When does Glass Transition (Tg) affect performance?

[stick1]

I have an optical assembly that is bonded with a UV curing adhesive. Testing is showing some unacceptable position shifts of the optic components after exposure to cold & hot temperature cycling. Closer inspection of the adhesive specifications shows that the Tg of the adhesive is inside of the operating temperature of the optics assembly. Even though the manufacturer claims stable performance on either side of the Tg, the shift appears to be creep or stress relaxation induced. Should allowing the glass transition temperature of the adhesive to be within the operating temperature limits ever be permitted (in the case of sensitive optical joints)?

 

[IRstuff]

I think the answer is clearly not, although, I would tend to think that there might be additional issues in your particular example. Given that the "creep" has to be produced by some sort of asymmetric stress, one needs to identify where it's comnig from and why. Additionally, its surprising that the optics could be that intolerant of creep, particularly if the optics are assembled in a normal fashion, as there are usually larger fluctuations in optics positioning than could be accounted for by a creep in optics that are glued together TTFN