None. A panel of similar dimensions (width, thickness, height) will have close to the same weight, with vacuum bagged glass being a tad above 2 tonne/m3 and alu being about 2.7 tonne/kg (but the alu is also a bit stiffer allowing for slightly tiner panels). Panel stiffness drives required thickness. GFRP and alu are close there together.Anyone know the weight difference between aluminum and standard fiberglass per square ft? Say on a cowling etc?
Carbon is way lighter, at maybe 1300 kg/m3 and also way stiffer. The biggie however is that with a composite cowl (or any shape) you can easily add a local stiffener and optimize thickness to what's needed. Just forget about that with any sheet metal unless you have deep pockets.
That, a thousand times over. Such extreme weight-saving measured are just a waste of money.Spending too much to save a tiny bit of weight could actually arrest the whole project. One has to weigh (pun intended) the value of some stuff. If it's going to cost you $3000 more, say, to cover with Oratex instead of Poly-Fiber in order to save ten pounds, I wouldn't do it unless I had really deep pockets. You'll just be a lot longer building the airplane, and as the build time lengthens, the likelihood of finishing it diminishes.
It's hard enough just matching the designer's target weight. Most plans-built homebuilts end up too heavy simply because builders weren't careful to copy exactly the designer's plans. They add a few instruments or a heavier tailwheel or a bigger engine or a few more coats of paint to get that deep gloss. Or they decide that those 1/4" x 1/4" rib capstrips are too fragile and make them 5/16" square instead. Or make them 1/4" thick by 3/4" wide so they can glue the fabric to the ribs to save stitching. Or they use Home Depot wood that has cockeyed grain that splits when you bend it so they have to make it bigger or start patching it or reinforcing it. Or they use oak just to be sure it will be strong enough. Or a metal prop instead of the specified wooden prop because they got a deal on Ebay. One has to watch for the "savings" in time or money that most often end up costing way more than one intended, and resist the temptation to modify or upgrade or spiff up. Weight that creeps in also has a way of making an airplane's CG end up outside the specs so that more weight is necessary, in the form of ballast, to bring it back where it belongs.
"Low weight" is not a metric that one should optimize blindly. The virtue of lowering weight is a viscious circle of lower weight, lower engine power, smaller engine required, less fuel required etc. The Dyn'Aero MCR4S is a good example. Outperforms a C172 with a lower stall, higher cruise, smaller engine (100 hp Rotax) by going for a very low weight, thus lowering cost.
If you're committed to building a Cassutt, build it reasonably, not ridiculously light. Because if you look at the money/effort expenses of such extreme lightening, you'd be better off by designing and building a composite wing, a composite (monocoque) fuselage, or just a from-scratch composite plane. sub 400 lbs empty for a starterless, non-electric O200 powered ship then doesn't sound impossible. But then, you could also put on a VW-derivative, cash out the money saved on the engine, have the same performance as a "real" cassutt and have a lot of money left over for fuel...