Gear spacing

[daniel_dsouza]

Hello CD,

Summer is almost upon us, and our team is gearing down for some R and R.

When designing gearboxes, I've read that the space between gears should be slightly larger than the sum of their radii. (Please correct me if I'm wrong)

So my question is, how much spacing have you found works? What factors does the spacing depend on? (number of teeth, radius, profile, etc.)

Thanks,
Daniel D.

[Thad House]

For C-C distance, the only things that matter are number of teeth and pitch. The formula is

(teeth1+teeth2)/(2*Pitch)

For FRC many gearboxes add .002-.003 to that, in order to increase tolerance. Because our gearboxes see such low runtimes that is ok.


Here is a great thread with some more information
http://www.chiefdelphi.com/forums/sh...ad.php?t=95413

The main purpose of having a gap is to prevent jamming the gears together. This happens because the gear teeth aren't a constant pitch radius relative to the bore of the gear, the gear shaft isn't straight, the inside race of a ball or roller bearing is not in line with the axis, or the teeth of the gear are made incorrectly.

Mostly these errors should require only 0.010 inches total of extra spacing, but it depends on the gears and bearings. On larger gears with oil lubrication, it is useful to not trap the oil at the bottom of the teeth, causing pressure and heat buildup as the oil is compressed and ejected. In watch gears extra gap is to prevent tiny particles from getting trapped and jamming the gear train.

In many gears, the clearance is cut into the gear by making the teeth narrower, so the optimum distance is the sum of the pitch radii.

[Mk.32]

Free calculator that does it for you. http://wcproducts.net/how-to-gears/

[Michael Blake]

(N1+N2)/(2P)+.003" is all you need to know.

--Michael Blake

[Adam.garcia]

I have designed quite a few gearboxes, two of which have been manufactured by our team.

The general rule of thumb is summarized on WCProduct's Website:

Originally Posted by RC

For a single stage gearbox (custom): +.000 - +.001 backlash
For any other gearboxes: +.002 - +.003 is recommended.

For our gearbox spacing, we used this gearbox calculator which RC made:

http://wcproducts.net/how-to-gears/

Hope this helps!

[DampRobot]

Like the other posters, I've always used .003" adder. However, this has caused some division on my team. Others believe that it's important to have dead spacing, while I contend that it increases efficiency, and runs better in FRC applications.

Also, we've had mixed results using a .003" adder on worm gears. Proceed with caution, if you're not using spur gears and want to put in an adder.

A search for spur gear design leads to more info. This is from a search for spur gear kinematics:

Fellows Gear shaper the art of Generating Gears
http://www.youtube.com/watch?v=A3X8cuJKyns


I'm guessing if you are looking for spacing, you should probably look more into gear design - it's got a lot of formulas, but it isn't very difficult.

[Jeffy]

Originally Posted by sst.thad

For C-C distance, the only things that matter are number of teeth and pitch. The formula is

(teeth1+teeth2)/(2*Pitch)

For FRC many gearboxes add .002-.003 to that, in order to increase tolerance. Because our gearboxes see such low runtimes that is ok.


Here is a great thread with some more information
http://www.chiefdelphi.com/forums/sh...ad.php?t=95413

Because it hasn't been explicitly stated:

# of Teeth*pitch is your pitch diameter.

If you use the two pitch radii (diameter/2) then add the tolerance (as stated above) of .002-.003", you will have a nice smooth gearbox.

http://www.youtube.com/watch?v=3LdlSAN1yks to amuse gear makers.

I forgot - the other reason for backlash (the extra distance) is to allow for the expansion of gears due to heating. This is not likely a problem for FRC, but in high-power gear boxes, particularly with oil lubrication, the gears heat and expand more than the gear-boxes. This would be especially bad for aluminum gears in a steel housing. The other case can also be true - if the gearbox is operated significantly below room temperature and the gear case is aluminum and the gears are steel, the distance between the bores in the case will decrease more than the pitch radii decrease.

The benefit to interference or zero clearance fits is that when the gear train reverses, there is no hysteresis between the input and the output. Some times anti-backlash gears are used with split spring-loaded gears to allow for tolerance without allowing backlash.

Otherwise, tight fits just load up the bearings and scrape up the gear tooth faces, while increasing the power losses in the drive train.

[JamesTerm]

Originally Posted by three_d_dave

http://www.youtube.com/watch?v=3LdlSAN1yks to amuse gear makers.

Thanks for providing these links. I hope many teams will come here and see this. I would like to ask a couple of follow-up questions. If teeth get stripped from gears from rigorous driving, can one assume the spacing is too far out? Also anyone care to share some guide as to how much tensile strength must be used to hold gears in place to avoid stripping or slipping?

[thefro526]

There have been a handful of threads on this subject before, and have proven that +.003 spacing is pretty much fool proof on most FRC applications.

Sometimes you'll encounter a situation where you're getting off the beaten path at which point, you'll need to deviate from the tried and true method of +.003 spacing. Usually, this is in higher reduction gearboxes where you're trying to minimize backlash. At this point you'll need to look at all of the part fits, including how well the gear fits on the shaft, how well the shaft fits into the bearings, etc. Depending on what exactly is going on in your mechanism, you could find something like .015" of backlash just in the fit of your gear bore - at which point you definitely don't want to add anything to the gear spacing.

Originally Posted by DampRobot

Also, we've had mixed results using a .003" adder on worm gears. Proceed with caution, if you're not using spur gears and want to put in an adder.

We build a worm box mid-season to cut down on the backlash in our system and to have a passive braking mechanism and ran exact C-C on between the worm and worm wheel shafts. In our case, we were using a harden steel worm with a bronze worm gear (16DP if anyone cares) and the gear ended up 'cutting' the wheel as the box broke in. The fit between the worm and worm wheel was fantastic after the break in period* and there was no noticeable backlash in that stage of the transmission.

*I highly suggest running in any custom worm box on a bench - especially if it's a single start worm. IIRC, we ran ours via a drill for something like 10-15 minutes continuously before we started to notice gains in efficiency... Meaning that we'd have more than likely never actually broken it in during the season.

[daniel_dsouza]

That was fast.

Originally Posted by Michael Blake

(N1+N2)/(2P)+.003" is all you need to know.

--Michael Blake

I guess this pretty much sums this up.

Thanks everyone. Now to build a gearbox.