Where can we find more information about the various composite materials such as E-glass and S-glass. I am mostly interested in material properties and the forms in which it is commonly available.

From : Don Stackhouse

Often the suppliers have data sheets, and there are some books on the subject. There are some articles in the "Ask Joe and Don" section of our website that go into the typical material properties and how to apply them in design.

However, published data are of limited usefulness when working with composites.

When you buy a bar of aluminum alloy, you are getting a complete material that you then just have to shape into the object you wish to make from it. The company that made the aluminum bar goes to extraordinary lengths to make sure that it conforms as closely as possible to some specification. If you buy a slab of wood, the property variations will be wider than we typically see in metals, but once again it is a "complete" piece of material, and the general behavior of the wood will be within a reasonably predictable range.

However, when you're working with composites, you are not just making the part, but also you are in effect making the raw material the part is made from. Yes, you can get some general property data from various sources like the ones I mentioned above, but that is not for the particular layup schedule, the exact materials and weaves, the same resins, core materials and other supporting structure, geometric factors, and especially the manufacturing processes your part uses. Your material properties will quite likely differ from that of the materials in the published data, and the differences could be very significant.

The published data are fine for a starting point, but ideally you should develop your own set of allowable stresses. In all probability this means making some test parts, then carefully loading them till they fail and then comparing the way they failed and the local stresses at failure with what you initially predicted.

This is especially important with compressive loads. There are two main failure concerns under compressive loads, and both are important. There is pure crushing of the material, and then there is buckling failure.

Contrary to popular conceptions, the compressive strength of most common high-performance composites is really very good, in most cases a fairly high percentage of the material's tensile strength. I've seen typical numbers in the neighborhood of 80% or more.

Kevlar is the exception to this. Its tensile strength is higher than carbon fiber (which is one reason it's so fantastic for propeller blades), but its compressive strength is less than half that of garden-variety el-cheapo E-glass (which is why it really isn't the best choice for most other structures). I've heard that this is because the Kevlar molecule actually has a kink in it. The Kevlar molecules buckle, forcing the matrix material (typically epoxy) to carry all the load, which then causes the epoxy to crumble under compressive overload. Oddly enough, the Kevlar fibers normally do not break when this happens, and when the compressively failed structure is stretched back out, it usually still has most of its original tensile strength!

However, most compressive failures occur in buckling. Composites are very strong for a given cross-section in comparison to the wood that most of our airplanes used to be made of. It only takes a paper-thin layer of carbon as a spar cap to carry the loads that used to require a large, thick strip sliced from a chunk of a tree.

Unfortunately, the buckling strength of something is extremely sensitive to its thickness. That paper-thin layer of carbon in the upper spar cap that replaces the big, thick stick of spruce is far more vulnerable to buckling failure because of that low thickness. If it isn't extremely well supported by a shear web or a core or some other means, it will fail in buckling long before it is anywhere near its failure point due to pure compressive overload. Such a failure is really an indictment of a poor design; it did not use the basic properties of the material to their best advantage. I believe that much of the mistaken belief that composites have poor compressive strength is due to premature failures from buckling.

There are lots of factors involved in the calculation of the critical load for buckling. However, they can be quantified, and the results used to predict and optimize the strength of other structures of yours that have similar materials and construction.

We did some fairly extensive lab tests of our own structures and materials back when we developed some of our earliest sailplane designs, particularly the original 2-meter Monarch. We were able to determine allowable stresses of our own materials and processes from that data, and use it for the design of subsequent airplanes. The results have been very consistent, and today I can design a spar for a certain bending strength and normally get extremely close to the desired strength on the first try. Good things come to those who do their homework.

Don Stackhouse
DJ Aerotech