Point load on curved glass plane

[Jazzy47]

So I had an interesting stained glass design created for a personal project. The end result was a curved piece instead of a flat plane like you'd hang on your window. The stained glass artists cannot cut pieces out of curved glass. Instead they have to cut the colors out of flat glass, heat it until in a kiln it's plastic (but not melted), and let it fall onto a mold of the desired radius. I know intuitively it is more difficult to cut curved glass...if you score it and try to tap it out like a flat piece of glass, it will just break along a stress fracture instead of the nice shape you cut. What is the structural reason for this? How are stresses distributed differently on a curved surface than a flat one for a brittle material? Obviously a flat piece of glass you can lay nicely on a table whereas a curved piece would be unsupported in the center, but I feel there are reasons beyond this. We didn't really analyze stresses of curved objects much in my undergrad.

Thanks!

 

[ColinPearson]

was the piece stress relieved after it was laid on/in the mold? Is that even a thing with glass work like this? Maybe it's something to do with residual stresses from the cooling process? I definitely do not know, just a thought.

 

[SwinnyGG]

Is that even a thing with glass work like this?

Annealing is a critical component of glass manufacture, for things like windshields, safety glass windows, etc. Whether or not it's commonly performed on artisan projects is something I do not know.

Jazzy, the problem cutting curved glass pieces is related to the fracture mechanics of the material- in the manufacture of windshields, for example, the net outline is always cut before the glass is shaped.

Remember that glass is not just a brittle material- it is an amorphous (non-crystalline) solid. Crack growth is dictated by bonding planes in any material, and in a material such as glass they are completely random. You can score and break a piece of flat glass because the score confines crack growth to a single plane. If the base plane of the part is now curved, the score still confines crack propagation in only one plane.

 

[Jazzy47]

Colin they don't normally (flat projects) heat the glass. They just score the shape into the glass and then lightly tap it out with a small hammer. The heat is just to avoid cutting from a curved surface, and instead form the pre cut flat glass pieces around a mold to curve them to the desired radius. I don't think they worry about annealing or treatment processes, as the glass is aesthetic and not structural. Interesting point though. jgKRI interesting...so it is more about the random material properties of glass itself than the geometry...so would other more structured brittle materials not have this issue when curved? Does internal moment come into play because of the eccentricity from the curve? What should I look at to study stresses on curved surfaces? Is it an FEA problem, or can one apply classic analysis? Sorry for all the questions haha.

 

[ColinPearson]

Interesting stuff!