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

Had this exact headache with my motorhome build last year. Short answer: trust neither blindly, use both.

The SmartShunt is only as good as its calibration. If your charged voltage, Peukert exponent and tail current aren't dialled in properly, it'll drift. Mine was reading 20% out after a week of cycling before I sorted the settings.

Your BMS is reading cell voltages directly, so it knows when you're at the top or bottom — but SOC% from a BMS is often just a rough voltage-based estimate in the middle range, which is famously flat on LiFePO4. Basically useless between 20–90%.

What I do:

• Use the SmartShunt for day-to-day SOC in the middle range (coulomb counting is genuinely better there)
• Let the BMS act as the safety net — if it's screaming, something's actually wrong
• Sync the shunt at 100% every time I hit a proper full charge (absorption done, tail current hit)

The drift issue usually comes down to parasitic loads the shunt doesn't know about, or an incorrect battery capacity figure entered in the app. What capacity have you set it to, and is it the actual usable capacity or the nameplate figure?

What BMS are you running? Some talk to Victron via DVCC over Bluetooth or VE.Can and you can get them working together properly rather than fighting each other.

LindaCross66 | 847 posts | Gloucestershire

@DownsCruiser makes a solid point about calibration. One thing worth adding — the SmartShunt's SOC will drift over time if it never gets a proper synchronisation event. Make sure your "charged voltage" threshold is set correctly for your specific cells, not just a generic LiFePO4 figure. I had mine set 0.2V too low and it was syncing prematurely, so the shunt consistently read higher than reality.

Your BMS meanwhile is typically protecting against cell-level voltages, which is actually more granular information. If your BMS is cutting out before your shunt reckons you're empty, that's a red flag worth investigating — could indicate cell imbalance rather than either device being wrong.

Running both through Victron Connect or Venus OS gives you a decent cross-reference picture. That's where I'd start troubleshooting.

The SmartShunt wins on coulomb counting accuracy mid-cycle, but the BMS wins on knowing when to pull the plug — treat them like two different instruments: one's your fuel gauge, one's your engine warning light. 🚨

Had my Victron reading 47% while my BMS was screaming blue murder at low cell voltage on a cold Scottish morning — turned out one cell was sagging badly under load, something the SmartShunt couldn't possibly know.

Key tip: set your SmartShunt's tail current and charged voltage parameters precisely to your battery spec (Fogstar Drift cells have this documented), otherwise SOC drift compounds every cycle until the readings are basically fiction.

Good thread. Had this play out in my own setup — 280Ah Fogstar Drift LiFePO4 with a Daly BMS and a SmartShunt running in parallel. After a long overcast stretch, the two were reading nearly 12% apart.

The thing nobody mentions: the SmartShunt drifts if it never sees a proper full charge to re-synchronise. If you're doing partial cycles — common when EV charging is hammering your loads — that drift compounds over weeks. Mine was consistently pessimistic because the synchronisation voltage threshold wasn't being hit.

My fix: set the SmartShunt's "charged voltage" about 50mV below your actual absorption voltage so it resets more reliably. Then use the BMS reading as your low-end protection boundary, and the SmartShunt for everything in between.

Two instruments, two jobs. Neither replaces the other.

@Drift_Geek interesting you mention the Daly — that's exactly where I'd start questioning the BMS SOC figure. Daly's coulomb counting implementation is pretty rudimentary compared to what the SmartShunt is doing.

One thing nobody's touched on: synchronisation events matter enormously. The SmartShunt needs to hit its "charged" threshold cleanly to reset to 100%. If your absorption stage is cutting short, it'll drift pessimistically over time.

For my shepherd's hut setup I log both via Victron Cerbo and check for divergence weekly. When they're >5% apart, that's my cue to investigate rather than just trust one over the other.

Treat the BMS figure as a protection boundary, SmartShunt as your operational gauge. Different jobs.