Battery Ageing on a Leash
Lithium-ion cells begin to degrade the moment they step off the production line. Unlike in sapiens, where the prevailing notion (counter) is that an active lifestyle trumps a sedentary one (more on ageing in humans), the answer is more ambiguous for lithium-ion cells — degradation could be worse during either a rest phase or charge/discharge cycles, depending on specific conditions. Unravelling such mechanisms is key to taming degradation in cells.
Although lithium-ion cells are now the norm for EVs and a host of other applications, most datasheets lack information relevant for complex use-cases like ours, which include a lot of stop-start as well as various permutations of riding, charging, and resting. Moreover, it is well known that temperature plays a dominant role in governing the degradation rates, but cell ageing data at tropical temperatures like ours is usually scant. It is also no secret that battery packs (multiple cells in series and parallel) do not come cheap, and thus it became vital to decrypt this Daedalian system ourselves to understand what it takes to extend its life without compromising on performance.
So we built an arsenal…
of 500+ cell cyclers (Invincibles era type), housed in chambers set at different temperatures, and rigorously tested cells under a gamut of conditions. There is no universally accepted method of accelerated ageing in lithium-ion cells since total test time is a critical parameter, and hence we kicked this project off fairly early-on. This effort has translated into generating data equivalent of about 20 million km (and counting) of cumulative rides under varied (and often extreme) weather conditions.
How is this knowledge embedded into the Ather 450?
Our batteries come with a Battery Management System (BMS), whose job is to monitor, control and protect the packs. The main factors that influence battery ageing are its temperature, operating voltage range, and charge/discharge currents, each of which can be influenced by actions taken by the BMS. The know-how gained from internal cell tests is built into this intelligent BMS in the form of a bunch of algorithms, and the magic from this mix conjures up on the Ather 450 in several avatars.
Firstly, many of the control algorithms, as described in an earlier post, are designed to keep battery degradation rates to a low level. For instance, during an aggressive ride on a scorching summer day, one might begin to feel a slight de-rating in acceleration after traversing a fair distance. This is a way of steering battery temperatures away from the ‘high-degradation’ zones, while still ensuring that the user gets to his/her destination. There is a fine line between high performance and an acceptable degradation rate, and using our internal wealth of data, we are now able to push this envelope much further to allow for prolonged duration of maximum performance (motorcycle in a scooter’s skin, after all) on hot days as well.
Secondly, the BMS processes information from a multitude of sensors on-board to estimate and keep track of the battery’s health. As batteries age, their capacities reduce, and they produce more heat than before due to increased internal resistance. While improved range prediction and failure prevention/detection are obvious uses of knowing the pack’s health, this quantity also feeds back into the aforementioned control algorithms to adjust for the changed thermal properties of the pack — this is to ensure smooth transitions between modes over life (and not just for a new scooter). Moreover, since everything today is tuned such that we err on the conservative side, we might even relax a few limits at a later stage based on how the battery’s health evolves.
Thirdly, optimized charging protocols are another way of lengthening battery life. Since the resting voltage of the pack (directly correlated with range) also has an important say in the ageing game, we may roll out options for customers to regulate the amount of charge held in the battery as per their customary needs. Additionally, once the scooter is plugged-in for an overnight charge, we intend to modulate the charging cycle to reduce time spent in zones characterized by slightly higher degradation rates (after due permissions from the user). With respect to fast-charging, which is typically associated with faster ageing, we have deciphered ways of doing so without hurting the battery, and which also adapt to the battery’s health.
What’s in the pipeline?
Although we do not take chances with our battery’s health and leave no stone unturned to keep it in check (this post is not sponsored by any medical insurance folks), there is no way of completely arresting degradation — but we warranty that all battery packs will have at least 70% of original capacities at the end of 3 years, with no limit on the cumulative distance ridden. At some later point though, users may feel the need to replace batteries, for which we are currently estimating costs. While we speculate lithium-ion cell prices ‘x’ years from now (steadily falling and slated to continue doing so), explore options of buying back degraded batteries for use in less stressful second-life applications (semi-retirement-like), and incorporate more intelligence to extend battery lifespans, it can be safely assumed that the need for replacement is quite a distance away. In the meanwhile, we intend to keep pushing our cells to reconnoitre new frontiers, such as even faster charging times — all the while keeping battery ageing under a tight leash.
Originally published at https://blog.atherenergy.com on October 10, 2019.
