I'm building a switching power supply taking 12V to 19V at 3A max out. I've already built one of these using a Schottky diode for the rectifier, but this time I'm thinking of trying a high-side-switched N-channel as the rectifier to reduce losses. The end result is to power a small computer based on a battery input (range 9.9V -> 16.8V, 3S or 4S LiPo.)

The battery circuit is separately fused/switched; this is the boost power supply part only.

My question is: The MOSFET driver IR2183 has a "typical circuit" that ties the VS signal for the bootstrap capacitor to the load in a different way than I'm doing here. Specifically, it shows the load as connected between the source and drain of the two MOSFETs, whereas I tie them together to provide a single output. Will the 2183 still properly bootstrap the gate drive of Q4? My guess is "yes" but I'd like a double-check by someone who's ideally done this before :-)

MC34063 and IR2183 based boost converter

  • 2
    \$\begingroup\$ Looks fine to me. \$\endgroup\$
    – Dave Tweed
    Commented Apr 27, 2013 at 20:35

2 Answers 2


Here's the checklist

  • IR2183 has a bootstrap mechanism.
  • IR2183 has built-in circuit for shoot-through prevention. (If it didn't have a built-in one, you would have to add an external one.)
  • IR2183 can operating at frequency up to 400kHz. Operating frequency of MC34063 is 100kHz. In the circuit, MC34063 is set to operate at roughly 50kHz.

So, yes, it should work.


It may work, but badly.

IR2183 has built-in circuit for shoot-through prevention - true, but it lacks zero-cross detection. So especially after you had a short L1-charging pulse via Q3, the stored energy quickly is transferred via open Q4 to the C4 and c5 and then backwards - via the still open Q4 -> L1 -> +15V (low side Vcc)


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.