Does this setup contravene any regs or is it ok?

[markythesparky]

Personally, I would not have wired the lobby and the attic as one circuit. If I were to come across such an installation, I’d be wondering what the original installer was thinking of.

That's the point! He didn't!

The origin of any final circuit is the DB/CU.

Where did you get that definition from?

 

[johnduffell]

Some people are looking at this backwards. A circuit is not defined topologically(other then a ring). It doesn't matter how it branches and at what points, any collection of points arranged as a tree topologically can be a circuit.
If there's an ocpd and some wiring that forms a tree, then by definition that is a circuit. The ocpd defines the circuit. Doesn't matter if you combined 3 previous circuits into one ocpd, you've turned them into one circuit.
As hinted above, the only exception would be combining two rings into one. That wouldnt be a valid circuit as the whole circuit has to be arranged as a ring. However that would only be against the regs in the same way a lollipop would be, and could probably be a deviation.

 

[SparkyChick]

Your argument is quite compelling because it's made me think about it again and I was starting to question my own view, but having just re-read the definitions and 314.4 I'm sticking to my guns and this is why.

These are in fact two circuits and I will explain this later, and assuming this, these are not now electrically separate as they have been connected together.

Fundamentally, yes they are two circuits... or more precisely, they were two circuits as they were originally designed, installed and connected to two different protective devices at the origin.

To address your question about the origin, since there is no definition of origin for a circuit, it seems sensible to base it on that of 'origin of the installation' i.e. the point of supply of electrical energy, which in this case is the protective device the circuit is connected to.

The definition of 'final circuit' is there to differentiate it from a distribution circuit and in this context is largely a red herring I believe.

The mere act of changing the origin of one of these circuits to that of the other means they become one circuit. It isn't the cabling that defines a circuit, it's the origin. The cable is not a point of supply of electrical energy for the circuit, the protective device it is connected to is.

Taking your argument that the cables are the origin to the extreme, you're saying you can't have a spur off a ring final circuit that originates at the device that protects the ring conductors because they aren't the same cables, which I think you'll agree is a nonsense.

You seem to be conflicted.

Not conflicted, just sensible.

Yes I have ‘lumped’ two circuits together as ‘temporary stop gap solution’.
I did it as a temporary stop gap solution knowing that, in my view and yours presumably as you have inferred, it’s ok as a temporary fix. But if its ok as a temporary fix, by definition it isn't meant as a permanent solution.

When you combine two circuits in this manner, there is a risk that the protective device will be overloaded as a result and thus subject to nuisance tripping under normal use because the installation was previously divided to minimise the chances of that happening.

Hence I would consider it a temporary solution to get the customer back on-line ASAP whilst a more permanent solution is implemented.

However, in the case of two lighting circuits, it's unlikely (based on the OPs description) the OCPD at the origin is going to be subjected to an overload when they are combined, so this is an acceptable thing to do.

Therefore the man from Stroma is correct. It doesn’t comply with the intention of the regulation. It is electrically safe taking into consideration the anticipated loads on those circuits but is does not comply. My temporary fix was not left like it.

Very different situations as outlined above, so I disagree. Mr. Stroma was incorrect.

Theres nothing wrong with having a lighting circuit on a 16a breaker as long as the cable is sufficient.

I think I said that

I'm not sure what you mean by the ring on a 20a breaker though.

Basically a ring final circuit on a 20A circuit breaker, most likely as a result of installation methods (cable in insulation for example). I've seen it a couple of times. Wired in 2.5/1.5mm but due to installation methods, a 32A breaker runs the risk of overloading the cable (thermal effects) so a 20A breaker has been deemed suitable.

My question is would you wire two ring circuits into one 32a breaker allowing of course for the loads anticipated on those circuits and leave it like it permanently and sign a completion certificate saying that it fully complies with BS7671.

Would I do it? No. But based on 433.1.204, providing the cabling complies, it's questionable whether it would contravene any regulations.

 

[123]

On rewires or new builds I have always put the smoke alarms and a commonly used lighting circuit into the same OCPD. I feel it's better than a standalone circuit for the smoke alarm. Also often see the door bell and the downstairs lighting in the same OCPD, I don't see the issue?

 

[Ian1981]

It’s been made into one circuit just accept it and move on with life

 

[markythesparky]

Fascinating thread!

Fundamentally, yes they are two circuits... or more precisely, they were two circuits as they were originally designed, installed and connected to two different protective devices at the origin.

So if they were indeed two circuits by design but we can change that by connecting them together at the CU, can we then lump three or four or even more to one breaker, loads considered?
Also if you had a long run of say 6.0mm SWA on a 20a OCPD and it was drawing 11a, and you had another circuit on a 20a which was 2.5mm T&E and drawing 7a could these be lumped together as one circuit even though they are of different sizes and even types? What would go on the certificate in respect of cable size? Where does this end?

To address your question about the origin, since there is no definition of origin for a circuit, it seems sensible to base it on that of 'origin of the installation' i.e. the point of supply of electrical energy, which in this case is the protective device the circuit is connected to.

I thought you'd pick that one! So at the origin of the installation, is the 'point of supply' the incoming cables connect to the head, the main OCPD, if so which side of it, or the point at which the tail connects to the OCPD? It doesn't say so again you cannot go definitive on that. You could argue that the tail's connection to the OCPD is the point of supply of electrical energy and to follow that argument, the cable feeding the lighting circuit is too.

The definition of 'final circuit' is there to differentiate it from a distribution circuit and in this context is largely a red herring I believe.

Agreed.

It isn't the cabling that defines a circuit, it's the origin. The cable is not a point of supply of electrical energy for the circuit, the protective device it is connected to is.

Why? I fundamentally disagree with this. I believe that it is absolutely the cabling that defines the circuit. Are you saying that a breaker with no cable connected is still a circuit? We are getting down the semantics of what the definition is and I think this is where you and I differ.

Taking your argument that the cables are the origin to the extreme, you're saying you can't have a spur off a ring final circuit that originates at the device that protects the ring conductors because they aren't the same cables, which I think you'll agree is a nonsense.

The regs state specifically that this is allowed from a ring final circuit and so this is also a red herring.

Would I do it? No. But based on 433.1.204, providing the cabling complies, it's questionable whether it would contravene any regulations.

So why wouldn't you do it then?

Finally, if you are correct and I am not, then what is the meaning of 314.4?

 

[markythesparky]

I feel it's better than a standalone circuit for the smoke alarm.

Why?

 

[markythesparky]

It doesn't matter how it branches and at what points

However that would only be against the regs in the same way a lollipop would be, and could probably be a deviation.

Why then could a lollipop layout be against regs if it is not designed to be a ring final and as long as the OCPD is the correct rating for the cable.

 

[123]

Why?

Either on here or elsewhere their was a spark, who also worked part time for fire brigade. He said of some of the sights he'd seen and the damage to human life he'd seen, and how working smoke alarms could have prevented it. Part of his thinking was smoke alarms on commonly used lighting circuit could have prevented it. It's a logic and approach I agreed with then, and still do now.

You can't help stupid though - if they pull smoke alarm heads off and throw them in the bin with no intention of repairing them, nothing we can do, whether separate circuit or on with the lighting.

 

[markythesparky]

Either on here or elsewhere their was a spark, who also worked part time for fire brigade. He said of some of the sights he'd seen and the damage to human life he'd seen, and how working smoke alarms could have prevented it. Part of his thinking was smoke alarms on commonly used lighting circuit could have prevented it. It's a logic and approach I agreed with then, and still do now.

You can't help stupid though - if they pull smoke alarm heads off and throw them in the bin with no intention of repairing them, nothing we can do, whether separate circuit or on with the lighting.

But I don't see why the smokes are better being fed with another circuit. Is it because people are not likely to switch of a lighting circuit if there smokes are beeping?

 

[123]

But I don't see why the smokes are better being fed with another circuit. Is it because people are not likely to switch of a lighting circuit if there smokes are beeping?

In a nutshell, yes.

Domestic Smoke alarm maintenance can be a bit of a losing battle, folk seem to think if they change batteries when it bleeps, the fact it says replace by NOV 2010 means bugger all.

 

[johnduffell]

Why then could a lollipop layout be against regs if it is not designed to be a ring final and as long as the OCPD is the correct rating for the cable.

Well true, but i was meaning a 4mm "stick" with a 2.5mm loop all on a 32A mcb. If it were all 4mm or all on a 20A mcb the would be no point making it a lollipop

 

[SparkyChick]

Fascinating thread!

I quite like them as it gives you the chance to exercise your interpretation of the regs and get other opinions, regardless of whether they change your views or not.

It's one of the things I miss from my old job, the ability to quickly bounce ideas around and discuss them. Working solo isn't conducive to it... well, it's not if you don't want to end up in a straight jacket

So if they were indeed two circuits by design but we can change that by connecting them together at the CU, can we then lump three or four or even more to one breaker, loads considered?

I would say theoretically yes on the proviso that they are being moved to a breaker of the same rating as their original design or less. But loading aside there are other factors to consider like the ability to safely terminate the cables at the DB. There will be a point where it's just not physically possible to do it safely without say using a much larger cable and terminating the individual cables in a junction box.

Also if you had a long run of say 6.0mm SWA on a 20a OCPD and it was drawing 11a, and you had another circuit on a 20a which was 2.5mm T&E and drawing 7a could these be lumped together as one circuit even though they are of different sizes and even types? What would go on the certificate in respect of cable size?

In their own right, both cables would be adequately protected by a 20A OCPD, and the total combined load (assuming that is the maximum you've stated) would not result in nuisance tripping due to unintentional overload.

So yes, theoretically they could be combined. I'd record the difference in cable size by stating MIXED in the size columns and then expressing that on a continuation page. Same as I would for any circuit with mixed conductor sizes. The lollipop being a fine example - not that I've ever done this.

And of course providing due regard was given to the other parts of the regulations regarding division of the installation (i.e. minimise risks, inconvenience etc.).

However, would I do either of these things? I'm not sure. I might at a pinch (as an example, the situation where you need another breaker and cost is a major factor) but ordinarily I doubt I would.

Where does this end?

Well quite, I think it's another case of what the regs allow vs. what's widely accepted as good practice. Do I like seeing four conductors in an MCB... not really, is it strictly against the regs... probably not providing the other provisions of the regulations are catered for (such as overload protection, division of circuits etc.).

In this case, I can see the logic behind the OPs reason for doing it, would I have done the same? Probably not. Do I think it's acceptable? Well clearly yes I do, I just don't think it's good practice and I would probably have just left it as it was.

I thought you'd pick that one! So at the origin of the installation, is the 'point of supply' the incoming cables connect to the head, the main OCPD, if so which side of it, or the point at which the tail connects to the OCPD? It doesn't say so again you cannot go definitive on that.

I'm just not trying to overthink it too much. A cable doesn't supply electricity, it carries it, so in this case I consider the source of power to be the protective device as the origin.

Why? I fundamentally disagree with this. I believe that it is absolutely the cabling that defines the circuit. Are you saying that a breaker with no cable connected is still a circuit? We are getting down the semantics of what the definition is and I think this is where you and I differ.

No, I would consider a breaker with no cable connected as a spare way.

The regs state specifically that this is allowed from a ring final circuit and so this is also a red herring.

Do they though? The only section that gives any definitive guidance about the topology of a ring circuit is 433.1.204, and it only specifies the topology in terms of acceptable cable loading.

The only place I've been able to find a statement to the effect it's allowed is in the OSG and it doesn't reference a regulation to back it up.

So why wouldn't you do it then?

Because I don't consider it to be good practice, plain and simple. Regardless of whether I consider it to be allowed by the regulations, it doesn't feel like it's something we should be indulging in at every opportunity

Finally, if you are correct and I am not, then what is the meaning of 314.4?

It basically requires that each final circuit is capable of being isolated completely from a single point. I can't say much more than that without saying pretty much what it says.

 

[Deleted member 26818]

Information regarding the origin of final circuits can be found in Appendix 15.

 

[markythesparky]

I’m sorry Spin, I cannot see the word ‘origin’ in App15 at all.