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

Had exactly this with my 100/30 last summer. Turned out the absorption voltage was set correctly but the absorption time was far too short — default settings on Victron kit assume a certain battery capacity and internal resistance that almost certainly won't match your actual bank.

Worth checking a few things:

• Tail current setting — if this is too high, the MPPT thinks the battery is full when it isn't. Drop it to around 1-2% of your battery capacity in Ah.
• Absorption time — are you using fixed or adaptive? Adaptive can be erratic on partly cloudy days because it calculates duration based on the bulk phase, which gets cut short by passing clouds.
• Battery temp — on colder days the Victron will naturally push float earlier if it's sensing (or assuming) low temps without a proper VE.Smart temperature sensor feeding it accurate data.

My setup is a 200Ah Fogstar Drift LiFePO4 paired with a 150/35, and I spent weeks chasing this before realising the tail current was set at the factory default of 4A, which was way too aggressive for my bank size.

Connect via VictronConnect and watch the live graphs during a full charge cycle — you'll see exactly when it transitions and why. The history tab showing absorption duration per day is genuinely useful for spotting patterns.

What battery chemistry are you running? Lead-acid, gel, and lithium all behave completely differently here, so the fix varies quite a bit depending on your setup.

@IslandVanLifer hit on the key issue. Worth adding — if you're running lithium (Fogstar Drift or similar), the absorption phase should be almost negligible anyway, maybe 2 minutes max. The early float drop becomes a non-issue because lithium absorbs charge right up to the voltage threshold without needing the trickle-in period lead-acid does.

For lead-acid users still seeing this: check the tail current setting in VictronConnect. If your bank is undersized relative to panel array, the tail current threshold gets hit prematurely. Default is often 2A, which on a small bank reads as "full" when it absolutely isn't.

Also — absorption time is adaptive on Victron MPPTs. Short previous discharge = shorter absorption next cycle. If you've been doing shallow cycles, the controller learns that. Check your history tab.

Really useful thread this — worth adding that the battery temperature plays a massive role too, especially if your setup is exposed to UK ambient conditions (a static caravan install like mine spends winter in genuinely cold air).

Victron's adaptive absorption algorithm will shorten absorption time dynamically if it thinks the battery charged quickly, which in cold weather can mean a partially discharged bank gets flagged as full prematurely.

Worth checking in VictronConnect → Battery settings → Adaptive absorption — try disabling it temporarily and setting a fixed absorption duration based on your actual battery capacity and typical depth of discharge.

@IslandVanLifer and @LH_Marine have covered the voltage and lithium angles well, so I won't rehash those.

Also — welcome to anyone new jumping in on this thread. These MPPT quirks catch everyone out initially; don't be discouraged, the diagnostic tools Victron provide are genuinely excellent once you know where to look.

Good point from @Nessa on temperature — though I'd add the other side of that coin worth checking is the battery sensor itself.

Had this exact problem in my cabin setup last winter. Victron was reading ambient air temp from the SmartShunt rather than actual battery surface temp. Result was the controller thinking the bank was warmer than it was, compensating by cutting absorption short.

Stuck a proper VE.Direct temperature sensor directly on the battery terminal. Problem gone.

Worth ruling out before you start fiddling with absorption time settings — you can chase that rabbit down a long hole otherwise. Check what your controller is actually using as its temperature reference first.

Good points all round, but nobody's mentioned state of charge going into the charge cycle as a variable yet.

If your batteries are only sitting at, say, 80% overnight (high overnight loads, poor insulation, whatever), Victron's adaptive absorption algorithm will cut absorption short because it thinks the bank charged quickly and therefore wasn't deeply discharged. It's working as designed — but the design assumption doesn't always match real-world usage patterns.

On my setup I had to switch from adaptive to fixed absorption time in VictronConnect, set to something appropriate for my actual bank capacity. Once I did that, the controller stopped second-guessing itself.

Worth logging a full charge cycle via the VRM portal or even just Bluetooth if you haven't got remote monitoring — the absorption duration column tells you immediately whether it's cutting short.

Really solid point from @LDVAdventure — incoming SoC is massively underrated as a variable here.

Worth adding another angle nobody's touched on: panel-to-battery wire losses. If your cabling is undersized or connections are corroding, the MPPT sees a higher voltage at its input terminals than the battery is actually receiving. It'll happily declare absorption complete whilst the battery is still genuinely hungry. Victron's VE.Smart networking (pairing with a BMV-712 or SmartShunt) solves this elegantly by feeding actual battery voltage back to the MPPT rather than relying on measured terminal voltage.

Had exactly this on my garden office install — 4mm² cable run that was borderline too long. Switched to voltage-sense via Bluetooth network and absorption periods immediately lengthened correctly. Check your voltage drop calculations, especially on longer cable runs. Victron's own documentation recommends keeping drop under 0.5% ideally.