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

Had exactly this head-scratcher last summer when I first commissioned the lithium bank in my cabin. The BMV-712 was reading 78% whilst the BMS app on my phone was showing 91%. Drove me slightly mad for a week.

After a lot of digging, the issue came down to calibration and what each device is actually measuring.

The BMV is coulomb-counting — it's tracking every amp flowing in and out from a reference point. If that reference point (your 100% sync condition) isn't set correctly, it drifts. Mine was syncing too eagerly at a voltage that didn't actually represent full charge on my Fogstar Drift cells.

The BMS, on the other hand, is typically using cell voltage curves to estimate SOC, which can be equally unreliable in the flat middle section of a LiFePO4 curve where 50% and 80% look virtually identical.

My honest take — neither is gospel on its own. What I do now:

• Trust the BMV for day-to-day trending and consumption tracking
• Use the BMS as a safety backstop, not an SOC gauge
• Let the system reach a genuine full charge periodically so the BMV can resync properly

The Victron community forum has some excellent threads on dialling in the charged voltage and tail current settings for a proper sync. Worth spending an evening there.

Curious what BMS you're running and whether it's giving you raw cell voltages — that would help narrow down which reading is further off.

@ForestJenny had exactly this with my narrowboat setup last year. Quick question before anything else — had you properly configured the BMV-712's battery capacity and Peukert exponent for lithium? Out of the box it's often defaulted to lead-acid settings which throws the coulomb counting right off.

Also worth checking: is your BMS doing cell-level voltage balancing and reporting SOC based on that, or purely coulomb counting like the BMV? They're fundamentally different methods so some divergence is almost expected.

What I ended up doing was syncing both to 100% after a confirmed full charge (watching for the tail current to drop right off) and seeing which drifted first. Mine turned out to be a dodgy shunt connection causing the BMV to miss small loads. Worth checking your shunt terminals are torqued properly?

@ForestJenny classic "two clocks, neither showing the right time" situation 😄

Short answer: neither is wrong, they're just measuring different things.

• BMV-712 does coulomb counting — tracks actual electrons in/out. Needs a proper synchronisation event (full charge to 100%) to stay accurate. If it never

@ForestJenny the BMV-712 is a coulomb counter — it measures what goes in and out — so if your charged state reference wasn't set correctly at commissioning, it'll drift from reality like a poorly-anchored boat. Your BMS is estimating SOC from cell voltages, which isn't terribly accurate mid-cycle due to lithium's notoriously flat discharge curve.

Practical fix: let the bank charge to 100% absorption, hold it there until tail current drops below ~1-2% of capacity, then let the BMV-712 synchronise — that's your accurate anchor point. After that, trust the BMV for daily tracking and use the BMS voltage reading mainly as a sanity check at the extremes.

Also double-check your Peukert exponent and charge efficiency factor in the BMV settings — factory defaults are often set for lead-acid, which will absolutely murder your accuracy on lithium.

@ForestJenny my Victron and my motorhome's BMS argued like a married couple for three weeks until I did a proper full charge cycle and let the BMV-712 sync at 100% — now they're basically best mates. 🤝