RCBO keeps tripping

[bernardgreen]

Of course it will - that's what RCDs do - but not if NO current is being used.

In theory a Neutral to GROUND fault could trip an RCD when no load current is flowing to a load between Live and Neutral. GROUND being ground and not the CPC. If the incoming neutral is not at ground potential then some current will flow from the incoming neutral, through the RCD and through the fault to ground. In most installations the CPC earth derived from the neutral will be a parallel path for the Neutral to GROUND cuurent but current sharing could easily result in 30 mA through the RCD.

 

[JohnW2]

In theory a Neutral to GROUND fault could trip an RCD when no load current is flowing to a load between Live and Neutral. GROUND being ground and not the CPC. If the incoming neutral is not at ground potential then some current will flow from the incoming neutral, through the RCD and through the fault to ground. In most installations the CPC earth derived from the neutral will be a parallel path for the Neutral to GROUND cuurent but current sharing could easily result in 30 mA through the RCD.

...particularly with a TT system, in which case a Neutral-to-CPC fault is, indeed, a true Neutral-to-earth one.

Kind Regards, John

 

[EFLImpudence]

.
Right, so...

It is possible with TN-S where the N-E impedance is very low

should read - "where the N-E impedance is relatively high".

In theory ...

Bernard: Why do you start with "In theory"?

 

[Spark123]

No, very low - I was talking about the supply characteristics.

 

[EFLImpudence]

No, very low - I was talking about the supply characteristics.

No. TNC-S is very low - i.e. N-E impedance = Zero (same conductor).

The situation in question arises in TNS because N-E impedance is higher than Zero - i.e. relatively high.

 

[Spark123]

You can have TN-S with relatively low N-E impedance, such as a farm building with its own transformer which isn't that uncommon up here.
Just offering an explaination to Crystal Ball who said he was looking at one that is TN-S.

A TN-CS N-E impedance is not zero wrt the consumer unit - the tails etc add a bit to it which is why you get the tripping when a relatively large appliance such as a shower is switched on with a fault on another circuit. Comparing a similar length of TN-S to a TN-CS then yes, the TN-S will have a higher N-E impedance.
Comparing a TN-S with another TN-S you can have high and low.

 

[EFLImpudence]

I think cross-purposes are in effect here.

My replies to you were with regard to an N-E fault tripping the RCD from stray voltage on the neutral from other installations, particularly because there could be a potential between N & E on TNS.


Crystal ball's post related to a 'normal' N-E fault tripping the RCD when different appliances were switched on.
This is the usual occurrence with a N-E fault and would be the same whichever method of earthing was employed.

 

[JohnW2]

I think cross-purposes are in effect here.

I'm sure you're right - this is, IMO, a seriously confused and confusing discussion, in that two totally different things are being discussed!

Kind Regards, John

 

[bernardgreen]

Bernard: Why do you start with "In theory"?

Because I haven't (yet) tested it in practise.

 

[Peet]

I once looked at a domestic installation for somebody where the immersion heater would sometimes trip out the RCD. Sometimes it would go days without doing it and then all of a sudden it would trip. The board was split into an RCD protected side and an unprotected side. The only thing I did to remedy the problem was to return the neutral of the circuit that the immersion was on to the correct neutral bar. Whoever had installed the wiring had taken the live off the MCB on the bank served by the RCD but connected the neutral to the non RCD side. After that the problem went.

BUT. This left me with one big question. How on earth did the RCD not trip with such a big imbalance on it? And why did it trip when it did? Similar to the OP it only tripped when a significant load was switched on.

Unfortunately I couldn't investigate it further.

Do RCDs become damaged and insensitive with previous high fault currents?

 

[EFLImpudence]

The RCD should have tripped when anything was switched on.
It could be faulty.

Do RCDs become damaged and insensitive with previous high fault currents?

Not with earth leakage faults nor should they become damaged with short circuit faults upto 6kA.

 

[Peet]

Because I haven't (yet) tested it in practise.

I have, and it does. I've seen it a couple of times where an RCD has tripped due to Neutral and Earth being accidentally touched together. The first time I saw it I deliberately setup the same 'fault' to verify it. Tiny differences in potential in very low impedance circuits lead to significant currents as I'm sure you'll know.

Something to be wary of if loss of RCD protected circuits is a problem and you're working on one that's turned off but uses single pole MCBs.

 

[Peet]

The RCD should have tripped when anything was switched on.

My thoughts exactly.

Have I come across (or read?) that sometimes high fault currents can cause a permanent imbalance in the device? The fault appearing as an 'offset' in its performance. So to the RCD it would appear that the circuit was permanetly leaking say 10mA with no load. Making it overly sensitive.

 

[EFLImpudence]

Have I come across (or read?) that sometimes high fault currents can cause a permanent imbalance in the device?

Not sure of the technicalities.
I would say it's just 'damaged'.
It could be tested.

The fault appearing as an 'offset' in its performance. So to the RCD it would appear that the circuit was permanetly leaking say 10mA with no load. Making it overly sensitive.

That would be the case but that is the opposite of the problem - not tripping when it should

 

[JohnW2]

Because I haven't (yet) tested it in practise.

I have, and it does. I've seen it a couple of times where an RCD has tripped due to Neutral and Earth being accidentally touched together.

I'm sure that we've all seen that (probably more times than we are willing to admit!) but it generally only happens if there is some load on the RCD in question. Some of the current from that load then flows through the N-E fault to earth, rather than through the N side of the RCD, hence creating an imbalance in the RCD which causes it to operate.

Bernard was talking about something different - a situation in which (althouigh he hasn't seen it happen) an N-E fault could theoretically result in an RCD operating even if there were no load on the RCD. As he described, this could arise if the supplier's neutral was at a significant potential above true earth and a fault between N and 'true earth' arose. That shouldn't happen with a TN-C-S supply, since (if proper main bonding is present) no 'earth' other than that derived from the supply Neutral should exist in the premises - but it could theoretically happen with other supply types. Have you seen that?

Kind Regards, John