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

Had exactly this issue with my static caravan setup last summer. The Fogstar Drift 100Ah I'm running has its own BMS reporting one thing via Bluetooth, and my Victron SmartShunt was telling a completely different story — sometimes 8-10% apart.

From what I've gathered, the SmartShunt is doing coulomb counting (tracking actual charge in/out), whereas most BMS units are estimating SOC primarily from cell voltage, which can be pretty unreliable in the middle of the charge range where LiFePO4 is notoriously flat.

A few things I found made a big difference to SmartShunt accuracy:

• Making sure the charged voltage threshold and tail current settings were dialled in properly — mine were still on defaults which weren't suited to my actual charge profile
• Ensuring the shunt is on the negative of the entire system so nothing bypasses it
• Letting the battery do a full charge cycle to let the SmartShunt re-sync

That said, the SmartShunt can drift over time if it never hits a proper full charge to reset itself. So if your system rarely reaches 100%, the BMS voltage-based reading might actually be more honest in some situations.

Curious what others think — do you trust one over the other, or do you cross-reference both? And does anyone using Victron's DVCC with a compatible BMS find the two eventually agree once properly configured?

@BoxerAdventure had the same head-scratching moment with my backup setup.

Quick question — have you properly configured the SmartShunt's Peukert exponent and charge efficiency factor in VictronConnect? Out of the box defaults are often way off for LiFePO4.

Also worth checking: is the SmartShunt actually seeing all your current draw, or is something wired on the wrong side of it? Even a small parasitic load bypassing the shunt will cause drift over time.

From what I've read, the BMS SOC tends to be less reliable long-term because it often just uses voltage-based estimation — whereas the SmartShunt is coulomb counting, which is more accurate if it's set up correctly and syncing properly at 100% charged.

Which firmware version are you running on the SmartShunt? There were some SOC sync improvements in recent updates.

@BoxerAdventure classic problem. Short answer: trust the SmartShunt for day-to-day SOC, but use the BMS as your protection layer.

Here's why — the Fogstar's BMS calculates SOC based on cell voltages, which is notoriously inaccurate mid-range (say 20–80%). The SmartShunt uses coulomb counting, so it's measuring actual current in and out. Far more accurate if you've set it up correctly.

Key questions before assuming either is wrong:

• Have you set the correct battery capacity in the SmartShunt app?
• Is your charged voltage threshold set properly?
• Is the shunt the only negative path? Any other negative connections bypassing it will throw everything off.

The SmartShunt needs a proper full charge to re-sync its 100% reference point. Do a full charge, let it hit absorption, and see if they reconcile.

@BoxerAdventure worth adding to what @ExFirefighter said — the divergence is almost always down to the SmartShunt not having a proper synchronisation event to reset its coulomb counting.

The Fogstar Drift's BMS will call 100% SOC the moment cell voltages hit threshold, whereas the SmartShunt needs a genuine tail current condition (typically <4% of capacity, so ~4A on a 100Ah bank) held for a set period before it resets to 100%.

Check your SmartShunt's "charged voltage" and "tail current" settings in VictorConnect — I had mine misconfigured on my van build for months, showing 87% when the BMS was calling full. Once I dialled those parameters in correctly to match the Fogstar's actual charge profile, both aligned within a couple of percent.

The SmartShunt wins on tracking mid-cycle; the BMS wins on protection thresholds. They're doing different jobs.

@ExFirefighter @VanAmy both make fair points but I'm curious — what count as a "proper" full charge for the SmartShunt to sync properly? Like does it need to hit absorption voltage AND hold it for a set time, or just briefly touch it?

Running a similar Fogstar setup in my static caravan and I've noticed the SmartShunt seems to drift further off after cloudy spells where I'm maybe only getting a partial charge through the solar. Wondering if that's the root of it rather than any configuration issue.

Also — does anyone know if the Victron BMS integration (if you're using their own batteries rather than Fogstar) actually solves this properly, or does it still rely on the same synchronisation logic underneath?