Narrowboat battery bank keeps dropping overnight even with solar topped up

Had the exact same drama on my 57ft widebeam last summer — turned out my Victron BMV-712 was lying to me because someone (me) had never properly set the battery capacity or Peukert exponent, so it thought the bank was full when it absolutely wasn't. 🙃

Few things worth checking:

• Parasitic draws — bilge pump float switch, fridge compressor cycling at 3am, that USB charger you forgot was plugged in... they all add up overnight when your Renogy panels have gone to sleep
• State of charge vs actual capacity — if your batteries are ageing, they'll accept a charge and look full but the usable capacity has quietly shrunk
• What chemistry are you running? LiFePO4 will hold voltage deceptively well until it suddenly doesn't, whereas AGM will sag more honestly
• Any chance your Fogstar cells (if that's what you've got) need a proper absorption cycle rather than just float-topping?

Also worth sticking a clamp meter on your negative bus at midnight to catch anything sneaky drawing current — narrowboats are notorious for random 12V accessories wired by twelve different previous owners with twelve different ideas about what a fuse is.

What's your panel wattage vs bank size ratio? Because depending on how much shade you're getting through locks and marinas, your solar might be putting in far less than you think even on a "sunny" day.

@StormyDrifter classic "garbage in, garbage out" situation with the BMV 😄

Worth also checking your charged voltage and tail current settings — if those are wrong, the BMV never properly registers a full charge, so it's perpetually pessimistic about state of charge.

Done this myself on my cabin bank. Had it set to factory defaults for months wondering why my Fogstar cells apparently hated me.

Also — parasitic loads on narrowboats are sneaky. Bilge pump float switches, gas detectors, inverter standby draw... all adds up overnight when solar's doing nothing. Stick a clamp meter on the negative before bed, see what's actually leaking out.

The Peukert thing matters more for AGM than LiFePO4 btw. If you've gone lithium, set it close to 1.05 and stop overthinking it.

@StormyDrifter the Peukert exponent is massively overlooked — for LiFePO4 it should be very close to 1.05, whereas flooded lead-acid typically sits around 1.25. If you've swapped chemistry at any point and forgotten to update that figure, your BMV will be wildly inaccurate under load.

Also worth auditing your parasitic loads overnight. Narrowboats are notorious for this — bilge pump float switches, 12V fridge compressors cycling, inverter standby draw. My setup on a widebeam was losing nearly 8Ah per night just from an old Waeco fridge's standby circuit.

Run a proper tail current calibration too. BMV-712 needs to see a genuine absorption finish before it'll reset the SOC to 100% — if your Victron MPPT is undersized for the bank capacity, it may never hit that threshold on a typical overcast UK day.

@StormyDrifter similar issue on my motorhome before I got the BMV settings dialled in properly.

One thing nobody's mentioned yet — have you set your tail current correctly? This is the percentage of charge current at which the BMV declares the battery "full" and resets the SOC to 100%. If it's too high, it'll call the battery full prematurely, and your overnight baseline is already wrong before you even start discharging.

For LiFePO4 I'd set it around 1-2% of capacity — so for a 200Ah bank that's 2-4A.

Also worth double-checking your synchronisation voltage threshold. If it's triggering a false 100% sync during a brief voltage spike from the solar, that'll completely skew your readings.

What capacity is your bank and what chemistry are you running?

@StormyDrifter been down this exact rabbit hole on my narrowboat — one thing nobody's flagged yet is tail current.

The BMV decides your battery is "full" based on a combination of voltage and how low the charge current has dropped. If your tail current is set too high, the BMV declares victory too early, the Peukert counter resets, and suddenly your 100% reading is actually 85%.

On my LiFePO4 bank (Fogstar Drift cells) I run tail current at around 1-2% of capacity. So a 200Ah bank needs it dropping below 2-4A before the BMV calls it full.

Check your History screen on the BMV too — if your "time since full charge" is creeping up even on sunny days, that's usually the culprit staring you in the face.