The need for a default assembly workbench

Have some feature requests, feedback, cool stuff to share, or want to know where FreeCAD is going? This is the place.
Forum rules
Be nice to others! Read the FreeCAD code of conduct!
User avatar
wandererfan
Veteran
Posts: 6268
Joined: Tue Nov 06, 2012 5:42 pm
Contact:

Re: The need for a default assembly workbench

Post by wandererfan »

adrianinsaval wrote: Sun Mar 26, 2023 4:05 pm I'm not sure fi this is correct, from what I understood earlier I think it's not desired to have standalone Part::Feature objects directly under the assembly container, so instead this would be App:Part too.
I don't know. The document says that a Part is something that "cannot be further decomposed" and App::Part can be decomposed into the LCS and the Bodies/Features inside it. It also says that an assembly contains at least 2 Parts, so if App::Part is "Assembly" then having only 1 component inside it would not be in agreement with the document. The document is pretty solidly in the conceptual domain and doesn't talk much, if at all, about the implementation domain.
I think it means much more than that, in other CAD systems you do some operations in the assembly, like mirroring, pocketing and can't remember what else. I think this has been mentioned before here too.
After reading a little more, I'd like to change that translation. In the case study, the bearing seat in component A and the outer race of component B are Assembly Features. There is another entity called ParametricAssemblyConstraint that says those 2 Assembly Features are coaxial. I don't know what the right translation would be - maybe it is as simple as a Face or Edge.

I haven't found anything yet about "operations", only data.
User avatar
saso
Veteran
Posts: 1920
Joined: Fri May 16, 2014 1:14 pm
Contact:

Re: The need for a default assembly workbench

Post by saso »

Here is the link to NIST STEP File Analyzer and Viewer with links to the standards, test models etc.

https://www.nist.gov/services-resources ... and-viewer STEP File Analyzer and Viewer
https://github.com/usnistgov/SFA

more...
http://www.ap242.org/ STEP AP242 Project
https://www.nist.gov/ctl/smart-connecte ... /step-nist STEP at NIST
https://www.nist.gov/services-resources ... ibrary-scl STEP Class Library (SCL)
https://www.mbx-if.org/cax/cax_introduction.php CAx Interoperability Forum (CAx-IF)
https://www.mbx-if.org/cax/cax_recommPractice.php CAx Recommended Practices
https://www.mbx-if.org/cax/cax_express.php CAx Express Schemas
https://www.mbx-if.org/cax/cax_stepLib.php Step File Library
https://www.mbx-if.org/cax/vendor_info.php AP242 Implementation Coverage
http://www.pdm-if.org/ STEP AP242 PDM Implementor Forum


And not related to this topic but maybe useful to someone, here is also the NIST IFC File Analyzer

https://www.nist.gov/services-resources ... e-analyzer IFC File Analyzer
https://github.com/usnistgov/IFA


I have used and know this for a long time but have to confess that I also did not use or study it in full detail, in the past we have been with colleagues mostly testing this by just importing and exporting different models from different cad tools we were using and checking how similar or different the result were.

To note, the STEP standard and interoperability covers of course much more then just assembly. For the geometry itself we can hope that the OCCT STEP importer/exporter that we are using is doing a good enough job, but that is not really related to this topic. The standard can and should be also useful to those that try to develop for example MBD (model-based definition) or GD&T (geometric dimensioning and tolerance) etc. but this are also different topics. For assemblies however, I hope it is clear that just throwing something together and calling it assembly will just not work for this.
Last edited by saso on Tue Mar 28, 2023 11:33 am, edited 11 times in total.
User avatar
onekk
Veteran
Posts: 6146
Joined: Sat Jan 17, 2015 7:48 am
Contact:

Re: The need for a default assembly workbench

Post by onekk »

maybe poking @yorik as he is working on IFC implementation...

Regards

Carlo D.
GitHub page: https://github.com/onekk/freecad-doc.
- In deep articles on FreeCAD.
- Learning how to model with scripting.
- Various other stuffs.

Blog: https://okkmkblog.wordpress.com/
freedman
Veteran
Posts: 3441
Joined: Thu Mar 22, 2018 3:02 am
Location: Washington State, USA

Re: The need for a default assembly workbench

Post by freedman »

I wrote this earlier in a post:
I hope we can get past using geometry for an attachment reference. Using a Sketch, origins or LCS you can have one icon to make an attachment.
I was hoping there might be some discussion so I will try again. Over the years I have used Solidworks, Inventor, ect. so I have used their assembly concepts. You pick geometry i.e. a hole then an object, then decide what constraint to add to the object. This works and it resembles A2+ and Asm3.

What bugs me about this is the code complexity, it seems like using geometry is more difficult to use, it requires more decision making, error handling, chance for error and challenging on curved surfaces.
One could say the geometry follows the sketch or source workflow and that is true but there is a down-side to that, if you change a sketch and that changed some geometry unexpectedly then you have a model error without knowing it. I have always felt like I'm walking on thin ice when I reference the edge of an object.

Using Geometry vs Sketch,Origin,LCS; Is there a difference in how robust a large model would be? Does anyone know what they use to design aircraft for instance? Lots of parts and many opportunities to make errors.
User avatar
saso
Veteran
Posts: 1920
Joined: Fri May 16, 2014 1:14 pm
Contact:

Re: The need for a default assembly workbench

Post by saso »

freedman wrote: Tue Mar 28, 2023 4:00 pm I hope we can get past using geometry for an attachment reference...
I plan to make a post that would review also some of this, but wanted to fist lay down some foundation on the asm/part tree structure... In short it is not one type or the other in reality it is more like a spectrum, so to say to have them all and be able to use them for different needs, from the more positional LCS, to general constrains, to kinematics and to dynamics... For large assemblies with bigger team LCS type is normally the way to go, maybe you can find some posts about it from the user Pauvres_honteux...
User avatar
FBXL5
Posts: 980
Joined: Sat Aug 03, 2019 8:45 pm

Re: The need for a default assembly workbench

Post by FBXL5 »

freedman wrote: Tue Mar 28, 2023 4:00 pm Does anyone know what they use to design aircraft for instance?
Aviation and automotive design rely on PDM systems that also control the placement of objects in the virtual space. (mainly Catia and NX in combination with several PDM systems)


While modelling, assembly WBs are used to position objects to each other and to a global coordinate system via constraints. When the data is transferred to the PDM system, the placement of each object within the global CS will be derived form the constraints and stored with the objects.

If you reload an assembly the default placement of the objects is rather controlled by the PDM system than by the constraints (the assembly document may have changed in the meantime). To change positions, assembly WBs are used again and then the new positions are again derived from the constraints when the Objects are retransferred to the PDM system.
GeneFC
Veteran
Posts: 5373
Joined: Sat Mar 19, 2016 3:36 pm
Location: Punta Gorda, FL

Re: The need for a default assembly workbench

Post by GeneFC »

Three totally off-topic, crude, and snarky comments deleted.
User avatar
LVAeronautics
Posts: 105
Joined: Sun Nov 20, 2022 6:21 pm
Contact:

Re: The need for a default assembly workbench

Post by LVAeronautics »

sliptonic wrote: Tue Mar 14, 2023 4:40 pm Can we start by agreeing that a default assembly workbench is in the best interest of our users and should be a strategic goal for the project?
Late poster to this thread, but agree 100% A common maintained assembly feature should be part of any standard FreeCAD load.

We are still wrapping our heads around FreeCAD and making huge strides with great results. This said, the majority of our assembly experience is with A2plus, but learning Assembly4, and seeing its strengths. Given the nature of our engineering project, Assembly 4 looks to be the better of all of them.
"Testing leads to Failure, and Failure leads to Understanding" -Burt Rutan
grd
Posts: 328
Joined: Wed Apr 13, 2022 5:13 am
Location: Eindhoven, The Netherlands

Re: The need for a default assembly workbench

Post by grd »

sliptonic wrote: Wed Mar 29, 2023 12:59 pm It's time to take a look at Assembly 4
Again a very thorough review. I like it and look forward to the rest.
About Nim. Latest Release 2.0.2. Here is Nim in 100 seconds and a Nim package. There are Qt and OCCT packages.
Post Reply