"Earth to neutral fault" with TN-C-S supply ?

[JohnW2]

You would think so, but as the manufacturers have stated the potential issue in their instructions they must be aware of a set of conditions that would create the issue.

As you say, since they have written it, one has to presume that they are aware of some way in which an N-E fault can stop the test button working.

However, even if (presumed( true), that's presumably only really relevant before the circuit has first been energised and loaded - since any significant load on a circuit with a significant N-E fault ought to result in the residual-current part of the device operating, shouldn't it?

We can speculate from now until domesday, but the answer should come from them in minutes.

"in minutes" ? You must have had much happier experiences than I have had in getting useful answers out of manufacturers!

 

[Deleted member 142255]

However, even if (presumed( true), that's presumably only really relevant before the circuit has first been energised and loaded - since any significant load on a circuit with a significant N-E fault ought to result in the residual-current part of the device operating, shouldn't it?

You would think so, but without knowing the actual circumstances that cause the fault, you can't be sure. To assume makes an "ass" out of "u" and "me".

 

[JohnW2]

You would think so, but without knowing the actual circumstances that cause the fault, you can't be sure. To assume makes an "ass" out of "u" and "me".

I would think that the only uncertainties (hence 'assumptions') relate to the impedance of the N-E fault and the magnitude of any loads. If both are 'significant', then I suspect that there wouldn't be an opportunity to 'test the test button', since the device would probably already be 'tripped'!

In any event, I don't see any 'safety' issue. If pressing the test button (ever) fails to result in the device operating, that is an indication that investigation (or circuit and/or device) by an electrician is required.

As for 'investigating (testing) the device', since I don't know how these things work, I have no idea as to whether it is possible to test the AFD functionality, although I suspect that (like over-current functionality) it is effectively impossible to test in a domestic environment - meaning it is very probably only the residual-current functionality that can be tested?

 

[FrodoOne]

Maybe I need to think more deeply but, at first sight, I don't get that suggestion.

As per the diagram below, as I see it the test current through the L side of the RCD will be the same whether or not there is an N-E fault downstream of the RCD, and the current through the N side of the RCD (zero if there are no loads) will also be the same whether or not there is a downstream N-E fault - so the residual current created by pressing the test button should be the same regardless of whether or not there is a downstream N-E fault/ What, if anything, am I missing?
View attachment 370919

Well, NO.

You wrote
"the current through the N side of the RCD will also be the same (as the current in the Line)
whether or not there is a downstream N-E fault."

A "downstream N-E fault" will place the PE Conductor in parallel with the Neutral.

If (say) they are both the same CSA, the resistance of the two paths in Parallel will be the same.

Hence, the current in the Neutral will be half the current in the Line,
with the other half of the Line current travelling via the PE Conductor.

 

[JohnW2]

Well, NO. ... You wrote
"the current through the N side of the RCD will also be the same (as the current in the Line) whether or not there is a downstream N-E fault."

NO You have misinterpreted my words (which perhaps were a little ambiguous) and have made it look as if your interpretation correctly reflated my intention by adding "(as the current in the Line)" to my quoted message. Adding a (correct!) clarification in parenthesis (as you did) into my quote, it should have read:

"....the current through the N side of the RCD will also be the same (as it was before the test button not been pressed) whether or not there is a downstream N-E fault."

... The point being that when one presses the test button, an additional test currant flows through the L side of the device, but the current through the N side remains the same as it was before the test button was pressed - hence creating a current imbalance ('residual current) for the device to detect - and that remains the cases regardless of whether or not a downstream L-N fault is affecting the current through N side of RCD (both before and when the button was pressed)

 

[FrodoOne]

NO You have misinterpreted my words (which perhaps were a little ambiguous) and have made it look as if your interpretation correctly reflated my intention by adding "(as the current in the Line)" to my quoted message. Adding a (correct!) clarification in parenthesis (as you did) into my quote, it should have read:



... The point being that when one presses the test button, an additional test currant flows through the L side of the device, but the current through the N side remains the same as it was before the test button was pressed - hence creating a current imbalance ('residual current) for the device to detect - and that remains the cases regardless of whether or not a downstream L-N fault is affecting the current through N side of RCD (both before and when the button was pressed)

OK.

BUT
Consider https://www.diynot.com/diy/attachments/1738000557204-png.370919/
(From where did this come ?)

There is still a circuit from the Neutral to the PE Conductor
and
back to the Neutral/Earth connection point, under these test button conditions
This circuit is in parallel with the Test circuit - although its resistance may be much higher,
depending on how far "down the line" the N-E fault is located.

Although that diagram shows the current to the Neutral as only a set of Blue Dots - not travelling through the Neutral side of the RCD -
there will also be a (lesser [?]) current through the Neutral side of the RCD to the N-E fault and back to the Neutral/Earth connection point.


An RCD "expects" to see equal currents (in opposite directions) in normal operation.
It detects a "fault" when the currents are not equal.

With a N-E fault condition, the Test current in the Neutral path through the RCD will be less than the current in the Line path
but,
also, it will be in the same direction.

I really don't know how that will affect the RCDs detection.

However, if the Test Level is set for an imbalance of approaching 30 mA,
any current (however small) travelling in the Neutral side of the RCD could affect that detection.

 

[JohnW2]

OK. ... BUT
Consider https://www.diynot.com/diy/attachments/1738000557204-png.370919/
(From where did this come ?)

I drew it myself

There is still a circuit from the Neutral to the PE Conductor
and
back to the Neutral/Earth connection point, under these test button conditions
This circuit is in parallel with the Test circuit - although its resistance may be much higher, depending on how far "down the line" the N-E fault is located.
Although that diagram shows the current to the Neutral as only a set of Blue Dots - not travelling through the Neutral side of the RCD -
there will also be a (lesser [?]) current through the Neutral side of the RCD to the N-E fault and back to the Neutral/Earth connection point.

Yes, fair enough. I suppose i was subconsciously ignoring that because the test current path is going to be connected close (electrically) to the point where the L and PE are joined (in a TN-C-S installation), such that probably only a very small proportion of the test current would go ('backwards') through the RCD - but I accept that at least some current will take that route, even if the L-N fault is fairly distant. However ....

... An RCD "expects" to see equal currents (in opposite directions) in normal operation. It detects a "fault" when the currents are not equal.

Indeed

With a N-E fault condition, the Test current in the Neutral path through the RCD will be less than the current in the Line path

Yes, as above, I agree that it will always be at least slightly slightly less, and that I was ignoring that. However ...

.... but, also, it will be in the same direction.
I really don't know how that will affect the RCDs detection.

When the currents are travelling in the same direction, they will surely add (just as if one winds a conductor N times around the jaw of a clamp meter, it fives a reading of N times the current in the conductor), won't they? If so, any test current travelling ('backwards') through the N side of the RCD would actually increase the residual current due to the test button having been pressed, hence making it more, not less, likely that pressing the test button would cause the device to trip.?

However, if the Test Level is set for an imbalance of approaching 30 mA, any current (however small) travelling in the Neutral side of the RCD could affect that detection.

It would - BUT, as above, as I see it the 'affecting detection' would be such as to make detection less likely only if the two currents were travelling in opposite directions through the sense coil of the RCD (and thus 'cancelling'). If some test current is travelling 'backwards' through the N side of the RCD, then I think the 'effect' would be to make detection more likely.

Of course, this is all assuming that the device is just an RCD - but, as has been pointed out, things might be different when there is also an AFDD in the melting pot - I have no idea about that. Indeed, we don't even know for sure (although I suspect that such is the case) that the test button is only testing the residual-current functionality.

 

[Deleted member 142255]

@JohnW2 I did contact them, there response was:

Thank you for your query.

An earth - neutral fault would prevent the test button from working.

If there is a neutral to earth fault in a circuit, when the RCD test button is pressed the current that should flow from the test button, to outgoing neutral, though the RCD coil via the incoming neutral, can flow instead along the final circuit neutral through the fault and to earth bypassing the coil in the RCD.

If the relative resistances of the two possible paths are balanced there then may not be enough current to trip the RCD.

By removing load connections the device can be proved.

 

[JohnW2]

@JohnW2 I did contact them, there response was:

Thank you for your query.​

An earth - neutral fault would prevent the test button from working.​

If there is a neutral to earth fault in a circuit, when the RCD test button is pressed the current that should flow from the test button, to outgoing neutral, though the RCD coil via the incoming neutral, can flow instead along the final circuit neutral through the fault and to earth bypassing the coil in the RCD.​

If the relative resistances of the two possible paths are balanced there then may not be enough current to trip the RCD.​

By removing load connections the device can be proved.​

Thanks. That is the same as FrodoOne suggested. However, as I pointed out to him, if some of the test current flows 'backards' through the neutral side of the device, I would expect that to add to the test current through the L side, therefore increasing residual-current (due to test button being pressed) and hence increasing, rather than decreasing, the chances of the device tripping. Is that not the case?

One interesting thing about their response is that is seems to imply that I was right in my suspicion that the test button only tests residual-current functionality.

Their response also seems to imply that the issue they perceive is nothing to do with the fact that an AFDD is part of the equation, in which case I would expect them to make the same comment in relation to any residual-current device - but, as far as I can make out, neither they nor anyone else seem to do so in relation to RCCBs or RCBOs!

Kind Regards, John

 

[Deleted member 142255]

However, as I pointed out to him, if some of the test current flows 'backards' through the neutral side of the device, I would expect that to add to the test current through the L side, therefore increasing residual-current (due to test button being pressed) and hence increasing, rather than decreasing, the chances of the device tripping. Is that not the case?

No mate, that is wrong. You can't have more in the neutral than in the phase conductor. It is the reverse current that equals out the live current and prevents the device operating.

One interesting thing about their response is that is seems to imply that I was right in my suspicion that the test button only tests residual-current functionality.

Yes, that is correct as you can see if you examine your earlier diagram.

Their response also seems to imply that the issue they perceive is nothing to do with the fact that an AFDD is part of the equation, in which case I would expect them to make the same comment in relation to any residual-current device - but, as far as I can make out, neither they nor anyone else seem to do so in relation to RCCBs or RCBOs!

Kind Regards, John

It would certainly look that way, unless they are fitted with some electronics to prevent it.

Hope that has helped you mate.

 

[JohnW2]

No mate, that is wrong. You can't have more in the neutral than in the phase conductor.

Of course one can't - but I never suggested that one could. I have been talking about the 'residual current' as seen by the sense coil when certain currents were flowing through L and N sides of the device.

It is the reverse current that equals out the live current and prevents the device operating.

Exactly. As I see it, for example ....

If 50 mA is flowing through the L side and 50 mA is flowing 'in the reverse direction' through the N side, then the sense coil will see a residual current of zero (hence no trip).​

​

If 50 mA is flowing through the L side and no current is flowing 'in the reverse direction' through the N side, then the sense coil will see a residual current of 50 mA (hence certainly should trip).​

​

If 50 mA is flowing through the L side and 25 mA is flowing 'in the reverse direction' through the N side, then the sense coil will see a residual current of 25 mA (hence maybe no trip).​

​

If 50 mA is flowing through the L side and 25 mA is flowing 'in the same direction' (i.e. 'backwards' in comparison with the usual situation) through the N side, then the sense coil will see a residual current of 75 mA (hence certainly should trip)​

It is the last of those situations which we are talking about.

Yes, that is correct as you can see if you examine your earlier diagram.

One obviously can't conclude that it is 'correct' on the basis of my having drawn the diagram that way, since I drew it on the assumption (maybe incorrect!) that the test button was only testing residual-current functionality!

It would certainly look that way, unless they are fitted with some electronics to prevent it.

It would, so why have we only seen these statements in relation to AFDDs, given that RCCBs and RCBOs are dramatically more prevalent>

Hope that has helped you mate.

See above!

 

[Deleted member 142255]

I gave you the response from Fusebox. If that doesn't satisfy you then why not address your queries to them at [email protected]

I didn't design the flipping thing so I will not guess at reasons. As you can see, they respond quickly.

Sorry, my bad. I thought the drawing came from somebody that knew how the thing was designed.

 

[JohnW2]

I gave you the response from Fusebox.

You did - but, as I've explained, (to you and FrodoOne) I am struggling to see how their explanation can be correct.

If you don't believe the concepts that I described in my previous post, you would only need a bit of time, together with some loads, cables and a clamp meter to demonstrate them for yourself.

If that doesn't satisfy you then why not address your queries to them at [email protected]

As I said before, and although I've grateful to you for having taken the time/trouble to ask them, I really don't have (and probably never will have) significant interest in anything to do with AFDDs, and they are seemingly not making the same statement in relation to any other residual-current devices.

Sorry, my bad. I thought the drawing came from somebody that knew how the thing was designed.

Not at all. On the assumption (seeming implicitly confirmed by their response to you) that the test button was only testing residual current functionality, I simply drew a diagram that would relate to the test button in any residual current device.

 

[Deleted member 142255]

Basically what they have said is if the neutral current equals the line current when the test button is operated, for whatever reason, yje decice will not trip.

 

[JohnW2]

Basically what they have said is if the neutral current equals the line current when the test button is operated, for whatever reason, yje decice will not trip.

Indeed - but they do not seem to understand that such is not true if the neutral current is travelling in the same direction (i.e. backwards' in comparison with the normal situation) as the direction of the line current.

As I've said, just take two loads and run both of their feeding L conductors (or N conductors) 'the same way around' through the jaw of a clamp meter, and the meter will show the sum of the two currents, not the difference between them