FreeCAD Forum

After seeing @ppemawm's fantastic work with FreeCAD version 0.20 and Assembly 4 (https://forum.freecadweb.org/viewtopic. ... 210a7f6ce6 ), I tried to redraw the frame with FreeCAD LinkBranch to try my turn to stay in PartDesign thanks to the new tools integrated by @realthunder.

Here is the result :

After integrating the dimensioned images of the project at scale 1 and in order to limit calculation times when the body comprises several Sweeps, I chose to build the frame in several bodies joined together at the end of the work by a Boolean union (operation included in PartDesign LinkBranch).
The principle of construction is of course the same as that of @ppemawm, except that LinkBranch makes it possible to remain in PartDesign.
I have also incorporated solid bars: the problem of tube joints can be circumvented by an internal Sweep of the main tube at the end of the course, but the files quickly become huge.

1st body :
2nd body :
3rd body :
Boolean union of bodies after symmetry of each (again, LinkBranch makes it possible to mirror several functions, or even all of them, in a single step.
The finished frame file is 212 MB.

One small note though: after constructing a Sweep, the following functions seem not to bind directly to the result (little bug?). So I regularly made intermediate Boolean Unions.

Nicely done. I always enjoy reviewing your excellent models.
I have not had any need for LinkBranch as yet but am encouraged by what I see in your work.

Perhaps the section and tube intersection tools from the Part workbench will be integrated eventually.

jpg87 wrote: ↑Wed Jun 01, 2022 7:50 am The finished frame file is 212 MB.

The size of your file is surprising, my frame file including the individual bodies (6) and their assembly is only 13.1 Mb.

ppemawm wrote: ↑Wed Jun 01, 2022 4:00 pm Nicely done. I always enjoy reviewing your excellent models.

Thank you very much, but I am far from the quality and the number of your productions!

Thank you very much, but I am far from the quality and the number of your productions!
As far as LinkBranch is concerned, I was especially attracted by Assembly 3 which corresponds better to the approach I had to assemblies, and the novelties of PartDesign, in particular the multiple symmetry which makes it much easier to do in one operation a global symmetry that is to say a symmetry of the whole of an object while remaining in PartDesign :

ppemawm wrote: ↑Wed Jun 01, 2022 4:00 pm The size of your file is surprising, my frame file including the individual bodies (6) and their assembly is only 13.1 Mb.

I will report this to @realthunder.

jpg87 wrote: ↑Wed Jun 01, 2022 7:50 am The finished frame file is 212 MB.

One small note though: after constructing a Sweep, the following functions seem not to bind directly to the result (little bug?). So I regularly made intermediate Boolean Unions.

Can you please share your full file? I'll check. The size could be due to your repeated union operation here.

realthunder wrote: ↑Sun Jun 05, 2022 3:00 am Can you please share your full file? I'll check. The size could be due to your repeated union operation here.

Here is the file of the 1st part already shared in the post https://forum.freecadweb.org/viewtopic.php?f=44&t=69178 .

https://www.dropbox.com/s/okx8ym5r0t1tc ... FCStd?dl=0

As someone with experience in tube bending, this frame makes me anxious...

ThanklessLiving wrote: ↑Sat Jul 16, 2022 6:42 pm As someone with experience in tube bending, this frame makes me anxious...

Do you mean there's some design flaw? Or that this sounds like a pain to construct?

jnxd wrote: ↑Sun Jul 17, 2022 5:07 am Do you mean there's some design flaw? Or that this sounds like a pain to construct?

Not question for me, but I think typical frames are mostly made of straight pipes with smaller radius bends in particular places.

Source of the pic.

jnxd wrote: ↑Sun Jul 17, 2022 5:07 am

ThanklessLiving wrote: ↑Sat Jul 16, 2022 6:42 pm As someone with experience in tube bending, this frame makes me anxious...

Do you mean there's some design flaw? Or that this sounds like a pain to construct?

Not that I see, it just looks complicated and difficult to get correct, depending on what machine and software and technologies you have available. There are so many bends that are very close together, with different and/or variable radius, coupled with a rotation of the workpiece between bends, plus some of the bends would likely need to be left long and coped to fit. You would need something like a rotary draw bender with the capability for variable radius bending and that's no small thing. At least the parts are simple enough that they doesn't require a machine that can bend in both directions in the same cycle because that's another expensive feature. The cherry on the cake would be a thousand little positioning slots and holes for all kinds of brackets but thankfully I cannot see them . . .

All in all nothing looks impossible, it can just be a bit complicated to get it set up to the point where the parts match the drawings. These are definitely not parts you can feasibly make in your garage...

Hello,
I decided to make a series of documents to show workflows to reinforce the toponaming problem already well advanced in LinkBranch by @realthunder .
Here are the pieces I drew and I am writing the comments and illustrations :
The links :
http://help-freecad-jpg87.fr/05_excomment_ind.php (english)

http://help-freecad-jpg87.fr/00fr/05_excomment_ind.php (français)