Défaillance interne / Production solaire à l'arrêt / Off Grid

Had a similar nightmare on the narrowboat last winter — woke up to zero solar production and the Multiplus sitting there doing absolutely nothing. Turned out to be an internal fault flag that had quietly killed the whole system overnight.

Setup at the time was:

• 4 × 175W panels (not massive, but enough for the boat)
• Multiplus 12/3000
• Fogstar Drift 200Ah LiFePO4
• Cerbo GX for monitoring

The Cerbo was showing a vague internal fault code but VRM wasn't giving me much to go on remotely. What actually fixed it was a full power-down — disconnected the battery, waited a proper 60 seconds (not the "count-to-ten" version), then brought everything back up in the right sequence. Panels last, obviously.

Worth checking:

1. Has the inverter thrown an internal fault in the history log? Pull it up in VictronConnect properly, not just the dashboard overview
2. Pylontech comms — if the BMS and Multiplus have lost their handshake, the inverter can just... stop. Check the VE.Bus to BMS communication is solid
3. Firmware — had a mate with a US3000C where a firmware mismatch between the Multiplus and the CCGX caused intermittent shutdowns

The Pylontech/Multiplus combo is generally rock solid but it's surprisingly sensitive to that BMS communication link. Anyone else seen this on a 48V system specifically? Curious whether it's more common at that voltage.

@MrBodge65 That error string looks like a Victron MPPT or Multiplus throwing a #67 or similar internal protection fault — the French strings are from the VEDirect/Venus firmware locale files.

Worth checking a few things before assuming hardware failure:

• VRM portal — does it log the exact fault code with timestamp?
• Venus OS version — there were known bugs in 2.xx builds causing spurious internal fault flags that cleared on reboot
• PV array isolation — a failed bypass diode in a panel can backfeed enough to confuse the MPPT into shutdown

On my boat I had the Cerbo GX log a phantom fault after a firmware update; a clean reboot of both the Cerbo and the Multiplus resolved it entirely. Hold the rocker switch on the Multiplus fully off for 30 seconds rather than a quick cycle.

What firmware version is running?

@MrBodge65 had exactly this on my shepherds hut setup last February — frozen morning, Multiplus 24/3000 just sat there with that same dead look. Turned out the internal temp sensor had flagged a protection fault and it basically locked itself out.

Worth checking VRM portal if you've got a Cerbo — the fault history usually tells you exactly what triggered it. Mine showed a #38 (input voltage out of range) which I'd never have spotted otherwise.

Also check your battery comms cable if you're running Victron lithium or Fogstar — a dropped BMS signal can cause the inverter to shut production as a safety measure. Reseating that cable sorted mine completely.

What got me in the van was assuming the fault was the inverter when it was actually the comms cable between the MPPT and the Cerbo that had worked loose — no communication meant the whole system went into a kind of paralysis. Victron's protection logic is conservative by design, which is usually a good thing until it isn't.

Worth pulling the VE.Direct cables and reseating them before you start digging into firmware or internal faults. I've seen @DefenderAdventure's #67 scenario too, but nine times out of ten on my setups it's been something embarrassingly physical rather than a genuine internal failure.

Check your ESS assistant settings as well if you're running one — a misconfigured grid setpoint can cause the Multiplus to just... sulk.

@AndyButler makes a good point about the comms cable — easy to overlook.

Had something similar at the cabin last winter. Before swapping anything out, I'd check whether VE.Direct or VE.Bus is actually passing data cleanly. A dodgy connection there can make the whole system look dead when it's not.

Also worth pulling up VictronConnect history logs — even if the display shows nothing, the MPPT often stores fault codes you can scroll back through. Saved me a wasted trip out once.

If the Multiplus has genuinely thrown an internal fault, a full power-down (including the DC side) for a few minutes sometimes clears it. Not always, but worth trying before assuming it's dead hardware.

@AndyButler good shout on the comms cable — caught me out too on the garden office build last year.

One thing nobody's mentioned: check your VE.Bus state in VE.Connect before assuming hardware failure. If it's showing "bulk" but production is zero, nine times out of ten it's a configuration issue rather than a physical fault.

Also worth pulling the event history from the MPPT — Victron logs the exact timestamp faults appeared, which can tell you if it's recurring overnight (temperature-related) vs a one-off.

If the Multiplus is flagging an internal fault specifically, try a full power cycle with the rotary switch (off → charger only → wait 30 seconds → back to inverter/charger). Clears phantom flags surprisingly often.

Fogstar batteries can also throw low-temp protection silently if your BMS isn't properly integrated — worth ruling out.