Victron BMV says 100% but cells are drifting apart — BMS cutting out overnight

Had almost exactly this scenario last winter with my shepherd's hut setup — Victron BMV reading 100% while the Fogstar Drift cells were clearly unbalanced, with the BMS tripping somewhere around 3am consistently.

The core issue is that the BMV is measuring pack voltage, not individual cell voltages. It genuinely has no idea what's happening at cell level. A pack can sit at a seemingly healthy resting voltage while one rogue cell is quietly racing ahead to its upper limit and triggering the BMS overvoltage cutoff.

A few things worth checking:

• Absorption voltage — is your charger actually holding absorption long enough for the balancer to do its work? Passive balancers only activate near top-of-charge, so if you're bouncing in and out of bulk all day you'll never fully balance.
• Balance current vs. drift rate — if cells have drifted significantly (say >50mV apart), most onboard balancers simply can't keep up. You may need an active balancer or a deliberate top-balance cycle via a bench power supply cell by cell.
• BMV Peukert settings — worth double-checking these are calibrated correctly, though that's more of a SOC accuracy issue than the root cause here.

With my EV charging load (I occasionally pull from the hut battery into a small inverter for a Zappi top-up experiment — long story), any imbalance gets exposed brutally fast under high discharge rates.

What BMS are you running and what's the cell chemistry? Lithium iron phosphate will mask drift far more aggressively than NMC due to the flat discharge curve. Makes diagnosis genuinely tricky.

Others here had this with Daly units — curious whether you're seeing similar patterns

@HelenPhillips classic 3am BMS drama — your batteries have better timing than my teenager 😄

Few things worth checking though:

• The BMV reads pack voltage, not individual cells, so it's blissfully ignorant of drift
• Fogstar Drift cells need proper top-balancing before first use — did yours get that treatment?

@HelenPhillips this is almost certainly a top-balancing issue combined with the BMV not knowing what it doesn't know — it's calibrated to voltage, not individual cell state.

The BMV sees pack voltage hitting absorption threshold and declares 100%, but if one cell is racing ahead, the BMS sees an overvoltage condition on that rogue cell and shuts everything down. Classic Fogstar Drift behaviour when cells haven't been properly top-balanced before commissioning.

What worked on my van build:

• Manually top-balance cells to within 5mV at 3.65V before installing
• Set your Victron charge voltage slightly lower (say 27.2V on a 24V system) to give the active balancer breathing room
• Check whether your BMS balancing current is actually sufficient — passive balancers on cheap units are notoriously slow

What BMS are you running? That'll determine whether active balancing is even an option here.

@HelenPhillips had almost the exact same circus with my cabin setup — Victron BMV was confidently lying through its teeth at 100% while one rogue Fogstar cell was basically phoning it in at 3.35V when the others were sat comfy at 3.65V 🙃

Worth checking your absorption time in the Victron charger settings — mine was way too short so the cells never got proper top-balance time before the BMV declared victory and went home early.

Also: does your BMS have active or passive balancing? Passive on a big pack is basically just vibes at that capacity.

@HelenPhillips one thing nobody's mentioned yet — what's your tail current set to in the BMV? If it's too high, the BMV declares 100% before the pack has actually finished absorbing, which masks the drift entirely.

On my narrowboat with Fogstar cells I had to drop tail current to around 1% of capacity before the SOC readings became trustworthy. Also worth checking your absorption voltage — if it's marginal, weaker cells never get properly topped up each cycle, and the drift compounds night after night until the BMS finally loses patience.

What does your BMS log show as the high/low cell voltages at the point of cutout? That'll tell you whether you're looking at a balancing problem or a genuine weak cell.