Let cams move a shaft in Inventor

[vidiii]

Hi,

I have made a small basic motor in inventor. I have constrained the drawings in a way that the motor runs correctly, however there is one error. The cams on the camshaft need to push up a shaft that is connected with the valves. Is there a possible way to let the cams push the valve up and down.

Screenshots of the problem:

http://users.pandora.be/ahmadi/tijdelijk/ScreenShot1.jpg

http://users.pandora.be/ahmadi/tijdelijk/ScreenShot2.jpg

http://users.pandora.be/ahmadi/tijdelijk/ScreenShot3.jpg

I hope to hear from you. I have tried many constraints in vain. Any ideas are welcome.

Kindest regards,
mab

 

[aardvarkdw]

What constraints have you tried? have you tried using a tangent constraint? Can you post the assembly?

David

 

[btrueblood]

Agree with aardvark, you need a tangent constraint b/n cam surface and follower. This will probably work better if the cam follower is a spherical ball or roller, not the flat plate you have shown.

Your next problem will be the spring - did you want it to deflect with the cam motion (acceleration), or is it just there to look pretty? I'm not sure how to link the spring length to an assembly parameter, but know it can be done.

 

[tmox]

i would think that the problem with the tangent constraint is the fact that the lobe has two different radii. you can probably pick one but, not both.
i suggest you check S.Dotson website for tutorials.
there is one for relative motion, where you define a plane, say on you shaft centerline, then using a formula relative to a known "clocked" position on the shaft relative to the bottom of you piston. Simply put, one o'clock - up .125, six o'clock - (max rise) up .5, 8 o'clock - down .38, etc.. Your part is constrained by a formula.
There is also a tutorial there on making the spring length adaptive so that it will work as well.

good luck.

 

[aardvarkdw]

I had completely forgotten about using functions. Thanks for reminding me tmox. I have to go and reread those tutorials now...

David

 

[btrueblood]

You may be right, Tmox, the cam profile I used was a spline, and it worked with a tangent constraint (but the modeller "saw" it as one continuous surface).