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

Had exactly this head-scratcher at my cabin setup last winter. Short answer: neither blindly, but favour your BMS for protection decisions and your BMV-712 for day-to-day monitoring.

Here's why they diverge:

The BMV-712 calculates SOC via coulomb counting — it tracks every amp in and out from a known reference point (typically a full charge). Errors accumulate over time, especially if your Peukert exponent or charge efficiency factor isn't dialled in correctly for your specific cells.

Your BMS, meanwhile, is likely doing a voltage-based estimation, which is notoriously unreliable mid-discharge on LiFePO4 given how flat the discharge curve is. Some more sophisticated BMS units (Daly being a common culprit for vagueness) do their own coulomb counting too, but calibration and synchronisation points vary wildly.

What I'd check:

• Is your BMV-712 set to sync at 100% SOC at the correct voltage/tail current for your battery? Fogstar Drift cells, for example, need fairly specific charge parameters
• Have you set the Peukert exponent to 1.05 for LiFePO4? Default assumes lead-acid
• What's your charge efficiency factor set to — should be around 99% for lithium

The two readings will naturally drift apart if the BMV isn't synchronising properly on full charges. Over time mine were 12% apart until I sorted the sync settings in VictronConnect.

What BMS are you running and what's your actual cell chemistry? That'll help narrow it down considerably. Others on here have had good results cross-referencing with a clamp meter during charge cycles to see which is tracking more accurately.

@TransitAdventure had the same drama with my static last summer. Fogstar Drift 200Ah was showing 87% on the BMS app, BMV-712 saying 94%. Drove me mad for weeks.

Turned out my shunt wasn't getting all the current — had a earth wire bypassing it. Classic install mistake. Once I sorted that, they tracked pretty closely.

Worth checking:

• Shunt wired correctly (literally every wire must pass through it)
• BMV tail current and charged voltage settings dialled in properly
• BMS SOC reset after a full charge

The BMV's coulomb counting drifts if it never gets a proper sync point. Make sure it's actually seeing a genuine 100% periodically or the error just compounds over time.

Neither reading is gospel until your install is solid tbh.

Good shout from @TransitAdventure on trusting the BMS for protection decisions.

One thing worth adding — the BMV-712 is only as good as your calibration. If you haven't nailed down your Peukert exponent and charged voltage settings properly, it'll drift all over the shop. Mine was reading 15% optimistic until I actually sat and tweaked those figures properly.

Also worth checking your tail current setting. If it's too high, the BMV thinks the battery's full before it actually is, then the whole SOC calculation unravels from there.

I've got a Fogstar 200Ah in my static and once I got those settings dialled in, the two readings sat within 3-4% of each other consistently. That's probably the best you'll realistically achieve between coulomb counting and a BMS doing its own thing.

The bit nobody mentions — BMV-712 drift over time is real. Mine was reading 15% higher than actual after about 8 months without a proper sync. Key thing is making sure your charged voltage and tail current settings are dialled in correctly, otherwise it never gets a clean 100% reference point to reset from.

On the motorhome I've got:

• Charged voltage set to 14.2V (LiFePO4)
• Tail current at 4%

Took me ages to figure out why it kept drifting. Was a dodgy setting from the start.

@ZFS_OffGrid what BMS are you running with that Fogstar? Some of the cheaper ones report SOC based on voltage alone which is basically useless, especially mid-charge. Coulomb counting like the BMV does is generally more reliable for monitoring, just needs that regular calibration sync to stay honest.

@OffGridMax is spot on about drift, but the fix is simpler than most people realise — synchronisation points.

The BMV-712 resets its SOC calculation when it detects a full charge (tail current drops below your configured threshold). If your system never truly reaches absorption/float properly, it never syncs, and the coulomb counting error compounds indefinitely.

Check your BMV settings: Charged voltage, Tail current, and Charged detection time. Most people leave these at defaults that don't match their actual bank.

My 400Ah Fogstar setup was reading 12% high until I tightened the tail current from 4% to 2% — sorted it immediately.

The BMS SOC meanwhile is often just voltage-based lookups, which are notoriously inaccurate mid-range on LiFePO4's flat discharge curve. Neither is perfect; used together correctly, they're actually quite solid.

Ran my BMV-712 alongside a Daly BMS on the narrowboat for two seasons — the BMV wins on granularity but the Daly has saved my cells twice when the Victron was still cheerfully reporting 23%.