Victron SmartShunt showing different SOC to my BMS — which one do I trust?

Had exactly this on my narrowboat last winter — drove me absolutely mad for weeks.

The short answer is: trust neither blindly, but lean on the SmartShunt for day-to-day decisions once it's properly calibrated.

Here's what was happening on mine. My BMS (a Daly unit on a 280Ah Fogstar cells build) was reporting SOC based on cell voltage, which sounds sensible but is actually quite crude — lithium voltage curves are notoriously flat through the middle range, so a small voltage wobble reads as a massive SOC swing.

The Victron SmartShunt does proper coulomb counting — it's literally measuring current in and out over time. Far more accurate in practice, provided you nail two things:

1. Charged voltage threshold — set this correctly so it knows when to reset to 100%
2. Tail current — the absorption tail setting tells it when the battery is genuinely full

Where people go wrong is never letting the bank reach a proper full charge, so the SmartShunt never gets a clean reset point and drifts over weeks.

My routine now: run a full charge to bulk + absorption completion at least once a week (easier said than done on cloudy stretches of the Shroppie, mind you), and the two readings stay within about 3% of each other most of the time.

What's your BMS brand? And are you on LiFePO4 or something else? Makes a big difference to which symptoms you're likely seeing.

@LisaKelly66 spot on about the calibration point — the SmartShunt is only as good as its starting reference.

One thing worth adding: the BMS SOC is often just voltage-based under the hood, which is pretty rough especially mid-charge. The SmartShunt does proper coulomb counting so it actually tracks what's gone in and out.

Had this exact headache setting up my shepherd's hut system with Fogstar Drift cells. The BMS was reading 80% while the SmartShunt said 64% — turned out the BMS just hadn't been properly synced at 100% since install.

Let it hit a proper full charge (absorb complete, current tail off), tick "synchronise to 100%" in the Victron Connect app, and they'll usually agree much better after that.

@LisaKelly66 @RobThompson classic situation — two devices staring at each other like two blokes in a pub both convinced they know the way to the motorway 😄

Genuinely though, I had this exact comedy duo situation on my static caravan setup with a Fogstar Drift battery. BMS said 87%, Smart

@LisaKelly66 @RobThompson genuinely curious — does the gap between the two readings drift over time or is it fairly static?

I'm setting up a Fogstar Drift 200Ah on my static caravan and already losing sleep over this exact scenario. My understanding is the SmartShunt does coulomb counting (fancy maths for "counting electrons in and out") whereas the BMS is typically doing voltage-based guesswork dressed up in a lab coat?

What I can't figure out is: if both are wrong in different ways, is there a sensible way to cross-reference them to get closer to reality? Or are we essentially triangulating nonsense?

Also — does the Peukert exponent setting in the SmartShunt VictronConnect app make a meaningful difference for LiFePO4, or is that more of a lead-acid concern?

@LutonAdventure great question — in my experience the drift is almost always cumulative rather than static. The SmartShunt's coulomb counting will creep away from reality if you've got any parasitic loads it doesn't see, or if the synchronisation threshold isn't set right for your battery chemistry. LiFePO4 in particular has a very flat discharge curve, which makes it harder for the shunt to know when it's genuinely full. I'd check your tail current and charged voltage settings first — those two alone sorted a 15% discrepancy I had on my setup. Once you nail a reliable sync point and eliminate any unmeasured loads, the gap should stay pretty tight between charges rather than wandering further off each cycle.