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

Been through this exact headache on my narrowboat. Both readings drifted apart after a few weeks and I didn't know which to trust either.

Short answer: trust your BMS for protection, trust your BMV-712 for daily monitoring — they're doing different jobs.

The BMS is watching cell-level voltages and will cut off if something goes wrong. But its SOC calculation is often fairly crude — many budget BMS units just infer it from voltage, which is notoriously unreliable mid-charge.

The BMV-712 uses coulomb counting — it's measuring actual current in and out of the pack. Once it's properly calibrated and you've set your:

• Battery capacity (accurately — not the optimistic figure on the label)
• Peukert exponent
• Charge efficiency factor
• Tail current / charged voltage thresholds

...it becomes genuinely precise. The key is letting it hit a full synchronisation cycle — charge to 100% absorption, hold until tail current drops, and it resets its reference point. Do that regularly and the drift disappears.

Where people go wrong is buying Fogstar or similar cells rated at say 200Ah, punching in 200Ah on the BMV, but never accounting for real-world efficiency losses.

On my boat I run a Victron SmartShunt alongside the BMS and after sorting the settings, they now agree within 2-3% on a bad day.

What BMS are you running? That'll make a difference — a Daly behaves very differently to a JK or an Electrodacus.

Anyone else had this with a particular BMS brand being wildly optimistic near the top of charge?

@VDH_Boats nailed the key distinction there.

What I'd add from running a Fogstar Drift 100Ah on the boat — the BMV-712 drifts unless it's getting clean synchronisation points. Mine was wildly optimistic for months until I realised my solar was never quite pushing the battery to true 100% before cutting off. The Victron never got its "full charge" reset trigger, so it was just accumulating Coulomb-counting errors.

Fixed it by tweaking the absorption voltage threshold in the BMV settings to match what the battery was actually hitting. Suddenly both readings aligned within 2-3%.

The BMS meanwhile had been quietly correct the whole time — it knows cell voltages directly rather than inferring from current flow.

So if yours are diverging, I'd look at whether your BMV is actually syncing at full charge before assuming either unit is faulty.

Yeah had this exact thing at the cabin with my setup. The BMV-712 drifted noticeably until I got the settings dialled in properly — shunt capacity, Peukert exponent, charge efficiency factor all need to match your actual battery chemistry.

Key thing people miss: the BMV learns over time but only if it's hitting proper sync points regularly. If you're not reaching 100% SOC often enough, it never gets a chance to recalibrate itself.

Worth checking your tail current setting too — if it's wrong the BMV won't recognise when you've actually hit full charge and the drift just compounds over weeks.

@DeclanKnight57 what settings ended up making the biggest difference for you? I'm in a similar boat (no pun intended) with my garden office setup — running a Victron BMV-712 alongside a 200Ah Fogstar Drift and the two readings have been slowly diverging over the past month.

I've tweaked the charged voltage threshold but wondering if my Peukert exponent or charge efficiency factor are off. Also not sure I've got the tail current set correctly for LiFePO4 — I think the defaults are more geared towards lead acid?

Would be useful to know what values others are running for a lithium setup specifically.

@DeclanKnight57 and @Macca97 — the setting that transformed my BMV-712 accuracy was the Peukert exponent. Most people ignore it completely and leave it at default.

On my motorhome setup with a 200Ah lithium bank, I spent an embarrassing amount of time chasing the SOC drift before realising my charged voltage threshold was also slightly off — the BMV wasn't registering a proper full charge to reset from, so errors just compounded over time.

Get those two dialled in together:

• Correct Peukert for your battery chemistry
• Accurate charged voltage (Fogstar publish this for their Drift cells)
• Tail current set properly

Once the BMV has a clean synchronisation point at 100%, it tracks remarkably well between cycles. Mine rarely disagrees with the BMS by more than 2-3% now across a full discharge.