On Thu, Sep 10, 2015, at 04:24 PM, Tom Easterday wrote: > > > On Sep 10, 2015, at 1:58 PM, John Kasunich <[email protected]> wrote: > > > …<a very long well thought out list of questions and suggestions that > ultimately pointed to the issue>... > > It was a good suggestion to forget about G33 and run the spindle my itself. > Even at max rpm (3000) we still have the same noisy spindle-speed-in signal. > So to make a long story short, the source of the variation in the > spindle-speed-in > is the VFD (which Peter Wallace essentially pointed out as well). But this > variation > is there at all speeds from 50rpm to 3000rpm. If I spin the spindle by hand > and > watch that signal it looks smooth as it should - the encoder is fine and > there is > no mechanical problem. The second I spin with the VFD I get the ripple in > the > signal. Also if I allow it to coast to a stop as John suggested the ripple > stops > as it coasts.
There are two possibilities: 1) The VFD is actually making the speed of the spindle wiggle around, due to some instability or something. Honestly I think this is less likely, especially if it appears to wiggle more-or-less the same at 50RPM and 3000RPM. To make the speed vary by several RPM several times per second means quite a bit of torque pulsation. If that is really happening, you want to solve it at the source. The torque pulsations and varying speed can't be good for surface finish. 2) The VFD is not actually changing the spindle speed, but it is inducing electrical noise into the encoder signals that is seen by the Mesa card as a speed variation. Again, you want to fix that at the source. > A question: We would like to perhaps smooth the velocity input that the > trajectory > planner uses for G33 commands. What signal does it use for the velocity in > spindle > coordinated motion? It isn’t motion.spindle-speed-in - we set that to zero > and G33 > happily did it’s thing. Is it a smoothed version of the ddt of > motion.spindle-revs? > We can’t filter motion.spindle-revs because that gets reset to zero but… It really isn't about velocity, it's about position. The position command to the Z axis tracks the position of the spindle (motion.spindle-revs). Turn off the spindle while it is in the middle of a G33 move (might need a finer pitch than your knurling, so that the spindle can coast to a stop before the G33 finishes). Then turn the chuck by hand. You should see the Z axis tracking the chuck position - I think it even tracks if you turn the chuck backwards, but I'm not 100% sure of that. -- John Kasunich [email protected] ------------------------------------------------------------------------------ Monitor Your Dynamic Infrastructure at Any Scale With Datadog! Get real-time metrics from all of your servers, apps and tools in one place. SourceForge users - Click here to start your Free Trial of Datadog now! http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140 _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
