Tom

If the RCD is connected to the supply side of the transformer it will not trip if you carry out the test on the secondary side of the transformer.

Also if the transformer does not have a connection to earth on the secondary side (an isolating transformer) the RCD will not trip as there is no earth fault path.

You can test the function of the RCD by doing an "up and down test".  This requires the RCD tester to do a 2 wire RCD test, your Megger  MFT will do this.Se the meter up to do an RCD 2 wire test. Place one probe on the incoming neutral to the RCD and the other probe on the outgoing line press the test button and take the reading.

This test only tests the correction function of the RCD it does not simulate a fault to earth. Before carrying out an RCD test you should carry out a loop impedance test first to ensure there is an earth path. Modern test instruments like your Megger MFT will abort the RCD test if they detect a voltage rise on the earth and give you a warning.

Hope this is helpful?

Regards

JP

1) Setting aside for a moment wondering what on earth what the fault condition is that the RCD on a non earthed transformer winding is going to protect against, the test method that is most likely to work is a variation of the one used to test an RCD downstream of another, without tripping the first one. That is to use the tester between the outbound live (like normal) and the incoming neutral (instead of earth) . This 'Z'  test (because of the directions of current flow) is very similar to how the test button works if you look at a device with the little sketch on the side.

3)Some designs of compact RCD/RCBO use the outbound wires as the winding in the current transformer , and the core is hidden in the plastic terminal tunnel and the screw terminals are in effect putting the test current in on the wrong side.

Testing from the load end or by vampire probes into the insulation of the outbound cables (naughty but useful) will give you a reliable trip.

I'll have a think about 2

Mike.

Edit that was a cross post - but broadly in agreement with JP anyway.

perhaps the easiest way to test the RCD itself is OK on the non-standard transformer is to take it out and wire it into a mains lead to test it.

However, the self test may not be working if the pole to pole voltage is not enough to operate the internals - it sounds like it might be a 220V RCD on  a more or less 110V split phase circuit - in which case it is really the wrong sort of breaker for that application.

Mike

Hi JP, The RCD is on the secondary side of the transformer and I'm glad to hear someone else say that the RCD won't trip without an earth path as that seemed like common sense to me! In an installation like this is the RCD actually going to give you any protection at all?

Thanks for your help!

Hi Mike, thanks for this that's very helpful I can't see what protection you are getting from having and RCD in this circuit and I've asked the question before of manufacturers without getting much of a useful answer..... Thanks for that test method, it's not something that I was aware of but it is something that I will definitely use going forward. As for your answer for number 3 that is definitely news to me and explains an awful lot. I'll investigate the best method for testing them going forward!

Tom

Hi again Mike, hadn't even thought to check the operating voltage (remembers now what assumption is the mother of....) so I'll go back and investigate. Sounds like they need a new one! Thanks again

tommatwalker said:

1) I have recently come across a couple of installations with RCDs installed on systems that are on isolating transformers. Testing to earth does not trip them, testing between line and neutral doesn't trip them but their test button does trip them. Can anyone recommend a reliable method of testing them using the meter to get a trip time for them? In this scenario if the test button does not trip them is that a sign that the device itself has failed entirely or just a quirk of the installation?

RCD test method 2 (upstream/downstream method) on page 98 of 2022 Edition of IET Guidance Note 3 Inspection and Testing.

tommatwalker said:

2) I have recently attempted to test an RCD connected to a centre tap transformer (outside of the UK) which didn't function at all. The transformer is fed off a 200V Delta Delta supply and appears to be centre tapped and at the RCD I'm getting 60V to earth from the line and 48V to earth from the Neutral. Again I can't find a reliable way to test this but in this instance the test buttom doesn't work either. Is this a sign that this RCD has failed and should be replaced or am I not testing it right?

Need to check the RCD is specified for the particular circumstance (voltages present etc)

tommatwalker said:

3) Intermittently when testing RCDs with my meter I can't get the RCD to trip when testing at the terminals but it will trip when I go further down the line. Can somebody explain to me why this would be as I've never found any rhyme or reason behind it!

There could be a number of reasons for this, including, if you are using 'RCD test method 1" (return the fault down the PE or cpc) leakage currents affecting the test. Regardless of test, could also be due to stored energy ... but if there are high "leakage currents", then testing to PE you will quite possibly see the capacitances that cause the "leakage current" as stored energy.

For point 3. I am aware that some makes of rcd stipulate not to test on the terminals of the rcd as it will not work. Sorry but i don't remember the make or makes. Maybe next time it happens make a note of the make and model and ask the manufacturer or google some instructions.

Gary

aligarjon said:

For point 3. I am aware that some makes of rcd stipulate not to test on the terminals of the rcd as it will not work. Sorry but i don't remember the make or makes. Maybe next time it happens make a note of the make and model and ask the manufacturer or google some instructions.

That's very interesting, because 'at the terminals' is exactly where PD IEC/TR 62350 Guidance for the correct use of residual current-operated protective devices (RCDs) for household and similar use recommends the test is carried out (using, I would add, the upstream/downstream test method)! Clause 6.3 says:

The testing current should be applied between the upstream and downstream terminal of the RCD.

Perhaps they are pointing out that, following installation, a test to PE when carried out at the terminals might not operate as expected (or the test may not indicate as being completed properly with instruments to BS EN 61557-6 due to stored energy on the load side of the RCD)?

If you do recall, aligarjon, please let us know.

Thanks guys that's really helpful information. I'll start having a look at the manufacturers of RCD that cause this issue and the one's that don't