Can I join 2 radials to make a ring?

[Longhorn]

I have 2 radial lines coming out of our CU in 2.5mm that used to run storage heaters, but have now been replaced with a double outlet on each (they are both powered from the same MCB). I have a couple of questions:

Can I join these together to turn it into a ring?
If so, would this constitute notifiable work?

Thanks in advance.

 

[Andy78]

I have 2 radial lines coming out of our CU in 2.5mm that used to run storage heaters, but have now been replaced with a double outlet on each (they are both powered from the same MCB). I have a couple of questions:

Can I join these together to turn it into a ring?
If so, would this constitute notifiable work?

Thanks in advance.

Any work done should be compliant with the wiring regulations. This includes testing to ensure safety and a certificate producing to prove this.

Notification is a separate and additional requirement to those of the wiring regulations.

 

[westward10]

There is no reason why this cannot be done but I see you have no electrical qualifications, why is it necessary to do this.

 

[Longhorn]

Thank for the replies. Westward10, you're absolutely right I have zero qualifications, hence asking here whether it is theoretically possible. It would not be my intention to do this myself, I'm just throwing the question out there to see if I'm overlooking something at this planning stage.

As a bit of background, the whole house is wired with radials. These two particular ones are relatively modern and my assumption would be that they would be infinitely more flexible with regards to extending their usefulness if they were part of a ring. My theory is that they could be joined together and have other sockets added, thus creating a modern ring where once there were just 2 radials.

 

[westward10]

Yes this can be done whether this makes it more flexible is debatable as radials themselves can be added to much more easily. Why not just split them onto their own 20A protective devices. What locations would they be serving.

 

[Longhorn]

Thanks Westward10. When you say radials can be added to more easily, do you mean adding more radials or adding more sockets to an existing radial? the latter seems to me akin to spurring off of a spur so I assume you mean the former.

I'm trying to keep disruption to a minimum so adding more radials would be a pain in the rump. Running a cable from the end of one one radial to the other goes right by where we could use another couple of sockets, so this seems to be the easiest way forward and something of a win-win.

 

[westward10]

If you create a ring final circuit and fit a 30/32A protective device you restrict yourself as to the amount of outlets on a single radial spur unless you fit a 13A fuse connection unit or alternatively you have to incorporate them into the existing ring final. A 20A radial is not so restrictive as you can just add on at various points as you like assuming the cables are suitably rated.

 

[Longhorn]

So if a spur was taken off of the existing 2.5mm radial and the radial was protected by a 20A MCB at the CU this would be perfectly safe and within regulations?

 

[westward10]

Yes assuming the cable is suitably rated.

 

[westward10]

You need to assess the potential loading of your new sockets, if they are in bedrooms or the lounge then loading is likely to be low so 20A would be ample but if it is a kitchen and heavier appliances are using them then a ring final circuit maybe a better option. A good electrician should give the correct advise.

 

[Sparkingmad]

Longhorn in your post you say that joining the 2 radials together to form a ring the wiring route runs past the room/s you will be wanting to add extra sockets to. So extending 2 radials to get sockets in positions you want or need is infact no harder than joining them together to form a ring. The question you also have to ask yourself is rcd protection for the modified circuit or circuits. The 2 points were they night storage heaters? Are they live or are they still on off peak supply? I would recommend going the same route as westward suggests not knowing what board you have.

 

[Longhorn]

Sparkingmad; you're absolutely right, simply extending the radials to add extra sockets would be marginally easier than creating an actual ring as it would save me joining them together, but I didn't know this was possible. My theorising, and the reason for the question, was based on my longterm believe of 2 things: 1) You can't spur off of a spur, and 2) the ring main was designed just after the war when copper was at a shortage and a ring can be made in 2.5, thus saving copper (I infer from this that radials were heavier gauge before the ring was invented).

Based on those 2 points I reasoned that as these radials are 2.5 (and thus similar to a spur in handling capability, albeit directly from the CU and not from an existing socket) it would be safer, and indeed more in keeping with ring spec, to just join them together and form a ring, which would also allow me to add more sockets ad hoc.

As far as loading goes, they are all upstairs right now, so mostly bedroom stuff (phone chargers, toothbrush charger, etc, and periodically the vacuum). Heaviest draw would be my iMac and standard lamps plugged into the walls. Although there is a possibility to incorporate a connection to my shed, so there may be a little heavier draw on it with certain power tools.

 

[Longhorn]

Yes assuming the cable is suitably rated.

Is the rating of a cable simply a question of its size? It's 2.5, does that make it suitably rated? Or does one 2.5 differ from another?

Please forgive the apparent schoolboy questions, I'm just trying to get a handle on how much can be achieved with the minimum disruption. These radials have been sympathetically installed and are pretty unobtrusive. If I can extend their usability and increase our safety in the house without ripping them out and effectively putting them straight back again I'm up for that.

 

[westward10]

2.5 is generally adequate although installation methods will be a factor. Altering a circuit will require some testing to ensure the existing and new parts provide suitable circuit protection and the provision for additional protection by way of a suitable rcd.

 

[telectrix]

as said before, easier to add to a radial as there's no restriction like with spurs off a ring. the other advantage of extending the radials and leaving them as radials is that if you have a fault on one and need to isolate it till rectified, the other one still works. put each on a 20A MCB

 

[Longhorn]

...put each on a 20A MCB

Each on its own MCB or can they share?

 

[telectrix]

better on separate MCBs so as to split them, this helps with division of circuits as per 17th edition, esp. if each is on a separate RCD/RCBO.

 

[Sparkingmad]

Each radial goes on its own mcb longhorn

What he was saying is if one radial has a fault you leave that one off and the other still works, if it was a ring main you would lose all the sockets until a professional can get out and test and locate fault and fix it

 

[Longhorn]

I don't have nearly enough MCBs for that. Apart from the kitchen (ring) the whole house is on radials, many of which are sharing MCBs. This is another reason for the ring idea. If I could join those newer radials to form a ring I could eradicate many of the older radials and free up some MCBs.

 

[Sparkingmad]

Sparkingmad; you're absolutely right, simply extending the radials to add extra sockets would be marginally easier than creating an actual ring as it would save me joining them together, but I didn't know this was possible. My theorising, and the reason for the question, was based on my longterm believe of 2 things: 1) You can't spur off of a spur, and 2) the ring main was designed just after the war when copper was at a shortage and a ring can be made in 2.5, thus saving copper (I infer from this that radials were heavier gauge before the ring was invented).

Based on those 2 points I reasoned that as these radials are 2.5 (and thus similar to a spur in handling capability, albeit directly from the CU and not from an existing socket) it would be safer, and indeed more in keeping with ring spec, to just join them together and form a ring, which would also allow me to add more sockets ad hoc.

As far as loading goes, they are all upstairs right now, so mostly bedroom stuff (phone chargers, toothbrush charger, etc, and periodically the vacuum). Heaviest draw would be my iMac and standard lamps plugged into the walls. Although there is a possibility to incorporate a connection to my shed, so there may be a little heavier draw on it with certain power tools.

This is way way before my time and the cable was imperial and not 2.5mm. I think the recommendations then still allowed the use of the BC plug adaptor to connect your iron to the light pendant . Plus the wire was tin plated back then and let's be honest you would only have had 1 maybe 2 single sockets in each room. Most people now have more sockets behind their TVs than they had in the whole house in the 40's and 50's. I'm sure there are a few here that remember those days

 

[Longhorn]

Well, I guess that's true of many of us here. If I'm safe just to extend the radials ad nauseam then it looks like we're in business. I'll make sure there's a 20A MCB on each.

 

[telectrix]

Well, I guess that's true of many of us here. If I'm safe just to extend the radials ad nauseam then it looks like we're in business. I'll make sure there's a 20A MCB on each.

best option would be a bigger CU. if you have. say. a 6 way, get a local spark to quote you for a 10/12 way dual RCD or RCBO board.

 

[Longhorn]

The board we have is a 10-way dual RCB MK board.

 

[telectrix]

that should be big enough for most houses. where you live? fawlty towers?

 

[Sparkingmad]

Might be worth getting a sparks in to test some of the other radials you may be able to convert some of those to ring mains to free up some room in that board. Lol

 

[Longhorn]

We seem to be going around in circles.

There are 10 MCBs, but one is the kitchen ring, one is the cooker, one is the alarm. So now we're down to 7. If each radial should have it's own MCB then theoretically we should have no more than 7 radials. Surely it would be better to have, say, an upstairs ring and a downstairs ring and maybe an outbuilding ring, and still have 4 MCBs left over for other things?

If I can connect the two newer radials as referenced in my original post I can add at least 3 more sockets along the way in very usable places, I just wanted to know it it was safe and legal to do this and whether it constituted notifiable work.

 

[telectrix]

altering circuits is not notifiable. a new circuit is.

 

[Deleted member 26818]

When rings were first introduced, sometimes two 15A radials were joined together to make the ring. If you were lucky, the two 15A fuses would be replaced with one 30A fuse.

 

[Mark42]

Hi Longhorn
What no one has explained yet is why radials can be safer: what you've heard about not spur-ing off spurs only applies to ring circuits, because rings are protected at 32A, so long lines of 2.5mm cable running spurs might mean - if multiple loads are plugged into the spurs - overloading that run of single 2.5mm cable. Genuinely possible, and genuinely dangerous.

Whereas radials on 2.5mm are never protected above 20A (sometimes 16A) so even if you have a hundred sockets on the thing, you can't overload the cable as the breaker would trip first. (Assuming the circuit is not stupidly long, but that's enough detail for now.)

Personally I dislike ring final circuits, and consider them outdated. Others will disagree. In your case I'd definitely go for keeping the two existing radials. You may add as many extra sockets locally as you wish. Make sure you have (or add) 30mA RCD protection. And do NOT connect the two radials together at any point.

Note: In my own house (which is 500 years old and made of wood) I did the opposite to what you're suggesting. I inherited only two 32A rings for the whole place. I started by splitting these rings, junking the 32A MCBs, and converting them to four radials, each with a 16A RCBO.

 

[Deleted member 26818]

Outdated?
If RFCs are outdated, what does that make Radials, which RFCs replaced?
It's not fashion week.

 

[Octopus]

Note: In my own house (which is 500 years old and made of wood) I did the opposite to what you're suggesting. I inherited only two 32A rings for the whole place. I started by splitting these rings, junking the 32A MCBs, and converting them to four radials, each with a 16A RCBO.

I'm guessing you didn't have much loading on them.................. 20A RCBO's may have been a better approach!

 

[Mark42]

I'm guessing you didn't have much loading on them.................. 20A RCBO's may have been a better approach!

Of course.
But whoever put the circuits in run the cables in and out of wattle and daub, up, down and all over the place, with tight bends: I just didn't trust the bloke, whoever it was.
I wanted to be sure that there was no chance, ever, of these hidden and buried wires running hot.
I added new 20A radial circuits for heavier loads, took the immersion heater off the old ring (aargh!) and installed all new kitchen circuits.

 

[Mark42]

Outdated?
If RFCs are outdated, what does that make Radials, which RFCs replaced?
It's not fashion week.

Indeed they did, to save copper.

But 2.5mm is now plentiful, cheap, and WWII is over

 

[kingeri]

Ring finals are not outdated, they are just another option for sparks to use where appropriate. These days I do install more radials than I used to, but rings are very versatile circuits, and where the building layout suits I still use rings when I need the 32A. Mainly kitchens and utilities these days. I don't like 32A 4mm^2 radials, the CCC is too close to the edge. The 32A 2.5mm^2 ring final has stood the test of time, and is far more flexible with installation methods. Fine, 20A radials for living rooms, bedrooms etc. are no problem, but for those areas which need higher current ratings, the ring final is still king for me.

 

[Mark42]

Ring finals are not outdated … rings are very versatile circuits, and where the building layout suits I still use rings when I need the 32A. Mainly kitchens and utilities these days. …

I’m not saying you’re wrong. It’s a possibly matter of personal preference.

Although I would argue that in kitchens and utility rooms, separate circuits for dishwasher, washing machine, tumble dryer, microwave, coffee machine and so on are preferable.

I concede that cost and time may be a consideration in commercial work, but that’s not the point I’m making.

… The 32A 2.5mm^2 ring final has stood the test of time, and is far more flexible with installation methods.

My main niggle with 32A rings is that a single point of failure (broken ring on one conductor - and surely every electrician has found this at some time) may result in a potentially dangerous cable overload which could sit undetected for years.

… I don't like 32A 4mm^2 radials, the CCC is too close to the edge.

I agree. Horrible, pointless things and awkward to connect accessories if the boxes aren’t deep enough. No argument there.

 

[Andy78]

My main niggle with 32A rings is that a single point of failure (broken ring on one conductor - and surely every electrician has found this at some time) may result in a potentially dangerous cable overload which could sit undetected for years.

I have found plenty of non-continuous ring conductors. I have never discovered any thermal damage to cable due to overload caused by this lack of continuity though.

It's a potential hazard that exists only in theory in my experience. It would be interesting to see a bench test however and see how 2.5mm cable performs in this overload situation.
Wonder if JW has ever done this ?

 

[Mark42]

I have found plenty of non-continuous ring conductors. I have never discovered any thermal damage to cable due to overload caused by this lack of continuity though.

It's a potential hazard that exists only in theory in my experience. It would be interesting to see a bench test however and see how 2.5mm cable performs in this overload situation.
Wonder if JW has ever done this ?

Hmm. I don't entirely disagree, but to extrapolate your argument would be to say it's OK to protect buried 2.5mm T&E at 32A. Would you ever do that?

It would indeed be interesting to run, say, two old-skool 3kw fires down a long bit of 2.5 T&E to see what actually happens. I do collect such things, but have enough on this week!

 

[Andy78]

Hmm. I don't entirely disagree, but to extrapolate your argument would be to say it's OK to protect buried 2.5mm T&E at 32A. Would you ever do that?

It would indeed be interesting to run, say, two old-skool 3kw fires down a long bit of 2.5 T&E to see what actually happens.

I would, and have installed 2.5mm cable protected by a 32A device. As a spur from a ring.

2 x 3kW fires on a bit of 2.5mm I would expect to be fine almost indefinitely with regards the cable. It is certainly just under the tabulated CCC of the cable.
If both were plugged into a double socket however, I would expect that to be the weak link as they are only rated at 20A.

 

[Mark42]

I would, and have installed 2.5mm cable protected by a 32A device. As a spur from a ring.
2 x 3kW fires on a bit of 2.5mm I would expect to be fine almost indefinitely with regards the cable. It is certainly just under the tabulated CCC of the cable.
If both were plugged into a double socket however, I would expect that to be the weak link as they are only rated at 20A.

Ha! Yes, correct of course, but you know that's not what I meant.

And sure, I once fused 0.5mm at 200A. For indicator lights in a main incomer switch panel. But I wouldn't use 0.5mm for 200A distribution, even if the budget was tight.

In the very specific case of a short spur or two, you are absolutely right in that the risk really is only theoretical, as no sensible person plugs two 3kW fires into the same double socket, or even in the same room. The radiated heat alone would set the curtains on fire before they needed to worry about the wiring.

My example was ill thought-out. How about a 3kW fire, a 1.5kW washing machine, a 1.5kW dishwasher, then a 1.5kW kettle on a broken long ring buried in insulation? Unlikely, but not impossible. And if it's said that such loads are very rare in practice, then why do we need 32A anyway?

I don't disagree that with quality modern cable, it would probably be OK for years. But I still say that any design which may need to take advantage of over-cautious specs and luck, is bad design!

 

[Andy78]

Ha! Yes, correct of course, but you know that's not what I meant.

And sure, I once fused 0.5mm at 200A. For indicator lights in a main incomer switch panel. But I wouldn't use 0.5mm for 200A distribution, even if the budget was tight.

In the very specific case of a short spur or two, you are absolutely right in that the risk really is only theoretical, as no sensible person plugs two 3kW fires into the same double socket, or even in the same room. The radiated heat alone would set the curtains on fire before they needed to worry about the wiring.

My example was ill thought-out. How about a 3kW fire, a 1.5kW washing machine, a 1.5kW dishwasher, then a 1.5kW kettle on a broken long ring buried in insulation? Unlikely, but not impossible. And if it's said that such loads are very rare in practice, then why do we need 32A anyway?

I don't disagree that with quality modern cable, it would probably be OK for years. But I still say that any design which may need to take advantage of over-cautious specs and luck, is bad design!

The new example given is not really relevant to highlighting the shortcomings of a ring final circuit either.
A 2.5mm twin and earth cable buried in insulation would likely not have been installed to comply with the regulations as this installation method would likely see the CCC below the required 20A at the end of the cable calculation.

I do get your point though and it's very refreshing that you are debating this issue with a bit of thought. Most ring haters really don't know why they do so.

As I say I'd like to see prolonged bench tests of 2.5mm cable under various overload conditions with arrangements simulating non-continuous ring final arrangements of various permutations before I was convinced there is an inherent problem with the standard arrangement.

 

[Sparkingmad]

Just to throw another curve ball 2.5mm t&e will carry 27A, how many sparks can honestly say they have checked at what overload current their protective devices operate at?
If you haven’t I suggest you try it for a giggle when you have time and inclination. 16A mcb’s operating after extended time with loads reaching 24+Amps. 32A mcb taking almost 60A before operating on overload.

 

[Richard Burns]

Just to throw another curve ball 2.5mm t&e will carry 27A, how many sparks can honestly say they have checked at what overload current their protective devices operate at?
If you haven’t I suggest you try it for a giggle when you have time and inclination. 16A mcb’s operating after extended time with loads reaching 24+Amps. 32A mcb taking almost 60A before operating on overload.

All this is accounted for in the cable rating values as it is a design characteristic of the MCB.
The bad situation is loading a cable at less than 1.45xIn for extended periods (small overload of long duration).
The 27A cable on a 32A MCB would have to be carefully overloaded to have a short term problem but the cable would have a reduced life in the long term as it would be running closer to its maximum temperature.
All the loading examples above will have little effect as they are short term loads, less than an hour. Having 7kW of lighting running 24 hours a day would be the killer. Heaters all have thermostats and switch off after a while, so the cable would warm then cool; possibly bad for thermal cycling but not too bad overall.

 

[Sparkingmad]

All this is accounted for in the cable rating values as it is a design characteristic of the MCB.
The bad situation is loading a cable at less than 1.45xIn for extended periods (small overload of long duration).
The 27A cable on a 32A MCB would have to be carefully overloaded to have a short term problem but the cable would have a reduced life in the long term as it would be running closer to its maximum temperature.
All the loading examples above will have little effect as they are short term loads, less than an hour. Having 7kW of lighting running 24 hours a day would be the killer. Heaters all have thermostats and switch off after a while, so the cable would warm then cool; possibly bad for thermal cycling but not too bad overall.

I agree Richard but the point I was making is that even though we may do the calculations for Design current and taking into account correction factors to determine our protective device rating to protect our cables and the circuit, the protective device can still allow continued overloading for more than just a short duration. The example I gave above would allow that circuit to operate at currents higher than it was intended to for longer periods than it should be possible too. It would not be unreasonable IMO to expect that mcb to allow that circuit to be overloaded by an extra 10A for quite a long time before it operated and continued overloading like this is possible until the mcb gets tired or weakened what is the unseen damage to the cable and detrimental effects could or would happen?

 

[Richard Burns]

I agree Richard but the point I was making is that even though we may do the calculations for Design current and taking into account correction factors to determine our protective device rating to protect our cables and the circuit, the protective device can still allow continued overloading for more than just a short duration. The example I gave above would allow that circuit to operate at currents higher than it was intended to for longer periods than it should be possible too. It would not be unreasonable IMO to expect that mcb to allow that circuit to be overloaded by an extra 10A for quite a long time before it operated and continued overloading like this is possible until the mcb gets tired or weakened what is the unseen damage to the cable and detrimental effects could or would happen?

This problem is most likely to occur on socket circuits where the load cannot be determined in advance and where say a standard 32A ring is installed in a commercial kitchen, a situation I have seen, then the circuit breaker could be routinely overloaded, if the load is over 46A then the circuit breaker will trip in less than 1 hour but during that hour the cable will be getting warm, but because it is only for one hour it should not suffer much (remembering that the cable can take its rated current 24hrs a day 365 days per year without serious degradation), it is the constant resetting of overloaded breakers that will take its toll on the cables and breaker and start to degrade them through thermal damage.