QUALIFICATION AND LIFE TESTING OF LI-ION VES 16 BATTERIES

In the frame of an ESA GSTP 5.2 activity (contract 4000105105), the qualification and life testing of a Saft range of Li-ion batteries based on VES16 cells and theirs autonomous simplified balancing system (SBS) has been carried out. In this abstract the development, qualification plan and successful results from all various tests conducted on the VES16 qualification battery modules are synthetized. Up to the present time, Saft batteries have been mainly utilizing for space applications high capacity cells, like the 45 Ah VES180 and the 35 Ah VES140 cells, targeting predominantly space missions in Geostationary Earth Orbit (GEO). However following the qualification and commercialization of the Saft 4.5 Ah VES16 cell in October 2011 [1] & [2], Saft has been developing and qualifying in the frame of this ESA GSTP 5.2 contract, VES 16 batteries for space missions, targeting both GEO and Low Earth Orbit (LEO) satellite missions. The electrochemistry of the VES16 cells used for the battery modules under the ESA qualification program is not novel. For VES16 cells, the Saft knowhow from large capacity space cells used for space applications, since SMART 1 mission in 2003, has been tailored for a cell with smaller capacity in order to facilitate the modular philosophy that has been deployed in this battery range. The VES16 battery design is based on standard building packs of different voltages, capacities and mechanical arrangements, in order to address most of the present and future space missions’ needs. The GSTP-5 tested VES16 battery designs are characterized by modularity, thus enhancing customization capabilities as per mission requirements and minimizing the number of relevant deltaqualification activities required for individual space missions. VES16 battery range, proposed by Saft, adopts a S-P configuration, since cells are connected in series (S) to form a string and strings were connected in parallel (P) to form a battery pack. The latter can individually form a battery module or a number of packs can be connected in parallel in order to form battery modules depending on capacity requirements. The Saft VES16 battery line offers batteries ranging from 6 to 10 S and 4 to 8 P configurations (Figure 1). This range includes a single deck also called “flat” (Figure 2) and a double deck (Figure 3) design aiming to fit any satellite accommodation. The aim of this qualification is not to test all the relevant modules but to test a number of representative ones for space applications, which have been called QM1 (2x10S8P), QM2 (10S5P) and QM3 (8S4P) detailed later in this document. Figure 1: Modules energy range from 384 Wh to 1280 Wh (EOCV = 4.1 V per cell) The range can be extended to higher voltages configurations by simply increasing the number of cells in series configuration. Indicatively, an 11S or 12S case can be derived from the 10S case by a simple homothetic arrangement. Figure 2: Single deck modules family DOI: 10.1051/ , 71617009 16 E3S Web of Conferences e3sconf/201 ESPC 2016 17009 (2017) © The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0 (http://creativecommons.org/licenses/by/4.0/). 1 BATTERY MODULES DESIGN Figure 3: Double deck modules family. The single deck battery module is an arrangement of cells, glued by resin, between a baseplate and a top plate. Connection to the cell can is made with tabs to the cell case, for the cathode, and to the cell top cap, for the anode. All strings of cells are parallel connected by bus bars. On the terminals of each individual cell a balancing system is implemented, called simplified balancing system (SBS). The principle of this balancing electronics system is based on cell current consumption above a given voltage threshold U0 (Figure 4). The system is fully autonomous and requires neither command nor telemetry. To achieve this balancing function, the SBS architecture is based on a simple circuitry which core component is a micro shunt. This micro shunt component quality level, selected by Saft in order to optimize SBS function, is QML-Q (Class 2 component with specific additional upgrade tests and characterization) as baseline. But the same component is qualified in class 1 (QML-V) and fully interchangeable.

Up to the present time, Saft batteries have been mainly utilizing for space applications high capacity cells, like the 45 Ah VES180 and the 35 Ah VES140 cells, targeting predominantly space missions in Geostationary Earth Orbit (GEO).However following the qualification and commercialization of the Saft 4.5 Ah VES16 cell in October 2011 [1] & [2], Saft has been developing and qualifying in the frame of this ESA GSTP 5.2 contract, VES 16 batteries for space missions, targeting both GEO and Low Earth Orbit (LEO) satellite missions.The electrochemistry of the VES16 cells used for the battery modules under the ESA qualification program is not novel.For VES16 cells, the Saft knowhow from large capacity space cells used for space applications, since SMART 1 mission in 2003, has been tailored for a cell with smaller capacity in order to facilitate the modular philosophy that has been deployed in this battery range.
The VES16 battery design is based on standard building packs of different voltages, capacities and mechanical arrangements, in order to address most of the present and future space missions' needs.The GSTP-5 tested VES16 battery designs are characterized by modularity, thus enhancing customization capabilities as per mission requirements and minimizing the number of relevant deltaqualification activities required for individual space missions.
VES16 battery range, proposed by Saft, adopts a S-P configuration, since cells are connected in series (S) to form a string and strings were connected in parallel (P) to form a battery pack.The latter can individually form a battery module or a number of packs can be connected in parallel in order to form battery modules depending on capacity requirements.
The Saft VES16 battery line offers batteries ranging from 6 to 10 S and 4 to 8 P configurations (Figure 1).This range includes a single deck also called "flat" (Figure 2) and a double deck (Figure 3) design aiming to fit any satellite accommodation.The aim of this qualification is not to test all the relevant modules but to test a number of representative ones for space applications, which have been called QM1 (2x10S8P), QM2 (10S5P) and QM3 (8S4P) detailed later in this document.The range can be extended to higher voltages configurations by simply increasing the number of cells in series configuration.Indicatively, an 11S or 12S case can be derived from the 10S case by a simple homothetic arrangement.The battery module can be customised to include the required type and number of thermal sensors, heaters and monitoring connectors, even thermostats if needed.The connectors can be rectangular or circular, redundant or not, with separated or mixed polarities, as specified by Battery General Design and Interface Requirements.
It is noticeable that the individual cell voltages are accessible on ground for any maintenance or health status check, by dedicated ground connectors, until the last days prior to launch.
The double-deck battery design is exactly the same as the single deck.The main principle is that the baseplate is shared between the two decks, so that the cells are glued on both sides of the baseplate.Therefore, the heart of the modules remains unchanged for all the VES16 batteries range.
This section describes the development and the design & analysis results at cell, SBS and battery level.The development plan is based on three main steps; the VES16 qualification plan, the SBS electronics qualification plan and the battery range design.The design of the modules was justified mainly by the electrical, electronical, thermal and mechanical analyses.
In the frame of the GSTP-5.2qualification Tests campaign, Saft completed these analyses, by detailed qualification tests results.
Detailed analyses have been conducted for single and double deck modules, for LEO and GEO cases.These analyses have demonstrated that the battery can sustain the various stresses generated by space environment, especially, the space missions' specific electrical, electronical, thermal and mechanical environmental requirements.
Hereafter, the main findings from the thermal, mechanical, reliability and radiation analyses are presented.
As a summary of thermal analysis, all the battery parts (cells, electronic components and wiring) remain within their derated values, (rated values in failed cases) with margins, for all 9 cases studied for single and double deck batteries.
The gradients within a string or within a module remain lower than 4°C, in all cases, as can be seen in Figure 7.

Design main justifications
Battery modules mechanical behaviour: The modal analyses provided the natural frequencies of the structures and cells, which were the following: -Above 400 Hz for the 10S8P module (QM1), -Above 1000 Hz for the 10S5P module (QM2), -Above 1090 Hz for the 8S4P module (QM3), The mechanical analysis showed that for random loads the structure and cells MOSyield are all positive and the minimum is locally for the cells' structure when the battery is loaded along the vertical-axis.

Reliability
The reliability analysis concluded that for a 12.5 years LEO mission, taking into account the loss of one string, a probability of success is 99,86 % and 99,99 % for the QM1 battery (2x10S8P) and the QM2 battery (10S5P) respectively.

Radiations:
The VES16 cell can withstand radiation levels up to 5 MRads.The only battery item that can be sensitive to radiations, is the SBS core component.
The component has been qualified, by ELDRS test, with 36 rad.s-1 rate, at 140 krads total dose.
In order to verify that this SBS core component will not be a limitation for the VES16 battery in orbit life, radiation and shielding analyses of the VES16 battery design studies have been carried out for the QM1 module.

Four mission cases have been considered:
• 12 years LEO mission at 700 km; • 12 years LEO mission at 1200 km; • 10 years MEO mission at 23616 km; • 15 years GEO mission at 35870 km.

Electronics:
The electronics SBS system is fully autonomous and requires neither command, nor telemetries.This system was developed with CNES support, to address all relevant space design requirements and issues.
The battery system has been modeled, using P-spice, following ESA experts approach guidelines, to assess its behaviour in worst case conditions @ EOL. [7] Figure 6: SBS model diagram The model has been correlated with real battery balancing test, achieved on the QM3 battery.
In the frame of the ESA GSTP 5.2 Qualification activity, three different batteries have been selected to be submitted to the qualification campaigns.
Those include two different single deck "flat" modules; a 8S4P (called QM3, shown in Figure 7) a 10S5P module (called QM2, shown in Figure 8) and also a 2x10S8P double deck module (called QM1, shown in Figure 9).In Table 1, the Qualification test matrix for the three modules is presented.As planned and foreseen, all three modules were submitted to characterization and environmental tests.QM3 battery is presently submitted to LEO life tests.
In Figure 10, the effect of discharge current while keeping a fixed charge rate C/10 is shown.

VES16
The main objective of the QM2 battery characterization test campaign was to characterise the battery at different charge / discharge rates and temperatures.The aim was to ease construction of a battery performance detailed meshing, for future batteries sizing, including not only cell, but also battery architecture contribution (parallel and series connections in particular).The characterization campaign has been completed successfully, for QM2 battery, and the battery has been authorised to pass through the environmental tests campaign as described in the following paragraphs.
Balancing system capability demonstration was performed, by introducing a 50 mV voltage deviation/unbalance between a single cell and the rest of the cells, in the module.The modules were then submitted to LEO (20% DoD), and GEO (70% DoD max), cycling in order to check the duration or number of charge / discharge cycles needed to re-equilibrate / balance the string voltages.
The 50 mV value, between average cell and unbalanced cell, was chosen since this value corresponds typically to the relevant unbalance obtained after 6 months of storage period.In Figure 13 and Figure 14, it is shown that the initial unbalance is recovered after less than 30 days in LEO, and less than 15 eclipse days in GEO cycling.Once electrical behaviour of the three batteries QM1, QM2 and QM3 was assessed, by the characterization tests completion, authorization was granted to continue the qualification tests sequences with thermal, mechanical and other Space missions' environments.
The QM1 module was submitted to environmental tests, consisting in: • Mechanical vibrations; • Shocks tests and; • Thermal vacuum cycles QM1 battery was tested in following mechanical vibrations: • Sine vibrations (24g) • Random vibrations (12.6 Grms).Three impacts were realized per axis, with battery discharges and battery voltage noise measurements, during shocks.
In conclusion, the QM1 environmental mechanical sequence has been performed successfully.
The QM1 battery was submitted to thermal environmental tests, in thermal vacuum chamber, in order to be as close as possible to VES16 batteries mission operating conditions.Battery leak rate was measured to 2.8 10-7 mbar.l/s,compliant to the leak rate criterion.
QM1 battery successfully passed all the Qualification tests Sequence.
The first resonance frequencies measured, in module width axis, were above 1000 Hz, which correlates well with the relevant simulations.Three impacts were realized, per axis, with battery discharges and battery voltage noise measurements during shocks.
In conclusion, the QM2 environmental mechanical sequence has been passed successfully.
The QM2 battery was submitted to thermal environmental tests in thermal vacuum chamber to be as close as possible to VES16 battery mission operating conditions.The resonance frequencies were above 1100 Hz, on all axes, which correlate well with the relevant simulations.
In short, the QM3 environmental mechanical sequence has been performed successfully, as shown by battery impedance and final functional tests results.
LEO life tests have been initiated with the QM3 battery, at the following conditions: • Real Time LEO, 20°C • DoD = 30% • Charge C/3 during 65 minutes up to 4.05 V per cell (or I < 80 mA) • Discharge C/2 during 35 minutes.
The above conditions were selected to be stringent enough to fit with mostly needed life test qualification conditions; but also to be comparable with existing Saft cell level running life tests.Relevant Saft Life test, at cell level, is referenced E1-07: 20000 cycles have been achieved on 3 cells and 18000 cycles achieved on 3 other cells.In Figure 23, the EoDVs for the 24 cells of QM3 battery during its LEO life test are shown.Despite the fact that this is only the beginning of the life test, the results obtained to this moment are satisfactory.As it is shown in Figure 30, there is only a 7.5 mV deviation between all cells at 3.65 V, at the end of each discharge.
The current life test is of great interest for potential future LEO applications since the QM3 battery is fully representative of a complete flight system.
All characterisation and environmental tests carried out on those 3 qualification batteries have been successfully carried out, which allow providing a complete battery designs ranges, qualified to stringent environments and available for the space community.This activity, supervised by ESA, has greatly helped Saft to reinforce its battery design performances justification.Today, more than 15 contracts, embedding VES16 batteries, have been awarded to Saft, by all European major primes and agencies both in LEO and GEO projects.
VES16 cells and batteries are cutting edge designed, multi-purpose missions and extended demanding needs, robust, modular, made for long life cycle with space mission required reliability.

Figure 11 :
Figure 11: Battery Charge & Discharge @ 30°Ccontinuous monitoring of individual cells All results obtained during electrical characterization of the QM1 battery were in line with expectations.

Figure 13 :
Figure 13: QM3 Battery voltage spread recovery after unbalance in LEO

Table 1 :
Qualification Test Matrix

Table 2 :
Thermal vacuum tests sequence