Sensorless Vector Control of Motor

Status
Not open for further replies.
I have a 400hp motor that is running in sensorless vector control. After the original tuning and commissioning, the system was put into service. The process required roughly a 70% load for the motor. After a few days, the motor load for the same process conditions jumped to 98%. If the drive is put into V/Hz mode, the current returns to 70%. Load readings were taken from the drive keypad. Any ideas on potential causes?
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
I have a 400hp motor that is running in sensorless vector control. After the original tuning and commissioning, the system was put into service. The process required roughly a 70% load for the motor. After a few days, the motor load for the same process conditions jumped to 98%. If the drive is put into V/Hz mode, the current returns to 70%. Load readings were taken from the drive keypad. Any ideas on potential causes?
When you say "load", do you mean kW or current?
 

mike_kilroy

Senior Member
Location
United States
sensorless mode should not pull more current than v/hz mode of course.

So it appears to be a clear indication the sensorless mode is unstable. your facts say marginally so, since it ran a few days ok.

the 'auto tune' in a lot of drives is iterative; i.e., it gets better each time you do it. so perhaps you could autotune again to tweak it in better.

if that does not solve it, perhaps you can turn some gains down real time and find the culprit. it is often either current loop gain too high or velocity loop. so if it was me, I would start with I loop prop lower, then if no change, put back as it was, then go to V loop integral then V loop proportional. this method should solve your instability.
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
sensorless mode should not pull more current than v/hz mode of course.

So it appears to be a clear indication the sensorless mode is unstable. your facts say marginally so, since it ran a few days ok.

the 'auto tune' in a lot of drives is iterative; i.e., it gets better each time you do it. so perhaps you could autotune again to tweak it in better.

if that does not solve it, perhaps you can turn some gains down real time and find the culprit. it is often either current loop gain too high or velocity loop. so if it was me, I would start with I loop prop lower, then if no change, put back as it was, then go to V loop integral then V loop proportional. this method should solve your instability.
I agree, and just want to add that some of the inconsistency is, at this time of year, sometimes attributable to the motor temperature when the initial autotune procedure is performed, and especially if you did a static tune (non-rotating). For vector control to work, the mP in the drive is comparing a created mathematical motor model against a feedback loop of performance. Flawed model = flawed performance. I always recommend to people that if you are going to do a static autotune when a motor is cold, run it in V/Hz for a few minutes before performing the Autotune, because slight differences in circuit resistance can make an amplified difference in motor performance under vector control. That's one reason why, when you select a rotating autotune, it often takes a few minutes of running the motor at full speed for the procedure to take place, they are reducing the effects of winding temperature on the readings. I'd be willing to bet that if you re-do the Autotune procedure after it has been running that you will see the current stabilize.
 
Status
Not open for further replies.
Top