Noise-Resolver vs Encoder

Looking for expert advice. We have a turning center that we replaced the 35 HP Dc Spindle motor and drive out with a 40 HP AC motor and drive. We are experiencing noise that presents it self as a kind of a chatter in the gear train of the spindle at start and stop as well as slower RPM's. We are currently using resolver feedback and am wondering if I should have encoder feed back (less noisy) ????????

Fell free to chime in and thank's in advance

If the resolver feedback is coming into linuxcnc you should be able to check the
magnitude of the noise by watching (or HAL-Scoping) the appropriate HAL pin
(also check for linearity around a full spindle turn)

This may also be a tuning issue (if there's a PID loop running the spindle for closed loop velocity or position control)

If the resolver is only tracking the spindle rotation and not used to close a
spindle velocity /position feedback loop then it cannot cause noise in the spindle gears

SRT wrote:

Looking for expert advice. We have a turning center that we replaced the 35 HP Dc Spindle motor and drive out with a 40 HP AC motor and drive. We are experiencing noise that presents it self as a kind of a chatter in the gear train of the spindle at start and stop as well as slower RPM's. We are currently using resolver feedback and am wondering if I should have encoder feed back (less noisy) ????????

Fell free to chime in and thank's in advance

There's no specific reason a resolver should be more noisy than an encoder. A good resolver
converter can be quite immune to noise, but proper shielding of the cables is
important. Probably 3 individually shielded cables should be used, with the shields
only grounded at the resolver reader end.

It is also possible the gear chatter has nothing at all to do with the resolver, and
everything to do with the cogging of the induction motor. The presence of rotor
slots shorting bars in the rotor make most induction motors cog, and it gets
worse at lower speeds. it will be absolutely worst when the VFD applies DC
braking right at the end of slowing down.

Jon

I'm not having any problem with noise to the resolver just wondering about noise from the resolver affecting the drive. We have been very careful wit our resolver wiring and routing and we are still getting the chatter in the gear train. It could very well be the cogging that Jon was talking about. We are not using a PID on the spindle.

The other problem we have is not being able to enable the sensor less vector with encoder feed back feature of our motor and our drive. When we tried that feature the chatter and or cogging was unbearable.

I ran into a very similar situation with a yaskawa motor and drive on a mill. In this case, the drive had weak components on the firing circuit for one phase.

Are the motor and drive matched?
Is the motor an induction motor or permenent magnent?
What brand of drive are you using? Model?


Pm motors run awful with a standard vfd . If you are using a drive that supports pm motors, the mode for pm must be enabled, then you have a bunch of pm only paremeters to set. Most drives that will run a pm motor do not support any kind of feedback,unless its say, a matched Fanuc .

We are using an Automation Direct Durapulse series drive, PN: GS3-4040, which supports sensorless vector w/encoder feedback, with a 40hp Baldor Vector duty motor with encoder feedback.


Thanks

Rick

A couple of things come to mind.

I have experienced low speed cogging when the carrier frequency is set to low,the induction voltage constant is set improperly or the induction requirements of the motor dont match the drives capabilities.

You can raise the carrier frequency and see if it helps,there is a formula for figuring the voltage constant, I dont have it off the top of my head. Finally you can add a reactor or choke to balance the inductance requirements. I would think that both Baldor and Automation Directs tech support would be helpful.