Technical information

Selecting Centrifugal Pumps

Copyright by
KSB Aktiengesellschaft
Published by:
KSB Aktiengesellschaft,
Communications (V5),
67225 Frankenthal / Germany
All rights reserved. No part of
this publication may be used,
reproduced, stored in or intro
duced in any kind of retrieval
system or transmitted, in any
form or by any means (electro
nic, mechanical, photocopying,
recording or otherwise) without
the prior written permission of
the publisher.
4
th
completely revised and ex
panded edition 005
Layout, drawings and
composition:
KSB Aktiengesellschaft,
Media Production V5

ISBN 0007844
4
Contents
1 Nomenclature ..................................................................6
2 Pump Types ................................................................89
3 Selection for Pumping Water ..........................................0
3.1 Pump Data ...............................................................................0
.. Pump Flow Rate ......................................................................0
.. Developed Head and Developed Pressure of the Pump .............0
.. Effciency and Input Power ......................................................0
..4 Speed of Rotation ....................................................................
..5 Specifc Speed and Impeller Type ..............................................
..6 Pump Characteristic Curves .....................................................
3.2 System Data .............................................................................6
.. System Head ...........................................................................6
... Bernoullis Equation .................................................................6
... Pressure Loss Due to Flow Resistances .....................................8
.... Head Loss in Straight Pipes ......................................................8
.... Head Loss in Valves and Fittings ..............................................
.. System Characteristic Curve .....................................................6
3.3 Pump Selection .........................................................................8
.. Hydraulic Aspects ....................................................................8
.. Mechanical Aspects ..................................................................9
.. Motor Selection .......................................................................9
... Determining Motor Power .......................................................9
... Motors for Sealless Pumps ......................................................
... Starting Characteristics ............................................................
3.4 Pump Performance and Control ...............................................4
.4. Operating Point .......................................................................4
.4. Flow Control by Throttling ......................................................4
.4. Variable Speed Flow Contol .....................................................5
.4.4 Parallel Operation of Centrifugal Pumps ..................................6
.4.5 Series Operation .......................................................................8
.4.6 Turning Down Impellers ..........................................................8
.4.7 Underfling of Impeller Vanes ..................................................9
.4.8 Preswirl Control of the Flow ...................................................9
.4.9 Flow Rate Control or Change by Blade Pitch Adjustment ........9
.4.0 Flow Control Using a Bypass ...................................................40
3.5 Suction and Inlet Conditions ....................................................4
.5. The NPSH Value of the System: NPSH
a
..................................4
.5.. NPSH
a
for Suction Lift Operation ...........................................4
.5.. NPSH
a
for Suction Head Operation .........................................44
.5. The NPSH Value of the Pump: NPSH
r
.....................................44
.5. Corrective Measures ................................................................45
3.6 Effect of Entrained Solids .........................................................47
4 Special Issues when Pumping Viscous Fluids ..................48
4.1 The Shear Curve ......................................................................48
4.2 Newtonian Fluids .....................................................................50
4.. Infuence on the Pump Characteristics ......................................50
4.. Infuence on the System Characteristics ....................................54
4.3 Non-Newtonian Fluids ............................................................54
4.. Infuence on the Pump Characteristics ......................................54
4.. Infuence on the System Characteristics ....................................55
Table of contents
5
5 Special Issues when Pumping Gas-laden Fluids ..............56
6 Special Issues When Pumping Solids-laden Fluids ..........57
6. Settling Speed ...........................................................................57
6. Infuence on the Pump Characteristics ......................................58
6. Infuence on the System Characteristics ....................................59
6.4 Operating Performance ............................................................59
6.5 Stringy, Fibrous Solids ..............................................................59
7 The Periphery ................................................................6
7. Pump Installation Arrangements ..............................................6
7. Pump Intake Structures ............................................................6
7.. Pump Sump ..............................................................................6
7.. Suction Piping ..........................................................................6
7.. Intake Structures for Tubular Casing Pumps ............................64
7..4 Priming Devices .......................................................................65
7. Arrangement of Measurement Points .......................................67
7.4 Shaft Couplings ........................................................................68
7.5 Pump Nozzle Loading ..............................................................69
7.6 National and International Standards and Codes .....................69
8 Calculation Examples
(for all equations numbered in bold typeface) ................7
9 Additional Literature .....................................................79
10 Technical Annex (Tables, Diagrams, Charts) .................80
Tab. : Centrifgal pump classifcation ...................................................8
Tab. : Reference speeds of rotation ....................................................
Tab. : Approximate average roughness height k for pipes ..................0
Tab. 4: Inside diameter d and wall thickness s in mm and weight of
typical commercial steel pipes and their water content ...........0
Tab. 5: Loss coeffcients for various types of valves and fttings ........
Tab. 6: Loss coeffcients in elbows and bends ....................................4
Tab. 7: Loss coeffcients for fttings ..............................................4/5
Tab. 8: Loss coeffcients for adapters .................................................5
Tab. 9: Types of enclosure for electric motors to EN 60 59 and
DIN/VDE 050, Part 5 ............................................................0
Tab. 0: Permissible frequency of starts Z per hour for electric motors ..0
Tab. : Starting methods for asynchronous motors ..............................
Tab. : Vapour pressure, density and kinematic viscosity of water at
saturation conditions as a function of the temperature .............4
Tab. : Infuence of the altitude above mean sea level on the annual
average atmospheric pressure and on the corresponding
boiling point .........................................................................4
Tab. 4: Minimum values for undisturbed straight lengths of piping
at measurement points in multiples of the pipe diameter D ......67
Contents
Tables
6
A m

Area
A m Distance between measuring point and pump
fange
a m, mm Width of a rectangular elbow
B m, mm Vertical distance from suction pipe to foor
C
v
gpm Flow coeffcient for valves, defned as the fow
of water at 60 F in US gallons/minute at a
pressure drop of lb/in

across the valve

c
D
Resistance coeffcient of a sphere in water fow
c
T
(%) Solids content in the fow
D m (mm) Outside diameter; maximum diameter
DN (mm) Nominal diameter
d m (mm) Inside diameter; minimum diameter
d
s
m (mm) Grain size of solids
d
50
m (mm) Mean grain size of solids
F N Force
f Throttling coeffcient of an orifce
f
H
Conversion factor for head (KSB system)
f
Q
Conversion factor for fow rate (KSB system)
f

Conversion factor for effciency (KSB system)

g m/s

Gravitational constant = 9.8 m/s

H m Head; discharge head

H
geo
m Geodetic head
H
s
m Suction lift
H
s geo
m Vertical distance between water level and pump
reference plane for suction lift operation
H
z geo
m Vertical distance between pump reference plane
and water level for positive inlet pressure
operation
H
L
m Head loss
H
0
m Shutoff head (at Q = 0)
I A Electric current (amperage)
K Dimensionless specifc speed, type number
k mm, m Mean absolute roughness
k Conversion factors k
Q
, k
H
, k

(HI method)
k
v
m

/h Metric fow factor for valves, defned as the

fow of water at 0 C in cubic metres per hour
at a pressure drop of bar
L m Length of pipe
L
s
m Straight length of airflled pipe
M Nm Moment
NPSH
r
m NPSH required by the pump
NPSH
a
m NPSH available
N
s
Specifc speed in US units
n min

(rpm) Speed of rotation

s

(rev/s)
n
q
min

Specifc speed in metric units

P kW (W) Power; input power
1
1
Nomenclature
Nomenclature
7
p
e
Pressure in suction or inlet tank
PN (bar) Nominal pressure
p bar (Pa) Pressure rise in the pump; pressure differential
(Pa N/m

)
p bar (Pa) Pressure (Pa N/m

= 0
5
bar)
p
b
mbar (Pa) Atmospheric pressure (barometric)
p
L
bar (Pa) Pressure loss
p
v
bar (Pa) Vapour pressure of fuid pumped
Q m

/s, m

/h Flow rate / capacity (also in litre/s)

q
air
% Air or gas content in the fuid pumped
Q
off
m

/h Flow rate at switchoff pressure

Q
on
m

/h Flow rate at startup pressure

R m (mm) Radius
Re Reynolds number
S m Submergence (fuid level above pump);
immersion depth
s mm Wall thickness
s m Difference of height between centre of pump im
peller inlet and centre of pump suction nozzle
T Nm Torque
t C Temperature
U m Length of undisturbed fow
U m Wetted perimeter of a fow section
V
B
m

Suction tank volume

V
N
m

Useful volume of pump sump

v m/s Flow velocity
w m/s Settling velocity of solids
y mm Travel of gate valve; distance to wall
Z /h Switching cycle (frequency of starts)
z Number of stages
z
s,d
m Height difference between pump discharge and
suction nozzles
Angle of change in fow direction; opening angle
Angle of inclination
Loss coeffcient
(%) Effciency
Pa s Dynamic viscosity
Pipe friction factor
m

/s Kinematic viscosity
kg/m

Density
N/m

Shear stress

f
N/m

Shear stress at yield point

Temperature factor; opening angle of a butter
fy valve; cos : power factor of asynchronous
motors
Head coeffcient (dimensionless head generated
by impeller)
1
Indices, Subscripts
a At outlet crosssection of
the system; branching off
Bl Referring to orifce bore
d On discharge side; at dis
charge nozzle; fowing
through
dyn Denoting dynamic com
ponent
E At the narrowest cross
section of valves (Table 5)
E At suction pipe or bell
mouth inlet
e At inlet crosssection of
system, e. g. in suction
or inlet tank
f Referring to carrier fuid
H Horizontal
in Referring to inlet fow
K Referring to curvature
L Referring to losses
m Mean value
max Maximum value
min Minimum value
N Nominal value
opt Optimum value; at best
effciency point (BEP)
P Referring to pump
p Referring to pressure
r Reduced, for cutdown im
peller or impeller vanes
s On suction side; at suc
tion nozzle
s Referring to solids
stat Static component
sys Referring to system /
installation
t Referring to impeller
prior to trimming
V Vertical
w Referring to water
z Referring to viscous fuid
0 Basic position, referred
to individual sphere
, , Consecutive numbers;
items
I, II Number of pumps oper
ated
Nomenclature
8
Table 1: Centrifugal pump classifcation
Number of stages Single stage Multistage
Shaft position Horizontal Vertical Horiz. Vertic.
Casing design Radial Axial Radial Axial Stage casing
Impeller entries
Motor type, Fig. ..
Dry (standardized)
motor a b c d e f g h
Magnetic drive i
Submerged dry rotor
motor (See ..) j k l m
Wet rotor motor
(See ..) n o p
2
Pump Types
Typical selection criteria for
centrifugal pumps are their
design data (fow rate or capac
ity Q, discharge head H, speed
of rotation n and NPSH), the
properties of the fuid pumped,
the application, the place of
installation and the applicable
regulations, specifcations, laws
and codes. KSB offers a broad
range of pump types to meet the
most varied requirements.
Main design features for classif
cation are:
the number of stages (single
stage / multistage),
2
the position of the shaft (hori
zontal / vertical),
the pump casing (radial, e. g.
volute casing / axial, e. g.
tubular casing),
the number of impeller entries
(single entry / double entry),
the type of motor (dry mo
tor / dry rotor motor, e. g.
submerged motor / wet rotor
motor, e. g. canned motor,
submersible motor).
These features usually determine
what a pump type or series
looks like. An overview of typi
cal designs according to classi
fcation features is given below
(Table and Figs. a to p).
Other pump classifcation
features include:
the mode of installation, which
is dealt with in section 7.,
the nominal diameter (for the
pump size, as a function of
the fow rate),
the rated pressure (for the
wall thickness of casings and
fanges),
the temperature (for example
for the selection of cooling
equipment for shaft seals),
the fuid pumped (abrasive,
aggressive, toxic fuids),
the type of impeller (radial
fow / axial fow depending on
the specifc speed),
the selfpriming ability,
the casing partition, the posi
tion of the pump nozzles, an
outer casing, etc.
a
b
Pump Types (Examples)
9
2
Fig 1 (a to p):
Centrifugal pump classifcation
acc. to Table 1
h g f
k j i
m l
p o
e d c
n
Pump Types (Examples)
0
3 Flow Rate Head Effciency Input Power
3
Selection for Pumping Water
This section applies mainly to
pumping water; the particulari
ties of pump selection for other
media are treated in sections 4,
5 and 6.
3.1
Pump Data
3.1.1
Pump Flow Rate
The pump fow rate or capacity
Q is the useful volume of fuid
delivered to the pump discharge
nozzle in a unit time in m

/s
(l/s and m

/h are also used in

practice, as are GPM in the US).
The fow rate changes propor
tionally to the pump speed of
rotation. Leakage fow as well
as the internal clearance fows
are not considered part of the
pump fow rate.
3.1.2
Developed Head and
Developed Pressure of
the Pump
The total developed head H of
a pump is the useful mechani
cal energy in Nm transferred
by the pump to the fow, per
weight of fuid in N, expressed
in Nm/N = m (also used to be
called metres of fuid)
)
. The
head develops proportionally
to the square of the impellers
speed of rotation and is inde
pendent of the density of the
fuid being pumped. A given
centrifugal pump will impart the
same head H to various fuids
(with the same kinematic viscos
ity ) regardless of their density
. This statement applies to all
centrifugal pumps.
The total developed head H
manifests itself according to
Bernoullis equation (see section
...) as
the pressure head H
p
propor
tional to the pressure differ
ence between discharge and
suction nozzles of the pump,
the geodetic head z
s,d
(Figs. 8
and 9), i.e., the difference in
height between discharge and
suction nozzles of the pump
and
the difference of the kinetic
energy head (v
d

v
s

)/g be
tween the discharge and suc
tion nozzles of the pump.
The pressure rise p in the
pump (considering the location
of the pressure measurement
taps according to section 7.!)
is determined solely by the pres
sure head H
p
along with the
fuid density according to the
equation

p = g [H z
s,d
(v
d

v
s

)/g]
(1)
where
Density of the fuid being
pumped in kg/m

g Gravitational constant
9.8 m/s

H Total developed head of the

pump in m
z
s,d
Height difference between
pump discharge and suction
nozzles in m (see Figs. 8
and 9)
v
d
Flow velocity in the discharge
nozzle = 4 Q/pd
d

in m/s
v
s
Flow velocity in the suction
nozzle = 4 Q/pd
s

in m/s
Q Flow rate of the pump at
the respective nozzle in m

/s
d Inside diameter of the re
spective pump nozzle in m
p Pressure rise in N/m

(for
conversion to bar: bar =
00 000 N/m

)
Highdensity fuids therefore
increase the pressure rise and
the pump discharge pressure.
The discharge pressure is the
sum of the pressure rise and the
inlet pressure and is limited by
the strength of the pump casing.
The effect of temperature on the
pumps strength limits must also
be considered.
3.1.3
Effciency and Input Power
The input power P of a pump
(also called brake horsepower)
is the mechanical power in
kW or W taken by the shaft or
coupling. It is proportional to
the third power of the speed of
rotation and is given by one of
the following equations:
)
In the US, the corresponding units are
ftlbf/lbm, i. e. foot head = foot
poundforce per pound mass; the
numerical value of head and specifc
work are identical.

P =
g Q H

in W =
g Q H

in kW =

Q H
in kW


000 67
(2)
3
where
Density in kg/m

in kg/m

in kg/dm

Q Flow rate in m

/s in m

/s in m

/h
g Gravitational constant = 9.8 m/s

H Total developed head in m

Effciency between 0 and < (not in %)
Table 2: Reference speeds of rotation
Number 4 6 8 0 4
of poles
Frequency Reference speeds for the characteristic curve documen
tation in min

(rpm)
For 50 Hz 900 450 960 75 580 480 45
For 60 Hz 500 750 60 875 700 580 500
The pump effciency is given
with the characteristic curves
(see section ..6).
The pump input power P can
also be read with suffcient ac
curacy directly from the charac
teristic curves (see section ..6)
for density = 000 kg/m

.
For other densities , the input
power P must be changed in
proportion.
Pumping media which are more
viscous than water (see sec
tion 4) or have high concentra
tions of entrained solids (see
section 6) will require a higher
input power. This is, for ex
ample, the case when pumping
sewage or waste water, see sec
tion .6.
The pump input power P is
linearly proportional to the
fuid density . For highdensity
fuids the power limits of the
motor (section ..) and the
torque limits (for the loading on
coupling, shaft and shaft keys)
must be considered.
3.1.4
Speed of Rotation
When using threephase current
motors (asynchronous squirrel
cage motors to IEC standards)
the following speeds of rotation
are taken as reference for pump
operation:
In practice the motors run at
slightly higher speeds (which
depend on the power output
and on the make) [], which the
pump manufacturer may con
sider for the pump design and
selection when the customer
agrees. In this case, the affnity
laws described in section .4.
are to be applied. The charac
teristic curves of submersible
motor pumps and submersible
borehole pumps have already
been matched to the actual
speed of rotation.
When using motor speed con
trollers (for example phase
angle control for ratings up to a
few kW or, in most other cases,
frequency inverters), gearboxes,
belt drives or when using tur
bines or internal combustion
engines as drivers, other pump
speeds are possible.
3.1.5
Specifc Speed and
Impeller Type
The specifc speed n
q
is a para
meter derived from a dimen
sional analysis which allows a
comparison of impellers of vari
ous pump sizes even when their
operating data differ (fow rate
Q
opt
, developed head H
opt
, ro
tational speed n at the point of
best effciency
opt
). The specifc
speed can be used to classify the
optimum impeller design (see
Fig. ) and the corresponding
pump characteristic curve (see
section ..6, Fig. 5).
n
q
is defned as the theoreti
cal rotational speed at which a
geometrically similar impeller
would run if it were of such a
size as to produce m of head
at a fow rate of m

/s at the
best effciency point. It is ex
pressed in the same units as
the speed of rotation. The spe
cifc speed can be made a truly
Effciency Input Power Speed Specifc Speed

High head
impeller
up to 25
Medium head
impeller
up to 40
Low head
impeller
up to 70
Mixed flow
impeller
up to 160
Axial flow
propeller
140 to 400 n
q
rpm
3
n
q
= n = n (3)
where Q
opt
in m

/s Q
opt
in m

/s = Flow rate at
opt
H
opt
in m H
opt
in m = Developed head at
opt
n in rpm n in rev/s = Pump speed
n
q
in metric units n
q
Dimensionless parameter
g Gravitational constant 9.8 m/s

Approximate reference values:

n
q
up to approx. 5 Radial high head impeller
up to approx. 40 Radial medium head impeller
up to approx. 70 Radial low head impeller
up to approx. 60 Mixed fow impeller
approx. from 40 to 400 Axial fow impeller (propeller)
Fig. 2: Effect of the specifc speed n
q
on the design of centrifugal
pump impellers. The diffuser elements (casings) of single stage
pumps are outlined.
Q
opt
/
(H
opt
/)
/4
Q
opt
(g H
opt
)
/4
Specifc Speed
dimensionless characteristic
parameter while retaining the
same numerical value by using
the defnition in the righthand
version of the following equa
tion []:
For multistage pumps the devel
oped head H
opt
at best effciency
for a single stage and for double
entry impellers, the optimum
fow rate Q
opt
for only one im
peller half are to be used.
As the specifc speed n
q
in
creases, there is a continuous
change from the originally
radial exits of the impellers to
mixed fow (diagonal) and
eventually axial exits (see Fig. ).
The diffuser elements of radial
pump casings (e.g. volutes) be
come more voluminous as long
as the fow can be carried off
radially. Finally only an
axial exit of the fow is possible
(e. g. as in a tubular casing).
Using Fig. it is possible to de
termine n
q
graphically. Further
types of impellers are shown in
Fig. 4: Star impellers are used in
selfpriming pumps. Periph
eral impellers extend the speci
fc speed range to lower values
down to approximately n
q
= 5
(peripheral pumps can be de
signed with up to three stages).
For even lower specifc speeds,
rotary (for example progressive
cavity pumps with n
q
= 0. to )
or reciprocating positive dis
placement pumps (piston
pumps) are to be preferred.
The value of the specifc speed
is one of the infuencing para
meters required when convert
ing the pump characteristic
curves for pumping viscous or
solidsladen media (see sections
4 and 6).
In Englishlanguage pump lit
erature the true dimensionless
specifc speed is sometimes des
ignated as the type number K.
In the US, the term N
s
is used,
which is calculated using gal
lons/min (GPM), feet and rpm.
The conversion factors are:
K = n
q
/ 5.9
N
s
= n
q
/ 5.6
(4)

Radial doubleentry impeller*)

Star impeller for side channel pump
(selfpriming)
Peripheral pump impeller for very low
specific speed (n
q
5 to 0)
*) Plan view shown without front shroud
Closed (shrouded) mixed flow impeller *)
Open (unshrouded) mixed flow impeller
Axial flow propeller
Radial impeller *)
3
Fig. 3: Nomograph to determine specifc speed n
q
(enlarged view on p. 80)
Example: Q
opt
= 66 m
3
/h = 18.3 l/s; n = 1450 rpm, H
opt
= 17.5 m. Found: n
q
= 23 (metric units).
Fig. 4:
Impeller types for clear liquids
Specifc Speed Impeller Types Characteristic Curves
3.1.6
Pump Characteristic Curves
Unlike positive displacement
pumps (such as piston pumps),
centrifugal pumps deliver a var
iable fow rate Q (increasing
with decreasing head H) when
operating at constant speed.
They are therefore able to ac
commodate changes in the
system curve (see section ..).
The input power P and hence
the effciency as well as the
NPSH
r
(see section .5.4) are
dependent on the fow rate.
4
90
80
70
60
50
40
20
30
80
70
60
50
40
0
5
10
20
30
10
0 20 40 60 80 100 120
Flow rate Q [m
3
/h] Flow rate Q [m
3
/h] Flow rate Q [m
3
/h]
140 160 0 100 200 300 400 500 0 500 1500 1000 2000 2500 3000 550
H
e
a
d

H

[
m
]
N
P
S
H
r

[
m
]
P
o
w
e
r

P

[
k
W
]
E
f
f
i
c
i
e
n
c
y

H

[
%
]
24
22
18
14
10
20
16
12
8
6
90
30
80
70
60
50
40
0
5
15
10
15
14
16
17
13
H
e
a
d

H

[
m
]
N
P
S
H
r

[
m
]
P
o
w
e
r

P

[
k
W
]
E
f
f
i
c
i
e
n
c
y

H

[
%
]
2
4
18
14
10
20
16
12
8
6
90
30
80
70
60
50
40
5
15
10
60
40
20
80
100
0
H
e
a
d

H

[
m
]
N
P
S
H
r

[
m
]
P
o
w
e
r

P

[
k
W
]
E
f
f
i
c
i
e
n
c
y

H

[
%
]
n = 2900 min
1
n = 1450 min
1
n = 980 min
1
Operating limit
>
L V
11
OPT

H
H
OPT
11
OPT

0
0
OPT
300
25
11
OPT

(
(
OPT
Operating limit for
low input power
for high
input power
25
25
300
300
150
150
70
70
40
40
11
OPT

NPSH
r
NPSH
r opt
25
25
300
300
150
70
40
150
300
70
40
25
3
Fig. 5: Effect of specifc speed n
q
on centrifugal pump characteristic
curves (Not drawn to scale! For NPSH
r
, see section 3.5.4).
Fig. 6: Three examples of characteristic curves for pumps of differing specifc speeds.
a: radial impeller, n
q
20; b: mixed fow impeller, n
q
80; c: axial fow impeller, n
q
200.
(For NPSH
r
see section 3.5.4)
Characteristic Curves
The relationship of these val
ues is shown graphically in the
pump characteristic curves,
whose shape is infuenced by
the specifc speed n
q
and which
document the performance of a
centrifugal pump (see Fig. 5 for
a comparison of characteristics
and Fig. 6 for examples). The
head curve of the pump is also
referred to as the H/Q curve.
The H/Q curve can be steep or
fat. For a steep curve, the fow
rate Q changes less for a given
change of developed head H
than for a fat curve (Fig. 7).
This can be advantageous when
controlling the fow rate.
5
H
e
a
d

H
Q
steep
Q
flat
Flow rate Q
Unstable
region
Steep characteristic curve
Flat characteristic curve
-
A
X
I
M
U
M
H
Fig. 7: Steep, fat or unstable characteristic curve
3
Characteristic Curves
H/Q characteristics normally
have a stable curve, which
means that the developed head
falls as the fow rate Q in
creases. For low specifc speeds,
the head H may in the low
fow range drop as the fow
rate Q decreases, i. e., the curve
is unstable (shown by the dash
line in Fig. 7). This type of
pump characteristic curve need
only be avoided when two
intersections with the system
curve could result, in particular
when the pump is to be used for
parallel operation at low fow
rates (see section .4.4) or when
it is pumping into a vessel which
can store energy (accumulator
flled with gas or steam). In all
other cases the unstable curve is
just as good as the stable charac
teristic.
Unless noted otherwise, the
characteristic curves apply for
the density and the kinematic
viscosity of cold, deaerated
water.
6
a
va
e
ve
e
ve
vd
vs
a
va
Hgeo
Hsgeo
zs,d
a va
pa pa
A B C
pe
D E
3
Fig. 8: Centrifugal pump system with variously designed vessels in suction lift operation
A = Open tank with pipe ending below the water level
B = Closed pressure vessel with free fow from the pipe ending above the water level
C = Closed pressure vessel with pipe ending below the water level
D = Open suction/inlet tank
E = Closed suction/inlet tank
v
a
and v
e
are the (usually negligible) fow velocities at position a in tanks A and C and at position e
in tanks D and E. In case B, v
a
is the non-negligible exit velocity from the pipe end at a .
System Head Bernoulli
3.2
System Data
3.2.1
System Head
3.2.1.1
Bernoullis Equation
Bernoullis equation expresses
the equivalence of energy in
geodetic (potential) energy,
static pressure and kinetic
energy form. The system head
H
sys
for an assumed frictionless,
inviscid fow is composed of the
following three parts (see Figs.
8 and 9):
H
geo
(geodetic head) is the
difference in height between
the liquid level on the inlet
and discharge sides. If the dis
charge pipe ends above the
liquid level, the centre of the
exit plane is used as reference
for the height (see Figs 8B and
9B).
(p
a
p
e
)/( g) is the pressure
head difference between the
inlet and outlet tank, applic
7
a
va
e
ve
e
ve
vd
vs
a
va
Hgeo
Hzgeo
zs,d
a va
pa pa
A B C
pe
D E
3
Fig. 9: Centrifugal pump system with variously designed vessels in suction head (positive inlet pressure)
operation. Legend as in Fig. 8.
System Head Bernoulli
able when at least one of the
tanks is closed as for B, C or
E (see Figs. 8B, C, E, 9B, C,
E).
(v
a

v
e

)/g is the difference in

the velocity heads between the
tanks.
For a physically real fow, the
friction losses (pressure head
losses) must be added to these
components:
H
L
is the sum of the head
losses (fow resistance in the
piping, valves, fttings, etc in
the suction and discharge lines
as well as the entrance and exit
losses, see section ...), and
is referred to as the system pres
sure loss.
The sum of all four components
yields the system head H
sys
:
H
sys
= H
geo
+ (p
a
p
e
) / ( g) + (v
a

v
e

)/g + H
L
(5)

where
all the heads H are in m,
all the pressures p are in Pa ( bar = 00 000 Pa),
all velocities v are in m/s,
the density is in kg/m

,
the gravitational constant is g = 9.8 m/s

.
8
Fig. 10: Pipe friction factor as a function of the Reynolds number Re and the relative roughness d/k
(enlarged view on p. 81)
3 System Head Pressure Loss Head Loss
5 6 8 10
3
10
4
2 3 4 5 6 8 10
5
2 3 4 5 6 8 10
7
2 3 4 5 6 8 10
6
2 3 4 5 6 8
0.007
0.008
0.009
0.010
0.012
0.014
0.016
0.018
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
DK 20
Fully rough (k > 0j
laminar turbulent
Re
crit
L
im
it
in
g

c
u
r
v
e
40
100
200
500
1000
2000
5000
10 000
20 000
50 000
H
y
d
ra
u
lic
a
lly
s
m
o
o
th
(k
=
0
j
L

=

6
4 R
e
Reynolds number 2E
P
i
p
e

f
r
i
c
t
i
o
n

f
a
c
t
o
r

L
100 000
The difference of the velocity
heads can often be neglected in
practice. When at least one tank
is closed as for B, C or E (see
Figs. 8B, C, E, 9B, C, E), Eq. 5
can be simplifed as
H
sys
H
geo
+ (p
a
p
e
)/( g) +H
L
(6)
and further simplifed when
both tanks are open as for A
and D (see Figs. 8A, D and 9A,
D) as
H
sys
H
geo
+ H
L
(7)
3.2.1.2
Pressure Loss Due to Flow
Resistances
The pressure loss p
L
is caused
by wall friction in the pipes and
fow resistances in valves, ft
tings, etc. It can be calculated
from the head loss H
L
, which is
independent of the density ,
using the following equation:
p
L
= g H
L
(8)
where
Density in kg/m

g Gravitational constant
9.8 m/s

H
L
Head loss in m
p
L
Pressure loss in Pa
( bar = 00 000 Pa)
3.2.1.2.1
Head Loss in Straight Pipes
The head loss for fow in
straight pipes with circular
crosssections is given in general
by
H
L
=
L

v

(9)
d g
where
Pipe friction factor according
to Eqs. () to (4)
L Length of pipe in m
d Pipe inside diameter in m
v Flow velocity in m/s
(= 4Q/pd

for Q in m

/s)
g Gravitational constant
9.8 m/s

3 Head Loss in Straight Pipes

For pipes with noncircular
crosssections the following
applies:
d = 4A/U (10)
where
A Crosssectional fow area
in m

U Wetted perimeter of the

crosssection A in m; for
open channels the free fuid
surface is not counted as part
of the perimeter.
Recommended fow velocities
for cold water
Inlet piping 0.7 .5 m/s
Discharge piping .0 .0 m/s
for hot water
Inlet piping 0.5 .0 m/s
Discharge piping .5 .5 m/s
The pipe friction factor has
been determined experimentally
and is shown in Fig. 0. It varies
with the fow conditions of the
liquid and the relative rough
ness d/k of the pipe surface. The
fow conditions are expressed
according to the affnity laws
(dimensional analysis) using the
Reynolds number Re. For cir
cular pipes, this is:
Re = v d/ (11)
where
v Flow velocity in m/s
(= 4Q/pd

for Q in m

/s)
d Pipe inside diameter in m
Kinematic viscosity in m

/s
(for water at 0 C exactly
.00 (0)
6
m

/s).
For noncircular pipes, Eq. 0 is
to be applied for determining d.
For hydraulically smooth pipes
(for example drawn steel tubing
or plastic pipes made of poly
ethylene (PE) or polyvinyl chlor
ide (PVC)) or for laminar fow,
can be calculated:
In the laminar fow region
(Re < 0) the friction factor is
independent of the roughness:
= 64/Re (12)
For turbulent fow (Re > 0)
the test results can be repre
sented by the following empiri
cal relationship defned by Eck
(up to Re < 0
8
the errors are
smaller than %):
=
0.09

(lg
Re
)

(13)

7
In Fig. 0 it can be seen that
the pipe friction factor depends
on another dimensionless para
meter, the relative roughness of
the pipe inner surface d/k; k is
the average absolute roughness
of the pipe inner surface, for
which approximate values are
given in Table . Note: both d
and k must be expressed in the
same units, for example mm!
As shown in Fig. 0, above a
limiting curve, is dependent
only on the relative roughness
d/k. The following empirical
equation by Moody can be used
in this region:
= 0.0055 + 0.5/ (d/k) (14)
For practical use, the head
losses H
L
per 00 m of straight
steel pipe are shown in Fig.
as a function of the fow rate Q
and pipe inside diameter d. The
values are valid only for cold,
clean water or for fuids with
the same kinematic viscosity, for
completely flled pipes and for an
absolute roughness of the pipe
inner surface of k = 0.05 mm,
i.e., for new seamless or longi
tudinally welded pipes. (For
the pipe inside diameters, see
Table 4).
The effect of an increased
surface roughness k will be de
monstrated in the following
for a frequently used region in
Fig. (nominal diameter 50
to 00 mm, fow velocity 0.8
to .0 m/s). The dark-shaded
region in Fig. 11 corresponds
to the similarly marked region
in Fig. 10 for an absolute rough
ness k = 0.05 mm. For a rough-
ness increased by a factor 6
(slightly incrusted old steel pipe
with k = 0.0 mm), the pipe
friction factor (proportional
to the head loss H
L
) in the
lightly shaded region in Fig. 0
is only 25% to 60% higher than
before.
For sewage pipes the increased
roughness caused by soiling
must be taken into considera
tion (see section .6). For pipes
with a large degree of incrusta
tion, the actual head loss can
only be determined experimen
tally. Deviations from the no
minal diameter change the head
loss considerably, since the pipe
inside diameter enters Eq. (9) to
the 5th power! (For example, a
5% reduction in the inside dia
meter changes the head loss by
0%). Therefore the nominal
diameter may not be used as
the pipe inside diameter for the
calculations!
0
Pie mareiial Conuirion o ie inreiioi
Sreel nev, seamless skin
aciu-cleaneu
galvanizeu
longiruuin- skin
ally velueu, Liruminizeu
galvanizeu
cemenreu
iivereu
useu, moueiarely iusry
sliglr inciusrarion
leavy inciusrarion
arei cleaning
AsLesros cemenr nev
Heavy clay (uiainage) nev
Conciere nev, uninisleu
virl smoorl inisl
Sun conciere nev, uninisleu
virl smoorl inisl
Reinoiceu conciere nev, virl smoorl inisl
All concieres useu, virl smoorl inisl
Neral ies uiavn
Glass, lasric
RuLLei ruLing nev, nor emLiirrleu
Woou nev
arei long exosuie ro varei
Nasoniy
1 m 5 10 50 100 500 1000 5000 10
4
0.001 k in mm 0.005 0.01 0.05 0.1 0.5 1 5 10
k
3
Table 4: Inside diameter d and wall thickness s in mm and weight of typical commercial steel pipes and their
water content in kg/m to ENV 10 220 (formerly DIN ISO 4200). D = outside diameter, s = wall thickness
All dimensions in mm Seamless pipe Welded pipe
Seamless Welded weight in kg/m weight in kg/m
DN D s * d s ** d Pipe Water Pipe Water
5 . .0 7. .8 7.7 0.95 0.5 0.866 0.46
0 6.9 .0 .9 .8 . . 0.4 . 0.46
5 .7 . 9. .0 9.7 .78 0.665 .56 0.69
4.4 .6 7. . 7.8 .55 .09 .7 .
40 48. .6 4. . 4.7 .9 .46 .6 .50
50 60. .9 54.5 . 55.7 4. . .9 .44
65 76. .9 70. .6 70.9 4.7 .88 5.4 .95
80 88.9 . 8.5 .9 8. 6.76 5.4 6.5 5.4
00 4. .6 07. . 07.9 9.8 9.00 8.77 9.4
5 9.7 4.0 .7 .6 .5 .4 .6 . .8
50 68. 4.5 59. 4.0 60. 8. 9.9 6. 0.
00 9. 6. 06.5 4.5 0. . .5 .8 4.7
50 7.0 6. 60.4 5.0 6.0 4.4 5. .0 54.
00 .9 7. 09.7 5.6 .7 55.5 75. 44.0 76.8
50 55.6 8.0 9.6 5.6 44.4 68.6 90.5 48. 9.
400 406.4 8.8 88.8 6. 9.8 86. 8.7 6. .7
500 508.0 .0 486.0 6. 495.4 5 85.4 77.9 9.7
600 60.0 .5 585.0 6. 597.4 84 68.6 9.8 80.
* above nominal diameter DN identical to DIN 448 ** above nominal diameter DN 5 identical to DIN 458
Table 3: Approximate average roughness height k (absolute rough-
ness) for pipes
Head Loss in Straight Pipes Dimensions and Weights of Steel Pipes

100
50
20
10
5
2
1
0.5
0.5 1 2 5 10 2 5 10
2
2 5 10
3
2 2 5 10
4
0.2
0.1
0.05
0.02
0.01
m
100 m
S
p
e
c
i
f
i
c

h
e
a
d

l
o
s
s

H
L
1.1
1.0
0.9
0.8
0 20 40 60 C
H
L
correction
for plastic pipe
T
e
m
p
e
r
a
t
u
r
e

f
a
c
t
o
r

J
Temperature t
m
3
/h
0.5 0.2 1 2 5 10 20 50 100 200 500 1000 2000 5000
Flow rate Q
l/s
4
.0

3
.5
3
.0

2
.5
2
.0

1
.2
5

1
.0

0
.8
0
.6
0
.5
0
.4

0
.3

d

=

1
5

m
m
2
0
2
5
3
2
4
0
5
0
6
5
8
0
1
0
0
1
2
5
1
5
0
1
7
5
2
0
0
2
5
0
3
0
0
3
5
0
4
0
0
1
.5
d

=

5
0
0

m
m
2
0

0
0
0
1
0

0
0
0
2
0
0

0
0
0
1
0
0

0
0
0
5
0
0

0
0
0
1

0
0
0

0
0
0
R
e

=

2

0
0
0

0
0
0
0LASTICPIPEANDDRAWNMETALTUBING
5
0

0
0
0
v
=
5
.0
m
/
s
100
50
20
10
5
2
1
0.5
0.5 1 2 5 10 2 5 10
2
2 5 10
3
2 2 5 10
4
0.2
0.1
0.05
0.02
0.01
m
100 m
S
p
e
c
i
f
i
c

h
e
a
d

l
o
s
s

H
L
m
3
/h
0.5 0.2 1 2 5 10 20 50 100 200 500 1000 2000 5000
Flow rate Q
l/s
d

=

1
5

m
m
2
0
2
5
3
2
4
0
5
0
6
5
8
0
1
0
0
1
2
5
1
5
0
1
7
5
2
0
0
2
5
0
3
0
0
3
5
0
4
0
0
5
0
0
6
0
0
7
0
0
8
0
0
9
0
0
1
0
0
0
1
2
0
0
1
4
0
0
1
6
0
0
1
8
0
0
d

=

2
0
0
0

m
m
4
.0

3
.5
3
.0

2
.5
2
.0

1
.2
5

1
.0

0
.8
0
.6
0
.5
0
.4

0
.3

1
.5
v
=
5
.0
m
/
s
2
0

0
0
0
1
0

0
0
0 5
0

0
0
0
2
0
0

0
0
0
1
0
0

0
0
0
5
0
0

0
0
0 1

0
0
0

0
0
0
2

0
0
0

0
0
0
R
e

=

5

0
0
0

0
0
0
.EWUNTREATEDSTEELPIPE
3
Fig. 11: Head losses H
L
for new steel pipes (k = 0.05 mm) (enlarged view on p. 82)
Fig. 12: Head losses H
L
for hydraulically smooth pipes (k = 0) (enlarged view on p. 83). For plastic pipe
when t = 10 C multiply by the temperature factor .
Head Loss in Straight Pipes

3
Fig. 13: Schematic representation of the valve designs listed in
Table 5
4 5
6 7 8 9 0
4 5
6 7 8 9
Head Loss in Straight Pipes Valves and Fittings
The head losses H
L
in plastic
(for example PE or PVC) pipes
or smooth drawn metal pip
ing are very low thanks to the
smooth pipe surface. They are
shown in Fig. and valid for
water at 0 C. At other tem
peratures, the loss for plastic
pipes must be multiplied with
a temperature correction factor
indicated in Fig. to account
for their large thermal expan
sion. For sewage or other un
treated water, an additional
0 0% head loss should be
taken into consideration for po
tential deposits (see section .6).
3.2.1.2.2
Head Loss in Valves and
Fittings
The head loss in valves and ft
tings is given by
H
L
= v

/g (15)
where
Loss coeffcient
v Flow velocity in a charac
teristic crosssection A (for
example the fange) in m/s
g Gravitational constant
9.8 m/s

Tables 5 to 8 and Figures to

5 contain information about
the various loss coeffcients for
valves and fttings for operation
with cold water.
The minimum and maximum
in Table 5 bracket the values
given in the most important
technical literature and apply to
valves which have a steady ap
proach fow and which are fully
open. The losses attributable to
straightening of the fow dis
turbances over a length of pipe
equivalent to x DN down
stream of the valve are included
in the value in accordance
with VDI/VDE 7 guidelines.
Depending on the inlet and
exit fow conditions, the valve
models used and the develop
ment objectives (i.e. inexpensive
vs. energysaving valves), the
loss values can vary dramatically.

T
a
b
l
e

S
:

I
c
s
s

c
c
e
f
f
i
c
i
e
n
t
s

Z

f
c
r

i
a
r
i
c
u
s

t
y
j
e
s

c
f

i
a
l
i
e
s

a
n
J

f
i
t
t
i
n
g
s

(
r
e
f
e
r
r
e
J

t
c

t
l
e

i
e
l
c
c
i
t
y

c
f

f
l
c
u

i
n

t
l
e

l
i
n
e

c
c
n
n
e
c
t
i
c
n

n
c
h
i
n
a
l

J
i
a
h
e
t
e
r

D
N
)
1
y

e

o

v
a
l
v
e

/

i
r
r
i
n
g


D
e
-

L
o
s
s

c
o
e

i
c
i
e
n
r

Z

o
i

D

=


s
i
g
n

1
5

2
0

2
5

3
2

4
0

5
0

6
5

S
0

1
0
0

1
2
5

1
5
0

2
0
0
S
l
i
u
e

u
i
s
c

v
a
l
v
e
s


m
i
n

1

0
.
1
(
u
L

=

D

)

m
a
x


0
.
6
5

0
.
6

0
.
5
5

0
.
5

0
.
5

0
.
4
5

0
.
4

0
.
3
5

0
.
3
R
o
u
n
u
-
L
o
u
y

g
a
r
e

v
a
l
v
e
s

m
i
n

2






0
.
2
5

0
.
2
4

0
.
2
3

0
.
2
2

0
.
2
1

0
.
1
9

0
.
1
S

(
u
L

=

D

)

m
a
x







0
.
3
2

0
.
3
1

0
.
3
0

0
.
2
S

0
.
2
6

0
.
2
5

0
.
2
3
B
a
l
l

a
n
u

l
u
g

v
a
l
v
e
s


m
i
n

3

0
.
1
0

0
.
1
0

0
.
0
9

0
.
0
9

0
.
0
S

0
.
0
S

0
.
0

0
.
0

0
.
0
6

0
.
0
5

0
.
0
5

0
.
0
4
(
u
L

=

D

)


m
a
x


0
.
1
5


m
i
n






0
.
9
0

0
.
5
9

0
.
3
S

0
.
2
6

0
.
2
0

0
.
1
4

0
.
1
2

0
.
0
9
B
u
r
r
e
i

l
y

P

2
.
5

1
0

m
a
x






1
.
2
0

1
.
0
0

0
.
S
0

0
.

0

0
.
6
2

0
.
5
6

0
.
5
0

0
.
4
2
v
a
l
v
e
s

P

1
6

2
5

m
i
n

4






2
.
0
4

1
.
S
0

1
.
5
5

1
.
3
0

1
.
0
S

0
.
S
4

0
.

5


m
a
x







2
.
5
0
`

2
.
3
0
`

2
.
1
0
`

1
.
9
0
`

1
.

0
`

1
.
5
0
`

1
.
3
0

G
l
o
L
e

v
a
l
v
e
s
,

o
i
g
e
u

m
i
n

5



6
.
0



6
.
0


m
a
x




6
.
S



6
.
S
G
l
o
L
e

v
a
l
v
e
s
,

c
a
s
r

m
i
n

6

3
.
0


m
a
x


6
.
0
C
o
m

a
c
r

v
a
l
v
e
s

m
i
n

0
.
3

0
.
4

0
.
6

0
.
6

1
.
0

1
.
1






1
.
1


m
a
x


0
.
3

0
.
9

1
.
9




1
.
9

2
.
2

2
.
2

2
.
3

2
.
5

2
.
5
A
n
g
l
e

v
a
l
v
e
s

m
i
n

S

2
.
0


m
a
x


3
.
1



3
.
1

3
.
4

3
.
S

4
.
1

4
.
4

4
.

5
.
0

5
.
3

5
.

Y
-
v
a
l
v
e
s


m
i
n

9

1
.
5


m
a
x


2
.
6
S
r
i
a
i
g
l
r
-
r
l
i
o
u
g
l

v
a
l
v
e
s

m
i
n

1
0

0
.
6


m
a
x


1
.
6
D
i
a

l
i
a
g
m

v
a
l
v
e
s

m
i
n

1
1

0
.
S










0
.
S


m
a
x


2
.

2
.

o
n
-
i
e
r
u
i
n

v
a
l
v
e
s
,

m
i
n

1
2

3
.
0











3
.
0
s
r
i
a
i
g
l
r

s
e
a
r


m
a
x


6
.
0











6
.
0

o
n
-
i
e
r
u
i
n

v
a
l
v
e
s
,

m
i
n

1
3

3
.
2





3
.
2

3
.

5
.
0

.
3

4
.
3
a
x
i
a
l


m
a
x


3
.
4

3
.
4

3
.
5

3
.
6

3
.
S

4
.
2

5
.
0

6
.
4

S
.
2

4
.
6

o
n
-
i
e
r
u
i
n

v
a
l
v
e
s
,

m
i
n

1
4

2
.
5

2
.
4

2
.
2

2
.
1

2
.
0

1
.
9

1
.

1
.
6

1
.
5
s
l
a
n
r
e
u

s
e
a
r


m
a
x


3
.
0

I
o
o
r

v
a
l
v
e
s


m
i
n

1
5






1
.
0

0
.
9

0
.
S

0
.

0
.
6

0
.
5

0
.
4


m
a
x







3
.
0
S
v
i
n
g

c
l
e
c
k

v
a
l
v
e
s

m
i
n

1
6

0
.
5




0
.
5

0
.
4


m
a
x


3
.
0
H
y
u
i
o
s
r
o

s

v

=

4

m
/
s


1

0
.
9



3
.
0


3
.
0

2
.
5

v

=

3

m
/
s








1
.
S



4
.
0


4
.
5

4
.
0

v

=

2

m
/
s








5
.
0



6
.
0


S
.
0

.
5
I
i
l
r
e
i
s



1
S





2
.
S
S
r
i
a
i
n
e
i
s



1
9





1
.
0
2
5
0

3
0
0

4
0
0

5
0
0

6
0
0

S
0
0

1
0
0
0

C
o
m
m
e
n
r






0
.
1






0
.
3
0
.
1

0
.
1
6

0
.
1
5

0
.
1
3

0
.
1
2

0
.
1
1

0
.
1
1
0
.
2
2

0
.
2
0

0
.
1
9

0
.
1
S

0
.
1
6

0
.
1
5

0
.
1
4
0
.
0
3

0
.
0
3

0
.
0
2





I
o
i

u
L

<

D

0
.
1
5





Z

=

0
.
4

r
o

1
.
1
0
.
0
6






0
.
0
6
0
.
4
0

0
.
3

0
.
3
3

0
.
3
3

0
.
3
3

0
.
3
0

0
.
2
S
0
.
5
6

0
.
4
S

0
.
4
0




0
.
4
0
1
.
1
0

0
.
9
0

0
.
S
3

0
.

6

0
.

1

0
.
6

`

0
.
6
3
`

`

A
l
s
o

o
i

P

4
0













3
.
0






Z

=

2

r
o

3

c
a
n

L
e

a
c
l
i
e
v
e
u


6
.
0

o
i

o

r
i
m
i
z
e
u

v
a
l
v
e




2
.
0
6
.
0

6
.
3

6
.
6

1
.
5

2
.
6


0
.
6


1
.
6



























4
.
3






A
x
i
a
l
l
y

e
x

a
n
u
e
u

a
s

i
o
m

4
.
6






D

1
2
5

1
.
5

3
.
0
0
.
4

0
.
4

(

.
0
)

(
6
.
1
)

(
5
.
5
)

(
4
.
5
)

(
4
.
0
)

(


)

I
n

g
i
o
u

s

3
.
0


0
.
4

0
.
3




0
.
3

S
v
i
n
g

c
l
e
c
k

v
a
l
v
e
s

v
i
r
l
-






3
.
0

o
u
r

l
e
v
e
i
s

a
n
u

v
e
i
g
l
r
s

2
)
2
.
5

1
.
2

2
.
2
4
.
0

1
.
S

3
.
4
6
.
5

6
.
0

.
0

2
.
S






I
n

c
l
e
a
n

c
o
n
u
i
r
i
o
n

1
.
0

I
o
i

u
L

<

D

o
o
r
n
o
r
e

1
)
o n - i e r u i n v a l v e s S l u r - o v a l v e s
1
)

I

r
l
e

n
a
i
i
o
v
e
s
r

s
l
u
r
-
o

u
i
a
m
e
r
e
i

u
L

i
s

s
m
a
l
l
e
i

r
l
a
n

r
l
e

l
i
n
e

c
o
n
n
e
c
r
i
o
n

n
o
m
i
n
a
l

u
i
a
m
e
r
e
i

D

.

r
l
e

l
o
s
s

c
o
e

i
c
i
e
n
r

Z

m
u
s
r

L
e

i
n
c
i
e
a
s
e
u

L
y

(
D

/
u
L
)
x

v
i
r
l

x

=

5

r
o

6
.
2
)

W
l
e
n

r
l
e

v
a
l
v
e

i
s

a
i
r
i
a
l
l
y

o

e
n
,

i
.
e
.

l
o
v

l
o
v

v
e
l
o
c
i
r
i
e
s
,

r
l
e

l
o
s
s

c
o
e

i
c
i
e
n
r
s

i
n
c
i
e
a
s
e

r
o

r
l
e

m
a
x

v
a
l
u
e
s
.

W
i
r
l

i
n
c
i
e
a
s
i
n
g

l
o
v

v
e
l
o
c
i
r
i
e
s

v

(
i
n

m
/
s
)

r
l
e

l
o
s
s

c
o
e

i
c
i
e
n
r
s

u
e
c
i
e
a
s
e

i
o
u
g
l
l
y

a
s

Z


3
/
v
.
S
e
e

I
i
g
.

1
3

o
i

u
e
s
i
g
n
s
.
3 Loss Coeffcients for Valves
4
1le Z value o rle single 90 elLov sloulu nor Le uouLleu, Lur only Le
mulrilieu Ly rle acrois inuicareu ro oLrain rle iessuie loss causeu Ly rle
comLinarion o elLovs illusriareu.
Table : Icss ccefficients Z fcr fittings
Combinations oI pipe bends (elbows):
Expansion |oints:
Bellovs-rye exansion joinr
virl/virlour guiue ie Z z 0.3/2.0
Comensarion ruLe Lenu Z z 0.6 ro 0.S
Cieaseu comensarion ruLe Lenu Z z 1.3 ro 1.6
Bellovs-rye comensarion ruLe Lenu Z z 3.2 ro 4
Inlet pipe Iittings:
Inler euge
Slai Z z 0.5 3 oi D = 5 60 45
Clameieu Z z 0.25 0.55 0.20 0.05 Z z 0.6 0. 0.S
1.S 1.6 1.4
D
Discharge pipe Iittings:
Z z 1 uovnsrieam o an auequare lengrl o sriaiglr ie virl an aioxim-
arely unioim velociry uisriiLurion in rle ourler cioss-secrion
Z z 2 oi a veiy iiiegulai velociry uisriiLurion, oi examle immeuiarely
uovnsrieam o a ie irring oi a valve, erc.
Conrinueu on nexr age
A 15 30 45 60 90
Suiace Suiace Suiace Suiace Suiace
smoorl iougl smoorl iougl smoorl iougl smoorl iougl smoorl iougl
Z oi R = 0 0.0 0.10 0.14 0.20 0.25 0.35 0.50 0.0 1.15 1.30
Z oi R = u 0.03 0.0 0.14 0.34 0.19 0.46 0.21 0.51
Z oi R = 2 u 0.03 0.06 0.09 0.19 0.12 0.26 0.14 0.30
Z oi R 5 u 0.03 0.06 0.0S 0.16 0.10 0.20 0.10 0.20
umLei o ciicum-
eienrial velus 2 3 3
Z 0.15 0.20 0.25
Rounu elLov
Welueu
Lenu
R
A
A
u
Table 6: Loss coeffcients in elbows and bends
3 Head Loss in Valves and Fittings Loss Coeffcients for Fittings
Note: For the branch fttings
in Table 7 and the adapters of
Table 8, one must differentiate
between the irreversible pressure
loss (reduction in pressure)
p
L
= v

/ (16)
where
p
L
Pressure loss in Pa
Loss coeffcient
Density in kg/m

v Flow velocity in m/s

and the reversible pressure
change of the frictionless fow
according to Bernoullis equa
tion (see ...):
p

= (v

)/ (17)
For accelerated fows (for ex
ample a reduction in the pipe
diameter), p

is always
negative, for decelerated fows
(e.g. pipe expansion) it is always
positive. When calculating the
net pressure change as the arith
metic sum of p
L
and p

, the
pressure losses from Eq. 6 are
always to be subtracted.
Often the socalled k
v
value is
used instead of the loss coef
5
a
a
R
K
R
K
1.2
0.8
0.4
0.4 0.8 1.2
0
0
L
o
s
s

c
o
e
f
f
i
c
i
e
n
t

Z
Elbow radius R
K
Duct width a
Radius on inner corner With turning
vane cascade
Radius on outer corner
3
Flow meters:
Short Venturi tube = 0
is referred to the velocity v at diameter D.
Diameter ratio d/D = 0.0 0.40 0.50 0.60 0.70 0.80
Area ratio m = (d/D)

= 0.09 0.6 0.5 0.6 0.49 0.64

Short Venturi tube 6 0.7 0. 0.
Standard orifce 00 85 0 4.5
Water meters (volume meters) 0
For domestic water meters, a max. pressure drop of bar is specifed for the
rated load. In practice, the actual pressure loss is seldom higher.
Branch fttings (of equal diameter)
Note:
The loss coeffcients
a
for the branchedoff fow Q
a
or
d
for the main fow
Q
d
= Q Q
a
refer to the velocity of the total fow Q in the branch. On the
basis of this defnition,
a
or
d
may have negative values; in this case, they
are indicative of a pressure gain instead of a pressure loss. This is not to be
confused with the reversible pressure changes according to Bernoullis equa
tion (see notes to Tables 7 and 8).
Q
a
/Q = 0. 0.4 0.6 0.8

a
0.4 0.08 0.47 0.7 0.9

d
0.7 0.0 0.4 0.5

a
0.88 0.89 0.95 .0 .8

d
0.08 0.05 0.07 0.

a
0.8 0 0. 0.7 0.7

d
0.7 0.9 0.09 0.7

a
0.68 0.50 0.8 0.5 0.48

d
0.06 0.04 0.07 0.0
Table 8: Loss coeffcients for adapters
Expansion Contraction
Type I II III IV
Type d/D 0.5 0.6 0.7 0.8 0.9
I 0.56 0.4 0.6 0. 0.04
= 8 0.07 0.05 0.0 0.0 0.0
II for = 5 0.5 0. 0.07 0.0 0.0
= 0 0. 0.7 0. 0.05 0.0
III 4.80 .0 0.88 0.4 0.
IV for 0 < < 40 0. 0.0 0.05 0.0 0.0
D d
v

D d
v

D d
v

D d
v

Fig. 14: Effect of rounding off
the inner and outer side of el-
bows in square ducts on the loss
coeffcient
Head Loss in Valves and Fittings Loss Coeffcients for Fittings and Flow Meters
D d D D
Standard orifice
D

v
d
v
Q
d
Q
d
Q
d
Q
a
Q
Q
a
Q
Q
d
Q
a
Q
45
45
Q
a
Q
fcient when calculating the
pressure loss for water in valves:
p
L
= (Q / k
v
)

. /000 (18)
where
Q Volume rate of fow in m

/h (!)
Density of water in kg/m

p
L
Pressure loss in bar (!)
The k
v
value is the fow rate in
m

/h which would result from a

pressure drop of bar through
the valve for cold water. It cor
relates with the pressure loss p
L

in bar with the fow rate Q in
m

/h. The notation k

vs
is used
for a fully open valve.
Conversion for cold water:
6 d
4
/k
v

(19)
where
d Reference (nominal) diameter
of the valve in cm (!)
Table 7 (continued)
6
Flow rate Q
S
y
s
t
e
m

h
e
a
d

H
s
y
s
System head characteristic H
sys
Dynamic component = H
L
+
v
a
2
- v
e
2
2g
Static component = H
geo
+
p
a
- p
e
g
Fig. 16: System characteristic curve H
sys
with static and dynamic
components
3
3.2.2
System Characteristic Curve
The system characteristic curve
plots the head H
sys
required by
the system as a function of the
fow rate Q. It is composed of
the socalled static and
dynamic components (see
Fig. 6)

.
The static component consists
of the geodetic head H
geo
and
the pressure head difference
(p
a
p
e
)/(g) between the inlet
10
3
5
2
10
2
5
2
10
1
5
2
1
0.5
0.2
0.1
0.5 1.0 0 0.5 1.0 0
J
0
= 45
60
74
90
J
0
J
v v
y
v
6
10
a
L
o
s
s

c
o
e
f
f
i
c
i
e
n
t

Z
Relative opening angle (J
0
-

Jj/J
0
Degree of opening y/a or relative lift y/DN
Fig. 15:
Loss coeffcients
of butterfy
valves, globe
valves and gate
valves as a
function of the
opening angle
or degree of
opening (The
numbers desig-
nate the types
illustrated in
Fig. 13)
Head Loss in Valves System Characteristic Curve

One must be careful to distinguish

between this use of static and
dynamic components and the pre
cisely defned static head and dy
namic head used in fuid dynamics,
since the dynamic component of
the system head curve consists of both
static head (i.e. pressure losses)
and dynamic head (i.e. velocity or
kinetic energy head).
7
120
100
50
57.5
40
30
20
10
8
3 4 5 Q m
3
/h
Q l/s
10
3 2 1 4 5 10 30 20 200 40 50 100
20 30 40 50 100 200 300 400 500 800
H
m
65-160
65-200
32-125
32-160
32-250.1 32-250 40-250 50-250 80-250
100-250
100-200 125-200
100-160
65-250
40-200 50-200 80-200
50-160
80-160
40-160
40-125 50-125 65-125
32-200 32-200.1
32-160.1
32-125.1
3
and outlet tanks, which are in
dependent of the fow rate. The
pressure head difference is zero
when both tanks are open to the
atmosphere.
The dynamic component con
sists of the head loss H
L
, which
increases as the square of the
fow rate Q (see section ...),
and of the change in velocity
head (v
a

v
e

)/g between the

inlet and outlet crosssections of
the system. Two points are suff
cient to calculate this parabola,
one at Q = 0 and one at any
point Q > 0.
For pipe systems connected
one after the other (series con
nection) the individual system
curves H
sys
, H
sys
etc. are
plotted as functions of Q, and
Fig. 17: Selection chart for a volute casing pump series for n = 2900 rpm
(First number = nominal diameter of the discharge nozzle, second number = nominal impeller diameter)
the heads for each fow rate are
added to obtain the total system
curve H
sys
= f(Q).
For branched piping systems the
system curves H
sys
, H
sys
, etc.
of the individual branches be
tween the fow dividers are each
calculated as functions of Q.
The fow rates Q

, Q

, etc. of
all branches in parallel for each
given head H
sys
are then added
to determine the total system
curve H
sys
= f(Q) for all the
branches together. The sections
before and after the fow
dividers must be added as for a
series connection.
System Characteristic Curve Selection Chart
8
100
50
40
30
20
10
6
1 2
0.3 0.4 0.5 1 2
3 4 5 10 20 Q m
3
/h
Q l/s
30
10
9
8
7
6
5
4
3
2
9
8
7
6
5
4
3
2
7
6
5
3
10
4
2
3
4
2
Pump size 1 Pump size 2 Pump size 3 Pump size 4
H
m
3 4 5
290
80 60 40 L/s 20 0
250 200 150 m
3
/h 100 50 0
290 250 200 150
Flow rate
m
3
/h 100 50
10
20
30
40
kW
46
2.5
4
6
8
26
30
40
50
60
70
0
H
e
a
d
m
m
N
P
S
H
P
o
w
e
r
219
208
199
190
180
219 mm
lmpeller
180-219
208
199
83.5
H %
82.5
80
75
75
77.5
70
65
60
190
180
3
3.3
Pump Selection
3.3.1
Hydraulic Aspects
The data required for selecting
a pump size, i.e. the fow rate Q
and the head H of the desired
operating point are assumed
to be known from the system
characteristic curve; the electric
mains frequency is also given.
With these values it is possible
to choose the pump size, the
speed of rotation and, if neces
sary, the number of stages, from
the selection chart in the sales
literature (see Figs. 7 and 9).
Further details of the chosen
pump such as the effciency ,
the input power P, the required
NPSH
r
(see section .5.4) and
the reduced impeller diameter
D
r
can then be determined from
Fig. 19: Selection chart for a multistage pump series for n = 2900 rpm
Fig. 18: Complete characteristics
of a centrifugal pump
Hydraulic Aspects of Pump Selection
9
3
the individual characteristic
curve (for example see Fig. 8).
If there are no specifc reasons
for doing otherwise, a pump
should be selected so that the
operating point lies near its
best effciency point Q
opt
(=
fow rate at which effciency is
highest, BEP). The limits Q
min

and Q
max
(for example due to
vibration behaviour, noise emis
sion as well as radial and axial
forces) are given in the product
literature or can be determined
by inquiry [].
To conclude the selection, the
NPSH conditions must be
checked as described in section
.5.
A multistage pump is chosen us
ing the same general procedure;
its selection chart shows the
number of stages in addition to
the pump size (Fig. 9).
For pumps operating in series
(one after the other) the devel
oped heads H

, H

, etc. of the
individual characteristic curves
must be added (after subtracting
any head losses which occur be
tween them) to obtain the total
characteristic H = f(Q).
For pumps operating in parallel,
the individual characteristics H

,
H

, etc. = f(Q) are frst reduced

by the head losses occurring up
to the common node (head loss
H
L
calculation according to sec
tion ...) and plotted versus
Q. Then the fow rates Q of
the reduced characteristics are
added to produce the effective
characteristic curve of a vir
tual pump. This characteristic
interacts with the system curve
H
sys
for the rest of the system
through the common node.
3.3.2
Mechanical Aspects
When selecting a pump the me
chanical aspects require atten
tion in addition to the hydrau
lics. Several examples are:
the effects of the maximum
discharge pressure and tem
perature of the fuid pumped
on the operating limits,
the choice of the best shaft
sealing method and cooling
requirements,
the vibration and noise emis
sions,
the choice of the materials of
construction to avoid corro
sion and wear while keeping
in mind their strength and
temperature limits.
These and other similar re
quirements are often specifc to
certain industries and even to
individual customers and must
be addressed using the product
literature [] or by consulting
the design department.
3.3.3
Motor Selection
3.3.3.1
Determining Motor Power
Operation of a centrifugal pump
is subject to deviations from
rated speed and fuctuations
in the fow volume handled,
and, consequently, changes in
the operating point (see section
.4.). In particular if steep
power curves are involved (see
Figs. 5 and 6), this may result
in a higher required pump input
power P than originally speci
fed. For practical purposes, a
safety allowance is therefore
added when the appropriate
motor size is selected. Safety
allowances may be specifed
by the purchaser, or laid down
in technical codes, see Fig. 0.
The safety allowances stipulated
by individual associations are
shown in the relevant type series
literature [] or the customers
specifcation.
When energysaving control
methods are used (e. g., speed
control systems), the maximum
Fig. 20: Drive power as a function of rated pump input power at the
operationg point
Example as per ISO 9905, 5199 and 9908 (Class I, II and III)
1
100
110
120
130
140
150
5 10 50 kW 20 100
%
D
r
i
v
e

p
o
w
e
r

r
e
l
a
t
i
v
e

t
o

p
u
m
p

i
n
p
u
t

p
o
w
e
r

u
n
d
e
r

r
a
t
e
d

c
o
n
d
i
t
i
o
n
s

i
n

%
Pump input power under rated conditions
Hydraulic Aspects of Pump Selection Motor Selection
0
3
power peaks which may pos
sibly occur must be taken into
account.
If a pump is selected for a
product with a density lower
than that of water, the motor
power required may have to be
determined on the basis of the
density of water (for example,
during the performance test or
acceptance test in the test bay).
Typical effciencies and power
factors cos of standardized IP
54 motors at 50 Hz are shown
in Fig. , and the curves of ef
fciency and power factor
cos as a function of relative
motor load P/PN in Fig. .
Table 9 lists types of enclosure
that provide protection of elec
tric motors against ingress of
foreign objects or water, and
of persons against accidental
contact.
The specifc heat buildup in
both electric motors and fexi
ble couplings during startup
as well as the risk of premature
contactor wear limit the frequen
cy of starts. Reference values
for the maximum permissible
number of starts Z are given in
table 0, unless otherwise speci
fed.
Submersible motor pumps (Figs.
j to m) are readyassembled
pump units whose motors need
not be selected individually [7].
Their electrical characteristics
are given in the type series
literature. The motor is flled
with air and can be operated
submerged in the product han
dled thanks to a in most cases
doubleacting shaft seal with a
paraffn oil barrier.
Table 9: Types of enclosure for electric motors to EN 60 59 and
DIN/VDE 050, Part 5
The type of protective enclosure is indicated by the IP code as follows:
Code letters (International Protection) IP
First digit (0 to 6 or X if not applicable) X
Second digit (0 to 8 or X if not applicable) X
Alternatively letters A, B, C, D and H, M, S, W for special purposes only.
Key to Protection of electrical Protection of persons against
digits: equipment against ingress of accidental contact by
solid objects
First 0 (not protected) (not protected)
digit > 50 mm in dia. back of the hand
> .5 mm in dia. fnger
> .5 mm in dia. tool
4 > .0 mm in dia. wire
5 protected against dust (limited wire
ingress permitted, no harmful
deposits)
6 totally protected against dust wire
Protection against ingress of water with harmful consequences
Second 0 (not protected)
digit vertical dripwater
dripwater up to 5 from the vertical
sprays (60 from the vertical)
4 sprays (all directions)
5 lowpressure jets of water
6 strong jets of water (heavy sea)
7 temporary fooding
8 permanent fooding
1
0.7
0.8
0.9
1.0
5 2 10 50 kW 20 300 100
H
cos J
Rated power P
N
E
f
f
i
c
i
e
n
c
y

H
P
o
w
e
r

f
a
c
t
o
r

c
o
s

J
2 poles
4 poles
Fig. 21: Typical effciencies and power factors cos of standard-
ized motors, IP 54 enclosure, at 50 Hz as a function of motor
power P
N
Table 0: Permissible frequency of starts Z per hour for electric motors
Motor installation Dry Wet (submersible motors)
Motors up to 4 kW 5 0
Motors up to 7.5 kW 5 5
Motors up to kW 5
Motors up to 0 kW 0
Motors above 0 kW 0 0
Motor Selection

3
0
0
0.2
0.4
0.6
0.9
1.0
0.5 1.0 kW 1.5
H
40 kW
5 kW
10 kW
]
cos J
E
f
f
i
c
i
e
n
c
y

H
P
o
w
e
r

f
a
c
t
o
r

c
o
s

J
Rated power P/P
N
2 poles
4 poles
Fig. 22: Curve of effciency and power factor cos of standardized
IP 54 motors plotted over relative motor power P/P
N
Submersible borehole pumps,
which are mostly used for ex
tracting water from wells, are
another type of readyassembled
units whose motors need not be
selected individually (Fig. p).
On these pumps, the rotor and
the windings are immersed in
water [7]. Their electrical char
acteristics and permissible fre
quency of starts are indicated in
the type series literature [].
3.3.3.2
Motors for Seal-less Pumps
Sealless pumps are frequently
used for handling aggressive,
toxic, highly volatile or valu
able fuids in the chemical and
petrochemical industries. They
include magneticdrive pumps
(Fig. f) and canned motor
pumps (Figs. n and o). A
magdrive pump is driven by a
primary magnetic feld rotating
outside its fameproof enclosure
and running in synchronization
with the secondary magnets in
side the enclosure [].
The primary component in
turn is coupled to a commercial
dry driver. The impeller of a
canned motor pump is mounted
directly on the motor shaft, so
that the rotor is surrounded by
the fuid pumped. It is separated
from the stator windings by the
can [7].
Sealless pump sets are generally
selected with the help of compu
terized selection programs, tak
ing into account the following
considerations:
The rotor is surrounded by
the fuid pumped, whose kine
matic viscosity (see section
4.) must be known, as it
infuences friction losses and
therefore the motor power
required.
Metal cans or containment
shrouds (for example made
of .460) cause eddy current
losses, resulting in an increase
in the motor power required.
Nonmetal shrouds in mag
drive pumps do not have this
effect.
The vapour pressure of the
fuid pumped must be known,
so as to avoid bearing damage
caused by dry running when
the fuid has evaporated. It is
advisable to install monitoring
equipment signalling dry run
ning conditions, if any.
Data on specifc fuid proper
ties such as its solids content
and any tendency to solidify
or polymerize or form incrus
tations and deposits, need to
be available at the time of se
lection.
3.3.3.3
Starting Characteristics
The pump torque Tp transmit
ted by the shaft coupling is
directly related to the power P
and speed of rotation n. Dur
ing pump startup, this torque
follows an almost parabolical
curve as a function of the speed
of rotation [0], as shown in
Fig. . The torque provided
by the asynchronous motor
must, however, be higher so as
to enable the rotor to run up to
duty speed. Together with the
voltage, this motor torque has a
direct effect on the motors cur
rent input, and the latter in turn
on heat buildup in the motor
windings. The aim, therefore, is
to prevent unwanted heat build
up in the motor by limiting the
runup period and/or current
inrush [] (see also Table ).
Motors for Seal-less Pumps Starting Characteristics

Table : Starting methods for asynchronous motors

Starting Type of Current Runup Heat build Mechani Hydraulic Cost Recommended Comments
method equipment input time up in motor cal loading loading relation motor designs
(mains load) during
startup
D. o. l. Contactor 48 I
N
Approx. High Very high Very high All Mostly limited to
(mecha 0.55 s 4 kW by energy
nical) supply companies
Star Contactor

of d. o. l. Approx. High Very high Very high .5 All; canned mo Usually stipu
delta combi values 0 s tors and sub lated for motors
nation mersible motors > 4 kW by
(mecha subject to a energy supply
nical) major drop in companies
speed during
switchover
Reduced Autotrans 0.49 times Approx. High High High 55 All No currentless
voltage former, the d. o. l. 0 s phase during
mostly values switchover
70% tap (gradually re
ping placed by soft
starters)
Soft Soft starter Continuous Approx. High Low Low 55 All Runup and run
start (power ly variable; 00 s down continu
electro typically ously variable
nics) I
N
via ramps for
each individual
load appllication;
no hydraulic
surges
Fre Frequency I
N
060 s Low Low Low Approx. All Too expensive to
quency inverter 0 use solely for run
inverter (power up and rundown
electro purposes; better
nics) suited for open
or closedloop
control
3
In the case of d.o.l. starting
(where the full mains voltage
is instantly applied to the mo
tor once it is switched on), the
full starting torque is instantly
available and the unit runs up
to its duty speed in a very short
period of time. For the motor
itself, this is the most favour
able starting method. But at up
to 4 8 times the rated current,
the starting current of the d.o.l.
method places a high load on
the electricity supply mains,
particularly if large motors are
involved, and may cause prob
lematic voltage drops in electri
cal equipment in their vicinity.
For motor operation on public
lowvoltage grids (80 V), the
regulations laid down by the en
ergy supply companies for d.o.l.
starting of motors of 5.5 kW
and above must be complied
with. If the grid is not suitable
for d.o.l starting, the motor can
be started up with reduced volt
ages, using one of the following
methods:
Star-delta starting is the most
frequent, since most inexpen
sive, way of reducing the start
ing current. During normal
operation, the motor runs in
delta, so that the full mains
voltage (for example 400 V) is
applied to the motor windings.
For startup, however, the wind
ings are starconnected, so that
the voltage at the windings is
reduced by a factor of 0.58 rela
tive to the mains voltage. This
reduces the starting current and
torque to one third of the values
of d.o.l. starting, resulting in a
longer startup process.
The motor runs up in star con
nection beyond pullout torque
up to the maximum speed of
rotation at point B in Fig. .
Then, switchover to delta is ef
fected and the motor continues
to accelerate up to rated speed.
During the switchover period of
approx. 0. s, the current sup
ply to the motor is interrupted
Starting Methods

Fig. 23: Starting curve for current I and torque T of squirrel-cage

motors in star-delta connection
( = star connection; = delta connection; P = pump)
3
0
0
100
200
300
0
100
200
300
%
400
500
50 100
%
B''
D'
D''
D
l
l
T
T
T
P
B'
B
% of n
synchr.
C
u
r
r
e
n
t

l
T
o
r
q
u
e

T
Motor speed n
and the speed drops. On pump
sets with a low moment of in
ertia (canned motors and sub
mersible motors), this speed re
duction may be so pronounced
that switchover to delta may
result in almost the full starting
current being applied after all,
same as with d.o.l. starting.
An autotransformer also serves
to reduce voltage at the motor
windings and unlike stardelta
starting allows selection of
the actual voltage reduction. A
70% tapping of the transformer,
for instance, will bring down
the startup torque and current
supplied by the mains to 49%
of the values for d.o.l. starting.
The fact that current supply is
never interrupted is another ad
vantage of autotransformers.
Soft starters are used for elec
tronic continuous variation of
the voltage at the motor wind
ings in accordance with the
dimmer principle. This means
that the startup time and start
ing current can be freely selected
within the motors permissible
operating limits (heat losses due
to slip!). Special constraints re
garding the frequency of starts
(contrary to Table 0) have to
be heeded [].
Frequency inverters (usually for
open or closedloop control)
provide a soft starting option
without the need for any addi
tional equipment. For this pur
pose, the output frequency and
voltage of the frequency inverter
(see section .4.) are increased
continuously from a minimum
value to the required value,
without exceeding the motors
rated current.
Starting Methods
4
3
3.4
Pump Performance and
Control [4], [6], [8]
3.4.1
Operating Point
The operating point of a centri
fugal pump, also called its duty
point, is given by the intersec
tion of the pump characteristic
curve (see section ..6) with
the system characteristic curve
(see section ..). The fow
rate Q and the developed head
H are both determined by the
intersection. To change the op
erating point either the system
curve or the pump curve must
be changed.
A system characteristic curve
for pumping water can only be
changed
by changing the fow resist
ance (for example, by chang
ing the setting of a throttling
device, by installing an orifce
or a bypass line, by rebuilding
the piping or by its becoming
incrusted) and/or
by changing the static head
component (for example, with
a different water level or tank
pressure).
A pump characteristic curve can
be changed
by changing the speed of rota
tion (see section .4.),
by starting or stopping pumps
operated in series or parallel
(see sections .4.4 or .4.5),
for pumps with radial impel
lers, by changing the impel
lers outside diameter (see sec
tion .4.6),
for pumps with mixed fow
impellers, by installing or
changing the setting of in
stalled preswirl control
equipment (see section .4.8),
for axial fow (propeller)
pumps, by changing the blade
pitch setting (see section
.4.9).
Please note: the effect of these
measures for changing the char
acteristic curve can only be pre
dicted for noncavitating opera
tion (see section .5).
3.4.2
Flow Control by Throttling
Changing the fow rate Q by
operating a throttle valve is the
simplest fow control method
not only for a single adjustment
of the fow rate but also for
its continuous control, since it
requires the least investment.
But it is also the most energy
wasting method, since the fow
energy is converted irreversibly
to heat.
Fig. 4 illustrates this process:
by intentionally increasing the
system resistance (for example
by throttling a valve on the
160
140
120
100
80
60
40
20
0
20 40 60 80 100 120
20 40 60 80 100 120
Q [%|
120
100
80
60
40
20
0
Q [%|
H [%|
B
2
B
1
P
2
P
1
P [%|
Pump charac-
teristic curve
Throttling
Surplus head
System head
requirement
System curve H
sys2
System
curve Hsys1
Power saved
Fig. 24: Change of the operating point and power saved by
throttling a pump whose power curve has a positive slope
Pump Performance Operating Point Throttling
5
3
pump discharge side) the origi
nal system curve H
sys
becomes
steeper and transforms into
H
sys
. For a constant pump
speed, the operating point B

on
the pump characteristic moves
to B

at a lower fow rate. The

pump develops a larger head
than would be necessary for
the system; this surplus head is
eliminated in the throttle valve.
The hydraulic energy is irrevers
ibly converted into heat which
is transported away by the fow.
This loss is acceptable when
the control range is small or
when such control is only sel
dom needed. The power saved
is shown in the lower part of
the fgure; it is only moderate
compared with the large surplus
head produced.
20
19
18
17
16
15
14
13
12
0.1 0.2 0.3 0.4 0.5 0.6 0.7
dBl
d
Area ratio (d
Bl
/dj
2
T
h
r
o
t
t
l
i
n
g

c
o
e
f
f
i
c
i
e
n
t

f
Fig. 25: Orifce plate and its throttling coeffcient f
The same is principally true
of the installation of a fxed,
sharp-edged orifce plate in the
discharge piping, which can be
justifed for low power or short
operating periods. The neces
sary hole diameter d
Bl
of the
orifce is calculated from the
head difference to be throttled
H, using the following equa
tion:
d
Bl
= f Q/ g H (20)
where
d
Bl
Hole diameter of the orifce
in mm
f Throttling or pressure drop
coeffcient acc. to Fig. 5
Q Flow rate in m

/h
g Gravitational constant
9.8 m/s

H Head difference to be
throttled in m
Since the area ratio (d
Bl
/d)

must be estimated in advance,
an iterative calculation is neces
sary (plotting the calculated vs.
the estimated diameter d
Bl
is
recommended so that after two
iterations the correct value can
be directly interpolated, see
example calculation 8.0).
3.4.3
Variable Speed Flow Control
At various speeds of rotation n,
a centrifugal pump has differ
ent characteristic curves, which
are related to each other by the
affnity laws. If the characteris
tics H and P as functions of Q
are known for a speed n

, then
all points on the characteristic
curve for n

can be calculated
by the following equations:
Q

= Q

. n

/n

(21)
H

= H

(n

/n

(22)
P

= P

(n

/n

(23)
Eq. () is valid only as long
as the effciency does not de
crease as the speed n is reduced.
With a change of speed, the op
erating point is also shifted (see
section .4.). Fig. 6 shows the
H/Q curves for several speeds of
rotation; each curve has an in
tersection with the system char
acteristic H
sys
. The operating
point B moves along this system
curve to smaller fow rates when
the speed of rotation is reduced.
Orifce Plate Variable Speed
6
If the system curve is a parabola
through the origin as for H
sys

in the example, the developed
head H according to Eq. () is
reduced to one fourth its value
and the required driving power
in Eq. () to one eighth its
value when the speed is halved.
The lower part of Fig. 6 shows
the extent of the savings P

compared with simple
throttling.
If the system curve is a parabola
with a large static head compo
nent as for H
sys
, it is possible
that the pump characteristic at
reduced speed has no intersec
Fig. 26: Operation of a variable speed pump for different system
characteristic curves H
sys1
and H
sys2

(Power savings P
1
and P
2
at half load each compared with simple
throttling)
160
140
120
100
80
60
40
20
0
0 Q [%|
100
80
60
40
20
0
0 Q [%|
H [%|
H
A2
H
A2
H
A1
P
2
P
1
H
A1
B
P [%|
]
n = 100%
90%
70%
60%
50%
n = 100%
90%
80%
70%
60%
n = 50%
P
2
P
80%
20 40 60 80 100 120
20 40 60 80 100 120
P
1
Head required
H
A2 stat
P
o
w
e
r

s
a
v
e
d
3
tion with it and hence, that no
operating point results; the low
er speed range is then of no use
and could be eliminated. The
potential power savings P

at a given fow rate Q are less
than for the system curve H
sys

as shown in the lower part of
the diagram [4]. The improve
ment compared with throttling
decreases as the static head
component H
sys,stat
increases
(i.e., for a lower dynamic head
component H
sys,dyn
).
Variation of the speed usually
means varying the electrical
driving frequency, which must
be considered when choosing
the motor. The expenditure for
variable speed drives is not low,
but it is amortized quickly for
pumps which are used often and
which are frequently required
to run at reduced fows with
small static head component
H
sys,stat
[8]. This is particularly
the case for pumps in heating
systems.
3.4.4
Parallel Operation of Centri-
fugal Pumps
Where one pump is unable to
deliver the required fow Q at
the operating point, it is possi
ble to have two or more pumps
working in parallel in the same
piping system, each with its
own nonreturn valve (Fig. 7).
Parallel operation of pumps is
easier when their shutoff heads
H
0
are all equal, which is the
case for identical pumps. If the
shutoff heads H
0
differ, the low
est shutoff head marks the point
on the common H/Q curve for
the minimum fow rate Q
min
,
below which no parallel opera
tion is possible, since the non
return valve of the pump with
smaller shutoff head will be held
shut by the other pump(s).
During parallel pumping it
must be kept in mind that after
stopping one of two identical
centrifugal pumps (Fig. 7),
the fow rate Q
single
of the re
maining pump does not fall
to half of Q
parallel
, but rather
increases to more than half. The
remaining pump might then
immediately run at an operat
ing point B
single
above its design
point, which must be considered
Variable Speed Parallel Operation
7
3
when checking the NPSH values
(see section .5) and the drive
power (see section ..). The
reason for this behaviour is the
parabolic shape of the system
characteristic H
sys
. For the same
reason, the reverse procedure of
taking a second identical pump
on line does not double the fow
rate Q
single
of the pump that
was already running, but rather
increases the fow rate less than
that:
Q
parallel
< Q
single
(24)
This effect when starting or
stopping one additional pump
is more intense when the system
curve is steeper or when the
pump characteristic is fatter. As
long as both pumps I and II are
running, the total fow rate
Q
parallel
is the sum of Q
I
and
Q
II
, i.e.:
Q
parallel
= Q
I
+ Q
II
(25)
To compute the characteristic
curve for parallel operation see
section ...
Starting or stopping individual
pumps operated in parallel does
save energy, but it allows only a
stepped control of the fow rate.
For continuously variable con
trol, at least one of the pumps
must be ftted with a variable
speed drive or a control valve
must be installed in the common
discharge piping [4].
If centrifugal pumps running
at fxed speeds and having un
stable characteristics (see Fig. 7
in section ..6) are run in par
allel, diffculties can arise when
bringing another pump online.
H
0
H
B
parallel
-
Q
parallel
Q
l
= Q
ll
= Q
parallel
/ 2 Q
parallel
= Q
l
+ Q
ll

System characteristic H
sys
B Operating point
H
0
Shutoff head
Characteristic of
pump l or pump ll
Characteristic of pump l + pump ll
B
single
Q
single
Flow rate Q
T
o
t
a
l

d
e
v
e
l
o
p
e
d

h
e
a
d

H
Fig. 27: Parallel operation of 2 identical centrifugal pumps with
stable characteristic curves
The problems occur when the
developed head H

of the pump
running is larger than the shut
off head (i.e., the developed
head at Q = 0) of the pump to
be started; the second pump is
unable to overcome the pressure
on its nonreturn valve (Fig. 8,
System curve H
sys
). Pumps
with unstable characteristics are
not suitable for such a low fow
operation. (For a lower system
curve H
sys
they would be per
fectly able to operate properly
since the developed head H

of the pump running is lower
than the shutoff head H
0
of the
pump to be started).
H
2 pumps
1 pump
Q
H
1
H
sys 1
H
sys 2
H
2
H
0
Fig. 28: Parallel operation of 2 identical centrifugal pumps with
unstable characteristics
Parallel Operation
8
3
3.4.5
Series Operation
In series operation, the pumps
are connected one after the
other so that the developed
heads can be added for a given
fow rate. This means that the
discharge pressure of the frst
pump is the inlet pressure for
the second pump, which must
be considered for the choice of
shaft seal and for the strength
of the casing. For this reason
multistage pumps are usually
used for such applications (ex
cept for the hydraulic transport
of solids, see section 6). They
do not pose these shaft sealing
problems.
3.4.6
Turning Down Impellers
If the fow rate or developed
head of a radial or mixed fow
centrifugal pump are to be
reduced permanently, the out
side diameter D of the impeller
should be reduced. The reduc
tion should be limited to the
value for which the impeller
vanes still overlap when viewed
radially. The documentation of
the pump characteristics (Fig.
8) usually shows curves for
several diameters D (in mm).
Impellers made from hard ma
terials, such as those used for
solidshandling pumps, or from
stainless steel sheet metal, as
well as single vane impellers
(Fig. 4) and star or periph
eral pump impellers cannot be
turned down. (The same is true
for underfling as described in
section .4.7). For multistage
pumps, usually only the vanes
D
t
D
r
D
1
but not the shrouds of the im
pellers are cut back. It is some
times possible to simply remove
the impeller and diffuser of one
stage of a multistage pump and
replace them with a blind stage
(two concentric cylindrical cas
ings to guide the fow) instead
of cutting back the impeller
vanes. Impellers with a non
cylindrical exit section are either
turned down or have only their
blades cut back as specifed in
the characteristic curve litera
ture (for example, as shown in
Fig. 9).
If the impeller diameter only
needs to be reduced slightly, a
rule of thumb can be applied.
An exact calculation cannot be
made, since the geometrical sim
ilarity of the vane angle and exit
width are not preserved when
turning down the impeller. The
following approximate relation
ship exists between Q, H and
the impeller diameter D to be
found (averaged, if required):
(D
t
/D
r
)

Q
t
/Q
r
H
t
/H
r
(26)
where subscript
t
designates the
condition before the reduction
Fig. 29: Contour for cutting
back the vanes of an impeller
with a mixed fow exit
of the impeller outer diameter
and index
r
the condition after
the reduction. The required
(average) reduced diameter re
sults as:
D
r
D
t
(Q
r
/Q
t
) D
t
(H
r
/H
t
)
(27)
The parameters needed to deter
mine the reduced diameter can
be found as shown in Fig. 0:
in the H/Q curve (linear scales
required!) a line is drawn con
necting the origin (careful: some
scales do not start at zero!) and
the new operating point B
r
. The
extension of the line intersects
the characteristic curve for full
diameter D
t
at the point B
t
. In
this way the values of Q and H
with the subscripts
t
and
r
can
be found, which are used with
Eq. (7) to fnd the desired re
duced diameter D
r
.
The ISO 9906 method is more
accurate, but also more involved
through the consideration of
the average diameter D

of the
impeller leading edge (sub
script

), valid for n
q
< 79 and
for a change of diameter < 5%,
as long as the vane angle and
the impeller width remain con
stant. Thus using the nomen
clature of Figs. 9 and 0:
Series Operation Impeller Diameter Reduction
9
3
"
T
1
T
"
R
1
R
(
T
$
T
(
R
$
R
u
s
i
n
g

E
q
.

2
8
u
s
in
g

E
q
.

2
6
T
o
t
a
l

d
e
v
e
l
o
p
e
d

h
e
a
d

H
Flow rate Q
Fig. 30:
Determination
of the reduced
impeller dia-
meter D
r

A solution is only possible when
D

is known and when a para

bola H ~ Q

is drawn through
the reduced operating point B
r

(with H
r
and Q
r
), not a line as
n
(D
r

)/(D
t

) = H
r
/H
t
= (Q
r
/Q
t
)

(28)
in Fig. 0, which intersects the
base H/Q curve for diameter D
t

at a different point B
t
(with dif
ferent H
t
and Q
t
).
Fig. 31: Under-fled vanes of a
radial impeller
Fig. 32: Characteristic curve set of a centrifugal pump with pre-swirl
control equipment, n
q
160
2
1
0
0 0.5 1.0 1.5
1.0
0.99
0.94
0.7
50
30
70
90
105
115
0.88
0.83
HH
opt
Operating limit
Pre-swirl control setting
Relative flow rate Q/Q
opt
R
e
l
a
t
i
v
e

d
e
v
e
l
o
p
e
d

h
e
a
d

H
/
H
o
p
t
3.4.8
Pre-Swirl Control of the Flow
For tubular casing pumps with
mixed fow impellers, the pump
characteristic can be infuenced
by changing the prerotation
in the impeller inlet fow. Such
preswirl control equipment is
often ftted to control the fow
rate. The various characteristic
curves are then shown in the
product literature labelled with
the control setting (Fig. ).
3.4.9
Flow Rate Control or Change
by Blade Pitch Adjustment
The characteristic curves of ax
ial fow (propeller) pumps can be
altered by changing the setting
of the propeller blade pitch. The
setting can be fxed and frmly
bolted or a device to change the
blade pitch during operation
can be used to control the fow
rate. The blade pitch angles are
3.4.7
Under-fling of Impeller
Vanes
A small, permanent increase of
the developed head at the best
effciency point (up to 4 6%)
can be achieved for radial im
pellers by fling the back sides of
the backwardcurved vanes, i.e.,
by sharpening the vanes on the
concave side, as shown in Fig.
. The shutoff head does not
change. This method is suitable
for minor fnal corrections.
Impeller Diameter Reduction Under-fling Pre-swirl Blade Pitch Adjustment
40
3
shown in the product literature
with their respective characteris
tic curves (see Fig. ).
3.4.10
Flow Control Using a Bypass
The system characteristic curve
can be made steeper by closing
a throttle valve, but it can also
be made fatter by opening a
bypass in the discharge piping
as shown in Fig. 4. The pump
operating point moves from B

to a larger fow rate B

. The
bypass fow rate is controlled
and can be fed back into the
inlet tank without being used
directly. From the point of view
of saving energy, this type of
control only makes sense when
the power curve falls for in
creasing pump fow rates (P

>
P

), which is the case for high

specifc speeds (mixed and axial
fow propeller pumps).
For these types of pumps, con
trolling the fow by preswirl
control or by changing the blade
pitch is even more economical,
however. The expenditure for a
bypass and control valve is not
small [4]. This method is also
suitable for preventing pumps
from operating at unacceptably
low fow rates (see operating
limits in Figs. 5 and 6c as well
as in Figs. and ).
0
0 0.5 1.0 1.5
7
24
2
1
HH
opt
= 0.83
0.88
0.94
0.99
1.0
11
16
20
Operating limit
Blade pitch setting
Relative flow rate Q/Q
opt
R
e
l
a
t
i
v
e

d
e
v
e
l
o
p
e
d

h
e
a
d

H
/
H
o
p
t
160
140
120
100
80
60
40
20
0
20 40 60 80 100 120 Q [%|
120
100
80
60
40
20
0
20 40 60 80 100 120 Q [%|
H [%|
B
1
B
2
P
2
P
1
P [%|
-
Bypass flow rate Useful flow rate
Surplus head
Head required
by the system
System curve
without bypass
Pump
characteristic
System curve
with bypass
Power saved
Fig. 33: Characteristic curve set of an axial fow pump with blade
pitch adjustment, n
q
200
Fig. 34: Characteristic curves and operating points of a pump with
a falling power curve and fow control using a bypass. (For a radial
fow pump the power curve would increase towards the right and
this type of control would cause an increase in power input, see
Fig. 5).
Blade Pitch Adjustment Bypass
4
3
3.5
Suction and Inlet Conditions
[3]
NPSH = Net Positive Suction
Head
3.5.1
The NPSH Value of the
System: NPSH
a
The NPSH
a
value is the differ
ence between the total pressure
in the centre of the pump inlet
and the vapour pressure p
v
,
expressed as a head difference
in m. It is in certain respects a
measure of the probability of
vaporization at that location
and it is determined only by
the operating data of the sys
tem and the type of fuid. The
vapour pressure of water and
other liquids are shown in Table
and in Fig. 5 as a function
of the temperature.
C
a
r
b
o
n
d
is
u
lp
h
id
e
A
c
e
t
o
n
e
B
e
n
z
o
l
n
-
B
u
t
a
n
e
E
t
h
a
n
e
E
t
h
a
n
o
l
D
i
e
t
h
y
l

e
t
h
e
r
P
r
o
p
a
n
e
i
-
B
u
t
a
n
e
B
e
n
z
o
l
P
h
e
n
o
l
T
o
l
u
o
l
A
n
i
l
i
n
e
M
e
t
h
a
n
o
l
A
c
e
t
o
n
e
A
c
e
t
i
c

a
c
i
d
F
o
r
m
i
c

a
c
i
d
G
l
y
c
e
r
i
n
C
a
r
b
o
n

d
i
s
u
l
p
h
i
d
e
A
m
m
o
n
i
a
S
u
l
p
h
u
r

d
i
o
x
i
d
e
C
a
r
b
o
n

t
e
t
r
a
c
h
lo
r
id
e
B
e
n
z
o
l
100
50
40
30
20
10
5
4
3
2
1
0.5
0.4
0.3
0.2
0.1
-50 0 100 200 C 300
bar
Temperature t
v
a
p
o
u
r

p
r
e
s
s
u
r
e

p
v
Fig. 35: Vapour pressure p
v
of various fuids as a function of the
temperature t (for an enlarged view see page 84)
Suction and Inlet Conditions NPSH Available
4
Table 12: Vapour pressure p
v
, density and kinematic viscosity of water at saturation conditions as a
function of the temperature t
3 NPSH Available Data for Water
t p
v

C bar kg/m

mm

/s
0 0.006 999.8 .79
0.00656 999.9
0.00705 999.9
0.00757 000.0
4 0.008 000.0
5 0.0087 000.0
6 0.0095 999.9
7 0.000 999.9
8 0.007 999.8
9 0.046 999.7
0 0.07 999.6 .07
0.0 999.5
0.040 999.4
0.0496 999.
4 0.0597 999.
5 0.070 999.0
6 0.086 998.8
7 0.096 998.7
8 0.006 998.5
9 0.096 998.4
0 0.07 998. .004
0.0485 997.9
0.064 997.7
0.0808 997.5
4 0.098 997.
5 0.067 997.0
6 0.060 996.7
7 0.0564 996.4
8 0.0779 996.
9 0.04004 995.8
0 0.044 995.6 0.80
0.0449 995.
0.0475 994.9
0.0509 994.6
4 0.058 994.
5 0.056 99.9
6 0.05940 99.5
7 0.0674 99.
8 0.0664 99.9
9 0.0699 99.6
40 0.0775 99. 0.658
4 0.07777 99.8
4 0.0898 99.4
4 0.0869 99.0
44 0.0900 990.6
45 0.0958 990.
46 0.0085 989.8
47 0.06 989.
48 0.6 988.9
49 0.76 988.5
50 0.5 988.0 0.55
5 0.960 987.7
5 0.6 987.
5 0.49 986.7
54 0.500 986.
55 0.574 985.7
56 0.6509 985.
57 0.7 984.7
58 0.846 984.
59 0.905 98.7
60 0.990 98. 0.474
t p
v

C bar kg/m

mm

/s
6 0.086 98.6
6 0.84 98.
6 0.85 98.6
64 0.9 98.
65 0.50 980.5
66 0.64 980.0
67 0.7 979.4
68 0.856 978.8
69 0.98 978.
70 0.6 977.7 0.4
7 0.5 977.
7 0.96 976.6
7 0.54 976.0
74 0.696 975.4
75 0.855 974.8
76 0.409 974.
77 0.489 97.7
78 0.465 97.0
79 0.4547 97.5
80 0.476 97.8 0.65
8 0.49 97.
8 0.5 970.6
8 0.54 969.9
84 0.5557 969.4
85 0.5780 968.7
86 0.600 968.
87 0.649 967.4
88 0.6495 966.7
89 0.6749 966.0
90 0.70 965. 0.6
9 0.78 964.7
9 0.756 964.0
9 0.7849 96.
94 0.846 96.6
95 0.845 96.9
96 0.8769 96.
97 0.9095 960.4
98 0.940 959.8
99 0.9776 959.0
00 .0 958. 0.95
0 .0878 956.8
04 .668 955.5
06 .504 954.0
08 .90 95.6
0 .47 95.0
.56 949.6
4 .66 948.0
6 .7465 946.4
8 .868 944.8
0 .9854 94. 0.460
.44 94.5
4 .50 99.8
6 .9 98.
8 .544 96.5
0 .70 94.8
.8668 9.
4 .040 9.4
6 .4 99.6
8 .47 97.9
40 .64 96. 0.60
t p
v

C bar kg/m

mm

/s
45 4.55 9.7
50 4.760 96.9
55 5.4 9.
60 6.80 907.4 0.890
65 7.008 90.4
70 7.90 897.
75 8.95 89.
80 0.07 886.9 0.697
85 .4 88.4
90 .55 876.0
95 .989 870.
00 5.550 864.7 0.579
05 7.45 858.7
0 9.080 85.8
5 .06 846.6
0 .0 840. 0.488
5 5.504 84.0
0 7.979 87.
5 0.65 80.6
40 .480 8.6 0.40
45 6.54 806.5
50 9.776 799.
55 4.47 79.8
60 46.944 784.0 0.9
65 50.877 775.9
70 55.055 767.9
75 59.487 759.4
80 64.94 750.7 0.79
85 69.76 74.6
90 74.45 7.
95 80.0 7.7
00 85.96 7.5 0.49
05 9. 70.8
0 98.694 690.6
5 05.6 679.
0 .90 667. 0.6
5 0.57 654.0
0 8.64 640.
40 46.08 609.4 0.45
50 65.7 57.4
60 86.74 54.4 0.60
70 0.5 448.4
74. 5.60 6.0 0.490
Density of sea water
= 00 040 kg/m

4
3
3.5.1.1
NPSH
a
for Suction Lift
Operation
For suction lift operation (Fig.
8) the pump is installed above
the suctionside water level. The
value of NPSH
a
can be calcu
lated from the conditions in the
suction tank (index e) as follows
(see Fig. 6)
NPSH
a
= (p
e
+ p
b
p
v
)/( g) + v
e

/g H
L,s
H
s geo
s (29)
where
p
e
Gauge pressure in suction tank in N/m

p
b
Absolute atmospheric pressure in N/m

(Table : consider
effect of altitude!)
p
v
Vapour pressure in N/m

(in Table as absolute pressure!)

Density in kg/m

g Gravitational constant, 9.8 m/s

v
e
Flow velocity in the suction tank or sump in m/s
H
L,s
Head loss in the suction piping in m
H
s geo
Height difference between the fuid level in the suction tank or
sump and the centre of the pump inlet in m
s Height difference between the centre of the pump inlet and
the centre of the impeller inlet in m
p
v
, t, , v
e
p
v
, t, , v
e
s'
H
s geo
Open sump
p
b
p
e
= 0
Closed tank
p
b
+ p
e

Reference plane
Fig. 36: Calculation of the NPSH
a
for suction lift operation for
horizontally or vertically installed pumps
For cold water and open sump
(Fig. 6, on the left) at sea level
this equation can be simplifed
with suffcient accuracy for most
practical purposes to
NPSH
a
= 0 H
L,s
H
s geo
s
(30)
The correction using s is only
necessary when the centre of
the impeller inlet (which is the
decisive location for cavitation
risk) is not at the same height
as the centre of the pump inlet
(= reference plane). In Fig. 6,
H
s geo
must be lengthened for
the pump on the left by the
value s (i.e., same sign for H
s geo

and s!). When s is unknown,
it can usually be estimated with
enough accuracy by examining
the pumps outline drawing.
Table 13: Infuence of the altitude above mean sea level on the an-
nual average atmospheric pressure and on the corresponding boiling
point (1 mbar = 100 Pa)
Altitude above Atmospheric Boiling point
mean sea level pressure p
b
m mbar C
0 0 00
00 989 99
500 955 98
000 899 97
000 795 9
4000 66 87
6000 47 8
NPSH
a
for Suction Lift Operation
44
3
3.5.1.2
NPSH
a
for Suction Head
Operation
For operation with positive inlet
pressure (also called suction
s'
p
v
, t, , v
e
p
v
, t, , v
e
H
z geo
Open sump
p
b
p
e
= 0
Closed tank
p
b
+ p
e

Reference plane
Fig. 37: Calculation of the NPSH
a
for suction head operation for
horizontally or vertically installed pumps
head operation), the pump is
installed below the liquid level.
Eqs. (9) and (0) change by
replacing H
s geo
with +H
z geo

and then read:
NPSH
a
= (p
e
+ p
b
p
v
)/( g) + v
e

/g H
L,s
+ H
z geo
s (31)
where
H
z geo
Height difference between the fuid level in the inlet tank and
the centre of the pump inlet in m
For cold water and open tanks
(Fig. 7, on the left) at sea level
this equation can also be simp
lifed for all practical purposes
to:
NPSH
a
= 0 H
L,s
+ H
z geo
s
(32)
The comments on s as outlined
in section .5.. apply analog
ously.
3.5.2
The NPSH Value of the Pump:
NPSH
r
When the inlet pressure drops,
cavitation bubbles start to de
velop in the pump long before
their effects become apparent in
the hydraulic performance. One
must therefore accept the pres
ence of a small amount of cavi
tation bubbles in order to oper
ate economically. The permis
sible amount of cavitation can
be defned with certain criteria.
Often a head drop of % result
ing from cavitation is accepted.
Fig. 8 shows how this point
is identifed: At a constant fow
rate and constant speed of rota
tion, the NPSH
a
of the test loop
is reduced until the pumps dis
charge head has fallen by %.
Other criteria for the cavitation
limit can also be used, such as
the increase in sound level due
to cavitation, the amount of
material erosion or a certain
reduction in pump effciency.
To avoid impermissible cavita
tion conditions, a minimum
NPSH value is required, which
is shown (in units of m) in the
NPSH
r
curves below the H/Q
1
0
NPSH
r NPSH
Q = const.
n = const.
3%
H
H
non-cavitating
First appearance of
cavitation bubbles
Fig. 38: Experimental determination of the NPSH
r
for the criterion
H = 0.03 H
non-cavitating

NPSH
a
for Suction Head Operation NPSH Required
45
3
characteristics (see Fig. 8). The
reference plane is the centre
of the impeller inlet (Fig. 9),
which can vary by the height s
from the reference plane of the
system, for example for vertical
pumps (see Figs. 6 and 7).
So as to avoid exceeding the
given cavitation limit, it is nec
essary that
NPSH
a
> NPSH
r
(33)
Fig. 40 shows this graphically
at the intersection of the NPSH
a

and NPSH
r
curves. If the NPSH
requirement is not fulflled, the
developed head will quickly de
crease to the right of the inter
section (i.e. at larger fow rates),
which produces a cavitation
breakdown curve. Prolonged
operation under these condi
tions can damage the pump.
0
Sg
0
Sg
0
Sg
0
Sg
0
Sg
0
Sg
H
NPSH
A
1
B
A
2
H
sys
Q
1
Q
2
Q
NPSH
a (2j
NPSH
a (1j
H/Q curve
NPSH
r
Fig. 39: Position of the reference point P
s
for various impellers
Fig. 40: Cavitation breakdown
curves A
1
and A
2
of the H/Q
curve in the case of insuffcient
NPSH
a
: An NPSH defcit exists
in the singly hatched (case 1)
and cross-hatched regions
(case 2). After increasing
NPSH
a
(1) to NPSH
a
(2), the
pumps useful operating range
is increased from Q
1
to Q
2
and
the operating point B can now
be reached.
3.5.3
Corrective Measures
The numerical values of
NPSH
a
and NPSH
r
are based
on the fxed design geometry
of the system and of the pump,
which cannot be changed after
the fact, and on the particular
operating point. It follows that
a subsequent improvement of
the NPSH
a
> NPSH
r
condi
tion in an installed centrifugal
pump system is only possible
with major design and fnancial
expenditure for the pump or
the system. Options include:
increasing H
z geo
or reducing
H
s geo
(by mounting the tank at
NPSH Required Corrective Measures
46
)NDUCER
3
a higher level or installing the
pump at a lower point), mini
mizing the pressure losses in the
inlet piping H
L,s
or replacing
the pump. In the latter case,
using a special lowNPSH suc
tionstage impeller or installing
Fig. 41: Sectional drawing of a pump with an inducer (detail)
Q
opt
Flow rate Q
Pump characteristic curve
NPSH
r
without inducer
N
P
S
H

v
a
l
u
e

o
f

t
h
e

p
u
m
p
D
e
v
e
l
o
p
e
d

h
e
a
d

H

o
f

t
h
e

p
u
m
p
NPSH
r
with inducer
Fig. 42:
Effect of an inducer on the
NPSH
r
an inducer (propeller in front of
the impeller, Fig. 4) can keep
the costs of the improvement
within limits (a rebuild of the
pump is unavoidable, however).
It must be kept in mind that the
NPSH
r
reduction by the inducer
does not have an impact on the
entire fow range of the pump in
question, but only on a certain
part of the range (see Fig. 4).
The resistance to cavitation ero
sion can be increased by choos
ing more suitable (and more
expensive) materials for the
impeller, in particular for larger
pump sizes.
In one special case, the elimina
tion of an NPSH problem is
quite simple: For closed fow
loops (for example in a heating
system), the system pressure can
simply be increased to improve
the NPSH
a
, as long as the sys
tem is designed to cope with the
higher pressure.
NPSH Required Corrective Measures
47
3
3.6
Effect of Entrained Solids
If the water to be pumped (for
example, domestic waste water,
rainwater or mixtures) contains
small amounts of entrained
solids, special impeller and
pump types are used (for ex
ample with cleaning covers or
special shaft seals) [].
Fig. 4 shows the most common
impeller designs for these types
of waste water. For pumping
sludge, nonclogging channel
impellers can be used up to
% solids content, single vane
impellers up to 5%, free fow
impellers up to 7% and worm
type impellers for even higher
concentrations.
Since single vane impellers can
not be turned down to adjust
the operating point (see section
.4.6), this type of pump is
often driven using a belt drive
(see Fig. 59g).
Allowances added to the drive
power are not shown in Fig.
0, but rather in the product
literature [], since they depend
not only on the drive rating but
also on the impeller design and
specifc speed. For example, for
single vane impellers pumping
domestic waste water or sewage
the following power reserves are
recommended:
up to 7.5 kW approx. 0%
(kW)
from kW approx. 0%
from 0 55 kW approx. 5%
above 55 kW approx. 0%
When assessing the head losses
in the piping (see section ...),
adequate allowances have to be
added [].
To avoid blockages in the pipes
for waste water with high solids
concentrations, a minimum fow
velocity of . m/s in horizon
tal pipes and m/s in vertical
pumps should be maintained.
(Exact values can only be de
termined experimentally!). This
is of particular importance for
variable speed drives [].
Effect of Entrained Solids Impeller Types for Pumping Waste Water
geschlossenes Einschaufelrad *) fr
Abwasser mit festen oder langfaserigen
Beimengungen
geschlossenes Kanalrad *) fr feststoff
haltige oder schlammige nicht gasende
Flssigkeiten ohne langfaserige Bei
mengungen
Schneckenrad fr Abwasser mit groben, festen
oder langfaserigen Feststoffen oder fr
Schlmme mit 5 bis 8% Trockensubstanz
Freistromrad fr Flssigkeiten mit groben
oder langfaserigen Feststoffen und Gas
einschlssen
Impeller Types for Pumping Waste Water
Fig. 43a: Closed single vane
impeller for waste water con-
taining solid or stringy sub-
stances
Front view shown without shroud
Fig. 43b: Closed non-clogging
channel impeller for sludge or
non-gassing liquids containing
solids without stringy compo-
nents
Fig. 43c: Free fow impeller
for fuids with coarse or stringy
solids and gas content
Fig. 43d: Worm type impeller
for waste water containing coarse,
solid or stringy substances or for
sludge with up to 5 to 8% solids
content
Fig. 43e: Diagonal impeller for
waste water containing solid,
stringy or coarse substances
Front view shown without shroud
48
4
4
Special Issues when
Pumping Viscous Fluids
4.1
The Shear Curve
Viscosity is that property of a
fuid by virtue of which it of
fers resistance to shear. Fig. 44
shows this process. In a fuid,
a plate with a wetted surface
area A is moved with speed v
0

parallel to a stationary wall at a
distance y
0
.
The movement requires that
the resistance force F be over
come, which can be expressed
as a shear stress = F/A. If the
wall distance y
0
, the velocity v
0

or the type of fuid is changed,
then the shear stress also
changes in proportion to the
velocity v
0
or inversely propor
tional to the distance y
0
. The
two easily identifed parameters
v
0
and y
0
are combined to yield
the shear gradient v
0
/y
0
.
Since the viscosity of the fuid
causes a shear stress not only
at the walls, but rather at every
distance from the wall within
the fuid, the defnition of the
rate of shear is generalized as
v/y (change of velocity per
change of distance). Just as for
the shear stress , it is not the
same for all wall distances y.
During an experiment, pairs of
values and v/y are measured
and can be plotted as a func
tion, the socalled shear curve
(Fig. 45).
When the shear curve is a
straight line going through the
origin:
= v/y (34)
v
0
y
0
Plate
Wall
F
v/y
Fig. 44: Velocity profle between
a plane wall and a moving par-
allel plate.
F = Towing force
v
0
= Towing speed
y
0
= Distance to wall
v/y = Rate of shear
D
B
S
D
a b
N
S
T
B
= T
f
+ H
uv
uy
T
N
= H
uv
uy
uv / uy 0 uv / uy 0
T
T

T
f
Plastic limit T
f
Fig. 45: Overview of the shear behaviour of viscous fuids
a without, b with a plastic shear limit
f
N Newtonian, B Bingham, S pseudo-plastic, D dilatant fuids
the constant factor of propor
tionality is referred to as the
dynamic viscosity with the units
Pa s. Fluids with this type of
curve (for example water or all
mineral oils) are normally vis
cous or Newtonian fuids, for
which the laws of hydrodyna
mics apply without restriction.
If the shear curve is not a
straight line through the origin,
the fuid is a nonNewtonian
fuid, for which the laws of
hydrodynamics apply only in
a limited fashion. One must
therefore strictly differentiate
between these two cases.
Since the quotient of dynamic
viscosity and density is often
used in fuid dynamic relation
Pumping Viscous Fluids Shear Curve
49
4
ships it is defned as the kine
matic viscosity
= / (5)
where
Kinematic viscosity in m

/s
Dynamic viscosity in Pa s
(= kg/sm)
Density in kg/m

(for numeri
cal values see Fig. 48)
For water at 0C, = .00
0
6
m

/s. For further numerical

values see Table . The units
centistokes = mm

/s, degrees
Engler E, Saybolt seconds S"
(USA) and Redwood seconds
R" (UK) are no longer used and
can be converted to m

/s using
Fig. 46.
Independently of the discussion
above, viscosity varies with
K
i
n
e
m
a
t
i
c

v
i
s
c
o
s
i
t
y

N

E
n
g
l
e
r
,

S
a
y
b
o
l
t

s
,

R
e
d
w
o
o
d

s
,

B
a
r
b
e
y

c
m
3
/
h

1
2
2
4
4
6
6
8
8
10
2
4
6
8
10
2
10
-6
2 4 6 8 10
-5
2 4 6 8 10
-4
2 4 6 8 10
-3
10
-2
2
4
6
8
10
3
2
4
6
8
10
4
m
2
s
B
a
r
b
e
y

c
m
3
/
h

D
e
g
r
e
e
s

E
n
g
l
e
r
R
e
d
w
o
o
d

s
e
c
o
n
d
s
S
a
y
b
o
l
t

s
e
c
o
n
d
s
Kinematic viscosity N in
Fig. 46: Conversion between
various units of kinematic vis-
cosity
Values required to
DIN 51507 (transformer oils)
DIN 51603 (fuel oils)
DIN 51601 (Diesel fuel)
ISO viscosity classification
to DIN 51519
G
e
a
r o
ils
A
u
to
m
o
b
ile
g
e
a
r o
ils
H
y
d
ra
u
lic
o
ils
H
L
, H
L
P
L
u
b
ric
a
tin
g
o
ils
L
-A
N
L
u
b
ric
a
tin
g
o
ils
C
L
L
u
b
ric
a
tin
g
o
ils
C
L
u
b
ric
a
tin
g
o
ils
C
L
P
D
ie
s
e
l fu
e
l
F
u
e
l
o
ils

E
L
F
u
e
l
o
ils

L
F
u
e
l
o
ils

M
F
u
e
l
o
ils

S
S
u
p
e
r
h
e
a
t
e
d
s
t
e
a
m
c
y
lin
d
e
r
o
ils
Z
D
S
u
p
e
r
h
e
a
t
e
d

s
t
e
a
m

c
y
lin
d
e
r

o
ils

Z
A
S
u
p
e
r
h
e
a
t
e
d

s
t
e
a
m

c
y
lin
d
e
r

o
ils

Z
B
T
r
a
n
s
f
o
r
m
e
r

o
ils
M
o
to
r o
ils
T
u
rb
in
e
o
ils
T
D
V
B
, V
B
L
C
o
m
p
re
s
s
o
r o
ils
V
C
, V
C
L
, V
D
L
,
R
e
frig
e
ra
tin
g
m
a
c
h
in
e
o
ils
K
C
R
e
frig
e
ra
tin
g
m
a
c
h
in
e
o
ils
K
A
Lubricating
oils BC
Lubricating
oils BB
Lubricating
oils BA
1500
1000
600
500
400
300
200
60
50
40
30
20
10
15
mm
2
/s
100
80
6
5
4
3
0 50 100 150 C 200
8
Temperature t
K
i
n
e
m
a
t
i
c

v
i
s
c
o
s
i
t
y

N
Fig. 47: Kinematic viscosity of various mineral oils as a function of
the temperature (enlarged view on page 85)
temperature: at higher tempera
tures almost all liquids become
thinner; their viscosity de
creases (Figs. 47 and 48).
The dynamic viscosity can be
measured for all liquids using a
rotating viscometer to determine
the shear curve. A cylinder ro
tates with a freely chosen speed
in a cylindrical container flled
with the liquid in question. The
required driving torque is meas
ured at various speeds along
with the peripheral speed, the
size of the wetted area and the
distance of the cylinder from the
wall.
Pumping Viscous Fluids
50
4
4.2
Newtonian Fluids
4.2.1
Infuence on the Pump
Characteristics
The characteristic curves of a
centrifugal pump (H, and P
as functions of Q) only start
to change perceptibly at vis
cosities above > 0 0
6
m

/s
and only need to be corrected
with empirical conversion fac
tors above this limit. The two
E
th
yl a
lc
o
h
o
l (m
e
th
yl a
lc
o
h
o
lj
A
c
e
to
n
e
Diethyl ether
t = 100 C
N = 2.01 mm
2
/s
t = 98.3 84.2 72.5 44.5 C
N = 15.8 7.76 4.99 2.33 mm
2
/s
t = 92.5 C
N = 2.35 mm
2
/s
t =18.3 50 70 C
N = 11.87 3.32 1.95 mm
2
/s
1.8
1.5
1.0
0.5
0
kg
dm
3
D
e
n
s
i
t
y

C
a
r
b
o
n

t
e
t
r
a
c
h
lo
r
id
e
B
enzol
n-Butane
i-B
u
ta
n
e
E
th
a
n
e
P
ro
p
a
n
e
A
m
m
o
n
ia
A
m
m
o
n
i
a
D
ie
th
y
l e
th
e
r
B
e
n
z
o
l
To
lu
o
l
Aniline
P
h
e
n
o
l
A
n
i
l
i
n
e
F
o
rm
ic

a
c
id
F
o
r
m
i
c

a
c
i
d
Heavy w
ater
P
h
en
o
l
A
c
e
tic
a
c
id
M
e
t
h
a
n
o
l
A
c
e
t
o
n
e
C
a
rb
o
n
d
is
u
lp
h
id
e
C
a
r
b
o
n

d
is
u
lp
h
ide
T
o
l
u
o
l
160 100 C 0
Temperature t
-100
1.8
1.5
1.0
0.5
0
mm
2
s
K
i
n
e
m
a
t
i
c

v
i
s
c
o
s
i
t
y

N
A
c
e
t
i
c

a
c
i
d
S
u
lp
h
u
r
ic

a
c
id
Fig. 48: Density
and kinematic vis-
cosity of various
fuids as a function
of the temperature
t (enlarged view
on p. 86)
most well known methods are
that described in the Hydraulic
Institute (HI) Standards and
that of KSB. Both methods use
diagrams containing the conver
sion factors which are applied
in a similar manner, but differ
in that the KSB method not only
includes the parameters Q, H
and but also the signifcant
infuence of the specifc speed
n
q
(see section ..5). The HI
method (Fig. 49) is based on
measurements with n
q
= 5 to
0 and gives the same numerical
results as the KSB method
(Fig. 50) in this narrow range.
The KSB method is based on
measurements with n
q
from
6.5 to 45 and viscosity up to

z
= 4000 0
6
m

/s. The use

of both diagrams is explained
with the examples shown in
them [9].
The fow rate Q, the total
developed head H and the ef
Newtonian Fluids Viscosity and Pump Characteristics
5
4
1.0
0.8
0.6
1.0
0.8
0.6
0.4
4
0.2
4
8
6
1
0
3
0
6
0
2
0
1
5
25
25 50 100 200 500
500 300 200 100 50 l/s 30 40 20 10 6
m
3
/h 1000 2000
80
150
4
0
1
0
0
2
0
0
1
.
0

1
0
-
6
1
0

1
0
-
6
1
0
0

1
0
-
6
1
0
0
0

1
0
-
6
4
0
0
0

1
0
-
6
2
.
0
2
0
3
0
4
0
6
0
8
0
2
0
0
3
0
0
2
0
0
0
3
0
0
0
4
0
0
6
0
0
8
0
0
m
2
/
s
3
.
0
4
.
0
6
.
0
8
.
0
k
Q
k
H
k
H
0.6 Q
0.8 Q
1.0 Q
1.2 Q
1 Flow rate Q
H
e
a
d

H
K
in
e
m
a
t
ic

v
is
c
o
s
it
y

N
C
o
n
v
e
r
s
i
o
n

f
a
c
t
o
r

k
H
C
o
n
v
e
r
s
i
o
n

f
a
c
t
o
r
s

k
Q
.

k
H
3
2
Fig. 49: Determination of the conversion factors k using the Hydraulic Institute method. Example
shown for Q = 200 m
3
/h, H = 57.5 m, = 500 10
6
m
2
/s
Viscosity and Pump Characteristics Conversion Factors
5
Curve No. 1 2 3 4 5
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0.4
0.5
0.6
0.7
0.8
0.9
1.0
f
H
fH
f
Q
nq, W = 45
30
20
10
10
20
30
6.5
6.5
0.5
0.6
0.7
0.8
0.9
1.0
f
H
fH
nq, W = 45
6
0
0
0
3
5
0
0 2
9
0
0
2
0
0
0 1
7
5
0 1
4
5
0 1
1
6
0
9
6
0
8
7
5
7
2
5
5
0
0
mm
2
s
1 1.5 2 3 4 5 6 8 10 15
4000
5000 3000 2000 1000
2000 1000
500 300 400 200 100 50 30 40 20 10 5 3 4
0.5 0.3 0.4
2 1
500 300 400 200 100 50 30 40 20 10 5 3 4 2 1
10000
3000
2500
2000
1500
1000
800
600
500
400
300
250
200
150
100
80
60
50
40
30 20
300
400
200
150
100
80
60
50
40
30
20
15
10
8
6
5
4
3
2
1.5
1
m
3
/h
l/s
nq, w = 5 10 15 20 25 30 35 40 45
Curve No. 1 2 3 4 5 5 4 3 1
4
3
2
1
H
e
a
d

H
W
.

o
p
t

i
n

m
Kinematic viscosity N
z
in
Flow rate Q
W. opt
S
p
e
e
d

i
n

r
p
m
Fig. 50: Determination of the conversion factors f using the KSB method. Example shown for
Q = 200 m
3
/h, H = 57.5 m, n = 1450 rpm, = 500 10
6
m
2
/s, n = 2900 rpm, n
q
= 32.8
4 Viscosity and Pump Characteristics Conversion Factors
5
fciency , which are known for
a singlestage centrifugal pump
operating with water (subscript
w), can be converted to the val
ues for operation with a viscous
medium (subscript z) as follows:
Q
z
= f
Q
Q
w
(36)
H
z
= f
H
H
w
(37)

z
= f

w
(38)
The factors f are designated
with k in the HI method; both
are shown graphically in Figs.
49 and 50. In Fig. 50 the speed
of rotation n of the pump must
be considered in the diagram
and the specifc speed n
q
of the
pump impeller must be known,
for example from Fig. or
Eq. .
With these factors the known
performance for water can be
converted to refect operation
with a viscous fuid. The con
version is valid for the range
0.8 Q
opt
< Q < . Q
opt
(39)
A simple calculation can thus be
done for three fow rates, with a
single exception:
At Q = 0.8 Q
opt
,
H
z
= .0 f
H
H
w
applies
(but H
z
never is > H
w
!).
At fow rate Q = 0, simply set
H
z
= H
w
and
z
=
w
= 0. A
worksheet or spreadsheet calcu
lation as shown in Fig. 5 can
simplify the conversion.
After the power is calculated at
the three fow rates (in the fow
range according to Eq. 9) using
(OWTO&INDTHE/PERATING0OINT
'IVEN
Flow rate Q
w
m
3
/h
Total developed head H
w
m
Speed of rotation n 1/min
Kinematic viscosity N
z
m
2
/s
Density
z
kg/m
3
Gravitational constant g 9.81 m/s
2
#ALCULATION
Q/Q
opt
= 0 0.8 1.0 1.2 -
Q
w
0 m
3
/h
H
w
m
H
w
0 -
n
q, w
from section - - - 1/min
3.1.5
f
Q, w
from Fig. 50 - -
f
H, w
- -
f
H, w
- -
Q
z
= Q
w
f
Q, w
0 m
3
/h
H
z
=

H
z
H
w
f
H, w
0
P
z
kW
1j
lf H
z
becomes larger than H
w
, it should be set to H
w
.
2)
These are four values on the H
z
/Q and Q/H
z
curve and three points on the
Q/P
z
curve. Plot versus Q.

z
g H
z
Q
z
H
z
1000 3600
= H
w
= H
w
f
H, w
1,03 H
w
f
H, w
H
w
f
H, w
1)
m
from the
pump cha-
racteristic
for 4 points
2)
Fig. 51: Spreadsheet for calculating the pump characteristics for a
viscous fuid using the KSB method (enlarged view on p. 87)
4
P
z
=
z
g H
z
Q
z
/ 000
z
(40)
where

z
Density in kg/m

Q
z
Flow rate in m

/s
g Gravitational constant
9.8 m/s

H
z
Total developed head in m

z
Effciency between 0 und
P
z
Power in kW (!)
all the characteristic curves can
be plotted over Q
z
using the
or 4 calculated points, as shown
in Fig. 5 on page 54.
For the inverse problem, i.e.
when the operating point for the
viscous fuid is known and the
values for water are to be found
(for example when choosing a
suitable pump for the requested
operating point), the water val
ues are estimated and the solu
Viscosity and Pump Characteristics Conversion Factors Conversion
54
4
tion is approached iteratively
using f
Q
, f
H
and f

in two (or
sometimes three) steps.
For specifc speeds above n
q
20
the more realistic KSB method
results in smaller power require-
ments; below this limit the cal-
culated required driving power
according to HI is too small [9]!
4.2.2
Infuence on the System
Characteristics
Since the laws of fuid dynam
ics retain their validity for all
Newtonian fuids, the equations
and diagrams for calculating
the pipe friction factor and the
loss coeffcients for valves and
fttings are also applicable to
viscous media. One must simply
substitute the kinematic viscos
ity of the viscous liquid
z
for
the water viscosity
w
when cal
culating the Reynolds number
Re = v d/. This yields a lower
Reynolds number, and a larger
friction factor
z
results from
Fig. 0. (Note: the infuence
of the wall roughness can now
often be ignored because of the
larger boundary layer thickness
in the fow.) All of the pressure
losses in the pipes, valves and
fttings calculated for water in
accordance with section ...
are to be increased by the ratio

z
/
w
.
Fig. 5 is also suitable for
general practical use: the dia
gram provides a fast way of de
termining the pipe friction fac
tor
z
as a function of the fow
rate Q, pipe inside diameter d
and kinematic viscosity
z
. It
must be kept in mind, however,
that the coeffcient
w
for water
in this diagram is only valid for
hydraulically smooth pipes (i.e.
not for roughsurfaced pipes)!
The corresponding
w
can be
used to calculate the ratio
z
/
w
.
Since the static component of
the system characteristic curve
H
sys
(Fig. 6) is not affected by
viscosity, the dynamic com
ponent of the system character
istic for water can be redrawn
as a steeper parabola for a vis
cous fuid.
4.3
Non-Newtonian Fluids
4.3.1
Infuence on the Pump
Characteristics
Since the local velocity gradients
in all the hydraulic components
of a pump are not known, a cal
Fig. 52: Conversion of the pump characteristics for water to that for
a viscous fuid
60
70
50
40
30
20
10
0 50 100 150 200 250
0
60
70
80
50
40
30
20
10
0
40
30
50
20
10
0
H
m
m
3
/h Q
0 50 100 150 200 250 m
3
/h Q
0 50 100 150 200 250 m
3
/h Q
H

P
kW
H
z
H
zopt
H
w
H
wopt
P
z
P
w
H
z
H
w
Q
z
o
p
t
Q
w
o
p
t
H
wopt
H
zopt
0.8 1.0 1.2 Q/Q
opt
P
zopt
P
wopt
Viscosity and Pump/System Characteristics Non-Newtonian Fluids: Pump Characteristics
55
4
8
0
0
6
0
0
4
0
0
3
0
0
2
0
0
1
5
0
1
0
0
0
1
2
5
m
m
1
0
0
8
0
6
5
5
0
4
0
3
2
2
5
P
i
p
e

i
n
s
i
d
e

d
i
a
m
e
t
e
r

d
1
2
4
6
8
1
0
1
4
2
m
3
/
h
6
8
1
0
2
4
2
6
8
1
0
3
4
2
6
8
1
0
4
0.010
0.011
0.012
0.013
0.014
0.015
0.016
0.017
0.018
0.020
0.022
0.024
0.026
0.028
0.030
0.035
0.040
0.03
0.035
0.04
0.05
0.06
0.08
0.10
0.12
0.15
0.20
0.30
0.40
0.50
0.60
1
0 -
3
1
0 -
5
1
0 -
6
m
2
/s 5
5
2
1
0 -
4
5
2
2
P
i
p
e

f
r
i
c
t
i
o
n

f
a
c
t
o
r

L
z
T
u
r
b
u
l
e
n
t

f
l
o
w
P
ip
e

f
r
ic
t
io
n

f
a
c
t
o
r

L
z
L
a
m
in
a
r

f
lo
w
K
i
n
e
m
a
t
i
c

v
i
s
c
o
s
i
t
y

N
W
a
t
e
r

in

h
y
d
r
a
u
lic
a
lly

s
m
o
o
t
h

p
ip
e
P
ip
e
in
s
id
e
d
ia
m
e
te
r d
F
l
o
w

r
a
t
e

Q
Fig. 53: Finding the pipe friction factor
z
for viscous liquids
Example: Q = 200 m
3
/h; d = 210 mm;
z
= 5 10
4
m
2
/s
4.3.2
Infuence on the System
Characteristics
When the shear curves are not
straight lines of constant linear
viscosity, one must divide them
into sections and determine the
coeffcient (= stiffness number)
and the exponent n (= structural
number) for each section indi
vidually (easiest when plotted
on doublelogarithmic scales).
Using a special diagram (analo
gous to Fig. 0), which shows
the pipe friction factors
z
as a
function of the generalized
Reynolds number Re
n
for vari
ous exponents n, the value of

z
can be read and the system
curve H
sys
determined for a par
ticular fow rate Q. Since this
process is very laborious, in par
ticular because of the need for
multiple iterations, it cannot be
recommended for general use.
Just as for the pump character
istics, in most cases diagrams
with a narrow range of applica
tion based on experience with
a particular fuid are used to
fnd the head loss H
L
. The more
the application differs from the
particular conditions of the dia
gram, the more uncertain will
the head loss analysis become,
so that in such cases the experi
ence of the design department
must be tapped.
culation of the infuence of non
Newtonian fuids on the pump
characteristics is not generally
possible. Only for a limited
number of special fuids, such as
fbre pulp, is a prediction based
Non-Newtonian Fluids Pump/System Characteristics
on knowledge gained during
years of experience with this
fuid feasible. The selection of a
suitable pump must therefore be
done by the design department.
56
5
5
Special Issues when Pumping
Gas-laden Fluids
Unlike dissolved gases, a non
dissolved gas in a liquid (ex
pressed as a volume percentage)
can change the design para
meters, the characteristic curve
and the general performance of
a centrifugal pump dramatically,
as shown in Fig. 54 using a non
clogging impeller pump as an
example. The gas content may
be caused by the production
process itself but also by leaking
fanges or valve stems in the suc
tion line or from airentraining
vortices in an open sump inlet
when the water level is too low
(see section 7.).
0
0
0
0
10
50
10
20
m
26
80
%
kW
14
100 200 m
3
/h 300 340
q
L
= 0%
q
L
= 0%
q
L
= 0%. 2.8%
2.8%
4.1%
5.5%
2.8%
4.1%
4.1%. 5.5%
5.5%
5.5%
6.9%
8.3%
9.6%
8.3%
9.6%
2.8%
0%
11%
11%
11%
6.9%
2.8%
0%
6.9%
8.3%
9.6%
Total suction-side flow rate Q
s
P
o
w
e
r

P
E
f
f
i
c
i
e
n
c
y

H
H
e
a
d

H
Fig. 54: Infuence of non-dissolved air on the operation of a non-
clogging impeller pump when pumping pre-treated sewage (open
three-channel impeller, D = 250 mm, n = 1450 rpm, n
q
= 37)
q
air
= Gas volume in suction piping as % of the mixture.
In the centrifugal force feld
of an impeller, the gas bubbles
tend to accumulate in certain lo
cations and to disturb the fow
there. This effect is reinforced
the more the pump operates
at a reduced fow rate, since
the lower velocities exert less
carrying force on the gas,
the smaller the impeller inlet
diameter is, since the throt
tling effect of the gas volume
is increased,
the lower the specifc speed n
q
of the pump impeller is, and
the lower the speed of rota
tion of the pump is.
These effects cannot be cal
culated. When signifcant gas
volumes are expected in the
pumpage, the following meas
ures can be useful:
A suffciently large settling
tank in the suction line can
allow the gas to separate out
of the fuid and thus mitigate
its disturbing effects.
Pipes which are used to fll
an open inlet sump must end
below the liquid level so that
there is no free fall of water
that might entrain air bub
bles in the tank. In addition, a
baffe can prevent the entry of
vortices in the suction piping
(see Figs. 64 and 65).
Lowfow operation of the
main pump can be prevented
by installing a special part
load pump. When this pump
is only needed occasionally,
it is advantageous to use a
selfpriming pump (whose ef
fciency is lower, though).
A gas removal line in front
of the impeller hub requires
a vacuum system, is only of
limited use for large gas quan
tities and disturbs the normal
operation of the pump.
In the pump, open impellers
(see Fig. 4) with few vanes are
advantageous, as is the instal
lation of an inducer (Fig. 4).
Without any special precau
tions, nonclogging impeller
pumps (Fig. 4) can pump up
to %
vol
and free fow impel
lers up to 6 to 7%
vol
of gas.
If a large gas content is to be
expected under normal oper
ating conditions, side channel
pumps or waterring pumps
(positive displacement prin
ciple) operate more reliably.
Pumping Gas-laden Fluids
57
6
Special Issues When Pumping
Solids-laden Fluids
6.1
Settling Speed
Solids (which are heavier than
water) are most easiest to pump
when their settling speed is low
est and their fow velocity high
est. Because of the large number
of infuencing parameters, the
settling speed can only be calcu
lated based on simplifying as
sumptions: the settling speed of
a single sphere in an unlimited
space (subscript 0) results from
force equilibrium as
4 g d
s

s

f

w
s0
=
c
D

f

(41)
where
w
s0
Settling speed in m/s
g Gravitational constant
9.8 m/s
d
s
Diameter of sphere in m
c
D
Resistance coeffcient of the
sphere dependent on Re
s

s
Density of the solid in kg/m

f
Density of the fuid in kg/m

Re
s
= w
s0
d
s
/
f
(42)
where

f
Kinematic viscosity of the
liquid in Pa s
The settling speed w
s0
is shown
graphically in Fig. 55.
.2 .3 .4 .5.6 .8
0.2
0.3
0.5
0.4
0.7
10
1.0
0.1
0.1 1.0 10 100
w
s
o

[
m
/
s
|
d
s
[mm|
Water
t = 15 C
R
e
s

=

1
0
5
1
0
4
1
0
3
1
0
2
4
0
0
0
5
0
0
0
D
s
=
1
5
0
0
k
g
/m
3
2
0
0
0
2
5
0
0
3
0
0
0
3
5
0
0
6
0
0
0
7
0
0
0
8
0
0
0
The solids concentration:
c
T
= Q
s
/(Q
s
+ Q
f
) (43)
has a large effect, where
c
T
Flowbased solids concen
tration (transport concen
tration)
Q
s
Flow rate of the solid in
m

/s
Q
f
Flow rate of the fuid in
m

/s
The solids concentration and
the boundary effect of the pipe
walls reduce the settling speed
considerably because of the mu
tual repulsion of the particles,
approximately according to the
following empirical relationship
w
s
= w
s0
( c
T
)
5
(44)
The effect of an irregular par
ticle shape cannot be estimated;
the shape may differ substan
tially from that of a sphere.
The effect of the particle size
distribution can also hardly be
calculated. Fig. 56 shows an
example of the distribution of
Fig. 55: Settling speed w
s0
of individual spherical particles (spherical
diameter d
s
) in still water
Pumping Solids-laden Fluids Settling Speed 6
100%
90
80
70
60
50
40
30
20
10
0.001 0.01 0.1 1 10 mm
d
50
d
s
Mass fraction
Fig. 56: Example of a particle size distribution
58
particle sizes d
s
plotted logarith
mically for the portion which
passed through a sieve of a given
mesh size. Transported solids
are almost always composed of
particles of various sizes, so that
the size distribution has a more
or less distinct Sshape. To sim
plify the analysis, it can be as
sumed that the particle size for
50% mass fraction, designated
d
50
, is representative of the
mixture. This assumption is the
most important source of dis
parities in the planning phase.
After all these assumptions
and gross approximations, no
exact predictions of the effects
of solids on the fow behaviour,
the system curve, the total de
veloped head and the effciency
of pumps are to be expected.
The design and selection of
pumps for solids transport must
therefore be left to experts who
have suffcient experience with
similar cases. Even then, ex
periments are often necessary
to attain a measure of certainty.
Only certain general tendencies
can be stated.
6.2
Infuence on the Pump
Characteristics
The solids behave differently
under the infuence of the cen
trifugal force feld in an impeller
than the carrier fuid (usually
water) does. The solids cross the
streamlines or collide with and
rub against the walls of the fow
passages. They thus reduce the
head H produced in the impeller
by the difference H.
Experimental data exist on the
effects of the particle diameter
d
s
, the concentration c
T
and the
density
s
of the solids as well
as the specifc speed n
q
. The
H/H = c
T
/

Re
s
(.8/n
q
)

(
s
/
f
) (45)
where
c
T
Transport concentration according to Eq. 4
Head coeffcient of the pump; here approx. =
Re
s
Reynolds number of the solids fow according to Eq. 4
n
q
Specifc speed of the pump according to Eq.

s
Density of the solid in kg/m

f
Density of the fuid in kg/m

m
= c
T

s
+ ( c
T
)
w
(47)
where

m
Average density in kg/m

s
Density of the solid in
kg/m

w
Density of water in kg/m

c
T
Transport concentration ac
cording to Eq. 4
Since the pressure rise in the
pump is the product of the
density and the developed head
(which is reduced when trans
porting solids), two independent
infuences are at work in Eq. 46:
the increased average density
due to the presence of the sol
ids, and the reduced developed
head (H H). Both changes
are caused by the solids concen
tration, but they have opposite
effects, since the density raises
the pressure while the head def
cit decreases it. Therefore, no
general prediction can be made
as to whether the pump pres
sure rise will be higher or lower
than the water curve when the
solids concentration increases.
Heavy, smallgrained solids
(for example ores) are likely to
Pumping Solids-laden Fluids Pump Characteristics
6
When conveying solids hydrau
lically, the pump characteristic
curve needs to be shown as de
veloped pressure p versus fow
rate Q, not as developed head
H, since the average density
m

of the solids / water mixture (in
contrast to pumping clean wa
ter) is not constant. As simpli
fcations, the geodetic head dif
ference z
s,d
between the pump
inlet and discharge as well as
the velocity head difference
(c
d

c
s

)/ g are ignored, i.e.,

the static head is set to equal the
total head (H
p
H):
p =
m
g (H H) (46)
where

m
Average density of the solids /
water mixture given by
Eq. 47 in kg/m

g Gravitational constant
9.8 m/s

H Total developed head in m

H Head reduction according
to Eq. 45 in m
p Pressure in N/m

(to convert to bar:
bar = 00 000 N/m

)
The average density of the mix
ture is given by
empirical relationship for the
relative head reduction H/H is
approximately
59
produce an increase, while light,
large solids (for example coal)
and low specifc speeds tend to
decrease the pressure.
6.3
Infuence on the System
Characteristics
When the fow velocity drops,
solids tend to settle to the bot
tom of horizontal pipe runs and
collect on the pipe wall. The
fow resistance increases and
the free fow passage becomes
smaller, so that despite the
decreasing fow rate, the fow
resistance can actually increase.
This results in the unusual
shape of the system curves as
shown in Fig. 57. The minimum
in the curves measured at vari
ous concentrations is a sure sign
that a solids accumulation is
taking place and that the pipes
will soon be clogged. The curve
minimum is therefore generally
considered to be the lower limit
of operation. Exact predictions
are only possible with suffcient
experience or by experiment.
6.4
Operating Performance
Fig. 57 shows the typical be
haviour of a centrifugal pump
transporting solids through a
horizontal pipe: with increasing
concentration, the intersection
of the pump and system charac
teristic curves shifts to ever low
er fow rates, so that the lower
limit of operation could be ex
ceeded. To avoid this, a control
system must intervene promptly.
Throttle valves would be subject
to high wear, however, so only
a change of rotational speed
remains as a feasible control
30%
20%
10%
0%
c
TA
20%
10%
0%
c
TP B
0
B
10
B
20
p
P
p
sys
Operating limit
B
0, 10, 20
Steady-state operating points
(Clean liquidj
(Clean liquidj
D
e
v
e
l
o
p
e
d

p
r
e
s
s
u
r
e

$
p
P

o
f

t
h
e

p
u
m
p
P
r
e
s
s
u
r
e

l
o
s
s

$
p
s
y
s

i
n

t
h
e

s
y
s
t
e
m
Flow rate Q
method for the hydraulic trans
port of solids. Speed control has
an additional advantage: when
the head developed by the pump
impeller drops as the impeller
wears, it is possible to compen
sate by merely increasing the
speed.
In vertical pipes, the settling of
the solids poses much greater
risk, since the pipe can suddenly
become plugged if the fow falls
below the minimum required,
even if only due to stopping the
pump.
The high erosion rates when
pumping granular solids are
the decisive parameter for the
design of the pumps used. An
example of their typical robust
design is shown in Fig. 58. The
risk of erosion also limits the
permissible operating range to
near Q
opt
.
High solids concentrations put
constraints on the use of centri
fugal pumps; the limit values can
only be found by experience.
The above considerations should
have convinced the reader that
the selection of pumps for
hydraulic solids transport is
risky without a solid base of
experience and should be left to
experts who do this frequently!
6.5
Stringy, Fibrous Solids
If long, stringy solids are present
in the fow, problems can oc
cur, in particular for axial fow
(propeller) pumps, when these
materials (plant fbres, plastic
sheets and rags for example) are
caught on the propeller blade
leading edge and accumulate
Pump Characteristics System Characteristics Operating Performance Stringy, Fibrous Solids
6
Fig. 57: Pressure developed by the pump p
P
and pressure losses of
the system p
sys
for various solids concentrations (concentrations
c
Tsys
, c
TP
) of the fow Q. The developed pressure p
P
= f(c
T
) can
also increase with increasing concentration c
TP
for solids with high
density (here shown decreasing for 10 and 20%).
60
Fig. 58: Typical centrifugal
pump for the hydraulic trans-
port of solids
there. The consequences are an
increasing head loss and power
input, which continue until the
driving motor must be stopped
due to overloading.
The problem can be solved by
slanting the leading edges of the
propeller blades backwards by
shifting the individual profle
planes during blade design, just
as for a backswept airfoil. Dur
ing operation, the fbres can
slide along the blade leading
edge until they are shredded in
the clearance gap at the outside
Solids-laden Fluids Operating Performance Stringy, Fibrous Solids
7
Figs. 59 a to o: Typical installation arrangements
a b c
f g h
k l m
diameter of the propeller and
fushed out. These selfcleaning
blades are called ECBs
(= ever clean blades) [5].
Untreated municipal sewage
often contains textiles which
tend to form braids and plug
impellers with multiple vanes or
other fowdividing devices.
Single vane impellers, worm
type (screw) impellers, or free
fow impellers (see Fig. 4) are
the better choice for these ap
plications.
6
6
7
The Periphery
7.1
Pump Installation
Arrangements
Pump installation arrangements
are design features in which
pumps of the same type (in
general of the same series) may
differ. Figures 59 a to o provide
typical examples of the most
frequent installation arrange
ments for horizontal and verti
cal centrifugal pumps [].
The major parameters classify
ing the pump installation ar
rangement are:
the position of the shaft, i. e.
horizontal or vertical (see
Figs. a and b, also i and c or h
and f),
the arrangement of the feet,
i. e. underneath or shaft
centreline (see Figs. d and e),
the mode of installation of the
pump set, i. e. with or without
foundation (see Figs. b and f),
the arrangement of the drive,
i. e. on its own or a common
baseplate or fanged to the
pump (see Figs. g, a, h and i),
the weight distribution of the
pump and drive,
the arrangement of the dis
charge nozzle on tubular cas
ing pumps (see Figs. k, l, m
and n),
the environment of the pump
casing, i. e. dry or wet (see
Figs. b and o).
7.2
Pump Intake Structures
7.2.1
Pump Sump
Pump sumps (or suction tanks)
are designed to collect liquids
and be intermittently drained
if the mean inlet fow is smaller
than the pump fow rate. The
sump or tank size depends on
the pump fow rate Q and the
permissible frequency of starts
Z of the electric motors, see sec
tion ....
The useful volume V
N
of the
pump sump is calculated using:
V
N
= Q
in

Q
m
Q
in

(48)

Q
m
Z

where
Z Max. permissible frequency
of starts per hour
Q
in
Inlet fow in m

/h
Q
m
= (Q
on
+ Q
off
) /
Q
on
Flow rate at switchon
pressure in m

/h
Q
off
Flow rate at switchoff
pressure in m

/h
V
N
Useful volume of pump
sump including potential
backwash volume in m

The maximum frequency of

starts occurs when the fow rate
Q
m
is twice the incoming fow
Q
in
. The max. frequency of
starts per hour is therefore:
Periphery Pump Installation Arrangements Intake Structures Pump Sump
d e
i j
n o
6
Z
max
= Q
m
/4V
N
(49)
With dirty liquids, solids must
be prevented from being depos
ited and collecting in dead zones
and on the foor. Walls arranged
at a 45, or better still 60 angle,
help prevent this (Fig. 60).
7.2.2
Suction Piping
The suction pipe should be as
short as possible and run with a
gentle ascending slope towards
the pump. Where necessary, ec
centric suction piping as shown
in Fig. 6 should be provided
(with a suffcient straight length
of pipe upstream of the pump
L d) to prevent the formation
of air pockets. If, on account
of the site conditions, ftting an
elbow immediately upstream of
the pump cannot be avoided,
an accelerating elbow (Fig. 6)
helps to achieve a smooth fow.
For the same reason, an elbow
with multiple turning vanes (see
Fig. 6) is required in front of
doubleentry pumps or pumps
with mixed fow (or axial fow)
impellers unless this is impos
sible because of the nature of
the medium handled (no stringy,
fbrous solids, see 6.5).
45 to 60
0.5 d
E
d
E
Suction
pipe
Fig. 60: Inclined sump walls to
prevent deposits and accumula-
tion of solids
Periphery Pump Sump Suction Piping
Fig. 61: Eccentric reducer and branch ftting to avoid air pockets
Fig. 62: Flow-accelerating
elbow upstream of a vertical
volute casing pump with high
specifc speed
Fig. 63: Intake elbow with mul-
tiple turning vanes upstream of
a double-entry, horizontal
volute casing pump (plan view)
Fig. 64: Installation of a positive
defector in the intake chamber
of a submersible motor pump
7
The suction and inlet pipes in
the suction tank or pump sump
must be suffciently wide apart
to prevent air from being en
trained in the suction pipe; posi
tive defectors (Figs. 64 and 65)
should be provided, if required.
The mouth of the inlet pipe
must always lie below the liquid
level, see Fig. 65.
If the suction pipe in the tank
or pump sump is not submerged
adequately because the liquid
level is too low, rotation of the
medium might cause an air
entraining vortex (hollow vor
tex) to develop. Starting as a
funnelshaped depression at the
surface, a tubeshaped air cavity
forms within a short period of
time, extending from the surface
to the suction pipe. This will
cause the pump to run very un
steadily and the output to de
crease. The required minimum
submergence (minimum depth
6
7 Suction Piping Minimum Submergence
Wrong
lnlet
pipe
Positive deflector
Suction
tank
Suction pipe
Fig. 65: Piping arrangement in the suction tank /
pump sump to prevent air entrainment
d
E
6 d
E
5.5 d
E
d
E
d
E
v
E
S
B
0.5 d
E
d
E
v
E
S
B
v
E
S
B
DN B
mm
65 80
80 80
100 100
150 100
200 150
250 150
300 200
400 200
500 200
Fig. 66: Clearances between wall and suction pipe in
the suction tank or pump sump according to relevant
German regulations. S
min
, as shown in Fig. 67.
2 suction pipes arranged side by side require a
distance of 6 d
E
.
1.5
1.0
2.0
m
0.8
0.6
0.5
0.4
0.3
0.1 0.05 0.2 0.3 0.4 0.5 0.6 0.8 1.0 m
2
1
0
.
5
vs

=

3

m
/
s
2
0
0
0
1
5
0
0
Q

=

1
0
0
0

m
3
/
h
8
0
0
6
0
0
5
0
0
4
0
0
3
0
0
2
0
0
1
5
0
1
0
0
8
0
6
0
5
0
4
0
3
0
2
0
1
5
1
0
3
0
0
0
S S S
S S S
d
E
d
E
M
i
n
i
m
u
m

s
u
b
m
e
r
g
e
n
c
e

S
m
i
n
Inlet diameter d
E
Fig. 67: Minimum submergence S
min
of horizontal and vertical suction pipes (with and without entry
nozzle) required for suction tanks to avoid hollow vortices (to Hydraulic Institute standards)
64
of immersion) is specifed in Fig.
67, the minimum clearance be
tween suction pipes and walls /
sump foor in Fig. 66. (Special
measures must be taken for tu
bular casing pumps, see 7..).
The minimum submergence S
min

can be read from Fig. 67 as a
function of the intake diameter
d
E
(this is the pipe inside dia
meter of straight, fangeless
pipes) or, where available, the
inlet diameter of the entry
nozzle and the fow rate Q. It
can also be calculated according
to the following equation given
by the Hydraulic Institute:
S
min
= d
E
+ . v
s

d
E
(50)

g


where
S
min
Minimum submergence in m
v
s
Flow velocity
= Q/900 p d
E

in m/s,
recommended to m/s
but never exceeding m/s
Q Flow rate in m

/h
g Gravitational constant
9.8 m/s

d
E
Inlet diameter of suction
pipe or entry nozzle in m
7 Suction Piping Air-entraining Vortex Minimum Submergence Intake Structures
Suction pipe
Raft
Fig. 68: Raft to
prevent air-
entraining hol-
low vortices
At fow velocities of m/s, the
minimum submergence levels
specifed by the relevant Ger
man regulations agree well with
the data given above [].
Wherever the required mini
mum submergence cannot or
not always be ensured, meas
ures as shown in Figs. 68 and
69 have to be taken to prevent
airentraining vortices. Irrespec
tive of the aspects mentioned
before, it should be checked
whether the submergence levels
also meet the NPSH
a
require
ments laid down in .5..
Round tanks with tangential
inlet pipes are special cases but
used frequently. The liquid dis
charged via the inlet pipe causes
the contents of the tank to ro
tate. For this reason baffes as
illustrated in Fig. 70 should be
provided.
7.2.3
Intake Structures for Tubular
Casing Pumps [1]
For tubular casing pumps, the
minimum submergence and the
design of the intake chamber
are of particular importance
because impellers with high spe
cifc speeds react very sensitively
to uneven inlet fows and air
entraining vortices.
Fig. 7 shows the arrangement
of suction pipes in intake
chambers of tubular casing
pumps.
Refer to Fig. 7 for the mini
mum water level required for
open, unlined intake chambers
Fig. 69: Use of swirl preventers
Suction pipe
Axial splitter
Axial splitter
Radial splitter
Radial
splitter
Tangential
inlet
Tangential
inlet
To pump
To pump
To pump
Baffle
Fig. 70: Use of swirl preventers in cylindrical
tanks to ensure smooth fow to pump
65
with and without entry cones or
calculate it using the following
equation:
S
min
= 0.8 d
E
+ .8 v
s

d
E

g
(51)
d
s
d
E
0.75 d
E
(
2

2
.
5
)

d
E
S
(0.3 0.5) d
E
4 d
E
Entry cone
Fig. 71: Suction pipe arrangement in intake chambers of tubular cas-
ing pumps. S
min
as shown in Fig. 72
d
E
(1.5 1.65) ds
2 suction pipes arranged side by side require a distance of > 3 d
E
7 Intake Structures Priming Devices
where
S
min
Minimum submergence in m
v
s
Flow velocity
= Q / 900 p d
E

in m/s
Q Flow rate in m

/h
g Gravitational constant
9.8 m/s

d
E
Inlet diameter of bellmouth
in m
Lined or covered intake
chambers or Kaplan intake
elbows are more expensive, but
allow pump operation at lower
submergence levels [].
Irrespective of the aspects
mentioned before, it should be
checked whether the submer
gence levels also meet the
NPSH
a
requirements laid down
in .5..
7.2.4
Priming Devices
Most centrifugal pumps are not
selfpriming; i. e. their suction
pipes and suctionside cas
ings must be deaerated prior
to startup unless the impeller
is arranged below the liquid
level. This often inconvenient
procedure can be avoided by
installing a foot valve (function
ing as a nonreturn valve) at the
suction pipe mouth (Fig. 7).
Deaerating is then only neces
sary prior to commissioning and
after a long period of standstill.
A closed suction tank (static
tank) serves the same purpose,
in particular when contami
nated liquids are handled (it
does, however, increase the fow
losses and therefore reduces the
NPSH
a
). The tank is under
negative pressure and mounted
upstream of the pump suction
nozzle (Fig. 74). Prior to com
missioning it must be flled
with liquid. When the pump is
started up, it empties the tank,
and the air in the suction or
siphon pipe is drawn into the
suction tank across the apex un
til the liquid to be pumped fol
66
lows. After the pump has been
stopped, the tank is reflled with
liquid via the discharge pipe
either manually or automati
cally. The air stored in the tank
escapes into the suction pipe.
The suction tank volume V
B

depends on the suction pipe
volume and the suction lift
capacity of the pump:
V
B
= d
s

p
L
s

p
b
4
p
b
gH
s
(52)
where
V
B
Suction tank volume in m

d
s
Inside diameter of the air
flled inlet pipe in m
L
s
Straight length of airflled
piping in m
p
b
Atmospheric pressure in Pa
( bar = 00 000 Pa)
Density of the liquid han
dled in kg/m

g Gravitational constant
9.8 m/s

H
s
Suction lift of pump in m
according to equation
1.5
1.0
m
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.15
0.1
0.1 0.2 0.3 0.4 0.5 0.6 0.8 1.0 m
0
.
7
5
0
.
5
0
0
.
2
5
4000
3000
2000
15
1000
800
600
500
400
300
150
100
80
60
50
40
30
20
10
Q = 200 m
3
/h
1500
VE
=
1
.
5
m
/
s
1
.
0
S
d
E
M
i
n
i
m
u
m

s
u
b
m
e
r
g
e
n
c
e

S
m
i
n
Inlet diameter d
E
Fig. 72: Minimum submergence S
min
of tubular casing pump suction
pipe to avoid hollow vortices
Fig. 73: Foot valve (cup valve)
with suction strainer
7 Priming Devices Suction Tank
d
s
H
s geo
Suction gate valve
with water cup
Liquid level in
operation
Strainer
Suction tank
vent hole
Liquid level prior
to start-up
Priming pipe
Shut-off valve
Non-return valve
Connect barrier fluid supply line!
Fig. 74: Suction tank arrangement
67
H
s
= H
s geo
+ H
L,s
(53)
where
H
s geo
Vertical distance between
water level and pump ref
erence plane for suction
lift operation in m, see
Fig. 6
H
L,s
Head loss in the suc
tion piping in m (refer to
...).
As H
L,s
is in most cases notably
smaller than H
s geo
, Eq. 5 can
be neglected and H
s
equated
with H
s geo
. In this case, Fig. 75
provides a much faster way of
fnding the required tank size.
For safety reasons the suction
tank volume should be multi
plied by a factor of to .5, or
by a factor of up to in the case
of smaller pumping stations.
The liquid pressure must never
reach its specifc vaporization
pressure at any point in the sys
tem.
7.3
Arrangement of Measure-
ment Points
In order to achieve a certain
accuracy in pressure and veloc
ity measurement, the fow must
be smooth and regular at the
measuring points. Therefore,
undisturbed straight lengths of
piping need to be arranged up
stream and downstream of the
measurement point(s), as shown
in Fig. 76 and indicated in Table
4. All pipe components which
may impede a straight, parallel
and nonswirling fow of liquid
are considered a disturbance.
Relevant German regulations
(VdS Association of German
7 Suction Tank Measurement Points
Fig. 75: Graph to determine the size of the suction tank. Follow the
numbers from 1 to 4 for selection. A safety factor of 3.0 has al-
ready been considered in the above diagram (head losses H
L,s
in the
suction piping were neglected).
0.03 0.05 0.1 0.2 0.3 0.5 2 3 5 1 1.5 20 30 60 10
30 50 100 200300 500 1000 l
15 m
3
300 200 150 100 80 60 50 40 30 20 mm 400 600
1
4
1
2
3
4
5
7
6
8
1
0
1
5
2
0
1
2
.
5
1
7
.
5 9
0
2
4
6
1
3
5
7
Inside diameter of suction pipe
Suction tank volume
S
t
r
a
i
g
h
t

l
e
n
g
t
h

o
f

p
i
p
i
n
g

L
s

[
m
]
2
M
a
x
.

s
u
c
t
i
o
n

l
i
f
t

H
s

[
m
]
3
Table 14: Minimum values for undisturbed straight lengths of piping
at measurement points in multiples of the pipe diameter D
Source Distance from Undisturbed
pump fange pipe length
A
s
/D A
d
/D U
s
/D U
d
/D
VdS 09S 0.5 .0 .5 .5 Inservice measurement
ISO 9906 .0 .0 5+n
q
/5 Acceptance test measurement
68
7 Measurement Points Shaft Couplings
D
D
A
s
U
s
U
d
A
d
Property Insurance Companies)
stipulate pipe lengths in mul
tiples of the pipe diameter for
inservice measurements, while
ISO 9906 specifes pipe lengths
for acceptance test measure-
ments. The data from both
sources are listed in Table 4.
If the required straight pipe
lengths cannot be provided, the
measuring results are likely to
be less accurate. Consequently,
pump fanges are not suitable as
measurement points.
The pressure measuring points
should consist of a 6 mm dia
meter hole and a weld socket
to ft the pressure gauge. Even
better still are annular measur
ing chambers with four drilled
holes spread evenly across the
circumference.
7.4
Shaft Couplings
In centrifugal pump engineer
ing, both rigid and fexible shaft
couplings are used. Rigid coupl
ings are mainly used to connect
Fig. 76: Arrangement of pressure measurement points up- and
downstream of the pump
shafts in perfect alignment, since
the smallest degree of misalign
ment will cause considerable
stress on the coupling and on
the adjacent shaft sections.
Fig. 77: Flexible (left) and highly fexible coupling
Flexible couplings to DIN 740
are elastic, slipfree connect
ing elements between drive and
pump which accommodate axi
al, radial and angular misalign
ment and damp shock loads.
Flexibility is usually achieved
by the deformation of damping
and rubberelastic spring ele
ments whose life is governed
to a large extent by the degree
of misalignment. Fig. 77 shows
two of the most common types
of fexible shaft coupling. Fig.
78 shows a spacer coupling
between a volute casing pump
and drive. It permits removal
of the pump rotating assembly
without having to dismantle the
suction and discharge piping or
move the pump casing or drive
(back pullout design).
Fig. 78: Pump with spacer coupling compared with normal coupling
69
7.5
Pump Nozzle Loading
A centrifugal pump mounted
on the foundation should not
be used as an anchorage point
for connecting the piping. Even
if the piping is ftted to the noz
zles without transmitting any
stresses or strains, forces and
moments, summarized as nozzle
loading, will develop under
actual operating conditions
(pressure and temperature) and
as a result of the weight of the
liquidflled piping. These cause
stresses and deformation in the
pump casings, and above all
changes in coupling alignment,
which, in turn, may affect the
pumps running characteristics,
the service life of the fexible
elements in the shaft coupling,
as well as the bearings and me
chanical seals. For this reason
limits have been defned for per
grouted baseplate, broken line
for pumps on nongrouted base
plates).
7.6
National and International
Standards and Codes
A series of national standards
and other technical codes have
been introduced in Germany
since the early sixties which
govern the dimensions, manu
facture, design, procurement
and use of centrifugal pumps.
Many of the requirements laid
down have been included in
European and international
standards and codes. Drawn up
by both operators and manu
facturers, these are now well
established in virtually all sec
tors of industry using or pro
ducing pumps. The most impor
tant standards are tabulated in
Fig. 80 on page 70.
7 Pump Nozzle Loading Standards and Codes
1200
1000
800
600
400
200
0
600
500
Nm
400
300
200
100
0
0 10 20 30 40 50 60 kg 70 80
F
v
,

m
a
x
M
m
a
x
M
m
a
x
F
v
,

m
a
x
F
H
,

m
a
x
F
H
,

m
a
x
z
x
x
z
y
y
P
e
r
m
i
s
s
i
b
l
e

m
o
m
e
n
t
s

M
m
a
x

a
t

t
h
e

f
l
a
n
g
e

r
e
f
e
r
e
n
c
e

p
l
a
n
e
P
e
r
m
i
s
s
i
b
l
e

f
o
r
c
e
s

F
H
,

m
a
x

a
n
d

F
v
,

m
a
x
Pump weight x
Discharge nozzle dia.
lmpeller outer dia.
lnlet
Outlet
Fig. 79: Permissible moments M
max
at the fange reference plane,
as well as permissible forces F
H,max
(at x,z plane) and F
V,max
(in y
direction) to ISO 1599 for single-stage volute casing pumps made of
ferritic cast steel or nodular cast iron at room temperature. Lower
numerical values apply to austenitic cast steel, lamellar graphite cast
iron or higher temperatures.
missible nozzle loading [].
As the loading profle for each
pump nozzle is made up of three
different forces and moments, it
is not possible to specify theo
retical nozzle loading limits for
all conceivable combinations.
Therefore, operators either need
to check whether the nozzle
loading imposed by the system
is still within the pumps permis
sible limits, or have to contend
with the considerably reduced
general limits specifed in sev
eral national and international
standards and codes (EURO
PUMP brochure Permissible
fange forces and moments for
centrifugal pumps, 986; API
60; ISO 599).
Fig. 79 shows the permissible
nozzle loading for singlestage
volute casing pumps to ISO
599 (solid line for pumps on
70
7
F
i
g
.

8
0
:

N
a
t
i
o
n
a
l

a
n
d

i
n
t
e
r
n
a
t
i
o
n
a
l

s
t
a
n
d
a
r
d
s

a
n
d

c
o
d
e
s

f
o
r

c
e
n
t
r
i
f
u
g
a
l

p
u
m
p
s

(
l
a
s
t

u
p
d
a
t
e
:

2
0
0
5
)
I
S
O

9
9
0
6
R
o
t
o
d
y
n
a
m
i
c

p
u
m
p
s

H
y
d
r
a
u
l
i
c

p
e
r
f
o
r
m
a
n
c
e

a
c
c
e
p
t
a
n
c
e

t
e
s
t
s

G
r
a
d
e
s

1

a
n
d

2

I
S
O

5
1
9
8
C
e
n
t
r
i
f
u
g
a
l
m
i
x
e
d

f
l
o
w
a
n
d

a
x
i
a
l
p
u
m
p
s

C
o
d
e

f
o
r
h
y
d
r
a
u
l
i
c

p
e
r
f
o
r
m
a
n
c
e
t
e
s
t
s
;

P
r
e
c
i
s
i
o
n

g
r
a
d
e
D
I
N

2
4
2
5
9
-
1
P
u
m
p
s
;

B
a
s
e
p
l
a
t
e
s

f
o
r

m
a
c
h
i
n
-
e
r
y
;

D
i
m
e
n
-
s
i
o
n
s
S
c
o
p
e

o
f

A
p
p
l
i
c
a
t
i
o
n
a
n
d

R
e
s
p
o
n
s
i
b
i
l
i
t
i
e
s
D
i
m
e
n
s
i
o
n
a
l

S
t
a
n
d
a
r
d
s
:

P
u
m
p
s

a
n
d

A
c
c
e
s
s
o
r
i
e
s
C
o
d
e
s

a
n
d

S
p
e
c
i
f
i
c
a
t
i
o
n
s
V
D
M
A
G
e
r
m
a
n

E
n
g
i
n
e
e
r
i
n
g
F
e
d
e
r
a
t
i
o
n
P
u
m
p
C
o
m
m
i
t
t
e
e
D
I
N
G
e
r
m
a
n

S
t
a
n
d
a
r
d
s

I
n
s
t
i
t
u
t
e
M
e
c
h
a
n
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g

S
t
a
n
d
a
r
d
s
C
o
m
m
i
t
t
e
e
,

P
u
m
p
s
C
E
N
C
o
m
i
t

E
u
r
o
-
p

e
n

d
e

N
o
r
-
m
a
l
i
s
a
t
i
o
n
E
u
r
o
p
e
a
n

S
t
a
n
d
a
r
d
s

C
o
o
r
d
i
n
a
t
i
o
n

C
o
m
m
i
t
t
e
e
T
C

1
9
7

P
u
m
p
s
I
S
O
I
n
t
e
r
-
n
a
t
i
o
n
a
l

O
r
g
a
n
i
z
a
-
t
i
o
n

f
o
r

S
t
a
n
d
a
r
d
i
-
z
a
t
i
o
n
T
e
c
h
n
.

C
o
m
m
.
T
C

1
1
5
/
P
u
m
p
s
D
I
N

E
N

7
3
4
S
i
d
e

c
h
a
n
n
e
l

p
u
m
p
s

P
N

4
0
;

R
a
t
e
d

p
e
r
-
f
o
r
m
a
n
c
e
,

m
a
i
n

d
i
m
e
n
-
s
i
o
n
s
,

d
e
s
i
g
n
a
t
i
o
n

D
I
N

E
N

7
3
3
E
n
d

s
u
c
t
i
o
n

c
e
n
t
r
i
f
u
g
a
l

p
u
m
p
s

(
r
a
t
-
i
n
g

P
N

1
0
)
;

w
i
t
h

b
e
a
r
i
n
g

b
r
a
c
k
e
t
;

R
a
t
e
d

p
e
r
-
f
o
r
m
a
n
c
e
,

m
a
i
n

d
i
m
e
n
-
s
i
o
n
s
,

d
e
s
i
g
n
a
t
i
o
n

s
y
s
t
e
m

D
I
N

E
N

2
2
8
5
8
E
n
d

s
u
c
t
i
o
n

c
e
n
t
r
i
f
u
g
a
l

p
u
m
p
s

(
r
a
t
-
i
n
g

1
6

b
a
r
)
;

D
e
s
i
g
n
a
t
i
o
n
,

r
a
t
e
d

p
e
r
-
f
o
r
m
a
n
c
e
,

m
a
i
n

d
i
m
e
n
-
s
i
o
n
s

I
S
O

2
8
5
8
E
n
d

s
u
c
t
i
o
n
c
e
n
t
r
i
f
u
g
a
l
p
u
m
p
s

(
r
a
t
-
i
n
g

1
6

b
a
r
)

D
e
s
i
g
n
a
t
i
o
n
,
r
a
t
e
d

p
e
r
-
f
o
r
m
a
n
c
e

a
n
d

d
i
m
e
n
-
s
i
o
n
s
I
S
O

3
6
6
1
E
n
d

s
u
c
t
i
o
n
c
e
n
t
r
i
f
u
g
a
l
p
u
m
p
s

B
a
s
e
p
l
a
t
e
a
n
d

i
n
s
t
a
l
l
a
-
t
i
o
n

d
i
m
e
n
-
s
i
o
n
s
I
S
O

3
0
6
9
E
n
d

s
u
c
t
i
o
n
c
e
n
t
r
i
f
u
g
a
l
p
u
m
p
s

D
i
m
e
n
s
i
o
n
s
o
f

c
a
v
i
t
i
e
s
f
o
r

m
e
c
h
a
n
i
-
c
a
l

s
e
a
l
s

a
n
d

f
o
r

s
o
f
t
p
a
c
k
i
n
g
I
S
O

9
9
0
5
T
e
c
h
n
i
c
a
l

s
p
e
c
i
f
i
c
a
-
t
i
o
n
s

f
o
r

c
e
n
t
r
i
f
u
g
a
l

p
u
m
p
s

C
l
a
s
s

I
I
S
O

5
1
9
9
T
e
c
h
n
i
c
a
l
s
p
e
c
i
f
i
c
a
-
t
i
o
n
s

f
o
r
c
e
n
t
r
i
f
u
g
a
l
p
u
m
p
s

C
l
a
s
s

I
I
I
S
O

9
9
0
8
T
e
c
h
n
i
c
a
l
s
p
e
c
i
f
i
c
a
-
t
i
o
n
s

f
o
r
c
e
n
t
r
i
f
u
g
a
l
p
u
m
p
s

C
l
a
s
s

I
I
I
A
P
I

6
8
2
S
h
a
f
t

S
e
a
l
i
n
g
S
y
s
t
e
m
s

f
o
r
C
e
n
t
r
i
f
u
g
a
l
a
n
d

R
o
t
a
r
y
P
u
m
p
s
A
P
I

6
1
0
C
e
n
t
r
i
f
u
g
a
l
P
u
m
p
s

f
o
r
P
e
t
r
o
l
e
u
m
,
P
e
t
r
o
c
h
e
m
i
c
a
l

a
n
d

N
a
t
u
r
a
l
G
a
s

I
n
d
u
s
t
r
i
e
s
D
I
N

E
N

2
3
6
6
1
E
n
d

s
u
c
t
i
o
n

c
e
n
t
r
i
f
u
g
a
l

p
u
m
p
s
;

B
a
s
e
p
l
a
t
e

a
n
d

i
n
s
t
a
l
l
a
-
t
i
o
n

d
i
m
e
n
-
s
i
o
n
s
E
N

1
2
1
6
2
L
i
q
u
i
d

p
u
m
p
s

S
a
f
e
t
y

r
e
-
q
u
i
r
e
m
e
n
t
s

P
r
o
c
e
d
u
r
e

f
o
r

h
y
d
r
o
-
s
t
a
t
i
c

t
e
s
t
-
i
n
g
E
N

1
2
6
3
9
L
i
q
u
i
d

p
u
m
p
s

a
n
d

p
u
m
p

u
n
i
t
s
;

N
o
i
s
e

m
e
a
s
u
r
e
-
m
e
n
t
;

T
e
s
t

c
l
a
s
-
s
e
s

2

a
n
d

3
E
N

1
1
5
1
C
i
r
c
u
l
a
t
o
r
s

w
i
t
h

i
n
p
u
t

p
o
w
e
r
s

u
p

t
o

2
0
0

W

f
o
r

h
e
a
t
i
n
g

s
y
s
-

t
e
m
s

a
n
d

s
e
r
v
i
c
e

w
a
t
e
r

h
e
a
t
i
n
g

s
y
s
-
t
e
m
s

f
o
r

d
o
m
e
s
t
i
c

u
s
e

R
e
-
q
u
i
r
e
m
e
n
t
s
,

t
e
s
t
i
n
g
,

m
a
r
k
i
n
g
E
N

8
0
9
P
u
m
p
s

a
n
d

p
u
m
p

u
n
i
t
s

f
o
r

l
i
q
u
i
d
s
;

G
e
n
e
r
a
l

s
a
f
e
t
y

r
e
-
q
u
i
r
e
m
e
n
t
s
D
I
N

2
4
2
9
9
-
1
P
u
m
p

n
a
m
e
-
p
l
a
t
e
s
;

G
e
n
e
r
a
l

s
p
e
c
i
f
i
c
a
-
t
i
o
n
s

D
I
N

2
4
2
9
6
P
u
m
p
s

a
n
d

p
u
m
p

u
n
i
t
s

f
o
r

l
i
q
u
i
d
s
;

S
p
a
r
e

p
a
r
t
s
;

S
e
l
e
c
t
i
o
n

a
n
d

p
r
o
c
u
r
e
-
m
e
n
t
D
I
N

2
4
2
7
3
P
u
m
p
s

a
n
d

p
u
m
p

u
n
i
t
s

f
o
r

l
i
q
u
i
d
s
;

M
a
t
e
r
i
a
l
s

a
n
d

p
r
o
d
u
c
t

t
e
s
t
s
V
D
M
A
2
4
2
5
2
C
e
n
t
r
i
f
u
g
a
l

p
u
m
p
s

w
i
t
h

w
e
a
r

p
l
a
t
e
s
,

P
N

1
0
(
w
a
s
h

w
a
t
e
r

p
u
m
p
s
)
,

w
i
t
h

b
e
a
r
i
n
g

b
r
a
c
k
e
t
;

D
e
s
i
g
n
a
t
i
o
n
,

r
a
t
e
d

p
o
w
e
r
s
,

d
i
m
e
n
s
i
o
n
s

V
D
M
A
2
4

2
7
9
C
e
n
t
r
i
f
u
g
a
l

p
u
m
p
s
;

T
e
c
h
-
n
i
c
a
l

r
e
q
u
i
r
e
-
m
e
n
t
s
;

M
a
g
-
n
e
t
i
c

d
r
i
v
e

a
n
d

c
a
n
n
e
d

m
o
t
o
r

p
u
m
p
s
E u r o p eW o r l d w i d e
I n t e r n a t i o n a l
F e d e r a l R e p u b l i c o f G e r m a n y
A
m
e
r
i
c
a
n

P
e
t
r
o
l
e
u
m

I
n
s
t
i
t
u
t
e
V
D
M
A
2
4
2
5
3
C
e
n
t
r
i
f
u
g
a
l

p
u
m
p
s

w
i
t
h

l
i
n
e
d
c
a
s
i
n
g

(
l
i
n
e
d
p
u
m
p
s
)
;

s
i
n
g
l
e

e
n
t
r
y
,

s
i
n
g
l
e

s
t
a
g
e
,

w
i
t
h

a
x
i
a
l

i
n
l
e
t
;

R
a
t
e
d

p
o
w
e
r
s
,
d
i
m
e
n
s
i
o
n
s

D
I
N

E
N

2
4
2
5
0
C
e
n
t
r
i
f
u
g
a
l

p
u
m
p
s
;

N
o
m
e
n
-
c
l
a
t
u
r
e

a
n
d

c
o
m
p
o
n
e
n
t

n
u
m
b
e
r
s
V
D
M
A
2
4
2
9
2
L
i
q
u
i
d

p
u
m
p
s
;

O
p
e
r
a
t
i
n
g

i
n
s
t
r
u
c
t
i
o
n
s

f
o
r

p
u
m
p
s

a
n
d

p
u
m
p

u
n
i
t
s
;

S
t
r
u
c
-
t
u
r
e
,

c
h
e
c
k

l
i
s
t
,

w
o
r
d
i
n
g

f
o
r

s
a
f
e
t
y

i
n
s
t
r
u
c
t
i
o
n
s
V
D
M
A
2
4
2
6
1
-
1
P
u
m
p
s
;

D
e
s
i
g
n
a
t
i
o
n
s

b
a
s
e
d

o
n

f
u
n
c
t
i
o
n

a
n
d

d
e
s
i
g
n

f
e
a
t
u
r
e
s
;

C
e
n
t
r
i
f
u
g
a
l

p
u
m
p
s
V
D
M
A
2
4
2
7
6
L
i
q
u
i
d

p
u
m
p
s

f
o
r

c
h
e
m
i
c
a
l

p
l
a
n
t

Q
u
a
l
i
t
y

s
p
e
-
c
i
f
i
c
a
t
i
o
n
s

f
o
r

m
a
t
e
r
i
a
l
s

a
n
d

c
o
m
p
o
-
n
e
n
t
s
D
I
N

E
N
1
2
7
5
6
M
e
c
h
a
n
i
c
a
l

s
e
a
l
s

M
a
i
n

d
i
m
e
n
-
s
i
o
n
s
,

d
e
-
s
i
g
n
a
t
i
o
n

a
n
d

m
a
t
e
r
i
a
l
c
o
d
e
s
D
I
N

E
N

I
S
O
9
9
0
6
R
o
t
o
d
y
n
a
m
i
c

p
u
m
p
s

H
y
d
r
a
u
l
i
c

p
e
r
f
o
r
m
a
n
c
e

a
c
c
e
p
t
a
n
c
e

t
e
s
t

G
r
a
d
e
s

1

a
n
d

2
D
I
N

E
N

1
2
7
2
3
L
i
q
u
i
d

p
u
m
p
s

G
e
-
n
e
r
a
l

t
e
r
m
s

f
o
r

p
u
m
p
s

a
n
d

i
n
s
t
a
l
l
a
-
t
i
o
n
s
;

D
e
f
i
n
i
-
t
i
o
n
s
,

v
a
r
i
-
a
b
l
e
s
,

s
y
m
-
b
o
l
s

a
n
d

u
n
i
t
s
D
I
N

E
N

1
2
6
3
9
L
i
q
u
i
d

p
u
m
p
s

a
n
d

p
u
m
p

u
n
i
t
s

N
o
i
s
e

m
e
a
s
u
r
e
-
m
e
n
t

T
e
s
t

c
l
a
s
s
e
s

2

a
n
d

3
D
I
N

E
N
1
2
2
6
2
C
e
n
t
r
i
f
u
g
a
l

p
u
m
p
s
;

T
e
c
h
n
i
c
a
l

d
o
-

c
u
m
e
n
t
a
t
i
o
n
,

t
e
r
m
s
,

s
c
o
p
e

o
f

s
u
p
p
l
y
,

q
u
a
l
i
t
y
D
I
N

E
N

1
2
0
5
6
-
4

G
r
a
v
i
t
y

d
r
a
i
n
-
a
g
e

s
y
s
t
e
m
s

i
n
s
i
d
e

b
u
i
l
d
-
i
n
g
s

P
a
r
t

4
:

S
e
w
a
g
e

l
i
f
t
i
n
g

u
n
i
t
s
;

L
a
y
o
u
t

a
n
d

d
i
m
e
n
-
s
i
o
n
i
n
g

D
I
N

I
S
O
9
9
0
5
(
C
l
a
s
s

I
)
D
I
N

I
S
O

5
1
9
9
(
C
l
a
s
s

I
I
)
D
I
N

I
S
O

9
9
0
8
(
C
l
a
s
s

I
I
I
)
C
e
n
t
r
i
f
u
g
a
l

p
u
m
p
s
;

T
e
c
h
n
i
c
a
l

s
p
e
c
i
f
i
c
a
-
t
i
o
n
s
D
I
N

E
N

8
0
6
-
1
a
n
d

-
2
T
e
c
h
n
i
c
a
l

s
p
e
c
i
f
i
c
a
-
t
i
o
n
s

f
o
r

d
r
i
n
k
i
n
g

w
a
t
e
r

s
y
s
-
t
e
m
s
D
I
N

1
9
8
9
R
a
i
n
w
a
t
e
r

h
a
r
v
e
s
t
i
n
g

s
y
s
t
e
m
s
D
I
N

E
N
1
2
0
5
0
S
e
w
a
g
e

l
i
f
t
-
i
n
g

u
n
i
t
s

f
o
r

t
h
e

d
i
s
p
o
s
a
l

o
f

w
a
s
t
e

w
a
t
e
r

f
r
o
m

b
u
i
l
d
i
n
g
s

a
n
d

p
r
e
m
i
s
e
s
;

D
e
s
i
g
n

a
n
d

t
e
s
t
i
n
g

p
r
i
n
-
c
i
p
l
e
s

D
I
N

2
4
4
2
0
-
1
S
p
a
r
e

p
a
r
t
s

l
i
s
t
s
;

G
e
n
e
r
a
l
D
I
N

2
4
4
2
0
-
2
S
p
a
r
e

p
a
r
t
s

l
i
s
t
s
;

F
o
r
m

a
n
d

s
t
r
u
c
-
t
u
r
e

o
f

t
e
x
t

f
i
e
l
d
D
I
N

1
9
8
8
-
5
T
e
c
h
n
i
c
a
l

s
p
e
-
c
i
f
i
c
a
t
i
o
n
s

f
o
r

d
r
i
n
k
i
n
g

w
a
-
t
e
r

i
n
s
t
a
l
l
a
-
t
i
o
n
s
,

p
r
e
s
-
s
u
r
e

b
o
o
s
t
-
i
n
g

a
n
d

r
e
d
u
-
c
i
n
g

s
y
s
t
e
m
s
D
I
N

1
9
8
6
D
r
a
i
n
a
g
e

s
y
s
t
e
m
s

f
o
r

b
u
i
l
d
i
n
g
s

a
n
d

p
r
e
-
m
i
s
e
s
D
I
N

E
N

7
3
5
O
v
e
r
a
l
l

d
i
m
e
n
s
i
o
n
s

o
f

c
e
n
t
r
i
f
u
-
g
a
l

p
u
m
p
s
;

T
o
l
e
r
a
n
c
e
s

D
I
N

2
4
2
5
1
M
u
l
t
i
s
t
a
g
e

c
e
n
t
r
i
f
u
g
a
l

p
u
m
p
s
;

D
r
a
i
n
a
g
e

p
u
m
p
s

w
i
t
h

h
e
a
d
s

u
p

t
o

1
0
0
0

m

a
t

a

r
a
t
e
d

s
p
e
e
d

o
f

1
5
0
0

r
p
m
Standards Codes Specifcations
7
n
q
= n Q
opt
/ H
opt
/4
= 900 (00/600) / 57.5
/4
= 900 0.6 / 0.88 = .8 rpm
or
= (n/60) Q
opt
/ (gH
opt
)
/4
= 48. (00/600) / 9.8 57.5
/4

= 48. 0.6 / 5.7 = .8 (dimensionless)
8
8.
Calculation Examples
Sought: The pressure differen
tial between the discharge and
suction sides indicated by the
pressure gauges.
(Taking z
s,d
= 50 mm into ac
count presupposes that the pres
sure gauges are ftted exactly
at the respective nozzle levels
to keep the same difference in
height. If they are mounted at
the same level, z
s,d
must be set
to zero. Refer to paragraph 7.
and ISO DIS 9906 for the cor
rect location of the pressure
measurement taps).
Flow velocities
v
d
= 4 Q /pd
d

= 4 (00/600) /p0.08

= . m/s
v
s
= 4 Q /pd
s

= 4 (00/600) / p0.0

= 7.08 m/s.
According to Eq. (1) the pressure differential is:
p = g [H z
s,d
(v
d

v
s

) / g]
= 998. 9.8 [57.5 0.50 (.

7.08

)/( 9.8)]
= 54 576 Pa = 5.5 bar
According to Eq. (2) the input power is:
P = g Q H /
= 998. 9.8 (00 / 600) 57.5 / 0.85
= 7 46 W = 7.5 kW
8.2
Input Power
Given: The data as per exercise
8..
Sought: The input power P.
8.3
Specifc Speed
Given: The data as per 8.; the
specifc speed n
q
is calculated
using Eq. (3)
The consecutive numbers of the
calculation examples in this
chapter are identical to the num
bers of the respective equations
in the text. For example, the ap
plication dealt with in exercise
8. refers to Equation ().
8.1
Pressure Differential
Given: A volute casing pump
Etanorm 80 00 with char
acteristic curves as per Fig. 8,
speed of rotation n = 900 rpm,
impeller diameter D

= 9 mm,
operating point at the point of
best effciency: Q = 00 m

/h,
H = 57.5 m, = 8.5%, water
temperature t = 0 C, density
= 998. kg/m

. Nominal
nozzle diameters DN
d
= 80;
DN
s
= 00; inside nozzle dia
meter d
d
= 80 mm, d
s
= 00 mm
[]. Height difference between
suction and discharge nozzles
z
s,d
= 50 mm, Fig. 8.
Calculation Examples
7
8.5
Bernoullis Equation
Given: A centrifugal pump
system as shown in Fig. 8 with
tanks B and D, designed for a
fow rate of Q = 00 m

/h for
pumping water at 0 C. The
dischargeside tank is under a
pressure of 4. bar (positive
pressure), the suction tank is
open to atmosphere, v
e
0. The
geodetic difference in height is
.0 m; the welded discharge
piping has a nominal diameter
of DN 00 (d = 0. mm acc.
to Table 4). The system head
loss is .48 m.
Sought: The system head H
sys
.
Eq. (5) gives:
H
sys
= H
geo
+ (p
a
p
e
) / ( g) + (v
a

v
e

) / g + H
L
where
Density = 998. kg/m

according to Table
Pressure in tank B: p
a
= 4. bar = 40 000 Pa
Pressure in tank D: p
e
= 0
(p
a
p
e
) / ( g) = 40 000/(998. 9.8) = 4.89 m
v
a
= 4 Q / (600 p d

) = 4 00/(600 p 0.0

)
= .60 m/s
(v
a

v
e

)/g = (.60

0)/( 9.8) = 0. m
H
geo
= .00 m
H
L
= .48 m
H
sys
= 57.50 m
8.9
Head Loss in Pipes
Given: The data as per 8. and:
suction pipe DN 00, d =
00. mm according to Table 4,
length = 6.00 m, average abso
lute roughness k = 0.050 mm.
Sought: The head loss H
L
according to Fig. or Eq. (9).
The diagram in Fig. gives: H
L
= .00 6.00/00 = 0.060 m
The calculation according to Fig. 0 would be more complex and
involved, but also absolutely necessary for other roughness values.
Relative roughness d / k = 0. / 0.050 = 40
According to Eq. (), the Reynolds number is Re = v d /
where
= .00 0
6
m

/s,
v = Q / A = (Q/600) 4 / (pd

) = (00 / 600) 4 / (p 0.0

)
= .60 m/s,
Re = v d / = .60 0.0 / 0
6
= .7 0
5
.
From Fig. 0, d / k = 40 = 0.06.
Eq. (9) gives
H
L
= (L / d) v

/ g
= 0.06 (6.00 / 0.0) .60

/ 9.8 = 0.060 m
8 Calculation Examples
7
Sought: The inside diameter of the orifce plate d
Bl
.
Eq. (20) gives
d
Bl
= f Q / (g H) with f according to Fig. 5.
As an iterative calculation is necessary, d
Bl
is estimated in the frst
instance, and this value is compared with the calculated diameter. If
the two values differ, the value selected for the second estimate lies
between the initially estimated and calculated diameters.
The following is calculated frst of all:
Q / g H = 00 / 9.8 5.0 = 5.4 m.
First estimate d
Bl
= 70 mm; (d
Bl
/ d)

= 0.709; f = .;
Result: d
Bl
= . 5.4 = 65. mm
Second estimate d
Bl
= 68 mm; (d
Bl
/ d)

= 0.670; f = .9;
Result: d
Bl
= .9 5.4 = 68.9 mm
Third estimate d
Bl
= 68.4; (d
Bl
/ d)

= 0.679; f = .8;
Result: d
Bl
= .8 5.4 = 68.4 mm
For a faster solution, it is recommended to plot the calculated versus
the corresponding estimated diameters in a diagram so that the third
estimate already provides the fnal result in the intersection of con
necting line and diagonal, see adjacent diagram.
8.15
Head Loss in Valves and
Fittings
Given:
The suction pipe described in 8.9, including
a slide disc valve DN 00,
a 90 elbow with smooth surface and R = 5 d,
a foot valve DN 00
and a reducer DN 00 / DN 00 according to Table 8,
type IV with an opening angle of = 0.
Sought: The head losses H
L
.
According to Table 5, the loss coeffcient of the slide disc
valve is = 0.0
Acc. to Table 6, the loss coeffcient of the 90 elbow is = 0.0
Acc. to Table 5, the approx. loss coeffcient of
the foot valve is = .0
Acc. to Table 5, the loss coeffcient of the reducer is = 0.
The total of all loss coeffcients is = .5
Eq. (15) then gives the following head loss:
H
L
= v

/ g = .5 .60

/ ( 9.8) = 0.8 m
69
68
70
67
66
65
69 68 70 67 66 65
C
a
l
c
u
l
a
t
e
d

d
i
a
m
e
t
e
r

d
B
l

[
m
m
|
Estimated diameter d
Bl
[mm|
X
X
X
1
3
2
E
s
t
im
a
t
e
d

d
ia
m
e
t
e
r




C
a
lc
u
la
t
e
d

d
ia
m
e
t
e
r
8
Calculation Examples
8.20
Orifce Plate
Given:
The pump described in exercise
8. is provided with a welded
discharge pipe DN 80, the in
side diameter being d = 8.
mm. The discharge head is to be
constantly throttled by
H = 5.00 m.
74
8 Calculation Examples
8.21
Change of Speed
Given:
The pump speed as per 8. (op
erating data with index

) is to
be reduced from n

= 900 rpm
to n

= 450 rpm.
Sought: The data for fow rate Q

, discharge head H

and driving
power P after change of speed.
Eq. (21) gives
Q

= Q

(n

/n

) = 00 (450 / 900) = 00 m

/h
Eq. () gives
H

= H

(n

/n

)

= 57.5 (450 / 900)

= 4.4 m
Eq. () gives
P

= P

(n

/n

= 7.5 (450 / 900)

= 4.69 kW,
on the assumption that the effciency is the same for both speeds.
8.27
Turning Down Impellers
Given:
The fow rate of the pump at
BEP described in 8., i.e. Q
t
=
00 m

/h, is to be reduced to
Q
r
= 5 m

/h by turning down
the original impeller diameter
D
t
= 9 mm.
Sought: The reduced diameter D
r
and the discharge head H
r
at BEP
after turning down the impeller (H
t
= 57.5 m).
Eq. (27) gives
D
r
D
t
(Q
r
/ Q
t
) = 9 (5 / 00) = 80 mm
Eq. (6) gives
H
r
H
t
(Q
r
/ Q
t
) = 57.5 5 / 00 = 8.8 m
8.29
NPSH
a
for Suction Lift
Operation
Given:
The centrifugal pump system
described in exercise 8.5 plus
the following data: place of in
stallation 500 m above M. S. L.;
H
L,s
(refer to exercises 8.9 and
8.5) = 0.9 m; H
s geo
= .00 m;
v
e
0. The pump described in
8. is installed horizontally with
an open suction tank, as shown
in Fig. 6. According to Fig. 8,
the pumps NPSH
r
is 5.50 m at
a fow rate of Q = 00 m

/h.
Question: Is NPSH
a
suffcient?
According to Eq. (29),
NPSH
a
= (p
e
+ p
b
p
v
)/( g) + v
e

/g H
L,s
H
s geo
s
where
Gauge pressure in suction tank p
e
= 0
Atmospheric pressure p
b
= 955 mbar = 95 500 Pa acc. to Table
Vapour pressure p
v
= 0.07 bar = 7 Pa acc. to Table
Density = 998. kg/m

acc. to Table
(p
e
+ p
b
p
v
)/( g) = (0 + 95 500 7) / (998. 9.8) = 9.5 m
v
e

/g = 0
H
L,s
= 0.9 m
H
s geo
= .00 m
s = 0, as the centre of the impeller inlet is at the same
height as the centre of the pump inlet.
NPSH
a
= 6. m
With an NPSH
r
of 5.50 m,
NPSH
a
is larger than NPSH
r
in this case and therefore suffcient.
75
8 Calculation Examples
8.31
NPSH
a
for Suction Head
Operation
Given: The pump system de
scribed in exercise 8.9 is be op
erated in suction head operation
with a closed tank as shown in
Fig. 7. The system data are as
follows: place of installation
500 m above M. S. L.; H
L,s
(re
fer to exercises 8.9 and 8.5) =
0.9 m; H
z geo
= .00 m; ve 0.
The pump described in 8. is
installed horizontally with a
closed suction tank, as shown
in Fig. 7. According to Fig. 8,
the pumps NPSH
r
is 5.50 m at
a fow rate of Q = 00 m

/h.
Question: Is NPSH
a
suffcient?
According to Eq. (31)
NPSH
a
= (p
e
+ p
b
p
v
) / ( g) + v
e

/ g H
L,s
+ H
z geo
s
where
Gauge pressure in
suction tank p
e
= 0.40 bar = 40 000 Pa
Atmospheric pressure p
b
= 955 mbar = 95 500 Pa acc. to Table
Vapour pressure p
v
= 0.07 bar = 7 Pa acc. to Table
Density = 998. kg/m

acc. to Table
(p
e
+ p
b
p
v
) / ( g)
= (40 000 + 95 500 7) / (998. 9.8) = 5.4 m
ve

/g = 0
H
L,s
= 0.9 m
H
z geo
= .00 m
s = 0, as the centre of the impeller inlet is at
the same height as the centre of the pump inlet.
NPSH
a
= 7.04 m
With an NPSH
r
of 5.50 m,
NPSH
a
is larger than NPSH
r
in this case and therefore suffcient.
8.36
Pump Characteristics When
Pumping Viscous Fluids
Given: A mineral oil with a
density of
z
= 0.897 kg/m

and a kinematic viscosity of
z
=
500 06 m

/s is to be pumped
by the centrifugal pump de
scribed in 8.; characteristic
curves according to Fig. 9.
Sought: The characteristics for
discharge head, effciency and
input power when pumping this
viscous fuid, using the spread
sheet calculation as per Fig. 5.
The data for handling water
(index w) are required frst to
fnd the conversion factors:
Flow rate at BEP Q
w, opt
= 00 m

/h
Head at BEP H
w, opt
= 57.5 m
Optimum effciency
w, opt
= 0.85
Power P
w, opt
= 7.5 kW
Speed n = 900 min

Specifc speed (as per exercise 8.) n

q
= .8
Kinematic viscosity
z
= 500 0
6
m

/s
Density of mineral oil
z
= 897 kg/m

The three conversion factors f

Q
= 0.84, f
H
= 0.88, f

= 0.6 are
taken from Fig. 5.
The calculation is continued using the table below:
Q/Q
opt
0 0.8 .0 .
Q
w

refer to
0 60 00 40 m

/h
H
w

Fig. 8
66.5 6.0 57.5 5.0 m

w
0 0.8 0.85 0.805
Q
z
= Q
w
f
Q
0 4.4 68 0.6 m

/h
H
z
= H
w
= .0 H
w
f
H
= H
w
f
H
= H
w
f
H

66.5 56. 50.6 44.9 m

z
=
w
f

0 0.50 0.58 0.499

P
z
=
z
Hz Q
z
/ (
z
67)
6.8 40. 44. kW
To calculate the power P
z
, the values for fow rate Q
z
in m

/h and
the density
z
in kg/m

are inserted into the equation.

These calculated points can be used to plot the characteristic curve
for a viscous fuid, cf. Fig. 5 and Fig. 8 (this chart is applicable for
handling water with an impeller diameter of 9 mm).
76
8 Calculation Examples
8.45
Head Reduction for Hydro-
transport
Given: Grit with a density of
z = 700 kg/m

and an average
particle size of d
s
= 5 mm is to
be pumped in cold water
(kinematic viscosity
f
=
.00 . 0
6
m

/s) at a concentra
tion of c
T
= 5% with a centri
fugal pump (hydraulic data as
per 8., specifc speed n
q
= ,
head coeffcient = .0).
Sought: The head reduction H/H at H = 57.5 m.
According to Fig. 55, the settling speed w
s0
of a single sphere under
the conditions described above is 0.5 m/s. Thus, the Reynolds
number is Re
s
= w
s0
d
s
/
f
= 0.5 0.005 / .0 0 6 = 500.
The head reduction is calculated using Eq. (45):
H/H = c
T
/

Re
s
(.8/n
q
)

(
s
/
f
)
= (0.5 / .0)

500 (.8 / )

(700 / 000 )
= 0.5 .6 0.046 .70 = 0.6
H = 0.6 57.5 = 9. m
Under the above conditions the pump discharge head of H
w, opt
=
57.7 m would be reduced by 6%, i.e. 57.5 9. = 48. m.
8.47
Average Density
Given: Hydrotransport as de
scribed in exercise 8.45.
Sought: The average density

m
and its effect on the pump
discharge pressure; will it rise
or fall?
According to Eq. (47) the average density is
m
= c
T

s
+ ( c
T
)
f
where

f

w
= 998, kg/m

for water at 0 C

m
= 0.5 700 + 0.85 998. = 5 kg/m

The pressure differential according to equation (46)

p
=
m
g (H H)
= 5 9.8 (57.5 9.) = 59 700 Pa = 5.94 bar
This is higher than the discharge pressure for handling water
(p = 5.5 bar) as per exercise 8.. Hence, the characteristic curve
p = f (Q) has increased by % for hydrotransport of solids.
8.48
Pump Sump
Given: The pump sump for a
pump as per 8. with the fol
lowing data:
Inlet fow Q
in
= 0 m

/h
Flow rate at switchon pressure
Q
on
= 0 m

/h and
Flow rate at switchoff pressure
Q
off
= 50 m

/h
The maximum permissible
number of startups of a pump
unit is given in Table 0 (section
..., dry motor with P
> 0 kW, in this case Z = 0/h).
Sought: The useful volume V
N
of the pump sump according to equa
tion (48) (all fow rates in m

/h):
V
N
= Q
in
(Q
m
Q
in
) / (Q
m
Z)
where
Q
m
= (Q
on
+ Q
off
) / = (0 + 50) / = 85 m

/h
V
N
= 0 (85 0) / (85 0) = 4. m

/h
77
8
Calculation Examples
8.50
Minimum Submergence
Given: The vertical unfanged
suction pipe according to 8.9
and Fig. 8D, inside pipe diame
ter d = d
E
= 0. mm at a fow
rate of Q = 00 m

/h.
Sought: The minimum submergence S
min
in the open suction tanks.
The fow velocity v
s
in the suction pipe inlet is
v
s
= Q/A = (Q/600)/(p d
E

/4) = (00 / 600) (p 0.0

/4) = .60 m/s

Eq. (50) gives the minimum submergence as
S
min
= d
E
+ . v
s
d
E
/ g
= 0.0 + . .60 0.0 / 9.8
= 0.75 m.
The same result can be obtained faster from the diagram in Fig. 67.
Fig. 66 provides the required distance to the wall with > 0. m,
the channel width with > .6 m and the distance to the foor with
> 0.50 m.
8.52
Suction Tank Volume
Given: A centrifugal pump sys
tem, data according to 8. and
8.9, including a suction tank as
per Fig. 74. The straight length
of the airflled suction pipe
DN 00 (inside diameter d
s
=
0. mm according to Table
4) is L
s
= .00 m, with H
s geo

= .60 m (= vertical distance
between pump reference plane
and water level for positive inlet
pressure operation). The atmos
pheric pressure p
b
= 989 mbar =
98900 Pa; density of the water
at 0 C = 998. kg/m

, vapour
pressure p
v
= 7 Pa.
Sought: The volume of the suction tank according to Eq. (52):
V
B
= (d
s

p /4) L
s
p
b
/ (p
b
g H
s
)
The suction lift Hs is defned by Eq. (5):
H
s
= H
s geo
+ H
L,s
Given is H
s geo
= .60 m, the suction pipe head loss H
L,s
is to be cal
culated from H
L,s
and H
L,s
as follows:
) Head loss H
L,s
of the pipe as per 8.9:
H
L,s
= (L / d
s
) v
s

/ g
where
= 0.06 from 8.9
L = H
s geo
= .6 m (not .0 m because the elbow length is taken
into account in H
L,s
)
d
s
= 0.0 m.
v
s
= .60 m from exercise 8.9.
H
L,s
= 0.06 (.60 / 0.0) .60

/ ( 9.8) = 0.06 m
H
L,s
covers the 80 elbow ( x 90 elbow according to Table 6 as
in 8.5) and inlet pipe fttings according to Table 7.
Loss coeffcient of 80 elbow (factor .4) = .4 0.0 = 0.4
Loss coeffcient of inlet pipe ftting (broken inlet edge) = 0.0
H
L,s
= S v
s

/ g = (0.4 + 0.0) .60

/ ( 9.8) = 0.044 m
) The total head loss H
L,s
= H
vs
+ H
L,s
= 0.06 + 0.044 = 0.070
m
and therefore
H
s
= H
s geo
+ H
L,s
= .60 + 0.07 = .67 m
The example shows that the head loss H
L,s
(= 0.070) can be neglect
ed for short suction pipes, since H
s geo
(.60 m) is considerably high
er. This simplifes the calculation. The volume of the suction tank V
B

can be calculated using Eq. (5) or can simply be determined using
the graphs of Fig. 75 (provided the head loss H
L,s
is neglected).
78
8 Calculation Examples
V
B
= (d
s

p / 4) L
s
p
b
/ (p
b
gH
s
)
= (0.0

p/4) .0 98 900 / (98 900 998. 9.8 .67)

= 0.4 m

The chosen tank size is .8 times the volume of 0.40 m

(cf. example
in Fig. 75).
Check
The lowest pressure is = p
b
gH
s
= 7 88 Pa
The vapour pressure is 0.07 bar = 7 Pa
This means the pressure does not fall below vapour pressure during
venting.
79
9 Additional Literature
[] Product literature (KSB sales literature)
[] KSB Centrifugal Pump Lexicon
[] Cavitation in Centrifugal Pumps. KSB publication No.
08.05
[4] Gebudetechnik von KSB. Pumpenregelung und Anlagenau
tomation. Planungshinweise. (Building Services Products from
KSB. Pump Control and Plant Automation. Planning Informa
tion). KSB publication No. 00.04 (005)
[5] Bernauer J., M. Stark, W. Wittekind: Improvement of Propeller
Blades Used for Handling Liquids Containing Fibrous Solids.
KSB Technische Berichte e (986), pp. 6
[6] Bieniek K., Grning N.: Controlling the Output of Centrifugal
Pumps by Means of Electronic Speed Control. KSB Technische
Berichte e (987), pp. 6
[7] Bieniek K.: Submersible Motors and Wet Rotor Motors for
Centrifugal Pumps Submerged in the Fluid Handled. KSB Tech
nische Berichte e (987), pp. 9 7
[8] Holzenberger K., Rau L.: Parameters for the Selection of
Energy Conserving Control Options for Centrifugal Pumps.
KSB Technische Berichte 4e (988), pp. 9
[9] Holzenberger K.: A Comparison of Two Conversion Methods
Applied to the Characteristics of Centrifugal Pumps While
Pumping Viscous Liquids. KSB Technische Berichte 5e (988),
pp. 45 49
[0] Holzenberger K.: How to Determine the Starting Torque Curve
of Centrifugal Pumps by Using Characteristic Factors. KSB
Technische Berichte 6 (990), pp.
[] Kosmowski I., Hergt P.: Frderung gasbeladener Medien
mit Hilfe von Normal und Sonderausfhrungen bei Kreisel
pumpen (Pumping Gasladen Fluids by Standard and Special
Design Centrifugal Pumps). KSB Technische Berichte 6
(990), pp. 4 9
[] Schreyer H.: Glandless Chemical Pump with Magnetic Drive.
KSB Technische Berichte 4e (988), pp. 5 56
9.
Additional Literature
80
10
10.
Technical Annex
Fig. 3: Nomograph to determine the specifc speed n
q

Example: Q
opt
= 66 m
3
/h = 18.3 l/s; n = 1450 rpm, H
opt
= 17.5 m. Found: n
q
= 23 (metric units).
Specifc Speed
8
F
i
g
.

1
0
:

P
i
p
e

f
r
i
c
t
i
o
n

f
a
c
t
o
r

a
s

a

f
u
n
c
t
i
o
n

o
f

t
h
e

R
e
y
n
o
l
d
s

n
u
m
b
e
r

R
e

a
n
d

t
h
e

r
e
l
a
t
i
v
e

r
o
u
g
h
n
e
s
s

d
/
k

5
6
8
1
0
3
1
0
4
2
3
4
5
6
8
1
0
5
2
3
4
5
6
8
1
0
7
2
3
4
5
6
8
1
0
6
2
3
4
5
6
8
0
.
0
0
7
0
.
0
0
8
0
.
0
0
9
0
.
0
1
0
0
.
0
1
2
0
.
0
1
4
0
.
0
1
6
0
.
0
1
8
0
.
0
2
0
.
0
3
0
.
0
4
0
.
0
5
0
.
0
6
0
.
0
7
0
.
0
8
0
.
0
9
0
.
1
D

2
0
F
u
l
l
y

r
o
u
g
h

(
k

>

0
j
L
a
m
i
n
a
r
T
u
r
b
u
l
e
n
t
R
e
c
r
i
t
L
i
m
i
t
i
n
g

c
u
r
v
e
4
0
1
0
0
2
0
0
5
0
0
1
0
0
0
2
0
0
0
5
0
0
0
1
0

0
0
0
2
0

0
0
0
5
0

0
0
0

H
y
d
r
a
u
l
i
c
a
l
l
y

s
m
o
o
t
h

(
k

=

0
j
L

=

6
4
R
e
R
e
y
n
o
l
d
s

n
u
m
b
e
r

2
E
P i p e f r i c t i o n f a c t o r L
1
0
0

0
0
0
10 Pipe Friction Factor
8
F
i
g
.

1
1
:

H
e
a
d

l
o
s
s
e
s

H
L

f
o
r

n
e
w

s
t
e
e
l

p
i
p
e
s

(
k

=

0
.
0
5

m
m
)

10
Head Loss in Steel Pipes
1
0
0
5
0
2
0
1
0521
0
.
5
0
.
5
1
2
5
1
0
2
5
1
0
2
2
5
1
0
3
2
2
5
1
0
4
0
.
2
0
.
1
0
.
0
5
0
.
0
2
0
.
0
1
m
1
0
0

m
S p e c i f i c h e a d l o s s H
L
m
3
/
h
0
.
5
0
.
2
1
2
5
1
0
2
0
5
0
1
0
0
2
0
0
5
0
0
1
0
0
0
2
0
0
0
5
0
0
0
F
l
o
w

r
a
t
e

Q
l
/
s
d

=

1
5

m
m
2
0
2
5
3
2
4
0
5
0
6
5
8
0
1
0
0
1
2
5
1
5
0
1
7
5
2
0
0
2
5
0
3
0
0
3
5
0
4
0
0
5
0
0
6
0
0
7
0
0
8
0
0
9
0
0
1
0
0
0
1
2
0
0
1
4
0
0
1
6
0
0
1
8
0
0
d

=

2
0
0
0

m
m
4
.
0

3
.
5
3
.
0

2
.
5
2
.
0

1
.
2
5

1
.
0

0
.
8
0
.
6
0
.
5
0
.
4

0
.
3

1
.
5
v

=

5
.
0

m
/
s
2
0

0
0
0
1
0

0
0
0
5
0
0
0
0
2
0
0
0
0
0
1
0
0
0
0
0
5
0
0
0
0
0
1
0
0
0
0
0
0
2
0
0
0
0
0
0
R
e
=
5
0
0
0
0
0
0
.
E
W

U
N
T
R
E
A
T
E
D

S
T
E
E
L

P
I
P
E
8
1
0
0
5
0
2
0
1
0521
0
.
5
0
.
5
1
2
5
1
0
2
5
1
0
2
2
5
1
0
3
2
2
5
1
0
4
0
.
2
0
.
1
0
.
0
5
0
.
0
2
0
.
0
1
m
1
0
0

m
S p e c i f i c h e a d l o s s H
L
1
.
1
1
.
0
0
.
9
0
.
8
0
2
0
4
0
6
0
C
H
L

c
o
r
r
e
c
t
i
o
n

f
o
r

p
l
a
s
t
i
c

p
i
p
e
T e m p e r a t u r e f a c t o r J
T
e
m
p
e
r
a
t
u
r
e

t
m
3
/
h
0
.
5
0
.
2
1
2
5
1
0
2
0
5
0
1
0
0
2
0
0
5
0
0
1
0
0
0
2
0
0
0
5
0
0
0
F
l
o
w

r
a
t
e

Q
l
/
s
4
.
0

3
.
5
3
.
0

2
.
5
2
.
0

1
.
2
5

1
.
0

0
.
8
0
.
6
0
.
5
0
.
4

0
.
3

d

=

1
5

m
m
2
0
2
5
3
2
4
0
5
0
6
5
8
0
1
0
0
1
2
5
1
5
0
1
7
5
2
0
0
2
5
0
3
0
0
3
5
0
4
0
0
1
.
5
d

=

5
0
0

m
m
2
0
0
0
0
1
0
0
0
0
2
0
0
0
0
0
1
0
0

0
0
0
5
0
0
0
0
0
1
0
0
0
0
0
0
R
e
=
2
0
0
0
0
0
0
0
L
A
S
T
I
C

P
I
P
E

A
N
D

D
R
A
W
N

M
E
T
A
L

T
U
B
I
N
G
5
0

0
0
0
v

=

5
.
0

m
/
s
F
i
g
.

1
2
:

H
e
a
d

l
o
s
s
e
s

H
L

f
o
r

h
y
d
r
a
u
l
i
c
a
l
l
y

s
m
o
o
t
h

p
i
p
e
s

(
k

=

0
)


F
o
r

p
l
a
s
t
i
c

p
i
p
e

w
h
e
n

t

=

1
0

C

m
u
l
t
i
p
l
y

b
y

t
h
e

t
e
m
p
e
r
a
t
u
r
e

f
a
c
t
o
r

10
Head Loss for Hydraulically Smooth Pipes
84
10
C
a
r
b
o
n

d
i
s
u
l
p
h
i
d
e
A
c
e
t
o
n
e
B
e
n
z
o
l
n
-
B
u
t
a
n
e
E
t
h
a
n
e
E
t
h
a
n
o
l
D
i
e
t
h
y
l

e
t
h
e
r
P
r
o
p
a
n
e
i
-
B
u
t
a
n
e
B
e
n
z
o
l
P
h
e
n
o
l
T
o
l
u
o
l
A
n
i
l
i
n
e
M
e
t
h
a
n
o
l
A
c
e
t
o
n
e
A
c
e
t
i
c

a
c
i
d
F
o
r
m
i
c

a
c
i
d
G
l
y
c
e
r
i
n
C
a
r
b
o
n

d
i
s
u
l
p
h
i
d
e
A
m
m
o
n
i
a
S
u
l
p
h
u
r

d
i
o
x
i
d
e
C
a
r
b
o
n

t
e
t
r
a
c
h
l
o
r
i
d
e
B
e
n
z
o
l
100
50
40
30
20
10
5
4
3
2
1
0.5
0.4
0.3
0.2
0.1
-50 0 100 200 C 300
bar
Temperature t
v
a
p
o
u
r

p
r
e
s
s
u
r
e

p
v
Fig. 35: Vapour pressure p
v
of various fuids as a function of the temperature t
Vapour Pressures
85
10
Values required to
DIN 51507 (transformer oils)
DIN 51603 (fuel oils)
DIN 51601 (Diesel fuel)
ISO viscosity classification
to DIN 51519
G
e
a
r

o
i
l
s
A
u
t
o
m
o
b
i
l
e

g
e
a
r

o
i
l
s
H
y
d
r
a
u
l
i
c

o
i
l
s

H
L
,

H
L
P
L
u
b
r
i
c
a
t
i
n
g

o
i
l
s

L
-
A
N
L
u
b
r
i
c
a
t
i
n
g

o
i
l
s

C
L
L
u
b
r
i
c
a
t
i
n
g

o
i
l
s

C
L
u
b
r
i
c
a
t
i
n
g

o
i
l
s

C
L
P
D
i
e
s
e
l

f
u
e
l
F
u
e
l

o
i
l
s

E
L
F
u
e
l

o
i
l
s

L
F
u
e
l

o
i
l
s

M
F
u
e
l

o
i
l
s

S
S
u
p
e
r
h
e
a
t
e
d

s
t
e
a
m

c
y
l
i
n
d
e
r

o
i
l
s

Z
D
S
u
p
e
r
h
e
a
t
e
d

s
t
e
a
m

c
y
l
i
n
d
e
r

o
i
l
s

Z
A
S
u
p
e
r
h
e
a
t
e
d

s
t
e
a
m

c
y
l
i
n
d
e
r

o
i
l
s

Z
B
T
r
a
n
s
f
o
r
m
e
r

o
i
l
s
M
o
t
o
r

o
i
l
s
T
u
r
b
i
n
e
o
i
l
s

T
D
V
B
,

V
B
L
C
o
m
p
r
e
s
s
o
r

o
i
l
s
V
C
,

V
C
L
,

V
D
L
,
R
e
f
r
i
g
e
r
a
t
i
n
g
m
a
c
h
i
n
e

o
i
l
s

K
C
R
e
f
r
i
g
e
r
a
t
i
n
g
m
a
c
h
i
n
e

o
i
l
s

K
A
Lubricating
oils BC
Lubricating
oils BB
Lubricating
oils BA
1500
1000
600
500
400
300
200
60
50
40
30
20
10
15
mm
2
/s
100
80
6
5
4
3
0 50 100 150 C 200
8
Temperature t
K
i
n
e
m
a
t
i
c

v
i
s
c
o
s
i
t
y

N
Fig. 47: Kinematic viscosity of various mineral oils as a function of the temperature t
Kinematic Viscosity
86
10
E
th
y
l a
lc
o
h
o
l (m
e
th
y
l a
lc
o
h
o
lj
A
c
e
to
n
e
Diethyl ether
t = 100 C
N = 2.01 mm
2
/s
t = 98.3 84.2 72.5 44.5 C
N = 15.8 7.76 4.99 2.33 mm
2
/s
t = 92.5 C
N = 2.35 mm
2
/s
t =18.3 50 70 C
N = 11.87 3.32 1.95 mm
2
/s
1.8
1.5
1.0
0.5
0
kg
dm
3
D
e
n
s
i
t
y

C
a
r
b
o
n

t
e
t
r
a
c
h
l
o
r
i
d
e
B
e
n
zo
l
n-Butane
i-
B
u
ta
n
e
E
th
a
n
e
P
r
o
p
a
n
e
A
m
m
o
n
ia
A
m
m
o
n
i
a
D
ie
t
h
y
l e
t
h
e
r
B
e
n
z
o
l
T
o
lu
o
l
A
n
ilin
e
P
h
e
n
o
l
A
n
i
l
i
n
e
F
o
r
m
ic

a
c
id
F
o
r
m
i
c

a
c
i
d
Heavy w
a
te
r
P
h
e
n
o
l
A
c
e
tic
a
c
id
M
e
t
h
a
n
o
l
A
c
e
t
o
n
e
C
a
r
b
o
n
d
is
u
lp
h
id
e
C
a
r
b
o
n

d
i
s
u
lp
h
id
e
T
o
l
u
o
l
160 100 C 0
Temperature t
-100
1.8
1.5
1.0
0.5
0
mm
2
s
K
i
n
e
m
a
t
i
c

v
i
s
c
o
s
i
t
y

N
A
c
e
t
i
c

a
c
i
d
S
u
l
p
h
u
r
i
c

a
c
i
d
Fig. 48: Density and kinematic viscosity of various fuids as a function of the temperature t
Density and Kinematic Viscosity
87
10
(OWTO&INDTHE/PERATING0OINT
'IVEN
Flow rate Q
w
m
3
/h
Total developed head H
w
m
Speed of rotation n 1/min
Kinematic viscosity N
z
m
2
/s
Density
z
kg/m
3
Gravitational constant g 9.81 m/s
2
#ALCULATION
Q/Q
opt
= 0 0.8 1.0 1.2 -
Q
w
0 m
3
/h
H
w
m
H
w
0 -
n
q, w
from section - - - 1/min
3.1.5
f
Q, w
from Fig. 50 - -
f
H, w
- -
f
H, w
- -
Q
z
= Q
w
f
Q, w
0 m
3
/h
H
z
=

H
z
H
w
f
H, w
0
P
z
kW
1j
lf H
z
becomes larger than H
w
, it should be set to H
w
.
2)
These are four values on the H
z
/Q and Q/H
z
curve and three points on the
Q/P
z
curve. Plot versus Q.

z
g H
z
Q
z
H
z
1000 3600
= H
w
= H
w
f
H, w
1,03 H
w
f
H, w
H
w
f
H, w
1)
m
from the
pump cha-
racteristic
for 4 points
2)
Fig. 51: Spreadsheet for calculating the pump characteristics for a viscous fuid using the KSB method
Viscous Fluids Pump Characteristics
88
10
Velocity Head
V
e
l
o
c
i
t
y

h
e
a
d

v
2
/
2

g

a
s

a

f
u
n
c
t
i
o
n

o
f

f
o
w

r
a
t
e

Q

a
n
d

i
n
s
i
d
e

p
i
p
e

d
i
a
m
e
t
e
r

d
F
l
o
w

r
a
t
e

Q
V e l o c i t y h e a d
F
l
o
w

r
a
t
e

Q
V e l o c i t y h e a d
89
10
Velocity Head
V
e
l
o
c
i
t
y

h
e
a
d

d
i
f
f
e
r
e
n
t
i
a
l

(
v
2
/
2

g
)

a
s

a

f
u
n
c
t
i
o
n

o
f

f
o
w

r
a
t
e

Q

a
n
d

i
n
s
i
d
e

p
i
p
e

d
i
a
m
e
t
e
r
s

d
1

a
n
d

d
2
F
l
o
w

r
a
t
e

Q
F
l
o
w

r
a
t
e

Q
V e l o c i t y h e a d d i f f e r e n t i a l
V e l o c i t y h e a d d i f f e r e n t i a l
90
11
Physical Sym Units Units not
Recom
Comments
dimension bol
SI units Other units
to be used
mended

(not complete)
any longer
units
Length l m Metre km, dm, cm, m Base unit
mm, m
Volume V m

dm

, cm

, mm

, cbm, cdm m

litre ( l = dm

)
Flow rate, capac Q, m

/s m

/h, l/s l/s and

ity, volume fow V

/s
Time t s Second s, ms, s, ns, s Base unit
min, h, d
Speed of rotation n /s /min (rpm) /s, /min
Mass m kg Kilogram g, mg, g, Pound, kg Base unit
metric ton hundred The mass of a commercial
( t = 000 kg) weight commodity is described as
weight.
Density kg/m

kg/dm

kg/dm

The term
und spezifce gravita must no
kg/m

longer be employed, because

it is ambiguous
(see DIN 05).
Mass moment of J kg m

kg m

Mass moment, . order

inertia
Mass rate of fow m

kg/s t/s, t/h, kg/h kg/s and t/s

Force F N Newton kN, mN, N, kp, Mp, N kp = 9.8 N. The weight
(= kg m/s

) force is the product of the

mass m by the local gravi
tational constant g.
Pressure p Pa Pascal bar kp/cm

, at, bar at = 0.98 bar

(= N/m

) ( bar=0
5
Pa) m w.c., = 9.8 0
4
Pa
Torr, mm Hg = . mbar
mm w.c. = 0.098 mbar
Mechanical , Pa Pascal N/mm

, N/cm

kp/cm

, N/mm

kp/mm

= 9.8 N/mm

stress (= N/m

)
(strength)
Bending mo M, N m kp m, N m kp m = 9.8 N m
ment, torque T
Energy, work, W, J Joule kJ, Ws, kWh, kp m J und kJ kp m = 9.8 J
quantity of heat Q (= N m kW h = kcal, cal, kcal = 4.868 kJ
= W s) 600 kJ WE
Total head H m Metre m l. c. m The total head is the work
done in J = N m applied to
the mass unit of the fuid
pumped, referred to the
weight force of this mass
unit in N.
Power P W Watt MW, kW, kp m/s, PS kW kp m/s = 9.8 W
(= J/s PS = 76 W
= N m/s)
Temperature T K Kelvin C K, deg. K Base unit
difference
Kinematic m

/s St (Stokes), m

/s St = 0
4
m

/s
viscosity E, cSt = mm

/s
Dynamic Pas Pascal P (Poise) Pa s P = 0. Pa s
viscosity second
(= N s/m

)
Specifc speed n
q

n
q
= n
Q
opt
(g H
opt
)
/4
Sl units (m und s)
11. Excerpt of Important Units for Centrifugal Pumps
9
9
KSB Aktiengesellschaft
D-67225 Frankenthal (Pfalz) / Germany
Telephone +49 6233 86-0 Fax +49 6233 86-3401 www.ksb.com
0
1
0
1
.
5
/
5
-
1
0



/



0
9
.
0
6



/



O
t
t
w
e
i
l
e
r
e 46,
ISBN 3-00-017841-4