This is part of the same problem with do you 'PAT' equipment that is screwed to the wall, and when is fixed wiring part of an appliance.
The honest answer is that it is not always clear - the extremes are clear, we do not do an EICR inside a self-contained item like a washing machine, nor do we treat the light switch as a portable appliance. The grey areas arise with equipment that is spread out - in the home the central heating system is one, that is sort of fixed wiring, but the voltages and currents between thermostats tanks and boilers may be 'mains level logic' or DC, controls to pumps may be via a boiler that puts a triac in the neutral to switch it.
Problems also arise with things like stage lighting, where gear in racks may well plug in but is not really an appliance, and the mains flows though it to somewhere else via more fixed wiring.
I kind of agree that BS7671 should wind it's neck in from areas where it clearly makes no sense, but right now that is not how it is written (and I have a few comments for the DPC that point to that as well).
But there cannot be a hole where no-one knows who is responsible.
For now I suggest that the practical approach for the OP is a bit hybrid, and is to measure the Zs up to the point it stops being normal mains (presumably the input to the inverter). However verify R2 (earth continuity) all the way to the motor body (and if the inverter cable is screened this should be solidly earthed). And take a look at the inverter instructions and see that there is nothing else that should be done.
Take care that some inverter designs have large capacitors on the DC bus, and one of the many things that can happen in a fault is that the capacitor remains charged long after the input side ADS has operated. (by large in this context I mean that 1/2Cv2 is more than a couple of joules and the stored energy is therefore potentially lethal)
This is part of the same problem with do you 'PAT' equipment that is screwed to the wall, and when is fixed wiring part of an appliance.
The honest answer is that it is not always clear - the extremes are clear, we do not do an EICR inside a self-contained item like a washing machine, nor do we treat the light switch as a portable appliance. The grey areas arise with equipment that is spread out - in the home the central heating system is one, that is sort of fixed wiring, but the voltages and currents between thermostats tanks and boilers may be 'mains level logic' or DC, controls to pumps may be via a boiler that puts a triac in the neutral to switch it.
Problems also arise with things like stage lighting, where gear in racks may well plug in but is not really an appliance, and the mains flows though it to somewhere else via more fixed wiring.
I kind of agree that BS7671 should wind it's neck in from areas where it clearly makes no sense, but right now that is not how it is written (and I have a few comments for the DPC that point to that as well).
But there cannot be a hole where no-one knows who is responsible.
For now I suggest that the practical approach for the OP is a bit hybrid, and is to measure the Zs up to the point it stops being normal mains (presumably the input to the inverter). However verify R2 (earth continuity) all the way to the motor body (and if the inverter cable is screened this should be solidly earthed). And take a look at the inverter instructions and see that there is nothing else that should be done.
Take care that some inverter designs have large capacitors on the DC bus, and one of the many things that can happen in a fault is that the capacitor remains charged long after the input side ADS has operated. (by large in this context I mean that 1/2Cv2 is more than a couple of joules and the stored energy is therefore potentially lethal)
