Faulty rcd?

Marbles · Jul 8, 2012

Domestic installation, rcd tripped on supply to garage which then feeds a barn 25 metres away via swa. supply cable to garage is swa and tests out fine on insul res and continuity as does swa to barn. connect any load to con unit in barn even with no earth and rcd trips? Circuits in barn show healthy insul res readings? any ideas chaps?

Taylortwocities · Jul 10, 2012

adamh · Jul 11, 2012

Taylortwocities said:
That is NOT how you test an RCD.

The test button is a very basic functional test.

From the sound of it "someone" has done a board swop and mixed up the neutrals.
That "someone" also has not done the required tests on the existing and the altered installation.

if the test button wasn't functioning would you class this as a satisfactory condition for an rcd? i think not!!!

I was doing a test the other day on a council property changed main s/w for an 80A / 30mA RCD as soon as the starter on the 2d or 5ft fluorescent fitting tried to fire the rcd would trip. Ir tests showed no problem but on a ramp test the RCD wasnt tripping untill 150mA also the test button wouldn't work! I swapped the rcd still the same problem, with the rcd isolated from all final c/cts the test button would work! it turned out to be a parallel path for the neutral with there being 2 supplys to the flat , 1 being economy 7 and with there being only 1 terminal for neutral in the time switch for the immersion and night heating whoever fitted it just doubled them up. So I would say checking the test button might not be a bad place to start as I'm sure you would do this anyway as part of testing???

Taylortwocities · Jul 11, 2012

adamh said:
if the test button wasn't functioning would you class this as a satisfactory condition for an rcd? i think not!!!

Indeed it is not a satifactory condition. My response was as a result of the interchange relating to if the RCD had been tested and the response was

yeah test button is functioning

.

If I remember my training, the RCD is tested re ½, 1x and 5x the nominal op current as part of the live tests. These results go on the cert. The working of the test button is a functional test. and (again if i remember my training) the functional test on an installation is the very last thing that takes place after all the other dead and live tests have been carried out.

I agree, an RCD will be rejected if the test button does not work because that is the user's only belt and braces method of knowing that the RCD functionally works.