Motion study do not move a component to its required position

[coisne]

Currently, I had a project with a multiple part box (a bottom, a middle plate and a top cover). When I move these parts manually, everything is fine, contacts are good and components are joining at their planned position.

I also checked that there is no obvious obstruction, and it's fine (at least one millimeter for the tightest fitting).

When I play the Motion Study, the top cover do NOT go to its required position (joint set to zero), and is "stuck" around 6 mm above - therefore the box is not correctly closed.

Worst: once back in modeler, when I drag it manually, it still seems "stucked" at this position and cannot be moved down easily... But if I move it up and down again, then it goes well again and I'm able to close the box again.

Any clue? Why motion study do not do as planned? If I have really an unwanted contact, how can I check it and see what is currently blocking my components from moving as expected?

[TheCADWhisperer]

Can you File>Export your *.f3d file to your local drive and then Attach it here to a Reply?

[coisne]

If possible, I would prefer to not link the whole project...

My problem is:

- Something is blocking the movement, and I need to identify what - didn't see anything special in F360 when a component push or is blocky by another component, is there a way to highlight these contacts?

- Motion study have a particular behavior (already found some strange behaviors in it with highly constrained joints) with overlapping joints (all joints shares a common axis), and I then need to know why it acts this ways and/or how to solve it properly.

If it's absolutely required, I may export/redo something to show it without my full project, but since I'm new in F360 and 3D modeling - my industrial drawing bases were 26 years ago, that's a lot! - it's probably something "obvious" that I missed... Like I missed to create FIRST a component BEFORE creating bodies in it rather than try to copy/paste/move bodies between components.

TIA.

[TheCADWhisperer]

Create a simplified dummy assembly that would require same Joints/Motions.

This is the technique that as worked for me for 30+ years of CAD, 9 times out of 10 when doing this exercise the solution is discovered.

For the odd time the solution is not discovered - Attach the dummy assembly here.

[coisne]

Here you go:

https://a360.co/337yWH3

This very simple case shows the problem, despite I build it roughly with boxes instead of (quite) complex sketches.

You can try: in modeler, the top cover MUST go to the bottom's cleat and touch it. The top cover also "push" properly the middle plate if required.

In the saved motion study, it doesn't.

[jhackney1972 · Accepted solution]

In your requirements, TOP component must move down and rest on the BOTTOM component.  This would intersect the Middle component.  This motion is not possible in your current model because you have Contact Sets On for all components.  I am attaching two models, the first - Blocked Motion Study - JRH1 has the motion study moving the MIDDLE component to rest on the BOTTOM component and the TOP component moving to rest on the MIDDLE component.  The second - Blocked Motion Study - JRH2, I have removed the Contact sets and both the TOP and MIDDLE component move to rest on the BOTTOM component.  I was not sure, from your stated requirement, which one you wanted.  I added a Sectional Analysis just for clarity.

[coisne]

The JRH2 is non-pertinent for my use, they're real objects and cannot pass through each others. The JRH1 is what I was looking for.

But I do not understand WHY... On my real design, the top is set (in motion study) to go to 0 mm. It's stuck around 6 mm above its required position - at 0 mm, it should "push" the middle plate and stay stuck there.

Regarding what you did in JRH1 design, I set the motion study to go to 6 mm (not -6 mm as I thought first) and it works, it's joining as expected.

But clearly, I don't understand. Why the instruction initially used (zero) does not work as intended? Apart the contact set (which is not reached in the blocked position, top is way above the middle plane!) the middle plane is in no way involved in this joint, it should be only a mechanical limitation to the movement... And it doesn't work as expected, and worst, even produce an undefined behavior - at least from my viewpoint.

Do you have an explanation? I'm glad to have solved my motion study, but it's frustrating to not understand why it doesn't work as expected, since I was planning to use motion study to "push" everything to limits and see if everything is fine... For me, it's still not THE solution, but only a bypass.

If I cannot have better than a bypass, fine, but I would prefer to know exactly what happens behind the scene to produce such strange results. If you don't know, please tell anyway, it's still an answer - and maybe a Fusion 360 bug, after all.

Thanks in advance.

[jhackney1972 · Accepted solution]

First of all remove all Contact Sets, they will just get in the way.  Motion Studies simply mirror what you have set for each joint motion limit.  The MIDDLE component limit is set to 0 and it works fine.  The TOP component must stop when it get to the top of the MIDDLE component so the motion limit of travel is set to 10mm, the thickness of the MIDDLE component flange.  The the Motion Study will use the value of the motion limit. Model is attached.

[coisne]

OK... After a bit of thinking, I found my misinterpretation:

• Dragging components in the Design environment is NOT the same as using "Drive joints" command.
• Motion study do exactly what "Drive joints" do, NOT what dragging components do.
• Driving a joint to an INVALID position does NOTHING. The component do NOT move AT ALL. And that's the problem, precisely.
• Motion study is NOT global, but totally step by step process without ANY notion of previous context.

In my case (real design): the movement was done to go from 200 mm (step 0) to 0 (step 75), while 6 mm (and not 0) was in fact the last valid position.

The parametric equation of this joint is obviously P(x)=200-8.x/3 with a step of 2.67 (8/3) mm.

Remember, once a value is invalid, component do NOT move...

So, here we are:

• P(72)=8 mm, OK (because above 6 mm limit), component is moved.
• P(73)=5.33 mm, INVALID (because below 6 mm limit). Component stays at previous position, 8 mm.
• P(74)=2.67 mm, INVALID. Component stays at 8 mm.
• P(75) and after=0 mm, INVALID. Component stays at 8 mm.

And that's exactly what my problem was...

@jhackney1972, thanks a lot.