• State Answer Suggested
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 81 subscribers
  • Views 767 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

ACIM Identification problems with Motorware 14

Richard Chapman
Intellectual 640 points
Other Parts Discussed in Thread: MOTORWARE

Hi Chris and all

We have Motorware 14  running on our own "28069M" hardware. We have successfully identified several PMSM motors and I think we understand the process reasonably well.

We have been wrestling with identification of a 1 HP 220 volt 3 phase ACIM. I will give some background and symptoms and then I also have some specific questions.

We have "fiddled" with most of the user.h parameters in order to get the identification to run - but whatever we do - we don't get very "usable" numbers from the process. In particular - we have varied the PWM frequency, the ISR frequency and the following:

#define USER_MOTOR_RES_EST_CURRENT (1.0)

#define USER_MOTOR_FLUX_EST_FREQ_Hz (5.0)

In general - we find that results are more repeatable with higher ISR and PWM frequencies. Still-  whatever we do - the numbers we get have rather high magnetization current numbers (around 3 amps) - and the motor (if it spins at all) will not reach full speed. If we manually reduce the magnetization current to around 2 amps - the motor behaves much better. In fact we have empirically adjusted the magnetization current until we have something that works - but this isn't a very satisfactory solution.

A couple of specific questions:

1) The comments and documents tell us that for ACIM we need to enter a calculated number for:

#define USER_MOTOR_RATED_FLUX (0.8165*230.0/60.0)

However - the identification process appears to ignore whatever we put here and makes up its own number. The numbers it produces are usually lower (around 2 V/hz) than we would expect from the recommended calculation (around 3 V/hz) ... 

Has the ACIM motor ID process changed so it no longer needs us to input the calculated rated flux?

2) The motor is intended for a fan. We understand we should do the identification under no load - with the fan removed . If we do this - the numbers we get will not allow the motor to spin at all. If we do the identification with the fan attached - we get numbers that allow the motor to spin - though not at full speed. Can anyone suggest why this should be the case.

3) Can anyone suggest  why our magnetization numbers are always too high - even when wee ID with no fan attached?

 

Thanks.

 

Richard.

 

 

 

 

 

  • 0
    Intellectual 640 points

    Hi again Chris...

    Further to the above problem report...

    We have been assuming we can identify motors - including ACIM motors - using LAB09 code - which is the code we use most. Could this be our problem? We assumed that the identification code was common to all labs. I note that there was a problem with ACIM identification in Motorware 11 and earlier. Is it possible this was fixed only in LAB02 code - or not fixed in all labs?

    Thanks Chris.

    Richard.

  • 0 in reply to Richard Chapman
    TI__Expert 5180 points

    Yes please use lab 2b to identify ACIM.

    The flux value is the target flux, assuming that the motor was designed to produce rated torque at a rated voltage of 230 VAC and rated frequency of 60 Hz. Please adjust those two value for your motor. Since this is a fan, the rated torque could be at a much higher frequency, so you might end up with something like:

    #define USER_MOTOR_RATED_FLUX (0.8165*220.0/100.0)

    The reason why the magnetizing current keeps increasing is because internally the code is trying to reach the value set in USER_MOTOR_RATED_FLUX. The reason why could be because lab 2b needs to be used, since it sets some specific things for ACIM identification.

    So please rerun motor Id with lab 2b, and also share your user.h, and current plots while identifying the motor are always useful to get some hints.

    -Jorge

  • 0 in reply to Jorge Zambada
    Intellectual 640 points

    Hi Jorge

    Thanks for the advice and sorry for the late reply.
    We did find LAB02B produced more consistent and repeatable results for identifying ACIM motors.

    In future - we will use LAB02B based code for ACIM identification.

    Let us know if you intend to implement the LAB02B ACIM identification code in other labs at some time in the future.

    Richard.