Is this what you are describing?

For the 2nd case - yes exactly that.Similarly where the insert is downstream of the main bond but before some other location that may requires bonding (e.g. a bathroom), where faults from elsewhere in the installation provide the 'other potential' rather than true Earth.

As an aside I did some experiments with plastic fittings in copper pipe a while ago - when filled with my local (fairly soft) water, I was getting readings in the region of 10kΩ for a simple straight joint, and higher for Tee fittings. There's likely to be a lot of variation with water chemistries (basic hardness to anti-corrosion additives in central heating circuits) - but like your 600Ω it can often fall between two stools of being too high to provide positive bonding (e.g. <<1Ω) and too low to provide isolation (e.g. >23kΩ).

   - Andy.

gkenyon said:

(v) any metal pipework and other extraneous metal entering the location (not IN the location) is connected back to ME.

It's seems the regs are satisfied if extraneous CPs have an insulating section on entry to a building, or are effectively connected to the MET on entry to the building.

Which make its sound like as long as I confirm that, my duty is done.
But I wonder about these EXT CP as they travel through the building. 

On a periodic inspection. I R2 wonder lead pipe work in bathrooms. This is where I get into doubt.
I know the pipe work satisfies the regs on entry into the building, but does not in a location.

Because everything is generally hidden and tiled over, we only have out meter readings to go on.
So I get a 600Ω reading on pipework back to the MET or a high sup bond reading in the location.

What to do?? Not worry and let the RCD take care.

But  we not satisfying note 6  701.415.2 as we are no longer effectively connected to the MET?

Thank you .  Yes thats the exact problem, what to do?

what to do?

Slightly flippantly, not to lie awake at night worrying ! It is unavoidable that there will be fault voltages somewhere, but

1) Realise that if the system is well designed, all these dangerous voltages are only present between the instant the fault develops, and the time it takes the ADS in the form of an RCD/ RCBO or whatever to trip the supply off. When that is working  properly, that duration is a small fraction of the half a heartbeat period or so that a shock needs to last to have a sporting chance of killing someone..

2) What is the chance of any-one even being there and  touching both things during that short dangerous interval anyway ?

3) Most accidental contact shocks are not good connections to large areas of wet skin with one contact either side of the heart,  so the resistance presented is higher due to dry skin and smaller contact areas, and at the same time,  the current path is probably not hand to hand.

what to do?

Minimize the risk in places where folk are wet and not fully dressed  such as bathrooms and swimming pools, and use bonding to move the risk out to other places that are less risky.

And, do the periodic tests of the RCD function please. 

Mike.

JohnE said:

So would 'A' would be what you described?

With the advice to bond just outside, A is in effect what I was describing.

B would also require a bond to the cpc of the equipment (if it were supplementary local equipotential bonding). Otherwise, B might as well be C.

C is the situation requiring RCD protection (the "exception" if you will).

600 Ω is very low - it would perhaps be equivalent to << 10 mm of plastic pipe with tap water, or 30 mm of central heating pipe, based on the resistances in this IEE WIring Matters article: www.voltimum.co.uk/.../201207138449.pdf