Another Cam Grinder

[hanermo3]

Absolutely !
You are using geometric scaling.

All machines, more or less, use the same ideas, to get better precision, resolution, repeatability and accuracy.

The same, even better, is common and used on modern cnc machines.
The support guide blocks are relatively far away from each other.
Say 500 mm midpoint (== 1 m bed VMC, typical, 45-55 mm linear guides).
Each might have an error of say 0.005 mm max.
But the midpoint will have much less error, => 0.002 mm, especially over 4 blocks.
This is geometric averaging.
You trade size and volume for accuracy.

The longer your lever arm//error, the lesser your error at the controlled point aka cutting tip.
In practice, it gets better since mass works for You, by always preloading (bias) in the same direction.
This is one way how it is relatively easy to make more accurate components, given some time, area, money for components, and a bigger machine.

It is quite easy to make a flat 2-5x more accurate, than the guide you have.
Similar applies to everything, including screws, that need special techniques.
e. Offset cam followers made 1 micron accurate screws 50 years ago.
SIP machines and Moore jig bores were the famous ones.

At that time it was very hard, and very expensive .. due to measuring mostly.
Today, it is vastly easier.

I spent 12-14 years so far learning to make very accurate cnc screws.

I am of the opinion that the cams I am grinding are accurate within close tolerances because the use of the oversized master cam being made using CAD and CNC reduces any geometrical errors during the grinding by possibly a factor of 5.

As I said I am no cam expert, maybe somebody who knows this subject better than myself can enlighten me.

Peter J.

 

[Charles Lamont]

The accuracy of the master cam is only one aspect of the matter. The geometry of the mechanism is also critical to its faithful reproduction.

Not long ago I completed a rocking frame grinder, shown here:

http://www.homemodelenginemachinist.com/showthread.php?t=25362&page=7

In a purely mechanical cam grinder, the geometry of the mechanism is crucial. My grinder provides the correct geometry by using the principle of similarity. (For those whose maths may be a little rusty, that is things of identical shape but scaled to different sizes.)

Firstly the master cam is scaled up by a factor of 5, and works on a 5:1 ratio of arm length.

Secondly, the follower shoe for the master cam has a radius 5 times that of the grinding wheel.

Thirdly, obviously, the grinding wheel and follower are mounted at a 5:1 ratio from the rocker pivot.

Finally, and this is the crucial step that ensures the geometry remains correct, missed in several designs I have seen, the contact points ("top") of the follower and wheel are in a straight line with the pivot point. In my grinder this is achieved by setting the pivot point at the same height as the top of the wheel.
Any deviation from this straight line would result in the 5:1 arm length ratio being compromised, and probably other errors I have not bothered to think through.

The height of the follower is adjustable to provide the depth of cut, but it ends up the same height as the top of the wheel when the cam is to size.

This geometry is straightforward and fairly easy to understand, though also quite easy to misunderstand. It would be perfectly possible to design a rocking frame grinder that used a flat follower shoe, but, for accuracy, that would require the master cam profile to be modified in a way that I did not want to spend time getting my head round.

 

[bluejets]

The way it was overcome by George Punter was to make the "following mechanism" linear in motion.

The necessity came from the need to produce replacement complex cams from old motorcycle engines for which he has made a name for himself.

As I said, I not knocking the design. My grinder follows the same as yours and it is quite satisfactory for the work.

What I was pointing out for any who may come along later, that there is an alternative that will produce "exact shape" cams of "any size" from "any size" blank. The one George punter built will do exactly that.

 

[ICEpeter]

Hello Jeff (I assume that is your correct first name, if not I apologize),
As the saying goes, there is more than one way to skin a cat (Apology to the cat)

As long as the method and equipment used accomplishes the desired result regardless of the approach, its goal was reached and it can be considered a success. I wouldn't rate one or another approach inferior or superior in comparison because of the ingenuity of us model engineers we travel different roads to get to the same destination. Cheers.

Peter J.

 

[john_reese]

A truly beautiful job. Did you build the tool & cutter grinder?