• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 32 subscribers
  • Views 650 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Original question:

DRV8313: sinusoidal commutation and other questions

DRV8313: Is the "built-in" deadband time (if any) sufficient for synchronous commutation (PWM-ing the INx inputs while the ENx inputs are asserted)?

Michael Arensdorf
Prodigy 50 points
Part Number: DRV8313
Other Parts Discussed in Thread: DRV8320, , MOTORWARE, DRV8312

The "Switching Characteristics" of the datasheet indicate t-dead = 90ns.  Is this a deadband time that is generated within the driver?

Also, there is a sentence "Synchronous rectification is achieved by applying a pulse-width-modulated (PWM) input signal to the INx pins while driving".  The sentence appears to be incomplete.  Should it end with "...while driving the ENx inputs high"?

I'm trying to replace the DRV8320 HAL code used in the InstaSPIN labs targeting a LaunchPadXL/BOOSTXL_DRV8320 combination with code to control a DRV8313 instead.  I see in HAL_setupPWMs() (hal.c) that a single EPWM peripheral is used to drive each GHx/GLx input pair.  The A output drives GHx and the B output drives GLx.  Interestingly the B output is forced to be continuously asserted (via the Action Qualifier) and the deadband generator is used to create an "Active High Complimentary (AHC)" signal which forces the B output to de-assert whenever the A output is asserted, plus additional deadband time.  I'm wondering if, for the purpose of controlling the DRV8313, the A output could drive the INx and the B output could drive the ENx in the same way that they drive the GHx and GLx on the DRV8320 but without the deadband operation, i.e. ENx are asserted continuously.  Alternatively, perhaps the ENx could be driven by GPIO instead, so it could be de-asserted independently of the EPWM peripheral.

  • 0
    TI__Expert 6065 points
    The expert for this part has been notified and will help you tomorrow when he is back in office.
  • 0 in reply to Sanmesh Udhayakumar
    TI__Guru 65190 points
    Michael,

    Yes, the driver internally inserts dead-time as needed. Let me check on the Synchronous rectification note.

    As for the Instaspin code, that's supported by the C2000 team, I'll move this post to their forum so they can help.

    Regards,

    -Adam
  • 0 in reply to Adam Sidelsky
    TI__Guru** 115420 points
    You might refer to the PWM configuration in HAL_setupPWMs() in hal.c of the reference labs for DRV8312 kit in motorWare as below folder, the PWM configuration should be used for DRV8313 also.
    C:\ti\motorware\motorware_1_01_00_18\sw\solutions\instaspin_foc\boards\drv8312kit_revD

    For DRV8312/DRV8312, the PWM bypass the deadband and force the PWM for low side to high.
  • 0 in reply to Yanming Luo
    Prodigy 50 points

    OK so I have everything "wired up" in the software to drive the DRV8313 rather than the DRV8320.   Fairly straight forward.  However, our circuit board was designed such that the current sense circuits are not biased towards 1.65V, but rather are biased to ground.  I know that accurate current measurement, both positive and negative, are required for the EWPM trip zones functionality.... Is it necessary for the FAST algorithm etc. as well?  Or will FAST etc. work without current measurements to drive a sensorless BLDC?

  • 0 in reply to Michael Arensdorf
    TI__Guru** 115420 points
    It's necessary, InstaSPIN-FOC (FAST) must have accurate three-phase current and voltage sensing signals to drive a sensorless BLDC/PMSM/ACIM. If not have these signals, InstaSPIN-FOC (FAST) can't work.