Indeed (as we have discussed) - although, as I said, the 'requirement' is pretty meaningless if that 'length' is measured from arbitrary point (such as entry into the CU/DB, which could be a long distance from the origin of the installation (and, I suspect, in practice rarely less than 0.5m). My comments were, as I presume you understand, in response to scousespark's suggestion ...The length is not to protect the cable, but to ensure the spd is able to do its job.
As I said before, if the concern which has led to the 'length requirement' is in relation to high frequency (rapid rise-time) transients, then electrical consideration might suggest that the connection from the origin of the installation to the SPD (and everything else) should perhaps be as long as possible (i.e. with a minimum, not maximum length requirement) - since it would surely be better to attenuate the transients before they get to the SPD (and everything else), rather than make it easier for them to get to the SPD (and hope that it can 'deal with them'), isn't it?The length of cable comes from BS7671, regulation 534.4.8 which states the total length of the cables (live + PE) should preferably be no longer than 0.5 metres, but should never be longer than 1 metre. I assume this is based on research to ensure the cables are not damaged during the spike.
Kind Regards, John