NMOS inverter

[alphacat]

Hello.
Given the following NMOS inverter, I need to find VOH.
Could you please help me on that?
Thank you.

**broken link removed**

 

[Hayato]

Considering a very small load, I'd choose 15V for VOH, given that Vin = 0V.

Anyway.

ML is an active load, it is probably biased in the saturation region (remember, for a FET saturated = BJT active). So it will act as a resistence.

Remember that the current through ML = the current through MI = I Total.

And, Vout = VDS (MI) = 15V - VDS (ML).

If MI is at cutoff (no current), then I total = 0 and VDS(ML) = 0 => Vout = 15V

Of course, as you increases Vin, the VDS(ML) is going to increase very little, but you would have to make the calculations. (Remember ID (ML) = ID (MI)).

 

[alphacat]

Hi Hayato.

ML is an active load, it is probably biased in the saturation region

I think that you meant that it is in the linear region.
I believe its in the linear region (VGS(L) - VTn(L) < VDS(L)) since:
ID(L) = 0A
=>
VGS(L) <= VTn(L)

If MI is at cutoff (no current), then I total = 0 and VDS(ML) = 0 => Vout = 15V

Why do you conduct that if ID(L) = 0A then VDS(L) = 0V?
It is enough that VGS(L) = VTn(L) (or just a bit less than VTn(L)) for ML not to pass current through its channel.

So Vout doesnt need to reach 15V for ID(L) to be 0A.

 

[alphacat]

Here is a simulation I performed.
As you can see, Vout never reaches 15V.

**broken link removed**

Here, Vout reaches max voltage of 10.41V.
I assume that this is VOH of the circuit, isnt it?

 

[Hayato]

FETs have 2 regions. Triode and Saturation.

Ok, your simulation is not correct, because you are using 3-terminal MOSFETs, so that you have body-effect (reduced gain). (I'm not saying your result is wrong, just the way the simulation was made).

To simulate this circuit you MUST use 4-terminal MOSFETs and connect the body to the GND. And do not forget to consider the parameters given by the exercise.