Positive Q Current Creates Negative Speed

[derekcook]

Hello, 

I have been running fine for around 9 months now using KMS. I got a new motor, and applying positive Q current to the new motor produces negative speed. Negative Q current produces positive speed. However, if I flip the sin pins on the resolver, then the resolver does not match the motor phasing, and the motor does not move. 

Temporarily, I multiplied the q reference current that goes into the current block by -1. This resolves my problem. However, is there a setting or better way to do this in KMS? It seems KMS expects positive Q current to produce positive speed, and if it does not, it just keeps asking for more Q current.

[linestream-adam]

Derek,

When you say negative speed, can you be more specific. Is the speed feedback in KMS negative or is it positive and the motor is rotating in the opposite direction you expect? Is KMS able to regulate the negative speed or is it running away?

First initial guess is that the alignment between the resolver and the motor isn't the same as your previous motors. Have you tried swapping two motor phases instead of the resolver phases?

[derekcook]

Thanks for the response Philip and Adam. 

The motor is spinning in the direction I expect, but the resolver is counting down instead of up, causing the speed feedback from the encoder block to be negative. KMS is not able to regulate the speed. It seems KMS just keeps asking for more q current. We've tried swapping both the motor and the resolver phases and have not been able to get the combination where the motor is spinning the right direction with positive speed. This is why I multiplied Q current going into the current block by -1 to correct the problem.

[linestream-adam]

Ok, it sounds like there is an alignment issue between the resolver and the motor. Or as you said the resolver is counting down instead of up.

Sounds like you need to modify the resolver block so that it is generating the electrical angle the way KMS expects. KMS expects that the electrical angle is an LQ24 variable that varies between 0 and 1 for each pole pair where 0 is aligned with the pole itself.

[derekcook]

Thanks for the help Adam! This would fix the issue as well as inverting the Q current.

[philip_drake]

Hello Derek, 

When I've encountered the request for positive speed resulting in the motor spinning in the wrong direction I have had to reverse two of the motor poles.  I remember that during parameter measurement with a sensor input,  that a pop up suggestion from KMS ask you to do just that. 

I'm afraid I cannot help you with a software solution. linestream-adamreynolds‌ may be able to assist. 

Best Regards,

Philip