Dimensional creep from residual stresses 1

[dwillms]

Hi All,

Has anyone any experience with dimensional creep (distortion) as a result of residual stresses over the *long term*? Objects do distort in the short term as a result of forming or machining operations from residual stresses, but are these changes long-term stable, or will the material continue to creep (however minute) as a result of these internal stresses. Is there a name for this phenomenon?

Another way of phrasing my question is: I understand how materials creep under external applied stresses, but will it creep as a result of its internal stresses when not under any (obvious) applied loading?

 

[Lcubed]

It depends heavily on the material, the temperature, the part geometry, and the type and intensity of the residual stress, but this kind of relaxation can continue until the residual stress is near zero.

Sometimes the part reaches equilibrium and ceases to change, but is then machined or processed in some way which causes a relatively large and/or fast dimensional change. Aluminum extrusions are notorious for this kind of behavior.

Hope this helps.

 

[dwillms]

Thanks Lcubed,

It just so happens that I'm working on an aluminum extrusion and we're deliberating a stress relief anneal following some post-extrude machining operations (removing 4.5% of the part mass).

Dimensional changes to the part can be accomodated following machining and before assembly, however once the part is in the field we can't tolerate any movement (even submicron).

Might an anneal be necessary?

 

[Lcubed]

Hi, Dwillms,
The machining operations and stock removal are probably relieving residual stresses and causing dimensional changes due to relaxation, rather than inducing new residual stresses which might be troublesome at a later time. It is likely that you should anneal prior to machining, to eliminate the locked-in stresses present in the extrusion. Then, depending on the type of machining you are contemplating, you may be able to operate on the part without inducing any new residual stresses, or you may need to re-anneal after machining.

This is qualitative comment only; the details of the annealing process are outside of my expertise. I'd look to a metallurgist, a heat-treatment expert, or someone similar, for more detailed advice. I've gotten good advice from Alcoa in the past; you might check with them.

Sub-micron sounds like a tough requirement to meet. Good luck.

 

[IRstuff]

How are you going to handle the thermal expansion of the aluminum?

Perhaps you should be using something like invar?

TTFN

 

[dwillms]

Hi IRstuff,

Well, thermal expansion is nice and linear (for uniform temperature) and can be predicted out of the model. Similarly, the effects of gravity can be calculated out. Its the unpredictable, non-linear stuff that we're worried about.