Radials vs Rings

I have two standard 30A rings for downstairs/upstairs power which I am planning to update soon to overcome some legacy DIY issues. I've seen it mentioned a few times that radials are more in vogue these days, which I guess gives 3 options:

1. Keeping each ring as 32A/2.5mm
2. Split each ring into two 20A/2.5mm radials
3. Convert each ring to 32A/4mm radial

In principle all these would be options, although (3) would be the biggest change. The cable cost, RCBO positions, cable costs and testing requirements are not particular concerns - I am trying to understand what factors would influence the best approach.

thanks, Radweld

(for info, this will probably form part of a larger, notified project and I'll do this work myself if the BCO is reasonable)

radweld said:

I have two standard 30A rings for downstairs/upstairs power which I am planning to update soon to overcome some legacy DIY issues. I've seen it mentioned a few times that radials are more in vogue these days, which I guess gives 3 options:
1. Keeping each ring as 32A/2.5mm
2. Split each ring into two 20A/2.5mm radials
3. Convert each ring to 32A/4mm radial
In principle all these would be options, although (3) would be the biggest change. The cable cost, RCBO positions, cable costs and testing requirements are not particular concerns - I am trying to understand what factors would influence the best approach.

Click to expand...

Prefacing this with the fact that I'm not an electrical professional, my view is that f you are happy with the capabilities of your present setup, then I personally would not think there's any real point in moving away from (1). The radial vs. ring arguments (of which you'll undoubtedly see some!) are more theoretical than anything else, and the vast majority of UK domestic wiring still uses rings.

Were you to consider (2), I would suggest that you should give some thought into the location of sockets and 'what might get plugged into what'. A 20A radial theoretically cannot (should not be asked to) simultaneously supply two loads approaching 13A - so I guess you should not really design a 20A circuit if the placement of sockets (and 'behaviour' of the occupants) is such that there was a reasonable chance that two 'full' loads would be applied to that one circuit simultaneously - particularly since a 20A MCB or RCBO would allow 26A to be drawn for appreciable periods of time.

Kind Regards, John

Although you can have a 32A radial with 4mm cable the cable would need to be clipped direct or a type able to take more current than standard twin and earth.

Using ali-tube cable does allow slightly more current than PVC twin and earth as does mineral insulated but in most cases to use radials means reducing to 2.5 mm cable.

The splitting of rings into two radials is not just a case of splitting. You need first to measure the earth loop impedance at furthest socket. Often you will find either the volt drop or loop impedance will not allow you to just split them.

You must realise setting up a system is not just a case of pulling in cables and connecting them up. A final ring with 2.5mm cable and a 32A MCB type B can have a maximum of 106 meters of cable. It was around 80 meters until regulations on volt drop changed in 2008.

It is true the main reason for ring mains has now gone. The idea was to use less cable but because of the 1/3 rule and the way cable loops in and out of the ceiling very little cable is saved. Also the use of IT equipment and the earth leakage means using 4 or 6 radials instead of 2 or 3 final rings used more RCBO's so reduces the chances of tripping. However also increases the price.

Also so often a final ring is broken and if the same happens with a radial it fails safe. i.e. stops working.

4mm cable with a radial has max length of around 52 meters
2.5mm cable about 32 meters. Hence why splitting a ring at 106 meters would need 4 radials to still comply not 2 but it would depend if original was on the limit or not.

The normal socket by MK will take two 6mm cables but many other makes will not so if using a 6mm radial which will be required if "Reference Method 100# (above a plasterboard ceiling covered by thermal insulation not exceeding 100 mm in thickness)" is used then you will have to stick to MK and not have any spurs.

Ali-tube is rated around 34.5A for ref method 100 with 4mm so you could use that. But oddly the rating is not given in BS7671 so you may need to prove to the inspector that it complies. I like ali-tube as you can wire the supply to fridge/freezer without RCD but once you see what it costs you may change your mind. Also in my area you can only buy by the role and unless complete rewire then likely one does not need a whole role.

Keep it simple, leave it as 30 / 32a Rings in 2.5mm T&E.

Changing to 4mm / 6mm radials won't gain you anything with domestic loading, it will just cost a fortune and be a nightmare to connect up, if you want to start chopping in 44mm deep backboxes to get the correct bend radii on 4mm or even 6mm you are a better man than I am.

Keep it simple, you know it makes sense.

CW.

you may want to create a new radial for the kitchen.
but leave the rest as rings

radweld said:

I am trying to understand what factors would influence the best approach.

Click to expand...

This is a genuine reply, not a sarky one:

The best approach is the one that you feel is best for you, given your understanding of the advantages and disadvantages of each (which includes having an informed and educated view on what are advantages and disadvantages), and given your knowledge of layout and the way you'll use the circuits.

Not being funny, but you must be able to do the design.

JohnW2 said:

Prefacing this with the fact that I'm not an electrical professional... The radial vs. ring arguments ... are more theoretical than anything else, and the vast majority of UK domestic wiring still uses rings.

Click to expand...

The radial versus ring argument is anything but academic.
The real reason rings are used is because few electricians are prepared to design an installation, preferring to opt for tradition rather than thinking it through.

ericmark said:

A final ring with 2.5mm cable ...

It is true the main reason for ring mains has now gone.

.. using 4 or 6 radials instead of 2 or 3 final rings

Also so often a final ring is broken...

Click to expand...

At least get the terminology correct before you start waffling on... and on and on and on....

dingbat said:

JohnW2 said:

Prefacing this with the fact that I'm not an electrical professional... The radial vs. ring arguments ... are more theoretical than anything else, and the vast majority of UK domestic wiring still uses rings.

Click to expand...

The radial versus ring argument is anything but academic.
The real reason rings are used is because few electricians are prepared to design an installation, preferring to opt for tradition rather than thinking it through.

Click to expand...

I guess that depends upon what one understands by 'academic'. RFCs have been in use in the UK for something like 60 years, and still serve purpose pretty well, despite the vast increase in the number of powered appliances etc. which have appeared in most households during those years. No-one is denying that there are pros and cons of RFCs. However, it's all very well talking about design, as if it were a magic answer, but proper design of a final circuit can only really be done when it only serves hard-wired loads. The moment you have sockets, your designer's hands are somewhat tied behind youir back, since you have to do some guesswork. I frankly doubt that there are many cases (in a standard domestic setting) in which a 'design process' will result in a clear-cut 'answer'ás to whether one should utilise a radial or ring final circuit; there will nearly alawys be pros and cons, and also personal opinions/preferences.

There is also the matter of practicalities. I've already voiced my reservation about multiple '20A radials' (which is a sort-of design+guesswork issue). If one moves from that to 4mm² radial circuits, there is the hassle of trying to properly and safely manage and terminate 4mm² cables, which (particularly with normal depth back boxes) I feel sure must bring some degree of potentially decreased safety with it. I am, in principle,a great believer in radial circuits, but have to say that these practical issues perhaps influence my thinking.

dingbat then said:

ericmark said:

A final ring with 2.5mm cable ...

Click to expand...

At least get the terminology correct before you start waffling on... and on and on and on....

Click to expand...

If you're commenting on the difference between 'final ring' and 'ring final', I would suggest that is perhaps a little on the petty side, particularly in a 'DIY' forum

Kind Regards, John

When the 17th Edition came out I went to an IET talk on the changes. In this talk the speaker referred to 106 meters as new limit of cable to be used in a ring. I went home and tried to see where he got the 106 meters from and failed. It took a message on the IET forum before I was able to understand how these figures were arrived at and this after I had been an electrician for 40 plus years.

Once I got my head around it then I started to see the problem with a radial. Or in fact any supply to sockets with less than a 32A supply.

The problem is we are instructed to expect a load of 20 amp at the furthest point from the supply and the rest (12 amp) even spread. This means a 20 amp radial has to be designed to take a load of 20 amp at furthest point. Where we calculate the ring as 26 amp even though it has a 32 amp MCB.

So to replace one ring takes three radials and the cable lengths on radials is very easy to exceed.

If I was wiring my own house from scratch yes I would use radials. And I would have one radial per room except for the kitchen/Utility room where I would likely use either a sub main or ring.

But I can sit down and work out lengths, volt drop, and loop impedances. And then after installing I can measure the loop impedance and be sure it falls within the limits.

I will not pretend I sit there with a calculator and work it out each time. I sat down once and wrote a spread sheet which I now use to check sizes against lengths.

But for the DIY guy well just can't see that happing? If you ask most DIY people what is loop impedance you get a blank expression. In fact some electricians don't really understand. They know how to dress mineral insulated cable and how to tighten a gland without marking it. But many installation electricians know how to read a plan and do as they are told and the design is down to the electrical engineer.

Years ago the "House Basher" was the lowest grade of electrician. But times have changed and now they have to be far more than a simple electrician. Ask them to program a PLC and yes they will likely fail but ask what 1.44 ohms mean and they will tell you straight away. Where both the maintenance and installation electrician will look with blank face and once explained say not a fair question!

Today with so many regulations not only BS7671:2008 but also the Part P, M, and F building regulations I think it unlikely any DIY guy will be able to design an installation.

So why try. Just follow the tried and tested method that's been used now for around 60 years.

radials are the new black

LOL.

There are arguments for both and perhaps an spread of both might be an answer .
Each have their own merits and problems and circumstances might determine each case

Taking a look at this and download the Summer 2008 "IET ring circuit Meeting" might give you food for thought.

http://electrical.theiet.org/wiring-matters/2008.cfm

dingbat said:

At least get the terminology correct before you start waffling on... and on and on and on....

Click to expand...

As far as I can make out, rings circuits have many practical advantages over radial/tree circuits

• longer cables possible
  narrower cables needed
  shallow back boxes
  two CPCs
  simple geometry (no more than one outlet on an unfused spur)
  a degree a fault tolerance (poor connection less likely to overheat)

OK the fault tolerance is controversial as a broken ring is not adequately protected and not apparent to the home owner. But a poor rather than broken connection at the beginning of a radial circuit must carry all the current for the rest of the circuit and surely that can burn the cable. I suppose this damage on a radial circuit is visible in the back box but damage in a ring circuit might occur in buried cable.

While sorting out wiring, I would consider burying cable in plaster/filler with at least one diameter between cables. I wouldn't want to use plastic conduit in plaster and avoid grouping two cables in a single conduit.

Thanks for the responses. The consensus seems to be 'why change from rings?' and that makes sense. I've seen comments that RFCs will at some point be dropped from the regs so was a little concerned that I should take this opportunity to adopt a more 'modern' approach.

cheers, Radweld

radweld said:

Thanks for the responses. The consensus seems to be 'why change from rings?' and that makes sense.

Click to expand...

As I said at the very start, that would certainly be my view.

radweld then said:

I've seen comments that RFCs will at some point be dropped from the regs so was a little concerned that I should take this opportunity to adopt a more 'modern' approach.

Click to expand...

I really don't think you should let that influence or concern you. Sure, RFCs are a fairly strange concept, more-or-less unique to the UK, and almost certainly would never arisen from engineering considerations - as I understand it, it was the post-war desire to minimise cable use that caused them to appear.

I therefore suppose that it's possible that the day will come when RFCs will disappear from the regs (Heaven forbid, maybe in the name of EU 'harmonisation', or somesuch!). However, I'm quite sure that won't happen for a very very long time and, even if/when it does, it would not be retrospective. Indeed, since RFCs are so ubiquitous in the UK, if it did happen, it wouldn't surprise me at all if (at least for many years) the regs allowed extensions and modifications to existing RFCs (rather than requiring that any changes to a circuit had to bring it up to current regs). I very much doubt that any of that is going to happen in many/most of our lifetimes.

Even though, as above, RFCs would probably never have arisen out of engineering considerations alone, I don't think there are strong engineering reasons for getting rid of them; as you know, and have seen, there are pros as well as cons - and they are 'serving purpose' in millions of installations. People may try to argue that the requirements of electrical installations will be different in the future but, in terms of the size of loads, we could now well be over the peak of curve. Whilst there was a dramatic increase in power-consuming electrical appliances in the third quarter of the 20th century, and maybe a bit beyond, we are now in an era in which the minimisation of power consumption is the flavour of the day.

Kind Regards, John.