It is a known phenomenon and why hoods, trunks and roofs on cars are not welded to their reinforcement skeletons, but attached with dabs of soft flexible adhesive.

Only the edges are physically attached so that the panel may evenly distort with thermal expansion.

 

Truth.

 

Blue, you are making assumptions with something that is welded. I am unaware of how you (or whoever) welded yours, however, while it is was you that observed it is not some irrefutable fact. Thermal expansion is not magical and HAZ is only rarely similar in mechanical properties to the rest of the piece. That welding can and most often does introduce stresses into the piece(s) being joined is most likely why you had noticeable deformation and by altering the reinforcement (lowering it) such that it had a larger area to flex over is most likely why it became less noticeable.

As to your later comment, you would be incorrect. The adhesive is a replacement for welding and is a cost saving tool during manufacturing. If you look more closely at your hood you will also notice adhesive pads that connect the ribbing to the skin. It is neither soft nor flexible in its cured state although I do not know if different chemistries are used in to get those effects. I do know that in its typical (and new) state that when properly used the metal will tear before a joint of that adhesive separates. Other vehicles use spot welds or even more of the adhesive. If they did not there would be deformation, to include bulging, depending on the differences in pressure/temperature/etc. Furthermore, the structure would be both less secure and strong were both pieces not connected at least intermittently at more than just the edges.

So I reiterate that the care of repair is the major drive of these witnessed oddities.

 

Thermal expansion is why welding, to certain degree, ***** with everything. The metal goes in molten and at like 3000 degrees, then it cools, freezes off and continues cool. So, guess what, it shrinks. After it cools and shrinks to room temp there are now induced stresses in the metal, due to this shrinkage. Welding is an awesome tool but this phenomena can't be ignored.

This is why GP's roof is "pulling".

 

And then slather some bondo over that, add in a little flex, and it pops off overtime.

 

how did this end up looking when done?

 

I haven't taken any pictures recently, but if you read the post where I posted the picture, I welded too much at one time and it got warped. Even if I welded it more patiently, the amount of welding needed to do the job would pull so much it would warp it anyways, probably just less than mine warped. It would require something pushing the roof up or some sort of support that you could clamp it down in that desired shape to fix it