Had something very similar on my narrowboat last winter. Spent weeks convinced it was temperature-related because that's the obvious culprit, but it turned out to be a cell imbalance issue that only revealed itself under low overnight loads.
The BMS was tripping on a low cell voltage deviation rather than overall pack temperature or SOC. One cell was lagging slightly behind the others, and at 2-3am when everything had been drawing slowly for hours, it would finally drift far enough out of tolerance to trigger protection.
A few things worth checking:
- Connect via Bluetooth and log the individual cell voltages over a full overnight cycle — don't just check in the morning after it's recovered
- What's your resting SOC before bed? If you're regularly sitting below 30% overnight, a weak cell will show itself
- Any chance of a parasitic draw you've not accounted for? Even 0.5A continuous adds up across 8 hours
Also worth checking your battery protect settings if you're running one in line — sometimes the low voltage disconnect threshold interacts oddly with the Victron BMS logic and you get a cascade trip that looks like a BMS fault but isn't quite.
What firmware version are you on? There were some known quirks in earlier Victron Smart Lithium firmware around the protection thresholds that got tidied up. Worth checking the Victron community forum too — their support is generally decent at responding to BMS behaviour questions.
Would be interested to hear what your cell voltage spread looks like if you can pull the data. Others on here might have seen the same pattern.