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

The questions you have asked including why you don't understand. ... I can't be bothered with this nonsense, if you won't ask the manufacturers, who fully understand the issue, then I don't have the time and energy to continue this.
I have just sent a message to the manufacturer, and will let you know if I get a useful response.

As for whether the manufacturers actually do "understand the issue", you will realise that I have my doubts. about that - and no-one (not even you) have yet indicated anything that is incorrect about my reasoning).

It seems to be unqueried by everybody but you.
... which I find pretty strange. If my belief about what seems to be a pretty basic electric principle is incorrect, then I would hope that some of these 'everybody but me' people would say so, and explain why I'm incorrect. However, if my understanding of these 'pretty basic electrical principles is correct', I find it hard to believe that none of the electricians would also have that same understanding - in which case, again, I would hope that at least some of then would say so.

Some people seem to believe everything they read, without question.
 
Sponsored Links
.... I can't be bothered with this nonsense, if you won't ask the manufacturers, who fully understand the issue, then I don't have the time and energy to continue this. It seems to be unqueried by everybody but you.
One more try ... do I take it that, for whatever reason, you do not believe (and/or understand) the explanation I've given about the unimportance of the direction of current flow through the two sides of a residual current device - an understanding of which seems to indicate that what the people at FuseBox wrote to you is simply not correct?

If so, consider the below diagram of a ('bench') setup in which the supply line is connected to one end (the 'supply' end in my diagram) and the supply neutral is connected to the other end (the 'load' end in my diagram). I have attempted to show the path, hence direction, of the test current flow with arrows.

With that arrangement, when one presses the test button, the entire test current travels through both the L and N sides of the device, so that the current in the N side is therefore identical to that in the N side - but the current is travelling in the same direction through both sides of the device..

Do you believe that, despite what I have explained, the device will not trip when one presses the test button, because those two identical currents flowing through the RCD will 'cancel'?

1738123040249.png
 
That illustrates my very point - that such would surely make absolutely no sense, given that there is an intentional connection between neutral and earth at the origin of a TN-C-S installation ?

They therefore presumably must mean an N-E fault on the circuit downstream of the device. However, whilst that might result in an RCD tripping (if there were other loads on the CU), I can't see how it could stop the device failing to response to pressing the test button, can you?
I concur
 
The reason why the manufacture has came up with this because I think this can be an issue with RCD test buttons, if the neutral is linked to earth at the origin and a fault links N-E after the RCD, then there is a current path which bypasses the sense coil in the RCD. Very small or no current flows in the coil so no trip.
 
Sponsored Links
I will admit not read through all the replies, but I am not sure how to test between neutral and earth in my own house. Did not think about it until this thread, I would have, in the past, simply turned off the DNO isolator and tested between the earth and neutral. However, turning off the DNO isolator will result in the inverter connecting earth and neutral together. There are now an assortment of isolators, two built into the consumer units, so turning that off should allow all in the consumer unit to be tested, but also two DC isolators, and at least one built into the inverter.

I assume the link with a power failure between neutral and earth is only for the UPS supply, but I have never looked.
 
I will admit not read through all the replies, but I am not sure how to test between neutral and earth in my own house. Did not think about it until this thread, I would have, in the past, simply turned off the DNO isolator and tested between the earth and neutral.
Even most TT supplies will have continuity between N and E without the intention of being connected.

I find that test useless as you'll have something bonded/connected somewhere upstream.
 
This would cause the problem.

View attachment 371004
Just a picture from the internet.

Have Fusebox built their RCDs the wrong way round?
Are any made that way round?
Yes, that is exactly what they have done. I couldn't post the picture they added to the email yesterday.
 

Attachments

  • Fusebox RCD.jpg
    Fusebox RCD.jpg
    5.9 KB · Views: 10
Just a quick comment, yes I understand Johns confusion. as said if N & E have a "bolted fault" somewhere in the circuit then N current will share in some proportion between N & E conductors thereby thru the N coil. So the balance of 100% thru L and 100% thru N and 0% thru E will be disturbed and skewed accordingly.
It might be more helpful to think as of L and N as being +ve and -ve in this instance at this time because in a pure resistive circuit they act so at any one moment in time although changing polarity 50 or 60 times per second in succession. When we talk about N reversing its flow it tends to muddy the waters.
I don`t see much value in the difference between TNC- S being N joined to E very near the incomer to the installation or TNS being probably further away therefore more resistive (or impedive) to varying amounts of ohmic fractions.
an NE fault will cause an overbalance only if sufficient L - N results in something near the trip current (say around 15 to 30 mA of the RCD stated threshold) and with none or very little L flowing it would be unlikely to trip if the L is not passing current but could be likely if N & E are sharing something from another L circuit which actually is unenergized.
In short, regardless of AFDD or not an RCD could be skewed one way or another by NE link on the outgoing circuit and result in de-sensitisation or over-sensation of the RCD protected circuit we are considering and that could be the case either in operation of the test button or regardless of it.
So, at all times, we must strive to achieve near a full balance or thereabouts and reduce paths of imbalance as far as practical.
To suggest TNS is less likely to cause problems then TNC-S whilst might sometimes be a factor can be a bit of a Red Haddock.
 
This would cause the problem.
Indeed it would. I think you've probably 'cracked it'. Well done and thanks!

Just a picture from the internet.
Having just looked, there are indeed diagrams of both methods of wring the test button out there.However, I've only ever previously noticed those which introduce the 'test resistor' between outgoing L and incoming N - and that'show it's always been with every RCD/RCBO I've taken apart. However, the existence of the diagrams like the one you posted presumably mean that some manufactures do it that way.

I think that everything I've written and suggested above remains true in terms of those devices that are wired 'the other way' (in comparison with your diagram ( i.e. like all those I've ever taken apart.) - but, as you imply, that ceases to be the case with the 'other method'.

Have Fusebox built their RCDs the wrong way round?
I don't think there is realy a 'wrong and right', in the sense that, in the absence of any faults in the connected circuit, the test button will work as intended with either arrangement. I need to think further but, off the top of my head, the only difference I can think of is behaviour in the presence of an N-E fault in the connected circuit.

Are any made that way round?
Per what Bazza has now told us, it seems that FuseBox ones probably are. However, someone said earlier that a similar statement (about the effect of N-E faults) appears from some other manufacturers, so FuseBox are probably not alone.

I wonder if, rather than being 'wrong', FuseBox (and probably some others) are actually being 'clever'/thoughtful, in providing a test button which not only tests the device but which can also detect (quite possibly 'otherwis 'unknown') N-E faults in the connected circuit - thereby triggering the need to 'call an electrician' ?

KInd Regards, John
 
Yes, that is exactly what they have done. I couldn't post the picture they added to the email yesterday.
[ the site seems to be behaving fairly normally today, after all the pain of yesterday! ]

Thanks for clarifying. Everything noow makes sense, but we don't yet know how widely this 'unusual' way of wiring vthe test button extends beyond FuseBox.

As I've just written, it's just conceivable that those manufactures who do this do so deliberately so that the failure of the tripping when the button is pressed will alert the house owner to the fact that an electrician needs to investigate, even when the reason is an N-E fault on the connected circuit, rather than malfunction of the device itself.
 
In a sound circuit if the test current is 30mA then if, say, 9mA is returning via this leakage then the RCD will not trip even though the 9mA may be unavoidable acceptable leakage.
 
Just a quick comment, yes I understand Johns confusion. as said if N & E have a "bolted fault" somewhere in the circuit then N current will share in some proportion between N & E conductors thereby thru the N coil. So the balance of 100% thru L and 100% thru N and 0% thru E will be disturbed and skewed accordingly.
Indeed - but as I've said, the 'confusion' only arose because it now seems that the test buttons of some devices (including FuseBox ones) are wired differently from how I thought they always were (from outgoing L to incoming N), and what has been the case with every residual current device I've ever taken apart.

IF the test button is wired the 'other' (seemingly 'usual') way, then there is no confusion and I think that everything I wrote remains correct.

I think I'm going to ask FuseBox why they do it that way.

Kind Regards, John
 
Last edited:

DIYnot Local

Staff member

If you need to find a tradesperson to get your job done, please try our local search below, or if you are doing it yourself you can find suppliers local to you.

Select the supplier or trade you require, enter your location to begin your search.


Are you a trade or supplier? You can create your listing free at DIYnot Local

 
Sponsored Links
Back
Top