Batteries in Human Spaceflight (part 2)
In the first part of this two-part series, we looked at the rules, regulations, and realities of how space batteries are engineered to be maximally energy dense and minimally risky. In this second part, Kush Sutaria takes us through some existing battery designs used in space and discusses their pros and cons.
In this article we’ll cover:
Overview of the International Space Station (ISS).
The ISS’s battery upgrade from Nickel-Hydrogen to Lithium-ion.
Comparison of Nickel-Hydrogen and Lithium-ion.
Performance analysis of the ISS’s batteries, taken from operational data.
How private space companies are lowering the cost of battery development.
Batteries used in spacesuits.
Thermal runaway testing and internal short circuit cells.
Let’s get spacey!
The International Space Station
The International Space Station (ISS) is the largest modular space station in low earth orbit (LEO) and is used for microgravity and space environment research. The ISS is an international project involving five space agencies and is estimated to be the single most expensive item ever constructed, costing roughly $150B USD as of 2010, with operations and maintenance costs at over $1B USD per year.

In LEO, each orbit of the earth is around 90 minutes, which means the spacecraft sees 45 mins of sunlight and 45 minutes of darkness per orbit. The first part of this series has some info on the harsh conditions in LEO (for example, temperatures can range from -65°C to +125°C). When in the sun, solar panels power the spacecraft and charge the batteries, which then power the spacecraft during the dark portion of the orbit. The ISS uses batteries on the exterior of the spacecraft. This minimizes risk to the crew in the event of a battery failure but does mean a spacewalk is needed to service or replace a battery. Because a battery is designed to be replaced, they are classified as ORUs (Orbital Replacement Units).
ISS Li-ion Battery Upgrade
The ISS originally used Nickel-Hydrogen batteries, but these were replaced with Li-ion over a period of several years. From 2016 to 2021, 48 Ni-H2 battery packs were replaced with 24 Li-ion battery packs. This replacement took 4 flights of the Japanese HTV cargo vehicle and 13 astronauts conducting 14 spacewalks.
Li-ion has several advantages over Ni-H2. The below charts compare typical space rated Ni-H2 batteries to the GS Yuasa LSE110 Lithium Cobalt Oxide space qualified battery cell.
Specific energy and energy density are superior for Li-ion, however Ni-H2 boasts much higher cycle life. At 80% depth of discharge cycling, the GS Yuasa cell can last around 5000 cycles before degrading to 80% capacity. This can of course be much improved by lowering the depth of discharge. Other points to note are that Li-ion is generally more expensive and is susceptible to thermal runaway (we talked about thermal runaway in the first part of the series). The Ni-H2 batteries NASA used also required pressure vessels to accommodate the pressurized hydrogen (which Li-ion does not) and required charging according to a pressure-temperature algorithm.
The timeline for the replacement:
2009 - feasibility study
2011 - development approved
2014 - production started
2016 - replacement begins
2021 - replacement finished

Each Li-ion battery pack has an adapter plate and interfaces with a Battery Charge Discharge Unit (BCDU) which provides the main charge/discharge path, as well as telemetry and commands. The adapter plate likely serves as the mechanical interface to the already existing BCDUs on the station, so that the new Li-ion batteries can continue to use the same power and data paths that are on the BCDUs that were designed for the original Nickel Hydrogen batteries. There are 2 internal heaters per cell for thermal conditioning and NASA data (below) shows the cells operating between 17° and 21°C on orbit. The packs are expected to have a service life of over 10 years, and as of October 2018, they had already undergone 10,000 cycles.

A comparison of old vs new is provided below:


The voltage range of the Li-ion packs appears to be higher than the old Ni-H2s and the total capacity is significantly higher. The decision to not use more than 1 cell in parallel was likely because the cells are already high capacity and adding more cells in parallel increases wiring complexity. When multiple cells are connected in parallel, there is also a potential risk of a shorted cell in the pack causing a short circuit across all other cells connected in parallel with it.

ISS Battery Performance
The latest update on battery performance was in 2018, with NASA saying that batteries are performing well after 10,500 cycles. NASA data suggests that the cells are being operated between 3.97V and 3.85V. End of discharge voltages (EODs) are within 10mV of each other with cell temperatures within 5℃ of each other. This suggests uniform degradation and good cell matching. Such a narrow voltage window ensures high cycle life by limiting the damage which is done at very high and low states of charge.
Initial and annual on-orbit capacity tests have been performed, with some data provided below.

These results are inline with expectations from the manufacturer.
Commercial spacecraft batteries
Commercial space companies tend to not give away too much design information on their batteries to protect IP, but there is a shift towards using more commercially off-the-shelf cells and chemistries, with 18650s already in use and 21700 designs now being tested. Generally, human-rated spacecraft designs use proven technology (18650s) and are hesitant to use newer designs until thorough safety testing has occurred. Most space designs so far have used 18650 or prismatic form factors. Commercial space ventures need to keep costs and development time low while still maintaining high standards of safety and performance, as they also need to make a profit, unlike government space programs. This is achieved by use of COTS parts (commercially off-the-shelf available), streamlined processes and procedures, and utilizing new developments in technology. Space companies don’t have the resources or know-how to manufacture or develop their own cells, so buying as much as they can externally helps to lower costs, especially with the constantly decreasing battery prices which come with the proliferation of EVs and stationary storage.
Spacesuit Batteries
NASA plans to return astronauts to the moon by 2026 and part of this means having spacesuits - or Extravehicular Mobility Units (EMUs) to walk on the surface. The current Li-ion battery used in spacesuits for spacewalks on the ISS uses 80 2.4Ah Moli 18650 cells arranged in a 16P5S (16 cells in parallel and 5 in series) configuration. The total battery pack has a capacity of 35Ah (650Wh) in a usable voltage range of 16 - 20.5V. The cell design has been thoroughly tested and is unlikely to sidewall rupture (remember this from part 1?). However the pack has insufficient vent paths for thermal runaway products, and adjacent cells are not protected well enough from neighboring cells to avoid thermal runaway propagation (remember this too?). This means that a single cell thermal runaway could cause other cells in the pack to also go into thermal runaway, leading to a cascading effect, and potentially destroying the entire pack and causing a large fire. A single cell thermal runaway is unlikely, thanks to thorough screening and using the cells within their operational limits, however, you can never escape the possibility that an internal defect from manufacturing could cause an internal short. As of 2016, this battery had been successfully used on over 22 spacewalks.


NASA has awarded contracts to 2 commercial space companies to build spacesuits for the moon, Axiom Space and Collins Aerospace, so it is up to these companies to redesign the battery to be compliant with all the regulations we talked about in the first part of the article, as the current battery is not.
NASA did some work to explore potential designs which would be passively propagation resistant (remember that from the first part?). These designs utilize an Aluminum heat spreader between the cells, which acts at structural support and dissipates heat generated by cells. Cells also have appropriate separation to minimize the risk of propagation and the cells have mica, shrink wrap and air gap insulation to protect them from overheating in case a nearby cell goes into thermal runaway. Fuses made of Nickel are used on the negative side of the cell. The design achieves 191Wh/kg whilst being passively propagation resistant. More info can be found here.

Interestingly, because the heat sink is so efficient at removing heat from the cell, during testing, NASA had trouble using a heater to get a cell to go into thermal runaway. To solve this, NASA developed an ISC (internal short circuit cell) device which uses a meltable wax pad within the cell to trigger the cell into thermal runaway. The idea is that only a small amount of heat is needed to melt the wax, which then creates an internal short between the two electrodes. This method ensures a more reliable and controlled thermal runaway initiation and is more representative of a field failure of a cell. These devices can be implanted into cylindrical and pouch cells. These cells are now available on the market in various form factors for testing thermal runaway.


Learning from space
Technologies from space programs often find their way into everyday life products and batteries should be no exception. Space batteries are designed to be extremely safe, due to the heightened consequences in case of a failure, and reducing mass is of critical importance. Cost reduction is now also a driving factor for private space companies. These aspects make the lessons learnt directly applicable to other vehicle types, such as EVs and electric aircraft. Hopefully, the developments pioneered for space batteries can trickle their way down into the consumer market.