RCD Failing 5x Test

[securespark]

Following on from my RCD fault thread, I'm starting a new one.

While fault-finding on Friday, I found the MEM RCD was failing the 5x test, so ordered a new one.

After curing the neutral short (effectively the socket which had the fault was fed via a live from the RCD side & a neutral from the incomer side), the 5x failure cleared up, and the unit sailed through the autotest with flying colours.

I think I still have a lot to learn about RCD's....

Anyone know why?

 

[Adam_151]

Can I answer a question with a question?

Whats the prupose of 514-12-02 ?

 

[Adam_151]

Anyhow, thought you said you'd check the neutrals?

Was it wired in singles or something and only live from immer. circuit?

 

[RF Lighting]

Whats the purpose of 514-12-02 ?

How many customers actually read it, never mind actually do as it says

 

[securespark]

As for checking neutrals, I did, but as I explained in the other thread, there was some sort of short between two neutrals connected to each bar.

When that was cleared, everything was fine.

 

[Spark123]

So maybe not all the trip current was passing through the RCD? Also maybe the RCD was a bit sticky and needed operated a few times to get it going freely again?

 

[bernardgreen]

I think I still have a lot to learn about RCD's....

Anyone know why?

I have tried to write this to suit both engineers and DIYers.

The basic idea of the RCD is that is trips when the current in the live conductor through the RCD is different from the current in the neutral conductor through the RCD. This can be either L>N ( live greater than neutral ) or N>L ( neutral > live )

[1] The most common fault is the live current is higher than the neutral current due to a live to to earth fault or person causing a current from live earth. This current does not return through the RCD on the neutral so un-balance occurs.

[2] A neutral to earth fault normally has no effect until an appliance on that feed is turned on. All the current to the appliance passes through the RCD on the live conductor but only some of the current passes back through the RCD on the neutral with the rest of going through the earth conductor back to the neutral at the neutral earth bond at the incomer.

Both these [1] and [2] un-balance the RCD in the same direction. Namely the live current through RCD is greater than the neutral currrent through the RCD.

The test button normally creates an increase in the live current.

[3] With a wiring error providing a feed with neutral from the RCD and live that is not through the RCD the current through an appliance will create an un-balance with the neutral current greater than the live current through the RCD.

It would not be impossible for a [3] to create an N>L unbalance that would cancel out the effect of the L>N un-balance(s) from a fault from [1] and/or [2].

The N>L un-balance from [3] would reduce the effect of the test L>N un-balance.

If the load wired as in [3] was 20 milliamps ( 5 watts ) and the test introduced a 30 milliamp "fault" the RCD would only see only 10 milliamp un-balance and would not trip for the test.

Bernard
Sharnbrook
UK

40 years electronics designer