Resolving high Zs on Ring Final Circuit

Yes, reasonably. Given the new CU, you could split the rings half way round into 16 A radials, but that may not gain very much.

I would speak to the DNO. With a Ze of 0.65 ohms they are not obliged to do anything, so there is likely to be a cost involved. When I was converted to TNC-S (for a slightly excessive Ze) the link between N and E was made under the side lawn.

Downside is the dreaded broken N (as well as any cost).

If you are installing a modern CU, will you not have RCD / RCBO protection for these circuits? If so, you are not limited to the Zs (with the 0.8 “rule of thumb”) in Tables 41.2 to 41.4.

Since 17th Edition, RCDs can be used for automatic disconnection of supply.

Where RCDs are used, volt-drop limits the circuit length. See Table 7.1(i) in the OSG for examples of circuit length difference you can have with RCDs on TN-S.

I assume they can do this by modifying connections at the supply head, they need to visit anyway to install an isolator and move the meter a bit to make room for a new consumer unit.

Moving the meter and fitting an isolator is usually a job for the supplier (or their contracted meter operator) rather than the DNO. Generally the DNO stops at the outgoing terminals of the cut-out. In some circumstances the DNO will alter things beyond the cut-out, e.g. re-positioning a meter at the same time as re-locating the cut-out - but won't do so unless it's incidental to a proper job of theirs. I doubt that the supplier/MOP will be able to provide PME - as that requires knowledge of the state of the DNO network - not everywhere is ‘PME ready’.

+1 for Graham's point that with an RCD, ADS isn't going to be an issue.

   - Andy.

Do you have any non RCD protected submains for which the 0.64 will be a problem.? 

As I see it, it is not a killing matter as it is,  the fault current to the furthest socket is ~ 200 amps, so I agree a 32A B type may not fire instantly under extreme low voltage but it will still go quite fast, and the RCD part should go with a resounding thwack for the safely of life stuff.

Another thing to check if you have not already is to see how much of the 0.64 ohms is the water pipes and associated, maybe picking up via the earth terminal next door, and how much really is the company earth that is supposed to be the main one - it may not be as good as you think.

Your mileage with the DNO may vary..  many use 0.8 ohms as a upper action limit.

mike.

Ze of 0.65 ohms

What's the installation's general position w.r.t. the LV distribution network? 0.65 might be reasonable for somewhere at the end of a country lane the opposite end off the village from the transformer, but if it was more in the middle of town I might be wondering if it's more of an early warning that something is rusting through. If it is the latter, then there may be more problems to come.

   - Andy.

AJJewsbury: 
What's the installation's general position w.r.t. the LV distribution network? 0.65 might be reasonable for somewhere at the end of a country lane the opposite end off the village from the transformer, but if it was more in the middle of town I might be wondering if it's more of an early warning that something is rusting through. If it is the latter, then there may be more problems to come.

If it's PILC, might the lead be cracking at the junction?

That was assumed to be the problem in my case, not least because people had been parking directly over it on the grass verge. ?

Thanks all for the responses, which help clarify my thinking. 
By the way I measured Ze with the earth isolated from the installation not that there was much change when I did reconnect to include the water bonding in to the equation.

The house is in a reasonable size village, I don't know where the substation is but I wouldn't have thought it was that far away.

I think I will just go with my plan as it is rather than ask for conversion to PME as it sounds like the conversion would mean digging up the driveway and if I have RCD's in place not an issue anyway.