Worm Drive - Gear ratio really accurate?

Thread Starter


Joined Oct 17, 2009
I'm using a worm drive on a stepper motor and the cog has 74 teeth.

As everything suggests, a worm drive gear ratio purely the number of teeth on the cog to 1. (74:1). So should require 74 turns of the input shaft to get 1 revolution of the output shaft.

Now here comes the strange bit....

My worm drive requires 75.25 turns to do one full revolution.

Is there any reason for this? pitch angle perhaps?


Joined Nov 12, 2008
Hi Bluebrakes,

Could you provide a simple drawing of your setup?

You are correct, a worm gear to, say, a spur gear is 1 to # of spur teeth. This assumes the drive motor is turning the worm gear and the spur gear is being turned by the worm gear (direct contact). I don't know which gear is being driven by what, whether the gears in question are turning a shaft attached to other gears on the opposite end, etc., thus the request for a drawing.

If this is a simple motor-connected-to-worm-gear-turning-spur-gear and the revolutions are off, then it could be the pitch angles of the gears are not a match or the gears are not spaced apart correctly (look up pitch circles).


Joined Sep 30, 2009
If there is axial play in the worm shaft/gear it will show up as extra turns when changing direction. It will 'climb the gear' until the play is taken up. Gear boxes with out thrust bearings built-in are notorious for this. With the worm wheel out of the box there should be only a small amount of axial play in the worm gear. Around .001-.002 inch would be good. Any more than this will show up in the free-play of the gears.


Joined Dec 8, 2009
Is this a true worm and worm gear set-up, or is the
worm driving a spur or helical gear?

If you have a worm turning a spur or helical gear, the pitch on the
gear (Circular Pitch) probably does not match the "lead" (Axial Pitch)
of the worm. Therefore the "gear" will not advance exactly
one tooth for every revolution of the worm. This is common
in non-critical and lower cost assemblies.

A true matched worm and worm gear combination will not
have this problem.

Information available on the web, or find a copy of the
"Machinery's Handbook".

A few accurate measurements of the two members involved
will probably identify the pitch and lead of the gear and worm.
After identifying the two components, the exact ratio could
be determined.
Last edited:


Joined Feb 11, 2008
A stepper motor does 5 full phase cycles per rev, (or 200 steps/rev in full step mode).

If you have driven it with n-1 phase cycles you would be driving it with only 49 cycles but counting that as 50.

The result would be (74/49)*50 = so that it would *appear* to have taken about 75.5 motor revs to get a full worm rotation.

Check your software that is counting motor revs if you have a n-1 error. :)