Using wiring for attic light socket as plug socket?

[Dave Plowman (News)]

Quite - my ground floor lighting circuit has been 10 amp since I installed it some 40 years ago. Due to a rather large load on it. ;-) The feeds are all 1.5mm - switch drops 1mm.

[John Rumm]

Its not a departure at all - its one of the standard circuits listed in the On Site Guide:

(you will note there is also a standard 16A lighting circuit - although that is very unlikely to be used in a domestic situation)

Sounds like a very useful learning experience. Keep in mind though that for subtle details that was several issues of BS7671 ago, and stuff has (and will continue to) evolve.

Alas in those days many professional electricians had very little clue about equipotential bonding (even the term "earth bonding" is a misnomer when you think about it - earthing and bonding are two different systems designed to add protection in two different ways). Not aided much by the

15th edition regs that also over egged the requirements somewhat.

You may find this useful:

The key requirement in situations where EQ bonding is used is to establish if the metalwork in question is capable of bringing a potential into the equipotential zone. If its not, then there is no point in bonding it. Hence why things like sinks and bath do not require bonding in spite of so many electricians believing they do.

Many of these things are chosen for backward compatibility as much as fundamental theory.

Depends on how you define "manage". It an installation that does not provide adequate discrimination[1] in the event of a fault

[1] i.e. the requirement that the effect of a fault be localised to it. Plunging an entire house into darkness due to a lamp failing and tripping a MCB is not acceptable today. (it never really was, but the implications of it had not really been considered at the time)

I sense an anti 10A fuse agenda here ;-)

I think you will find that modern lighting accessories are fine on 10A circuits (note that even the limitation on the "small" version of the normal lamp holders was removed).

(and TBH a 6 way CU is woefully inadequate for many places these days)

If doing it now, it would be quite common to do that, and also add an extra ring circuit for the kitchen / utility room - since the modern compliment of kitchen appliances can rapidly use all 7.2kW capacity of a ring circuit.

Would it not have been simpler to provide a dedicated circuit for this purpose? Then you would *know* its not going to be tripped regardless of what happens elsewhere.

Its not uncommon to find all sorts of strange stuff in existing installations - especially if they are ageing and have been hacked about a bit in the following years.

(I recall digging a 5A circular junction box out of the plaster above the sink in the kitchen of a neighbour's property. (yes it was the screw terminal type that is intended to remain accessible for maintenance))

I would expect that as long as you fill in the test results correctly they will be happy (they probably don't even look at the name of the contractor on top of the page!)

Your suspicion is this case is unfounded - or at least not for this reason. Needless to say Part P has complicated the "certification" process somewhat in subsequent years.

Quite possibly, although you would need to consider what you are attempting to achieve, and also the unintended consequences of the change.

Keep in mind that when providing over current protection for any circuit, there are two discrete issues that need to be addressed: overload, and fault handling. (i.e. long term over current resulting from applying more load than the circuit design current, and the very short term effects from massive over current caused by someone nailing through the cable etc). Traditionally these two functions are provided by the same protective device at the origin of the circuit[2]. If you are using one of the standard circuits, then both of these are allowed for in the design; The protective device nominal rating is lower than that of the "as installed" cable current carrying capacity[3], and someone has already done the sums to make sure the cable won't melt while carrying 1000A for the few ms necessary to open the MCB on its magnetic trip).

So for example changing a "whole house" 10A lighting circuit to a "normal" type B 6A MCB on the grounds that modern lamps no longer require the current may be fine. But you are starting with a circuit that was already adequately protected. As consequence of changing the protective device, you have also reduced the fault current tolerance of the MCB from 50A to 30A (type B MCBs need 5x nominal current to trip on the "instant" part of their response curve)[4]. A result you may now find is that when one of the few remaining incandescent lamps does fail, its far more likely to trip the whole circuit. (and trips and falls in the house kill orders of magnitude more people each year than are electrocuted!) (note that this is easy enough to mitigate in this particular example since you could use a type C MCBs instead)

[2] this does not always have to be the case - fault current protection must be provided at the origin, while overload protection may be delegated elsewhere or in some cases may not be required at all. [3] Remember that the "clipped direct" current carrying capacity of even 1.0mm T&E is 16A

[4]

While that all sounds nice and warm and fuzzy, can you actually justify it on technical merit?

Agreed in most cases... although keep in mind the increasing use of ancillary loads on lighting circuits - e.g. extractor fans for forced ventilation in kitchens, loos, bath/shower rooms etc, localised space heating with "heater" lamps etc.

[John Rumm]

If you think about it, the drop lead to most BC lamp holders is in 3A flex...

(you are correct however in that it is "protected". This is one of those situations I alluded to in my previous post, where overload protection may be delegated or dispensed with in certain circumstances)

They are still acceptable, however you would be very unlikely to use one in a new installation for several reasons.

Modern trends in insulating properties makes their use more difficult. The big downside with a BS3036 rewireable is the slower fusing time. This requires a x 0.725 de-rating factor be applied to the current carrying capacity of the cables used in the circuit they protect. With the effects of the extra thermal insulation added to the equation, you would frequently find that you would need to increase the conductor CSA to compensate - hence unnecessary cost in both materials and time spent wiring.

Old style fuse boxes typically don't have any provision for RCDs, hence you would need to house these separately, and also use multiple fuse boxes to provide adequate discrimination.

Much depends on who is doing the resetting... Its much harder for someone to wrap tinfoil round a tripped MCB or stick a nail in it instead ;-)

Alas, outside the shrinking population in places like this group, the number of them that can DIY seems to be ever diminishing.

Its a difficult assertion to validate one way or the other. However it might be worth considering that the vast majority of house fires started "electrically" are actually the result of the misuse of appliances, or use of faulty appliances. Fixed wiring faults come *way* down the list.

[John Rumm]

When you consider that many a 3A fuse will happily supply nearly 5A indefinitely, that's over 1kW of available power ;-)

Hmmm nice bit of logic.... oh, that's really hot - let's insulate it and make it much hotter - using flammable insulation!

[John Rumm]

Shame whoever wired this place did not think of that... they did one 5A (Type 1 MCB) and then worked out it was not going to hack the 40 or so lamp fittings on the ground floor, then split of two rooms on a second one. Even then the first circuit would not tolerate all the lights on for long!

[Tim Watts]

On Wednesday 12 February 2014 16:19 John Rumm wrote in uk.d-i-y:

Indeed.

When designing my circuits to work with a 10A Type C breaker, I require the trip times (loop impedances) and max current capacity of the wiring to meet the standards for 10A type C lighting *right to the lamp fitting*.

If the lamp fitting itself is sub standard, then it is usually a negligible risk as

a) it tends to be self protecting (hard to fit a lamp that will overload it);

b) unlikely to be a danger even if it smoked (the exception being flush mount ceiling fittings or you live in a hippy den with fabric all over the ceiling!);

c) As John suggested, if you look at the wires in many fittings, they would not even be adequately protected by a 6A Type B breaker.

d) The bits that are most at risk of causing a fire (cables running thorough the building often via bits of wood, piles of fluff and sawdust etc) are designed and protected correctly.

Then you come to bathroom fans and other things that are commonly fed from a lighting circuit and they all want a 3A FCU fitted regardless of circuit protection!

Cheers,

Tim

[Dave Plowman (News)]

I was just thinking the 'Grand Designs' usual of thousands of low voltage downlighters must need some thinking about - certainly before LEDs. When I wired my place the usual was a single bulb pendant per room. ;-)

[dennis]

You need less lights and use the CH to heat the place.

[John Rumm]

People here before were fond of the multi armed dangly thing with three or five spokes and a lamp on each... they soon ramp up the lamp count if not used sparingly!

[John Rumm]

Judicious use of low energy lamps, and a few fitting changes, and bulb wattage reductions has tamed the excess a little. Lots of high ceilings though, so they take a reasonable amount of lighting to be comfortable.

[Johny B Good]

Well thanks for that revealing information. It's additional knowledge I can retain for future reference. Obviously, I'm not going to uprate the fusing on my owm lighting circuits since the efficacy of the 5A fusing has amply proved itself in my case over the past 3 decades.

Still and all, the whole point of a single fuse at the CU to protect a lighting circuit is essentially to minimise the risk of a house fire. Being 'plunged into darkness' is very rarely as fatal as smoke inhalation or worse.

Of course there is always the counter argument in the automotive case where it is (or at least, was) deemed preferable to eliminate a sudden blackout of the headlamps due to a fuse fault by foregoing any protective fusing whatsoever, making the risk of an electrical vehicle fire higher as a result. The idea being that it is better to still be able to safely navigate along a narrow country road on a dark night for as long as it takes to bring the vehicle safely to rest and permit a rapid bailout.

I believe modern practice with vehicular wiring is to fuse each headlight independantly now that the fusebox is housed within the less harsher environment of the passenger compartment, usually just under the dashboard over the driver's footwell.

Well yes but you _know_ why. Even now, the idea of using a 10A fuse over a 5A fuse for lighting circuit protection still leaves me with a feeling of uneasiness.

I'm familiar with the 'moving goalposts' syndrome. Obviously a fine tuning of the safety versus utlity compromise. At least you still have the option 'to make it safer'.

Funny you should mention that. :-)

We needed to have a downstairs toilet and shower room extension just after moving in for the benefit of the XYL's elderly parents whom we had moved in with us (the whole reason for buying such a capacious property in the first place).

What we landed up doing was to convert the woefully inadequate kitchen (probably an earlier conversion from a cloakroom) into the downstairs shower room and use the extension to create a new kitchen.

I remember recovering the 2.5mm FT&E cable and sockets (along with a redundent porcelain DC fuse holder and fuses that had simply been left in situ) after the old kitchen space had been gutted prior to its conversion.

In the meantime once the kitchen extension had been completed, the electrical contractor added a 2 way wylex fuse box to extend a seperate kitchen ring main and supply a cooker point feed. The lighting was served by the addition of one extra fluorescent tube luminary to the existing ground floor lighting circuit.

Overloading the kitchen ring main has never been a problem. I suppose it is just possible to overload it if we were to switch the 3KW kettle on whilst using the "George Formby", toasting bread , microwaving some food and 'Brevelling' some tasty sandwiches whilst we (the XYL & I) each run a Dyson over the respective halves of the kitchen dining room floor areas when the noise of the dishwasher will be nicely drowned out. We may even have enough 13A sockets without having to disconnect the fridge and the freezer nor cheat by plugging into the 13A cooker point socket.

I'm sure that would be a good 8 or 9KW's worth right there (maybe even 10). The strange thing is, we've never felt impelled to "Work so hard at cooking and cleaning as to make our eyes bleed" Not even at Christmas time when it was 'our turn' to host the 'christmas family do'.

Perhaps we simply don't posses enough of the latest energy consuming 'Must Have' kitchen gadgets to bless us with such overloading problems.

Think it through. What's important? having a dedicated circuit that can still go faulty despite its splendid isolation from the other ring mains or a choice of three alternative sockets less than an extension lead's length away. Remember, this sort of eventuality ideally should never happen or at least no more than once in a lifetime so there's no good reason to over engineer a solution when a perfectly good one is already to hand.

Just the fact that lower fuse ratings result in lower fire hazard risks. I wasn't concerned over the electric shock hazard which remains unchanged when simple fuses provide the protection.

If my floor standing cooling fan is anything to go by, I'd imagine a small vent axia type of fan wouldn't draw much more than 50 watts. As for 'heater lamps' you're probably looking at 500W max (in a suitable lamp fitting) or else 1KW for a radiant 'bathroom heater' that's mounted high up and operated by the obligatory pull cord.

The more powerful radiant heater would be better served by a 5A fused junction box spur tapped off the ring main. Quite frankly, I wouldn't feed such loads from a lighting circuit other than maybe a vent axia fan or two. You seem to be trying to justify the higher 10A fuse rating with extra appliances that would be better served from a ring main supply (via a suitably fused connection box).

My own personal view is that the the primary function of a CU is safety and, to this end, selecting a lower fuse rating option over a higher one when conditions allow maximises its effectiveness at minimising the risk of a house fire.

The risk can never be totally eliminated whatever size fuse is chosen but the risk does become more improbable as you lower the fuse rating used on any existing circuit. Take my 15A ring main as an example of this. I've never ever had to replace the 15A fuse link with a 30A one because of 'overload' issues (nor for any other reason).

The 15A fuse link has sufficed quite nicely for the loading this particular ring main is subjected to. IOW, its utility hasn't suffered as a result of the lower fuse rating so I've gained an extra margin of safety 'for free' by this simple expedient.

I suspect that most domestic 10A lighting circuits could be just as well served by a 5A fuse link or 6A MCB these days simply by virtue of the extensive use of CFL lamps in place of the tungsten filament GLS lamps they'd originally been sized for. It's certainly worthy of some consideration if you have 10A lighting circuits in your home.

[meow2222]

The oldest vehicle I had had no fusing whatever. The original fusebox with space for 2 fuses was present, but not connected.

Exceeding 30A on a ring circuit isnt a problem, theyre designed to deliver much more for limited periods. 30A is the continuous rating. We regularly used to heavily exceed 30A at the last house, but nothing got hot or blew.

NT

[John Rumm]

I don't think anyone was suggesting that you do. Just highlighting that it is an allowable and properly designed option.

Missing the point I think... fusing good; single lighting circuit bad!

The chances of any lighting circuit causing a fire in the first place are vanishingly small, but having a pair of them (and in consequence a lower average current load on each) can only improve the odds further.

So if a house is wired with a single 10A circuit, you would get a much greater improvement in safety by splitting it into two circuits (of whatever rating), rather than just keeping it as it is and lowering the fuse size.

There are a few permitted cases where fusing may be omitted on similar grounds - i.e. where failure of the supply is more dangerous than the circuit being damaged by over heating etc. (unlikely to crop up domestically though!)

Whilst I agree with the general principle of using protective thresholds of no bigger than they need to be, I also recognise that the absolute size of the fuse is not the sole guide to how safe a circuit is.

Possibly - but that does not always equate to a lower current protective device. Nuisance trips from "close to tripping" protective devices can in themselves harbour dangers.

[snip]

With kitchens you can generally ignore the kettle and microwave etc - they are not usually on long enough to pose a problem. The big white things with heater elements like washing machine, dishwasher, tumble drier etc can swallow more "long term" capacity. Some single ovens are also often plugged into a kitchen ring circuit these days.

Keep in mind the circuit is *designed* to be able to cope with short term overloads of several kW without any damage. A normal 32A MCB will happily supply 45A for some time, and even 50A for 10 mins or so. This reflects the case that the wiring itself also requires some time to reach a temperature where it will start to suffer damage.

Most people don't. Although many have enough that it would become an issue if the kitchen was sharing a circuit with large parts of the rest of the house as well.

Whichever ring you power it from, its sharing with other appliances, which may cause a nuisance trip. Without sharing, there is no possible cause for a trip other than catastrophic failure of the freezer itself!

(it would also allow the provision of a freezer supply that does not have RCD protection, which eliminates another set of trip scenarios from an application where enhanced shock protection is not an issue)

[big snip]

Yup fans are typically no more than tens of watts at most. Most heater lamps are usually under 750W. Although there are a few over 1kW. E.g:

A 6A circuit gives you 1380W to play with. So that would only allow another two lamps (allowing the required 100W per lighting position) on the circuit if using a 6A one.

Other common reasons for larger capacity circuits would be a largish bank of strip lights - since they can have a fairly heavy switch on surge.

I find no need to "justify" the circuit - they exist, and are appropriate for some situations. I was just pointing out some of the reasons that the "normal" 1380W may not offer the required capacity.

Nothing wrong with the logic... although you need to focus attention on the right things. Adding RCD protection would give a much more significant reduction in the likelihood of of house a fire cause by an electrical source than say worrying about (properly installed and configured) 10A lighting circuits. As would Fitting mains powered interlinked smoke alarms.

Any circuit is capable of supplying ample current to heat something enough to cause a fire... Quality of terminations and cable insulation are more relevant.

Your 15A rewireable will supply 28A pretty much indefinitely anyway - in these days of central heating, most ring circuits outside a kitchen are less heavily loaded in the past (lots of small appliances rather than heavy load space heating etc)

Personally I don't have any 10A ones, although as I alluded to elsewhere, one would have actually been a quite a good choice here due to the huge number of lamps employed on the ground floor! (when we moved in, turning on half the downstairs lights all at once would trip the 5A MCB after about 10 mins) (I have replaced the CU since)

Now I have three 6A Type C MCB circuits for the house, another for the outside lights, another in the garage / workshop. (I also have battery backed emergency lighting fed from the adjacent normal lighting circuits, since there are no street lights here, and it can be *very* dark)

[chris French]

In message , Johny B Good writes

Of course the opposite is also easy....

[Johny B Good]

The manufacturer probably decided the risk with 6 or 12 volt wasn't worth the expense and trouble of 'fuse protection' (plenty of other hazards to consider) in a box on wheels that could be rather more swiftly evacuated than a typical house (plus, one might hope, at least one of the occupants would be wide awake and alert to such a danger).

ISTR, that 2.5mm FT&E has a continuous rating of 27A (depends on how the cables are routed (trunking with other cables versus single run clipped onto beams and joists with adequate air space).

The 27A rating is only 3A shy of the full 30A fuse rating but on a ring main it would be difficult to persuade more than 27A to flow over the shortest section between the CU and the nearest socket (but not impossible under the right (wrong?) circumstances).

The remaining 3A is modest enough that in most installations this would be supplied over the 'scenic route' via the rest of the ring main wiring.

What helps is the 13A fusing of the plugtops limiting the current draw from a twin gang outlet to 26A (the regs might have something to say in regard of the use of 3 and 4 gang outlets - I just don't know).

In most domestic properties, the risk of overloading the shortest run to the socket nearest the CU is vanishingly small so the use of 27A rated cabling in a 30A fused ring circuit is deemed acceptable in the regs (it's only a 10% overload at its worst after all).

[Adam Funk]

Has fuse manufacturing not improved in the past 70 years?

[Johny B Good]

I appreciate that. I was pointing out that just because I could uprate to 10A fuses without violating the regs, I wasn't about to do so (you're not obliged to fuse up to the allowable limits if you don't need to).

If I'd been aware that use of 10A fusing was permitted on lighting circuits when I was rewiring the house, I'd probably have chosen that option (but I've no doubt, after replacing all bar one incandescent lamp with CFLs, I'd have long since downrated to 5A fusing by now anyway).

I agree that there's more to having two lighting circuits simply to limit the maximum current draw. It's just that there are a lot of quite modern 3 bedroom properties (less than 40 years old) relying on only a single lighting circuit (along with a single ring main) and I thought this was standard practice. I suppose it was by the cost concious "Barret Homes" of the building industry at the time.

You'll get the same safety benefit regardless of whether you simply replace the 10A fuse with a 5A one or split the lighting into two seprately fused circuits, each fused at 5A. In the latter case you'll have improved utility over the single circuit and retain the same safety level against faults in the fittings and lamp cordage.

If we ignore the bare faced lie that CFLs have 5 times the luminous efficiency and assume a more reasonable 4 times improvement, a total relamping of the original filament lamps with equivilent light power CFLs will reduce the load current to only 25% of the original requirements used to define the size of fuse.

For anyone who has replaced most of their incandescent filament lamps with CFLs or LEDs, they can probably simply replace the 10A fuse with a 5A one without detriment to the utility of the lighting circuit.

There _just_ might be a reference to life support system power in the home but other than that I wouldn't have thought the regs would allow any such 'dispensation' whatsoever.

"House Owner Rage"?

Well, it helps that the kitchen's ring main is not cursed with the washing machine and tumble drier loads, they live in the utility room right next to the downstairs shower/toilet and are plugged into the ground floor ring main circuit.

I see your point but my setup is protected by simple fuses with no ELCBs in sight so I don't have the problem of 'nuisance trips' to contend with. In any case, a freezer can go unattended for 24 hours before it just has to be powered back up.

I think we'd notice a ring main outage well within that time limit and pay the basement a visit if only to check the fuse box, allowing us to swap the freezer to an adjacent working socket if need be which is not even one step away from the fuse board.

That chest freezer has long since been sent to the council tip. I don't think we even bothered to relocate it to the newly built kitchen (BICBW on this point) electing instead to simply buy a new more energy efficient unit more in keeping with our 'brand new' kitchen.

As I said, not an issue in my case.

Well, you're offering a justification regardless. I'm not saying you're wrong to provide such justification, just pointing out that such justification doesn't apply in my case since, like most homes connected to the gas main we enjoy the benefit of central heating which makes such additional loads redundent.

For those with electric only homes, I can see the justification to use 10A fusing on the lighting circuits to increase the utility over and above the bare minimum of lighting only.

I might even consider uprating the fuse to allow the use of such a heating lamp if I decided to improve the comfort of taking a bath whilst a CH fault had forced us back to using the immersion element for heating the water. TBH, I think I'd forego such luxuries on the grounds that it'll 'help concentrate the mind' on getting the CH repaired in a timely fashion.

I made sure to include such redundency when the CH was installed but it did rather help that the previous owner had installed a hot water tank designed for both immersion and central heating which reduced the parts cost significantly of the CH installation - it was only a matter of plumbing the heat exchanger coils into the CH system.

We've got plenty of the stand alone battery powered smoke alarms scattered throughout the house. That's a 'last line of defence' measure to protect life and limb (but not goods and chattels - any saving of those would simply be a side effect of the primary protection).

You make a very good argument for my case. :-)

You've built a system that meets your requirements, seemingly taking the more cautious approach with the lighting circuits. I guess you've been playing "Devil's Advocate" in offering the counter argument to my assertions which were initially prompted by my ignorance of the regs in regard of permissable lighting circuit fuse ratings.

I've learnt something new so the whole discussion hasn't been a pointless waste of time and I'm sure it's given others something to ponder upon which might prove to be of some use in the future.

[Capitol]

Seems to be a case of extreme verbal diarrhoea!

[meow2222]

th space for 2 fuses was present, but not connected.

Maybe. But I doubt it following a convo I had with them :) Trying to find o ut what the engine was, the oldest guy that had been there forever said 'we used to put whatever was lying around in those, could be anything.' I dont think vehicle hazards were considered much in postwar designs.

er much more for limited periods. 30A is the continuous rating. We regularl y used to heavily exceed 30A at the last house, but nothing got hot or blew .

Its routine, but not a problem. Think about it, washing machine dishwasher kettle microwave... we often exceed 30A.

NT

[meow2222]

AFAIK fuse wire is still the same material (tinned copper) it was 100 years ago. Its drawn differently, but afaik the result is the same. 3/13A cartridges have only changed in very minor details since their design in the 40s. BICBW.

NT