AC servo - encoder timing.

[KCress]

I have a Yaskawa AC, (brushless DC motor), servo drive. The bearings failed after years of service. The the drive was disassembled, bearings replaced, and the drive put back together. This includes pressing the encoder back onto the shaft.

Sweet! Everything looks great and feels nice.

Power it up and... The shaft turns about one revolution and shuts down. Does this everytime. We stick an identical motor/encoder unit on and it runs fine.

I'm getting a feeling that some sort of alighnment is required between the encoder and the motor position so the motor controller knows how to sequence the power to the motor. However no indexing marks were ever noted or made, and there seems to be no easy way to adjust the alighment. Everything gets pressed on and off the shaft.

Does this alignment issue ring true to you?
Any bench top suggestions or proceedure you know of?

[Al_The_Man]

You have to detect or find out what feedback/commutation system is used, if it is hall effect simulated pulses on the encoder, I have done it for these kind of systems using a double beam scope.
Al.

[KCress]

Hi Al.

To be even reasonably smooth and powerful BLDC motors need position information fed back to their controllers.

Normally it's something like a Hall effect switch(s). But I can see that if I were designing a system that had to have an accurate position encoder on it for other reasons, I'd try to use that same encoder to provide my motor position too.


So you're agreeing that the encoder does have to be "aligned"?

[Al_The_Man]

Depends whether the motor is AC sinusoidal or BLDC, often both motors use the same method, except AC sometimes use it momentarily at start up.
The modern method of simulating hall effect devices is the pulses are included as separate tracks on the encoder disk.
I have outlined the method in previous posts, I can post it again if you need to align using hall type outputs.
Al.

[KCress]

That was very useful! I did not realize that was what those other tracks are.

I would greatly appreciate you reposting it, or directing me to where it was last posted. Or just some key words to search with as I searched much on this question before starting this thread.

[duklatd]

Hi


I see you have a YASKAWA motor . please ley me know what kind of motor to check on the feed bak unit. or call yaskawa @ 1-800-yaskawa and ask for Duke

[Al_The_Man]

Originally Posted by KCress That was very useful! I did not realize that was what those other tracks are. I would greatly appreciate you reposting it, or directing me to where it was last posted. Or just some key words to search with as I searched much on this question before starting this thread.

I could not find the previous thread, but basically I fed the encoder with 5v to enable the commutation signals to be output, specifically Hall A, then connect a 1k resistor to each of the three stator leads and connect all other ends of the resistors to one point and this will be the reference point for one of the 'scope channels, IOW, one channel of the scope is connected to the hall A output and the other channel connected to A phase of the motor, next I couple a DC motor to the AC motor shaft in order to back feed it at a fairly constant rpm, this generates an AC waveform as per the PDF.
The encoder has to be rotated until Hall A coincides with Phase A.
Al.

[KCress]

Thanks AL.

I can manage that.