Você está na página 1de 12

Three-Dimensional Modeling Shock-Wave

Interaction with a Fin at Mach 5

Amjad A. Pasha

Arabian Journal for Science and


Engineering

ISSN 2193-567X
Volume 43
Number 9

Arab J Sci Eng (2018) 43:4879-4888


DOI 10.1007/s13369-018-3210-6

1 23
Your article is protected by copyright and
all rights are held exclusively by King Fahd
University of Petroleum & Minerals. This e-
offprint is for personal use only and shall not
be self-archived in electronic repositories. If
you wish to self-archive your article, please
use the accepted manuscript version for
posting on your own website. You may
further deposit the accepted manuscript
version in any repository, provided it is only
made publicly available 12 months after
official publication or later and provided
acknowledgement is given to the original
source of publication and a link is inserted
to the published article on Springer's
website. The link must be accompanied by
the following text: "The final publication is
available at link.springer.com”.

1 23
Author's personal copy
Arabian Journal for Science and Engineering (2018) 43:4879–4888
https://doi.org/10.1007/s13369-018-3210-6

RESEARCH ARTICLE - MECHANICAL ENGINEERING

Three-Dimensional Modeling Shock-Wave Interaction


with a Fin at Mach 5
Amjad A. Pasha1

Received: 1 August 2017 / Accepted: 19 March 2018 / Published online: 27 March 2018
© King Fahd University of Petroleum & Minerals 2018

Abstract
The three-dimensional single-fin configuration finds application in an intake geometry where the cowl-shock wave interacts
with the side-wall boundary layer. Accurate numerical simulation of such three-dimensional shock/turbulent boundary-layer
interaction flows, which are characterized by the appearance of strong crossflow separation, is a challenging task. Reynolds-
averaged Navier–Stokes computations using the shock-unsteadiness modified Spalart–Allmaras model is carried out at Mach
of 5 at large fin angle of 23◦ . The computed results using the modified model are compared to the standard Spalart–Allmaras
model and validated against the experimental data. The focus of the work is to implement the modified model and to study
the flow physics in detail in the complex region of swept-shock-wave turbulent boundary-layer interaction in terms of the
shock structure, expansion fan, shear layer and the surface streamlines. The flow structure is correlated with the wall pressure
and skin friction in detail. It is observed that the standard model predicts an initial pressure location downstream of the
experiments. The modified model reduces the eddy viscosity at the shock and predicts close to the experiments. Overall,
the surface pressure using modified model has predicted accurately at all the locations. The skin friction is under-predicted
by both the models in the reattachment region and is attributed to the poor performance of turbulence models due to flow
laminarization.

Keywords High-speed flows · Shock wave · Turbulent boundary layer · Shock-unsteadiness · Separation bubble · Turbulence
modeling · Single fin · Compressible flows · Computational fluid dynamics

List of symbols
b1 Shock-unsteadiness damping parameter Subscripts
Cf Skin friction coefficient 0 Stagnation condition
cb 1 Shock-unsteadiness parameter n Normal to shock wave
M1n Upstream Mach number normal to shock w Wall condition
z 2+ Wall-normal distance to the nearest point ∞ Freestream condition
in wall coordinates
δ0 Boundary-layer thickness upstream of inter- Abbreviation
action CFL Courant-Friedrichs-Lewy
μT Eddy viscosity SA Spalart–Allmaras
ν Kinematic molecular viscosity
ν̃ Modified turbulent kinematic viscosity
1 Introduction

The most fundamental three-dimensional shock/boundary-


layer interaction is generated on a single-fin configuration. It
B Amjad A. Pasha consists of a flat plate with a sharp fin mounted perpendicular
aapasha@kau.edu.sa
to it. The oblique shock wave generated by the fin interacts
1 Department of Aeronautical Engineering, with the turbulent boundary layer on the plate and results
King Abdulaziz University, Jeddah 21589, Saudi Arabia in flow separation. The three-dimensional vortical flow thus

123
Author's personal copy
4880 Arabian Journal for Science and Engineering (2018) 43:4879–4888

generated alters the inviscid flow pattern. Additional shock of the experiments and under-predicted the skin friction in
waves, expansion regions and free shear layers are generated the reattachment region. Thivet [12] computed three differ-
that result in a complex flow in the interaction region. Practi- ent single-fin configuration cases with M∞ = 3, θ = 15◦ ,
cal applications of single-fin configuration include scramjet M∞ = 4, θ = 20◦ and M∞ = 4, θ = 30.6◦ , using the
inlets, where the oblique shock generated by the cowl inter- two-equation standard and modified k−ω turbulence mod-
acts with the side wall boundary layer [1]. els. The prediction of secondary vortex region was improved
The single-fin shock boundary-layer interaction flows are using modified k−ω model, hence improving the wall data
characterized by localized regions of increased pressure, skin in this region. The review articles [14–18] discusses different
friction and heat transfer rate. Prediction of these surface single-fin configurations flow physics and wall data in detail.
properties is important in the design of scramjet inlets. In The review articles discuss the experimental study for single-
addition, the flow distortion caused by the interaction can fin configurations with deflection angles of 8◦ , 12◦ , 15◦ , 16◦ ,
degrade the performance of these inlets significantly. There- 18◦ , 20◦ , 24◦ , 31◦ and Mach numbers in the vicinity of 3, 4,
fore, the aerodynamic loads generated from these interactions 5 and 8. The Reynolds number based on the boundary-layer
play a significant role in the structural integrity of hyper- thickness upstream of the interaction region lies in the vicin-
sonic vehicle [2,3]. Computational fluid dynamic approach ity of 2 × 105 for most of the cases. Surface oil flow patterns,
is a useful tool to understand the complex three-dimensional wall pressure and skin friction measurements have been stud-
flow pattern in these shock-wave/boundary-layer interactions ied. These configurations have been computed using RANS
and to predict its influence on the wall data. The direct method using algebraic, one- and two-equation turbulence
numerical simulation and large eddy simulation demand a models for predicting the surface pressure and skin friction.
large number of grid points at high Reynolds number flows The computations for 20◦ at M∞ = 4 showed that the k−ω
leading to large computational resources and computational model showed a delayed separation and higher values of
times to capture the fine features of shock-wave/turbulent wall pressure and Cf in the plateau of conical recirculation
boundary-layer interaction cases [4–7]. As an engineer- region [18]. The standard k− and Spalart–Allmaras mod-
ing approach, Reynolds-averaged Navier–Stokes (RANS) els show similar trend to k−ω model predictions [14,16].
method is applied in the present work along with one- Similar trend in wall data was observed by these turbulence
equation turbulence models to compute these flowfields. models for the weaker interaction case of 15◦ at Mach 3 [17].
Experiments and computations were carried by many The computations using RANS method for single-fin flows
authors [8–13] for single-fin geometry. Kubota et al. [8] with an angle of α = 20◦ and free stream Mach number
based on the experimental study, proposed a flow model of of 3.0 predicted some of the features of crossflow vortices
for single-fin geometry with deflection angle θ = 15◦ and and shock waves [15]. The three-dimensional contour plots
M∞ = 2.3. They showed that for this weak shock/boundary- were plotted using eigen values of the velocity gradient field.
layer interaction, a strong and stiff vortex is formed attached The flow consists of the vortical structure with the ellip-
to a fin and a weak stretched vortex is formed above it, tical cross section in the core of this cone. This vorticity
attached to the flat plate. As the interaction is weak, no sheet lifts up at separation line and forms the conical vor-
lambda shock was observed in this case. Alvi et al. [9] per- tex.
formed experiments for single fin with deflection angle θ = Several modifications have been proposed in the liter-
16◦ and M∞ = 2.95. The proposed flowfield model depicted ature [19], namely compressibility correction, realizability
an inviscid shock which bifurcated to form a separation shock constraint, rapid-distortion correction and length-scale cor-
and a rear shock, representing a lambda structure in a cross rection. The implementation of pressure gradient across
plane perpendicular to the flow direction. The separated con- shock waves to the Spalart–Allmaras model, modified wake
ical vortical flow region is formed underneath the lambda function to the Baldwin–Lomax model and limitation of pro-
shock structure. A slip line is generated from the triple point duction of turbulent kinetic energy to k− model is also used
of intersection of three shock waves, and a set of compression to improve the wall pressure in shock/boundary-layer inter-
and expansion waves are formed between the shear layer and action flows [20–22]. Their performance varied from one test
the slip line. Panaras [11,13] computed using RANS code for case to another.
deflection angle of θ = 20◦ and M∞ = 3.0. The wall pres- The shock-unsteadiness modified Spalart–Allmaras model
sure data were improved using modified Baldwin–Lomax of Sinha et al. [23] has shown potential in improving separa-
turbulence model, compared to the standard Baldwin–Lomax tion bubble prediction in two-dimensional and axisymmetric
model. Edwards et al. [10] studied the performance of four flows [23,24]. The application of shock-unsteadiness modi-
different one-equation turbulence models at M∞ = 8 and fication by Sinha et al. [23] to Spalart–Allmaras model was
θ = 15◦ . Among them, the standard Spalart–Allmaras model limited to simple compression corner geometry at supersonic
has shown to predict the surface properties close to the exper- Mach number of 2.8 at deflection angle 24◦ . The standard
iments. This model predicted flow separation downstream Spalart–Allmaras model predicted lower values of eddy vis-

123
Author's personal copy
Arabian Journal for Science and Engineering (2018) 43:4879–4888 4881

cosity and resulted in initial pressure location upstream of Deflection


experiments. Therefore, a larger separation bubble size is Fin angle, θ
wall Single−Fin o
predicted; consequently, the wall pressure and skin friction 23
are not predicted accurately as compared to the experiments. Inviscid
shock
The shock-unsteadiness modified Spalart–Allmaras model wave Fin−tip Flow
corrects the amplification of eddy viscosity across the shock.
The higher values of eddy viscosity predicted by the mod- z
ified model push the initial pressure location close to the x Origin
experiments and predict accurate separation bubble size. y

mm

152 m
Both the standard and modified Spalart–Allmaras models

mm
m

mm
mm
182

mm
162
under-predicted the skin friction coefficient in the reattach-

m
142

m
122

92 m
ment region as compared to the experiments. The modified

102

82 m
Turbulent
Spalart–Allmaras model showed higher values of skin fric- boundary
layer
tion as compared to the standard Spalart–Allmaras model. Sections Flat−plate
Later, Pasha et al. [24] applied the shock-unsteadiness
modified Spalart–Allmaras model to axisymmetric cone- Fig. 1 Three-dimensional single-fin configuration with a fin mounted
flare geometry at significantly higher Mach numbers of on the flat plate. The surface measurements [26] were taken along the
11–13 with two flare angles, 36◦ and 42◦ . The higher dashed lines
pressure rise across flare-shock results in largely separated
flows. In contrast, to simulate separated flows at supersonic
K and stagnation pressure P0 = 2.12 MPa were taken in
Mach numbers, it is comparatively difficult to simulate such
the reservoir of the Ludweig tube experimental facility. The
flows at hypersonic Mach flows. The computed results were
free stream temperature and pressure corresponding to their
compared to the experimental data. The standard Spalart–
stagnation values are T∞ = 68.3 K and p∞ = 4008.5
Allmaras turbulence model suppressed flow separation at
N/m2 . A fin of height = 100 mm normal to the flat plate
the corner and therefore did not reproduce the correct shock
is taken. The fin tip is placed at a distance of 286 mm
structure in the interaction region due to a high level of tur-
downstream of the flat plate edge. Both the flat plate and
bulence predicted at the shock wave. By comparison, the
fin wall are maintained under the isothermal condition of
shock-unsteadiness modified Spalart–Allmaras model damp-
300 K. The flow is turbulent, upstream of the interaction
ens the turbulence amplification at the shock. It thereby
region with a unit Reynolds number of Re1∞ = 37 × 106
reproduced the separation bubble size observed experimen-
m−1 . Wall data like pressure, skin friction and heat trans-
tally and matched the wall pressure measurements well. The
fer rates were measured at different cross sections in the
shock-unsteadiness model also predicted the flow topology
shock-wave/turbulent boundary-layer interaction region (see
in the interaction region correctly. The modification had a
Fig. 1). The undisturbed turbulent boundary-layer proper-
negligible effect on weak interactions, where the standard
ties were measured on the flat plate at a distance of 20 mm
Spalart–Allmaras model performed reasonably well.
upstream of the fin tip. The boundary-layer thickness δ of 3.8
In this paper, an attempt is made to extend the shock-
mm, displacement thickness = 1.6 mm, momentum thick-
unsteadiness modified Spalart–Allmaras to three-dimensional
ness = 0.16 mm and skin friction Cf = 1.35 × 10−3 is
single-fin configuration flows and study the flowfield in
measured.
detail. First, the test case is described, which is followed by
the simulation methodology. Next, the computed flowfield
using modified Spalart–Allmaras model is explained for the
strongest shock strength case with fin deflection angle of 23◦ 3 Simulation Methodology
and M∞ = 5. Next, the computed wall pressure and skin fric-
tion using modified model [23] and the standard model [25] The three-dimensional Reynolds-averaged Navier–Stokes
are compared with the experimental results [26]. equations [27] are used in the numerical simulations. The
turbulence model equations are fully coupled to the mean
flow equations. The shock-unsteadiness modified Spalart–
2 Test Case Allmaras model of Sinha et al. [23] and its standard ver-
sion [25] are used for calculating the eddy viscosity. A
The schematic of the single-fin configuration used in the compressible correction to standard Spalart–Allmaras model
experiments of Schulein [26] is shown in Fig. 1. The fin is has been proposed by Catris et al. [28] by modifying the
inclined with a deflection angle of θ = 23◦ to the flow direc- diffusion laws in the turbulence model, but this strategy com-
tion at M∞ = 5. The stagnation temperature of T0 = 410 plicates the numerical implementation for three-dimensional

123
Author's personal copy
4882 Arabian Journal for Science and Engineering (2018) 43:4879–4888

flows [29]. Therefore, the compressibility correction is not


used in the present work. The governing equations are dis-
o
cretized in a finite-volume formulation where the inviscid 23
fluxes are computed using a modified low-dissipation form of
the Steger–Warming flux-splitting approach. The method is Top
Fin−wall
Inlet
second-order accurate both in stream-wise and wall-normal
directions. The viscous fluxes and the turbulent source terms
are evaluated using central difference method. More details i k
of the numerical method are given in Ref. [30]. The code is j
Exit
capable of running on parallel machines and has been used Flow
Flat−plate
successfully in several supersonic and hypersonic applica-
tions [23,24,31,32].
The shock-unsteadiness modified Spalart–Allmaras model
of Sinha et al. [23] accounts for the effect of unsteady shock Extrapolation
motion in a steady mean flow. The shock-unsteadiness cor-
rection is achieved by adding a source term of the form Fig. 2 Computational domain with appropriate boundary conditions
−cb 1 ρ̄ ν̃ Sii in the transport equation for ρ̄ ν̃, where ρ̄ is mean
density, ν̃ is modified turbulent kinematic viscosity, Sii is
mean dilatation and shock-unsteadiness parameter wall boundary conditions identical to those listed above. The
formulation of two-dimensional code used in the flat plate
4 2  numerical simulation is similar to, as described in Sect. 3.
cb 1 = (1 − b1 ) − c1 . (1)
3 3 The value of the momentum thickness reported in the exper-
iments [26] is matched to obtain the mean flow and the
Note that this additional term is effective only in regions turbulence profiles at the inlet boundary of the computational
of strong dilatation and therefore does not alter the standard domain.
Spalart–Allmaras model elsewhere. Here, b1 is model param- A single-block grid is generated using a code, and a careful
 = 1.25+0.2(M −1). The model parameter b
eter and c1 1n 1 grid refinement study is performed by systematically varying
represents the coupling between the unsteady shock motion the number of grid points in each direction, as well as refining
and the upstream velocity fluctuations, and is given by, the cell size in critical regions. The origin is taken at a tip of
   the single-fin, and the grid is stretched exponentially in the
b1 = max 0, 0.4 1 − e1−M1n (2) upstream and downstream directions of origin with initial
grid size of 1 × 10−4 m. A structured mesh with exponential
It is zero for subsonic flows without shock waves, and it stretching normal to the plate and fin walls are used to span
reaches an asymptotic value of 0.4 for high Mach numbers. the computational domain. Wall pressure and skin friction
The detailed formulation of shock-unsteadiness modified coefficient are found to be sensitive to the computational
Spalart–Allmaras model and its implementation for two- mesh and are used to identify a grid converged solution. First,
dimensional compression corner and axisymmetric cone- the number of points in the wall parallel direction is refined
flare flows at supersonic and hypersonic Mach numbers is and it is observed that among the three grids, the 100 ×
presented in Refs. [23,24]. In the current work, the modifi- 110 × 110, 140 × 110 × 110, and 200 × 110 × 110, the
cation is applied to the three-dimensional flows which are last two grids match. Next, 140 × 110 × 110 grid is taken
more difficult to simulate as compared to their counterpart and only the distance of the first cell center from the wall
two-dimensional flows. is successively reduced by halve-times from its initial value
The computational domain and boundary conditions for of 2 × 10−6 m. Wall pressure and Cf variation indicate that
three-dimensional single-fin configuration are identified in 1 × 10−6 m is sufficient for an accurate solution. The z 2+
Fig. 2. The freestream conditions taken in the simulations varies between 0.06 in the undisturbed boundary layer to
are T∞ = 68.3 K and p∞ = 4008.5 Pa. At the fin wall a maximum value of z 2+ < 0.6 in the shock/boundary-layer
and flat plate, a constant wall temperature of 300 K and no- interaction region. This value of the z 2+ is sufficient to capture
slip velocity condition is applied. An extrapolation boundary the high gradients of the mean and turbulent variables in the
condition is assigned at top and exit planes. The freestream boundary layer in the wall-normal direction, especially at
and wall boundary conditions for the turbulence model vari- the reattachment region. In the next step, using 140 × 110 ×
ables are taken as ν̃∞ = 0.1ν∞ and (νT )w = 0. The inlet 110 grid, the number of points in the transverse direction is
profiles for the computations are obtained from separate increased. It is observed that variation in wall pressure and
two-dimensional flat plate simulation at the freestream and skin friction for 140 × 160 × 110 and 140 × 200 × 110 is less

123
Author's personal copy
Arabian Journal for Science and Engineering (2018) 43:4879–4888 4883

Fig. 3 Flow solution in a p/p∞


single-fin shock/boundary-layer
Fin-wall
interaction in terms of the 12
computed surface streamlines Planar 10
on the plate. The shock structure shock 7
is shown by pressure contours at 4
x-sections = 92 and 182 mm 1
Separation
shock
Reattachment
line (R1)
Flat-plate

Fin-tip
(origin)

Separation X

line (S1) Y

than 3%. Finally, the number of points in the flat plate normal Separation z/δ0 = 0.8
direction is increased to arrive at the 140 × 160 × 160 grid. Fin shock z/δ0 = 0.25
Surface properties variation shows that further refinement to wall
140 × 160 × 200 grid points yields less than 2% variation in
wall pressure and skin friction. A converged grid of 140 × Pressure
160 × 160 is obtained with 140 points along the stream-wise contours
Z

direction, 160 points transverse to the flow and 160 points


Y
normal to the plate. X

In the present computations, a CFL of 0.05 is used at the


beginning and it is gradually increased to 10 in the first 3500
iterations. It is further increased to 100 at 6000 iterations
and to 1000 at 8000 iterations. A maximum CFL of 5000 is Reattachment Separation
line Vortex Streamline
surfaces line
used after 10,000 iterations. The corresponding time steps core
vary from 2 × 10−12 s for the initial iterations to 2 × 10−5 s
at steady state solution. It takes 190 cpu-hours to reach the Fig. 4 Computed stream surfaces starting at z/δ0 = 0.8 and z/δ0 =
0.25 showing vortex region, and the pressure contours at x-section =
steady state solution in 30,000 iterations. The time step was 82 mm, as viewed from the downstream direction
calculated using the relationship τ = CFLz min /(u + a),
where z min is the minimum cell size and, u and a refer to
the local velocity and speed of sound [33]. tion shock. The flow separates at the primary separation
line S1 and attaches at the primary reattachment line R1
near the fin wall. The streamlines converge at separation
4 Flow Physics line S1, and the fluid moves upwards normal to the plate
and then turns in a counter clockwise direction to form a
In this section, the computed results using modified Spalart– helical flow as shown in Fig. 4. Two stream surfaces origi-
Allmaras model are discussed. An isometric view of the nating at z/δ0 = 0.8 and z/δ0 = 0.25 are shown, where z is
flow solution for the fin is shown in Fig. 3. The pressure the normal distance from the plate and δ0 is the undisturbed
contours are taken at two x-sections of 92 and 182 mm boundary-layer thickness. The fluid impinges on the plate at
to identify the shock structure. The flow pattern is indi- the reattachment line from the top, making the streamlines
cated by the streamlines taken in the cell adjacent to the diverge in either direction. The inviscid shock wave in Fig. 4
flat plate and behave similarly to the skin friction lines. bifurcates into a lambda structure and encloses the vortex
A planar shock wave generated by the fin interacts with region.
the turbulent boundary-layer on the plate. It separates the The computed stagnation pressure contours in Fig. 5
boundary-layer and results in the formation of the separa- indicate that an inviscid shock bifurcates and forms a sep-

123
Author's personal copy
4884 Arabian Journal for Science and Engineering (2018) 43:4879–4888

M
Inviscid shock 5.0
0.3 3.8
Rear 2.5
shock Triple point 1.3
Expansion 0.1
2.6 region
Separation
0.2 shock

z/x
2.3 2.8 Shear-layer
3.2 4.9
Fin 4.1
vortex 4. 4
0.1 3. 6 2.6 3.9
2.8
3.4
2.4 2.5
2.8

3.0
1.5 3.3 2.6 2.9
2.6
2.5 3.0 3.0 3.0 3.4 3.7 3.3 3.1 2.3
0
0.5 0.6 0.7 0.8 0.9
y/x
Fig. 5 Computed stagnation pressure contours at x-section = 122 mm Fig. 7 Enlarged region of computed Mach contours at x-section =
122 mm

near the corner region of fin wall and plate. The impinge-
ment of the supersonic jet (entropy layer) emanating from
the triple point of shock–shock interaction results in peak
values of pressure as indicated in the region marked in
Fig. 6. A small fin-vortex is formed when the fluid com-
ing from the inviscid region interacts with fin wall and
turns in the clockwise direction, as viewed from the down-
stream direction. Similar, corner-vortex was observed in the
vapor screen images of the experiments [8]. The schematic
sketches of these flowfield features are explained in detail in
Refs. [9,18].

Fig. 6 Enlarged region of computed pressure contours at x-section =


122 mm
5 Comparison of Computed Wall Data with
Experiments
aration shock and a rear shock, and results in the formation
of lambda structure in a cross plane perpendicular to the Figures 8 and 9 show the computed pressure and speed
freestream flow direction. The shear layer is formed over contours, overlapped with the wall pressure and skin fric-
the vortex region, which emanates from the separation point. tion on the flat plate at x-section = 122 mm. The distance
It interacts with the rear shock wave and then rolls up and along the y-axis is normalized with the corresponding x-
turns back to form a tongue. An entropy layer is gener- section distance measured from the fin tip. The surface data
ated from the triple point of the intersection of three shock are taken along the dashed lines as shown in Fig. 1. The
waves and a set of compression, and expansion waves are wall pressure remains constant in the undisturbed boundary-
formed between the shear layer and the slip line (see Fig. 7). layer before the interaction region. The separation shock
A secondary flow separation region was observed in the affects the upstream flow at the point of influence U, and
experiments [26], whereas in the present computation, it the wall pressure rises effectively across the separation
is not predicted. The computed pressure contours in Fig. 6 shock at primary separation point S1. It remains constant
indicate that an expansion fan is generated when the rear in the separated vortex flow and rises to peak values at pri-
shock wave reflects from the shear layer. The computed mary reattachment R1. The wall pressure then decreases
Mach contours in Fig. 7 shows that a type-IV shock-shock away from the reattachment region and rises near the fin-
interaction [34] is observed at the triple point. An alter- plate junction. The skin friction does not vary significantly
nate increase and decrease in Mach contours in the region in the unperturbed boundary layer before the region of
between shear layer and jet represents the weak compres- influence U. The boundary layer is compressed across the
sion and expansion waves. The flow remains supersonic in separation shock wave, and hence it increases the velocity
the separated region. Small subsonic pockets are observed gradient and thereby increases skin friction. The skin fric-

123
Author's personal copy
Arabian Journal for Science and Engineering (2018) 43:4879–4888 4885

delayed separation at y/x = 1.14. A similar trend is observed


by the standard model at all other x-sections as shown in
Table 1. A vortex region is calculated based on the distance
between S1 and R1. The standard Spalart–Allmaras model
predicts a small vortex region of 74 mm as compared to the
experimental value of 82 mm. There is a difference of 11%
in the vortex region between the Spalart–Allmaras model
and the experimental value. The skin friction coefficient pre-
dicted by the standard model in Fig. 11 matches close to the
experiments, except in the secondary flow separation region.
The model also under-predicts the peak values in the reat-
tachment region.
The shock-unsteadiness parameter cb 1 as described in
Eq. 1 is a function of the upstream normal Mach num-
Fig. 8 Computed pressure contours overlapped with wall pressure at ber M1n and needs to be evaluated at each shock wave
x-section = 122 mm
so as to implement the shock-unsteadiness correction. For
two-dimensional compression corner flows [23] and axisym-
metric cone-flare flows [24], an average value of M1n was cal-
culated and based on it cb 1 was calculated and implemented.
This is the drawback and limitation of the shock-unsteadiness
Spalart–Allmaras model. The three-dimensional shock struc-
ture in the single-fin configuration is quite complex. It is not
easy to find the orientation of the different shock waves and
the inclination of the upstream flow at each shock. There-
fore, it is a difficult task to calculate the mean value of M1n .
An alternate approach is to use different values of cb 1 in the
current single-fin case to improve the flow predictions. A
higher magnitude of cb 1 yields a larger separation, and cb 1 =
− 0.2 is found to match the experimental separation location
closely and therefore is used in the present simulations. A
similar approach by Gaitonde et al. [22] was used for sim-
Fig. 9 Computed speed contours overlapped with skin friction coeffi-
cient at x-section = 122 mm ulating three-dimensional double-fin shock/boundary-layer
interaction. The turbulence model constants were varied by
limiting the value of production term in standard k- turbu-
lence model. A smaller value of model constant resulted in
lower turbulent kinetic energy. Hence, the computed solution
resulted in a larger separated region and matched well with
the experimental flowfield and wall pressure.
In the present case, a shock-unsteadiness parameter of
cb 1 = − 0.2 is chosen to yield a larger separation and is
found to match the experimental initial pressure rise location
closely as indicated in Fig. 10. The shock-unsteadiness cor-
rection reduces the turbulent eddy viscosity in the region of
Fig. 10 Comparison of computed wall pressure at x-section = 152 mm the separation shock. This causes an increase in the length
with experiments [26] using standard Spalart–Allmaras model [25] and
modified Spalart–Allmaras model [23] of the separation shock and hence brings the separation
point predictions close to the experiments. The same trend is
observed in the axisymmetric flows over cone-flare cases at
tion reaches a peak value at the reattachment point R1 due hypersonic Mach numbers in Ref. [24].
to the high values of the velocity gradient and hence shear Figures 12 and 13 indicate that the modification predicts
stress. lower values of μT /μ∞ in the region of y/x  1.2, as com-
Figure 10 indicates that at x-section = 152 mm, exper- pared to the standard Spalart–Allmaras model. Hence, the
iments give an initial pressure rise location at y/x = 1.22, modified model improves the initial pressure rise location S1
where as the standard Spalart–Allmaras model predicts a (see Fig. 10) to around 6%. Also, the pressure distribution

123
Author's personal copy
4886 Arabian Journal for Science and Engineering (2018) 43:4879–4888

Table 1 Comparision of
x-section (mm) S1 R1
primary separation point S1 and
reattachment point R1 using Experiment SA Modified SA Experiment SA Modified SA
standard Spalart–Allmaras (SA)
and shock-unsteadiness 82 1.3 1.24 1.3 – 0.5 0.5
modified SA model with the 92 1.27 1.22 1.27 – 0.51 0.51
experiments [26] at different 122 1.23 1.16 1.21 – 0.51 0.51
x-sections on the flat plate
152 1.22 1.14 1.2 0.52 0.52 0.52
162 – 1.13 1.18 – 0.51 0.51
182 – 1.12 1.15 0.52 0.52 0.52

R1. Panaras [21] attributed this under-prediction of skin fric-


tion due to the poor performance of turbulence models. The
models indicated lower values of turbulence inside the sepa-
ration vortices, making the flow almost laminar in this region.
Also, it is the incapability of one-equation model to capture
the peak velocity gradients in the reattachment region, though
you put more number of grid points in this region to capture
it. Therefore, both the models underestimate the skin friction
than the experimental results. Similar observation was made
in the numerical simulations of Edwards et al. [10] where the
Fig. 11 Comparison of computed skin friction at x-section = 122 mm
standard Spalart–Allmaras model under-predicted the skin
with experiments [26] using standard Spalart–Allmaras model [25] and
modified Spalart–Allmaras model [23] friction in the reattachment region for flow over single-fin
configuration with Mach 8 and θ = 15◦ . More advanced
500 μT/μ∞ two-equation turbulence models [35–37] can be applied to
0.2
14000 capture the velocity gradients and hence predict the peak skin
9500
friction at the reattachment region accurately. Currently, this
z/x

5000
500 is beyond the scope of work. The modified Spalart–Allmaras
0.1
00 model predicts vortex region length between S1 and R1 to
70 14500 4500
500
2000 10500 1000
2500 4000 8500
3500
6000 2500 be 79 mm with a difference of 5% to the experimental value.
0.6 0.8 1 1.2 The computed locations of primary separation point S1 and
y/x
reattachment point R1 are compared with the experimental
Fig. 12 Computed normalized eddy-viscosity contours at x-section = data at different x-sections in Table 1. Overall, the modified
122 mm using standard Spalart–Allmaras model [25] Spalart–Allmaras model matches to the experimental data at
all locations.
500 μT/μ∞
14000
0.2 9500
z/x

5000

0.1
500 6 Conclusion
00
65 14500 4500
500 1000
2000 10500
2500 3000 8500 6000 2000 Reynolds-averaged Navier–Stokes based computations were
3000
0.6 0.8 1 1.2 carried out to investigate the three-dimensional shock/boundary-
y/x
layer interaction in a single-fin configuration at Mach 5 with a
Fig. 13 Computed normalized eddy-viscosity contours at x-section large deflection angle of 23◦ . The shock-unsteadiness modi-
= 122 mm using shock-unsteadiness modified Spalart–Allmaras
model [23]
fied Spalart–Allmaras model and its standard version are used
in the computations. The inviscid shock generated by the fin
interacts with the boundary layer on the adjacent flat plate
is well predicted in the reattachment region and the corner and results in a formation of a complex region. The viscous
region of the plate fin junction by the modified model as effects cause the bifurcation of inviscid shock and result in
compared to the standard model. the formation of a lambda shock structure, one edge being the
Figure 11 depicts that the modified model over predicts the separation shock and the other being the rear shock. Type-IV
skin friction between y/x = 0.65 and 0.72, whereas it under- shock-shock interaction results from the interaction of these
predicts the skin friction by 42% at the reattachment region shock waves. The lambda shock encloses a cross flow coni-

123
Author's personal copy
Arabian Journal for Science and Engineering (2018) 43:4879–4888 4887

cal vortex. A shear layer emanates from the separation point 10. Edwards, J.R.; Chandra, S.: Comparison of eddy viscosity-
and interacts with the rear shock wave and then rolls up and transport turbulence models for three-dimensional shock-separated
flowfields. AIAA J. 34(4), 756–763 (1996)
turns back to form a tongue. An entropy layer is generated 11. Panaras, A.G.: The effect of the structure of swept-shock-
at the triple point, and a set of compression and expan- wave/turbulent boundary-layer interactions on turbulence model-
sion waves are embedded in it and the shear layer. These ing. J. Fluid Mech. 338, 203–230 (1997)
flow features influence the wall pressure and skin friction 12. Thivet, F.: Lessons learned from RANS simulations of shock
wave/boundary layer interactions. AIAA Paper, p. 583 (2002)
and a correlation between them is explained. The standard 13. Panaras, A.G.: Calculation of flows characterized by extensive
Spalart–Allmaras model predicts initial pressure location crossflow separation. AIAA J. 42(12), 2474–2475 (2004)
downstream of experiments. The shock-unsteadiness cor- 14. Delery, J.; Marvin, J. G.; Reshotko, E.: Shock-wave boundary layer
rection leads to an improvement in prediction of the initial interactions. AGARDograph No. 280. ISBN 92-835-159-6 (1996)
15. Panaras, A.G.: Review of the physics of swept-shock/boundary
pressure rise location. This leads to an accurate prediction of layer interactions. Prog. Aerosp. Sci. 32, 173–244 (1996)
vortex size and hence the shock structure. The skin friction is 16. Knight, D.D.; Degrez, G.: Shock wave turbulent boundary layer
under-predicted at reattachment region in comparison with interactions in high mach number flows—a critical survey of
experiments by both the modified model and its standard ver- current numerical prediction capabilities. AGARD Advis. Rep.
319(2), 1.1–1.35 (1998)
sion. This behavior is attributed to the poor performance of 17. Knight, D.; Yan, H.; Panaras, A.G.; Zheltovodov, A.: Advances in
these models due to the prediction of laminar flow in this CFD prediction of shock wave turbulent boundary layer interac-
region. Further improvements to the present computations tions. Prog. Aerosp. Sci. 39(2–3), 121–184 (2003)
are envisaged by simulating the more advanced two-equation 18. Babinsky, H.; Harvey, J.K.: Shock Wave-Boundary-Layer Interac-
tions. Cambridge University Press, Cambridge (2011)
turbulence models. 19. Roy, C.J.; Blottner, F.G.: Review and assessment of turbulence
models for hypersonic flows. Prog. Aerosp. Sci. 42(7–8), 469–530
Acknowledgements The author would like to thank Prof. Krishnendu (2006)
Sinha from Department of Aerospace Engineering, Indian Institute of 20. Ma, L.; Lu, L.; Fang, J.; Wang, Q.: A study on turbulence
Technology Bombay, for his valuable guidance, help and selflessness transportation and modification of Spalart–Allmaras model for
time investment to accomplish this work. I would like to thank our Chair- shock-wave/turbulent boundary layer interaction flow. Chin. J.
man Dr. Khalid A. Juhany and Prof. Mahmood Khalid for supporting Aeronaut. 27(2), 200–209 (2014)
this work. 21. Panaras, A.G.: Turbulence modeling of flows with extensive cross-
flow separation. Aerospace 2(3), 461–481 (2015)
22. Gaitonde, D.V.: Progress in shock-wave/boundary-layer interac-
tions. Prog. Aerosp. Sci. 721, 80–99 (2015)
23. Sinha, K.; Mahesh, K.; Candler, G.V.: Modeling the effect of shock-
unsteadiness in shock/turbulent boundary-layer interactions. AIAA
References J. 43(3), 586–594 (2005)
24. Pasha, A.A.; Sinha, K.: Shock unsteadiness model applied to hyper-
1. Van Wie, D.M.: Scramjet inlets. In: Curran, E.T., Murthy, S.N.B. sonic shock wave/turbulent boundary-layer interactions. J. Propul.
(eds.) Scramjet Propulsion, Progress in Astronautics and Aeronau- Power 28(1), 46–60 (2012)
tics, pp. 447–511. Institute of Aeronautics and Astronautics Inc., 25. Spalart, P. R.; Allmaras, S. R.: A one-equation turbulence model
Reston (2000) for aerodynamic flows. AIAA Paper, p. 439 (1992)
2. Bose, D.; Brown, J.L.; Prabhu, D.K.; Gnoffo, P.; Johnston, C.O.; 26. Schulein, E.: Optical skin friction measurements in short-duration
Hollis, B.: Uncertainty assessment of hypersonic aerothermody- facilities. AIAA J. 44(8), 1732–1742 (2006)
namics prediction capability. J. Spacecr. Rockets 50(1), 12–18 27. Wilcox, D.C.: Turbulence Modeling for CFD, 2nd edn, pp. 491–
(2003) 492. DCW Industries, La Canada (2000)
3. Marvin, J. G.; Brown, J. L.; Gnoffo, P. A.: Experimental database 28. Catris, S.; Aupoix, B.: Density corrections for turbulence models.
with baseline CFD solutions: 2-d and axisymmetric hyper- Aerosp. Sci. Technol. 4(1), 1–11 (2000)
sonic shock-wave/turbulent boundary-layer interactions, NASA 29. Deck, S.; Duveau, P.; d’Espiney, P.; Guillen, P.: Development and
TM2013216604 (2013) application of Spalart–Allmaras one-equation turbulence model to
4. Garnier, E.; Adams, N.; Sagaut, P.: Large Eddy Simulation for three-dimensional supersonic complex configurations. Aerosp. Sci.
Compressible Flows. Springer, Berlin (2009) Technol. 6(3), 171–183 (2002)
5. Georgiadis, N.J.; Yoder, D.A.; Vyas, M.A.; Engblom, W.A.: Sta- 30. Sinha, K.; Candler, G. V.: Convergence improvement of two-
tus of turbulence modeling for hypersonic propulsion flowpaths. equation turbulence model calculations. AIAA Paper p. 2649
Theor. Comput. Fluid Dyn. 28(3), 295–318 (2014) (1998)
6. Yang, Z.: Large-eddy simulation: past, present and the future. Chin. 31. Pasha, A.A.; Sinha, K.: Shock-unsteadiness model applied to
J. Aeronaut. 28(1), 11–24 (2015) oblique shock-wave/turbulent boundary-layer interaction. Int. J.
7. Fang, J.; Yao, Y.; Zheltovodov, A.A.; Lu, L.: Investigation of Comput. Fluid Dyn. 22(8), 569–582 (2008)
three-dimensional shock wave/turbulent-boundary-layer interac- 32. Pasha, A.A.: Study of parameters affecting separation bubble size
tion initiated by a single fin. AIAA J. 55(2), 509–523 (2017) in high speed flows using k-ω turbulence model. J. Appl. Comput.
8. Kubota, H.; Stollery, J.: An experimental study of the interaction Mech. 4(2), 95–104 (2018)
between a glancing shock wave and a turbulent boundary layer. J. 33. Nompelis, I.: Computational study of hypersonic double-cone
Fluid Mech. 116, 431–58 (1982) experiments for code validation. Thesis (Ph.D.), University of Min-
9. Alvi, F.S.; Settles, G.: Physical model of the swept shock nesota (2004)
wave/boundary layer interaction flowfield. AIAA J. 30(9), 2252– 34. Edney, B. E.: Anomalous heat transfer and pressure distributions
2258 (1992) on blunt bodies at hypersonic speeds in the presence of an imping-

123
Author's personal copy
4888 Arabian Journal for Science and Engineering (2018) 43:4879–4888

ing shock. FFA Rept. 115, The Aeronautical Research Institute of 37. Roy, S.; Pathak, U.; Sinha, K.: Variable turbulent Prandtl number
Sweden, Stockholm (1968) model for shock/boundary-layer interaction. AIAA J. 56(1), 342–
35. Quadros, R.; Sinha, K.; Larsson, J.: Turbulent energy flux generated 355 (2018)
by shock/homogeneous turbulence interaction. J. Fluid Mech. 976,
113–157 (2016)
36. Quadros, R.; Sinha, K.: Modelling of turbulent energy flux in
canonical shock-turbulence interaction. Int. J. Heat Fluid Flow 61,
626–635(2016)

123