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

Mine showed a 12% difference for weeks until I realised my Peukert exponent was set for lead-acid on a lithium bank — classic.

The short answer is: trust neither blindly, use both as a conversation.

Here's my rough take after years of faffing with Fogstar Drift cells and a Daly BMS:

• BMS SOC is often just voltage-based maths dressed up in a percentage costume — looks confident, frequently lying
• BMV-712 is coulomb counting, so it's only as good as its charge efficiency factor and whether you've let it sync properly at 100%

The BMV wins long-term if you've calibrated it properly and you're regularly hitting a full absorb/float cycle so it can re-anchor itself. If you're never fully charging (common with solar in a British winter, obviously 😅), it drifts like a unloved narrowboat.

My workflow:

1. Set your charged voltage threshold and tail current correctly in the BMV
2. Let it sync at 100% whenever you hit a proper full charge
3. Use the BMS figures mainly as a sanity check and for cell-level balance info

What BMS are you running? Some are genuinely dreadful at SOC — anything to add, folks? Curious whether anyone's had better luck with the JK BMS figures specifically.

Great point @BrookLover, and worth adding that the BMV-712 is only ever as good as its synchronisation events. If your charged detection voltage or tail current thresholds aren't dialled in properly for lithium, it'll drift over time and never fully reset its reference point. I had mine showing 94% when the BMS was screaming at me that cells were balancing at 100% — turned out my "charged voltage" was set 0.2V too low so it never triggered a sync.

Also worth checking your shunt is the only path current can take — any bypass, however small, will compound the inaccuracy over days and weeks. The two units together tell a much richer story than either alone.

Really useful thread this. Quick question for the group — on my narrowboat setup I've got a Fogstar Drift 200Ah with its own BMS, and the BMV-712 sitting alongside it.

What's everyone using as their sync trigger? I've currently got mine set to sync at 98% charged with a tail current of 4%, but I wonder if that's too generous and it's never actually hitting a proper full sync in real-world conditions?

Also — does the BMS on a Fogstar communicate anything useful externally, or is it essentially a black box? Feels like I'm flying half-blind compared to some of the fancier setups I've seen with Victron Smart lithiums where everything talks to each other via VE.Direct.

Is there a point where you'd just ditch the BMV entirely and rely solely on the BMS, or is that asking for trouble?

@HollyGaz the Fogstar Drift's built-in BMS communicates SOC based on cell voltage curves, which is reasonable but can drift over time without a proper full-charge reset. The BMV-712 meanwhile is coulomb-counting from whatever its last sync point was.

On my narrowboat I run both and cross-reference constantly. The key thing I've found: make sure your BMV syncs at a genuine 100% SOC — i.e., absorption complete, tail current below your set threshold. If you're running solar and rarely hitting a proper full charge (common in winter), both readings will wander independently.

Set your BMV charged voltage to match what the Drift's BMS considers "full," and set a realistic tail current (I use 1% of capacity, so 2A on a 200Ah bank). Once they're singing from the same hymn sheet, the differences should collapse to 2-3%.

@HollyGaz on the narrowboat — worth knowing that coulomb counting (what the BMV does) drifts over time without clean sync points. If you're not hitting 100% regularly or your shunt install is dodgy, it'll wander.

My static van setup I've got a Victron BMV-712 alongside a Fogstar 200Ah and honestly the two readings rarely agree perfectly. I just use the BMV as my day-to-day gauge and let the BMS handle protection cutoffs. Different jobs really.

Main thing — check your:

• Peukert exponent (set to 1.05 for lithium, not the lead-acid default)
• Charged voltage threshold
• Tail current setting

Get those wrong and your BMV will confidently lie to you all day. Took me an embarrassing amount of time to sort mine.

@HollyGaz in my shepherd's hut I had the same head-scratching moment early on. What sorted it for me was making sure the BMV gets a proper full sync regularly — mine resets cleanly when the Fogstar hits absorption and holds there. If you're running the boat loads constantly it might never get that clean top-up to sync against.

Also double-check your charged voltage threshold in the BMV settings — needs to match what your charger actually delivers, not just a ballpark figure. A few mV out and it won't register as "full" and sync properly.

Once I tightened those two things up, both units tracked each other within 2-3% which feels realistic given the two different measurement methods.