I have the following circuit components on a PCB with an N-channel MOSFET.
I want to convert the circuit to P-channel MOSFET controlled by an ESP pin with minimum external components. Maybe a few resistors or diodes, I want to avoid another FET.
A problem is that ESP pin won't handle voltages >3.6V even though I could use open-drain (high impedance when not used) I can't be certain that this might not change from SW and damage the pin

Any ideas ?

 

Hi Jimis,
Welcome to AAC.
The problem with your circuit is that the PMOS will not switch OFF.
To turn it Off the Gate would have to be +13V.
Have you a general purpose transistors in your parts box.?
E

 

Hi Jim,
Adding a NPN transistor as shown in this image.
The +3.3V from the ESP will turn On the PMOS.
E

 

Thanks for the reply Eric.
I understand that a transistor would solve my problem but because the pcb already exists I was thinking if there is a simpler way to avoid difficult components to place like the transistor. Maybe resistors or diodes. The actual circuit on the PCB is the following (N-channel which I want to make P-channel)

 

Thanks for the reply Eric.
I understand that a transistor would solve my problem but because the pcb already exists I was thinking if there is a simpler way to avoid difficult components to place like the transistor. Maybe resistors or diodes. The actual circuit on the PCB is the following (N-channel which I want to make P-channel)

If the pcb already exists then your source and drain pins will be in the wrong place.

 

So let me guess...

The original design uses a NMOS as a high side switch and you are not getting the needed voltage at the load.

You need to show more of the schematic.

 

I was thinking if there is a simpler way to avoid difficult components to place like the transistor.

Off-hand, don't see anyway to avoid adding a transistor.
Just dead-bug glue it to the PCB and wire to the P-MOSFET.

 

Hi Jim,
As this a prebuilt PCB, I guess you are aware if you do not invert the ESP input signal to the NMOS the output sense into the load will be inverted?
E

 

I didn't want to complicate it but here it is.
Mains power supply feeds U1 and U3 which charges the Lead Acid Bat (1st pic). There is leakage current from the 12V to the battery that's why diode Vdrop is -ve but it's not a problem.

When mains PS is off the circuit (pic 2) still works powering the peripherals from the Lead Acid Bat (although Vgs is -ve !) but the voltage drop at FET is 1.88V. Obviously I had to use a P-ch FET.

 

I didn't want to complicate it but here it is.
Mains power supply feeds U1 and U3 which charges the Lead Acid Bat (1st pic). There is also a charging current from the 12V to the battery that's why diode Vdrop is -ve but it's not a problem.

When mains PS is off the circuit (pic 2) still works powering the peripherals from the Lead Acid Bat (although Vgs is -ve !) but the voltage drop at FET is 1.88V. Obviously I had to use a P-ch FET.

Why do you need to switch the power from the battery?

 

I want to be able to switch everything off to save battery. Needs to be able to kill itself.
Now that I see it I will have to modify Eric's circuit so that when ESP shuts off the P-FET it should stay that way until Mains power comes back up

 

Scratch my last post Eric's circuit should work fine

 

Can you have a look at my Spice project ? Doesn't seem right.
I tried to import the N-ch and P-ch MOSFETs but I might not have done it correctly

 

Hi Jim,
Try this and we will edit as edit required.

E

 

Hi,
This sim is using an actual PMOS.
I am searching for the model MOS stated in your original sim.
E

 

If you make R2=1k it will make Vgs go out of the limits that si2301cds (PMOS) can handle (+/- 8V)

 

Hi Jim,
Choose R1 and R2 values to give the acceptable voltage range for the PMOS.
Post your asc sim when ready.
E.

Is there a reason you have chosen those two MOS types?

 

Hi Jim,
I have created models of the MOSFET's from your zip file.
This is what LTSpice shows.
NoteR1/2.
E

 

I have those FETs. I don't think the models are correctly inserted in SPICE. Those voltages don't look right.
With 3V3 the NMOS (DMN3023L) Vds should be very low. Now it shows 11V !
Unless my calcs are wrong R1 and R2 should be as I have it my project.

 

Please add the component files also in the zip file. They are issing