Well the "input" is connected to the grid.
Is it? The instructions eric posted says:
"..... The Solar iBoost+ output must be the only supply to the immersion heaters ...."
... which I took to mean that the 'input' was only ever from the solar installation (probably DC), never from the grid. Did you interpret it differently?
The way I see it there are three plausible scenarios for how the output is set up.
1. The output is floating. No RCD will have any effect on faults with the output wiring.
Agreed. I've said that a number of times.
2. The output is independently referenced to earth. A RCD on the input side will not see faults on the output side.
True, but, if it were the case, then an RCD on the
output side would see downstream faults.
3. The output is referenced to the input, which in turn is referenced to earth at the substation. A fault on the output side will drive imbalance currents through any RCD supplying the input side.
As above, I read it as indicating that the supply never came from the grid, hence not connected to the substation. That does not, of course, preclude the possibility that the input (or output) might be deliberately connected to earth within the installation.
If the output were earth referenced, whether directly or indirectly, then, as per (2), an RCD on the
output side would see downstream faults. However, contrary to what you say, I don't think an RCD on the input side would see any 'output side'faults, would it (since the non-earthed side of the input would presumably be isolated from the output)?
I don't know which one this device is, but I strongly suspect 3.
As above, I'm not so sure about that, not the least because I'm far from convinced that 'the grid' is part of the equation. However, like you, I do not 'know'.
Kind Regards, John