main 'Trip' switch opening on change to night rate

[EFLImpudence]

The RCD is tripping because the current flowing through the Neutral terminal of the load has two parallel routes back to the Neutral terminal at the substation which is connected to the Ground.

Well, to be pedantic, the RCD is tripping because its Neutral is carrying a different amount of current (a nominal 30mA less) than the line.
It has no knowledge of the Earth and/or its pd to the Neutral.

That is all I have been saying.

 

[John D v2.0]

Come on guys, we're arguing semantics! I say it's the number of eggs in a small box, you say it's half a dozen

 

[JohnW2]

Come on guys, we're arguing semantics! I say it's the number of eggs in a small box, you say it's half a dozen

Hmmmm. Unlike so many similar discussions which, I agree, are purely semantic/terminological, I'm no so sure about this one.

EFLI does not seem to accept and/or understand that, in order to get an N-E current leak through a fault (which results in an L/N imbalance at the RCD, hence a trip), there simply has to be an N-E pd at the point of the fault. To my simple mind, that's not semantic, but, rather 'Electricity 101', so perhaps I'm missing something about what he's saying?

Kind Regards, John

 

[JohnW2]

Well, to be pedantic, the RCD is tripping because its Neutral is carrying a different amount of current (a nominal 30mA less) than the line.

Indeed, and I assume that we are all agreed about that.

It has no knowledge of the Earth and/or its pd to the Neutral.

It doesn't have such knowledge, and it doesn't need to. As you say, it only knows about the L/N current imbalance. However, (assuming the imbalance is due to an N-E fault) we do have an interest in the pd across the fault, since it that which causes the N-E current, hence the L/N imbalance, which is the only thing that the RCD 'sees'.

Kind Regards, John

 

[EFLImpudence]

It's nothing directly to do with potential differences

That's what I have been saying.

That's ALL I have been saying.

 

[JohnW2]

That's ALL I have been saying.

Maybe that was your intent, but you have been writing things like:

.... It is just the shared current through parallel paths which causes the RCD to trip even if there is NO pd between N and E.

In practice (i.e. forgetting about someone bypassing one side of the RCD with a bit of wire, which is 'cheating'), the only place for current to 'leak to' (from either N or L) is earth - and in the case of a leak from N (i.e. an N-E 'leak') if there is no N-E pd, there can be no no N-E leakage current. Hence, in practice, if an N-E (rather than L-E) fault causes an RCD to trip, then there must be an N-E pd.

Kind Regards, John

 

[EFLImpudence]

Ok - but my first reply was in response to this -

could be capacitance+resistance between N/E on an existing circuit, causing the extra load of the heaters to only just go over the tripping threshold when the neutral voltage increases relative to earth.

which I thought implied that it was the pd between N and E which causes the RCD to trip.

 

[John D v2.0]

If only I'd spent an extra 30 seconds re wording that message, I'd have saved hours of discussion!

 

[JohnW2]

Ok - but my first reply was in response to this - .... which I thought implied that it was the pd between N and E which causes the RCD to trip.

Ah! To me, he was merely 'implying' that it was the N-E pd which resulted in the N-E current - which, in turn, caused the L/N current imbalance in the RCD which caused it to trip.

Kind Regards, John

 

[JohnW2]

If only I'd spent an extra 30 seconds re wording that message, I'd have saved hours of discussion!

It does rather seem that way However, don't feel bad, since I, for one, understood exactly what you were saying/implying without the need for any re-wording!

Kind Regards, John

 

[EFLImpudence]

Ah! To me, he was merely 'implying' that it was the N-E pd which resulted in the N-E current - which, in turn, caused the L/N current imbalance in the RCD which caused it to trip.

Without labouring the point , you are implying the same.

It is not the N - E pd itself which causes the trip - it is the fault between N and E - as you, yourself, said "It's nothing directly to do with potential differences".

There may be a pd between N and E, as there is between L and E, but with no fault the RCD will not trip.



However, why can there not be parallel conductors sharing the current with no pd between them?

 

[John D v2.0]

The PD is between the point where the parallel path diverges to where it joins again. So between n and e was misleading for me to say, actually I was using it as a proxy for the difference between the point of fault and the potential at the substation where the neutral is earthed.
The path is long, so the PD can be noticeable, but in your simpler, shorter case, the PD between the two ends of the parallel path would be fractions of a volt.

To take it to an extreme, if the n conductor through the rcd was superconducting ie no resistance, then there would be no PD and therefore adding your link would not trip the RCD. If the link was also superconducting, not sure what would happen! Zero divided by zero?

 

[JohnW2]

Without labouring the point , you are implying the same. ... It is not the N - E pd itself which causes the trip - it is the fault between N and E - as you, yourself, said "It's nothing directly to do with potential differences".

I'm becoming more inclined to agree with JohnD (v2) that this is getting 'semantic' - or, at least, linguistic. We're all saying essentially the same thing, but you don't seem to like other people's wording!

There may be a pd between N and E, as there is between L and E, but with no fault the RCD will not trip.

Of course. However, this discussion has been entirely about N-E faults (actually N-CPC) - i.e. N-E (not L-E) and 'fault' (not 'normal').

However, why can there not be parallel conductors sharing the current with no pd between them?

We're not talking about the pd between the two 'parallel' conductors (N and CPC), which would obviously fall to a low value when they were connected via a fault (zero pd if the fault were of 'negligible impedance'), but, rather, the potential between the downstream end of the neutral and earth. Given that both the N and the CPC are ultimately connected to earth (one via the RCD and the other not), no current would flow through either of them unless there was a non-zero N-E potential at the downstream end.

The magnitude of the N-E potential at the downstream end of the neutral (8.6V in my example) is then an indicator of what current will flow through the CPC when an N-CPC fault arises - and the higher that N-E pd, the greater will be that bit of the total 'return' current that flows down the CPC (rather than the N conductor), unbalancing the currents through the RCD.

However, this rather complicated discussion does not tell us anything other than what we already knew - that the higher the current legitimately flowing through the load (and normally all through the neutral), the higher will also be the current that flows down the CPC (reducing the current going through the N-side of the RCD) in the presence of an N-E fault!

Kind Regards, John

 

[Diver Fred]

An Update to the problem.

By Thursday morning I had checked all the house wiring twice, including lifting numerous floorboards and correcting some errors inserted by a previous tenant (I know who did it as the present tenant professes not to be interested in DIY) but still could not find the cause of the problem. I then noticed that the pre-payment meter had been changed in the summer of this year. On contacting the suppliers I also found out that the person who changed the meter had reported that a feed wire from a Sealed splitter box was 'illegal'. Following that report a chap was sent out to 'correct' the wiring by removing the so-called illegal wire. This 10mm wire was routed behind the electric base board so couldn’t be easily traced.


The correction by-passed the neutral side of the trip to the Night-Storage heaters fuse box - which explains why the RCD was Tripping when the night rate was switched in.

When I spoke to the electric suppliers they said that the meter was faulty and they would arrange for a replacement - time scale 3-4 hours - in fact less than one hour. The technician
was the person who discovered the error and corrected it very quickly.

I'm now waiting for a reply to my letter to them asking for an explanation why the wire was thought to be illegal, why when it was removed the whole house system was checked and also for some recompence for the discomfort to my tennant and for my time in investigating the issue.

W