Drawing Constraints

[wendert]

I am having difficulty with an assembly in terms of actually constraining the assembly. I have a series of linkages that are coupled so that I can drive the assembly with a pneumatic cylinder. However, the last constraint couples the cylinder with a Rod Clevis to tie it all together. I have tried to constrain the model with an insert constraint, but that did not work. So, I next tried to constrain the 2 items together axially with a MATE. This worked to line up the items with respect to the axis of each item, but I still require to constrain the items to set a position of the clevis on the shaft. I am unable to do so. I know it is not an issue of the geometry not being able to actually accomplish this because I can manually move the 2 pieces through each other while constrained axially. Any ideas? Has anyone else experienced this issue with linkages? In dealing with linkages, the actual gemoetry will allow for 2 positions to be observed, and I wonder if this is why I am having such a time of it. Any help is greatly appreciated!

 

[cadranger]

Hi,

I take it you want to fully constrain the assembly. That is, allow no movement. With any linkage, you should constrain the parts to actually allow them to move like they would in real life. Use the insert constraint for pins and bolts in the mating parts. You've used the mate contraint for axis (piston in cylinder for example). Linkages generally use rotational movement and linear movement in combinations. Once all the parts are linked and allowed to rotate and slide together, you should select the best constraint the will stop the movement. Sometimes its a mate or flush with offset between origin axis planes of two parts or just between two surfaces. Using the angular constraint alone to do this is risky. Inventor doesn't respect angles very well. As you might see, the solution for a 10 degree angle can also be 180 degrees opposed. Inventor lets you state the angle but if the parts are closer to the 180 opposed solution it will use that. Once you find the ideal (there is always more than one solution)constraint, you can then use the "drive the constraint" tool. This allows you to set the links in motion with rotational and linear limits set by you so that the parts don't move through each other. You can check for interference between parts if you require a set movement but are unsure if the parts are actually touching. If you are still having trouble you can send me the assembly.

Good Luck

 

[wendert]

I had done as you have suggested to some degree. The difference being that The geometry of the assembly does not really lend itself to the final constraint that would disallow movement. The assembly is a series of linkages that don't really have a end or intermediate position that is easily defined. I built the assembly with both insert and mate constraints, but the final constraint that was to tie the assembly together would not take. I modeled the driving cylinder as one part, and adjust the stroke by changing a parameter within the part to cause the cylinder to extend and retract. So, in essence this part should have acted to fully constrain the assembly, yet give me the ability to move or drive the assembly as I changed the parameter. But, it seemed to me that Inventor did not know what to do because of the multiple solutions that were possible with the assembly itself. Have you ever run into a situation where the constraint would not take due to multiple solutions. Additionally, I have not used any angular constraints because of exactly what you stated.
I will work toward a solution on my end, but I might take you up on sending the assembly to you if I can't resolve the situation.
Thank you!

 

[cadranger]

I'm not sure what your cylinder part (ipt) is doing when you change a dimension within it. My guess is you're relying on adaptivity to do the work. Don't. Another guess is that maybe your linkage assembly is built in such a way that there is no best solution for movement. It's kind of like putting an extra link in a window wiper arm. That extra link could flop about and still do its job. Your link movements should be predictable or else you should question their value in the assembly. The fact that the final constraint is not taking suggests some sort of inaccuracy between surfaces or edges preventing proper mating. Keep me informed.

 

[mekt]

If you are not solved yet the following may, hopefully, help.

Creat work-points at the specific locations of the parts you want to assemble. then mate this work-point (or sketched work axes)

Hope I made my best to help.

Good luck.