As we seem to be in a bit of a quiet period between revisions to BS 7671, I thought it might be a good opportunity to review what we already have and see if it could be improved.
There are a few of the existing regulations where the wording doesn't entirely make sense to me - I'm sure others have had similar problems. So what's your "favourite" cryptic regulation? What's wrong with it and (if you can) how could it be improved?
- Andy.
As we seem to be in a bit of a quiet period between revisions to BS 7671, I thought it might be a good opportunity to review what we already have and see if it could be improved.
To be honest, it's probably more efficient to comment on up-coming HD 60364 series standards, as these will be incorporated into BS 7671 in the future. At this stage, we are looking at accepted HD 60364 series standards, for which the UK is obliged,. by agreement with CENELEC, to adopt in BS 7671 at the next available update.
The only other thing we can do, is correct errors, or try and address ambiguities ... but if there is an issue with understanding the technical intent of the requirement in the relevant HD 60364 series document, that has to go back to CENELEC (and possibly IEC) committees to get a steer, or, sadly, wait until after that is ratified in the international standard!
For openers I'd like to nominate 462.1 (isolating/switching N):
This is the subject of a Special National Condition (SNC) in the Harmonized standard ... so, what does that mean?
Well, we can resolve the issue nationally, but if it results in an adjustment to the BS 7671 requirement, that JPEL/64 deems is a modification to the technical intent of the SNC, then we will have to apply to CENELEC for a change to the SNC, and potentially this has to happen prior to amending it in BS 7671.
it's probably more efficient to comment on up-coming HD 60364 series standards
Very interesting point. What's the process for HDs? - do they go through a DPC process like BS 7671 or are changes usually sponsored by national committees? (From purely an administrative view, with such a wide potential audience, I can see an advantage in having things filtered a bit...)
...that JPEL/64 deems is a modification to the technical intent of the SNC
So, in the case of 462.1, perhaps the first stage is to work out what the technical intent was. Yes, I know, ultimately only a court of law can decide on the meaning, but presumably there should be someone in the UK who knows what was originally intended (the current wording isn't that old IIRC) - the wording of the 2nd part especially simply doesn't make sense to me (mea culpa no doubt, but I'm unlikely to be the only one, and if a significant number of those that are meant to implement the regs can't understand what is required, we have a problem, regardless of the semantic correctness).
- Andy.
do they go through a DPC process like BS 7671 or are changes usually sponsored by national committees? (From purely an administrative view, with such a wide potential audience, I can see an advantage in having things filtered a bit...)
Yes, they are put out for public comment through the same BSI Standards Development Portal as BS 7671.
So, in the case of 462.1, perhaps the first stage is to work out what the technical intent was.
It is as simple as it's not necessary to switch the neutral in cases where bonding is in place, and either (a) or (b) applies.
(a) and (b) are in effect the same thing ... it's a reliably earthed TN system. BS 7671 can't second-guess a distributor or legislation, but with respect to (b) see Regulation 114.1 (2nd para).
So, if there is no bonding, you must switch the neutral, or it's not a TN system, you must switch the neutral (unless there's another exception ... there are odd exceptions for TT systems dotted around).
Second interpretation
In a TN-C-S or TN-S installation in which protective equipotential bonding has been installed, the neutral conductor may be left unswitched provided that either condition (a) or condition (b) applies.
It follows that if there is no protective equipotential bonding (e.g. because the other mains services arrive in plastic pipes) the neutral conductor may be left unswitched even if neither condition (a) nor condition (b) applies.
I think I'm allied with this one.
Some of this is the language of standardization, though ... which is dictated to us by ISO, IEC and BSI. I agree that the language used is not ideal for users of the standard, although we do provide training in the industry that is supposed to address that.
Some of this is the language of standardization, though ... which is dictated to us by ISO, IEC and BSI. I agree that the language used is not ideal for users of the standard, although we do provide training in the industry that is supposed to address that.
I did wonder at the beginning of this thread, whether there might be an element of linguistics. Some languages have moods which do not exist in others, which may apply particularly to conditional clauses. So, something may be clearer in one language than another.
It is as simple as it's not necessary to switch the neutral in cases where bonding is in place, and either (a) or (b) applies.
That bit I think I understand the words (even if I have queries about the logic), it's the "the neutral conductor is reliably connected to Earth by a low resistance to meet the disconnection times of the protective devices according to the requirements of Chapter 41" that's got me confused.
Say I have a rural PME (or PNB) supply where the supply's electrode has a resistance of close to 20Ω - how does that mesh with Chapter 41?
- Andy.
Say I have a rural PME (or PNB) supply where the supply's electrode has a resistance of close to 20Ω - how does that mesh with Chapter 41?
see Regulation 114.1 (2nd para).
In a TN system, the earth fault path does not include Earth, so I'm not sure how a particular earth electrode resistance value helps ...
Say I have a rural PME (or PNB) supply where the supply's electrode has a resistance of close to 20Ω - how does that mesh with Chapter 41?
see Regulation 114.1 (2nd para).
In a TN system, the earth fault path does not include Earth, so I'm not sure how a particular earth electrode resistance value helps ...
I think that there is another facet to this discussion, which was not obvious to me initially.
It is easy to think of protective equipotential bonding in terms of 411.3.1.2. However, that is only concerned with extraneous-conductive-parts.
The definition of equipotential bonding (page 39) includes exposed-conductive-parts as well. So, the bolts which attach the earth bar inside a DB to the casing provide not only support, but also equipotential bonding.
It follows that virtually all modern domestic and light commercial installations will have some bonding.
461.2 does little more than re-state Reg 8 of ESQCR 2002.
Given the exception in 461.2, 462.2 would seem to apply almost entirely to TT and IT systems.
It follows that virtually all modern domestic and light commercial installations will have some bonding.
Agreed, where ADS is used.
see Regulation 114.1 (2nd para).
Yes, but that only mentions the reliability of that connection, not its resistance.
In a TN system, the earth fault path does not include Earth, so I'm not sure how a particular earth electrode resistance value helps ...
Exactly, so what are they trying to get at with the phrase "the neutral conductor is reliably connected to Earth by a low resistance to meet the disconnection times of the protective devices according to the requirements of Chapter 41" when the that resistance has no relevance to disconnection times?
- Andy.
The definition of equipotential bonding (page 39) includes exposed-conductive-parts as well.
But is that just covering the case of supplementary bonding (to/between exposed-conductive parts) rather than c.p.c.s. in general - given that a very significant proportion of the supply voltage can be dropped along a c.p.c. during a fault (and in the case of distribution and larger final circuits, for a significant length of time).
- Andy.
But is that just covering the case of supplementary bonding (to/between exposed-conductive parts) rather than c.p.c.s. in general - given that a very significant proportion of the supply voltage can be dropped along a c.p.c. during a fault (and in the case of distribution and larger final circuits, for a significant length of time).
Since BS 76781:2008, bonding to e-c-p's is only applied 'to each building' ... but all we are doing here, is highlighting potential ambiguities in the web of related requirements, not a single requirement.
It's still possible for faults [somewhat depending on definition] to go via intermediate substances and then part of earth before finding that good [enough] return path...
Earth has become a DMZ where we want a clear separation of concerns, but reality prevails that it's a difficult zone to monitor and protect. (De-Militarized Zone; like electron traps in a band-gapped semiconductor..]
I hesitate to speak in such knowledgeable company - and I know I'm not adding much value to you guys, but plenty read and don't reply -
I'm quite sure this reg has been around a long time - since at least the 16th edition or in other words - the turn of the century.....;-) A while anyway.....
I've always interpreted it to mean that as long as it's a TN system of some sort - either inside (Like flats where the N_E connection is sometimes inside the property - very like a PNB - at or near the tenants meter) or a classic TNS or TNC-S system - where the N_E bond is at the transformer and at in the case of a TNC-S multiple points between transformer and the properties it serves -outside the property - you don't have to switch the neutral - leaving - if its a TT system you do have to switch the neutral.
I've always thought the TT bit was really what this reg was trying to indirectly point at; there you do have to switch the neutral too.
The earth and metal bits in most TN systems remain at neutral potential anyway - so unless you switch the Terra neutral (TN conductor) it'd be useless to switch the neutral elsewhere. And in this same reg it starts off - you cant switch the TN conductor.
Note 559.5.5 does say a circumstances where the N needs to be switched.......which is interesting but I'd guess that's to prevent a current carrying conductor on another phase being tampered with while a this phase is getting worked on. The same could be said for high level warehouse heaters for example - I'd guess.
What I've not come across before - perhaps because I do commercial city work mostly - is a TN system with up to 20 ohms (Ze I'd assume?) this to me would require the property to be treated as if its a TT surely or the DNO needs to come and sort it out?
I've always interpretted "........low resistance to meet the disconnection times of the protective devices....." to mean - as long as the Ze is good enough that it doesn't affect the MCCBs, MCBs, BS88s - then the Neutral need not be switched.
I don't think I mind the wording.....but id maybe phrase it more like I've done.. what what. ;-)
What I've not come across before - perhaps because I do commercial city work mostly - is a TN system with up to 20 ohms (Ze I'd assume?)
I think what Andy means is that the transformer is earth spiked at around 20 ohms, which is a good reading for a TT system. The multiple earth rods/plates/meshes etc along the route bring the reading down somewhat, then the N and E are separated at the premises supply intake, which mean when we do an Earth Loop impedance test, it is actually measuring the N conductor, with its multiple earth connections, rather than the transformer at source, which would give a reading of 20 ohms.
Please follow up if I'm badly misinformed!
I'm quite sure this reg has been around a long time - since at least the 16th edition or in other words - the turn of the century.....;-) A while anyway.....
It has indeed - at least the general gist of it...
In my 16th:
460-01-04 Combined protective and neutral (PEN) conductors shall not be isolated or switched.
Except as required by Regulation 476-01-03, in TN-S or TN-C-S systems the neutral conductor need not be isolated or switched, where the neutral conductor can reliably be regarded as being at earth potential. For supplies which are provided in accordance with the Electricity Safety, Quality and Continuity Regulations 2002, the supply neutral conductor (PEN or N) is considered to be connected with earth by a suitably low resistance.
which I think says pretty clearly what we all think it should say. (476-01-03 is the one for a DP main switch in household single phase systems)
The bit for the suitably low resistance "to meet the disconnection times of the protective devices according to the requirements of Chapter 41" is a later addition - and that's the bit that's giving me the most trouble!
What I've not come across before - perhaps because I do commercial city work mostly - is a TN system with up to 20 ohms (Ze I'd assume?) this to me would require the property to be treated as if its a TT surely or the DNO needs to come and sort it out?
Ah, no, not Ze. (Not in my head at least, which might well be where I'm going wrong). Ze (and indeed Zs) in a TN system are the metallic earth fault loop - which doesn't rely on the general mass of the Earth (unlike a TT system). To me the words "connected to Earth" mean the connection to the general mass of the Earth (as per the Part 2 definitions), rather than merely to the PE conductor - i.e. includes the resistance of the supply electrodes(s) - with are often in the 'not exceeding 20Ω' range.
Maybe there's some subtle difference between the 16th use of "earth" and the 18th "Earth" (the latter more definitely referring to the part 2 definition).
- Andy.
I think what Andy means is that the transformer is earth spiked at around 20 ohms
Exactly! (sorry, crossing posts)
- Andy.
We're about to take you to the IET registration website. Don't worry though, you'll be sent straight back to the community after completing the registration.
Continue to the IET registration site
