Kirk Wallace wrote: > If it makes any difference, I was cutting air, but I can imagine that > there are other hardware load sources. Ok, well, that excuses the spindle. It seems that the peak error was .0005", so I wouldn't get too concerned about it. You could look to see if the INPUT to the Z drive had all those rapid fluctuations, or if it was only in the PID error value. If the input was basically a straight line, then it is just the noise of the PID following at that particular velocity. If it IS in the commanded velocity to the Z axis, then it means at that particular spindle RPM, the encoder is suffering from the quantizing effect of the encoder. We are working on something to make a better velocity estimate from encoders, might have it working by the Fest, but I'm not sure that will be of any benefit to a spindle-sync'ed movement. > I should do more testing on > cutting real threads, but my guess is that what I have is good enough, > and with some effort could get better. I certainly can use much less > lead in, with higher RPM. > It looked pretty good to me. I really have to try this out sometime soon. I think I already have the code on my backup CNC computer.
Jon ------------------------------------------------------------------------------ Crystal Reports - New Free Runtime and 30 Day Trial Check out the new simplified licensign option that enables unlimited royalty-free distribution of the report engine for externally facing server and web deployment. http://p.sf.net/sfu/businessobjects _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
