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

Mine showed a 12% difference until I realised the BMV-712 hadn't been properly synced after I swapped cells — basically it was counting from a fantasy number it invented, which is exactly what my ex used to do with the household budget.

Short answer: trust neither blindly, but here's my rule of thumb for my shepherd's hut setup —

• BMS is great for cell-level protection (it'll pull the plug before anything goes catastrophically wrong)
• BMV-712 is better for trend data and actual consumed Ah over time, once it's properly calibrated

The classic fix is letting the bank reach 100% full charge (proper absorption, not just bulk), at which point the BMV-712 auto-syncs and resets its SOC. After that mine stayed within 2-3% of the BMS reading consistently.

Also worth checking:

• Is your Peukert exponent set correctly for your battery chemistry?
• Is the shunt wired so everything runs through it? (I had my Fogstar battery's BMS bypass routing round mine for ages like a sneaky little gremlin)

For emergency backup purposes I personally watch the BMV-712 for daily decisions and let the BMS do the actual heavy lifting on protection.

Anyone else running Victron kit alongside a smart BMS — does yours ever just... disagree for absolutely no reason on a Tuesday?

@PanelRoger ha, "fantasy number" is exactly the right phrase for it.

Had the same scenario last spring when I reconfigured my Fogstar Drift 200Ah — the BMV-712 was stubbornly convinced it knew better than reality. The key thing people miss is that the BMV learns its sync point from a fully charged, absorption-complete cycle. If you swapped cells mid-cycle or without reaching float, it's just extrapolating from wherever it happened to be.

My approach now:

• Trust the BMS for cell-level accuracy
• Trust the BMV-712 for trend and consumption data
• Force a manual sync once you've hit a genuine 100% charge confirmed by your Victron MPPT via VictronConnect

The two should converge within a few cycles after that. If they don't, your charged voltage threshold in the BMV settings probably needs adjusting to match what your BMS actually considers full.

The BMV-712 is a coulomb counter — it tracks electrons in and out from a sync point, so if that sync point is wrong, every subsequent reading is precisely, confidently, wrong; your BMS meanwhile is measuring actual cell voltages directly, so for a freshly reconfigured pack, trust the BMS until the BMV earns it back.

Key fix: force a manual sync on the BMV-712 when your pack is genuinely at 100% (fully absorbed, tail current dropped to your configured value) — Settings → Battery → Synchronise SoC to 100%.

After one proper charge cycle with correct charged voltage, tail current, and charge detection time dialled in, the two should converge to within a couple of percent. If they still disagree wildly after that, your Peukert exponent or charge efficiency factor needs attention.

@OldSailor is right about the coulomb counter thing, and that's actually why I lean toward trusting the BMS for absolute SOC and the BMV-712 for relative tracking between syncs.

My setup has a Fogstar Drift 200Ah and I've found the two readings drift apart mostly when:

• Peukert losses aren't dialled in correctly on the BMV
• The charged voltage / tail current sync thresholds need adjusting

Worth checking your charged voltage setting in the BMV — if it's set too high for a LiFePO4 profile it might rarely (or never) sync automatically, so it just keeps accumulating error.

Once I got mine syncing reliably at full charge the two units now sit within 2-3% of each other, which I'm comfortable with. Any more than that and something needs investigating rather than just picking the number you prefer.

@OldSailor and @CableTieWarrior have covered the fundamentals well, so I'll add a practical angle from living aboard.

I actually run both and treat them as complementary rather than competing:

• BMS = absolute SOC reference (cell-level chemistry, knows when you're truly empty)
• BMV-712 = trend and consumption data (far better for monitoring discharge rates, load tracking, etc.)

The key setting most people miss on the BMV is the Peukert exponent and charge efficiency factor — get those dialled in for your specific cells (LiFePO4 wants ~1.05 and ~99% respectively) and the drift becomes negligible.

Force a proper sync by letting the BMS reach its 100% threshold with absorption complete, then let the BMV catch that tail current drop below your sync threshold. After that mine stays within 2-3% of the BMS for weeks.