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

Had this exact same battle on my boat last summer — the BMV reckoned I had a full 87% whilst the BMS was waving a red flag at 71%, and I didn't know whether to celebrate or panic.

Short answer: trust neither blindly, use both as a cross-reference.

The BMV-712 is a coulomb counter — it's measuring every amp going in and out, but it drifts over time if your "charged full" sync point isn't set correctly. If your charged voltage, tail current, and charged detection time settings are off, it'll never properly reset to 100% and the error compounds.

Your BMS, meanwhile, is calculating SOC from cell voltages — which is pretty accurate at the extremes (nearly full or nearly empty) but notoriously wobbly in the middle of the charge curve, especially with LiFePO4 where the voltage is famously flat.

A few things worth checking on the BMV side:

• Peukert exponent — set it to 1.05 for lithium, not the lead-acid default
• Charge efficiency factor — should be 99% for LiFePO4
• Synchronisation — is it actually hitting full charge and resetting?

On the BMS side, which unit are you running? Some of the cheaper ones (no offence to anyone's Fogstar or Daly setups) have SOC algorithms that are frankly optimistic at best.

Once I corrected my BMV settings, both units settled within about 4% of each other, which is close enough for a floating tin can.

What BMS are you running, and what are your current BMV settings? Others will have been through this — **would love to hear if anyone's found a

@SussexBoater the discrepancy usually comes down to which parameter is actually drifting.

BMV-712 calculates SOC via coulomb counting — it's accumulating tiny measurement errors over time. If it hasn't had a proper sync point (i.e., a genuine 100% full charge where tail current drops correctly), it'll wander off.

BMS figures are typically voltage-based, which has its own problems — voltage is a rubbish SOC indicator under load.

Neither is gospel, frankly.

What I'd check first:

• Has your BMV's charged voltage and tail current been set correctly in VictronConnect?
• When did it last fully sync?

For my garden office setup I had exactly this — BMV was optimistic by ~15% because the tail current threshold was set too high and it was syncing prematurely.

Fix the sync parameters first, then decide which reading to trust.

Been here myself with a 280Ah LiFePO4 bank + Victron BMV-712 combo.

Honest answer: trust neither blindly — they measure different things.

The BMV does coulomb counting, so if your charged/discharged efficiency settings are slightly off, error compounds over time. Mine drifted badly until I nailed the Peukert exponent and set a proper tail current for sync.

The BMS is watching cell voltages directly, so when it screams at 71% it's probably seeing a weak cell dragging the pack down early.

That 16% gap @SussexBoater suggests one of:

• BMV needs a full sync (let it hit absorption properly)
• You've got cell imbalance the BMS is catching
• Your BMV's charged voltage threshold is set too low

Check your lowest cell voltage at that 71% BMS reading. If it's already at 3.0V you've got an imbalance problem, not a calibration problem.

Same battle at my static caravan last winter, 200Ah Fogstar Drift cells with a JK BMS alongside the BMV-712.

What finally cracked it for me was tail current calibration. My BMV was set to a 2% tail current threshold from its lithium defaults — meant it called "full" whilst the BMS was still seeing elevated current and hadn't registered a proper top-balance.

Once I tightened that tail current figure down to 1% and let the bank do a proper full cycle before trusting either reading, both instruments converged to within 3-4% of each other.

The BMV is brilliant for trend tracking — I watch the ampere-hour consumption religiously. The BMS is my hard protection line. Neither reading alone tells the whole story, but together they've kept me honest through several power cuts now.

Great thread this — @SussexBoater your 16% discrepancy is actually worth digging into properly rather than just picking a winner.

One thing nobody's mentioned yet: when did you last do a full synchronisation cycle on the BMV? If it hasn't seen a genuine tail current end-of-charge recently, it'll drift significantly over time. LiFePO4 flat voltage curve makes this worse than with lead-acid.

Also worth checking your Peukert exponent setting in the BMV — factory default is often wrong for lithium. Should be closer to 1.05 rather than the lead-acid 1.25.

My money's usually on the BMS for real-time cell-level accuracy, but a well-calibrated BMV gives better trend data over time. Run them both, understand what each is actually measuring, and you'll get the full picture. Neither instrument is broken here — they're just answering slightly different questions.