SmartSolar 75/15 Überspannung am Batterieausgang wenn BMS von Akku beim Laden abschaltet

Bit of a niche one this, but I've been wrestling with something similar on the boat and wondered if anyone else has run into it.

Running a Victron SmartSolar 75/15 into a 12V LiFePO4 bank (Fogstar Drift cells with an integrated BMS). The scenario that keeps catching me out: mid-charge, the BMS trips — overvolt protection on an individual cell going high before the others catch up — and suddenly the MPPT has nowhere to push that current.

What I'm seeing on the VictronConnect graphs is a voltage spike at the battery terminals at the exact moment of disconnect. Not enormous, but enough to make me uncomfortable about what it's doing to the controller's internals over time.

My understanding is the MPPT's output capacitors absorb some of it, but with a decent panel behind it (I'm running ~180W of Renogy mono on the cabin roof) there's a fair bit of energy looking for somewhere to go in that instant.

A few things I'm considering:

• Adding a small AGM buffer across the battery bus to act as a snubber
• Checking whether Victron's BMS assistant firmware logic helps here (though the 75/15 has limited assistant support)
• Properly balancing the cells so the BMS trips less frequently in the first place

The last point is probably the real fix honestly — the spikes are a symptom of an unbalanced pack more than a controller problem.

Has anyone dealt with this on a narrowboat or similar DC-only system where you can't just rely on an inverter/charger's smarter handshake with the BMS? Curious what solutions people have landed on.

@AshChild yeah this is a known gotcha with Victron MPPTs and LiFePO4 — when the BMS disconnects under charge, the controller suddenly loses its load and the output voltage spikes. Can damage the MPPT itself too if it's bad enough.

Few things worth checking:

• Enable the "BatteryLife" algorithm — helps but doesn't fully solve it
• Better fix is a Victron compatible BMS (e.g. JK or Daly with Victron comms) so it signals the MPPT to back off before disconnecting
• Or wire a small lead-acid buffer battery in parallel — absorbs the spike

Had this exact issue on my narrowboat build. Ended up going with a JK BMS with the Victron port, connects via VE.Direct and the MPPT throttles down gracefully. Night and day difference.

What BMS are you running currently?

So this is exactly why I ended up wiring a dummy load (just a small 12V bulb) across the battery terminals on my shepherd's hut setup — gives the MPPT somewhere to dump current when the BMS throws a strop.

Crude? Absolutely. Effective? Seems to be.

What I'm curious about though — has anyone actually quantified the overvoltage spike? Like, are we talking briefly kissing 16V or something properly alarming? Because I've seen conflicting numbers thrown around and I can't tell if my little 100W Renogy panel setup is at serious risk or whether I'm being paranoid.

Also @OakSpirit — does the Victron "BatteryLife" algorithm make any difference here, or is that completely irrelevant once the BMS has already disconnected? Always wondered if that feature was doing anything useful in my situation.

@Robbo that's an interesting workaround with the dummy load — does it create any meaningful parasitic drain day-to-day though?

On the boat I've been wondering whether fitting a Victron Battery Protect in the right place would help absorb that voltage spike, or whether it would just suffer the same fate as the MPPT itself when the BMS pulls the plug?

Also curious — is this specific to the 75/15 or does it affect the larger SmartSolar units too? I've got a second install planned for the tiny house using a 100/20 and want to know if I need to account for this in the design before I commit to a wiring layout.

Has anyone actually measured the spike voltage? Wondering how far above absorption setpoint it actually climbs in practice.

@RetiredChef2 the dummy load drain is pretty minimal in practice — I used a 1W resistor on my motorhome setup for a while, draws next to nothing.

That said, the cleaner fix IMO is using Victron's VE.Direct or setting a proper tail current so the MPPT knows when to back off before the BMS gets trigger-happy. Also worth checking your absorption voltage isn't creeping too close to the BMS cutoff threshold — that's what caused my issues.

Fogstar cells in particular seem to have a fairly tight top-end tolerance compared to some other LiFePO4 packs I've used.