OK, is definitely the mosfet part of the block the one creating the issue... If I apply pressure in the centre of the area is all good and dandy, the moment I apply pressure on the extremes all goes pete tong...

Have probed with the fluke (in continuity mode) and I can't find a derivation anywhere (tried with the header adjacent to the block first then the main ATX header).

The block is also completely insulated except the areas that have thermal tape or the NB.

I've noticed that the stock mosfet cooler has 0.5mm tape towards the exterior and 1 mm tape in the centre, will try to source some more 0.5mm tape (snow permitting) and try that way...