Clumsy and lagging joints

[wx188]

Hello, 

I was trying to recreate f16 main landing gear in F360 with functional joints, and I think I messed up the workflow. 

I started with making the "structural" parts first, when now I think about it, I should've had started with moving parts first and make mounts for them later. 

Now I'm stuck with weird geometry, but thats not the main concern, the worst part is that the joints are super laggy.

Sometimes they don't work at all (for example (see the pic below) this joint does not work from 15-30 degrees of motion, and it's just glitched out, no idea why and how to fix it)

This whole part is pretty rushed and not made with the best practises, but I was pushing my limites when it comes to making complex joint assemblies (I am not used to this type of things so I had to rethink and change many things throughout the process).

Some joints are ball ones, but if they were anything else the whole thing would stop working.

This is (see the picture below) how I imagine the part joints. Rev stands for revolute, and the yellow question mark is the rigid tube connecting to the wheel hub (it has no wheel now, but whatever :p); I'm not sure how it should be connected.

When I try to apply these joints to my part, it's just not moving at all.

I am looking for tips with making parts like this (the workflow, and the approach to joints; best practises) in the future.

Maybe someone can fix it or at least give me an idea of how it should be recreated to work like I imagined (without laggy joints).

Thank you for your insights in advance.

[TheCADWhisperer · Accepted solution]

@wx188 

First thing I noticed is that Sketch1 is missing dimensions and not fully defined.

When I started adding the missing dimensions, they did not make logical sense.

For a mechanism like this to work in CAD I would expect everything to be ultra-precise.

In the real world we cannot manufacture perfect parts so there are clearances between parts.

In the CAD world we sort of can, and we can model the clearances, but then Contact solving becomes an issue.

In the CAD world there are ways around this, but the discussion becomes very involved.

The easiest example I can describe is a door hinge.

In the CAD world we only need one hinge with a Revolute Joint. 

But in the real world we usually have two and sometimes 3 hinges on the door.

In other CAD softwares doing analysis I would use one Revolute Joint and one Point-Line Joint for the two hinges to avoid having to do Contact Solving to account for clearances.

In truth - this mechanism might not be all that complex to get to solve motion fluidly, but I didn't dig any deeper after examining the first sketch.

[wx188]

Hi,

Well I never ever thought that sketch would be messed up.

What I did is created the sketch (pretty long ago) using the grid because I could not find any other better way for it at that time.

The "diamond" shape is stupidly complex for how simple it is.

I have thought about making something like that:

But went with easier way and just used the background tiles; than immedatly after I created the next components.

I'm surprised the walls of the "diamond square" are different lenght.

Thank you for your observation, I will correct that for sure.

[TrippyLighting]

---

@wx188 wrote:

...

I will correct that for sure.

---

I haven't looked at every sketch in the design, but the three or four sketches I did look at were not fully defined (fully dimensioned and constrained). I would guess not one single sketch is fully defined. My recommendation would be to correct that before moving to joints.