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

Had this exact head-scratcher on my narrowboat last winter. Both my Fogstar 200Ah LiFePO4 BMS and the Victron SmartShunt were confidently telling me completely different things — one said 74%, the other 61%. Maddening.

After a fair bit of digging, here's what I landed on: the SmartShunt is generally the more trustworthy day-to-day SOC reading, provided you've set it up properly — correct battery capacity, Peukert exponent, charge efficiency factor, all of that. It's measuring actual coulombs in and out, sitting on the negative bus and watching everything.

The BMS SOC, on the other hand, tends to be estimated from cell voltages, which on LiFePO4 is notoriously unreliable across that big flat middle section of the discharge curve. A tiny voltage difference can swing the reading wildly.

That said, neither is infallible. The SmartShunt drifts over time if it never gets a proper synchronisation point — i.e. a full charge to let it reset to 100%. I found mine had been drifting for weeks because my solar wasn't quite hitting absorption properly through all the grey days.

The fix that worked for me: trust the SmartShunt, but make sure it's syncing correctly on full charges. Keep the BMS reading as a secondary sanity check — if they diverge by more than 10-15%, something needs investigating rather than ignoring.

Curious whether anyone running Victron kit with a different BMS brand has found a way to get them talking to each other properly? I've seen mentions of CAN bus integration but never gone down that rabbit hole.

TrevorEvans replied:

@HazelDawn Classic narrowboat winter dilemma! The short answer is — trust neither blindly, but lean toward the SmartShunt if it's been properly calibrated.

The BMS SOC is often just voltage-based estimation, which is notoriously unreliable with LiFePO4 given that incredibly flat discharge curve. The SmartShunt uses coulomb counting (measuring actual current in and out), which is inherently more accurate provided you've set the correct battery capacity, Peukert exponent, and — crucially — the charged voltage threshold so it knows when to reset to 100%.

Where it commonly goes wrong is the shunt not being the single path for all current. Any loads or charging sources bypassing it completely skews the count.

Worth checking your wiring — every single positive connection should route through that shunt, no exceptions. What's your setup look like for charging sources?

DodgyRoamer replied:

@HazelDawn The root cause is almost always Peukert's law and temperature compensation — two things BMS units and coulomb counters handle very differently.

Your Fogstar BMS measures cell voltages and infers SOC from the discharge curve. Your SmartShunt counts coulombs in/out but needs a correctly configured Peukert exponent and charge efficiency factor for LiFePO4 (typically 1.05 and 99% respectively — check your VictronConnect settings).

Key questions:

• Has the SmartShunt ever seen a proper full synchronisation point (i.e., tail current reached at full charge)?
• Is your battery temperature compensation set correctly for winter conditions?

In my garden office setup I run both, but I treat the SmartShunt as ground truth only after a confirmed full charge sync. The BMS SOC readout on Fogstar units tends to drift badly after several partial cycles.

Calibrate the SmartShunt properly first — then compare again.

GrumpyBuilder replied:

SmartShunt wins, full stop. The BMS SOC is usually just voltage-based nonsense dressed up as a percentage — LiFePO4 flat discharge curve makes that completely unreliable in the middle range.

SmartShunt is coulomb counting. It's measuring current in and out. Still not perfect — needs a proper sync point at 100% (full charge, tail current drops off) — but it's actually doing maths rather than guessing from voltage.

Main thing people cock up: the charged voltage and tail current settings in VictronConnect. Most leave them at defaults which are rubbish for Fogstar cells specifically. Sort those out and your SmartShunt will be far more trustworthy.

My cabin setup had the same argument between BMS and SmartShunt for months until I dialled the charge parameters in properly. Night and day difference.

ShaunWard replied:

@GrumpyBuilder I'd push back a bit there — it's not quite that simple. The SmartShunt is brilliant at coulomb counting but it drifts if you haven't given it a proper full charge sync recently. On a narrowboat in winter you're often running the engine for shorter periods and solar is minimal, so the shunt may not have seen a genuine 100% reference point in weeks. Meanwhile the BMS knows the cell voltages intimately and will catch things the shunt misses entirely, like cell imbalance skewing your usable capacity.

What I'd actually do @HazelDawn is check when your SmartShunt last synchronised — you'll see it in the VictronConnect history. If it's been a while, do a proper full charge and let it reset. After that the two figures should start converging reasonably well.