Hi,
I want to make a simulation of a heap of particles with a lifting cylinder. How can I make my cylinder fixed/motionless during 1 second, then lifting, and then fixed/motionless again, please?
My MFiX version is 23.3.2.
Thank you in advance
granuheap_2023-11-30T165347.942220.zip (8.4 MB)
Hi @Marie-Anne, welcome to the MFiX forum
Have you studied the mixer_3d or conveyor3d tutorials? It uses a “keyframe” data file data_kf_0001.txt to specify the speed of the mixer/conveyor, which can change with time. Per the comment in usr_mod.f
! Keyframes are data tables provided by user to perform user-!
! defined operations such as ramping or other time-dependent !
! operations
Alternately, instead of implementing the code to read from the keyframe file, you could hard-code the transition times directly in usr_mod.f. This would be simpler but would require recompiling the solver whenever you want to change the transition times or speeds.
Please feel free to follow up here if you have additional questions.
– Charles
Please see the attached example, where the cylinder moves up and down based on the translation defined in the keyframe file. This is based off mixer_3d as Charles suggested.
granuheap_2023-12-01T093416.720918.zip (15.6 MB)
1 Like
Shaken, not stirred.
Thank you very much for you reply, sorry for the late answer. I tried to run the zip file you attached but I got this error and I cannot find the (null) file…
Any help please
(null) is not a file, it indicates that the debugger was not able to identify the source of the error. Very odd. Did you make any changes to the code or did you run it unmodified? I presume you are running on Windows?
Thank you for your answer
I didn’t make any changes, I only built the solver selecting “shared memory parallel” and “build solver in parallel”.
Yes I’m running on Windows
Please try to build the serial solver (uncheck “shared memory parallel”) to see if this is an issue with SMP.
I unchecked SMP and now it works, thank you!
Something odd here. This case runs on Linux in either serial or SMP mode (4 threads).
On Windows, I confirm that the serial run succeeds, but SMP crashes.
With 1 SMP thread, I see the same message Marie-Anne reported - stack overflow in (null). But with more than 1 thread, the SMP solver exits right after printing Pre-Processing geometry for DES., as if SMS mode were active (it is not) - there is no other error or warning message. SMP on Windows has not been tested as thoroughly as SMP on Linux, but it should work - I have no idea what’s going wrong here, I’ll follow up if I figure it out!