[jcolley]

Well, previous discussion on the subject has piqued my interest in timing belts.

For a fourth (A) axis, would it be feasable to use a timing belt arrangement with a couple of reduction stages to provide rotary motion, yet still have sufficient strength to prevent a largish and unbalanced rotary stage from moving under the force of a cutting tool?

I have looked at harmonic drives and like the backlash specs, but for a larger axis, you really need a huge (read expensive) gear set.

I like the worm wheel method, but to provide low backlash, either you get rid of the gear clearance or use a dual worm or gear in an anti-backlash arrangement.

A belt seems like it could provide the backlash I'm after, but I would need about a 100:1 reduction ratio (in a couple of stages most likely).

Any thoughts from you proponents of belts?

Jim

[andy_ck87028]

An interesting question and I must admit I am purely answering to keep the thread going as I am sure there are people with some useful input on this.

I would love to know what people come up with.

Andy

[turmite]

Hi,

I designed a 4th axis and had a friend make it for me. I used the belt system and if you are turning anything very far out from center you will get movement....well, at least I did and still do, but to a lesser degree. I eventually had to go to a 17:1 ratio and that is all I could build for or I would have it 50:1. I make rifle stocks and take fairly heavy cuts on the roughing passes. If the bit makes any kind of a plunge on the toe of the buttstck you can see the stock twist a little, but it has not cause me any realy problem.

I use a 300 oz/in stepper and if I had a 450 oz it would probably be ok.

Mike

[Tormach]

A belt would be a troublesome solution if you're looking to the high forces of metal cutting. Minimum would be 25mm Kevlar. It might work for woodworking. You then also get into the addition backlash and compression added by the pinion bearings.

A simple worm gear will get you about 30 arc seconds of backlash if it's well cut. That's less than 0.001 on an 8" diameter. Haas is no better than 15 arc seconds. A worm gear is also self locking, so the individual impact forces of each cutting tooth are not seen at the drive motor. If you use a more efficient transmission like a belt and gearbox or a harmonic drive, the cutting will effect the motor driver. If you use a high line count encoder on a servo in combination with a belt drive, I'd be willing to bet that you could see encoder counts of disturbance with each flute impact when cutting at lower speeds. A good AC servo driver like Indramat or Rockwell will allow you to watch scope style plots of servo error on a setup like that.