Supplementary Bonding Revisited

[JohnW2]

In a recent thread …

It is whether it is an extraneous-c-p (or exposed-c-p) to the more stringent equipotential zone of the bathroom which has to be limited by supplementary bonding to 50V in the event of a fault.

I might be missing something as well as I'm not sure why you are asking this after all this time. … Have you discovered something which has made unnecessary what we have been doing for ever?

As I wrote briefly in that thread, it was only when I found myself writing in that thread that I realised that I need to try to get my head around some confusions and uncertainties about things which I had always ‘accepted’ without question - but I suspect I am just being dim or, at least, not thinking clearly enough! I’ve tried to think about this and express myself in some haste, so this may be a bit confused/confusing (as well as very lengthy) …..

As you (and I) often say, a common problem which many people seem to have in relation to some of the regulations relating to bonding (be it ‘main’ or ‘supplementary’ bonding) is that any and every regulation which talks about details of bonding only applies if there actually are some extraneous-c-ps to bond.
Assuming you still agree with that, I assume that you will agree that, then, as you know, in relation to bathrooms etc. 701.415.2 says …

Local supplementary protective equipotential bonding according to Regulation 415.2 shall be established connecting together the terminals of the protective conductor of each circuit supplying Class I and Class II equipment to the accessible extraneous-conductive-parts, within a room containing a bath or shower, including the following: ….

Per the above, that regulation presumably can only apply if there are any extraneous-c-ps (‘extraneous’ to the bathroom), since what it is ‘requiring’ would be impossible in their absence.

In order to decide whether any conductive parts entering a bathroom are ‘extraneous-c-ps’ the most obvious thing to do is to look to BS7671’s definition, which, as you know, is …

Extraneous-conductive-part. A conductive part liable to introduce a potential, generally Earth potential, and not forming part of the electrical installation.

As I see it, the only potential that one could credibly imagine, say, an internal pipe introducing into a room would be the potential of the MET - do you believe that ‘counts’ and qualifies it as being an ‘extraneous-c-p’ as far as the room is concerned? I suppose it must - otherwise it’s unlikely that anything would ever count as extraneous to the room (but not extraneous to the property)!

However, if I understand correctly, you don’t really use the BS7671 definition to decide whether something is an extraneous-c-p (and may therefore need bonding) but, rather you use the criteria of 415.2.2 as a “test” of whether something is ‘extraneous’ (and may hence need bonding).

If that is correct, then, having now reflected (maybe for the first time!) on this … the criteria of 415.2.2 provide a test of whether supplementary bonding, if present, is adequate (although it’s hard to see that any credible SB bonding conductor could fail to satisfy the criteria!), but I’m struggling to see how it can be of any value in determining whether (currently absent) SB is required (i.e. whether something qualifies as an extraneous-c-p).

Consider an L conductor within something in a bathroom coming into contact with an exposed-c-p of the item. The potential of that exposed-c-p will (prior to a protective device operating) rise to a proportion of the supply voltage (above MET potential), the magnitude of that pd depending on the respective CSA’s of the L and CPC of the supply cable - by my reckoning, with a 230V supply, roughly 142V for 2.5/2.5mm² cable, 139V for a 1.5/1.0mm² cable and 115V for a 1.0/1.0mm² cable - certainly all well above 50V.

If an extraneous-c-p is (as commonly will be the case) in electrical continuity with something bonded to the MET (or, at least, earthed to the MET), then there will be such a pd (142V, 139V or 115V - all well above 50V) between the faulty exposed-c-p and the extraneous-c-p - despite the fact that, when measured, it is very likely that the resistance between exposed- and extraneous-c-ps (or between extraneous-c-p and MET) would be well below the ‘pass’ criteria of 415.2.2. Hence, unless I’m missing something, such a ‘test’ is meaningless in the absence of SB, isn’t it?

In other words, I can’t see why (in the absence of SB) a low resistance between a conductive part which enters the room and an exposed-c-p (or the MET) can, in itself, provide any reassurance that SB is ‘not required’ (or that the part should not be considered to be an extraneous-c-p).

I am therefore more than a bit confused by this one of the criteria (in 701.415.2) for omission of SB in bathrooms being permissible …

(vi) All extraneous-conductive-parts of the location are effectively connected to the protective equipotential bonding according to Regulation 411.3.1.2.
NOTE: The effectiveness of the connection of extraneous-conductive-parts in the location to the main earthing terminal may be assessed, where necessary, by the application of Regulation 415.2.2.

… since, as above, I can’t see why (e.g. in the fault scenario I described) the situation would be any safer (in the absence of SB) even if the resistance form extraneous-c-p to MET were low enough to satisfy the 415.2.2 criteria (or, come to that if it were near zero!). In fact, if anything (albeit totally trivial), in the situation I described the ‘risk’ would tend to increase as that resistance reduced, because there would be less limitation of the current that could flow through a person, and the risk would decrease if the resistance were much higher than the 'maximum' required to satisfy 415.2.2, wouldn't it?! I’m confused!

I was going to ask whether I am I perhaps misunderstanding the way in which you use the criteria of 415.2.2. as a ‘test’, but it seems that I would have to ask the same of 701.415.2!

As above, I think that I must be missing a lot here, or just being dim!

Kind Regards, John

 

[ericmark]

The thing that comes to mind is the Emma Shaw case. With this case a fault was transmitted through the internal metalwork of the internal walls, and the idea of bonding is to stop this, had the water pipes and walls been the same polarity there would have been no problem. So there are a number of things which could have been done, including not bonding the water supply.

We did at one point go daft with bonding, metal window frames etc. But in the main what we want is when some thing like Emma Shaw happens is we are not found guilty, so there is a tendency to go OTT with bonding, it is a case of watch my back.

To me central heating is a real problem, we are running metal pipes room to room, which can transmit a fault room to room, and with all the boss white etc, we really have no idea if the pipes are bonded, soldered joints OK, but compression fittings we simply don't know.

 

[EFLImpudence]

First of all John, before trying to answer your post.

What is the actual physical contact that supplementary bonding is assumed to prevent getting a shock?

It is often said that SB is needed because there are wet bodies with wet feet in a bathroom.

However, surely the wet feet only matter with a hand contacting a live exposed-c-p and the feet on the floor, if earthed - it's not going to be hand to live exposed-c-p and foot on extraneous-c-p (standing on pipe) is it? So SB has no effect on this situation.

If it is one hand to live exposed-c-p and other hand on extraneous-c-p then the wet feet do not matter and it is no different than a kitchen where SB is not required.

 

[bernardgreen]

However, surely the wet feet only matter with a hand contacting a live exposed-c-p and the feet on the floor, if earthed - it's not going to be hand to live exposed-c-p and foot on extraneous-c-p (standing on pipe) is it? So SB has no effect on this situation.

It could be a hand touching an earthed pipe or tap in the shower and the water in the shower tray having some conductive path to an item that is not at Earth potential. A leaky shower tray and the ceiling rose in the room under the shower being one possible scenario.

 

[EFLImpudence]

As you (and I) often say, a common problem which many people seem to have in relation to some of the regulations relating to bonding (be it ‘main’ or ‘supplementary’ bonding) is that any and every regulation which talks about details of bonding only applies if there actually are some extraneous-c-ps to bond.
Assuming you still agree with that, I assume that you will agree that, then, as you know, in relation to bathrooms etc. 701.415.2 says … Per the above, that regulation presumably can only apply if there are any extraneous-c-ps (‘extraneous’ to the bathroom), since what it is ‘requiring’ would be impossible in their absence.

Yes, obviously if there are no such parts then the regulations do not apply.
However, if there are extraneous-c-ps but no exposed-c-ps then the regulations also do not apply as there is also nothing to which to bond them.

In order to decide whether any conductive parts entering a bathroom are ‘extraneous-c-ps’ the most obvious thing to do is to look to BS7671’s definition, which, as you know, is …

As I see it, the only potential that one could credibly imagine, say, an internal pipe introducing into a room would be the potential of the MET - do you believe that ‘counts’ and qualifies it as being an ‘extraneous-c-p’ as far as the room is concerned? I suppose it must - otherwise it’s unlikely that anything would ever count as extraneous to the room (but not extraneous to the property)!

However, if I understand correctly, you don’t really use the BS7671 definition to decide whether something is an extraneous-c-p (and may therefore need bonding) but, rather you use the criteria of 415.2.2 as a “test” of whether something is ‘extraneous’ (and may hence need bonding).

No, it is a test of whether its resistance to the MET indicates it is an extraneous-c-p.

If that is correct, then, having now reflected (maybe for the first time!) on this … the criteria of 415.2.2 provide a test of whether supplementary bonding, if present, is adequate (although it’s hard to see that any credible SB bonding conductor could fail to satisfy the criteria!), but I’m struggling to see how it can be of any value in determining whether (currently absent) SB is required (i.e. whether something qualifies as an extraneous-c-p).

Well, 415.2.2 does do that but the test is carried out to determine whether SB is required.

Consider an L conductor within something in a bathroom coming into contact with an exposed-c-p of the item. The potential of that exposed-c-p will (prior to a protective device operating) rise to a proportion of the supply voltage (above MET potential), the magnitude of that pd depending on the respective CSA’s of the L and CPC of the supply cable - by my reckoning, with a 230V supply, roughly 142V for 2.5/2.5mm² cable, 139V for a 1.5/1.0mm² cable and 115V for a 1.0/1.0mm² cable - certainly all well above 50V.

Ok.

If an extraneous-c-p is (as commonly will be the case) in electrical continuity with something bonded to the MET (or, at least, earthed to the MET), then there will be such a pd (142V, 139V or 115V - all well above 50V) between the faulty exposed-c-p and the extraneous-c-p - despite the fact that, when measured, it is very likely that the resistance between exposed- and extraneous-c-ps (or between extraneous-c-p and MET) would be well below the ‘pass’ criteria of 415.2.2. Hence, unless I’m missing something, such a ‘test’ is meaningless in the absence of SB, isn’t it?

I do not know what you mean by the test being meaningless because its results are satisfactory?
Why would the resistance always be well below the pass criteria?
A long CPC from metal light to MET, CPC or bond to pipe, then pipe to bathroom may easily be higher than the pass criteria.

In other words, I can’t see why (in the absence of SB) a low resistance between a conductive part which enters the room and an exposed-c-p (or the MET) can, in itself, provide any reassurance that SB is ‘not required’ (or that the part should not be considered to be an extraneous-c-p).

If the resistance between exposed-c-p and extraneous-c-p is within the pass criteria - because, for example, all services are on the other side of the bathroom wall - then SB will not required.
If a resistance is already negligible (or low enough) what is bonding the parts going to achieve?

I am therefore more than a bit confused by this one of the criteria (in 701.415.2) for omission of SB in bathrooms being permissible …

… since, as above, I can’t see why (e.g. in the fault scenario I described) the situation would be any safer (in the absence of SB) even if the resistance form extraneous-c-p to MET were low enough to satisfy the 415.2.2 criteria (or, come to that if it were near zero!). In fact, if anything (albeit totally trivial), in the situation I described the ‘risk’ would tend to increase as that resistance reduced, because there would be less limitation of the current that could flow through a person, and the risk would decrease if the resistance were much higher than the 'maximum' required to satisfy 415.2.2, wouldn't it?! I’m confused!

Surely it is the Volt-drop caused by the resistance that is important.
If there is no potential difference then there will be no current through the person - unless you consider the points in my previous post about earthed wet feet.

I was going to ask whether I am I perhaps misunderstanding the way in which you use the criteria of 415.2.2. as a ‘test’, but it seems that I would have to ask the same of 701.415.2!

Well, they are both the same. 701.415.2 states ...according to 415.2.2 - but then says a couple of things which are different which in my opinion should be ignored.

I think you seem to be using the test in 415.2.2 the wrong way round but I do not understand why you don't come to the same conclusion.

 

[EFLImpudence]

It could be a hand touching an earthed pipe or tap in the shower and the water in the shower tray having some conductive path to an item that is not at Earth potential. A leaky shower tray and the ceiling rose in the room under the shower being one possible scenario.

Yes, but that is a separate issue which could be avoided by inserting a length of plastic pipe to isolate the bathroom pipes - but then someone might intentionally earth those pipes in case a nearby cable decided to deform itself and attach its line conductor to the pipe.

 

[JohnW2]

First of all John, before trying to answer your post. What is the actual physical contact that supplementary bonding is assumed to prevent getting a shock?

A very good question. However, since SB (when if/required) seeks only to minimise PDs between exposed- and/or extraneous-c-ps, I suppose one has to assume that it is the scenario of simultaneous physical contact with two such things that SB seeks to address.

... It is often said that SB is needed because there are wet bodies with wet feet in a bathroom. .... However, surely the wet feet only matter with a hand contacting a live exposed-c-p and the feet on the floor, if earthed - it's not going to be hand to live exposed-c-p and foot on extraneous-c-p (standing on pipe) is it? So SB has no effect on this situation. .... If it is one hand to live exposed-c-p and other hand on extraneous-c-p then the wet feet do not matter and it is no different than a kitchen where SB is not required.

Exactly. Maybe the regulations are not intended to make sense?

Kind Regards, John

 

[bernardgreen]

Maybe the regulations are not intended to make sense?

YES they are not

 

[JohnW2]

No, it is a test of whether its resistance to the MET indicates it is an extraneous-c-p.

I don't understand your "No". How does what you have written differ from my statement (to which you are responding) which read ...

However, if I understand correctly, ... you use the criteria of 415.2.2 as a “test” of whether something is ‘extraneous’...

Well, 415.2.2 does do that but the test is carried out to determine whether SB is required.

I realise that's what you have always said but I am now observing that, having reflected upon this, (a) it appears to be intended as a test of whether SB (if present) is adequate and (b) if SB is not present, does not seem helpful in deciding whether SB may be required/helpful (as explained in the rest of my post, and below).

If the resistance between exposed-c-p and extraneous-c-p is within the pass criteria - because, for example, all services are on the other side of the bathroom wall - then SB will not required. If a resistance is already negligible (or low enough) what is bonding the parts going to achieve?

I thought I explained, at least in relation to the scenario I was considering. As I said, although the difference is theoretically trivial, in that scenario it is theoretically the case that the lower that resistance, the greater the theoretical 'need' for SB (to minimise the risk of electric shocks). So let me try again ...

Consider, in the absence of SB, there was an L-E fault in some Class I item (supplied by 2.5/1.5mm² cable) which, as I said, would seemingly cause (until the fault was cleared) the potential of the exposed-c-p of that item to rise to about 142V above MET potential. If one were to simultaneously touch that exposed-c-p and an extraneous-c-p which had a low resistance connection to the MET, then one would experience a roughly 142V shock (plenty high enough to be potentially fatal). The current through the person would be dependent upon the R1 of the circuit PLUS the resistance between the extraneous-c-p and the MET PLUS the resistance of the person's body. In terms of the total resistance of the 'shock path', the latter would swamp the other two but, at least in theory, the lower the resistance from extraneous-c-p to MET, the lower that total resistance, hence the higher the shock current through the person.

However, if SB were present, joining the exposed- and extraneous-c-ps, the PD between them would obviously be negligible, hence little/no shock

So, to answer your question as to what SB would achieve if there were a very low measured resistance between extraneous-c-p and MET ... as I see it (in the scenario I have postulated) it would 'achieve' a reduction in the voltage across the body of a victim from ~142V to something close to zero.

What is wrong with my reasoning?

I think you seem to be using the test in 415.2.2 the wrong way round but I do not understand why you don't come to the same conclusion.

I've said much the same of you ... i.e. that you appear to be using a test of whether SB (if present) is 'adequate' to determine whether SB is required (when it is not yet present).

Kind Regards, John

 

[JohnW2]

Yes, obviously if there are no such parts then the regulations do not apply. However, if there are extraneous-c-ps but no exposed-c-ps then the regulations also do not apply as there is also nothing to which to bond them.

I forgot to reply to this bit.

Yes, and I would think that exposed-c-ps in bathroom are pretty rare animals. Certainly in terms of my bathrooms, there are no exposed-c-ps other than one brass shaver socket in one of them and, thinking of other bathrooms I know, no exposed-c-ps come to mind (other than, perhaps, Class I light fittings on 'out-of-reach' ceilings).

So that's even less situations in which a requirement for SB could possibly apply.

Kind Regards, John

 

[EFLImpudence]

I don't understand your "No". How does what you have written differ from my statement (to which you are responding) which read ...

415.2.2 is nothing to do with determining whether the parts are extraneous-c-ps.
It is resistance to the MET - like for Main bonding on parts entering the premises if <23kΩ (or whatever you believe is the relevant figure).

I realise that's what you have always said but I am now observing that, having reflected upon this, (a) it appears to be intended as a test of whether SB (if present) is adequate and (b) if SB is not present, does not seem helpful in deciding whether SB may be required/helpful (as explained in the rest of my post, and below).

The R<50V/Ia is used on a new installation to determine if SB is required. On an existing installation a satisfactory measurement will indicate that either SB is not required or is already present and adequate.

I thought I explained, at least in relation to the scenario I was considering. As I said, although the difference is theoretically trivial, in that scenario it is theoretically the case that the lower that resistance, the greater the theoretical 'need' for SB (to minimise the risk of electric shocks). So let me try again ...

I don't think that is accurate as a negligible resistance will mean the Voltage is the same at both points (won't it?).
The greater the resistance the greater the Volt drop and potential difference so SB reduces it.

Consider, in the absence of SB, there was an L-E fault in some Class I item (supplied by 2.5/1.5mm² cable) which, as I said, would seemingly cause (until the fault was cleared) the potential of the exposed-c-p of that item to rise to about 142V above MET potential. If one were to simultaneously touch that exposed-c-p and an extraneous-c-p which had a low resistance connection to the MET, then one would experience a roughly 142V shock (plenty high enough to be potentially fatal).

Only if the person were earthed (by wet feet as I commented above); not between the two touched parts.

( The current through the person would be dependent upon the R1 of the circuit PLUS the resistance between the extraneous-c-p and the MET PLUS the resistance of the person's body. In terms of the total resistance of the 'shock path', the latter would swamp the other two but, at least in theory, the lower the resistance from extraneous-c-p to MET, the lower that total resistance, hence the higher the shock current through the person.

However, if SB were present, joining the exposed- and extraneous-c-ps, the PD between them would obviously be negligible, hence little/no shock

Again, it depends if you are talking about a shock to earth or just touching two points at the same potential.

So, to answer your question as to what SB would achieve if there were a very low measured resistance between extraneous-c-p and MET ... as I see it (in the scenario I have postulated) it would 'achieve' a reduction in the voltage across the body of a victim from ~142V to something close to zero.

You keep saying to the MET. Both parts are connected to the MET so it is the resistance between the two parts which matters.

What is wrong with my reasoning?

Only, I think, that you are thinking SB is to do with and prevent current through a person to earth rather than just through a person between two live parts during a fault.

My points about wet feet not being the consideration where SB is concerned - or have I got it all wrong?[/QUOTE]

 

[JohnW2]

I don't think that is accurate as a negligible resistance will mean the Voltage is the same at both points (won't it?).

I think this is what you're probably missing ...

What you say would only be true if there were no appreciable current flowing through any part of that 'negligible resistance', but is far from the case in the fault scenario I described.

The path ('of negligible resistance') we are talking about is (at least in the scenario I was describing) that from an extraneous-c-p via bonding and/or CPCs to the MET and hence via the CPC of the circuit in question to the exposed-c-p. There are thus essentially two components (in series) to that path - firstly from extraneous-c-to the MET and secondly from the MET to the exposed-c-p.

Considering the latter component (resistance of the CPC of the circuit in question), it's resistance may well, in some senses, be 'negligible' but, under the fault condition I described, there would be 'hundreds of amps' flowing through it - such that, as I said, if it were 2.5/1.5mm² cable, there would be about 142V 'voltage drop' across it - the exposed-c-p would therefore be at a potential of around 142V above MET potential.

In contrast, since no current was flowing through it, there would be no VD across the other component of the path resistance (that from extraneous-c-p to MET) - the extraneous-c-p would therefore be at MET potential.

Hence, under that fault scenario, the potential of the exposed-c-p would be ~142V above MET potential, whilst the extraneous-c-p would be at 0V above MET potential - hence a potential difference of ~142V between them (despite the 'negligible resistance' path between them).

I would hope that's clear, but I'll try to produce a diagram to illustrate the point (probably with some 'actual numbers').

Kind Regards, John

 

[EFLImpudence]

Right - - - so - - -

what then is the mechanism that (supposedly) limits touch Voltage to 50V and how does it work?

 

[JohnW2]

Right - - - so - - -

Does that mean that you understand and agree with what I wrote - namely that, in the absence of SB, a dangerous PD can exist between an extraneous-c-p and an exposed-c-p even if the measured resistance between the two is "negligible" (and that SB would essentially eliminate that risk)?

[ I'll be writing a little more about that, particularly the "negligible resistance" shortly ]

what then is the mechanism that (supposedly) limits touch Voltage to 50V and how does it work?

As I implied, having looked at that 'test' (415.2.2) for years without much question, it's only in the past couple of days that I have been moved to think properly about it, and having now done so, I find it difficult to make any sense of what it is trying to achieve.

As far as I can make out, if an installation 'passes' that test (as written), all that would seem to ensure is that if the maximum current that can flow without a protective device operating (e.g. IΔn for an RCD, 5 x In for a B-curve MCB) flows between the exposed-c-p and extraneous-c-p through the entire path whose resistance has been measured, then the maximum potential difference which can exist between those two parts is 50V.

So far, that's just a statement of Ohm's Law. However, what I've never thought about previously is the question of how on earth a situation could ever arise in which the maximum current allowed by protective devices (or, indeed, any appreciable current) could ever "flows between the exposed-c-p and extraneous-c-p through the entire path whose resistance has been measured". Am I missing something?

Turning to real-world situations which I can envisage arising, I hope I have demonstrated to your satisfaction that if an L-E fault arises in a Class I item (fed by 2.5/1.5 mm cable), then a PD of the order of 140V could theoretically come to exist between the exposed- and extraneous-c-ps no matter how low the measured resistance between the two.

So, unless there is a flaw in my reasoning, even if the installation 'passes' that test, in the absence of SB it's still possible that a dangerous 'touch voltage' (which I take to mean the PD between exposed- and extraneous-c-ps) can arise in the event of a fault. If so, what (if anything) does 'passing the test' achieve'?

If my argument is correct, then the only way to avoid dangerous PDs between exposed- and extraneous-c-ps would be to install SB (regardless of any measurements), wouldn't it? Whether one wants/needs to eliminate the risk of such PDs in the presence of RCD protection (which obviously is 'required' in bathrooms) is perhaps a different matter.

Any thoughts?

Kind Regards, John

 

[EFLImpudence]

I'm not sure any more, but -

how can there be a PD between exposed- and extraneous-c-ps if the resistance between them is negligible; either because of bonding or shortness of connections?