Can't get assembly3 to work

[bambuko]

but still nothing....

Yes, assembly 3 is very powerful workbench but takes some getting used to
I like it very much (see: https://forum.freecadweb.org/viewtopic. ... 47#p555958)

This is very good page if you are looking for help and inspiration: http://help-freecad-jpg87.fr/02_ass_ind.php

I took your file and made some changes
It goes together OK (have a look at it) but I still have questions about this particular constraint, so I asked about it on github
(BTW like I have said earlier in this thread I wouldn't use axial alignment, but that's another story...)

as3test(bambuko).FCStd

(29.31 KiB) Downloaded 21 times

Screenshot from 2022-04-15 22.35.35.png (78.55 KiB) Viewed 1337 times

PS it might be an idea to change the title of this thread? to - "assembly 3 axial alignment constraint" or something like that?
(helps others searching for the topic?)

[FBXL5]

Part of the problem is that neither OP or you have shared your files so I don't know what have you chosen for the elements used in the constraint?
Have you used edges, cylindrical faces or planar faces?

For the record: I used the cylindrical faces for the elements.

But my aim was to give a pad on the back and tell OP that he did nothing wrong so far...

Yes, assembly 3 is very powerful workbench but takes some getting used to

Whenever I try to build a kinematic assembly I find myself translating wikipages or extending some original pages if there is too much content missing.
Right now I'm trying to create a python script for a dialog window to control the actuator of a kinematic assembly and to take an image sequence while moving. The hardest part is to comment properly and to describe it on a wiki page.

I like your locomitive model. I've heard about the obsession with railway scale models on the britsh isles.

[lcorley]

When I open as3test(bambuko).FCStd, the constraint does not work. If I use the move tool, the bar can be moved in any direction. There must be something wrong with my setup. But I'm using the 0.20.28697 (Git) version un-modified. I just unzipped it and ran it.

FBXL5, when you mention messages about remaining DOF, where do you see these?

regards,
Leon

[bambuko]

When I open as3test(bambuko).FCStd, the constraint does not work....

what do you mean "it doesn't work" - can you be more precise/descriptive?
I open it in 0.20 perfectly fine:

Screenshot from 2022-04-16 08.15.08.png (92.67 KiB) Viewed 1217 times

this is what I use:

Code: Select all

OS: elementary OS 6.1 Jólnir (Pantheon/pantheon)
Word size of FreeCAD: 64-bit
Version: 0.20.28671 (Git) AppImage
Build type: Release
Branch: (HEAD detached at fa059e7)
Hash: fa059e7b7e87c77408737903a294f076b97eb71b
Python 3.9.12, Qt 5.12.9, Coin 4.0.0, OCC 7.5.3
Locale: English/United Kingdom (en_GB)
Installed mods: 
   * DynamicData 2.46.0
  * ToolViewReverse_ToolBar
  * sheetmetal 0.2.49
  * Manipulator 1.4.3
  * fasteners 0.3.40
  * ThreadProfile 1.81.0

If I use the move tool, the bar can be moved in any direction...

Which "move tool" are you using?
I know in my early days I was also confused about possibility of using https://wiki.freecadweb.org/Std_TransformManip
to mess up assemblies
Remember also, that Axial Alignment constraint leaves assembly part with two degrees of freedom (you can rotate and slide constrained part along the axis)

There must be something wrong with my setup. ...

more likely confusion I have tested using the same setup...

....when you mention messages about remaining DOF, where do you see these?

in report view (I assume you have it enabled? toggled ON?)

report view.png (54.59 KiB) Viewed 1217 times

[bambuko]

...But my aim was to give a pad on the back and tell OP that he did nothing wrong so far...

I was pretty confused and had a lot of questions in the beginning as well so no problem at all

I also had a reply from the boss, confirming my advice earlier on in this thread
(I have asked on githhub: https://github.com/realthunder/FreeCAD_ ... ssues/1208)
because I was beginning to doubt my sanity faced with your questions guys )
and here is the reply:

Screenshot from 2022-04-16 08.30.35.png (140.18 KiB) Viewed 1213 times

[FBXL5]

when you mention messages about remaining DOF, where do you see these?

in report view (I assume you have it enabled? toggled ON?)

Yes, it was in the report view.

reply from the boss

And so we have to wait patiently until the developers of the solver improve their application...

[bambuko]

...wait patiently until the developers of the solver improve their application...

I personally go back to my original recommendation of using different constraint.
It may not feel "intuitive" but it actually works and is a lot easier (proof of the pudding being in eating - that's why I have attached my model).
Plane Coincident constraint is much better choice (particularly when combined with element offset, and occasional element flipping when necessary).

Screenshot from 2022-04-16 13.41.49.png (28.31 KiB) Viewed 1138 times

Seriously - it works, it is robust, doesn't require additional constraints to fully constrain it.
All the joints in my loco motion are done this way...
Why would you insist on Axial alignment?

[lcorley]

I'm using axial constraint because that's what was used in JOKO eng's demo video to create a swivel joint.
I'm using the "Axial Move Part" tool in the Assembly3 WB. Again, because that's what's in the demo.

But I'm beginning to think I have a deeper problem with my installation.

I tried several constraint types -- none worked.
I get red colors on the elements, not orange or black.
Nothing shows up in the report view.

I erased and re-unzipped my installation - same response

Code: Select all

OS: Windows 10 (10.0)
Word size of FreeCAD: 64-bit
Version: 0.20.28697 (Git)
Build type: Release
Python 3.8.13, Qt 5.12.9, Coin 4.0.0, OCC 7.5.3
Locale: English/United States (en_US)
Installed mods: 
  * A2plus 0.4.56
  * Assembly3 0.11.3
  * Assembly4 0.11.10
  * fcgear 0.4.56
  * LCInterlocking
  * Manipulator 1.4.3
  * sheetmetal 0.2.49

I think I'll try a different computer.
I'll report back in and hour or so.

regards,
Leon

[lcorley]

OK, different computer.
Unzipped to desktop
Opened Freecad

Code: Select all

OS: Windows 10 (10.0)
Word size of FreeCAD: 64-bit
Version: 0.20.28697 (Git)
Build type: Release
Python 3.8.13, Qt 5.12.9, Coin 4.0.0, OCC 7.5.3
Locale: English/United States (en_US)

Loaded as3test(bambuko).FCStd
Mouse over Axial Alignment constraint in model tree.
Elements turn orange in view
Report view says-

10:08:43 9.881496 <asm3.gui> gui.py(86):
Traceback (most recent call last):
File "<string>", line 577, in _catch
File "C:\Users\adryc\Desktop\FreeCAD_weekly-builds-28697-Win-Conda_vc14.x-x86_64\bin\lib\site-packages\freecad\asm3\gui.py", line 64, in _setElementVisible
FreeCADGui.Selection.updateSelection(vis,obj,subname)
TypeError: argument 1 must be bool, not list

Cleared report view
Deleted and re-applied AxialAlignment constraint.
Report view says-

10:34:52 26.095037 <asm3.gui> gui.py(86):
Traceback (most recent call last):
File "<string>", line 577, in _catch
File "C:\Users\adryc\Desktop\FreeCAD_weekly-builds-28697-Win-Conda_vc14.x-x86_64\bin\lib\site-packages\freecad\asm3\gui.py", line 64, in _setElementVisible
FreeCADGui.Selection.updateSelection(vis,obj,subname)
TypeError: argument 1 must be bool, not list

Tried PlaneAlignment with the same result.
Maybe I should try a fresh download of the distribution.

[bambuko]

...Mouse over Axial Alignment constraint in model tree.
Elements turn orange in view
Report view says-

Code: Select all

10:08:43  9.881496 <asm3.gui> gui.py(86): 
Traceback (most recent call last):
  File "<string>", line 577, in _catch
  File "C:\Users\adryc\Desktop\FreeCAD_weekly-builds-28697-Win-Conda_vc14.x-x86_64\bin\lib\site-packages\freecad\asm3\gui.py", line 64, in _setElementVisible
    FreeCADGui.Selection.updateSelection(vis,obj,subname)
TypeError: argument 1 must be bool, not list

it's not you
I did the mouse over bit and got the same message
Don't know what it is
I will create simple model in 0.20 just in case it is some problem between Master and Link branch
back in a tick