No contour plot showing for CCX results

Tim_Gowing

Hello Victor,
I am enjoying the new version 5, especially the paint selection, unbelievably productive, a game changer, many thanks for this amazing development.
I produced an analysis including CCX contact, with ease using the paint tool, it solved using jeff's bConverged v2.8 solver, although the result contours would not show upon completion of solving. I tried starting Mecway and opening only the CCX results and again the values show in the rainbow legend but no contours on the mesh. The results display ok using Jeff's Windows version of Calculix CGX.
I hope that it's just me missing the flick of a switch, otherwise I can send the files.
Kind regards, Tim

Sergio

Hi Tim, looks like you have done a nolineal FEA, in that cases you must move the time slider on the lower side of the Mecway's window to see the results. In CGX you are seeing the time increment 10, but in Mecway the time 0.

Regards!

By the way, very nice model, can I ask you what is this?

Tim_Gowing

Hi Sergio,
Thank you for the slider hint; embarrassingly I've asked that here before....

The model is half of the base of a fluidised bed in a test rig; air is passed through the particle bed and the particles can behave a bit like a fluid.

I've read on one of these threads that you're running ccx v2.10:
'Calculix ccx2.9 built with msys2/mingw64 vs. bConverged ccx2.8p2'.
Have you compiled ccx 2.10 64-bit within Windows 10 using the pointers from kwip?
https://github.com/ddfem/ccx_mingw64
http://hillyuan.blogspot.co.uk/2013/10/build-calculix261-in-mingw64.html

I've been patiently waiting for the 2.10 build from Jeff Baylor because his work has been very reliable over several years and versions. I would really like to be using 2.10 though, do you have a procedure for the 2.10 build?
Regards, Tim

Sergio

Tim, I just donwload the win64 executable from some place that kwip has uploaded, and it works very very well. If you work with contacts I would advice to switch. Also CGX have some valuable improvements (like moving between increments directly with pg-up/pgdown keys, that in your case is needed).

http://mecway.com/forum/discussion/163/calculix-ccx2-9-built-with-msys2-mingw64-vs-bconverged-ccx2-8p2#latest

In fact I have an instalation of the 2.8x from Jeff Baylor were I have put directly the both executables (ccx.exe and cgx.exe) from kwip, and work very very smoth with a notable diference in time to solve from 2.8.

Tim_Gowing

Sergio,
Thank you for your recommendation of the 2.10 executable of kwip. I just tried it and the analysis shown in the contour plots above ran in 11 minutes versus 31 minutes for 2.8. I will investigate further, I haven't looked at the results back to back yet, although I'm not expecting any issue.
Much appreciated.
Tim