2
\$\begingroup\$

I am trying to design a power supply for portable device and stumbled upon a problem that is probably trivial, but it's a tough one for me (having no formal electrical education).

The device is powered from battery, but if I plug USB connector, the P channel MOSFET in Q1 (MF5853CS) closes and disconnects battery and the device is powered through Schottky diode in Q1. I pretty much copied the design in Microchip's AN1149 (Designing A Li-Ion Battery Charger and Load Sharing System With Microchip’s Stand-Alone Li-Ion Battery Charge Management Controller).

The design is working fine when battery is fully charged, but if it's not charged (3.8 V and lower) and I unplug the USB connector, there is a short drop in voltage to around 3.04 V which is too low for U2 (power supply supervisor CAT809, RESET is low if voltage lower than 3.08 V) and it in turn turns off my buck converter for around 140 ms and that causes other problems further down in my design. My load is approx. 100 mA

enter image description here enter image description here

In my quest to solve this, I placed Schottky diode D1 and lowered value of resistor R18, but that introduced another problem and that is upon connecting USB power, Schottky diode in U8 gets shorted out and now conducts in both directions (diode testing with my meter show 0.19 V in one and 0.75 V in opposite direction. Why that happens is beyond me.

enter image description here

I could probably get around it if I connect the U2 supervisor directly to battery, but it doesn't seem right. Or maybe if I lower the inrush current for capacitor C7 with a diode and resistor would that help mitigate the drop of voltage?

And now after I wasted number of U8 components (MOSFET + Schottky) I'm more than ever curious why that happened and how to prevent that.

\$\endgroup\$
3
  • \$\begingroup\$ Welcome! _ 0.19 V in one and 0.75 V in opposite direction”_ In circuit or loose? It should not have broken. \$\endgroup\$
    – winny
    Feb 22, 2023 at 21:03
  • \$\begingroup\$ @winny thank you! you are correct, I took it out of the board and the shottky is ok, but I think the mosfet is dead. I measured 0.7V and around 2.7kohm between drain and source in both directions. And above all I have 2.7V on VBUS with no USB connected, just from battery :( \$\endgroup\$
    – zdenek
    Feb 23, 2023 at 10:02
  • \$\begingroup\$ "but I think the mosfet is dead" Don't. Measure and verify instead. \$\endgroup\$
    – winny
    Feb 23, 2023 at 11:00

1 Answer 1

0
\$\begingroup\$

If you measured the diode drop across the Schottky diode, it is also in anti-parallel with the body diode of the MOSFET, which would read about 0.7V. You might be able to add a large capacitor across C7 to hold the voltage up during switch-over. 100 mA at 5V is equivalent to 50 ohms, so 1000 uF would be a 50 ms time constant - probably more than long enough. You might want to add a diode and resistor in series with the capacitor as follows to avoid excess current surge.

schematic

simulate this circuit – Schematic created using CircuitLab

Transition hold-up circuit

\$\endgroup\$
1
  • \$\begingroup\$ thank you. on my first version of the board without the D1 diode (and on the second one before I connect USB) I get reading (0.2V) only in one direction. I took U8 out of the board and the shottky is ok, but I think the mosfet is dead. I measured 0.7V and around 2.7kohm between drain and source in both directions. And above all I have 2.7V on VBUS with no USB connected, just from battery.... I had tried to add large capacitor before, but it was just trial and error, no math behind it, and I stopped at 470uF cap (and without the resistor to control current) I will try larger capacitor then. \$\endgroup\$
    – zdenek
    Feb 23, 2023 at 10:26

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.