I am designing power supply which must output following voltages at a maximum of 1.5 A stepped down from a 24 V DC supply.

  • 24 V
  • 13-15 V
  • 0-5 V

Waveform for one of the inputs requires following timing diagram.

enter image description here

PWM seemed intuitively like the easiest solution as single 24 V source can be used to generate all the needed voltage levels.

My trouble is finding the adequate component to do the switching. Components I found by browsing for integrated packages on the TI website so far either can't handle the voltage or switching frequency is too slow.

This transistor though seems like it may be just right.


Its features include voltage up to 60 V, current up to 4 A and 110 ns for single wave.

Assuming I follow design below is it possible for me to obtain the voltage values I desire?

enter image description here

Additionally can PWM be applied in such way that 24 V signal is on for 800us and the drops to 5 V?

Issue I see possibly arising is PWM being an averaged value over a period of time from where it's generated and unable to respond to quick changes such as those desired by that timing diagram.

  • \$\begingroup\$ Why is L2 present? It complicates things (like the right side is not grounded anymore) and basically L1 and L2 are in series. I'd remove L2 and make L1 200 uH. In theory you could indeed control M2 using a PWM signal but making the right PWM signal so that you will get the right voltages across the load resistor will be a challenge. For accuracy of the voltages you would want a feedback loop but that might interfere with the speed you want to achieve. It might look easy now but trust me, it is not. \$\endgroup\$ – Bimpelrekkie Feb 5 '19 at 12:12
  • \$\begingroup\$ Thanks for your input. It does seem too easy. Why is feedback loop necessary? If rail is at 24V and I supply 50% duty ratio it will be 12V, if i supply 20.8% duty ratio it will be 5V. What other factors can make it so the output voltage is not only the product of the PWM duty ratio? \$\endgroup\$ – mega_creamery Feb 5 '19 at 12:55
  • 1
    \$\begingroup\$ If rail is at 24V and I supply 50% duty ratio it will be 12V For Rload = infinite that would be true. But when that is not the case then currents flow, though the Rdson of the MOSFET and the ESR of the inductor which cause a voltage drop. Since the current can be up to 1.5 A these can not be ignored. Without feedback the output voltage will be unpredictable. Look at designs of DCDC converters, nearly all of them use feedback. \$\endgroup\$ – Bimpelrekkie Feb 5 '19 at 13:01

Your Answer

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

Browse other questions tagged or ask your own question.