Proceedings of the 7th International and 45th National Conference on Fluid Mechanics and Fluid Power (FMFP)

December 10-12, 2018, IIT Bombay, Mumbai, India

FMFP2018– PAPER NO. 078

Effect of Number of Buckets on Flow Characteristics in Pelton Turbine

Vishal Gupta Vishnu Prasad Ruchi Khare

Department of Mechanical Professor Assistant Professor
Engineering, Department of Civil Engg, Department of Civil Engg,
Radharaman Group of Institutes M.A. National Institute of Technology, M.A. National Institute of
Ratibad, Bhopal, India Bhopal, M.P., India Technology, Bhopal, M.P., India
Email: [email protected] Email: [email protected] Email: [email protected]

Abstract based on empirical relations like Taygun [Nechleba 1957]

The number of buckets is important design parameter for may not give good performance of runner. Hence, it is
Pelton turbine. It affects water flow characteristics on needed to assess the performance of Pelton turbine runner
bucket and performance of runner. The number of with different number of buckets. In the present paper,
buckets depends on the pitch diameter of bucket, the numerical flow simulation is carried out in an existing
rotational speed of runner and jet ratio. So, optimum six jet Pelton turbine model to study the effect of number
number of buckets is essential for the best performance of of buckets on local flow parameters like velocities, flow
turbine. The decrease in number of buckets will lead to angles, flow distribution of water in bucket and also on
loss of water as jet may not strike the buckets and leave performance of Pelton turbine runner. The efficiencies
without any impact, whereas, the increase in number of obtained by numerical simulation of the existing 20
buckets will add to the excess weight, manufacturing cost buckets turbine runner are compared with the
and interference of jet. In present case, transient flow experimental values for validation and are found to be in
simulation of an existing six jet Pelton turbine model has good agreement with experimental results.
been carried out for three solidities of runner. Commercial
code ANSYS CFX has been used for the analysis. The II. GEOMETRIC MODELING AND BOUNDARY
efficiency from CFD results of existing runner solidity CONDITIONS
has been compared with the experimental model test
The existing Pelton turbine model has six jets. The jet
results at different operating regime of turbine.
diameter and pitch circle diameter are 32.4mm and
175.5mm respectively. The runner of Pelton is axi-
Keywords: Pelton turbine; solidity; multiphase flow.
symmetric and hence geometry of half of runner having
three jets is considered for simulation. The buckets are
I. INTRODUCTION also symmetrical about the splitter and therefore half jet
Pelton turbine is a low specific speed turbine which is and bucket are modeled due to limitation of
used for high head and low discharge sites. Pelton turbine computational facility. The modeling has been done in
has the series of buckets fitted to the periphery of circular ANSYS Workbench. The modeled geometry of half of
disc mounted on rotating shaft. In this turbine, all stator having 3 jets is given in Fig.1 and half of the rotor
available energy of water is converted into kinetic energy domain geometry with 10 buckets is shown in Fig.2.
or velocity head. The water is issued from nozzle in the
form of jet. The jet of water exerts torque on runner
causing it to rotate. A lot of work has been reported for
design and analysis of distributor [Perrig et al. 2006],
nozzle shape leading to change in jet shape [Matthias et
al. 2004], bucket profile [Rygg, 2013]. But very few
authors have studied for effect of number of buckets on
the performance and flow distribution of Pelton turbine
runner [Židonis et al. 2015].
Generally numbers of bucket are so chosen that
complete jet of water may be utilized in doing the work
[Peron et al. 2006]. The selection of number of buckets Fig.1: Geometry of stator domain
1
 Blade loading coefficient: The water flow in runner
subjected to change in dynamic pressure at the surface of
bucket from splitter to exit of the bucket due to change in
flow direction [Fiereder et al. 2010].
Pressure variation along surface of bucket in
dimensionless form is termed as blade loading coefficient
and expressed as:
P (VII)
Cp 
1
C12
Fig.2: Geometry of quarter of existing rotor domain (20 2
buckets) Velocity coefficient:
C
Kc  (VIII)
The complete flow domain is discretised into tetrahedral 2 g  H
and prismatic 3-D elements for simulation. The prismatic Loss coefficient:
elements are used near to bucket surface for proper W
resolution of boundary layer. K  2
L (IX)
W
1
BOUNDARY CONDITIONS Discharge coefficient:
Stator domain containing jet is kept stationary. Rotor 4Q
domain has been set different rotational speeds. The KQ  2 (X)
  B  Z  2 g  H
analysis is carried out by taking SST turbulence model
because of its ability to capture the flow with sharp
curvatures and runner rotation [Gupta et al. 2016]. III. RESULTS AND DISCUSSIONS
Transient flow simulation is chosen with time step VALIDATION
corresponding to 1° runner rotation for all the cases. Air The comparison of CFD results with experimental results
and water are taken as working fluids with a reference has been done for hydraulic efficiency only at similar
pressure of 1 atmosphere. The jet inlet has been defined operating conditions for existing runner with 20 blades for
as inlet with water velocity corresponding to 50 m head. validation [Gupta et al. 2016] (given in Table 1). The
The jet symmetry, stator symmetry and rotor symmetry rated speed and discharge of turbine model is 820 rpm
have been defined as symmetry and rotational periodicity and 122 lps respectively. The values of speed and
has been applied at both the symmetric ends of stator and discharge coefficients are 0.48 and 0.77 respectively.
Table 1: Comparison of results
rotor. The opening is specified all around stator and rotor.
The transient rotor-stator interface is specified in between Speed coefficient 0.43 0.45 0.48 0.50 0.53
stator and rotor domains.
Discharge coefficient 0.77 0.77 0.77 0.77 0.77
FORMULAE USED
Computed efficiency (%) 84.07 87.76 88.03 86.24 84.51
Theoretical power input:
PI    g  Q  H (I) Experimental efficiency (%) 90.00 91,00 91.50 91.00 90.00
Numerically computed power output:
2  nT Deviation (%) 6.59 3.56 3.79 5.23 6.10
PO  (II)
60 It is seen that deviation in efficiency at the rated operating
PO regime from both approach has been around 3.5% and
Hydraulic efficiency:  H  x100 (III) increases at over and under loading of turbine. The
PI
deviation between simulation and experimental values of
Time step corresponding to 1° runner rotation: efficiency may be because of the losses which are not
60 accounted properly in CFD.
t  (IV)
360  n Further, it is seen from Fig.3 for 20 bucket runner that
Torque coefficient: torque coefficient at rated operating regime increases
T from 0 to 0.20 upto 50° rotation. This indicates that the
KT  2 (V) water jet from nozzle reaches to runner during this
 C  B  L  D
1
rotation. After 50° rotation, the torque coefficient
Speed coefficient:
becomes nearly constant with small fluctuations. Runner
 N D experiences high value of torque when the bucket is
B  (VI)
60 2  g  H normal to the jet and its value reduces when the bucket
moves away from normal position. Hence average value
2
of torque from 50° to 100° runner rotation was considered of buckets increases except near splitter. It is maximum
for torque and power calculation. around the 0.70 of stream wise location due to sharp
change in flow direction.
EFFECT ON TORQUE
0.50
The flow simulation has been carried at rated operating

Blade loading
regime for three solidities having 18, 20 and 22 buckets. 0.40

coefficient
0.25 0.30 18
Torque coefficient

0.20 0.20 20
0.15 0.10
0.10 18 0.00
0.05 20 0.0 0.5 1.0
22 Streamwise
0.00
Fig. 5: Blade loading variation at mid span
0 50 100
EFFECT ON ABSOLUTE VELOCITY
Runner rotation (°) It is observed from Fig.6 that absolute velocity variation
at mid span for rated speed gradually decreases from inlet
Fig. 3: Runner torque variation to outlet. The decrease in absolute velocity from inlet to
outlet indicates the change of kinetic energy into
From Fig.3, it is observed that torque on runner is affected
mechanical energy [Kotousov,2005]. The difference
by number of buckets. There are large fluctuations in between inlet and outlet velocity is found to nearly
torque on runner with 18 bucket and fluctuations reduce independent of the number of buckets.
as the number of buckets is increased. It may be due to 1.20
reduction in time to reach the bucket normal to jet as the Absolute velocity
1.00
number of buckets increases. coefficient
0.80
0.60 18
EFFECT ON EFFICIENCY 20
0.40 22
The variation of efficiency with speed in Fig.4 for
different runner solidity predicts that there is parabolic 0.20
variation of efficiency with speed for all solidities 0.00
resembling efficiency variation of Pelton with speed in 0.0 0.5 1.0
experimental results. The maximum efficiency is Streamwise
achieved nearly at same speed. The efficiency is found to
Fig.6: Variation of absolute velocity at mid span
be maximum for runner with 20 number of buckets.
1.02 VARIATION IN FLOW PATTERN
1.00 18
Normalized

0.98
efficiency

22
0.96
20
0.94
0.92
0.42 0.47 0.52
Speed coefficient φ
Fig.4: Variation in normalized efficiency
There is drastic drop in efficiency with 22 buckets while
with 18 buckets, efficiency is more at lower speed and
less at higher speed as compared to 20 buckets. This may
be due to loss of water in less number of buckets and
interference of flow between buckets in case of more Fig.7: Variation in water volume fraction at mid span
buckets. It is seen that each jet is striking buckets normally after
EFFECT ON BLADE LOADING passing through cut out. The flow interruption by
The blade loading at mid span of bucket for different preceding and successive buckets increases with increase
solidities (Fig.5) at rated speed, is uneven as water moves in number of buckets.
from splitter to exit of bucket. It increases as the number
3
The water volume fraction in Fig.7 clearly indicates that The flow in axial direction decreases as the number of
less water remains on buckets as the number of buckets buckets increases while flow increases in radial inward
decreases. The buckets are not fully emptied before the direction with increase in number of buckets.
next jet. The volume of water remaining on bucket
decreases with increase in number of buckets. Nomenclature
It is evident from coefficient of different velocity B bucket width (m)
components in Table 2 that all velocities except tangential C absolute velocity of water at any point (m/s)
decreases from inlet to outlet. These velocities are same at C1 absolute velocity of water at runner inlet (m/s)
inlet but some have slight deviation at outlet as number of D runner diameter (m)
bucket changes. The maximum torque, friction loss and g acceleration due to gravity (m/s2)
efficiency are found to be for existing runner as shown in H net available heat at inlet of nozzle (m)
Table 3. The water distribution in Table 4 indicates that n rotational speed of runner (rpm)
water flow is maximum in axial direction and it decreases P static pressure at any point (Pa)
with increase in number of buckets. The flow in radially PI input power supplied from jet to runner(Watt)
inward direction increases as number of buckets increases PO power available at the shaft (Watt)
but in radially outward direction, flow is found to be Q discharge (m3/sec)
lowest for 20 buckets. The flow angles are found to be T torque available at runner (N-m)
increasing as number of buckets increases. W1 whirl velocity of flow at inlet
Table 2: Average values of velocity coefficients W2 whirl velocity of flow at outlet
Number of buckets Z number of buckets
18 20 22
Inlet Outlet Inlet Outlet Inlet Outlet
ρ density of fluid (kg/m3)
Relative velocity
0.51 0.44 0.51 0.46 0.51 0.44
coefficient REFERENCES
Absolute velocity
coefficient
1.00 0.15 1.00 0.18 1.00 0.18 [1] A. Perrig, A. François, J-L Kueny, M. Farhat, and E.
Whirl velocity Parkinson. "Flow in a Pelton turbine bucket: numerical
1.00 0.07 1.00 0.06 1.00 0.08 and experimental investigations." Journal of fluids
coefficient
Tangential engineering, Vol. 128, no. 2, pp. 350-358. (2006)
velocity 0.49 0.49 0.49 0.49 0.49 0.49 [2] H.B. Matthias & O. Promper, "Numerical simulation
coefficient
of the free surface flow in Pelton turbines"
Flow angle (° ) 0.00 17.45 0.00 21.10 0.00 23.58
Proceedings of 6th International Conference on
Hydraulic Machinery and Hydrodynamics Timisoara,
Table 3: Average values of non-dimensional parameters
Number
Romania, pp. 119-124 (2004)
KT KL KQ ϕB ηH [3] J.R. Rygg, CFD Analysis of a Pelton Turbine in
of buckets
18 0.76 0.862 0.077 0.48 87.87 OpenFOAM. Master Thesis, Trondheim, Norwegian
20 0.77 0.899 0.077 0.48 88.03 University of Science and Technology (2013).
22 0.75 0.874 0.077 0.48 85.87 [4] A. Židonis and G.A. Aggidis, "State of the Art in
Numerical Modelling of Pelton Turbines". Renewable
Table 4: Water distribution at exit of runner and Sustainable Energy Reviews, Vol.45 pp. 135-144
Number of Axial Radically Radically (2015)
buckets (%) inward (%) outward (%)
18 91.80 7.08 1.12 [5] M. Peron, E. Parkinson, L. Geppert, T. and Staubli,
20 89.93 9.09 0.98 "Importance of Jet Quality on Pelton Efficiency and
22 88.06 10.84 1.10 Cavitation", Proceedings of 7th International
Conference on Hydraulic Efficiency Measurements,
IV. CONCLUSIONS IGHEM, Milan, Italy (2008)
The torque and efficiency are highest for 20 buckets. [6] M. Nechleba, Hydraulic Turbines: Their Design and
The fluctuation in torque is seen more for less number of Equipment, Penerbit Artia, (1957).
buckets. The maximum efficiency has occurred nearly at [7] V. Gupta, V. Prasad, and R. Khare. "Numerical
same speed coefficient. The blade loading is found to be simulation of six jet Pelton turbine model." Energy
increasing with increase in number of buckets and it is 104: 24-32 (2016).
maximum near the middle of stream length. The decrease [8] R. Fiereder, S. Riemann and R. Schilling, "Numerical
in average values of relative and whirl velocities from and Experimental Investigation of the 3D Free Surface
inlet to outlet confirms characteristics of Pelton turbine. Flow in a Model Pelton Turbine", Proceeding of 25th
The deflection of jet decreases with increase in number of IAHR symposium on Hydraulic Machinery and
buckets. The flow of water is maximum in axial direction. Systems, Timisoara, Romania (2010).