Im trying to create a power supply with an addjustable output between 0-12V or 0-3A (CV and CC adjustable). To make sure that the components don’t get damaged I'm trying to implement a protection for over current and voltage. When current higher then 3A appears the MAX4172 will act as a current sense amplifier and the MAX9038 as a comperator with a reference voltage (reference itself still needs some work, also I know it has an internal reference). Since the output here is 5V and I want to make it variable I used the two transistors to amplify it to the source voltage (or at least close to). This seems to work nicely as seen in the picture except that the MOSFET doesnt work at low voltages, google told me its the Vto value for the MOSFET that is causing the problem. The application that it will be used for might require a high current and low voltage (0.5V in some cases), but I would like to actually have that output voltage of course! Is there any way to make this possible or any suggestions for good P-channel MOSFETs for this case? (let’s say as low as ~0.3V is required)

Since the actual power supply has to be able to regulate current and voltage I also need overvoltage protection. I've seen crowbar circuits which seem to work but bring the voltage and current to 0 (from testing a while ago at least), is there a way to make this work just like my current limiting, by limiting the output voltage to 12V?

Also, since I normally don't do too much regarding power supplies is there any other obvious problem that I've missed with the circuit thats not gonna work realistically? (also sorry for the horrendous loooking simulation, finally got it working somehow and last time I moved stuff I broke it again :P)

enter image description here


1 Answer 1


Issues I see with this design:

  • MAX4172 input voltage range issue. The supply of your MAX4172 is + 5 V while the sense inputs are at +12 V. That is not going to work, you will have to supply the MAX4172 with at least 12 V also.

  • The circuit around Q1 and Q2 will probably work fine. However I would use a simpler circuit, I'd use one NMOS to make M2 switch on. Then the polarity of the output of the MAX9028 will be incorrect, to solve this, swap the inputs of the MAX9082 so that its output becomes low when there's an overcurrent.

  • \$\begingroup\$ Thanks for your feedback, supply for MAX4172 was something i didnt think of yet so thats good to know! looking at the data sheet though it uses a resistor to ground on the output only. I tried adding resistors at different locations but it didnt really seem to help.. where would adding a small resistor help then? I knew about the mosfet circuit but completely forgot i could just change polarity by swapping MAX9082 inputs.. dumb me. i implemented this and it seems to work (i assume its also a little safer in case any of the current measurement circuit fails, cause low means off) \$\endgroup\$
    – Parimo
    Commented Oct 23, 2020 at 10:29
  • \$\begingroup\$ You're right, I was thinking that the output of the MAX4172 only sinks current but it does not, it is a source. So my comment on that is wrong, I'll remove it. \$\endgroup\$ Commented Oct 23, 2020 at 10:36
  • \$\begingroup\$ Do you know of a fix for the MOSFET not turning on at low voltages? Is this something i can even fix or is it a limitation of components? Would love to be ble to regulate up to voltages of around 300-400mV if possible. \$\endgroup\$
    – Parimo
    Commented Oct 26, 2020 at 17:32
  • \$\begingroup\$ Be more specific, do you mean an output voltage of 300 mV? For PMOS M2 to be able to conduct, it needs a Vgs > Vt = 1 V. When the Drain and Source voltage of M2 are 300 mV then you would need the Gate at -700 mV, so you would need a negative supply voltage for that. I suggest that you study designs of lab supplies (Google: "lab supply circuit") to see how to regulate to low output voltages. General remark: you might be falling into the trap of wanting more than your current experience level allows you to. \$\endgroup\$ Commented Oct 26, 2020 at 20:24
  • \$\begingroup\$ Basically i need to be able to protect something by applying a voltage of 300mV over it (with other serie resistors). The other resistors can be small which means the voltage drop of 300mV is the output voltage. So i need to regulate atleast from 300mV Managed to make this work in LTSpice by applying a voltage of -3V over the source of the MOSFET you suggested me to add. Most lab supplies seem to add an LDO after the transformer, is this possible when using a flyback converter instead of an iron core transformer? Or should i look into IC's like MAX1044 (turning positive supply into negative)? \$\endgroup\$
    – Parimo
    Commented Oct 27, 2020 at 10:50

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.