Anyone else finding their MPPT controller underreporting SOC in cold weather?

Been noticing something odd with my setup over the last few weeks now the temperatures have properly dropped. I've got a Victron SmartSolar 100/30 paired with a 200Ah lithium (LiFePO4) from Fogstar, and the SOC reading on the VictronConnect app seems to be consistently reading lower than it should be — sometimes showing 60% when the resting voltage suggests it's closer to 80%.

From what I can gather, LiFePO4 cells do behave differently in the cold and the voltage curve flattens out even more than usual, which probably throws off the calculations. Temperatures here in the East Midlands have been dropping to around 2–4°C overnight in the van, and I'm wondering if that's enough to cause a noticeable drift. I've got the battery temp sensor fitted and it is reporting correctly, so it's not like Victron is completely unaware of the conditions.

Has anyone else seen this with a similar setup, or is there something in the settings I should be tweaking? I did wonder whether recalibrating the SOC after a full charge cycle might help bring it back in line, or whether this is just something you have to live with over winter.

CarolWatson | 847 posts

@Russ1992 Yes, absolutely seen this with my own setup! The thing to check is whether you've got the battery temperature sensor connected and properly configured in VictronConnect. Without it, the MPPT is essentially guessing at SOC based on voltage curves calibrated for room temperature — LiFePO4 voltage characteristics shift noticeably in the cold, so it reads lower than actual.

Also worth checking your absorption and float voltage settings — some people drop these slightly in summer and forget to adjust back. The Victron app lets you set temperature compensation, which makes a real difference once it's properly configured.

What temperatures are you seeing at the battery itself? If it's dropping below about 5°C you might also want to consider whether your Fogstar has low-temperature charge protection kicking in, as that can cause confusing behaviour too.

Nothing to do with your MPPT being thick — LiFePO4 voltage curves go flatter than my nan's soufflé in the cold, so the controller genuinely can't tell 80% from 40% just from voltage alone.

VitoProject | 1,203 posts

@Russ1992 The flat voltage curve issue @Lefty72 mentioned is exactly why SOC estimation from voltage alone is basically useless with LiFePO4 below about 10°C. Your SmartSolar is doing coulomb counting and voltage-based estimation — when those two methods disagree in the cold, it hedges conservative. Annoying but not actually broken.

More importantly: check your charge parameters. Fogstar recommend slightly reduced charge voltages in cold conditions, and if your absorption voltage is set for 25°C ambient you're potentially stressing the cells anyway. Worth setting up a temperature-compensated profile in VictronConnect if you haven't already — the SmartSolar supports it natively.

My 280Ah Eve cells do exactly the same thing every November without fail. I've stopped caring about the SOC percentage and just watch watt-hours instead.

Sam1975 | 312 posts

Worth adding something nobody's mentioned yet — the resting voltage your MPPT reads immediately after a charge cycle will be artificially elevated when it's cold, so it thinks the battery is fuller than it actually is. Then as you draw load, voltage sags faster than expected and the controller scrambles to catch up with reality.

I noticed this badly last January with my static caravan setup. The Fogstar 200Ah I'm running showed 97% SOC at 08:00, then apparently dropped to 61% by lunchtime without much load at all. Confused me for weeks.

The fix that actually helped me was letting the battery rest 20-30 minutes before the MPPT takes its voltage snapshot. Some of the Victron charge profile settings let you tweak absorption timing, which indirectly helps this.

MaxFrost | 87 posts

Had almost identical confusion on my boat last winter — the Victron app was showing 40% when the battery was practically full after a proper charge cycle.

One thing worth adding: are you using the Victron battery profile or a custom one? I found that selecting the wrong battery preset skewed the SOC readings noticeably, even before cold temperatures complicated things further.

Also worth checking — does your Fogstar battery have a BMS that reports SOC separately via Bluetooth? If so, which figure do you actually trust when the two disagree? That's been my ongoing headache. The BMS and the MPPT rarely agree in cold conditions.

GlenDoug | 641 posts

One thing worth being precise about — your MPPT isn't really designed to track SOC accurately, it's a charge controller first. The SOC readout is essentially voltage-based inference with some current integration bolted on. For proper SOC visibility you want a dedicated shunt like the Victron BMV-712 or the SmartShunt. That's what I run in my garden office setup alongside the SmartSolar, and the two don't always agree, particularly after a cold night.

The MPPT will also apply temperature compensation to its charging voltages if you've enabled that — worth checking your VictronConnect settings to confirm it's actually active. If it's not compensating properly it could be slightly undercharging, which would then show as lower SOC regardless of the reporting method.

Two separate issues, easy to conflate.