Shower Cable in wall

[hilme]

Thank you all for your comments, an amazing response! I will consider the insulation situation once I get the plasterboard off, but will likely upgrade the cable that runs to the shower.

The thing you refer to as the "shower pull switch" is actually a double pole switch in the airing cupboard outside the bathroom. The cable from there runs to a junction box (similar to this: link) in the loft, and from there down into the wall and to the shower. I would expect to replace the section between the shower and the junction box. Does that sound OK?

On the earthing/bonding/equipotential side:

There is a single RCD for the whole of the CU for all circuits in the house.

The bathroom also contains:
1. heated towel rail, taken from a fused spur run through the wall.
2. a non IP rated light fitting on the ceiling outside zone 2
3. a fuse box (the type that would normally be fixed to the wall) that is under the floor boards that runs some switched recessed lights in the kitchen ceiling below.
4. A radiator
5. A number of other central heating and mains cold and DHW 15mm pipes under the floor boards that run to other rooms, including those that run to the basin.
6. A shower.
7. A shower head riser bar - to mount the shower head on, plus a shower curtain rail.


Which of the above metal items should be bonded together / or to earth? As an aside I am also going to be running to the bath two new copper pipes from a pump outside the bathroom (driven from a different ring). It has plastic connectors, so the new pipes cannot be bonded to anywhere. That's not good is it?

Thanks again,

Hilme

 

[EFLImpudence]

On the earthing/bonding/equipotential side:

There is a single RCD for the whole of the CU for all circuits in the house.

If your main bonding is in place AND all circuit disconnection times are met then NO supplementary bonding is required.

The bathroom also contains:
1. heated towel rail, taken from a fused spur run through the wall.
2. a non IP rated light fitting on the ceiling outside zone 2
3. a fuse box (the type that would normally be fixed to the wall) that is under the floor boards that runs some switched recessed lights in the kitchen ceiling below.
4. A radiator
5. A number of other central heating and mains cold and DHW 15mm pipes under the floor boards that run to other rooms, including those that run to the basin.
6. A shower.
7. A shower head riser bar - to mount the shower head on, plus a shower curtain rail.


Which of the above metal items should be bonded together / or to earth? As an aside I am also going to be running to the bath two new copper pipes from a pump outside the bathroom (driven from a different ring). It has plastic connectors, so the new pipes cannot be bonded to anywhere. That's not good is it?

It is not a question of "yes,yes, yes,no, etc."
The need can only be determined by measurement of the resistance between simultaneously accessible parts.

 

[EFLImpudence]

You may feel it's a bit pedantic, but the reason I commented/challenged' is because you keep making this statement that the only parts which need bonding are those which are "already earthed".

I think it makes the bonding requirement clearer to people who don't understand.

My main point is that the concept of bonding does not relate only to things which 'are earthed', but to anything which has the theoretical capacity to acquire/'introduce' a potential, which is not necessarily earth potential.

Does that not refer to the same parts introducing a voltage because of a fault elsewhere?
Are you suggesting that isolated parts (with no connection to earth, either MET or ground) can sometimes introduce a potential?

Let's face it, if you could guarantee that the things in question were all 'already earthed', in the literal sense, then there would be no need for any bonding - since they would be guaranteed to always be at the same potential (earth potential).

But they aren't - because of impedance. That's the point.
Don't forget we are talking here about supplementary bonding.

Bonding is only necessary because of the possibility that one of the parts in question might acquire a potential other than earth potential (which could only happen if the connection to earth was, or became, 'imperfect'.

Not with you here.
Are you suggesting that bonding is only necessary in case an earth or bond becomes loose or corroded?

Don't forget that, even when defining what things need main bonding, the regs do not define an extraneous-c-p as a conductor which is "liable to introduce earth potential". Instead, they say "liable to introduce a potential, usually earth potential".

See above.

 

[Scoby_Beasley]

The thing you refer to as the "shower pull switch" is actually a double pole switch in the airing cupboard outside the bathroom. The cable from there runs to a junction box (similar to this: link) in the loft, and from there down into the wall and to the shower. I would expect to replace the section between the shower and the junction box. Does that sound OK?

Ignoring the earthing argument that is going on, and on...............

You'd be far better off running the new cable from the switch to the shower, without the junction box (think! every connection/joint is a potential problem).
While you are doing the job, you may as well do it correctly!!

 

[Aragorn84]

Not only that, but its a 30A JB, and thus is presumably already overloaded from the existing showers current requirements?

 

[JohnW2]

...you keep making this statement that the only parts which need bonding are those which are "already earthed".

I think it makes the bonding requirement clearer to people who don't understand.

Provided they understand what you mean by 'earthed', that could be true. I think it may be your other usual statement that one has to ascertain whether bonding is necessary by 'testing' that has the greater capacity to confuse them (see below).

...anything which has the theoretical capacity to acquire/'introduce' a potential, which is not necessarily earth potential.

Does that not refer to the same parts introducing a voltage because of a fault elsewhere?

Not necessarily ‘elsewhere’. The worst case scenario is for the fault to occur at the local ‘part’, since then potential difference (due to current in the CPC) between that part and the MET (hence most other CPCs) would be at its greatest. The only ‘elsewhere’ fault that could result in a (lesser) similar effect would be one whose fault current shared some of the same CPC path back to the MET. In any other ‘elsewhere’ situation, there would be no current in the CPC from the part in question, hence it would be at the same potential as the MET (hence nearly all other CPCs and exposed-c-ps).

Are you suggesting that isolated parts (with no connection to earth, either MET or ground) can sometimes introduce a potential? ... But they aren't - because of impedance. That's the point. Don't forget we are talking here about supplementary bonding.

I think that some of the potential confusion arises because of differences between the main bonding and supplementary bonding situations. With main bonding, the issue being addressed is that of an extraneous-c-p being at true earth potential, whereas the MET/CPCs may, under certain fault conditions, be above true earth potential. With supplementary bonding, it’s almost the opposite – the issue being addressed then is that, under certain fault conditions, the potential of an exposed-c-p be above MET potential (hence above the potential of most CPCs and exposed-c-ps in the installation).

Put another way, in the supplementary bonding situation, I would say that the theoretical situation which would most need bonding would be that in which (obviously incorrectly, and shouldn’t happen) an exposed-c-p was not ‘already earthed’. That’s another reason why I think that a blanket statement that the need for bonding can be ascertained by ‘testing for a low impedance path to earth’ might possibly confuse some people.

Kind Regards, John

 

[EFLImpudence]

Provided they understand what you mean by 'earthed', that could be true.

I think that is all most people understand by 'earthed'.

I think it may be your other usual statement that one has to ascertain whether bonding is necessary by 'testing' that has the greater capacity to confuse them (see below).

Well, it will but if it makes them realise bonding isn't applied because a metal part is just there.
Spoons do not need bonding but they would if they were earthed.
How else it is to be done if not by testing?

I think that some of the potential confusion arises because of differences between the main bonding and supplementary bonding situations.

Obviously but only if it is not understood.
So, if not understood, is that confusion?

With main bonding, the issue being addressed is that of an extraneous-c-p being at true earth potential, whereas the MET/CPCs may, under certain fault conditions, be above true earth potential.

Yes.

With supplementary bonding, it’s almost the opposite – the issue being addressed then is that, under certain fault conditions, the potential of an exposed-c-p be above MET potential (hence above the potential of most CPCs and exposed-c-ps in the installation).

Yes.

The issue is potential difference. It doesn't matter which way round it is. The same conductor will equalise it.

Put another way, in the supplementary bonding situation, I would say that the theoretical situation which would most need bonding would be that in which (obviously incorrectly, and shouldn’t happen) an exposed-c-p was not ‘already earthed’.

Now, you're doing a xxxxx (better not say) by thinking of a single occurrence, which is itself a fault, which could be the one time the safety measures can't work.
In this situation, though, the exposed c-p should not be unconnected and would be discovered by the first test for supplementary bonding.

In the case of the MP's daughter it would have been better had her cutlery hook been earthed and washing machine not but no one would advocate that for domestic premises.

While it is, though, an unearthed exposed c-p would be better not bonded but would also be discovered should anyone attempt to bond the CPC.
So my little rule still applies.

That was a silly example.

That’s another reason why I think that a blanket statement that the need for bonding can be ascertained by ‘testing for a low impedance path to earth’ might possibly confuse some people.

I've never said testing for a low impedance path to earth. I just say by testing.

That will obviously confuse people who don't know but, nevertheless is the only way to determine if supplementary bonding is required.

R = 50V / Ia will and does confuse people but it cannot be ignored when discussing supplementary bonding.

How should people be advised on bonding?

 

[JohnW2]

Provided they understand what you mean by 'earthed', that could be true.

I think that is all most people understand by 'earthed'.

I suspect that some would interpret ‘earthed’ as relating to an explicit bit of G/Y installed by an electrician or plumber, and might not think of a situation in which the pipework were ‘incidentally earthed’ - but maybe I'm wrong.

With main bonding, the issue being addressed is that of an extraneous-c-p being at true earth potential, whereas the MET/CPCs may, under certain fault conditions, be above true earth potential.

Yes.

With supplementary bonding, it’s almost the opposite – the issue being addressed then is that, under certain fault conditions, the potential of an exposed-c-p be above MET potential (hence above the potential of most CPCs and exposed-c-ps in the installation).

Yes. The issue is potential difference. It doesn't matter which way round it is. The same conductor will equalise it.

Indeed it will. However, in one case (main bonding) the need for bonding arises because something is 'already earthed' (connected to true earth), whereas in the other case (supplementary bonding) the need for bonding arises if the impedance of the path to earth is too high to keep the potential close enough to earth under certain fault conditions.

Now, you're doing a xxxxx (better not say) by thinking of ... That was a silly example.

I know it was a silly example, but I deliberately chose it to underline my point - that the conceptual need to equalise potentials is at its greatest for things which are not earthed, which is why I think some might be confused by being told that the only things which need supplementary bonding are those which are 'already earthed'.

I've never said testing for a low impedance path to earth. I just say by testing. That will obviously confuse people who don't know but, nevertheless is the only way to determine if supplementary bonding is required. R = 50V / Ia will and does confuse people but it cannot be ignored when discussing supplementary bonding.

I know you didn't say 'low impedance path to earth', but that (lower that the threshold you mention) is what you meant, isn't it?

How should people be advised on bonding?

It’s obviously almost moot, since it is probably going to become increasingly rare to have a situation in which the requirements for omitting supplementary bonding are not satisfied. However, if the requirement for bonding does apply, then I would have thought that the decision should be based on inspection and judgement, rather than ‘testing’.

I suppose my main concern about relying on testing to determine whether bonding is required is that such testing necessarily relates only to the situation at the time of testing, which cannot be guaranteed to persist. Consider the situation in which you had tested the metal pipework within a bathroom and found a very high resistance to earth. Do I take it that you would conclude that the pipework didn’t need to be bonded? However, that situation would only arise as a result of some interruption (probably by plastic pipework) in the metallic path to ‘earth’(usually ‘earthed’ or bonded pipework). It would be only too easy for future plumbing work, perhaps very remote from the bathroom, to change that situation and result in the bathroom pipework being ‘reconnected’ to earth. I have to say that if it were me, and I knew that there was mixed plastic and metal pipework in the house, then I would probably want to see that metal pipework in the bathroom bonded, even if at the time of testing it appeared to be ‘floating’, unless I could (somehow!) be absolutely certain that no future plumbing work could alter the situation.

Kind Regards, John

 

[hilme]

Thank you for help with the shower, I will run 10mm2 through the wall (without packing in insulation) and to the switch, provided I think the switch connections can take it!

Isn't earthing/bonding complex?!

Here is a window on my thoughts from what has been written...

Firstly some key terms:

Bonding = electrically connecting one object to another
Earthed = EITHER connected electrically to the earth provided at the CU directly (as is my rising main) OR by bonding it to another object that is earthed OR by the nature of the object it is naturally connected to "earth", be that via metallic pipework, or by being partially buried in the ground (metal conservatory frame??!).
CPC = the earth of twin and earth.

So here's what I have understood...

1. Not every metallic object needs bonding or earthing, floating potential is fine, unless it can be dragged up to another potential by something or someone touching it.

2. If an object could possibly end up at a potential that is different to some other conductor nearby then they should be bound together to prevent that potential difference.

That all seems to make sense, and suggest that there are times when bonding or earthing is not required and won't improve safety.

Are there any likely/real situations in which you would be less safe to bond any/all conductors to earth? Any additional connections will only reduce the impedance (is resistance is equivalent to impedance at low frequency?) to earth and so decrease tripping times.

I know that if you had something held at a high potential then you would be better of bonding nearby things to it and NOT to earth, but I would like to think that the high potential object had already been earthed and therefore the RCD (or even MCB) would have gone already.

So 2 questions remain:
1. Can I safely bond the water pipes (Hot, cold, and central Heating supply and return) together? And also to an earth from a ring that is used to run the electric towel rail? If so what size conductor should I use?

2. I am adding 2 new water pipes to the bathroom, one cold feed from the plastic header tank (via plastic bodied pump) which will not connect to (but may touch) any other metal part on its journey to the bath taps. The other pipe is coming from the existing hot water tank (via the same plastic bodied pump) also into the bath taps. Should they have any bonding / earthing?


Do I write too much?

Thanks again for your many replies.

Hilme

 

[Scoby_Beasley]

Have you a test/multimeter?

If, from your main earth terminal (usually in the CU) to structural metalwork (pipework) has a resistance of greater than 22Kohms then 'earthing' is not required. If two adjacent 'earthed' (and can be touched at the same time) components have a reading of less than 0.05ohms between them then they should be bonded together.

A mixer tap would join them. Obviously if we are talking water outlets! I'm not suggesting you put a mixer tap into an electric fire supply!!

Edited!!!

Added missing K

 

[JohnW2]

Isn't earthing/bonding complex?! Here is a window on my thoughts from what has been written... Firstly some key terms:
Bonding = electrically connecting one object to another

Yes, with the pupose of ensuring that there is no potential difference between them.

Earthed = EITHER connected electrically to the earth provided at the CU directly (as is my rising main) OR by bonding it to another object that is earthed OR by the nature of the object it is naturally connected to "earth", be that via metallic pipework, or by being partially buried in the ground (metal conservatory frame??!).

Yes, in the way in which EFLI has been using the term 'earthed', that is correct. In electrical terms, the purpose of 'earthing' something is to facilitate the satisfactory operation of a protective device (fuse, MCB, RCD) in the case of certian types of fault.

CPC = the earth of twin and earth.

That is one (the most common) form of CPC. A CPC ("Circuit Protective Conductor") serves to provide the 'earthing' ás decribed above.

So here's what I have understood... 1. Not every metallic object needs bonding or earthing, floating potential is fine, unless it can be dragged up to another potential by something or someone touching it.

Correct apart from the "unless...." bit. Any floating metal could be brought up to a high potential in the matter you describe, but still doesn't need bonding.

2. If an object could possibly end up at a potential that is different to some other conductor nearby then they should be bound together to prevent that potential difference.

As above, not if it is 'floating'.

That all seems to make sense, and suggest that there are times when bonding or earthing is not required and won't improve safety.

Indeed- and, as often discused, there are situations in which 'unnecessary' bonding can actually result in a deterioration of safety (see below).

Are there any likely/real situations in which you would be less safe to bond any/all conductors to earth?

Yes, as I've just mentioned ... if you were to simultaneously touch something live and a floating conductor, you would not get a shock (not via that route, anyway!). If, instead of being floating, that conductor were connected (directly or indirectly) to earth, then you would get a shock.

I know that if you had something held at a high potential then you would be better of bonding nearby things to it and NOT to earth, but I would like to think that the high potential object had already been earthed and therefore the RCD (or even MCB) would have gone already.

The classic example is of a vaccum cleaner with a frayed faulty cable being used in a bathroom. Touch that and also, say, a 'floating' metal bath, you get no shock. If the bath were connected to earth, you would get a shock.

So 2 questions remain:1. Can I safely bond the water pipes (Hot, cold, and central Heating supply and return) together?

That, in itself, would do no harm - although if you have metal plumbing, they weill all be 'well joined' already.

And also to an earth from a ring that is used to run the electric towel rail? If so what size conductor should I use?

The metal towel rail will be earthed via its CPC. If you join that to the pipes then, if they were 'floating' (very unlikely with metal plumbing), and hence not needing bonding, then, as above you would theoretically be creating a shock hazard which wouldn't otherwise exist.

2. I am adding 2 new water pipes to the bathroom, one cold feed from the plastic header tank (via plastic bodied pump) which will not connect to (but may touch) any other metal part on its journey to the bath taps. The other pipe is coming from the existing hot water tank (via the same plastic bodied pump) also into the bath taps. Should they have any bonding / earthing?

This is the one which EFLI may want to debate. However, don't forget that it's very probable that your bathroom will fulfill the requirements for not needing any supplementray bonding at all.

Do I write too much?

Probably - but I am the very last person who is in a position to criticise that

Kind Regards, John

 

[EFLImpudence]

Isn't earthing/bonding complex?!

Yes.

Bonding = electrically connecting one object to another

Not exactly.
It is joining parts which are already electrically connected so that the resistance between them is very low in order that there shall be no (or negligible) potential difference between them.

Earthed = EITHER connected electrically to the earth provided at the CU directly (as is my rising main)

So that the protective device shall operate should they become live.

OR by bonding it to another object that is earthed OR by the nature of the object it is naturally connected to "earth", be that via metallic pipework, or by being partially buried in the ground (metal conservatory frame??!).

Sort of - that is Main Bonding.

CPC = the earth of twin and earth.

Correct.

1. Not every metallic object needs bonding or earthing, floating potential is fine, unless it can be dragged up to another potential by something or someone touching it.

Yes but if that is possible it would not be floating - isolated.

2. If an object could possibly end up at a potential that is different to some other conductor nearby then they should be bound together to prevent that potential difference.

Correct - or another part which has become live.

That all seems to make sense, and suggest that there are times when bonding or earthing is not required and won't improve safety.

If wrongly applied it will introduce a hazard which otherwise would not be present.

Are there any likely/real situations in which you would be less safe to bond any/all conductors to earth? Any additional connections will only reduce the impedance (is resistance is equivalent to impedance at low frequency?) to earth and so decrease tripping times.

You are not bonding to earth. You are bonding parts together.
Impedance is the Alternating Current equivalent of Direct Current resistance.
When you say all conductors, you mean all CPCs.

I know that if you had something held at a high potential then you would be better of bonding nearby things to it and NOT to earth,

Depends here what you mean by earth.

but I would like to think that the high potential object had already been earthed and therefore the RCD (or even MCB) would have gone already.

That is earthing.

So 2 questions remain:
1. Can I safely bond the water pipes (Hot, cold, and central Heating supply and return) together? And also to an earth from a ring that is used to run the electric towel rail? If so what size conductor should I use?

2. I am adding 2 new water pipes to the bathroom, one cold feed from the plastic header tank (via plastic bodied pump) which will not connect to (but may touch) any other metal part on its journey to the bath taps. The other pipe is coming from the existing hot water tank (via the same plastic bodied pump) also into the bath taps. Should they have any bonding / earthing?

For example -

Firstly -
If the resistance is greater than 23,000Ω between parts and the Main Earthing Terminal they should not be bonded.

IF your shower has a 40A mcb then -
Measure the resistance between simultaneously accessible parts.
1) If the resistance is below 0.25Ω (50/(40x5)) they do not require bonding because they are deemed to be bonded well enough in themselves. (bonding anyway will not matter)

2) If the resistance is between 0.25Ω and 23,000Ω they should be bonded together and to the CPCs of all the circuits in the bathroom.

Do I write too much?

Not if you are interested.

 

[JohnW2]

If two adjacent 'earthed' (and can be touched at the same time) components have a reading of less than 0.05ohms between them then they should be bonded together.

Hmmm. You seem to be talking about two things that are already very effectively 'bonded'. How much lower than 0.05Ω do you thinking your bonding cable would make it?!

Kind Regards, John

 

[Scoby_Beasley]

Hmmm. You seem to be talking about two things that are already very effectively 'bonded'. How much lower than 0.05Ω do you thinking your bonding cable would make it?!

Kind Regards, John

TBH. something pulled out of GN3 which I now can't find.

 

[JohnW2]

Hmmm. You seem to be talking about two things that are already very effectively 'bonded'. How much lower than 0.05Ω do you thinking your bonding cable would make it?!

TBH. something pulled out of GN3 which I now can't find.

I can but repeat my "Hmmm" - it really makes absolutely no sense to me.

The guideline maximum for a main bonding conductor (end to end) is 0.05Ω - is that perhaps what you're thinking of?

Kind Regards, John