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.

  • TI Thinks Resolved

CCS/DRV8323: update from DRV8305 to DRV8323

Part Number: DRV8323

Tool/software: Code Composer Studio

Hello everyone, I am again with a question on DRV8323.
Driver output  looks like this 

I can not understand where these short signals come from. At the input of the driver, the signal is not interrupted.

Green middle point of the bridge. blue gate high FET.

The output of DRV8305 looks better.

Ch2.

  • Andrew,

    Sorry for the delay here.

    Is this occurring on your own board or the EVM?

    What FET is being used and can you send me a picture of your register configuration?

    Regards.

    -Adam
  • In reply to Adam Sidelsky:

    I use CSD88599Q5DC. I use DRV8323RH. I tried IDRIVE = Hi-Z 240
    IDRIVE = 75 kΩ to DVDD 520, IDRIVE = 18 kΩ to DVDD 1140, IDRIVE = to DVDD 2000.

    U7 DNP

    R15,R16 = 0R.

  • In reply to Andrew Buckin:

    Andrew,

    Was the result the same if you used lower iDrive vs. higher iDrive? This could help us rule out an iDrive issue. The images shown, which iDrive setting was used. If iDrive is too high, it could cause other phases/nodes to unintentionally turn on which could cause this issue to appear.

    Another possible issue is that the parallel FET architecture is causing this issue. Could you remove the gate resistors on LS and HS for either U6 or U7 and see if this behavior is any different?

    Regards,

    -Adam
  • In reply to Adam Sidelsky:

    The second transistor is not installed.

    Switching in one channel is induced on the gate in another channel?
    Does Switching open a high transistor in another channel and then the driver closes it?

  • In reply to Andrew Buckin:

     I looked three channels in the gates. In one of the channels visible fluctuations. 0-RPM. IDRIVE - 2A.

  • In reply to Andrew Buckin:

    Andrew,

    Please send us your remaining schematic, if concerned, please email it to adam(dot)sidelsky(at)ti(dot)com.

    Regards.

    -Adam
  • In reply to Adam Sidelsky:

    We have a standard scheme.  :)EVO BLDC 6050C.PDF

  • In reply to Andrew Buckin:

    Andrew,

    What is your peak and continuous current per phase?
    We notice a small amount of VM capacitance near the DRV, can you probe VM near DRV with scope?
    Also please probe nFAULT same time.
    Could you share scope images of the input while the motor is spinning at 1000RPM?
    The 1000RPM images shown, are these during spin up or with the motor consistently spinning 1000RPM?
    Would you be willing to share the layout?
    What is the application here? Are the output wires long? Are the FETs far away from the DRV?

    Regards.

    -Adam
  • In reply to Adam Sidelsky:

    Thanks for the advice,

    I watched. nFAULT - everything is clean. We want current up to 70A.
    I set GAIN to 40V/V and #define USER_ADC_FULL_SCALE_CURRENT_A (82.5)
    I set PWM frequency to 60-80KHz. fluctuations gone. :)

    How set the dead time between phases in MotorWare?
    Is now the dead time depends on the frequency of the PWM?

    Where can I find recommendations (formulas) on the PWM frequency for motors with a small inductance.

    thank
    Andrew.

    PS:

    I returned to PWM to 15 KHz, reduced two define to 0.5.

    The signals have become readable and beautiful.  :)


    // # define HAL_PWM_DBFED_CNT 1
    #define HAL_PWM_DBFED_CNT (uint16_t) (0.5 * (float_t) USER_SYSTEM_FREQ_MHz) // 2 usec


    //! \ brief Defines the PWM deadband rising edge count (system clocks)
    //!
    // # define HAL_PWM_DBRED_CNT 1
    #define HAL_PWM_DBRED_CNT (uint16_t) (0.5 * (float_t) USER_SYSTEM_FREQ_MHz) // 2 usec


    Now I repeat all the tests.
    Maybe something else forgot and missed.

    :)

  • In reply to Andrew Buckin:

    I did the first tests. A beautiful signal in the GATE and phase does not mean that the motor is working well. :(

    I made a speed comparison regarding the delay. And back to delay 2us or 3us.

    I did not expect that with "field weakening" a big difference in speed. To change the direction of rotation with the "field weakening" with great acceleration is generally impossible.

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.