772 Chapter 9 Frequency Response

The formula for R CL can be seen to be simply

R CL = R L || R o = R L′ (9.156)

The formula for R gs is the most difficult to derive, and the derivation should be performed
with the hybrid- π model explicitly utilized. The result is

R sig + R s
R gs  ------------------------------------------
- (9.157)
⎛ ro ⎞
1 + g m R s ----------------
⎝ r o + R L⎠

When R sig is relatively large, the frequency response will be dominated by the Miller
multiplication of C gd . Another way for saying this is that C gd R gd will be the largest of the
three open-circuit time constants that make up τ H ,

τ H = C gs R gs + C gd R gd + C L R CL (9.158)

enabling us to approximate τ H as

τ H  C gd R gd (9.159)

and correspondingly to obtain fH as

1
fH  ------------------------ (9.160)
2 π C gd R gd

Now, as R s is increased, the gain magnitude, A M = G m R′L , will decrease, causing R gd to

decrease (Eq. 9.154), which in turn causes fH to increase (Eq. 9.160). To highlight the trade-
off between gain and bandwidth that R s affords the designer, let us simplify the expression for
R gd in Eq. (9.154) by assuming that G m R′L  1 and G m R sig  1,

R gd  G m R′L R sig = A M R sig

which can be substituted in Eq. (9.160) to obtain

1
fH = ---------------------------------- (9.161)
2 π C gd R sig A M

which very clearly shows the gain–bandwidth trade-off. The gain–bandwidth product remains
constant at

1
Gain–bandwidth product = A M fH = ------------------------ (9.162)
2 π C gd R sig
 9.9 Other Wideband Amplifier Configurations 773

In practice, however, the other capacitances will play a role in determining fH , and the gain–
bandwidth product will decrease somewhat as R s is increased.

EXERCISE

9.30 Consider a CS amplifier having g m = 2 mA/V , r o = 20 kΩ , R L = 20 kΩ , R sig = 20 kΩ ,

C gs = 20 fF , C gd = 5 fF , and C L = 5 fF . (a) Find the voltage gain A M and the 3-dB frequency
fH (using the method of open-circuit time constants) and hence the gain–bandwidth product. (b)
Repeat (a) for the case in which a resistance R s is connected in series with the source terminal with
a value selected so that gm Rs = 2 .
Ans. (a) −20 V/V, 61.2 MHz, 1.22 GHz; (b) −10 V/V, 109 MHz, 1.1 GHz

9.9.2 The CD–CS, CC–CE and CD–CE Configurations

In Section 7.6.1 we discussed the performance improvements obtained by preceding the CS
and CE amplifiers by a buffer implemented by a CD or a CC amplifier, as in the circuits
shown in Fig. 9.39. A major advantage of each of these circuits is wider bandwidth than that
obtained in the CS or CE stage alone. To see how this comes about, consider as an example
the CD–CS amplifier in Fig 9.39(a) and note that the CS transistor Q 2 will still exhibit a
Miller effect that results in a large input capacitance, C in2 , between its gate and ground.
However, the resistance that this capacitance interacts with will be much lower than R sig ;
the buffering action of the source follower causes a relatively low resistance, approximately
equal to a 1 ⁄ g m1 , to appear between the source of Q 1 and ground across C in2 .

VCC VDD
VDD

I2 I2
I2
Q1 Q1
Q1

Q2 Q2 Q2

I1 I1 I1

(a) (b) (c)

Figure 9.39 (a) CD–CS amplifier. (b) CC–CE amplifier. (c) CD–CE amplifier.
774 Chapter 9 Frequency Response

Example 9.15

Consider a CC–CE amplifier such as that in Fig. 9.39(b) with the following specifications: I 1 = I 2 =
1 mA and identical transistors with β = 100, fT = 400 MHz, and C μ = 2 pF . Let the amplifier be
fed with a source Vsig having a resistance R sig = 4 kΩ , and assume a load resistance of 4 kΩ. Find the
voltage gain A M , and estimate the 3-dB frequency, fH . Compare the results with those obtained with a
CE amplifier operating under the same conditions. For simplicity, neglect ro and rx .

Solution
At an emitter bias current of 1 mA, Q 1 and Q 2 have
g m = 40 mA/V
r e = 25 Ω
β 100
r π = ------ = --------- = 2.5 kΩ
gm 40
g gm
C π + C μ = -----m- = ----------
-
ωT 2 π fT
–3
40 × 10
= --------------------------------------6- = 15.9 pF
2 π × 400 × 10
C μ = 2 pF
C π = 13.9 pF
The voltage gain A M can be determined from the circuit shown in Fig. 9.40(a) as follows:
R in2 = r π 2 = 2.5 kΩ

R in = ( β 1 + 1 ) ( r e1 + R in2 )
= 101 ( 0.025 + 2.5 ) = 255 kΩ

V b1 R in 255
-------
- = ---------------------
- = ------------------ = 0.98 V/V
V sig R in + R sig 255 + 4
V b2 R in2 2.5 -
-------- = ---------------------
- = -------------------------- = 0.99 V/V
V b1 R in2 + r e1 2.5 + 0.025
V
-------o- = – g m2 R L = – 40 × 4 = – 160 V/V
V b2
Thus,
V
A M = -------o- = – 160 × 0.99 × 0.98 = – 155 V/V
V sig

To determine fH we use the method of open-circuit time constants. Figure 9.40(b) shows the cir-
cuit with Vsig set to zero and the four capacitances indicated. Capacitance C μ 1 sees a resistance
Rμ 1 ,
R μ 1 = R sig || R in
= 4 || 255 = 3.94 kΩ
 9.9 Other Wideband Amplifier Configurations 775

Rsig Vb1
Q1 Vo

re1 Vb2 RL
Vsig  Q2

Rin
Rout1 Rin2

(a)

Rsig
Q1

Cp1 Cm2
Cm1 RL
Q2
Cp2

(b)
Vo

Cm
Rsig
RL

Vsig  Cp


(c)

Figure 9.40 Circuits for Example 9.14: (a) the CC–CE circuit prepared for low-frequency, small-signal
analysis; (b) the circuit at high frequencies, with Vsig set to zero to enable determination of the open-circuit
time constants; (c) a CE amplifier for comparison.

To find the resistance R π 1 seen by capacitance C π 1 we refer to the analysis of the

high-frequency response of the emitter follower in Section 9.7.2. Specifically, we adapt Eq.
(9.133) to the situation here as follows:
R sig + R in2
R π 1 = ---------------------------------
-
R sig R in2
1 + ------- - + ---------
r π 1 r e1
4000 + 2500
= ---------------------------------------- = 63.4 Ω
4000 2500
1 + ------------ + ------------
2500 25
776 Chapter 9 Frequency Response

Example 9.15 continued

Capacitance C π 2 sees a resistance R π 2 ,

R π 2 = R in2 || R out1
R sig
= r π 2 || r e1 + --------------
β1 + 1
4000
= 2500 || 25 + ------------ = 63 Ω
101
Capacitance C μ 2 sees a resistance R μ 2 . To determine R μ 2 we refer to the analysis of the frequency
response of the CE amplifier in Section 9.5 to obtain
R μ 2 = ( 1 + g m2 R L ) ( R in2 || R out1 ) + R L
4000
= ( 1 + 40 × 4 ) 2500 || ⎛ 25 + ------------⎞ + 4000
⎝ 101 ⎠
··
= 14,143 Ω  14.1 kΩ

We now can determine τ H from

τH = Cμ 1 Rμ 1 + Cπ 1 Rπ 1 + Cμ 2 Rμ 2 + Cπ 2 Rπ 2
= 2 × 3.94 + 13.9 × 0.0634 + 2 × 14.1 + 13.9 × 0.063
= 7.88 + 0.88 + 28.2 + 0.88 = 37.8 ns
We observe that C π 1 and C π 2 play a very minor role in determining the high-frequency response. As
expected, C μ 2 through the Miller effect plays the most significant role. Also, C μ 1 , which interacts
directly with ( R sig || R in ) , also plays an important role. The 3-dB frequency fH can be found as fol-
lows:
1 1
fH = ------------ = ---------------------------------------
- = 4.2 MHz
2 πτ H 2π × 37.8 × 10 –9

For comparison, we evaluate AM and fH of a CE amplifier operating under the same conditions.
Refer to Fig. 9.40(c). The voltage gain A M is given by

R in
A M = ---------------------
- ( –gm RL )
R in + R sig
rπ
= ------------------
- ( –gm RL )
r π + R sig
2.5
= ---------------- ( – 40 × 4 )
2.5 + 4
= – 61.5 V/V
R π = r π || R sig = 2.5 || 4 = 1.54 kΩ
R μ = ( 1 + g m R L ) ( R sig || r π ) + R L
= ( 1 + 40 × 4 ) ( 4 || 2.5 ) + 4
= 251.7 kΩ
 9.9 Other Wideband Amplifier Configurations 777

Thus,
τH = Cπ Rπ + Cμ Rμ
= 13.9 × 1.54 + 2 × 251.7
= 21.4 + 503.4 = 524.8 ns
Observe the dominant role played by C μ . The 3-dB frequency fH is

1 1
fH = ------------ = ------------------------------------------
- = 303 kHz
2 πτ H 2 π × 524.8 × 10 – 9

Thus, including the buffering transistor Q 1 increases the gain, A M , from 61.5 V/V to 155 V/V—a factor
of 2.5—and increases the bandwidth from 303 kHz to 4.2 MHz—a factor of 13.9! The gain–bandwidth
product is increased from 18.63 MHz to 651 MHz—a factor of 35!

9.9.3 The CC–CB and CD–CG Configurations

In Section 7.6.2 we showed that preceding a CB or CG transistor with a buffer implemented
with a CC or a CD transistor solves the low-input-resistance problem of the CB and CG
amplifiers. Examples of the resulting compound-transistor amplifiers are shown in Fig. 9.41.
Since in each of these circuits, neither of the two transistors suffers from the Miller effect,
the resulting amplifiers have even wider bandwidths than those achieved in the compound
amplifier stages of the last section. To illustrate, consider as an example the circuit in

VCC

VCC C1 VDD

B1
I Vi Q1 I
C1 C2 Vo E Vo
B1 B2 B2
Vi Q1 Q2 Q2 VBIAS Vi Q1 Q2

E C2 Vo
Rin
I
2I 2I

VEE VEE VSS

(a) (b) (c)

Figure 9.41 (a) A CC–CB amplifier. (b) Another version of the CC–CB circuit with Q2 implemented
using a pnp transistor. (c) The MOSFET version of the circuit in (a).
778 Chapter 9 Frequency Response

Fig. 9.41(a).10 The low-frequency analysis of this circuit in Section 7.6.2 provides for the
input resistance,
R in = ( β 1 + 1 ) ( r e1 + r e2 ) (9.163)
which for re1 = re2 = re and β 1 = β 2 = β becomes

R in = 2r π (9.164)

If a load resistance R L is connected at the output, the voltage gain Vo ⁄ Vi will be

V α2 RL
- = 1--- g R
----o- = ------------------ (9.165)
Vi r e1 + r e2 2 m L
Now, if the amplifier is fed with a voltage signal Vsig from a source with a resistance R sig ,
the overall voltage gain will be
Vo 1 R in ⎞
- = --- ⎛ --------------------
------- - (g R ) (9.166)
Vsig 2 R in + R sig⎠ m L
⎝
The high-frequency analysis is illustrated in Fig. 9.42(a). Here we have drawn the
hybrid-π equivalent circuit for each of Q 1 and Q 2 . Recalling that the two transistors are
operating at equal bias currents, their corresponding model components will be equal (i.e.,
r π 1 = r π 2 , C π 1 = C π 2 , etc.). With this in mind the reader should be able to see that Vπ 1 =
– Vπ 2 and the horizontal line through the node labeled E in Fig. 9.42(a) can be deleted. Thus
the circuit reduces to that in Fig. 9.42(b). This is a very attractive outcome because the
circuit shows clearly the two poles that determine the high-frequency response: The pole at
the input, with a frequency f P1 , is
1
f P1 = ------------------------------------------------------------ (9.167)
C
2 π ⎛ -----π- + C μ⎞ ( R sig || 2r π )
⎝2 ⎠

and the pole at the output, with a frequency fP2 , is

1
f P2 = ------------------- (9.168)
2 π Cμ RL
This result is also intuitively obvious: The input impedance at B 1 of the circuit in Fig. 9.42(a)
consists of the series connection of r π 1 and r π 2 in parallel with the series connection of C π 1
and C π 2 . Then there is Cμ1 in parallel. At the output, we simply have R L in parallel with
Cμ .
Whether one of the two poles is dominant will depend on the relative values of R sig and
R L . If the two poles are close to each other, then the 3-dB frequency fH can be determined
either by exact analysis—that is, finding the frequency at which the gain is down by 3 dB—
or by using the approximate formula in Eq. (9.68),

1 1
fH  1 -----2- + -----2- (9.169)
f P1 f P2

10
The results derived for the circuit in Fig. 9.41(a) apply directly to the circuit of Fig. 9.41(b) and with
appropriate change of variables to the MOS circuit of Fig. 9.41(c).
 9.10 Multistage Amplifier Examples 779

Rsig B1 C1

Vsig

B2 C2 RL

Vo

(a)
Rsig B1 C2


Vsig RL Vo

(b)

Figure 9.42 (a) Equivalent circuit for the amplifier in Fig. 9.41(a). (b) Simplified equivalent circuit. Note
that the equivalent circuits in (a) and (b) also apply to the circuit shown in Fig. 9.41(b). In addition, they can
be easily adapted for the MOSFET circuit in Fig. 9.41(c), with 2rπ eliminated, Cπ replaced with Cgs, Cμ
replaced with Cgd , and Vπ replaced with Vgs.

EXERCISE

9.31 For the CC–CB amplifier of Fig. 9.41(a), let I = 0.5 mA, β = 100 , C π = 6 pF , C μ = 2 pF , R sig =
10 kΩ, and R L = 10 kΩ . Find the low-frequency overall voltage gain AM , the frequencies of the
poles, and the 3-dB frequency fH. Find fH both exactly and using the approximate formula in Eq.
(9.169).
Ans. 50 V/V; 6.4 MHz and 8 MHz; fH by exact evaluation = 4.6 MHz; fH using Eq. (9.169) =
5 MHz.

9.10 Multistage Amplifier Examples

We conclude this chapter with the frequency-response analysis of the two multistage ampli-
fiers we studied in Section 8.6. As we shall see, these are relatively complex circuits: Sim-
ply replacing each transistor with its high-frequency, equivalent-circuit model will make it
exceedingly difficult for pencil-and-paper analysis, and will most certainly not lead to any
analysis and design insight. Rather, we will use the knowledge and experience we have