• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 81 subscribers
  • Views 223 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.

TMS320F28052F: PMSM motor fan system low speed noise

Tan Ma
Prodigy 20 points
Part Number: TMS320F28052F
Other Parts Discussed in Thread: MOTORWARE, TMS320F28069F

Hello Sir or Madam,

This is Tan Ma, I am an electrical engineer. 

We have designed a motor drive based on TMS320F28052F to control a 60 Poles PMSM motor fan system (HVLS fans).

Right now the issue is my company is using the motorware 13, and at the low speed (10-25 rpm) the motor and fan systems have a grinding noise. I have tried several ways to eliminate those noise, such as adding deadtime compensation, decreasing the deadtime, modifying the current and speed loop PI gains, increasing the current and voltage feedback measurement resistors to increase the accuracy, modifying the PWM switching frequency and so on. But the noise is still there.

My firmware is developed base on the instaspin_foc labs for hvkit_rev1p1 development kit. Last week, I tried the motorware 18 firmware and used TMS320F28069F DSP. I just applied the lab 10a on the hvkit_rev1p1, and there is no noise at all. I have double checked the motor parameters and settings to make sure it is the same. 

After that, I have compared the current feedback waveforms. In closed loop mode, the TI hvkit_rev1p1 current feedback waveform is better than my drive. But it open loop mode (voltage/hertz) mode, my drive has a better current feedback signal. Therefore, I don't think it is the quality of the feedback measuring signals.

So right not I am not clear that I couldn't eliminate the noise because of motorware version differences (V18 vs V13) or because of different DSP (TMS320F28052F (fixed point) VS TMS320F28069F (floating point)). 

 

Thank you very much and wait for you guys' reply. Any help will be appreciated. 

Dr. Tan Ma

 

  • 0
    TI__Guru** 109810 points

    Do you use the same PWM and control frequency for motor control with your own board or TI High voltage kit?

    Do you check the phase voltage sensing signals? And what pole frequency is the voltage sensing circuit on your own board?

    What the USER_MAX_VS_MAG_PU is used in your project? And do you implement the over-modulation?

  • 0 in reply to Yanming Luo
    Prodigy 20 points

    Hello Yanming, thanks for the reply.  I have tested both TMS320F28052F and TMS320F28069F on hvkit_rev1p1. Both used the lab 10a. TMS320F28052F couldn't start the motor even they the setting is same.

    Do you use the same PWM and control frequency for motor control with your own board or TI High voltage kit? 

    Yes, I use the same PWM frequency and control frequency.

    Do you check the phase voltage sensing signals? And what pole frequency is the voltage sensing circuit on your own board?

    I checked the voltage sensing signals, the pole frequency is : USER_VOLTAGE_FILTER_POLE_Hz  (353).

    What the USER_MAX_VS_MAG_PU is used in your project? And do you implement the over-modulation?

    Since the noise happens in the very low speed (10-30 rpm), over-modulation will influence the performance?  USER_MAX_VS_MAG_PU  setting:  (1.0)

  • 0 in reply to Tan Ma
    TI__Guru** 109810 points
    Tan Ma said:
    Yes, I use the same PWM frequency and control frequency.

    What PWM and control ISR frequency are you using in the F28052F project?

    Did you run the motor identification on the F28052F board? Did it have the same or very similar identified motor parameters as using a high-voltage kit? If not, the issue should be from the current or voltage sensing signals on your own board.

    Did you measure the phase waveform with an oscilloscope? Any difference between your own board and high voltage kit?

  • 0 in reply to Yanming Luo
    Prodigy 20 points

    --What PWM and control ISR frequency are you using in the F28052F project?

    The PWM and ISR control frequency are both 15K.

    --Did you run the motor identification on the F28052F board? Did it have the same or very similar identified motor parameters as using a high-voltage kit? If not, the issue should be from the current or voltage sensing signals on your own board.

    I did the motor identification on the F28052F board. I used the same motor parameters on both drives. 

    --Did you measure the phase waveform with an oscilloscope? Any difference between your own board and high voltage kit?

    I have measured the phase waveform, as I mentioned before: when tested with FOC and closed loop, high voltage kit has better current waveform. But when tested with open loop, my board had better waveform. Therefore, I don't think it is the issue of the measurement circuit. 

    One more thing I need to mention: On the high voltage kit, I have tested 28069F and 28054F. By using 28054F, I couldn't start this 60 poles motor, but I could run a 10 poles pmsm motor (smaller HP). Then I tried another 28 poles PMSM motor with the correct motor parameter, and again I couldn't start the motor. So is there any thing I need to pay attention when using 28054F to control PMSM motor with more poles?   BTW, the 60 poles and 28 poles motors could be controlled by my board, and the 60 poles motor just have low speed grinding noise issue which I want to solve.

  • 0 in reply to Tan Ma
    TI__Guru** 109810 points

    Do you use the embedded PGA of F28054F for current sensing? 1AxisMtrPfc-F2805x-[R2.1].pdf If yes, you can refer to attached schematic. And you might tune the speed and current PI controller using different Kp&Ki at low speed. If possible, improve the voltage sampling resolution by changing the voltage sensing circuit.