Simulate Early, Simulate Often... In Rhino
Just wondering if there are any improvements, stabilization, or features to look forward to. I see hints of topographical optimization on the Intact Solutions web page. Is that still in the works?Is that a screen shot of Scan-and-Solve?
I'm not really interested in Cloud based solutions as I paid good money to have a solution working directly within Rhino.
Thanks and happy 2020.
Tags:
Hello Aaron,
We have many exciting developments happening behind the scenes. Leveraging DARPA-funded research, we are greatly increasing speed and adding new types of physics. This development involves a complete rewrite of the core simulation code (dubbed Intact.Simulation).
While we're working on a new version for Rhino, what capabilities do you use in Scan&Solve Pro? What if there was a Grasshopper plug-in, would that be useful?
Michael
P.S. The topology optimization you see on our webpages is from Desktop Metal's Live Parts. It uses our simulation API along with biology-inspired techniques to guide the optimization process. Here's a collection of videos that highlight its features and operation:
https://www.youtube.com/channel/UCLUGj_RYvkUor_JA-wPaIkg/search?que...
Thanks for the check-in Michael,
Live Parts looks really cool.
The work I do with Scan&Solve Pro is primarily with wood and orthotropic materials. I have been incorporating carbon fiber into the instruments we build and have found the Total deflection and Von Misses simulations to be the most helpful. I can not use the danger level option because it doesn’t apply to wood at this point in time.
There are many very expensive FEA programs out there and I am assuming serious engineers are using Solidworks or Catia etc. So, the audience Scan&Solve is catering to, I think, would be primarily the non-classically trained tinkerers and designers, like me. There are a lot of structural concepts and solutions that are intuitive and I frequently use the program to vet that. However, there are structural problems that are too complex to intuit and I would love for the program to allow me to better explore them. Since I am not a mechanical engineer, I imagine that I lack some useful basic engineering concepts that I would be able to apply to the Scan&Solve results. As it stands, I haven’t changed any of the parts I have designed based on what Scan&Solve has shown me. It would be nice if the program was more helpful in that regard. The ability to refine my geometry and remove excess weight from a part is one of the biggest goals for me. I don’t necessarily need the program to make me a new NURBs object, but even generating a texture to represent the part would be a good start. I have tried using the Von Misses stresses and Principal Tension and Compression views for this but it generates totally different results and doesn’t give a cohesive direction to explore. What does the computer do for turning a solid beam into a truss? Will the program naturally generate an I beam? Now, if you curve that same beam, how would the truss shape change, if at all?
Concerning the program features that exist now, for both wood and carbon fiber, being able to tell what the grain direction and orientation are is complicated. The coordinate arrows don’t really help because, I don’t know what the red, green, and blue are representing, they only demonstrate the direction at intervals, and I can’t see where I have grain runout on a part I am working on. Absolutely nothing I am designing has a simple straight structural line! A texture showing the long run of the grain lines over the part would be great.
Having a plywood material selection would be useful too. I can build up a plywood component now but having a selection for face-grain direction, face veneer thickness, and total number of plies would let the computer do it much faster.
I have never used Grasshopper. It seems interesting for dynamically generating components but it’s not really useful to me in the design space. There are a bunch of topographical optimization plugins for it but the appeal of Scan and Solve was its ability to work within Rhino. If it is necessary to use Grasshopper for generating or ‘growing’ topographically optimized structures then I guess I would have to learn but I would rather be able to stay within Rhino.
Thank you. I hope the time you have put into the program has been rewarding.
Aaron
Grasshopper plug-in would be fantastic! Bit I agree with William below, I'm not so interested in cloud-based simulation. Grasshopper plug-in Good!
Cheers!
Has there been any more progress since march on an update? Rewriting the core code sounds like it might be a multi-year endeavor, but I'm sure we'd love to hear about any progress or interim enhancements and especially stability improvements.
© 2024 Created by Michael Freytag. Powered by