Phyiscal gear generation is a serious business. I'm curious how smooth or self-leveling a RepRap'ed gear will be without some postprocessing. You may see some serious binding and it'll almost definitely make a racket. I know Forrest has a program and is going to attempt a gear in HDPE.
Now, mind you, I've been living in the Land of the Giant Gears for the last two months (big project at work, doing 80+ hours a week). They have two types of machines that generate gears. One is a reciprocating blade cutter (best known is the MAAG name). the other has a worm gear that's been segregated into cutting blades. Both work by moving their cutter from the top to the bottom of the workpiece. The MAAG needs about a hundred strokes per tooth and perhaps four passes to finalize the gear. A rotary hobber needs two somewhat slower passes (roughing and finishing). Still a lot faster. It can't do single piece herringbone gears, though.
Now, I can't remember the web site, but astronomers who build their own telescopes regularly make their own worm gear sets. Some use a tap to cut the driven gear via a standard tap. I remember one that used a section of allthread with slots cut in it as a rotary cutter. He had a metal lathe, clamped the allthread in the chuck and made a freewheeling rotary holder for an aluminum blank. Just turn it on, engage the cutter in the material, and push it in a little bit for each rotation until you're fully engaged. It's somewhat important in high accuracy applications to use a section of the same allthread as you plan to drive it with as sizes can vary somewhat.
I'm working on a little math to determine if it would be possible to easily angle the allthread to achieve spur gears. Otherwise, some accurate Z positioning would allow for some easily duplicated helical gears.A rough blank could be made on a RepRap (possibly with some form of post-smoothing for the outer surface), or otherwise cut from pre-extruded solid plastic stock. My initial calculations based off the dimensions of a 3/4"-10 allthread make me think it'll be more trouble than its worth to attempt spur gears with this method.
The most critical component of this method is actually accurate blank diameter. Modern gearmaking machines positively drive the blank, so if the diameter is a little off, it's not a great concern. If you actually drive the blank with the cutter (helical/worm gear method), your external dimension must be accurate or you'll double thread the blank. Say, you're trying to make a 40 tooth gear with my 3/4"-10 allthread helical cutter. You'll need a 4" circumference blank.
Hmmm... time to fire up Alibre...
[Edit]
Actually, found a good example among the telescopers!
[
www3.telus.net]
A RepRap gear would probably make a great near-cut blank, and a tap system would clean it up sufficiently. The only difference I'd make is putting a vertical height adjustment system in. This would allow you to change your center position of the cutter and clean a worm gear out into a helical.
Edited 1 time(s). Last edit at 09/15/2007 07:11PM by SOI Sentinel.
