0
\$\begingroup\$

I am trying to develop a schematic with minimal components. The board would drive a N-MOSFET from a microcontroller, and I am wondering if I can omit the gate resistors (pulldown for "default" off, series from microcontroller to gate).

The transistor is essentially charging a capacitor whose charge will be used to provide high current for a very short time (20mA at 3V for 15ms to downstream).

The rationale behind omitting the gate resistors is the following:

  • 4.7kOhm current limiting resistor (VCC from a 3V coin cell, limit high current which reduces coin cell life). This would mean low drain current.
  • Transistor very close to microcontroller, expected low parasitic noise.

My question is: would this design work?

enter image description here

\$\endgroup\$
7
  • \$\begingroup\$ Are you sure a N FET works as Q1? The more voltage there is on capacitor, the less there is Vgs to keep transistor on, right? Even an ideal P FET that turns fully on would take roughly 2.35 seconds to charge the 100uF capacitor to 99% via 4k7 resistor. Also do you think 100uF is large enough to give 20mA for 30ms? It would ideally need 6V change on capacitor. \$\endgroup\$
    – Justme
    Dec 22, 2020 at 8:34
  • \$\begingroup\$ I computed the current using I = C*dV/dt, where C=100uF, dV=3V , dt=20ms. You're right, it needs about 2s to charge after 5tau, which is fine. I'm not sure I understand the rest of the reasoning. \$\endgroup\$
    – Mihai Galos
    Dec 22, 2020 at 8:39
  • \$\begingroup\$ EDIT: You're right, redid my calculations, it would drive 20mA for 15ms. \$\endgroup\$
    – Mihai Galos
    Dec 22, 2020 at 8:41
  • 1
    \$\begingroup\$ @MihaiGalos When the N-FET turns on, it's source will begin rising toward VCC. With the gate voltage fixed, coming from the micro, this means the gate-source voltage will decrease, preventing the FET from fully turning on. \$\endgroup\$
    – Unimportant
    Dec 22, 2020 at 8:45
  • 1
    \$\begingroup\$ you can use your MCU as a pull down, so as long as the brief power-on uncertainty period doesn't blow anything up if the FET is one state or the other, it's fine. \$\endgroup\$
    – dandavis
    Dec 23, 2020 at 0:09

1 Answer 1

Reset to default
1
\$\begingroup\$

No, it would not work.

And it is not about gate pull-down or series resistors, these could be left off, as nothing bad should happen even if the gate floats for a few milliseconds after connecting power and before the MCU configures the pin as output. This floating can also be pevented by if there is a large resistor, like in the order of 1 megaohm to keep the idle state.

The problem is that the FET should be of P type instead of N type to switch it on and off reliably.

\$\endgroup\$
2
  • \$\begingroup\$ Thank you for the answer. How about if the capacitor and current limiting resistor are on the high-side (connected to VCC) of the transistor, and the low-side connected to GND? - would driving via N-Channel MOSFET work then? \$\endgroup\$
    – Mihai Galos
    Dec 22, 2020 at 9:08
  • 1
    \$\begingroup\$ Yes that would work. \$\endgroup\$
    – Justme
    Dec 22, 2020 at 9:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.