ZS for an motor run off an inverter

By-;pass the inverter - link it out, then, disconnect the motor leads from the motor and test at the end of the motor leads at the motor.

You could but it doesn't mean very much because it will never happen. The Zs to the inverter input terminals is the thing to test, that is as far as normal CPD operation can occur. If you short the output to Earth, or the motor becomes faulty, the electronics will limit the current to a safe level, and probably the CPD will be unaffected. Faults in the electronics could operate the CPD, but an Earth fault is unlikely and you have just measured that anyway. VSDs are really a kind of additional protection, which is fully electronic, and testing them is rather beyond normal MFTs. The Zs to the input is necessary to be OK for normal CPD operation. BTW they limit fault currents to protect themselves, and to some extent the Motor, but the motor does not have an overload or starter, as these functions pass to the VSD. All the settings are there in the manual, along with maximum speed, max current, minimum speed, start time, overload current, normal operating current, electronic braking, and usually a host of other stuff. There is nothing else to worry about!

Thank you. So the out going part of the circuits fault protection is inbuild in the inverter drive? What about soft starts as they are also electronic?

The Zs to the inverter input terminals is the thing to test

Agreed - but I'd perhaps add a continuity test for PE to the motor itself.

   - Andy.

You can test the Zs on the motor fed by a soft-start with the motor running, with your loop tester. A continuity test of the invertor or whatever motor Earth to the switchgear would be normal, as the flexible conduit can hide many sins! This will probably not work at all with an inverter supply and may well damage the tester. Supplementary bonding of fixed machinery is also a good idea, Earth wires seem to have a habit of failing in machinery power arrangements for some reason, probably because they do not show up for notice as power failure does.

In terms of the recommendation that the electronics will limit the current to a safe level, this may well be true, but caution is required.


BS 7671 contains the answer as to what to do ... and from previous experience, davezawadi might not like it (sorry David ?).


Where reliance on electronic equipment such as power converters to provide electric shock is necessary, this is covered by Regulation 419.2 of BS 7671, which, in addition to stating the voltage the output has to achieve in the relevant time given in Chapter 41 for protection against electric shock, states the following:



So, the answer is "Follow manufacturer's instructions". And if they don't provide any for this purpose, you can't rely on that method of protection against electric shock according to BS 7671.

So if the inverter cannot offer fault protection how can we tell this complies to bs7671 in regards to ZS?

The only other approach, without using RCD, is to demonstrate you've followed the inverter manufacturer's instructions for protective devices and installation conditions ... which is nearly the same thing.


In the absence of instructions, it's really not possible. Part of the selection process.

No, I don't really like it Graham, because the manufacturer does not give instructions on this, because it is not relevant. The output (motor) side of an inverter is only connected to the mains supply by electronics, there is no direct connection at all. This means that Zs at the motor has no meaning, but the CPD will provide the final circuit protection against overload or Earth fault if the electronics fail in some way.  The electronics provide all the protective functions necessary for a motor fault, usually including phase disconnection, current-limiting / disconnection on fault, overload, etc. Obviously the motor itself needs an Earth connection, but it has no direct effect on the disconnection time of the primary circuit. That phrase in BS7671 is a wonderful "catch-all" phrase, but it has no proper meaning. Inverters check the motor parameters on power-up automatically (they need more information than any electrician could ever measure) to work at all. If any parameter is out of range, not symmetric, etc. they will not power the motor.


As I said above, the only valid measurement which can be made is the Zs at the inverter input. If you think otherwise you are welcome to explain how and what instrumentation is to be used, what the limits are, and how you get a loop reading through the electronics, which essentially isolate the motor from the input voltage anyway. I would also like you to explain what the BS7671 phrase really means because it is easy to say but impossible to achieve. You and I both know there are a number of such phrases in the book, it is time they were removed by policy. Inverters are one of the most complex but common devices in wiring systems, particularly after the latest EU directives on motors. The lack of understanding shown in BS7671 is amazing and should be corrected with an explanation. You will be saying that a loop figure for class 2 power supply outputs is needed next!

Mr Jack

The inverter does provide fault protection as I explained above. I also do not understand Grahams comment on shock protection, as bonding and Earthing are still required in the usual way. It is true that there is no shock protection for contact with an inverter output and Earth at the same time, but this scenario should be impossible with correct installation. For RCD lovers, there is the possibility that one might work in the inverter output, but is so far outside the design specification that what happened on fault would be unknown. Nuisance tripping would be a major problem too, as the output waveforms do not necessaraly sum to zero at all times, and the fast edges at several kHz would have unknown effects. The average RCD tester would not work because the waveform is at zero for a proportion of the time in normal operation, and this would probably trigger the timing circuit as a trip.


It is also debatatable as to whether inverter wiring is covered by BS7671, as it is really an appliance, connected to the fixed wiring system. However, I am going to leave that problem alone!