EEL Akkus mit Seplos BMS zeigen falschen SOC an und MP2 schaltet ab

Had this exact headache with my shepherd's hut setup last winter — Fogstar Drift cells with a Seplos BMS reporting 10% SOC while the pack was sitting at 51.4V, which on a 48V LiFePO4 system is nowhere near empty.

The Multiplus-II was dutifully throwing a low battery shutdown like an obedient but thick labrador, completely ignoring the fact there was plenty of juice left.

Turned out my DVCC settings in VenusOS were the culprit — the MP2 was trusting the BMS SOC figure over actual voltage, which sounds sensible in theory but falls apart when your Seplos hasn't been properly calibrated after a full charge cycle.

A few things worth checking:

• Full charge calibration — the Seplos needs to hit a proper top-balance to reset the SOC counter; mine was never fully charged after initial install
• DVCC / SVS / STS flags in Cerbo/VenusOS — experiment with which signals you're letting the Multiplus trust
• BMS firmware — Seplos have quietly pushed updates that improve coulomb counting accuracy
• Cell drift — if one cell is running away, the BMS panics the whole pack based on that outlier

Spent an embarrassing number of evenings poking at this with a VE.Direct cable and a mug of tea before it clicked.

Anyone else running a Seplos Mason with a Victron stack hit this? Curious whether it's more common with prismatic cells that haven't been properly top-balanced before commissioning — suspect that's the root cause for most of these phantom-low-SOC reports.

@CamperCarl that 51.4V on a 48V LiFePO4 pack is essentially fully charged territory — something's badly miscalibrated in your Seplos config.

Few things worth checking:

• Cell count set correctly in the BMS? If it thinks it's a 15S pack when you've got 16S, the SOC maths goes completely sideways
• Capacity setting — does the BMS know the actual Ah figure of your Fogstar cells?
• The Seplos needs a proper full calibration cycle — charge to absorption, hold until current drops to ~5A tail, then let it learn 100% SOC

The MP2 shutting down is almost certainly reacting to the BMS's dodgy SOC signal via DVCC rather than actual cell voltage. Check your VE.Bus System Configurator minimum SOC threshold too.

Had similar grief on my shepherd's hut with a different BMS — always turned out to be setup parameters rather than a faulty unit.

@MrBodge65 is absolutely right on the voltage. 51.4V across a 16S LiFePO4 pack is ~3.21V/cell — firmly mid-to-upper charge territory, certainly not 10% SOC.

The culprit here is almost certainly the Seplos BMS losing its coulomb-counting calibration. This happens when the pack never reaches a proper full-charge synchronisation point — the BMS needs to see all cells hit the top-balance voltage simultaneously to reset the SOC to 100%.

A few things to check:

• Cell-level voltages via the Seplos app — are any cells diverging significantly?
• Charge parameters in your Victron MPPT/Multiplus — if absorption voltage is set too low, the pack never triggers the full-charge flag
• The Seplos SOCSYNC setting specifically

The MP2 shutdown is almost certainly the DVCC/SOC-based low-SOC cutoff responding to the reported 10%, not the actual pack state. Fix the calibration first.

@DevonDweller @MrBodge65 worth checking whether the Seplos has been set up with the correct battery chemistry profile — I've seen units ship defaulted to a voltage curve that doesn't match LiFePO4's flat discharge plateau properly, which makes the coulomb counting go haywire after a few cycles.

On my motorhome setup with a Victron Cerbo talking to a Seplos via CAN, the SOC calibration only sorted itself once I forced a full charge-to-CV-absorb cycle and let it reset the 100% anchor point. The BMS needs to "see" a proper top before its SOC tracking becomes trustworthy.

Also double-check your DVCC settings on the Multiplus II — if the MP2 is seeing a dodgy SOC signal over CAN it'll honour it and shut down prematurely regardless of actual pack voltage.

@HazelDweller that's a really good shout on the chemistry profile — had something similar on my static caravan setup where the Seplos was defaulting to a generic lithium profile rather than LiFePO4 specifically. Completely threw the SOC calculation off.

Quick question for @CamperCarl though — have you done a full capacity calibration cycle since installing? My understanding is the Seplos needs to see at least one proper charge-to-full then discharge-to-low cycle before its coulomb counting settles down properly. Did yours ever actually reach the top balancing voltage, or was it cutting out before that point?

Also worth checking — is the Multiplus II talking to the Seplos via DVCC or just using the voltage/current sensing independently? That could explain why the MP2 is acting on dodgy SOC data rather than the actual pack voltage.