Most of us in North America have setup that conforms to rapid shut down requirement, where the grid tie inverter would communicate with shut down devices or optimizers and shut off the power pretty much right at the panel. This prevents our solar setup to provide backup power when the grid is down (arguable when the power is most valuable), it also makes it hard to just unhook the PV string from the grid tie inverter and plug it into another off-grid inverter to provide backup power.
I want to discuss a setup where we would use a Y split connector at the panel before the optimizer / shutdown devices to allow a subset of the panels to connect to a separate off-grid inverter in parallel. This off-grid inverter would only be connected with the grid is out and would be feed power into a manual transfer switch to the house. Don't worry about the transfer switch part for now, the focus here is whether the parallel MPPT connection would work.
I want to discuss a setup where we would use a Y split connector at the panel before the optimizer / shutdown devices to allow a subset of the panels to connect to a separate off-grid inverter in parallel. This off-grid inverter would only be connected with the grid is out and would be feed power into a manual transfer switch to the house. Don't worry about the transfer switch part for now, the focus here is whether the parallel MPPT connection would work.
- Is it safe to connect 2 MPPT to the same panel, or special provision would have to be made to ensure only one MPPT is connected at a time. I have tested that is is OK to connect a separate MPPT in parallel with the Solar Edge Optimizers when they are in shut down mode. However, what would happen when the grid comes back on and the Solar Edge optimizers turn on while the off-grid inverter's MPPT input is also active? I don't expect them to work nicely and "share the power", but I wonder if there would be damage (SE Optimizer, concept of operation).
- Is it OK to connect 2 (or 3) panels in series while each of them are connected to their own optimizers? See blue line in diagram. When the optimizers are off, they behave almost like an open circuit so I think that is OK (can be wrong though). But when the optimizers are on, would this series connect cause problems? If it is NOT OK to do this while the optimizers are on, the none would have to run the lines for each panel down to a place where a switch can be added in-line to only connect them in series when the optimizers are off.