Cascade Op Amp Circuits

CDR Charles B. Cameron

October 6, 2006

1 Background 2 Cascade Connection

This document addresses three related topics: Figure 1 shows two circuits that do not include op
amps.
• When we connect the output of one circuit with
The Thévenin-equivalent voltage for Circuit 1 is
gain G1 to the input of a second with gain G2 ,
we cannot simply compute the overall gain as R2
G = G1 G2 . It would be simple if we could do it, VTH1 = vo1 = vi , (1)
R1 + R2 1
but we cannot.
its Thévenin-equivalent output resistance is
• If the two circuits include op amps, then there is
a condition that, if satisfied, would let us com- RTH1 = R3 + R1 ||R2 , (2)
pute the overall gain as G = G1 G2 .
• There is a way to guarantee that the condition is and its gain is
satisified, so we really can compute G = G1 G2 . R2
G1 = . (3)
The connection of op amp circuits in the manner de- R1 + R2
scribed above is known as cascading them. It lets
Similarly, the Thévenin-equivalent voltage for Cir-
us analyze each circuit separately and use the results
cuit 2 is
in a simple manner to build more elaborate circuits,
greatly easing both the analysis of existing circuits RL
and the design of circuits of our own. VTH2 = vo2 = vi , (4)
R4 + RL 2

vi1
RTH R4
R1
R4
vo2
R3 vi2 vo2
vo1 +
vTH1 RL
RL
R2 −

(a) Circuit 1 (b) Circuit 2

Figure 2: Use of the Thévenin-equivalent circuit for
Figure 1: Two circuits without op amps. Circuit 1 to obtain the overall gain.

1
 ROUT1 ROUT2
+ + + + + +
VRIN RIN + GV VRIN RIN + GV v
vi − 1 RIN1 − 2 RIN2 o
− −1 1
− −2 2
− −

Figure 3: Two cascaded amplifiers, 1 and 2. Each amplifier i is assumed to have its own distinct gain Gi ,
input resistance RIN i , and output resistance ROU T i .

its Thévenin-equivalent output resistance is model of an op amp. However, we do not assume that
the gains G1 and G2 are large, nor do we assume that
RTH2 = R4 ||RL , (5)
the input resistances RIN 1 and RIN 2 are large, nor
and its gain is do we assume that the output resistances ROU T 1 and
ROU T 2 are small.
RL
G2 = . (6) It is easy to see, though, that by the voltage divider
R4 + RL rule
The Thévenin-equivalent circuit for Circuit 1 can RIN 2
be used as shown in Figure 2 to obtain the overall VRIN 2 = G1 VRIN 1 (10)
ROU T 1 + RIN 2
gain: RIN 2
VRIN 2 = G1 vi . (11)
RL ROU T 1 + RIN 2
vo2 = vTH1
RTH + R4 + RL It follows that
µ 1 ¶µ ¶
RL R2
= vi1 . vo = G2 VRIN 2 (12)
(R3 + R1 ||R2 ) + R4 + RL R1 + R2
(7) RIN 2
= G2 G1 vi (13)
ROU T 1 + RIN 2
From this it is easy to see that the overall gain is
µ ¶µ ¶ so the overall gain of the two circuits combined is
R2 RL
G= . RIN 2
R1 + R2 (R3 + R1 ||R2 ) + R4 + RL G = G1 G2 (14)
ROU T 1 + RIN 2
(8)  
Note that this is not at all the same as 1
= G1 G2  . (15)
µ ¶µ ¶ R
1 + ROU T1
R2 RL IN 2
G1 G2 = . (9)
R1 + R2 R4 + RL If we could find a way to force
 
This is unfortunate because if it were the same we
could get G very easily by first obtaining G1 and G2  1 =1
R
(16)
separately and then multiplying them together. 1 + ROUIN
T1

then the overall gain would simply be

3 Cascade Amplifiers G = G1 G2 . (17)
Now consider the arrangement in Figure 3, where This is the simple form that was impossible when we
two blocks containing amplifiers are connected in cas- did not have amplifiers in the circuit, as in Figure 1.
cade. Each amplifier circuit is similar to the non-ideal

2
 I R2
R1 R2


− R2
+ vo
D I RIN C ROUT
vi R1 i

TEST
vi


 + ↓ + +
Figure 4: A non-inverting op amp configuration. I R1 VD RIN +− AV vTEST
− −
D
−
4 Meeting the Condition
Figure 6: Applying the non-ideal op amp model to
We can satisfy the condition of (16) if we require that
the inverting op amp configuration in order to deter-
ROU T 1 ¿ RIN 2 . There are two indpendent ways of
mine its output resistance. Applying a test voltage
meeting this condition. We can either make ROU T 1
at the outputs, calculating the resulting test current,
very small or we can make RIN 2 very large.
and taking their ratio lets us calculate the output
Consider the non-inverting op amp configuration resistance ROUT = vtest /itest .
of Figure 4. Because the op amp has a very large in-
put resistance (almost no current enters its inverting
and non-inverting terminals), using this configuration
allows us to meet the condition of (16).
Now consider instead the inverting op amp config- To obtain the output resistance, we assume that
uration of Figure 5. Recognizing that the presence of the input vi = 0, effectively putting R1 and RIN in
negative feedback around the op amp causes the volt- parallel. Then we use Kirchoff’s current law at node
age at the inverting terminal to be a virtual ground, D, apply Ohm’s law, and rewrite the equations using
the only resistance seen at the input terminal is that nodal voltages. Note that VC = AVD . We get
due to R1 . This resistance is not necessarily very
large, so when using this configuration as the second
of two cascaded amplifier circuits we could only meet IR1 − IRIN − IR2 = 0 (18)
the condition of (16) by insisting that the output re-
sistance of the preceding circuit be very small. VD VD vD − vTEST
− − − =0 (19)
We might consider the output resistance of the in- R1 RIN R2
vTEST
verting configuration itself. Certainly its input resis- vD = ³ ´. (20)
tance is not satisfactory. We can obtain the output R2 R11 + R1IN + R12
resistance by applying the non-ideal op amp model,
as shown in Figure 6.

With this information we can calculate the test cur-

R2 rent:
R1
−
vi + vo
vTEST − VC
iTEST = − IR2 (21)
ROUT
vTEST − AVD vTEST − VD
= − (22)
Figure 5: An inverting op amp configuration. ROUT R2

3
  
vTEST  A
iTEST = 1− ³ ´  (23)
ROUT R2 R11 + R1IN + R12
 
vTEST  1
− 1− ³ ´
R2 R 1
+ 1 + 1
2 R1 RIN R2
(24)
µ
1 1
= vTEST −
ROUT R2

1/R2 − A/ROUT 
+ ³ ´ . (25)
R2 R11 + R1IN + R12

Now we have what we need to calculate the output

resistance:
vTEST
Ro = (26)
iTEST
1
= 1/R
. (27)
1
− 1
+ “ 2 −A/ROUT ”
ROUT R2 R2 R1 + R1 + R1
1 IN 2

Now A is very large—on the order of 105 or 106

or more, so the denominator of this expression is a
very large negative number. As a result, Ro is a very
small negative number: practically zero.
In particular, it is much smaller than the output
resistance ROUT of the op amp itself.
This kind of analysis of the output resistance of an
amplifier that uses an op amp generally shows that
such circuits have a very small output resistance.
Thus whether we use a non-inverting op amp con-
figuration to achieve large input resistance or obtain
a very low output resistance by using an op amp, the
conditions of (16) are met. This lets the very conve-
nient calculations of (17) apply.