Terry Neilson wrote:
> Jon,
> That was a good explanation of the possible problems that can occur when
> using linear scales for position feedback. What I still don't understand
> though, is how the knee milling machines in the shop that have linear
> scales, continue to run every day with very good accuracy, and no problems
> with servo hunting issues.
Well, ballscrews can have a very long life.  If properly lubed and swarf 
is kept out, then
they can run for decades before developing significant backlash.  So, I 
didn't want to say that only a
brand-new $5000 super-precision ballscrew will work.  Not at all.  But, 
a lot of machines I hear of
being retrofitted do have significant backlash in the ballscrews.
>  I would think that a rotary encoder reading the
> rotary travel of a ball screw would cause a cumulative inaccuracy that
> increases with increasing travel of the table.
No, certainly NOT cumulative!  How could that be?  The ballscrew doesn't 
have a cumulative
error, and the encoder reads the rotation fo the screw.  if the 
screw/nut assembly has backlash,
then the encoder will not know about that, and there will be "lost 
motion" when the direction reverses.
This "lost motion" will be equal to the amount of backlash, but that is 
not going to be cumulative,
it will be repeatable every time you come back to the same position from 
the same direction.

Or, are you talking about something completely different than the 
backlash issue?  Then, yes, there
is a cumulative error in the leadscrew, but how big is that?  It can be 
quite small on ground ballscrews,
or fairly bad on cheap rolled screws.  Most commercial machine tools 
will have ground screws that
are pretty accurate.  Much better than 0.001" per foot is not usually 
warranted as the ways aren't
straight enough to be able to tell the difference.  If you are worrying 
over drilling two holes to within .001"
location and over a foot apart, you really should not be thinking of 
using a used, retrofitted machine tool.
>  A linear scale is measuring
> the actual movement,so in theory should be more accurate especially at
> greater distance.
> Can you tune out errors in ball screws using EMC, and get repeatable
> accuracy for most of the length of travel?
>   
Yes, EMC has a machine error compensation table.  You have to be able to 
home the machine
for this to work.  You then map out the position error in each 
direction, typically using gage blocks
and an indicator, and then enter the deviation in a table.  But, this 
doesn't address servo loop
control and stability.  That is an entirely different issue.

Note that while the linear scale is quite accurate measuring linear 
movement, the machine ways
may not be moving in perfectly straight lines, or the axes may not be 
perfectly orthogonal to each other.
Stuart Stevenson has been working on a kinematics module to make these 
corrections, and I gather he
has it working.  He has the precision alignment instruments to make the 
measurements, most of us do
not.


Jon

------------------------------------------------------------------------------
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to