• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 82 subscribers
  • Views 1209 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.

LAUNCHXL-F28379D: Trip Zone with Deadband

Jens H
Prodigy 130 points
Part Number: LAUNCHXL-F28379D

Hello,

I would like to use the LaunchXL-F28379D to control the switches of a half-bridge inverter. I would like to use the half-bridge in Triangular Current Mode with ZVS. For detecting the lower current boarder I´m using an external Comparator. The Comparator trips when I reach the boarder. When reaching this boarder, the switches should be switched as fast as possible.

My idea was to connect the output of the external Comparator to the TripZone. If the Comparator goes low, the TripZone is activated and deactivates for example switch S1 that is controlled by ePWM-Module EPWM1A. After a specified dead time (e. g. 30ns) the other switch (S2) should be activated over Module ePWM1B.

The Problem is, that if I use the Trip Zone, I can not use the Deadband-Module in the ePWM-Module to get a specified dead time between turning S1 off and turning S2 on, right?

How can I switch the switches S1 and S2 using the external Comparator Signal as fast as possible? If possible it should happen without integrating the CPU. So using an Interrupt and an ISR is not an option for me.

Are there other modules that can be used? Or is it possible to use the TripZone with Deadband? Is it possible to adress the Action Qualifier without using the CPU?

I would be glad if you could help me with my question.

Best Regards,

Jens

  • 0
    TI__Mastermind 24095 points

    Hi,

    Jens H said:
    The Problem is, that if I use the Trip Zone, I can not use the Deadband-Module in the ePWM-Module to get a specified dead time between turning S1 off and turning S2 on, right?


    Did you look at the possibility of using T1/T2 events added in type-4 PWM?
    Please refer to Figure 15-21. Possible Action-Qualifier Actions for EPWMxA and EPWMxB Outputs
    Also refer to AQCTLA2/B2 registers

    T1/T2 (Trip, DC events) can now be used to set events (set/clear PWM output) in action qualifier module.
    So - for your configuration, you'll have the deadband set as needed for normal operation.
    When certain comparator event occurs - you'll configure S1 to turn off immediately.
    S2 - via T1 configuration in Action Qualifier, can be used to set, this will go through the usual deadband and turn on S2 after deadband delay.

    Basically, the T1/T2 feature is added to address requirements like your system.




  • 0 in reply to Subrahmanya
    Prodigy 130 points

    Hi,

    Thank you very much, this helps a lot.

    Is there an example in which T1/T2 events are used?

    Best regards,

    Jens

  • 0 in reply to Jens H
    TI__Mastermind 24095 points

    Hi,

    There are no examples available currently for this events.
    But there is a plan to release one early next year with T1/T2 events used to implement peak current mode control.

  • 0 in reply to Subrahmanya
    Prodigy 130 points

    Hi,

    I tried to use the T1/T2 function as you described. But it doesn´t work yet. My trip over the TripZone directly, that turns off Switch 1 over ePwm1A works. But the trip over the AQ using T1, that should turn on Switch 2 over ePwm1B doesn´t work. I´ve written to the following registers:

        EALLOW;
        EPwm1Regs.AQTSRCSEL.bit.T1SEL = 0b0100;        // select TZ1 as an event source
        EPwm1Regs.AQCTLB2.bit.T1U = 0b10;                 // Action when event occurs on T1 in UP-Count, 10: Set: force EPWMxB output high.
        EPwm1Regs.AQCTLB2.bit.T1D = 0b10;                 // Action when event occurs on T1 in Down-Count, 10: Set: force EPWMxB output high.
        EDIS;

    Is this correct? Or do I have to use DCBEVT1 somehow? Are T1/T2 somehow connected to DCAEVT1/DCAEVT2?

    As this seems to be a newer feature, how do I know, if my board is capable of doing this?

    Best regards,

    Jens

  • 0 in reply to Jens H
    TI__Mastermind 24095 points

    Hi,

    Jens H said:
    Is this correct? Or do I have to use DCBEVT1 somehow? Are T1/T2 somehow connected to DCAEVT1/DCAEVT2?


    Is PWM1B not derived from PWM1A using deadband in your configuration?
    Are you generating PWM1B independently from PWM1A?
    Above feature suggestion would work if PWM1A and PWM1B are complimentary outputs (PWM1A goes low with comparator event and PWM1B turns ON with dealy based on the same event). So, I would not expect any independent PWM configuration in PWM1B in Action Qualifier.

    From the way you've configured, PWM1B would be independently set/clear with T1 events. Are you expecting 2 TZ1 events in T1U and T1D?
    What is the trip zone configuration for PWM1B in this case?

  • 0 in reply to Jens H
    TI__Mastermind 24095 points

    Hi,

    Jens H said:

    As this seems to be a newer feature, how do I know, if my board is capable of doing this?



    There is no additional hardware needed to support this feature. This is part of the EPWM configuration.