Hertzian stress or?

[DYV1973]

LS,

To clamp a measuring device onto a round steel structure, the choice was made to used a kind of "knife". Since the contact area is very small, it is possible to exceed the yield stress of the metal (knife and structure are made from identical materials) and obtain a very stable contactpoint.
Initially we used the formulas for Hertzian stress to calculate the stresses. But the radius of the "knife" blade is difficult to determine and has a large impact on the result. Also plastic deformation will occur, and the contact surface will not be a linecontact anymore.

Does anyone have a suggestion how to solve this?

Thanks,
DYV

 

[rb1957]

if you want to attach a flat thing to a round thing, and not have either of them to deform (adapting to the other), then you need an adaptor.

this could be a solid piece, like a machined piece profiled to suit both parts, or it could be a formable piece, like epoxy shim.

 

[DYV1973]

To have the best possible and stable mounting, we are aiming for plastic deformation. Which means that local stresses should be high enough to exceed the yield stress. The problem is how to calculate, as we have a short line contact in axial direction.

 

[sdra2]

Do you have Roark? My 6th Ed has a good discussion on Hertzian stresses, their limitations (infinite lengths, types of stress obtained, etc), safe working stresses, knife edge contact, etc.

Do you have access to non-linear FEA code? Might this be required?

Can't quite picture what it is you are trying to achieve, but "a radius that is difficult to determine" that has "a large impact on the result" would seem to be an intractable problem. Is it simply the case that it is difficult for you to determine, and that you could get accurate measurements done somewhere else?