Hi all,
Anyone else experiencing a strange behaviour of Reference Points moving arbitrarily away from the associated referenced surfaces ?
This seems to happen after the solver has completed it’s run. In version 2.4.3.
The strange part is that the surface is still referenced in the selection window. However, the X,Y,Z corodinates have changed
I am sorry for my late reply but could you share the file? What units are you using an how large is your model (the scale bar is not visible). Are the results in the same units?
Thank you for the model. I think it also shows the problem where the file gets corrupted when opened in a newer version. I hope this will enable me to find both bugs.
Looking at the model, I think the unit system got mixed up. The point positions are scaled by a factor of 1000. But I cannot reproduce the reference points flying away. Did it happen only once, or multiple times? Do you remember what you did exactly? I fixed the RPs’ positions, ran the model again (deactivated contacts), and when I open the results, the RPs keep their positions.
But I discovered that for larger models, the selection cache was not working efficiently enough, so I fixed it, and now opening RP properties is much faster.
It happened once, but I cant recall exactly when. I think it was after i ran the analysis.
I will keep an eye on it and see if it happens again.
Thank you @Matej
Please do. And if happens again try to reproduce it.
this just happened to my assembly. To reproduce? (still testing) create an assembly, then apply rigid constraints to several floating points. After running the points will move. The analysis this happened on was an uncoupled thermal analysis where some rotation is expected. one point is constrained in XY but allowed to rotate while the Z and XY rotations are constrained elsewhere. I don’t know if this will repeat. I can’t share the model but I can try to see if it happens on a different sharable model.
More info, the analysis failed:
increment 1 attempt 2
increment size= 2.500000e-01
sum of previous increments=0.000000e+00
actual step time=2.500000e-01
actual total time=2.500000e-01
iteration 1
Using up to 16 cpu(s) for the stress calculation.
Using up to 16 cpu(s) for the heat flux calculation.
*ERROR in u_calloc: error allocating memory
variable=au1, file=mafillsmmain.c, line=183, num=1319376000, size=8
Job failed - no results exist.
Process elapsed time: 560.318 s So I clicked results anyway, which then went to step 1. the points lost their location when I clicked on step 2. One more tidbit is that the points where created with “center of gravity”. The points did not move when I switch them to coordinates.
It appears that if I update the geometry …import step, swap geometry and re-mesh I get the same behavior with the reference points flying off into space: as an example X went to -66998.6072662893 mm when it was close to 0 to 50mm on the part. This time the create by/from region was set to coordinates. This is with PPM V2.5.1
