# Ring or radial . . . That is the question

Yes, but their point is you should have all the sockets on the ring in case one day you want to spur off one of them.

I.e. not have any spurs now.
But you dont have spurs off a radial. You can branch off a radial anywhere. Thus a branch is not a spur and one may sxtend the circuit from any point.

Unless you branch off each socket upstairs then you'd have to have two cables at all but one of the sockets anyway - which I don't see the issue in personally.

Branching from the design stage is wrong. As is introducing JBs in the design stage

Yes, but their point is you should have all the sockets on the ring in case one day you want to spur off one of them.
I.e. not have any spurs now.
But you dont have spurs off a radial.
Well, you can - branch and spur have the same definition.
However, I would call a radial spur one that has a smaller cable than the main circuit.

You can branch off a radial anywhere.
You can spur off a ring anywhere.

Thus a branch is not a spur
...but our definitions say they are the same.

and one may extend the circuit from any point.
extend with no restrictions, yes.

Branching from the design stage is wrong.
Why do you say that? I see absolutely nothing wrong with branches in a radial, any more than I see anything conceptually wrong with spurs from a ring. Indeed, if one socket on a radial connects to two downstream ones, there's no way one could decide that one of those connections was 'the radial' and the other was 'a branch'.

Kind Regards, John

Thus a branch is not a spur
...but our definitions say they are the same.
As often discussed, in the case of a radial one cannot define either 'a branch' or 'a spur' (as distinct from the radial itself). A radial may 'branch' (i.e. split into two at a 'Y-junction'), but (certainly if the CSA of both cables is the same), there's no way one can say that one of the cables is a continuation of the radial and the other is 'a branch' (or a spur). The simple fact is that 'a radial circuit' does not have to be linear - it can be 'branching'.

Kind Regards, John

We have definitions today which say what is a spur and what is a radial which have changed from when I started. I called a fused spur a radial, as it is a circuit in it’s own right as has it’s own over current device. But today with have fused spur, and non fused spur. We only call it a radial when the proactive device is in the distribution unit. Not a clue why the change?

But when designing an installation we have a number of considerations. For example, volt drop, ability to operate the protective device, cost, inconvenience, and many more.

So you have to consider distance, in some houses I have seen two rings split left and right side of house rather than upper and lower floor as this reduces the volt drop.

If we want the lights not to fail when a socket circuit is tripped we have to consider how to split circuits, using two RCD’s the normal way is to arrange the circuits so the lights and sockets for every room are from different RCD’s. So if the sockets have two circuits, then lights must also have two circuits unless all lights on one RCD and all sockets on the other, which would mean fault finding is a problem when there is only one RCD for all sockets.

You can use RCBO’s but that increases the cost. As to 4 mm² for radial well that depends how the cable is run. For reference method 100 4 mm² is only rated at 27 amp, so in the main it is 6 mm² for a 32A radial.

In the main the problem is when the 13A socket was designed it was designed to go with a ring final system, and once you leave that system you run into a host of problems, getting the thicker cables into the sockets is one of them, to have a 2.5 mm² ring and a spur from the ring also in 2.5 mm² is easy enough, but a 6 mm² radial with a 2.5 mm² spur one has a problem with getting the cables in the terminals although with MK it can be done, but also ensuring three cables of unequal size are clamped well enough.

The design current is also a consideration, it seems we are allowed to consider for volt drop that the design current in a ring final, or a 32A radial is 26A, that considers that 20A is drawn form centre with ring final, or end with radial, and the remaining 12A is equally drawn over the whole length. However if you reduce from 32A you still have that 20A drawn form centre with ring final, or end with radial, so the design current means with radials with 16A, 20A or 25A MCB’s they don’t have a design current allowing as much cable to be used as with a ring. Of course to a special you could select your own design current, so a 20A radial of say 30 meters could have one socket at the end, but 10 sockets near the origin, so you could design with 13A at end and rest within first 10 meters. However not likely that is really the case.

Designing a kitchen, or utility room, with tumble drier, washing machine, oven, dish washer, etc. You need to show this will not cause either too much volt drop, or regular tripping, or too much current on one end of the ring.

In the main we do the design in our head, we did a house like this before, and last time the loop impedance figures were well within, or just on the edge, so we consider splitting or combining circuits based on our experience, rather than sitting down and working it all out. With the unusual we may work it out, but rule of thumb is one reel of cable for one ring, never start a second reel and likely it will be OK.

It is hard to show that it has exceeded the volt drop, on paper we see that with an incomer of 0.35Ω we should at mid point on the ring final be getting 0.94Ω but even if 1Ω it would be hard to be 100% sure that this was due to too much cable or errors in reading. But as it gets further and further out there reaches a point where one has to concede that one has made an error and you have to compensate by making some changes.

With industrial units we have to be very careful, I had a shrink wrap machine fail to work on one socket, but was OK with other sockets, and it resulted in pulling in replacement cables. But getting caught out with domestic is rare. Even if it is wrong, who is going to find out?

The problem with a MCB is the magnetic part will either trip or not trip as the impedance increases, where with the fuse, it simply took a little longer. So with the 30A fuse (BS 1361) 1.15Ω was the figure quoted, but if exceeded then it just took slightly longer, but with a 32A MCB 1.44Ω is on the edge so volt drop could cause it to not operate on the magnetic part so we are told to reduce it to 95% of the calculated figure to ensure it is within limits, as to exceed it will reduce tripping times from 0.01 seconds to 10 seconds when the thermal part of the trip will work. This is why we have to measure the loop impedance, back in the day of the fuse, we could get away with so much more.

To have a radial to every room other than kitchen and utility room does have some advantages specially using RCBO for each, a fault only effects one room. But it will cost so much more.

As to 4 mm² for radial well that depends how the cable is run. For reference method 100 4 mm² is only rated at 27 amp, so in the main it is 6 mm² for a 32A radial.
No, it's not. Don't use method 100.

Last edited:
That certainly is a list of solutions looking for problems.

Almost none of those things will be encountered in the 'normal' domestic installation.

We have definitions today which say what is a spur and what is a radial which have changed from when I started. ....... Not a clue why the change?
What were the definitions of spur and radial when you started?

Kind Regards, John

A radial was a supply from an over current device where the size of the cable is the same from the over current device to the socket.
A Spur was from a cable of a larger size than the spur and only feed one device.
So the cable to a FCU could be a spur, or the FCU could be on the radial or ring, but the cable from a FCU is a radial and is a new circuit.
With that it means ALL spurs have only one device connected to them, as after the FCU it's no longer a spur.

The problem is we call it a ring final, and that word final means it is the last circuit of the system, if you use FCU's then the ring is no longer the final circuit as you have further circuits coming from the ring.

The referring to the cable after the FCU as a spur I still feel is wrong, however in BS7671:2008 in the appendix it is called a spur. It points to the radial from a FCU and states; "Spur using an FCU connected directly to the ring" and also "Spur circuit using an FCU" I did not think F was a vowel? so may be it's an error?, but it clearly calls the cable after the over current protection device a spur. May be this is why wiring after the FCU is not considered as a new circuit under the Part P rules?

"Spur. A branch from a ring or radial final circuit." now with a radial it can branch out at any point and I would only consider it a spur if the branch was a thinner cable than the main radial, but that is not in the description. Circuit. An assembly of electrical equipment supplied from the same origin and protected against over current by the same protective device(s). Well after the FCU it is protected by a different over current device to the radial or ring it is fed from, so it is a new circuit.

When I stated we spurred from the buzz bars to distribution units, that short run of cable was not protected by an over current device at the origin, always less than 3 meters, hence why I have stated a spur should be less than 3 meters, but a radial from a FCU can be as long as you want.

Now this brings in another odd thing, with a 4mm² radial I could fit a FCU and continue in 4mm² to a socket/s some distance away, the 13A fuse would reduce the loop impedance required and as a result allow the distance. Now that is a spur off the main radial but in 4mm².

So I would say yes we spur off a ring or radial with a FCU but the cable after the FCU is a radial.

It's only the words, what we can do has never altered, under amendment 3, I become skilled not competent, but my skill has not changed it is the definition which has changed.

A radial was a supply from an over current device where the size of the cable is the same from the over current device to the socket.
A Spur was from a cable of a larger size than the spur
I would agree with that - so would App 15.
and only feed one device.
Why only one device?

So the cable to a FCU could be a spur, or the FCU could be on the radial or ring, but the cable from a FCU is a radial and is a new circuit.
I would agree with that as well.
However, the position of the FCU is surely irrelevant; it still protects the whole spur - from overload if not SC; if no overload possible then it it not actually required.
Can a new circuit still not be a spur.

With that it means ALL spurs have only one device connected to them, as after the FCU it's no longer a spur.
Surely a spur (from main circuit) could have all sorts connected to it. It would still be a spur.

The problem is we call it a ring final, and that word final means it is the last circuit of the system, if you use FCU's then the ring is no longer the final circuit as you have further circuits coming from the ring.
True, but then BS7671 is full of errors.

The referring to the cable after the FCU as a spur I still feel is wrong, however in BS7671:2008 in the appendix it is called a spur.
Yes, but can't the new circuit be a spur?
It points to the radial from a FCU and states; "Spur using an FCU connected directly to the ring" and also "Spur circuit using an FCU" I did not think F was a vowel? so may be it's an error?,
I don't know about Flintshire but down here F begins with an E.
but it clearly calls the cable after the over current protection device a spur.
It is, isn't it? It can also be a circuit under the definition.
May be this is why wiring after the FCU is not considered as a new circuit under the Part P rules?
I don't think logic was involved in the notification requirements.

"Spur. A branch from a ring or radial final circuit."
It doesn't define 'branch' though.
now with a radial it can branch out at any point and I would only consider it a spur if the branch was a thinner cable than the main radial,
So would I and App 15.
but that is not in the description.
Clearly inconsistent, then so not worth debating, really.
Circuit. An assembly of electrical equipment supplied from the same origin and protected against over current by the same protective device(s). Well after the FCU it is protected by a different over current device to the radial or ring it is fed from, so it is a new circuit.
I agree but that doesn't mean the definition of spur is changed. Can you not have a spur from a spur?

When I stated we spurred from the buzz bars to distribution units, that short run of cable was not protected by an over current device at the origin, always less than 3 meters, hence why I have stated a spur should be less than 3 meters, but a radial from a FCU can be as long as you want.
No. you always mix up the requirements in 433 and 434.
It is protected by an OPD not at the origin so no different.

Now this brings in another odd thing, with a 4mm² radial I could fit a FCU and continue in 4mm² to a socket/s some distance away, the 13A fuse would reduce the loop impedance required and as a result allow the distance. Now that is a spur off the main radial but in 4mm².
Oooh, crafty. It may not be a spur; it could be the main circuit. How do you tell?
According to definitions it could be a spur or branch (no definition). Is a tree all branches apart from the trunk?
Therefore, a radial may, after the first junction, be all branches or spurs.

So I would say yes we spur off a ring or radial with a FCU but the cable after the FCU is a radial.
No, disagree, all branch - but still a new circuit.

It's only the words, what we can do has never altered, under amendment 3, I become skilled not competent, but my skill has not changed it is the definition which has changed.
That may be true but not necessarily so in all cases.

A radial was a supply from an over current device where the size of the cable is the same from the over current device to the socket. A Spur was from a cable of a larger size than the spur
I would agree with that - so would App 15.
It's just words. I personally find it a bit confusing to refer to part of a (branching) radial circuit as "a radial" in its own right - to me that implies that it is a separate circuit although, in the absence of an intervening additional OPD, it isn't. Although I accept it's not something one would do, if you agree with all that eric has written, does that mean that you would not regard a single 6mm² cable originating from a 2.5mm² ring (to supply sockets) as a "a spur"?
The problem is we call it a ring final, and that word final means it is the last circuit of the system, if you use FCU's then the ring is no longer the final circuit as you have further circuits coming from the ring.
True, but then BS7671 is full of errors.
Now it's more than "just words". As often discussed, if (per literal interpretation of BS7671 Part 2 definition) we are to believe that an FCU creates "a new circuit" (and in the absence of any definition of 'a circuit' in the Building Regs), we would have to conclude that adding a fused spur to a ring final is now notifiable in E&W, although still explicitly not notifiable in Wales.
"Spur. A branch from a ring or radial final circuit."
It doesn't define 'branch' though.
Indeed and, as often discussed, in the absence of a CSA change it is impossible to define 'a branch' (let alone a spur) of/from a radial circuit.
Circuit. An assembly of electrical equipment supplied from the same origin and protected against over current by the same protective device(s). Well after the FCU it is protected by a different over current device to the radial or ring it is fed from, so it is a new circuit.
See above, and implications in terms of notification.

My personal view is that (and per current usage), in relation to a ring, a 'spur' should have it's topolgical meaning (i.e. a connection from the ring which is not part of the ring), whether or not their is a CSA change or an intervening OPD (i.e. a fused or unfused spur). In teh case of a radial circuit, neither 'spur'nor 'branch' (as opposed to 'the radial' itself) really have any meaning in the absence of CSA changes and/or an intervening OPD. If there is a CSA reduction and/or an intervening OPD, I suppose there is a case for calling it a 'spur' from a radial circuit.

Kind Regards, John

A radial was a supply from an over current device where the size of the cable is the same from the over current device to the socket. A Spur was from a cable of a larger size than the spur
I would agree with that - so would App 15.
It's just words. I personally find it a bit confusing to refer to part of a (branching) radial circuit as "a radial" in its own right - to me that implies that it is a separate circuit although,
Whilst it is just words, by our definition, it is though.
It has different requirements for Zs, VD etc.
in the absence of an intervening additional OPD, it isn't.
True.
Although I accept it's not something one would do, if you agree with all that eric has written, does that mean that you would not regard a single 6mm² cable originating from a 2.5mm² ring (to supply sockets) as a "a spur"?
Well, it's different with rings, isn't it? I would consider it a spur or a branch.

The problem is we call it a ring final, and that word final means it is the last circuit of the system, if you use FCU's then the ring is no longer the final circuit as you have further circuits coming from the ring.
True, but then BS7671 is full of errors.
Now it's more than "just words". As often discussed, if (per literal interpretation of BS7671 Part 2 definition) we are to believe that an FCU creates "a new circuit" (and in the absence of any definition of 'a circuit' in the Building Regs), we would have to conclude that adding a fused spur to a ring final is now notifiable in E&W, although still explicitly not notifiable in Wales.
That's a good point about the BR not having a definition, so it's up to you whether to notify or not.

"Spur. A branch from a ring or radial final circuit."
It doesn't define 'branch' though.
Indeed and, as often discussed, in the absence of a CSA change it is impossible to define 'a branch' (let alone a spur) of/from a radial circuit.
What about they're all branches except the 'trunk'?

If there is a CSA reduction and/or an intervening OPD, I suppose there is a case for calling it a 'spur' from a radial circuit.
That's just my thinking - especially as we call it a fused spur -
but then, some people wrongly call the FCU itself a spur.

Whilst it is just words, by our definition, it is though. It has different requirements for Zs, VD etc.
I don't really understand that. No matter what one calls the bit of the cable concerned, requirements for both Zs and VD will remain the same. If there is an intervening OPD (which is a different matter), then that will change the Zs requirement - but even that wouldn't change VD requirement.
Although I accept it's not something one would do, if you agree with all that eric has written, does that mean that you would not regard a single 6mm² cable originating from a 2.5mm² ring (to supply sockets) as a "a spur"?
Well, it's different with rings, isn't it? I would consider it a spur or a branch.
So would I (in fact, I would call it a spur) - but by eric's definition, you would not be able to call it a 'spur', since it did not have reduced CSA.
That's a good point about the BR not having a definition, so it's up to you whether to notify or not.
.. except that the final arbiter of interpretation ('the Courts') would, in the absence of a definition in the Building Regs themselves, look for some other 'accepted definition' of a circuit - and would probably end up directly (or indirectly, via 'experts') looking to the BS7671 definition.
What about they're all branches except the 'trunk'?
One could say that - but I'm not sure that it would offer any benefit over simply saying that they were all parts of a (branching) radial circuit - which is how I regard them!

Kind Regards, John

I don't really understand that. No matter what one calls the bit of the cable concerned, requirements for both Zs and VD will remain the same. If there is an intervening OPD (which is a different matter), then that will change the Zs requirement - but even that wouldn't change VD requirement.
The Zs requirement for a 3A fused circuit is a lot higher than that for 32A.
May be I'm generalising too much but the VD for a fan or a light may be less stringent than that for main circuit which may supply all sorts of things.

So would I (in fact, I would call it a spur) - but by eric's definition, you would not be able to call it a 'spur', since it did not have reduced CSA.
As you said, just words.

.. except that the final arbiter of interpretation ('the Courts') would, in the absence of a definition in the Building Regs themselves, look for some other 'accepted definition' of a circuit - and would probably end up directly (or indirectly, via 'experts') looking to the BS7671 definition.
We shall never know.

One could say that - but I'm not sure that it would offer any benefit over simply saying that they were all parts of a (branching) radial circuit - which is how I regard them!
I didn't realise benefits were the consideration. It's just semantics.

As a spur is defined as a branch, it doesn't matter.

#### DIYnot Local

Staff member

If you need to find a tradesperson to get your job done, please try our local search below, or if you are doing it yourself you can find suppliers local to you.

Are you a trade or supplier? You can create your listing free at DIYnot Local

Replies
9
Views
13K
Replies
16
Views
4K
Replies
17
Views
2K
Replies
9
Views
11K
Replies
30
Views
21K