FEM Pipeline units on last dev versions

About the development of the FEM module/workbench.

Moderator: bernd

mac_the_bike
Posts: 42
Joined: Sun Jun 30, 2019 12:56 pm

Re: FEM Pipeline units on last dev versions

Post by mac_the_bike »

Code: Select all

OS: Ubuntu 22.04 LTS (GNOME-Flashback:GNOME/gnome-flashback-metacity)
Word size of FreeCAD: 64-bit
Version: 0.20.29177 (Git) AppImage
Build type: Release
Branch: (HEAD detached at 0.20)
Hash: 68e337670e227889217652ddac593c93b5e8dc94
Python 3.9.13, Qt 5.12.9, Coin 4.0.0, Vtk 9.1.0, OCC 7.5.3
Locale: English/United Kingdom (en_GB)
Select the eigenmode you want to display further in the tree and then create a new result pipeline from this (Menu Results -> Post pipeline fro result).
I have tried this. I have chosen x displacements from mode 6 and 7 and I get the SAME max/min values for the 2 pipelines. The contour plots look different.
Also, as I said previously, there is no way of telling which set of results are associated with any particular pipeline.
More importantly, it should be impossible to highlight 2 result pipelines at the same time.
This way you see WHERE the min/max is.
I now understand what you are saying.
I also don't see quantities under Calculate. This is how your example it looks in FreeCAD 0.20:
I get a panel "Hints user defined equations", see attachment, _1.
The pipeline allows you to visualize all available result scalars and vectors. What result do you miss?
Principal stresses are not available in the pipeline, see attachment, _2.
Your example shows that the units are incorrect in the pipeline results. I will have a look.
This is just for interest.
If you were looking the displacements of Mode 6, then this is a Rigid Body Mode, RBM.
If you imagine 6 "Basic RBMs", i.e. rigid body translations in the x, y and z directions and rigid body rotations about the x, y and z axes, then the calculated RBMs will be linear combinations of these "Basic RBMs". If you perform an analysis with a different mesh, then in all probability the RBMs will be different linear combinations of the "Basic RBMs".

I have attached a file which will extract the displacements from an ".frd" file and sort them. The number in the last column corresponds to the mode or stressing case numbers, i.e. 1 is the first, etc. The first occurrence of, say, 1 in the last column is the smallest displacement and the last occurrence of 1 is the largest for the first case. For convenience, rename the file to be: awk.1.sh.
invoke: ./awk.1.sh /tmp/fcfem_hegj7cei/FEMMeshNetgen.frd x
Attachments
awk.1.sh.py
(607 Bytes) Downloaded 26 times
pipeline5_1.jpg
pipeline5_1.jpg (76.53 KiB) Viewed 1486 times
pipeline5_2.jpg
pipeline5_2.jpg (67.31 KiB) Viewed 1486 times
User avatar
uwestoehr
Veteran
Posts: 4961
Joined: Sun Jan 27, 2019 3:21 am
Location: Germany
Contact:

Re: FEM Pipeline units on last dev versions

Post by uwestoehr »

mac_the_bike wrote: Thu Jul 14, 2022 11:27 am I have tried this. I have chosen x displacements from mode 6 and 7 and I get the SAME max/min values for the 2 pipelines. The contour plots look different.
Yes, we have 2 known bugs here:

- the pipeline does not tell you later on which set of results it depends. Workaround: after you created it, label it accordingly to know later from what mode you created it.

- selecting different pipelines, don't update the scale bar. Workaround: open the pipeline change the result type and then subsequently the initial result type. This will update the scale.

This year many pipeline bugs were fixes, but there are still many :(
I get a panel "Hints user defined equations", see attachment, _1.
Yes, but this just lists every possible result. No analysis will have all possible results.
Principal stresses are not available in the pipeline, see attachment, _2.
Indeed. Can you please report this in our bugtracker at GitHub?
mac_the_bike
Posts: 42
Joined: Sun Jun 30, 2019 12:56 pm

Re: FEM Pipeline units on last dev versions

Post by mac_the_bike »

Indeed. Can you please report this in our bugtracker at GitHub?
This is not possible because I don't know all the possible types of data that can be plotted in the results pipeline and those which the results pipeline doesn't recognize.
Yes, but this just lists every possible result. No analysis will have all possible results.
This is a list of the result types that can be manipulated - I don't think it implies that all the quantities are available to you. When you double click on the "Results" object, then results that are not available are "greyed" out.

uwestoehr
Thanks for your patience with my ramblings. I hope, and I bet you hope, that this is my last communication.
user1234
Veteran
Posts: 3338
Joined: Mon Jul 11, 2016 5:08 pm

Re: FEM Pipeline units on last dev versions

Post by user1234 »

I was also at first a little bit mad, but in the end @uwestoehr is right. Unify the input and output make sense. When an other FEM solver or unit system occurs, it is much easier to adapt. I am glad, that he started working on this.

Greetings
user1234
User avatar
uwestoehr
Veteran
Posts: 4961
Joined: Sun Jan 27, 2019 3:21 am
Location: Germany
Contact:

Re: FEM Pipeline units on last dev versions

Post by uwestoehr »

uwestoehr wrote: Sat Jul 16, 2022 4:03 pm - selecting different pipelines, don't update the scale bar. Workaround: open the pipeline change the result type and then subsequently the initial result type. This will update the scale.
I opened a bug: https://github.com/FreeCAD/FreeCAD/issues/7230

and here is a fix:
https://github.com/FreeCAD/FreeCAD/pull/7231
User avatar
uwestoehr
Veteran
Posts: 4961
Joined: Sun Jan 27, 2019 3:21 am
Location: Germany
Contact:

Re: FEM Pipeline units on last dev versions

Post by uwestoehr »

mac_the_bike wrote: Wed Jul 20, 2022 10:30 am This is not possible because I don't know all the possible types of data that can be plotted in the results pipeline and those which the results pipeline doesn't recognize.
Just report that the 3 stresses are not available for the pipeline while they are available in the CCX result mesh.
Thanks for your patience with my ramblings. I hope, and I bet you hope, that this is my last communication.
:shock: This is the wrong approach!

We are here to discuss to make FreeCAD better. And you found a new bug and you are suffering from 2 pipeline bugs. I don't understand why you are upset by yourself. You did not insult people nor were rude, just reported your issues - how it should be.
User avatar
uwestoehr
Veteran
Posts: 4961
Joined: Sun Jan 27, 2019 3:21 am
Location: Germany
Contact:

Re: FEM Pipeline units on last dev versions

Post by uwestoehr »

uwestoehr wrote: Thu Jul 21, 2022 1:40 am Just report that the 3 stresses are not available for the pipeline while they are available in the CCX result mesh.
I did this now: issue #7242

------
For the upcoming FreeCAD 0.20.1 release, I could almost fix issue #7230
User avatar
uwestoehr
Veteran
Posts: 4961
Joined: Sun Jan 27, 2019 3:21 am
Location: Germany
Contact:

Re: FEM Pipeline units on last dev versions

Post by uwestoehr »

uwestoehr wrote: Sat Jul 23, 2022 4:25 pm ... report that the 3 stresses are not available for the pipeline while they are available in the CCX result mesh.
I did this now: issue #7242
That will be fixed for FC 0.20.1 too. It was my mistake to allow a last minute change that added a regression.
(see https://forum.freecadweb.org/viewtopic. ... 66#p612366)
User avatar
uwestoehr
Veteran
Posts: 4961
Joined: Sun Jan 27, 2019 3:21 am
Location: Germany
Contact:

Re: FEM Pipeline units on last dev versions

Post by uwestoehr »

I think this thread can be marked as resolved as soon as FreeCAD 0.20.1 is out in some days.

I am proud to say that this will be the first release in which at least the known pipeline bugs are all fixed.

There is of course room for improvement and there might be further issues.
Post Reply