DNO removed RCD on TT system

[ban-all-sheds]

Hmm.

15 years ago the 16th Edition was in force, and it contained provisions pretty similar to the current ones about minimising inconvenience and taking due account of the operation of any single protective device.

I still am surprised, given what we can see, and what you have said about the separate circuits, that more use was not made of RCBOs, or of multiple RCDs within the DBs.

 

[JohnW2]

Hmm. ... 15 years ago the 16th Edition was in force, and it contained provisions pretty similar to the current ones about minimising inconvenience and taking due account of the operation of any single protective device. ... I still am surprised, given what we can see, and what you have said about the separate circuits, that more use was not made of RCBOs, or of multiple RCDs within the DBs.

True. However, I suppose that one of the issues is that, with a TT installation, unless every final circuit is protected by an RCD or RCBO (and maybe even then - see comment about my installation below), one will still have to have an up-front TD RCD - in which case the whole installation (if single phase) will (as you recently pointed out) get taken out by any N-E fault on a circuit protected by a SP RCBO. One could, I suppose, argue that N-E faults are sufficiently rare that this would not represent a serious 'inconvenience'.

In the OP's case, of course, there appear to be two phases - so, as I said, at least that affords the possibility of giving each of them their own up-front RCD, thereby reducing 'inconvenience'. That's like my set-up, with 3 phases, each with their own up-front RCDs (primarily to give required fault protection to the lengthy distribution circuits). I only have two or three RCBOs, but they are all SP.

Kind Regards, John

 

[westie101]

though the other option that has been used in this situation, often with off-peak heating) is to bring in 2 services both on the same phase.
This is where the load exceeds the capability of one service cable, cut-out & meter.
Though the neutral being combined may just suggest split single

 

[JohnW2]

though the other option that has been used in this situation, often with off-peak heating) is to bring in 2 services both on the same phase. This is where the load exceeds the capability of one service cable, cut-out & meter. Though the neutral being combined may just suggest split single

Isn't that last point more convincing than you imply? If the line conductor of one service was not adequate for the load, then surely a single neutral would not be adequate for the total load, either, even if two lines (of same phase) were brought in?
Edit: I didn't look at the photo properly, so what I've just written (and probably some of the things I wrote earlier) is probably nonsense! It would be true if there were only one income neutral (which is what I was thinking), but that's not the case. There are separate neutrals to (and from) the meters, with the neutrals being combined after the meters - which seems pretty odd.

Kind Regards, John

 

[finalcoat]

JohnW2

If 2 single phase RCDs were used, would these need to be installed before the neutrals are combined i.e. before the isolator?

 

[JohnW2]

JohnW2 ... If 2 single phase RCDs were used, would these need to be installed before the neutrals are combined i.e. before the isolator?

Please see edit of my previous message - I didn't look at photo properly.

If 2 'single phase' RCDs were installed, it would be necessary to get rid of the combining of the neutrals and, instead, feed them separately to the circuits supplied by the corresponding L cables - which might make it a bigger job than I was envisaging. Since there are only two Ls, there is a spare way in that isolator (it's designed for 3L+N), so the two neutrals could (instead of being joined in that block) be switched separately by that isolator. Once that was done, the two RCDs could be fed from the respective L/N pairs coming out of the isolator.

Kind Regards, John

 

[Risteard]

If, per the title of this thread, it is really is a TT system then, as your surmise, every circuit must have RCD protection

Not strictly true. A TT system with a suitably low impedance which can be guaranteed is not actually required to have an RCD for fault protection.

 

[JohnW2]

If, per the title of this thread, it is really is a TT system then, as your surmise, every circuit must have RCD protection

Not strictly true. A TT system with a suitably low impedance which can be guaranteed is not actually required to have an RCD for fault protection.

True - but, in a 'domestic environment (even in a 'large country house'), it is extremely unlikely that a TT earth will have "a suitably low impedance which can be guaranteed".

IIRC, even the DNO earth rods are only required to have an impedance of around 10Ω (which itself would be far too much), and at least one of those will be part of a customer's EFLI.

Kind Regards, John

 

[JohnW2]

deleted (done it again - quoted myself when I intended to edit!)

 

[ban-all-sheds]

True. However, I suppose that one of the issues is that, with a TT installation, unless every final circuit is protected by an RCD or RCBO (and maybe even then - see comment about my installation below), one will still have to have an up-front TD RCD - in which case the whole installation (if single phase) will (as you recently pointed out) get taken out by any N-E fault on a circuit protected by a SP RCBO.

Well - there don't appear to be any lengthy sub-mains in this installation.

So the requirement to have everything on some sort of RCD could have been done by splitting the DB into multiple sections, or installing multiple DBs, so that only a subset of circuits would be lost if one tripped.

 

[JohnW2]

True. However, I suppose that one of the issues is that, with a TT installation, unless every final circuit is protected by an RCD or RCBO (and maybe even then - see comment about my installation below), one will still have to have an up-front TD RCD - in which case the whole installation (if single phase) will (as you recently pointed out) get taken out by any N-E fault on a circuit protected by a SP RCBO.

Well - there don't appear to be any lengthy sub-mains in this installation. So the requirement to have everything on some sort of RCD could have been done by splitting the DB into multiple sections, or installing multiple DBs, so that only a subset of circuits would be lost if one tripped.

Yes, of course - but, as I said, even without the distribution circuits issue (like I have), that is only true if one does have "everything on some sort of RCD" - else there will be a need for an up-front RCD (hence potential problems with N-E faults and SP RCBOs).

However, as I also said, in the case of the installation we're discussing, there are two separate phases (at least, supplies) which could be treated separately, even if they each had to have 'up front RCDs' - so one would thereby avoid a single fault taking out the entire installation.

Kind Regards, John

 

[ban-all-sheds]

Yes, of course - but, as I said, even without the distribution circuits issue (like I have), that is only true if one does have "everything on some sort of RCD" - else there will be a need for an up-front RCD (hence potential problems with N-E faults and SP RCBOs).

But you don't need a single up-front one. With an installation the size of the one in the photo you could have half a dozen - as long as all circuits are on an RCD you're fine.

NE faults with SP RCBOs will still result in the upstream time delayed RCD tripping, but that won't take out the whole house, only some of it.

 

[JohnW2]

Yes, of course - but, as I said, even without the distribution circuits issue (like I have), that is only true if one does have "everything on some sort of RCD" - else there will be a need for an up-front RCD (hence potential problems with N-E faults and SP RCBOs).

But you don't need a single up-front one. With an installation the size of the one in the photo you could have half a dozen - as long as all circuits are on an RCD you're fine.

Sure - as I've said, the installation in question lends itself to the first step in that direction, given that there aretwo separate supplies/phases. However, on the basis of what we've been told, there is a serious paucity of RCD/RCBO protection of any sort (only for bathrooms and outside circuits) let alone the sort of structured arrangement you are talking about.

NE faults with SP RCBOs will still result in the upstream time delayed RCD tripping, but that won't take out the whole house, only some of it.

Indeed - as above. The ideal, of course, is to have every final circuit separately RCD/RCBO protected, then there will usually be no need for any 'upstream TD RCD', hence no problem with SP RCBOs. I have that situation in terms of final circuit protection but, at least in theory, still need the TD RCDs to protect the distribution circuits.

Kind Regards, John

 

[AdrianUK]

though the other option that has been used in this situation, often with off-peak heating) is to bring in 2 services both on the same phase. This is where the load exceeds the capability of one service cable, cut-out & meter. Though the neutral being combined may just suggest split single

Isn't that last point more convincing than you imply? If the line conductor of one service was not adequate for the load, then surely a single neutral would not be adequate for the total load, either, even if two lines (of same phase) were brought in?
Edit: I didn't look at the photo properly, so what I've just written (and probably some of the things I wrote earlier) is probably nonsense! It would be true if there were only one income neutral (which is what I was thinking), but that's not the case. There are separate neutrals to (and from) the meters, with the neutrals being combined after the meters - which seems pretty odd.

Kind Regards, John

Lets hope this is a split phase (230-0-230) supply otherwise thats' one seriously overloaded combined neutral conductor! (and isolator switch & isolator in the fuseboard & fuseboard neutral bus)

If it IS a split phase supply .... then is that fuseboard rated for 460V? .... It looks like a standard three phase board which, strictly speaking, is probably only rated for 400V between phases......

Looking at the photos... it doesn't look like every third way is vacant, so the installer must have paralleled up two of the fuseboard phases onto one of the incoming supplies.... I'm not sure the installer really understood the supply arrangements

 

[JohnW2]

Lets hope this is a split phase (230-0-230) supply otherwise thats' one seriously overloaded combined neutral conductor! (and isolator switch & isolator in the fuseboard & fuseboard neutral bus) ... If it IS a split phase supply .... then is that fuseboard rated for 460V? .... It looks like a standard three phase board which, strictly speaking, is probably only rated for 400V between phases.....

I can't really disagree with any of that - but would it be normal to bring in a split-phase supply as two separate cables, each with a neutral?

Looking at the photos... it doesn't look like every third way is vacant, so the installer must have paralleled up two of the fuseboard phases onto one of the incoming supplies....

That's certainly one possibility. Give or take the '460V rating' issue, would there necessarily be anything wrong with that? What are the top two bits - they look like single-phase boards to me. Might the two supply phases not be split between those three 'sub-boards', with just one of the supply phases going to the 3-phase board (with all three phases of the board paralleled), in which case there would not be a 460V issue? ... would there be anything wrong with that (provided the two phases were vaguely balanced, for the sake of the single neutral)?

I'm not sure the installer really understood the supply arrangements

Maybe not - but, there again, nor do we with any certainty! If it is a split-phase supply, it's possible that what we are looking at is quite reasonable, isn't it?

Kind Regards, John