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.

TMS320F28069: MotorID Failing with High DC Bus Voltage

Part Number: TMS320F28069

Hi,

I have some custom hardware based closely on HVKIT.

My motor is a BLDC, ~310VDC, Rs = 0.98R, Ls = 0.003H

If my DC bus voltage is about 100VDC I can ID the motor no issues, if my DC bus voltage is > ~100V then the ID fails with Rs_Ohm being some large number (1.0 e+28)

Some user.h settings:

USER_MOTOR_RES_EST_CURRENT = 1.0

USER_MOTOR_IND_EST_CURRENT -1.0

USER_MOTOR_MAX_CURRENT = 3.5

USER_MOTOR_FLUX_EST_FREQ_Hz = 10

USER_R_OVER_L_EST_FREQ_Hz = 300

USER_PWM_FREQ_kHz = 10

I feel the problem may be related to my slow 10KHz PWM, but I don't seem to have any success even if I change it to 15KHz.

Out of interest, this motor ID's on HVKIT.

Any suggestions.

Thanks

Stomp!

  • Was the motor failed to identify parameters on both customized board and TI HV-Kit, or only customized board? Did you have any current waveform to show the failed ID? What rated frequency and current is the motor? You might increase USER_MOTOR_FLUX_EST_FREQ_Hz and USER_MOTOR_RES_EST_CURRENT based on the motor specification.
  • Hi Yanming,

    Ahhhh.. I set some manual user parameters and I can spin the motor up to 4K RPM, only up to a DC Bus voltage of 100V (its a 310V motor).

    The current waveform shows spikes at ~130us intervals (twice PWM frequency?). The spikes I think are real, as I'm using 0.1in spaced scope probe.

    I think there is a minor issue somewhere on the MOSFET bridge side..

    The U Phase current waveform at 4K RPM:

    The U Phase voltage waveform at 4K RPM:

    The U Phase voltage waveform at 1K RPM (Note, motor still spins, but makes some funny noises).

    I'll make this as resolved, just because it now looks like a hardware issue on my side rather than a control software issue.

    Thanks anyhow!

    Stomp!

  • Hi Yanming,

    Just one other thing I noticed. I'm not sure if its directly related to the current spikes (which I've improved slightly by increasing the gate turnoff resistors a bit).

    On a very zoomed out view with the top trace being the U phase voltage, and the bottom phase being the upper U gate driver. I am wondering you can determine why the PWM is stopping for such a long period of time?.

    When I increase DC Bus > 100V the PWM chopping more evident.

    Thanks

    Stomp!

  • I suppose the issue should be from the h/w board if you can't meet the same problem on TI HVkit, please make sure the dc-bus voltage, phase voltage, phase current sensing are good on this customized board and set right scale current & voltage value in user.h.
    For this pwm chopping issue, please check there is any fault to trig the driver or C2000 protection to disable the pwm output.
  • Hi Yanming,

    As always, thanks for your time.

    Can you help me on this topic of the ADC sample point.

    My gate driver is has a 400ns - 800ns Propagation delay, I have a small resistor near the gate of each MOSFET, which maybe adds another 5nS.
    This means I could have from 405nS to 805nS from PWM High to MOSFET On. On HVKIT the driver has 700nS to 1.9uS.

    So my question is about the ADC sample point. I currently have a SOC delay of 9 for the TMS320F28069 part. Is there any way I can trigger an external pin to watch on the scope when the ADC is sampled? I have a really strange feeling that by having the faster driver I need to modify the ADC sample window or conversion delay.

    Is there any guidance documents or examples on this?

    Thanks
    Stomp!.
  • You may refer to the chapter 17 of instaSPIN user's guide (SPRUHJ1G) if you are using shunt resistor for current sampling. All the example projects support 3-shunt current sampling by default.