Anyone else had grief with a Victron SmartSolar 100/30 dropping into absorption too early on cold mornings?

My 200Ah lithium (LiFePO4) setup has been behaving oddly since the weather turned. Running a Victron SmartSolar 100/30 into a 24V battery bank with two 175W panels on the roof of my Transit conversion. During summer it was spot on, but now we're getting those crisp 4–5°C mornings I'm noticing the controller is jumping into absorption phase after only pulling maybe 15–20Ah back in, even when the bank was sitting at around 60% SOC overnight.

I've got the absorption voltage set to 28.4V and float at 27.2V, which is bang on for my Winston cells according to the spec sheet. The SmartSolar is connected via Bluetooth and I can watch it in the VictronConnect app — the voltage does genuinely hit 28.4V briefly on those cold mornings, presumably because the internal resistance of the cells climbs in low temps. So the controller thinks it's done when it really isn't.

Has anyone found a practical fix for this without going full-blown MPPT temp compensation trickery? I did wonder about dropping absorption voltage slightly for winter, maybe to 28.0V, and relying on a longer absorption time instead. Would that actually help or am I barking up the wrong tree?

Hey @BarryCrane, classic cold weather headache this one. Worth checking your absorption voltage threshold - LiFePO4 cells have a lower internal resistance when cold, so the controller can see voltage rise faster than expected and think it's already hit absorption.

Have you set the battery temperature compensation to zero? That setting is designed for lead-acid and can actually work against you with lithium. Also double-check your tail current setting if you're on a custom profile - dropping it slightly can help stop premature transitions.

What BMS are you running? Some cheaper units restrict charge current in the cold which artificially inflates the voltage reading at the MPPT terminals. Could be the BMS rather than the Victron itself causing the confusion.

@BarryCrane my SmartSolar was doing the exact same thing on the shepherd's hut last January — turned out the battery temperature sensor wire had worked itself loose, so the MPPT was essentially guessing and guessing wrong like a blindfolded dart player. Check your VE.Direct connection and if you're not running a Smart Battery Sense or temp sensor, that's your culprit right there — it'll default to room temp assumptions which in a cold Transit at 6am is absolutely useless.

@BarryCrane mine was doing something similar on the boat last February — turned out the MPPT was reading phantom voltage spikes off cheap cable runs and thinking it'd hit absorption already, so check your wiring resistance before you go mad tweaking the charge profile itself.

Hey @BarryCrane, just to add to what @Chunk and @SamFrost are getting at — have you got a battery temperature sensor connected to the MPPT? Without one, the controller's just guessing based on ambient temp at the unit itself, which on a cold Transit roof can be wildly different to your actual battery bank temp. The SmartSolar will apply temperature compensation and can shift voltage thresholds quite aggressively. Worth grabbing a Victron Smart Battery Sense or wiring in the optional temp sensor. Also double-check your absorption voltage is set correctly for LiFePO4 in VictronConnect — should be around 28.4V for 24V, not the default lead-acid profile. Easy to overlook if the unit came pre-configured.

@BarryCrane worth checking your absorption voltage threshold in the VictronConnect app — on a 24V LiFePO4 bank it should typically be around 28.4V, but if it's been left at the AGM default of 28.8V or higher, the MPPT can interpret early morning resting voltage combined with any small surface charge as being closer to absorption than it actually is. Cold cells also have higher internal resistance, so they'll show an inflated voltage under light load. Pop into VictronConnect, head to the charger settings, and double-check your charge algorithm is set correctly for lithium rather than defaulting to a lead-acid profile. Also worth confirming your bulk-to-absorption transition current threshold if you're on a newer firmware version — that setting catches a lot of people out.

Good shout from @BurnSpirit on the absorption voltage, and @BoatMark's phantom voltage point is worth investigating too. One thing nobody's mentioned yet — have you checked whether your panels are wired in series or parallel? On a cold morning with frost, Voc can spike noticeably higher than the rated spec, which can confuse the MPPT's calculations temporarily. Also worth having a look at your charge history in VictronConnect — if you tap into the battery icon you can see whether the controller is actually logging unusually short bulk phases. That'll tell you pretty quickly whether it's genuinely hitting absorption voltage early or whether it's a sensor/reporting oddity. What firmware version are you running on the SmartSolar? There were a few quirks in older builds that Victron quietly sorted in later releases.

Good points all round from the others. One thing I haven't seen mentioned yet — check whether your SmartSolar is picking up panel voltage bleed through the PV input at night or early morning. On my boat setup I had the controller waking with a slightly inflated battery reading before the bank had properly settled. The fix was enabling the "tail current" setting in VictronConnect, which meant absorption only kicked in once the charge current actually dropped to a meaningful threshold rather than reacting purely to voltage. Made a real difference through the colder months.

Had this exact issue last winter on my narrowboat setup. Worth checking your temperature compensation settings in VictronConnect — if you've got a Smart Battery Sense or temp sensor connected, make sure the compensation coefficient is set correctly for LiFePO4 (should be around 0mV/°C, as most lithium BMS units handle temp compensation internally). If Victron's MPPT is also applying temperature compensation on top of what the BMS is doing, you can end up with the absorption threshold being hit artificially early on cold mornings. @CallumHobbs might be heading somewhere similar with his point. Definitely worth a look before anything else.

@Curly63 beat me to the temp compensation point so I'll leave that alone.

One thing worth checking — are you using the VE.Direct cable or just Bluetooth to monitor? I've seen the app show slightly different voltage readings vs what the controller actually sees at the terminals.

Also, what are your absorption time settings? On my cabin setup I dropped the fixed absorption time right down for LiFePO4 — the default settings are really meant for lead acid and can confuse things, especially when the bank isn't fully depleted to begin with.