They sound smooth enough at PID=100, but that was low speed, what do they sound like at full operating speed?
On Wed, 2 Feb 2022 at 14:19, Thaddeus Waldner <thadw...@gmail.com> wrote: > > copy-paste from terminal: > uname -a > Linux linuxcnc 4.19.71-rt24-v7l+ #1 SMP PREEMPT RT Fri Jan 1 21:15:16 GMT > 2021 armv7l GNU/Linux > As I said, this is the sd card image straight from the LinuxCNC website. I > wanted to ask, is that the best route to go? I’m not terribly impressed > with the latency numbers, though they appear to work when I don’t run > anything else (i.e. LinuxCNC doesn’t abort). I plan to get this machine > into service (operated by others) as soon as possible. > > Here is a short video that illustrates the problem. > https://youtu.be/DdYYpdC51Vs > > I did try with P=0, as suggested by Sam, and it was also a big improvement > over the default configuration. It may have had a bit more of an aggressive > whine, but stable, to my ears. > > When I have more time, I could use my scope to run an fft on the step > signal to get data that is better than “sounds like bad bearings”. > > Configs are attached. They are mostly straight from the pncconf wizard, > with the z axis removed. I may have fat-fingered the servo period in this > copy, just so you know. > > On Wed, Feb 2, 2022 at 4:32 AM Sam Sokolik <samco...@gmail.com> wrote: > > Could you post your configs? Could your hal functions be out of order? > I > > run pid steppers with RPI and don't notice this. > > > > Sam > > > > On Wed, Feb 2, 2022, 3:01 AM andy pugh <bodge...@gmail.com> wrote: > > > > > On Wed, 2 Feb 2022 at 07:44, Thaddeus Waldner <thadw...@gmail.com> > wrote: > > > > > > > 2) why is servo thread jitter even making into the stepper pulse > train > > > with a Mesa Ethernet card and no communication errors? Do I have > anything > > > misconfigured? > > > > > > The idea of using a PID is that it compensates for the fact that the > > > servo thread updates don't get to the card exactly every servo period. > > > Sometimes the feedback - calculate - send of the required step rate > > > will run early, sometimes late. > > > > > > I am hoping that PCW might pop up to explain exactly why the PID helps > > > here. It's not something I have paid that much attention to as I > > > rarely use steppers. > > > > > > It might be interesting to try a non-pid config. (or just set P to > > > zero which has about the same effect) > > > > > > -- > > > atp > > > "A motorcycle is a bicycle with a pandemonium attachment and is > > > designed for the especial use of mechanical geniuses, daredevils and > > > lunatics." > > > — George Fitch, Atlanta Constitution Newspaper, 1912 > > > > > > > > > _______________________________________________ > > > Emc-users mailing list > > > Emc-users@lists.sourceforge.net > > > https://lists.sourceforge.net/lists/listinfo/emc-users > > > > > > > _______________________________________________ > > Emc-users mailing list > > Emc-users@lists.sourceforge.net > > https://lists.sourceforge.net/lists/listinfo/emc-users > > _______________________________________________ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users > _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
