Ok, but would you do it if neither is liable to introduce a potential?
No, I wouldn't, since there would be no point (like bonding two spoons together
). However, I think this has become a discussion about what "liable to" means. I take it to mean 'could'/'might' (under certain circumstances), whereas you seem to take it to mean 'does now'. I suspect that we may have to 'agree to disagree' about that one.
I must say that I would argue that my interpretation seems more consistent with the reg's definition of an extraneous-c-p. If they meant a conductive part which
did "introduce a potential, generally Earth potential" (which is what you would determine, for one point in time, by your testing), wouldn't it just say that, rather than "liable to"?
That is less likely than introducing a permanent negligible impedance path to earth which might cause one to have a severe shock.
You can't really say that as a generalisation, since it depends totally upon how likely the "liable to" event actually is.
Up to you but a 1% chance of that is surely better than the now 100% chance of causing the above.
It's not a 100% chance of having a severe shock - it's a 100% chance of that shock happening IF someone simultaneously touched it and something else which (due to a fault) was live, at the very time that fault was present, which is an extremely unlikely occurrence.
The problem is that, as so often, we are talking about, and try to compare, two minute risks. If one (a) omitted all (required) main bonding OR (b) earthed every bit of fixed metal in a house (I'll let you off the spoons and forks
) then, in either case, I would think that the probability of anyone suffering a severe shock as a result would be incredibly small (since it requires not only a fault, but a victim who is unlucky enough to be touching 'the wrong two things' at the very time the fault is present), wouldn't you?
... but are you actually saying that, if you saw my metal LPG pipe entering my cellar about 18" below ground level, you would not bond it because when you 'measured it' (during a dry period) it had a very high resistance/impedance to earth?
Good - so what is the difference?
That's what the rest of my paragraph (which you didn't quote) went on to explain ....
I suppose you are bonding them in the English meaning of the word but not by what we usually mean by Main or Supplementary bonding and not in the sense of any requirement of the regulations ....
I agree that it's not BS7671's meaning of Main or Supplementary bonding, and is not a requirement, but it's a common concept in relation to the wider scope of electrics and electronics. For example, the 'wrist strap and fly lead' worn by people handling static-sensitive electronics is a form of equipotential bonding.
... but surely you are doing a Bernard and guarding against the Blue Moon event rather than the more likely events when you would not want them joined.
Am I? Is what I am 'guarding against' any more of a 'Blue Moon event' than the possible (but incredibly unlikely) events that concern you in relation to 'unnecessarily earthed metal'?? As above, the real problem is that we are all talking about incredibly unlikely events - but that's also what many of the regs are designed to 'guard against'.
In fact, I would say that the thing I am 'guarding against' in my workshop is essentially the same thing that you are 'guarding against' when you advise against 'unnecessarily earthing' things. I am considering the (far from uncommon) situation in which one of the metal bench tops is at earth potential (because a Class I item is sitting on it) and a fault arises in something on the other bench, rendering it 'live'. Very unlikely, I agree, but I don't think any less likely than the situation you are contemplating, is it? In your situation, you can advise against the deliberate earthing of the metal. However, in my workshop I can't do that, since any earthing is incidental/accidental - but what I can do is ensure that the two bench tops could not (in the event of a fault) come to have significantly different potentials.
Kind Regards, John