Available online at www.sciencedirect.

com

ScienceDirect
Procedia Engineering 116 (2015) 97 – 102

8th International Conference on Asian and Pacific Coasts (APAC 2015)

Department of Ocean Engineering, IIT Madras, India.

Wave Energy Harvesting Turbine: Performance Enhancement

Paresh Halder, Abdus Samad*

Wave Energy and Fluids Engineering laboratory, Department of Ocean Engineering, IIT Madras, Chennai, Pin-600036, India

Abstract

The performance of an oscillating water column (OWC) wave energy system depends upon several factors
including its turbine performance. OWC turbines have inherently low efficiency and low operating range. Through
this work, the shape of a Wells turbine was tried to modify to enhance the efficiency and enlarge the operating
range. A numerical investigation with steady state flow condition has been reported to change blade thickness from
hub to tip. A new concept to combine NACA0015 and NACA0024 at the hub, the midspan and the tip section
were developed and the blade was produced through smooth polynomial so that the blade will have different
thickness along the span. A commercial code ANSYS-CFX® v14.5 was used for the simulations. The turbulence k-
ω SST model was adopted and the reference turbine performances were compared with existing results reported in
the literatures. It was found that a blade thinner at midspan gives lesser flow separation, which helps increasing the
turbine performance.

© 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
© 2014 The Authors. Published by Elsevier B.V.
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review
Peer- underresponsibility
Review under responsibility of organizing
of organizing committee
committee of APAC
, IIT Madras 2015, Department
, and International of Ocean of
Steering Committee Engineering,
APAC 2015 IIT Madras.

Keywords: Wells Turbine, Axial flow turbine, Wave energy conversion

Introduction

Wells turbine, which is an axial flow self-rectifying low-pressure air turbine, rotates continuously in one direction
by the reversing flow or bi-directional flow. The turbine is used in an oscillating water column (OWC) wave

* Corresponding author. Tel.: +0-091-44-2257 4826; fax: +0-000-000-0000 .

E-mail address:samad@iitm.ac.in

1877-7058 © 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer- Review under responsibility of organizing committee , IIT Madras , and International Steering Committee of APAC 2015
doi:10.1016/j.proeng.2015.08.269
98 Paresh Halder and Abdus Samad / Procedia Engineering 116 (2015) 97 – 102

energy system. The system performance greatly depends upon the turbine’s power production capability under
different wave conditions. The wave condition varies in each wave cycle as well as varies with weather condition.
The turbine blades are constructed using symmetric airfoil and the angle of attack on the airfoil is 90o. Curran and
Gato (1997) compared single-plane and multi-plane wells turbines. Torresi et al. (2008) reported an accurate
description of the steady three-dimensional flow-field in a high solidity Wells turbine. Taha and Sawada (2010)
compared computational and experimental results with tip clearance to chord length ratios of 0.0056 and 0.0111. It
was shown clearly that the tip clearance (TC) significantly influences the turbine performance. Halder and Samad
(2014, 2015) studied the effect of uniform tip gap and casing treatment to delay flow separation and to increase
turbine performance.
Several authors reported effect of blade-profile shape modification to improve overall efficiency of the Wells
turbine numerically or experimentally (Table 1). Takao and Okuhara (2013) reported that increasing blade
thickness hub to tip enhanced the turbine performance and improving the stall characteristics. However, no
information was found in the literatures on different profiles to delay stall.
In the present work, a numerical study for a new concept to vary blade profile thickness from hub to tip of a
Wells turbine has been reported. The results were compared with experimental and numerical results. Different
profile thickness for wide flow coefficients were simulated and details analysis has been reported.
Table 1: Blade-profile modifications to enhance Wells turbines performance.
Design modification Advantage Description Profile
o
Mohamed et al., Increased power output (average Incident angle varied: 5 to 14 NACA0021
(2011) relative gain: +11.3% Improved
efficiency: 1%
Raghunathan and The NACA0021 produced the peak Thicker and modified aerofoil blades NACA0024,NACA0021,
Tan, (1985) efficiency. Efficiency drop: ~10% with improved performance of the NACA0015H,NACA0015,
blade roughened blade. turbine. NACA0012
Suzuki and Arakawa, Efficiency improved at an angle of fan-shaped blades with different NACA0021, NACA0012
(2008) attack <7o. Stall angle= 10º was sweep angles
smaller.
Takao et al., (2006) Peak efficiency higher Optimum blade profile: NACA0015 NACA0015,NACA0020CA9,H
SIM 15-262133-1576
Thakker and Higher power output Preferable rotor blade profile CA9 NACA0015,NACA0020,CA9,
Abdulhadi, (2007) HSIM 15-262133-1576
Takao and Okuhara, Improved efficiency and stall The blade thickness increases NACA0015,
(2013) characteristics gradually from hub to tip NACA0020,NACA0025

Nomenclature

h Hub-to-tip ratio TC Tip clearance

LE Leading edge T Shaft torque
N Speed of rotor, rpm t Blade thickness
PS Pressure surface Utip Rotor velocity
Q Volume flow rate V Axial velocity
RANSE Reynolds averaged Navier-Stoke equations z Number of rotor blade
RB Rotor blade ω Rotational speed
Rmid Mid Span radius ρ Density
Rtip Tip radius η Efficiency
s Turbine solidity * Non-dimensional parameter
∆P* Pressure drop coefficient
∆P* Stagnation pressure drop
T* Torque Coefficient
 Paresh Halder and Abdus Samad / Procedia Engineering 116 (2015) 97 – 102 99

Computational Methodology

As stated earlier, the Wells turbine profile was varied from hub-to-tip. Fig. 1 shows the variation of profile
thickness using NACA0015 and NACA0024. The left most cases is the original blade (NACA0015) and it has
same profile all along the span. Similarly, case A has NACA0024 profile. The other profiles (cases B, C, D and E)
are mixed profiles from hub-to-tip, which makes the blade thinker or thicker at different sections. The profiles in
between the hub and the midspan, and midspan and tip were produced through a third order polynomial curve.
The turbine performance was numerically investigated by solving the 3D incompressible Reynolds Averaged
Navier-Stokes equations (RANSE) and k-ω SST turbulence model. The RANSE were discretised by a finite
volume method. The governing equations used for the simulations were continuity, momentum and energy. A finite
volume based solver ANSYS CFX® v14.5 was used to solve the equations. High resolution and first order
numerical schemes were implemented. The solutions were considered converged once the maximum residual
values were reached to a value of 10E-5.
Table 2 shows the specification of the turbine, which rotates at a constant velocity (=2000rpm). A single blade
passage with periodic boundary conditions was used for the computations. Table 3 shows meshing and boundary
conditions. The computational domain was divided into four and six times of the blade chord length along the
upstream and downstream of the rotor, respectively (Fig. 2).

Tip Tip Tip

Mid Mid Mid

Hub
Hub Hub

Original blade A B C D E

Fig. 1. Variation of blade profile thickness from hub-to-tip

Table 2. Specifications of rotor.

Parameter Dimension
Blade profile NACA0015,NACA0024
Blade chord length, C 0.125m
Blade number, z 8
Blade thickness, t 15% of C
Solidity 0.644m
Tip radius 0.3m
Hub radius 0.2m
Hub/Tip ratio 0.67m
Mean radius 0.25m
Tip clerance of the chord length 1mm

Result and Discussion

Present numerical results were validated with an existing results available in the literatures (Curran and Gato,
1997; Torresi et al., 2008) (Fig. 3) and match well with the results of original blade (Fig. 3). Fig. 3(a) shows the
torque coefficient (T*) variation with different blade profiles from hub-to-tip. The torque coefficient is highest for
100 Paresh Halder and Abdus Samad / Procedia Engineering 116 (2015) 97 – 102

a blade having midspan-thickness lower (NACA0015) while the tip and hub sections were wider (NACA0024).
Fig. 3(b) shows that the peak efficiency is almost constant for all the cases.
Table 3. Meshing and boundary conditions.
Parameter Description
Flow domain Single blade
Interface Periodic
Mesh/Nature Unstructured
Fluid Air
Turbulence model k-Ω SST
Inlet Velocity
Outlet Pressure
Hub Wall
Casing Wall
Blade Wall
Residual RMS criteria 1x10-5
Mass imbalance 0.001

Fig. 2. Computational domain

a 0.3 Present result b

A
B
0.6
C
D
E
0.2 Exp, Curran and Gato (1997)
CFD, Torresi et. al(2004) 0.4
T*

Present result
η

A
B
0.1
0.2 C
D
E
Exp, Curran and Gato (1997)
CFD, Torresi et. al(2004)
0.0 0.0
0.0 0.1 0.2 0.3 0.4 0.0 0.1 0.2 0.3 0.4
U* U*
Fig. 3. (a) Torque coefficient; (b) Efficiency

Fig. 4 shows the static pressure distribution and blade-loading curve at the midspan of the blade passage. At
lower flow coefficients, the flow is attached and there is not much variation in pressure on the blade surfaces. The
airfoil profiles are taken at Midspan. At U*=0.275, the blade E is showing higher pressure on suction surface and
hence this gives a delayed separation. The separated flow can have lower power transfer to the blade and hence has
 Paresh Halder and Abdus Samad / Procedia Engineering 116 (2015) 97 – 102 101

lower torque (Fig 3(a)). It can also be found that the torque is higher at higher flow coefficient and hence it gives a
wider operating range. This observation is consistent with an existing study (Mohamed et al., 2011).

Flow Flow
a b c 2

Cp
-1 O
α=4.29
NACA0015
-2
E

α=4.29 (U*=0.075) -2 -1 0 1 2
X/C
4

Cp
-4
O
α=12.68
-8 NACA0015
E
-12
α=12.68 (U*=0.225) -2 -1 0 1 2
X/C

-4
Cp

-8 O
α=15.38
-12 NACA0015
E
-16
α=15.38 (U*=0.275) -2 -1 0 1 2
X/C

Fig. 4. Static pressure distribution: a) Original blade (NACA0015), b) Modified blade (E) and (c) blade loading with original and Modified
blade (E)

Fig. 5 shows the streamline pattern at the midspan of the blade passage. At lower flow coefficient (U*=0.075),
there is no circulation at the trailing edge and the streamlines are attached to the blade surfaces. As the flow
coefficient increases, flow separation occurs at the trailing edge. With higher flow coefficient (U*=0.275), the flow
circulation is found at the trailing edge. However, the size of the circulation is much larger in case of standard
aerofoil (NACA0015). As a result, the torque is lower in the original aerofoil (Fig. 3(b)). It was already reported
that increasing blade thickness enhances turbine performance and improves the stall characteristics (Takao and
Okuhara, 2013).
102 Paresh Halder and Abdus Samad / Procedia Engineering 116 (2015) 97 – 102

a b c

NACA0015

Fig. 5. Streamline line on the midspan of the flow passage: (a) U*=0.075; (b) U*=0.225; (c) U*=0.275;

Conclusion

A novel concept to vary blade thickness from hub-to-tip of a Wells turbine blade was reported and the blade
performance has been evaluated. Reynolds averaged Navier-Stoke equations were solved with k-w SST model. The
numerical results were validated with the existing experimental results.
It was found that a lower thickness at the midsection produces larger torque and hence larger power. The reason
behind the performance enhancement was the delayed flow separation and increased blade loading.
It was found that the peak efficiency was not changed much because of blade shape change. An increase
operating range of the turbine was also found through this investigation.
By further shape modification with larger number of parameters, it can produce further power enhancement.

References

Curran, R., Gato, L.M.C., 2005. The energy conversion performance of several types of Wells turbine designs. Proceedings of the Institution of
Mechanical Engineers, Part A: Journal of Power and Energy 211, 133–145. doi:10.1243/0957650971537051
Halder, P., Samad, A., 2014. Tip Leakage Effect on Wells Turbine, in: 1st International Conference on Renewable Energies Offshore. Lisbon,
Portugal.
Halder, P., Samad, A., 2015. Casing Treatment of a Wave Energy Extracting Turbine, in: International Conference on Water Resources, Coastal
and Ocean Engineering. NIT Surathkal, Karnataka, India.
Mohamed, M.H., Janiga, G., Pap, E., Thévenin, D., 2011. Multi-objective optimization of the airfoil shape of Wells turbine used for wave
energy conversion. Energy 36, 438–446. doi:10.1016/j.energy.2010.10.021
Raghunathan, S., Tan, C.P., Ombaka, O.O., 1985. Performance of the Wells Self-Rectifying Air Turbine. Aeronautical Journal 89, 369–379.
doi:10.1016/0142-727X(85)90026-8
Suzuki, M., Arakawa, C., 2008. Influence of Blade Profiles on Flow around Wells Turbine. International Journal of Fluid Machinery and
Systems 1, 148–154. doi:10.5293/IJFMS.2008.1.1.148
Taha, Z., Sugiyono, Sawada, T., 2010. A comparison of computational and experimental results of Wells turbine performance for wave energy
conversion. Applied Ocean Research 32, 83–90. doi:10.1016/j.apor.2010.04.002
Takao, M., Okuhara, S., 2013. Wells Turbine for Wave Energy Conversion - Improvement of Stall Characteristics By the Use of 3-Dimensional
Blades - 9, 1–5. doi:10.1299/jfst.2014jfst00
Takao, M., Thakker, A., Abdulhadi, R., Setoguchi, T., 2006. Effect of blade profile on the performance of a large-scale Wells turbine for wave-
energy conversion. International Journal of Sustainable Energy 25, 53–61. doi:10.1080/14786450600593295
Thakker, A., Abdulhadi, R., 2007. Effect of Blade Profile on the Performance of Wells Turbine under Unidirectional Sinusoidal and Real Sea
Flow Conditions 2007. doi:10.1155/2007/51598
Torresi, M., Camporeale, S.M., Strippoli, P.D., Pascazio, G., 2008. Accurate numerical simulation of a high solidity Wells turbine. Renewable
Energy 33, 735–747. doi:10.1016/j.renene.2007.04.006