This is a presentation I gave at the LS-DYNA users group conference in Koblenz, Germany today. I’d welcome any questions.
There was a notable increase of interest in IGA over last year at this conference. There were nearly twice as many talks (5 this year, 3 last year), and there were about 25 people in the room (about double last year). DynamorePres_presented.pdf (4.1 MB)
Very cool, @matt!
Two questions:
Regarding the bracket on page 37, I can’t quite make-out the colorbar, does that say “Displacement”? i.e. were you able to run a simulation?
Do you have a tutorial / video example of bringing a T-Spline model from Autodesk into Coreform? e.g. What file type would one export from Speedform / F360? How would one import? T-Junctions ok? Is the continuity translated exactly? etc
Hey Greg
Yes this was a displacement. We fixed the bottom holes and pulled up on the top hole. This is just a shell model so I don’t know that it is a particularly realistic simulation but it did run!
Good idea about a tsplines video tutorial. We have a high level process video I can share in the meantime. Or Chris can if he beats me to it. But we just export via iges. Tjunctions are fine, we have a proof of concept code that sews the iges nurbs patches back together, detects creases and continuity, and builds the uspline. This needs more testing and time to become more robust, we are also working with Autodesk to make this more seamless. I don’t believe we have this exposed in the GUI yet. Is this workflow of interest to you?
Yes this was a displacement. We fixed the bottom holes and pulled up on the top hole. This is just a shell model so I don’t know that it is a particularly realistic simulation but it did run!
Nice! Mostly couldn’t tell if that was a “reparameterization accuracy” colormap or a colormap associated to a simulation. I figured if it was a simulation, it was probably shells.
Good idea about a tsplines video tutorial. We have a high level process video I can share in the meantime. Or Chris can if he beats me to it. But we just export via iges. Tjunctions are fine, we have a proof of concept code that sews the iges nurbs patches back together, detects creases and continuity, and builds the uspline. This needs more testing and time to become more robust, we are also working with Autodesk to make this more seamless. I don’t believe we have this exposed in the GUI yet. Is this workflow of interest to you?
I’m not sure the workflow is of particular interest to me, a Cubit/Trelis workflow is much more interesting to me. But Cubit doesn’t mesh T-Junctions yet, so in the meantime I might be interested in exploring T-Junctions ad-hoc via F360.
