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

Had this exact headache last winter with my cabin setup. Short answer — neither is perfect, but they're measuring different things which is why they disagree.

The BMV-712 is doing coulomb counting from the shunt, so it's tracking charge in vs charge out. Dead accurate if you've got your settings dialled in correctly (peukert exponent, charged voltage threshold, tail current). If those are off, the drift compounds over time.

Your BMS is typically working off cell voltages and its own internal model. Depending on the BMS quality, that can be quite rough — especially mid-range where LiFePO4 voltage curves are almost flat.

What I found helped:

• Make sure the BMV syncs properly at full charge — that resets the drift
• Set your tail current low enough that it actually detects full charge and syncs
• Don't trust either reading heavily between about 20–80% on LFP, voltage-based readings are basically useless there

For day-to-day I watch the BMV-712. It's the more reliable instrument if configured right. The BMS SOC I treat more as a sanity check or early warning system.

What BMS are you running? Some of the cheaper units (looking at you, generic Alibaba boards) have genuinely terrible SOC algorithms. If you're on something like a Daly or JK that might explain the gap.

Anyone else found a reliable way to get the two to agree consistently? Curious whether integrating via Victron's Cerbo changes anything for people.

Great question to kick off with @Lefty31 — welcome to the forum! 🎉

To add something practical: the BMV-712's coulomb counting drifts over time and needs a proper synchronisation point to reset accurately. If your battery never reaches a genuine 100% SOC (i.e., absorption and float fully completed), it'll accumulate error.

Your BMS meanwhile is typically working from cell voltages, which can be misleading mid-charge or under load due to voltage sag/rise.

My approach with my van setup:

• Trust the BMS for cell-level protection (that's its actual job)
• Trust the BMV-712 for longer-term SOC trends, but make sure your charged voltage and tail current settings match your actual battery chemistry

Worth double-checking your BMV's programmed capacity matches your real usable capacity too — Fogstar cells often have slightly more real-world capacity than the rated figure.

Really good thread this. To add some practical nuance — the BMV-712's accuracy degrades over time if it never gets a proper full synchronisation point. In my motorhome setup with Fogstar Drift lithium cells, I found the BMV was drifting 8-12% low after a few weeks of partial cycling because it rarely hit 100% SOC to reset itself.

The fix that sorted mine: set the "charged voltage" threshold carefully in the BMV settings — it needs to match what your BMS considers full, not just your charger's absorption voltage. They're often slightly different.

@Lefty31 — what battery chemistry are you running? That makes a significant difference to which device you'd lean on. For LiFePO4 specifically, I'd weight the BMS reading more heavily during the flat mid-section of the charge curve where coulomb counting tends to accumulate errors fastest.

@SimonKelly that's the real kicker isn't it — my Fogstar Drift cells almost never hit absorption long enough to trigger a proper sync, so my BMV-712 just drifts off into fantasy land like it's writing a novel.

@CopperMaker this is exactly my situation on the boat — Fogstar cells sitting at what the BMS calls 97% but the BMV-712 has drifted miles off because it rarely sees a proper sync point.

Practical question for the thread: is there a reliable way to force a manual sync on the BMV-712 without waiting for a full absorption cycle? I've been tempted to just hit the reset when I know the bank is genuinely full, but unsure if that's the right approach or whether I'm just masking the drift rather than fixing it.

Also wondering whether anyone runs both a BMV and a Daly/JK BMS and has found a sensible way to cross-reference the two readings as a sanity check rather than trusting either in isolation.

@RussWebb that resonates completely — spent three winters troubleshooting exactly this on my narrowboat before I finally cracked it.

The piece nobody mentions: your BMV-712 has a synchronisation threshold in the settings. Mine was defaulting to 100% SOC trigger at a voltage that my Fogstar cells genuinely never reach under normal charging.

Go into the BMV settings and drop that synchronisation voltage to match where your cells actually sit when full — I set mine to 27.0V on a 24V system, combined with a tail current of 2% of capacity. After that, the two readings converged considerably.

The BMS is still your safety device. The BMV is your fuel gauge. Get the sync point right and they'll at least be speaking the same language.