View Full Version : Backlash in rotary motion

02-08-2010, 02:20 PM
I know that ball screws are used to eliminate backlash in linear motion devices. I am wondering how backlash is delt with in rotary motion devices such as rotary tables that are used forth 4th axis on cnc machines.

02-08-2010, 02:44 PM
There are probably other devices, but a harmonic drive is one of the "no backlash" solutions for rotary motion. I put that in quotes because everything has backlash at some level, but it is extremely low for a harmonic drive.

I bought a couple of these for use on a 4th axis CNC project. They're getting more expensive on eBay since more people realize what they're good for.

There's also ways to set up your worm gears to be preloaded to reduce backlash. They have to be made very carefully for it to work well enough without adding a lot of friction though.



Rich Carlstedt
02-08-2010, 02:58 PM
Worm screws with a pre-load address the issue you mentioned.
Wear is not a consideration, as the "resolver" or "encoder" determines the location, ergo , wear never changes the position desired

The Artful Bodger
02-08-2010, 03:12 PM
One means is to use duplicate gears preloaded agaist each other by some sort of adjustment or ever just a strong spring.

02-08-2010, 03:24 PM
Recirculating balls have been used in auto steering boxes so it shows that it could be done for tables as well. I don't have enough experience with different makes to know if it has be applied there.

02-08-2010, 04:22 PM
There are three components of "back-lash" in a rotary table:
- end-play in the worm (adjustable);
- the quality of the worm>wheel meshing (adjustable); and
- the table bearings (not normally adjustable).

If "cutter climbing" is a real problem, put some "drag" on the table with a "braking" device or use the table clamps.

Otherwise use a less aggressive "climb" milling or switch to conventional milling.

Worm end-play and worm>wheel mesh adjustment do the job most times.