Victron Multiplus keeps dropping to float too early — anyone else had this?

Had this exact issue on my narrowboat setup last winter. The Multiplus was declaring float after maybe 2–3 hours on absorption, even though the Fogstar Drift LiFePO4 cells were nowhere near fully charged — sitting around 95% by my Victron BMV-712's reckoning.

Turned out my absorption voltage was set slightly too high, so the charger was "seeing" the target voltage being hit before the cells were actually balanced. Dropping absorption voltage by about 0.1V and extending the absorption time manually in VictronConnect sorted it for me.

A few things worth checking:

• Absorption voltage — is it correctly set for your battery chemistry?
• Tail current setting — if this is too high it'll cut to float prematurely. For LiFePO4 I've seen recommendations around 1–2% of battery capacity
• Battery temperature — cold batteries behave oddly and can confuse the charge algorithm
• BMS communication — if you're running DVCC with a compatible BMS, sometimes the BMS is actually the one calling float early

Worth connecting via VictronConnect and watching the charge parameters in real time during a full cycle. The live graphs are genuinely useful for spotting where it's going wrong.

Curious whether others are seeing this more on LiFePO4 or on AGM/gel setups too? My previous lead-acid bank never had this problem, so I wonder if it's partly down to the flatter charge curve on lithium catching the Multiplus out.

@AshSeeker Classic Victron/LiFePO4 mismatch — the Multiplus is using tail current to exit absorption (typically 2-5% of battery capacity), but if your Fogstar Drift BMS is already smoothing the current draw internally, the charger sees a falsely low tail current and legs it to float prematurely.

Check your charger.current_cutoff setting in VEConfigure — I'd nudge the absorption tail current threshold down to around 1-2% of your bank's Ah rating, and make sure your absorption voltage is actually hitting the cells (3.45-3.50V per cell is your friend here).

Also worth enabling the adaptive absorption algorithm with a fixed minimum absorption time of at least 1 hour as a safety net — stops it bolting early on partially discharged cycles too.

@AshSeeker had the exact same nightmare on my static van with a Fogstar Drift 200Ah — the fix that saved my sanity was bumping the absorption time to a fixed duration in VictronConnect rather than letting it exit on tail current, because LiFePO4 internal resistance is so low it hits that threshold embarrassingly early.

Set a repeated absorption time of around 1 hour per 100Ah as a starting point, then tweak from there — my 200Ah bank needed about 2 hours to genuinely top off.

Also worth checking your charge voltage is actually set to 14.2V (or whatever your Fogstar spec says), not the default lead-acid profile lurking in there causing confusion — Victron's defaults are brilliant for lead, less so for lithium until you've had a proper dig through the settings. 🔧

@OldSailor is right about tail current, but worth adding — if you're on VE.Connect you can actually watch this happen in real time. Check the battery current graph during absorption and you'll likely see it nosediving well before the cells are full.

On my shed setup with a 280Ah Fogstar Drift I found the culprit was an overly optimistic C-rate assumption baked into Victron's defaults. The tail current threshold was triggering against the charger's rated output rather than actual battery capacity.

Bump your absorption voltage up slightly to 3.55V/cell if you haven't — LiFePO4 has that flat discharge curve that fools voltage-based logic badly. Combined with adjusting the tail current percentage in VEConfig it made a significant difference without needing a BMS-driven signal.

@AshSeeker one thing nobody's mentioned yet — check your battery temperature. LiFePO4 internal resistance drops significantly when cold, and on a narrowboat in winter you're potentially seeing genuine tail current being reached faster than expected simply because the cells are performing differently at 5–8°C versus 20°C. The Multiplus isn't necessarily misconfigured; it may be reading the situation accurately but your actual charge acceptance is genuinely reduced.

My tiny house setup had identical symptoms until I added a BMV-712 with temperature compensation properly configured. Worth cross-referencing the Multiplus charge log timestamps against ambient temperatures overnight — if early float correlates with colder mornings, that's your culprit rather than a software misconfiguration.

Fogstar Drift cells are decent but their datasheet charge curves assume nominal temperature ranges.