For my project I want to allow the user to either use a 12 or 24VDC power adapter if needed but I dont want to make 2 seperate DC inputs. Neither do I want to use a buck-boost because I dont want to regulate the 12V voltage but most bucks only work with Vin>Vout by ~1V.

Right now I am using the setup in the picture to get the 24V to 12V, the load side needs min. 5A so thats why I dont want to use any linear regulator aswell.

A user told me I could make a relay or a circuit to bypass the regulator when the voltage is at a specific level. But that's where I am stuck and would like to know a good and easy way check the Vin or the regulated Voltage < 12V and if so bypass it and just use the 12V directly. I read a bit about comparator ICs that could do that but my knowledge is really limited here. If you could provide me with some guidance I would really appreciate that.


I tried to think of something myself but can't really say if that would work.


simulate this circuit – Schematic created using CircuitLab

Voltage regulator


1 Answer 1


There are buck switching converter chips designed to use a PMOS instead of a NMOS for the upper pass element. This allows 100% duty cycle, at the cost of a little efficiency loss due to the PMOS. Some switcher chips can also do 100% duty cycle with a NMOS, by using a charge pump. So you got a more efficient NMOS for the switching converter.

Converters capable of 100% duty cycle will just keep the top FET fully ON when Vin is too low for regulation (ie, 12V in, 12V out) so it's just the RdsON between input and output.

If you are going to design a switching converter anyway, and it looks like it since you put a schematic in your question, this could be a simpler solution to your problem...

  • \$\begingroup\$ It's Rds_on and the inductor series resistance between input and output. \$\endgroup\$
    – jpa
    Commented May 2, 2021 at 16:23
  • \$\begingroup\$ I tried to do that with a LM25085MY on TI Webbechn but even with super small RdsON and the inductor (which increased footprint extremly) and a series Rsense it wanted I get about -0.2V for 12V. If I simulated it for 24V now it get's up to 14V which is horrible. Can't seem to get it right. \$\endgroup\$
    – Daniel Do
    Commented May 3, 2021 at 9:40

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.