Trying my fingers on a very old SMPS circuit (15+ years) section. The open loop test circuit generates 25 kHz pulse train (12 V - 0 V) with 50% duty cycle as designed in the actual SMPS circuit. SK2996 has 3-4 V threshold voltage. I have kept the secondary open during the entire testing. Is a dummy load required for the flyback to work correctly m?

Q1. On the first run, the circuit (secondary & all other sections detached) heats up the NTC (not shown) near the AC supply (230 V, 50 Hz) and the 317 V DC supply drops to almost 100 mV but none of the component malfunctioned. What possible must be happening? Something related to the pulse transformer itself?

Q2. Next I remove the sense resistor to remove the MOSFET from circuit, but the 1.1 V at the gate (don't know how the voltage leaked as the MOSFET was cold tested OK) fried the UC3845B output totem pole section. Why? Is the PWM IC venerable once the MOSFET fails in actual SMPS circuit?

Q3. Any thumb rule how to safely and correctly trigger a MOSFET biased to 317 V DC in an SMPS circuit (search shows 12-16 V volt switching only) using UC3845B controller?

Open Loop Circuit enter image description here


enter image description here

  • \$\begingroup\$ ”Is a dummy load required for the flyback to work correctly m?” Yes, you must! \$\endgroup\$
    – winny
    Commented Jan 11, 2019 at 9:41

1 Answer 1


Just a quick idea of what I would do.

Replace all of your components in the Drain circuit with a single resistor. Maybe something that would allow 100ma to flow from the 317 v circuit. Check the Drain circuit with a scope when you are driving Gate with Sq wave. Does Drain look right? Square wave? Is there much ripple on the 317 v supply?

Then I would reduce that resistor and draw more current and try again. Does it look right? Square wave? Is there much ripple on the 317 v supply?

  • \$\begingroup\$ Seems like a brilliant idea ( sorry can't vote). The resistor would be 3.3KOhm, 1W . What if there is ripple? How the actual circuit ( should draw few amperes I supposed) will be deduced latter? PWM with 12V too high? \$\endgroup\$
    – ark1974
    Commented Aug 11, 2018 at 2:11
  • \$\begingroup\$ Does a power supply ripple a factor here? I have checked on scope ( across the bleeder resistor) seems pretty smooth thought. \$\endgroup\$
    – ark1974
    Commented Aug 11, 2018 at 2:20
  • \$\begingroup\$ If there is excessive ripple, it might indicate that you are drawing a very large current. \$\endgroup\$
    – Rp Rp
    Commented Aug 11, 2018 at 14:41
  • 1
    \$\begingroup\$ I did not realize that this is a flyback converter. In a FORWARD converter you transfer power (current) to the secondary load, WHILE the primary transistor is flowing current. In a FLYBACK converter you are storing energy in the magnetic field of the transformer while the primary transistor is drawing current and whne the primary transistor turns off that energy in the magnetic field "flys back" and the secondary load (outfitted with a diode) takes the current. What happens if there is no secondary load to take that energy ??? \$\endgroup\$
    – Rp Rp
    Commented Aug 11, 2018 at 14:48
  • 2
    \$\begingroup\$ It comes back to the primary and presents itself with the opposite polarity that was originally presented to it by the primary transistor. Seems like that could potentially damage the primary transistor and anything else in that circuit. You must connect the secondary load. \$\endgroup\$
    – Rp Rp
    Commented Aug 11, 2018 at 14:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.