Propulsion and Power Research 2016;5(3):236249

HOSTED BY http://ppr.buaa.edu.cn/

Propulsion and Power Research

www.sciencedirect.com

ORIGINAL ARTICLE

Stability management of high speed axial ow

compressor stage through axial extensions of bend
skewed casing treatment
Dilipkumar Bhanudasji Alonea,n, S. Satish Kumara, Shobhavathy M. Thimmaiaha,
Janaki Rami Reddy Mudipallia, A.M. Pradeepb, Srinivasan Ramamurthyc,
Venkat S. Iyengara

a
Propulsion Division, CSIR-NAL, Bangalore 560037, India
b
Aerospace Engineering Department, IIT Bombay, Powai 400076, India
c
NCAD, CSIR-NAL, Bangalore 560017, India

Received 24 November 2014; accepted 5 January 2016

Available online 12 September 2016

KEYWORDS Abstract This paper presents the experimental results to understand the performance of
moderately loaded high speed single stage transonic axial ow compressor subjected to various
High speed
compressor; congurations of axial extensions of bend skewed casing treatment with moderate porosity. The
Bend skewed casing bend skewed casing treatment of 33% porosity was coupled with rectangular plenum chamber of
treatments; depth equal to the slots depth. The ve axial extensions of 20%, 40%, 60%, 80% and 100% were
Axial extensions; used for the experimental evaluations of compressor performance. The main objective was to
Stall margin improve- identify the optimum extension of the casing treatment with reference to rotor leading edge which
ments; results in maximum stall margin improvements with minimum loss in the stage efciency. At each
Peak stage efciency; axial extension the compressor performance is distinctive. The improvement in the stall margin was
Hot wire probe; very signicant at some axial extensions with 4%5% penalty in the stage efciency. The
Stall; compressors stage shows recovery in terms of efciency at lower axial extensions of 20% and 40%
Abrupt with increase in the peak stage efciency. Measurements of ow parameters showed the typical
behaviors at near stall ow conditions. Hot wire sensor was placed at the rotor upstream in the tip
region to capture the oscillations in the inlet axial and tangential velocities at stall conditions. In the
absence of casing treatment the compressor exhibit abrupt stall with very high oscillations in the

n
Corresponding author. Tel.: +9180 25051527.
E-mail address: dilipbalone@nal.res.in (Dilipkumar Bhanudasji Alone).
Peer review under responsibility of National Laboratory for Aeronautics
and Astronautics, China.

2212-540X & 2016 National Laboratory for Aeronautics and Astronautics. Production and hosting by Elsevier B.V. This is an open access article under the
CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

http://dx.doi.org/10.1016/j.jppr.2016.01.009
Stability management of high speed axial ow compressor stage through axial extensions of bend skewed casing treatment
inlet axial and tangential velocity of the ow. The extents of oscillations reduce with bend skewed
casing treatment. Few measurements were also performed in the plenum chamber and salient
results are presented in this paper.
& 2016 National Laboratory for Aeronautics and Astronautics. Production and hosting by Elsevier B.V.
1. Introduction
High speed axial ow compressors are integral parts of
aero-engines because of their ability to produce relatively
higher pressure ratio per stage with reasonably good
efciency. The main drawback of these machines are
relatively lower operating range because of the instabilities
like rotating stall and surge in the absence of control tools.
These instabilities are normally triggered at off-design
conditions and will be catastrophic in nature if they are
not controlled and allow growing. Bend skewed casing
treatment with 33% porosity coupled with plenum chamber
is designed and used for evaluating the performance of
single stage axial ow compressor with stage total pressure
ratio of 1.36 at 12,930 rpm, corrected speed was used for
the experimental analysis. Casing treatments is the powerful
old techniques used for controlling the instabilities in tip
strong rotors. Researcher across the globe [124], designed
several types of casing treatment ranging from simple
honeycomb, circumferential grooved, axial grooved and
recently bends skewed casing treatments to improve the
stall margin of the compressor. However, the criterion for
the selection of casing treatments for the tip critical rotors
was proposed given by Greitzer et al. [1]. Venzel and Moss
[2] studied the effect of circumferential grooved and slotted
casing treatment on the performance of multistage com-
pressors. They found the stall pressure ratio was slightly
lower than the baseline and the overall compressor ef-
ciency with the grooved casing was not appreciably differs
from the baseline, but it was 1 to 2 points lower with the
slotted casing. In case of radial tip distortion stall pressure
ratio was slightly improves with both the casing treatments.
The casing treatment used was not effective in extending
the stall margin of the compressor. Donald Urasek [3]
studied the effect of circumferential grooved casing treat-
ment on the performance of two stage high pressure ratio
fan and found stall margin improvement without affecting
the efciency. Slotted casing was more effective in enhan-
cing the stall margin of the compressor at the cost of
efciency penalty which is conrmed by Guruprasad [5]
through experimental results. He also conrmed that higher
stall margin gains with minimum efciency penalty can be
achieved by optimizing various geometric parameters
of casing treatments. Parametric studies on the axial slot
casing treatment was carried out by many researcher [612]
to understand its potential benets numerically and experi-
mentally on low speed as well as high speed compressors.
237
This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Almost everyone conrm the benets of this type of casing
treatment but at the cost of degradation in the efciency.
Axial skewed slotted casing treatment was then modied to
give birth to another very interesting casing treatment called
bend skewed casing treatment by Junqiang Zhu et al. [12
16]. This type of casing treatment has a combinations of
axial as well as bend slots. Experimental and numerical
results of an isolated rotor showed considerable improve-
ment in the stall margin without efciency penalty. The
axial skewed casing treatment improves the stall margin by
27.66% and degrades the compressor efciency by 3%
whereas the bend skewed groove achieves 18.09%
improvement with negligible loss in the compressor ef-
ciency at lower axial extensions. The bend skewed casing
treatment weakens the circulation in the tip region because
of the rear skewed slot. Gourdain et al. [17], Schnell et al.
[18,19], Voges et al. [20] studied bend skewed casing
treatment numerically and experimentally and conrm the
positive benet of this type of casing treatment in enhan-
cing the stall margin by periodical injection of energized
uid out of the casing slots and stabilizing the tip clearance
vortex.
From the above literatures it is clear that the bend skewed
casing treatment has a potential to improve the stall margin
with minimum penalty on the efciency. However, the
research does not give the substantial understanding of
behavior of this type of casing treatment with plenum
chamber in the high speed compressor stage environment.
Experimental data presented by Junqiang Zhu et al. [1216]
is on subsonic compressor with the isolated rotor environ-
ment, in which effect of presence of stator will not be
highlighted. The rotor stator gap also plays signicant role
with casing treatment retrotted above the rotor as reported
by Gourdain et al. [17]. Also, the casing treatment geometry
used by Junqiang Zhu is having higher porosity of 67%.
There is a gap in the research to understand the behavior of
lower porosity bend skewed casing treatment in the transonic
environment which is very close to the real engine applica-
tions. This fact motivated us to take the rigorous experi-
mental research on the lower porosity bend skewed casing
treatment and understand the compressor stage performance
with different geometrical parameters. Dilipkumar Alone
et al. [2124], generated exhaustive experimental data to
understand off-design unsteady behavior of the transonic
compressor stage and also evaluated an impact of various
geometrical parameters of bend skewed casing treatments
like porosity, axial locations, and an un-explored parameter
238 Dilipkumar Bhanudasji Alone et al.

Nomenclature Greek letters

C axial chord at rotor tip (unit: mm) temperature ratio, E TTsealevel

atm
, Tsealevel E 288.15 K
CT casing treatment pressure ratio, E Psealevel , Psealevel E 1.01325 bar
Patm

CF choke ow condition efciency

DC direct current difference
FFT fast Fourier transform v isentropic index, E 1.4
m mass ow rate (unit: kg/s)
m corrected corrected mas p ow rate (unit: kg/s) Subscripts
m corrected m
N rotor rotation (unit: rpm) SC smooth casing
Ncorrected corrected rotation (unit: rpm) CT casing treatment
Ncorrected pN t tip radius
NSF near stall ow condition peak peak or maximum
Pr stage total pressure ratio atm atmospheric
R local blade radius (unit: mm) s static condition
RMS root mean square 0 total condition
SF stall ow condition 1 rotor inlet condition
SC smooth casing 2 rotor exit condition
TI turbulence intensity (unit: %) 3 stator exit condition
V axial velocity (unit: m/sec)
0.2C 20% rotor axial chord coverage CSIR Council of Scientic and Industrial Research
0.4C 40% rotor axial chord coverage NAL National Aerospace Laboratories
0.6C 60% rotor axial chord coverage NCAD National Civil Aircraft Development
0.8C 80% rotor axial chord coverage
1C 100% rotor axial chord coverage

plenum chamber on the transonic axial compressor. From the
experimental data it is conrmed that even with lower
porosity bend skewed casing treatment at optimum axial
coverage in combinations of plenum chamber it is possible to
overcome the drawback of conventional casing treatment in
improving the stall margin at the cost of efciency penalty.
An experimental result shows signicant improvement in
stall margin with increase in compressor stage peak ef-
ciency. This paper presents interesting results to understand
the effects of axial extensions of bend skewed casing
treatment coupled with plenum chamber on the performance
of single stage high speed axial ow compressors. Experi-
mental results presented here are steady as well as unsteady
state using hot wire anemometers. In addition, this paper
present few interesting measurements in the plenum chamber
located above the casing treatments slots using hot wire
probe. In addition this is rst ever attempt to understand the
ow behavior in the plenum chamber of the bend skewed
casing treatment. The presence of plenum chamber and
unsteady measurements in the plenum chamber bring
uniqueness and novelty to this investigation compared to
other paper reported in the literatures.
stage compressor is powered by DC motor with power rating
of 1.15 MW@1000 rpm. The power to the compressor stage
was fed through 1:18 step up high speed gear box and
electronic Torquemeter. Motor speed was precisely con-
trolled using feedback controller. Compressor swallows the
air mass ow through the ow metering inlet Bellmouth; get
compressed in the rotor and discharged to the atmosphere
through volute casing and 18 m long and 2 feet diameter
exhaust ducting. Venturimeter located at the downstream of
exhaust ducting is also used to measure the air mass ow
rate. Buttery exit throttle valve provided at the exit of the
ducting was used for controlling the airow rate through the
compressor stage for generating the performance map. In
addition to main buttery valve small bypass ow control
valve was used for ne control of the mass ow near stall.
Solenoid actuated emergency surge relief valve was used to
prevent the compressors going into the instabilities.
Table 1 Specications of the compressor stage.
Stage Single transonic

Stage pressure ratio 1.36

2. Test facility descriptions Corrected mass ow rate 23 kg/s
Corrected speed 12930 rpm
Rotor blade 21 (transonic)
Experimental analysis was carried out in open circuit Stator blade 18 (subsonic)
single stage moderately loaded transonic axial ow com- Tip relative Mach number 1.15
pressor facility of the CSIR-NAL, Propulsion Division; Tip diameter 450 mm
India. The test facility has an overhang rotor. The single
Stability management of high speed axial ow compressor stage through axial extensions of bend skewed casing treatment
Figure 1 Photograph of the experimental test set up.
Figure 2 Bend skewed casing treatment.
Compressor stage pressure ratio was measured using four
total pressure rakes placed at the stator exit and one total
pressure rake placed at the rotor inlet. Total pressure rake
cover the entire annulus height of the compressor at the
inlet and at the stator exit. Stator exit rake have 15 points
each, whereas inlet rake has 21 total pressure points.
Experimental test rig is equipped with circumferentially
mounted static pressure taps for the wall static pressure
measurements at rotor inlet, rotor exit and at the stator exit.
K-type pre-calibrated thermocouples were used for the ow
temperature measurements at the rotor inlet and at the stator
exit. High speed electronic torquemeter was used for
measuring the power input to the compressor stage. Single
component hot wire probe was used for detecting the
instabilities and for measuring the uctuations in the inlet
axial and tangential velocities at stall mass ow rate. The
hot wire probe was calibrated using hotwire probe cali-
brator. Three hole aerodynamic probes was used for
mapping the variations of the various ow parameters at
the rotor exit at choke ow and near stall ow conditions.
The comparative variations of these ow parameters are
studied along the radial direction for the smooth casing and
four axial extensions of the bend skewed casing treatments.
The ow parameters show very distinct behavior in the tip
regions for all the casing treatment and smooth casing. The
239
Figure 3 Casing treatment retrotted above the rotor at 20%, 40%
and 100% axial extensions. (a) 20% axial extension, (b) 40% axial
extension, and (c) 100% axial extension.
probe movement was precisely controlled near the casing
and hub wall using probe actuators and DC stepper motor.
All the experiments were conducted at the same time
early morning to ensure clean, dust free and similar ambient
conditions. During this time other facilities like wind
tunnels and high speed test rigs were not in operational.
This ensures the precise acquiring of the steady and
particularly noise free unsteady data. The sound levels
were within 45 dB in the control room which is tted with
acoustic liners wherein all the electronics, data acquisition
system and rig control is located.
Technical specications of the compressor test facility
are given in Table 1.
The photograph of the compressor test facility is shown
in Figure 1.
3. Bend skewed casing treatment descriptions
Casing treatment designed for an experimental program
is bend skewed type. It has equi-spaced 105 bend slots with
uniform width which results in porosity of 33%. The slot
width was kept equal to maximum thickness of the rotor
blade. Bend skewed casing treatments slots are having
240 Dilipkumar Bhanudasji Alone et al.

Figure 4 Compressor performance map with solid casing and different axial extensions of bend skewed casing treatments. (a) Stage efciency
against corrected mass ow rate and (b) stage total pressure ratio against corrected mass ow rate.

frontal axial section and then 451 skewed rear section. Both
the slots were inclined by 451 in the radial direction, such a
way that ow emerging from the treatments slots is in
counter-clockwise direction. The axial length of the casing
treatment is kept equal to axial chord of the rotor tip. The
depth of the slots is kept in such a way to get depth to slot
width ratio of 2.9. Casing treatment was coupled with the
plenum chamber of equal depth above the casing slots.
Depth of the groove and plenum chamber was 11 mm.
Schematic of the casing treatment ring is shown in Figure 2.
Figure 3 shows the casing treatment and plenum chamber
ring retrotted above the rotor at 20%, 40% and 100% axial
extensions. Figure 5 Change in stalling mass ow rate.

smooth casing. Based on the compressor performance

4. Results and discussions
This section is focus on the experimental results showing
the sensitivity of the compressor stage to the various axial
extensions of the bend skewed casing treatment. The
compressor performance is discussed in the form of
compressor operating envelop and compared with the
map stall margin in terms of displace of stall mass ow
rate to lower side is also evaluated. Followed by overall
performance detailed variations of the rotor exit ow
parameters will be discussed. Lastly behavior of the inlet
axial and tangential velocities, plenum chamber velocity
measured by hot wire will be presented and discussed. All
the experimental measurements were carried out with
Stability management of high speed axial ow compressor stage through axial extensions of bend skewed casing treatment
Figure 6 Change in peak stage pressure ratio.
Figure 7 Change in peak stage efciency.
measurement accuracy of 71%. Flow angle measurement
using aerodynamic probe were carried out within 71/21
measurement accuracy.
4.1. Compressor performance map
Compressor performance map is generated to estimate
the stable operating range of the compressors stage with
smooth casing and casing treatment for ve different axial
extensions. Compressor stage was operated at different
constant speed lines and mass ow rate was precisely
controlled using exit throttle valve till the instability occurs.
Mass ow rate is corrected using sea level and local
ambient conditions for plotting.
Performance map is plotted in terms of stage total
pressure ratio and efciency against corrected mass ow
rate for different operating speeds. Figure 4 shows the
performance map of the compressor stage. From the
performance map it is clear that with all the casing
treatment axial extensions compressor stage stalls at lower
mass ow rates compared to smooth casing except for the
100% axial coverage and at 80% design speed. Percentage
displacement in the mass ow stalling mass ow rate is
241
calculated using Eq. (1). The compressor stage efciency is
evaluated using the compressor power output and power
input measured using torquemeter using Eq. (2). Percentage
reduction in the stalling mass ow rate with casing
treatments is determine the stall margin improvement.
Percentage change in the peak stage pressure ratio and
peak stage efciency is also calculated using Eqs. (3) and
(4) respectively.
From the performance map shown in Figure 4, it is
observed that in the presence of smooth casing peak stage
total pressure ratio of the compressor stage increases from
1.09 to 1.242 with increase in the operating speed from
50% to 80%. However the peak stage efciency decreases
from 86% at 50% speed to 82.2% at 80% design speed. In
the presence of smooth casing the reduction in the peak
stage efciency is very signicant with increase in the
operating speed. This reduction in the efciency is the
results of higher losses due to higher viscous effects which
normally increased with the tip speed. Compressor stage
stalled at higher mass ow rate with smooth casing. Stable
operating range of the compressor stage with smooth casing
estimated is 22% at 50% speed which further reduces to
17% at higher operating speed of 80%.
Figures 57 show the variations of stalling mass ow
rate, stage peak pressure ratio and efciency gain/loss with
casing treatments.
Change in mass flow rate %
h i
1
m Corrected  Stall m Correctedm Corrected
SC  m Corrected  CT
 CT
 100
Compressor stage efficiency %
h 1
i
mCpT 01 Pr
1 2
Compressor power output
 100
Compressor power input PInputTorquemeter
Change in peak pressure ratio %
h i
Pr  Pr 3
Prpeak peak PrCTpeak  SCpeak  SC  100
Change in peak stage efficiency %
4
peak peak  CT  peak  SC
From Figure 5 it is clear that with 100% axial extension
with increase in the operating speeds compressor stalls at
relatively higher mass ow rate and thus less effective in
increasing the stable operating range of compressor. At 80%
speed it results in negative performance and stalls at higher
mass ow rate than the smooth casing. Drop in peak pressure
ratio also increases with speeds and can be seen from Figure 6.
This loss in total pressure further reduces the peak stage
efciency. Penalty in the efciency is higher at lower speed but
compressor shows the improvement at higher speed by
minimizing the losses associated with the viscous effects at
higher operating speeds and can be seen from Figure 7. At
80% axial extension and at all the operating speeds except
70% compressor stalled at lower mass ow rate and results in
signicant gain in the stall margin, but again in terms of
242 Dilipkumar Bhanudasji Alone et al.

pressure loss and efciency penalty. This extension at 70%

speed produces lowest stall margin of 3% with highest penalty
of 7% in the peak stage efciency. At 60% extensions
compressor stage shows some sign of recovery in the
performance with signicant improvement in the stall margin
and minimal reduction in the stage efciency at 70% speed.
Maximum improvement of 32% in the stall margin is at 50%
speed with 3% drop in the peak stage efciency. At 60%
operating speeds it results in 30% improvement and main-
tained the same efciency of that with smooth casing. Beyond
70% speed compressor stage shows recovery in terms of stable
operating range, pressure recovery and efciency gain, with
9.43% improvement in stalling mass ow rate at 1.6% Figure 9 Rotor exit axial Mach number.
improvement in the peak stage efciency.
At lower axial extensions of 40% and 20% compressor
stage shows tremendous improvements in its performance. At
20% axial extension stalling mass ow rate gradually increases
with increase in the operating speed and thus results in slightly
lower stall margin improvement at higher operating speeds. At
this coverage and at 50% speed the reduction in the stall mass
ow rate is 31.15% compared to 10.48% at 80% design speed.
Compressor stage shows recovery in the stage pressure ratio
which is increases with increase in the speed. As a result, at
20% axial coverage compressor stage shows gain in the peak
stage efciency over entire speed range. The improvement in
the stage efciency is 3.5%. At 40% axial extension peak stage
pressure ratio decreases marginally from 60% speed onwards
with maximum drop at 80% speed. Even though the pressure
ratio decreases it always maintained above the smooth casing
at all the speeds. Decrease in the stage pressure ratio causes Figure 10 Rotor exit total pressure ratio.
marginally lower peak stage efciency at 40% extension
compared to 20% extension. This shows that with casing
treatment retrotted at lower axial extensions of 20% and 40%,
compressor stage allow stable operation at higher incidence
and thus results in slightly higher pressure ratio compared to
other extensions and solid casing.

4.2. Rotor exit ow parameters

Rotor exit ow parameters are evaluated at the choke

ow and near stall ow conditions for the smooth casing

Figure 11 Rotor exit static to total pressure ratio.

and all the extensions of casing treatment at 80% design

Figure 8 Rotor exit absolute ow angle.
speed. Their comparative behavior is studied along the
span-wise direction. The variations of these ow parameters
are discussed at near stall ow condition only. All these
ow parameters are evaluated by using aerodynamic probe
in non-nulling mode. Probe was traversed with automatic
probe traverse mechanisms with ne control of the probe
movement in the wall boundary layers. The parameters are
Stability management of high speed axial ow compressor stage through axial extensions of bend skewed casing treatment 243

discussed here are rotor exit absolute ow angle, axial
Mach number, total and static pressure ratio.
4.2.1. Rotor exit absolute ow angle and axial Mach number
Figure 8 shows the variations of rotor exit absolute ow
angle. A variation is plotted along the blade height as non-
dimensional radial height (R/Rt). Absolute ow angle gives the
degrees of the variation of absolute swirl imparted by the rotor
at the rotor exit. Prole shows gradual decrements in the swirl
as a result of lower blade curvature along the blade span till
90% of the blade span. It is also functions of axial velocity
which is increases till 90% blade span as shown in Figure 9. In
the tip region swirl again increases due to the boundary layer
effects and decrease in the axial velocity caused by ow
retardations close to the wall.
The variation in the ow angle from rotor hub to tip is
signicant at near stall conditions because of closure of exit
throttle valve. Closure of exit valve reduces the mass ow rate
through the compressor stage which reduces the axial Mach
number, which further cause increase in the inlet incidence.
The increase in the incidence further increases the ow
deviations at the rotor exit and thus increase the ow angle.
For the smooth casing ow angle increases up to 521 in the tip
whereas it was 481 in the hub section. Flow angle increases
with the casing treatment extensions as results of reductions in
the axial velocity and can be seen from Figure 9.
At 40% axial extension the ow angle is more because of
lower stalling mass ow rate as conrmed from Figure 5.
This retardation is very rapid in the tip zone. Flow angle
measured in the tip region with this coverage is close to 601.
Another interesting observations is that even though the
100% axial coverage results in lower stall margin improve-
ments, probe measurements shows the ow retardations
caused by the slots throughout the blade span, due to more
exposure of the casing slots and plenum to the rotor. The
performance of 60% and 80% extensions falls in between
40% and 100% axial coverage.
4.2.2. Rotor exit total pressure and static to total ratio
Figures 10 and 11 show the comparative variations of rotor
exit total pressure and static to total pressure ratio respectively
at near stall ow condition. Total pressure prole is skewed in
the tip region and skewness is increased with casing treatment
compared to smooth casing. Total and static, both the pressure

Figure 12 Comparative variations of rotor inlet axial and tangential velocity for the smooth casing at stall condition. (a) Variations of rotor inlet
axial and tangential velocities, (b) variations of rotor inlet axial and tangential velocities for six rotor revolutions, (c) FFT rotor inlet axial and
tangential velocities, and (d) exploded view of FFT showing the dominating stalls frequencies.
244 Dilipkumar Bhanudasji Alone et al.

increases toward the tip till 90% span, beyond which it start
drooping because of casing wall effects. Increase in the total
and static pressure in the span wise directions is due to more
work imparted by the dynamic action of the compressor rotor
because of higher local peripheral speed. Smooth casing shows
marginally higher pressure ratio throughout the blade height as
with smooth casing compressor stalls at higher mass ow rate
compared to all the extensions of casing treatments. However
for all the casing treatment extensions static pressure is higher
than the smooth casing as with casing treatments compressor
stage stalls at a lower mass ow rate and thus improves the
diffusion and can be observe from Figure 11. Diffusion is
highest in the tip region with casing treatments.
4.3. Rotor inlet axial and tangential velocities for
smooth casing and casing treatment at stall condition
During the stall conditions the inlet velocity uctuates in
the axial and tangential directions. The uctuations are
predominant in the rotor inlet and at the tip section.
Behavior of the axial and tangential velocities is studied
using single component hot wire probe for the smooth
casing and casing treatments. In stall, behavior of the axial
velocity is studied for four axial extensions of the casing
treatment and tangential velocity is studied for the two axial
extensions.

Figure 13 Comparative variations of rotor inlet axial velocity for the smooth casing and casing treatment extensions at stall ow conditions. (a)
Variations of rotor inlet axial velocity for the smooth casing and four axial extensions of bend skewed casing treatment, (b) variations of rotor inlet
axial velocity for the smooth casing and four axial extensions of bend skewed casing treatment for the six rotor revolutions, (c) FFT of rotor inlet
axial velocity for the smooth casing and four axial extensions of bend skewed casing treatment, and (d) exploded view of FFT showing dominating
stall frequencies.
Stability management of high speed axial ow compressor stage through axial extensions of bend skewed casing treatment 245

4.3.1. Smooth casing
When the compressor stage is tted with smooth casing it
stalls at relatively higher mass ow rate compare to casing
treatments. The compressor stage shows abrupt nature of
the stall as after the peak pressure ratio even smaller closure
of the exit throttle valve, compressor enters into the instable
mode. The behavior of the inlet axial velocity is studied and
compared with the other axial extensions. The unsteady
data plotted here for the 60% design speed due to the
limitations of hot wire sensor. Figure 12 shows the behavior
of instantaneous inlet axial and tangential velocities mea-
sured at the rotor inlet in the tip region, with the help of
single component hot wire probe. Unsteady data is acquired
with sampling frequency of 50 kHz. From Figure 12(a) and
(b) is very clear that the oscillations in the axial and
tangential velocities are very rapid and high for the smooth
casing and also can be conrm from the FFT plots shown in
Figure 12(c) and (d). The inlet axial velocity varies from
188 m/s to 10 m/s very rapidly and thus indicate high
intensity of stall. The tangential velocity also present and
show the signifying dominances. Its magnitude almost
reaches the tangential speed of the rotor and few occasions
it matches with the magnitude of axial velocity.
The presence of tangential components gives the active
nature to the stall cell. Compressor possesses abrupt nature
of stalling characteristics with smooth casing. Growth of
tangential velocity increases the rotor inlet blockage sig-
nicantly and causing the rotor incidence to increase
stalling value. As plotted in the FFT results shown in
Figure 12(c) and (d), it is clear that stall cells travels at
70 Hz which is 53% of the normal stalling frequency at this
speed. From Figure 12(b) is clearly seen that for the six
rotor revolutions that correspond to 0.045 s, there are three
distinct axial and tangential velocity peaks. One stall cells
travels corresponds to two rotor revolutions, indicate that
stall cell travels at nearly half the rotor speed.
At stall conditions the amplitude of oscillations of the
axial velocity is signicantly higher compared to that of

Figure 14 Comparative variations of rotor inlet tangential velocity for the smooth casing and casing treatment extensions at stall ow conditions.
(a) Variations of rotor inlet tangential velocity for the smooth casing and four axial extensions of bend skewed casing treatment, (b) variations of
rotor inlet tangential velocity for the smooth casing and two axial extensions of bend skewed casing treatment for the six rotor revolutions, (c) FFT
of rotor inlet tangential velocity for the smooth casing and two axial extensions of bend skewed casing treatment, and (d) exploded view of FFT
showing dominating stall frequencies.
246 Dilipkumar Bhanudasji Alone et al.

Figure 15 Variations of turbulent Intensity.

tangential velocity. For axial velocity the amplitude of

oscillations is almost touching 30 m/s whereas for in
tangential directions it is merely 7.5 m/s. From this we
can conclude that axial components are still dominating in
the stall regions. This could be because of location of the
measurement plane which is 1 chord away from the rotor
leading edge. Trends shows that very close to rotor leading
edge tangential component may be dominating over the
axial components. At this operating condition with smooth
casing Turbulence Intensity calculated is 41.75% which is
comparatively higher than all the casing treatment exten-
sions and can be conrm from Figure 15.

4.3.2. Axial extensions of bend skewed casing treatment

Behavior of the inlet axial velocity and tangential
velocity at stall conditions is studied for different axial
extensions of casing treatments. The variation of the
tangential velocity is compared with the smooth casing.
The location of the hot wire probe was kept same for all the
measurements. Figure 13 shows the comparative behavior
of inlet axial velocity for smooth casing and 20%, 40%,
60%, 100% axial extensions of bend skewed casing
Figure 16 Flow behaviors in the plenum chamber at choke ow and
treatment. From Figure 13(a) and (b), it is clear that with
stall conditions at 60% speed. (a) Plenum chamber velocity behavior at
all the extensions of casing treatments compressors stalls at choke ow and stall conditions, (b) exploded view showing variations
lower mass ow rate and thus enhanced the stall margin of for six rotor revolution, and (c) FFT of velocity signals.
the compressor stage at this speed. At stall conditions the
lowest mean axial velocity of 62 m/s measured is with 20%
axial coverage compared to 92 m/s with smooth casing. speed and with 20% coverage it travels at 73 Hz, which
Amongst all the casing treatment extensions, highest axial corresponds to 55% of the rotor speed. The amplitude of
velocity at stall is 68 m/s with 100% extension, whereas
oscillations reduces with the casing treatment compared to
other two axial extensions of 60% and 40% it falls in
that of with smooth casing. Oscillations are highest with
between 20% and 100% axial extensions. With increase in
the axial extensions from 20% to 100% stall cells slowed 60% axial extensions as for this axial extension the bend
down and travels at slightly lower speed and can be seen part of the casing treatments come over the rotor leading
from the FFT analysis shown in Figure 13(c) and (d). edge. For this extension this bend part is located above the
At 100% axial extensions stall cell travels at 60 Hz which sensitive portion of the rotor chord and thus causing sudden
corresponds to 45% of the rotor speed, at 60% extension it change in the upcoming streamlines. This increases the
travels at 63 Hz which corresponds to 48% of the rotor oscillations in the ow and thus also increases the turbulent
Stability management of high speed axial ow compressor stage through axial extensions of bend skewed casing treatment
intensity compared to other axial extensions and can be
seen from Figure 15. Turbulent intensity based on the axial
velocity is calculated using the mean and root mean square
(rms) values of axial velocity and using Eq. (5)
V rms
Turbulent Intensity % 5 from Figure 16(a) and exploded view shown in Figure 16
V mean
Tangential velocity measurements are shown in
Figure 14. Comparative behavior of tangential velocity is
studied only for the axial extensions of 20% and 40%. The
ow behavior is compared with the smooth casing. From
Figure 14(a) and (b) it is clear that with casing treatment is
in place, the uctuations in the tangential velocity reduced
signicantly. Casing treatment at these axial extensions is
able to remove the blockage created by the tangential
velocity at stall. At 40% axial extension also the appearance
of tangential components of the velocity is very small
compared to smooth casing. Further placing the treatment at
20% axial extension the oscillations in the tangential
disappeared and can be conrmed from the FFT results
shown in Figure 14(c) and (d). At 20% axial extensions and
at stall condition aerodynamic blockage created due to the
boundary layer growth/separations ahead of the rotor
leading edges are effectively eliminated. This phenomenon
allows compressor stage to operate at higher incidence
without going into abrupt stall as with the smooth casing.
These mechanisms results in progressive nature of stall
which further improve the pressure recovery of the com-
pressor stage and can be seen from the compressor
performance map shown in Figure 4.
4.3.3. Velocity measurements in the plenum chamber
Variation of the ow velocity emerging from the casing
slots in the plenum chamber is measured for the 100% axial
extensions. Probe was placed at half the plenum depth from
the top wall. The ow in the plenum chamber is very
complicated as the ow was neither coming in axial nor in
the tangential direction. Single component probe may not
give the actual picture of the complex and swirling ow
entering the plenum chamber through the casing slots.
Probe was rotate to get the actual ow direction and nally
aligned to the direction where the velocity of the ow was
highest. To conrm the direction of swirling ow entering
the plenum ow visualizations was done using tufts. Flow
was entering the plenum chamber from the rear section of
the casing treatment slots at an angle of 451 and going
towards the front section of the slots. The measurements
were carried out at 100% axial extensions at choke ow and
stall ow condition. Figure 16 shows the variation of ow
velocity in the plenum chamber. At choke ow condition
the mean velocity measured in the plenum chamber was
40 m/s, which was lower than the main stream velocity at
this condition.
The pressurized uid from the rear section of the blade
was entering the chamber through the slots and leaving
through other slots. The severity of the ow entering the
plenum chamber was more in the stall condition due to
247
additional back pressure. The plenum chamber is acting as a
bridge for channelizing the ow from high pressure to low
pressure side. At stall condition and at this coverage plenum
chamber was getting lled with the uid and thus the ow
velocity was almost reaches to 1 to 2 m/s and can be seen
(b). From these gures it is quite clear that in addition to
pressure difference there was periodic injection of the uid
into the plenum chamber as rotor blade passes the casing
treatment slots. Periodic injections of the uids can be
observed from the velocity peak. Each 0.016 s represent the
time required for one stall event with 100% axial extensions
which is slightly higher than the normal stalling frequency
of half of the rotor speed. With this axial extensions stall
cells travels at 45% of the rotor speed and can be seen from
FFT analysis shown in Figure 16(c).
Turbulent intensity was calculated at this condition at
choke ow and stall conditions. The intensity of turbulence
was 11% at choke condition whereas it reaches to 66% with
stall conditions. From this we can conclude that the ow
entering the plenum is highly swirly and thus increase the
turbulence level. This turbulence further causes mixing
losses which results in the total pressure loss and degrada-
tions in the efciency of the stage with this axial extension.
5. Conclusions
From the experimental studies on the axial extensions of
bend skewed casing treatments following conclusions are
drawn.
1. All the extensions of casing treatment do not produce
the fruitful results. Axial extensions of 100% at 80%
speed compressor stage stalls at higher mass ow rate
and results in negative stall margin.
2. At 80% axial extensions the at all the operating speeds
excepts 70% compressor stall with lower mass ow rate
and results in signicant gain in the stall margin, but
again in terms of pressure loss and efciency penalty.
3. At 60% extensions compressor stage performance
shows signicant improvement with maximum improve-
ment of 32% in the stall margin at 50% speed with 3%
drop in the peak stage efciency.
4. Beyond 70% speed compressor stage shows recovery in
terms of stable operating range, pressure recovery and
efciency gain.
5. At lower axial extensions of 40% and 20% compressor
stage performance improves drastically. At 20% axial
extensions stall margin improvement at lower operating
speed is higher than at higher operating speed. Com-
pressor stage shows recovery in the stage pressure ratio
which is increases with increase in the speed.
6. Rotor inlet hot wire measurements shows that with
smooth casing inlet axial velocity varies very rapidly
and thus indicates the very strong nature of stall. The
presence of the tangential velocity in the stall gives the
248
active nature to the stall cell. Growth of tangential
velocity increases the rotor inlet blockage signicantly
causing the rotor incidence to increase stalling value.
7. For the smooth casing stall cells travels at 53% of the
rotor speed. At stall conditions the amplitude of oscilla-
tions of the axial velocity is signicantly higher
compared to that of tangential velocity.
8. At stall conditions with 20% axial coverage compressor
stall at lowest mean axial velocity of 62 m/s compared
to 92 m/s with smooth casing. Amongst all the casing
treatment extensions, highest axial velocity at stall
condition is 68 m/s with 100% extension.
9. With increase in the axial extensions from 20% to 100%
stall cells slowed down by traveling at slightly lower
speed. At 100% axial extensions stall cell travels at 45%
of the rotor speed, at 60% extension it travels at 48% of
the rotor speed and with 20% coverage it travels at 55%
of the rotor speed.
10. When the bend skewed casing treatment is in placed the
uctuations in the tangential velocity reduces signi-
cantly. At 20% and 40% axial extensions casing
treatment remove the blockage created by the tangential
velocity at stall condition. These further results in
progressive nature of stall with this type of casing
treatment.
11. It will be interesting to carry out in depth unsteady
computational analysis to understand the ow physics
associated with this casing treatment coupled to axial
ow compressor stage in near future.
12. The main limitations of this experimental study were a
compressor stage performance evaluation restricted up
to 80% design. The performance evaluation could not
carry out at design speed due to rotor dynamics issues. It
will be interesting to see behavior of compressor stage
in the presence of bend skewed casing treatment
conditions at design condition in near future.
13. Another limitation is the unsteady measurements using
hot wire probes. The measurements were restricted till
60% design speed to avoid potential damage to the hot
wire probe. Specically design hot wire probe or
unsteady probe can be used for higher speed in near
future.
14. Also, the specically designed multipoint probe can be
used to resolve the unsteady ow eld in the bend
skewed casing treatment slots in near future.
Acknowledgments
Authors take this opportunity to thank Director, CSIR-
NAL, for funding the research program through elevenths
ve year plan and allow publishing the results. Indefatig-
able hard work of Mr. Joseph Arul Raj and Mr. G.S. Ravi,
Team members of AFCR are genuinely accredited. We
would also like to express thanks to Head, Propulsion
Division and Electrical Section, NWTC, Belur for their
support and encouragement. Also special thanks are due to
Dilipkumar Bhanudasji Alone et al.
Mr. Veera Sheshakumar and Mr. Vijay Tijare for helping in
rotor dynamics aspects of the test rig.
References
[1] E.M. Greitzer, J.P. Nikkanen, H.D. Joslyn, Fundamental
criterion for the application of rotor casing treatment, Trans.
ASME J. Fluid Eng. 101 (1979) 237243.
[2] L. Venzel, J.E. Moss Jr, Effect of casing treatment on
performance of multistage compressor, NASA TMX-3175,
Jan., 1975.
[3] D. Urasek, Effect of casing treatment on performance of a
two stage high pressure ratio fan, NASA-TP-1409, Feb.,
1779.
[4] D. Royce W. Osborn. Effect of tip clearance on overall
performance of transonic fan stage with and without casing
treatment, NASA TMX-3479, 1977.
[5] S.A. Guruprasad, Experimental investigations on the inu-
ence of axial extension and location of outer casing treatment
on the performance of an axial ow compressor, in:
Proceedings of the 4th International Symposium AIF, Dres-
den, Germany, 1999.
[6] H. Takata, Y. Tsukuda, Stall margin improvement by casing
treatment, its mechanism and effectiveness, ASME J. Eng.
Power 1 (1977) 237243.
[7] H. Fujita, H. Takata, A study on congurations of casing
treatment for axial ow compressors, Bull. JSME 27 (230)
(1984) 16751681.
[8] G.D.J. Smith, N.A. Cumptsy, Flow phenomenon in com-
pressor casing treatment, Trans. ASME, J. Eng. Gas.
Turbines Power 106 (3) (1984).
[9] P.C. Adams, G.D.J. Smith, The effect of axial slot casing
treatment geometry on the stall margin improvement and
efciency of a low speed axial ow compressor, ISABE 87-
7036, 1987.
[10] R. Emmrich, H. Hnen, R. Niehuis, Time resolved investiga-
tions of an axial compressor with casing treatment, part-1:
experiment, GT2007-27581, in: Proceedings of the ASME
Turbo Expo: Power for Land, Sea and Air, Montreal Canada,
2008.
[11] R. Emmrich, R. Kunte, H. Hnen, R. Niehuis, Time resolved
investigations of an axial compressor with casing treatment,
part 2: experiment, GT2007-27582, in: Proceedings of the
ASME Turbo Expo: Power for Land, Sea and Air, Montreal
Canada, 2008.
[12] R. Emmrich, H. Hnen, R. Niehuis, Time resolved experi-
mental investigations of an axial compressor with casing
treatment, J. Turbomach. 131 (2009) 011018-1-9.
[13] J. Zhu, W. Chu, The effects of bend skewed groove casing
treatment on performance and ow eld near end wall of an
axial compressor, AIAA 2005-809, in: Proceedings of the
43rd AIAA Aerospace Sciences meeting and exhibit, 1013
January, Reno, Nevada, 2005.
[14] X. Lu, W. Chu, Y. Zhang, J. Zhu, Experimental and
numerical investigation of a subsonic compressor with
bend-skewed slot-casing treatment, Proc. IMechE Vol. 220
Part C: J. Mech. Eng. Sci. (2006).
[15] X. Lu, J. Zhu, C. Nie, W.H. Huang, The stability-limiting in
a subsonic axial ow compressor and its passive control with
casing treatment, GT2008-50006, in: Proceedings of the
Stability management of high speed axial ow compressor stage through axial extensions of bend skewed casing treatment
ASME Turbo Expo: Power for Land, Sea and Air, Berlin,
Germany, 2008.
[16] X. Lu, W. Chu, J. Zhu, Y. Zhang, Numerical investigations
of the coupled ow through a subsonic compressor rotor and
axial skewed slot, J. Turbomach. 131 (2009) 011001-1.
[17] N. Gourdain, M. Montagnac, J.F. Boussuge, Numerical
simulation of an axial compressor with non-axisymmetric
casing treatment, in: Proceedings of the 2nd European
Conference for Aerospace Sciences (EUCASS).
[18] M. Voges, R. Schnell, C. Willert, R. Mnig, M.W. Mller, C.
Zscherp, Investigation of blade tip interaction with casing
treatment in a transonic compressor, part 1: particle image
velocimetry, GT2008-50210, in: Proceedings of the ASME
Turbo Expo: Power for Land, Sea and Air, Berlin, Germany,
2008, Journal of Turbomachinery 133 (2011) 011007-1.
[19] R. Schnell, M. Voges, R. Mnig, M.W. Mller, C. Zscherp,
Investigation of blade tip interaction with casing treatment in
a transonic compressor, part 2: numerical results, GT2008-
50212, in: Proceedings of the ASME Turbo Expo: Power for
Land, Sea and Air, Berlin, Germany, 2008, Journal of
Turbomachinery 133 (2011) 011008-1.
[20] M. Voges, C. Willert, RMnig, H.P. Schiffer, The effect of a
bend-slot casing treatment on the blade tip ow eld of a
transonic compressor rotor, GT2013-94006, in: Proceedings
of the ASME Turbo Expo, Turbine Technical Conference
and Exposition, San Antonio, Texas, USA, 2013.
249
[21] D.B. Alone, S. Satish Kumar, M.T. Shobhavathy, J.R.
Reddy, A.M. Pradeep, S. Ramamurthy, S.V. Iyengar, Per-
formance characterization of the effect of axial positioning of
bend skewed casing treatment retrotted to a transonic axial
ow compressor, GT2014-26099, in: Proceedings of the
ASME Turbo Expo, June 1620, Dusseldorf, Germany,
2014.
[22] D.B. Alone, S. Satish Kumar, M.T. Shobhavathy, J.R.
Reddy, A.M. Pradeep, S. Ramamurthy, S.V. Iyengar, Experi-
mental investigation on the effect of porosity of bend skewed
casing treatment on a single stage transonic axial ow
compressor, GT2014-26102, in: Proceedings of the ASME
Turbo Expo, June 1620, Dusseldorf, Germany, 2014.
[23] D.B. Alone, S. Satish Kumar, M.T. Shobhavathy, J.R.
Reddy, A.M. Pradeep, S. Ramamurthy, S.V. Iyengar, On
understanding the effect of plenum chamber of a bend
skewed casing treatment on the performance of a transonic
axial ow compressor, GT2014-26103, in: Proceedings of
the ASME Turbo Expo, June 1620, Dusseldorf, Germany,
2014.
[24] D.B. Alone, S. Satish Kumar, M.T. Shobhavathy, J.R.
Reddy, A.M. Pradeep, S. Ramamurthy and S.V. Iyengar,
Improvement of moderately loaded transonic axial compres-
sor performance using low porosity bend skewed casing
treatment, Int. J. Rotating Mach. 2014 (2014), 14 pp. Article
ID 625876, http://dx.doi.org/10.1155/2014/625876.