I've had a bit of a search but can't find posts on the forum from people who have practical experience of body making using these two similar processes:

http://www.rqriley.com/frp-foam.htm

The first one is essentially a boat process applied to a road vehicle.

http://www.autospeed.com/cms/A_110990/article.html

On the second one I can't work out if the foam was dissolved out of the final body or not.

I realise both these methods wont have the potential for perfect parts in the way the buck - mould method would, but I keep being drawn back to them due to the reduced cost to get a one off body and I'm fairly sure that the time to make the body will be quicker too.

I'd like to hear opinions from people that have practical experience in working these methods.

I'm not going to be aiming for a ultra-gloss type finish anyway. Ideally I'd like the finish you find on the inside of bagged parts on the outside, then finished with flat paint. Kind of a rat-rod approach to FRP on a sports car. Sound stupid? It probably is.

the man you need to speak to is

Smart51

He's been building his cool little car in just this way and simply tearing through the build. I'm sure he will be able to give you plenty of practical advice. Personally I've built large (3m) r/c plane wings and 2m floatplane floats this way and got superb results. The foam should always be left in place as it is part of the structure.

Linky

Look at this thread here. It isn't quite as the Robert Q Riley method but it is similar.

In theory it is simple. Polyurethane foam carved to the shape you want. GRP inside and out. Body filler to smooth. Job done.

In practice I've found it more tricky. Sanding curves into the foam cause it to warp. I've taken to fibreglassing the inside of the foam sheets before carving the outside to the required shape. Even this isn't fool proof as the outer layer of GRP shrinks as it cures pulling the panel out of shape. I've had to cut the inner face of the GRP to bend it and then GRP over the cut to pull things back to how I want them. This spoils the curve on the outside a bit so more filler in places than I'd like. One of the panels is now so heavy I'm going to use it as a buck to pull a mould to make a lighter single skinned GRP panel to replace the heavy original.

On the plus side, it is very rigid. More than a single skinned GRP panel of the same weight.

Another warning. Woven rovings don't bend over radiused foam nearly as well as CSM. This leads to the rovings lifting off the foam leaving air bubbles. Bridging paste for car body returns is great stuff for building up the surface of large radiuses (20mm) into small radiuses (3mm). In other places, I've laid the GRP flat over the surface then chamfered the edge of the foam away from the GRP and filled with the bridging compound, then glassed over the edge, giving a nice 90° corner.

It is not difficult, but it is messy and time consuming. Don't expect a perfect suface on the GRP but body filler up from the GRP to get a nice finish. Once painted, you won't see what's underneath.

I think some resins are more prone to shrinkage than others, and some sort of filler in the resin can help also?

I realise now that in the electric car example the polystyrene blocks were bonded to the inner composite panels of the chassis which makes the chassis a body one glued lump.
This won't work for me but is an interesting approach, I wonder about the durability of polystyrene in that application as well.

Actually this article describes what I've had in mind for a while:

http://www.mci.i12.com/carbon/petrol_tank.htm

60 deg C I could do just by wrapping in black plastic and leaving on drive for an afternoon