On 04/06/2015 12:29 PM, Gene Heskett wrote: > On Monday 06 April 2015 14:12:31 John Alexander Stewart wrote: >> Hi all; >> >> Finally fixed the Oldham couplers for my X/Y axes on my latest build, >> and was measuring backlash for inserting in the LinuxCNC ini file. >> >> This is for a "G0704" style of mill conversion, using one of those >> kits from the old KeilingCNC place. >> >> X/Y/Z is 0.1mm, 0.24mm 0.1mm or, 2.5 thou, 9 thou, 2.5 thou. > The 2.5 thou is a bit loose, bigger balls in the nut perhaps? > > The 9 thou Y indicates either the thrust bearing at the solidly anchored > end of the screw is duff, or perhaps not properly assembled. I was able > to fix one of those in a cheap 16mmx5 Chinese screw on my toy lathes z > by padding the outer ring separation with some half thou alu foil shims. > Several of them stacked. Tricky to cut too. Seems to be holding up so > far. > >> Is this considered "ok" or "out of range"? >> >> I've put these figures in my mill ini file, and it's currently running >> the X and Y axes back and forth; will measure it again after a bit of >> wearing in. >> >> Any thoughts/comments? >> >> (as an aside, Andy Pugh's feedback for spindle soft-gearing works a >> treat - thank you again Andy) >> >> John. > Let us know what you found John, it helps the next person with a similar > problem. > > Cheers, Gene Heskett I've been dealing with a loose Y axis on my cinci contourmaster; this machine is set up so one can turn either the ballscrew or the servo motor to adjust position. The so-called lock on the servo end was loose and so the backlash was self-adjusting;out to unusable amounts. Ugh. I'm in the process of a dual fix ... complete rebuild the mount for the ballscrew so I can lock it down tight.... and add a 10" 5 um glass scale for reference. For now I will connect just the glass scale with it's attendant tuning problems (just not quite enough counts) but the future holds a config where P and D are fed from a 40K count/in encoderon the ballscrew and I is driven by the glass scale in like manner to Stu's big machine.
Some years ago I was asked to fab a bunch of blocks for a US First robotics team. Set them up and ran as a batch. When finished I miked them to see how I'd done. SD was 0.001 but the dimensions were all large by about 2 thou. This on a machine (mazak) with 0.0025 backlash on both X and Y. Appropriate cutter comp might have landed these in the right range if it made a difference. ;-) Just personal experience. Dave ------------------------------------------------------------------------------ BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT Develop your own process in accordance with the BPMN 2 standard Learn Process modeling best practices with Bonita BPM through live exercises http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual- event?utm_ source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_campaign=VA_SF _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
