I have employed a cowboy electrician: What now?

[JohnW2]

Before you do anything drastic, call him to a meeting have a frank conversation about all you have discovered and give him the opertunity to put it right.

Whilst that it what a lawyer or mediator would undoubtedly say (and makes a fair bit of sense), there are situations like this in which the very last thing the customer wants is for the individual in question to go anywhere near their electrical installation again (as in "putting it right").

Kind Regards, John

 

[winston1]

He claimed he had not left the earths unsheathed, and insisted that it was okay to have a radial off a ring as long as the number of sockets on that radial does not exceed the number on the ring (which is nonsense). He does not seem to be listed under the competent persons scheme

Well it is easy enough to check whether earths are sheathed or not. Have you actually done so (photos). A single radial off a ring should only contain one socket, but the total number of sockets on spurs can indeed equal the number on the ring.

By the way a cowboy is a man who rounds up cattle on horseback.

 

[JohnW2]

A single radial off a ring should only contain one socket, but the total number of sockets on spurs can indeed equal the number on the ring.

Unless you know something that I don't, I don't think there is any regulation which restricts the number of sockets on spurs (other than the restriction relating to unfused spurs). The OSG does not represent 'regulations'.

Kind Regards, John

 

[EFLImpudence]

Although that option is not suggested in the 'guidance' regarding ring finals in the regs, it makes good electrical sense - but only up to a point. A problem could arise if the several sockets on a 4mm² unfused spur drew a high total current from one point on the ring - which, if fairly close to one end of the ring, could overload the cable of the short leg (which would be contrary to the guidance in the regs). For that reason, I'm not convinced that those who invented the 'dispensation' for ring finals will have intended/wanted an unfused spur supply more than one (single or double) socket, even if the cable used for that spur were adequate.

It is therefore down to correct design which is allowed.

No different than a myriad of other things which may be done wrongly.

 

[EFLImpudence]

but the total number of sockets on spurs can indeed equal the number on the ring.

Some bungalows have all the sockets on spurs and none on the actual ring.

 

[JohnW2]

It is therefore down to correct design which is allowed. ... No different than a myriad of other things which may be done wrongly.

That's generally true - but we have to remember that, in relation to ring finals, we are talking about some things which would not be allowed by the (other) regulations, were it not for the specific dispensation in 433.1.204.

One could argue, as per recent discussions here, that what is 'allowed' should depend upon the designer's assessment of what currents are 'likely' (i.e. "design current"). On that basis, despite ('the guidance of') Appendix 15, if a designer was confident that it was unlikely that the total current in the spur would ever exceed 20A for appreciable periods, that a 2.5mm² unfused spur from a ring final supplying multiple sockets would/should then be acceptable, shouldn't it?

We see arguably 'silly' examples here. Because of ('informative') Appendix 15, we see people who have an unfused spur supplying one single socket (or one load via an FCU) being told that it would 'not be allowed' for them to take a 'spur from that spur' to supply something via an FCU with a 3A fuse. 'Sensible design' would not say that that was 'not allowed', would it?

Kind Regards, John

 

[Nozzle]

He agreed to install thicker than standard cable 2.5mm to 4 hifi sockets – I asked for 10mm - and invoiced me for the thicker cable, but instead installed standard cable 2.5mm cable.

OT alert

Do you like to buy gold plated and oxygen free mains cables and gold plated HDMI cables too? Most likely from that Russ Andrews bunch.

Nozzle

 

[EFLImpudence]

It is therefore down to correct design which is allowed. ... No different than a myriad of other things which may be done wrongly.

That's generally true - but we have to remember that, in relation to ring finals, we are talking about some things which would not be allowed by the (other) regulations, were it not for the specific dispensation in 433.1.204.

Yes but 433.1.204 begins with, within commas, "with or without unfused spurs" and nowhere does it mention that it applies to the spurs.
If it did apply, do you contend that a 4mm² spur with two double sockets would require 13A protection or the 4mm² is not allowed even though its CCC is >20A.
I was going to ask - would you also contend that a ring circuit wired in 4mm² (not for factor correction reasons) has to comply with 433.1.204? - but that doesn't make sense because it can't NOT.

One could argue, as per recent discussions here, that what is 'allowed' should depend upon the designer's assessment of what currents are 'likely' (i.e. "design current"). On that basis, despite ('the guidance of') Appendix 15, if a designer was confident that it was unlikely that the total current in the spur would ever exceed 20A for appreciable periods, that a 2.5mm² unfused spur from a ring final supplying multiple sockets would/should then be acceptable, shouldn't it?

No, because he wouldn't be confident as you cannot guarantee it if being used by a customer or in a commercial setting.
Yes, if you should wish to do it your own home and remember the requirement.

We see arguably 'silly' examples here. Because of ('informative') Appendix 15, we see people who have an unfused spur supplying one single socket (or one load via an FCU) being told that it would 'not be allowed' for them to take a 'spur from that spur' to supply something via an FCU with a 3A fuse. 'Sensible design' would not say that that was 'not allowed', would it?

I would say that example is alright because there is still 16A protection and even if someone changes it to 13A that is still 26A and alright.
However, were it a double socket then, no.
I know someone could change it to a double socket but that would be their fault just as they may add a socket before a properly protected spur.
We can't do everything to negate mistakes or ignorance.

 

[JohnW2]

Yes but 433.1.204 begins with, within commas, "with or without unfused spurs" and nowhere does it mention that it applies to the spurs. If it did apply, do you contend that a 4mm² spur with two double sockets would require 13A protection or the 4mm² is not allowed even though its CCC is >20A.

It's not the spur itself that is the issue, but what it is fed from. An unlimited number of sockets is obviously acceptable if all wired in 4mm² cable all the way back to a 32A MCB in the CU (i.e. a standard 32A radial). However, what would you say about the acceptability of that circuit if I told you that the last 2 metres of cable back to the CU was to be in 2.5mm² cable, in parallel with 48 metres of 2.5mm² cable (i.e. such that 96% of the total current went through the 2m length of 2.5mm cable). Would you still regard that circuit as acceptable?

I was going to ask - would you also contend that a ring circuit wired in 4mm² (not for factor correction reasons) has to comply with 433.1.204? - but that doesn't make sense because it can't NOT.

Quite. As you say, by definition, such a circuit would be compliant with 433.1.204.

No, because he wouldn't be confident as you cannot guarantee it if being used by a customer or in a commercial setting.

Indeed - but that's the whole discussion we've been having recently. The reality is that, without a crystal ball, a designer cannot know how a sockets circuit is going to be loaded - but the general view seemed to be that, given that truth, a designer is allowed to work on the basis of what (s)he believes is "likely". As discussed, were that not the case, it would not ever be possible to design a sockets circuit which had multiple sockets, particularly not a 20A radial, since there could be no certainty that the load would never exceed the In of the OPD.

I would say that example is alright because there is still 16A protection and even if someone changes it to 13A that is still 26A and alright.

Maybe, but that doesn't alter the fact that we see people being told that it is 'not allowed' If you feel that "26A is still alright" (presumably assuming Method C installation) would you feel that (despite Appendix 15), it would be 'alright' to have an unfused spur supplying two single sockets?

Kind Regards, John

 

[EFLImpudence]

Yes but 433.1.204 begins with, within commas, "with or without unfused spurs" and nowhere does it mention that it applies to the spurs. If it did apply, do you contend that a 4mm² spur with two double sockets would require 13A protection or the 4mm² is not allowed even though its CCC is >20A.

It's not the spur itself that is the issue,

...but that's what we're talking about.

but what it is fed from. An unlimited number of sockets is obviously acceptable if all wired in 4mm² cable all the way back to a 32A MCB in the CU (i.e. a standard 32A radial). However, what would you say about the acceptability of that circuit if I told you that the last 2 metres of cable back to the CU was to be in 2.5mm² cable, in parallel with 48 metres of 2.5mm² cable (i.e. such that 96% of the total current went through the 2m length of 2.5mm cable). Would you still regard that circuit as acceptable?

No, of course not but I keep saying that is for the designer to consider.

Indeed - but that's the whole discussion we've been having recently. The reality is that, without a crystal ball, a designer cannot know how a sockets circuit is going to be loaded - but the general view seemed to be that, given that truth, a designer is allowed to work on the basis of what (s)he believes is "likely".

All socket circuits must, by nature, be like that.

As discussed, were that not the case, it would not ever be possible to design a sockets circuit which had multiple sockets, particularly not a 20A radial,

You cannot design a socket circuit , other than avoid what may be obvious as wrong - high demand spur(s) near one end (as you say) , a huge kitchen where the demand is bound to be too high.
Apart from single 20A demand circuits (which may be added to by the unknowing) I can think of no reason to install a 20A radial.
People talk as if 2.5mm² is all there is available.

Splitting into two 20A radials is often quoted as a solution, albeit temporarily permanent, for a fault in a ring circuit so it is fundamentally flawed, then.

since there could be no certainty that the load would never exceed the In of the OPD.

Again, that is the nature of socket circuits.

I would say that example is alright because there is still 16A protection and even if someone changes it to 13A that is still 26A and alright.

Maybe, but that doesn't alter the fact that we see people being told that it is 'not allowed' If you feel that "26A is still alright" (presumably assuming Method C installation) would you feel that (despite Appendix 15), it would be 'alright' to have an unfused spur supplying two single sockets?

Yes, electrically, it obviously is - leaving aside the fact that 'no one' 'knows' you're not supposed to put two 13A loads in a double socket.
We see people being told all sorts of things.

I cannot believe Appendix 15 did not show two singles because people are supposed to know about the double socket 'limit' and therefore would know it is wrong.
Perhaps they did not show it because it is obviously alright and everyone should know that.
More likely, it is not shown because it is not possible to show everything.
It does not show two single sockets in 15B; what does that mean - not allowed?

Why is Appendix 15 there at all?
433.1.204 makes no mention of it.

 

[PBC_1966]

There was a time when the Wiring Regs. stated quite clearly that a non-fused spur may feed two single sockets or one double. Then a revision made it quite clear that a non-fused spur may feed only one single socket or one double.

Whatever technical issues might exist in either case, the stated rules were clear and concise. Now, apparently, the rules themselves are silent on the issue and there's an "informative appendix" which is vague to the point of not being very informative at all.

It seems to me that in its current incarnation, BS7671 has lost its way somewhat.

 

[JohnW2]

An unlimited number of sockets is obviously acceptable if all wired in 4mm² cable all the way back to a 32A MCB in the CU (i.e. a standard 32A radial). However, what would you say about the acceptability of that circuit if I told you that the last 2 metres of cable back to the CU was to be in 2.5mm² cable, in parallel with 48 metres of 2.5mm² cable (i.e. such that 96% of the total current went through the 2m length of 2.5mm cable). Would you still regard that circuit as acceptable?

No, of course not but I keep saying that is for the designer to consider.

OK, so are you now revising your previous statement to say that you only consider a 4mm² unfused spur (from a 2.5mm² ring final) supplying multiple sockets is only OK if it originates more than X% of the ring length from one end of the ring? ... and, if so, what is your value of 'X'?

... but the general view seemed to be that, given that truth, a designer is allowed to work on the basis of what (s)he believes is "likely".

All socket circuits must, by nature, be like that.

Exactly my point.

Apart from single 20A demand circuits (which may be added to by the unknowing) I can think of no reason to install a 20A radial. People talk as if 2.5mm² is all there is available.

As I said recently, I would personally be hesitant to install a multiple-sockets 20A radial unless it was only for my use.

I cannot believe Appendix 15 did not show two singles because people are supposed to know about the double socket 'limit' and therefore would know it is wrong. Perhaps they did not show it because it is obviously alright and everyone should know that. More likely, it is not shown because it is not possible to show everything.

You seem to be missing something. It's not just a question of Appendix 15 'not showing/mentioing' two single sockets (as you say, it is not possible to show/mention everything) - but, rather, that it explicitly says "An unfused spur should feed one single or one twin socket outlet only".

Why is Appendix 15 there at all? 433.1.204 makes no mention of it.

Very good question. Paul has indicated that, in previous versions of the BS7671, much of which is now in ('informative') Appendix 15 (including this business about what can be supplied by unfused spurs) existed in the body of the regulations. One can but speculate about why they have moved it from being 'required' to be only 'guidance'.

Kind Regards, John

 

[EFLImpudence]

OK, so are you now revising your previous statement to say that you only consider a 4mm² unfused spur (from a 2.5mm² ring final) supplying multiple sockets is only OK if it originates more than X% of the ring length from one end of the ring? ... and, if so, what is your value of 'X'?

I don't think I said that.
I haven't worked out the value but you have several times but probably more than the 2m in your above example.

but, rather, that it explicitly says "An unfused spur should feed one single or one twin socket outlet only".

Oh, I'm sorry. I been thinking for myself for too long and completely forgotten it actually said that.

Next question, then, may be rather obvious - Why only one single?

 

[EFLImpudence]

It seems to me that in its current incarnation, BS7671 has lost its way somewhat.

Yes, probably been altered too quickly and too often that related parts get forgotten.

A bit like when editing a long post.



There is a corrigendum just issued, did you know?
http://electrical.theiet.org/wiring-regulations/updates/index.cfm

 

[PBC_1966]

Next question, then, may be rather obvious - Why only one single?

Presumably the thinking was that with two singles, which may be some distance apart or even in different rooms, there was a much greater chance of two 3kW loads being plugged in than with a double socket.

There is a corrigendum just issued, did you know?
http://electrical.theiet.org/wiring-regulations/updates/index.cfm

Interesting. On two completely unrelated issues, is that saying that BS7671 no longer accepts the installation of BS196 connectors?

And in table 44.3, what's the 277/480V specification doing there? That's an American standard system.