Content uploaded by Abdelmadjid Boudjemai
Author content
All content in this area was uploaded by Abdelmadjid Boudjemai on Oct 07, 2015
Content may be subject to copyright.
Thermal Analysis of The Alsat-1 Satellite Battery
Pack Sub Assembly into the Honeycomb Panel
A. Boudjemai
Centre of Satellite Development (CDS)-
ASAL, Space Technology Research
Division BP.: 4065 Ibn Rochd USTO
POS 50 Ilot T12 Bir El Djir -31130-
Oran – Algeria.
a_boudjemai@yahoo.fr
R. Hocine
Department of Electronics Engineering,
BP 1505 EL M’Naour, University of
Sciences and Technology of Oran
Algeria.
rak_hocine@yahoo.fr
M.N. Sweeting
Surrey Satellite Technology Limited,
GU2 5XH, UK.
m.sweeting@surrey.ac.uk
.
Abstract—Information on battery problems can be useful in
guiding research to improve battery technology. Problems that
are serious or reoccur are the obvious ones to concentrate on.
Observed problems can be caused by more than one
phenomenon. However the problem that was observed on the
Alsat-1 battery module where some cells were damaged, and the
damage was caused by the extreme temperatures, an explanation
can be brought to this problem is that the two different inserts
(simple insert and hard insert) are very close which caused an
increase in heat flux. The inserts used to support the battery in
the honeycomb panel have a serious impact on the conduction
from the solar panel to the battery pack, and so the temperature
of the solar panel closer to the battery determines its
temperature. Thus for this reason the simulation was performed
to see carefully this phenomenon caused by the thermal coupling
of the surrounding inserts and the important feedback from the
results obtained in order to avoid all risky design in the future on
the Algerian satellites such as Alsat-1B. Thermal simulations
showed that the adjacent inserts cause thermal interference and
the adjacent inserts are highly sensitive to the effect of high
temperatures. The different results obtained in this paper are
very helpful in the preliminary design stage of the spacecraft and
which is a very promising application in the design of satellite
electrical power subsystem.
Keywords—Alsat-1; Battery; cells; Honeycomb plate;
Finite Element; Temperature; Adhesive.
I. INTRODUCTION
Alsat-1 is an Algerian earth observation satellite (90kg)
which evolves in a sunsynchronous retrograde circular orbit
(SSO). Designed to be part of a constellation for daily disaster
monitoring, Alsat-1is equipped with two banks of cameras
giving a total of 600km field of view at 32 meters ground
sampling distance in three spectral bands: Red, Green and Near
Infra-Red. In November 2002, the 28th, at 06:07am GMT,
Alsat-1was successfully launched by COSMOS-3M from the
cosmodrome of Plesetsk in Russia into a 700 km SSO orbit and
is now fully operational with five-year satellite mission lifetime
[1].
Alsat-1 is three axes stabilized earth observation satellite in
imaging mode and evolves in a BBQ mode out of imaging
time. The attitude determination and control subsystem gives a
good attitude pitch/roll/yaw stability during imaging (=5 m°/s)
and the Orbit filter provides a maximum track error of the
scene position of 5km (GPS On during one orbit a day).
The thermal control of Alsat-1presents unique challenges to
the thermal engineer since the mass; power and volume
available are all very limited. Regardless of these problems and
the extreme environments and changing power conditions
(internally and externally) of the satellite the subsystems must
still be maintained within the specified temperature limits.
The general packages layout in Alsat-1 is given by Fig. 1.
Fig. 1. Packages layout in Alsat-1.
The primary power to the satellite is supplied via 4 solar
aluminium honeycomb panels (see Fig. 2). The power from
each of the four solar panels is fed into a dedicated battery
charge regulator (BCR), i.e. one BCR per solar panel. The
output of the BCRs is connected to a 22 cell, 4Ah NiCd
battery, the power distribution module (PDM) input and the
power conditioning module (PCM) input. The solar arrays and
BCR outputs are isolated from each other using one blocking
diode per BCR [2].
The solar cells used on Alsat-1 were ENE (Belgium) type
single junction GaAs/Ge cells, mounted on aluminium face
sheet aluminium honeycomb substrate. The cells provide on
average 19% conversion efficiency at ambient temperature.
The solar panels were conFig.d as follows: the panel substrates
were made by the TRB [2] consisting of 20mm aluminium core
honeycomb with 0.5mm aluminium face skins front and rear.
The front of the panels has an insulating layer of 75 μm kapton
bonded by TRB before delivery to SSTL. The cell lay down
design of the four panels was identical consisting of 6 strings of
48 cells in series (a total of 288 cells per panel).
Fig. 2. Alsat-1 solar aluminium honeycomb panels.
This paper describes the thermal behaviour of the Alsat-
1satellite battery module. Simulation results are presented and
discussed. The spacecraft has completed over seven years of
operation on orbit. In this the temperature impact on the battery
pack was analyzed and discussed.
The 3-D finite element model of the honeycomb plate with
the six inserts has been developed in Patran/Nastran. A new
approach of the insert with an adhesive model was introduced
into this study using finite element analysis.
THE BATTERY BEHAVIOUR AND OPERATION
OPTIMIZATION
The main cause for concern was the battery pack
temperature. The battery temperature become much lower than
it was expected, something that was not predicted by the
model. The cause for this behaviour is the asymmetry of the
SFF, which was supposed to be small but that showed it to be
very relevant. The inserts used to support the battery in the
honeycomb panel have a serious impact on the conduction
from the solar panel to the battery pack (see Fig. 3), and so the
temperature of the solar panel closer to the battery determines
its temperature. This has led to the spacecraft having to fly in a
specific direction, to keep the +Y solar panel warm [3]. Some
asymmetry was expected but its extent was not realised and so
the battery operates either too hot or too cold relatively to the
model (see fig. 4). This might be a useful feature if there is the
need to cool down the battery at latter stages of the mission.
inserts used to support
the battery in the honey comb panel
inserts used to support
the battery in the honey comb panel
Fig. 3. Inserts used to support the battery in the honeycomb panel.
Fig. 4. Alsat-1 battery cells configuration.
Something that is not easy to see on Fig. 5 is the behaviour
of the temperatures of the battery on the early stages of the
mission. In the early days of the mission, the temperature of the
battery went below 0ºC, for two very short periods.
Fig. 5. Battery behaviour on the early stages of the mission.
After six years of the satellite operations on orbit we
noticed the battery degradations under the temperature effect
(see Fig. 6), the solutions of the operation optimization of the
battery are considered which are:
• Thermal control of the Alsat-1 battery using attitude
control (Yaw angle): For the purpose of reducing the
operational battery temperature and improving its
performance, we made an alteration to the Alsat-1
spacecraft's attitude in order to avoid having the solar rays
directly hitting the facet of the satellite accommodating
the battery module. This was implemented by rotating the
spacecraft 180 degrees around its Z axis using the second
Alsat-1 ADCS model (ADCSM2) which based on the 6-
state Kalman filter [4 - 7].
• Actually, Alsat-1 attitude is estimated the quaternion
attitude vector and the body rates vector using another
attitude filter, quaternion version and more specifically,
the 6 state. This method allowed us to reduce the
operational battery temperature from 35° C to around 15°
C.
• Initially in the first ADCS model (ADCSM1), the Alsat-1
attitude was estimated the attitude and rate parameters
using the small Euler angles filter – small libration.
• Battery charge management using software control: The
principle of this method is to control the battery
temperature via the control of its charge current. Using
battery management software to control the charge current
of the battery has allowed a more accurate adjustment of
the current values using feedback from temperature
telemetry readings and also by controlling the charge
duration. This is not the case with hardware battery
charge, which only uses electronic circuitry to control the
charge current coarsely.
The battery environment in orbit (temperature, voltage and
current) was carefully checked all the time using this tool.
Since the battery pack cells are charged during sunshine and
discharged in the shadow, therefore their respective voltage
increases and decreases periodically. As an example, Fig. 7
shows the voltage for cell 19 in its healthy state. The telemetry
data was taken from the 30th August 2005 file and clearly
shows that the cell was charging and discharging properly. The
voltage was increasing from 1.2 volts (nominal) to 1.45 volts
(max) [8].
Fig. 6. Battery temperature.
Fig. 7. Cell 19 healthy 30/08/2005.
Fig. 8 shows, a year later and for cell 19, a sign of defect. In
fact, cell 19 started acting as a resistor from 30th October 2006.
The discharge cell voltage was hitting a lower value well below
the nominal value. A close monitoring of cell 19 data showed
later that it was deteriorating slowly and finally it went to the
short circuit state on the 22nd March 2008 with a voltage of 0
volt, see Fig. 9.
Fig. 8. Cell 19 acting as a resistor30/10/2006.
Fig. 9. Cell 19 short circuit 22/03/2008.
II. TOTAL MODEL OF A HONEYCOMB PLATE
In this section we analyze by the finite element method the
battery support, it is a composite honeycomb plate with six
inserts, four inserts (B,D,E and F) for the battery assembly and
the remainder inserts (A and C) it is for momentum wheel
assembly which is adjacent to the battery pack as indicated in
fig. 3. The hexagonal honeycomb, have been widely used in
the manufacture of the aerospace structures due to their
lightweight, high specific bending stiffness and strength under
distributed loads [9-12]. The total elements and nodes of the
FEM models are 74904 elements and 77468 nodes for a
complete honeycomb sandwich plate.
Fig. 10 shows the full FEM of the sandwich honeycomb
panel, inserts position (all dimensions are in mm) and adhesive.
A
B
C
D
E
F
Insert 1
Insert 2
Insert 3
Insert 4
Insert 5
Insert 6
(a)
(b)
Fig. 10. Full FEM honeycomb plate: (a) FEM core with insert, (b) a
complete FEM honeycomb model.
Multiple boundary conditions are used in this study. We
apply each time either a gradient of temperature, or the solar
density flux arriving at a surface of the honeycomb plate (see
Fig. 11) so that to simulate the orbital condition. The different
types of thermal loading used in this analysis are as follows:
• A fixed temperature (2°C, 40°C and 60°C) on the top face
of the honeycomb plate model.
• A 1378 W/m² direct radiation refers to the solar flux
arriving at a surface of the honeycomb plate, and 5 W was
applied on the honeycomb plate which represents the
dissipated power in the Alsat-1 battery.
(T, q)
Fig. 11. Honeycomb sandwich plate subjected to heating Temperature and
heat source q over entire upper surface.
III. RESULTS AND DISCUSSION
The thermal analysis is done on the honeycomb plate
including the adhesive and inserts using Msc Patran/Msc
Nastran softwares. The mappings of thermal results onto a
honeycomb plate model are given in this section.
The thermal temperature is a critical parameter in the
mechanical design of space applications. However, the effects
of the temperature on the electronic components carried by the
honeycomb structures generally come from several sources.
This is why these equipments are often heated significantly by
the power dissipated within the devices (self heating) and by
the power dissipated in adjacent inserts (thermal coupling).
All simulations were done according to Fig. 10.a, with A,
B, C and D represents the adjacent inserts.
The results presented in Fig. 12 to 14 are those obtained
with a honeycomb plate with six inserts and with two adjacent
inserts. The analysis was carried out under the software Msc
Patran and Msc Nastran.
Fig. 12 shows the results of the honeycomb plate subjected
to thermal heating. The coloured fringes give the amplitude of
the temperature vector describing the shape of each case. The
red colour corresponds to maximum temperature.
Progressively with simulation, the effect of the heat transfer
by conduction in the plate is noticeable. Indeed, the
temperature on the level of the two adjacent inserts of the plate
increases, causing a transfer of heat which comes to heat the
electronics components carried by the panel (see fig. 15).
We notice that the increase or decrease of temperature
depends on the temperature imposed in the boundary
conditions. This is due to the existence of temperature
variations on the satellite orbit.
We also note that the distance between the inserts plays an
important part in increasing the heat transfer in the coupling
region (see fig. 16). Another issue is when heat travels through
the core, most of it is conducted through the cells walls (see
fig. 17), which furthermore contribute to increasing the heat
transfer in the coupling region.
The presence of a large amount of heat in the coupling area
is also due to the dissipated power by the equipments carried
by the honeycomb plate (see fig. 18).
We note that the adhesive model play an important role in
the results and in the heat transfer through the inserts as show
in the fig. 19.
A specific thermal control of the alsat-1b battery is a
necessary process for removing excessive heat from inside
battery pack in order to keep the battery components within a
safe operating temperature.
Fig. 12. Temperature Profile Distributions in °C (for Msc patran plot-
boundary temperature T=60°C).
Fig. 13. Temperature Profile Distributions in °C (for Msc patran plot-
boundary temperature T=40°C, P=1378w/m2).
Fig. 14. Temperature Profile Distributions in °C (for Msc patran plot-
boundary temperature condition, T=2°C, P=5w).
Fig. 15. The two adjacent inserts.
Fig. 16. The heat transfer in the coupling region.
Fig. 17. Heat transfer through the cells walls.
Fig. 18. The heat transfer through the honeycomb core.
Fig. 19. The heat transfer through the adhesive.
IV. CONCLUSION
In light of this study, the thermal coupling problem between
two adjacent inserts of a honeycomb plate was analyzed in
order the show the battery pack assembly in the satellite with
the adjacent equipment.
From the results we can conclude that the satellite
subsystem assembly can have a consideration impact in the
equipments functionality.
The clearance and thermal interference between the
adjacent inserts has an important influence on the satellite
equipments (such as the electronics box), which can cause the
satellite equipments failures.
It is important to note that the inserts used to support the
battery in the honeycomb panel have a serious impact on the
thermal conduction from the solar panel to the battery pack, for
this the primary ‘lesson learned’ is the need to fulfil more
stringent requirements for the battery thermal design and to
find the best inserts placement in the honeycomb panel for next
Algerian microsatellite production. Thermal control is critical
to ensure batteries provide the desired electrical performance
and long life.
REFERENCES
[1] A. Boudjemai, Alsat-1 Mechanical description report, Surrey Satellite
Technology Limited, November, 2001.
[2] M. Bekhti et al., Power system design and in orbit performance of
Algeria’s first micro satellite Alsat-1, vol. 78, no7, pp. 1175-1180,
Electric Power Systems Research, Elsevier, 2008.
[3] Luis Gomes, Alsat-1 Thermal document, Surrey Satellite Technology
Limited, March, 2003.
[4] A.M. Si Mohammed et al., Three-axis active control system for gravity
gradient stabilised microsatellite,pp.796-809, Acta Astronautica 64 796–
809, 2009.
[5] A.M. Si Mohammed et al., Analytic Solution of Nadir Attitude Pointing
for LEO Microsatellite, pp. 159-164, WSEAS TRANSACTIONS on
applied and theoretical mechanics, ISSN: 1991-8747, Issue 8, Volume 2,
2007.
[6] A.M. Si M ohammed et al., Yaw Phase Mode Attitude Control Using Z
Wheel for LEO Microsatellite , pp. 165-171, WSEAS transactions on
applied and theoretical mechanics, ISSN: ISSN: 1991-8747, Issue 8,
Volume 2, 2008.
[7] A.M. Si Mohammed et al., Simulation of microsatellite attitude using
Kalman filtering in orbit results, Simulation Modelling Practice and
Theory 16, 257–277, Elsevier, ISSN: 1569-190X, 2008.
[8] M. Bekhti el al., Temperature effects on satellite power systems
performance, pp. 57-62, Advances in Communications, Computers,
Systems, Circuits and Devices, ISBN: 978-960-474-250-9, 2010.
[9] A.Boudjemai, et al., Inserts thermal coupling analysis in hexagonal
honeycomb plates used for satellite structural design, Applied Thermal
Engineering 67 (2014) 352-361, Elsevier 2014.
[10] A. Boudjemai, R. Amri, A. Mankour, M.H. Bouanane, D. Boutchicha,
Modal analysis and testing of hexagonal honeycomb plates used for
satellite structural design, Materials and Design 35 (2012) 266-275.
[11] A.Boudjemai, M.H. Bouanane, L. Merad and M.A. Si Mohammed,
Small Satellite Structural Optimisation Using Genetic Algorithm
Approach, 3rd International Conference on Recent Advances in Space
Technologies, RAST '07, pp. 398-406, Istanbul, Turkey, ISBN: 1-4244-
1057-6, © 2007 IEEE.
[12] A.Boudjemai M.H. Bouanane A. Mankour H. Salem R. Hocine R. Amri,
Thermo-mechanical design of honeycomb panel with fully-potted inserts
used for spacecraft design, 6th International Conference on Recent
Advances in Space Technologies, pp. 398-406, ISBN: 978-1-4244-
9616-7, © 2013 IEEE.
