Understanding Screw or Worm Gears

Post Reply
User avatar
Site Admin
Posts: 30
Joined: Sat Mar 10, 2018 3:25 pm
Location: Fort Gatlin, Francis Langhorne Dade Territory

Understanding Screw or Worm Gears

Post by colSteve » Sun Mar 11, 2018 4:37 pm

Normal gears are lined up so the tooth on one gear pushes directly on the tooth on another gear. Screw gears slide along the tooth of the gear they are driving. A normal gear is like lifting a motorcycle straight up. Or like lifting it with a pry bar for leverage, when the gear being pushed is larger than the gear pushing it. A screw gear is like dragging a motorcycle up a ramp. Except the motorcycle has no wheels.

Normal gears are easy to understand, like a lot of levers all connected to a central point. One lever for each tooth. When the gear is bigger, the distance from the tooth to the center is longer. This is like having a longer pry bar, or a wrench or bolt cutters with a longer handle. It makes it easier for the smaller gear to spin the larger gear. But you also have to push twice as far, so the larger gear does not spin as fast. The torque basically doubles, and the speed is cut in half.


Notice there is also a little bit of teeth sliding in and out against teeth on the other gear. And both gears have bearings, so the shafts have to spin against the bearings. So some energy is wasted in things rubbing against each other. This is why the torque does not quite double when you add a gear to cut the speed in half. And the more gears you need to get the desired output speed and torque, the more weight you add, and cost to build it.


A worm gear is like a ramp wrapped around a shaft. As the shaft spins, the ridges on the left drag against the teeth on the gear on the right. Even if the ramp were flat, so the gear on the right did not spin at all, some energy is lost just dragging. The steeper the ridges, the steeper the ramp. But dragging something along a ramp that rises slightly is not much harder than dragging it along a flat surface. The torque is the energy that is left over after the energy spent just on dragging. So if the angle of the ramp doubles, it gets a little harder and the torque goes down a little. But the torque is not cut in half. And the speed does not double.

When you drag something, it gets a little easier once you get it moving. When it is sitting still, you have to give it an extra jerk to pull it loose and get it going. If you can't overcome that initial stick, then the result is the same on a flat surface or inclined ramp: zero. Suppose making the ramp half as steep is what it takes for you to be able to overcome that initial stick. The speed of the gear on the right won't be cut in half. It will go from zero speed to actually moving. Making the screw half as steep won't double your torque once the gear is moving. But it could more than double the power of the screw to get moving at all, from being stalled.

Suppose you need more power to overcome a screw gear that can't move the load at all. In this case, making the screw half as steep might double the size of the load you can get moving. It might double the stall torque, even if it does not double the spinning torque. The higher the gear ratio, the flatter the ramp, the less change in spinning torque from making the ramp flatter. So going from 40 to 80 gear ratio, spinning torque might only go up 20%. But going from 1 to 2 gear ratio, spinning torque might go up 80%. But stall torque might go up more than 20%, perhaps 50% going from 40 to 80. But at a lower gear ratio, there is not as big a difference between stall and spinning torque. Going from 1 to 2 gear ratio, stall torque might also go up almost 80%.

Below are some links that show screw gear torque and friction on a ramp. The main takeaway, is there are too many variables to guess torque and stall torque of a screw assembly with much confidence. Changes in lubrication or gear wear could make a large difference. So you have to rely on the data provided by the manufacturer, and then try it out and see if it does what you want.

https://www.engineersedge.com/gears/scr ... lation.htm
http://www.rushgears.com/tech-tools/wor ... and-torque
https://evmc2.wordpress.com/2014/07/07/ ... tor-types/
http://thecraftycanvas.com/library/onli ... em-solver/

Post Reply