Consumer unit feed / main tails more than 3m

I think that is my point. How do they manage to do that; i.e. limit the protection of their fuse to a short distance when the identical second fuse protects the whole length?

Ah, but who says it's always going to be identical. DNO fuses aren't necessarily a constant - within my memory the 'norm' has gone from 60A to 80A to 100A and back to 80A (at least in these parts) . With the coming demand for EVs and heat pumps tomorrow might soon be different again. It's not uncommon for DNOs/suppliers to be asked for upgrades on existing installations - so what's the DNO/supplier to do then? At the moment if you ask for an upgrade to 100A they look at the visible bit of the meter tails up to the CU or whatever and if they're less than 25mm² they say no. Simples. So what happens if the tails disappear into the building and re-appear at one (or more) CUs miles away? The DNO then can't be sure that their change won't make thing less safe than it was before without having to do a lot of faffing about on the consumer's side which they're usually very disinclined to do.  With domestic customers they can't just say "we'll do what you ask, on your head be it" - as any court will say the professional DNO/supplier owes a duty of care to the 'ordinary' customer who the courts won't expect to necessarily understand the subtleties of CCC or ADS. So they invent an administrative procedure that avoids the problem for them.

Are you saying that the distributor disagrees "that their fuse affords protection to the part of the installation between the origin and the main distribution point of the installation"

As an aside, one at least does say exactly that (I'll try to dig out which one) - they offer no agreement at all and their documentation explicitly says to treat the situation as if the meter tails were unprotected and apply BS 7671 434.2.1, which as it happens gives a result very similar to what the other DNOs would agree too.

   - Andy.

Maybe someone can explain this? 

Not a standard I'm familiar with and I can't see the context, but my guess is that I think they're saying that if you have one protective device protecting two difference c.s.a. cables, that the total length that can be protected varies depending on proportions of the lengths of the two types of cables - as added together they need to meet the maximum permissible loop impedance.

For instance, say your protective device can protect either 10m of a 'thick' cable, or 5m of a 'thin' cable (before increasing loop impedances mean that disconnection times get too long), then you could have:

0m of 'thick' cable feeding 5m of 'thin' cable, or

5m of 'thick' cable, feeding 2.5m of 'thin' cable, or

10m of 'thick' cable, feeding 0m of 'thin' cable, or

anything in between, so long as the proportions still add up to 1 - so 25% of your thick cable allowance means you can have at most 75% of your thin cable allowance, or 33% vs 67% or 99% and 1%.

  - Andy.

Not a standard I'm familiar with and I can't see the context, but my guess is that I think they're saying that if you have one protective device protecting two difference c.s.a. cables, that the total length that can be protected varies depending on proportions of the lengths of the two types of cables - as added together they need to meet the maximum permissible loop impedance.

For instance, say your protective device can protect either 10m of a 'thick' cable, or 5m of a 'thin' cable (before increasing loop impedances mean that disconnection times get too long), then you could have:

0m of 'thick' cable feeding 5m of 'thin' cable, or

5m of 'thick' cable, feeding 2.5m of 'thin' cable, or

10m of 'thick' cable, feeding 0m of 'thin' cable, or

anything in between, so long as the proportions still add up to 1 - so 25% of your thick cable allowance means you can have at most 75% of your thin cable allowance, or 33% vs 67% or 99% and 1%.

  - Andy.