1
\$\begingroup\$

I'm trying to design a power latch circuit for a microcontroller. The typical power latch circuit is powered by a button, but I want to power mine with a small accelerometer which sends out an interrupt. The typical power latch circuit also makes the microcontroller power itself. The problem in my scenario: the interrupt is only 3 milliseconds long and the microcontroller needs to long to boot up, meaning I have to rely on other components to keep the power latch high until the microcontroller is powered. I came up with this circuit so far and wanted to know if it is acceptable or complete garbage :D Power Latch Circuit that normally "stays on" and is switched off on GPIO HIGH

The idea behind this is to reverse normal power latch dynamics with the GPIO pins, i.e. the MCU doesn't have to have a HIGH-pin to ensure it is staying enabled, but it only relies on a HIGH-pin for power-off (S1).

Please let me know if this circuit is fine, because I'm pretty much just getting started with all this symbol schematic stuff. Thanks a lot!

\$\endgroup\$
1
  • \$\begingroup\$ I believe LT makes a single-chip solution for this sort of thing; the search term you want is "monostable" or perhaps "pulse stretcher". \$\endgroup\$
    – Hearth
    Commented Oct 8, 2023 at 22:35

1 Answer 1

1
\$\begingroup\$

Probably simplest latch for you purpose is this. To trig the latch with accelerometer it has to have a push-pull output (not an open-collector).

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$
2
  • \$\begingroup\$ i'm trying to simulate this and it seems the only way the N-MOSFET stays on is via the MCU pulling its gate high. Is that the intended & expected behaviour? I feel like it shouldn't be. As a note /& correction: my accelerometer has a 3.3V (VCC) INT pin with configurable push-pull / open-drain. EDIT: nevermind... i forgot to adjust the MOSFET values. It does work like expected. EDIT2: I also had to change both R1 and R2 to 100k, otherwise the simulation wouldn't work. \$\endgroup\$
    – M4GG1
    Commented Oct 9, 2023 at 5:23
  • \$\begingroup\$ If accelerometer has open drain you can connect its output to M2 drain to trigger the latch. \$\endgroup\$ Commented Oct 9, 2023 at 5:29

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.