On Thursday, November 17, 2011 12:05:10 PM [email protected] did opine: > > Or maybe a Sherline or MaxNC. > > I feel sure I am wrong but from reading this mailing list I get the > impression that servo drive is superior to steppers By superior I mean > more accurate
This is likely where you will get differing opinions. Unless you have preloaded ball screws and other parts of the driving mechanics similarly backlash-free, and have mapped the screws for errors, the backlash inherent in most machines will IMO make this argument moot. My backlash ridden acme screws, with split nuts for lash control, is a much larger source of error than the servo/stepper argument can ever be. For instance, with 20 tpi screws, a microstepping driver currently running at 16x, I have an inch divided by 20=0.050" per turn of the screw. My motors are the usual suspects, 400 step/rev motors, so a full step then is that .050"/400=0.000125" for a full motor step. Now assuming a perfect mapping between steps (like that will ever happen), and operating at 16 microsteps, then I have a theoretical accuracy of 7.8125e-06" in the xy directions! Can servos match it? Probably with careful setup using very expensive measurement tools. Unless making clock and watch parts like Ian is, likely overkill. Can my machine actually do it? (insert laugh track here) Of course not. But it has milled the seats for the ball bearings hidden inside the nut carrier for my z axis, and which fit well enough that I drilled a hole calculated to hit the greasing hole in those shielded bearings, through which about an oz of grease has been injected, and the grease that escaped all came out through the gaps in the bearings shield. So one could claim that it can and has milled pretty close to a true circle. The phrase that comes to mind is one an ex bro-in-law was fond of, it was good enough for the girls he went with. ;) Backlash comp tweaking means I can get something under a thou for a few hours wear after a few hours of tweaking backlash settings in the ini file, with a restart of emc each time I edit a backlash figure, a very time consuming effort when the backlash motion calc isn't split, but all to one side of the position, and according to Murphy, always the wrong side according to my .0005" dial indicator, so I have to re-zero each time I restart. :) It would be helpful if the .ini files value was /2, and applied as correction to both directions so a dial indicator could be setup and centered, and it then remained so after the restart. Or, perhaps the position file could have a + or - sign appended to value record the direction it was last moved, and the recovered position based on that? One of my wishes is that at some point in 2.5, emc's axis display front end grows a pulldown menu that will allow backlash to be tweaked in real time and saved in a file for reload at rebooting emc time. This would be nice as would the ability to map the screw for both faces of the screw effectively giving us a dynamic backlash compensating means along with the error map. This would be even handier for those with ball screws that are showing signs of wear or maybe have been over-stressed while moving. That of course is assuming that one has the means to make such accurate measurements. As a hobbyist, the cost of such measurement tools far exceeds my mostly SS funded budget. For me, when drilling holes at x intervals, the screws I have _are_ my long distance standard. And it has worked very well for me, so far, the mistakes that have been made, and there have been quite a few, are all mine. ;-) I guess my point of all this rambling drivel is, that when looking for more accuracy, you should attack the biggest errors first, particularly when they are hundreds of times larger than this particular argument probably is. I would posit that even with good ball screws, the mechanical errors will exceed the electronic errors rather handily, from thermal growth in the machine if for no other reason. I suppose one could take the machines temperature and setup yet another table it rereads when the temp change is noted by emc. Whether that is worth it again depends on the scale of what you are doing. Potentially quite valuable for Stewart and his monster cinci, but IMO serious overkill for carving watch parts. Or a trigger guard bow for a BP rifle out of solid brass, the last job I did with mine. > Has anyone outfitted a Sherline with servos > > Richard Cheers, Gene -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) My web page: <http://coyoteden.dyndns-free.com:85/gene> Yow! And then we could sit on the hoods of cars at stop lights! ------------------------------------------------------------------------------ All the data continuously generated in your IT infrastructure contains a definitive record of customers, application performance, security threats, fraudulent activity, and more. Splunk takes this data and makes sense of it. IT sense. And common sense. http://p.sf.net/sfu/splunk-novd2d _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
