That sounds correct, the manufacturer instructions says dedicated breaker and that also makes the math sane wrt how much current can make it onto the circuit.
How does the AIO know to turn on the charger and absorb the extra solar? Or is the idea the extra solar only ever comes on when EVSE is on.
the AIO just stays on, grid-forming 240vac at the sub-panel, powered from solar day-time and 48vdc battery night time. The micros are on when the 48v battery pack is sufficiently high SOC, and obviously when sunny. The two parallel power source into the sub-panel are from the AIO & micros, the EVSE is a load.
OK. Regardless of what control scheme you use, there is nothing in this setup that ensures no power goes into the AIO. There isn’t an “AC diode” in the AIO to block power from flowing back in. An AC coupling capable AIO is engineered to tolerate this (presumably by setting up as a charger).
the micros are the only other power source, and they are connected only when an EV is plugged in with sufficient demand. The high SOC cut-off turns off the micros to make sure they never produce more than the EV is pulling. The low SOC threshold turns on the micros back on.
I don't believe those have a setting for how much AC coupling there is. They tell you to never AC couple something capable of outputting more than the charging current supported by the AC charger and battery. As such, once the SoC drops below the level where the battery is eligible for CC at that charge current, then it is safe to turn the relay back on.
There should not be any back-feed to AIO. The AIO simply diverts excess production (from the DC-PVs) to charge the battery, up to the float voltage set point. It pulls from battery as needed to meet load. My bench test consisted a 48vdc pack, an AIO without AC-IN (MPP), an M215, feeding the grid. There were 2-PV configuration, the DC-PV feeds the AIO PV-inputs. The other PV connected directly to the M215 then to 240vac grid. The purpose was to determine whether the AIO AC-OUTPUT was clean enough for the M215 to turn on and perform grid-tie feed. I tried with 3-different M215/M250 and was pleasantly suprised they all grid-tie fed. I could gen up a quick block diagram if that would help. Thank you so much for the discussion and comments, really don't like smoke
that is exactly the intention for the two trigger thresholds, but design/intention are usually not 100% fool-proof until actual fabrication.
quick & dirty block. A 900-w resistive heater was used as load. A clamp meter was used to measure current, and the readings were as expected. This bench test was done last year to confirm the MPP AIO can turn on a micro. Planning to repeat with recently ordered PowMr AIO soon.
the load of 900-w totally overwhelmed the micro production, which maxes out ~200-w. The AIO made up the deficit from its solar + battery. The AIO automatically adjusts the AC-output current. Please elaborate on the current switch ? didn't really use one.
the SoC logic: micros OFF @high Soc/ON @low SoC, is to ensure the load (EVSE) exceeds the micros production. The thinking is that the AIO battery is drawn to help the micros, the AIO constantly adjusts between charging/draining as a function of the output load, thus acting as a buffering.
www.victronenergy.com
thought to update, though sorta hi-jacked this thread. So got my cheapy AIO 240 (PowMr 5.5K48) on the bench, first thing was to remove the N-G screw thus allows split-phase generation via transformer. Been running various functional tests, so far so good. Was glad the AC-OUT can turn on an old M215. The test load is a 100-w incandescent bulb, with a router speed controller to vary the draw. The AIO can throttle battery consumption as expected with variable load. What was/is surprising is that the 48v battery gets charge from the AC-OUT line, with excess power produced by the M215 micro. Another surprise finding yesterday is the M215 doesn't need the Neutral to work.
