• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 83 subscribers
  • Views 574 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.

CCS/TMS320F28027: TMS320F28027: Piccolo HVPSFB(PCMC) FET Switching timing

user5277039
Intellectual 640 points
Part Number: TMS320F28027

Tool/software: Code Composer Studio

Dear Ti community

I'm developing PCMC on HVPSFB.

To understand FET switching timing,I made time chart.

I insert DCAEV1 the middle of EPWM1 up count and down count.

(SR_mode = 2)

But,the time chart was as follows.

Q2,Q3,Q5,Q6 are strange.

So,I want you to point a mistake out.

  • About Q2,Q3.

First,COMP_A = 310,COMP_B = 20.

Count up EPWM2_CNT and soon,EPWM2_B is high because of COMP_B match.

After that EPWM2_B is low because of 1st DCAEV1 and soon be high because of COMP_B match.

Then,1st interrupt start and EPWM2_action change.

COMP_A/B_Shadow don't change(A:310→310   B:20→20), so COMP_A/B don't change at 3rd SYNC. 

Then 2nd interrupt start and EPWM2_action change.

Before 4th SYNC , COMP_A/B_Shadow  change(A:310→20   B:20→310), so COMP_A/B  change at 4th SYNC. 

At 4th SYNC,EPWM2_B is low because of CNT = 0.

And soon ,EPWM2_A is high because of COMP_A match .

Then 3rd interrupt start and EPWM2_action change.

At 6th SYNC,EPWM2_A is low because of CNT = 0.

And soon ,EPWM2_A is high because of COMP_A match .

After 6th SYNC,EPWM2_COMP_A/B_shadow change(A:20→310   B:310→20).

COMP_A/B change at 4th DCAEV1 and EPWM2_A is low because of 4th DCAEV1 .

And soon ,EPWM2_B is high because of COMP_B match .

Then 4th interrupt start and EPWM2_action change.

Before 8th SYNC,EPWM2_COMP_A/B_shadow change(A:310→20   B:20→310).

So COMP_A/B change at 8th SYNC and EPWM2_B is high because of CNT = 0 .

And soon ,EPWM2_A is high because of COMP_A match .

・・・

  • About Q5,Q6

First,COMP_A = 310,COMP_B = 20.

When count start ,EPWM4_B is high because of CNT = 0.

And soon , EPWM4_B is low because of COMP_B match.

At 1st DCAEV1 , EPWM4_B is high because of CNT = 0.

And soon , EPWM4_B is low because of COMP_B match.

Then,1st interrupt start and EPWM4_action change.

And soon  EPWM4_action change before 2nd SYNC.

At 2nd SYNC , EPWM4_A is high because of CNT = 0.

After 2nd SYNC, EPWM4_COMP_A/B_shadow change(A:310→20  B:20→310).

At 2nd DCAEV1 , COMP_A/B change,and soon EPWM4_B is high because of COMP_A match.

After that, EPWM4_A/B don't change.

Because the condition to be low EPWM_A/B is only when CNT = 0 between interrupt and SYNC(that means EPWM4_action②,④).

  • 0
    TI__Genius 15735 points
    Hi user5277039,

    Sorry for the late reply, just want to let you know I am looking into it. Is the question why the time chart you get is incorrect?

    Regards,
    Chen
  • 0 in reply to Chen Jiang ti
    Intellectual 640 points

    Yes,I'd like to know a reason why the time chart is incorrect.

    But ,I get  the cause only a little.

    When,EPWM1 is up ,interrupt starts CNT=232.

    And when is down, interrupt starts CNT = 74.

    But,when I tyied using a 「ExperimentersKits」,

    the interrupt starts CNT = 245~251 and CNT = 55~61.

    In that case,while NOP waiting,SYNC occurs.

    As a result, it becomes correct.

    Why the interrupt start delayed?

  • 0 in reply to user5277039
    TI__Genius 15735 points
    Hi user5277039,

    In the code of HVPSFB project, the period of epwm2 is 300. Can you explain why you use 150 instead 300 in your example? Thanks.

    Regards,
    Chen
  • 0 in reply to Chen Jiang ti
    Intellectual 640 points

    To calculate excel , It's supposed that DCAEV1 occuers at 150 ,450,750 ・・・count.

    EPWM2 is reseted at DCAEV1 and SYNC, isn't it.

    And, the wrong time chart is settled.

    I want to answer about interrupt timing.

  • 0 in reply to user5277039
    TI__Genius 15735 points
    Hi user5277039,

    I will be back to office on 3/21 and will follow this thread at that time. Sorry for the inconvience.

    Regards,
    Chen
  • 0 in reply to user5277039
    TI__Genius 15735 points

    Hi user5277039,

    Yes, you are right about the SYNC.

    Do you have any updates of this issue?

    "When,EPWM1 is up ,interrupt starts CNT=232.

    And when is down, interrupt starts CNT = 74."

    This is the original settings in the HVPSFB code, isn't it?

    Also I don't know whether you have modified the code when you said "But,when I tyied using a 「ExperimentersKits」"

    Regards,
    Chen

  • 0 in reply to Chen Jiang ti
    Intellectual 640 points

    I don't modified the code.

    I set the breakpoint at the code 「PUSH   AR1H:AR0H」 because the code is the first line of interrupt.

    As the result, the debug stop at the breakpoint .

    Then  I check the EPWM1 CNT .

    CNT should be 232 or 74, but CNT = 245~251 and CNT = 55~61.

    I want to understand the delay.

  • 0 in reply to user5277039
    TI__Genius 15735 points

    The reason is it takes approximately 15 cycles to enter the ISR in assembly based on the hardware. Thanks.

    Regards,

    Chen