Victron BMV-712 showing different SOC to my BMS — which one do I trust?

Running into exactly this on my boat setup at the moment. Fogstar Drift 100Ah with its own BMS, paired with a BMV-712 — and they're regularly 8-12% apart, sometimes more after a heavy draw.

From what I've been reading, the BMV works on coulomb counting (tracking actual amps in/out) which sounds great in theory, but it drifts over time if your peukert exponent or charge efficiency factor isn't dialled in correctly. Most people seem to leave these at defaults and then wonder why the numbers wander off.

The BMS, on the other hand, is reading cell voltages directly and estimating SOC from that — which has its own problems because LiFePO4 has such a flat discharge curve that voltage is a pretty rubbish indicator across most of the range.

My current thinking is:

• Trust the BMS for protecting the cells (that's its actual job)
• Trust the BMV for longer-term trend tracking once it's been properly calibrated
• Use a full charge cycle to let the BMV sync up properly — it should reset to 100% when charge current drops to your tail current threshold

Has anyone actually gone through the full BMV calibration properly for LiFePO4? I adjusted my charged voltage to 14.2V and tail current to 4% but I'm not convinced I've got it right.

Also curious whether anyone running a Victron Lynx Smart BMS finds the two play nicely together — presumably Victron's own ecosystem talks to itself better than third-party combinations?

Hey @12V_King — classic dilemma this one! Short answer: trust the BMV-712 for day-to-day SOC, but don't ignore the BMS entirely.

The Fogstar's BMS is primarily there for protection (over/under voltage, overcurrent etc.) rather than accurate coulomb counting. The BMV-712, once properly configured with the correct battery capacity and Peukert exponent, is a far more sophisticated SOC tracker.

Key thing to check — have you set your charged voltage threshold and tail current correctly in the BMV settings? These determine when it resets to 100%, and if they're off, errors compound over time. For LiFePO4 I'd suggest around 13.5V charged voltage and 4% tail current as a starting point.

The bigger divergences after heavy draw are almost certainly Peukert-related — worth tweaking that setting.

What @RayTaylor said is solid, but I'll add something from bitter experience running a similar Fogstar/Victron combo at my cabin.

The gap you're seeing after heavy draw is almost certainly a Peukert effect issue. The BMS calculates SOC using a fairly simple coulomb-counting model that doesn't always account for how hard you're hammering the cells. The BMV-712 has a dedicated Peukert exponent setting — worth dialling that in properly for lithium (I run mine at 1.05).

Also check your charged voltage and tail current thresholds in the BMV settings. If those aren't calibrated correctly, the 712 never gets a clean synchronisation point and the drift compounds over time.

Once I sorted both, my two readings sit within 2% consistently — even after running the EV charger overnight off the battery bank.

The bit nobody mentions: that 8-12% gap usually narrows dramatically once you've let the BMV-712 learn your battery properly.

Mine was all over the shop for the first month in the motorhome. The fix was forcing a proper full charge cycle — I mean genuinely full, tail current down to 1-2% of capacity — so the BMV could nail its sync point. After that it settled.

Also worth checking your Peukert exponent setting. Fogstar's lithium cells don't behave like lead-acid, and the default BMV settings assume old-school chemistry. Dial it back toward 1.05 and you'll see the heavy-draw discrepancy shrink considerably.

The BMS is protecting the cells. The BMV-712 is measuring them. Two different jobs. Stop asking one to do the other's work.

BMV-712 wins every time for SOC — it's doing proper coulomb counting while the Fogstar BMS is basically just guessing from voltage curves like it's 1987.

That said, the real villain here is your synchronisation point — if the BMV never sees a proper full charge (absorption and tail current dropping below ~2% of capacity), it'll drift worse than my motorhome's compass near a service station.

Check your charge settings in VictronConnect: Charged voltage and Tail current are the two knobs that matter most. Fogstar Drift wants roughly 14.6V absorption with tail current around 1-2A per 100Ah.

@MountainHermit is right about the learning curve — mine was 15% out for weeks until I realised my solar controller was terminating absorption too early.