With reference to your article about Castlepoint car park (NCE 18 May) there are several points raised that are worthy of comment.
While temperature differential is an important consideration in precast car park design, tolerances and drainage falls can also mean that planks do not lie 'true' on their seatings.
Bearing spalling in RC support members will occur unless active measures are used to keep load away from the unreinforced 'covercrete' outside the rebar cage in the supporting toes.
I have never seen a prestressed oor plank with end spalling due to bearing problems though, and I doubt that the damage to the planks shown in your diagram has actually occurred due to the presence of prestressing tendons right to the end of the units.
To suggest that bearing spalling could lead to collapse seems excessive. For this to occur, the precast planks would have to pass the rebar cage inside the 'damaged' supporting toes on their way down.
Given that the structure is tied with an insitu topping and cannot spread appreciably, this appears to be almost impossible.
Precast concrete can offer a durable and economic solution for car parks, and I hope that this troubled project will not prevent the continued use of well detailed precast concrete car parks in the future.
P R Bracey (M), Member of IStructE task group for 'Design recommendations for multi-storey and underground car parks', 17 Hazel Grove, Winchester, SO22 4PQ