Second-Life Batteries: When a Used Pack Is an Asset, and When It Is a Liability

A battery retired from an electric vehicle or a heavy industrial duty typically still holds a large share of its original capacity. The arithmetic is compelling: energy that has already been paid for once, available again at a fraction of the price of new cells, for applications far less demanding than the one it left. Second-life storage is not a marketing idea — it is a real and growing part of the stationary storage market, and we are asked about it regularly.

It is also the area where we most often advise caution, for one reason: a second-life pack is not a cheap new pack. It is an unknown pack. Everything that makes new cells predictable — a datasheet, a known production date, a documented duty history, a warranty — is either missing or must be reconstructed at your own cost.

What actually arrives

A batch of retired modules is rarely homogeneous. Cells that left the factory matched have since diverged, because the ageing they experienced was not uniform: position in the pack, thermal gradients, depth-of-discharge history and charge rates all leave their mark differently on each cell. Two modules with the same measured capacity can have very different internal resistance, and therefore very different behaviour under load and very different heat generation.

That spread is the central engineering fact of second-life work. It means the useful question is never "how many kWh did we buy?" but "how much usable, safely deliverable energy does this assembly have, and how fast is the spread widening?"

Three questions before a used pack has any value

First, state of health — measured, not claimed. That means capacity testing at a defined rate and temperature, internal resistance measurement per module, and a self-discharge check over a meaningful rest period. A short open-circuit voltage reading proves almost nothing; a cell with an internal defect can present a perfectly normal terminal voltage.

Second, history and provenance. Was the pack retired on a scheduled fleet replacement, or after a fault, a thermal event, or an accident? Mechanically damaged cells are not a discount opportunity, they are a hazard. Where the original BMS logs are available, they are worth more than any single measurement you can take afterwards.

Third, traceability of the cell type itself. Chemistry, format and original ratings determine what the pack can safely do next, and they determine your fire strategy, your enclosure and your charge profile. If nobody can tell you what the cells are, our answer is straightforward: we do not build with them.

The BMS requirement goes up, not down

There is a persistent assumption that because a second-life system is cheaper and less critical, it can carry simpler electronics. The opposite is true. A new, matched pack is forgiving; a repurposed one is defined by its imbalance. It needs per-cell or per-module monitoring, conservative voltage and temperature windows, and balancing capability sized for a divergence that will keep growing over the pack's remaining service life. This is precisely where active balancing earns its cost: passive balancing bleeds the strongest cells down to the weakest as heat, which in a mismatched second-life assembly means throwing away a meaningful share of the capacity you bought. Redistributing charge instead of burning it is the difference between a usable second-life system and a disappointing one.

Where it fits, and where it does not

Second-life batteries suit applications that are low C-rate, tolerant of derating, and non-critical in the safety sense: peak shaving and self-consumption on a commercial site, buffering behind a slow charger, off-grid or seasonal storage where floor space is available and an oversized bank is acceptable. Generous sizing is the friend of a repurposed pack — running it gently is what makes the remaining life long.

We do not recommend second-life cells for medical equipment, for telecom or data-centre backup where hold-up time is a hard commitment, for high C-rate duty, or anywhere the runtime figure is part of a safety case. In those applications the value of the battery is the certainty that it will deliver on the worst day, and certainty is exactly what a used pack cannot supply cheaply.

Count the real cost

The honest metric is cost per usable kWh over the remaining life, not price per nameplate kWh. Into that number go testing and grading labour, sorting and rejection losses, new enclosures and interconnects, a more capable BMS, and the compliance work that repurposing triggers — international standards work exists specifically for evaluating repurposed batteries, and reusing cells means you, not the original manufacturer, own the resulting product's safety case and warranty. Sometimes that total still comfortably beats new LiFePO4. Increasingly often, with new cell prices where they are, it does not. It is worth doing the calculation before committing to the idea.

If you are evaluating a second-life proposal, or weighing it against a new-cell system, we are happy to review the duty cycle and the numbers with you. Write to info@lmelectro.com — we respond within 24 hours.