Acta Astronautica 93 (2014) 291303

Contents lists available at ScienceDirect

Acta Astronautica
journal homepage: www.elsevier.com/locate/actaastro

Aerothermodynamic and stability analyses of a deployable

re-entry capsule
Valerio Carandente n, Gennaro Zuppardi, Raffaele Savino
Department of Industrial Engineering (DII)Aerospace Division, University of Naples Federico II, Italy

a r t i c l e i n f o abstract

Article history: Recent research projects, in the field of atmospheric re-entry technology, are focused on
Received 11 April 2013 the design of deployable, umbrella-like Thermal Protection Systems (TPSs). These TPSs are
Received in revised form made of flexible high temperature resistant fabrics, folded at launch and deployed in
25 June 2013
space for de-orbit and re-entry operations. In the present paper two possible spherecone
Accepted 19 July 2013
Available online 26 July 2013
configurations for the TPS have been investigated from an aerodynamic point of view. The
analyzed configurations are characterized by the same reentry mass and maximum
Keywords: diameter, but have different half-cone angles (451 and 601). The analyses involve both the
Atmospheric re-entry evaluation of thermal and aerodynamic loads and the assessment of the capsule long-
Deployable heat shield
itudinal stability. The aerothermodynamic analysis has been performed for the completely
Longitudinal stability
deployed heat shield in transitional and continuum regimes, while the longitudinal stability
Thermal and aerodynamic loads
has been analyzed in free molecular, transitional and continuum regimes, also taking into
consideration the heat shield deployment sequence at high altitudes.
& 2013 IAA. Published by Elsevier Ltd. All rights reserved.

1. Introduction goals and to offer the possibility to safely recover the payload
and potential data. These are the reasons why novel concepts
In recent years there has been an increasing interest for of deployable aerodynamic decelerators for de-orbit and re-
small space platforms (micro- and nano-satellites), in entry purposes have been proposed by the University of
particular for Earth's environmental protection, prevention Naples, in collaboration with several Italian small, medium
of natural disasters and homeland security. Other applica- and large enterprises [14].
tions include scientific experiments or technology, space These concepts are not completely new in the aero-
exploration, observation of the universe, astrophysics, space community, that already dedicated previous efforts
biology or physical sciences in microgravity. in the study of inflatable or deployable systems for atmo-
Reducing size, mass and power implies a significant spheric re-entry [59]. This kind of capsules, in fact, can be
reduction of costs and of development time, increasing easily accommodated in launch vehicles in folded config-
accessibility to space and sustaining frequent launches. uration and, when deployed, exhibit a low ballistic coeffi-
Unfortunately this seems to be in contrast with the increasing cient (i.e. the ratio between the capsule mass and its
problem of space debris. In addition, as a consequence of the surface, times the drag coefficient). A lower ballistic
miniaturization, these systems require more sophisticated coefficient implies a larger deceleration in the upper part
solutions to achieve ambitious scientific and technological of the Earth atmosphere, offering as advantage the reduc-
tion of the aero-thermal peak loads and, consequently, a
much higher reliability of the re-entry phase.
n
Correspondence to: Department of Industrial Engineering (DII), Examples of inflatable systems proposed and already
University of Naples Federico II, P.le V. Tecchio 80, 80125 Napoli, Italy.
tested are the Inflatable Re-entry and Descent Technology
Tel.: +39 0817682358; fax: +39 0817682351.
E-mail addresses: [email protected], (IRDT) [5] and the Inflatable Re-entry Vehicle Experiment
[email protected] (V. Carandente). (IRVE) [6]. An inflatable Nanosat De-orbit and Recovery

0094-5765/$ - see front matter & 2013 IAA. Published by Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.actaastro.2013.07.030
292 V. Carandente et al. / Acta Astronautica 93 (2014) 291303

Nomenclature Nc number of collision cells

Ns number of simulated molecules
a1, a2 variable parameters for Eq. (1) S reference surface, m2
g gravity acceleration, m/s2 T temperature, K
h altitude, m V velocity, m/s
m satellite mass, kg angle of attack, rad
mcs mean collision separation, m flight path angle, rad
p pressure, Pa mean free path, m
q_ convective heat flux, W/m2 air density, kg/m3
r trajectory radius of curvature, m skin friction, Pa
rn nose-cone radius of curvature, m half-cone angle, rad
s curvilinear abscissa, m
t time, s Subscripts
x spatial coordinate along the satellite axis, m
y spatial coordinate perpendicular to the cont continuum regime
satellite axis, m f fluid dynamic
B ballistic parameter, kg/m2 fmf free molecular flow regime
CD drag coefficient s simulated
CMz longitudinal moment coefficient 1 free stream condition
KnD overall Knudsen number 0 at the stagnation point
M Mach number
N number density, m
3

System has been specifically designed for CubeSat pay- nose is always aligned with the free stream. For the two
loads by Andrews Space [7]. analyzed configurations, the longitudinal stability is also
Beside those, different concepts for re-entry systems investigated considering three intermediate steps during
based on mechanically deployable heat-shields exist. In the TPS deployment sequence.
1990 a deployable capsule was developed using an The computations in rarefied flow regimes, i.e. at
umbrella-like heat-shield, made of silicon fabrics and altitudes higher than 100 km, have been performed by
called parashield [8]. A similar satellite called Bremsat means of the Direct Simulation Monte Carlo (DSMC)
was studied in 1996 at the University of Bremen [9]. method. In continuum flow regime, i.e. at altitudes below
The solutions considered in the present work consist of 60 km, computational results have been obtained by Com-
a hemispherical rigid nose-cap and a conical deployable putational Fluid Dynamics (CFD). At intermediate altitudes
umbrella-like TPS made of flexible high temperature computations have been performed by both DSMC and CFD
resistant fabrics. In the mission scenario, the de-orbit and thus the comparison between the two methods has
maneuver can be accomplished taking advantage from the been also reported.
aero-brake deployment (i.e. without propulsion systems) The work is organized as follows. In Section 2 the
and, if required, the re-entry path can be controlled by vehicle configurations under investigations are introduced
means of the variable aerodynamic drag resulting from the and described. The computing methods and their valida-
variable frontal area (i.e. the variable ballistic parameter), in tion are reported in Section 3. Section 4 deals with the
order to correctly target the system in the selected landing DSMC aerodynamic analysis of the different capsule con-
area [2]. Two possible architectural configurations for the figurations encountered during the TPS deployment in the
spherecone capsule have been considered. They mainly range of altitudes between 100 and 150 km. The atmo-
differ in the half-cone angle (451 and 601) and for the spheric re-entry trajectories are presented in Section 5.
deployment mechanism. The two configurations will be The aerothermodynamic analyses performed in transi-
here labeled as TPS-45 and TPS-60, respectively. tional and continuum regimes, for significant flight condi-
The work is intended to assess the thermal and aero- tions selected along the re-entry trajectories, are reported
dynamic loads in the most severe conditions along the in Section 6. The main conclusions are summarized in
capsule re-entry path and its longitudinal stability in free Section 7.
molecular, transitional and continuum regimes. The study
of the longitudinal stability, in particular, is fundamental to 2. System conguration
verify that the satellite does not assume a wrong attitude
during re-entry, which could compromise the effective- The capsule consists of a cylindrical structure containing
ness of the TPS. A favorable situation can be achieved if the all the subsystems necessary for the on-orbit mission and for
satellite shows longitudinal stability with respect to the the re-entry phase, umbrella-like frameworks, off-the-shelf
nominal equilibrium condition around zero angle of attack ceramic fabrics for the conical deployable heat shield and
and instability for the reverse attitude. In this case, the available ceramic materials (e.g. silica, alumina or zirconia)
satellite is aerodynamically self-stabilized and the capsule for the rigid hemispherical nose. The necessary subsystems
 V. Carandente et al. / Acta Astronautica 93 (2014) 291303 293

Fig. 1. Geometrical characteristics of the capsule configurations (dimensions in mm): (a) TPS-45 and (b) TPS-60.

Fig. 2. Sketch of the deployment process for TPS-45: (a) Step 1, 7.501; (b) Step 2, 18.51; (c) Step 3, 35.01; and (d) Step 4, 45.01.
294 V. Carandente et al. / Acta Astronautica 93 (2014) 291303

include parachute, beacon, OBDH, AOCS, IMU, GPS receiver, the longitudinal stability, five plausible locations
batteries and sensors. A total re-entry mass of 15 kg has been (labeled in Fig. 1 with letters from A to E) have been
assumed. As discussed before, half-cone angles () of 451 and considered for the Center of Gravity (CG). Three reference
601 are considered for the TPS. The radius of curvature of points are located along the capsule axis and two of them
the nose for each of the two configurations is such that the along the normal to the axis, according to Fig. 1 and
conical umbrella is tangent to the nose-cap when the Table 1.
deployment process is complete.
Fig. 1 shows the geometrical configurations under
investigation. When the TPS is completely deployed the Table 1
base diameter of the two configurations is 1.08 m, while Center of gravity positions.
the cylindrical structure has a diameter of 30 cm and a
total length of 60 cm. Figs. 2 and 3, respectively, show the Point A Point B Point C Point D Point E
configurations of TPS-45 and TPS-60 at three stages
x [m] 0.20 0.30 0.40 0.30 0.30
during the opening phase up to the final configurations. y [m] 0.00 0.00 0.00 0.075
0.075
In addition, in the following analyses concerning

Fig. 3. Sketch of the deployment process for TPS-60: (a) Step 1, 16.01; (b) Step 2, 36.01; (c) Step 3, 50.01; and (d) Step 4, 60.01.
 V. Carandente et al. / Acta Astronautica 93 (2014) 291303 295

3. Computational methods and 2D at about 77 and 65 km altitude, respectively),

practically satisfied the above mentioned requirements.
3.1. Direct Simulation Monte Carlo (DSMC) The validation analysis also evaluated the adequacy of
the number of collision cells (Nc) and of the simulated
It is well known that the Direct Simulation Monte Carlo molecules (Ns) automatically stated by the code. The
(DSMC) method [10,11] is currently the only possible tool convective heat flux has been chosen as the merit para-
for the solution of rarefied flowfields from free molecular meter for evaluating the quality of the results, being the
to continuum low density regimes. most sensitive parameter in DSMC computations. Fig. 4
DSMC considers the gas as made up of discrete mole- shows that, at an altitude of 76.4 km, increasing the
cules. It is based on the kinetic theory of gases and number of collision cells up to 25% and doubling the
computes the evolution of millions of simulated mole- number of simulated molecules, an excellent match
cules, each one representing a large number (say 1015) of among the heat flux profiles is obtained.
real molecules in the physical space. Intermolecular and
moleculesurface collisions are also taken into account.
The computational domain is divided in cells, used for 3.2. Computational Fluid Dynamic (CFD)
selecting the colliding molecules and for sampling the
macroscopic fluid-dynamic quantities. The solution of the hypersonic flowfield in continuum
The most important advantage of the method is that it regime has been carried out by FLUENT [18]. In this case the
does not suffer from numerical instabilities and does not rely flowfield around the capsule has been assumed to be laminar
on similarity parameters, like Mach and Reynolds numbers. and in chemical non-equilibrium. Air has been simulated as a
On the other hand, it is inherently unsteady. A steady reacting mixture composed of five neutral species (O2, N2, O,
solution is achieved after a sufficiently long simulated time. N and NO) implementing same Gupta's kinetic model [14] for
The DSMC codes used in the present work are DS2V chemical reactions considered in the DSMC codes. The
[12] and DS3V [13] for 2D and 3D analyses, respectively. numerical simulations have been performed with a den-
Both codes consider air as a gas made up of five neutral sity-based, time implicit, and numerical resolution scheme
reacting species (O2, N2, O, N and NO) and rely on the built- through a controlvolume-based technique. The AUSM
in GuptaYos-Thompson [14] chemical model, consisting of (Advanced Upstream Splitting Method) scheme has been
17 reactions. The codes are sophisticated. As widely employed for convective numerical fluxes. This procedure
reported in literature [1517], sophisticated codes imple- has been successfully used for similar problems in previous
ment computing procedures which guarantee a higher works by Savino et al. [1922].
efficiency and accuracy with respect to basic DSMC codes. The computational grid (Fig. 5) was generated by the
The user can verify that the number of simulated molecules software GRIDGEN. The grid is composed of about 25,000
and collision cells are adequate by the online visualization of
the ratio between the molecule mean collision separation
(mcs) and the mean free path () in each collision cell. In
addition, the codes allow the user to change during the run,
the number of simulated molecules. The mcs/ ratio has to
be less than unity everywhere in the computational domain,
but Bird [12,13] suggests 0.2 as a limit value for an optimal
quality of the run. In addition, the codes give the user
information about the stabilization of the runs by means of
the profile of the simulated molecules number as a function
of the simulated time. The stabilization of a DSMC calcula-
tion is achieved when this profile becomes jagged and
included within a band defining the standard deviation.
Besides, the stabilization of the runs from a fluid dynamic
point of view is provided by the ratio of the simulated time
(ts) and the time needed to cross the computing region at
the free stream velocity (tf). A rule of the thumb suggests to
consider that a fluid-dynamic computation is reasonably
stabilized for ts/tf E10. As reported in Table 2, all runs Fig. 4. Heat flux profiles along the TPS-45 surface evaluated by DSMC for
reported in the present work (3D at 150 and 100 km altitude different values of Ns and Nc (h 76.4 km).

Table 2
Quality report for DSMC runs.

h 150 km h 100 km h E77 km h E 65 km

Maximum mcs/ E5.0  0
3 E9.0  10
1 E2.0  10
1 E3.0  10
1

Minimum ts/tf E40 E30 E9.0 E7.0
296 V. Carandente et al. / Acta Astronautica 93 (2014) 291303

Fig. 6. Heat flux profiles along the TPS-45 surface evaluated by CFD for
different mesh finesses.

flight conditions corresponding to the re-entry trajectory

at an altitude of 76.4 km. Table 3 shows that the main
fluid-dynamic parameters calculated using a mesh with
Fig. 5. Computational grid for the TPS-45. 25,000 cells are very similar to those obtained with a
coarser mesh characterized by about 15,000 cells, and with
a finer mesh characterized by about 70,000 cells. In
Table 3 addition, Fig. 6 shows the heat flux profiles along the
CFD grid convergence analysis. TPS-45 surface by the three different meshes.
E15,000 E25,000 Error E70,000
cells
Error [%]
cells [%] cells 4. Aerodynamic and stability analyses in hypersonic
rareed regime
CD 1.0084
5.6600 1.0689
0.20582 1.0667
p0 [Pa] 1088.7
0.091768 1089.7 +0.045884 1090.2
q_ 0 [kW/m2] 422.32 +3.7081 407.22
2.9075 395.38 In order to evaluate the drag coefficient, the ballistic
parameter and the stability characteristics of the deploy-
able capsules at high altitudes (i.e. at 100 and 150 km), a
number of computations have been carried out by the
quadrilateral cells, whose density becomes larger DS3V code. Considering that at h100 km both TPSs can
approaching the capsule. In particular, the nominal CFD be considered completely deployed, three intermediate
computational grid used for simulations has been obtained steps of the TPS deployment sequence have been taken
adapting a coarser mesh in the shock layer. This choice into account for each configuration only at 150 km. At
provides the best resolution in the most critical region of these altitudes the overall Knudsen number (KnD1) ranges
the flowfield where, in particular, high temperature and from 0.11 to 42. According to Moss [23], a general defini-
velocity gradients must be considered. tion of the transitional regime considers the overall
Different conditions were set for the computational Knudsen number between 10
3 and 50. Both completely
domain boundaries: a pressure far-field condition was set deployed TPSs are therefore in high rarefied regimes, very
for the boundary labeled as 1, in order to model free close to the free molecular flow regime. Table 4 reports
stream conditions, symmetry conditions for the symmetry free stream inputs to DS3V and some operative parameters
surface (labeled as 2), no-slip conditions for the surface of at the altitudes under investigation.
the capsule (labeled as 3) and a pressure outlet for the Fig. 7(a) and (b) shows the drag coefficients computed
downstream surface (labeled as 4). The free stream con- at four different altitudes: h65 km and 77 km by CFD,
ditions have been imposed according to the re-entry h 100 km and 150 km by DSMC for TPS-45 and TPS-60,
trajectories parameters reported in Section 5. The no-slip respectively. The drag coefficient variations computed by a
assumption on the wall was justified by the nearly con- modified version of the global bridging formula proposed
tinuum regime, because CFD analyses have been per- by Thales Alenia Space-Italy (TAS-I) [24] are also reported
formed only for Knudsen numbers less than 4  10
3. for completeness as Eq. (1). CD,cont and CD,fmf represent the
A grid convergence analysis, in order to ensure the drag coefficients computed in continuum and free mole-
solution independence from the mesh fineness, has been cular flow regimes (i.e. at 65 km and 150 km altitude,
performed (see Table 3). The percentage error among respectively). Finally, a1 and a2 are variable parameters
significant parameters (namely, the drag coefficient, the allowing one to tune up the bridging formula on the basis
stagnation pressure and convective heat flux), computed of the computational results. A good accordance, in this
using three different grids, has been evaluated for the case, has been obtained assuming a1 equal to 0.095 and
 V. Carandente et al. / Acta Astronautica 93 (2014) 291303 297

Table 4
Input data to DS3V and operative parameters.

h [km] 1 [kg/m3] N1 [m
3] T1 [K] p1 [Pa] V1 [km/s] M1 KnD1

9 16
4
150 2.08  10 5.20  10 634 4.50  10 0.760 13.4 42.0
100 5.59  10
7 1.19  1019 196 3.20  10
2 0.750 26.3 0.11

Table 5
Ballistic coefficients for TPS-45, h 150 km.

Step 0 Step 1 Step 2 Step 3 Step 4

CD 2.46 2.06 2.05 2.04 1.96

S [m2] 0.0707 0.151 0.320 0.655 0.915
B [kg/m2] 86.2 48.1 22.9 11.3 8.38

Table 6
Ballistic coefficients for TPS-60, h 150 km.

Step 0 Step 1 Step 2 Step 3 Step 4

CD 2.46 2.10 2.09 2.09 2.08

S [m2] 0.0707 0.246 0.579 0.806 0.915
B [kg/m2] 86.2 29.1 12.4 8.93 7.89

Fig. 7. Bridging functions for the drag coefficient profiles, as functions of

the free stream Knudsen number, for TPS-45 (a) and TPS-60 (b). Fig. 8. Ballistic parameter profiles as functions of the deploying step
(h 150 km).

0.18 for TPS-45 and TPS-60, respectively and a2 equal to 1.0 provides the measure of the capability to aerodynamically
in both cases control the capsule during the de-orbit phase.

 As far as the longitudinal stability is concerned, the
KnD1 angle of attack ranged from 01 to 101 for the computation
C D C D;fmf
C D;cont a2 1
KnD1 a1 of the stability derivatives (dCMz/d) relative to the nom-
inal equilibrium condition around the zero angle of attack
Tables 5 and 6 report the ballistic parameter (B) for TPS-45 and from 1701 to 1801 for the stability evaluation in the
and TPS-60, respectively, during the deployment phase, reverse attitude. The stability derivative has been numeri-
while Fig. 8 graphically shows its variation, starting from cally approximated by the finite difference in the interval
the folded configuration (Step 0) up to the completely 101.
deployed TPS (Step 4). It is evident that the ballistic The reference system has been assumed in accordance
coefficient is reduced, for both configurations, of about with Fig. 1, being the z-axis such as to form a right-handed
one order of magnitude. The variation of this parameter triad with x and y.
298 V. Carandente et al. / Acta Astronautica 93 (2014) 291303

As well known, in this case, a negative value of the

stability derivative identifies a stable condition (the more
negative the derivative, the more stable the system),
considering that a negative moment coefficient represents
a pitching moment. The analysis has been carried out
considering as poles the five different positions for the
Center of Gravity (CG) listed in Table 1 and as reference
length the longitudinal dimension of the capsule (i.e.
60 cm). At each deployment step, the reference surfaces
reported in Tables 5 and 6 have been also considered.
Fig. 9(a) and (b) shows the longitudinal moment
coefficient (CMz) as a function of the angle of attack during
the TPS deployment at 150 km altitude and assuming the
center of gravity located in point B. Fig. 9(a) clearly shows
that TPS-45 is longitudinally stable during the entire
deployment process with respect to the equilibrium con-
dition around zero angle of attack. At the same time, it is
unstable with respect to the reverse equilibrium condition,
as reported in Fig. 9(b). This situation is very favorable
because implies the aerodynamic self-stabilization of the
satellite. An opposite behavior is registered for TPS-60.

Fig. 10. Stability derivative profiles with respect to the equilibrium

condition around zero angle of attack for TPS-45 as a function of the
deployment step for different CG positions at h 150 km, in nominal
(a) and reverse (b) attitudes.

Fig. 10(a) shows the stability derivative (dCMz/d)

profiles at 150 km altitude as a function of the TPS-45
deployment step with respect to the equilibrium condition
around zero angle of attack and for the center of gravity
locations listed in Table 1. Fig. 10(b) refers to the reverse
equilibrium condition. These figures confirm that TPS-45
exhibits a favorable stability also for different locations of
the center of gravity. As expected, at each deployment
step, as the center of gravity moves forward, the stability
derivative decreases.
Fig. 11(a) and (b) refers to TPS-60 and are analogous to
Fig. 10(a) and (b). Even for different locations of the center
of gravity TPS-60 exhibits an unfavorable stability behavior
in highly rarefied regime.
Tables 7 and 8 report the stability derivatives for the
two TPSs at 100 km altitude and for the above mentioned
different center of gravity locations. Also in this case the
TPS-45 configuration shows a higher stability than TPS-60
Fig. 9. Longitudinal moment coefficient (CMz) profiles as functions of the
angle of attack () at different deployment steps in nominal (a) and
with respect to the equilibrium condition around zero
reverse (b) attitudes. The center of gravity is located in point B and angle of attack. In addition, TPS-45 is also less stable than
h 150 km. TPS-60 with respect to the reverse equilibrium condition.
 V. Carandente et al. / Acta Astronautica 93 (2014) 291303 299

Since in this case there are two stable equilibrium condi- 5. Re-entry trajectories and test conditions
tions, a stabilization technique (e.g. spin, 3-axis) is
preferable. The re-entry trajectories have been computed numeri-
cally integrating the dynamic equations of motion
reported in Eq. (2) by means of Euler's method. The initial
conditions for the re-entry trajectory at h120 km are,
V7550 m/s and
0.51
8
V 2 C D S 2
>
> dt
 2 m 
>
dV

g sin
V
2B
g sin
< 2
V d Vr
g cos 2
>
>
dt
>
: dh V sin
dt

The aerodynamic drag coefficient (CD) has been assumed

equal to 1 for TPS-45 and to 1.56 for TPS-60, in accordance
with the results obtained by CFD calculations in continuum
regime. Fig. 12(a)(d) shows the re-entry trajectories in
terms of the free stream velocity (a), free stream Knudsen
number (b), stagnation point pressure (c) and stagnation
point convective heat flux (d) profiles for TPS-45 and TPS-60.
The stagnation point pressure has been preliminary evalu-
ated assuming a pressure coefficient equal to 2, as usual for
spherecone capsules in hypersonic regime. The stagnation
point convective heat flux variation along the trajectory has
been estimated by Tauber's engineering formula [25]
reported in Eq. (3). This formulation provides a preliminary
evaluation of the order of magnitude of the stagnation-point
heat flux in hypersonic re-entry flows. A more detailed
estimation of the convective heat flux distribution around
the capsule is carried out with CFD and DSMC in Section 6.
r
3
q_ 0 1:83  10
4 V 3
rn

It is possible to notice that the maximum stagnation-

point pressure condition is met around 65 km (ranging
from about 1500 Pa for TPS-60 to about 2300 Pa for TPS-
45). The maximum stagnation point heat flux condition is
met around 77 km (being about 325 kW/m2 for TPS-60
and about 480 kW/m2 for TPS-45). The integrated heat
loads at the stagnation point for TPS-45 and TPS-60 are
about 93 and 61 MJ/m2, respectively.
Fig. 11. Stability derivative profiles with respect to the equilibrium
condition around zero angle of attack for TPS-60 as a function of the
At the maximum stagnation pressure conditions (h
deployment step for different CG positions at h 150 km, in nominal (a) 63.5 km for TPS-45 and h65.7 for TPS-60) KnD1 is in the
and reverse (b) attitudes. >order of 3  10
4. At the maximum stagnation heat flux

Table 7
Stability derivatives [1/deg] with respect to the equilibrium condition around zero angle of attack for the two TPSs at h 100 km.

dCMz/d [deg
1] CG  A CG  B CG  C CG  D CG  E

TPS-45
1.1  10
2
7.0  10
3
3.0  10
3
6.4  10
3
7.6  10
3

TPS-60
4.9  10
3
1.6  10
3 1.7  10
3
9.4  10
4
2.3  10
3

Table 8
Stability derivatives [1/deg] with respect to the reverse equilibrium condition for the two TPSs at h 100 km.

dCMz/d [deg
1] CG  A CG  B CG  C CG  D CG  E

TPS-45 9.2  10
4
5.0  10
4
2.0  10
3
1.8  10
4
1.0  10
3

TPS-60
2.6  10
3
4.8  10
3
7.0  10
3
4.2  10
3
5.4  10
3
300 V. Carandente et al. / Acta Astronautica 93 (2014) 291303

Fig. 12. Velocity (a), free stream Knudsen number (b), stagnation point pressure (c) and stagnation point convective heat flux (d) profiles as functions of
altitude.

Table 9
Input data to DS2V and FLUENT.

h [km] 1 [kg/m3] N1 [m
3] T1 [K] p1 [Pa] V1 [km/s] M1

79.4 (TPS-60) 2.03  10
5 4.24  1020 199.8 1.20 6.00 21.1

76.4 (TPS-45) 3.20  10
5 6.69  1020 205.5 1.90 5.97 20.7
65.7 (TPS-60) 1.49  10
4 3.12  1021 231.4 9.90 3.35 11.0
63.5 (TPS-45) 2.00  10
4 4.18  1021 237.4 13.6 3.40 11.0
50.0 (TPS-45) 1.02  10
3 2.14  1022 270.7 79.8 1.08 3.30
50.0 (TPS-60) 1.02  10
3 2.14  1022 270.7 79.8 0.732 2.20

conditions (h76.4 km for TPS-45 and 79.4 for TPS-60) KnD1 6. Aerothermodynamic analysis of atmospheric re-entry
is in the order of 5  10
3. At these altitudes and for both
TPSs, the flow is in continuum low density regime. Being in All numerical simulations considered fully-catalytic
this case the flow regime borderline between continuum and wall conditions, assuming a constant wall temperature of
transitional, the aerodynamic analyses have been performed 300 K. These assumptions have been made both for DSMC
using both CFD and DSMC codes. Table 9 reports the free and CFD calculations to get estimations for the cold-wall
stream parameters (from the US Standard Atmosphere 1976) convective heat flux.
and some operative test conditions for all the simulations Fig. 13 reports the comparison between CFD and DSMC
carried out. calculations regarding the most significant thermal and
 V. Carandente et al. / Acta Astronautica 93 (2014) 291303 301

Fig. 13. Convective heat flux profiles along the capsule surface at the maximum q_ 0 condition (a); pressure profiles along the capsule surface at the
maximum p0 condition (b), and skin friction distribution along the capsule surface at the maximum q_ 0 condition (c).

mechanical loads acting on the surfaces of the analyzed Finally, Table 10 confirms that TPS-45 has a higher
capsules. stability with respect to the equilibrium condition around
In particular, Fig. 13(a) shows that the convective heat zero angle of attack than TPS-60, also in continuum regime
flux at the stagnation point for TPS-45, mostly due to the (KnD1 E8  10
5). Stability derivatives with respect to the
smaller radius of curvature, is about 47% higher than the reverse equilibrium condition for the two TPSs are
one computed for TPS-60. reported in Table 11. It can be noticed that the TPS-60 is
Also in Fig. 13(b) the higher pressure values are undesirably more stable than the TPS-45 around the
registered for TPS-45 (the percentage increase between reverse equilibrium condition.
the two configurations is about 58%).
Fig. 13(c) shows the skin friction distribution along the
capsule surface for TPS-45 and TPS-60. In this case DSMC 7. Concluding remarks
results overestimate CFD results, due to the completely
different computational approach. An overall good agree- In the frame of the development of deployable Thermal
ment between CFD and DSMC analyses has been achieved. Protection Systems (TPSs) made of light and cheap fabric,
In Fig. 14 some contour distributions for some signifi- two possible configurations have been analyzed for a
cant flowfield variable have been reported for TPS-45 in recoverable micro/nano satellite.
the maximum stagnation-point heat flux condition. From The stability analysis in hypersonic, highly rarefied
this figure it can be noticed that, as well known, DSMC regime (i.e. at 150 km altitude) demonstrated that the
calculations overestimate the stand-off distance and the TPS configuration characterized by a half-cone angle of 451
shock wave thickness with respect to CFD, resulting in a (TPS-45) is longitudinally stable with respect to the equili-
smoother variation of the flowfield variables along the brium configuration around zero angle of attack and
stagnation line. unstable with respect to the reverse equilibrium attitude.
302 V. Carandente et al. / Acta Astronautica 93 (2014) 291303

Fig. 14. Pressure (a), temperature (b) and velocity (c) distributions in the flowfield past TPS-45 configuration at the maximum q_ 0 condition.

Table 10
Stability derivatives [1/deg] with respect to the equilibrium condition around zero angle of attack for the two TPSs at h 50 km.

dCMz/d [deg
1] CG A CG  B CG  C CG  D CG  E

3
3
3
2
TPS-45
4.5  10
3.3  10
2.1  10
1.0  10 3.7  10
3
TPS-60
2.6  10
3
2.3  10
3
2.0  10
3
1.1  10
2 6.3  10
3

Table 11
Stability derivatives [1/deg] with respect to the reverse equilibrium condition for the two TPSs at h 50 km.

dCMz/d [deg
1] CG A CG  B CG  C CG  D CG  E

3
3
3
2
TPS-45
1.9  10
1.8  10
1.7  10
1.2  10 8.5  10
3
TPS-60
3.3  10
3
3.3  10
3
3.4  10
3
1.3  10
2 5.9  10
3

This capsule is therefore aerodynamically self-stabilized. References

This behavior is verified both for different positions of the
center of gravity and during the entire deployment phase. [1] E. Bassano, R. Savino, R. Lo Forti, A. Ferrarotti, C. Richiello, G. Russo,
An opposite behavior has been observed for a capsule R. Aurigemma, F. Punzo, P. Dell'Aversana, IRENEItalian Re-Entry
NacellE for Microgravity Experiments, in: Proceedings of the 62nd
characterized by a TPS half-cone angle of 601 (TPS-60). In International Astronautical Congress, October 37th, 2011, Cape
transitional and continuum regimes TPS-45 confirmed to Town, IAC-11.A2.7.7., 2011.
be more favorable than TPS-60 from a stability point [2] V. Carandente, G. Elia, R. Savino, Conceptual design of de-orbit and re-
entry modules for standard CubeSats, 2nd IAA Conference on Uni-
of view. versity Satellite Missions and Cubesat Workshop, February 47th, 2013,
Aerothermodynamic analyses verified an overall good Rome, 2013.
agreement between CFD and DSMC in transitional regime. [3] MISTRAL Research Project: Proposal for the Aerospace Campania
District, DAC Feasibility Study, 2012.
From the thermal and aerodynamic loads point of view
[4] R. Savino, V. Carandente, Aerothermodynamic and feasibility study
TPS-60 would be more favorable than TPS-45, thanks to of a deployable aerobraking re-entry capsule, Fluid Dyn. Mater.
the higher drag coefficient and radius of curvature. None- Process 8 (4) (2012) 453477.
theless, this advantage would not be justified by the [5] D. Wilde, S. Walther, Inflatable re-entry and descent technology
(IRDT)further developments, in: Proceedings of the 2nd Interna-
different behavior of the capsules from a longitudinal tional Symposium of Atmospheric Re-entry Vehicles and Systems,
stability point of view. Arcachon, France, 2001.
 V. Carandente et al. / Acta Astronautica 93 (2014) 291303 303

[6] M.C. Lindell, S.J. Hughes, M. Dixon, C.E. Wiley, Structural analysis and [16] G.A. Bird, M.A. Gallis, J.R. Torczynski, D.J. Rader, Accuracy and
testing of the Inflatable Re-entry Vehicle Experiment (IRVE), in: efficiency of the sophisticated direct simulation Monte Carlo algo-
Proceedings of the 47th AIAA/ASME/ASCE/AHS/ASC Structures, rithm for simulating noncontinuum gas flows, Phys. Fluids 21
Structural Dynamics, and Materials Conference, May 14th, 2006, (2009).
Newport, RI, AIAA-2006-1803, 2006. [17] M.A. Gallis, J.R. Torczynski, D.J. Rader, G.A. Bird, Convergence
[7] J. Andrews, K. Watry, K. Brown, Nanosat de-orbit and recovery behavior of a New DSMC algorithm, J. Comput. Phys. 228 (2009)
system to enable new missions, in: Proceedings of the 25th Annual 45324548.
AIAA/USU Conference on Small Satellites, August 811th, 2011, [18] Fluent 6.3 User's Guide, Fluent Inc., Lebanon, USA, 2006.
SSC11-X-3, 2011. [19] D. Paterna, R. Monti, R. Savino, A. Esposito, Experimental and
[8] D.L. Akin, The parashield entry vehicle concept: basic theory and numerical investigation of Martian atmosphere entry, AIAA J.
flight test developments, in: Proceedings of the 4th Annual AIAA/ Spacecr Rockets 39 2 (2001) 227236.
Utah State University Conference on Small Satellites, August 27 [20] R. Savino, M. De Stefano, Fumo, aerothermodynamic study of ultra-
30th, 1990, Logan, UT, 1990. high temperature ceramic winglet for atmospheric test, AIAA J.
[9] M. Wiegand, H.J. Knigsmann, A. Small Re-entry Capsule BREMSAT
Thermophys. Heat Transfer 22 4 (2008) 669676.
2, in: Porceedings of the 10th AIAA/USU Small Satellite Conference,
[21] R. Monti, R. Savino, M. De Stefano, Fumo, matching flight conditions
September 1619th, 1996, Logan, UT, 1996.
on sharp leading edges in Plasma wind tunnels, AIAA J. Thermophys.
[10] G.A. Bird, Molecular Gas Dynamics and Direct Simulation Monte
Heat Transfer 21 3 (2007) 660664.
Carlo, Clarendon Press, Oxford, 1998.
[22] R. Monti, M.De Stefano Fumo, R. Savino, Thermal shielding of a
[11] C. Shen, Rarefied Gas Dynamic: Fundamentals, Simulations and
reentry vehicle by ultra high temperature ceramic materials, AIAA J.
Micro Flows, Springer-Verlag, Berlin, 2005.
Thermophys. Heat Transfer 20 3 (2005) 500506.
[12] G.A. Bird, Visual DSMC program for two-dimensional flows, The
DS2V Program User's Guide Version 4.5. [23] J.N. Moss, Rarefied Flows of Planetary Entry Capsules, Secial course
[13] G.A. Bird, Visual DSMC program for three-dimensional flows, The on Capsule Aerothermodynamics, May 1995, Rhode-Saint-Gense,
DS3V Program User's Guide Version 2.5. Belgium, AGARD-R-808, 1995, pp. 95129.
[14] R.N. Gupta, J.M. Yos, R.A. Thompson, A Review of Reaction Rates and [24] A. Schettino, R. Votta, P. Roncioni, M. Di Clemente, M. Gerritsma, C.
Thermodynamic Transport Properties for an 11-Species Air Model Chiarelli, D. Ferrarella, Aerodynamic and Aerothermodynamic Data
for Chemical and Thermal Non-Equilibrium Calculations to 30,000 K, Base of EXPERT Capsule, West-East High Speed Flow Field Con-
NASA TM 101528, February 1989. ference, November 1922nd, 2007, Moscow, Russia, 2007.
[15] G.A. Bird, Sophisticated Versus Simple DSMC, in: M. Ivanov, [25] M.W. Tauber, A Review of High Speed, Convective, Heat Transfer
A. Rebrov (Eds.), Proceedings of the 25th International Symposium Computation Methods, NASA TP 2914, 1989.
on Rarefied Gas Dynamics, 354, Saint Petersburg, Russia, July 21st
28th 2006, p. 349.