Offset sprocket/cam in linear->rotational system? 2

[slmechanis]

Easy picture example

A pneumatic piston pulls back on a tray, which has chain attached to either end, wrapped around a sprocket, causing a shaft to rotate.

My thought is that if I offset the shaft and turn the sprocket into a sort of cam, I can increase low end torque and high end velocity, but I have no idea how to calculate this as the chain is completely wrapped around the sprocket it isn't clear where I should be calculating force from, i'm assuming that I should be calculating mechanical advantage against the part of the sprocket closest to the side that is being "pulled" from the chain, so in the image example the shaft in the new design would be giving the desired behavior.

 

[MikeHalloran]

In either case, how do you get the chain to pass itself?

Maybe you need to build a fast throwaway prototype to understand the problem.


Mike Halloran
Pembroke Pines, FL, USA

 

[tbuelna]

"Easy picture example"

Should have quit one sentence sooner.

 

[slmechanis]

@mike: there would be either 2 sets of teeth, or a guide to guide the chain past itself in its final iteration (or straight machining a sprocket in a spiral pattern), I'm not overly worried about the chain folding over itself since it works in the prototype, this was largely a math question since i'm not quite getting the reaction time I want and would rather not start making concessions for something I could fix pretty easily this way.

my instincts tell me that at any given point I should be calculating mechanical advantage from where the pulling side of the chain first contacts the sprocket at that particular moment in time, a real quick mockup I made seems to behave in that way but I'd rather know the real math for fine tuning before I invest in it rather than just wing it.

 

[3DDave]

What happens to the loose chain? The path length for the chain changes during the stroke so some place the chain will go slack, leaving a lot of hysteresis for non-trivial amounts of eccentricity.

 

[MikeHalloran]

This seems the sort of thing you could do with a bead chain or a dial cord.
Not so much, with a roller chain.


Mike Halloran
Pembroke Pines, FL, USA

 

[gruntguru]

Assuming the chain is paralell to the force and the motion of the tray, the lever arm will be the vertical displacement from the shaft axis to the lowest point on the pitch circle of the sprocket.

je suis charlie

 

[slmechanis]

Alright thats what I thought, thanks Grunt

 

[3DDave]

The chain won't be parallel except at one point in the rotation. The displacement to the side as it changes from that will change the path length; but the chain isn't elastic enough to account for it.