International Journal of
Advanced Engineering
Website: www.ijetae.com (ISSN 2250-
Volume 3, Issue 8, August 2013)
The Study of End Losses in a Three Dimensional Rectilinear
Turbine Cascade
Vinod Kumar Singoria1, Samsher2
1
Research Scholar, 2Professor, Department of Mechanical Engineering, Delhi Technological University (DTU), (Formerly
Delhi College of Engineering) Bawana Road, Delhi –India-110042.
Abstract--Three dimensional geometry of rectilinear cascade
of four reaction blades is created using Gambit 2.4.6® software
and solved using CFD commercial software FLUENT 6.2.16 ®.
Air with an inlet velocity of 102 m/s is passed through the
cascade. The cascade is open to atmosphere at the exit. Initially,
both surfaces of the blades of the cascade are kept as smooth
and secondary loss is analyzed. This secondary flow loss is then
compared with the blades on which roughnesses of 250, 500,
750 &1000 µm were applied on suction and pressure surfaces
individually as well as on both the surfaces together. It is
observed that in a cascade having both surfaces smooth, mass
averaged total loss is 10.53%. Whereas this loss gets almost
doubled and becomes 19.72% when a roughness of 1000 µm is
applied on both the surfaces. When roughness of 1000 µm is
applied separately on the suction and pressure surfaces of all
the blades, the mass averaged total loss is 17.96% and 12.33%
respectively. These total losses are segregated into secondary
flow loss or end loss and profile loss. It is observed that
contribution of secondary loss is 4.94%, 5.98% and 2.75%
respectively for smooth and roughness of 500 µm is applied
over pressure & suction surfaces separately. The same is
5.79%, 2.47% & 4.37 % when roughness of 1000 µm is applied
over pressure & suction surfaces & over both surfaces together
respectively. Therefore, the secondary loss is maximum with
cascade having pressure surfaces rough and minimum in which
suction surfaces of blades are roughened. However, the effect
of roughness on secondary loss in case of blades having both the
surfaces rough is of intermediate values. The trend of increase
in losses is insignificant with the increase in magnitude of
roughness (beyond 500µm) for all types of application of
roughness.
Keywords-- cascade, energy loss, secondary flow loss, profile
loss, roughness.
I. INTRODUCTION
The turbine is a heat engine designed specifically to
convert the thermal or potential energy of working substance
into rotational kinetic energy, which in turn drives a
generator to generate electricity. As the working substance
flows through the turbine its temperature and pressure vary
from stage to stage.
782
Emerging Technology and
2459, ISO 9001:2008 Certified Journal,
Each stage of a turbine employs fixed and moving blades
having an airfoil shape that is optimized for the inlet fluid
conditions. In a closed cycle steam turbine, as it operates,
there is a continuous buildup of various impurities in the feed
water with time. Contact between contaminated steam/water
with metal surfaces in stationery and moving parts results in
corrosion, erosion, and deposition [Samsher, 2007]. The net
result is a continuous deterioration in performance and
reliability of turbine. Efficiency loss occurs when erosion,
deposition or corrosion results in increased surface
roughness. This effect is more marked in the high pressure
(HP) or intermediate pressure (IP) than in the low pressure
(LP) stage of turbine due to high Reynolds number in HP
turbine. In low pressure turbine, generally, the Reynolds
number is below the critical value.
The viscous diffusion in the flow through turbine cascade
results in the decrease in integrated flux of total pressure
through the cascade. Since this decrease in total pressure flux
is related to the amount of kinetic and potential energy loss
in the cascade, hence this pressure flux is termed as ‗total
pressure loss‘ or simply ‗loss‘. This total pressure loss has
significant effect on the efficiency of the cascade and hence
it should be minimized.
Losses of various kinds, adversely affects the efficiency of
turbines. The two major losses encountered in the cascade
are termed as ‗profile loss‘ and ‗end loss‘ or secondary loss.
Profile loss: The profile loss is the loss due to boundary
layer on the blade surface and trailing edge thickness. When
a fluid flows over a solid surface, the fluid velocity changes
from zero at the wall (for stationary wall) to its free stream
value. This fluid layer in the neighborhood of the solid
boundary where the effect of fluid friction (viscous effects)
is predominant is called the boundary layer. The flow outside
the boundary layer can be considered frictionless or potential
flow. The boundary layer separates near trailing edge as it
has some definite thickness. The separation points shift
towards leading edge depending upon the extent of adverse
pressure gradient [Samsher, 2002].
International Journal of Emerging
Engineering
Website: www.ijetae.com (ISSN 2250-
Volume 3, Issue 8, August 2013)
Build-up of various impurities, erosion, and corrosion
introduces the roughness over the surface of the blades,
which in turn greatly affect magnitude of profile loss through
the turbine blades.
End wall or secondary losses: The term secondary flows
refers to the three- dimensional vortical flow structures that
develop in blade passages due to high turning of the flow and
non-uniform inlet total pressure profiles. The work output is
significantly affected due to the Secondary Flow. Secondary
flows are superimposed upon the mean flow in transverse
plane of the primary flow. Secondary flows cause to generate
a non-uniform flow at exit of the blade row thereby
efficiency of the blade row downstream gets reduced.
Roughness over the surface of the blades also significantly
affects the secondary losses. Other than profile losses,
secondary losses also contribute significantly to the total loss
[Yahya, S. M.,1983], if the blade is short.
Formation of vortex and Secondary flow phenomenon:
The formation of vortex in a turbine system largely affects
the secondary loss. The incoming boundary layer separates
upstream of the leading edge, forming a horseshoe vortex.
This vortex is consisted of two legs. The leg formed at the
pressure side and suction side of the blade are called pressure
vortex and suction vortex respectively. The legs so formed
differ from each other. The suction side leg is affected mainly
by curvature of the suction surface. Whereas the pressure
side leg is affected by curvature and pressure difference
between pressure side and suction side of adjacent blade of
the corresponding flow channel. The formation of a suction
and pressure side leg and passage vortex in a wind tunnel
experiment is shown in figure 1. The pressure side vortex leg
is usually increases towards the exit of the cascade.
Langston et al. [1977] was among the first to study the
evolution of secondary flows using hot wire and flow
visualization techniques to qualitatively assess flow patterns
at boundary layer, near the end wall region of a cascade.
According to Langston, the incoming inlet boundary layer
splits into two streams, one moves towards the pressure
surface and other towards suction surface. Adverse pressure
gradient is resulted in 3-D flow separation and horseshoe
vortex formation. Passage vortex is formed due to boundary
layer and pressure gradient across the blade passage and
rotates in anticlockwise direction. Cross flow is observed at
end wall, because of the blade to blade pressure gradient.
Suction surface leg rotates in opposite direction of pressure
surface leg and consequently termed as counter vortex.
783
Technology and Advanced
2459, ISO 9001:2008 Certified Journal,
The pressure surface leg of horseshoe vortex merges and
strengthens the passage vortex. Later, different experiments
were performed by Marchal et al. [1977], Sieverding et al.,
[1983], Wang et al. [1997] and Sharma et al. [1987] which
complied with the conclusions of Langston et al. [1977].
Wang et al. [1997] concluded that pressure side vortex
moves towards the suction side and merge with passage
vortex at approximately one fourth of the distance from the
leading edge.
There are two main designs for leading edge geometry: the
fillet and the bulb for reducing secondary flow loss. Young
J. Moon et al. [2000] analyzed the effect of end wall fencing
for reducing the secondary flow using k-ξ turbulence model.
They also justified the optimized positioning of the endwall
fencing for reducing the secondary flow losses, because the
end wall fencing prevents the merging of pressure side horse
shoe vortex with the passage vortex and hence total pressure
loss decreases.
Arun K. Saha et al. [2008] analyzed the turbulent flow
through a three dimensional non-axisymmetric blade passage
and observed that endwall contouring reduces the pitch wise
pressure gradient near the endwall which reduces the chances
of flow separation. SonodaToyotaka et al. [2009] use axis-
symmetrical end wall contouring method for reducing the
secondary losses in high pressure turbine having low aspect
ratio. They investigated the effect of three types of end wall
contouring: 1) only hub contour, 2) only tip contour and 3)
hub and tip contour and observed that hub contouring, the tip
contouring and the hub and tip contouring all reduce the mass
averaged overall loss by 4%, 5%, and 10%, respectively, as
compared to the base line.
Brear et al. in [2010] strived to reduce the pressure surface
separation by modifying the leading edge geometry. They
observed that increasing the blade thickness at the pressure
surface decrease the strength of secondary flow by increasing
the momentum near the wall. Shih et al. [2003] observed
effects of leading-edge airfoil fillet on the flow in a turbine.
The increased size of the stagnation zones on the endwalls
about the airfoil‘s leading edge lowers the flow speed and
velocity gradients there, which in turn reduces turbulence
production. G. I. Mahmood et al. [2007] studied the
secondary flow structure in a blade passage with and without
leading edge fillet and observed that the size and strength of
the passage vortex become smaller with the fillets.
T. Korakianitis et al. [2010] has proposed a direct
design method based on specifying blade surface-
International Journal of Emerging
Engineering
Website: www.ijetae.com (ISSN 2250-
Volume 3, Issue 8, August 2013)
curvature distributions so as to minimize the chances of
flow separation. Qi Lei et al. [2011] analyzed the effect
of leading edge modification on the secondary loss. They
used vortex generator for introducing counter rotating
vortex which oppose the passage vortex and hence
reduce the secondary flow losses.
Much work has been done to understand the
occurrence and modeling of secondary flow and end loss
phenomenon. Moreover researchers had tried to reduce
the secondary loss in any cascade in order to get higher
aerodynamic efficiency of the power plant. It is a well-
established fact that roughness over the blade surface
increases the profile loss and corresponding total loss for
a given cascade. But effect of roughness on the secondary
flow and corresponding losses has not been studied
much. In this paper results of computational study of
flow through Three Dimensional Rectilinear Turbine
Cascade in general and secondary loss and its
dependence on roughness of the blades of the Cascade in
particular is presented.
II. MODELLING
The friction losses and the secondary losses are
generated as the fluid flows through blades passages in
the turbine. The roughness on aerofoil surfaces affects
these losses greatly. The present computational study for
finding End Losses in a Three Dimensional Rectilinear
Turbine Cascade has been carried out using
Computational Fluid Dynamics (CFD). The brief of CFD
software used and boundary conditions is presented here.
2.1 CFD Simulation
CFD is a computational tool that solve the
fundamental nonlinear differential equations (mentioned
below) that describe fluid flow (Navier-Stokes and
allied equations), for predefined geometries and
boundary conditions.
i. Conservation of momentum (Navier-Stokes equation)
p ij
( ui ) ( uiu j ) gi
t xj xj xj
(1)
Technology and Advanced
2459, ISO 9001:2008 Certified Journal,
ii. Conservation of energy (Energy Equation)
+
E (ui ( E p))=
( )
t xi
T u(
h +
xi keff xi j' j' j j' j )
ij eff + Sh (2)
iii. Conservation of mass (Continuity equation)
( ui ) Sm (3)
t xi
FLUENT 6.2.16 ® converts unsolvable governing
equations (Navier-Stokes equations) to a solvable set of
algebraic equations for a finite set of points within the space
under consideration. FLUENT 6.2.16 ® uses iteration
technique for finding detailed information with respect to
velocity, pressure, temperature, and chemical species etc.
Using Gambit2.4.6® as preprocessor, three-dimensional
model of a typical blade cascade geometry has been made
[Samsher, 2002]. Virtual prototype is prepared allocating
proper boundary conditions representing the actual flow
behavior. Thereafter the computational software FLUENT
6.2.16 ® is used as solver & post processor for flow
simulation.
In turbomachinery flow is affected by rotation, three-
dimensionality, curvature, separation, free
stream turbulence, compressibility, large scale
unsteadiness, heat transfer and other effects. Fluid flows of
practical relevance are mostly turbulent. Turbulence models
approximate these transport processes in terms of mean
flow field by empirical formulations. Therefore, in view to
Fi obtain a better result, turbulence modeling is chosen due to
its being realistic.Turbulence models modify the original
unsteady Navier Stokes equations by introduction of mass
averaged fluctuating components to produce Reynolds
Mass averaged Navier Stokes (RANS) equations.
784
International Journal of Emerging
Engineering
Website: www.ijetae.com (ISSN 2250-
Volume 3, Issue 8, August 2013)
The most widely used models for turbomachinery
application is the - model [Shih et al., 2003]. In this
model, the turbulent kinetic energy ( ) and the energy
dissipation rate ( ) are considered as the properties, which
govern the turbulent flow phenomena. The Realizable
turbulence model of Shih et al. [2003] has been selected for
solution of present problem.
This model is expected to provide more accurate results
since it contains additional terms in the transport equations
for and that are more suitable for stagnation flows and
flows with high streamline curvature.
2.2. Description of Computational Domain
A computational study of flow through Three
Dimensional Rectilinear Turbine Cascade for smooth profile
and profile having roughness of 500 µm was carried out
[Vinod Kumar Singoria et al., 2012] results of which were
compared with values of percentage loss coefficients
measured along the pitch by Samsher [2002]. Results of
computational study and experiment were found to be in
agreement as shown in Figure 2. The same inlet and outlet
boundary conditions are taken as compared in experimental
setup.
After validation, the model was taken for investigation of
effect of roughness on the secondary flow loss as percentage
of total loss. The set of following conditions are taken for the
present study of secondary flow loss:
i. both surfaces of the blade of the cascade are smooth ii.
roughness of 250, 500, 750 & 1000 µm is applied
separately over:
a. Entire pressure surfaces of all blades (PSR )
b. Entire suction surfaces of all blades (SSR )
c. Both surfaces (suction & pressure sides) of all blades
(BSR) .
Keeping in view the limitation of computational facility
the cascade of a reaction profile (6030) consists of three flow
channels using four test blades are placed in rectilinear
cascade test section with appropriate stagger angle, chord,
pitch, and inlet fluid flow angle. A three dimensional model
of the profile 6030 was created, with the help of
785
Technology and Advanced
2459, ISO 9001:2008 Certified Journal,
Gambit2.4.6® and the dimensions of the model were kept
same as the experimental setup of Samsher [2006]. Shape
and specifications of turbine blade 6030 cascade model is
shown in Figure 3. Various boundary zones so created from
various faces & boundary types assigned to each of them is
also shown in figure 3. Flow is assumed to be symmetric
about the mid span plane. In order to make 3-D model to
represent the actual flow, first of all a 2D model was created
and then this 2-D model was converted into 3-D by sweeping
the faces of the 2-D model by blade height. After creating the
volume which is subjected to fluid flow meshing of the same
is done.
2.3 Boundary and Operating Conditions
The atmospheric temperature is assumed to be constant at
27 °C, as it varied from 20°C to 35 °C in experiment
performed by Samsher [2002]. The velocity at the inlet is
given as 102 m/s. The pressure outlet value at exit is assigned
as zero gauge pressure, as the exit is directly exposed to
atmosphere. The exit measurement plane is at 15 % distance
of chord distance. Initially blade surfaces were kept smooth
and results were obtained. In addition to these input
conditions for study of secondary flow loss, roughness of
250, 500, 750 and 1000 µm were applied on pressure and
suction surfaces individually and then on both the surfaces
together to see the effect of roughness on secondary flow.
Figure 4 shows a 2D view of passage between two blades of
a rectilinear Cascade.
2.4 Energy Loss Calculation
The phenomenon of secondary flow is observed only near
the end wall. In the mid span of the blade, profile loss is
dominant. Secondary loss at the end walls includes loss from
the boundary wall on the endwall wetted surface, loss due to
flow separation and diffusion of passage secondary vortex.
Energy loss coefficients represent total energy loss while
fluid flow is taking place along the cascade from inlet to
outlet. Energy loss coefficient at mid span where flow is 2-
dimensional represents profile loss, whereas near the walls,
it represents total loss i.e. profile loss plus secondary flow
loss. To segregate secondary loss at the end wall, profile loss
at the mid space was subtracted from the total loss. The value
of energy loss coefficient is calculated using equation-4.
1
International Journal of Emerging Technology and Advanced
Engineering
Website: www.ijetae.com (ISSN 2250- 2459, ISO 9001:2008 Certified Journal,
Volume 3, Issue 8, August 2013)

P2s 1
computed along the complete blade span starting from
y
endwall surface at zero mm to other endwall surface at 95
1 mm height. In order to visualize the flow near the end wall,
1 PP0101 the measurements near the endwalls were taken at small
PP02 1 distances & secondary losses were computed. For first 10
P2s mm height from bottom end wall, measuring points are 2 mm
apart from each other. Thereafter, it was computed at every
= 2s P01 5 mm interval till 85 mm blade height. Finally for the last 10
mm height it was again computed at every 2 mm interval.
Table 1 shows the various intervals at which total loss
coefficients were found.
P01 1 PP2s The total (combined) losses in a blade cascade are
estimated by the energy loss coefficient , which is
1 1 PP0101 PP022s 1
essentially the sum of profile loss coefficient & end loss
PP201s 1 coefficient as given by Kostyuk and Frolov (1988) in
equation 6.
01
(total) = (pr) + (sec) (6)
(4)
To calculate a single value of energy loss coefficient, the 3.1 Effect of Roughness on Total Energy Loss Coefficients:-
mass average value of loss coefficient was calculated using First, mass averaged total energy loss coefficients have been
determined using values of total and static pressures at exit
s It is expected that total energy loss increases till both end

0
y Vady of the blade compared to its value at mid span, but humps are seen just before both the ends. The presence of

humps = s at hub and casi ng occur

the relationas shown in equation 5 [Yahya, S. M., 1983].
Vady
0 a roughness of 250μ is shown in Figure 5. End losses or
(5)
Where is the mass average loss coefficient, Va is the
axial velocity, is the density of air, S is the pitch distance
and dy is the elemental length in pitch wise direction.
III. RESULT AND DISCUSSION
The pressure and velocity vector were analyzed at
appropriate locations in order to understand behavior of the
flow through cascade. Various values of pressure (total as
well as static pressure) at inlet & exit are taken along the
whole blade span. Mass averaged total loss coefficients were
because of the formation of vortex
and total pressure at inlet. Mass averaged total loss
coefficients were computed along the complete blade span.
A plot of total energy loss coefficients for cascade PSR with
secondary losses are obtained by subtracting mid span value
of profile loss from individual mass averaged loss value
along the blade height.
cores that leads to increase in local energy loss coefficient.
Comparison of loss coefficient in case of smooth, Both
surfaces Rough (BSR), Pressure Surface Rough (PSR),
Suction Surface Rough (SSR) Cascades for roughness 250
µm has been shown in Figure 6.
It is observed from the Figure 6 that the loss coefficient is
high at hub and casing due to the endwall boundary layers.
The local increase in loss coefficient is observed due to the
secondary flow cores near the hub and casing. It is also seen
that profile loss is lowest for smooth blade and increases

786
International Journal of Emerging
Engineering
Website: www.ijetae.com (ISSN 2250-
Volume 3, Issue 8, August 2013)
when roughness is applied over pressure, suction surface
separately and together in the same order. The trend of
energy loss coefficient for other roughnesses i.e. 250, 500,
750 and 1000 µm is same. Hence, they are not shown in
graphical form here to avoid repetitiveness but the results are
summarized in table 2. When the roughness of 250 m was
applied over pressure and suction surface separately and
together, the profile loss coefficient increases to 11.71, 15.97
and 17.10% respectively compared to 10.53% in case of
smooth. It can be concluded that the same roughness is
significantly detrimental when applied over suction surface,
whereas it increases marginally compared to smooth blade
value when roughness is applied over pressure surface. The
combine effect of roughness over pressure and suction
surface is seen when the case of roughness over both the
surfaces is applied together.
3.2 Effect of roughness on Secondary loss :
It is evident from results shown in Figure 6 and Table 2
that difference in percentage values of total energy loss &
Profile loss gives percentage of secondary loss in total
energy. In case of smooth blade secondary loss is 4.94%.
Since the total energy loss is also increased with application
of roughness, in spite of absolute change in secondary loss,
it is not truly reflected in the percentage change in secondary
loss when the absolute change in secondary loss is non-
dimensionalised with total loss with the same roughness.
Therefore, the percentage of secondary loss in total energy
loss is shown in last column of table 2, has been non-
dimensionalised with the total loss in case of smooth blades.
Presence of roughness over the different part of blade
affects the secondary loss differently. Secondary loss
increases in case of PSR blade cascades compared to smooth
blade.
The secondary loss further increases slightly if the
roughness over pressure surface is increased from 250 to 500
µm. Roughness on pressure surfaces of PSR cascade makes
boundary layer thicker and thus strengthen the pressure side
vortex and hence passage vortex due to which the secondary
loss is increased. The Pressure side leg of horse shoe vortex
is affected by curvature and pressure difference between
pressure side and suction side of adjacent blade of the
corresponding flow passage thereby contribute more to the
Secondary losses. It can be concluded that presence of
roughness over pressure surface increases secondary losses.
This may be due to favorable effect of pressure side leg on
the passage vortex. This is in line with the energy loss
787
Technology and Advanced
2459, ISO 9001:2008 Certified Journal,
coefficient where roughness over pressure surf ace is least
detrimental. On the contrary, the magnitude of secondary
loss decreases in case of BSR & SSR cascades with increase
in roughness from 250 to 500 µm. The secondary losses
reduces to 2.94, 2.75 and 2.46% respectively when
roughness of 250 µm, 500 µm and 750 µm applied over
suction surface. This may be due to strengthening of suction
side counter vortex, which mixes with the passage vortex and
reducing the effect of passage vortex. Therefore, the
secondary losses are lower in this case. Increase in roughness
further enhance counter vortex and hence secondary losses
are further reduced. Profile loss due to roughness over
suction surface also favors these results. Magnitude of
secondary loss at roughness value of 1000 µm was found to
be approximately equal to that of 750 µm for all types of
cascades i.e. PSR, SSR & BSR.
Table 2 shows that profile loss for BSR cascade is
maximum followed by SSR and PSR. In smooth type of
cascade profile loss is happened to be minimum. It was
observed that the mass averaged secondary loss as
percentage of total loss in smooth blade cascade was 4.94%.
When 250 µm roughness was present on both surfaces
together this loss was 4.65%. The secondary loss further
reduces to 4.37% when roughness over both the surfaces is
increased to 500 µm. However, the roughness of 250 µm
applied separately over pressure and suction surfaces results
in secondary loss of 5.41% and 2.94%, respectively. Thus, it
is clear that roughness over pressure surface strengthen the
passage vortex, whereas suction side roughness weakens the
passage vortex by stronger suction side counter vortex.
When the roughness over pressure and suction surface
applied together, the effect of counter vortex is more
pronounced, thus causing secondary loss to decrease in this
case.
The effect of different levels of roughnesses (250, 500,
750 &1000 µm) over various surfaces of blades on the profile
and secondary loss are summarized in Table 2.
3.3 Contour plots of total pressure distribution:
Total pressure distribution for BSR cascade over entire
computational domain very close to end wall and at mid span
are shown in Figure 7 and Figure 8. After entering cascade
section total pressure drops due to viscous effect and other
losses over the cascade section. At exit of cascade wakes are
formed where total pressure drops significantly. However in
core flow region, pressure drop is insignificant. At
significant distance from trailing edge intermixing of core
International Journal of Emerging Technology and Advanced
Engineering
Website: www.ijetae.com (ISSN 2250- 2459, ISO 9001:2008 Certified Journal,
Volume 3, Issue 8, August 2013)

flow and wake takes place. At very close to end wall region gi Gravitational Body Force
wake bands are much broader & are much more diffused at
very exit from blade trailing edge because of end wall
boundary layer interaction as seen in Figure 7. Whereas, Fi External Body Force
wakes are very distinct at mid span, Figure 8.
ij Stress Tensor
IV. CONCLUSIONS
• The pattern of variation of energy loss coefficient in Keff Effective Thermal Conductivity
spanwise direction (y/S) is same for smooth as well as
rough blades. Moreover the Figure 6 shows that energy Jj‘ Diffusion Flux
loss coefficient is least for smooth blades and it reaches
the maximum value in case the roughness is introduced Sh Source term includes heat of chemical reaction
on pressure as well as suction surface together.
• It is observed that applying roughness on blade surface T Temperature
definitely increases the profile loss as well as total
energy loss coefficient. But the mass averaged E Energy term
secondary loss is lower than the secondary loss in a
smooth cascade in case of SSR & BSR and higher in h Enthalpy
case of PSR compared to smooth blade case.
• SSR has least secondary losses may be because of more P2s Static pressure at outlet
flow separation of boundary layer and thus mixing of
passage vortex with suction side counter vortex takes Po1 Total pressure at inlet
place resulting into least secondary losses.
• Secondary Loss in BSR is more than SSR because the Po2 Total pressure at outlet
roughness over pressure surface promotes passage
vortex, whereas roughness over suction surface Ratio of specific heats for air
strengthen counter vortex present as suction surface
leg. Therefore, we can say that secondary losses in the
y Local energy loss coefficient
case of pressure surface are more that of suction surface
for the same roughness level.

• Due to the end wall boundary layers the loss coefficient
is high at hub and casing.
• Humps are observed near the hub and casing due to
secondary vortex which increases the local energy loss
coefficient.
NOMENCLATURE
REFRENCES
[1 ] Brear Michael J., Hodson Howard P., Gonzalez Palom and Harvey
Neil W., 2002, ―Pressure Surface Separations in Low-Pressure
Turbines—Part 2: Interactions With the Secondary Flow,‖ Transactions
of the ASME, 124, 402-409.
[2 ] Korakianitis T. and Hamakhan I. A., 2010, ―Aerodynamic
Performance Effects of Leading-Edge Geometry in Gas-Turbine
Blades,‖ Applied Energy, 87,1591–1601.

Density [3 ] Langston L. S., Nice M.L. and Hooper R.M., 1977,

―ThreeDimensional Flow within a Turbine Cascade Passage, " ASME
Journal of Engineering for Power, 99, 21–28.
ui Velocity vector [4 ] Lei Qi, ZhengpingZou, Peng Wang, Teng Cao and Huoxing Liu,
2011, ―Control of Secondary Flow Loss in Turbine Cascade by
Sm Momentum Source Term Streamwise Vortex ,‖ Computers & Fluids, 54, 45-55.
[5 ] Mahmood, G.I. and Acharya S., 2007, ―Experimental Investigation
P Static Pressure of Secondary Flow Structure in a Blade Passage With and Without

788
International Journal of Emerging
Engineering
Website: www.ijetae.com (ISSN 2250-
Volume 3, Issue 8, August 2013)
Leading Edge Fillets,‖ ASME Journal of Fluids Engineering, 129, 253-
262.
[6 ] Marchal P., and Sieverding C.H., 1977, ―Secondary Flows Within
TurbomachineryBladings,‖ Secondary Flows in Turbomachines,‖
AGARD-CP-214, 11, 1–19.
[7 ] Moon, Young J. and Koh Sung-Ryong, 2000, ―Counter-rotating
streamwise vortex formation in turbine cascade with endwall fencing,‖
Computers And Fluids, 30, 473-490.
[8 ] SahaArun K. and AcharyaSumanta, 2008, ―Computations of
Turbulent Flow and Heat Transfer Through a Three-Dimensional
Nonaxisymmetric Blade Passage,‖ ASME Journal of
Turbomachinery, 130, 031008-1- 031008-10.
[9 ] Samsher, 2002, ―Effect Of Blade Surface Roughness on Profile Loss
and Exit Angle in A Rectilinear Steam Turbine Cascade,‖ Ph.D. Thesis,
Mech. Engg. Dept. IIT Delhi.
[10 ] Samsher, 2007, ―Effect Of localized roughness over reaction &
impulse Blades on loss coefficient‖ Journal of Power and energy, part
A IMechE, U.K. 221, 21-32
[11 ] Sharma O.P. and Butler T.L, 1987, ―Predictions of Endwall Losses
and Secondary Flows in Axial Flow Turbine Cascades,‖ ASME Journal
of Turbo machinery, 109, 229-236.
[12 ] Shih T. I-P. and Lin Y.L, 2003, ―Controlling Secondary-Flow
Structure by Leading-Edge Airfoil Fillet and Inlet Swirl to Reduce
Aerodynamic Loss and Surface Heat Transfer,‖ Transactions of the
ASME, 125, 48-56.
789
Technology and Advanced
2459, ISO 9001:2008 Certified Journal,
[13 ] Sieverding C.H and, Bosch P. Van den, 1983, ―The Use of Coloured
Smoke to Visualise Secondary Flows in a Turbine-Blade Cascade, "
ASME Journal of Fluid Mechanics, 134,85-89.
[14 ] SonodaToyotaka, Hasenjäger Martina, Arima Toshiyuki and Sendhoff
Bernhard, 2009, ―Effect of End Wall Contouring on Performance of
Ultra-Low Aspect Ratio Transonic Turbine Inlet
Guide Vanes,‖ ASME Journal of Turbomachinery, 131, 011020-1-
011020-11.
[15 ] Vinod Kumar Singoria, Deepika Sharma, Samsher, 2012 ―Effect of
roughness on secondary flow in a rectilinear turbine‖ proceedings of
conference organized by YMCA University of science & technology,
Faridabad, Haryana, India, 132-141.
[16 ] Wang H.P., Olson S.J., Goldstein R.J., Eckert E.R.G., 1997, ―Flow
visualization in a linear turbine cascade of high performance turbine
blades,‖ ASME Journal of Turbomachinery, 119, 1-8.
[17 ] Yahya, S. M., 1983, Turbines, compressors and fans, Tata
McGrawHill, New- Delhi.
Figures and graphs
[18 ] Vinod Kumar Singoria, Samsher, 2013, ―Mechanism,
Characterization, Pattern and Effect of Roughness over Turbine
Blade: A Review‖ International Journal of Engineering and Innovative
Technology (IJEIT), 2, 191-200.
International Journal of Emerging Technology and Advanced
Engineering
Website: www.ijetae.com (ISSN 2250- 2459, ISO 9001:2008 Certified Journal,
Volume 3, Issue 8, August 2013)

790
International Journal of Emerging Technology and Advanced
Engineering
Website: www.ijetae.com (ISSN 2250- 2459, ISO 9001:2008 Certified Journal,
Volume 3, Issue 8, August 2013)

791
International Journal of Emerging Technology and Advanced
Engineering
Website: www.ijetae.com (ISSN 2250- 2459, ISO 9001:2008 Certified Journal,
Volume 3, Issue 8, August 2013)

Figure 3: schematic diagram and specifications of turbine blade 6030 cascade

792
International Journal of Emerging Technology and Advanced
Engineering
Website: www.ijetae.com (ISSN 2250- 2459, ISO 9001:2008 Certified Journal,
Volume 3, Issue 8, August 2013)

Figure 4: Measurement plane at 15% of the chord

793
International Journal of Emerging Technology and Advanced
Engineering
Website: www.ijetae.com (ISSN 2250- 2459, ISO 9001:2008 Certified Journal,
Volume 3, Issue 8, August 2013)

794
International Journal of Emerging Technology and Advanced
Engineering
Website: www.ijetae.com (ISSN 2250- 2459, ISO 9001:2008 Certified Journal,
Volume 3, Issue 8, August 2013)

795
International Journal of Emerging Technology and Advanced
Engineering
Website: www.ijetae.com (ISSN 2250- 2459, ISO 9001:2008 Certified Journal,
Volume 3, Issue 8, August 2013)

Figure 7: Total pressure distributions in wake region very close to end wall o f BSR cascade having Roughness of 500 𝞵m.

Figure 8: Total pressure distributions in wake region at mid span of BSR cascade having Roughness of 500 m.

Table 1

796
International Journal of Emerging Technology and Advanced
Engineering
Website: www.ijetae.com (ISSN 2250- 2459, ISO 9001:2008 Certified Journal,
Volume 3, Issue 8, August 2013)

Non dimensional distance on blade span

S.N Distance from Non Dimensional S.N. Distance from Non Dimensional
bottom wall mm) Distance bottom wall Distance
(mm)
1 14 50
0 0 0.53
2 2 0.02 15 55 0.58
3 4 0.04 16 60 0.63
4 6 0.06 17 65 0.68
5 8 0.08 18 70 0.74
6 10 0.11 19 75 0.79
7 15 0.16 20 80 0.84
8 20 0.21 21 90 0.95
9 25 0.26 22 87 0.92
10 30 0.32 23 89 0.94
11 35 0.37 24 91 0.96
12 40 0.42 25 93 0.98
13 45 0.47 26 95 1

797
International Journal of Emerging Technology and Advanced
Engineering
Website: www.ijetae.com (ISSN 2250- 2459, ISO 9001:2008 Certified Journal,
Volume 3, Issue 8, August 2013)

Table 2
Summary of Total Energy Loss, Profile Loss and Secondary Loss for smooth, SSR, PSR & BSR cascades for different levels of roughness
Location Roughness Percentage of Percentage Contribution percentage of Second
roughness osecondary loss
levels Total energy loss Profile loss in t loss relative to total l
in total energy of energy of the air (C=A
in smooth blades
( m) the air (A) (mid span value)
(D=100*C/A)

Smooth 10.53 10.01 0.52 4.94

PSR 250 11.71 11.14 0.57 5.41

500 12.24 11.61 0.63 5.98

750 12.33 11.73 0.61 5.79

1000 12.33 11.73 0.61 5.79

SSR 250 15.97 15.66 0.31 2.94

500 17.78 17.49 0.29 2.75

750 17.96 17.7 0.26 2.47

798
International Journal of Emerging Technology and Advanced
Engineering
Website: www.ijetae.com (ISSN 2250- 2459, ISO 9001:2008 Certified Journal,
Volume 3, Issue 8, August 2013)

1000 17.96 17.7 0.26 2.47

BSR 250 17.10 16.61 0.49 4.65

500 19.42 18.95 0.46 4.37

750 19.72 19.25 0.46 4.37

1000 19.72 19.25 0.46 4.37

799