• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 82 subscribers
  • Views 699 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/TMS320F28335: buck converter protected reactive time , ZCD / OCP

Wei Lee
Prodigy 80 points
Part Number: TMS320F28335

Tool/software: Code Composer Studio

before I use DELAY_US function for Over current protection (OCP) . To avoid the inductor current that reach OCP will turn high side driver off then the current will be ongoing lower. once lower than OCP , the Epwm will immediately turns high side driver on. I afraid of this system will be unstable, because the Epwm will still turn on-off-on-off.

Now I using over current window during I_OCP*100% to I_OCP*90% ,but It still can't turn off immediately ,reaction must be 20~30uS .

even if I put it on "void main(void)" or " interrupt void adc_timer_isr(void)".

Zero current detection (ZCD) , if the inductor current is going negative ,the low side MOSFET must be turns off immediately (< 1uS)

Over current protection (OCP) , if the inductor current is going too much,the high side MOSFET must be turns off immediately (< 1uS).

BTW I have studied the "C2000 Digital Power BoosterPack"  that said "In this project the two PWM outputs are driven low immediately on a comparator event to protect the power stage."

I don't know how many "driven low immediately" time.

###################################################################################

if(IL < 2048) //IL < 0 , ZCD
{
EPwm1Regs.AQCSFRC.bit.CSFA = 2; //HG on
EPwm1Regs.AQCSFRC.bit.CSFB = 1; //LG off
OCP=0;
}
else if(IL > 4095) //IL > 3 ,OCP
{
EPwm1Regs.AQCSFRC.bit.CSFA = 1; //HG off
EPwm1Regs.AQCSFRC.bit.CSFB = 2; //LG on
OCP=1;
}
else if(IL > 3891 && OCP==1) //IL > 2.7 , OCP window
{
EPwm1Regs.AQCSFRC.bit.CSFA = 1; //HG off
EPwm1Regs.AQCSFRC.bit.CSFB = 2; //LG on
}
else //IL >0 A and IL < 3A  , feedback ---error amp
{
OCP=0;
EPwm1Regs.AQCSFRC.bit.CSFA = 0;
EPwm1Regs.AQCSFRC.bit.CSFB = 0;

Vbo_err = Vbu_cmd - Vo; // cmd 400

PWM1_Duty = Vbo_cmpa1 + (Vbo_err-Vbo_err1)*1 + (Vbo_err+Vbo_err1)*1;
Vbo_err1=Vbo_err;

Vbo_cmpa1=PWM1_Duty;


if( PWM1_Duty > PWM1_Period ) //limit duty
{
PWM1_Duty = 0.9*PWM1_Period;
}
if( PWM1_Duty < 1 )
{
PWM1_Duty = 0;
}

EPwm1Regs.CMPA.half.CMPA =PWM1_Duty;
EPwm1Regs.CMPB =PWM1_Duty;
}

#############################################################################

  • 0
    TI__Genius 10820 points

    Hi,

    Based on the high 20-30 us delays, it seems like the behavior you are seeing is as if OCP has been implemented in software. When on-chip PWM trip protection (hardware) mechanism is used, a reaction time less than a few 10s of ns is expected. This is what is used in the BoosterPack solution.

    Please refer to the BoosterPack documentation and the PWM trip sub-module description in the Technical Reference Module for more information.

    Hrishi

  • 0 in reply to Hrishi Nene
    Prodigy 80 points

    Hi Hrishi,

    the following waveforms are 1kHz signal. the reactive time is about 30uS. 

    then I using 100kHz the same as ePwm frequency that will be unstable.

    and I would like try to use the BoosterPack documentation's method,but F28335 hasn't comparator .......

  • 0 in reply to Wei Lee
    TI__Genius 10820 points
    It looks like the green trace is your fault signal. This signal can be used directly as a TZ (trip zone input) on the F28335 device. You can then use one-shot or cycle-by-cycle trip mechanism. In this case your reactive time should be of the order of 20-40ns.

    Hrishi