Centrifugal PuMP

Dr. Ir. Harinaldi, M.Eng

Mechanical Engineering Department
Faculty of Engineering University of Indonesia
Pumping System in an Industry

Centrifugal Pump
Construction and Component
Casing

Volute
- area enlarge along flow direction
- create uniform velocity distribution
Diffuser
- large size centrifugal pump
- guide vanes surround the impeller
- fluid flow decelerated while
directed to enter the volute
Working Principles

sure
s
pre

gy
er
en
tic
ne
Ki
Fluid
Installation
Inlet head : Outlet head : Total head developed by the pump:
pi Vi 2 po Vo2 po  pi Vo2  Vi 2
 Z i  Z o H   Z o  Z i 
g 2 g g 2 g g 2g

H  H s   losses
 H s  h fi  h fo  hin  hout

H = manometric head
hfi = friction loss at inlet
hfi = friction loss at outlet
hin= inlet loss
hout = outlet loss
Impeller
Theoretical Assumptions:
 No tangential flow in
the blade passage
 Impeller blades are
infinitely thin
 No Velocity variation
across impeller width
 Analysis only at inlet
and outlet
 Radial inlet flow
Head and Flow Capacity H - Q
Theoretical Head Rise / Euler Head

h  E 
U 2C x 2   U1C x1 
g
1

2g
    
C22  C12  U 22  U12  W22  W12 
Flow Capacity/Flow Rate

Q  2r1Cr1b1  2r2Cr 2b2

slip factor : slip
'
C x2
s 
Cx 2
 STODOLA PROPOSAL
C x  e
STODOLA PROPOSAL
If the number of blades is Z, and impeller circumference is
2r2 then the distance between blades is 2r2/Z = 2e/sin 
Then :
Other Slip Factor
Stodola  20o <  < 30o

 s  1
 sin  2 
Z 1  Cr 2 U 2 cot  2 
Buseman  30o <  < 80o

s 
 A  BCr 2 U 2  cot  2 
1  Cr 2 U 2 cot  2 
A and B are function of  2 , Z and r2 r1
Stanitz 80o <  < 90o
0.63
 s  1
Z 1  Cr 2 U 2 cot  2 
Example
The impeller of a centrifugal pump has backward-facing blades inclined at
30o to the tangent at impeller outlet. The blades are 20 mm in depth at the
outlet, the impeller is 250 mm in diameter and it rotates at 1450 rpm. The flow
rate through the pump is 0.028 m3/s and a slip factor of 0.77 may be
assumed. Assume also the blades of infinitesimal thickness. Determine the
theoretical and actual head developed by the impeller, and the number of
impeller blades

Solution:
Flow Capacity/Flow Rate

Q  D2Cr 2b2
 Cr 2  Q D2b2
Cr 2  0.028  0.250.02
Cr 2  1.78 m/s
For ideal outlet velocity triangle = 30o

Wx 2  Cr 2 tan 30o  1.78 / tan 30o  3.08 m/s

U 2  D2 N 60   0.251450 / 60  19 m/s
C x 2  U 2  Wx 2  19  3.08  15.92 m/s
Theoretical (Euler) head

U 2C x 2  U1C x1
E  C x1  0( flow enters radially at inlet )
g

E
1915.92  30.83 m (ans.)
9.81
Actual head with slip

C x' 2   s .C x 2
 E N   s .E  0.77 30.83  23.74 m (ans.)
Number of blade
 s  1   sin  2  Z 1  Cr 2 U 2  cot  2 
  
0.77  1   sin 30o Z 1  1.78 19 cot 30o 
 Z  8.15  8 (ans.)
Pump Losses
1. Mechanical friction power 2. Impeller (Disc) friction
loss, Pm power loss, Pi

3. Leakage and recirculation 4. Casing power loss, Pc

power loss, Pl
Pump Losses
1. Mechanical friction power loss, Pm
Pump Losses
2. Impeller (Disc) friction power loss, Pi

Head loss : hi
Flow rate : Qi Pi = g Qi hi
Pump Losses
3. Leakage and recirculation power loss, Pl

Head across impeller : Hi

Leakage flow rate : q = Qi - Q
Pl = g qi Hi
Pump Losses
4. Casing power loss, Pc

Head loss : hc
Flow rate : Q Pc = g Qhc
Pump Losses
H-Q Diagram
Efficiency

Fluid power developed by pump gQH

Overall Efficiency   o 
shaft power input Ps

Fluid power at casing outlet

Casing Efficiency 
Fluid power at casing inlet
Fluid power at casing outlet

Fluid power developed by impeller - Leakage loss
gQH H
 c  
gQH i H i

Flow rate through pump Q Q

Volumetric Efficiency   v  
Flow rate through impeller Q  q Qi
Efficiency
Fluid power at impeller exit
Impeller Efficiency 
Fluid power supplied to impeller
Fluid power at impeller exit

Fluid power developed by impeller  impeller loss
gQi H i Hi
 i  
gQ i H i  hi  H i  hi 
Fluid power supplied to impeller
Mechanical Efficiency 
Power input to the shaft
gQi hi  H i 
 m 
Ps
Actual head developed by pump
Hydraulic Efficiency 
Theoretical head developed by impeller
H H
 H  
H i  hi  E
Efficiency Relation

 H   ci
 o   ci v m   H v m
Pump Shaft Power, Ps

Ps  Pm  g hi Qi  hc Q  H i q  QH 

Driven Motor Shaft Power, PM

Transmission Efficiency, T

PS
Ps  T PM PM 
T
Pump’s Characteristic Curve
E  U 2 U 2  Q A cot  2  g
E  K1  K 2Q

EN  K1  K 2Q s

hshock  K3 Q  QD 
2

where :
QD is design flowrate

h f  K 4Q 2
Effect of Flow Rate Variation
Inlet velocity

Outlet velocity
Q;H Q;H
Effect of Blade Outlet Angle
C x 2  U 2  Cr 2 cot  2

for  2  90o E  U 2 U 2  Cr 2 cot  2  g

E  U g   QU
2
2 2 cot  2 gA  H  a  bQ
for  2  90o H a
for  2  90o H  a  bQ
Effect of Blade Outlet Angle
Theoretical
characteristic curves

Actual
characteristic curves
Flow in the Discharge Casing
Volute Casing
Function:
1. Collector
2. Diffuser

Deviation in capacity from

the design condition will
result in a radial thrust (P):

P  495 KHD2 B2

where : K  0.36 1  Q QD 
2

Function:
P = radial force (N)
H = Head (m)
D2 = peripheral diameter (m)
B2 = impeller width (m)
Circular section to
reduce losses due to
friction and impact
Flow in the Discharge Casing
Vaneless Diffuser
Flow in the Discharge Casing
Vaneless Diffuser Continuity:
m  ACr  2rbCr  2r2b2  2Cr 2
Cr  r2b2  2Cr 2 rb
Conservation of angular momentum:
C x  C x 2 r2 r  usually C x  Cr
Then: C  Cx
Radius, r 
C  C x 2 r2 r Outlet kinetic energi 

tan  '2  C x 2 Cr 2  cons  tan  '

rd
tan  ' 
dr
Then:
   2  tan  ' lnr r2 
  angle of diffuser
Flow in the Discharge Casing
Vaned Diffuser
Able to diffuse the outlet
kinetic energy at:
 Higher rate
 Shorter length
 Higher efficiency

Number of vanes on the

diffuser ring:
 Greater number  better
diffusion but more friction
loss
 Square cross section of
diffuser channel  max rh
 Number of diffuser vanes
have no common factor
with the number of
impeller
Flow in the Discharge Casing
Contribution of each section of the pump to total head
Cavitation in Pumps
Vapour bubbles formation of the liquid as the local
absolute static pressure of a liquid falls below the vapour
pressure
 occurs mainly at the suction side (at the eye of impeller as the
velocity increases and pressure decreases)
 Local pitting of impeller  cavitation erosion
 Noise
 Decrease pump efficiency
Net Positive Suction Head (NPSH)
The difference of total suction head in the impeller inlet
side (impeller eye) above the vapour pressure

 pi Vi 2 pvap 
NPSH      all pressures are absolute
 g 2 g g 
 A measure of the energy available on the suction side of the
pump
 A measure to indicate the occurrence of cavitation

Cavitation Parameter (Toma Cavitation Number)

 pi Vi 2 pvap 
   
NPSH  g 2 g g 
 
Head Developed by pump H
Net Positive Suction Head (NPSH)
NPSH Required (NPSHR)
 Net Suction Head as required by the pump
in order to prevent cavitation for safe and
reliable operation of the pump.
 The required NPSHR for a particular pump
is in general determined experimentally by
the pump manufacturer (will vary
depending on the size and speed of the
pump) and a part of the documentation of
the pump.

Measurement of NPSHR by 3%
head reduction

Example of pump documentation

Net Positive Suction Head (NPSH)
NPSH Available (NPSHA)
 The Net Positive Suction Head
made available the suction
system for the pump.
 The NPSHA can be determined
during design and construction,
or determined experimentally
from the actual physical system
and calculated with the Energy
Equation

Energy at 1 = Energy at 2 + Energy lost between 1 and 2

p1 p2 V22 p2 V22 p1
 z1     losessinlet     z1   lossesinlet
g g 2 g g 2 g g
At inlet p2 = pi ; V2 = Vi and  lossesinlet = hin + hfi, then:
NPSH available at impeller inlet :
p1 pvap
NPSH A    z1  hi  h fi
g g
Cavitation ~ NPSH
To avoid cavitation in a pump operation
NPSH A  NPSH R or A R
Suction Specific Speed
A function due to cavitation that influences the efficiency
Dimensionless suction specific speed

NQ1/ 2
N suc    f  , N suc 
g NPSH 3 / 4
Cavitation parameter

Ns

 NPSH 
3/ 4
  3/ 4
N suc H 3/ 4

Similarity Laws
NPSH 1
  N1 N 2  D1 D2    1  2
2 2

NPSH 2
Example
When a laboratory test was carried out on a pump, it was found that, for a
pump total head of 36 m at discharde of 0.05 m3/s, cavitation began when the
sum of the static pressure plus the velocity head at inlet was reduced to 3.5
m. The atmospheric pressure was 750 mmHg and the vapour pressure of
water 1.8 kPa. If the pump is to operate at a location where atmospheric
pressure is reduced to 620 mmHg and the vapour pressure of water is 830
Pa, what is the value of the cavitation parameter when the pump develops the
same total head and discharge? Is it necessary to reduce the height of the
pump above the supply, and if so by how much?