We've had some great feedback to the Ask an Engineer videos, and while we assure you we ARE getting a microphone and will find a quieter space to film than the factory, Vladim Shapiro took the opportunity to scientifically test our coke can strength test.
The models we used in the Strength and Structure video are available for free download so he tried structural analysis on them using free simulation software. Scan&Solve is a Rhino plugin that can do structural analysis on any b-rep or meshed solid model directly without any preprocessing.
Here's his video, remember to watch with CAPTIONS ON to follow along.
Pretty neat right? You can try it yourself for free with an evaluation version by going to Scan&Solve. Thanks Vadim!
If you have any 3D printing questions you would like answered by our 3D printing engineer Matthew Hagan please email email@example.com
I think that you are correct. If I actually drank soda, I would know that it is 12oz and not 8oz
We should have checked the mass of the can more carefully, e.g see this post http://www.elmhurst.edu/~chm/vchembook/121Adensitycoke.html
So thanks for catching this!
This will affect the numerical values in all the tests, but not the outcomes or the conclusions in the video. The stresses and displacements will increase in proportion to the applied load. The high stresses will be in the same locations, and the shape of the deformations will remain the same. This is what "linear elasticity" means.
A reference like http://en.wikipedia.org/wiki/Beverage_can and no need to drink anything.
The video says the material approaches the limits, but still stays below them. That was the give away, specially after first theory was discarded as it didn't match at all. Second one matches in place but still not in outcome. With the correct value it would mark some zones red and, if the software supports it, simulate the breaking. But imagine if the mass was correct, the video says it should not break.
Indeed. The correct mass is roughly 1.5 times more, so the computed stresses and displacements would be 1.5 more. Thus the danger level of 0.86 becomes 1.29, and you would see the the red at all points where the values exceeds 1. I posted an image of the solution with 3.375N applied at the link below. (It is still not exactly what the weight should be, but is much closer.)