Rolling a cylinder on a stationary cylindrical surface

[Factao]

Hi,

I was trying to reproduce the motion of the thumb when I encountered an constraining issue. Essentially, I need to roll a cylinder on a stationary cylindrical surface. It should be easy with a tangent constraint, but it does not seem to work. I will join the files in question, if these are required. 

Here are the more complex explanation: 

I have a ball joint, that is keeping in place by two movable guides. On one of these guides, there another attachable guide that is guiding the rotation of the ball joint in function of the angular position of the guide on which it is attached, with the help of a cylinder rolling on its surface. This should recreate the complex motion of the thumb metacarpal with only two actuator. 

I cannot make the cylinder roll on the surface, it keep being tangent with an imaginary line , creating a planar movement instead of the expected one. The axial rotation of the ball joint is not supposed to be directly constrained by any other constraint.

So, Am I trying to constrain the impossible or did I failed to constrain?

Also: I am using inventor 2018 pro

[andrewdroth · Accepted solution]

I think as the ball spins the pin axis (and therefor surface) won't be square to the slot in rotation guide. Essentially I think the rotation guide surface would need to twist to stay parallel the the rotation pin.

I can't say for sure, nut my gut tells me something is going to bind in this scenario.

Maybe constraining a reference surface sphere that is placed inside the ball part to the guide surface would allow it to solve?

[JDMather]

How much experience do you have with Environments>Dynamic Simulation and Rolling Cylinder on Plane Joint?

[Factao]

@JDMather  None, I am kind of a freshman engineering student that is a little bit in advance. I'd learned pretty much everything that I know on my own and I never needed the dynamic stimulation before. Would it be possible with it?

@andrewdroth  And, as for the parallelism between the pin and the surface in question, I will see this this evening, but the plane on which the pin is sketched seem to be perpendicular to the surface, no matter what. But, with the intended rotation movement, It would make sense that it does get a little angular variation.

And just to be sure, did you saw a constraint that goes against what I am trying to do?

[JDMather]

I will try to post example tomorrow, but first thing I notice is interference between your parts.

Inventor doesn't care about this interference, but I do.  I will have to remodel the parts.

[Factao]

I will modify that for you, Do you care about tolerance and surface contact, or is it just the cylinder that go within the matter?

[JDMather]

Unless I didn't look at this long enough - the radius of  your "yokes" is smaller than the radius of the sphere.

[Factao]

Actually, yes, this is how it is supposed to be. This mechanism, once complete, will hold the ball part, so the yoke need a surface that is in contact with an imaginary circle of smaller diameter that is created if you cut the sphere with a plane that is not on its center, but parallel to its radius. I did correct this on these attached part, it should be a better representation of the reality.

[JDMather]

I have to get some other work done and don't time to work more on this - but the Attached should give you some ideas.

Do not move in assembly environment (if you do - you can reset position by un-suppressing Home Position constraints).

Go to Dynamic Simulation and click Play.

Do you have a link to a reference that you are using for this project?

[johnsonshiue]

Hi! This case can be solved by using Solid Sweep in Inventor 2020 and Transitional Constraint. I don't believe the Extrusion Cut provides the correct geometry allowing the rod to slide through. I will share an example using Solid Sweep. But, it will be on 2020, not 2018.

Many thanks!

[johnsonshiue]

Here is the 2020 example with solid sweep. It should work better.

Many thanks!

[Factao]

@johnsonshiue  Thank you for this. However, using this solution jammed every piloting option. Is there is a way to avoid this with the constraint of the assembly?

@JDMather Thank you for the dynamic simulation. I see where you are going with this, I will complete it. However, there is no way to do this with the assembly constraint? It is a bit quicker than having to remake a new simulation tho illustrate every single motion.

[Factao]

Well, gentleman, I think I partially solve my issue by using a "sketch driven" assembly. I will join the files finished files.

Essentially, the initial issue isn't solved at all, nothing has been resolved, but it work. Let's hope that Autodesk  get some industrial spying done on Dassault, so then we can all happily start constraining with sketch.

More seriously, Is there is a way to some kind of dynamical interferences analysis? I would love to be able to know if the pin is approximately tangent whit the guide and make a verification of this last one, just avoid machining something of completely wrong.

[IgorMir]

Well Tristan - before expecting some action from Autodesk - have you grounded at least one part in your assembly? Which is pretty much a must for Inventor assemblies. Last time I had a look at it - they all were free.

Cheers,

Igor.

---

@Factao wrote:

Well, gentleman, I think I partially solve my issue by using a "sketch driven" assembly. I will join the files finished files.

Essentially, the initial issue isn't solved at all, nothing has been resolved, but it work. Let's hope that Autodesk  get some industrial spying done on Dassault, so then we can all happily start constraining with sketch.

More seriously, Is there is a way to some kind of dynamical interferences analysis? I would love to be able to know if the pin is approximately tangent whit the guide and make a verification of this last one, just avoid machining something of completely wrong.

---

[JDMather]

---

@Factao wrote:

@JDMather Thank you for the dynamic simulation. I see where you are going with this...

---

I have no idea where you are going with this?

I asked you to provide reference information, but you did not.

You mention Dassault, but not software used (I use Dassault - SolidWorks each and every day).

You are using an ancient release of Autodesk Inventor? (2018?)

Are you familiar with Contact Sets?

Are you familiar with 2D Contact Joints?

Are you familiar with 3D Contact Joints?

More information is needed about your true Design Intent.

[Factao]

@JDMather In the order:

-I think you misunderstood what I was saying. You provide me an incomplete dynamic simulation, which I am grateful you did, since you did not and do not have any obligation to help me. I then told you that I have an idea about what you where going with what you provided to me. If you do not see where I am going with this special joint, please refer at the precision on the design intent below.

-As for the asked reference informations, I answered every question you asked me. I do not know if my answer were good enough, but you did not asked for precision after I responded to you, so I thought they were. My apologize if they weren't helpful or if I did not used the forum correctly and failed to transmit these information.

-As for Dassault, I was joking about how easy it can be to constrain with sketch driven assembly in Solidwork, which we cannot do in inventor. This is by no mean a relevant information, just a little bit of humour about an absent feature of Inventor.

-Yes, I was using Inventor 2018. I did not found (nor really searched) a way to freely update Inventor 2018 to a more recent version, such as Inventor 2020. I did manage to get this done, and I am now using Inventor 2020.

-No, I am not familiar with contact set, but It seem to be what could be the "dynamic interference analysis" I need. I am going to look this up tomorrow. 

-As I previously told you, no, I never used the dynamic simulation before, I never really needed It. Thus, I have no familiarities with both 2D and 3D contact joint.

-The intention behind this design is to recreate the motion of third articulation of the thumb, which should be just over the wrist. I did create this to be an easy to control ball joint like articulation, but I had difficulties to constrained the axial rotation of the joint, which is why I did this post. I you want a better understanding of what I am doing,  If you want to visualize these motion, I am inviting you to watch the following video(https://www.youtube.com/watch?v=zyl6eoU-3Rg&t=322s), from 4:05 to 5:45, there is a great explanation about the actuations of this particular joint (the carpometacarpal joint). 

@IgorMir  I did not find the option before. I was "grounding" by constraining to the tree plane of the origins, thank you for the tips. Also, I am not expecting Autodesk to take any action. I am not in position to expect such a thing, I barely use Inventor to a fraction of its full potential. I was joking about an absent feature that their concurrent have. 

[JDMather]

---

@Factao wrote:

... how easy it can be to constrain with sketch driven assembly in Solidworks, which we cannot do in inventor. 

---

Again, no reference cited or SolidWorks assembly provided.

I will provide example in Inventor in a bit, but I suspect you will then say, "That is not what I intended," which will lead to a game of 20 questions to narrow down to true Design Intent.

[JDMather]

I suspect that there is a plethora of functionality that you have not yet discovered.

[Factao]

@JDMather  Honestly, I never required constraining with sketch in Inventor before, so I thought it was quite harder than with Solidwork, since I saw a complaint or two online, but it look like I'd been in the wrong. 

As for what I get from this example you just shown, I think I must learn to use the dynamic simulation if I want to stimulate any kind of complex motion, and I will definitely try using the contact set option to see how it will behaves in a realistic scenario, rater than constraining everything and hoping that everything is going to work perfectly.

As for the design intent, if you are talking about the how the model behaves when dimensions are modified, I haven't done a great job on these parts. I think it is to early to refine the model yet, since it is one of the many considered options, despite being the one that I am probably going for, so if these are required to solve my issue, I will do a preliminary version of them.

If these are not the design intent you are talking about, then may you define them? I am not studying engineering in an english environment, so perhaps I know what you are talking about, but it just have a completely different name when I studied it.

[Factao]

After some reflexion on this problems, The pin was not constantly parallel with the guide.