Failure of Aluminum Tube Fired from a Gun

[swmann]

An Aluminum 7075-T6 tube is fired from a gun with a mass on the front end. Total weight is 10.132 lbs. Yield stress (compressive) taken from MIL-HDBK-5H is 68 kpsi. The peak acceleration is about 18,000 g's.

Here are the specs.
Tube
d_inner= 2.225 in.
d_outer= 2.736 in.
length = 16.55 in

Gun d_inner = 2.913 in
Gun material is AISI 4340

Taking pinned/pinned end conditions the slenderness ratio is 18.8. This puts it as an intermediate column for buckling (or short column depending on the text). For intermediate column buckling I used the Johnson formula which gives the critical stress as 64 kpsi.
The problem is that the stress generated at peak acceleration is about 92 kpsi and the tube did not fail.
The applied stress is even much more than the yield and ultimate stress for the material.
I checked and rechecked all the data. The radial strain at max acceleration provides 0.005 clearance all around. My only guess is that the tube is unable to deform enough laterally to reach the strain point that corresponds to the yield stress point on the stress-strain curve.
If this is the case how much stress can the tube take before something gives and what would be the most likely mode of failure.

Thank You for any information

 

[MikeHalloran]

The tube is fully supported by the gun barrel during launch. It's not going to buckle in the usual sense.

The first permananet deformation I'd expect to see is a decrease in axial length, and an increase in diameter. It wouldn't necessarily be a _gross_ deformation. If you can recover a launched tube without further damage, you might detect such a deformation in a tube that "didn't fail". I don't know what constitutes a failure in your application.

At some higher acceleration or thinner wall, I'd expect to see the walls fold, accordion- style, but the transition level would be very sensitive to manufacturing variations.



Mike Halloran
NOT speaking for
DeAngelo Marine Exhaust Inc.
Ft. Lauderdale, FL, USA

 

[hacksaw]

are you calculating the dynamic stresses? suspect that the critical loads for column stability are rate dependent

 

[Cockroach]

Interestingly enough, this problem has been recently solved with other threads. Take a look at wall stresses induced by internal loading (i.e. pressurization) under influence of axial forces. Your gun barrel acts very, very similar to this situation.

With some experience in military applications, obtain the peak pressure as a result of solid propellant being converted to gas. Next calculate the force required to acclerate the mass to your required number of g's. This is the force experienced by the pressure vessel wall.

Now apply the Von Mises-Hencky Equation for axial loading, this is the equation covered in the other thread. (Piping Applications) I will work out some numbers based on your recent input and post the results later.

Kenneth J Hueston, PEng
Principal
Sturni-Hueston Engineering Inc
Edmonton, Alberta Canada