Applied Pipeline Hydraulics

APPLIED/PIPELINE HYDRAULICS 1
Chapter One
APPLIED/PIPELINE HYDRAULICS
Most chemical engineering processes involve fluids flow from one location to the other.
Finished products manufactured (in fluid form) are transported from plant sites to the
discharge points where end users have access to use such products. Raw materials,
intermediate and semi-finished products (in liquid/gaseous form) move either by gravity
and or are transported by expenditure of energy from one unit to the other for quality
control purposes and to maintain product specifications. This chapter considers basic
engineering principles required to handle fluid in plant operations. Detailed transfer
phenomena are not presented here, but could be obtained in literature cited and in Perry’s
Chemical Engineers Handbook. Pipeline hydraulic design procedures are presented. A
typical design module for pressure vessel with the application of ASME CODES and
PVELITE Software is presented in this chapter.
1.1 Commonly Used Terms/Parameters

The ratio of the weight of a given volume of a substance in vacuum at a given
temperature to the weight of an equal volume of water in vacuum at a given temperature
is known as the absolute specific gravity (Licker, 2003).
The weight of a unit volume, γ and the density (the mass of a unit volume)  are related
by:
γ= g (1.1)
where g = 9.81 m/s2 is the acceleration due to gravity

Since the unit of mass in the S I system (kg) is the mass of the International prototype of
the kilogram, while the unit of force in the MK (force)S system (kgf) is the weight of the
same prototype then the density in SI units (kg/m3) is numerically equal to the unit
weight in MK(force)S units (kgf/m3).
The relative density (specific gravity) ∆ is the ratio of the density (unit weight) of a
substance to the density (unit weight) of water,
2 CHEMICAL ENGINEERING UNIT OPERATIONS, SYNTHESIS AND BASIC DESIGN CALCULATIONS
 
∆= = (1.2)
 w  w
The density of a mixture of liquids whose mixing is not affected by any appreciable
physicochemical changes can be calculated approximately by assuming that the volume
of the mixture equals the sum of the volumes of its components:
1 x1 x2 (1.3)
   ...
mix 1 2
where x1 , x 2  mass fractions of the mixture components
1 ,  2  densities of the mixture and its components
Similarly, the density of a suspension is calculated by the following:

1 x (1  x)
  (1.4)
susp c ld
where x = mass fraction of the solid phase in the suspension
 c and ld  densities of the solid (crystalline) and liquid phases respectively
On the basis of Clapeyron+ equation, the density,  of a gas at the temperature, T and
pressure P can be calculated by the equation:
T OP M 273P
  O  (1.5)
TP O
22.4 TP O
Where M/22.4 kg/m3 = density of the gas in standard conditions (denoted by the
superscript O), i.e at T = 273.15K and PO = 760 mmHg = 1.013 x 105 Pa; M = molar mass
of the gas, kg/kgmol, T = temperature, K.
Note: The pressures P and PO must be expressed in the same units.
The density of a mixture of gases is expressed as follows:
mix  11  22  ... (1.6)
where 1 ,  2  volume fractions of the components of the gas mixture
1 ,  2  respective densities of the components

+
In the Clapeyron equation PV = RT for 1 kmol of a gas, the molar gas constant, R =
(POVO)/TO = 760 x 133.34 x 22.4/273 = 8310 J/Kmol.K
A type of stress which is exerted uniformly in all direction; its measure is the force
exerted per unit area is called pressure.
The pressure, P of a column of liquid having a height, h with a liquid of density  can be
expressed as follows:
P  gh (1.7)
Equation (1.7) gives us the following relationships between the various units of pressure:
1 atm = 760 mmHg =  gh = 13600 x 9.81x0.76 = 1.013 x105 Pa
1 atm (technical atmosphere) = 1 kgf/cm2 = 104 kgf/m2

= 9.81x 104 Pa = 735 mmHg = 104 mmH2O
1.2 The fundamental equation of hydrostatics is expressed as:

P = PO +  gh (1.8)
The force due to the pressure of a liquid on a flat wall, F (in N) is:
P = (PO +  g hW) A (1.9)
Where PO = pressure on the surface of the liquid, Pa

hW = depth of submergence of the centre of gravity of the wall under the level of the
liquid, m
A = area of the wall surface, m2
 = density of the liquid, kg/m3
g = acceleration due to gravity, = 9.81 m/s2

Hydrostatic test details the determination of the burst resistance or leak tightness of a
fluid component or system by imposing internal pressure. This phenomenon is applied in
pigging of pipeline systems; pressure vessel test of maximum allowable load and others.
The units of dynamic viscosity, often called viscosity, are:
in the S I System [  ] = N .s = Pa .S = kg
m2 m .s
in the M K (force) S System [  ] = kgf 2.s

m
in the CGS system [  ] = dyn .2 s = g = P (poise)

cm cm . s
the relationship between the customarily used viscosity units is:
1 cP = 10-3 Pa.s = 1 kgf . s

9810 m 2
Viscosity-gravity constant is an index of chemical composition of crude oil; defined as

the general relation between specific gravity and Saybolt universal viscosity; the constant
is low for paraffinic crude oils and high for naphthenic crude oils.
The kinematic viscosity,  (in m2/s) is related to the dynamic viscosity  by the
expression:
 
 (1.10)

The values of  for liquids at different temperatures can be determined by means of

nomograph. In the absence of experimental data, the dynamic viscosity of many organic
liquids (at 20 OC) can be determined approximately according to the following empirical
equation (Souders, 1938), (Planovsky & Kafarov, 1944):

Log log u    An   B  3  2.9 (1.11)
10 M
where  = dynamic viscosity of a liquid at atmospheric pressure and 20 OC, mP
 = density of the liquid, kg/m3
M = molar mass, kg/kg mol

A = number of identical atoms in a molecule of the organic compound
n = numerical value of the atomic constant found in Table 1.1
B = correction for the grouping of the atoms and the nature of the bond between them
given in Table 1.2
It is convenient to use a nomogram to determine  on the basis of Equation (1.11). The

quantity log (log  ) in the nomogram is denoted by y. If y is positive, the right hand
scale should be used, and if it is negative, the left hand scale is used. A definite value of
O
 - the dynamic viscosity at 20 C – corresponds to each value of y.
Viscosity index (VI) is an arbitrary scale used to show the magnitude of viscosity
changes in lubricating oils with changes in temperature.
1.3 To determine the dynamic viscosity of a gas at different temperatures, the

nomogram in Fig.1.1 can be used. The dynamic viscosity of a gas mixture can be
calculated according to the following approximate equation:
M mix 1M 1  2 M 2 (1.12)
   ...
 mix 1 2
Fig 1.1: Nomogram for determining the dynamic viscosities of gases at p = 1atm (the
data may also be used with a permissible approximation for pressures of several
atmospheres) (Source: Pavlov et al., 1979 p 591)
1-O2; 2 – NO; 3 – CO2; 4 – HCl; 5 – Air; 6 –N2; 7- SO2; 8 – CH4; 9 – H2O; 10 – NH4; 11 – C2H6; 12 – H2;
13 – C6H6; 14 – 9H2 + N2; 15 – 3H2 + N2; 16 – CO; 17 – Cl2.
Conversion to SI: 1 cP = 1 x 10-3 Pa.s

Where Mmix, M1, M2… = molar masses of the mixture and separate components
mix , 1 , 2 ….. = corresponding dynamic viscosities
1,2 ,.... = volume fractions of the components in the mixture
The following empirical equation, which has been checked for a number of gas mixtures
(coke-oven gas, producer gas, etc.) at atmospheric pressure, is also used to calculate mix
for gases:
( 11 )  M1Tcr1  ( 2 2 )  M 2Tcr 2  ... (1.13)

max 
 1 M1Tcr   2 M 2 Tcr 2  ...
where Tcr1,Tcr2… are the critical temperatures of the components, k.  1 , 2 are the
volume fractions of the components in the mixture. The other symbols are the same as in
Equation (1.12)
When using Equation (1.13) all the values of the viscosities  1  2 ... must be taken for
the same temperature T, and the viscosity of the mixture,  mix is obtained for this
M Tcr
temperature. Values of for selected gases can be obtained from tables.
The change in the dynamic viscosity of gases with the temperature is expressed by the
equation:
(1.14)
1.5
273 C  T 
T  0  
T  C  273
where  0  dynamic viscosity at 0 OC
T = Temperature, K
C = Sutherland’s constant [0-26] Pavlov et al., 1979
For a mixture of normal (non- associated) liquids, the value of
mix, can be calculated by the expression:
log mix  x1 log1  x2 log2  ... (1.15)
where 1 ,  2 ,  dynamic viscosities of the separate components
x1, x2, .= mole fractions of the components in the mixture

In accordance with the additivity of the fluidities of the components, the dynamic
viscosity of a mixture of normal liquids is determined by the equation:
1 1 2
   ... (1.16)
mix 1 2
Where 1 , 2 ... are the volume fractions of the components in the mixture.
The dynamic viscosity of dilute suspensions  susp can be calculated by the following
equations:
When the volume concentration of the solid phase is less than 10 %
 susp = ld (1  2.5) (1.17)
When the volume concentration of the solid phase is up to 30 % Weinspach (1969)
 susp = ld 0.59

(1.18)
(0.77  )2
where  ld  dynamic viscosity of the pure liquid

  volume fraction of the solid phase in the suspension
1.4 Equation of rate of flow

1.4.1 The volumetric rate of flow of a fluid (liquid or gas) Qv (in m3/s) is:
Qv = vA (1.19)
The mass rate of flow of a fluid, Qm (kg/s) is:
Qm = QV  = vA  (1.20)
where A = cross-sectional area of the flow, m2

v = mean velocity of the flow, m/s
 = density of the fluid, kg/m3

2
For a round pipe, A in Equation (1.19) is substituted with  d and we have:
4
QV = 0.785 d2v (1.21)

where d is the internal diameter of the pipe, m.
The equation of continuity for an incompressible fluid flowing through a pipe with
variable cross-sectional area is:
Qv = v1A1 = v2A2 = v3A3 = … (1.22)
1.4.2 Buoyancy
The same principles used above to compute hydrostatic forces can be used to calculate
the net pressure force acting on completely submerged or floating bodies. These laws of
buoyancy, the principles of Archimedes, are that:
1. A completely submerged body experiences a vertical upward force equal to the

weight of the displaced fluid; and
2. A floating or partially submerged body displaces its own weight in the fluid in
which it floats (i.e., the vertical upward force is equal to the body weight).
The line of action of the buoyancy force in both (1) and (2) passes through the centroid of
the displaced volume of fluid; this point is called the center of buoyancy. (This point need
not correspond to the center of mass of the body, which could have non uniform density.
In the above it has been assumed that the displaced fluid has a constant g. If this is not the
case, such as in a layered fluid, the magnitude of the buoyant force is still equal to the
weight of the displaced fluid, but the line of action of this force passes through the center
of gravity of the displaced volume, not the centroid.)
If a body has a weight exactly equal to that of the volume of fluid it displaces, it is said to
be neutrally buoyant and will remain at rest at any point where it is immersed in a
(homogeneous) fluid.
1.5 Principal dimensionless numbers for flow through pipes and

channels
1.5.1 The Reynolds number characterizing the hydrodynamic conditions and serving as a
measure of the forces of inertia and internal friction (viscosity) in a flow is given as:
Re = vd  vd (1.23)

 
where v = mean velocity of the flow, m/s

d = diameter of the pipe, m
3
 = density of the liquid, kg/m
 = dynamic viscosity, Pa. s
 = kinematic viscosity, m2/s

Flows through straight pipes are characterized by the following values of the Reynolds
number:
Laminar flow Re  2000
Transition flow 2000  Re  3000
10000
8000 Turbulent flow
6000
Recr
Laminar flow
4000
2000
0.001 0.002 0.003 0.004 0.005 0.006 0.007

d/D
Fig. 1.2: Dependence of Recr in coils on the ratio d/D
Developed turbulent flow Re  3000

In practice it has been observed that flows through bends, coils have critical value of the
Reynolds number, Recr that is higher than for straight pipes. This depends on the ratio
d/D, where d is the internal diameter of the coil pipe and D is the diameter of the coil
turns. A relationship of this is shown in Figure 1.2
The curve represents the transition region. For flows through pipes that are not round, the
equivalent diameter equal to four times the hydraulic radius is introduced into the
expression for Re.
The hydraulic radius (hydraulic mean depth) rh is the ratio of the area of the flow cross
section A to the wetted perimeter P:
rh = A (1.24)
P
For a round pipe completely filled with a liquid we have the expression:
rh =  d  d
2
(1.25)
4 d 4
Consequently, the equivalent diameter deq can be used instead of the diameter for flows
through pipes whose cross section is not round.
deq = 4 rh = 4 A (1.26)
P
1.5.2 The Froude number, which is a measure of the ratio of the forces of inertia and
gravity in a flow, is given as:
Fr = v2 (1.27)
gd
where v is the velocity of the flow, m/s

Other researchers expressed the Froude number as:
v (1.28)
Fr 
gd 0.5
1.5.3 The Euler number, which is a measure of the ratio of the forces of pressure and
inertia in a flow, is:
P
Eu  (1.29)
v 2
where ∆P is the pressure difference (loss of pressure) spent for overcoming the hydraulic
resistance, Pa
1.5.4 The Bernoulli equation for a non – viscous (ideal) incompressible liquid is
given as:
z1 
P1 v1

2
P v
 z2  2  2
2
(1.30)
g 2g g 2g
For a viscous (real) incompressible liquid, the following expression is applied:

2 2
P1 v1 P v (1.31)
z1    z 2  2  2  h1
g 2 g g 2 g
where z = geometrical (position ) head, m

P = piezometric (static) head, m
g
v2 = velocity (dynamic) head, m

2g
h1 = hydraulic loss head, m

The adiabatic flow of an ideal gas flowing through a frictionless conduit or a constriction
(such as an orifice nozzle, or valve) can be analyzed with the energy balance. The total
energy balance is similar to that of the liquid and is given as follows:
∆h + g ∆z + ½ ∆v2 = q + w
For adiabatic flow in a constant area duct, the governing equations can be formulated in a
more generalized dimensionless form that is useful for the solution of both subsonic and
supersonic flows. The resulting expressions and how to apply them are not shown here,
but can be obtained from the work of Darby, (2001).
1.5.5 The relationship between the mean velocity, vmax in a pipe is:
a) For laminar flow conditions, v = 0.5vmax
b) For turbulent flow conditions the ratio v/vmax depends on the value of the Reynolds
number Re vmaxd

v/vmax
0.8
0.7
0.6
0.5
0.4
0.3
103 104 105 106 107 108
Re
Fig. 1.3 Dependence of ratio v/vmax on Re
From Figure 1.3 one can assume for turbulent flow conditions that v approximately
equals from 0.8 vmax to 0.9vmax, although the ratio v/vmax may exceed 0.9 at high values of
Re.
1.6 The velocity of a discharge of a liquid v (m/s) from a small orifice in

the bottom or a wall of a vessel with a constant level of the liquid in
the vessel is calculated by the equation:
v =  2gH 
0.5
(1.32)
where  = dimensionless velocity coefficient
g = acceleration due to gravity, m/s2

H = height of liquid level above centre of orifice, m
If the pressure on the surface of the liquid in a vessel (Pa, Po) and the pressure in the
space into which the stream is flowing out of the vessel (P, Pa) are different, then the
quantity H1 = H + (Po – P)/  g, where  is the density of the flowing out liquid, kg/m3,
should be substituted in Equation (1.32) for H.
The volumetric rate of flow of a liquid Qv (m3/s) flowing through an orifice with the area
Ao (m2) at a constant level of the liquid in the vessel and at Po = P is:
Qv = Cd Ao (2gH)0.5 (1.33)
where Cd is a dimensionless discharge coefficient. It is the product of the velocity

coefficient  and the coefficient of contraction 
Cd =  (1.34)
do
d

man
Fig 1-4: Measuring the rate of flow with an orifice plate

1.6.1 The time,  (s) needed for emptying an open vessel with a constant cross sectional
area A through an orifice with the area Ao can be calculated by the equation:
2 AH 0.5 (1.35)

C d Ao 2 g 
0 .5
where H is the initial level of the liquid over the orifice, m

1.7 Measurement of the rate of flow of a fluid using normal measuring orifice plate
(Fig. 1-4)
The volumetric rate of flow of a fluid Qv (m3/s) is given by the equation:
0.5 0.5
Qv =  P   (   )  (1.36)
Cd kAo  2   Cd kAo 2gH man 
     
where Cd = discharge coefficient of the normal measuring orifice plate in a smooth not
rough pipe (Table 1.1)
k = roughness coefficient allowing for the roughness of the pipe walls (the mean values
of k for pipes are given in Table 1.2. For hydraulically smooth pipes k = 1.
Ao = 0.785 do2 = cross –sectional area of the measuring plate orifice, m2
do = diameter of the orifice, m
H = difference between the levels of the liquid in the differential manometer connected to
the orifice plate, m
man = density of the liquid in the differential manometer, kg/m3

 = density of the fluid flowing through the pipe, kg/m3
Since the discharge coefficient of the measuring orifice plate, Cd depends on Re  vd and

the value of Re is not known beforehand then when measuring the rate of flow Qv it is
necessary to determine the mean value of Cd for the given (do/d)2 from Table 1.2. After
calculating Qv, the value of Re can be determined. The value, Cd can then be calculated.
1.8 Measuring the rate of flow of a fluid with the aid of a Pitot-Prandtl
tube
From the stagnation pressure equation for an incompressible fluid, velocity is expressed
as follows:
(1.37)
0 .5 0 .5
 2     P  PS  
v    Po  Ps    2 g  o  

      
where v = velocity of the fluid

Po = the stagnation pressure
Ps = the static pressure of the fluid at the elevation where the measurement is taken.
V2
2g
Figure 1.5: Pitot tube V,

P
The basic tube, as shown in Figure 1.5 has two concentric tubes. The end of the center
tube is open and pointed into the flow. The other tube is closed at the end but has holes
in the side that are perpendicular to the flow, to sense static pressure. The flow is
calculated from the difference between the ram pressure in the center tube and the static
pressure in the outer tube.
An annubar is a more accurate device based on the pitot tube concept. It employs several
velocity sensing openings subjected to the fluid flow. The resulting ram and static
pressures are averaged to obtain an accurate indication of the flow rate (Durgaiah, 2004).
For a compressible fluid, the above incompressible fluid equation can be solved if the
Mach Number ≤ 0.3
1.9 Mach Number

The speed of sound C in an ideal gas is given by:
C = KRT 0.5 (1.38)
C
where K = p
C v
Cp, Cv = heat capacity of gas at constant pressure and volume respectively, J/(mol·K)
This shows that the acoustic velocity in an ideal gas depends only on its temperature.
The Mach number, Ma is a ratio of the fluid velocity, v to the speed of sound, c.
Ma = v/c (1.39)
1.10 Venturi Meters

The Venturi Tube, also known as a low-loss flow tube, works by the same principles as
an orifice plate. The venturi tube has a gradual reduction in diameter to its narrowest
point, the vena contracta, and then a gradual expansion back to the original pipe diameter.
This device is more expensive to manufacture than an orifice plate, but it has a much
lower permanent pressure loss. The internal path of the venturi meter is machine-finished
and specifically contoured to minimize this pressure loss. Permanent losses of the order
of 5 percent of the measured pressure differential are obtainable. A typical Venturi meter
is illustrated in Figure 1.6.
Fig 1.6: The critical dimensions of a typical venturi tube (Source: Spink (1975))
1.11 The power, P (in KW) consumed by the motor of a pump (or a fan) is
calculated by the equation:
Q v P Q v gH p
P  (1.40)
1000 1000
where Qv = volumetric rate of low of the fluid m3/s
P = increase in pressure created by the pump (fan) in the flow being

handled and equal to the total hydraulic resistance of the network, Pa
Hp = head created by the pump, m
 = density of the fluid, kg/m3

g = 9.81 m/s2 = acceleration due to gravity
 = overall efficiency of the pump (fan) equal to the product of the
efficiencies of the pump, p, the transmission tr and the motor m,  = p tr m.
For pump handling liquids, equation 1.40 holds for any values of P.
Table 1.1: Discharge coefficient of Orifice Plates (Pavlov, 1979)

Re= Values of discharge coefficient cd for (do/d)2 equal to
vdp
0.05 0.1 0.2 0.3 0.4 0.5 0.6 0.7

5000 0.6032 0.6110 0.6341 - - - - -
10000 0.6026 0.6092 0.6261 0.6530 0.6890 0.7367 0.7975 -
20000 0.5996 0.6050 0.6212 0.6454 0.6765 0.7186 0.7753 0.8540
30000 0.5990 0.6038 0.6187 0.6403 0.6719 0.7124 0.7650 0.8404
50000 0.5984 0.6032 0.6168 0.6384 0.6666 0.7047 0.7553 0.8276
100000 0.5980 0.6026 0.6162 0.6359 0.6626 0.6992 0.7472 0.8155
400000 0.5978 0.6020 0.6150 0.6340 0.6600 0.6950 0.7398 0.8019
d = internal diameter of pipe, m; v = mean velocity of fluid in pipe, m/s

do = diameter of orifice of a normal orifice plate, m
Table 1.2: Values of the roughness coefficient (Kafarov, et al., 1974)
V Pipe Values of coefficient k for (do/d)2 equal to
diameter,
m 0.1 0.2 0.3 0.4 0.5 0.6 0.7
0.05 1.0037 1.0063 1.0082 1.0118 1.0144 1.0172 1.02
0.10 1.0024 1.0045 1.0064 1.0065 1.0108 1.013 1.0148
0.20 1.0017 1.0023 1.0034 1.004 1.0052 1.006 1.007
0.30 1.0005 1.001 1.001 1.001 1.001 1.001 1.001
When using Eq. 1.40 to calculate the power consumed by a pump or fan, the quantity
becomes:
P = Pvel + Pfv + Plift + Padd (1.41)
where Pvel = expenditure of pressure for creating a velocity of the flow at the outlet on
the network (the velocity in the suction space equals zero).
Pfr = loss of pressure for overcoming the friction resistance in the pipes.
Plr = loss of pressure for overcoming local resistances.
Plift = pghlift = expenditure of pressure for lifting the fluid.
Padd = P2-P1 = difference between pressures in the delivery space (P2) and in the
suction space (P2).
1.12. The expenditure of pressure for creating a velocity of flow is:

v2 (1.42)
 Pvel 
2
The drag force FD on objects immersed in a large body of flowing fluid or an object
moving through a stagnant fluid is given as:
C D V 2 A (1.43)
FD 
2
where CD = drag coefficient (Fig. 1.7)

V = velocity (m/s) of the undisturbed fluid
A = protected area (m2) of bluff objects such as spheres,
ellipsoids, disks and plates, cylinders, ellipses, air foils with axes
perpendicular to the flow.
For flat plates placed parallel with the flow:
CD = 1.33
Re 0.5
10 4
 Re  105 x 5  (1.44)
CD =
1.33
Re1 / 7

106  Re  109  (1.45)
The characteristic length in the Reynolds number (Re) is the length of the plate parallel
with flow. For bluff objects, the characteristic length is the larger linear dimension
(diameter of cylinder sphere, disk etc) which is perpendicular to the flow.
1.13 The loss of pressure for overcoming friction resistance in straight

pipe and channels is found as follows:
A. Isothermal Flow
The temperature of the fluid flowing through a pipe is constant.

The equation used for calculation is
L v2 (1.46)
Pfv  
d eq 2
or in the dimensionless form (for a given roughness of the pipe wall).

Eu = CRemSL (1.47)
where  = dimensionless pipe resistance coefficient (its value depends on the

flow conditions and on the roughness of the pipe wall e).
deq= equivalent – diameter, m (for round pipes, deq = d).
L = length of the pipe, m
v = velocity of flow, m/s
 = density of fluid kg/m3

 Pfv
Eu 
v2
Table 1.3: Average values of pipe wall thickness of certain materials

Pipes e, mm
Steel pipes and tubes, seamless and welded, with insignificant 0.2
corrosion
Old corroded steel pipes and tubes 0.67 and

above
Pipes and tubes of roofing steel coated with varnish or drying oil 0.125
Iron water main pipes that have been in service 1.4
Aluminum technical smooth pipes and tubes 0.015 – 0.06
Finished seamless pipes and tubes of brass, copper and lead; glass 0.0015-0.01
pipes and tubes
Concrete pipes; a smooth surface with float work 0.3 -0.8
Concrete pipes; a rough surface 3–9
Petroleum pipes in medium conditions of service and pipes for 0.2

saturated steam
Steam pipes functioning periodically 0.5
Air pipes for compressed air from a compressor 0.8
Condensate pipes functioning periodically 1.0
Re = vd eq 

S1 = L
d eq
Fig. 1.7: The Moody Diagram showing dependence of the pipe resistance coefficient λ on
the Reynolds number Re and the degree of roughness deq/e (Frank, 1999)
The dash line in Fig 1.7 shows the boundary of the so-called self-similar region in
which the pipe resistance coefficient λ does not depend on the Reynolds number Re and
is determined only by the value of the ratio deq/e.
Fig. 1.8: Drag Coefficient versus Reynolds Number for a smooth flat plate parallel to the
flow. (Schlichting, 1979)
Table 1.4: Values of equivalent diameter deq and the coefficient A for selected sections with laminar flow
Shape of section de A Shape of deq A

q section
Circle with d 64 a/b = 0.1 1.81a 85

diameter d
Square with side a a 57 a/b =0.25 1.6a 73
Equilateral 0.58a 53 a/b =0.5 1.3a 62

triangle, with side
a
Ring of width a 2a 96 Ellipse

(a=minor
semi-axis, b=
major semi-
axis):
Rectangle with a/b = 1.55a 78

sides a & b 0.1
a/b ≈ 0 2a 96 a/b = 0.3 1.4a 73
a/b = 0.5 1.3a 68

Fig. 1.9: Flow coefficients for concentric orifices with corner taps (Frank, 1999)
The following equations can also be used to determine resistance coefficient:
I. Laminar flow (Re < 2300)
The pipe resistance coefficient  does not depend on the roughness of the wall of
a pipe, but depends only on Re:
for round pipes,  = 64 (1.48)

Re
A
 
Re
Values of A for sections having various shapes are given in Table 1.4
For isothermal laminar flow of fluids through pipes the loss of pressure due to
friction can also be calculated according to the Hagen – Poiseuille law:
vL (1.49)
Pfr  32
d2
II. Turbulent-flow Re > 2300

For smooth pipes:  = 0 . 316
0 . 25
(for Re < 100,000)
Re
 
e for dimensionless geometrical.
deq
To calculate the pipe resistance coefficient  in rough pipes, the following

equation is used (Nayyar,1999):
1    6.810.9  (1.50)
  2log    
0.5
3.7  Re  
B. Non-Isothermal Flow
In non-isothermal flow, when the liquid flowing through a pipe becomes heated or cooled
(the temperature of the pipe wall differs from that of the liquid), the right-hand sides of
Equations (1.45), (1.46) and (1.48) should be multiplied by the dimensionless correction
factor x.
For laminar conditions:
x =  Pr w 
1
3   Gr Pr 
0 .15
 (1.51)
  1  0 . 22  lq lq  
  Re  
 Pr ld    lq  
for turbulent conditions in hydraulically smooth pipes:

1
 P
x   rw


3
(1.52)
 Pr ld 
Where Reld, Prld, Grld = Reynolds, Prandtl, and Grashof numbers calculated for the
average temperature of the liquid.
Prw = Prandtl number calculated for the liquid at the temperature of the pipe wall.
For gases, the value of the Prandtl number Pv remains constant with a change in the
temperature, and the correction factor x for gases determined by Equation (1.52) equals
unity.
With elevation of the temperature, the values of the Prandtl number Pr decrease for
dropping liquids (Fig. 1.10). Hence the correction factor x determined by Eq. (1.52) is
greater than unity when a liquid becomes cooled (tw < tld) and is less than unity when a
liquid is heated (tw>tld).
Fig.1.10: Values of Prandtl number for liquids (Soo, 1970). Substances for the various
points are shown on the table below.
Substance Point Substance Point Substance Point

No No No
Acetic acid, 100 % 15 Diethyl ether 28 Methyl 20

alcohol,100 %
Acetic acid, 50 % 9 Ethyl acetate 24 Methyl alcohol, 10

40 %
Acetone 25 Ethyl alcohol, 13 Octane 33

100 %
Ammonia, 26 % 14 Ethyl alcohol, 8 Pentane 26

50 %
Amyl acetate 31 Ethyl 29 Sulphuric acid, 1

bromide 111 %
Aniline 5 Ethyl iodide 27 Sulphuric acid, 2

98 %
Benzene 22 Ethylene 36 Sulphuric acid, 4

glycol 60 %
Butyl alcohol 11 Glycerine, 6 Toluene 23

50 %
Carbon disulphide 30 Heptane 32 Water 17
Carbon 18 Hydrochloric 21 Xylene

tetrachloride acid, 30 %
Chlorobenzene 35 Isoamyl 3
alcohol
Chloroform 34 Isopropyl 7
alcohol
Substances for Fig 1.10
1.14 The loss of pressure because of friction in a bent pipe Pbent is

greater than in a straight pipe Pstv by the relation
Pbent = Pstv (1.53)
The dimensionless correction factor  > 1 is calculated by the formula

d (1.54)
  1 3.54
D
where d = internal diameter of the pipe; D = diameter of a turn of the bend.

The loss of pressure for overcoming the resistance of local obstacles (bends, or elbows,
membranes, shut-off fittings, sudden expansions or contractions (reducers), etc] is
determined as follows:
The loss pressure in each local obstacle is the sum of two losses: the loss due to friction
and the additional loss due to the change in direction or the cross-sectional area of the
flow. Since the entire length of the pipe is taken into account when calculating the pipe
friction resistance Pfr in Equation (1.41) is the sum of these additional pressure losses.
V 2
Pfr   2
(1.55)
where  is a dimensionless local resistance coefficient (its values for selected local
obstacles are given in Tables ( Perry, 2008).
Another way of calculating the loss of pressure in a local obstacle is equated to the
hydraulic resistance of a straight pipe of an equivalent length:
Leq V 2 V 2 (1.56)
Pfr    n
d 2 2
where Leq = nd = equivalent length of a straight pipe having the same

hydraulic resistance as the given local obstacles.
n = equivalent resistance (dimensionless) whose values are given in Table 1.5
Table 1.5: Equivalent resistance of screwed fittings and valves
Fittings n Equivalent Fittings
resistance
45o elbows 15 Couplings Negligible
90o elbows (standard- 32 Unions Negligible

radius)
90o square elbows 60 Gate valves (open) 7
180o close return bends 75 Globe valves 300

(open)
Tees (used as elbows, 60 Angle valves 170

entering valves) (open)
Tees (used as elbows, 90

entering branch)
Source: Perry,2008
1.15 For a network consisting of piping, having a constant cross section and
constant velocity of the fluid along the length of the piping, the total hydraulic
resistance of the network Pn according to Equation 1.41 is:
V2  L 
Pn  1    ghlift P2  P1  (1.57)
2  deq 
or using the second way of calculating the losses due to local obstacles:
V 2  L  Leq  (1.58)

Pn  1   ghlift  P2  P1
2  deq 
1.16 The hydraulic resistance of a bank of pipes or tube with the lateral flow of a
stream around them (Fig. 1-11) is found as follows:
square pitch:
3.23
Eu = b(3+4.5m)  S1  Re 0.28 (1.59)
d 
m m
) S2 (b) S2
(a 
S1 S1 S1
  
    
(a) (b)
Figure 1.11: Tube bank arrangements (a) square pitch; (b) triangular pitch
Triangular pitch:
at S1 S
 2
d d
Eu = b(2+3.3m) Re-0.28 (1.60)

at S1 S
 2
d d
Eu = b(2.7 + 1.7m) Re-0.28

(1.61)
In Equations (1.60) and (1.61), the following notation is used:
b = correction factor depending on the angle of attack  (the angle between the
axis of the pipe and the direction of flow), determined from Table 1.6.
m = number of pipe rows in a bank in the direction of the flow
d = external diameter of a pipe
S1 and S2 = lateral and longitudinal pipe pitches.
Table 1.6: Correction factor obtained from angle of attack

, 90 80 70 60 50 40 30 10
deg
b 1 1 0.95 0.83 0.69 0.53 0.38 0.15
The flow velocity is calculated for the narrowest section of a bank AB, the values
of the physicochemical constants are taken for the mean temperature of the flow.
The Re is calculated according to the external diameter of a pipe.
1.17 Hydraulic resistance of Shell-and Tube Heat Exchangers

For the tube space (tube side) of a heat exchanger, and also for the inter tubular
space (shell side) without lateral baffles as shown in Figure 1.12, we have;
Figure 1.12: Heat exchanger (tube-

shell ) without lateral baffles
nL V 2 V 2 (1.62)
P  
d eq 2
  2
where L = length of one pass, m

n = number of passes
The remaining symbols are the same as in the preceding equations. The following
values are adopted for the coefficients of local resistances b in shell-and-tube heat
exchangers:
Tube side
Inlet or outlet heeder 1.5

180o bend between passes or sections 2.5
Tube inlet or outlet 1.0
Shell side
Shell side inlet or outlet 1.5
180o bend through baffle in shell side 1.5
90o bend in shell side 1.0
When the shell side has baffles (Fig. 1.11) the hydraulic resistance in it is
determined by Equations (1.59) – (1.61)
The velocities of a fluid in pipe connections are usually close to those in the tubes
or on the shell side. If the velocity in the pipe connections is greater, then the
losses of pressure .at the inlet to a heat exchanger and at the outlet from it are
calculated according to the velocities in the connections.
* The following formula is also used
3m (1.63)
o 
Re0.2
where m is the number of tube rows in the direction of the flow at the inlet to a
heat exchanger and at the outlet from it are calculated according to the velocities
in the connections.
1.18 Hydraulic Resistance of Scrubber Packings

The resistance of a layer of a dry (non-irrigated) packing with the height H is:
H Vg 
2
Pdry   (1.64)
d eg 2
where  = dimensionless résistance coefficient when a gas passes through a layer

of the packing.
H = height of packing layer, m
deq = equivalent diameter, m
Vg = velocity of the gas through the voids of the packing (actual), m/s
 = density of the gas, kg/m3

The equivalent diameter can be expressed through the characteristics of the

packing its free volume Vfree in m3 of voids per m3 of packed volume (numerically
equal to the unit void area in a column section in m2/m3) and the unit surface area
 m2 of dry packing per m3 of packed volume:
deq = 4V free (1.65)


The actual velocity of a gas Vg is found through the fictitious velocity,  (related
to the total cross section of the apparatus) by the equation:
 (1.66)
Vg 
V free
Soo (1970) had established that for a randomly dumped packing of rings:
at Re g  40 we have  
140 (1.67)
Re g
at Re g  40 we have   16 (1.68)
0.2
Re g
vg deq 4
where Reg   (1.69)
 
The calculations of the hydraulic resistance of irrigated packings are more

complex. These could be found in the section of mass transfer absorption of this
book (Chapter 2).
1.19 Hydraulic resistance of plate column apparatuses

The resistance of a bubble plate P is computed as the sum of three addends.
P = Pdry + P + Pg-lg (1.70)
where Pdry = resistance of a dry plate
P = resistance due to the forces of surface tension
Pg-lg = resistance of the froth layer on a plate

The resistance on a dry plate is:
v h2  g
Pdry =  (1.71)
2
where vh = velocity of the gas in the slots of the cap or the holes of a
plate, m/s
g = density of the gas, kg/m3
 = resistance coefficient equal to:

for bubble-cap plates ------- 4.5 to 5.0
for perforated (sieve) plates
– with a free section of the holes of 7-10 % 1.82
- ditto 11-25 % 1.45
For grid trays 1.4-1.5
The resistance due to the forces of surface tension is
4
P  (1.72)
d eq
where  = surface tension, N/m

deq = equivalent diameter of a hole, m
4A
- for bubble-cap plates deq =
P
(A is the area of the clear cross-section of a slot, P is the perimeter of the
slot)
- for sieve and perforated plates and tray s deq equals the diameter of a hole,
for grid trays – the double width of a slot.
The resistance of the froth layer at the same velocities of the gas that are used in
plate mass-exchange columns is determined as follows: (a) on a bubble-up plate
 e 
Pg lg  1.3K lg  L   h  g (1.73)
 2 
where g = acceleration due to gravity m2/g

k = relative density of the froth layer (foam) in practice it is
assumed that K ~ 0.5
ld = density of the liquid, kg/m3

L = distance from the top edge of the slots to the weir crest, m, (Figure
1.11)
e = height of a slot, m
h = height of the liquid level above the weir crest, m

(b) on a sieve plate
g-lq = 1.3gkld (hw + h) (1.74)

where hw = height of the weir crest, m
The quantity h is determined by the formula for outflow over a weir with account taken
of the density of the froth:
2
 Q v , lq  3
(1.75)
 h   
 1 . 85 Pk 
where Qv-ld = volumetric rate of flow of the liquid, m3/s

P = weir perimeter, m
K = 0.5
o o S
o oo
oo o d
o o
o
Figure 1.13: Schematic view of bubble plate
1.20 Pipeline Hydraulics

Hydraulic analysis of pipeline is performed to determine the possible range of operational
parameters.
For a given pipe size, fluid properties and flow rate the hydraulic analysis should provide
pressure and temperature profiles along the pipeline for steady state and transient
conditions.
Data obtained from hydraulic analysis should be able to address:

 Surge pressure during shut down of a liquid line
 Turn down limitations and inhibition or insulation requirements to prevent wax or
hydrate formation (deposition).
 Effect of flow conditions on the efficiency of corrosion inhibitors.
 Liquid catching and slug control requirements at the downstream of the phase
lines.
1.20.1 Pipe type and usage:

Pipe materials-ductile iron, concrete, steel, fibre- glass, PVC and polyolefin.
Plastic pipe and tubing design: Design pressure for plastic gas piping system or nominal
wall thickness for a given design pressure shall be determined by the equation (ASME
CODE, 2007):
t (1.76)
P  2S   0.32
(D  t)
Where D = specified outside diameter, in

P = design pressure, psig
S = for thermostatic pipe and tubing long term hydrostatic strength
determined in accordance with the listed specification at a
temperature equal to 73 0F, 100 0F, 120 0F, or 140 0F; for re-inforced
thermosetting plastic pipe, 11000 psi is used in practice.
t = specified wall thickness, in
1.20.2 Flow Equations and Flow Regimes
GAS HYDRAULIC CALCULATIONS

Commonly used equations for calculating hydraulic data for gas pipe lines (Campbell,
1992):
1.20.2.1 Panhandle A
0.5394
 2 0.0375G(h2  h1 )(Pavg )2  
1.0778  ( P1 )  ( P2 )2   (1.77)
T  2.6182  
 Tavg Zavg  
Qb  435.87 b  D E 
 Pb  G0.8539LTavg Zavg
 
 
 
1.20.2.2 Panhandle B (McAllister, 1993)

0.51
 2 0.0375G (h2  h1 )( Pavg ) 2  
1.020  ( P1 )  ( P2 ) 2   (1.78)
T   Tavg Z avg  
Qb  737 b  D 2.53 E   0.961 
 b
P

G LT Z
avg avg 
 
 
where Qb = flow rate, SCFD

Pb = base Pressure, Psia
Tb = base Temperature, oR
Tavg = average gas Temperature, oR
P1, P2 = inlet and outlet pressure respectively, psia
G = gas specific gravity (air =1)
L = line length, miles
Z = average gas compressibility
D = pipe inside diameter, in
h2 = elevation of terminus of line, ft
h1 = elevation at origin of line, ft
Pavg = average line Pressure, psia
E = efficiency factor, E = 1.0 for new pipe with no bends, fittings or pipe diameter
changes,
E = 0.95 for very good operating conditions typically through 12-18 months,
E = 0.85 for unfavorable operating conditions;
E = 0.92 for average operating conditions
1.20.3 Weymouth Equation (Boyun and Ghalambor, 2005)

0 .3
 T   ( P ) 2  ( P2 ) 2 
Q  433 .5 b   1  D
2.667
E (1.79)
 Pb   GLTZ 
2 ( P1P2 ) 
Pavg   P1  P2   P2  (1.80)
3 P1 
Pavg is used to calculate gas compressibility factor, Z

where Q = flow rate, SCFD
Tb = base temperature, oR
Pb = base pressure, psia
G = gas specific gravity, (air = 1)
L = line length, miles
T = gas temperature, oR
Z = gas compressibility factor
D = pipe inside diameter, in
E = efficiency factor
1.20.4 Liquid Pipe Lines (ASME CODE)
Internal design pressure is determined by the relation:
St
P2 EF (1.81)
D
where P = internal design pressure, lb/in2 gauge

S = specified minimum yield strength, lb/in2 (can be obtained from table)
t = nominal wall thickness, in
D = nominal outside diameter of the pipe, in
E = weld joint factor (see table)
F = design factor of 0.72
1.20.5 Calculating Pressure Drop

An equation for calculating the pressure drop in flow line is (Gargliadi and Liberatore,
2000):
V P2  P1   2
 fLV2   X (1.82)
gD
1
V 

Or P  Pf (1.83)
  X
 
has been developed which enables one to calculate the pressure drop between any two
points in a liquid system, the said section containing no device to add or remove work.
P = pressure
V = volume or specific volume
v = velocity in conduit
g = gravitational constant
f = frictional factor
 = density
1.20.6 Liquids Hydraulic Calculations

Typical equations used for calculating pressure drop and flow rates for pipe lines.
Allowable working pressure for carbon steel pipe is obtained from the equation
(Campbell, 1992):
t
PD
u c (1.84)
2S  YP
Where t =
wall thickness, in
P =
allowable working pressure
D =
outside diameter, in
S =
allowable stress, psi
Y =
coefficient
u =
under thickness tolerance, in
c =
allowance for threading, mechanical strength, and corrosion, in
from tables, Y = 0.4
u = 0.125 x t (u 12.5 % of t)
c = 0.050-inch
S-Values are:
Spec API 5L GRADE A- 25,000 psi
Spec API 5L GRADE B – 29,750 psi
Spec API 5LX GRADE X – 42,000- 35,700 psi
S t  0.057 (1.85)
P  1.75
D  0.04  0.7t
1.20.7 Pressure Drop Calculations for Rubber Hose

The general flow equation for rubber hose is (Crocker, 2000):
0.0134fSLQ2 (1.86)
P 
d5
where ∆P = pressure drop, psi

F = friction factor
S = specific gravity
L = length of line, ft
Q = flow, gpm
d = inside diameter of line, in

3160 Q (1.87)
NR 
Vc  d
where NR = Reynolds Number

Q = flow, gpm
Vc = fluid viscosity, centistokes
d = internal diameter of line, in.
Friction Factor, f is determined by the relation:
For laminar flow, f  64 (1.88)
NR
For turbulent flow, we refer to relevant friction factor and Reynolds Number charts (see
Fig 1.7).
1.20.8 Typical equations Used for Calculating Pressure Drop and Flow Rates for
Pipe Lines
(a) For crude oil, Shell/MIT equation is used (Boyun and Ghalambor, 2005):
P
0.241 f  s  Q 2
(1.89)
d5
For viscous flow; f = 0.00207 x (1/r) and for turbulent flow; f = 0.0018 + 0.00662
x(1/r)0.355
where r = 0.0119 x Q/dv
(b) For gasoline, Benjamin Miller equation is used:
B = (0.1692)(d5P/S)0.5(log(d3SP/z2) + 4.35) (1.90)
(c) For gasoline, T. R. Aude equations are also used:
B = 0.871 K (1/z0.104)((P0.552d2.656))/S0.448) (1.91)
P = ((Bz0.104S0.448)/(0.871Kd2.656))1.812 (1.92)
(d) For water system hydraulics, Hazen- Williams is used (Crocker, 2000):
Q = (0.148 d2.63P0.54C)/S0.54 (1.93)
where Q = barrels per day

B = barrels per hour
P = Pressure loss, psi per mile

d = inside diameter of pipe, in
z = absolute viscosity, centipoises
S = specific gravity
K = efficiency factor (Aude’s formula)
F = friction factor
r = Reynolds Number modified, Re/7742
v = kinematic viscosity, centistokes
C = Hazen-Williams Coefficient (120 for new steel pipe; 100 for ordinary 15
year old steel; and 150 for fibre glass pipe)
1.20.9 Equivalent Stresses

The wall thickness may be derived from hoop stress considerations based on design
factors; should be such that the longitudinal shear and equivalent stresses in the pipe wall
under functional and environmental loads DO NOT exceed certain values.
ANSI/ASME B31.4, ANSI/ASME B31.8

Two types of load considered as defined in ANSI/ASME B31.8
Functional loads (Article AB 41.32)
Environmental load (Article AB 41.33)
Equivalent stress is defined as:

Seq = (Sh 2 + SL2-SLSh +3SS2)1/2 (Von Mises Equation) (1.94)
where Seq = equivalent stress

Sh = hoop stress (due to pressure)
SL = longitudinal stress (due to pressure, thermal expansion and bending)
S = combined shear stress (due to torque and shear force).
1.20.10 Hoop Stress is defined mathematically as:

PD (1.95)
Sh 
2t
Longitudinal Stress:
Fully restrained pipeline
SL = µ (Sh-P)- Eα(T2 –T1) (1.96)

Fully unrestrained pipeline
SL 
Sh Mb
 (1.97)
2 Z
Combined Shear Stress:

T 2 Fs (1.98)
Ss  
2Z A
Where,
P = pipe line internal pressure
D = pipe line internal diameter
t = wall thickness
µ = Poisson’s ratio
M = Modulus of elasticity
α = Linear coefficient of thermal expansion
T1 = Pipe line installation temperature
T2 = Pipe line operation temperature
Mb = bending moment applied to the pipe line
Z = pipe section modulus
T = Torque applied to the pipe line
Fs = Shear force applied to the pipe line
A = pipe wall cross section area
1.20.11 Temperature Profile

Prediction of temperature distribution in the flowing system may be as important as the
pressure profile. It affects many design considerations surrounding the line.
Crude oil pour points, gas hydrates, vapor-liquid phase behavior and water content
of gas are all temperature sensitive. Therefore a reliable temperature prediction at the
early step in the calculation procedure is imminent.
The temperature at any point is predictable through the use of First Law of
Thermodynamic and energy balance. If potential and kinetic energy changes are ignored;
and work equals zero in the pipe section involved- this first law balance for a steady state
system reduces to:
∆H = Q (1.99)
The enthalpy of a gas increases with decreasing pressure; it is almost independent of

pressure for a liquid. So, the enthalpy at any point in the line depends both on pressure at
that point and the heat energy lost or gained through the pipe wall.
From heat transfer principle,

Q = UA∆Tm (1.100)
For circular pipe, the equation becomes:

Q = (U)(  d L)(∆Tm) (1.101)
Where: U = overall heat transfer coefficient


= 3.1416
d = pipe diameter
L = pipe section length = L2-L1
T1 T2
T1
ln
T
∆Tm =
log mean ∆T = 2
Q = heat lost in line section of length L

∆T1 = T1-Tg
∆T2 = T2-Tg
Tg = normal ground or water temperature at that point (far enough away
from pipe to be undisturbed by it).
If phase changes occur in the line section, the following equation applies.
  V  
P
T2
2
h   mCpdT   V  T   dP (1.102)

T1
P1   T P 
Equation (1.102) relates enthalpy to specific heat (CP) and P, V and T. The second term is
zero, or essentially so, for an ideal gas and a liquid. For a gas the equation shows that the
enthalpy increases as the pressure declines.
If the pressure drop in line segment (dP) is small compared to the absolute pressure,
the second term of Equation (1.102) is small compared to the first term. For a gas line at a
pressure above 3.5 MPa (500 psi), ignoring this second term may not be important. There
are thus three basic solutions for the ∆h equation.
 Use an enthalpy-PVT equation of state program

 Approximate the second term by assuming a series of Joule-Thomson
expansions.
 Ignore the second term completely.
One can approximate Method (2) by translating the change of enthalpy with pressure into
temperature change per unit length. Then one can obtain Equation (1.103):
  J 
 (T1  JL1 )   Tg  a   (1.103)
ln    aL
 T  JL    T  J  
 2  
a  
2 g
 
Metric English
Where: Cp = heat capacity KJ/kg,oC Btu/lb,oF
a = du/B q ᵧ Cp
L1 = distance from initial point m ft
L2 = distance from final point m ft
L = L2 –L1 m ft
o o
T1 = temperature at L1 C F
o o
T2 = temperature at L2 C F
P = gas sp gr (air=1.0) - -
o o
Tg = ground or water temperature C F
o o
J = Joule-Thomson coefficient C/m F/ft
U = overall heat transfer coefficient, kJ/h.m2.oC Btu/hr.ft2,oF
q = gas flow rate, std volumes 103m3/h Mscf/h
B = constant 408 24.4
Considering the Joule-Thomson effect, it is possible for the gas temperature to be less
than that of the ground. This is likely to occur with long lines possessing a large pressure
drop or at regulator stations. For lean, pipeline-quality natural gases the cooling due to
Joule-Thomson expansion is approximately 0.004 to 0.005 oC/kPa at pipelines operating
near 70 bar.
Then the quantity ‘’J’’ is zero and Equation (1.103) becomes
 T T
ln 1 g

  d U  L (1.104)
T T  B  q  C
 2 g  p
Equation (1.104) is simple to solve.
1.20.12 Flow Efficiency

For gas flow problems as encountered in oil field operations, the FANNING EQUATION
for pressure drop may be used. A modified form of this equation employing units
commonly used in oil field practice is:
F  L  MSFD   STZ
2
P (1.105)
20000  d 5  Pav
and for determining of the friction factor F;
20.14MSFDS (1.106)
Re 
d 
where P = pressure drop, (psig)

F = friction factor, dimensionless
L = length of pipe, ft
MSFD = gas flow at standard condition
S = specific gravity of gas
T = absolute temperature (oF +460)
Pav = average flowing pressure, psia
Re = Reynold’s Number, dimensionless
µ = viscosity, centipoises
In actual practice, empirical flow formulas are used by many to solve the gas problems of
field production operations.
The Weymouth formula is the one most frequently used since results obtained by its
use agree quite closely with actual values. Recent modification of the formula by
including Compressibility Factor, Z made the formula applicable for calculation of high
pressure flow problems.
The modified equation is as follows (Campbell, 1992):

Ts  P 2  P2 2  (1.107)
Qs  433.45   d 2.667   1 
Ps  LSTZ 
where Qs = rate of flow of gas in ft3/24 hrs measured at standard conditions
D = internal diameter of pipe, in
P1 = initial pressure, psi absolute
P2 = terminal pressure, psi absolute
L = length of lines in miles
S = specific gravity of flowing gas (air =1.0)
T = absolute temperature of flowing gas
Ts = standard absolute temperature
Ps = standard pressure, psia
Z = Compressibility factor
Values of depression d2.667 for different inside diameter of pipe are given in tables.
Flow efficiency is determined by:
v
 f  actual (1.108)
vdesign
f
where = flow efficiency
Vactual = actual flow velocity
Vdesign = design flow velocity
NOTE: Presently, SOFTWARE is being used for simulation of flow efficiency in gas
and liquid flow systems.
1.21 PV Elite Vessel Analysis Program: Input Data

Pressure vessels are designed with the support of PV Elte software. A typical input data is
presented here.
Design Internal Pressure (for Hydrotest) 230.00 psig
Design Internal Temperature 450 F
Type of Hydrotest not Specified
Hydrotest Position Horizontal
Projection of Nozzle from Vessel Top 0.0000 in.
Projection of Nozzle from Vessel Bottom 0.0000 in.
Minimum Design Metal Temperature -20 F
Type of Construction Welded
Special Service None
Degree of Radiography RT 1
Miscellaneous Weight Percent 0.
Use Higher Longitudinal Stresses (Flag) Y
Select t for Internal Pressure (Flag) Y
Select t for External Pressure (Flag) Y
Select t for Axial Stress (Flag) N
Select Location for Stiff. Rings (Flag) N
Use Hydrotest Allowable Unmodified Y
Consider Vortex Shedding N
Perform a Corroded Hydrotest N
Is this a Heat Exchanger No
User Defined Hydro. Press. (Used if > 0) 0.0000 psig
User defined MAWP 0.0000 psig
User defined MAPnc 0.0000 psig
Load Case NP+EW+WI+FW+BW

Load Case 2 NP+EW+EE+FS+BS
Load Case 3 NP+OW+WI+FW+BW
Load Case 4 NP+OW+EQ+FS+BS
Load Case 5 NP+HW+HI
Load Case 6 NP+HW+HE
Load Case 7 IP+OW+WI+FW+BW
Load Case 8 IP+OW+EQ+FS+BS
Load Case 9 EP+OW+WI+FW+BW
Load Case 10 EP+OW+EQ+FS+BS
Load Case 11 HP+HW+HI
Load Case 12 HP+HW+HE
Load Case 13 IP+WE+EW
Load Case 14 IP+WF+CW
Load Case 15 IP+VO+OW
Load Case 16 IP+VE+OW
Load Case 17 IP+VF+CW
Load Case 18 FS+BS+IP+OW
Load Case 19 FS+BS+EP+OW
Wind Design Code ASCE-7 93
Design Wind Speed 70.000 mile/hr
Exposure Constant C
Importance Factor 1.
Roughness Factor 1
Base Elevation 0.0000 ft.
Percent Wind for Hydrotest 33.

Use Wind Profile (Y/N) N
Damping Factor (Beta) for Wind (Ope) 0.0100
Damping Factor (Beta) for Wind (Empty) 0.0000
Damping Factor (Beta) for Wind (Filled) 0.0000
Seismic Design Code UBC 94

UBC Seismic Zone (1=1,2=2a,3=2b,4=3,5=4) 0.000
UBC Importance Factor 1.000
UBC Soil Type S1
UBC Horizontal Force Factor 3.000
UBC Percent Seismic for Hydrotest 0.000
Consider MAP New and Cold in Noz. Design N
Consider External Loads for Nozzle Des. Y
Consider Code Case 2168 for Nozzle Des. N
Material Database Year Current w/Addenda or Code Year
Complete Listing of Vessel Elements and Details:

Element From Node 10
Element To Node 20
Element Type Elliptical
Description Head1
Distance "FROM" to "TO" 0.1600 ft.
Inside Diameter 144.00 in.
Element Thickness 1.0000 in.
Internal Corrosion Allowance 0.0000 in.

Nominal Thickness 0.0000 in.
External Corrosion Allowance 0.1250 in.
Design Internal Pressure 230.00 psig
Design Temperature Internal Pressure 450 F
Design External Pressure 15.000 psig
Design Temperature External Pressure 100 F
Effective Diameter Multiplier 1.2
Material Name SA-516 70
Allowable Stress, Ambient 20000. psi
Allowable Stress, Operating 20000. psi
Allowable Stress, Hydrotest 26000. psi
Material Density 0.2830 lb./cu.in.
P Number Thickness 1.2500 in.
Yield Stress, Operating 31750. psi
UCS-66 Chart Curve Designation B
External Pressure Chart Name CS-2
UNS Number K02700
Product Form Plate
Efficiency, Longitudinal Seam 1.
Efficiency, Circumferential Seam 1.
Elliptical Head Factor 2.
Detail Type Nozzle
Detail ID Noz N1 Fr10
Dist. from "FROM" Node / Offset dist 0.0000 in.

Nozzle Diameter 20. in.
Nozzle Schedule None
Nozzle Class 300
Layout Angle 0.
Blind Flange (Y/N) N
Weight of Nozzle ( Used if > 0 ) 403.75 lb.
Grade of Attached Flange GR 1.1
Nozzle Matl SA-516 70
Element To Node 30
Element Type Cylinder
Description SHELL

Detail Type Saddle
Detail ID Lft Sdl
Dist. from "FROM" Node / Offset dist 2.6600 ft.
Width of Saddle 10.000 in.
Height of Saddle at Bottom 108.00 in.
Saddle Contact Angle 120.
Height of Composite Ring Stiffener 23.000 in.
Width of Wear Plate 12.000 in.
Thickness of Wear Plate 0.7500 in.
Contact Angle, Wear Plate (degrees) 120.
Detail Type Saddle
Detail ID New Sdl
Width of Saddle 10.000 in.
Height of Saddle at Bottom 108.00 in.
Saddle Contact Angle 120.
Height of Composite Ring Stiffener 23.000 in.
Width of Wear Plate 12.000 in.
Thickness of Wear Plate 0.7500 in.
Contact Angle, Wear Plate (degrees) 120.

Detail Type Nozzle
Nozzle Schedule 80
Nozzle Class 300
Layout Angle 0.
Nozzle Matl SA-106 B
Detail Type Nozzle
Nozzle Schedule 80
Nozzle Class 300
Layout Angle 180.
Grade of Attached Flange GR 3.5
Nozzle Matl SA-106 B
Element To Node 40
Element Type Elliptical

Description
Elliptical Head Factor 2.
PV Elite 2008 ©1993-2008 by COADE Engineering Software
Fig. 1.14: Design of Pressure vessel (incomplete) – Software Application
Fig. 1.15: Design of Pressure Vessel (complete)- Software Application

EXAMPLES
Example 1.1: The flue gases from a typical furnace installation in process plant are
discharged to the atmosphere through a chimney 19 m high. The composition of the gases
is CO2 -12.7%, O2 -4.9%, N2 -77.5 %, H2O – 4.9% (by volume). The cross sectional area
of the chimney equals that of the horizontal flue. The average temperature of the gases in
the flue and the chimney is 250oC. Find the velocity of the gases if the sum of the
resistance coefficients of the flue and chimney is:
Lfl H
    27.3
deq d
Where λ = pipe resistance coefficient

Lfl = length of flue
deq = equivalent diameter of the flue
d = internal diameter of the chimney
H = height of chimney
  sum of the local resistance coefficient
Solution: We determine the density of the gas in standard conditions with the relation -
M 1
ρo = = (44 x0.127 + 32 x0.049 + 28 x 0.778 + 18 x 0.049)
22.4 22.4
= 1.328 kg/m3
The density of the gas at 250 oC is:
TO 273
o =
ρfl = ρ T 1.328 x = 0.693 kg/m3
273  250
Let us take the average temperature of the air in Port Harcourt to be 28o C.
The density of air at this temperature is
273
ρ air = 1.293 x = 1.173 kg/m3
273  28
Considering Bernoulli equation with the reference plane at the level of the flue
v2 v2
PbarI   PbarII    fl gH  PI
2 2
Where Pbar1 = barometric pressure at the level of the horizontal flue
Pbar 11 = barometric pressure at the top edge of the chimney

V = velocity of the gas
H = height of the chimney
P
1 = total pressure losses (total resistance of the flue and the chimney
The left-hand side of the Bernoulli equation considers a section drawn through the
beginning of the flue, and the right-hand side – through the top edge of the chimney.
The barometric pressure at the level of the horizontal flue is:
Pbar1 = Pbar 11 + ρair g H
Inserting this value into the Bernoulli equation, we obtain
Pbar 11 + ρair g H + v = Pbar 11 + v 2

2
+ ρfl g H + ∆P1
2 2
whence
P
1 = g H( ρair – ρfl) (a)
P
The resistance of the flue and the chimney, 1 can be calculated by the equation:
2
 v
P    
L H

fl
 
1 d d 2
 eq 
or with a view to the initial conditions of the example:
2
 v
fl
P  27.3
1 2 (b)
From equations (a) and (b), we obtain

2
 v
fl
27.3 
2 = g H( ρair – ρfl)
Introducing numerical values yields;
2
0.693 v
27.3   ( 1.173  0.693 )  9.81  19
2
( 1.173  0.693)  9.81 19

 2  3.075
V= 27.3 0.693 = 3.075 m/s
Example 1.2: The reading of the water differential manometer of a Pitot-Prandtl tube
(Fig. 1.4) installed along the axis of a horizontal air pipe is 13 mm. determine the rate of
flow of the air if its temperature is 40 oC, the pipe diameter is 159 x 6 mm, and there is
straight section 7 m long before the tube. The air is at atmospheric pressure.
Solution: The density of the air at 40 oC is:
ρ air = 1.293 x 273 = 1.13 kg/m3
273  40
The maximum axial velocity of the air is determined from the relation:
Vmax = 2  Pvel = 2 x13 x 9 .81 = 15.07 m/s

 1 .13
The following value of Reynolds number corresponds to this velocity:
V max d  15 . 07 x 0 . 147 x1 . 13
Re = = = 132 000
 0 . 019 x10  3
The length of the straight section of flow stabilization before the Pitot tube should be at
least 40 diameters: 40 x 0.147 = 5.9 m
The condition is observed because there is a straight pipe section 7 m long. From the
graph or plot v/vmax against Re (Fig 1.2) for Re = 132000, the ratio of the average
velocity to the maximum one v/vmax ≈ 0.85.
Hence the average velocity is v =0.85 x 15.07 =12.8 m/s
The rate of flow of air becomes:

Qv = 0.785 x 0.1472 x 12.8 =0.217 m3/s = 780 m3/h
Example 1.3: To study a certain production process experimentally in laboratory
conditions a geometrical model of the industrial apparatus was fabricated to a scale of 1
in 10. In the industrial apparatus, the working substance is hot air (100 oC, atmospheric
pressure) flowing with a velocity of 3 m/s. In the laboratory model it is intended to use air
at atmospheric pressure having a temperature of 22 oC as the working substance. Is it
possible in these conditions to obtain complete hydrodynamical similarity of the
industrial apparatus and the model, and what must the velocity of the air in the model be?
Solution: To observe hydro-dynamical similarity, with similar boundary conditions, the
Reynolds and the Froude numbers for the model and the industrial apparatus must be
equal (the Euler number in this case is not a governing factor).
i.e we must have Re1 = Re2
v1 L1 P1 v 2 L 2 P2
i.e = (i)
1 2
2 2
v1 v2
Fr1 = Fr2 ; i.e = (ii)
gL 1
gL 2
In (i) and (ii) subscript ‘1’ relates to the industrial apparatus and subscript ‘2’ to the
model.
We have, v1 = 3 m/s; L2 = 0.1 L1
273
ρ1 = 1.29 x = 0.945 kg/m3
373
273
ρ2 = 1.29 x = 1.19 kg/m3
295
µ1 = 0.0215 c P; and µ2 = 0.0185 c P
Using these values in eqn (i) we obtain
3 L1  0.945 v2  0.1 L1 1.19


0.0215 0.0185
3  0.945  0.0185
V2 =
0.0215  0.1  1.19
= 20.5 m/s
Substituting in eqn (ii) yields
2
32 v
 2
L1 0 .1L1
V2 = 0.9 = 0.946 m/s
The results show that the observance of complete hydro-dynamical similarity i.e the
simultaneous similarity of the resistance forces and the gravity forces in the model and
the industrial apparatus in the given conditions is impossible. We will have to limit
ourselves to approximate modeling of the process as is done in practice in the majority of
cases, i.e observe only one condition, (either Re = idem or Fr = idem) depending on
which of these conditions is more important.
If for example, the friction forces are more important, i.e the Reynolds number, Re then
the velocity of the air in the model should be taken equal to 20.5 m/s.
Let us find the conditions in which it is possible to obtain simultaneous similarity of the
friction forces and gravity forces in the model and the industrial apparatus.
v1 L1 v 2 L2
From the conditions; 
1 2
Where  is the kinematic viscosity, it follows that
v1 v1L 2

 2  2 L1
From the condition that Fr = idem,

2 2
v1 v2
=
gL 1
gL 2
v1 L1

v2 L2
Both conditions will be observed if

v
v 1 L 2 = L 1 i.e 1
= 3
L 1
v 2 L 1 L 2
v 2 L 2
This last expression shows the required condition of the simultaneous similarity of the
friction forces and gravity forces.
1.8 References

1 ASME CODE SECTION VIII, DIVISION 1: Rules for construction of Pressure
vessels, 2007 CASTI Publishing Inc. Canada.
2 Boyun, G. and Ghalambor, A. (2005): Natural Gas Engineering Handbook. Gulf
Publishing Company, Texas, 77046, USA
3 Campbell, J. M. (1992): Elements of Field Processing.
4 Crocker, S.(2000): Piping Hand Book-Generic Design Considerations; Part B.
5 Darby, R. (2001): Chemical Engineering Fluid Mechanics, Marcel Dekker, Inc.
New York, Second Edition
6 Durgaiah, D. R. (2004): Fliud Mechanics and Machinery. New Age Int. Publishers,
Daryaganj New Delhi, India
7 Frank K. (1999) Mechanical Engineering Handbook
8 Gargliadi, M. G. and Liberatore, L. J. (2000):: Piping Hand Book,-Piping Systems
Part C. McGraw- Hill
9 Kafarov, V. V., Perov, V. L. and Meshalin, V. P. (1974): ‘Principles of the
Mathematical Modeling of Chemical Engineering Systems’, Moscow, Khimiya
10 Licker, M. D. (2003): Dictionary of Engineering, McGraw Hill, 2nd Edition
11 Maloney, J. O (2008): Perry’s Chemical Engineers Handbook 8th Edition. 2008 Mc

Graw –Hill.
12 McAllister, E. W. (1993): Pipeline Handbook. 3rd Edition. Gulf Publishing

Company, Houston, Texas.
13 Nayyar. M. L. (1999): Piping Hand Book- Piping Fundamentals Part A. Betchel
Corp.
14 Pavlov, K. F., Romankov, P. G.and Noskov, A. A (1979): Examples and Problems

to the Course of Unit Operations of Chemical Engineering, Mir Publishers,
Moscow
15 Planovsky, A. N. and Kafarov, V. V. (1944): Khim. Prom, 8, 19
16 PV Elite 2008 ©1993-2008 by COADE Engineering Software
17 Schlichting, H.(1979): ‘Boundary Layer Theory’ 7th ed. McGraw-Hill, New York.
18 Soo, S. L. (1970):Fluid Dynamics of Multiphase Systems. Waltham. Mass.,
Blalsdell Publishing Co.
19 Souder, M. J. (1938): Am. Chem. Soc., 60, 154.
20 Spink, L. K. (1975): (Principles and Practice of Flow Meter Engineering. 9th
Edition. Page 24. Published by Foxboro Company)
21 Weinspach, P. M. (1969): Chem. Ing. Techn., 41, 5-6: 260-265

Applied Pipeline Hydraulics

Uploaded by

Copyright:

Available Formats

Applied Pipeline Hydraulics

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Applied Pipeline Hydraulics

Uploaded by

Copyright:

Available Formats

1.1 Commonly Used Terms/Parameters

where g = 9.81 m/s2 is the acceleration due to gravity

where x1 , x 2  mass fractions of the mixture components

1 ,  2  densities of the mixture and its components

Similarly, the density of a suspension is calculated by the following:

where x = mass fraction of the solid phase in the suspension

 c and ld  densities of the solid (crystalline) and liquid phases respectively

mix  11  22  ... (1.6)

where 1 ,  2  volume fractions of the components of the gas mixture

1 ,  2  respective densities of the components

1 atm (technical atmosphere) = 1 kgf/cm2 = 104 kgf/m2

1.2 The fundamental equation of hydrostatics is expressed as:

Where PO = pressure on the surface of the liquid, Pa

g = acceleration due to gravity, = 9.81 m/s2

in the M K (force) S System [  ] = kgf 2.s

in the CGS system [  ] = dyn .2 s = g = P (poise)

the relationship between the customarily used viscosity units is:

1 cP = 10-3 Pa.s = 1 kgf . s

Viscosity-gravity constant is an index of chemical composition of crude oil; defined as

The values of  for liquids at different temperatures can be determined by means of

 = density of the liquid, kg/m3

M = molar mass, kg/kg mol

It is convenient to use a nomogram to determine  on the basis of Equation (1.11). The

1.3 To determine the dynamic viscosity of a gas at different temperatures, the

Conversion to SI: 1 cP = 1 x 10-3 Pa.s

mix , 1 , 2 ….. = corresponding dynamic viscosities

1,2 ,.... = volume fractions of the components in the mixture

( 11 )  M1Tcr1  ( 2 2 )  M 2Tcr 2  ... (1.13)

where  0  dynamic viscosity at 0 OC

For a mixture of normal (non- associated) liquids, the value of

mix, can be calculated by the expression:

log mix  x1 log1  x2 log2  ... (1.15)

where 1 ,  2 ,  dynamic viscosities of the separate components

x1, x2, .= mole fractions of the components in the mixture

 susp = ld (1  2.5) (1.17)

When the volume concentration of the solid phase is up to 30 % Weinspach (1969)

 susp = ld 0.59

where  ld  dynamic viscosity of the pure liquid

1.4 Equation of rate of flow

where A = cross-sectional area of the flow, m2

 = density of the fluid, kg/m3

QV = 0.785 d2v (1.21)

1. A completely submerged body experiences a vertical upward force equal to the

1.5 Principal dimensionless numbers for flow through pipes and

where v = mean velocity of the flow, m/s

 = dynamic viscosity, Pa. s

 = kinematic viscosity, m2/s

Laminar flow Re  2000

Transition flow 2000  Re  3000

0.001 0.002 0.003 0.004 0.005 0.006 0.007

Fig. 1.2: Dependence of Recr in coils on the ratio d/D

Developed turbulent flow Re  3000

where v is the velocity of the flow, m/s

For a viscous (real) incompressible liquid, the following expression is applied:

where z = geometrical (position ) head, m

v2 = velocity (dynamic) head, m

h1 = hydraulic loss head, m

1.6 The velocity of a discharge of a liquid v (m/s) from a small orifice in

where  = dimensionless velocity coefficient

g = acceleration due to gravity, m/s2