Novel Diaphragm App--overdesigned?

[nnovelli]

A basic question for you pressure guys: Will a diaphragm with convolutions undergo more movement before reaching yield stresses than one with a straight section?

My situation:
I'd like to use a diaphragm to mount a camera onto the side of a custom vacuum dewar to impart some adjustability: +/- 1mm of axial movement from the unflexed position. The ID is 3.4", and the OD should be between 5.5" and 7".

Roark's equations say that I'm getting close to yield stresses in Aluminum and Stainless if I use a diaphragm with no convolutions, in thicknesses of .012 - .09". I am hoping that by spec'ing some waves in it, I can get a better safety factor.

Possible?

 

[JStephen]

If I'm understanding your geometry right, I believe a diaphragm with convolutions would undergo more movement before yielding. It would also hold less pressure before yielding. If it's loaded by pressure, it's a step in the wrong direction. If movement is fixed, and you need flexibility to account for the movement, then you improve things.

 

[nnovelli]

Thanks, That's very helpful. Our movement is fixed--the excursion of the diaphragm will be controlled by our manual alignment of the camera.

I was worried that pressure would cause "bellying" between the effective ID and OD of the diaphragm, but I believe it won't add enough stress to cause yield. Is there a rule of thumb for how much additional movement convolutions add?