Study of silt erosion on performance of a Pelton turbine

M.K. Padhy
a,
*
, R.P. Saini
b
a
ITER, S O A University, Bhubaneswar, India
b
Alternate Hydro Energy Centre, Indian Institute of Technology Roorkee, Roorkee, India
a r t i c l e i n f o
Article history:
Received 15 May 2010
Received in revised form
28 October 2010
Accepted 30 October 2010
Available online 8 December 2010
Keywords:
Efciency loss
Pelton turbine
Silt concentration
Silt size
Operating hour
Jet velocity
a b s t r a c t
The erosion of turbine components causes the decrease in turbine efciency; increase in the maintenance
cost and down time for turbine repair. All these factors contribute to the plant cost and cause the revenue
losses. It is, therefore, essential to study the amount of efciency loss owing to the erosion of turbine
components due to the silt particles coming in water. The present study has been carried out to investigate
the effect of silt erosion for different silt laden parameters and operating parameters such as silt concen-
tration, sizeof silt particles, jet velocityandoperating hours of theturbineonefciencyloss inPeltonturbine
buckets.
It has been found that turbine efciency loss depends strongly on silt and operating parameters. A
correlation has also been developed as a function of silt parameters and operating condition for percentage
of efciency loss. The developed correlation may be useful for turbine manufactures in order to predict the
efciencyloss of a Peltonturbine at the manufacturing stage of a turbine under a givensilt content water for
a particular site.
2010 Elsevier Ltd. All rights reserved.
1. Introduction
A number of hydro electric project sites in the Himalayan ranges
and in the northeastern region of India face severe silt erosion
problems in turbines, which over a period of time drastically reduce
the overall efciency of power generation system. A study of
several existing hydro electric power stations located in these
regions has conrmed the severity of silt erosion on the critical
underwater parts of turbines and other components of the hydro-
power stations [1]. Himalayan rivers contain very high sediment
concentration during the monsoon season. Major components of
this sediment are hard abrasive sand and silt which severely
damage the turbine components [2e4].
For efcient operationof hydropower plants, inorder tomeet the
electricitydemand, thehydroenergyis storedeither inreservoirs for
dam based schemes or settling basins for run-of-river schemes.
These reservoirs or settling basins are lled with sediments over
a period of time. This problem must be taken care of by sediment
settling systems inpower plants. However, lot of unsettledsediment
pass through the turbines every year and turbine parts are exposed
to severe erosion. The erosion of hydro turbine components is
a major problem for the efcient operation of hydropower plants.
These problems are more prominent in power stations which are of
run-of-river types. The problem is aggravated if the silt contains
higher percentage of quartz, which is extremely hard. Nepal is also
facing severe silting problem in hydropower plants with specic
sediment yield of about 4240 tonnes/km
2
/year [5].
Silt erosion is a result of mechanical wear of components due to
dynamic action of silt owing along with water. However, the
mechanism of erosion is complex due to interaction of several
factors viz. particles size, shape, hardness, concentration, velocity,
impingement angle, properties of material and so on. The silt laden
water passing through the turbine is the root cause of silt erosion of
turbine components which consequently leads to a loss in ef-
ciency. This affects output, abetting of cavitation, pressure pulsa-
tions, vibrations, mechanical failures and frequent shut downs.
Since silt erosion damage is on account of dynamic action of silt
with the component, properties of silt, mechanical properties of the
component in contact with the ow and conditions of ow.
Therefore, these are jointly responsible for the intensity and
quantum of silt erosion. The erosion damages are to some extent
different for Pelton and Francis turbines. In case of Pelton turbines,
needle, seal rings in the nozzles and runner buckets, splitter are
most exposed to sand erosion. In case of Francis turbines runner
vane, guide vane cascade and the labyrinth rings are exposed to
wear. The hydraulic machines, working under medium and high
heads are normally exposed to erosion. High head Francis and
Pelton turbines are highly affected by sand erosion. Even low head
* Corresponding author. Tel.: 91 9439500444.
E-mail address: mamatapadhy@rediffmail.com (M.K. Padhy).
Contents lists available at ScienceDirect
Energy
j ournal homepage: www. el sevi er. com/ l ocat e/ energy
0360-5442/$ e see front matter 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.energy.2010.10.060
Energy 36 (2011) 141e147
Kaplan turbine and propeller turbines are also found to be eroded
in river water with high sediment contents. Bucket, nozzle and
needle in case of impulse turbines and guide vanes, faceplates,
runner blades and seal rings in case of reaction turbines are most
affected parts.
Different investigators [6e8] developed correlations for erosion
rate of turbine runners as a function of different parameters such as
silt concentration, quartz content, particle size, and ow velocity.
These correlations developed by different investigators, were based
on experiments conducted on different type of test rigs.
Takgi et al. [9] conducted the hydraulic performance tests on
a Francis turbine model with sediment laden ow. They observed
the turbine best efciency decreased in direct proportion to the
increase in solids concentration.
An extensive literature review on the erosion models, experi-
mental investigations on erosive wear and case studies of erosion of
hydro turbine components were carried out by Padhy and Saini [10].
Brekke [11] in a case study of Driva Power Plant observed that
erosive wear of facing plates on head cover and lower cover of
a Francis turbine leads to an increased leakage between the guide
vane facing and covers and caused around 10% relative efciency
loss in Driva Power Plant. In case of Pelton turbine, Brekke [12]
observed that the height of splitter gradually decreases due to
erosion which was proportional to the erosion depth in the bucket
and the highest loss in efciency was at best efciency point (BEP).
Pradhan [13] predicted efciency loss for suspended sediment
concentration in Jhimruk river during the peak monsoon, where its
average values ranged fromabout 20,000 to as high as 60,000 ppm.
Bajracharya et al. [14] predicted relation between the erosion
rate of spear of Pelton turbine nozzle and the reduction of efciency
from the eld data and the efciency reduction was expressed in
the form of correlation.
It is revealed fromliterature survey that a number of researchers
have conducted experiments to study the effect of silt parameters
on erosive wear. Most of these experiments are on small size
samples in different types of test rigs to simulate the ow condi-
tions in the turbine. However, actual ow conditions and the
phenomenon of erosive wear are too complex to simulate. In order
to take into account the factors (i) the actual ow conditions, (ii)
operating hours of the turbine, (iii) the curvature of the Pelton
turbine bucket, (iv) the variation of the size and the concentration
of the eroding particles and (v) jet velocity, the present investiga-
tion has been carried out for actual ow conditions over Pelton
buckets of a small scale Pelton turbine.
2. Experimental study
An experimental setup is designed and fabricated to carry out
the proposed investigation. Fig. 1 shows a schematic of the test
setup. The setup consists of; (i) Pelton turbine runner, (ii) Water
Nomenclature
C silt concentration, ppm
H the head to the turbine, m
P
o
the electrical power output, kW
Q the discharge through the turbine, m
3
/s
S silt particle size, mm
V velocity of ow, m/s
C
w
fraction of solid by weight
g acceleration due to gravity, m/s
2
t operating hour, h
h
g
the efciency of the generator (0.96 for the
considered generator)
h
o
the efciency of turbine, %
h
m
turbine peak efciency with sediment laden ow
h
w
turbine peak efciency with clean water
g specic weight of water, kN/m
3
Fig. 1. Schematic of experimental setup [15]. Description: 1: Pelton turbine runner; 2: Water tank; 3: Stirrer; 4: Cooling jacket; 5: Service pump; 6: Penstock pipe; 7: Spear valve;
8: Control valve; 9: Channel with weir; 10: Generator set; 11: Pressure transducer; 12: Control panel with ballast load; 13: Driving motor; 14: Motor for driving stirrer.
M.K. Padhy, R.P. Saini / Energy 36 (2011) 141e147 142
tank, (iii) Stirrer, (iv) Cooling jacket, (v) Service pump-motor set,
(vi) Penstock pipe, (vii) Spear valve and nozzle, (viii) Control valve,
(ix) Channel with weir, (x) Generator and a control panel with
resistive load and (xi) the instrumentation. Details of experimental
setup, instrumentation, methodology for experimentation and data
generated for silt erosion under different conditions are discussed
in relation to previous studies carried by the authors [15].
The most important part of the setup is the Pelton turbine
runner. Pelton turbine runner having pitch circle diameter of
245 mm, mounted with 16 numbers of buckets is used for the
experiments. The Pelton turbine buckets made up of brass are
chosen as specimen to get measurable amount of erosion in a short
time period. The weight of each specimen bucket is 205 1.5 g.
Shaft of the runner is directly coupled with the generator shaft. A
photograph of the runner used under the present study is shown in
Fig. 2. Based on the results obtained during experimentation, effect
of silt parameters and operating parameters on the efciency of
Pelton turbine has been discussed in this paper.
3. Method of efciency measurement
In order to compute the efciency of Pelton turbine the input
to the turbine in terms of head and discharge was calculated. A
generator was coupled with the runner shaft to generate the elec-
tricity as output. The output of the turbine was recorded by consid-
ering the efciency of generator. A control panel was connected to
the generator to record the power generation of the generator. The
control panel consisted of a Voltmeter, a Wattmeter and electric
bulbs as ballast load. Power output from the generator was moni-
tored and recorded continuously with the help of Wattmeter having
a least count of 10 W.
Input to the turbine was maintained constant throughout the
experiment by keeping the head and ow as constant to the
turbine. Discharge from the turbine under different heads was
measured with the help of a rectangular notch. A free fall of water
fromthe weir to the tank was ensured. A pointer gauge was used to
measure the height of water over the notch. The discharge of the
pump corresponding to each head was recorded. Then the channel
was closed at the turbine outlet and discharge fromthe turbine was
allowed to enter into the tank. A control valve was set to maintain
the constant head and ow during a particular set of readings.
By considering discharge and head readings, turbine input was
computed. Input to the turbine was calculated from the head and
discharge from the turbine by using the following expression:
P
i
g H Q (1)
The efciency of turbine was calculated as follows:
h
o

P
o
g H Q h
g
100 (2)
where H is the head to the turbine, m; P
o
is the electrical power
output, kW; Q is the discharge from the turbine, m
3
/s; h
o
is the
Fig. 2. Photograph of Pelton turbine runner.
Table 1
Range of parameters.
Parameters Range
Concentration 5000e10,000 ppm
Silt size Upto 355 mm
Jet velocity 26.62e29.75 m/s
Operating time 8 h run for each set
y = 1034.6x
-0.0492
R
2
= 0.9562
700
750
800
850
900
950
1000
1050
0 1 2 3 4 5 6
Mass loss,g
P
o
w
e
r

o
u
t
p
u
t
,
W
Fig. 3. Effect of bucket mass loss on turbine power output.
y = 0.6187x
-0.0479
R
2
= 0.9813
0.4
0.45
0.5
0.55
0.6
0.65
0 1 2 3 4 5 6
Mass loss, g
E
f
f
i
c
i
e
n
c
y
,

o

Fig. 4. Effect of bucket mass loss on turbine efciency.
M.K. Padhy, R.P. Saini / Energy 36 (2011) 141e147 143
efciency of turbine, %; h
g
is the efciency of the generator (0.96
for the considered generator); g specic weight of water (kN/m
3
)
It was observed from the experiments that the power output
decreasedwiththebucket mass loss. Therefore anattempt was made
to establish a correlation for efciency loss as a function of bucket
mass loss for different silt parameters and operating parameters.
Pelton turbine runners are designed either for casting of the disc
and buckets in one piece (i.e. monocast) or the disc and the buckets
separately. Single cast of complete runner is the preferred and
commonly used method for the Pelton turbines in modern power
plants where the turbine units are for high power generation and of
bigger sizes. However, in case of defect found in one bucket, the
whole runner has to be rejected. Moreover, for manufacturing of
runners with disc and buckets separately is of advantageous as in
case of defect found in any bucket, the individual bucket is to be
discarded. Different methods are adopted for xing the individual
buckets to the disk. The stem of the bucket is supposed to be
designed and manufactured of different shapes.
The stem of the bucket is relatively heavier in comparison to the
cup (bucket without stem). In order to discuss output and efciency
loss of turbine with respect to percentage mass loss of bucket, mass
of cup was considered. Accordingly, the percentage mass loss of cup
was determined by considering the mass of cup alone. During
experimentation, parameters related to input and output of the
turbine were measured. The values of these measured parameters
were used to nd out the efciency of the turbine under different
conditions considered for this investigation. As discussed above,
output and efciency of the turbine were determined. The effect of
mass loss of buckets on the power output and efciency of the
turbine is discussed.
Basic mass volume method was adopted to determine the mass
of the bucket without the stem. A measuring ask lled with water
was used for nding out the mass ratio. The bucket was dipped
completely and the volume of water was recorded. Again the stem
of the bucket only was dipped in water and the volume of water
was recorded. Mass of the bucket without stem was calculated
using following relationship:
Mass of bucket without stem
Mass of bucket with stem

Volume of bucket without stem

Volume of bucket with stem
(3)
Percentage mass loss was determined with respect to the total
mass of bucket without stem.
4. Range of parameters
Under the present study, effects of three parameters such as; silt
concentration, silt size and jet velocity on Pelton turbine efciency
loss were investigated. Sample of sand was collected from river
Bhagirathi (India) near the head works of Maneri Bhali Hydro
Electric Project, Phase I, which is one of the most silt affected power
houses in India. The quartz content in the sand sample was more
than 90% [8]. Sand was dried in an oven for twenty four hours and
then sieved to different sizes before mixing with water. The range
of the parameters considered for the present investigation is given
in Table 1. Error analysis carried out for the study was reported
earlier by authors [15].
5. Results and discussion
Effect of bucket mass loss on output and efciency of turbine is
shown in Figs. 3 and 4. Fig. 3 shows the power output versus bucket
mass loss. It was found that power output of the turbine decreased
with mass loss of the bucket. It was observed that the rate of power
loss initially found to be more. However rate of power loss in later
V = 28.32 m/s
t = 8 h
0.6
0.8
1
1.2
1.4
1.6
1.8
2
4000 5000 6000 7000 8000 9000 10000 11000
Concentration,ppm
P
e
r
c
e
n
t
a
g
e

e
f
f
i
c
i
e
n
c
y

l
o
s
s
,

302 micron
215 micron
135 micron
45 micron
Fig. 5. Percentage efciency loss versus silt concentration.
V = 28.23 m/s
t = 8h
0.5
0.7
0.9
1.1
1.3
1.5
1.7
1.9
2.1
0 50 100 150 200 250 300 350
Size, micron
P
e
r
c
e
n
t
a
g
e

e
f
f
i
c
i
e
n
c
y

l
o
s
s
,

10000 ppm
7500 ppm
5000 ppm
Fig. 6. Percentage efciency loss versus silt size.
C = 5000 ppm
S = 135 micron
t = 8 h
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
26 26.5 27 27.5 28 28.5 29 29.5 30
Jet velocity, m/s
P
e
r
c
e
n
t
a
g
e


E
f
f
i
c
i
e
n
c
y

l
o
s
s
,

Fig. 7. Percentage efciency loss versus jet velocity.

y = 4.0108x
0.726
R
2
= 0.9986
0
2
4
6
8
10
12
0 0.5 1 1.5 2 2.5 3 3.5 4
Percentage mass loss
P
e
r
c
e
n
t
a
g
e

e
f
f
i
c
i
e
n
c
y

l
o
s
s
Fig. 8. Effect of percentage mass loss on percentage efciency loss of turbine.
M.K. Padhy, R.P. Saini / Energy 36 (2011) 141e147 144
stage was found to be at lower rates and became asymptotic. Similar
trend was observed in case of turbine efciency as shown in Fig. 4.
As discussed earlier, presentation of effect of silt parameters on
turbine output and efciency would be more relevant if these were
discussed with percentage mass loss of bucket. Accordingly, the
effect of silt parameter on percentage efciency loss has been dis-
cussed as follows.
Fig. 5 shows percentage efciency loss of turbine with silt
concentration for different values of silt size and a xed value of jet
velocity. For a givenvalueof silt particlesize, efciencyloss was found
to be about 2 percent with concentration value from 5000 ppm to
10,000 ppm. Rate of percentage efciency loss was found to be
signicant inthis case. However, rateof percentageefciencyloss was
found to be nominal with silt size as shown in Fig. 6.
Fig. 7 shows the percentage efciency loss of turbine with jet
velocity for silt concentration 5000 ppm and for xed value of
mean silt size of 135 mm. Rate of percentage efciency loss was
found to be 0.25e0.40 with a jet velocity variation from 26 m/s to
30 m/s, which is to be considered signicant.
Fig. 8 shows percentage efciency loss with percentage mass
loss of the bucket. It was found that about 8% efciency loss
occurred against a bucket mass loss of 3.5%. This shows that ef-
ciency of the turbine decreased signicantly with bucket mass loss.
An attempt was made to develop a correlation for efciency loss
of turbine as a function of silt size, silt concentration, jet velocity
and operating hours of the turbine.
6. Development of correlation
It was observed that bucket mass loss and alteration of bucket
shape played a critical role in the Pelton turbine efciency loss. The
system designer may require the correlations for efciency drop of
Pelton turbine under the actual conditions having silt laden water
ow. Under the present investigation correlation for efciency loss
as a function of system and operating parameters was developed
from the experimental data. As discussed earlier, the efciency loss
was strongly dependent on the silt parameters, silt size (S), silt
concentration (C), water jet velocity (V) and operating hours of the
turbine (t).
Saini and Saini [16] mentioned that it was feasible to adopt
a statistical approach for the development of correlations from
experimental data. The correlation for efciency loss from the
experimental data was developed by following the procedure given
by Saini and Saini [16]. Sigma Plot (version 10.0) software was
employed for carrying out the regression analysis. Different plots
are shown on a linear scale, however in order to have the best curve
tting, data on log scales were employed while carrying out the
regression analysis. It was found that the regression of data deals
with rst order polynomials.
Following this approach, the present correlation is developed.
The percentage efciency loss is presented as a function of
system and operating parameters and can be written as:
Log (t)
0.2 0.4 0.6 0.8 1.0
L
o
g

(

%
)
-1.8
-1.6
-1.4
-1.2
-1.0
-0.8
-0.6
-0.4
-0.2
At =
Fig. 9. Plot of log(h
%
) versus log(t).
Log (S)
1.6 1.8 2.0 2.2 2.4 2.6
/
t
)
-2.0
-1.8
-1.6
-1.4
-1.2
-1.0
A S
t

=
Fig. 10. Plot of log(h
%
/t
0.75
) versus log(S).
Log (C)
3.7 3.8 3.9 4.0
L
o
g

(

%
/
(
t
0
.
7
5
S
0
.
0
4
5
)
)
-1.7
-1.6
-1.5
-1.4
-1.3
-1.2
-1.1
-1.0
-0.9
Fig. 11. Plot of log(h
%
/t
0.75
S
0.099
) versus log(C).
Log (V)
1.42 1.43 1.44 1.45 1.46 1.47 1.48
-5.00
-4.95
-4.90
-4.85
-4.80
-4.75
-4.70
-4.65
-4.60
Fig. 12. Plot of log(h%/t
0.75
S
0.099
C
0.93
) versus log(V).
M.K. Padhy, R.P. Saini / Energy 36 (2011) 141e147 145
h
%
f S; C; V; t (4)
Fig. 9 shows the rst order regression of the log scale data of
operating time versus efciency loss, which shows the average
value of m (average slope of lines) equal to 0.75. Therefore ef-
ciency of the turbine can be represented as:
h
%
anti log C
1
t
0:75
(5)
or
h
%
A
1
t
0:75
(6)
In order to induce the effect of silt size parameter, the values of
h
%
=t
0:99
were calculated from the data of normalized erosive wear
obtained experimentally and plotted against log value of particle
size considered for the experimentation as shown in Fig. 10. The
correlation obtained is presented in the following:
A
1
Anti log C
2
S
0:099
(7)
or,
h
%
t
0:75
Anti log C
2
S
0:099
(8)
or,
h
%
t
0:75
A
2
S
0:099
(9)
Similarly, the effect of silt concentration parameter is induced in
the Eq. (17) by representing coefcient A
2
as a function of silt
concentration. These values were plotted against respective silt
concentration values as shown in Fig. 11. The correlation obtained is
presented as follows:
A
2
Anti log C
3
C
0:93
(10)
or,
h
%
t
0:75
S
0:099
Anti log C
3
C
0:93
(11)
or,
h
%
t
0:75
S
0:099
A
3
C
0:93
(12)
In order to induce the effect of velocity parameter in the above
equation, A
3
was developed as a function of jet velocity from the
rst order regression of the data on log scales. Values of C
4
and m
were found to be as 9.6146 and 3.40 respectively. These values are
plotted against the value of jet velocity as shown in Fig. 12. Finally
the correlation obtained is in following form:
h
%
t
0:75
S
0:099
C
0:93
Anti log C
4
V
3:40
(13)
h
%
t
0:75
S
0:099
C
0:93
A
4
V
3:40
(14)
The nal form of the correlation for percentage efciency loss of
rated efciency was obtained as follows:
h
%
2:43 10
10
t
0:75
S
0:099
C
0:93
V
3:40
(15)
Fig. 13 shows the comparison between observed value and
predicted value of efciency loss with respect to the mass loss of
bucket. A good agreement was observed and an average absolute
percentage deviation between the experimental and calculated
data for efciency loss was found within 10%.
7. Conclusion
Based on the experimental investigation it was found that the
silt concentration, silt size, jet velocity and the operating hours of
the turbine were strong parameters for efciency loss in Pelton
buckets. It was concluded that the percentage efciency loss of
a Pelton turbine increased with an increase in the silt concen-
tration, silt size and jet velocity. It was observed that the turbine
power output was decreasing with the mass loss of Pelton
buckets. The rate of power loss was initially found at a higher rate;
however the rate of power loss was found to be at lower rates and
became asymptotic after operating the turbine over a long period
of time.
Based on experimental data obtained, a correlation was devel-
oped for percentage efciency loss as a function of silt concentra-
tion, silt particle size, jet velocity and operating hours of the
turbine. The developed correlation might be useful for turbine
manufacturing industries in order to predict the efciency loss of
a Pelton turbine at the manufacturing stage of a turbine under
a given silt content water at a particular site.
References
[1] Khera DV, Chadhawork RS. Silt erosion.trouble for turbines. International
Water Power and Dam Construction 2001;53:22e3.
[2] Tong D. Cavitation and wear on hydraulic machines. International Water
Power and Dam Construction 1981;April.
[3] Pande VK. Criteria for developing silt erosion model test facility. Conference of
CBIP; 1987.
[4] Singh SC. Operational problems and development of a new runner for silty
water. International Water Power and Dam Construction 1990;November.
[5] Chaudhary CS. Impact of high sediment on hydraulic equipment of Mar-
syangdi hydropower plant. In: Proceedings of the international seminar on
sediment handling technique, NHA, Kathmandu; 1999.
[6] Krause M, Grein H. Abrasion research and prevention. Hydropower Dams
1996;4:17e20.
[7] Naidu BSK. Addressing the problems of silt erosion at hydro plants. Hydro-
power Dams 1997;3:72e7.
[8] Schneider C, Kachele T. Recent research results on predicting and preventing
silt erosion. In: Proc. of 1st int. conf. on silting problems in hydro power
plants, New Delhi, vol. III; 1999. p. 2e13.
[9] Takgi T, Okamura T, Sato J. Hydraulic performance of Francis turbine for
sediment-laden ow. Hitachi Review 1988;2.
[10] Padhy MK, Saini RP. A review on silt erosion in hydro turbines. Renewable and
Sustainable Energy Reviews 2008;12:1974e87.
[11] Brekke H. Discussion of Pelton turbine versus Francis turbines for high head
turbines. Colorado: IAHR; 1978.
[12] Brekke H. Design of hydraulic machinery working in sand laden water. In:
Duan CG, Karelin VY, editors. Abrasive erosion and corrosion of hydraulic
machinery. London: Imperial College Press; 2002.
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Percentage efficiency loss, observed
P
e
r
c
e
n
t
a
g
e

e
f
f
i
c
i
e
n
c
y

l
o
s
s
,

p
r
e
d
i
c
t
e
d
Fig. 13. Comparison of experimental values with predicted values of efciency loss.
M.K. Padhy, R.P. Saini / Energy 36 (2011) 141e147 146
[13] Pradhan PMS. Improving sediment handling in the Himalayas. Nepal: OSH
Research; 2004. October.
[14] Bajracharya TR, Acharya B, Joshi CB, Saini RP, Dahlhaug OG. Sand erosion of
pelton turbine nozzles and buckets: a case study of Chilime hydropower
plant. Wear 2008;264:177e84.
[15] Padhy MK, Saini RP. Effect of size and concentration of silt particles on erosion
of pelton turbine buckets. Energy 2009;34(10):1477e83.
[16] Saini RP, Saini JS. Heat transfer and friction factor correlations for articially
roughened ducts with expanded metal mesh as roughness element. Inter-
national Journal of Heat and Mass Transfer 1997;40(4):973e86.
M.K. Padhy, R.P. Saini / Energy 36 (2011) 141e147 147