Flight Simulation and Unsteady Numerical Drag Prediction for a Pitching-

Accelerating Non-winged Hypersonic Reentry Vehicle

Mostafa Khalil1, Mohamed S. Abdel-Gawad2, Mahmoud Y. M. Ahmed3

Aerospace Engineering Department, Military Technical College, Cairo, Egypt

Abstract:
A non-winged reentry vehicle, such as the upper stage of a ballistic missile, endures harsh
aerodynamic loads during the phase of reentry. This phase is characterized by sophisticated flight phenomena
including hypersonic speed, continuous variation of surrounding atmospheric conditions, and pitching
unsteady vehicle motion. During flight in this descent phase, drag is considered as the dominant aerodynamic
force which accurate prediction is crucial as far as reliable trajectory prediction and mission planning are
concerned.
This study aims to investigate drag on a generic bi-cone non-winged reentry vehicle resembling an
upper stage of a ballistic missile during descent. This is done by using both trajectory prediction and
numerical simulation of vehicle-air interaction. To obtain the vehicle dynamics and instantaneous flight
conditions through descending phase, a three-degree-of-freedom 3DOF pitch flight trajectory model is
utilized starting from point of separation. It was concluded that the vehicle experiences small-amplitude
pitching motion while descending from about 90 km to 30 km and accelerating in hypersonic regime (Mach
5.7: 7). Drag is then calculated via time-dependent numerical simulations using a commercial CFD solver.
The results show the flow field pattern evolution in addition to the temporal variation of drag on the vehicle
during the flight phase in concern.

I. Introduction:
An upper stage of a ballistic missile, that is designed to deliver its payloads at long ranges, is a re-
entry vehicle by definition. During the descent phase towards a target, this hypersonic vehicle needs to
accurately follow its predefined trajectory using on-board guidance, navigation, and control (GNC) devices.
One fundamental input to GNC is the aerodynamic characteristics of the vehicle. While lift plays a minor role
for such un-winged non-lifting vehicle, drag appears to be the dominant aerodynamic force acting on the
vehicle. An accurate estimation of drag on a reentry vehicle should account for a multitude of complexities.
The complexities of flow around the descending re-entry vehicles are owed to the continuous and sharp
variation in atmospheric conditions. Air pressure and density increase while temperature profile varies from
one layer to another. The flow changes from free molecular, to slip flow, and eventually to continuum flow
depending on Knudsen number [1,2]. In addition, the freestream speed increases as the vehicle accelerates
during descent while performing pitching unsteadiness. Finally, the flow around the vehicle is hypersonic
invoking extreme temperatures and heat flux, strong shock waves, and shock-boundary layer interaction.
The focus of previous studies on hypersonic vehicles is more oriented towards steady flight
conditions. For instance, the comprehensive database by Weiland [3] includes steady aerodynamic
characteristics of a wide variety of reentry vehicle configurations. On the other hand, studies dedicated to
understanding the unsteady aerodynamic features around complete missile and projectile configurations at
supersonic and low hypersonic speeds are available in the open literature. In the majority of these studies, the
focus was made on one form of unsteadiness. Due to the unavailability or high cost of wind tunnel
measurements, numerical simulation based on the CFD technique was adopted in these studies as a suitable
alternative to investigate hypersonic reentry vehicles. Numerical simulation tools based on Navier-Stokes
equations assume that the flow around the vehicle is a continuum flow of real gas; an assumption that applies
for flight altitudes below approximately 90 km in the standard atmosphere [4].

1 Lecturer of aerodynamics and flight mechanics

2 Researcher
3 Associate professor of aerodynamics
 For instance, pure pitching of hypersonic vehicles of various shapes was studied in [5-14], whereas
phugoid, flapping, and rolling unsteadiness were separately examined in [14], [15], and [16], respectively. In
contrast, the impact of pure longitudinal acceleration on hypersonic vehicle characteristics in Mach 1 to 5
freestream was addressed numerically by Gledhill et al. [17]. Yawing-spinning motion of hypersonic
projectiles with Mach numbers of 2.89 and 4.49 at a constant altitude was numerically studied by Cayzac et
al. [18]. In contrast, pitching-yawing for spinning finned projectile was examined by Sahu [19] at Mach 3 and
a spin rate of 2500 rad/sec. Liu et al. [20] investigated numerically the static and dynamic stability derivatives
of hypersonic waverider performing both pitching oscillation and longitudinal acceleration at constant flight
altitude and Mach.
To the authors’ best knowledge, unsteady aerodynamic characteristics of an un-winged hypersonic
vehicle in a real descent trajectory involving variable freestream conditions, longitudinal acceleration, and
pitching were not discussed in the open literature. This shortage is the main motivation of running research
conducted by the authors using numerical simulation taking into consideration all flight complexities stated
above. The objective is to understand the aerodynamics and estimate the drag acting on a hypersonic reentry
vehicle that has the form of a bi-cone representing the upper stage of a long-range ballistic missile. After
separation, the vehicle flies freely performing a combined pitching oscillation and longitudinal acceleration
during descent in a dense atmosphere with flight Mach numbers varying from 5.7 to 7. The study is conducted
by numerically simulating the unsteady continuum flow around the vehicle. Vehicle flight performance and
the corresponding freestream conditions are obtained by solving the trajectory problem from the point of
separation till impact on the ground.
The remainder of the paper is organized as follows. Aspects of the case study and research
methodology are explained in the next section. Discussions of the main findings follow and the paper wraps
up with the key conclusions.

II. Case Study:

The hypersonic vehicle in concern is a non-winged reentry vehicle representing the upper stage of a typical
ballistic missile [21]. It has the form of a bi-cone with a multi-step tip and an overall fineness ratio of 2.5 and
its gravity center is located 1.13D from its tip; the configuration and relative dimensions are shown in Figure
1. The carrier missile is a multi-stage ballistic missile that is launched vertically to hit remote ground
stationary targets. After separation of the booster and sustainer stages of the missile, the unpowered upper
stage continues its flight separately till the ground impact. The height, speed, and flight path angle of the
vehicle at the point of separation are approximately 39.3 km, 2020 m/s and 40°, respectively.

Figure 1 Configuration and dimensions of the case study model

III. Flight trajectory model:

The hypersonic vehicle separates during the missile ascent and continues a ballistic trajectory up to the
summit and then descends towards the target. Separation conditions are adopted as initial flight conditions for
trajectory prediction. A computer code is developed based on the (3 DOF-pitch) equations to estimate the full
unpowered trajectory of the vehicle from the separation point to the ground impact. In the model, as
illustrated in Figure 2, (X, Z) is the inertial frame of the pitch plane located at the launch site, and ( x, z) is the
body frame located at the vehicle center of gravity CG. The Earth is approximated to be flat and non-rotating.
The equations describing the unpowered vehicle motion in pitch plane affected by aerodynamic drag, lift,
pitching moment, and gravity are stated as [22],
 m V̇ =−mg sin γ −¿ q ∞ S ref C D ¿ (1)

α
mV γ̇ =−mgcosγ + q∞ Sref C L α (2)

[
I yy θ̈=q ∞ S ref Lref C αm α +C qm ( )]
˙
θd
V
(3)

Z
x

V
q
z a

O X

Figure 2 3-DoF flight model in the pitch plane

where V is the vehicle total velocity relative to the air stream. m , I yy , Sref , and Lref are the vehicle mass,
mass moment of inertia, reference area, and reference length, respectively. g and q ∞ are the gravitational
α
acceleration and freestream dynamic pressure, respectively. C D , and C L are the drag and lift slope
coefficients acting on the vehicle, respectively. , , and are the angle of attack, pitch angle, and flight-path
angle of the vehicle, respectively. γ̇ and θ̈ are the flight path rate and pitch angular acceleration, respectively.
q α
C m is the change in the pitching moment coefficients with respect to the pitching velocity, q , whereasC m is
the change in the pitching moment coefficients with respect to the angle of attack. Aerodynamic coefficients
and derivatives input to the trajectory model above are calculated using an engineering technique developed
by the research group [23] that was confirmed to yield reliable results [24].

As shown in Figure 3, the vehicle flight performance including the flight altitude and velocity as a function of
flight time are illustrated. The flight time zero denotes the instant of the vehicle separation. According to the
flight simulation results, the vehicle takes about 81s through the dense atmosphere (85-90 km altitude [25])
till the impact with the ground. The evolution of vehicle attitude pitch and flight path angles and the
corresponding pitch rate is is illustrated in Figure 4 over the entire flight time.

Figure 3 Vehicle flight performance from separation to ground impact

 Pitch rate [deg/s]
Pitch angle [deg]
-20 1.5
Flight path angle -25 1
-30 0.5
-35 0
-40 -0.5
-45 -1
Gama
[deg]

-50 -1.5
Theta
-55 -2
228 238 248 258 268 278 288 298 308
Flight time [s]

Figure 4 Vehicle flight dynamics through the descent in the dense atmosphere
The dashed frame in Figure 3 highlights the period of the trajectory of interest in the present study. This
period is characterized by the accelerating descent in the dense atmosphere, i.e., the vehicle velocity increases
while flying towards the ground. This period spans 51 seconds starting from the altitude of 87 km to the
altitude of 29.4 km. During the flight in the period in concern, the temporal variation in vehicle kinematics,
freestream dimensionless parameters, and attitude analysis are shown in Figure 5a, b, and c, respectively.

AOA [deg]
Velocity [m/s]

2050 4
Velocity
1950 0

1850 -4

1750 -8
228 233 238 243 248 253 258 263 268 273 278
Flight time [s]
(a)
Reynolds no.

Mach no.
1E+7 7.2
1E+6 6.9
1E+5 6.6
1E+4 6.3
1E+3 6
Re
1E+2 5.7
228 233 238 243 248 253 258 263 268 273 278
Flight time [s]
Pitch angle [deg]

(b)
Pitch rate [deg/s]

-20 1
Flight path angle

-25 0.5
-30 0
-35 -0.5
-40 -1
Gama Theta
[deg]

-45 -1.5
228 233 238 243 248 253 258 263 268 273 278
Flight time [s]
(c)
Figure 5 Results of flight trajectory simulation through the concerned period of study

III. Setup for numerical simulations and drag calculation:

A. Design of numerical simulation cases:
As inferred from the trajectory results above, the unsteadiness encountered by the vehicle is a combined
longitudinal acceleration and pitching while simultaneously descending. Longitudinal acceleration has an
average value of 5 m/s 2, while the pitching has an average period of 17 s and amplitude of about 3⁰. The full
pitching cycle starts at 262 s from separation as the vehicle altitude gets below 87 km; Figure 4.
Numerical simulation is planned to understand the aerodynamic behavior of the vehicle under acceleration,
and pitching while descending. For the sake of a better understanding of physics, two simpler subcases are
planned. In the first subcase (case#1), the vehicle is assumed to perform pure longitudinal acceleration while
descending for the entire period in concern (51 seconds). In the other subcase (case#2), the vehicle is assumed
to perform pure pitching while flying in cruise conditions (i.e. constant altitude and speed) for the whole
pitching period of 17 seconds. For the main case in which a combined acceleration-pitching is performed
(case#3), a representative period is simulated to economize the simulation budget. During the three-second
period starting at the altitude of 38 km, the vehicle performs one-quarter of the pitching cycle along with
longitudinal acceleration. This period is marked by the dashed frame in Figure 5 and is better illustrated
separately in Figure 6 below. Details of these three cases are listed in Table 1 below. It is worth mentioning
here that the results of case#1 were briefly reported in [26] and are presented here for the sake of
completeness. In addition to these three unsteady simulation cases, a set of steady simulations (case#4 to
case#12) are planned for the sake of comparison and analysis. Table 2 explains the setup for steady simulation
cases.
Mach no.

P. [Pa]
6.5 640

6.4 540

6.3 440

6.2
M P_inf 340
273 273.5 274 274.5 275 275.5 276
Flight time [s]

(a) Freestream Mach and pressure

AOA [deg]
Velocity [m/s]

2025 1
2020 0
2015 -1
2010 -2
2005 -3
273 273.5 274 274.5 275 275.5 276
Flight time [s]

(b) Vehicle velocity and incidence angle AOA

Figure 6 Temporal variation of vehicle kinematic and freestream conditions for the main case

Table 1 Design of unsteady numerical simulation cases

Operating conditions
Physical
Case # Case definition Altitude Incidence
time [sec] Mach
[km] angle [deg]

1 Pure linear acceleration while the descent 51 5.7 : 7 87 : 30 0

2 Pure pitching at cruise conditions 17 6 60 -3 : 3

3 Combined acceleration-pitching while descent 3 6.2 : 6.5 38 :34 0:3

Table 2 Design of steady numerical simulation cases

Operating conditions Operating conditions

Case # Incidence Case #
Mach Altitude [km] Mach Altitude [km] Incidence
angle [deg] angle [deg]
4 6.5 70 0 8 6 60 0
5 6 60 9 1
6 5.7 52 10 2
 7 6.5 33 11 3
12 4

B. Computational domain definition:

For zero-incidence cases, an axisymmetric 2D computational domain is constructed. Domain configuration,
extents, and boundary definitions are illustrated in Figure 7a. The domain is divided into nine blocks at each
discontinuity on the vehicle contour to generate a structured grid. The discretized domain in the vicinity of the
vehicle is shown in Figure 7b.

(a) Shape, extents, and boundary definitions, L is the model length

(b) Discretized domain

Figure 7 Features of the axisymmetric computational domain

The spatial resolution quality is assessed through grid sensitivity analysis. Four grids with different
resolutions are generated to simulate the flow around the vehicle at a representative case namely, Mach 6
freestream flow at standard sea sea-level conditions. The total drag coefficient is taken as a measure for of
solution quality; the results are shown in Figure 8. Based on these results, grid 3 with 91100 cells is adopted.
Drag coeff.

0.14
0.135 Chart Title
0.13
0.125
0.12
0.115
0.11
1 2 3 4
Grid ID
Figure 8 Grid sensitivity check results
For non-zero incidence cases, a half-full three-dimensional computational domain is constructed of two
interfaced semi-spherical subdomains inside each other such that their centers coincide at the vehicle gravity
center. The interface between the two domains permits the sliding of the inner grid with respect to the outer
one to perform the pitching motion. Figure 9a shows the shape and boundary conditions for the computational
domain. The inner domain is divided into eighteen blocks at each discontinuity on the vehicle contour while
the outer domain is divided into five blocks to generate a structured grid. First cell heights and cell counts
normal to the solid wall are taken identical to those in grid 3 used in the 2D computational domain which
quality is assured through grid sensitivity checks above. Figure 9b below shows the discretized domain as
well as a zoom-in on the vehicle.
C. Solver setup and boundary conditions:
A commercial CFD solver is used [27] in which the transient density-based, implicit, double precision
second-order accurate technique is implemented. Air is treated as an ideal gas while the laminar-turbulent
transitional flow is handled using the intermittency transition shear-stress transport (SST) k-𝝎 turbulence
model available in the solver. The vehicle surface is treated as an adiabatic no-slip wall. At the inlet
boundary, freestream Mach and static pressure and temperature are defined. At the exit, the freestream static
pressure and total temperature are specified. The boundary conditions for the freestream flow vary from one
case to the other according to Tables 1 and 2. For all steady cases, freestream conditions are directly defined
whereas for transient cases, user-defined functions are used.

(a) Shape, extents, and boundary definitions

(b) Discretized domain features

Figure 9 Features of the axisymmetric computational domain

For case#1, a user-defined function, UDF, is developed to account for the temporal variation in atmospheric
conditions and Mach number during accelerating descent based on trajectory results, Figure 5. For case#2,
another UDF is developed to account for the pitching motion of the inner domain with respect to the fixed
outer one. For the main case, case#3, the two UDFs are implemented to simultaneously account for
longitudinal acceleration, pitching, and descent.
D. Time step sensitivity analysis:
The temporal resolution of the transient simulations is decided based on time sensitivity checks in which
acceleration and pitching are treated separately since each has a separate UDF. For the pure acceleration case
(case#1), three-time step sizes namely 0.1, 0.01, and 0.001 of a second are examined and temporal variation
of vehicle total drag coefficient is adopted as a measure of solution quality. Based on the results shown in
Figure a, a time step size smaller than 0.01 seconds yields an insignificant effect on the solution quality. So, a
time step of 0.01 s is utilized to be used inthrough the simulation of case #1 simulation.
For the pure pitching case (case#2), four different time step sizes namely 0.1, 0.04, 0.02, and 0.01 of the
pitching period (17 seconds) are examined. Figure b shows the results of time step sensitivity analysis for
pure pitching based on the proposed approach. A time step size of 0.01 of the period ( i.e. 0.17 seconds) is
chosen for case#2. For the main case (case#3), the smaller time step size namely, 0.01 seconds is adopted for
Drag coefficient

case#3 simulation.

0.3104
0.3102
0.3100
0.3098
Δt=(1/10)
0.3096
sec
0.3094
1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2

Simulation time [s]

(a) Pure acceleration case
AOA [rad]

0.08 T/
10
T/
0.04 25

-0.04

-0.08
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Simulation time [s]

(b) Pure pitching case

Figure 10 Time step sensitivity analysis results

III. Results and Discussion:

A. Validation of the solver:
Wide-ranged steady aerodynamic characteristics experimental database of various reentry vehicles
configurations is was provided by Weiland [3]. To conduct the validation of the CFD solver, a case of a
slender bi-cone with a hemispherical tip configuration is selected. Figure below shows the configuration of
the model. The computational domain and solver setup have the same aspects explained previously. Three
simulation cases are undertaken at three different Mach numbers to compare with the available wind tunnel
results. Table 3 holds a comparison between the experimental and calculated drag coefficients at the given
Mach values. The small deviation is taken as an indication of solver validity in solving the flow problem in
concern.
 Figure 11 Configuration of the validation case

Table 3 Validation case results

Model drag coefficient
Mach Deviation (%)
CFD Wind tunnel [3]

2.53 0.3522 0.34 3.5

4.0 0.3412 0.35 2.5

5.96 0.2237224 0.23 2.7

B. Results for case#1, pure longitudinal acceleration while descent at zero incidence:
The flowfield features around the vehicle at a representative time instant; at 166 seconds from separation, are
illustrated in Figure . An overview of the domain is shown as pressure and Mach contours to the left and right,
respectively, of Figure .a. A close-up view to of Mach contours around the model is shown in Figure .b ,
while a zoom-in view of the vehicle's nose tip is shown in Figure .c with Mach and pressure contours at the
top and bottom halves of the view, respectively. The shock wave generated ahead of the body is oblique and
straight with a small curvature at the tip. Since the nose tip is rounded, the shock wave is detached at a small
distance ahead of the tip. A weak expansion wave can be shown at the shoulder of the first conical section
downstream of the tip. In addition, the thick boundary layer characterizing the hypersonic flow is evident. As
the boundary layer separates at the vehicle base, a shear layer is created that propagates downstream and
engulfs a large zone of recirculating air at the base of the vehicle.
 (c) Nose tip zoom-in, Mach contours (top), pressure contours (bottom)
Figure 12 Flowfield features at 166 sec. from separation

Drag on the vehicle increases significantly with time as the vehicle descends as shown in Figure . This is
definitely owed to the continuous rise in freestream total pressure and the subsequent increase in surface
pressure. In contrast, the drag coefficient decreases monotonically as the vehicle descends. Meanwhile, the
freestream density and, more significantly, velocity increase with rates higher than that of the drag.
Eventually, the drag coefficient decreases as the vehicle descends.

Figure 13 Temporal variation of drag and drag coefficient during accelerating descent

To analyze the drag behavior more deeply, drag components namely, forebody pressure drag ( D fp ), forebody
friction drag ( D ff ), and base pressure drag ( D bp ) are examined at four different time instances during descent.
These components are extracted from unsteady (US) simulation results and are listed in Table 4. For the sake
of comparison, the corresponding components based on steady simulations (S) (cases#4: #7, respectively) are
listed as well.

Table 4 Evolution of drag components during descent

Flight time from

Dfp [N] Dff [N] Dbp [N] Total drag [N]
separation [sec]
US1 S2 US S US S US S

166 2.06 2.57 4.38 4.59 0.57 0.55 7.02 7.715

176 8.33 7.99 8.38 7.66 2.49 2.21 19.19 17.86

186 32.46 31.35 16.2 14.42 12.43 11.47 61.1 57.24

 196 65.35 63.55 23.14 19.73 28.05 22.93 116.54 106.22

206 1006.72 971.1 163.34 176.7 381.75 359.71 1551.81 1507.5

As inferred from Table 4, the rise in drag during the longitudinal acceleration is a general trend for all
components along with the whole interval. The pressure drag on the forebody ( D fp ) is the dominant
component which represents about 32% of the total drag at the altitude of 77.8 km and increases to reach 65%
of the total drag value as the vehicle reaches the height of 32.5 km. Similarly, the pressure drag on the base (
Dbp ) increases as the vehicle descends to form about 25% of the total drag value. On the other hand, the
forebody friction drag ( D ff ) forms about 60% of the total drag at the altitude of 77.8 km and reach 11% of the
total drag value when the vehicle is 32.5 km above the ground. Compared with steady calculations, the
instantaneous drag values are slightly higher (2 – 10%) than their steady counterparts at the same freestream
conditions. This difference may be explained by the kinematics of the vehicle during acceleration. The slight
difference may be owed to the relatively small value of longitudinal acceleration which may make the
unsteady motion of the accelerating vehicle is a close similarity to a quasi-steady one.

C. Results for case#2, pure pitching at cruise conditions:

Figure below shows the evolution of key flow field features, Mach and pressure contours, at four different
positions of the cycle namely quarter, half, three-quarters, and full pitching period. Mach contours show the
structure of the boundary layer around the vehicle while pitching up. The boundary layer is thinner on the
upper meridian of the vehicle while pitching down and thicker in case of pitching up. It is noticed that the
boundary layer is always attached and does not separate on the leeward side during pitching. This may be
owed to the low pitching amplitude and the establishment of a laminar boundary layer along the vehicle
surface. For pressure contours, the flow field pressure distribution around the vehicle at different states during
the cycle is also shown. Pressure contours around the vehicle tip indicate the change in shock wave location
and strength on both sides of the vehicle at incidence. Both the shock wave and the expansion fan at the
conical section shoulder are stronger on the windward side than on the leeward side. The low frequency and
amplitude of the vehicle pitching motion assures that the flow around the vehicle is dynamically stable.

(a) at t = T/4
 (b) at t = T/2

(c) at t = 3T/4

(d) at t=T
Figure 14 Evolution of Mach contours around the whole vehicle (left), and pressure contours at the nose tip
(right) during a complete pitching cycle

The evolution of vehicle drag coefficient during two complete cycles of a regular pitching oscillation with a
period of 17 seconds is illustrated in Figure 15 below. The dashed line represents the variation of incidence
angle during the cycle whereas the markers indicate the static values of vehicle drag coefficient at the
respective instantaneous incidence angles (cases #8: #12).
Drag Coeff.

9.6 6
8.4 4
AOA [deg]

7.2 2
6
4.8 0
3.6 -2
2.4
1.2 -4
0 -6
262 267 272 277 282 287 292 297
Flight time [s]
Figure 15 Temporal evolution of drag coefficient during two complete pitching cycles
The drag performs two cycles per pitching cycle because of the symmetry of the vehicle. Starting at zero
incidence, drag attains a minimum value that increases rapidly as the vehicle pitches down reaching a
maximum value at the lower limit of incidence. During pitching up, drag drops to a minimum that is slightly
higher than that at the start. A clear drop in the maximum drag at the end of pitching-up is also evident. This
drop may be explained by the fact that the peak positive value of incidence is less than the negative one since
the pitching cycle is not symmetric as indicated in Figure 5.a. This drag peak is higher than the static drag
value at the same incidence angle. With respect to vehicle kinematics, a minor lag of about 0.17 seconds (1%
of the cycle period) is addressed in the drag values of the pitching vehicle in this cycle. This is possibly due to
the low amplitude and frequency of pitching.
To explore the drag components acting on the vehicle and their individual evolution with time during this
cycle of pitching, Figure shows the variation at four different instances during one cycle of drag components
namely, forebody pressure (FP), forebody friction (FF), and base pressure (BP) based on unsteady
simulations (-U). Corresponding steady drag components values (-S) are also shown for comparison.
0.1

0.09

0.08
Drag coefficient

0.07

0.06

0.05

0.04

0.03

0.02
T/4, -4 deg T/2, 0 deg 3T/4, 4 deg T, 0 deg

Time in periods, incidence angle

FP-U FF-U BP-U FP-S FF-S BP-S

Figure 16 Drag components variation during one complete cycle

The three components of drag are minimum when the vehicle is instantaneously at zero incidence and
maximum at the peak incidence angle. Clearly, the dominant (and most incidence-sensitive) component is the
forebody pressure drag while base pressure drag is the minor contributor and the least sensitive to incidence.
In addition, drag on the vehicle base is almost unresponsive to the attitude of the vehicle during pitching
because of the low pressure. The base drag variation during pitching is about 6 drag counts (a drag count =
0.0001) which signifies less than 2% of its average value. Since its unsteady and steady values are almost
equal, it can be treated as insensitive to the motion of the vehicle. On the other hand, the forebody surface
friction drag shows moderate variation during pitching; it is higher at incidence that than is it at no incidence.
The variation of friction drag reaches within 19 drag counts that represent 5% of its average value. Similar to
base drag, friction drag on forebody is less delicate to vehicle kinematics since its unsteady and static values
are almost equal (about two drag counts average difference). Finally, the vehicle forebody pressure drag is the
most sensitive component to variation in vehicle state during pitching. The variation of this component
reaches within 138 drag counts that representing about 16% of its average value. The difference between its
static and unsteady values echoes its sensitivity to vehicle kinematics. It can be also inferred that the drop in
total drag amplitude with oscillation is primarily owed to pressure drag on the vehicle forebody.

D. Results for case 3: Combined acceleration-pitching while the descent

The variation of drag acting on the vehicle as it performs pitching during accelerating descent for the three
seconds in concern is addressed in Figure 17. For the sake of comparison, the variation of drag during the
same period without pitching is plotted in dashed line.
 1500

Drag [N]
Drag (acc.)
1300 Drag (pitch-acc.)

1100

900

700
273 273.5 274 274.5 275 275.5 276
Flight time [s]
Figure 17 Evolution of drag on under pitch-accelerating descent

As expected, drag acting on the vehicle increases as it descends during the period in concern. Pitching has the
role of slightly reducing the instantaneous drag at zero incidence and increasing it at the peak pitch-up angle.
Overall, it can be inferred that the pitching motion has the impact of increasing the rate by which drag on the
vehicle increases with time. This trend can be viewed as a superposition of two factors namely, drag rises as
the freestream total pressure increases, and drag rises as the vehicle pitches up from zero to 4 degrees.

Conclusion:
The features and evolution of the flow field around a hypersonic reentry vehicle during the descent phase
was were presented in this study. The vehicle, that which is the upper stage of a ballistic missile, experiences
both longitudinal acceleration and pitching while descending towards the earthground. The trajectory from the
separation point is estimated using a trajectory prediction code. The flow around the vehicle is simulated
using a commercial CFD tool in three cases: pure longitudinal acceleration during descent, pure pitching at
cruise conditions, and combined pitching-acceleratingon while descent. Simulations account for the variation
of freestream properties during descent in a continuum standard atmosphere. The study findings can be
briefed in that, for the case in concern, as the vehicle descends, drag increases almost exponentially with time
as a result of the rise on in freestream total pressure while the drag coefficient, on the other hand, decreases
monotonically as the vehicle descends. In addition, upon comparing the steady and unsteady simulations
results, the phase shift and results difference, which is counted for the unsteady cases, is owed to the
kinematics of the vehicle itself either at the longitudinal acceleration or pitching oscillation unsteadiness.
For the case in concern, since the kinematics of the vehicle (acceleration and pitching frequency) are small,
only marginal differences can be addressed between steady and unsteady simulations. Also, the pressure drag
force on the forebody of the vehicle is the dominant component of the acting drag force, while the total drag
force is dramatically increases while the freestream is completely turbulent.
Simulation of the entire descent phase considering both longitudinal acceleration and lateral oscillation is
considered for future work. Aerodynamic heating effects on the vehicle during descent can also be addressed.

References:
[1] Anderson Jr JD 2010 Fundamentals of aerodynamics (Tata McGraw-Hill Education)
[2] Shen C 2006 Rarefied Gas Dynamics: Fundamentals, Simulations and Micro Flows (Springer
Science & Business Media)
[3] Weiland, C., "Aerodynamic Data of Space Vehicles," Springer Science & Business Media, 2014.
[4] Celenligil M C, Moss J N, and Bird G A 1989 Direct simulation of three-dimensional flow About the
AFE vehicle at high altitudes Rarefied Gas Dynamics: Theoretical and Computational Techniques
(National Aeronautics and Space Administration) pp 447-461
[5] Hemdan, H. T., “Similarity Solutions for Oscillating Pointed-nose Slender Axisymmetric Bodies—
part II: curved bodies,” Acta Astronautica, vol. 49, no. 11, pp. 611-626, 2001.
[6] Clay, J., and Walchner, O., “Hypersonic Stability Derivatives of Blunted Slender Cones,” AIAA
Journal, vol. 3, no. 4, pp. 752-754, 1965.
[7] Orlik-Rückemann, K. J., “Stability Derivatives of Sharp Wedges in Viscous Hypersonic Flow,” AIAA
Journal, vol. 4, no. 6, pp. 1001-1007, 1966.
[4] Orlik-Rückemann, K. J., “Oscillating Slender Cone in Viscous Hypersonic Flow,” AIAA Journal, vol.
10, no. 9, pp. 1139-1140, 1972.
[5] Brady, J., Levensteins, Z. and Lyons, W., “Hypersonic Drag, Stability, and Wake Data for Cones and
Spheres,” AIAA Journal, vol. 2, no. 11, pp. 1948-1955, 1964.
[6] Ericsson, L. E., “Universal Scaling Laws for Nose Bluntness Effects on Hypersonic Unsteady
Aerodynamics,” AIAA Journal, vol. 7, no. 12, pp. 2222-2227, 1969.
[7] Adams, J. and Griffith, B., “Hypersonic Viscous Static Stability of a Sharp 5-deg Cone at Incidence,”
AIAA Journal, vol. 14, no. 8, pp. 1062-1068, 1976.
[8] Khalid, M. and East, R., “High Mach Number Dynamic Stability of Pointed Cones at Small Angles of
Attack,” AIAA Journal, vol. 18, no. 10, pp. 1263-1265, 1980.
[9] Lijun, X., Yunjun, Y. and Zhou L., “High-Performance Computing of Periodic Unsteady Flow Based
on Time Spectral Method,” Procedia Engineering, vol. 99, pp. 1526-1530, 2015.
[10] Chou, Y. and Laitone, E., “Phugoid Oscillations at Hypersonic Speeds,” Aiaa Journal, vol. 3, no. 4,
pp. 732-735, 1965.
[11] Hui, W. and Tobak, M., “Unsteady Newton-Busemann Flow Theory. I-Airfoils,” AIAA Journal, vol.
19, no. 3, pp. 311-318, 1981.
[12] Ye, Y. Zhao, Z. and Tian H., “The Stability Analysis of Rolling Motion of Hypersonic Vehicles and
its Validations,” Science China Physics, Mechanics & Astronomy, vol. 57, no. 12, pp. 2194-2204,
2014.
[17] Gledhill, I., Forsberg, K. and Eliasson, P., “Investigation of Acceleration Effects on Missile
Aerodynamics Using Computational Fluid Dynamics,” Aerospace Science and Technology, vol. 13,
no. 4-5, pp. 197-203, 2009.
[18] Cayzac, R. Carette, E. and Thepot, R., "Recent Computations and Validations of Projectile Unsteady
Aerodynamics," Proceedings of 22nd International Symposium on Ballistics, 14-18 November 2005,
Vancouver, BC, Canada.
[19] Sahu, J., “Time-accurate Numerical Prediction of Free-Flight Aerodynamics of a Finned Projectile,”
Journal of Spacecraft and Rockets, vol. 45, no. 5, pp. 946-954, 2008.
[20] Liu, X., Liu, W. and Zhao, Y., “Navier–Stokes Predictions of Dynamic Stability Derivatives for Air-
breathing Hypersonic Vehicle,” Acta Astronautica, vol. 118, pp. 262-285, 2016.
[21] “DSSC report, DSSC-TR-2025B,” 2015.
[22] Chin, S. S., “Missile Configuration Design,” McGraw-Hill, New York, 1961.
[23] Abdel-Hamid, O. E., “MAC User Guide,” Generalized MAC ed, EAF-TRC,2004.
[24] El-Mahdy, L. A., Ahmed, M. Y. M., Mahmoud O. K., and Abdel-Hamid, O. E., “A Comparative
Study of Prediction Techniques for Supersonic Missile Aerodynamic Coefficients,” Journal of
Mechanical Engineering, vol. 14, no. 1, pp. 35-60, 2017.
[25] U. S. Atmosphere, “National Oceanic and Atmospheric Administration,” National Aeronautics and
Space Administration, United States Air Force, Washington, DC, 1976.
[26] Mohamed S. Abdel-Gawad, Mostafa S. Khalil, and Mahmoud Y. M. Ahmed, “Unsteady Numerical
Analysis of Drag on Non-Winged Hypersonic Vehicle during Re-entry,” IOP Conference Series:
Materials Science and Engineering, 610 (2019) doi:10.1088/1757-899X/610/1/012096.
[27] "FLUENT14.5.7 User’s Guide,” ANSYS Inc., Lebanon, NH, 2011.