This is my first time working with fiberglass. I'm working on a small fender for my race car. Ultimately, I'd like to do these and some other parts in carbon, but it seemed wise to spend some time practicing and learning with cheaper materials.
I made a mold of an existing fender using tooling gelcoat, fiberglass cloth and polyester resin. That worked really well, and I'm happy with the mold.
But, my first part did not turn out so well. In most places where there are relatively sharp bends, I'm getting large (1"-2" long) voids, which I assume are due to air bubbles forming where the 225g fiberglass cloth separates from the mold while curing.

Any suggestions on how to address this? Should I be using CSM instead of cloth? I've seen mention of others using spray adhesive to keep the glass in place, but I'm not sure how that might affect the mold release. (I'm using spray PVA.)

Original part (which was cleaned up after this shot, before working on the mold):


Top outside edge. The voids are clearly visible.



Leading edge with some clay and stuff still hanging on.

To be honest, the corner problem will get much worse when you start using carbon fibre. Not so much because of the fibre, but because of the different characteristics of epoxy resin. This is why basically everyone uses at least vacuum compression when working with carbon. But in my opinion, if you use vacuum compression, you already have a pump and therefore could do a proper infusion instead. This will give better and much more consistent results. The clear gelcoat used with carbon fibre won't help you at all, because you will end of having the same imperfections under the gelcoat, wich is even worse. Gelcoat is great for cosmetically appealing carbon parts, but only if you got a void free process for making the laminate.