0
\$\begingroup\$

My SMPS controller can supply about 250mA and 3.3V output.

There are rare but crucial moments when power demand goes above 250mA, voltage sags below 2.7V, and the MCU brown-out occurs.

The question is: if I add a 3V LDO, will it stay inactive at 3.3V until the droop reaches 3V before it regulates? Or is there some kind of a low side switch that will try to “regulate” the 3.3V down to 3V?

TIA

\$\endgroup\$
2
  • \$\begingroup\$ MandoMando , What is the duration of the crucial moments when required current is more than 250mA ? and what is maximum demand current during that duration? Since to support momemtary and sudden load , you can increase capacitance on the output 3.3V . \$\endgroup\$
    – Anil KG
    Jul 31, 2019 at 3:08
  • \$\begingroup\$ @AnilKG it's in the order of a couple of hundred milliseconds. Long enough that caps won't help but short enough that an LDO won't over heat. \$\endgroup\$
    – MandoMando
    Jul 31, 2019 at 14:10

1 Answer 1

3
\$\begingroup\$

I'd suggest you use a 3.3V LDO with a series SCHOTTKY diode such as the SDM40E20LS or SDM40E20LA dual diodes. These diodes have very low Vf values of around 310mV @100mA.

Connecting two in parallel will get you about 1A capability at 410mV drop.

enter image description here

The LDO may have some minor supply current (about 10mA) if your SMPS drops by 0.1V during normal operation, but otherwise should tend toward zero.

The two diodes in parallel won't share current exactly, but well enough to keep the load voltage well above your brownout level.

\$\endgroup\$
2
  • \$\begingroup\$ Unless your SMPS has an OR-ing diode or mosfet, I would suggest adding a diode to the SMPS output as well. In this schematic the LDO will provide power into your SMPS which can damage it, or at least throw off its feedback loop. \$\endgroup\$
    – Stiddily
    Jul 31, 2019 at 12:08
  • 1
    \$\begingroup\$ @Stiddily I'd suggest this is incorrect ….the SMPS is a Buck converter so you could add a diode to the input if you were so inclined, but I doubt that is necessary. The only condition that you would IMO need to investigate is the startup of the SMPS, since here the linear regulator would also be starting and will have its own risetime. Specifically addressing your concern about the feedback, the SMPS is already in an overcurrent mode (since the output voltage is dropping so it has already moved to maximum current delivery into its output capacitor, so no disturbance there. \$\endgroup\$ Jul 31, 2019 at 15:22

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.