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

Had exactly this circus on my narrowboat last spring. The short answer is: trust neither blindly, but lean on the SmartShunt for day-to-day navigation — provided you've set it up properly.

The BMS is watching cell voltages and protecting the pack from catastrophe. It's not really designed to be an accurate fuel gauge — that's not its job. Most budget BMS units (looking at you, every anonymous Chinese unit that ships with Fogstar DIY cells) calculate SOC from voltage curves alone, which is notoriously wobbly mid-charge.

The SmartShunt does Coulomb counting — it's actually measuring electrons in and out. The catch is it's only as good as your configuration:

• Charged voltage set correctly?
• Tail current dialled in?
• Peukert exponent appropriate for lithium (hint: closer to 1.05, not the lead-acid default)?

If those are wrong, your SmartShunt will drift like a canal boat with a fouled prop.

What I'd actually suggest is resetting the SmartShunt SOC at each full charge event and letting it count down from there. Over a week of typical use, mine agrees with the BMS to within 3-4%, which I consider acceptable for something floating around on the Bridgwater & Taunton.

What battery chemistry are you running, and which BMS? That'll change the advice considerably.

PeteGreen84 | 47 posts

Great shout @ExmoorNomad. One thing I'd add — the SmartShunt is only as good as your charged voltage and tail current settings. I had mine wildly out for months because I'd left the tail current at the default 4% rather than tuning it to my actual battery chemistry (LiFePO4 in my case).

Once I dialled that in and made sure the shunt was seeing every load and charge source, it became rock solid. The BMS SOC reading is often just a rough voltage-based estimate anyway, especially mid-cycle when the curve is so flat on lithium.

Worth connecting the SmartShunt to the Victron app and watching a full charge cycle — you'll quickly spot where the discrepancy creeps in.

Yeah @PeteGreen84 is onto something there. I had massive drift on my shepherd's hut setup until I sorted the tail current settings properly in the Victron app — default values are rubbish for LiFePO4.

BMS SOC is often just voltage-based under the hood, which is basically useless mid-cycle with lithium. SmartShunt doing proper coulomb counting will always be more accurate day-to-day.

My Fogstar Drift cells + SmartShunt combo has been rock solid once I set:

• Charged voltage to match actual absorption endpoint
• Tail current around 1-2% of capacity
• Peukert to 1.05 for lithium

The BMS figure is still worth a glance — if the two readings are wildly diverging and staying apart, that's a sign something's actually wrong rather than just calibration drift.

Yeah but there's another angle nobody's mentioned — peukert effect and temperature compensation. If your SmartShunt's peukert exponent isn't dialled in for your specific cells, the SOC reading drifts badly under high discharge loads. Running my van's compressor fridge and charging the EV off the same bank really exposed this.

My Fogstar Drift LiFePO4 cells behave differently to what Victron assumes out of the box. Took me three weeks of logging in VRM before I stopped second-guessing the readings.

BMS SOC algorithms are also wildly inconsistent between brands — some just count coulombs, some do voltage curves, some do both badly.

Point is: even with the SmartShunt set up "properly" as @ExmoorNomad suggests, properly is doing a lot of heavy lifting there. What exact settings did you use?

My Fogstar Drift cells laugh in the face of Peukert at these UK temperatures, @LuckySkipper — the real villain nobody's naming is parasitic loads sneaking past the shunt because someone wired the BMS directly to the chassis without routing through it properly. 🙃

Been there, done that, watched my SmartShunt confidently report 87% SOC while the lights were literally dimming.

Sort your wiring topology first, then argue about which number to believe.

@AnneButler fair point on the Fogstar cells, but I'd add one thing nobody's nailed yet — sync events.

Your SmartShunt only knows its SOC is correct when it hits a proper full charge and the tail current drops to whatever threshold you've set. If you're never properly topping out (common off-grid in winter with limited solar), the shunt drifts over weeks and starts lying to you.

I had this exact problem last January. SmartShunt said 72%, BMS said 58% — turns out I hadn't hit a genuine full charge in nearly 3 weeks.

Quick checklist:

• Set tail current to ~1-2% of capacity
• Check your "charged voltage" threshold is realistic
• Force a full charge periodically

Once I sorted mine, both readings converged within 3-4%.