Anda di halaman 1dari 6

Journal of Scientific &CHAKRABORTY


Vol. 72, June 2013, pp. 373-378


Performance prediction of Centrifugal Pumps with variations of blade number

Sujoy Chakraborty1, Kishan Choudhuri2, Prasenjit Dutta3, Bishop Debbarma4

Department of Production Engineering N.I.T Agartala, Tripura, India

Received 25 July 2012; revised 07 December 2012; accepted 28 March 2013

Centrifugal pumps are used extensively for hydraulic transportation of liquids over short to medium distance through
pipelines where the requirements of head and discharge are moderate. At present, the influence of blade number on inner flow field
and the characteristics of centrifugal pump have not been understood completely. The use of numerical analysis tools allow us to
obtain data in inaccessible positions for the experimentations. In this paper, a two dimensional numerical study of steady, static
pressure distribution and incompressible flow characteristics inside the passage with different numbers of blades of centrifugal
pump impeller has been carried out. The investigation focuses mainly on the efficiency of the pump. Centrifugal pumps with
impeller blades 5, 6 and 7 have been modeled and its efficiency at 3000 rpm is evaluated by FLUENT 6.3 software. The numerical
analysis displays that with the increase of blade number, the head and static pressure of the model increases, but the efficiency of
centrifugal pump varies with number of blades and shows maximum for 7 number of blade.
Keywords: Centrifugal pump, Blade number, CFD, Numerical analysis, Performance prediction.

From such literature, it was found that most previous
research, especially research based on numerical
approaches, had focused on the design or near-design
state of pumps. Few efforts were made to study the offdesign performance of pumps. Centrifugal pumps are
widely used in many applications, so the pump system
may be required to operate over a wide flow range in
some special applications. Thus, knowledge about offdesign pump performance is a necessity. On the other
hand, it was found that few researchers had compared
flow and pressure fields among different types of pumps.
Therefore, there is still a lot of work to be done in these
fields. A centrifugal pump delivers useful energy to the
fluid on pump age largely through velocity changes that
occur as this fluid flows through the impeller and the
associated fixed passage ways of the pump. It is
converting of mechanical energy to hydraulic energy of
the handling fluid to get it to a required place or height by
the centrifugal force of the impeller blade. The input
power of centrifugal pump is the mechanical energy and
such as electrical motor of the drive shaft driven by the
prime mover or small engine. The output energy is
*Author for correspondence

hydraulic energy of the fluid being raised or carried. In a

centrifugal pump, the liquid is forced by atmospheric or
other pressure into a set of rotating vanes. A centrifugal
pump consists of a set of rotation vanes enclosed within
a housing or casing that is used to impart energy to a
fluid through centrifugal force 1.
Computational Fluid Dynamics (CFD) is the analysis
of systems involving fluid flow, heat transfer and
associated phenomena such as chemical reactions by
means of computer-based simulation. The heat and mass
transfer, fluid flow, chemical reaction, and other related
processes that occur in engineering equipment, in the
natural environment, and in living organisms play a vital
role in a great variety of practical situations. Nearly all
methods of power production involve fluid flow and heat
transfer as essential processes. The same processes
govern the heating and air conditioning of buildings. Major
segments of the chemical and metallurgical industries
use components such as furnaces, heat exchangers,
condensers, and reactors, where thermo-fluid processes
are at work. Aircraft and rockets owe their functioning
to fluid flow, heat transfer, and chemical reaction. CFD
constitute a new third approach in the philosophical study
and development of the whole discipline of fluid
mechanics. In the seventeenth century, the foundations
of experimental fluid dynamics were laid in France and



England. The eighteenth and nineteenth centuries saw

the gradual development of theoretical fluid dynamics,
again primarily in Europe. As a result, throughout most
of the twentieth century the study and practice of fluid
dynamics involved the use of pure theory on the one
hand and pure experiment on the other hand. However
the advantage of high speed digital computers combined
with the development of accurate numerical algorithms
for solving physical problems on these computers as
revolutionized the way we study practical fluid dynamics
today. It has introduced a fundamentally important new
third approach in fluid dynamics i.e. the approach of
computational fluid dynamics.
CFD analysis is very useful for predicting pump
performance at various mass-flow rates. For designers,
prediction of operating characteristics curve is most
important. All theoretical methods for prediction of
efficiency merely give a value; but one is unable to
determine the root cause for the poor performance. Due
to the development of CFD code, one can get the
efficiency value as well as observe actual. The prediction
of behavior in a given physical situation consists of the
values of the relevant variables governing the processes
of interest. Let us consider a particular example. In a
combustion chamber of a certain description, a complete
prediction should give us the values of velocity, pressure,
temperature, concentrations of the relevant cherI1ical
species, etc., throughout the domain of interest; it should
also provide the shear stresses, heat fluxes, and mass
flow rates at the confining walls of the combustion
chamber. The prediction should state how any of these
quantities would change in response to proposed changes
in geometry, flow rates, fluid properties, etc.
Computational Fluid Dynamics is now an established
industrial design tool, helping to reduce design time scales
and improve processes throughout the engineering world.
CFD provides a cost-effective and accurate alternative
to scale model testing with variations on the simulation
being performed quickly offering obvious advantages.
The CFD occupies today a very significant place in the
disciplines of fluid mechanics and turbo machinery due
to the great progress in the development of numerical
methods and computing power. However, the initially use
of CFD tools to design a new machine represents a non
realistic procedure (Ar none, 1999) 2. The design of a
new machine (or upgrading an existing machine) would
require a great investment of time without guarantee of
success. Along with the introduction of CFD tools, its

incorporation of computer aided design (CAD) codes

has speeded up the design process because of a faster
geometry and grid generation (Asuaje, 2002) 3 .
Nevertheless, the problem always reduces down to the
selection of reasonable values for a number of geometric
parameters. At this point, the know-how, skills and
talent of the designer remain the principal ingredients
for designing and optimizing a machine. This study
assesses a two-dimensional numerical analysis of steady,
static pressure distribution and incompressible flow
characteristics inside the passage between 5, 6 and 7
number of blades of centrifugal pump impeller at 3000 rpm.
CFD code of commercial software Fluent 6.3 is used
to simulate the inner flow field under steady condition.
The standard k-e turbulence model and SIMPLEC
algorithm applied to solve the RANS equations. The
standard k-e model is a semi-empirical model based on
model transport equations for the turbulence kinetic (k)
and its dissipation rate (e). The model transport equation
for k is derived from the exact equation, while the model
transport equation for e is obtained using physical
reasoning and bears little resemblance to its
mathematically exact counterpart. In the derivation of
the k-e model, it is assumed that the flow is fully turbulent,
and the effects of molecular viscosity are negligible. The
simulation is steady and moving reference frame is
applied to take into account the impeller-volute interaction
due to convergence precision of residuals 10-5 . For
modeling of centrifugal pumps with impeller blade from
5 to 7, GAMBIT, a preprocessor of CFD code of
commercial software Fluent 6.3 has been used.
Mathematical Formulation

Mathematical model can be defined as the

combination of dependent and independent variables and
relative parameters in the form of a set of differential
equations which defines and governs the physical
phenomenon. In the following subsections differential
form of the governing equation are provided according
to the computational model and their corresponding
approximation and idealizations.
Governing Equations

The steady, conservative form of Navier-Stokes

equations in two dimensional form for the incompressible
flow of a constant viscosity fluid are as follows:




Table 1Pumps geometry

C U = 0


X- momentum:
(U U) (V U)
= - n + 1 ( U + U ) (2)
C R C2 U 2

Y- momentum:
( U V) ( V V )
2 V 2V
=- n + 1 (
C R C2 U 2

u D
X = x , Y = , P = P 2 ,U = u , V = v , R =
D n u

X, Y represents Global co-ordinate; U, V represents

Velocity along X, Y co-ordinate; Pn represents Pressure
in global co-ordinate; u, v represents velocity along x, y
co-ordinate; p represents pressure in local co-ordinate;
Re = Reynolds number.
Transport Equation for the Standard k- model

The simplest and most widely used two-equation

turbulence model is the standard k- model that solves
two separate transport equations to allow the turbulent
kinetic energy and its dissipation rate to be independently
determined. The transport equations for k and in the
standard k- e model are:


De = [(m + t ) e ] + C e (+G + C Gb) - C 2
1e k
2 k
Dt x
Where turbulent viscosity,

mt = cm ke

Circular arc
80 mm
168 mm
80 mm
52 mm



DK = [(m + t ) k ] + G + G - e - U
Dt x

Blade number
Inlet blade angle
Outlet blade angle
Shape blade
Impeller inlet diameter
Impeller outlet diameter
Volute casing
Inlet diameter
Volute tongue radius

In these equations, Gk represents the generation of

turbulence kinetic energy due to the mean velocity
gradients. Gb is the generation of turbulence kinetic
energy due to buoyancy. k and s are the turbulent
Prandtl numbers for k and , respectively. represents
density. YM represents the contribution of the fluctuating
dilatation in compressible turbulence to the overall
dissipation rate. C1 , C2 , C3 , C are constants. All
the variables including turbulent kinetic energy k, its
dissipation rate are shared by the fluid and the volume
fraction of each fluid in each computational volume is
tracked throughout the domain.
Pumps Geometry

The pump (impeller + volute) parameters are

presented in Table 1.
Boundary Conditions

Pressure inlet and pressure-outlet are set as boundary

conditions. As to wall boundary condition, no slip
condition is enforced on wall surface and
standard wall function is applied to adjacent region.
Grid Independent Test

The grid independence test has been done for 5, 6

and 7 bladed impeller centrifugal pump at 3000 (rpm)
rotational speed. In the grid independence test, maximum
total pressure has been taken as a criterion for
independence. Based on the different grids, analysis has
been made and it was observed that after refining the
grid from nodes 316798 for every blade at 3000 rpm,
results are not varying significantly. So, nodes 316798
have been used for further analysis.



Blade no. 5
Blade no. 6
Fig.1Static pressure distribution at the mid span for different impellers

Blade no. 5

Blade no. 6

Blade no. 7

Blade no. 7

Fig. 2Total pressure distribution at the mid span for different impellers

Numerical simulation
The CFD analysis of the centrifugal pump with 5, 6
and 7 bladed impeller at 3000 rpm has been shown below.
Simulation and analysis of inner flow field
Static pressure (Bar) distribution at 3000 rpm

Static pressure (Bar) distribution at the mid span of

the pump is shown in Fig.1 From Fig.1,
it can be seen clearly that for different blade number, the
static pressure gradually increase from impeller inlet to
outlet, the static pressure on pressure side is evidently
larger than that on suction side at the same impeller
radius. With the increase of blade number, the static
pressure at volute outlet grows all the time and the
uniformity of static pressure distribution at screw section
become worse and worse, but at diffusion section become
better and better. The impellers with different blade
number all have an obvious low pressure area at the

suction side of blade inlet. With the increase of the blade

number, the area flow pressure r egion grows
continuously, which indicates that the blade number has
significant effects of pumps characteristics.
Total pressure (Bar) distribution at 3000 rpm

Total pressure (Bar) distribution at the mid span of

the pump is shown in Fig.2 In Fig.2 total pressure
distribution of centrifugal pump with different
blade number s has been shown. From that
it can be seen clearly that for different blade number, the
total pressure gradually increases. It is also seen that
the pressure is less in the impeller inlet side whereas
pressure is more in the impeller outlet side.
Velocity vectors

The relative velocities inside the impeller are shown

in the fig.3. The flow rate is lower than the nominal one
and we could see the trend the separation on the suction


Fig. 3Relative velocities inside the impeller.


Fig. 4 - Absolute velocities near the tongue.

side of the trailing edge. From the figure it is clearly

visible the flow direction of the impeller. Fig 3. The
absolute velocity vectors near the tongue for a flow rate
greater than the nominal are represented in the fig.5.
Here is clearly visible the separation on the outlet side
and the blockage between the tongue and the impeller.
Curve characteristics

Fig. 5 shows the static pressure distribution along

the space between blades. The x- axis of the plot denotes
the distance along the circumference between the blades.
The lower part of the figure is towards the impeller inlet
and upper part of the figure is towards the impeller outlet
of the pump. As it is seen from the figure the pressure is
less in the impeller inlet side whereas pressure is more
in the impeller outlet side.

Fig. 5Static pressure (Bar) distribution at various positions

(near the blade region)

Table 2Predicted values of head and efficiency


Head and Efficiency calculation

Head H/m
Efficiency h/%

Head H of centrifugal pump is calculated

as follows 4:
H = out - in


Where pout is the total pressure of volute outlet,

pin is the total pressure of impeller inlet, r is density of
the fluid, d, and g is the gravity acceleration.
Total efficiency h is calculated as follows:

h = ( 1 + d + 0.03)
h v hh


number 5

number 6

number 7

where Pe is the water power and Pe=g QH, DPd is

the disk friction loss, calculation method is described in
Ref.5. hh is the hydraulic efficiency and hv is the volume
Results and discussion
Rotational speed and total pressure is an important
parameter to calculate head as well as the total
efficiency. The head and efficiency of pump model with
different blade number under design condition are shown
in Table.2. From Table 2, its easily visible that with the
increase of blade number the head increases. But the



efficiency of centrifugal pump varies with number of

blades. Here at 7 numbers of blades, the efficiency is of
maximum. So the optimum blade number of the model
pump in this paper for efficiency is 7 respectively. It is
observed that with the increase of blade number, the
static pressure is gradually increasing. With the increase
of blade number the static pressure at volute outlet grows
all the time.
The numerical studies on performance of centrifugal
pump were investigated by using the FLUENT 6.3
software. It has been observed that with the increase of
the blade number, the limitation between blade and flow
stream gets more, and also the area of low pressure
region at the suction of the blade inlet grows continuously
and the static pressure is gradually increasing .The
uniformity of static pressure distribution at screw section
become worse and worse, while at diffuser section, it
becomes better and better. With the increase of blade
number, the head of centrifugal pump grows all the time
and pressure too, but the change regulations of efficiency
is little bit complex. It varies with number of blades. Here
the optimum blade number of the model pump in this
paper for efficiency is 7 respectively.
Authors are very much grateful to the reviewers for
their valuable advice for significant modification and
better presentation of the paper.

Lamloumi H, Kanfoudi H & Zgolli R, Numerical Flow

Simulation in a Centrifugal Pump, Int Renew Energy Cong,
November, (2010) 5-7.
Arnone P, Boncinelli A, Munari & Spano E., Application of
CFD techniques to the design of the Ariane 5 turbopump, in
Proc. 14th AIAA Comp Fluid Dyn Conf, Norfolk, USA,(1999)
Asuaje M, Bakir F, Kouidri S, Noguera R, & Rey R, Validation
dune demarche de dimensionnement optimise des roues
centrifuges 2D par comparaison avec les outils de la simulation
nuerique (CFD), in 10eme Conf Annu de la Soci Canadienne
de la CFD, Windsor, Ontario, Canada, June (2002) 560565.











Minggao T, Shouqi Y & Houlin L, Numerical research on

performance prediction for centrifugal pumps[J]. Chin J of
Mech Eng, 23(1) (2010) 21-26.
TAN M, Prediction research on energy characteristics for
centrifugal pumps[D]. Zhenjiang: Jian Univ, (2008).
Houlin L, WANG Y, YUAN S, Minggao T, & WANG K,
Effects of BladeNumber onCharacteristics of Centrifugal
Pumps, Chin J of Mech Eng, 23 (2010).
Tuzson J, Centrifugal pump design, Joh Wil & So, ISBN 0-47136100-3, (2000).
Stepano A. J, Centrifugal and Axial Flow Pumps, Joh Wil &
So, New York, (1957).
Ferziger J. H. & Eric M. P, Computational Methods for Fluid
Dynamics, Sprin, Berl, Germany, (1996).
Parrondo-Gayo J. L., Erez J, & Francos J, The eect of the
operating point on the pressure uctuations at the blade passage
frequency in the volute of a centrifugal pump, Trans of the
ASME, J of Fluids Eng, 124 (2002) 784790.
Gonz alez J, Fern andez-Francos J, Blanco E, & SantolariaMorros C, Numerical simulation of the dynamic eects due to
impeller-volute interaction in a centrifugal pump, Trans of the
ASME, J of Fluids Eng, 124 (2002) 348355.
Blanco-Marigorta E, Fern andez-Francos J, Parrondo-Gayo J.
L, & Santolaria-Morros C, Numerical simulation of centrifugal
pumps, in Proc. ASME Fluids Eng Summ Conf (FEDSM 00),
Boston, Mass, USA, June (2000).
Yedidiah S, Science in the design of uid machines, in Proc.
ASME Fluids Eng Div Summ Meeting (FEDSM 02), Montreal,
Quebec, Canada, (2002).
Sloteman D, Saad A & Cooper P, Design of custom pump
hydraulics using traditional methods, in Proc. ASME Fluids
Eng Div Summ Meeting (FEDSM 01),New Or-leans, La, USA,
May (2001).
Japikse, D, Marscher, W. D, & Furst, R. D, Centrifugal Pump
Design and Performance, Conc ETI, Inc., Whit Riv Junc, VT, (1997).
Coutier-Delgosha, O, Fortes-Patella, R, Reboud, J.L & Hakimi,
N, Numerical simulation of cavitating flow in an inducer
geometry, 4th Euro Conf Turbo, Firenze, Italy, 20-23 march (2001).
Lohrberg H, Stoffel B, Fortes-Patella R & Reboud JL, Numerical
and Experimental Investigations on the Cavitating Flow in a
Cascade of Hydrofoils, Proc of the 4th Int. Symp. Cavi, Pasadena,
California, June (2001).
Stewart, W. L., Analysis of Two-Dimensional CompressibleFlow Loss Characteristics Downstream of Turbomachine Blade
Rows in Terms of Basic Boundary-Layer Characteristics, NACA
TN-3515, July (1955).
Gulich, J. F, Centrifugal Pumps, Sprin Verl, Berlin, Germany,