• State TI Thinks Resolved
  • Locked Locked
  • Replies 1 reply
  • Answers 1 answer
  • Subscribers 82 subscribers
  • Views 1696 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.

TMS320F28379D: External synchronization of EPWMs, eCAP and EPWM, Synchronizing two /three controllers

Radha Sree Krishna Moorthy
Prodigy 150 points

Part Number: TMS320F28379D

Hello,

I am looking to synchronize the EPWM modules of two different 28379 DSPs using a synchronization signal from a third 28379 DSP. As per the manual, the best possible solution is to use the EXTSYNCOUT and EXTSYNCIN through the X-bar for synchronization. 

1. With the aforementioned approach what should be the width of the pulse that is to be used for synchronization. Does the synchronization work based on the rising edge or the falling edge of the pulse ?

2. Is there an alternate way to do it via eCAP i.e., capture the rising the rising edge of the sync pulse using eCAP and generate an interrupt to reset the EPWM TB counter. Or should the synchronization pulse be generated through eCAP. 

3. Is there any other better way to do this to achieve good synchronization without significant delays ?

Thanks & Regards,

Radha

  • 0
    TI__Guru 57255 points

    Hi Radha,

    First, I apologize for the delay.  I'm answering for a different member of the team here.

    1) Please take a look at the "ePWM Timing Requirements" section of the datasheet (currently Table 5-59).  I would also recommend referring to the following E2E thread, as it will provide some extra detail.  The device in that older thread didn't have a crossbar, but the concept is still valid for your newer device).  This will be the lowest latency & most deterministic method.
    https://e2e.ti.com/support/microcontrollers/c2000/f/171/t/313720

    2) Using interrupts is valid, but it will add non-deterministic latency, since the taking of interrupts is non-deterministic.  It will also have higher latency.  If this is okay for your system, then this approach may be viable.

    Hopefully this helps!


    Thank you,
    Brett