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

Had this exact headache last year. Short answer — trust your BMS for protection, trust the SmartShunt for accuracy.

Here's my thinking: the BMS is watching cell voltages and will cut off if anything goes dangerously low or high, so it's your safety net regardless. But SOC percentage is where the SmartShunt earns its keep. It's doing proper coulomb counting, tracking every amp in and out, whereas most BMS units (especially cheaper ones) are just inferring SOC from resting voltage — which is notoriously unreliable mid-charge or under load.

That said, the SmartShunt needs to be properly calibrated or it'll drift over time. A few things worth checking:

• Have you set the correct battery capacity in VictronConnect?
• Is your shunt on the negative of the entire system, nothing bypassing it?
• Have you let it hit a proper full charge (absorption complete) so it can sync to 100%?

My Fogstar Drift 200Ah LiFePO4 and SmartShunt were about 12% apart when I first set up. Turned out I had a small load wired before the shunt. Sorted that and they've been within 2-3% ever since.

What BMS are you running? Some of the better ones (Daly Smart, JK) actually export SOC via Bluetooth and can be compared more meaningfully. Would be good to know the full setup before diagnosing further.

Anyone else had persistent drift even with everything wired correctly? Curious if it's a known issue with certain battery chemistries.

Midge93 | 47 posts

Good breakdown @NZ_Marine. Worth adding that the SmartShunt is only as good as its setup though — if you haven't dialled in the correct battery capacity, Peukert exponent, and charge efficiency factor, it'll drift badly over time.

Mine was reading 15% out until I realised I'd left the capacity at the default setting like a muppet.

Also make sure you're getting a proper full synchronisation cycle regularly — the SmartShunt resets its SOC to 100% when it detects a full charge, so if your charge tail current threshold is set wrong it might never sync properly.

Victron's own documentation is decent for walking through the calibration. What BMS are you running? Some of them do actually communicate SOC data reasonably well, particularly anything using CANBUS integration with Victron kit.

DerekDixon | 203 posts

@Midge93 nails it on setup — spent a whole weekend on the narrowboat last autumn getting mine dialled in properly. The two settings that made the biggest difference were charged voltage and tail current. Had mine set too high initially, so the SmartShunt never registered a full charge and drifted badly over a week of cruising.

Once I matched the charged voltage to what my Victron MPPT actually terminates at (rather than the battery's stated absorption voltage), everything synced up beautifully. The BMS and SmartShunt now rarely disagree by more than 2-3%.

The SmartShunt also needs a proper synchronisation point — if it never sees a genuine 100% charge, the coulomb counting errors compound over time. Worth forcing a full absorption cycle occasionally just to let it reset.

EssexNomad | 312 posts

Mine disagreed by 23% once — turned out the SmartShunt's charged voltage threshold was set for a lead-acid profile someone had copy-pasted from a YouTube tutorial filmed in a garage in Ohio. 🙄

FormerMariner1 | 89 posts

Following on from @EssexNomad's point about thresholds — can someone clarify the exact interaction between the "charged voltage" setting and the "tail current" setting in the SmartShunt? My understanding is that both conditions must be met simultaneously before it resets to 100%, but the VictronConnect documentation is genuinely ambiguous on this.

Specifically for a 12V LiFePO4 (Fogstar Drift 100Ah in my van conversion) — what voltage are people actually using? I've seen recommendations ranging from 13.2V to 14.0V, which is an enormous spread. Presumably setting it too low means the shunt resets prematurely and accumulates error over time?

Also — does the SmartShunt's Peukert exponent matter much for LiFePO4, given its relatively flat discharge curve compared to lead-acid? Or should that effectively be left at 1.00?