Steps for finding crude
properties, Oil & Gas J., pp 150 152, June 4, 1979, Pumping Heavy Crudes 3: Heat transfer explored in pipelining high-pour-point crude oil, Oil & Gas J., pp 110 111, June 18, 1979,
Pumping Heavy Crudes 4: Restart of heavy crude lines probed, Oil & Gas J., pp 105 106,
July 2, 1979, Pumping Heavy Crudes 5: Design of heavy crude facilities explored, Oil & Gas
J., pp 69 70, July 16, 1979.
[16] Rahimi, P., et al., 2009, 5th National Centre for Upgrading Technology (NCUT) Upgrading and
Refining Conference, Sep. 1416.
[17] Mohitpour, M., 1991, Temperature Computation in Fluid Transmission Pipelines, Proc. of
ETCE, Pipeline Engineering Symposium, ASME, Houston, TX, PD-Vol. 34.
[18] Holman, J. P., 1981, Heat Transfer, 5th Ed, McGraw Hill Book Co., New York, NY, USA.
[19] Edminster, W. C., 1972, Applied Hydrocarbon Thermodynamics, Gulf Publishing Co., Houston,
TX, USA. Vol. 2.
[20] ANSI/API RP 14E 2007, Recommended Practice for Design and Installation of Offshore Products Platform Piping Systems.
[21] Fox, R., McDonald, A., and Pritchard, P., 2004, Introduction to Fluid Mechanics, 6th Edition, John
Wiley & Sons, Inc. NJ., USA.
[22] Gray, J. C., 1976, How Temperature Affects Pipeline Hydrostatic Testing, Pipeline and Gas J.,
pp. 2630, Aug.

Downloaded From: http://ebooks.asmedigitalcollection.asme.org/ on 12/11/2015 Terms of Use: http://www.asme.org/about-asme/terms-of-use