Less modulus....a "stretchy" fabric.
Here's the problem. Prop rotates clockwise (viewed from the rear) so in reaction the engine rotates CCW on the mounts, around the crankshaft axis. Viewed from the left side, cylinders 2&4 move downward in relation to the cowl inlet.
Despite being encapsulated in a very flexible matrix (urethane rubber), the fiberglass fabric continues to do what you would expect; it is stiff in tension and compression. It does not allow the "tube" to deform into the desired smooth S-curve, and the result is more like thinwall tube buckling; wrinkled duct walls. Right now the objection is mostly cosmetic (it doesn't seem bother the low velocity flow of the large inlets), but this is an experiment. How can it be improved?
The primary force we wish to resist is ballooning due to dynamic pressure, 136 lbs per sq ft at 200 knots and sea level. One possibility would be to change the fiber orientation; all the fibers wrapped around the tube rather than 50/50. Instead of the current bidirectional fabric I could try a uni fabric or perhaps a wrap with 1" glass tape:
Another approach would be to simply thicken the rubber matrix cross section. The compressive properties of a fiber/matrix composite are heavily influenced by the matrix. This may be the reason standard fabric reinforced rubber automotive hoses resist buckling until bent severely.
A third possibility would be to mimic a standard hump hose:
Any other ideas?