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

Been wrestling with exactly this on my own setup. My Fogstar Drift 100Ah shows maybe 8-10% higher SOC on the built-in BMS app compared to what the BMV-712 is reading. Bit unsettling when you're trying to plan how much solar you've actually got in reserve.

From what I've gathered, the BMV relies on coulomb counting from a shunt — so it's tracking actual current flow in and out. Accuracy lives and dies by whether you've got the peukert exponent and charge efficiency factor dialled in correctly. Have people found the Victron defaults reasonable for LiFePO4, or do they need tweaking?

The BMS SOC, on the other hand, seems to use voltage curves primarily — which can be a bit wobbly in the middle of the charge range where LFP is notoriously flat.

My suspicion is the BMV is closer to reality if it's been properly synchronised after a full charge cycle. But I'm not 100% confident mine is set up optimally — I just followed the basic Victron Connect wizard.

Few questions for the room:

• Do you trust your shunt-based monitor or BMS more, and why?
• What tail current settings are you using to trigger a full sync on LFP?
• Anyone running both and found a way to get them agreeing consistently?

Would love to know if this is just a calibration issue or something more fundamental about how each system calculates SOC.

@SparkyHiker had exactly this gap on my setup last winter — Fogstar BMS was consistently optimistic compared to my BMV-712.

Short answer: trust the BMV-712, but only if you've properly configured it.

Key things to check:

• Charged voltage threshold set correctly (I use 27.4V on my 24V system)
• Peukert exponent — LiFePO4 should be around 1.05
• Tail current percentage dialled in

The BMS SOC is largely a voltage-estimation shortcut. It's notoriously unreliable mid-cycle because LiFePO4's discharge curve is so flat.

The BMV does proper coulomb counting, so it tracks actual amp-hours in/out. Far more accurate once you've let it sync through a few full charge cycles.

Mine took about a week of regular use before it settled into readings I trusted.

@SparkyHiker this took me a while to get my head around when I first set up my cabin system.

The BMS is essentially counting charge cycles internally, while your BMV-712 is measuring actual current flow at the shunt. The BMV wins that argument every time, provided you've got it configured properly — correct battery capacity, charge efficiency factor, and crucially a reliable synchronisation point so it knows when to reset to 100%.

The common culprit I've seen is the tail current setting. If the BMV isn't seeing a clean absorption-to-float transition, it never syncs and the drift compounds over days. Worth checking what you've set that at.

Trust the BMV for day-to-day decisions. The BMS is really there for protection, not precision monitoring.

Has anyone actually done a proper capacity test to verify which one is closer to reality?

I've got a similar discrepancy on my static caravan setup — not Fogstar but same kind of gap between the BMV-712 and the BMS app. What I'm wondering is whether it's partly down to the Peukert exponent setting on the BMV. Mine was still on the default lithium setting when I first installed it and I'm not convinced it was right for my specific cells.

Also — what's your charged voltage threshold set to on the BMV? If it's even slightly off, the 100% sync point drifts over time and the gap just compounds.

@SparkyHiker what tail current percentage are you using for the sync trigger?

@SmartSolarNerd capacity test is the gold standard but there's a quicker sanity check first — verify your BMV-712's shunt is actually seeing all your load and charge current. In my van build I had a secondary 12V circuit bypassing the shunt entirely which made the BMV read pessimistically. Also worth checking: has the BMV ever hit a proper synchronisation point? It only resets to 100% SOC when charge current drops below your configured tail current threshold and absorption time is met. If those parameters are set wrong it'll drift badly over time. The Fogstar BMS meanwhile is doing coulomb counting internally but starting from its own calibration assumptions about usable capacity. Neither is necessarily "right" — they're just counting from different baselines with different error accumulation.

@ExBrickie31 good shout on the shunt wiring — I had this exact problem with my garden office setup and it turned out I'd missed a small load that was bypassing the shunt entirely (a tiny always-on 12V socket I'd completely forgotten about). Took me ages to work out why my BMV kept drifting lower.

One thing nobody's mentioned yet: have you checked your charged voltage and tail current settings in the BMV? These determine when it calls 100% SOC. If those are off, it'll never sync properly and the drift compounds over time.

My Victron dealer told me most Fogstar cells need the tail current setting lower than the BMV default — worth digging into the Fogstar specs to double-check. What are your current BMV settings, @SparkyHiker?