Hi all!,
I have these 2 circuits, the one with the P-Mosfet works properly, the other with N-Mosfet doesn't work.

I use to On/Off a circuit pushing a button. Pressing the button the Mosfet starts drive and supply a microcontroller. One IO from the micro is put to HIGH and connect to "1" so the mosfet continue driving. If you press the button for a while, the signal goes to another IO to the micro and the software out a LOW to "1" disconnecting the circuit.

What I'm doing wrong?

 

Possibly, too many diodes. Remember, an emitter follower introduces 1 standard silicon diode voltage drop between the base and emitter. In the n-channel circuit, the signal at "1" goes through two diodes before reaching the gate. AND, that voltage drop is added to whatever the actual uC output pin voltage is. If the output is only 3 V, then the voltage at the gate is less than 2 V.

Look at the conduction curves in the 540 datasheet to see what the part requires for reliable switching. The solution might be to change to a "logic level" MOSFET that is designed to work with lower Vgs.

ak

 

the transistor @ the n mosfet is not able to lift the gate.

The current flows collector diode 1MOhm resistor to gnd.
This will disable the NPN transistor to generate >0.6V between base and collector hence no conductance.
rebuild with PNP and replace 1MOhm with a smaller one.
or
keep NPN connect collector to gate with resistor to +5

Picbuster

 

Hi all!,
I have these 2 circuits, the one with the P-Mosfet works properly, the other with N-Mosfet doesn't work.

I use to On/Off a circuit pushing a button. Pressing the button the Mosfet starts drive and supply a microcontroller. One IO from the micro is put to HIGH and connect to "1" so the mosfet continue driving. If you press the button for a while, the signal goes to another IO to the micro and the software out a LOW to "1" disconnecting the circuit.

What I'm doing wrong?

The N fet is being turned on by a Npn Emitter follower so it wont work.

Replace the transistor for a Pnp 2N2907 to pull the gate to positive.

 

The suggestions in posts 3 and 4 will invert the logic polarity. That is, it will take a logic 0 at the "1" input to turn on the n-MOSFET, and a logic 1 to turn it off.

or
keep NPN connect collector to gate with resistor to +5

This would make the manual pushbutton part of the circuit not work. The original circuit has a passive diode-OR circuit at the gate. That won't work if the gate is clamped to GND with an NPN switch.

ak

 

Correct I only addressed the fet problem not the working of the switches.
I assumed that Pepe333 is able to solve this pushbutton circuit.

Picbuster