I would like to use AC-Out1 to supply three final circuits from the battery in the event of a power cut.
The following setup is in place:
Main meter cabinet: SLS Cs35 → 32A NEOZED fuse to the garage sub-distribution board (6mm²) → 25A 300mA RCD → 25A MCB and then the MultiPlus 2 5000 (“grid current limit” set to 25A).
The following solutions would be conceivable:
Run 4mm² directly from the MP2 back to the main meter cabinet and install a changeover switch (SFT240), 40A 30mA RCD and 3 MCBs (2x 16A and 1x 10A) there.
Or run a cable from the MP2 to the sub-distribution board (garage), install the changeover switch, RCD and MCBs there, and then run an 8G2.5mm² cable (or split differently) back to the cabinet from there…
The question is: is there anything that would favour solution 1 or 2?
Regarding solution 1: In my opinion, the 4mm² cable should be sufficient… especially since no high-power appliances are connected to the final circuits… or would there be any objections to this?
If the system is configured correctly, it can send a message when a power cut occurs. You could also run a UPS alongside it, which can send emails when events occur as well, just to be 100% sure.
In a bad grid region where the victron system is critical, having gone through rolling outages of 14 hours per day, weak grid supply resulting in grid rejection etc, everything is on AC out 1.
That is a standard deployment here and in many other regions with unreliable grids.
In the last 7 years there was one failure - input relays - and the system continued to work fine with grid disconnected. Twice in that time have very sensitive devices restarted on cutover.
It is robust, and reliable.
Its use in marine environments is testament to how reliable victron is, I would be amazed if any grid has better uptime than a multi or quattro.
Iirc the efficiency curve is worse for grid parallel loads vs output-based loads.
In case of emergencies a changeover switch is there for manual failover. This is only used for maintenance.
I did once install a small ups to power the POE and wifi gear, so comms and alerts for issues come in.
I don’t bother now. When VRM reports no comms or the last update time starts to climb beyond the GX restart interval I know there is an issue - usually just an internet outage.
Not that this matters as the GX keeps working and my automations use local LAN access, not relying on internet APIs.
Basically, you’re right—however, the likelihood of the MultiPlus-II failing isn’t particularly high either. And as far as I know, the system simply switches to “Pass-through” in the event of many faults (e.g. with the Cerbo or BMS). Or how is that handled?