suppressing ringing on a short pulse

We have two AX1500s hooked to linear actuators from danaher motion.

We have noticed two issues.

The first is that there is a ~60Hz pulse of 10-12Vpp even after the actuator has reached its desired position. This is obviously the PID loop applying a small amount of power (as the pulse width is only a 0.5-1 microseconds) Is this an indication that I need to tighten my PID loop? Possibly some more hysteresis (in the form of lessening differential gain)?


Second issue: I can live with this pulse. It is not affecting the system as far as we can tell. However, the induction of actuators is causing ringing on this "PID pulse". The ringing spikes to 20V. This is causing problems with our other EMI sensitive components.

Since the pulse goes from 0-12V when a positive command is given, and 0 to -12V when a negative command is given, I am a little unsure on the circuit needed.

It is easy to suppress the ringing on outer edges of both pulses, but how do I suppress the noise about the 0 (without completely blocking the 0 to -12V pulses, or vice versa).

To phrase it differently, we can clip at 12V and -12V, but another method is needed for supressing the noise at the reference point.

Any pointers?

Thanks,

-steve

The 60Hz would be consitent with the controller's 16ms update rate. If you are running in position mode and there is pression on your actuator, it is normal that you would see some pulsing since the voltage would drop to 0 when destination is met. If at that point, something pushes back the actuator, then an error would appear, thus applying power at the next cycle. We do have a beta version of the firmware with a 4ms update rate that might make this pulsing less noticeable.
<P style="MARGIN: 0px">Regarding the spikes, they look kind of high in your case but we have no good recommendation on how to suppress them. A few suggestions: add snubber circuit (RC), keep M+ and M- wires closed together, twist the M+ and M- wires together, add shielding.

Thanks. I am still working on this problem, but it had slid down my task list.

I am going to look into using some sort of bi-directional suppression diode, possibly in conjunction with a RC filter.

I don't think changing the period to 4ms would help, as it is the ringing that causes our problems. I might be interested to see what the 4ms period would do to our control loop, however. I assume we can go back and forth between the versions?

I'll report back later