Victron Multiplus 12/3000 keeps dropping to float too early — anyone else?

Got a Multiplus 12/3000/120 in my shepherd's hut build, paired with 400Ah of Fogstar Drift LiFePO4. Been noticing the unit cuts to float when the batteries are sitting around 95-96% SOC according to the BMS. Should be hitting 100% before dropping off, surely.

Charge voltage is set to 14.2V absorption, 13.5V float, absorption time 1hr. Using the built-in charger from a Honda EU22i generator. The Fogstar BMS isn't sending any CVL/CCL signals — no DVCC, just standalone.

Anyone tuned their absorption settings to actually reach full on LiFePO4 without just guessing? Wondering if bumping absorption time to 2hrs would help or if I'm barking up the wrong tree entirely.

WingAndPrayer88 | 847 posts

@BatteryTim this is almost certainly your absorption voltage and/or time settings not being tuned properly for LiFePO4. The Multiplus defaults are designed around lead-acid, so it's likely hitting the absorption voltage ceiling and deciding it's done before the cells are properly topped off.

Worth checking your absorption voltage is set to around 14.2-14.4V and that you've got a decent absorption time configured — not just voltage-triggered cutoff. Also double-check your charge current isn't undersized relative to your 400Ah bank, as the Multiplus might be seeing the voltage rise quickly under light current and calling it finished prematurely.

Have you connected it to VictronConnect yet? The charge curve graph in the app is genuinely useful for seeing exactly what's happening in real time rather than guessing.

MasterProject | 134 posts

Had almost identical behaviour on my narrowboat setup with a Multiplus and LiFePO4. Worth checking whether your charge current tail setting is configured — the Multiplus can be set to end absorption when current drops below a threshold, and with lithium's flat charge curve it can trigger that cut-off surprisingly early.

In VictronConnect (or VEConfigure if you're on older firmware), look at the "repeated absorption" and "tail current" parameters. My Fogstar cells settled properly once I dialled the tail current down and extended minimum absorption time to around 30-60 minutes.

Also worth double-checking your battery's recommended charge profile directly with Fogstar — their Drift cells have specific recommendations that don't always match Victron's LiFePO4 presets out of the box.

TracyAllen | 2,341 posts

Worth checking your bulk-to-absorption transition current threshold in VEConfigure — the Multiplus will exit bulk early if charge current drops below a certain percentage of the programmed charge current setting, regardless of actual voltage. With LiFePO4's notoriously flat charge curve, that trigger can fire prematurely.

Also, your Fogstar's BMS may be throttling current slightly at higher SOC (perfectly normal behaviour), which then fools the Multiplus into thinking absorption is "done."

My fix on a similar setup was bumping absorption voltage fractionally to 14.2V and setting a fixed absorption time of 20-25 minutes rather than the adaptive algorithm, which frankly doesn't play nicely with LFP chemistry. The adaptive mode was designed around lead-acid assumptions.

Tail current setting of around 2-3% of battery capacity (so ~8-12A on your 400Ah bank) is where I'd start.

MasterSolar | 1,203 posts

One thing nobody's mentioned yet — your Fogstar Drift BMS is almost certainly reporting a slightly optimistic SOC compared to what's actually happening at the cell level. I ran into this on my static caravan build with very similar kit.

What actually sorted it for me was logging a few charge cycles via VRM Portal and watching the voltage curve rather than trusting the SOC percentage blindly. If you're genuinely hitting 95-96% voltage-wise before dropping to float, that's actually fairly normal LiFePO4 behaviour — the absorption plateau is brutally short compared to AGM.

Try extending your absorption time to a fixed 30-45 minutes rather than relying on the current-based cutoff. VEConfigure makes this straightforward. The Multiplus isn't broken — it's just optimistically efficient with LiFePO4's flat discharge curve confusing the algorithm.

SomersetOffGrid | 847 posts

@BatteryTim I'd add one more angle here — check whether your Multiplus has adaptive charging enabled in VEConfigure. This feature "learns" from previous charge cycles and can progressively shorten absorption time, which might explain why it's dropping to float earlier than expected. After a few cycles it can get quite aggressive about it.

Also worth double-checking your battery capacity setting in VEConfigure matches your actual 400Ah bank. If it's set lower, the calculated tail current threshold will be reached sooner than it should be.

Given you've got a shepherd's hut setup (presumably static rather than mobile), you might also benefit from connecting via VE.Bus Smart Dongle if you haven't already — proper monitoring makes diagnosing this sort of thing much easier rather than guessing blind.

AngliaOffGrid | 2,847 posts

@BatteryTim welcome to the forum — good first post, this is exactly the kind of detail that gets you useful answers.

One thing nobody's touched on: what absorption voltage have you actually set? For Fogstar Drift cells you want 14.2–14.4V for 12V. If it's lower than that, the Multiplus reaches target voltage sooner than expected and legitimately transitions early — it's not misbehaving, it's just been told the wrong target.

Check VEConfig → Charger tab → Absorption voltage first. Dead simple fix if that's the culprit.

On my narrowboat I had almost identical behaviour until I realised a previous owner had set 13.8V absorption — essentially barely above resting voltage for LiFePO4. Took me an embarrassing amount of time to spot.

MPPTGal | 634 posts

Had almost identical behaviour in my own shepherd's hut build last winter. What cracked it for me was checking the absorption time setting — mine was far too short and the Multiplus was simply timing out before the batteries were genuinely full, regardless of what the SOC display claimed.

In VictronConnect, look at "Repeated absorption time" and "Absorption duration." I bumped mine up and the early float drops stopped almost immediately.

@MasterSolar makes a fair point about BMS reporting too — the two issues can compound each other in a frustrating way.