Aerospace Science and Technology 36 (2014) 87–93

Contents lists available at ScienceDirect

Aerospace Science and Technology

www.elsevier.com/locate/aescte

Shock tunnel experiments on control of shock induced large

separation bubble using boundary layer bleed
R. Sriram ∗,1 , G. Jagadeesh 2
Department of Aerospace Engineering, Indian Institute of Science, Bangalore, Karnataka, 560012, India

a r t i c l e i n f o a b s t r a c t

Article history: Shock–Boundary Layer Interaction (SBLI) often occurs in supersonic/hypersonic flow fields. Especially
Received 12 January 2014 when accompanied by separation (termed strong interaction), the SBLI phenomena largely affect the
Received in revised form 4 April 2014 performance of the systems where they occur, such as scramjet intakes, thus often demanding the
Accepted 8 April 2014
control of the interaction. Experiments on the strong interaction between impinging shock wave and
Available online 13 April 2014
boundary layer on a flat plate at Mach 5.96 are carried out in IISc hypersonic shock tunnel HST-2. The
Keywords: experiments are performed at moderate flow total enthalpy of 1.3 MJ/kg and freestream Reynolds number
Hypersonic flow of 4 million/m. The strong shock generated by a wedge (or shock generator) of large angle 30.96◦ to
Separation control the freestream is made to impinge on the flat plate at 95 mm (inviscid estimate) from the leading
Shock–Boundary Layer Interaction (SBLI) edge, due to which a large separation bubble of length (75 mm) comparable to the distance of shock
Shock tunnel experiments impingement from the leading edge is generated. The experimental simulation of such large separation
bubble with separation occurring close to the leading edge, and its control using boundary layer bleed
(suction and tangential blowing) at the location of separation, are demonstrated within the short test
time of the shock tunnel (∼600 μs) from time resolved schlieren flow visualizations and surface pressure
measurements. By means of suction – with mass flow rate one order less than the mass flow defect in
boundary layer – a reduction in separation length by 13.33% was observed. By the injection of an array
of (nearly) tangential jets in the direction of mainstream (from the bottom of the plate) at the location of
separation – with momentum flow rate one order less than the boundary layer momentum flow defect
– 20% reduction in separation length was observed, although the flow field was apparently unsteady.
© 2014 Elsevier Masson SAS. All rights reserved.

1. Introduction

Interaction of a shock of sufficient strength with a boundary

layer effects in flow separation [5]. A shock impinging on a flat
plate boundary layer is a fundamental case of the interaction,
which is especially observed in high speed intakes at off-design
condition (at higher operating Mach numbers than the design con-
ditions). Particularly, at hypersonic speeds, the occurrence of such
interactions in scramjet intakes, when accompanied by flow sep-
aration, affects the performance of the entire system drastically,
thus calling for the control of the interaction at off-design. In Fig. 1. Schlieren image of intake flow field at off-design Mach number of 8 (Mahap-
such a case the separation occurs close to (or at) the leading edge atra and Jagadeesh [10]).
(where the boundary layers are insignificantly thin); the separation
bubble is of length, comparable to the distance of shock impinge- ment from the leading edge – termed large separation bubble in
the present study.
A schlieren image of such separation occurring in hypersonic
* Corresponding author. Tel.: +91 80 22932424.
intake (cowl plate) at off-design Mach number of 8 was presented
E-mail addresses: sriram@aero.iisc.ernet.in (R. Sriram), jaggie@aero.iisc.ernet.in
(G. Jagadeesh).
by Mahapatra and Jagadeesh [10]; the schlieren image is shown in
1
Research scholar. Fig. 1, with the indication of location of separation S, reattachment
2
Professor. R and the distance between them Lsep called separation length.

http://dx.doi.org/10.1016/j.ast.2014.04.003
1270-9638/© 2014 Elsevier Masson SAS. All rights reserved.
88 R. Sriram, G. Jagadeesh / Aerospace Science and Technology 36 (2014) 87–93
There are very few (and only recent) investigations on the separa-
tion at leading edge termed ‘separation with zero initial boundary
layer thickness’ at hypersonic (high enthalpy) flow conditions [14]
and to the best of our knowledge there are no reported works on
the control of such flow fields at hypersonic speeds in the litera-
ture.
The interactions and their control (of especially separation, us-
ing various techniques) at supersonic speeds are well understood
[4,15,22]. The understanding of complex 3-dimensional and un-
steady aspects of the interaction at supersonic speeds has grown
substantially with the advent of advanced flow diagnostics (like
PIV and supersonic anemometry) and computations (especially LES
and DNS). However, experimentation on the interaction at hyper-
sonic flow condition requires the simulation of associated moder-
ate to high flow total enthalpies. While hypersonic wind tunnels do
simulate the high Mach numbers, impulse facilities like shock tun-
nels are required to simulate the required total enthalpy, but only
for short test times (∼1 ms). There are however, only few shock
tunnel studies on shock wave boundary layer interactions reported
in the literature [2,3,11], due to the concerns regarding the evolu-
tion of separated flow within the short run time; simulation and
investigation of control of the interaction within the short run time
is still challenging.
There are two ways by which the interaction may be con-
trolled – one is by the manipulation of the boundary layer ahead
of the interaction; the other is by adopting control mechanisms
in the interaction zone itself (such as passive control, or bound-
ary layer bleed at shock foot itself) [4]. Since the boundary layer’s
resistance to the imposed adverse pressure is crucial in determin-
ing the interaction, the increase of boundary layer momentum (or
favourably modifying the boundary layer profile) upstream of the
interaction would increase the boundary layer resistance, thereby
tending to reduce the size of separation bubble or even prevent
separation. Techniques of this type are thus appropriate for strong
interactions (with separation, especially large separation bubbles).
They include vortex generators which enhance the momentum of
boundary layer flow by mixing with the outer flow [1,12,18,21];
boundary layer bleed through slots or holes which include suction
and blowing upstream of the interaction, manipulating the bound-
ary layer profile [6,7,17,19,20,23]; and also wall cooling (including
film and transpiration cooling) [2,8,9].
Boundary layer bleed is a classical technique for separation con-
trol, but there are challenges in implementing the technique to
shock boundary layer interactions, due to such concerns as bleed
location relative to the interaction and bleed slot/hole geometry
which can have effects on shock patterns. Especially, the bleed
through the holes (rather than slot) is 3-dimensional and under-
standing the flow control requires the study of the complex 3-D
flow field; this has largely been addressed by computational stud-
ies in the literature [7,17,20]. Bleed through micro-holes has also
been used as micro-air-vortex generators upstream of the interac-
tion [1,21]. Recently, boundary layer bleed has also been imple-
mented to control SBLI phenomena of a shock train [23]. Many
of the flow control studies reported in the literature are at su-
personic speeds; while the above mentioned studies addressing
the wall cooling are among the experimental investigations at hy-
personic speeds (including the data from among the few reported
shock tunnel experiments). There are few wind tunnel studies on
the control of hypersonic SBLI using bleed. Schulte et al. [19] re-
ported an extensive (hypersonic) wind tunnel study on the control
of impinging shock wave boundary layer interaction using bleed
(both suction and blowing) at Mach 6 (total temperature of 500 K).
Optimal location of bleed relative to shock impingement location,
bleed slot width and bleed slot angle, resulting in maximum re-
duction in separation length, were reported from among the ex-
periments in the study. Haberle and Gulhan [6] also reported hy-
Fig. 2. A typical pitot signal in HST-2 (at Mach 5.96).
personic (Mach 6) wind tunnel studies of scramjet inlet flow field
in the presence of bleed for the control of lip shock induced sep-
aration on the ramp. Though these experiments were reported at
high Mach numbers, shock tunnel experiments are required to un-
derstand the control of the interaction at higher total enthalpies
associated with hypersonic flows. It is with this backdrop that
shock tunnel experiments were initiated in order to understand
the hypersonic impinging shock wave boundary layer interaction
with large separation bubble and its control. The present paper re-
ports the shock tunnel demonstration of reduction in separation
length using bleed (both suction and blowing) at the location of
separation occurring close to the leading edge, at same freestream
Mach number and Reynolds number reported by Schulte et al. [19],
but at a relatively higher total enthalpy of 1.3 MJ/kg.
2. Experimental facility and test model
Experiments are performed in IISc hypersonic shock tunnel
HST-2 [10]; the shock tube is 50 mm in diameter, with 2 m driver
and 5.12 m driven sections, and the test section is of 300 mm ×
300 mm cross section and 450 mm length. It is a conventional
shock tunnel where the reflection of a propagating shock at the
end of the shock tube compresses the test gas to high pressure
and temperature (thus simulating the required enthalpy). The com-
pressed test gas is expanded through a nozzle (whose area ratio
simulates the required Mach number) into the test section of the
tunnel. Thus the hypersonic flow of required flow enthalpy is sim-
ulated in impulsive fashion, i.e. for a short run time. The end of
the shock tube is equipped with two fast response PCB pressure
sensors from which the shock speed and the reservoir pressure
after the shock reflection are measured. The total temperature af-
ter shock reflection is estimated from the shock speed. The total
pressure after the normal shock inside the test section is mea-
sured using a pitot probe; the pitot pressure is measured by a fast
response PCB sensor and all the pressure signals (including the sur-
face pressure signals) are recorded by the data acquisition system
NI PXI-6133 at a rate of 1 mega sample per second. A typical pitot
pressure signal is shown in Fig. 2.
The pitot signal also serves to indicate the test time of the tun-
nel. It may be seen that the pitot signal rises from 2.7 ms to 3.3 ms
(called rise time), after which it remains steady for ∼600 μs, till
3.9 ms. The time for which the pitot remains steady is the test
time of the shock tunnel, during which the time averaged pres-
sure gives the pressure behind a normal shock for the simulated
freestream conditions (the steady pitot pressure). From the mea-
sured shock tube conditions and the pitot pressure freestream con-
ditions are estimated using normal shock relations and are given in
Table 1. The uncertainty in freestream Mach number M ∞ is ±3.2%,
 R. Sriram, G. Jagadeesh / Aerospace Science and Technology 36 (2014) 87–93 89

Table 1
Freestream conditions.

M∞ p ∞ (Pa) T ∞ (K) Re∞ (×106 /m) h0 (MJ/kg)

5.96 1277 160 4 1.3

Fig. 4. Schlieren image of the impinging shock boundary layer interaction.

Fig. 3. Schematic of the test model.

in freestream pressure p ∞ and temperature T ∞ is ±4.4%, in total

enthalpy h0 is ±3%, and in freestream Reynolds number Re ∞ is
±8.3% (obtained from the uncertainties in measured shock tube
conditions and pitot pressure by the method described by Moffat
[13]).
The strong impinging shock is generated by a shock generator
(wedge) of angle 30.96◦ to the freestream (spanning 60 mm). The
shock is made to impinge on a flat plate 60 mm below the bottom
of the wedge, whose leading edge is fixed 10 mm behind (down-
stream) of the leading edge of the shock generator for the present
experiments (although the position can be varied). The wedge and
the flat plate are held by a fixture which mounts them in the
shock tunnel, as shown in Fig. 3. From Euler computations the in-
viscid shock impingement location was found to be 95 mm from
the leading edge, accounting of the slight curving of the impinging
shock due to the interaction with expansion fan from the rear end
of the wedge.
Time resolved schlieren flow visualizations using a high speed
camera (Phantom V-310) and surface pressure measurements us-
ing fast response PCB sensors (with uncertainty of ±1%) are the
flow diagnostics used to study the interaction and its control.
There are also few surface pressure measurements using Kulite
sensors (with uncertainty of ±0.5%) and closely spaced MEMS
pressure sensors made in-house [16] for the template case with-
out any flow control. Schlieren flow visualizations are obtained at
a rate of 10 000 frames per second using the high speed camera.
3. Results and discussions
A schlieren image of the flow field (after the establishment of
steady flow with separation) for the template case (without flow
control), pointing the various features of the interaction, is shown
in Fig. 4. The corresponding surface pressure distribution is shown
in Fig. 5, with the schlieren image cropped to appropriate scale
attached at the top for reference.
It is apparent that the separation bubble is as large as the
distance of shock impingement from the leading edge and that
the separation occurs close to the leading edge. The location of
separation is found out by the intercept of the (projected) separa-
tion shock on the flat plate surface, since the upstream influence
lengths are negligibly small compared to separation length (due to
thin boundary layers close to the leading edge). This is done by
locating points of intensity peaks corresponding to the separation
shock and fitting a line to that; the intercept of the line on the
plate gives the separation location (with uncertainty of ±0.05 mm
based on the distance of the farthest point from line fit). For the
above case the separation is located at 16 mm from the leading
Fig. 5. Surface pressure distribution on the flat plate.
edge. The reattachment is similarly (approximately) located from
the schlieren image as the foot of the reattachment shock, which
is around 90 mm from the leading edge for the above case. Thus
the separation length is estimated to be ∼75 mm. The separation
length is estimated similarly for other cases too from the respec-
tive schlieren images. A good correspondence may also be seen in
the surface pressure distribution, where the peak pressure is ob-
served at 90.5 mm from the leading edge.
Suction of the fluid from the surface tends to reduce the dis-
placement effect of the boundary layer, thereby making it more
resistant to adverse pressure gradients. In the present experiments
suction is created simultaneously at the location of separation (i.e.
at 16 mm from the leading edge) and at a location slightly down-
stream (by 5 mm) by means of oblique holes made on a (2 mm
thick) plate kept in flush with the flat plate surface above a suc-
tion chamber milled in the flat plate as shown schematically in
Fig. 6.
The suction chamber is evacuated by means of a turbo pump,
through a pipe of 1/2 in diameter that connects the chamber and
the turbo pump (kept outside the tunnel dump tank) through a
connecting port at the top of the tunnel dump tank. The holes
are maintained oblique with an angle of 60◦ to the flat plate sur-
face, so that the suction is also accompanied by a small increase in
streamwise velocity in the boundary layer close to the plate. There
are 16 holes in each row for a span of 60 mm as shown in Fig. 7.
Each hole is 2 mm in diameter.
The turbo pump maintains a vacuum of 10−5 mbar pressure,
such that due to the much higher static pressure (1277 Pa) above
90 R. Sriram, G. Jagadeesh / Aerospace Science and Technology 36 (2014) 87–93

Fig. 6. Schematic diagram of suction from the surface of the plate in the presence
of freestream and without impinging shock.
Fig. 8. Schlieren image of impinging shock boundary layer interaction in the pres-
ence of suction.

Fig. 7. Photographs of the model for suction.

the plate exposed to the flow, fluid is sucked inside the surface
through each hole at chocked mass flow rate. The estimate of the
total chocked mass flow rate with all the holes (32 holes in to-
tal) 4.2 × 10−4 kg/s; this may be compared with the mass defect Fig. 9. Schematic diagram of blowing from the bottom of the plate to the surface in
due to boundary layer displacement at the location of separation, the presence of freestream.
which is 5 × 10−3 kg/s for the entire span of the interaction (con-
servatively estimated by assuming the displacement thickness to
be the boundary layer thickness itself, since they are of the same
order). Hence, roughly 12% of the displaced mass is sucked. The
contribution to the momentum due to oblique holes is small (2 or-
ders lower than momentum defect) due to smaller chocked flow
velocities, and need not be considered. Since the boundary layer
suction at above specified magnitude does not greatly alter the
outer inviscid flow that may be visible in schlieren (by means of
waves) the attention may be directly turned towards the flow field
with impinging shock. The various features of the interaction in
the presence of suction, observed in the visualization after the es-
tablishment of steady flow field are indicated in Fig. 8.
The locations of suction are also indicated in the figure. It
is apparent that the separation with suction occurs considerably
downstream of the location for the template case; in fact after the
second array of orifices. The separation is approximately located Fig. 10. Photograph of the model used for blowing experiments.

at 26 mm from the leading edge, and with nearly the same loca-
tion of reattachment as the template case, the separation length is
∼65 mm with suction, 10 mm less than that of the template case.
Tangential blowing at the point of separation makes the value
of wall velocity gradient positive and hence non-zero, thus delay-
ing or avoiding flow separation. In the present study, the mass flow
for blowing at the surface of the plate is given from the bottom of
the plate itself. The schematic arrangement of the set-up for blow-
ing experiments adopted in the present study is shown in Fig. 9.
Since the bottom of the plate is inclined at an angle of 26.57◦
to the freestream, the pressure is ∼14.36 times more than the
freestream pressure. Thus if a bleed is to be given from the bottom
of the plate to the top, whose pressure is of the order of freestream
pressure a chocked flow may be expected (since the pressure ratio
is much greater than 0.528, which is required for chocking). In the
experiments, the bleed is given by means of holes (as with suction
experiments) rather than a 2-dimensional slot. It is not possible to
have a perfect tangential injection; hence to be closer, the bleed
hole must be oriented at an angle as small as possible to the flat
plate surface. The angle of 10◦ to the plate surface shown in the
figure was decided based on the machining constraint, since if the
hole is to be drilled at such low angles, in order that it ends at the
location of separation at around 15 mm from the leading edge on
the surface, the hole must start from the bottom of the plate, very
close to the leading edge. A photograph of the flat plate model
with oblique holes for blowing is given in Fig. 10.
It may be noted that the holes are only approximately aligned
spanwise, around the location of separation (for a span of 50 mm).
There are 12 holes, each of diameter 2.5 mm and since the hole is
highly oblique to the surface each hole may appear to be extending
for ∼10 mm in the streamwise direction in the photograph. How-
ever, attention to the precision of the hole locations is not given
presently, intending only to demonstrate the effect of providing
bleed from the bottom of the plate at approximate separation lo-
cation.
 R. Sriram, G. Jagadeesh / Aerospace Science and Technology 36 (2014) 87–93
Fig. 11. Schlieren image with injection from the bottom of the plate over the surface
and without impinging shock.
Estimate of chocked mass flow rate at each hole is 1.59 ×
10−4 kg/s; thus for 12 holes the estimate of chocked mass flow
rate of injection is around 2 × 10−3 kg/s, which is comparable with
the mass defect due to boundary layer displacement at the location
of separation. However, the mass defect due to boundary layer is
due to the induced y-component of velocity in the boundary layer.
The aspect which is more important with tangential blowing is
the momentum addition, to counter the momentum defect in the
boundary layer. The estimate of momentum defect (assuming mo-
mentum thickness to be boundary layer thickness itself since they
are of same order) at the location of separation is 7.67 N for the
entire span of the interaction. The momentum flow rate of each
jet estimated from the chocked mass flow and the chocked flow
velocity is around 6.6 × 10−2 N. Thus for 12 holes the estimate
of momentum injection rate is around 0.8 N; since the jet dimen-
sions are comparable with the momentum thickness, it may be
said that a perfect tangential injection adds a momentum which
is 10% of the deficit momentum. However, since the injection is
not perfectly tangential, it needs to be seen, if the injected jets
separate from the surface of the plate and reattach downstream.
A schlieren image during the test time in the presence of blowing
from the bottom of the plate on the plate surface (without imping-
ing shock) is shown in Fig. 11.
Downstream of the leading edge shock, a shock due to the
jet injection may be seen, which eventually merges downstream
with the leading edge shock. Further downstream, there are no
shocks from the surface of the plate, indicating that the injected
jet is attached to the flat plate surface. Thus it may be comfort-
ably supposed that the injection is tangential to the surface of the
flat plate. With impinging shock (and with blowing), the various
features observed in the visualization during the test time are indi-
cated in Fig. 12. Subsequent to leading edge shock the shock due to
injection is seen, which supposedly emanates approximately from
the location of separation for the template case. The separation
shock is apparently downstream of this location considerably, al-
though the shock does not seem to emanate from close to the
surface as with the other cases due to thick boundary layers on
fluid injection. Thus the separation location (and thus the sepa-
ration length) with blowing cannot be considered as accurate as
the other cases. The estimated separation location by extending
the separation shock to the surface is around 30–35 mm (consid-
erably downstream of the other cases) though it was found to be
oscillatory; the separation length was approximately 60 mm, with
the reattachment also slightly downstream of the location for other
cases. A further discussion on the trend and the observed unsteady
behaviour with blowing are presented subsequently.
It is evident from the shown schlieren images that the sepa-
ration length is considerably reduced by providing boundary layer
bleed. While with suction a 13.33% reduction is seen, with blowing
91
Fig. 12. Schlieren image of impinging shock boundary layer interaction in the pres-
ence of blowing.
Fig. 13. Comparison of the variation of separation location with time.
the furnished schlieren image shows a reduction by 20%. However,
the time resolved schlieren images with blowing showed unsteady
(nearly oscillatory) behaviour. A comparison of the time resolved
separation location for all the cases is shown in Fig. 13 (with the
test time as observed from pitot signal indicated).
It may be seen that only by the middle of the test time (around
3.6 ms) the separation location is found to reach a steady position
for the template case. With suction, the separation location seems
to be steady from the beginning of the test time itself; probably
due to smaller separation length, which requires lesser time to
evolve. However, with blowing though the separation shock arrives
at the average location by the beginning of the test time itself, it
tends to get back downstream, and after 3.5 ms again tends to
move upstream until 3.8 ms, while at the last frame during the test
time the shock again moves back downstream. While the reattach-
ment shocks are steady for both the template case and for suction
from the beginning of the test time, for blowing the reattachment
shock also displayed such unsteady behaviour.
A comparison of the temporal behaviour of separation and reat-
tachment locations for the case with blowing is shown in Fig. 14.
The location of reattachment shock also seems to follow the same
trend as the separation shock. The reason for this unsteadiness
is not known, though the fluctuations in pressure bringing about
variations in chocked mass flow rate may be speculated as a rea-
son. Nevertheless, in the present experiments, blowing reduces
the separation bubble length significantly despite the unsteadiness,
and the reduction is apparent in all time frames. It can further be
noted that (based on chocked mass flow estimates) the bleed mass
flow rate with blowing is one order more than that with suction.
However the mechanisms of control with blowing and suction are
92 R. Sriram, G. Jagadeesh / Aerospace Science and Technology 36 (2014) 87–93
Fig. 14. Comparison of the variation of separation and reattachment locations with
time for the flow field with blowing.
Fig. 15. Comparison of surface pressures with and without flow control.
different; while with (tangential) blowing it is the addition of mo-
mentum to the boundary layer that is expected to aid separation
control, with suction the control is achieved by the reduction in
boundary layer displacement. In the present experiments, with the
bleed achieved through orifices rather than (2-dimensional) slots,
the three dimensional aspects can also take prominence, aiding the
generation of streamwise vorticity. This could be more pronounced
with blowing, due to higher mass injection; the blowing through
the orifices thus also acts as air vortex generators, enhancing the
mixing of the outer flow with the boundary layer. The generation
of streamwise vorticity may also have effect on the boundary layer
stability and can play a role on the observed unsteadiness with
blowing.
A comparison of surface pressure distributions is given in
Fig. 15 (from only PCB sensors). The error bars are the maximum
and minimum measured values from all experiments for each case,
while the plotted values are the averages of all measured values.
It must be mentioned that the sensor at 85 mm which was little
upstream of the reattachment location sensed a high but unsteady
pressure, especially because of the larger sensing area (diameter
5.5 mm) of PCB sensor, due to which the edge of the sensor
gets very close to the reattachment. Despite the observed oscil-
lations in the reattachment shock itself, the pressure measured
by the sensor at 85 mm with blowing is lesser than that with
other cases; because the reattachment is apparently little down-
stream with blowing. The comparison of surface pressures inside
the separation bubble is shown in Fig. 16 (since separation was
close to leading edge it was not possible to measure pressures in
the vicinity of separation). It may be noted that with suction the
Fig. 16. Comparison of surface pressures inside separation bubble with and without
flow control.
plateau pressures are low. This may have to do with the possi-
ble reduction in surface pressure with suction due to reduction in
boundary layer displacement. The blowing however increases the
surface pressure levels, especially close to the injection location.
Even for the case without impinging shock the sensors read con-
siderably higher pressure than the freestream pressure. Thus, the
pressure in the separation bubble also seems to be higher than
the template case. Though the reduction in shock induced separa-
tion length with bleed in separation location (close to the leading
edge) is demonstrated within the short run time of the shock tun-
nel, further studies and refinements are required to understand
the various complicated aspects (such as the observed unsteadi-
ness with blowing). Further experiments are required in order to
understand the role of location, orientation and geometry of the
bleed holes and optimize them for better control of the hypersonic
SBLI.
4. Conclusions
The control of impinging shock induced large separation bubble
on a flat plate, by boundary layer bleed at the location of separa-
tion (close to the leading edge) was demonstrated within the short
test time of shock tunnel HST-2 at Mach 5.96 (at total enthalpy
1.3 MJ/kg and Reynolds number 4 million/m). The strong imping-
ing shock was generated by a shock generator of angle 30.96◦ to
the freestream. The separation length (without bleed) was 75 mm
for inviscid shock impingement location of 95 mm from the lead-
ing edge, with the separation occurring at 16 mm from the leading
edge. Suction in the region of separation, at a mass flow rate one
order less than the boundary layer mass defect at separation loca-
tion, reduced the separation length by 13.33%. Tangential blowing
at the location of separation, with momentum flow rate of the in-
jected fluid one order less than the boundary layer momentum
defect at the separation location, reduced the separation length by
20%, although the flow field was found to be unsteady within the
test time with blowing.
Conflict of interest statement
The authors declare that there is no conflict of interest with any
individual/organization for the present work.
Acknowledgements
The authors would like to thank the Defence Research and De-
velopment Organisation (DRDO), India, for their support of the
research works in Laboratory for Hypersonic and Shock wave Re-
search (LHSR), Dept. Aerospace Engineering, Indian Institute of Sci-
 R. Sriram, G. Jagadeesh / Aerospace Science and Technology 36 (2014) 87–93
ence, Bangalore. The support of the other members of LHSR is also
gratefully acknowledged.
References
[1] Mohd Y. Ali, Farrukh S. Alvi, Rajan Kumar, C. Manisankar, S.B. Verma,
L. Venkatakrishnan, Studies on the influence of steady microactuators on
shock-wave/boundary-layer interaction, AIAA J. 51 (2013) 2753–2762.
[2] M. Bleilebens, H. Olivier, On the influence of elevated surface temperatures on
hypersonic shock wave/boundary layer interaction at a heated ramp model,
Shock Waves 15 (2006) 301–312.
[3] J.-P. Davis, B. Sturtevant, Separation length in high-enthalpy shock/boundary
layer interaction, Phys. Fluids 12 (2000) 2661–2687.
[4] J. Delery, Shock wave/turbulent boundary layer interaction and its control, Prog.
Aerosp. Sci. 22 (1985) 209–280.
[5] J. Delery, J.G. Marvin, Shock-wave boundary layer interactions, AGARD-AG-280,
1986.
[6] J. Haberle, A. Gulhan, Internal flowfield, investigation of a hypersonic inlet at
Mach 6 with bleed, J. Propuls. Power 23 (2007) 1007–1017.
[7] T.O. Hahn, T.I.-P. Shih, W.J. Chyu, Numerical study of shock-wave/boundary-
layer interactions with bleed, AIAA J. 31 (1993) 869–876.
[8] M.S. Holden, R.J. Nowak, G.C. Olsen, K.M. Rodrigues, Experimental studies of
shock wave–wall jet interaction in hypersonic flow, AIAA Pap. No. 90-0607,
1990.
[9] M.S. Holden, S.J. Sweet, Studies of transpiration cooling with shock interaction
in hypersonic flow, AIAA Pap. No. 94-2475, 1994.
[10] D. Mahapatra, G. Jagadeesh, Studies on unsteady shock interactions near a
generic scramjet inlet, AIAA J. 47 (2009) 2223–2232.
[11] S.G. Mallinson, S.L. Gai, N.R. Mudford, The interaction of a shock wave with a
93
laminar boundary layer at a compression corner in high-enthalpy flows includ-
ing real gas effects, J. Fluid Mech. 342 (1997) 1–35.
[12] D.C. McCormick, Shock/boundary-layer interaction control with vortex genera-
tors and passive cavity, AIAA J. 31 (1993) 91–96.
[13] R.J. Moffat, Describing the uncertainties in experimental results, Exp. Therm.
Fluid Sci. 1 (1988) 3–17.
[14] J.N. Moss, S. O’Byrne, N.R. Deepak, S.L. Gai, Simulations of hypersonic, high-
enthalpy separated flow over a ‘tick’ configuration, AIP Conf. Proc. 1501 (2012)
1453.
[15] S. Raghunathan, Passive control of shock–boundary layer interaction, Prog.
Aerosp. Sci. 25 (1988) 271–296.
[16] S.N. Ram, Measurement of static pressure over bodies in hypersonic shock tun-
nel using MEMS-based pressure sensor array, M.Sc. (Engg) Dissertation, Dept
of Aerospace Engineering, IISc, Bangalore, 2011.
[17] M.J. Rimlinger, T.I.-P. Shih, W.J. Chyu, Shock-wave/boundary-layer interactions
with bleed through rows of holes, J. Propuls. Power 12 (1996) 217–224.
[18] M. Rybalko, H. Babinsky, E. Loth, Vortex generators for a normal shock/
boundary-layer interaction with a downstream diffuser, J. Propuls. Power 28
(2012) 71–82.
[19] D. Schulte, A. Henckels, U. Wepler, Reduction of shock induced boundary layer
separation in hypersonic inlets using bleed, Aerosp. Sci. Technol. 2 (1998)
231–239.
[20] T.I.-P. Shih, M.J. Rimlinger, W.J. Chyu, Three-dimensional shock-wave/boundary-
layer interactions with bleed, AIAA J. 31 (1993) 1819–1826.
[21] L.J. Souverein, J.-F. Debieve, Effect of air vortex generators on a shock wave
boundary layer interaction, Exp. Fluids 49 (2010) 1053–1064.
[22] P.R. Viswanath, Shock-wave–turbulent-boundary-layer interaction and its con-
trol: a survey of recent developments, Sadhana 12 (1988) 45–104.
[23] A. Weiss, H. Olivier, Shock boundary layer interaction under the influence of a
normal suction slot, Shock Waves 24 (2014) 11–19.