# ESP Auto Power Off Circuit

Thread Starter

#### MMM**MMM

Joined May 30, 2022
91
I have recently simulated the circuit demonstrated in the provided video using Multisim. While the circuit appeared to function correctly in the video, the simulation results indicate that it is not working as intended. The output remains high irrespective of the switch state. The left switch is used solely to simulate the scenario when the microcontroller sends a high signal. I would greatly appreciate any insights or guidance on identifying the issue with the circuit. Thank you in advance for your assistance.

#### crutschow

Joined Mar 14, 2008
33,334
Add a 1kΩ load from the MOSFET drain to ground.
Never leave a connection to a MOSFET floating (the oscilloscope proves little or no simulated load).

Note that the maximum Vgs threshold voltage of Q4 is 3V (for an ID of 250µA), so it may barely turn on with only 3.3V gate-source voltage.
Q4 should be a logic-level MOSFET with a max Vgs of ≤2V.

Thread Starter

#### MMM**MMM

Joined May 30, 2022
91
When S3 is closed, the circuit works as shown in the image

But when the switch S4 is closed and S3 is open, the circuit does not work, closing S4 is similar to send a high signal from the esp to just keep the power on. What I do is: first close S3, then close S4, and finally open S3. After opening S3, the voltage drops to zero

Regarding the P-MOS, thanks for grabbing my attention to this. But I need to understand something: Vds of Q4 is -3.3V, Vgs is -3 and the threshold is -3V. So, according to the saturation condition: |Vds| >= |Vgs| - |Vth| => 3.3 > 0. Thus, should it be on in this case or off? The condition is met.

Also, if you have any feedback regarding Q5 or any other component in the circuit please mention it. Thanks in advance.

#### crutschow

Joined Mar 14, 2008
33,334
But I need to understand something: Vds of Q4 is -3.3V, Vgs is -3 and the threshold is -3V. So, according to the saturation condition: |Vds| >= |Vgs| - |Vth| => 3.3 > 0. Thus, should it be on in this case or off?
It would be slightly on, since Vgs(ths) is defined as occurring at an Id of 250µA.
To be fully on with minimum on-resistance, Vgs should be at least 4.5V (below):

#### crutschow

Joined Mar 14, 2008
33,334
After opening S3, the voltage drops to zero
Q5 has a maximum Vgs(ths) of 4V.
It may not be turning on with only 3.3V applied.
You need to be using logic-level type MOSFETs with a max Vgs(ths) of ≤2V.
And you need to look at the data sheets of the parts your are using, so you don't makes such mistakes.

Thread Starter

#### MMM**MMM

Joined May 30, 2022
91
It would be slightly on, since Vgs(ths) is defined as occurring at an Id of 250µA.
To be fully on with minimum on-resistance, Vgs should be at least 4.5V (below):

View attachment 298342
Does MW4953D work instead of Q4?

Last edited:
Thread Starter

#### MMM**MMM

Joined May 30, 2022
91
Q5 has a maximum Vgs(ths) of 4V.
It may not be turning on with only 3.3V applied.
You need to be using logic-level type MOSFETs with a max Vgs(ths) of ≤2V.
And you need to look at the data sheets of the parts your are using, so you don't makes such mistakes.
Does FQP30N06L work instead of Q5?

#### crutschow

Joined Mar 14, 2008
33,334
Does MW4953D work instead of Q4?
Yes, that should work fine, since it is nearly fully turned on with Vgs=2.5V
Does FQP30N06L work instead of Q5?
That should also work, since the drain current is low (current through R9).
it wouldn't work if you wanted the minimum on-resistance at high current, which requires Vgs=5V or greater.