1
\$\begingroup\$

I am a bit confused. In saturation region, sometimes, an MOSFET is called voltage-controlled current source and sometimes it is called variable resistor where its resistance is controlled by the gate-source voltage. Are the two concepts equivalent? I meant MOSFET operates in saturation region not triode.

\$\endgroup\$
  • 3
    \$\begingroup\$ Where did you see someone say that a MOSFET can be used as a variable resistor in the saturation region? \$\endgroup\$ – The Photon Apr 18 '16 at 15:56
1
\$\begingroup\$

A MOSFET acts like a variable resistor in the linear region and a VCCS in the saturation region. In saturation, the VCCS has a small-signal output resistance due to channel length modulation, but it's not the same behavior as the linear region.

In saturation, the drain current is given by the equation:

$$I_D = k(V_{GS} - V_{T})^2(1 + \lambda V_{DS})$$

We can calculate the drain-source conductance (1/resistance) as:

$$G_{sat} = \frac {dI_D}{dV_{DS}} = \lambda k (V_{GS} - V_T)^2$$

In the linear region, the drain current is given by:

$$I_D = k[2(V_{GS} - V_T)V_{DS} - V_{DS}^2]$$

If \$V_{DS} << V_{GS} - V_T\$, this reduces to:

$$I_D = 2k(V_{GS} - V_T)V_{DS}$$

and the conductance is:

$$G_{lin} = \frac {dI_D}{dV_{DS}} = 2k(V_{GS} - V_T)$$

By themselves, \$G_{sat}\$ and \$G_{lin}\$ don't look all that different. To my mind, the main differences are:

  1. Saturation assumes that \$V_{DS} > V_{GS} - V_T\$, and you need a current source to model that.

  2. The linear region is used for switching. A resistor model is useful for estimating power dissipation when the switch is on.

  3. Lambda is typically very small (~0.01), so the saturation output resistance is a second-order effect. In the linear region, \$V_{DS}\$ is just as important as \$V_{GS}\$.

\$\endgroup\$
1
\$\begingroup\$

I'm pretty sure they are the same as it's a voltage controlled current source, but I've never seen that exact description. A current constraining source would be a resistor. Usually, as MOSFET as a resistor is an ohmic device, so I pushed through the math as if the device was exclusively a saturated MOSFET resistor (ie: mobility controlled device) and you get the following: $$R_{ON}=res\left(\frac{L}{t_{inv}W}\right)=\left(\frac{1}{n_{inv}q\mu_n}\right)\left(\frac{L}{t_{inv}W}\right)$$ This then becomes $$R_{ON}=\frac{1}{\mu_n C_{ox}\frac{W}{L}\left(V_{gs}-V_{th}\right)}$$

because \$ q n_{inv} \mu_n =Q_{inv}=C_{ox} \left(V_{gs}-V_{th}\right)\$ There's your zeroth order analysis.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.