Hi everybody,
I am new to Texgen so please excuse my problem, even if it is trivial.
I have created a textile model and exported it as volume mesh. Furthermore I have created periodic boundary conditions by *equation. When this model is analyzed with a linear elastic material, I’m receiving following error:
***WARNING: SOLUTION DEPENDENT STATE VARIABLES CANNOT BE APPLIED TO ELEMENT 1
BECAUSE ITS MATERIAL DEFINITION DOES NOT INCLUDE *DEPVAR
***ERROR: THE FIRST DISTRIBUTION ORIENTATION DIRECTION FOR ELEMENT 4679 IS ZERO
I don’t understand the warning, as I haven’t defined any user material so far. I thought I don’t need any state variables?
The actual error itself doesn’t make sense either, as I have included the *.ori and *.eld files:
*Distribution Table, Name=TexGenOrientationVectors
COORD3D,COORD3D
*Distribution, Location=Element, Table=TexGenOrientationVectors, Name=TexGenOrientationVectors, Input=5hs050.ori
*Orientation, Name=TexGenOrientations, Definition=coordinates
TexGenOrientationVectors
1, 0
*Initial Conditions, Type=Solution, Input=5hs050.eld
Has anybody a clue, how this error message is caused by?
Best Regards,
André
problem with orientation
Moderators: Martin, Developers
Re: problem with orientation
Hi Andre,
I can't see why you should have a problem with what you describe. Was the textile complex? The volume meshing doesn't always give the best results - particularly for textiles including vertical, or near vertical, yarns. I would suggest that you try the voxel mesh output and see if you have more success with that.
Otherwise take a look in the exported files and see if there is anything obviously wrong with the elements mentioned.
Hope that helps and that you find TexGen useful. If you still have problems an example would be useful.
It would be good to know how you are using TexGen and where you are based.
Louise
I can't see why you should have a problem with what you describe. Was the textile complex? The volume meshing doesn't always give the best results - particularly for textiles including vertical, or near vertical, yarns. I would suggest that you try the voxel mesh output and see if you have more success with that.
Otherwise take a look in the exported files and see if there is anything obviously wrong with the elements mentioned.
Hope that helps and that you find TexGen useful. If you still have problems an example would be useful.
It would be good to know how you are using TexGen and where you are based.
Louise
Re: problem with orientation
Hi Louise,
Thanks for your reply.
I've tried the voxel mesh for my 5-harness satin, but it ends up to be very coarse in terms of geometrical accuracy although it generates many elements.
With the volume mesh I figured out why there is the error: "THE FIRST DISTRIBUTION ORIENTATION DIRECTION FOR ELEMENT 4679 IS ZERO"
The entry for this element in the *.ori file is indeed zero: "4679, 0, 0, 0, 0, 0, 0"
When I built a new model of my 5-harness satin, this error comes up again with different elements.
Could it be, that there is some problem with the generation of the orientation file for very distorted elements?
Best Regards from Hanover
André
Thanks for your reply.
I've tried the voxel mesh for my 5-harness satin, but it ends up to be very coarse in terms of geometrical accuracy although it generates many elements.
With the volume mesh I figured out why there is the error: "THE FIRST DISTRIBUTION ORIENTATION DIRECTION FOR ELEMENT 4679 IS ZERO"
The entry for this element in the *.ori file is indeed zero: "4679, 0, 0, 0, 0, 0, 0"
When I built a new model of my 5-harness satin, this error comes up again with different elements.
Could it be, that there is some problem with the generation of the orientation file for very distorted elements?
Best Regards from Hanover
André
Re: problem with orientation
Hi Andre,
You do have to choose the voxel dimensions carefully to reflect the geometry of your textile. We've found the results to be good using this method.
As far as the volume mesh goes it's possible that there are problems when there are distorted elements. The only thing I can suggest is that you send me your tg3 file and details of the inputs you were using in the volume mesh dialog and I'll check what's going on in my debug version. There are certainly some places in the volume mesh algorithm which throw an assertion when running in debug but which just continue in the release version. My guess is that, in these cases, it may not then continue to calculate the orientation ( I didn't write this code so I'd have to check that).
Best regards,
Louise
You do have to choose the voxel dimensions carefully to reflect the geometry of your textile. We've found the results to be good using this method.
As far as the volume mesh goes it's possible that there are problems when there are distorted elements. The only thing I can suggest is that you send me your tg3 file and details of the inputs you were using in the volume mesh dialog and I'll check what's going on in my debug version. There are certainly some places in the volume mesh algorithm which throw an assertion when running in debug but which just continue in the release version. My guess is that, in these cases, it may not then continue to calculate the orientation ( I didn't write this code so I'd have to check that).
Best regards,
Louise