500A Victron Smart Shunt & Battery BMS not correct

Been having a similar head-scratcher with my shepherd's hut setup and wondering if anyone else has run into this.

Running a 12V 200Ah Fogstar Drift LiFePO4 with a Victron 500A SmartShunt, and the SOC readings between the two never quite agree. The BMS on the Fogstar is reporting around 87% whilst the SmartShunt is showing 94% — a 7% discrepancy that seems to shift around depending on charge/discharge cycles.

I've gone through the usual suspects:

• Peukert exponent — set to 1.05 as recommended for LiFePO4
• Charged voltage threshold — set to 13.8V
• Tail current — sitting at 4% (2% seemed too aggressive for my use case)
• Charge efficiency factor — tried 99% and 100%

The SmartShunt is the only thing on the negative bus apart from the battery itself, so I don't think current is sneaking past it anywhere.

My suspicion is that the two systems are using different reference points for what "full" actually means, and they drift apart from there. The BMS resets its SOC based on cell voltages internally, whereas the SmartShunt is purely coulomb counting and relies on that synchronisation event.

Has anyone managed to get these two to agree reliably, or is a small discrepancy just something you learn to live with? Also curious whether anyone's using Victron's VE.Direct connection to pull BMS data directly into Cerbo GX and whether that resolves it — I haven't gone down that route yet.

Would be good to know what others are seeing with their setups.

@SOC_Wizard classic Fogstar Drift behaviour — the BMS and the SmartShunt are basically two blokes arguing about who ate the last biscuit, neither one's wrong but they'll never agree unless you nail your charged voltage and tail current settings in the VictronConnect app.

Had this exact issue in my motorhome last summer. The root cause for me was the SmartShunt never getting a proper synchronisation point — it needs to see the battery hit 100% (fully charged, tail current dropping off) before its coulomb counting becomes reliable.

Check your charge settings in VictronConnect:

• Charged voltage — should be around 14.2V for LiFePO4
• Tail current — try 2-4% of capacity (so ~4-8A for 200Ah)
• Charged detection time — I use 3 minutes

Once the shunt gets a clean sync, it'll track much better. The Fogstar BMS SOC reading is calculated differently and they'll rarely agree perfectly — I just use the SmartShunt as my primary reference and ignore the BMS percentage after initial setup.

@TorFinn is right that it's a common Drift quirk, but it's usually fixable on the Victron side.

@SOC_Wizard this is almost certainly a charged voltage threshold misconfiguration on the SmartShunt. By default Victron sets it at 13.2V, but the Fogstar Drift's BMS considers itself full closer to 14.2–14.4V after absorption. The shunt never sees a valid sync point, so coulomb counting drift compounds over days.

In VictronConnect → SmartShunt settings, set:

• Charged voltage: 14.0V (just below your charger's absorption setpoint)
• Tail current: 4% (1% is too tight for LiFePO4)
• Charged detection time: 3 minutes

Also worth checking your negative busbar — if anything bypasses the shunt's negative terminal (even temporarily), the counting goes immediately out of sync. Had exactly this on my shepherd's hut build when I'd left a direct chassis earth from the inverter frame.

Once properly synced the Fogstar/Victron combo is genuinely solid.