CMOS Common Source Amplifier.
As was mentioned in Lecture 30, there are two different environments in which MOSFET amplifiers are found, (1) discrete circuits and (2) integrated circuits (ICs). We will now begin to look at the IC MOSFET amplifiers. There are three basic configurations of IC MOSFET amplifiers:
As was also mentioned in Lecture 30, large-valued resistors and capacitors are not often used in these IC environments. Instead, active loads are incorporated using MOSFETs as loads. In the amplifier circuits shown above, the active loads are actually the nonideal current sources. [Also notice that there are no bypass capacitors as we saw with discrete MOSFET (and BJT) amplifiers.]
We will look at all three of these amplifiers more closely over the next few lectures. The intention is to pair the discrete version of the MOSFET amplifier with its IC version. Since we’ve covered the CS amplifier in discrete form already, we’ll begin with the analysis of the CMOS CS amplifier.
CMOS Common Source Amplifier
An example of a complementary MOSFET amplifier is shown in
In this circuit, Q2 and Q3 form a PMOS current mirror. Because both PMOS and NMOS devices are used in this circuit, it is called a complementary MOS (CMOS) circuit. In addition to forming part of the current mirror, Q2 also functions as the current source load (aka active load) for Q1. For Q2 to be a current source, Q2 must operate in the saturation mode, of course. The output resistance ro2 of Q2 is
It is helpful to observe the characteristic curve for Q2 to understand its active-load role:
Referring to the CS amplifier circuit above in Fig. 6.18(a), when i = I REF then V GD2=2 (by symmetry with Q1). This implies that v = VSG, which is the Q point shown in Fig. 6.18(b). Furthermore, it is useful to observe the graphical construction of the transfer function vO/vI for this amplifier, as illustrated in Figs. 6.18(c) and (d) shown below. The drain currents of Q1 and Q2 are the same. The operating point of the amplifier is found
Furthermore, it is useful to observe the graphical construction of the transfer function vO/vI for this amplifier, as illustrated in Figs. 6.18(c) and (d) shown below. The drain currents of Q1 and Q2 are the same. The operating point of the amplifier is found from the intersection of the Q1 characteristic curve with the load curve of Q2 for a particular vGS1:
Collecting these intersections from this figure as vGS1 ( I v = ) changes, we can construct point-by-point the transfer characteristic curve for this amplifier:
From this plot, we can see that Region III shows a linear relationship between vO and vI. This is the region where the circuit of Fig. 6.18(a) can be used as a linear amplifier.
Small-Signal Voltage Gain and Output Resistance
Now we’ll determine the small-signal voltage gain and output resistance of this amplifier. The small-signal equivalent circuit for this CMOS CS amplifier is:
It is important to recognize that no small-signal model is needed for Q2 because its affect on the signal vo can be incorporated using the small-signal resistance ro2 as shown above. So, at the output
V0 = - gm1 Vgs2(r01||r02
while at the input
Vgs1= V t
Substituting (3) into (2) gives the open circuit small-signal voltage gain for the CMOS CS amplifier to be
or substituting for gm1, ro1, and ro2
Since ro1 and ro2 are usually large, this Avo gain is typically relatively large (approximately -20 to -100, or so). Neat! We have incorporated the effects of relatively large resistance for this amplifier without having to actually construct a large resistor. From the small-signal model we see from inspection that
Rout =r01||r r02
Summary for CMOS CS amplifier:
1. Very large input resistance.
2. Very large output resistance.
3. Potentially large small-signal voltage gain.
Example N33.1 (similar to text exercise 6.15). A CMOS CS amplifier shown in Fig. 6.18(a) is fabricated with W/ L=100 um/ 1.6 for all transistors. With kn' = 90 uA/V2, kn' = 30 uA/V2 IREF = 100 uA, VAn =8 V/ìm, and VAp= 12V/ìm, determine the following quantities: (a) Find gm1. The common expression for gm we use is
For a MOSFET in the saturation mode
Substituting (7) into (6) gives the transconductance for Q1 in terms of ID1 to be
This form of gm was actually used earlier in (5).] Because the amplifier is biased so that IREF= ID , then
(b) Find ro1.
(c) Find ro2.
(d) Find Avo.
This value represents the largest gain. The gain will be reduced when an actual load is attached to the amplifier