On Wed, 2009-04-08 at 05:29 +0000, Tom wrote: > Jon Elson <el...@...> writes: > > > > > > What is your count resolution (in the ini file, that is INPUT_SCALE)? > > Just guessing an encoder directly mounted on a 5 TPI leadscrew, that is > > 5 * 2000 = 10,000 counts/inch or 0.0001" per count. That is definitely > > on the low side. I am using 20,000 counts/inch on the X and Y of my > > Bridgeport, and it works, but doesn't give the tightest control. But, > > these were the encoders I could afford at the time (1000 cycle/rev, 4000 > > counts/rev). > > > > If you are not using FF1 and FF2, you should really explore them, they > > can help IMMENSELY. If you are, then increased encoder resolution > > should help. I know that I can tune my minimill, with the 128,000 > > counts/inch, to insanely small following error. Far less than the real > > errors (backlash, screw pitch errors, etc.) > > > > Jon > > Hi Jon, > > You guessed right, my INPUT_SCALE value is 10160 since my screws are metric. > > Yes, I use both FF1 & FF2. > > I am already looking into retrofitting the highest resolution encoders I can > afford. I think you are right about needing all the counts you can get for > good > accuracy. My ballscrews all have less than .001" backlash. > > Thanks! > Tom
Just in case it might be useful, I am using the USDigital 10000 pulse/rev. (2500 cpr) encoders on my HNC lathe: http://www.wallacecompany.com/cnc_lathe/HNC/00001-1a.jpg http://www.wallacecompany.com/cnc_lathe/HNC/00006-1a.jpg Current pricing seems to be about $50 for each axis if you make your own disk hubs and sensor mounts. I have my encoders on the end of my 5TPI ballscrews, giving 50000 pulses/1". ------------ Kirk http://www.wallacecompany.com/machine_shop/ ------------------------------------------------------------------------------ This SF.net email is sponsored by: High Quality Requirements in a Collaborative Environment. Download a free trial of Rational Requirements Composer Now! http://p.sf.net/sfu/www-ibm-com _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
