Numerical investigation on the effectiveness

of water injection method for mitigating

propeller cavitation
Cite as: AIP Conference Proceedings 2121, 060005 (2019); https://doi.org/10.1063/1.5115906
Published Online: 18 July 2019

Nasimul Eshan Chowdhury, Abdul Karim Miah, and Arafat A. Bhuiyan

ARTICLES YOU MAY BE INTERESTED IN

Improvement of thermal-hydraulic performance of compact heat exchangers with multi-

corrugated fin and oval tube arrays
AIP Conference Proceedings 2121, 030016 (2019); https://doi.org/10.1063/1.5115861

Experimental investigation on thermal treatment of local and marine based micro-algae for
production of biodiesel
AIP Conference Proceedings 2121, 120006 (2019); https://doi.org/10.1063/1.5115943

The effect of leading-edge serration at owl wing feathers on flow induced noise generation
AIP Conference Proceedings 2121, 060001 (2019); https://doi.org/10.1063/1.5115902

AIP Conference Proceedings 2121, 060005 (2019); https://doi.org/10.1063/1.5115906 2121, 060005

© 2019 Author(s).
 Numerical Investigation on the Effectiveness of Water
Injection Method for Mitigating Propeller Cavitation
Nasimul Eshan Chowdhury1,a), Abdul Karim Miah1, b) and Arafat A. Bhuiyan1, c)
1
Department of Mechanical and Chemical Engineering, Islamic University of Technology (IUT), Gazipur-1704,
Dhaka, Bangladesh.

a)
[email protected]
b)
[email protected]
c)
[email protected]

Abstract.Cavitation is a common problem for marine propellers. Introducing water to the cavitation prone area to
increase the operating pressure is one of the approaches to mitigate cavitation. The principle goal of this study is to
investigate the effectiveness of this method, depending on the positions of the injection source. The k-ω SST turbulence
model with the curvature correction and the Zwart cavitation model have been applied on the commercial CFD Code
ANSYS to investigate the issue. This combined model has been validated against experimental data and a reasonable
agreement has been observed. Finally, the final model has been computed on a modified INSEAN E779a propeller. The
method has been found less effective for water, as a working fluid.

INTRODUCTION
There are different types of cavitation, occurring in a marine propeller [1] and a number of methods to reduce the
cavitation formation for the marine propellers [2, 3]. Introducing mass to the cavitation prone areas of the propeller
is one of the approaches [4]. In this study two positions for the injection have been numerically compared. The k-ω
SST turbulence model and the Zwart cavitation model have been used for the simulation. To validate the combined
model, a dummy setup has been simulated and compared with the experimental data. Getting plausible results, the
model has been applied to the final setup.
The low pressure region takes a triangular shape, having the wider portion near the tips. It needs more tangential
velocity for the particle to travel near the tips. For this reason, the closer the position of the source to the tips, the
easier it is for the injected mass to reach the low pressure region. However, there is lacking for real life experiment
on the issue. Consequently, the study urges to undertake more researches for confirmation.

COMPUTATION RESOURCE
The simulation was done on a Core-i5, 7th generation Desktop, having 8GB RAM. The computation time was
around 5 days, for 7 seconds of simulation time span. The Fluent software from the Ansys software package has
been used for the computation. The mass and momentum conservation are solved, to obtain the velocity and
pressure fields. The numerical model applies implicit finite volume scheme [5].

8th BSME International Conference on Thermal Engineering

AIP Conf. Proc. 2121, 060005-1–060005-6; https://doi.org/10.1063/1.5115906
Published by AIP Publishing. 978-0-7354-1861-5/$30.00

060005-1
 NUMERICAL MODELS

Turbulence model
Turbulence model: As shear stress transport (SST) k–ω turbulence model is good in predicting the values of the
parameters, under adverse pressure gradients and separating flow [6]; many computations have used this model [7].
Thus, this model has been selected for the purpose. The model can be expressed in the form of equation 1.
k
( k )
(  k ) (  kui ) x j
   G k  Yk  Sk
t xi x j
(1)
Here, Gk is the generation of turbulence kinetic energy due to mean velocity gradients,  k is the effective

diffusivity of k , Yk is the dissipation of k , due to turbulence, S k is user-defined source term.

Cavitation model
There is number of models to capture the cavitation phenomenon. The most common models are Singhal
cavitation model, Schnerr and Sauer model and Kunz model [7]. Zwart-Gerber-Belamri model has been used in this
study, which can be applied in both mixture and Eulerian multiphase models [8].Zwart model assumes all the
bubbles in a system, having the same size and proposes the total interphase mass transfer rate per unit volume (R) is
calculated using the bubble density numbers (n) [8]. The mass change rate of a single bubble can be written as the
equation 2.
d B
R  n  (42B  )
dt
(2)
Here,  B = Bubbles Radius,  = Vapor Density
The key assumption is that the cavitation bubble does not interact with each other, which is plausible only during
the earliest stage of cavitation, during the formation of the nucleation site of the cavitation bubbles. Eventually, the
final form of this cavitation models can be expressed as the equation 3 and equation 4.
If, P  P
3 nuc (1   )  2 P  P
Re  Fvap
B 3 l (3)
If, P  P
3  2 P  P
Re  Fcond
B 3 l (4)
Here,  nuc = nucleation site volume fraction = 5×10 , -4
Fvap = evaporation coefficient = 50, Fcond =
condensation coefficient = 0.001,  B = 10-6 m, P = the local far-field pressure, P = saturation vapor pressure.

060005-2
 NUMERICAL MODEL VALIDATION

Dummy Model Setup

The physical model has been validated against an experiment of formation of cavitation for cone shaped
specimen, tested in a water tunnel. Figure 1(left) shows the grid view of the setup, consisting of 47357 elements and
Fig. 1 (right) the accuracy of the dummy simulation.

FIGURE 1. Mesh distribution of the dummy setup and the result

Final Numerical Model Setup

The initial pressure difference for both inlet and outlet have been set to zero, the outer curved surface has been
set as pressure outlet, the operating pressure is 101325 Pa and the other parameters have been kept as default. It has
been assumed that the buoyancy of bodies in water is negligible, no slip condition prevails on the propeller surface
and operating fluid is incompressible. The sliding mesh method has been selected for the rotating mesh of the
rotor.After obtaining stability, the time step was set to 10 -4 sec. Initially the setup was run without water injection, to
get a fully developed state. During this period the ports were signed as wall. The rotation of the propeller has been
increased step-by-step and finally set to 3000rpm. Afterwards, to capture the water injection phenomenon, the
boundary condition for the outer ports was changed to pressure inlet.

Geometry Profile
The specimen is a modified design of INSEAN E779a propeller. The important dimensions are shown in Table
1. Using the Solid works designing software eight circles have been drawn on the blades of the propeller.
TABLE 1.Propeller dimensions
Angle of attack (at the tip) tan-1(250/714) or, 19.2972rad
number of blades 4
skew angle 00
rake angle 4.560
pitch ration 250 mm
P/D 1.1

Grid Properties
The setup is divided into one stator, one rotor. The rotor is a disk with a cavity of profile of the propeller
geometry, as shown in Fig. 2.

060005-3
 FIGURE 2. Grid view of the final setup

170
Pressure (KPa)

130 1.5mm mesh size

2mm mesh size
90

50
38 43 48 53 58 63 68 73
Location on X direction (mm)

FIGURE 3.Mesh distribution for different grid sizes.

Two computations have been conducted for mesh size of 2mm and 1.5mm, on the surface of the propeller, with
automatic meshing method, as shown in Fig 3 (left). Getting satisfying result at the moment of .0012sec, as shown in
Fig. 3 (right), 1.5mm mesh size has been selected for the simulation, altogether having about 1+M number of
meshes.

RESULT AND DISCUSSION

The analysis of the water injection for the inner ports revealed that for the high rotation of the propeller the
injected water does not get enough time to spread over the blade, as the streamlines in Fig. 4 (left) showed.
However, the streamlines follow a spiral path while crossing the blade, as they deviates for, the imaginary red
circular path, in Fig. 4 (left). Similar diffusion problem has been found for the outer port water injection approach.
Moreover, like the inner port injection, for the outer port injection, the injected water follows a spiral path, as shown
in Fig. 4 (right). However, as mostly the cavitation occurs at the tips, indicating by the foggy patterns in Fig. 5and
there is very less influence of centrifugal force on the injected water, for the effective positioning of injection, the
ports should be as close to the tips as possible.

060005-4
 FIGURE 4.Water injection through the inner ports (left) and the outer ports (right)

FIGURE 5. The cavitation area, due to high rotation

On the other hand, it was found that the injected water quickly turns to vapor, as it enters the low pressure
region. The data in Fig. 6, has been taken along a straight line, axially passing through one of the outer ports, which
indicates the transformation of injected water into vapor due to sudden pressure drop. However, FIGURE 6 6 also
show that with the increase of inlet pressure, the size of the liquid phase region increases, resulting less cavitation.
This proves that by increasing the inlet pressure significantly, the mass injection approach decreases the cavitation
formation to some extent.
1.1

1
VF% of water

0.9

0.8
1300000 Pa
0.7
13000 Pa
0.6
50 100 150 200 250 300
Position (x10-5 m)

FIGURE 6. Comparison of different inlet pressures

However, reading for the same parameter along a circle, tangentially crossing the outer ports showed that for the
reasonable inlet pressure of 130,000 Pa, water injection has negligible influence in decreasing the cavitation.

060005-5
 1.2
Water injection No injection
1.0
Vf% of vapour
0.8
0.6
0.4
0.2
0.0
-0.2
0 50 100 150 200 250 300 350
Angural Position (Degree)

FIGURE 7. Comparison of different injection methods

CONCLUSIONS
The combined model has been found very robust to compute complex phenomena and very useful for cavitation
related studies. The water injection approach was found less effective, as the water turns into vapor by cavitation
and cannot increase the pressure in the low pressure region, even for the perfect port location. However, injecting
expandable substances might solve the problem. On the other hand, as the injected water turns into vapor,
decreasing the density, the centrifugal force cannot influence enough to radially defuse the injected water.
Consequently, the study states that the mitigation by water injection is not significant. However, further real life
experiment must be carried out to confirm the finding.

REFERENCES
1. HonFIIMS, E.I.J.C.-W. An introduction to propeller cavitation. 2015. https://www.iims.org.uk/introduction-
propeller-cavitation/.
2. Chekab, M.A.F., et al., Investigation of Different Methods of Noise Reduction for Submerged Marine
Propellers and Their Classification. American Journal of Mechanical Engineering, 2013. 1(2): p. 34-42.
3. Nwaoha, T.C., S. Adumene, and T.E. Boye, Modelling prevention and reduction methods of ship propeller
cavitation under uncertainty. Ships and Offshore Structures, 2017. 12(4): p. 452-460.
4. Lee, J.-H., et al., Reduction of propeller cavitation induced hull exciting pressure by a reflected wave from air-
bubble layer. Ocean Engineering, 2014. 77: p. 23-32.
5. Dular, M., et al., Experimental evaluation of numerical simulation of cavitating flows around hydrofoil.
European Journal of Mechanics-B/Fluids, 2005. 24(4): p. 522-538.
6. Hear-Stress Transport (SST) k-omega Model, 2018; https://www.sharcnet.ca/Software/Fluent6/html/ug/node487.htm.
7. Javadpour, S.M., et al., Experimental and numerical study of ventilated supercavitation around a cone
cavitator. Heat and Mass Transfer, 2017. 53(5): p. 1491-1502.
8. Cavitation Models. http://www.afs.enea.it/project/neptunius/docs/fluent/html/th/node343.htm.

060005-6