If you are going to sand your frame to the bare carbon, you really need to put some kind of UV protectant on the entire frame. Whether it's an automotive grade clearcoat or otherwise, if you don't UV protect it, over time the sunlight will begin to degrade the resin system, weakening the structural integrity of the frame.
You don't need UV protection on CF-EP-laminates. The fibres are intransparent for UV light, so the matrix won't geht any damange below the first fibre. You might get some yellowing on the outside of your part, but there is no significant loss in structural integrity
That might be different for GF or AF, because laminates made from these fibres are transparent or translucent.
One of just many pieces of evidence supporting my statement that exposed carbon fiber can degrade when exposed to UV light and moisture:http://me.eng.sunysb.edu/~compmech/downloads/N29.pdf
" onclick="window.open(this.href);return false;" onclick="window.open(this.href);return false;
Have you read the paper? The specimens only exposed to UV lose just a very little percentage of stiffness and strength. And, in fact, the positive effects of moisture at least compensate the negative effects of UV radiation. In addition the wall thickness is relatively low (but something usual for bicycle components), the negative effects will be the lower, the thicker the specimen is (like I said, the UV radiation just gets to the first fibre - behind the fibre there is just darkness
For me this paper points out, that it is not necessary to put on some coating on CF-EP parts, because the losses in E|| and R+|- are less than in most cases for examle numerical mistakes are. In addition E|- will decrease more than E||, so the effect of the decreasing R+|- is also at least partly compensated through redistrubution of stresses within the different layers.
Or, otherwise said, if the bike part has that low safety factors, that the very small effect of UV radiation comes to an effect, you should not ride it...
Concerning the influences of moisture, they are mostly positve, (break strain under Sigma|- + for example), so negative effects of moisture are overcompensated by positive effects in general. See http://books.google.de/books?id=GVN_V1o ... &q&f=false
fore more information. That's one of the standard books in the countrys where german is spoken, it's written by my professor, who was postgraduate with Puck and developed the newer version of the Puck criteria of failure together with Puck. You will hardly find someone who knows more about those things then he does