1. Home
  2. ⏣ General Metalworking
  3. Rotational accuracy required for gear cutting?

Rotational accuracy required for gear cutting?

I need to cut a small gear but was wondering how much angular accuracy is needed.

I have a 3rd axis on a stepper drive available & know even at 2000 steps/rev that is not good enough to just put the gear blank on the stepper motor & try to cut it, so I want to add the stepper to a rotary table.

The rotary tables I looked at have input ratios from 16:1 to 96:1 & I was wondering what would give me the required accuracy to make a usable gear?

The specific gear I am working on now is a 43T, 0.5 MOD, 0.886" OD so is fairly small.

Thanks, MikeB

Reply to
BQ340
Loading thread data ...

The higher the worm ratio, the finer the resolution.

Reply to
PrecisionmachinisT

43 teeth? Are you making a Merlin?

AFAIK the accuracy required rises with the speed and power they transfer, the speed due to rotational inertia. For a model you can cut the tooth gaps oversized.

I had a big problem with uneven residual play in the worm wheel and had to manually preload and lock the spindle for each cut. jsw

Reply to
Jim Wilkins

It is a gear for a cheap telescope drive, must be a bastard size. That is why I am making it, no one seems to sell it.

The old nylon gear is totally smeared, took me all day to identify it by measuring the mating gears & mocking up the gear pitch diameters & shaft spacing dimensions in Autocad as a double-check. (Good thing there are free gear profile DXF generators!)

I expect the cheaper rotary tables to be sloppy, but like you did, that can be worked around.

I will probably have to cut a couple of them before I get it right, but it will be nice to have the capability to be able to make gears for other future repairs too.

MikeB

Reply to
BQ340

What is the a minimum ratio that would be acceptable based on 2000 steps/rev?

The table I am considering has as 36:1 ratio.

MikeB

Reply to
BQ340

You might be able to properly identify the last parameter, the pressure angle, from an unsmeared gear.

Reply to
Jim Wilkins

Isn't there anything from Boston Gear that would work?

Reply to
Michael A. Terrell

Depends on the gear accuracy class you trying to achieve.

Reply to
PrecisionmachinisT

If you use a 36:1 dividing head each step is 36/43 of an input shaft turn, or 1674.4186 steps. You should distribute the over and under steps evenly so the last tooth space isn't too near or far from the first. A non-cumulative error of about one part in 3000 should be well below your machining tolerance on a gear 0.886" in diameter.

5/12 = 0.416667.

If you enter the data into a spreadsheet you can manually modify the step allocation to see how it comes out at the end. jsw

Reply to
Jim Wilkins

Good, thanks!

I'll get that table then.

I was reading up on my stepper drive & it claims it can mico-step to

4000 steps/rev so if that will actually work with my motors it should cut that error in 1/2.

MikeB

Reply to
BQ340

1674 or 1675 turns of the stepper rotates the gear OD by 0.0647". The rounding error from half a step amounts to 1/3350 of that, or ~20 microinches, insignificant compared to cutting the teeth at the wrong pressure angle. How accurate are the other molded plastic gears? jsw
Reply to
Jim Wilkins

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.