• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 81 subscribers
  • Views 444 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.

TMS320F28027: HRPWM on TMS320F28027

Andrey Sachko
Prodigy 250 points
Part Number: TMS320F28027
Other Parts Discussed in Thread: LM5109

Hello All,

I would like to ask TI people about HRPWM module on TMS320F28027 chip.Of course,I will read later ALL details about it. But now I quickly need to plan my system. Normally for Synchronous Buck Converter we drive upper transistor (signal EPWM1A) and signal EPWM1B for lower transistor is inverted version of signal EPWM1A with dead band intervals inserted. Dead Band intervals are formed by RED,FED counters in DB submodule.

But what about,if I switch to HRPWM mode? You say (I understand) that high resolution signal is formed only on upper EPWM1A signal.Then how I can get high resolution EPWM1B signal,with dead bands inserted? This is for Synchronous Buck Converter.

Does it mean,that I need to use only high resolution EPWM1A signal and then in external chip to form high resol EPWM1B with dead bands inserted?

We can, probably, use IR21834 for this task.

Thank you

Andrey Sachko

  • 0
    TI__Guru 57255 points

    Hi Andrey,

    For the F28027 device, there are largely three options:
    1) Use an external device to add the deadband to the PWM controlling the lower transistor.  This is what you've mentioned.
    2) Use ePWM1A for the upper FET and ePWM2A for the lower FET.  You will need to synchronize the two ePWM modules, but this is easily done.  You will then have two PWM outputs with HRPWM functionality.  During PWM updates, you'd need to write to both PWM registers appropriately.  It is a bit more painful, but not usually too bad.
    3) Use PWM1A and PWM2B to control the FETs.  Have PWMA be HR.  Utilize the deadband feature.  If used in this way, PWMxB's output will contain a few cycles of jitter (in PWM clock cycles) from the RED/FED settings that you configure.  Depending on your application/topology, this may or may not be an issue.

    Hopefully this helps.  You'll need to determine which of the three options is most appropriate for your project.

    [the problem is that the PWMxB output from the ePWM module does not have HRPWM capability;  note that this has been improved on newer devices]


    Potentially useful thread:
    https://e2e.ti.com/support/microcontrollers/c2000/f/171/t/341217


    Thank you,
    Brett

  • 0 in reply to Brett Larimore
    Prodigy 250 points

    Hello, Brett

    Thank you for the answer. It seems second approach is used in C2000 Digital Power BoosterPack. They use EPWM4A,EPWM5A signal pair to drive LM5109.

    I leave this thread open,so later I will ask a couple more questions on this topic.

    In which devices TI has made high resolution EPWMxB signals?

    Regards

    Andrey

  • 0 in reply to Andrey Sachko
    TI__Guru 57255 points

    Hi Andrey,

    "In which devices TI has made high resolution EPWMxB signals"

    Examples are the F2837xD, F2837xS, and F2807x families from C2000.  Any future C2000 device will almost certainly also have this capability.


    Thank you,
    Brett