1930's rewire

[plugwash]

I'm sure that a lot still install ring finals, but probably more 'out of habit' than any other reason. Installing a ring final is not really any easier/quicker than installing a radial.
If one installs 32A radials (with 4mm² cable) then the number of breakers, and hence required size of CU, is exactly the same as it would be with ring finals.

That is true, but it also means cable that is both substantially more expensive and substantially bulkier. If you are forced up to 6mm² (4mm² is quite marginal for 32A) or you want to have grid switches (which have smaller terminals than standalone accessories) it gets even worse.

20A radials meanwhile mean you are likely to want more circuits, especially in the kitchen. A 32A circuit can accommodate two full-size loads at once plus smalls. A 20A circuit on the other hand can only accommodate one full-sized load

 

[plugwash]

Why?
How? If one compares 'like with like' (32A radial vs 32A ring), one willk usually/often require roughly twice the length of (2.5mm²) cable for the ring

While there are cases where that may be true I think "usually/often" is pushing it. Sure a radial has one less leg to/from the CU but I would expect in a typical house the legs to/from the CU will represent a minority of the cable in the circuit.

 

[JohnW2]

That is true, but it also means cable that is both substantially more expensive and substantially bulkier.

4mm² is appreciably less than twice the cost of 2.5mm², so cable cost may (depending on topology of circuit) often be cheaper with a 4mm² radial. 4mm² is not significantly 'bulkier' (I often confuse the two when looking at them) and, for what it's worth, I find 4mm² somewhat nicer/easier to work with than 2.5mm.

If you are forced up to 6mm² (4mm² is quite marginal for 32A) or you want to have grid switches (which have smaller terminals than standalone accessories) it gets even worse.

All true, but I was talking about the general situation. As for "4mm² is quite marginal for 32A", so can (at least some parts of) 2.5mm² cable in a 32A ring final be 'marginal', so I'm not too impressed/.concerned with/about that argument.

20A radials meanwhile mean you are likely to want more circuits, especially in the kitchen. A 32A circuit can accommodate two full-size loads at once plus smalls. A 20A circuit on the other hand can only accommodate one full-sized load

Indeed so - that's why I say so often that I don't really like, and never would not install, a 20A radial (particularly, as you say, in kitchen).

Kind Regards, John

 

[JohnW2]

While there are cases where that may be true I think "usually/often" is pushing it. Sure a radial has one less leg to/from the CU but I would expect in a typical house the legs to/from the CU will represent a minority of the cable in the circuit.

I think you're quibbling. Yes, as I've said, it obviously depends on the topology of the circuit (location of sockets), A ring will inevitably involve at least a little more total cable length (since one could always have omitted the final leg back to the CU and called it a radial) but, according to the situation, it can vary from 'only slightly more' to 'twice as much', the average being somewhere between those extremes.

The difference will only be negligible if at least two (ideally most) of the sockets are very close to the CU (which may only apply to one of the 3 typical sockets circuits in an average house), and won't be the case if all the sockets are fairly distant from the CU (e.g. an 'upstairs' circuit fed from a downstairs CU).

Anyway, my point is that (in comparison with the other costs, particular of labour, which are essentially the same for both types of circuit) the difference in cable cost with the two methods is pretty trivial and not, in itself, something I think should be the deciding factor between the two types of circuit.

Kind Regards, John

 

[flameport]

Do a lot of electricians still use this method when rewiring etc, just because it is the easy/quick solution?

Some do, because of the usual excuses such as the old ways are best, can't be bothered to consider anything else, that's how it's always been done, the NICEIC/others said that's the way and so on.

Does rewiring with radials result in a lot more breakers and a much larger CU, even in a typical 3 bed semi?

Compared to the method of one ring for the whole house as intended in 1947, then yes it does.

However no choice today, as all socket outlets require 30mA RCDs, and the maximum leakage current permitted in normal use is 9mA, which could be just 3 or 4 pieces of connected equipment.
One or two circuits for the whole house isn't just undesirable, it's not compliant with BS7671. Neither is a couple of RCDs covering 50% of the house each for the same reason.

A ring final circuit is one specific type of circuit designed at a time when there was a need for it.
That time was over 70 years ago and all of the circumstances which led to it's design and subsequent use haven't been valid for decades.

 

[JohnW2]

.... However no choice today, as all socket outlets require 30mA RCDs, and the maximum leakage current permitted in normal use is 9mA, which could be just 3 or 4 pieces of connected equipment. ... One or two circuits for the whole house isn't just undesirable, it's not compliant with BS7671. Neither is a couple of RCDs covering 50% of the house each for the same reason.

We agree that there are no longer any particularly good reasons for using ring finals but, as for the rest ...

... as I've often said in the past, (with visitors to the forum in mind) it somewhat concerns me that you quite often express personal views (usually resulting from fairly 'extreme' interpretations of BS7671) that many widely-adopted practices are "not compliant with BS7671" - but express them in a manner which makes them sound like universally-accepted facts, rather than your opinions.

In terms of the above example, I suspect that very few people interpret BS7671 as requiring RCBOs for every circuit (or, at least, requiring more than two RCDs) - and BS7671 obviously doesn't (and never would or could) require no more than 3 or 4 outlets on a sockets circuit - which (given that designers don't have crystal balls) would be the only way of ensuring that 3 or 4 pieces of equipment with a total leakage >9mA could not be connected to the circuit.

Somewhat ironically (one might even suggest 'inconsistently') the implication of what you say is that you are advocating 20A sockets circuits. As I recently wrote, I have never been all that comfortable with them, particularly for kitchens, since (again given that designers don't have crystal balls), if there are just two (or more) 13A outlets, it is theoretically possible for the design current of the circuit to be exceeded by just two items of connected equipment - yet it would be common (probably 'usual') for such a circuit to have several 'double sockets' (an presumably always will be?).

Kind Regards, John

 

[securespark]

I have to pipe up in support of FP here.

I have a dislike of ring finals and have done ever since my apprenticeship. In 1947, there were good reasons for installing this type of circuit. Indeed, it was devised out of need.

But now?

To my mind, radials are superior in every way.

As I wrote in the other thread, 20A socket outlet circuits were often installed on the first floor of domestic buildings, as the demand was such it was not deemed necessary for anything bigger.

As for split boards, I think they are an abomination. I don't believe they comply and some at the IEE did not either. My personal opinion is that they were introduced as a cheap fix because there would have been uproar if the standard was all-RCBO boards instead.

 

[plugwash]

My personal opinion is that they were introduced as a cheap fix because there would have been uproar if the standard was all-RCBO boards instead.

I would argue the same is true of the "16th edition" split load boards that preceeded them, though I admit there was great inconsistency in how such boards were installed with some people only putting downstairs sockets and showers on the RCD and others putting everything except lights on the RCD.

 

[securespark]

Agreed, when I wrote "split boards", I was referring to all varieties!

 

[Harry Bloomfield]

The best way to wire a TV over a fireplace is not to, and to put it low down in a corner.

Yep, try sitting down in a comfortable armchair and staring at a point well above eye level, above the fireplace and your neck will soon tell you it is a bad idea. Eye level, at a comfortable seated height is best.

 

[JohnW2]

I have to pipe up in support of FP here. ... I have a dislike of ring finals and have done ever since my apprenticeship. In 1947, there were good reasons for installing this type of circuit. Indeed, it was devised out of need. .... But now? ... To my mind, radials are superior in every way.

As regards radials vs. rings, I personally don't have any strong feelings, either way, but I certainly agree that there are no longer any particularly good reasons for favouring rings.

As I wrote in the other thread, 20A socket outlet circuits were often installed on the first floor of domestic buildings, as the demand was such it was not deemed necessary for anything bigger.

..and, in turn, I indicated that I remain a little conceptually uncomfortable with 20A radials, given that a user can theoretically overload such a circuit if it only has one (double) socket, and certainly if is has more than just one double. However, I agree that it is probably 'reasonable' in low demand areas of a house - so it again comes down to judgement as to what is 'likely'.

As for split boards, I think they are an abomination. I don't believe they comply and some at the IEE did not either. My personal opinion is that they were introduced as a cheap fix because there would have been uproar if the standard was all-RCBO boards instead.

I certainly don't feel like that (I assume you mean dual-RCD boards, not those with a single RCD protecting only half of the circuits).

As I've been writing quite a lot recently, I do think that we (and the regs) are probably addressing issues which do not significantly exist - whether in the name of 'Nanny' or whatever. How often have you come across situations in which significant (let alone 'dangerous') problems have arisen because an installation was not an 'all-RCBO' one.

As I've also often said about 'Nanny', if we (or the regs) are so concerned about these perceived issues/'dangers', why on earth is there not a requirement for emergency lighting in domestic premises - since (albeit usually brief) 'power cuts' are, at least in my case, easily the most common cause of my being 'plunged into darkness' (or would be, were it not for my emergency lighting!)?

Kind Regardds, John

 

[securespark]

I feel very strongly.

After 30 years of testing and fault-finding gash, appallingly installed circuits, I absolutely detest ring final circuits.

When they die, I shall be the one cheering the loudest!

 

[JohnW2]

I feel very strongly. After 30 years of testing and fault-finding gash, appallingly installed circuits, I absolutely detest ring final circuits. When they die, I shall be the one cheering the loudest!

Yes, I realise that from what you've written. I'm a little surprised that your feelings are quite so strong/'passionate', but I obviously have not had the experiences you have had with the circuits.

I suppose what you mention is one valid point against rings - that, even if installed satisfactorily, they take more time to test and fault-find. Mind you, I doubt that (m)any electricians would, for example, charge less for an EICR in an 'all-radials' installation - so it's probably an advantage for the electricians which doesn't get passed on to the customers

Kind Regards, John