Replacing kitchen ring with 2 radials

Joined
9 Dec 2008
Messages
17
Reaction score
0
Location
London
Country
United Kingdom
Hi,
Posted 5 years ago about the electric installation in my kitchen... And I'm back about the same thing! Seems like damp walls have potentially damaged cables: an electrician visited, and confirmed i've got faults in 2 places. One of them can be easily accessed and fixed, the other requires removing kitchen worktop, chanelling into a damp wall, making good, etc...

He suggested an alternative which is to make the ring into 2 radials. See below a little diagram.

My worry is that he mentioned that each radial will have a 16A MCB (and get rid of the old ring's 32A).
I've read lots of posts, for example:
"A 2.5 radial can be protected by a 20A MCB subject to install conditions (method, length etc)"
Read more: //www.diynot.com/forums/viewtopic.php?p=667021#667021#ixzz3GKnLVWZn

"A 2,5 radial must be on no more than a 20A with a floor area covered of 50m2 or less."
Read more: //www.diynot.com/forums/viewtopic.php?t=70804#ixzz3GKonC3Qp

So I'm thinking a 20A MCB could be used. Is this correct?

With 16A, I'm worried that the MCB on Radial 1 could be triggered if the washing machine and dishwasher heat at the same time and we suddenly make a cup of tea and toast a slice of bread.
I don't have the option to lay a new cable and I don't want to dig the walls to repair the ring. If it is really a problem, only solution I imagine is to get rid of that double socket.

We would then be forced to plug kettle and toaster on Radial 2. But then maybe I would have to move Oven on its own circuit to ease the load (it's an easy route to the CU).

What do you guys think?

(I did engineering at school 15 years ago, so I'm technically minding and even though I won't do the work, I like to understand what tradesmen do in my house and the logic behind their decisions!)

Thanks,
Thierry

----------------------

All cables 2x2.5mm2+1.5mm2

Kitchen is roughly 16m2.

All other appliances (fridge freezer, combination microwave oven, induction hobs) have their own dedicated circuits.

Radial 1 :
- washing machine
- dishwasher
- 1 double socket (in which the kettle is plugged rated 3kW, and once a week the toaster at 900W)

dblSocket--1m--dishwasher--1m--washing machine---9m---CU

Radial 2:
- extractor fan
- normal oven
- 6 double sockets

dblSock--5m--dblSock--1.5m--dblSock--2m--dblSock--2m--dblSock--1m--fan+oven--5m--CU
 
Sponsored Links
A 16A radial for a kitchen is a poor solution for the reasons you have described. 20A would be better, and can normally be used with 2.5mm cables.

What is the problem with the existing circuit? Damp walls on their own do not cause damage to plastic insulated cables, so if there is a fault with a section of cable it suggests it has been physically damaged somewhere.

Can't really understand why it would be necessary to remove worktops to replace cables either - new cables can be run underneath the units and brought up behind the cupboards, the only damage being to a very small section of wall between the worktop and socket.

If you have other circuits for freezer etc., it's certainly possible that one or more of those could have a socket or two added to them.
 
A 16A radial for a kitchen is a poor solution for the reasons you have described. 20A would be better, and can normally be used with 2.5mm cables.
Indeed, to be frank, I'm not particularly happy with even 20A radials in kitchens, since that's the place where most of the 'large loads' are, and just two of them, let alone more, can theoretically be too much for a 20A circuit.
What is the problem with the existing circuit? Damp walls on their own do not cause damage to plastic insulated cables, so if there is a fault with a section of cable it suggests it has been physically damaged somewhere. ... Can't really understand why it would be necessary to remove worktops to replace cables either - new cables can be run underneath the units and brought up behind the cupboards, the only damage being to a very small section of wall between the worktop and socket.
Yes, I agree - I thought all that myself. It would be helpful if we could have some more information about exactly what damage to the circuit the electrician thinks has occurred.

Kind Regards, John
 
If one does the calculations splitting a 106 meter ring into two 53 meter does not work the main reason is we are allowed to assume the 32 amps is split into 20 amp at centre and 12 amp even spread plus we are using 2x21 amp rated cables with a 32A MCB so we correct for that as well.

So step one when splitting a ring is to measure the loop impedance at the two ends and calculate the volt drop from those readings with different loads and from these readings we can work out what the maximum load permitted will be.

With RCD protection we are only really worried about volt drop but without that protection we also must ensure enough current can run to open the MCB being used with the magnetic part of the device.

We should when measuring the loop impedance measure both earth - line and neutral - line and work on the highest of the two. But in real terms with RCD protection the earth - line loop impedance can be much higher than the line - neutral reading.

In proper Blue Peter style here is one I did earlier entering the figures you can test to see what the maximum size will be. Default is the 106 meters of a ring. Change the loop impedance to 1.1 Ω and swap from ring to radial and move design current to 20 amp and the cable length is just 31 meters keeping the 11 volt drop permitted not 53 meters. Swap the loop impedance to 1.28 and design current to 16A then again 11 volt drop and we have 40 meters of cable permitted. Move to 1.8 Ω and 10 amp design current then jumps to 67 meters.

So it is possible when splitting a ring one may have one leg with a 10A MCB and other leg with a 20A MCB it's all down to the loop impedance measurement.

Without those loop impedance measurements or prospective short circuit current which amounts to same thing trying to work out what size MCB can be used is impossible.

Splitting the ring to two radials will increase the loop reading but if we assume no RCD then the earth loop impedance for a B type MCB is calculated at 5 times the current rating so simple maths a B32 MCB is calculated at 160 amp so simple ohms law 230/160 = 1.4375 ohms lets call it 1.44 ohms. With a 16A or B16 then 2.875 lets call it 2.9 ohm and with 20A then 2.3 ohm. So with no RCD these are the maximum figures it would be dangerous to exceed them.

With a RCD we are only looking at volt drop and although the permitted is 11 volt if your supply voltage was on the high side then you could exceed this limit without any real problem.

Off the top of my head I would say the fridge and freezer are the most likely items to have a problem with volt drop. The compressor can't start under load so there is an auto resetting overload which should it try to re-start under load will open and the time it takes to re-close should be enough for the pressure to drop before it re-tries. Except with those using inverters. So a volt drop can cause the fridge or freezer not to start this is why most state in the instructions they should not be powered with an extension lead.

If the overload device is operated too often it fails and I have seen this a lot when working in Algeria with AC units. We were always told when turning off generators we were to leave 5 minutes before turning back on again so as not the stress all the overloads in the AC units.

I think it is unlikely however even with the massive 28 volts it would drop with a 2.3 Ω reading using a B20 MCB that it would really cause too much of a problem. However the electrician has to follow the regulations and really he has to follow those guide lines.

Now I am going to completely turn around the argument. When my son lived on a narrow boat he was charged for power by the size of the MCB fitted to the shore supply. With an inverter drive washing machine set at coolest wash it would trip out the B4 MCB but would hold with a B6 MCB even though it would draw 12 amp when heaters cut in. The heaters were just not on for long enough to trip the 6A MCB. So even with a B16 MCB it is unlikely it will ever trip. However that also means with a B16 MCB the volt drop could exceed the calculated value so in real terms be it a B16 or B20 MCB the volt drop and the resulting problems with a fridge or freezer are in real terms the same.

Clearly the ELI still needs to be measured and should not exceed the values Fuses to BS 3036 we have to just take the book values can't really work these out. 30A = 1.09Ω and these values will be entered on the minor works certificate. Because the electrician has to complete the minor works certificate should he not comply with the regulations he is leaving documented proof he was aware of the fact. As a result many have that sharp intake of breath and jobs worth comment.
 
Sponsored Links
Thanks for the replies.

flameport/JohnW2: "What is the problem with the existing circuit? Damp walls on their own do not cause damage to plastic insulated cables, so if there is a fault with a section of cable it suggests it has been physically damaged somewhere."

The electrician told me about a neutral to earth fault.
In terms of physical damage, I can't say for sure. All I know is that workmanship at installation wasn't great. They were errors in the render they used, so they had to pull everything out including the cables and re-render. This might have slightly damaged the cable as they pulled them? Who knows. Also, some cables are squeezed between the wall and the floating wood flooring. Maybe the flooring expansion and contraction is damaging the cable? hard to tell again. Visually, neither the electrician nor I saw anything.
I assumed it's damp as the walls are constantly damp, but the electrician did say that insulated cables shouldn't be affected by damp (unless there is an existing cut into the cable and the lime of the plaster get into the copper over a long time).

flameport: "Can't really understand why it would be necessary to remove worktops to replace cables either - new cables can be run underneath the units and brought up behind the cupboards, the only damage being to a very small section of wall between the worktop and socket."

You are completely right. It's the upstand, not the worktop, that needs removing (a 10cm glass band glued to the wall). I just wanted to avoid possible damage to that, and the whole making good palava (making sure tradesmen put waterproof render and not bonding for example), avoid if possible.

ericmark: If one does the calculations....
That was quite something to try and understand!
In my house, I've got 2 RCDs protecting 2 halves of all circuits.
In my case,Radial 1 is about 11m. Using your volt drop calculator, if the MCB is 20A, that means design current 20A. Trying to keep Volt drop under 11v. A Zs of 1.1ohms gives a max length of 31m, and with my 11m I'm well within that, so it's just a matter of making sure the Zs is indeed 1.1ohm or less.
My Radial 2 is 17m, so the same logic applies.
I assume, the shorter the cable, the more chance of having a small Zs.
Are you saying the electrician should mesure Ze and Zs, and from that, work out the size of the MCB? When he visited, he certainly didn't. Maybe he'll do it when he does the minor work certificate, but then it's too late to put a 20A MCB instead of 16A!

So... if the Zs is satisfactory, would 2 radials on 20A MCBs be a) legal and b) not constantly tripping ?

Thanks again.
 
There is a problem in that until he splits the ring he can't measure the loop impedance so he has to guess then check his guess was correct. He could in theroy measure the resistance and calculate but in practice likely he will just take a chance and swap the MCB if it shows as not complying.

Length of cable can be a surprise where it goes 1/3 way into room before coming down to next socket.

So easy way out is 16A MCB's as unlikely these will have a problem with the loop impedance. However once split with the B32 still in place he could measure the loop impedance in seconds and then select the size of breaker to use.

To use a low ohm meter and calculate is time consuming to plug in a loop impedance tester is quick and easy so although we should do dead tests we often just do the live test.

I would not really expect any problems using B16 MCB's but as long as it tests out then using B20 is not a problem.

Since RCD protected it is really only the volt drop we are worried about and in real terms even if the volt drop is too high it is unlikely to cause a problem so if it was my house I would fit B20 MCB's what ever the loop impedance showed.

But I am pointing out he could have a valid reason for saying it has to be a B16 and without the loop impedance readings we can't say if B20 is technically permitted or not.

If I was doing the job and we could make the system live then I would measure loop impedance of ring which would give a good idea what it is likely to be when split. However since there is a fault likely he can't measure loop impedance until after fault it removed. Except with a low ohm meter and calculator which will likely cost you more in time than cost of swapping MCB if there is a problem once it's all up and running.
 
flameport/JohnW2:.... The electrician told me about a neutral to earth fault. In terms of physical damage, I can't say for sure. All I know is ..... I assumed it's damp as the walls are constantly damp, but the electrician did say that insulated cables shouldn't be affected by damp (unless there is an existing cut into the cable and the lime of the plaster get into the copper over a long time).
I suppose anything is theoretically possible. You certainly could not get a neutral-earth (or any other type) of fault in a cable because of damp unless it had suffered mechanical damage, which not only completely penetrated the outer sheath but which also completely penetrated the insulation of the neutral conductor. Such damage is extremely unlikely in the absence of a direct penetrating injury (e.g. by nail, screw or drill). Do I take it from what you say that your electrician has isolated the piece of cable between two sockets and confirmed that this piece of cable (and only this piece of cable) is responsible for the neutral-earth fault?

You indicated that there were initially two faults, one of which was accessible and could be rectified, but this other one which was less accessible for repair. What was the nature of the fault/damage/whatever that was accessible for repair?
ericmark: If one does the calculations....
That was quite something to try and understand! ... Are you saying the electrician should mesure Ze and Zs, and from that, work out the size of the MCB? When he visited, he certainly didn't. Maybe he'll do it when he does the minor work certificate, but then it's too late to put a 20A MCB instead of 16A!
Whilst most of what eric wrote is theoretically correct, none of it is likely to be an issue in relation to any normal house. With the sort of cable lengths likely to be seen in a normal house (and certainly the lengths you mention in yours), a 20A radial circuit using 2.5mm² cable is most unlikely to be a problem, in terms of either Zs of voltage drop. Your electrician would certain have to measure Zs (and calculate voltage drop) to confirm that it was acceptable (as he would for any circuit), but it is very unlikely that it would not be OK. That leaves us uncertain as to why he has advised 16A rather than 20A MCBs (indeed, depending upon the routing of the cable, if one is available for your CU, even 25A MCBs might be acceptable). The simplest way to understand his thinking is obviously to ask him - it could be that, in haste, he simply divided 32A by two and got 16A, without considering that 20A would probably also be OK!
So... if the Zs is satisfactory, would 2 radials on 20A MCBs be a) legal and b) not constantly tripping ?
Provided only that the current-carrying-capacity of the cable, with the installation method used, is at least 20A (which was already a requirement when it was a ring, so is hopefully true!), it would certainly be compliant with regs (hence ‘legal’); as I’ve said, if the cable installation method allowed it (primarily, none of the cable buried in insulation), even a 25A MCB might well be compliant.

As for tripping, that obviously depends upon what get plugged in. As I said before, I’m personally a bit uneasy about even 20A circuits in kitchens, particularly when an oven is being fed by them - and the difference between 16A and 20A is not all that great. In terms of ‘overloading’ the circuit, just two 2.5kW loads would exceed 20A, and three 2kW ones certainly would. As for tripping, that’s much less likely. A 20A MCB will allow 22.6A to flow indefinitely, 29A to flow for about an hour and currents appreciably more than 29A to flow for shorter periods, without tripping. However, one obviously is not meant to ‘push’ MCBs in that way, not the least because it could well affect their life expectancy.

The nicest solution would obviously be to have the ring ‘mended’. Particularly if you are going to go with radials (and maybe even if you don’t), it would also be worthwhile considering moving the oven onto its own, new, circuit (which you said would be fairly easy), since that would obviously appreciably reduce the potential current demand on the sockets circuits, thereby making 16/20A circuits more reasonable.

Kind Regards, John
 
Do I take it from what you say that your electrician has isolated the piece of cable between two sockets and confirmed that this piece of cable (and only this piece of cable) is responsible for the neutral-earth fault?
[...]
You indicated that there were initially two faults, one of which was accessible and could be rectified, but this other one which was less accessible for repair. What was the nature of the fault/damage/whatever that was accessible for repair?

And I thought I included all the details! Just shows you guys know what you're talking about...

Yes, the electrician isolated a fault at the end of what I called Radial 1, between 2 sockets. This wall is the super damp wall that needs channelling if we were to restore the ring. He didn't tell me what the fault was there.

The other fault he identified is at the begining of what I called Radial 2, between the fan+oven and the first double socket. This one he identified as Neutral/Earth fault. This is easy to repair as he can lay a cable from the CU in the cellar, drill the wall to the back of the ovens, and we're there. So it's not so much repairing the damage, it's more running a new cable and forget the old one I suppose. And if he drills a slightly bigger hole, he can put 2 cables: 1 for the radial (fix fault), 1 for the new dedicated oven circuit!

As for why he chose 16A, I did ask this morning if he could fit 20A and if he reckons it's a) enough for a kitchen and b) safe/follows regulations. We'll see what he says.

if the cable installation method allowed it (primarily, none of the cable buried in insulation), even a 25A MCB might well be compliant.

What about burried in plaster and cement? and squeezed between wood flooring and floor insulation? Sound like my installation doesn't allow it! And if in doubt, can't assume it's fine.

Particularly if you are going to go with radials (and maybe even if you don’t), it would also be worthwhile considering moving the oven onto its own, new, circuit (which you said would be fairly easy), since that would obviously appreciably reduce the potential current demand on the sockets circuits, thereby making 16/20A circuits more reasonable.

Noted, makes sense.

Well, guys, I know a lot more than I did before. Thanks for the time spent writing all this stuff.

Thanks again,
Thierry
 
And I thought I included all the details! Just shows you guys know what you're talking about... Yes, the electrician isolated a fault at the end of what I called Radial 1, between 2 sockets. This wall is the super damp wall that needs channelling if we were to restore the ring. He didn't tell me what the fault was there.
OK. So this is the 'problem' one, and we don't know what sort of fault it is (neutral-earth or whatever) - is that correct? Do I take it that he is sure that, whatever the nature of the fault, it is definitely in the cable between the two sockets mentioned?
The other fault he identified is at the begining of what I called Radial 2, between the fan+oven and the first double socket. This one he identified as Neutral/Earth fault. This is easy to repair as he can lay a cable from the CU in the cellar, drill the wall to the back of the ovens, and we're there. So it's not so much repairing the damage, it's more running a new cable and forget the old one I suppose. And if he drills a slightly bigger hole, he can put 2 cables: 1 for the radial (fix fault), 1 for the new dedicated oven circuit!
Fair enough. I had rather hoped that he had already fixed that one, so that you would be able to tell us what he had discovered to have been the cause of the N-E fault in that cable - but you clearly can't tell us that yet. What you say about the nature of the repair presumably assumes that you are going install a separate circuit for the oven - to actually 'fix the fault' would presumably require replacing the cable between the oven (is that a socket?) and another socket, not a cable from the CU.
As for why he chose 16A, I did ask this morning if he could fit 20A and if he reckons it's a) enough for a kitchen and b) safe/follows regulations. We'll see what he says.
We'll be interested to hear what he says. As I've said, I suspect that he may just have overlooked' the possibility of fitting a 20A MCB!
if the cable installation method allowed it (primarily, none of the cable buried in insulation), even a 25A MCB might well be compliant.
What about burried in plaster and cement? and squeezed between wood flooring and floor insulation? Sound like my installation doesn't allow it! And if in doubt, can't assume it's fine.
'Buried in plaster/cement', on the surface of a wall or under floorboards (the commonest situations) are all fine, and, if it were all installed in one (or more) of those ways, would give 2.5mm² cable a current-carrying-capacity of 27A, adequate for a 25A MCB, let alone a 20A one.
Noted, makes sense. ... Well, guys, I know a lot more than I did before. Thanks for the time spent writing all this stuff.
Thanks again,
You're welcome. Keep us posted!

Kind Regards, John
 
OK. So this is the 'problem' one, and we don't know what sort of fault it is (neutral-earth or whatever) - is that correct?

Correct.

Do I take it that he is sure that, whatever the nature of the fault, it is definitely in the cable between the two sockets mentioned?

I believe so.

What you say about the nature of the repair presumably assumes that you are going install a separate circuit for the oven - to actually 'fix the fault' would presumably require replacing the cable between the oven (is that a socket?) and another socket, not a cable from the CU.

You are more aware of the situation than I am it seems. You are completely right.

You don't happen to work in East London by any chance?!

Regards,
T
 
John, you have been Frank a couple of times in the past week. Are you suffering from a split personality? ;)

As for the faulty cable, I dug out the following story for you. Basically, I had installed a new piece of cable from a 20A breaker to an FCU. A while later, after some very heavy rain, it developed a fault and started tripping the RCD.

I had a fault on the parents-in-law's install.

RCD tripping. Isolated it to a circuit which was just a 20A radial with a towel rail on it. The fault turned out to be a pinhole in the cable, letting in moisture.

Who'd have thought it? But that's the sort of thing you're up against.

I suspect the fault was either on the cable to start with or had been inadvertently made during installation.

But the fault lay dormant until dampness revealed it. Could this have been the case in the OP's kitchen?
 
OK. So this is the 'problem' one, and we don't know what sort of fault it is (neutral-earth or whatever) - is that correct?
Correct.
Do I take it that he is sure that, whatever the nature of the fault, it is definitely in the cable between the two sockets mentioned?
I believe so.
Thanks for confirming.
What you say about the nature of the repair presumably assumes that you are going install a separate circuit for the oven - to actually 'fix the fault' would presumably require replacing the cable between the oven (is that a socket?) and another socket, not a cable from the CU.
You are more aware of the situation than I am it seems. You are completely right. You don't happen to work in East London by any chance?!
'Fraid not - and, in any event, contrary to what you have probably assumed, I am not an electrician. However it sounds as if (with a possible 'moment' in relation to the 16A/20A business) your electrician is pretty much 'on the ball' - and hopefully his responses to your latest questions, coupled with what you've been told here, will help to reassure you about that! Of course, it remains possible that he knows something that we don't which means that you are 'stuck' with 16A MCBs if you split the ring.

Kind Regards, John
 
The fault turned out to be a pinhole in the cable, letting in moisture.
[...]

I suspect the fault was either on the cable to start with or had been inadvertently made during installation.

But the fault lay dormant until dampness revealed it. Could this have been the case in the OP's kitchen?

And there's me convinced my (not so brilliant) builders have maybe fixed their dodgy cabling by reconnecting cables with a connection block and then bury it all under plaster... ;)

could have been a small "invisible" fault after all.
 
John, you have been Frank a couple of times in the past week. Are you suffering from a split personality? ;)
I think you'll probably have to explain that one to me - I haven't been aware of ever having been Frank (although I am sometimes pretty frank in what I say/write!) :)
As for the faulty cable, I dug out the following story for you. Basically, I had installed a new piece of cable from a 20A breaker to an FCU. A while later, after some very heavy rain, it developed a fault and started tripping the RCD. ... I suspect the fault was either on the cable to start with or had been inadvertently made during installation. ... But the fault lay dormant until dampness revealed it. Could this have been the case in the OP's kitchen?
Yes, of course that is possible (although, I think we can probably agree, a very rare happening). As I wrote:
I suppose anything is theoretically possible. You certainly could not get a neutral-earth (or any other type) of fault in a cable because of damp unless it had suffered mechanical damage, which not only completely penetrated the outer sheath but which also completely penetrated the insulation of the neutral conductor. Such damage is extremely unlikely (note that is not the same as 'impossible'!) in the absence of a direct penetrating injury (e.g. by nail, screw or drill).
Kind Regards, John
 
The pin hole problem also happened to me. Submersible pump had water in it seal oil tested OK all dried out glands remade and replaced went down again.

Second time it was my job not wanting to be also caught out I got some re-enterable compound and a car tyre valve and after I refitted everything I pumped 20 psi into the pump to test.

About three meters up the flex I found a really small hole and the compound was ussing out.

It then clearly set inside the flex (180 mm²) and when replaced the fault never came back.

On another job it was far more serious. I went to the switch room and did not like the smell so called in the chemist. With full BA set it was found seals had gone on a hydrofluoric acid pump and the acid had got into cable and syphoned into the switch room. Rather deadly.

It did however generate a lot of work. Every cable was cut and a epoxy resin joint used so it could not happen again.
 

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.

Select the supplier or trade you require, enter your location to begin your search.


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

 
Sponsored Links
Back
Top