XY-table design using linear motors & cross roller bearings 1

[nstcook]

Hello everybody!

Presently starting to design an XY table for short but fast strokes of a few mm at a time. The peak acceleration is to be about 120 m/s^2 with a 5 kg load. Using 6mm dia. crossed rollers. My major concern is dynamic stiffness of the table. Also Yaw and Pitch errors should be within 5 arc seconds over the total travel range of about 70 mm for each stage. The unusual part about this project is that the linear motor driving the bottom stage (X stage) is located directly below the stage and exerts a huge vertical magnetic force (~200 kg) on the stage, which would heavily preload the linear bearing. Any inputs in this matter would be greatly appreciated; especially as regards mounting, preloading from the side, appropriate materials etc.
Thermal management is also important since the motor is very close to the crossed roller rails and any differential expansion between the stationary rails and the moving rails on the X-plate would change the preload. Your comments and suggestions on these and other relevant issues are welcome.
Thank You!
-nstcook [sig][/sig]

 

[LTD]

Quick notes, I'm rushed but:

Graphite / BMI or Graphite / PEEK, or Graphite / PMR15 composite table.

Some graphite fiber is VERY good at conducting heat along the fiber... Expensive option, but your application doesn't sound cheap.

Graphite (carbon) composites have very very low thermal expansion (CTE) rates. Some can be made to be Zero.

Ceramic / graphite machine base? Again near zero CTE, so your heat problems minimize before you ever try to cool it.

Note: this approach minimizes variability TOO.

Ceramic bearings - Rollers are just barely available.
Ceramic balls - skidding at 12g's is not good, ceramics are lighter
Hollow rollers ! Litterally these are roller bearings where the rollers are tubes - low interia, interesting pre-load.

Seperate the table from the motor with a stand-off push-pull link, and have it in it's own rail-riding carriage?
This seperates the function of the table and the motor - but it makes your gadget longer.

If magnetic field is a problem, check out the interesting properties of bismiuth! Fun with physics. [sig][/sig]

 

[rgv]

What's this design for? Do you work for a semiconductor equipment company? Just asking... I'm familiar with wirebonders and diebonders. We have all sorts of XY tables. Anyway, if you're concerned with magnetic attraction but don't want to sacrifice stiffness, aluminum is always the choice. See you later. [sig][/sig]