Why is my RCD tripping when doing a Zs test and the tester is on no trip

Might be worth adding some more information into the question.  Eg Type A or AC or B. Trip time 30mA or 100mA or 300mA or is it time delayed.  Make and model of tester could also be helpful.  Are all items disconnected from the circuit in question?

A Zs test, which measures earth fault loop impedance, can sometimes cause an RCD (Residual Current Device) to trip due to the relatively high test current used. This is because the test current, typically between 20A and 25A, can exceed the tripping threshold of some RCDs, particularly those with lower trip currents or older models

modern 'no trip' testers dont use a 20A test current, but there are 2 techncical approaches. The oldest 'D-lock' was one trade name for example, used a Dc circulating in the N-E loop to saturate the core of the RCD, so that the L-E AC would not trip it.

This largely does not work with modern  A type RCDs designed to response to pulsing DC, and certainly does not work with type B that respond to steady DC by design.

The other method is to use a test current lower than the 30mA that trips the RCD, say 10mA, but there are problems here. Firstly the voltage drop due to a low-ish Zs, say half an ohm, and a 10mA test current is very small in that example 5mV, and easily be lost on a 230V supply with lots of load switching etc, that may actually cause step voltage drops far more than that. And a few mV is the sort of voltage associated with dissimilar metals, damp and corrosion, and to a lesser extent with thermocouple effects - metaphorically it is looking for an small effect in the long grass.


So the instrument spends some seconds keying the test current on and off in a pseudo-random pattern and tries to correlate that pattern with observed steps and bumps in the mains voltage and correlating the L-N and L-E voltage changes to to deduce a value the changes it is creating and calculating a value for  loop resistance.
And even so,  in an installation with a bit of standing leakage, where the RCD is near to tripping anyway, then even a 10mA test current may still be enough to take it into tripping.

So a low current  test may not always be no-trip, and it may not be that accurate, especially for low Zs values. There is a lot to be said for looking at the voltage drop for a high current test, and if there is no RCD to worry about the results will be far less fruit machine like.

Mike.

Thanks 

Thanks