Victron BMV says 100% but cells are drifting apart — BMS issue or something else?

Seen this exact thing on my narrowboat build last winter. The BMV was sat cheerfully at 100% whilst my Fogstar 280Ah cells were telling a different story through the BMS app — two cells lagging noticeably behind the others at the top of charge.

Root cause in my case was passive balancing not keeping up. The BMS (a JK unit) was only balancing above 3.45V per cell, and the bulk charge was pushing through so fast that the cells hit absorption, the charger backed off, and balancing never got enough time to do its work properly.

A few things worth checking in your situation:

• What's your charge rate relative to capacity? High C rates can mask drift that only shows at rest
• Is your BMS balancing threshold set appropriately? Dropping mine to 3.40V made a meaningful difference
• Are the cell interconnects torqued correctly? Loose busbars cause resistance imbalance that looks like cell drift

The BMV only sees the pack as a whole — it's essentially blind to what's happening at cell level. So 100% SoC from Victron is entirely possible whilst the cells are genuinely out of balance. The two systems aren't contradicting each other, they're just measuring different things.

Worth doing a slow top-up charge at a lower current and watching the BMS app in real time. You'll quickly see whether it's a balancing issue or whether one cell is genuinely weaker.

Anyone else running JK BMS units with Victron kit — curious whether you've tweaked the balancing parameters much?

@LochWalker classic Victron/BMS disconnect this — the BMV is measuring pack voltage and has no idea what's happening at cell level.

Few things worth checking:

• Absorption time — is your charger actually holding the cells at full voltage long enough for the BMS to balance? Passive balancers need the cells sitting at top-of-charge for a good while
• Tail current threshold in the BMV settings — if it's declaring 100% too early, the cells never get properly topped
• Balancing current on your BMS — Fogstar packs often ship with fairly weak passive balancers (30-60mA typically), which is almost useless if cell drift has already developed

Had virtually identical behaviour on my boat last spring. Dropped absorption voltage slightly and extended the absorption time — counterintuitive but gave the balancer more dwell time at the top. Sorted itself over about two weeks of cycling.

What BMS are you running alongside the Fogstar cells?

@LochWalker the drifting cells are almost certainly not fully top-balanced. The BMV reaches 100% via voltage + tail current — it has no visibility into individual cell voltages whatsoever, as @PanelEwan rightly notes.

What you need to do:

1. Charge to absorption, then hold until current drops to ~0.5–1% of capacity
2. Watch your BMS app — lagging cells need longer at top voltage to allow the passive balancer to work through them
3. If your BMS balancer current is only 50–100mA (very common on budget units), this process takes days not hours

Worth checking whether your Fogstar cells were storage-charged at different states when delivered — mine arrived slightly mismatched and needed a full 48-hour top-balance session on a bench charger before installation.

Active balancers like the Daly Smart or a dedicated Heltec unit will sort persistent drift far more efficiently than passive bleeding.

@LochWalker both @PanelEwan and @DevonDweller have nailed it, but here's the bit nobody's mentioned — your absorption voltage might be too low to ever trigger proper balancing in the first place.

If you're running passive balancing via the BMS, those resistors only bleed down the high cells when they're actually at top voltage for long enough. Rush through absorption and you're leaving the job half done.

Stick your charger on a proper bulk/absorption cycle — 3.55–3.60V per cell held until tail current drops below ~1% of capacity — and let it sit. Done this myself with Fogstar cells; took three full cycles before they settled within 10mV of each other.

The BMV will happily lie to you forever if your synchronisation parameters aren't tight. Check your "charged voltage" and "tail current" settings in the BMV config whilst you're at it.

Really good points from everyone above. One thing worth adding — check whether your BMS is actually allowing the cells to reach full absorption. Some BMS units will trip the charge FET early if any single cell hits the high voltage cutoff before the others have caught up, which means the lagging cells never get a proper top-up. It becomes a self-reinforcing problem; they fall further behind each charge cycle.

If your BMS has adjustable cell OVP settings, try nudging the per-cell cutoff up very slightly (within safe LFP limits, obviously — 3.65V absolute maximum) so the pack can stay in absorption long enough for the weaker cells to catch up. Even a few extra minutes makes a real difference with stubborn drift. Worth logging a few charge cycles through the BMS app to see exactly when it's cutting out.