NT wrote:

I think there are two things at issue he

One is semantics - yes you could argue that if you are changing nothing,
then installing bonding is not "required".

Thats a true fact, not semantics

Not really - the wiring regs still say its required. The fact that there
is no legal compulsion to update the exiting install to follow those
requirements does not make the requirement go away.

We could include extra verbage at the top that highlights what is meant
by "required" etc, but then why single out this article? Few of the
others contain similar. Its not as if this one is focussed on specific
advice on a task like changing a CU or installing outside power feeds.
Its just an overview of the right things to do in bathrooms and a heads
up as to why they are different from other rooms.

However, if no change is
being made or contemplated, why refer an article on bathroom electrics
in the first place? Chances are if you are looking for information on
this, then you plan to make some alterations, at which point this work
becomes required.

Many times people get a PIR done that says you need to fit equi
bonding, and they mistakenly believe it to be so. Our article should
tell them the truth, that there is no requirement to fit it unless
youre carrying out electrical work.

PIRs don't include an in depth analysis of the implications of things
like missing bonding. In some cases you could argue that a particular
installation has very few if any failure modes where EQ bonding would
mitigate. Equally however, there may be some that make it very well
worth having.

By all means tell em that they are not legally obliged to carry out the
work; but if the PIR has highlighted a fault that should be fixed,
unless they are in a position to carry out enough analysis to say if
there is a better cost benefit trade off to be had elsewhere, they ought
to carry out the fix. Any other advice would be reckless.

A couple of example scenarios:

e.g.

1) 16th edition PMR head end, small shower room, plastic pipes
everywhere and the only electrical circuits in use are the lighting one,
and a spur from a upstairs ring circuit to feed a towel rail. The lights
are SELV downlighters. The actual risk of serious shock in this
circumstance is negligible. Since mains never manifests in the room
anywhere other than the towel rail, and there are no other extraneous
conductive paths (the CPC of the lighting circuit does not present in
the room - the metalwork of the down lighters is floating, the switch is
non conductive and out of reach on a string). According to to the Regs
Bonding is still required, however the implications of ignoring the
requirement are to all intents nil as things stand.

2) 15th edition install, re-wireable fuses, TT head end (overhead wire
supply), no bonding at all (main or supplementary), a VO ELCB is
connected inline with the main earth provided by a gas pipe. There are a
number of borrowed neutrals and neutral earth faults. a 13A socket is
installed beside the sink to allow a hairdrier etc be plugged in.
Bathroom has CH piped towel radiator, metal pipes to all taps and
cistern, and an electric shower, plus fan heater, and some class 1
luminaries on a low ish ceiling controlled by a nice polished chrome
wall switch next to the shower. In this case there are any number of
fault scenarios that could present serious and life threatening shock
hazards i.e. faults that could leave high touch voltages on metal
surfaces for extended periods, with ready access to an independent earth
close to hand. Now obviously a PIR would highlight all manner of
shortcomings here. The long term solution would probably be a package of
measures including a re-wire or at least a new head end. However
installing main and supplementary bonding PDQ would be cheap and
effective way to lower the rooms status from "death trap" to something
more acceptable.

Theres nothing wrong with also adding a section of whether we think
you should, but I dont think we ought to tell porkies by saying yes
you're required to do it.

Depends on your interpretation of required. See comments above.

Also if we are addressing competent DIYers seeking to
improve bad things on their electrical system, then encouraging them to
add bonding when its required but missing, is a "good thing" IMHO. Its
cheap, easy to do, and depending on the circumstances can make a life or
death difference should you be unlucky enough to have something go bang
unexpectedly.

So opinion is mixed. If, as we did once before, guess the average cost
of installing it, multiply by the number of houses, and divide by the
number of lives it saves per year it works out at a phenonmenally high
price per life, when many other simple domestic works for the same
money/time would yield far greater safety improvement. It just isnt
warranted on safety grounds.

Statistical exercises like this can be useful for setting government
policy[1]. However we are not really concerned with overall impacts,
targets, costs to the nation etc, and are focussing on one electrical
installation. What is best for the majority has no bearing on what is
best for Joe Internet Searcher, 32 Acacia Avenue who has just moved into
a place and had a PIR throw up reams of faults that he does not
understand; missing bonding being one of them. The only sensible advice
we can give in a general purpose guidance article is to say you need to
be lead by the professional opinion of the people you have paid for
advice and here is what the bonding does and how it works if you want to
know. The implications of it not being there may be trivial, or may be
very serious, we can't tell you from here.

(Note also this is partly a circular argument. We have very safe
electrics in the UK as a general rule - both fixed, and appliances (the
former more so than the latter). Deaths are very rare, and injury
relatively rare. Part of the reason for this is that we do specify and
install safety measures like EQ bonding, RCD protection and a host of
other measures. Its almost impossible to say how many extra injuries etc
we would have had without these measures, other than a casual
observation that numbers of deaths etc have been falling for years (at
least until the intro of part p anyway), and electrical installations as
a whole have been getting better as more are updated and rewired to
current standards).

Why not...

Lets say install cost £20 materials, 4 hours labour for a diying
novice, including going and getting the bits.
Total cost £20 + 4x7-15, perhaps £40 as a balpark figure
= £60 total

So the question that matters to me is, could that £60 save the life of
someone in the family should something go wrong? The answer is usually yes.

So the next question is, how likely is it that something will go wrong
in a dangerous way, with my particular installation, that means I would
get to "use" that £60s worth of investment and hence make it worth
spending here and not on mitigating some other risk. If the answer is
"high to moderate" then you spend the £60. If the answer is "I don't
know" the next question is can I get sufficient expert analysis at less
cost, that would tell me if its going to make a difference. If the
answer is no, then you spend the £60. If the answer is yes then you
spend the money on that instead, and then possibly spend the £60
afterwards as well depending on the outcome.

x20 million houses = £1.2 billion

If this saves one life every 5 years, thats a cost of 6 billion per
life saved, which is orders of magnitude higher than a whole swathe of
more practical constructive measures one can take.

We are not talking about spending billions, since we are not writing an
article to advice policy makers. We are talking to the individual ones
looking at your £60 bill.

Secondly, there is the issue of severity. I expect based on comments you
have made in the past that you don't consider the lack of "required"
bonding to be of particular concern. However as I see it (and as Adam
highlights, as any form of proper inspection would see it), it is
classed as a fairly severe fault.

That only shows a recognised flaw in such testing, namely that the

Its a flaw in the process that one accepts since it controls the cost of
such testing. A PIR plus the cost of remedial action is in many many
cases going to be significantly less that the cost of a test with a
detailed analysis and risk assessment to go along with it. Also doing
the work once often saves repeating the exercise later to justify that
something non standard is actually ok in this circumstance.

results given are sometimes entirely unrealistic and to encourage
needless spending.

Agreed; and sometimes the results will be highly realistic, and
encourage spending on something that needs doing as a matter of urgency.

How do you differentiate without detailed technical knowledge?

Perhaps there is a whole new article there; "My PIR said X, how much
notice should I take?".

[1] although following the logic, part P would never have been
introduced - which shows up the disparity between stated and real agenda.


--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/