Poll - Ring final circuits with high integrity earthing

[ban-all-sheds]

With respect, I think that what both you and BAS are doing is interpreting what you believe they are saying, or trying to say.

I believe that when they say "two individual protective conductors" they mean "two individual protective conductors".

For some reason you think they mean something else.

 

[ban-all-sheds]

You're merely restating the view that you have stated dozens of times.

Yes, because it hasn't sunk in for you yet.

IMO, the fact that 543.2.9 requires a ring final to have a ring CPC does not preclude that (one) ring CPC being regarded as two "individual protective conductors" (when viewed from any socket)

But it can't be viewed as two individual ring protective conductors, can it?

as far as 543.7.1.203(iii) is concerned.

And as far as 543.7.1.203(iii) is concerned, it has to be able to be viewed as two individual ring protective conductors.

You are assuming some link between those two regulations which just doesn't exist in terms of what the regulations actually say.

Am I?

Is there no link?

Is 543.2.9 not part of 543?

Does 543.7.1.203(iii) not actually say that each of the two individual CPCs must comply with Section 543?

FGS, John - how much longer do you think you can credibly continue to deny that if Reg A says a thing T must have a property P, and Reg B says that you must have two Ts and that each one must comply with Reg A that the overall requirement is not 2 Ts each having property P?

Mind you - I'm not sure why I'm bothering to ask you - you're very quick to complain when other people don't answer your questions, but I've lost count of the number I've asked you which you have totally ignored.

 

[JohnW2]

I'm very busy at the moment, so haven't got the time to plod through all your recent posts. However, I think this point is important in relation to my view ....

You keep talking about "paths to earth"...

I do - because I believe that is the primary concept of the whole of 543.7.

... and you believe that to go from standard to HI a radial has to have its number increased, i.e. making its number "high" when compared to standard

I do.

... but a ring does not have to have its number increased, and that the same number as standard can be called "high"?

As I said when I adapted your table, if a 'normal' ring final has 'separate terminals', I believe that it is already "HI" - such that no change/increase in anything is necessary to make it "HI". Although, if I'm right, they seem to have made a right hash of writing it, I believe that is what 543.7.2.201(i) is trying to say.

If I'm wrong, what exactly in the point of 543.7.2.201(i) ?

Kind Regards, John

 

[JohnW2]

The more I look at 543.7.2.201, the more confused I get as to the meaning of the requirement in its preamble for compliance with (by implication, all parts of) 543.7.1.

Consider 543.7.2.201(ii)(a) (a radial with its protective conductor "being connected as a ring"). Unless that protective conductor is ≥10mm² (or ≥4mm² and mechanically protected), the only way that it could comply with 543.7.1.203, would be via (iii) - and that would only be possible if a single protective conductor "connected as a ring" were considered to be acceptable as "two individual protective conductors" - a suggestion that was strongly contested in the early parts of this discussion. The emphasis of the argument for ring finals has now somewhat changed, to focus mainly on 543.2.9 - but that only becomes relevant because of the reference to "complying with (all of) 543" in 543.7.1.203(iii).

I know it doesn't help in relation the the question of what the regs "actually say", but it seems to me that 543.7.2.201 probably didn't intend to require compliance with 543.7.1.203, and that 543.7.1.203(iii) didn't intend to require compliance with 543.2.9.

Kind Regards, John

 

[SimonH2]

Would you quote 543.7.1.203(iii) and highlight where it says "circuit" in "circuit protective conductor"

543.7.1.203 The wiring of every final circuit and distribution circuit intended to supply one or more items of equipment, such that the total protective conductor current is likely to exceeed 10mA, shall have a high integrity protective connection complying with one or more of the following:-

(i)...

(ii)...

(iii) Two individual protective conductors, ...

Are you going to claim that if it says a circuit must have a protective conductor, that that is not a circuit protective conductor?

Yes, I will claim that. "protective conductor (PE)" is not the same as "circuit protective conductor (cpc)". There are two different entries in the definitions - or hadn't you noticed that ?
I cannot think of a situation where a cpc is not also a PE, but I can think of situations where a PE is not necessarily a cpc.

For an analogy, consider the "debates" about earthing vs bonding. Where the regs require equipotential bonding but not directly earthing. The fact that in most cases it's going to be hard or impossible to apply equipotential bonding without also earthing the bonded component doesn't turn "bonding" into "earthing".

Is it an insult?
Do you really believe that just because you can write what looks like a simple question there has to be a simple answer?

Go on - answer the simple question "When did you stop beating your wife?"

Ah, the old diversionary tactic again. You know really well that the question I asked did have a simple answer - just that you didn't want to give it. So you try and divert attention away from that by changing the topic. You do it a lot.

Perhaps you could show where I've stated (or even implied) that every simple question must have a simple answer ?

You do realise that I was saying "no, of course not" to your question "So you agree that....", and not "no, of course 543.7.1.203(iii) does not in any way require either of those conductors to be a ring"?

So you are still saying that the words "Two individual protective conductors, each complying with the requirements of Section 543" includes the word "ring" or something with the same meaning ?

If you cannot highlight the word "ring" in that text then regulation 543.7.1.203(iii) does not require the PEs to be rings. There may be another eg which does - but it's not 543.7.1.203(iii)

Do you agree that if for HIE for a radial circuit one has to double the normal number of paths from each socket to earth then it is common sense, logical etc that for HIE for a ring circuit one has to double the normal number of paths from each socket to earth?

No.
What is logical is that if you have a circuit topology which has non-redundant connections, then something needs to be added to apply redundancy - ie a second path. If the circuit topology is such that redundancy is already present, then further redundancy is not necessary.
Taking the specific example, would you compare the PE redundancy of a radial where the end earth connection has been extended and taken back to the MET thus closing a ring, and that of an RFC (with one cpc) ? Assuming both use the same termination methods - ie using separate terminals in both accessories and MET - is there any difference in either topology or redundancy between the resulting PE "rings" ?

Also consider the specific statement that (with some conditions relating to circuit protection) that where two radial circuits are present, it is acceptable to link the earth terminations of the end points together so that the CP forms a ring. That is most definitely not "doubling up" the number of CPCs - it brings the earthing arrangements of the two radials directly into alignment with what most of us believe to be a sensible approach to HI earthing on an RFC.

So we have two radials, lets assume wired in T&E, and by choice or design we arrange for the last devices to be in close proximity so we can easily just drop a bit of earth wire between them.
Do we now agree that both radials (subject to separate terminations etc) now meet the requirements for HI earthing ? And that this has a single ring of protective conductors from the MET, via all the accessories, and back to the MET ?

 

[SimonH2]

And a bit more for BAS ...

Lets take our industry, IT. Suppose you are sat in a meeting with various people reviewing power supplies for a number of small installations.

At installation A you observe that while you have a modular UPS, it only has one power conversion module installed - so if this fails you will lose power to the supported equipment. You all agree that adding a single extra module, thus doubling the number of modules, will give you redundancy in the event of a power module failure.

At installation B you observe that there is a larger load, and this is supported (without redundancy) by a similar UPS which has 3 power modules installed. What do you agree is the required action to introduce redundancy in the event of a single power module failure ?
Do you :
a) Apply the logic that because you had to double the number of modules in A, you must double the number of modules in B - and thus install an extra 3 modules for a total of 6.
or b) Apply the logic that adding only 1 module (for a total of 4) will give you redundancy and therefore adding the extra 2 is not required ?

So no, it's not logical to say that because A means doubling something, then to achieve the same result with B means doubling something.

 

[ban-all-sheds]

As I said when I adapted your table, if a 'normal' ring final has 'separate terminals', I believe that it is already "HI"

If so then you can tell us which of (i) to (v) in 543.7.1.203 it complies with.

Which one is it?

such that no change/increase in anything is necessary to make it "HI". Although, if I'm right, they seem to have made a right hash of writing it

Or, what about....

... you are not right and they have not made a hash of it?

I believe that is what 543.7.2.201(i) is trying to say.

First, please notice that 543.7.2.201 says that the high integrity protective connection has to comply with 543.7.1.

Which means that it has to comply with 543.7.1.203.

Which means that 543.7.2.201 says that the high integrity protective connection has to be one of

• A single circuit protective conductor having a cross-sectional area of not less than 4mm² complying with the requirements of Regulations 543.2 and 543.3
• A single copper circuit protective conductor having a cross-sectional area of not less than 10mm² complying with the requirements of Regulations 543.2 and 543.3, the circuit protective conductor being enclosed to provide additional protection against mechanical damage, for example, within a flexible conduit
• Two individual circuit protective conductors each complying with the requirements of Section 543. The two circuit protective conductors may be of different types. e.g. a metal conduit together with an additional conductor of a cable enclosed in the same conduit
• etc

If I'm wrong, what exactly in the point of 543.7.2.201(i) ?

To allow a ring final to be wired with a fat, or chubby+protection cpc as per 543.7.1.203(i) or (ii). If you were using singles in conduit, you could use 2.5mm² for L&N, each as a ring, and 4mm² for the CPC, also as a ring.

If you have 1.5mm² for your CPC then you drop down through the list in 543.7.2.201 until you get to (iii) and you drop down the list in 543.7.1.203 until you get to (iii). And you install two individual cpcs, each one of which complies with Section 543, i.e. each one of which is a ring, as required by 543.2.9.

 

[JohnW2]

I believe that is what 543.7.2.201(i) is trying to say.

First, please notice that 543.7.2.201 says that the high integrity protective connection has to comply with 543.7.1. ... Which means that it has to comply with 543.7.1.203. ... Which means that 543.7.2.201 says that the high integrity protective connection has to be one of ...

Yes, that's what is says. However, what I said was that I believe 543.7.2.201(i) is trying (unsuccessfuly) to say something different.

If I'm wrong, what exactly in the point of 543.7.2.201(i) ?

To allow a ring final to be wired with a fat, or chubby+protection cpc as per 543.7.1.203(i) or (ii).

That's not really an answer to my question, because that would still be true even if none of 543.7.2 existed. So I have to ask again, what exactly in the point of 543.7.2.201(i)? To my mind, it is clearly trying (but failing) to give special consideration to a ring final with a single cpc ring which wouldn't exist if 543.7.2 didn't exist.

Kind Regards, John

 

[ban-all-sheds]

Consider 543.7.2.201(ii)(a) (a radial with its protective conductor "being connected as a ring"). Unless that protective conductor is ≥10mm² (or ≥4mm² and mechanically protected),

Why should it not be? It's an option, there are other ways to compliance.

the only way that it could comply with 543.7.1.203, would be via (iii) - and that would only be possible if a single protective conductor "connected as a ring" were considered to be acceptable as "two individual protective conductors" - a suggestion that was strongly contested in the early parts of this discussion.

True, if you choose not to take the option of 543.7.1.203 (i) or (ii) you are going to have to choose between the remaining options, 543.7.1.203 (iii), (iv) or (v). And by choosing (iii) you have chosen not to use one of the acceptable arrangements described in 543.7.2.201(ii)(a). Not the end of the world - you still have other options you may choose from.

The emphasis of the argument for ring finals has now somewhat changed, to focus mainly on 543.2.9 - but that only becomes relevant because of the reference to "complying with (all of) 543" in 543.7.1.203(iii).

543.2.9 does not apply to radials, so you don't have to comply with it for a radial.

It does not apply to radials where you need HIE and your chosen method for that is to make the CPC a ring. But if you do want do use a ring then it will have to be either a 543.7.1.203 (i) or (ii) one.

Or, of course, you could use two individual CPCs for your radial circuit, as per 543.7.1.203(iii), and they would not have to be ≥10mm², they would not have to be ≥4mm² and mechanically protected, and they would not have to be rings - they would simply have to be two individual CPCs, each one just like the one CPC you would have for the same radial circuit if you did not need HIE.

I know it doesn't help in relation the the question of what the regs "actually say", but it seems to me that 543.7.2.201 probably didn't intend to require compliance with 543.7.1.203, and that 543.7.1.203(iii) didn't intend to require compliance with 543.2.9.

You only think it doesn't help, in fact you only create the issue of it not helping, when you decide that the regulations should not be read as written because they don't mean what they say and don't say what they mean.

Basically your position seems to be "I don't think that the regulations mean what they actually say, and therefore I have problems in determining what they actually mean."


I can suggest a very simple, rational, logical and common-sensible way for you to resolve those problems

 

[JohnW2]

Consider 543.7.2.201(ii)(a) (a radial with its protective conductor "being connected as a ring"). Unless that protective conductor is ≥10mm² (or ≥4mm² and mechanically protected),

Why should it not be? It's an option, there are other ways to compliance.

Indeed, but you were only able to make that comment because you chopped the quote in the middle of my sentence, which went on to say "...the only way that it could comply with 543.7.1.203, would be via (iii)". I say that because, in most situations (certainly domestic ones), (iv) and (iv) are not realistic options - so, "unless one uses (i) or (ii), it essentially has to be (iii) - which, as you go on to say, you feel cannot be satisfied by a single CPC ring.

So that brings me back to asking what 543.7.2.201(ii)(a) is all about. Given that you don't feel it can comply with (iii) and given that (iv) and (v) will not usually be realistic options, do you really believe that the radial with a single CPC ring it relates to is only acceptable if wired in 10mm² (or protected 4mm²) cable? What is daft about that is that a radial without it's CPC arranged as a ring would seemingly be compliant if it's CPC were 10mm² (or protected 4mm²) - so what is 543.7.2.201(ii)(a) (or, come to that, any of 543.7.2) usefully saying?

Basically your position seems to be "I don't think that the regulations mean what they actually say, and therefore I have problems in determining what they actually mean."

I suppose that's not far off a statement of my position. For example, as above, since none of 543.7.2 appears to really achieve anything (that wouldn't be achieved without it, since it is all prefaced by a requirement to comply with 543.7.1), I certainly "have problems in determining what they actually mean" - even though I have my own ideas about what it may have been trying to say/achieve.

Kind Regards, John

 

[EFLImpudence]

Yes, I will claim that. "protective conductor (PE)" is not the same as "circuit protective conductor (cpc)".

The definition of "protective conductor" encompasses: a CPC, the Earthing conductor and Bonding conductors.

There are two different entries in the definitions - or hadn't you noticed that ?

Yes, there are more than two.
There are definitions for:
A Protective conductor, a Circuit protective conductor, Earthing conductor and Bonding conductor.
The Protective conductor must meet the requirements of one of the other three depending on which of the three it is.

I cannot think of a situation where a cpc is not also a PE, but I can think of situations where a PE is not necessarily a cpc.

Exactly. So, if it does not meet the requirements for a cpc then it is inadequate for that purpose and not compliant.

For an analogy, consider the "debates" about earthing vs bonding. Where the regs require equipotential bonding but not directly earthing.

There shouldn't be 'debates' merely explanations for those who do not understand.
They will never require both. The purposes are different.

The fact that in most cases it's going to be hard or impossible to apply equipotential bonding without also earthing the bonded component doesn't turn "bonding" into "earthing".

I would say ALL cases but you will not be 'also earthing' because unless a conductive part IS already earthed, i.e. may be an extraneous-conductive-part, then it will not require bonding.
The bonding conductor does not, however, become a cpc because it will not be part of an electrical circuit.

So you are still saying that the words "Two individual protective conductors, each complying with the requirements of Section 543" includes the word "ring" or something with the same meaning ?

It does not but it refers to "each complying with the requirements of Section 543" which does.

If you cannot highlight the word "ring" in that text then regulation 543.7.1.203(iii) does not require the PEs to be rings. There may be another eg which does - but it's not 543.7.1.203(iii)

So, you are disregarding "each complying with the requirements of Section 543" which does.



The rest of your post is logical but that is not what the regulations say.
As ring circuits are a special case with special regulations, perhaps this is another.

 

[JohnW2]

Much as I don't want to play down the importance of "what the regulations actually say", could we perhaps put that issue aside for a moment and talk about our personal opinions, based on knowledge of the electrical issues, as to what we personally believe would be reasonable as providing "high integrity earthing".

As for my personal view, I believe that the most crucial aspect of increasing the 'integrity' of earthing is to have redundancy of independent paths to earth from each point in the circuit (including separate terminals for the multiple paths). To have just two paths is probably normally adequate (although more than two obviously would increase 'integrity' even more) - and I believe that two parallel paths can be provided by two parallel cpcs in a radial circuit, or by having a single cpc arranged as a ring (for either ring or radial circuits).

I am personally far less happy with relying on 'high CSA cpcs', alone, to provide HIE, since they do not provide any redundancy, and I feel sure that disconnection at terminations is a far greater risk than mechanical damage along the length of a cpc (particularly when it is part of a multicore cable). I would therefore personally not be comfortable to rely on something like 543.7.1.203(i) or (ii) to provide "HIE", which I do not personally regard as 'safe' (when HIE is required).

Regulations aside, does anyone have the same 'personal view' as me?

Kind Regards, John

 

[ban-all-sheds]

Yes, I will claim that. "protective conductor (PE)" is not the same as "circuit protective conductor (cpc)". There are two different entries in the definitions - or hadn't you noticed that ?

Yes, I have noticed that.

I have not noticed, I must admit, 543.7.1.203 using the term "protective conductor (PE)". Is this because I've missed something which is there, or because you are seeing things which are not?

But I will tell you what I have noticed, and that is written English, which therefore construes as written English.

I have also noticed ".. every circuit .. shall have a .. protective conductor ..". So that makes it a circuit protective conductor.

cf:

".. every table .. shall have 4 legs ..". That will make them table legs.

".. roofs .. shall be covered with tiles ..". That will make them roof tiles.

".. drains .. shall be fitted with covers ..". That will make them drain covers.

Do you see? Or do you now want to start arguing against reading the regulations as if they were written in English?

Ah, the old diversionary tactic again.

Not so.

You know really well that the question I asked did have a simple answer

No - I know really well that it did not. I was unsure if you knew that, or if you were even familiar with the concept that what can be written as a simple question may not have a simple answer.

- just that you didn't want to give it. So you try and divert attention away from that by changing the topic. You do it a lot.

No, I was not diverting or changing the topic.

I was trying to show you that what can be written as a simple question may not have a simple answer.

Perhaps you could show where I've stated (or even implied) that every simple question must have a simple answer ?

You asked a question which you said had a simple, "yes or no" answer.

It did not.

So at the very least you were not recognising that not all "simple" questions have "simple" answers.

I'll again ask you a simple question:

"Simon, when did you stop beating your wife?". And I'll tell you that it is simple, and that it has a simple answer, which is the date you stopped.

Do you get it now?

So you are still saying that the words "Two individual protective conductors, each complying with the requirements of Section 543" includes the word "ring" or something with the same meaning ?

Yes - something with the same meaning, which is the stated requirement to comply with the requirements of Section 543 which means a stated requirement to comply with the requirements of 543.2.9.

If you cannot highlight the word "ring" in that text then regulation 543.7.1.203(iii) does not require the PEs to be rings. There may be another eg which does - but it's not 543.7.1.203(iii)

You are being utterly ridiculous.

543.7.1.203(iii) requires compliance with Section 543.

Section 543 contains 543.2.9.

Therefore 543.7.1.203(iii) requires compliance with 543.2.9.

Therefore 543.7.1.203(iii) requires that each of the two individual cpcs of HIE ring circuits be rings themselves.

That's the way the regulations work.

Do you seriously expect me, or anybody else, to believe that you really think that every single requirement for {thing T} has to be explicitly repeated every time a requirement for T is prescribed, and that unless this is done none of the others apply?

Do you really think that if Regulation A1 says T must have property P1, and must comply with all the other regulations in Section A, and Section A has a Regulation A2, which says that T must have property P2, that if you've ticked the box for P1 you don't have to bother with P2 because A1 does not explicitly say that T must have P1 and P2?

Do you have any idea of how large and totally unusable the regulations would become if that were the case?

Please let us know if you do think that, because then we would all see that you have without doubt completely lost the plot, and we can all just ignore your barking mad ramblings.

Because that's what they are, even though you will no doubt try and portray that as an insult.

543.7.1.203(iii) does not require them to be rings because it doesn't explicitly say so even though it does explicitly say they have to comply with another regulation which explicitly requires that?

Sorry, but that really, really is barking mad.

Do you agree that if for HIE for a radial circuit one has to double the normal number of paths from each socket to earth then it is common sense, logical etc that for HIE for a ring circuit one has to double the normal number of paths from each socket to earth?

No.

Fair enough.

What is logical is that if you have a circuit topology which has non-redundant connections, then something needs to be added to apply redundancy - ie a second path. If the circuit topology is such that redundancy is already present, then further redundancy is not necessary.

So you think it is logical that "high integrity" should mean "more integrity" for one sort of circuit but "same integrity" for another.

Fair enough.

Taking the specific example, would you compare the PE redundancy of a radial where the end earth connection has been extended and taken back to the MET thus closing a ring, and that of an RFC (with one cpc) ?

Of course.

With the former you have done something to increase the integrity. In the latter you have not.

Assuming both use the same termination methods - ie using separate terminals in both accessories and MET - is there any difference in either topology or redundancy between the resulting PE "rings" ?

No, there isn't. But in the former that topology represents increased integrity or redundancy compared to "normal", and in the latter it does not.

In the former it makes a change which makes it reasonable for there to be the concept of "high integrity" as something distinct from "normal", in the latter it does not. But the concept of "high integrity" as something distinct from "normal" is applied by the Regulations to RFCs.

If you are right, and that actually there are no such requirements, do you have a logical, simple even, explanation of why they didn't just exempt RFCs in 543.7?

Also consider the specific statement that (with some conditions relating to circuit protection) that where two radial circuits are present, it is acceptable to link the earth terminations of the end points together so that the CP forms a ring. That is most definitely not "doubling up" the number of CPCs

Indeed it is not. It is changing the topology to increase the redundancy of connections. If the circuit started out needing that level of redundancy then when you need to increase it you need to change the topology. The way which the regulations say that shall be done is to double up the number of cpcs.

It really does. That's what "two individual" means - two is double the number one.


[quoteSo we have two radials, lets assume wired in T&E, and by choice or design we arrange for the last devices to be in close proximity so we can easily just drop a bit of earth wire between them.
Do we now agree that both radials (subject to separate terminations etc) now meet the requirements for HI earthing ? And that this has a single ring of protective conductors from the MET, via all the accessories, and back to the MET ?[/QUOTE]
I agree that it meets the requirements for HI earthing for radials, not rings.

 

[JohnW2]

My goodness - does that get the prize for the longest post in the forum? I gave up after the first screenfull or so!

 

[ban-all-sheds]

Lets take our industry, IT. Suppose you are sat in a meeting with various people reviewing power supplies for a number of small installations.

At installation A you observe that while you have a modular UPS, it only has one power conversion module installed - so if this fails you will lose power to the supported equipment. You all agree that adding a single extra module, thus doubling the number of modules, will give you redundancy in the event of a power module failure.

At installation B you observe that there is a larger load, and this is supported (without redundancy) by a similar UPS which has 3 power modules installed. What do you agree is the required action to introduce redundancy in the event of a single power module failure ?
Do you :
a) Apply the logic that because you had to double the number of modules in A, you must double the number of modules in B - and thus install an extra 3 modules for a total of 6.
or b) Apply the logic that adding only 1 module (for a total of 4) will give you redundancy and therefore adding the extra 2 is not required ?

But what if the redundancy you seek is also against failure, or removal, of one of the two power supplies into the room? I had a customer once who had N+1 power in his servers, where N+1 = 3. One weekend, when the electricians came in to do some work, they switched one circuit off, and he lost half his servers. They hadn't long recovered from all that when, the following weekend, they lost the other half.

What if you want to maintain power module redundancy even when you're operating on only one supply (hint - have you ever seen the lengths they go to in air traffic control installations?)

Surely we haven't found a question which does not have an answer as simple as you think?

So no, it's not logical to say that because A means doubling something, then to achieve the same result with B means doubling something.

But is it actually illogical to say that if Scenario A+ means 2xA that B+ means 2xB?

Starting with a clean sheet you might very well be able to design a B+ ≠ 2B, but you are not. You are starting with a regulation which says that α+ = 2α, and that is a logical approach.