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

Having this exact same debate with myself right now on the narrowboat! 🙈

My Fogstar Drift 200Ah LiFePO4 has its own BMS reporting one SOC and my Victron SmartShunt is consistently showing about 8-12% lower. It's doing my head in trying to figure out which one to actually trust for day-to-day decisions.

From what I've read so far, the SmartShunt is doing coulomb counting — so it's tracking actual current in and out — whereas the BMS figure is often more of a voltage-based guesstimate, especially on cheaper units. But then the SmartShunt can suffer from drift over time if it's not getting a proper full charge to reset the 100% sync point.

A few things I'm wondering:

• Does your SmartShunt have the charged voltage and tail current set correctly in VictronConnect? Mine took a few tweaks to get right
• How old is the discrepancy — did it appear gradually or was it there from day one?
• Is your BMS a "proper" comms-enabled one like a JK or Daly with actual Bluetooth monitoring, or just the basic protection board type?

I ended up trusting the SmartShunt more once I dialled in the settings properly, but I still feel like I'm flying a bit blind when the two disagree significantly.

Would love to hear from anyone who's integrated their BMS data directly into Victron's ecosystem via DVCC — apparently that can help the two "agree" with each other? Has anyone on here actually got that working properly on a 12V boat setup?

@LindaClark90 I had exactly this on my static caravan setup with a Fogstar battery and SmartShunt.

Key question — has your SmartShunt been properly calibrated? Specifically:

• Is the charged voltage threshold set correctly for LiFePO4 (around 13.6V rather than the default lead-acid setting)?
• Have you set the tail current correctly — LiFePO4 charges differently to lead-acid and the defaults are usually wrong out of the box
• Has it had a proper full synchronisation cycle yet?

The BMS SOC reading is often just a voltage-based estimate, which is notoriously unreliable in the flat middle section of a LiFePO4 discharge curve.

Once I got the SmartShunt parameters dialled in via the Victron Connect app, the two readings came within 2-3% of each other. The SmartShunt coulomb counting is generally more reliable once calibrated properly.

What are your current tail current and charged voltage settings?

@LindaClark90 classic SmartShunt calibration gremlin! The SmartShunt is doing coulomb counting — it's meticulously tracking every electron in and out, but it drifts unless it gets a proper synchronisation event (i.e., hitting 100% at a genuine full charge).

The critical question: is your SmartShunt actually seeing a full absorption cycle complete before it resets to 100%? On a narrowboat with variable solar and hotel loads, it often thinks it's full before it truly is.

Check in the VictronConnect app:

• Charged voltage threshold — should be ~3.45V per cell (13.8V for 12V system)
• Tail current setting — typically 4% of capacity for LiFePO4

My cabin setup was 15% out until I tightened those two parameters. After that, SmartShunt and BMS agreed within 2%. The SmartShunt can be more accurate than the BMS — but only if calibrated properly.

Right, so I lived this exact soap opera on my Ducato-based motorhome for about three weeks before I cracked it.

The villain in my story was the charged voltage threshold in the SmartShunt settings. Mine was set 0.2V too high, so it never registered a proper "full" sync point. The SmartShunt was essentially a lost tourist — counting away diligently but with no idea where it started from.

Once I matched the charged voltage threshold to what my Fogstar BMS actually considers 100% (check the Fogstar docs — it's not always what you'd assume), the two readings snapped together within a couple of percent within a few cycles.

@RetiredEngineer61's question about calibration is the right thread to pull. @DailySolar's coulomb counting explanation is spot on — but coulomb counting only stays accurate if it gets a reliable anchor point regularly.

Really interesting thread, been following this as I've got a similar situation brewing with my own setup.

One thing nobody's mentioned yet — what's your Peukert exponent set to in the SmartShunt? LiFePO4 behaves very differently to lead-acid here, and if it's still on the default setting it could be introducing cumulative errors over each cycle.

Also worth checking your charged voltage threshold and tail current settings. If the SmartShunt isn't recognising a full charge correctly, the drift between the two readings will just keep getting worse over time.

@LindaClark90 narrowboat use is particularly tricky because your charging profile is probably quite irregular — engine alternator, shore power, solar all potentially feeding in at different times which makes accurate coulomb counting harder.

Has anyone actually verified which is closer to reality by doing a proper capacity test?

Great thread, this catches so many people out! One thing worth adding that nobody's mentioned yet — the charged voltage and tail current settings in your SmartShunt are absolutely critical for LiFePO4.

If those aren't dialled in correctly, the SmartShunt never properly triggers a "full" synchronisation point, so the coulomb counting drift just compounds over time and you end up with that 8-12% discrepancy @LindaClark90 is seeing.

For a Fogstar Drift specifically, I'd suggest setting charged voltage around 14.2V with a tail current of roughly 2-4% of your battery capacity — so about 4-8A on your 200Ah. Once it gets a clean sync at full charge regularly, you should see both readings converge nicely within a cycle or two.

Also worth checking your shunt is the only path between negative terminal and load — even a small bypass will throw the counting off entirely.