TESTING AN RCBO AT A LIGHT SWITCH

I suppose that 643.8 applies. I test R.C.D.s by plugging a tester into a 13 Amp socket outlets sometimes, so I can not think of a reason that you can't test an R.C.B.O. at a light switch as long as any loads that may be connected are disconnected that may affect the test results.


Z.

Hello Mike.


In my experience and to the best of my knowledge, testing an RCBO from the remote end of a connected circuit is in order.  One possible issue to be aware of is if there are any earth leakage paths in the connected circuit.  If the connected circuit is in good condition, the RCBO should trip as expected on the various tests.  If the RCBO tests produce unexpected results, then the problem is most likely to be some level of earth leakage  in the connected circuit.  The best course of action then is to disconnect the circuit at the RCBO and repeat the RCBO tests, which should then be ok.  The requirement will then be to investigate and repair the circuit defect.


Regards,


                  Colin Jenkins.

If the circuit has high earth leakage then the R.C.B.O. would be tripping off at regular intervals so would no doubt be reported pronto. Why not just test the R.C.B.O. with the Test button? 643.8.


Z.

If you are just replacing a broken light switch are you really going to dismantle the consumer unit for testing?


Andy

What sort of fault are you testing for? - if it is to be sure the light switch is earthed, then that is a a sensible thing to verify, especially if you are working on the lihgting circuit in that room, or there is a problem with it.

However, so long as there is CPC contiuity, and the RCD trips at some point, I'd not too get hung up  on a specific accurate answer - for example if the RCD still trips at In/2, well that may just mean there is already a lot of leakage present, quite possible with a large office full of electronically ballasted lamps, and the tester pushes it over the top.

Equally a large capacitive load may make the power seem to hold up longer than expected after tripping, so it may fail the 40ms test and appear  a few tens of msec slow.

To verify the RCD functions perfectly  on its own yes it needs to be isolated from such effects, but unless you have good reason to suspect it of being faulty, this is is just a lot of extra aggro for  something that may introduce more new problems than it solves.

Of course, if the RCD never trips, even at X2 or X5 or if there is no CPC to the fitting at all, then you have found something that will need looking at, and as part of that an RCD test is then worth it.

Thanks  mapj1.

No particular fault in mind.

It just struck me that periodic testing Zs at the light switch is very convenient.

So why not, for the extra minute or so, test the RCBO.

Some colleagues have been adamant that RCBOs must be tested at the D.B.

Regards,

Mike.