[Scot808]

I have some spare parts that might potentially be able to be used to make a light duty 4th axis. I have an R8 spindle shaft and could fabricate a housing and probably find an adapter for a 3-4" chuck. I realize this thing won't be very rigid. My question is if a timing belt setup is a viable means to drive it. I have some HTD5 pulleys with 9mm wide belts. I think I can only get about a 4:1 reduction. This already isn't sounding too promising. However, any advice, thoughts, etc. would be appreciated.
Thanks!

[handlewanker]

Hi, a 4th axis is a valuable item no matter what the reduction ….if it works.

You should be able to get pulleys to make it at least 1:6 reduction.

Depending on how good you make the belt drive.....9mm is not all that rigid when it comes to tensioning it to reduce backlash......probably a wider belt would be better......also a lot depends on what you want to use it for so the stepper type will have to be an important feature.

You can use deep groove ball races (simpler design) but angular contacts would be better (much more complicated).

I wouldn't bother with the R8 feature as it's only capable of holding specific diams like cutters etc.....a 3 jaw chuck would probably cover all your needs...….spindle height is important to allow the jaws to protrude as the chuck rotates without hitting the table and the hole through the spindle needs to be as big as you can get it to allow stock size through etc.
Ian.

[spumco]

Try a double-reduction setup. Use a jackshaft and build it in to the spindle housing:

- Motor has small pulley driving jackshaft
- Jackshaft has large pulley and a small pulley
- Jackshaft small pulley drives large pulley on 4th axis spindle.
- 3:1 on the motor to jackshaft, and 4:1 on the spindle nets you 12:1 total reduction.

A 1.8 degree stepper will give .15 degrees per step (9 arc-minutes) at full step, and .015 at 10x microstepping.

At 10x microstepping and a 2" radius (for your 4" chuck), that means a (theoretical) resolution of 0.0005." Not bad.

Assuming your motor is a stepper with a max useable RPM of 1000, your A-axis rapids will be 83RPM. That's way fast for an indexer.

The HTD pulleys reportedly have some backlash but I'm not enough of a power transmission guru to determine how much - if any - you'll see.

While the HTD belt & pulley you have can handle fairly high loads, they are likely going to be 'springy' compared to 15/20/25mm wide belts regardless of tooth profile or pitch. The springy part means you'll be limited in rigidity if you are drilling or milling in such a way that the operation produces rotational force on the 4th axis.

A clamp or brake on the 4th spindle would cure that, but then you've got to cobble up something else.

A 20mm or 25mm wide double-reduction pulley set, using a GT, GT2, S5M, or AT5 tooth profile will be significantly more rigid and (probably) with less backlash than your HTD 5M-9 setup. And this is where a stepper will work well as the holding torque at zero or low speed is probably better than a similar sized servo with no built-in brake.

If you want to see a monster of a belt-driven 4th axis, have a look at what this guy's done:

WebPageMain


Oddly enough, I'm in the middle of a 4th axis build myself. My chinese worm-drive 6" 4th just has too much backlash and it's not adjustable - with my trunion table I can only use it like a motorized indexer that I still have to indicate after every move. So I've scored a big hollow-shaft harmonic drive off ebay and am building an offset belt-driven version. (Belt drive so I can have a hollow spindle for longer parts)

I was thinking about a belt-drive like you're discussing, but I didn't want to deal with making a spindle and precision bearing pockets. Good luck, and post up as you make progress.

-Ralph