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

Had this exact headache last summer when I was commissioning the setup in my motorhome. Running a 200Ah Fogstar Drift LiFePO4 with the built-in BMS, and the SmartShunt was telling me something completely different — sometimes 8-10% apart.

Here's what I landed on after a fair bit of head-scratching:

Trust the SmartShunt for day-to-day monitoring, but let the BMS be your safety net.

The BMS is measuring cell voltages and protecting against over/under discharge. It's not really designed to be an accurate fuel gauge — it's a guardian. The SmartShunt, when properly configured, is doing proper coulomb counting and is far better at tracking actual capacity over time.

The key word there is properly configured. My readings were all over the place until I:

• Set the charged voltage correctly (I had it too low)
• Dialled in the tail current percentage
• Let it complete a proper full charge cycle to synchronise

After that, the two figures got much closer — within 2-3% most of the time.

One thing worth noting: LiFePO4 has that notoriously flat voltage curve, which makes voltage-based SOC estimates (what most BMS units use) pretty unreliable in the middle range anyway.

Has anyone else found their BMS brand makes a difference here? I've heard some of the JK BMS units are much better at SOC accuracy than the cheaper options. Curious whether those running Daly or Overkill are seeing the same discrepancy. And does anyone bother syncing them, or just pick one and stick with it?

@SmartSolarGuy classic calibration drift issue this one.

The SmartShunt is only as good as its synchronisation events — it needs a genuine full charge (absorption complete, tail current dropped below your configured threshold) to reset its 100% reference point. If that's not happening reliably, the SOC just drifts further off over time.

On my narrowboat I trust the SmartShunt for day-to-day coulomb counting but treat the BMS voltage-based SOC as a sanity check at the extremes. The BMS knows cell-level voltages; the shunt doesn't.

Key things to check:

• Is your charged voltage setting in the shunt config actually matching absorption setpoint?
• Tail current — LiFePO4 has a very flat charge curve, set this low (~2-4% of capacity)
• Peukert exponent should be 1.05 for LiFePO4, not the lead-acid default

Sort those three and they'll likely agree within 3-5%.