Anyone else struggled to get accurate readings from a Victron SmartSolar in cold weather?

Been running a 400W setup on my static van up in the Lakes — two 200W panels wired in series feeding into a Victron SmartSolar MPPT 100/20. Works brilliantly in summer but since the temperatures dropped last month I've noticed the SOC readings on the VictronConnect app seem all over the place. Battery is a 100Ah lithium (a Fogstar Drift, for what it's worth).

Yesterday morning it was about 3°C overnight and the controller was reading 67% SOC when I woke up, but within 20 minutes of the panels getting any light at all it jumped to 91%. I hadn't used anything overnight beyond a small 12V fan heater on its lowest setting. Doesn't feel right to me. I've got the battery temperature sensor fitted and the charge profile set to lithium, so I don't think it's a settings issue.

Wondering whether this is just the MPPT struggling to get a proper baseline voltage reading in the cold, or whether it's something specific to how the Fogstar BMS reports state of charge at low temps. Has anyone else seen this with a similar setup, particularly those of you running systems year-round in northern England or Scotland where nights get properly chilly? Would love to know if there's a fix or if I'm just going to have to live with dodgy morning readings until spring.

My Victron SmartSolar once told me my tiny house battery was "float" at 9am in January — mate, that's not float, that's hypothermia.

Great thread @CumbrianExplorer — cold weather definitely throws the SmartSolar's readings off sometimes. Worth checking your battery temperature compensation settings in the VictronConnect app. If you've not got a Smart Battery Sense or a temperature sensor fitted, the controller is just assuming ambient temp, which in Cumbrian winters could be wildly off from actual battery temperature. The MPPT will adjust absorption/float voltages based on that assumed temp, so you can end up with incorrect charge curves.

Also worth noting that cold panels actually produce higher voltage — so your series string could be pushing closer to that 100V input limit on frosty mornings than you'd expect. Might be worth keeping an eye on that in the live data.

A Smart Battery Sense is fairly cheap and made a noticeable difference to my readings once fitted. Solved most of my winter weirdness overnight.

Really relate to this — had the same head-scratching moment on my narrowboat last winter. The Victron app was confidently reporting bulk charge when I knew the bank wasn't anywhere near full.

One thing worth trying: have you calibrated the battery voltage compensation settings in VictronConnect? There's a temperature coefficient option that makes a noticeable difference once you actually dial it in properly. Easy to overlook.

Also wondering — are your panels picking up any shading from nearby trees or the van roof edge? Cold plus partial shading can send the MPPT into a right confused state, reading even stranger than usual.

@CotswoldCruiser that "float at 9am" scenario sounds like it could be a dodgy battery temp sensor rather than the controller itself — worth checking whether yours is even connected?

Something that caught me out — the SmartSolar's temperature compensation only works properly if you've actually got a temperature sensor connected (the VE.Direct one or via a BMV). Without it, the controller assumes a fixed ambient, usually around 25°C, and your absorption/float voltages will be wrong in either direction.

Worth pulling up the history data in VictronConnect and checking whether your logged voltages actually match what the battery needed at that temperature. Lead-acid in particular shifts about 4mV per cell per degree C, so a cold morning in the Lakes could easily skew things significantly.

Running a 200Ah LiFePO4 bank here via Fogstar cells — I added a BMV-712 partly just for this reason. The battery voltage readings became noticeably more trustworthy once the temperature data was feeding through properly.

Good point from @DucatoProject there. Worth adding that if you're using the internal temperature sensor rather than an external one, it's measuring the controller's own temperature rather than the battery bank — which in a cold static van could be sitting somewhere significantly warmer than your batteries. I found this out the hard way on my own setup. The controller might be tucked inside an insulated cupboard whilst your batteries are in an external locker taking the full brunt of a Cumbrian winter night. That temperature differential can cause the compensation to actually work against you. If you haven't already, proper external temperature sensor wired directly to the battery is well worth the few quid. Makes a noticeable difference to how the charging profile behaves in sub-zero conditions. The SmartShunt also picks this up if you're running one of those in the network.