3
\$\begingroup\$

Recently I came across a schematic of a synchronous buck converter using normal transistors in stead of MOSFETS, like this: Synchronous buck using transistors

I was always under the impression that when the top switching device turns off, the inductor generates a large negative voltage because it wants to resist the change in flow of current. That's why the reverse biased freewheel diode is there in a asynchronous buck regulator, so the current can keep flowing.

How does the above circuit work then? The bottom PNP transistor, taking the place of the freewheel diode, cannot conduct current in that direction afaik. Current can flow emitter to ground but not from ground to emitter, can it ?

I built the circuit and it appears to work, measuring point before the inductor, where both emitters meet (excuse the poor quality): Synchronous buck using transistors scope image

Although I'm not sure what I'm seeing here. There appears to be a dip below ground on the negative cycle. At first I thought this was the inductor's negative voltage spike, somehow being cut off by the transistor. But the same is happening on the positive cycle, with a little 'bulge' above VCC.

Could anyone enlighten me about how this works and what is happening?

\$\endgroup\$
  • \$\begingroup\$ I guess the top NPN would stay on until the PNP conducts enough, so the diode's role is taken up by the NPN. \$\endgroup\$ – peufeu Aug 6 '17 at 19:45
  • \$\begingroup\$ I can't see Q1 turning on ever. SFAICT T1 is doing all the work. does it get hot? \$\endgroup\$ – Jasen Aug 7 '17 at 0:11
  • \$\begingroup\$ Replace Q1 with a diode and T1 with a P-MOSFET, then you're cool. \$\endgroup\$ – Harry Svensson Aug 7 '17 at 1:55
  • \$\begingroup\$ @Jasen : Neither transistor gets hot. The bottom transistor clearly does something as removing it changes te waveform into a nearly flat line that stays high. \$\endgroup\$ – Unimportant Aug 7 '17 at 10:14
3
\$\begingroup\$

Current can flow emitter to ground but not from ground to emitter, can it ?

Collector and emitter can work in reverse. It's an NPN or a PNP so junctions are somewhat interchangeable but you don't get great β but you do usually get quite low saturation voltages.

On the other hand, the circuit has poor efficiency because the top transistor (NPN) is an emitter follower and will always drop pretty much 0.7 volts or greater unless the base drive is lifted above the incoming power rail.

I would never sanction using this circuit - MOSFETs are far more suited to synchronous buck conversion (generalism alert!).

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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