For a battery powered ultra low consumption application I designed a classical high side switch using the complementary MOSFET FDMA1032CZ. this is the schematic:

enter image description here

The idea is simple:

  • before shipping, the circuit consumption is nearly 0[A] (only MOSFET and battery leaks),
  • then the user powers on the circuit using switch S1,
  • the microcontroller starts running and activates Q1-N and Q1-P via "PWR_ON" signal,
  • the circuit is then continuously powered on by the battery through the mosfet switch.

I need assistance to calculate the optimal value for both pull-up and pull-down resistors (R1 and R2). For reminder, those 2 resistors are mandatory to set voltage level when circuit is powered on.

In my opinion those resistors must be :

  1. as large as possible to minimize current when the 2 mosfets are on,
  2. but not too big to make sure the 2 mosfets can switch fully on.

When looking at datasheet, it appears that the maximum MOSFET gate body leakage IGSS_max = +/-10[uA] and the minimum gate threshold voltage VGSth_min = +/-0.6[V].

According to previous point 2), the resistor value should NOT be greater than R_max = VGSth_min/IGSS_max = 0.6/10e-6 = 60[kohm].

For a battery voltage of 3V, this would result in a total consumption through the resistors of: I = 2x V_BAT/R = 2x 3/60e3 = 100[uA], which is far too big...

I would appreciate if someone could confirm and/or correct my calculation.


It seems your calculations are correct.

This IGSS is untypically high for mosfets. I think this is because of zener diodes on gate. Typical value is ±100 nA.

|improve this answer|||||

The common reasons for including that resistors are as follows:

R2: To avoid leaving the gate of Q1-N in a floating state and thus avoid unwanted conduction in specific situations . For example, when the microcontroller is powered but not yet executed the code that configures the I/O system (in your case, to write a high value on PWR_ON) or in forced reset caused by noise and watchdog activation.

R1: As Q1-N cannot impose a high level in the Q1-P gate, the turn-off of this mosfet depends exclusively on the discharge of its GS capacitor through R1.

The exact value of each resistor is not critical. However, they influence the mosfet turn-off time (Q1-P) and determine the battery consumption when conducting (Q1-N and Q1-P). In fact, depending on the circuit used not even a resistor between G and S for Q1-P is necessary (or between the base and emitter the case of a bipolar transistor). This is true when using well defined voltages levels on gate-source. See figure below, for alternatives:

Power Control

|improve this answer|||||

Your Answer

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

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