This question already has an answer here:

I'm trying to design a 24V to 12V step down converter. Power efficiency is not that important to me, but I want to come up with the cheapest solution possible. I don't need isolation neither. So I though I would just come up with the simplest buck converter. Something like this

enter image description here

Sw2 would of course be an appropriate power MOSFET. Also the schottky diode is arbitrary, just used the first thing available in Circuit Lab.

Would such a simple buck converter be the cheapest solution for my application, and would it work, assuming that ripple is not a concern for my application?

EDIT: Second part of the question was if it works with such a simple feedback loop. There is a similar question which is already answered so I removed that part.



marked as duplicate by Olin Lathrop, Dave Tweed Feb 9 '16 at 18:34

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

  • \$\begingroup\$ What is your current requirement? Is it really 30A? \$\endgroup\$ – uint128_t Feb 9 '16 at 17:18
  • \$\begingroup\$ Yep. 30A peak. Say 15-20A continuous. \$\endgroup\$ – Ali Alavi Feb 9 '16 at 17:19
  • \$\begingroup\$ Is the load really similar to a resistor? It is actually possible to run a buck converter without feedback if the load is invariable. In your case, something a little over 50% duty cycle would do it. If you want precisely regulated output voltage, then of course you need feedback of some sort. \$\endgroup\$ – mkeith Feb 9 '16 at 18:35
  • \$\begingroup\$ @mkeith No, the resistor is just for simulation. The load is variable and can be anything. \$\endgroup\$ – Ali Alavi Feb 9 '16 at 18:37
  • \$\begingroup\$ @Olinlathrop I edited the question and edited the duplicate question out \$\endgroup\$ – Ali Alavi Feb 9 '16 at 18:43

While I'm not 100% sure on cost, a buck converter would very likely be the best solution for your problem.

As for your circuit working, it would work right up until your Vout reached 12V; then the MOSFET switch should open, but all the stored energy in the inductor would continue charging the capacitor to some overvoltage level determined by inductor/capacitor size ratios (and instantaneous load).

Now, with a sufficiently small inductor value, a small shunt resistor and sufficiently high capacitor value, that could end up being a small enough difference to not matter. Depending, of course, on what error factor you're willing to accept in your 12VDC output.

For a better solution than trying to brute-force your circuit with a self-oscillating converter, the TL494 controller looks like it'd work well for you (using appropriately high-power MOSFET(s) for the switching, of course), and it's incredibly inexpensive. :)

A read-through of the datasheet for that controller should tell you what you'll need for how to set up your converter based around it.


I would prefer something like an synchronous buck controller. You can connect external FETs, (n channel) and the current is mainly Independent from your controller. You need something like a charge pump or bootstrap circuit to drive the high switch. I would prefer a synchronous solution, because 20A over the power diode are about 20W. With a Low Side mosfet, power dissapation of the switch can be in the range of one to two watts.


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