You are correct regarding the stiffness v. weight. I would add that I recently did a couple of the type tests you are talking about and the flaps were ALOT stiffer. But the weight is a real concern. Part of the problem is that, if strength was the end criteria, then veil wouldnt be the best solution. The random orientation of the fiber, and the short fiber lengths limits the achievable strength. In typical GRP (glass fiber reinforced plastic)construction, chopped strand mat is only used as a thickness builder--gets some panel thickness between the woven strength layers (spacing here yields panel stiffness)
For controlling twist in a long slender flap, the proper material might be two layers of unidirectional layed at a 30 degree bias. Or a piece of .75 Oz flat woven fabric cut and layed at a 45 degree. ( The same way you might orient balsa grain in a laminated panel.)
The trouble is, like you said, the weight can go up fast. I came the conclusion that this would be an appropriate application for maybe a built -up sheeted flap with very thin skins and a CF wrap. Once the CF skin is used, all of the bending and twisting loads become tension in the skin, and compression in the core (monocoque) so applying this stuff over solid parts becomes heavier than required. Alot of the core function is to simply maintain the spacing of the stressed skin. (and I'm sure the structural guys here will call me to task for not mentioning the core-to-skin shear as well)
Used in that way for hollow stab/el, built up flaps, and light fuse sections really makes sense, even with epoxy. But getting high fiber content ratios is nearly impossible to do without vacuum bagging. Otherwise, painting on epoxy and trying to remove excess or sand off weight is not very effective overall.
Also, the difference in strength or stiffness between reinforcing with .5 or .6 oz CF versus regular old glass fiber of same weight can be difficult to evaluate or take advantage of unless very consistent and optimum fiber content ratio can be created--nearly impossible by hand.
Another real problem with CF over balsa is that, even with small amounts the skin becomes so strong that in bending, the core simply crushes. The popular full-scale solution is to use end-grain balsa for core since it is quite strong in that orientation. Also, since usable layup thicknesses are so thin for our purposes, small process variations in fiber content or thickness can create very big stress concentrations and these also can lead to core failure.
All in all though, carbon fiber is a wonderful product. It does take alot of insight to optimize its potential however.
Having been involved with fiber reinforced composite engineering, design, and application for thirty plus years, I actually got to see the introduction of Kevlar, carbon fiber and some others.
Anyone for synthetic saphire or silicon carbide fiber reinforcement?? Yes, its out there.
just some thoughts.
I'm glad to have these forums to discuss some of this stuff--its hard to find other concentrations of modelers interested in these things.