testing RCD at Consumer unit

[ianphilip007]

hi guys

now i usually test RCD,s via the nearest plug socket as i feel it is a much safer and easier way to carry out a live test .

But just been reading GN3 and it states that they should be tested on the actual load side of the RCD with all loads disconnected. im assuming this means to test between the output terminals L/N of the RCD and the earth bar with all Mcb's off and main switch on. or have i got it wrong and the mcb's have to be on and circuits under no load???

and also i am correct in thinking only one test is needed for all circuits covered by the RCD not individual rcd test for each circuit
??

many thanks

 

[i=p/u]

:17:i think your right ian phillip

 

[telectrix]

he is right. main sw. and RCD on. MCBs off.

 

[ianphilip007]

is that your prefered method tel ?? instead of a local socket with all loads removed or switched off at the wall ??

 

[sparky-s-w]

yea the regs do state, to be done at the CCU. I questioned this myself as it is easy with the plug addapt of most multi funct testers, just to do a auto test. but having said that if your doing a full board with multiple circuits then its quicker to do one test (per RCD) at the board.
Also your readings might be slightly off testing remote from the board, due to other influences.

 

[telectrix]

i usually do it the easy way, all loads unplugged and auto test at a socket close to the CU. testing at the RCD may be recommended, but iy's too easy for a lead to slip contact while you're resetting the RCD. test aborted.

 

[ianphilip007]

yes i guess it might be quicker as your at the C.U. for both rcds and saves walking back to the C,U. to reset, just seems safer to use a socket outlet that is covered by each RCD than working on a live board especially with main switch on. but your right the regs do state the test should be done at the C.U.and any other influences should be omitted.

 

[ianphilip007]

i usually do it the easy way, all loads unplugged and auto test at a socket close to the CU. testing at the RCD may be recommended, but iy's too easy for a lead to slip contact while you're resetting the RCD. test aborted.

totally agree with that tel, main reason i use a local socket.

 

[Octopus]

I test at the board and at the "extreme" socket on socket circuits.

 

[telectrix]

trust you to be extreme. LOL.

 

[kingeri]

I understand that the RCD test is testing the functionality of the RCD itself, and not the circuit, but what Murdoch says makes sense to my small brain, ensuring the RCD trips within limits at the extremity of the circuit.

 

[topquark]

I understand that the RCD test is testing the functionality of the RCD itself, and not the circuit, but what Murdoch says makes sense to my small brain, ensuring the RCD trips within limits at the extremity of the circuit.

That's what your max Zs values are there for, to prove that the device will operate at the extremity of the circuit. Your proving that the device conforms to spec on disconnect times. Otherwise you'd have to test at the extremity of all circuits protected by the RCD.