Testing structural foam sandwiches

The shell of the center section is made of a honey comb sandwich. Very stiff and still light, however, very work intensive: The core has to be cemented to the face sheets and the inner face sheet has to be installed in a geled state. This is not a problem, when the shell is only a couple of square centimeters large, but gets really involved when it’s several square meters. That’s why we want to make the next shells with structural foam cores. At least one step less is needed to produce a good sandwich.

Large unsupported shells tend to buckle when a certain critical shear stress is exceeded. Buckling does not mean automatically failure, but can be avoided by properly choosing the flexural stiffness of the shell. Usually the stiffness is estimated from old measurements conducted by Idaflieg in the 1980’s. Those values are fine, but they were not measured in sandwiches and they certainly do not consider how well or bad we realize the sandwich. This is why I prefer making own test specimens and measuring the mechanical properties myself.

The flexural stiffness D depends strongly on thickness of the shell (to the power of three). The cores differ slightly in thickness, so that comparing directly D is not meaningfull. This is why it is better to consider the effective elastic modulus of the face sheets. Having this modulus, the thickness can be chosen so that the necessary flexural stiffness is achieved.

The diversity of materials is high: PVC, PMI, carbon, glass, aramid, etc. There are also many epoxy resins. We decided, thus, to test several mixtures. Once we started, we took it very seriously and have tested four different core materials, over five different face sheets and three core shapes:


The measurement setup follows Herbert Funke’s approach: The shell is used to seal a vacuum chamber and the pressure difference creates a uniform load. The maximum displacement allows to estimate the flexural stiffness of the plate and the shear modulus of the core. Weight and fiber volume fractions are determined by weighting the raw materials.

In theory the displacement should depend linearly on the pressure difference. Our measurments show that up to a neglible non-linear deviation this is true:

Stiffness is mostly determined by the slopes of the above curves. The flatter the better. The test specimens differ in weight and in achieved effective modulus. We are still evaluating different core materials. So far, we’ve made test specimens for the following foams:

  • AIREX® C70.55


PVC foams (AIREX®) can uptake quite some energy and have consequently a high strength, while PMI foams (ROHACELL®) tend to be stiffer but somewhat more brittle. This is in particular true for the 31 IG-F. Evonik has invested research into creating a strong PMI foam, which is equally good as aramid honey comb. HERO is the result of this research and it aims mainly to fullfil the requirements of aerospace industry. RIMA is a small cell sized foam with very low resin uptake designed specifically for vacuum resin infusion. The specs show a very reasonable strength and being an industrial non-approved material, the price should be much lower than for HERO. Thanks to Evonik, we got some samples. They are ready to be measured.

AIREX® is compared to ROHACELL® cheaper.  So why not taking AIREX? ROHACELL is lighter with equal mechanical properties and the quality of the foaming is much better. All sheets of C70.55 we’ve purchased have imperfections, which can be of substantial size:


We have not fnished our evaluation and have not decided which structural foam to use. Both AIREX and ROHACELL are still in…

3 thoughts on “Testing structural foam sandwiches

  1. Pingback: Cabin’s shell | hortenmicrolight

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s