That sounds like you could have to much Kd. Try reducing it. Do you have
an oscilloscope? An oscilloscope really helps when tuning servos. As the
UHU doesn't have an analog output you would have to probe the motor
drive signals. Put a simple R-C filter in line with the probe to filter
out the 19.5kHz PWM frequency.
Are your encoders mounted directly to the motors?
Les
On 02/06/2021 07:20, John Dammeyer wrote:
I really don't know if this is the right forum for this question but in working
with the mill over the last few months every once in a while the DC Brushed
Servo Y axis makes this horrible noise like a stepper motor at the harmonic
instability. Usually I've just cycled power on the servo and then everything
is fine again.
Today it wouldn't go away. I added a bunch of oil onto the lead screw and it seemed to suppress it a bit.
The back story is this. I bought two Brushed DC Servos 10 years ago and built up 4 of the HP_UHU kits. The US Digital encoders were crap and I lost position by a little bit on each move. I changed to Henrik Olsson's PIC awesome based replacement module for the UHU controller and replaced the US Digital with CUI encoders. Noise problem gone. No missing pulses. Repeatable operation.
One Servo got hotter than the other though. Some resistance and current measurement and I discovered that although the servos were supposed to be the same they weren't. At least not internally. By this time they were more than 8 years old so no chance of replacement. I put the hotter motor on the Y axis which has to move the X table and X Table Pivot assembly and tuned it for smooth operation.
Start-up when cold appears to make the noise show up more easily. I think it's just a servo tuning thing. I'm still running ACME screws and bronze nuts. Conversion to ball screws is a year or so away.
So the question is: if there is a sort of resonance noise at around 30 IPM it could be a loose or worn acme nut since extra oil seemed to make it less. But is a tuning parameter I could 'tweak' to prevent this? Here are the parameters I can adjust. This was from a file in the DC Servo folder. It may not be how they are set at the moment but I'm wondering what parameter might make a slight difference for this sort of oscillation.
Suggestions?
Thanks
John Dammeyer
****************************
(P)-Kp: 3000 (3000)
(I)-Ki: 25 (25)
(D)-Kd: 25000 (25000)
(T)-Ti: 1
(V)-V.ff: 0
(A)-A.ff: 2000
(X)-Multiplier: 1
(L)-Servo Rate: 1500Hz
(E)-Error limit: 150
(C)-Max PWM: 740
(F)-Error scale: 0
(O)-Output Offset: 0
(Y)-Encoder filter: 78kHz
(J)-AD Region: 2
(K)-AD Scale: 256
(B)-Integrator Limit: 450
(W)-Fault input filter: 1
(S)-Save settings
-Peak error (abs): 15
"ELS! Nothing else works as well for your Lathe"
Automation Artisans Inc.
www dot autoartisans dot com
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users
