Materials Today: Proceedings 37 (2021) 2153–2157

Contents lists available at ScienceDirect

Materials Today: Proceedings

journal homepage: www.elsevier.com/locate/matpr

Impeller design and CFD analysis of fluid flow in rotodynamic pumps

Chandrasekaran M. a,⇑, Santhanam V. b,⇑, Venkateshwaran N. b
a
Vels Institute of Science, Technology and Advanced Studies, Chennai 600117, India
b
Rajalakshmi Engineering College, Chennai 602105, India

a r t i c l e i n f o a b s t r a c t

Article history: The intricate internal flows in the rotodynamic water pump impellers can be well predicted by using
Received 27 June 2020 computational fluid dynamics and mathematical modeling. An impeller is a radial type rotor that
Received in revised form 23 July 2020 increases the acceleration of the fluid due to the centrifugal action and the fluid is pumped through
Accepted 27 July 2020
the casing to a higher potential head. The efficiency and performance of the centrifugal pump can be
Available online 15 September 2020
enhanced by optimizing certain design parameters of the impeller. In this work, the impeller design
and development of potential flow calculations for the liquid flow using the MATLAB software was per-
Keywords:
formed. NX-CAD was used to model the impeller and pump. CFD analysis in ANSYS CFX showed the flow
Impeller design
Rotodynamic pumps
of liquid through the variable cross-section of the volute casing of the pump. Modification in impeller
MATLAB design showed that the overall performance of the pump can be improved. The program developed in
NX CAD the MATLAB also served as a type of calculator for determining the type of pump that can be used for var-
ANSYS CFX ious industrial purposes based on the velocity of the liquid, head and discharge rate.
Ó 2020 Elsevier Ltd. All rights reserved.
Selection and peer-review under responsibility of the scientific committee of the International Confer-
ence on Newer Trends and Innovation in Mechanical Engineering: Materials Science.

1. Introduction with various design parameters of the guide vanes. They showed
that the overall size of the pump can be reduced for the same
The requirements for energy efficient and low cost pumps had capacity due to improved aerodynamics.
driven the scientific community to develop new models for pump The impeller parameters significantly affect the overall operat-
design. Due to increasing demand for low carbon footprint several ing performance of the rotodynamic pumps as they provide the
researchers [1–6] have attempted to develop innovative materials, necessary pressure to the fluid. Careful analysis of the impeller
new processes and modifications in designs. In this work a modifi- profile can significantly improve the performance and efficiency
cation in design and selection process had been attempted to of the pump. Raj et al [12] had reviewed the application of various
reduce the energy waste in the rotodynamic pumps. The Impeller CFD techniques studied by the researchers, it was reported that
of the Centrifugal pump usually has vanes fitted between the two equation k-e turbulence model along with unsteady Navier–
shrouds or plates. The crown plate has a suction eye and the base Stokes equations (Reynolds-averaged) were giving closer predic-
plate is mounted which is keyed to a shaft. The impeller without tions on the centrifugal pump. It was also reported that further
the crown plate is known as the non-clog or semi-open impeller. studies on impeller volute interaction and fluid flow study in
Several researchers [7–10] have attempted to optimize the perfor- volute casing can improve the performance of the centrifugal
mance and improve the overall efficiency of the rotodynamic pumps. The work done by Gurupranesh et al [13] predicted the
pumps by using computational fluid dynamics (CFD) analysis and performance of the pump by using virtual prototyping and CFD
varying the impeller design parameters suitably. Junaidi et al analysis by fluid flow simulation package. The impeller vane profile
[11] had shown that the vortices and non-uniformity in the fluid was created by using solid works software using point by point
flow can be reduced by introducing inner guide vanes at the method, and the parameters such as pressure distribution, static
entrance of the centrifugal pump, they performed CFD analysis pressure head, turbulence, fluid velocity and direction of flow.
The results were closely in agreement to that of the actual values.
Zhou et al [14] had investigated the three different types of cen-
⇑ Corresponding authors. trifugal pumps with different blades by using CFX software. The
E-mail address: [email protected] (M. Chandrasekaran).

https://doi.org/10.1016/j.matpr.2020.07.637
2214-7853/Ó 2020 Elsevier Ltd. All rights reserved.
Selection and peer-review under responsibility of the scientific committee of the International Conference on Newer Trends and Innovation in Mechanical Engineering:
Materials Science.
M. Chandrasekaran et al. Materials Today: Proceedings 37 (2021) 2153–2157

problem was simulated by using standard two equation turbulence

j – e model. It was reported that the comparison of computational
results was in good agreement with that of actual values for
twisted blade arrangement but vary significantly with straight
blade arrangement. Nigussie et al [15] had used 3d modeling and
CFD simulation to determine the velocity profiles and pressure dis-
tribution. Rajendran et al [16] had used ANSYS CFX to obtain the
pressure plots, flow pattern and blade loading in the impeller pas-
sage. The delivery head predicted by CFD for the design flow rate
was 9.45 m. Tan et al [17] reported the use of direct and inverse
iteration methods to solve the continuity and boundary equation
of the fluid flow. Three types of pumps with different wrap angles
were numerically simulated, and results indicated that highest effi-
ciency and larger head was obtained for the larger blade wrap
angle. The work done by Rajmane et al [18] exhibited that model-
ing and simulation of impeller can result in precise prediction of
pump parameters for better efficiency. Kim et al [19] used response
surface methodology (RSM) to design optimized impeller that sat-
isfied design specifications, and it was verified by numerical anal-
ysis. Minggao et al [20] studied the effect of different impeller
outlet diameters on the flow field in six different types of centrifu-
Fig. 2. Impeller 3D model in NX CAD.
gal pumps. It was reported that the change of outlet parameters
directly affected the total pressure and static pressure at the volute
and at the impeller. Also, the predicted simulation results were in
accordance with the cutting law with good accuracy. The above lit-
erature shows that the studies are focused on the improvement in
the performance of the machines [21–25] by using different
approaches [26–32] for various applications and processes.
In the present study the 3D model of the impeller in a centrifu-
gal pump had been created using the NX CAD software using the
circular arc method. MATLAB coding was used to plot the charac-
teristic curve of a centrifugal pump at different speeds. Finally,
CFD analysis was performed by using ANSYS CFD to predict the
velocity profile and pressure distribution in the impeller.

2. Centrifugal pump impeller design

2.1. Design procedure

The CAD model of the impeller was done using NX CAD soft-
ware. the design procedure involves the computation of required

Fig. 1. Generation of blade profile. Fig. 3. Flowchart for re-plot of the pump curve.

2154
M. Chandrasekaran et al. Materials Today: Proceedings 37 (2021) 2153–2157

Table 1 center and the length is equal to the second radius of curvature,
Boundary Conditions Used in CFD Analysis. with H as center and BH is radius, an arc was drawn till the arc
S. No. Type of Boundary Condition Values meets the next circle. This procedure was repeated until the curve
1 Mass flow rate 0.35 m3/kg meets the external diameter of the impeller. The curve was then
2 Inlet Pressure 1 bar off-set with internal blade thickness at the entry of the blade and
3 Rotation speed 2000 rpm external blade thickness at the exit of the blade as shown in Fig. 2.
4 Number of Iterations 625

2.2. MATLAB programming for selection of pump

water path by using the laws of fluid motion. Fig. 1 shows the gen-
eration of blade profile. Initially two concentric circles representing MATLAB program was used to plot the pump characteristic
the outer and inner diameter of the impeller were drawn. In the curve which can be used as a tool for selection of pump based on
next step, five concentric circles were drawn at equal distance the speed and delivery head. The pump performance curves were
starting from the inner circle to the outer circle. Then, a line was converted into a digital data using Engauge digitizer software,
drawn from point A at an angle of the inlet blade angle till the dis- then, the data points were derived from the pump curve and
tance of first radius of curvature value. With G as center and AG is exported to the excel sheet using .csv extension. This file was then
radius, an arc was drawn till the arc meets the next circle [29–34]. loaded into the MATLAB program in the form of an array and the
Another line was drawn from B, which passes through the previous data was re-plotted again in the form of a graph. The program also
calculates all the parameters of the pump using the formulae

Fig. 4. Mass flow rate distribution.

Fig. 5. Velocity gradient of the liquid flow.

2155
M. Chandrasekaran et al. Materials Today: Proceedings 37 (2021) 2153–2157

3. Results and discussion

The 3D model of the impeller was created using NX CAD soft-

ware which was used as input in ANSYS simulation. Figs. 4–6 show
the results given by the CFD simulation. The CFD analysis showed
the mass flow rate distribution along various cross section of the
pump casing, while the velocity gradient distribution depicts the
changes in the velocity of the fluid along the various cross section
of the pump casing.
Characteristic curves are essential to be familiar with for the
selection of suitable pump. Fig. 6 shows the re-plot of the pump
performance curve using MATLAB. The design spot on the diameter
curve indicates the selection of the optimum kind of pump for a
particular application with the best efficiency available. The
Fig. 7 shows the plot of speed from 2000 rpm to 2400 rpm in steps
of 50 rpm increment versus discharge rate. The design modifica-
tion, which was done on the blades of the impeller showed an
increase in the overall efficiency which was greater by 10% than
the previous pumps.
The program that had been developed in the MATLAB helps the
Fig. 6. Re-plot of pump curve in MATLAB. industrialists in the ease of choosing the best type of pump that
can be used, according to the requirement of the customer. This
also helps in reducing the time in choosing the pump.

4. Conclusions

The vane profile was developed in NX-CAD software by circular

arc method. The CFD analysis showed the distribution of various
parameters along various cross section of the pump which indi-
cated the proper flow of the fluid. The MATLAB program helped
in choosing the best suitable pump according to the requirement
and the application. A framework was built to perform simplified
pump calculation and for displaying actual pump performance
curve data for manufacturer. A systematic procedure was devel-
oped to digitize experimental data from images to make use of in
the program. Linear interpolation was used to calculate realistic
pressure drop data based upon the simplified pump model and
manufacturer’s model. This framework can be applied to extract
and store the performance model of 100 s of pump model that
can be used for commercial purpose. Such a tool can be useful
for designers and analyst in selecting the pump that meets the
design and cost requirement.

Declaration of Competing Interest

Fig. 7. Plot of speed versus discharge. The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared
to influence the work reported in this paper.

obtained from the velocity triangle and then links the data points References
obtained with the pump performance curve and the data obtained
from the formulae. The final graph obtained was used to choose the [1] S. Ramasubramanian, M. Chandrasekaran, S. Baskar, A. Dowhithamaran, Mater.
pump with best efficiency, mass flow rate and pressure head. The Today (2020), https://doi.org/10.1016/j.matpr.2020.05.637.
[2] Jishuchandran, K. Manikandan, R. Ganesh, S. Baskar, Int. J. Ambient Energy
flow chart of the program used in this study is presented in Fig. 3. (2020) 1–16, https://doi.org/10.1080/01430750.2020.1725637.
[3] S. Baskar, V. Vijayan, I.I. Premkumar, D. Arunkumar, D. Thamaran, Mater Today.
(2020), https://doi.org/10.1016/j.matpr.2020.05.352.
[4] B.S. Kumar, V. Vijayan, N. Baskar, Mech. Mech. Eng. 20 (2016) 347–354.
[5] D. Arunkumar, M. Ramu, R. Murugan, S. Kannan, S. Arun, S. Baskar, Mater
2.3. CFD analysis of the fluid flow in the pump Today. (2020), https://doi.org/10.1016/j.matpr.2020.01.220.
[6] D. Arunkumar, M. Ramu, R. Murugan, S. Kannan, S. Arun, Sanjeevi Baskar, Mat
ANSYS CFD had been used in this study to simulate the fluid Today. 1–5. https://doi.org/10.1016/j.matpr.2020.01.220
[7] S.R. Shah, S.V. Jain, R.N. Patel, V.J. Lakhera, Procedia Eng. 51 (2013) 715–720.
flow in the centrifugal pump, the flow was assumed to be turbu-
[8] L. Hedi, K. Hatem, Z. Ridha, Int. Renewable Energy Congress (2010) 300–304.
lent and the standard j – e model was used to simulate the fluid [9] N. Dash, A.K. Roy, K. Kumar, Mater Today 5 (2018) 4460–4466.
flow. Standard wall conditions were considered for the volute [10] J.P. Yuan, Y.W. Zhu, A.X. Ge, J. Adv. Mater. Res 945 (2014) 914–923.
casing, impeller and pump walls. The input velocity was assumed [11] M.A. Junaidi, N.L. Kumari, M.A. Samad, G.S. Ahmed, Mater Today 2 (2015)
2073–2082.
to be constant and the factors considered in the analysis is given [12] P.P. Raj, P.S. Gupta, P.P. Kumar, G. Paramesh, K. Rohit, Mater Today 21 (2020)
in Table 1. 175–183.

2156
M. Chandrasekaran et al. Materials Today: Proceedings 37 (2021) 2153–2157

[13] P. Gurupranesh, R.C. Radha, N. Karthikeyan, IOSR-JMCE, (2010) 2278–1684. [25] R. Muraliraja, J. Sudagar, R. Elansezhian, A. Raviprakash, R. Dhinakaran, V.S.
[14] W. Zhou, Z. Zhao, T.S. Lee, S.H. Winoto, Int. J. Rotating Mach. 9 (2003) 49–61. Shaisundaram, M. Chandrasekaran, Arabian J. Sci. Eng. 44 (2019) 821–828.
[15] T. Nigussie, E. Dribssa, Int. J. Eng. Res. Gen. Sci. 3 (2015) 668–677. [26] R. Muraliraja, R. Elansezhian, Trans. he IMF 93 (2015) 126–132.
[16] S. Rajendran, D.K. Purushothaman, Int. J. Eng. Res. Technol. 3 (2012) 2278– [27] R. Muraliraja, D. Sendilkumar, D. Elansezhian, Int J Electrochem Sci 10 (2015)
2281. 5536–5547.
[17] L. Tan, B. Zhu, S. Cao, H. Bing, Y. Wang, Chin. J. Mech. Eng.-En 27 (2014) 171– [28] S. Baskar, V. Vijayan, S. Saravanan, A.V. Balan & A. Godwin Antony,
177. International Journal of Mechanical Engineering and Technology, 9 (2018)
[18] S.M. Rajmane, S.P. Kallurkar, Int. Res. J. Eng. Technol. 2 (2015) 984–988. 91–96.
[19] J.H. Kim, H.C. Lee, J.H. Kim, S. Kim, J.Y. Yoon, Y.S. Choi, J. Mech. Sci. Technol. 29 [29] A. Godwin Antony, V.Vijayan, S.Saravanan, S.Baskar, M.Loganathan,
(2015) 215–225. International Journal of Mechanical Engineering and Technology, 9 (2018)
[20] T. Minggao, L. Houlin, W. Yong, W. Kai, D. Liang, J. Irrig. Drain E-ASCE 5 (2009) 681–691.
314–318. [30] S. Saravanan, A. Godwin Antony, V. Vijayan, M. Loganathan, S. Baskar,
[21] V. S Shaisundaram, M. Chandrasekaran, Mohan Raj, S., & Muraliraja, R. International Journal of Mechanical Engineering and Technology, 1 (2019)
International Journal of Ambient Energy, 41(2020), 98–104. 785–790.
[22] V.S. Shaisundaram, M. Chandrasekaran, M. Shanmugam, S. Padmanabhan, R. [31] S. Dinesh, A. Godwin Antony, K. Rajaguru, V. Vijayan, Mech. Mech. Eng. 21
Muraliraja, L. Karikalan, Int. J. Ambient Energy (2019) 1–5. (2017) 17–28.
[23] V.S. Shaisundaram, M. Chandrasekaran, S. Mohan Raj, R. Muraliraja, T. [32] S. Dinesh, K. Rajaguru, V. Vijayan, A.Godwin Antony, Mechanics and
Vinodkumar, Int. J. Ambient Energy 40 (2019) 699–703. Mechanical Engineering , 20(2016) 451–466.
[24] V.S. Shaisundaram, L. Karikalan, M. Chandrasekaran, Int. J. Vehicle Struct. Syst.
11 (2019).

2157