Multiplus-II battery SOC

Right, so this is something that's been nagging at me for a while and I reckon I can't be the only one dealing with it.

Running a Multiplus-II on the narrowboat with a 12V LiFePO4 setup (Fogstar Drift cells, if anyone's curious), and the SOC reading on VRM has become essentially decorative. Completely fictional. It'll show 47% whilst the BMS is reporting the cells are sitting pretty at 98%. Occasionally it drifts the other way and panics at 12% when everything's absolutely fine.

The core problem, as far as I can tell, is that the Multiplus can't properly synchronise its SOC counter unless it sees a genuine 100% absorption cycle — which LiFePO4 never really gives it, because the voltage curve is so flat and absorption time is so short. The thing never gets a proper "full tank" reference point to reset from, so the drift compounds over time.

What's actually helped me somewhat:

• Connecting the Multiplus to the BMS via a DVCC setup so it takes state-of-charge data directly from the battery rather than guessing
• Making sure the charge parameters in ESS are actually tuned for lithium rather than left on some generic profile
• Running a full manual sync through VRM when everything's demonstrably charged

Still not perfect though, and I'm suspicious my shunt calibration is also part of the problem.

Anyone else fighting this on a 12V lithium system? Particularly interested if anyone's sorted it properly with a Victron SmartShunt rather than relying on the Multiplus internal counter — that seems like the more logical architecture to me but I've been putting off the additional wiring.

@Boycie25 the Multiplus-II's SOC estimation is notoriously wobbly without a proper shunt talking to it directly. On my shepherd's hut build I spent weeks scratching my head before I wired in a Victron SmartShunt and connected everything through VE.Direct. Night and day difference — the Cerbo GX suddenly had something real to work from rather than just guessing from voltage curves.

The problem with LiFePO4 is that the voltage plateau is so flat the Multiplus genuinely struggles to infer SOC from it alone. It's not a bug exactly, just physics working against you.

If you haven't already, also check your battery capacity setting in VEConfigure matches your actual Fogstar Drift capacity — people often leave it at default and wonder why the numbers drift over time.

Interesting thread - I've been wrestling with similar questions across a few different setups (shepherd's hut and a van conversion).

@Boycie25 one thing I haven't seen mentioned yet: are you using the VE.Bus BMS or communicating via VE.Can? The communication pathway between your BMS and the Multiplus-II seems to matter as much as having a shunt at all.

Specifically I'm curious:

• Is your BMS actually transmitting SOC data, or just cell voltages?
• Have you checked what the Multiplus-II is actually receiving in VictronConnect vs what the BMS reports natively?

My sceptical side wonders whether some of these SOC drift issues are really a calibration problem rather than a hardware gap. Has anyone done a proper full-cycle test to validate what the shunt reports against actual measured capacity? That seems like the baseline before throwing more hardware at the problem.

@Boycie25 the piece nobody mentions is that the Multiplus-II's internal SOC model resets its calibration reference every time it hits a proper absorption cycle — but with LiFePO4's flat discharge curve, it rarely gets the voltage signal it needs to confirm that's actually happened.

What fixed mine on the motorhome was pairing a Victron SmartShunt with a properly configured BatteryLife setting in VE.Configure, then telling Venus OS to trust the shunt's coulomb counting over the inverter's voltage guesswork. Night and day difference.

One gotcha though — make sure your charged voltage threshold in the SmartShunt settings matches your actual BMS charge termination voltage, not Victron's default. Fogstar Drift cells especially; the BMS cuts off earlier than most people expect.

Had the same headache on my static caravan setup. The fix that actually worked for me — BMV-712 with a proper shunt, wired correctly so everything runs through it. Once VE.Smart networking tied the BMV to the Multiplus-II, the SOC settled right down.

@LiFePO4Nerd is onto