Victron lfp internal balance start voltage per cell?

Been thinking about this a lot lately after pulling apart an old battery bank and reusing the cells in a new setup.

Got a few of the older Victron LFP packs — the ones with the 3-pin comms lead and internal balancing circuitry baked in. They've done their time in my previous system and are showing the usual signs of age: capacity drift between cells, resting voltages that don't quite match up anymore.

My question is around the internal balancer trigger voltage. From what I can piece together, the passive balancing only kicks in above a certain per-cell threshold — somewhere around 3.4–3.5V if memory serves — but I've never found a definitive figure in Victron's documentation for these older non-smart units. The newer Smart LFP cells are well documented, but these older ones feel a bit like they were quietly swept under the rug.

Has anyone actually measured when the bleed resistors start doing their thing? I'm wondering whether my charge profile is even getting the cells into that window long enough to let the balancer do its job, or whether I'm just flogging a dead horse with cells this worn.

Might be worth running them through a slow top-balance externally before trusting the internal balancer to sort things out — but curious what others reckon.

Also, if anyone's repurposed similar older Victron packs into a DIY build alongside a proper external BMS (I've got a Daly sitting on the shelf), did you bother keeping the internal comms lead connected or just ignore it entirely?

Would love to hear from anyone who's been down this road.

@LazyBodger those older Victron packs typically start passive balancing around 3.45V per cell if I recall right — though it varies slightly by firmware version on the BMS.

Worth connecting via VE.Direct and pulling the data if you can, some of the older packs will actually report cell voltages individually. That'll tell you more than any spec sheet.

When I rebuilt a bank last year I found two cells in one pack were sitting noticeably low at rest — would've never spotted it otherwise.

One thing to watch — if you're mixing those cells into a new pack, make sure your new BMS balance threshold aligns with whatever the cells were previously conditioned to. Fogstar cells I've used are pretty tolerant but older Victron cells can be fussier.

What capacity are the packs you're pulling apart?

@DalesWanderer's figure sounds about right from what I've seen too.

Worth noting though — with those older internal BMS packs the balance current is pretty tiny, like 50–100mA. So if your cells are significantly out of balance from being pulled apart and rebuilt, don't expect the internal balancer to sort it quickly on its own.

If the drift is more than a few percent between cells I'd manually top-balance them first before relying on the internal balancer to catch up. Saves a lot of grief later.

Had a similar situation reusing some pulled cells in my van build and the internal balancer just couldn't keep up until I did a proper top-balance session first. 🔋

From my own experience repurposing older cells into a bank for my garden office build, I'd add that the balance start voltage matters less than you'd think if your cells are already well-matched going in.

Where I got caught out was assuming the internal balancer would fix a dodgy cell — it won't. It'll just run warm trying to drag the others down to match it.

@LazyBodger if you're reusing pulled cells, I'd strongly recommend doing a proper capacity test on each one first and grouping them by actual capacity before you even think about what the BMS balancing threshold is set to. A clapped-out cell in an otherwise healthy pack will cause you no end of grief regardless of where balancing kicks in.

Renogy's battery monitor made spotting the rogue cell in my setup much easier than I expected.

@CallumHobbs curious what you found with cell divergence over time after reassembly — that's often where the real issues creep in with repurposed cells.

One thing worth adding: even if the internal balancing kicks in at the right voltage, those older Victron packs can have quite a low balance current (sometimes as little as 50mA). If your cells have drifted significantly in capacity, passive balancing at that rate will take ages to bring them back into alignment — or may never fully catch up during a normal charge cycle.

Might be worth logging individual cell voltages with something like a Victron BMV or even a basic cell monitor before committing to a new build around them. Saves a nasty surprise down the line.

@DODQueen raises a really important point there. I'd add that in my experience the balance start voltage on those older Victron internal BMS units typically kicks in somewhere around 3.40–3.45V per cell, but the frustrating thing is it's not user-adjustable on the older packs — it's baked into the firmware.

What I'd watch for @LazyBodger is that if your cells have drifted significantly in capacity after being pulled from the original pack, the balancer simply won't have enough headroom to correct large divergences before the high-voltage cutoff triggers. It's a fairly passive, top-balancing approach.

If divergence is already noticeable at rest, you might be better off doing a manual top balance before trusting the internal balancer to sort it out on its own. Saved me a lot of grief on a similar rebuild last year.