Big Clumsy Cam Followers don't work for me

[Bengy2]

I am working on a design involving a cam and follower. The nearest approximation of the shape of the cam is an ellipse and the follower continuously rotates around the inside of this ellipse (it is an internal cam). The problem I have is that there is very little space to accommodate the follower, or to be more exact the motion envelope of the follower. I was thinking if I could line the internal surface of the Cam with ball bearings, or equivalent, then I wouold not have to use a follower with incorporated bearings thus allowing the assembly to be more compact. My question is has anyone ever seen anything like this done before. The problems I am having are, amongst others,:

a)How to maintain the ball bearings in position around the internal surface of the Cam

b)How to design a follower that will ride smoothly across the ball bearings and not create a "bumpy" effect

Any replies would be gratefully received

 

[aamoroso]

depending on loads you may be able to use a Permaglide strip attached to the walls of the path. INA makes them

INA Permaglide

Fill what's empty. Empty what's full. And scratch where it itches.

 

[Bengy2]

Thanks for the reply aamoroso, but yes the loads will be substantial and I would have thought it would have to be borne by a mechanical bearing of sorts. I will check out your suggestion anyway. Any further replies from anyone would be most welcome.

 

[zekeman]

What kind of motion are you expecting out of the follower?
It may be possible to get this done with an external cam, rather than the slot cam you are using.

 

[Bengy2]

Thanks for the reply zekeman. I am trying to obtain a uniform rate of motion of the reciprocating yoke of a Variable Scotch Yoke Mechanism(obviously the yoke will not have uniform velocity as it is has to stop and reverse direction at its limits). I have spent a huge amount of time trying to solve this problem, including searching this site, with nothing yet that is satisfactory. I believe that the only solution that suits my situation is to condition the input to the reciprocating crank to give the desired rate of reciprocating motion. The simplest and most reliable way to do this, I feel, is by a cam. My problem is that space is very tight and I need to keep the motion of the follower within the area of the slot cam so an external cam is not ideal, I have considered this already. It should also be noted that loading of the follower on the wall of the cam slot will occur on the external wall only during operation, there is no need for a "slot". Maybe this information might spark a brainwave in someone out there.

 

[Barry1961]

When stuck with a "impossible" design problem try to stay open to totally different methods of doing it. Often the goals can be redefined to make the problem very simple. Make sure you have not defined a want as a need.

Do you really need the exact move profile you have in your design?

Can you move the cam to somewhere that is not so space limited?

An eccentric ball bearings like harmonic drives uses. With this the outer race would flex and provide the linear motion.

A linear ballscrew actuator and program the move profile.

A rack and pinion with eccentric drive.

A gear motor with crank and program the move profile.

Use two small cam followers.

Run a roller chain/kevlar belt around two sprockets in an ellipse and connect to chain/belt.

Run steel on bronze and call it a wear item.

Barry1961

 

[zekeman]

If you want to get help on this or any other problem, you must be very explicit of your specs and the constraints. We are not mindreaders and and cannot be helpful from vague descriptions of the problem.If words are not forthcoming, perhaps a sketch could be forwarded.
As a general comment to others who do not get satisfactory answers here,try to be clearer in posing your problem at the very least. I know engineers sometimes have difficulty in this area, but do the best you can.

 

[DavidLeech]

Once again many thanks to all for your contributions, I will take some time to consider the options. Thanks zekeman for your advice re posting questions, I will make a better attempt if I need to post the same query again.

Bengy2

 

[zekeman]

Looking over your earlier post, since you want a controlled oscillation of a scotch yoke, did you consider making the slot of the scotch yoke a cam-like slot so that you could engage it with an ordinary oscillatory crank? Rothbart's cam book, "Cams",Wiley, 1956 shows such a mechanism on P 16.
Good luck on this.

 

[Bengy2]

zekeman, I have not had a chance to look at the book you suggest yet but the Scotch Yoke I am working with needs to be variable - the length of stroke needs to be adjustable. I have considered introducing a cam like slot in the yoke but this will only work if the length of stroke remained fixed.

Barry1961 has made the point that when stuck with a "impossible" design problem try to stay open to totally different methods of doing it. I believe I have done this but still find no alternative which is adequate. I will produce a diagram and post it here in the coming days (maybe under a new thread). Maybe some fresh ideas will come from this.

Thanks to all for you help,
Bengy2

 

[unclesyd]

Food for thought.

http://www.cunningham-ind.com/index.htm

 

[Sparweb]

Fitting within your original ideas, you could try needle bearings instead of balls - it would be more compact and "smooth". But I'm with Barry: try thinking a bit more out of the box.

Steven Fahey, CET