For what ever reason, am I'm sure other people's machines are like this as well, by backlash is different in one direction than the other.  In my case, I need .0018 comp in the +y, and .0032 in the -y.
Differential bl compensation would be easy to implement and very useful!
Thanks,
Eric

Differential bl compensation would be easy to implement and very useful!
You would think that is the case. Ask Greg who worked on backlash for the SS for a few years.

If the backlash you note is the same/ repeatable  for the most of the travel then i am thinking that your thrust bearing / assembly
and mounting  need to be looked at closely and worked on. Get rid of the problem since in the long run you will be a happier camper.

RICH

Hi Rich,
Since Mach3 already has the algorithm in place for BL compensation,  I think the only logic that would be needed is to use one BL compensation variable if transitioning in one direction, and the other BL compensation variable in the other direction.   Once that logic is in place, all that is left is the GUI change, and adding the two new variables to the .xml input/output. 
If I had the source code, I would love to add it for you!

Regarding the SS, that is why I went back to the parallel port.  I got tired waiting for Greg to fix his implementation.

I'm using a LMS mini mill.  My backlash is coming from a bad nut/screw match.  I'm going to purchase larger balls for it to see if I can reduce the backlash some more.  Measured, the BL is pretty close from both directions, but in use, I need .0043" comp in Y+ and .001" in Y- to get a smooth circle.
Everything it tight, and I tossed the rubber spiders for the lovejoys and machined new ones out of delrin so there is no compression going on there.  I can't fathom why I get need different numbers in use.


Regards,
Eric Sanders
http://www.compufoil.com