Hello,
I am currently testing SwiftComp in conjunction with gmsh. Suddenly I come across inconsistencies. So far, this "error" has not appeared.
If I let my model calculate locally, the solution file of the dehomogenization will be wrong, or at least built differently than if I run it here in the cdmHUB. The displacements are output normally and can also be displayed in Gmsh, but strains and stresses are not.
I created the model on my local computer and uploaded it. The input is exactly the same. Unfortunately I could not find an error after studying the manual several times. Besides, it worked at first. To exclude further errors I repeated the example from the manual with a CommonSG, the same phenomenon occurs.
I have attached the two result files, as well as my input.
Hi Hendrik,
To run SwiftComp with Gmsh in local computer, you need to use SwiftCompG.exe instead of SwiftComp.exe. Otherwise, you will get error dehomogenization results for stress and strain. I would suggest you to use online version on cdmHUB.
________________________________
---- Emailed forum response from liu1512@purdue.edu
Hey Xin,
thank you very much for your lighning-fast response.
Isn't there a way to get SwiftCompG.exe? Of course I can use the online version, but it is much more convenient and less time consuming to have all data on my local maschine.
Hi,
You need to ask Dr. Yu for SwiftCompG.exe. However, I still recommend the online version, it should be faster than the local version and more robust. You can use file transferring tool like FileZilla to upload and download files between local machine and cdmHUB. It is as convenient as your local machine once you familiar with the tools.
________________________________
---- Emailed forum response from liu1512@purdue.edu
If you want to run SwiftComp locally and if you have access to Ansys, Abaqus, or Nastran, it is better to use those GUIs instead of the rather limited gmsh GUI.
---- Emailed forum response from wenbinyu@purdue.edu
Hendrik Hagedorn @ on — Edited @ on
Hello,
I am currently testing SwiftComp in conjunction with gmsh. Suddenly I come across inconsistencies. So far, this "error" has not appeared.
If I let my model calculate locally, the solution file of the dehomogenization will be wrong, or at least built differently than if I run it here in the cdmHUB. The displacements are output normally and can also be displayed in Gmsh, but strains and stresses are not.
I created the model on my local computer and uploaded it. The input is exactly the same. Unfortunately I could not find an error after studying the manual several times. Besides, it worked at first. To exclude further errors I repeated the example from the manual with a CommonSG, the same phenomenon occurs.
I have attached the two result files, as well as my input.
I am grateful for any suggestion.
Structure
996 B Click to download
Hendrik Hagedorn @ on
local solution
local solution
131 KBClick to download
Hendrik Hagedorn @ on
I am grateful for any suggestion.
"HUB" solution
148 KBClick to download
Xin Liu @ on — Edited @ on
Hendrik Hagedorn @ on
Hey Xin,
thank you very much for your lighning-fast response.
Isn't there a way to get SwiftCompG.exe? Of course I can use the online version, but it is much more convenient and less time consuming to have all data on my local maschine.
Xin Liu @ on — Edited @ on
Wenbin Yu @ on — Edited @ on