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

Had exactly this situation with my cabin setup last winter and it genuinely had me scratching my head for a while.

My BMV-712 was reading around 74% whilst the BMS was showing 81% — not a massive difference but enough to matter when you're relying on it for emergency backup and don't want any nasty surprises at 2am.

From what I gathered after a fair bit of reading:

• The BMV-712 tracks SOC via coulomb counting (measuring current in/out), so it's only as accurate as its calibration and how well you've set the charge efficiency factor and Peukert exponent
• Your BMS typically calculates SOC from cell voltages, which can drift at the "flat" part of the discharge curve on LiFePO4

Neither is perfect, essentially. The BMV tends to drift over time if it rarely hits a proper full charge to reset itself, whereas the BMS voltage-based method gets wobbly mid-range.

What sorted mine out was making sure the BMV was doing a proper sync — it needs to see the battery genuinely full (all cells balanced, current tapering right off) to reset its 100% reference point. Once I let it do a full absorption cycle on my Fogstar cells, the two readings came much closer together.

Worth checking:

• Is your charged voltage threshold set correctly in the BMV settings?
• Has the battery actually reached a proper full charge recently?

Curious whether anyone has found one consistently more reliable than the other for daily use, particularly with LiFePO4? And does anyone run both readings on a Cerbo GX display — does that help or just create more confusion?

@PartnerNomad had this exact thing with my shepherd's hut setup. Honestly? I trust the BMV-712 over the BMS most of the time.

The BMS SOC is often just estimating from voltage, which is notoriously unreliable mid-cycle. The BMV does proper coulomb counting — it's actually measuring what's gone in and out.

Main thing to check though:

• Is your shunt correctly installed (all loads/charge sources going through it)?
• Have you done a proper sync/calibration recently — let it hit 100% at full charge?

If the shunt isn't seeing everything, it'll drift. Mine went way off until I realised my Renogy MPPT was wired before the shunt. 🤦

Once wired properly and synced, my BMV has been rock solid. BMS SOC still wanders about a bit.

Great topic @PartnerNomad. Worth pointing out that the discrepancy often comes down to how each device calculates SOC. The BMV-712 uses coulomb counting — measuring actual current in and out over time — whereas most BMS units estimate SOC primarily from cell voltage, which can be quite optimistic, especially at mid-range states of charge where the voltage curve is relatively flat.

A few things to check: has your BMV been properly calibrated with the correct battery capacity and Peukert exponent? And crucially, when did it last sync to 100%? If it's not hitting the tail current threshold during a full charge, it'll drift low over time.

That 7% gap is actually fairly typical in my experience. Neither device is perfect, but a well-calibrated BMV-712 with accurate settings will generally give you a more reliable picture of true usable capacity.

@TU_Power is bang on — but the real villain here is usually a dodgy charged voltage threshold on your BMV-712, meaning it never resets its coulomb counting to 100% properly and drifts over time like a tired sailor on night watch.

Check three things:

1. Charged voltage — set it ~0.2V below your actual absorption voltage
2. Tail current — typically 1-2% of your battery's Ah capacity
3. Charged detection time — 3 minutes usually works

Your Fogstar/Victron BMS is reading cell-level voltages and estimating SOC from that, which is notoriously unreliable mid-cycle on LiFePO4 due to the flat discharge curve.

Long story short: a properly calibrated BMV-712 wins every time — it's actually measuring electrons in and out, not guessing from voltage like the BMS is.

Had this exact scenario play out in my motorhome last spring. Both devices were "correct" — they just disagreed on when the battery was last truly full.

The BMV-712 only resets to 100% when it sees your configured charged voltage held for the tail current duration. If those settings don't match your actual battery chemistry, it drifts over weeks until the numbers look nonsensical.

What fixed it for me: I synchronised both manually to 100% immediately after a proper absorption-complete charge, then left them alone for a fortnight. After that the gap had narrowed to under 2%.

@OldSailor is pointing at the right culprit. Check your tail current setting on the BMV too — mine was left at the Victron default, which was far too conservative for my Fogstar cells.

@RustyTinker hit something important there — they're both "correct" in their own world.

What settled it for me was understanding that my Fogstar Drift cells have a remarkably flat discharge curve. Between about 20% and 80% SOC, the voltage barely moves. So any device relying heavily on voltage interpolation — and the BMS leans that way more than people realise — is essentially guessing in that middle band.

The BMV-712 wins that fight hands down, provided your shunt is properly sized and your tail current setting is dialled in correctly. Mine was set at 4% from factory — dropping it to 1% transformed the accuracy after full-charge synchronisation.

Trust the coulomb counter (BMV) for your day-to-day working range. Let the BMS do its actual job — protection. They're different tools.