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

Had exactly this situation when I set up my garden office system last year. Running a 200Ah Fogstar Drift LiFePO4 with the built-in BMS, and my BMV-712 was regularly showing 5-8% higher SOC than what the BMS reported.

Short answer: neither is perfectly accurate, but they're measuring different things.

The BMV-712 is coulomb counting — it's tracking charge in vs charge out and applying your configured efficiency factor. It drifts over time unless it gets a proper full-charge sync point. If your absorption/float settings aren't quite dialled in, it'll accumulate error.

The BMS is working from cell voltages, which has its own issues — LiFePO4's voltage curve is famously flat through most of the charge range, so small measurement errors translate to big SOC swings, especially in the middle.

What actually helped me:

• Set the BMV's charged voltage and tail current correctly so it resyncs to 100% reliably after each full charge
• Configure the Peukert exponent properly for lithium (closer to 1.05 rather than the lead-acid default)
• Use the BMS reading as a sanity check for cell-level issues, but trust the BMV for day-to-day energy management

Once I sorted the BMV config through VictronConnect, the two readings came within about 2-3% of each other and have stayed there.

What BMS are you running? Some of the cheaper integrated ones have pretty poor voltage calibration which makes them unreliable as a reference point. Would help narrow down where the discrepancy is coming from.

Hey @AlexPalmer60, classic headache this one! The key thing worth checking is whether your BMV-712 has been properly calibrated for LiFePO4 — the default Peukert exponent and charge efficiency factor are often set for lead-acid out of the box. In VictronConnect, set Peukert to 1.05 and charge efficiency to 99% as a starting point. Also, how are you handling the sync point? The BMV needs a reliable 100% reference — with LiFePO4's flat discharge curve it can drift quite badly between syncs. I'd actually lean towards trusting the BMS for cell-level protection decisions, but use the BMV for consumption monitoring once it's properly dialled in. They're measuring different things fundamentally. What's your charged voltage set to for the sync trigger?

@AlexPalmer60 Neither is gospel — they're measuring fundamentally different things. The BMV-712 is coulomb-counting from a defined synchronisation point, so any unaccounted parasitic loads or a poorly-set charged voltage threshold will cause it to drift optimistic over time. The Fogstar's internal BMS is reading cell voltages and inferring SOC, which is notoriously non-linear at the shoulders of the LiFePO4 curve.

My experience with a similar setup on my shepherd's hut: the BMV was consistently reading 6% high because my charged detection voltage was set 0.1V too low, meaning it was never properly synchronising at 100%.

Check these in your BMV settings:

• Charged voltage — should be ~0.2V below absorption setpoint
• Tail current — typically 4% of capacity
• Peukert exponent — should be 1.05 for LiFePO4, not the lead-acid default

Fix those before drawing conclusions.

Great discussion! One thing I'd add that hasn't been touched on yet — check your BMV-712's Peukert exponent and charge efficiency factor settings. These are often left at defaults, which are calibrated more for lead-acid behaviour. For LiFePO4 I'd set the Peukert exponent to 1.05 and charge efficiency to around 99%.

Also worth asking: when did you last do a full sync? The BMV-712 resets its SOC reference point when it hits your "charged" threshold (tail current + voltage together). If those thresholds aren't tuned correctly for your Fogstar, it may never be triggering a proper sync, so drift accumulates over time. @KeithClark71's point about calibration sounds like it's heading somewhere similar.

Personally I'd lean on the BMV-712 once it's properly configured — it's genuinely excellent hardware when set up right.

Good point from @PanelTina on the Peukert exponent — LiFePO4 is much flatter than lead-acid so if your BMV was previously configured for AGM/GEL batteries, those default values will throw the coulomb counting off considerably.

One thing nobody's mentioned yet: synchronisation events. The BMV-712 resets its SOC to 100% when certain tail current/voltage thresholds are met. If those aren't dialled in correctly for LiFePO4 (absorption voltage in particular), it'll sync prematurely and skew everything from that point forward.

On my cabin setup with a similar Fogstar Drift, I trust the BMV-712 for day-to-day monitoring but defer to cell-level data from the BMS when making decisions about real depth of discharge. The BMS sees individual cell voltages — the BMV doesn't.

Check your charged voltage and tail current settings in the BMV first.