Cylindrical joint challenge

[trhodesL6HUL]

Is anyone able to recreate this problem?

Figure 1: There are two cylindrical joints among the three 5-way connectors. One between the left and center connector, and one between the right and center connector.

After applying a third cylindrical joint between the left and right connector (circled in red), Fusion 360 can't seem to compute it. Figure 2 below shows what happens:

I've seen far more complicated designs in Fusion 360, so it's probably me doing something wrong. Does anyone have a solution for this? I've attached the Fusion file to this post.

[Anonymous]

Can't you simply select the circular edge and extrude. Thus by three extrusions couldn't you complete the joints.

[trhodesL6HUL]

I don't need 3 extrusions. I have tried inserting cylinders in the openings before adding the cylindrical joint, but it still does not work when you try to add the last (third) joint. Are you able to recreate the same error? what's in the picture, or the "could not compute" error?

[Anonymous]

Can't you move the parts closer along the axis using move command. When the are apart by a small distance use the joint command . It may work. Try.

[Anonymous]

Anyway I'll try myself and let you know of the results I obtained ."Nothing is impossible "

[trhodesL6HUL]

I have tried that, and many other things. Let me know if you figure it out, or if you agree that it's a bug.

[TrippyLighting]

I tried this out yesterday and at first I thought that the angles/math might be off. That appears not to be the case.

This might be a borderline case that presents a particular challenge for the  joint system.

Lets see what @kgrunawalt and @jeff_strater have to say.

[Anonymous]

Are you expecting an assembly as in the figure with a cylindrical joint in the marked portion. I have used move command to move the parts.

[kgrunawalt]

I've played with this. The hairball solution is valid but the solver should not pick it and be locked up with it. There should be an infinite number of solutions as the parts spread out. My guess is that the hairball solution reduces some error that accumulates with the distance from the center. My measurements don't turn up the culprit however. Everything looks good.

I'll look into it.

[trhodesL6HUL]

Nathan said he's created an internal report (FUS-32367) for this. He's not sure if it's a bug or a limitation. See post in Support: https://forums.autodesk.com/t5/support/possible-cylindrical-joint-bug/td-p/7047834

[kgrunawalt]

This does seem like an issue.

(Edit) Actually, if work point distance is not along axes, then explains failing 3rd cylindrical. But why are workpoints not coincident. Hmmmm.

[kgrunawalt]

Anyway, here is a workaround. Instead of creating the 3rd cylindrical joint that fails:

1. Lock distance of one of the others
2. Align point with line using Align command and SHIFT key to rotate cylindricals so that point lies on line
3. Fully lock the joints
4. Capture the position
5. Repeat with more connectors (each new one requires one cylindrical)

This is fully parametric in that you can either modify geometry to change distance between connectors or modify the cylindrical offsetz parameters (same result).

[Anonymous]

Isn't this not the solution or are the coordinates wrong?

[trhodesL6HUL · Accepted solution]

The solution was to add more decimal places to the angle in sketch 1. I do not know why it worked when trying to connect 3 connectors, but it worked. See here: https://forums.autodesk.com/t5/inventor-forum/possible-cylindrical-joint-mate-bug/m-p/7051733/highli...