variable width pulse delay

Discussion in 'General Electronics Chat' started by ako, Mar 14, 2009.

  1. ako

    Thread Starter Member

    Mar 14, 2009
    10
    0
    Can anybody help me how to implement a fixed delay variable pulse width circuit? The input should be a 15V to the circuit is a pwm signal. The output should be the same in duty cycle but delayed for a fixed amount of time. However it should respond to changes in the duty cycle of the input signal. Is that possible?
     
  2. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    What is the frequency of your PWM signal?
    What is the range of the duty cycle (mimimum to maximum)?
    How much delay do you need? Does it need to be adjustable?
    Why do you want to do this?
     
  3. ako

    Thread Starter Member

    Mar 14, 2009
    10
    0
    I am trying to use the parallel outputs of a uc3864 resonant controller to switch a two-phase interleaved converter. Our frequency is set at 200k and min duty cycle at 40% and max at 50%. I just want the other output to be delayed so that the two phases don't switch simultaneously. I plan to delay the other pulse by 50% of the period of 200k so i think it should be delayed by about 2.5us. Appreciate help. Thanks.
     
  4. ako

    Thread Starter Member

    Mar 14, 2009
    10
    0
    I think I'm just looking for a phase shifter circuit.. 180% phase shift.
     
  5. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    A logic inverter comes to mind, but I don't think that's what you want. You need to post waveforms of the input, and the proposed output, for 40% and 50% input duty cycles.
     
  6. ako

    Thread Starter Member

    Mar 14, 2009
    10
    0
    Hi. Waveform attached. I am assuming that it is running at 40% duty right now. Just don't mind the art. The green lines project the maximum 50% duty cycle.:)
     
  7. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    Did you look at UC1861,63 in the same datasheet as UC3864? I think it does what you want.
     
  8. ako

    Thread Starter Member

    Mar 14, 2009
    10
    0
    I was thinking about it before but abandoned the idea. The dead time for 61/63 isn't fixed. It is the sum of the on-time and twice the width of the one-shot. Varying the load would also vary the on-time which in turn also varies their respective off-times, which I believe further complicates things. I think the 61/63 application is for half bridge type circuits.
     
Loading...