New Wiring Regulations

[Flyfisher]

While equi-potential bonding seems like a good idea, is it not almost certain that, in these days of modern plastic plumbing, it is almost inevitable that such bonding will be compromised sometime, somewhere?

I dont know if you mean bonding of incoming services or bonding of metalwork in bathrooms.

I wasn't being specific, just thinking that it's generally a good thing to bond all metalwork together, certainly within a normal touching radius (2m?), so that any conductive surfaces you could touch would all be at the same potential.

Would it be correct to say that, yes, loads of earth bonding is a good thing (if in doubt, earth it?)

Its widely accepted that many items are slightly safer left unearthed. Earthing is not perfect, especially on TT systems where the CPC (earth wire) voltage can rise to high voltage during a fault.

Why is nothing ever simple? :roll:

RCDs are not 100% reliable, do not limit shock current (as is often fondly imagined), and do not act at all on the minority of L-N shocks. However they still offer a magnitude of safety improvement under identical conditions.

Shock current: I thought RCDs are designed to keep any shock below the fatal time-vs-current curve for the average person.

Presumably, a L-N shock would be a rare occurence in the normal course of events, especially with equi-potential bonding in place. A finger in a lamp socket is always possible but I'm not sure how dangerous it would actually be as the current flow would only be through the finger tip (and if it wasn't the RCD would operate). Highly unpleasant though and, I guess, possibly a killer for the infirm.

One of the problems with them is that conditions are not identical, as users too often take full advantage of this apparent extra safety, and start doing things they never would have before, like using power tools outdoors in mild drizzle etc, thus negating any safety benefit and more.

That reminds me of the old argument against seatbelts and all the other safety features in modern cars - they make drivers feel so safe that they take more risks. Hence the old joke about mounting a long sharp spike in the centre of the steering wheel, which would certainly make any driver be very careful :wink:

 

[tony]

however you only get a shock if there is a difference in potential(voltage), so if everything is connected together it all has the same voltage and you are unable to get a shock.

regards
tony

 

[Flyfisher]

Yes, that was my understanding of the benefit of equi-potential bonding.

 

[Lime]

Shock current: I thought RCDs are designed to keep any shock below the fatal time-vs-current curve for the average person.

Presumably, a L-N shock would be a rare occurence in the normal course of events, especially with equi-potential bonding in place. A finger in a lamp socket is always possible but I'm not sure how dangerous it would actually be as the current flow would only be through the finger tip (and if it wasn't the RCD would operate). Highly unpleasant though and, I guess, possibly a killer for the infirm.

An RCD does not limit the current flowing through you before it operates. It should limit the time it flows as long as the current is over 30mA.
(or whatever it is rated at)
The damage to the human body depends on the path of the current through it.
some parts of the body (through the head for example) are much more vulnerable to shock than other parts.

Any shock should be a rare occurence in the normal course of events.

Given that L-E shocks are curtailed by RCDs (hopefully) L-N shocks are the shocks most likely to kill.

An L-N shock is just as likely whether equi-potential bonding is in place or not.

 

[tony]

so if you were ordering this work to be done in your house would you tell the electrician you dont want bonding.?

as its his name on any certificate and he is resonsible for the installation i think the electrican would put it in anyway, a few clamps and a couple of metres of cable.

i installed it in my house.

regards

tony

 

[biffvernon]

I was talking to someone last night who has a house in France where he's just been doing some re-wireing. It seems regulations there are rather different so we concluded that they must have a different sort of electricity. We couldn't work out what happens in the cross-channel interconect where the two types of electricity meet and mix, but the wine he had brought back was quite nice.

 

[Penners]

I was talking to someone last night who has a house in France where he's just been doing some re-wireing. It seems regulations there are rather different so we concluded that they must have a different sort of electricity.

Quite true. French electricity is different, in that it only works when it feels like it. And never between 12.30pm and 3.30pm.

 

[Penners]

Why is nothing ever simple?

Well - speaking as he who posed the original question, Flyfisher, I think you've summed the situation up very nicely.

So, if I can pick the bones out of what everyone has said...(and noting that my installation is the type that has an earth cable running from the consumer unit earth block out to a pin in the earth - is that TT?), I should ensure that all my bathroom metal fittings are bonded together in 10mm2 earth cable, with the end of the run connecting back directly to the earth block in my consumer unit. No other bonding is required.

Is that right?

The earth cable run back to the CU would pass very close to the stainless steel sink in the utility room. Should I make a connection that too?

 

[MdB]

It seems regulations there are rather different so we concluded that they must have a different sort of electricity. We couldn't work out what happens in the cross-channel interconnect where the two types of electricity meet and mix

French electric is 230V AC rather than our 240V AC and they get around the problems of connecting them together by making the link DC rather than AC with some big converters either end (DC also helps with transmission). We get about 5% of total UK power from France this way.

 

[tony]

Penners,

Not quite right.

10mm2 to main water stopcock, on the consumer side of the stopcock, before any t, branch etc.

10mm2 to main gas pipe at least 600mm after the meter on the consumers side of the meter.

10mm2 to central heating/oil services.

Bonding in bathroom,

my bathroom for example has

1 standard pendant
1 light switch located outside in hall
1 inline duct fan mounted in loft
1 radiator
1 sink
1 bath
1 mixer shower
1 towel radiator.

I have taken a 4mm2 cable from the towel radiator to the bath pipes to the sink pipes to the radiator and then to the light switch.

the light switch, pendant and fan are on the same circuit so i dont have to loop to each of them, i didnt bond the shower as the pipes are inaccessable behind tiles, but the pipes to the shower emerge at the bath and are soldered .

Earth clamps on pipes have to be accessable.
On my radiator it has a little lip at the bottom and i use a special earth clamp that is u shaped that tightens on the metal and provides a good earth, these are called ezy earth clamps or something the electrical wholesaler tlc sells them

the cable must be continious so dont cut the cable but strip about 1 inch of the outer insulation with a stanley knife and wrap around the terminal, the reason for this is if a plumber comes in and alters pipework and takes the clamp off the cable is still in one piece, if you cut it and then connect 2 ends up the chances are it might be broken.

Its not a difficult task and i think its a worthwhile task for safety sake.

Ps if you have a airing cupboard next to the bathroom, put some clamps across all the pipes as well., for the sake of a few pounds

cant be too safe with electricity

regards

tony

 

[Flyfisher]

We get about 5% of total UK power from France this way.

. . . of which, quite a high proportion is generated using nuclear power, I understand.

 

[Lime]

We get about 5% of total UK power from France this way.

. . . of which, quite a high proportion is generated using nuclear power, I understand.

Quite!
All of France's spare electricity comes from nuclear power.
I believe the electric current always flows in our direction.
What happened to the idea that Paris and London would balance each other out because they were in different time zones?

 

[NT]

Lime

Unfortunately grossly unsafe situations can occur in new installations too if alterations are made with no thought to the consequences.
Small apparently insignificant changes can be potentially disastrous to the integrity of an electrical safety system.

The whole point behind the design of modern electrical systems is that they are not vulnerable to this problem. They are designed with several safety barriers in place for precisely this reason, not just one.

Even the water supply pipe coming into the house can be a source of danger.

yes, all incoming services are, primarily due conduction of remote lightning strike.

Ensuring the bathroom metal fittings are all at the same potential (even if live) is a sensible precaution for safety.

This does not follow.

Firstly concrete floors are somewhat conductive (google Ufer if you dont believe me) and remain earthed regardless of what the equi bonding does. In an ancient install, crossbonding assorted unearthed items merely reduces the risk of accessory to accessory shock and increases the risk of accessory to floor shock.

Secondly, and much more importantly, the number of lives expected to be saved by bathroom bonding new installs is less than 1 per decade. When you look at what it costs, and the return, you'll quickly notice that really any realistic safety policy using this money could return 100-1000x more lives saved. Spend that money instead on mandated non-slip floor varnishes for varnished floors and stairs if you want a real return. Hunting rabbits in a land with no rabbits is a bit pointless.


Flyfisher wrote:

I wasn't being specific, just thinking that it's generally a good thing to bond all metalwork together, certainly within a normal touching radius (2m?), so that any conductive surfaces you could touch would all be at the same potential.

The problem is
a) it doesnt cause that to be so.
b) it introduces risk as well as removing risk.
c) There is no significant risk to begin with

Shock current: I thought RCDs are designed to keep any shock below the fatal time-vs-current curve for the average person.

Not any, no. They dont act at all on L-N currents.
And within safe i*t, no, but theyre a big improvement in this respect. If for some reason your shock current were especially high, or to a sensitive region, their quick but not infinitely quick response would not save you. They also are not fail safe, so RCDs occasionally fail to do anything, regardless of whose frying tonight.

Presumably, a L-N shock would be a rare occurence in the normal course of events

yes, relatively. There are still old neutral chassis appliances out there in use though. I dont mean the modernish (1980s) non-exposed live/neutral chassis stuff, but rather the much earlier transformerless valve radios ('universal' sets) some of which have exposed metalwork connected to the mains neutral. Then there are the various miswired appliances with neutral cases, and miswired sockets etc.


No-one has yet died from a BC lampholder. ES do kill occasionally.

One of the problems with them is that conditions are not identical, as users too often take full advantage of this apparent extra safety, and start doing things they never would have before, like using power tools outdoors in mild drizzle etc, thus negating any safety benefit and more.

That reminds me of the old argument against seatbelts and all the other safety features in modern cars - they make drivers feel so safe that they take more risks. Hence the old joke about mounting a long sharp spike in the centre of the steering wheel, which would certainly make any driver be very careful

But you see the difference, that the car spike one is obviously false, whereas the former is unfortunately the reality. This is one of the reasons why I'm not convinced RCDs will actually save any lives.


tony

however you only get a shock if there is a difference in potential(voltage), so if everything is connected together it all has the same voltage and you are unable to get a shock.

Equi bonding doesnt do that, it leaves your wet/damp ground floor at earth potential, along with the wet walls.

If we changed standard TT earthing practice over to using Ufer earths that problem would be history, then equi bonding would become much more effective. Plus Ufer earths are a good deal safer in the first place.


tony

so if you were ordering this work to be done in your house would you tell the electrician you dont want bonding.?

no, and I think the reasons why are simple enough to see. That does not mean its across the board use is soundly justified on safety grounds, any more than general use means that for any other regulation.


NT

 

[NT]

biffvernon

I was talking to someone last night who has a house in France where he's just been doing some re-wireing. It seems regulations there are rather different so we concluded that they must have a different sort of electricity. We couldn't work out what happens in the cross-channel interconect where the two types of electricity meet and mix, but the wine he had brought back was quite nice.

they have a mix of red currents and white. But at least they make grape mileage out of it.


NT

 

[NT]

Penners

my installation is the type that has an earth cable running from the consumer unit earth block out to a pin in the earth - is that TT?),

yup. For you an RCD is a basic must, not an extra safety feature, because your relatively high impedance earthing system is not effective on its own.

I should ensure that all my bathroom metal fittings are bonded together in 10mm2 earth cable, with the end of the run connecting back directly to the earth block in my consumer unit. No other bonding is required.

Bathroom metalwork should be bonded to each other with insulated 4mm^2, but this is not run back to earth anywhere. This is not earth bonding but equipotential bonding. For you such crossbonding has some minor safety value.

The earth cable run back to the CU would pass very close to the stainless steel sink in the utility room. Should I make a connection that too?

safer not connected on a TT system.


tony

Earth clamps on pipes have to be accessable.
On my radiator it has a little lip at the bottom and i use a special earth clamp that is u shaped that tightens on the metal and provides a good earth, these are called ezy earth clamps or something the electrical wholesaler tlc sells them

and you can mount these behind the rad so the clamp screw is accessible but the thing is mostly unseen. Pick the right connection points and the equi bonding can remain out of view, or at least unnoticed.

Its not a difficult task and i think its a worthwhile task for safety sake.

cant be too safe with electricity

Over a quarter of the population dies unnecessarily. Perhaps if they spend less time trying to crack down on zero risks they'd start addressing the real issues.


NT

 

[Lime]

Quote:
Its not a difficult task and i think its a worthwhile task for safety sake.
cant be too safe with electricity


Over a quarter of the population dies unnecessarily. Perhaps if they spend less time trying to crack down on zero risks they'd start addressing the real issues.

I totally agree.
Part P was brought in on a fanfare of safety hype which, relative to accidents on a daily basis, is absolutely false.
I have been told the death figures are 6-7 people a year caused by electrocution. (accidents on stairs are about 10 times this figure.)
I don't think part P will do anything to prevent death but it will make a closed shop lots more money.

What a pity the real cause of accidental death in this country is not more profitable: bad driving.
Do you know how many are killed every year?
3500
Compared to electrocution, preventing annual car deaths would save more lives than 365+ years of part P and all the other paraphernalia.

Safety needs to be considered in relation to relative risk.

 

[Flyfisher]

Ensuring the bathroom metal fittings are all at the same potential (even if live) is a sensible precaution for safety.

This does not follow.

Firstly concrete floors are somewhat conductive (google Ufer if you dont believe me) and remain earthed regardless of what the equi bonding does. In an ancient install, crossbonding assorted unearthed items merely reduces the risk of accessory to accessory shock and increases the risk of accessory to floor shock.

Surely cross-bonding will eliminate the risk of accessory-to-accessory shock? I take your point about the possible increase of accessory-to-floor shock, but this can be protected against by an RCD whereas an (unbonded) accessory-to-accessory shock could conceivably be a L-N shock if there are two faults, and also likely to be low resistance hand-to-hand rather than a high-resistance path to ground. OK, this might be scraping the barrel of probabilities, but surely eliminating a small risk for the sake of a bit of earth bonding is beneficial?

Safety needs to be considered in relation to relative risk.

True . . . until politics raises its ugly head, where being seen to be doing something is usually more important than being effective

 

[Mark]

Its interesting this new fangled electrikery stuff.

Earths and bonding seem to be a big issue. Looking at the niceic site, the top ten questions they are asked by their own members...
http://www.niceic.org.uk/approved/topten.html

half the questions relate to earths/bonding. look at q10 for bathrooms.

I have to say I have had different responses from different electricians when I have asked about our tt system (which I suspect has poor earthing). Also different web sites give different advice.

As I understand it, in laymans terms, we have Live - the supply, Neutral - which is in reality an electricity company provided earth for the return of the live circuit, and earth which is a safety connection.

main bonding which is intended to ensure that any fault in any electrical item results in a current flowing to earth rather than just energising the device. Main bonding connects the main earth to the gas and water pipes.

supplementary bonding or equipotential bonding is intended to make sure that there isnt a potential difference between things in an area - generally bathrooms. it connects things which are not part of the electrical installation but which could provide a path to earth which would lead to a shock if you touched say the radiator the faulty lightswitch at the same time.
there seems to be lots of confusion about whether or not to connect equipotential bonding to the main earth. So far as I can tell, the answer is yes. as i understand it though this could be dangerous with systems with poor earths because a fault could result in a voltage on the bonded items which could then earth through anyone touching them

I would be a lot happier if electricians could give me a consistent answer. saying 'well things changed between the 15th and 16th editions' isnt very helpful.

 

[Flyfisher]

Interesting link, and you've only got to read the listed answers to realise how a lot of the inconsistency arises - liberal use of 'usually', appropriate', 'normally' etc etc leaves plenty of scope for alternative interpretations.

Q&A number 1: can a washine machine or tumble dryer be installed in a bathroom?
Answer - yes, if all of BS7671 is met . . . part 6 requires additional safety measure for the shock risk in a bathroom . . . regulations permit such a machine in zone 3 (600-3000mm from a batch) provided it is suitable for the location . . . agreement of the manufacturer must be sought. In other words, yes if it's suitable and we leave it up to the manufacturer to take the risk on that one!

Q&A number 2: How close to a sink can a socket-outlet be installed?
Answer BS 7671 does not specify a minimum distance . . . but Reg 512-06-01 requires due account to be taken of all external influences . . . . all equipment to be of a design appropriate . . . . take account of the conditions likely to be encountered . . . Accessories used in domestic installations are not designed to be splashed and therefore not suitable for installation close to a sink or draining board . . . . NICEIC recommends that socket-outlets and other accessories should be located at least 300 mm, measured horizontally, from a sink or draining board, where they are unlikely to be splashed.

Credit to the NICEIC for at least trying to turn all that into a firm recommendation, even though they couldn't resist the 'unlikely' qualification - so what happens if your socket 300mm from the sink does get splashed and causes a problem? What a minefield.

Imagine being caught up in a lawsuit and trying to justify that your interpretations and considerations of the 'rules' are appropriate while being faced with an array of expert witnesses with opposing interpretations (because you can always find experts with opposing views!)? I'm sure this is why organisations end up hiding behind certificates of compliance in order to limit their liability (and satisfy their insurance companies) . . . . it's not my fault your honour, I've got this certificate . . .

The lunatics are taking over the asylum

 

[Penners]

Tony - thank you for your correction to my attempt to translate all of the views here into an action plan for my bathroom.

As regards main bonding of my TT system, I have no gas pipe, my water main rises out of the floor in alkathene - copper is connected to it about a foot above the floor, and my oil supply from the tank runs above ground (although in contact with the surface of the ground in places). Should I stilll bother with main bonding?

I understand your recommendations for the bathroom. In our utility room we have a stainless steel sink, a washing machine and a tumble dryer (both powered, of course, from 13A sockets), plus a ceiling light with normal wall switch and a radiator. Should I put in an equipotential bonding cable from the sink to the appliance casings to the 13A sockets to the light switch to the rad?

The airing cupboard is a considerable distance from the bathroom and the utility room. Should I still connect across all the pipes in there?

Many thanks for your continued help.