I ran a model last night that has a huge 115GB results file (explicit plastic model with fine mesh). It was trying to open when I left this morning but am now wondering if it is even possible with my 64GB of RAM?
Does PrePoMax load the whole results file into memory? I clearly need to refine my approach but I just would like to know if it can work.
thanks
Luke
You can also try opening it with GraphiX (CalculiX native pre- and postprocessor). But consider reducing the size of the output file by removing some output requests or, even better, reducing the frequency of output. CalculiX lacks more options to control output frequency in procedures like explicit dynamics but you can at least avoid saving results every increment since those analyses tend to consist of a large number of increments and thus the files can get huge even with small models.
I didn’t expect the model to actually run, it was the first try with an explicit step and I had no idea how long it would run. I haven’t run an explicit model in ages so I need to review the details about the setup.
thanks
Luke
After my recent tests with explicit dynamics in CalculiX, I can tell you that this procedure still requires lots of improvements. Several features seem to be unsupported or just cause errors (even plasticity in my case) and the analyses take quite a lot of time even for small models. You should adjust the incrementation carefully (taking into account mass scaling which can be enabled by specifying the desired minimum time increment) and significantly reduce the frequency of output. But the elimination of any convergence issues can be worth it.
Hi lukerickert,
Explicit dynamics + plastic model ?? 
. Is that possible?
Would you share your input deck with just a couple of elements.?
Regards
I haven’t noticed any limitation like this in the User’s Manual but I had to remove plasticity in my test models due to the “No result” error. Of course, in Abaqus, it works with plasticity without any issues.
it is a simple model looking at stress concentrations (no contact etc), it doesn’t really need to be run explicit but I want to see what it would do. It ran for 10 hours and produced results but I don’t know what yet as the file is so massive 
Luke
In my tests it has never worked with plastic material.
If you manage to open the file and you get plastic strain, please report. It would be an interesting progress.
I asked Guido about this. He said that Johnson-Cook plasticity is in development for the next release but regular plasticity should also work in dynamic explicit analyses. It works in my single-element test model but doesn’t in a more complex model involving projectile impact. I sent him this model and he said that he would take a look at it.
Thank you FEAnalyst. The next release looks promising
CalculiX 2.21 is available now and the bug with plasticity in my explicit dynamics example that I shared with Guido was fixed. If you encounter other potential issues with plasticity in those simulations, let me know and I will notify Guido.
Thanks Jakub and congratulations to you too for the new version.
I will run again some old files to see if they go well.
Reading through the forum it looks like there are some major improvements and fixes. It is the perfect icing on the cake after a year of work and contributions from the whole community.
Congratulate Don Guido for me if you have direct contact with him.
I have a similar problem opening a frd file (for post-processing) of about 2,6 Gb - NL static
After 30 min the wheel rotates…I stopped it
With another software just 2 minutes
I can read .frd result files up to 6 GB. It takes 2-3 minutes. But, I did not test PrePoMax on larger files. If possible share the file and I can give it a try.
I tried again and it runs…less than 1 min - Ryzen9 - 64 Gb RAM
Maybe yesterday something went wrong -
PS I take the opportunity to say that could be useful to change the unit in post-process mode or, at lest, to apply a factor to the results.
For example the frd file has unit lenght meter and dividing x 1000 I can obtain mm