Anyone else finding their MPPT controller wildly overestimates battery SOC in winter?

So I've been scratching my head over this for the past few weeks. I've got a Victron SmartSolar 100/30 paired with two 200W panels (wired in series, so around 44Voc) feeding into a 200Ah LiFePO4 battery from Fogstar. The system's been rock solid all summer, but since October the SOC readings have been all over the place — jumping to 100% after barely an hour of weak winter sun even though the battery was clearly at 60-70% the night before.

From what I can tell, the MPPT is hitting absorption voltage (14.2V in my case) pretty quickly on a cold, lightly loaded morning and just declaring the job done. The thing is, at low temperatures and low charge currents, LiFePO4 cells do sit at a higher resting voltage anyway, so I get why it's confused — but it's making it really hard to trust the display for any kind of planning.

I've been playing around with the absorption time settings in VictorConnect and wondering whether bumping it up, or maybe lowering the absorption voltage slightly for winter, would actually help. Has anyone found a config that works well for UK winters specifically? I'm also wondering if adding a proper battery monitor like the Victron BMV-712 would just solve the problem entirely by tracking amps in and out rather than relying on voltage.

Would love to know how others are handling this — especially if you're running LiFePO4 rather than AGM, as most of the guides I find online still seem to be written with lead-acid in mind.

Hey @SophieClark, yes, absolutely seeing this! I'd wager your controller is relying too heavily on voltage-based SOC estimation rather than coulomb counting, and in winter the lower panel output means the battery rarely hits proper absorption properly, so the controller thinks it's fuller than it is.

With LiFePO4 the flat discharge curve makes voltage-based readings even more unreliable than with lead-acid.

A couple of things worth trying — have you enabled the "tail current" setting in VictronConnect? Getting that dialled in correctly makes a surprising difference. Also, pairing your SmartSolar with a Victron BMV battery monitor (or the SmartShunt) would give you proper coulomb counting and genuinely accurate SOC. The two talk to each other over Bluetooth which is rather handy. Makes winter management so much easier once you've got it sorted! 🙂

@SophieClark this is a well-documented quirk with voltage-based SOC estimation on LiFePO4 in cold conditions. The flat discharge curve already makes voltage a poor proxy for SOC, but cold temperatures suppress cell voltage further, so the BMS/controller sees artificially depressed voltage and misreports capacity.

On my boat setup I've got a Victron BMV-712 doing coulomb counting as the primary SOC reference — the SmartSolar then defers to that via the VE.Smart network rather than doing its own voltage guesswork. Makes a substantial difference in winter.

Worth checking in VictronConnect whether you have a temperature sensor fitted — without one, the charge algorithm isn't compensating properly for temperature either, which compounds the problem. The SmartSolar 100/30 supports the optional Victron temperature sensor dongle, or you can get the data via a compatible BMS with VE.Smart broadcasting.

Coulomb counting paired with temperature compensation is really the only reliable solution here.

Something worth checking in your Victron Connect app — have you set the battery capacity correctly under the charger settings? If that's wrong, the coulomb counting will be miles off from the start.

Also, are you using a BMV-712 or SmartShunt alongside this? The SmartSolar on its own is fairly limited for accurate SOC. Once I added a SmartShunt to my system, the readings became dramatically more trustworthy. The MPPT and shunt talk to each other over VE.Smart networking and the SOC figure actually reflects real-world charge state rather than just guessing from voltage.

One more thing — check your tail current setting. If it's too low, the controller thinks the battery is full before it actually is, which compounds the winter problem @JulieEvans and @BreezyHermit are describing.

@SophieClark had exactly this on my narrowboat last January. The flat discharge curve on LiFePO4 makes voltage-based SOC almost meaningless, especially once temps drop below about 5°C — the resting voltage shifts enough to throw everything off.

What actually helped me was enabling the VE.Smart Networking feature and pairing a BMV-712 as the "master" SOC source. The SmartSolar then defers to the shunt-based reading rather than guessing from voltage. Night and day difference in accuracy.

Also worth checking your absorption and float voltage thresholds — Victron's defaults lean toward AGM territory and may need nudging down slightly for LiFePO4 in winter, otherwise the controller thinks it's hit full charge far too early.

What battery brand are you running? Some Fogstar and similar cells have notably different cold-temp voltage behaviour compared to the spec sheet.

Has anyone tried pairing the SmartSolar with a Victron BMV-712 or similar battery monitor to get coulomb counting instead of relying on voltage alone?

On my boat I was getting exactly this kind of drift over winter — the MPPT was reading 80%+ SOC when the BMV was telling a completely different story. Once I trusted the BMV as the primary SOC reference it made a huge difference to how I managed loads.

@SophieClark what's your current setup for actually measuring SOC — are you relying solely on what the SmartSolar reports, or do you have a dedicated shunt-based monitor in the loop?

@EwanCole58 yes, exactly this — BMV-712 is the proper fix. Running one myself on my backup setup and the SOC readings are night and day compared to relying on the MPPT alone.

Worth noting though: make sure you configure the tail current and charged voltage settings correctly in the BMV or it'll never sync properly and you're back to square one. Caught me out first time round.

Also @SophieClark — cold temps actually reduce your usable capacity too, so even accurate coulomb counting will show you "losing" capacity in winter. That's normal for LiFePO4, not a fault.

@EwanCole58 and @MoorCamper have basically nailed it — the BMV-712 is the answer here. One thing worth adding though: make sure you calibrate the shunt correctly and set your battery capacity, Peukert exponent, and charge efficiency factor properly in the VictronConnect app. Out of the box defaults aren't optimised for LiFePO4. I've got mine set to ~1.05 Peukert and 99% charge efficiency on my Fogstar Drift cells and the SOC tracking is rock solid even through the worst Scottish winters. Takes 20 minutes to configure properly but makes a massive difference.

Been running a BMV-712 on my boat for a couple of years now and it's transformed SOC accuracy, especially through the dark months when you're barely hitting absorption.

One thing nobody's mentioned yet — make sure you nail the Peukert exponent and charge efficiency factor in the BMV settings. Defaults are fine for lead-acid but LiFePO4 needs tweaking (charge efficiency closer to 99%, Peukert around 1.05). Makes a noticeable difference to how accurately it tracks.

Also worth syncing the BMV tail current setting properly — that's often where SOC drift creeps in on lithium.