How to animate a 3D mechanism

[zayas6PYNJ]

Hi all,

I'm hoping somebody could help me with this problem I'm having in the early stages of my design. The idea for the project is to create a mirror which can pivot or rotate back and forth via a handle that the operator moves from a distance, either this or possibly a mirror that they can slide from side to side (kind of like a crank-slider) to view what they need in order to work. I instantly thought of a bus door opening and closing and am working to create a handle like that. I'm struggling to see the motion of this though and figure out if it would move properly. I tried to animate it using techniques I learned from these videos but to no avail. 

https://www.youtube.com/watch?v=Y4VAJ5mNnng

https://www.youtube.com/watch?v=oTkNbi32ZNw

https://www.youtube.com/watch?v=zu6gMPZO86U <--- bus door opening

Can anybody help with animating something in 3D so that I can continue progressing on this project?

Thank you in advance!

[j.palmeL29YX]

Show us your 3D-model (.dwg) and we can tell/show you how to animate the mechanism (if possible). 

[leeminardi]

The bus door mechanism is a simple 4-bar linkage similar to my youtube video (your first link reference) if you consider just one of the doors.  The second door can be considered a 4-bar linkage as well that is driven by the first door (the result is a 6-bar linkage).  The mechanism is 2D (not 3D as you refer to it) as all the kinematic action happens in a plane.

I would make each of the components a 3D block (the door, handle, linkage). On a separate layer I would have the 2D geometry that is referenced by AutoCAD's parametric constraints similar to what I did in my tutorial. 

If you post a file with the 3D geometry I will  take a stab at it.

Lee

[zayas6PYNJ]

Here you go, I didn't do anything with realistic dimensions yet because I was more focused on getting this aspect of it right and then worrying about size. 

[zayas6PYNJ]

I actually struggling with this as well, in the second youtube video link I posted I was able to get that rotating motion but with the first I couldn't. When I was trying to coincide the different parts the solid parts would stay in place while only the line would move to be coincident with the other.

[j.palmeL29YX]

---

@zayas6PYNJ wrote:

 When I was trying to coincide the different parts the solid parts would stay in place while only the line would move to be coincident with the other.

---

The additional lines must be included in the block definitions. 

And also be aware: All lines (and other geometrie which is involved in the parametric) must be in the same Z-Level. 

(This is one of the big limitations in AutoCAD: Parametric works only in 2D). 

[j.palmeL29YX]

---

@zayas6PYNJ wrote:

Here you go,  

---

Thanks. 

It's not yet clear how the mechanism should work: 

- The green lever can turn around the magenta cylinder. But how should this cylinder move in the curved groove (what does caus this movement).
- How should the plate (cyan) move? Move planar along the large surface? Or turn around any (which) axis? Or what is your goal? 

[leeminardi]

Upon closer examination of the door mechanism what you have is a 5-bar linkage not a 4-bar linkage.  The red lines in the image below can be considered to be the links.

In the following image I have isolated the links of the mechanism.  The links are labeled in yellow and the joints in cyan.

A 4-bar linkage has one degree of freedom (1 DOF). That is, rotate one link and the position of the other links can be precisely determined.

Grubler's equation can be used (except for special cases) to determine the number of DOF of a planar mechanism.   The general equation is:

DOF = 3 * (L - 1 ) - 2 * J

where L is the number of links and J the number of joints.

For a 4-bar linkage this is DOF = 3 * (4 - 1) - 2 * 4 = 1

For the bus door mechanism we have

DOF = 3 * (5 - 1) - 2 * 5 = 2

What this means is that we have to consider the rotation of 2 links to determine the amount the door is open..  There's the handle link  (link #2)  AND the motion of the pin in the circular slot which can be modeled as an additional link (Link #1).

I do not have the time right now to rig this mechanism and there are some misalignments in your file that need to be fixed first.  The handle holes do not line up (see below).

Post a cleaned up version of your file and I will look at it later today.  You should also check that the position of the components make sense for the door open and closed positions.

[zayas6PYNJ]

Sorry for being clueless all, you're a big help though! I've been self-learning everything since graduating and don't have any real-life guidance so everything I learn is from forums like this, youtube, or trial and error so thank you bunches.

I'd like to be able to use the magenta cylinder on the left as a handle for the operator to grab and push along the curved groove. With this motion I'd like the mirror (cyan plate) to move back and forth or rather up and down so they can see what's going on out of their view (think of a dentist using a small mirror to see the backs of your teeth because they can't see it normally from where they sit). I'm not sure if this is possible so I suggested perhaps the mirror (would be a small size) being able to slide back and forth (or rather side to side) at a fixed angle rather than what I mentioned above and what's shown in the .dwg as a longer stationary mirror that is rotated back and forth. 

I hope this makes sense! I'm sorry if it doesn't but I can certainly try to clarify again if needed!

Very much appreciated!

[j.palmeL29YX]

I'd suggest a different sequence of the steps of your workflow. 

1.: Create a kinematic schema what the mechanism should do. (as attached dwg example).

2.: Add the needed parametric contraints (and dimensions) until you get the wished movements. (>>video<<)

3.: I necessary change dimensions until the mechanism works as expected.

4.: Now you can add Material to to each of the blocks to get real solid objects. Each block definition should include the appropriate geometry of the kinematic schema and also the 3D-Solids).

5.: After finished test the complete mechanism (e.g. check for collisions or other problems). 

=> Done.

[zayas6PYNJ]

Hi, 

Thank you for your response! Yes, the drawing I made is a really messy rough draft but I've updated the drawing file with a new drawing that isn't weirdly angled so everything should be properly aligned now. Kinematics was a difficult class for me so I hope you can bear with me, I appreciate all the help I can get and am excited to learn from my mistakes/shortcomings. I believe the motion should be correct, although I am picturing this as if I were grabbing the handle and when it goes through the circular groove the mirror on the other end of the pole will in a sense be pushed/pulled back and forth depending on where I am in that circular groove. 

[zayas6PYNJ]

I'll definitely try this thank you!

[leeminardi]

In your latest file you can consider the handle to have 2 links.  One from the center of the slot (point A) to the pivot on the slot (point B) and a link from B to C.  Consider AB to be link #1 and BC to be link #2.

A kinematic model of the mechanism using AutoCAD's parametric constraints would look like this. Note that it has 5 links (including the ground).

 

Note that the the two end vertices are locked to ground.  All the links are dimensioned so that their lengths do not change as the mechanism is rotated.

In the image below links 1 and 2 are rotated together about point A and you can see that the door slightly rotates.

The 2 links are rotated together again to a more "open" position.  As you can see the door opens more but perhaps not as much as desired.  The 4.6... dimension should probably be decreased.

The image below show the configuration when links 1 and 2 are rotated by different amounts.

In reality, there is friction in the system that affects which link rotates first.

You should play around with the link dimensions in the parametric model to get the action you desire. Then you can adjust the solid model to fit the revised link lengths and make blocks out of each component.

YOu can manually rotate link 1 or 2 by themselves of together to see the action.  I would need to modify my vlisp program animaterotation to animated the resulting mechanism.