Howard Chan wrote: > Dear all, > Thank you for them help! I solved the problem. I use John Kasunich > suggestion to check the encoders. It is really 8000 count/rev, I use it > to calculate the INPUT_SCALE and using HALmeter and SANYO DENKI > softeware to check the encoder pulse in same value. > > Yet there are another problem, how can I find the P value in PID setting. > I tried any value to set it but I dont know is it correct so I capture a > picture, please help me to check it. > http://myweb.polyu.edu.hk/~icwfchan/pid.png > P = 25.000 > I = 70.000 > D = 0.015 > FF1 = 0.091 > OK, you may need more P, I seems VERY high, I would expect a value more like 2 with your P of 25. D may need more or less, depending on the drive. Those spikes at the beginning and end are the worst error, and are not caused by EMC, in my experience. You may need to check the tuning within the drive to see if you can eliminate that. The error during the accel/decel can be helped with a VERY SMALL amount of FF2. Sometimes just .001 to .005 is sufficient. If your machine is set up in mm units, then the error trace there is .002mm per vertical division, so that isn't too bad.
Even though you are not using my servo controller system, the page I wrote on PID tuning might be helpful http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?PWM_Servo_Amplifiers The specific hal pins will have different names, and not all encoder interfaces provide the "delta" pin, which is change in raw encoder counts per servo sample cycle, which I find to be useful both for triggering halscope and for seeing where in the move (accel, coast, decel) we are. Jon ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users