Re: [CAD_CAM_EDM_DRO] Using linear encoders.
Posted by
Jon Elson
on 2004-03-18 22:40:21 UTC
Peter Reilley wrote:
back to the
computer so that a deadband roughly equal to the backlash can be set in the
software to stop the hunting.
A 1000
cycle/rev encoder gives 4000 counts per rev, if you count all transitions.
On a 5 TPI leadscrew, that is 20,000 counts/inch, or a resolution of .00005"
You won't get that resolution on a linear scale without a very expensive
interpolator. Most DRO scales are roughly .0002" (actually 5 uM), so
4 times greater resolution for roughly $100 - 150, while a standard 5 uM
30" scale will run $350 and up. Unless your ballscrew is of the
high-precision
type, however, the linear scale will be a good deal more accurate,
depending on
the accuracy of the screw.
Jon
>Can you use linear encoders in a servo system? This would be on aSure, you can use a linear scale. It is better to have the scale feed
>Bridgeport type machine with ball screws. I can see where any backlash,
>no mater how small, would result in a continual hunting for position.
>However, I know that there is at least one commercial CNC conversion system
>that uses the linear scales for feedback.
>
>
back to the
computer so that a deadband roughly equal to the backlash can be set in the
software to stop the hunting.
>Is this a reasonable thing to do or should I go with the rotary encodesYou will get a lot more resolution a lot cheaper with rotary encoders.
>mounted on the servo motors.
>
>
A 1000
cycle/rev encoder gives 4000 counts per rev, if you count all transitions.
On a 5 TPI leadscrew, that is 20,000 counts/inch, or a resolution of .00005"
You won't get that resolution on a linear scale without a very expensive
interpolator. Most DRO scales are roughly .0002" (actually 5 uM), so
4 times greater resolution for roughly $100 - 150, while a standard 5 uM
30" scale will run $350 and up. Unless your ballscrew is of the
high-precision
type, however, the linear scale will be a good deal more accurate,
depending on
the accuracy of the screw.
Jon
Discussion Thread
Peter Reilley
2004-03-18 10:40:56 UTC
Using linear encoders.
bull2002winkle
2004-03-18 12:05:23 UTC
Re: Using linear encoders.
Larry Ragan
2004-03-18 13:02:41 UTC
RE: [CAD_CAM_EDM_DRO] Re: Using linear encoders.
Jon Elson
2004-03-18 22:40:21 UTC
Re: [CAD_CAM_EDM_DRO] Using linear encoders.
bull2002winkle
2004-03-18 23:19:06 UTC
Re: Using linear encoders.