Is the pipework really metallic underground? Rigid pipework would be a pain to feed through ducts (presumably with 90 degree bends to arrive out of the ground vertically) - and all the pre-insulated tubing I've looked at has been plastic (usually PEX) which is much more flexible.

Could it be what we're looking at is a transition from plastic to copper, that's then covered with thermal insulation?

Test wise, the old school method would be to test against a temporary electrode (say a long screwdriver shoved into the mud outside) - even if it's off by a few hundred or even a thousand Ohms or so, it should still give you a sensible ball park reading, since you're looking for tens or hundreds of k Ohms.

   - Andy.

seconded. Plant a real extraneous conductive part in some real mud, and measure the resistance  to it - no need for power or a MET.

Biggest risk of bonding here is that you are introducing links between the neutrals of substations that would normally never meet - 1km is a 'long way' in LV land. During thunderstorms and so on very large voltages may be induced as well as the risk of very significant diverted neutral currents in normal operation.

Even if test show it is an extraneous conducting part,  serious consideration should be given to the use of insulating pipe sections and enclosures for the bit of visible metal as a safer  alternative to the normal 'bond everything in sight'. That might depend on the circulating fluid and its conductivity however.
Mike.

Hi, Thanks for the reply. Yes the pipe work is fully metallic, Its installed in trenches in the early construction phase and welded. You're right I shouldn't be that concerned about a few ohms because of a poor earth. I took your advise and used a long lead on multiple exposed parts from this building and buildings nearby e.g Cranes, Temporary power panels, MET next door. I got a mix of readings from 0.05mo to 0.02mo. 

0.02mo.

=20kΩ - so that makes it a simple answer then!

   - Andy.

Andy may be tongue in cheek. I took that to be 20 mΩ or 0.02 Ω.

Looks like the meter is set to Mega Ohms, 500 V. Reading of 0.02 M ohms 

If a current value of 30 mA is used in the equation , an item would not be considered an extraneous-conductive-part if the resistance (Rcp) is at least 7.67 kΩ ( 0.00767MΩ)

Ah yes, I need to go to Specsavers. 

If a current value of 30 mA is used in the equation

I think that would be an unusual choice by the designer - an installation where users could get noticeable shocks between parts (even if it's not quite enough to kill them, well 95% of them at least) wouldn't be considered satisfactory by most members of the public. I would have thought that most would have picked a much lower value - typically yielding required resistances between 23kΩ and half a MΩ.

   - Andy.

Indeed - 20-30 Kohms, reflecting 10mA or less into a full mains fault condition would be a more appropriate upper limit (*).  of course Full voltage fault conditions are rare in any one property but as more are connected in parallel, as happens is a system like this, the chances of something nasty happening in one or other increase the odds.

Mike

* And that sort of resistance is the kind of thing you can see from almost casual contact with ground such as the painted 'skids' of containers, or wet and mud caked tyres on farm trailers.

Even a toy tent peg manages a few k ohm or two.

Indeed - 20-30 Kohms, reflecting 10mA or less into a full mains fault condition would be a more appropriate upper limit (*).  of course Full voltage fault conditions are rare in any one property but as more are connected in parallel, as happens is a system like this, the chances of something nasty happening in one or other increase the odds.

Mike

* And that sort of resistance is the kind of thing you can see from almost casual contact with ground such as the painted 'skids' of containers, or wet and mud caked tyres on farm trailers.

Even a toy tent peg manages a few k ohm or two.