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.

PMP12072: PMP12072

Part Number: PMP12072
Other Parts Discussed in Thread: TIDA-00751

I have been working on a design based on the circuit shown in PMP12072.  I ran into a problem where the circuit was not shutting off the MOSFET's quick enough.  There was too much delay between the time the input signal went to zero and when the MOSFET's turned off.  So I saw the turn-off circuit shown in TIDA-00751 and did a copper spin to try this circuit.  Improvement was seen.  However it is not turning off as fast as I need or as fast as what is shown in the TI design files related to TIDA-00751.  I have compared the spec's for the MOSFET's shown in PMP12072 and TIDA-00751 and it doesn't seem like the differences should be causing as much turn-off delay as I am seeing.  From the time the input drops to zero and the output turns off is about 1.3msec.  Files are attached.  As always any comments/thoughts you can provide are greatly appreciated.

  • Larry,

    PMP12072 was designed to switch at 60Hz, so a high speed turn off was not necessary. D2/C2 acts as a peak detector for the secondary side of the transformer. The voltage on C2 is likely staying charged because the D2 diodes are reversed biased when transformer secondary polarity reverses. In other words, it does not actively discharge C2. C2 is being discharged by R6 (100k), which has a time constant of 1mS, which may explain why you are seeing a relatively long delay at turn off. Try decreasing the value of C2 (try 100pF, 100X smaller) to see if this speeds up the Q5/Q6 discharge circuit. But C2 should be large enough to allow the FET gates to stay high over the full period of controller U1.

  • John, thank you again for your assistance. The 100pF cap was tried and greatly improved the delay but the falling edge of the output waveform was a little ragged. I tried a 1000pF and that provided a clean edge. Everything seems to be working fine now.