Had almost the identical thing happen on my boat last winter — woke up to the Victron BMV-712 screaming at me showing 18% state of charge, yet I'd only had the anchor light and bilge pump float switch running overnight. Hardly what you'd call a heavy load.
Turned out it wasn't the BMS misbehaving at all. The culprit was a combination of two things:
- Cold cell temperatures dragging the actual usable capacity down significantly — lithium really does suffer below about 5°C
- A rogue background draw I hadn't accounted for. My chartplotter was staying in standby rather than actually powering off, pulling around 0.8A continuously
The BMS was actually doing its job correctly — it was reporting accurately, I just hadn't fully understood what I was looking at.
Worth checking a few things on yours:
- What's the cell temperature reading? If you're in a cold location right now, that alone could explain a lot
- Have you run a proper parasitic draw test with a clamp meter on the negative bus?
- Is the battery reporting actual cell voltages via the Victron app, or just a calculated SoC?
The Victron ecosystem is pretty transparent if you dig into the VictronConnect app — you can pull historical data and usually spot the anomaly pretty quickly. I'd start there before assuming the BMS itself is faulty.
What does your overnight consumption log actually show in VictronConnect? That'd be the first thing I'd want to see before drawing any conclusions.