Power switch using nmos and pmos

[jmhautbois]

Hi everyone,

My question seems to be posted frequently, but as far as I try, I can't find out my design error...
The goal is to have a temperature driven circuit, using a LM35, a schmitt trigger and nmos+pmos, I would like to let the VCC go or block it.
The simulation below simulates the LM35 using a sine wave, between 0 and 1V (0°C and 100°C) : it should change its state when temperature is above 70°C, and get back to normal when temperature goes down 50°C.


I am using LTSpice in order to help me validate the principles, and it works well, except the pmos usage.
As you can see, the output of nmos is good (5V when temperature is above 70°C, 0V when temp is less than 50°C) but the output of pmos is not ok : it is always around 5V...

What's wrong in my design ?

Thanks a lot for your help !

 

[dick_freebird]

For starters, your PMOS is upside down (rigged as a
source follower but the polarity is backwards). In a
real life setup you'd be conducting through the body
diode no matter what, getting about 4.5V +/-. You
also show no load resistance, so it would pull right
up to the supply.

 

[jmhautbois]

OK, I changed my PMOS, and know, added a load, and I always have 0V.
I am quite sure this is obvious...

- - - Updated - - -

OK, got it, I needed to specify a PMOS model.
When using a IRF7210, and a 1ohm load, I get 5A, which is what I am looking for !


Is this circuit OK, or should I add some specific parts in order to have a good design ?
I think about diodes, R, C, etc. ?