RCD tripping on new cu

[WayneL]

For WayneL

1) Each cct should be discreet. The neutrals are commoned at the CU, by disconnecting a neutral from the board with the breakers/main sw. off that neutral should have no connection to any other. (Borrowed neutrals are not solely found on the lighting ccts) This is a simple test to find which ccts may be interconnected

Interesting reading - unfortunately completely untrue.
A 'borrowed neutral' is not a case of two circuit's neutrals being connected together............so your tests won't work.

A borrowed neutral is when a fitting/accesory is fed by the live of one circuit and utilises the neutral of another circuit.

So, (using lighting), if you feed the landing light with the downstairs lighting 'line' conductor (at the switch), and then run the upstairs lighting circuit neutral to the rose, then you have 'borrowed' the upstairs lighting circuit neutral.

This doesn't mean that the upstairs and downstairs circuit's neutrals are connected together - they aren't!......so the continuity test between neutral and neutral bars won't detect it.

The best way to detect a borrowed neutral is to do a R1 + Rn test on the suspect accessory - i.e. connect 'line' and 'neutral' together at the light fitting and test for continuity at the board between 'line' and your 'suspect' neutrals.

2) An RCD test takes some current and diverts it to the earth path, if that path is of a high resistance the RCD will not trip as insufficient current will flow. Further some RCDs will detect a high earth and refuse to set. The OP has stated that both RCDs are tripping under load and yet will not trip when tested, as both RCDs are acting in the same manner it would seem that an external influence is acting here.

I hope that has explained my 'random' comments.

RCDs do not work like this - they operate by detecting an imbalance between 'line' and 'neutral' - nothing to do with 'earths', 'CPCs' or the like.

An RCD test is done by your meter - that's why you connect your probes between 'line' and 'cpc' - the meter simulates a leak to earths and times the trip.

Some RCDs (particularly RCBOs) have earth flyleads - I was told these are used by the 'test-button' to do a similar thing that your meter does to allow tripping when pressed. (don't know how true this is) - but the 'flyleads' can't be anything to do with the actual operation of the RCD - otherwise they'd all have them......and they wouldn't all connect to the same 'earth bar' - they'd have to be split.

As for the 'main earth' (as you put it) or Zs, being to high.....a 30mA RCD will operate in 40 mS with a Zs of 1667 ohms - How high do you think the OPs Zs is???

 

[truckster]

My error. I used the term borrowed when I should have said interconnected

The fly lead on an RCD is used as a reference point to the neutral so that it will trip if the upsteam neutral is lost, coversely a high Ze can have the same effect, some also use it to detect dc spikes I don't know how but there you go

 

[WayneL]

The fly lead on an RCD is used as a reference point to the neutral so that it will trip if the upsteam neutral is lost, coversely a high Ze can have the same effect, some also use it to detect dc spikes I don't know how but there you go

Not saying your wrong on this one - just not sure what you mean.......I'll have to have a think about it.

 

[dim_bulb]

intresting reading

i take it you have disconnected everything and done IR tests betwen L - N, N-E, L-E, and not just L/N -E?

When you find the cause can you let us all know

thanks luke

 

[WayneL]

The fly lead on an RCD is used as a reference point to the neutral so that it will trip if the upsteam neutral is lost, coversely a high Ze can have the same effect, some also use it to detect dc spikes I don't know how but there you go

I've thought about this bit and I can't see how it would happen - if the upstream neutral is lost, in anything but a TN-C-S system, then all current flow would cease, so why would the RCD trip.......it's only the same as isolating the 'line' conductor......I don't see what the 'earth' connection would do in this situation.

If the upstream neutral was lost on a TN-C-S system, everything would function as normal - but anything earthed/bonded would rise in potential..........I suppose this could be detected by the flylead?

Not getting the idea really.....maybe someone else could expand on the subject......because I would like to know why some RCDs have flyleads and others don't?

 

[WayneL]

How did you get on today, Araman?

 

[Araman]

Thanks to all of you who contributed to this sorry little saga.
I found the fault eventually today which was a borrowed neutral as most of you suggested. there were also a few self inflicted wounds which I may not have discovered without your good advice.

Thank you to all, my customer is happy, which is the most important thing.