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

Had exactly this with my 100/30 on the shepherd's hut build last summer. Turned out the default absorption time was far too short for my battery state — Victron's adaptive algorithm was seeing a voltage spike and calling it "full" prematurely.

Few things worth checking:

• BMS communication — if you're running a smart LiFePO4 (I use Fogstar Drift cells), make sure the MPPT is actually receiving accurate SoC data rather than guessing from voltage alone
• Tail current setting — this is the one most people miss. If your tail current threshold is set too high, it'll exit absorption early. I dropped mine to around 2% of battery capacity and it made a noticeable difference
• Wire losses — voltage sensed at the controller versus actual battery terminal voltage can fool the algorithm. Remote voltage sensing via the VE.Direct port sorts this properly

Also worth checking whether you've got BatteryLife enabled in VictronConnect. It's designed for partial SoC operation and will actively avoid full charges to preserve cycle life — which sounds exactly like what you're describing if you weren't expecting it.

On my emergency backup system in the house I run a SmartSolar 150/35 and had similar behaviour until I disabled BatteryLife and set a fixed absorption time of 2 hours. Much more predictable since.

What battery chemistry are you running and what does your absorption voltage look like in the history tab? That'll narrow it down considerably.

@JYT_Solar — oh mate, I had almost identical grief on the boat last spring. Spent three weekends convinced my Fogstar lithium cells were going duff before I twigged it was the controller being too clever for its own good.

The adaptive absorption algorithm basically learns from yesterday's charge cycle — so if your batteries were already fairly topped up the previous day (say, shore power overnight), it shortens today's absorption window dramatically. Then a grey morning rolls in and it's nowhere near full before float kicks in.

What sorted it for me: switching to fixed absorption time in VictronConnect, set to a proper duration for my bank size. Lost a bit of efficiency in ideal conditions but gained consistency.

Also worth checking your battery voltage sense wiring if you're running that — a dodgy connection there will confuse the algorithm rotten.

@JYT_Solar @Kingy — worth checking your tail current setting if you're on lithium. The MPPT drops to float when charge current falls to that threshold, and Victron's default is often set too high relative to your battery capacity. Rule of thumb is around 2-4% of battery Ah capacity.

On my cabin setup I dropped it from the default down to around 1-1.5% and it made a noticeable difference — batteries were actually reaching proper full charge before float kicked in.

Also double-check your absorption voltage is correctly set for your specific cells. Fogstar and similar cells have slightly different specs to what Victron's preset profiles assume. Connect via VictronConnect and set it manually rather than trusting the built-in profiles.

@RayWatson81 nailed the tail current point — I'd add that on my shepherd's hut Victron 100/50 I had to drop the tail current right down to around 2-4% of battery capacity before it stopped bailing out of absorption early like a nervous apprentice on his first day.

Also worth double-checking your battery voltage sense — if you're relying on controller terminals rather than a dedicated Ve.Smart network or external sense wires, voltage drop across your cabling makes the MPPT think the battery's fuller than it is.

Quick checklist:

• Tail current: 2-4% for lithium
• Absorption voltage set correctly for your specific cells (Fogstar cells can be fussy)
• Enable Ve.Smart networking if you've got a BMV-712 — transforms accuracy overnight

@RayWatson81 @MarineGeoff — good shout on tail current. One thing I'd add from my garden office setup: temperature compensation can quietly throw this whole equation off too. If your battery sensor isn't fitted (or is positioned badly), the MPPT misreads what the battery actually needs and bails out of absorption early.

Spent a confused fortnight last autumn convinced my Fogstar 200Ah was playing up. Turned out the SmartShunt temp sensor was reading ambient air rather than the battery casing. Moved it, problem mostly vanished.

Worth pulling up your MPPT history graphs in VictorConnect — you can see exactly when it's transitioning and whether the voltage curve actually justifies it. Usually tells the story pretty quickly.