Can fuselages be reinforeced with nylon stockings?

subtitled : a short course in designing composite structures?

From : Don Stackhouse

Regarding the current thread about reinforcing fuselages and other structures with nylon stockings, I'd like to add some relevant info on composite structure design. I will cover several subjects, but I promise to eventually end up back on the original subject of nylon/epoxy reinforced fuselages.

A composite structure is a mixture of materials, typically a matrix and a fiber of some sort. In general, the fiber is some strong material which is there to carry the loads, and the matrix is there to hold the fibers in position.

It is very important that the stiffnesses and strengths of the components be properly matched. The stiffness of the fiber should be substantially higher than the matrix, and the "strain to failure" (how far you can stretch it before it breaks, usually specified in inches per inch of original size, or in %) of the matrix should be higher than the fiber's. An example of this would be concrete reinforced with steel. Concrete by itself has good compressive strength, but low tensile strength. By adding steel rods, mesh, etc., the steel can carry the tensile loads. Since the steel is stiffer than the concrete, as you stretch the structure the steel picks up most of the load.

Consider now the case of concrete reinforced with rubber rods. As you stretch this structure the stiffer concrete picks up the loads much faster than the rubber, and in fact the concrete will have long since broken and crumbled away (because its strain to failure is much less than the rubber) before the rubber has stretched far enough to carry hardly any of the load. While the rubber may in fact be capable of supporting the required load, it contributed almost nothing to the strength of the original structure. The structure would have been stronger if you had used the space occupied by the rubber to add more concrete! This structure does have the interesting property of holding all its fractured pieces together after the failure, which might be useful in some cases.

In the case of fiberglass/epoxy, the glass fibers are stiffer than epoxy, so the fibers carry the load. In the case of carbon/epoxy this is also true.

Kevlar is a little more complicated. Its stiffness is higher than epoxy, so it also tends to carry the load, but its strain to failure is higher than most resins, so the structure fails when the epoxy has been stretched to its strain limit in tension. The Kevlar fibers are still carrying the load at this point, it's just that the epoxy can't stretch any further. Since the Kevlar is not at its strain limit yet, the structure usually turns into a rope at this point.

The situation for Kevlar in compression is a bit different. The Kevlar molecule actually buckles under compressive load, so although its compressive stiffness is high, its compressive strength is low (only about 40% of the strength of ordinary "E" type fiberglass). Initially the Kevlar fibers carry the load, but when the molecules buckle, that load is shifted to the epoxy, which then crumbles. The Kevlar molecules retain almost all of their tensile strength after failing in compression, so once again the structure turns into a rope, at a load somewhat higher than pure epoxy.

At the risk of re-igniting an old flame war, what about Kevlar vs. glass fuselages? Well, composite fuselages are normally thin shells, and usually fail in landing situations by compressive buckling of the skin (not necessarily the fibers). The Kevlar has a much lower compressive strength than glass, but it also has a lower density, so a Kevlar layup is quite a bit thicker than a glass layup of the same weight. Compressive buckling typically occurs in thin shells at a load far below the compressive strength of the material itself, and is sensitive to stiffness and VERY sensitive to wall thickness. In thin shells subject to buckling of the skin, the extra stiffness and wall thickness of the Kevlar laminate can offset its lower material compressive stiffness and delay the buckling of the shell walls, so that the strength of the overall structure in compression might be higher than the same structure of similar weight done in fiberglass. I believe that this is the explanation for the test results the Anonymous Waco Dude reported last year for his crash tests of glass vs. Kevlar fuselages. If you had a structure that failed more in pure compression rather than buckling of the skin, such as certain types of cross-section contours and some sandwich structures, the results could favor a fiberglass layup. You can only make specific conclusions for specific designs on this one.

In general, however, if the structure is failing due to skin buckling you are not getting the fullest advantages possible from the material. It's possible that a re-design to eliminate the buckling failures could increase strength while reducing weight. Whether or not that can be done for a reasonable price is another matter entirely!

Now, what about nylon/epoxy? (see, I didn't forget the original thread!) Well, nylon is LESS stiff than epoxy, so the epoxy does most of the load carrying. I like the way Burt Rutan put it at one of his seminars, when discussing Dynel/epoxy, which is a similar structural situation. In his opinion, the Dynel was a great way to hold the epoxy in place while it cured. The epoxy does all the real work of carrying structural loads in this system. You would do about as well by just applying a layer of epoxy, although the fabric helps give you a nice, smooth, uniform thickness to the layer, which is also important. Although you are getting some additional strength from the nylon/epoxy, you would probably get more strength for about the same weight by using 3/4 oz. glass cloth instead of the nylon stocking.

One advantage to nylon/epoxy over fiberglass/epoxy is that the nylon/epoxy is closer to the stiffness of the wood, so you are getting more benefit from the wood structure. Glass/epoxy, kevlar/epoxy and carbon/epoxy are all so stiff that the original wood structure underneath is carrying very little load. The wood's primary contribution in this situation is to prevent buckling failure of the thin composite skin applied to it. In general, anytime you start mixing very stiff things with relatively flexible things you have to be very careful that the stiff things don't try to carry the entire load by themselves, wasting any contributions of the less stiff components and possibly creating nasty stress concentrations in the process. As some of the makers of rigid airships in the 1920's in Britain found out the hard way, it's entirely possible to make a structure stronger in one place and cause the overall structure to be weaker as a result.

The bottom line is that composites have tremendous capabilities to provide strength in ways not possible with conventional materials, and when you have increased capabilities you also have more opportunities to get yourself in trouble. If you don't do your homework, you can end up with a structure that at best doesn't use the materials to their fullest advantage, and at worst can be weaker than what you started with.

Don Stackhouse @ DJ Aerotech