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

Had this exact headache last year. Short answer — trust neither blindly, use both together.

The BMV-712 does coulomb counting, so it's only as accurate as its shunt calibration and how well you've set your charged voltage / tail current. If those are off, the drift compounds over time. Mine was reading 8% high after a few weeks until I tweaked the settings properly.

Your BMS is working off cell voltages, which sounds more "real" but voltage-based SOC is notoriously sloppy in the middle range — especially with LiFePO4 where the curve is basically flat from 20–90%.

What I do now:

• Use the BMV for day-to-day monitoring (trends, watts in/out, etc.)
• Use the BMS cell voltages to sanity check at the extremes (near full or near empty)
• Let a full charge cycle re-sync the BMV periodically

Worth double-checking your BMV's Peukert exponent and charge efficiency factor settings too — defaults aren't always right for lithium.

What BMS are you running? Some of the cheaper ones have dodgy voltage reporting anyway, which throws everything off. And are you on a Fogstar or similar prismatic setup?

Would be good to hear what others are doing — curious if anyone's found a slicker way to keep them in sync.

@DodgyDrifter that's useful context, cheers.

Quick question though — what's typically causing the drift between the two over time? Mine started pretty close but my BMV-712 is now showing maybe 8% higher SOC than the BMS on my Fogstar Drift 100Ah.

I've got a garden office running off this setup so accuracy actually matters when I'm deciding whether to fire up the kettle or conserve what's left.

Is the fix as simple as doing a full charge cycle to let the BMV recalibrate, or is there something I should be adjusting in the charged voltage / tail current settings? Currently have charged voltage set at 14.2V and tail current at 4% — not sure if those are sensible defaults for LiFePO4.

@DaiLewis69 biggest culprit in my experience is peukert effect not being set correctly on the BMV. Lithium is basically flat at 1.00–1.05, but if it's still on the lead-acid default it'll miscount under high loads.

Second thing — parasitic drain. My narrowboat has a few always-on items (bilge pump float switch, some LED indicators) that the BMV sees but the BMS doesn't always account for in its SOC model.

Also worth checking your charged voltage threshold and tail current settings in the BMV. If it's not resetting to 100% cleanly at the end of a charge cycle, the drift compounds over days. I had mine out by 12% after a week of overcast weather on the cut before I sorted that.

Fogstar Drift cells with a Daly BMS here — once those settings were dialled in, both readings sit within 2–3% pretty consistently.

@BoatEwan is right on Peukert, but the other silent killer is synchronisation events — if your BMV is set to sync at 99% voltage trigger and your BMS is calling 100% SOC at a different voltage, they'll diverge every single charge cycle like two drunk blokes arguing over a map.

On my static van setup with Fogstar Drift cells, I set the BMV charged voltage to exactly match what the BMS considers full — took me an afternoon of muttering but sorted the 12% drift I was seeing overnight.

Also check your tail current setting. Too high and the BMV syncs prematurely, too low and it never syncs at all. Victron's defaults assume you're not me, apparently.

@Rob1963 raises something I learned the hard way charging my shepherd's hut setup. The sync threshold was catching me out constantly — BMV kept resetting SOC at what it thought was full, but the BMS was still showing cells a bit unbalanced at top-end.

Fixed it by being more conservative with my charged voltage detection and tail current settings. Once those aligned better with what the BMS was actually seeing at genuine full charge, the two readings converged nicely and stayed within about 2-3% of each other.

Worth logging both over a few full cycles with VictronConnect — you can usually spot exactly when they diverge and work backwards from there. Mine always drifted most during my EV charging sessions when loads were unpredictable and the shunt was seeing some fairly chaotic current profiles.