Numerical_investigations_on_effect_of_wear-ring_cl+(1)
Numerical_investigations_on_effect_of_wear-ring_cl+(1)
Numerical_investigations_on_effect_of_wear-ring_cl+(1)
Abstract
A typical submersible well pump was investigated in this article. The whole flow field of submersible well pump was
numerically simulated by computational fluid dynamics software. The influence of clearance of wear-rings on the external
characteristic and internal flow field was analyzed through comparing the calculation results with experimental results.
The result of the numerical simulation shows that changing clearance of front wear-ring has a greater impact on pump
performances than changing clearance of back wear-ring, and the head and efficiency of pump decrease with the increase
in the size of clearance. Especially when the size of clearance is larger than 0.5 mm, decreasing becomes more obvious.
When the front and back wear-ring size of the clearance comes to 1.0 mm, the efficiency decreases from the highest
point of 75.31% to 65.44% at rated flow, and the head of pump decreases about 3.5 m. When the size of clearance is
0.2 mm, reverse-flow will appear in the front shroud cavity of the impeller, and leakage from back wear-ring through the
balance hole into the impeller, which has a little influence on the flow field of the impeller inlet.
Keywords
Submersible well pump, wear-ring of impeller, clearance, numerical simulation, leakage
Creative Commons CC-BY: This article is distributed under the terms of the Creative Commons Attribution 4.0 License
(http://www.creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of the work without
further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/
open-access-at-sage).
2 Advances in Mechanical Engineering
to obtain the flow characteristics of the balance hole This article presents a simulation based on a compu-
area with the experimental method.7–9 tational domain including front and back wear-ring
Baskharone10 used computational fluid dynamics gaps, inter-stage clearance, and gains the simulation
(CFD) methods to simulate the leakage flow and study results by changing front wear-ring gap or back wear-
its influence on the internal clearance leakage ring gap, compares and analyzes the numerical
flow of multistage pump, and compared with the exist- simulation results with the test results, and explores the
ing impeller leakage model. A research method by influence of deference size of wear-ring gap on charac-
Li et al.11 to carry out the hydraulic performance teristics and inner flow field.
through changing the size of wear-ring clearance by
test, and the effect of different wear-ring gaps on the
performance through changing the viscosity of oil, was Physical model and numerical method
studied. A labyrinth wear-ring was used for numerical Physical model
simulation by Hirano et al.,12 which analyzed the influ-
The type of pump model in this article is 200QJ80-22
ence of the wear-ring clearance leakage and flow on
with two stages as shown in Figure 1(a), for which the
force of impeller.
rated flow is 80 m3/h, rated head is 22 m, rated speed is
Numerical model
In this article, the three-dimensional (3D) modeling of Figure 2. Test ring of submersible well pump.
the inlet section, the impeller, the space guide vane, the
front and back sealing rings, the clearance, and the out- Figure 7. A turbine flowmeter was used to measure the
let section is carried out through the Pro/E software flow Q and the precision of turbine flowmeter is
platform. The whole mesh generation process was car- 6 0.3%. Speed n is measured by a tachometer
ried out in the ICEM 13.0 software, and the two-stage (PROVA RM-1500, Taiwan). During the experiment,
whole flow field was meshed with structural grid. The only one dynamic pressure transmitter (CYG1401) was
sketch of the impeller-structured mesh is shown in used to measure the outlet pressure. The precision of
Figure 1(b). In this study, the grid sizes are less than CYG1401 is 6 0.2%.
2.0 or the grid numbers are large than 3.12 3 106. The overall measurement uncertainty is calculated
Figure 1 gives a general view of the model pump 2D by the square root of the sum of the squares of the sys-
assembly drawing and the mesh in the impeller.13,14 tematic and random uncertainties,16 and the calculated
The rotating part and the stationary part with the result of expanded uncertainty of efficiency is 0.5%.
multi-reference frame model in FLUENT software In the pump performance test, pump efficiency was
were based on CFD platform. The channel flow field normally defined as follows
was set as 3D incompressible steady viscous flow, using
standard k 2 e two-equation turbulence model. rgQH
SIMPLE algorithm was used for the pressure and velo- h= 3 100% ð1Þ
PS
city coupling. Turbulence flow near the wall is treated
as the standard wall condition. The velocity at the inlet where h is the pump efficiency, Q is the rated flow
of the first impeller is set to be irrotational, the center (m3/s), H is the head of the pump (m), and Ps is the
section of the inlet pressure is set as the reference point output power of the motor (W)
of pressure, and the relative pressure is 0; the outflow is
set as free flow. It is assumed that the solid wall is Pout Pin
H= ð2Þ
no-slip, which means the velocity in different directions rg
on the wall is 0. The convergence accuracy is 1025, and
the rotational speed is 2850 r/min.14 where Pin is the inlet total pressure and Pout is the outlet
total pressure (the Pin and Pout unit is Pa).
The test results were compared with the results of
Test results numerical simulation. The gap values used in simula-
The experimental study of the model is able to not only tion are in agreement with the test pump. The results
verify the reliability of CFD numerical calculation but are shown in Figure 3.
also analyze the difference in the performance of differ- Figure 3 showed the comparison between the numer-
ent clearances. So, the prototype model was manufac- ical results and the test results. As can be seen from
tured and all the sizes of clearance are 0.50 mm, and Figure 3, the numerical results are consistent with the
tested by test rig shown in Figure 2. Experiments were changing trends of experimental data. The numerical
done in an open-type pump system, which have the results are higher than the test results, and the main
identification from Jiangsu Province of China. The test result is that the numerical calculation does not take
rig is composed of two parts, namely, the data acquisi- into account the mechanical loss of the bearing and the
tion system and the water circulation system.15 The volume loss caused by seal gap leakage, making the
data acquisition system changes all kinds of physical numerical value of the head and efficiency higher than
quantities at different conditions, while the water circu- the test value. In general, the comparison of numerical
lation system supplies the necessary environment for results with the experimental results shows the accuracy
submersible well pump. The test rig is shown in and feasibility of the numerical calculation.
4 Advances in Mechanical Engineering
Figure 4. Relative changes in (a) head and (b) efficiencies in different clearances at rated flow.
Shi et al. 5
Figure 5. Comparison of performance characteristic under different clearances: (a) pump efficiency and (b) head.
Figure 6. 2D streamline distribution in area of front wear-ring of first stage at rated flow.
Flow field flow did not have a greater impact on the impeller inlet,
and the second impeller front wear-ring clearance area
Front wear-ring clearance area flow field is also the case. With the increase in gap sizes, the resis-
Changes in the wear-ring clearance will affect the inter- tance by the wear-ring clearance to flow from the front
nal flow field of the pump. Therefore, in order to impeller shroud cavity into the impeller inlet is reduced.
explore the internal flow field by changing wear-ring The reverse-flow in the front impeller shroud cavity gra-
clearance sizes, take the flow field of the impeller as the dually disappears. In the front impeller shroud cavity
research object in this article and take the front wear- was basically no reverse-flow when clearance value up
ring flow field area to analyze the design condition. to 0.7 mm, but the more flow through the front wear-
The streamlined distribution of the first and second ring clearance back to the impeller inlet, instead, the
wear-ring clearance regions is, respectively, shown in large leaks have a great impact on the impeller inlet so
Figures 6 and 7; it can be seen clearly from the chart that the hydraulic loss and volume loss is more serious.
that the flow in the front impeller shroud cavity gener- Larger clearance leakage made the flow field of impeller
ates a rotary motion when the clearance is 0.2 mm. disorderly, and the hydraulic efficiency and the volume
Impeller exit near the front impeller shroud cavity area efficiency were decreased, so as to decrease the effi-
also has a reverse-flow. Due to the outflow from the ciency and the head of the pump.
impeller of high-pressure water flow into the pump cav- Figures 8 and 9 show the static pressure contour dis-
ity, the front wear-ring clearance is smaller, and the tribution in the area of front wear-ring of first stage
leakage of liquid is greatly hindered. So, the front and second stage at rated flow, respectively. From the
impeller shroud cavity is full of reflux, but the loss has figure, we can clearly observe the pressure changes in
no effect on the main flow of the pump. Smaller leakage the clearance areas with clearance sizes increasing. The
6 Advances in Mechanical Engineering
Figure 7. 2D streamline distribution in the area of front wear-ring of second stage at rated flow.
Figure 8. Static pressure distribution in the area of front wear-ring of first stage at rated flow.
Figure 9. Static pressure distribution in area of front wear-ring of second stage at rated flow.
greater the clearance size, the more obvious the pres- outlet area of clearance, high pressure is spreading
sure gradient in the clearance. The low pressure zone in out as the clearance sizes increasing, in this case, not
the guide vane cavity outward diffusion, and in the only the volume loss is serious, but also affects the
Shi et al. 7
Figure 10. Comparison of balance hole area static pressure under different back wear-ring clearances at rated flow: (a) first stage
and (b) second stage.
pressure uniform distribution of the flow field inside the leakage through the balance hole has less impact on
the impeller. In the range of machining, smaller clear- the flow field of the impeller inlet and has few hydraulic
ance sizes are better, because the smaller clearance sizes losses. The leakage increased with the increasing sizes
can reduce the hydraulic loss and volume loss, so as to of wear-ring clearance; the flow through the balance
improve the performance of the pump. hole has a greater impact on flow field of the impeller
back shroud, and the flow field inside the impeller is
Balance hole area flow field disordered and has great hydraulic loss. Larger leakage
volume also led to increased volume loss, and the effi-
In order to balance the axial force of the impeller, the ciency and head also decreased significantly. The results
impeller back shroud has balance holes generally. And shown in Figure 10 are consistent with the results
the leakage from the balance hole will cause the flow shown in Figure 4. Therefore, it is very important to
field of impeller inlet to disorder and increase the strictly control the sizes of the clearance for the inner
hydraulic loss. Figure 10 shows the comparative analy- flow field structure and the optimum design of hydrau-
sis of static pressure contour in the balance hole area, lic performance.
where the clearance values of both impeller front
shroud and inter-stage clearance are 0.2 mm, and the
Conclusion
clearance values of back shroud are 0.20, 0.35, 0.50,
0.70, and 1.00 mm, respectively. The numerical results are consistent with the experi-
It can be seen from Figure 10 that when the size mental results. The numerical calculation does not take
clearance of impeller back shroud is 0.2 and 0.35 mm, into account the mechanical loss and the volume loss,
8 Advances in Mechanical Engineering
so the test value is slightly lower than the numerical centrifugal pump with different diffusers. J Fluid Eng: T
value. ASME 2012; 134: 071102.
The wear-ring clearance sizes of the impeller are 3. Gao B, Wang Z, Yang L, et al. Effect of wear-ring clear-
taken as 0.20, 0.35, 0.50, 0.70, and 1.00 mm, respec- ance on performance and flow characteristics of centrifu-
tively, and the numerical results of whole pump under gal pump. J Drain Irrig Mach Eng 2017; 35: 13–17.
4. Guan XF. Modern pumps theory and design, vol. 4. Beij-
different clearances were obtained. With the increase in
ing, China: China Astronautic Publishing House, 2011.
the wear-ring clearance of impeller back shroud, the
5. Kang C, Yang X and Yu XJ. Inter-stage interaction and
leakage increases; the flow through the balancing hole unsteady flow characteristics in multi stage centrifugal
has a greater impact on the impeller back shroud and pump. J Drain Irrig Mach Eng 2015; 33: 566–571.
has great hydraulic loss, which seriously disorders the 6. Kurokawa J, Matsumoto K, Matsui J, et al. Perfor-
flow field inside the impeller. mance improvement and peculiar behavior of disk fric-
The reverse-flow is formed when the front wear-ring tion and leakage in very low specific speed pumps. In:
clearance is small. In the export of clearance, the speed Proceedings of the 20th IAHR international symposium on
increases as the clearance increases. With the increase in hydraulic machinery and systems, Charlotte, NC, 6–9
the clearance sizes, the reverse-flow in the impeller front August 2000. Nihon Kikai Gakkai Ronbunshu, B Hen/
shroud cavity gradually disappears, while the leakage Transactions of the Japan Society of Mechanical Engi-
of wear-ring clearance increases; it has a great effect on neers Part B.
flow field of the impeller inlet. 7. Xiao LQ, Li YB, Liu Y, et al. Effect of wear-ring clear-
ance on cavitation flow and performance of centrifugal
pump with inducer. J Drain Irrig Mach Eng 2016; 34:
Declaration of conflicting interests 657–664.
The author(s) declared no potential conflicts of interest with 8. Gulich JF. Disk friction losses of closed turbo machine
respect to the research, authorship, and/or publication of this impellers. Forsch Ing 2003; 68: 87–95.
article. 9. Metwally IA and Gastli A. Correlation between eddy
currents and corrosion in electric submersible pump sys-
tems. Int J Therm Sci 2008; 47:800–810.
Funding 10. Baskharone EJ. Primary/leakage flow interaction in a
The author(s) disclosed receipt of the following financial sup- pump stage. J Fluid Eng: T ASME 1999; 121: 133–138.
port for the research, authorship, and/or publication of this 11. Li WG, Fei ZT and Cai YX. Effect of clearance of wear-
article: This project was supported by the Priority Academic ring on the centrifugal oil pump. Pump Tech 2004; 5:
Program Development of Jiangsu Higher Education 7–13.
Institutions (PAPD), Industry Academia Research and 12. Hirano T, Guo Z and Gordonkirk R. Application of
Prospective Joint Research Project of Jiangsu Province computational fluid dynamics analysis for rotating
(2015064-08), supported by the Open Research Subject of machinery—part II: labyrinth seal analysis. J Eng Gas
Key Laboratory of Fluid and Power Machinery (Xihua Turbine Power 2005; 127: 820–826.
University), Ministry of Education (szjj2016-069), and the 13. Fluent, Inc. Fluent user’s guide. New York: Fluent, Inc.,
Graduate Innovation Project of Jiangsu Province in China 2003.
(CXZZ13_0677) 14. Wang FJ. Computational fluid dynamics analysis: CFD
principles and application. Beijing, China: Tsinghua Uni-
versity Press, 2004.
References 15. Helmut B, Helmut J, Hoi Y, et al. Numerical simulation
1. Stel H, Sirino T and Ponce FJ. Numerical investigation of low specific speed American Petroleum Institute pumps
of the flow in a multistage electric submersible pump. J in part-load operation and comparison with test rig
Petrol Sci Eng 2015; 136: 41–54. results. J Fluid Eng: T ASME 2012; 134: 024501 (9 pp.).
2. Zhou L, Shi WD, Lu WG, et al. Numerical investiga- 16. ISO 9906:1999. Rotodynamic pumps—hydraulic perfor-
tions and performance experiments of a deep-well mance acceptance tests: grades 1 and 2.