Because of the holidays, TI E2E™ design support forum responses will be delayed from Dec. 25 through Jan. 2. Thank you for your patience.

  • State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 82 subscribers
  • Views 323 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.

TMS320F280039C: PWMxA and PWMxB with phase shift by deadband module

Wayne Huang
Genius 15365 points
Part Number: TMS320F280039C
Other Parts Discussed in Thread: SYSCONFIG, C2000WARE

Dear Champs,

I am asking this for our customer.

In TRM 20.7.2 Dead-band Submodule Additional Operating Modes, there is a note 

".... Phase shifting B-channel with respect to the A-channel: When PWMxB is derived from PWMxA using the DEDB_MODE bit and by delaying rising edge and falling edge by the phase shift amount. When the duty cycle value on PWMxA is less than this phase shift amount, PWMxA’s falling edge has precedence over the delayed rising edge for PWMxB. It is recommended to make sure the duty cycle value of the current waveform fed to the dead-band module is greater than the required phase shift amount."

I tried to test DEDB_MODE and found it works when deadband is smaller than duty, but it does not work (no PWM output) when duty is smaller than deadband.

I used Sysconfig in CCS 12.2 with c2000ware V4.03 in below settings.

Question:

It says "It is recommended to make sure the duty cycle value of the current waveform fed to the dead-band module is greater than the required phase shift amount."

Is it RECOMMEDNED or REQUIRED?

The user wants to generate "PWMxA’s falling edge has precedence over the delayed rising edge for PWMxB.". Is it possible?

  • +1
    TI__Expert 8051 points

    Hello Wayne,

    A colleague of mine tested the use case of a duty shorter than the DB on the F28388D, which uses the same ePWM as the F280039C. According to that thread, located here, it seems like it should be worded as a hard requirement. If you feel that the current wording is unclear or misleading, that's understandable, please submit a document feedback ticket via the link at the bottom of each page of the TRM! These help us locate and resolve these concerns.

    As for the user's request- to clarify, they're asking for a setting where the arrival of a falling edge of ePWMxA prevents the rising edge of ePWMxB? The easiest way to implement this that I can think of would be to utilize the device's CLB, overriding the output of ePWMxB with the Boolean result of (ePWMxB AND ePWMxA). This is assuming that the two signals are not derived from the same source, in which case, I believe that this could be accomplished with just the logic settings of the DB.

    Please let me know if this is not sufficient, and/or if you need any more assistance!

    Regards,

    Jason Osborn

  • 0 in reply to Jason Osborn
    TI__Genius 15365 points

    Dear Jason,

    The user only wants to generate PWMxA which leads to PWMxB during run time, say 135, 150, 180,  degree. And of course, PWMxA and PWMxB cannot both be ON at the same time.

    We believe CLB can make it but it will be more complicated.

    The users wonders if this is doable by this deadband module.

    But it seems the answer is NO.

    Yes, I think the doc description is not clear here.

    Will submit a feedback later.