Assembly joint process recommendation

[Anonymous]

Hello,

I was struggling with making an assembly of many parts so I simpified it and attached it here.  

I have a chain of 4 links that I want to assemble between 5 pivots.  The pivots at the ends of the chain are to be grounded; 2 pivots are free to translate, and the central pivot is constrained to vertical motion only.  All of the link ends rotate on the pivots.  I want the assembly to move with the proper kinematic motion when one of the links at the end of the chain is rotated around its end pivot.  

This is my construction process so far:

1.  Make intersecting planes to define the location of the initial pivot axes.

2.  Create axes for the initial pivot points.

3.  Draw all pivots in 1 sketch at the axes points and symmetrically extrude them.

4.  Sketch the initial link design between 2 pivots.

5.  Extrude the link sketch and create a component from it.

6.  Copy the link component and paste into the top level assembly.

7.  Move the new link between the next 2 pivots and orient it roughly inline with the pivots.

8.  Repeat this for the remaining links.

What I want to do now is create joints such that the links will rotate on the pivots and also maintain axial position on the pivots.  When I tried this with the real assembly it did not go well.  How do you recommend I go about making the joints between the links and pivots?

Thanks,

Stephen

[wilkhui]

Hi Stephen,

What difficulties are you facing, why didn't it go well for you?

Is the below roughly what you're trying to achieve?

Cheers,

Indy

[Anonymous]

Yes thank you.  The joint type shoud be revolute or cylindrical.  

The real model is more complex.  I know the initial XY locations of all of the pivots.  However, I do not know the exact lengths of the links between them.

The link is made from 2 rod-end connectors,  and 1 rod between them.  I didn't model that in the simple model.  

What I want to do is put the rod-end connectors onto the pivots with cylindrical joints; then, assemble the link rods between the link-end connectors using cylindrical joints.  Thus, the rods should be connected between the link-ends and oriented correctly between the pivots.  I'm assuming the assembly solver could figure this out.  Do you think so?

I attached some pictures of the real model assembly with descriptions, and also the model archive.  In the model the rods are not yet connected to the rod-ends, that is what I am trying to accomplish.  When I add these cylindrical joints, the components I selected as grounded move their orientation and they should not move or rotate.  

Specifically if I try to make a cylindrical joint between LINK #2 and ROD END #3 then all of the other components attached to ROD END #3 by joints rotate with it.  But the SLIDER RAIL is grounded so none of them should move, just the ROD END #3.

[Anonymous]

Hi I hope you can continue to help me.

I reverted to the simple link model, I attached it again.  Here is what I did so far:

1. I made the slider rail and the slider block components.  Then I made a slider joint between them.

2. I grounded the slider bar.

3. I made a rigid joint between the slider block and the pivot3.

4. I assembled the link1 betweem the first pivot1 & pivot2 using revolute joints.  

5. I rotated the first revolute joint 10 degrees to make it easier to select the joint origins for the link2 connection.

6. I made a revolute joint between pivot2 and link2.

7. I tried to make a revolute joint between the link2 and the pivot3, but then I get a joint warning error.

I assumed when I made the last joint that Fusion 360 would automatically adjust the joint angles and positions to move the joints into proper position.  Is this not possible?

Thanks

[burnandreturn]

When you connect your link rods to your tierod ends I would use a ridgid joint.  A cylindrical joint will fall apart I would think.  Unless you place limits and that is time consuming and not neccessary. 

[wilkhui]

Hi Stephen, sorry for taking a while to respond.

I think there's probably an incorrect face selected somewhere because if I delete Rev6 and try to add it again it works fine and Fusion automatically chooses a cylindrical joint:

Hope this helps!

Indy

[Anonymous]

I was using a cylidrical joint because I didn't know what the accuracy the assembly solver required to make the joint work, so by making it cylindrical, it leaves the axial dof undefined in case there was any error in moving the part to the right location.  That was my concer, I don't know if it's valid.  In the real model I made the joints between the rod ends and rods to be cylindrical also as I wanted to leave the sliding dof open incase I had any error in the rod length.  I was trying to get the program to define the rod length for me.  I thought it might be easy to do and very accurate as it would solve for the locations exactly.  Do you know if this might work?   Thanks.

[Anonymous]

Thanks I didn't get to try it yet but I will tomorrow.  My process was to make joint origins for both parts that I was going to join.  Is it better to just choose the surfaces like you did here?  Is this the "As Built" method?  I need to locate the links on the pivots axially at some difference that's why I chose revolute.  I will try more things tomorrow.  Thanks again.

[wilkhui]

Hi Stephen,

Sorry for not being more clued up on terms, what's the 'As Built' method? Edit: That should have been obvious, d'oh! I tend to model things the way that they would be manufactured so yeah, I suppose this is an interpretation of the 'As Built' method.

Regarding setting up joints, I just choose the surfaces first because I find it simpler so my advice would be to use the workflow that you prefer 🙂

Indy

[Anonymous]

I got the test parts to be assembled correctly, and it animates correctly when you drag one of the end links.  Thanks.  I attached the model if you're interested.  Now I need to get the same thing to work on the real assembly (CreeperFeetMechAssy1...).  I went through it and changed all of the revolute joints to cyindrical joints.  I also didn't know when grounding a component you have to ground all of the sub components seperately, that was causing a problem. I got the assembly correct, but I have one problem left, the shaft does not rotate.

The shaft was the first component created.  If you go to the Cyl2 joint in the timeline you see when I put in the bearing block just before it.  The shaft should rotate in the bearing block because of the Cyl joint but it does not.  You should be able to rotate the shaft by dragging the pulley in the middle it has a rigid joint to the shaft. 

This is a very simple problem with a simple solution I'm sure.  Can you check the creation of the Cyl2 joint in CreeperFeetMechAssy?

Thanks

[Anonymous]

Hi, I got the assembly joints to work all together now.  Thank you for your help.  I could have used your feedback earlier today but I got it to work by the end of the day.  It took a long time to figure out.  I think the issue was some of the initial componenets had sub-componenets, and I had to make them a rigid group, even though the sub-component was the only item in the sub-assembly.  Thanks very much for your help

[wilkhui]

Hi Stephen, sorry about leaving you hanging but I can't help the difference in our timezones (I'm on GMT +1). I'm glad that everything worked out in the end.

Indy