Hi everyone,

I think I have found a suitable solution and/or compromise.

Ned, thanks for your reply. I would especially appreciate your comments
& thoughts on the following, since you have designed cam mechanisms.

After Bobs reply, I did not think I had enough space to have the
cam-follower roller "rotate with" the shaft, & put a bearing on each
side of the shaft. I had thought of this before, but the available
space is so small, I just did not think there would be a stock bushing
small enough.

However, I found a self lubricating plastic bushing from www.igus.com
(part # GSM-0203-03), 2 mm ID X 3.5 mm OD X 3 mm long. I do have "just
enough" space to put one of these miniature bushings on each side of
the shaft. I used the online computer program at the Igus site, their
"expert system" under "online tools" to calculate bushing life under
the most severe conditions, and even with shock loads and "edge
loading" selected in the program, they showed excellent life. Since the
cam-follower load is in the center of the two plastic bushings, the
maximum cam-follower load of 168 pounds is reduced to 84 pounds per
each of the plastic bushings. The bushings are only 0.32 cents each in
quantities of 500, so its cost effective.

I will still use some 3/16" OD, W-1 or D2 drill rod as a cam-follower
roller, but now, the roller ID will have a press fit on the 2 mm OD
shaft, so the roller rotates with the shaft. The shaft will then be
supported by a miniature Iglide self lubricating plastic bushing at
each end.

I think the cam-follower should roll well using this method, since the
Iglide bushings have very low friction. Plus, I still have a 3/16" OD
roller with a smaller shaft. This also eliminates the problem with the
roller having a thin wall, since the roller ID is now pressed onto the
2 mm OD shaft, and it's really like a solid roller having no wall.

With the previous method of just using a 3/16" OD hardened steel
bushing oscillating on a 1/8" OD hardened shaft, even if it had
acceptable wear, I just don't know how good it would actually "roll". I
tend to think it would not have reliably rolled well, and there would
have been sliding between the cam and roller OD.

I think this new method will work well, and seems to be about the best
solution I have found. However, I am interested what everyone thinks.
Since the cam is oscillated manually by hand and the speeds are slow,
using the new method where the cam-follower "rotates with" the shaft,
and the shaft is supported by a bushing at each end, do you think I can
get away with not hardening the drill-rod roller ? Perhaps I should
harden it anyway, in case there is still some small degree of sliding
between the cam and roller OD. I supose it's hard to get truly "pure
rolling action" 100% of the time.

I was thinking of possibly using some "belt dressing" on the cam curve
to increase friction between the Cam curve and roller OD, to insure the
roller always rolls in the shaft and there is no or very minimal
sliding between the cam and roller OD.

http://www.lpslabs.com/Products/Lubr...t_Dressing.asp

The belt dressing lasts for a long time on fast rotating & hot belts,
so perhaps it might last plenty long on a hand operated device that
pivots slow and never gets hot.

Does anyone have any other ideas for a friction coating on the cam
curve or roller OD that would last and handle high loads ? My machinist
suggested to "glass bead blast" the cam curve in increase traction
between the cam and roller OD.

Bob, thanks allot for your reply, because you got me thinking again of
having the cam-follower rotate with the shaft, & putting a bushing on
each side of the shaft. Then, when I was looking through the igus
catalog, I found the miniature bushing and put everything together. I
did not know they made bushings that small, that could support loads
14,310 PSI with a PV of 12,000 dry.

Thanks again to everyone that has replied, especially Ned and Bob. I
have so many constraints on this design, because of the limited space,
I was really starting to feel boxed in. It always helps to get feedback
from other people, and that's why forums like this are great.

Sincerely,
John