Operational Amp...
Operational Amp...
Operational Amp...
aerational Amplifiers
10.1 Operationa Amplifiers 10.3 Feedback
Parameters ofPractical Op-Amps
10.2
NTRODUCTION
he nreceding chapter, we saw how the coupling and bypass capacitors determine
then
he lower cutoff frequency of an amplitier. If such capacitors can be eliminated,
m amplifier's frequency
response can go down to dc (zero frequency). Because its
frequency response cannot go all the way down to de, an amplifier that employs couplingg
capacitors is known as an ac amplifier. In contrast to this, if coupling is accomplished
drectly without the use of capacitors, we say that the amplifier is directly coupled
and refer to it as a direct-coupled (dc) amplifier. Note that the two ways in which
the term "dc" is used are, in essence, synonymous. Provided that a dc (direct-coupled)
amplifier has no bypass capacitors, its frequency response will go down to dc (zero
frequency). The elimination of coupling and bypass capacitors from amplifier design
has an extremely important consequence-amplifiers can be more easily fabricated on
IC chips. As a result, amplifiers-in particular, operational amplifers-can be made small
n size and economical in price.
The differential amplifier is a direct-coupled device that is typically the input stage
0 an operational amplifier (op amp.) Because it requires no capacitors tor coupling
can be produced small
ages, the op amp lends itself nicely to IC fabrication. It for numerous linear
S7e and high in reliability. Operational amplifiers are useful and
nonlinear applications as well. They are versatile predictable-
ns,and
they are the basic analog IC.
to
euack can be produced in a cireuit or system by returning portion ofthe output
a
Fig
But due to the ampliñer, v, =A(V2-Vi)=-Avj, so
(10.3)
and substituting into Eq. 10.2, we get
Vj O-
oVo Rin -
Ro
-oVo Ra-(+R)-R, -4+R,)+
Thus,
V20 I)A (-vi) -Ra's (10.4)
Vo
V20- R+(1/A)(R +R)
Drill Exercise 10.1
For the ideal-amplifier circuit given in Fig. 10.2, suppose that R, = 1 k2, R, = 10 k2,
(a)
(b)
each
Fig. 10.1 (a) Operational-amplifier circuit A=100,000, and v, =1 V. (a) Find v,V, i, and i, (6) Find the power absorbed by
symbol, and (b) operational-amplifier model. resistor, the independent voltage source, and the ideal amplifier (i.e., the dependent
voltage source).
Example 10.1
ANSWER (a)- 10.0 V, 0.10 mV, 1.0 mA, - 1.0 mA;
Let us find
v, for the ideal
amplifier circuit shown in
this circuit is shown in Fig. 10.2a. The explicit 1om o 10.0 mW, 1.0 mW, -
10.0 mW
Fig. 10.26. (b) 1.0 mW, -
In the circuit
given in Fig. 10.2, the
noninverting input is at the reference polenh that the gain Abecomes
Circuit such as that in Fig. 10.2, let us consider the
case
Tnat15, VOV. Furthermore, since node v, and node v, are constrained DIrarily large. When A^o, from Eq. 10.4 we have that
sources Dy ve
(independent and dependent,
Currents only at node
v respectively), in using nodal anay e sum
(the inverting input). Since the (10.5)
current, by KCL, amplifier inpui n0
is called
negative
feedback. terminals
must be at the same
obtain
cntial, theri v,
Vy, ofd
eauivaletly, since both input
negative input.
Such a
connection
we must hava. st
cquatiOn,
We
mitstituting this fact ininto the
because of the
infinile
gain propcriy,
the umnlia.cdbnck
bctwcen
In anop-amp circuit, dircctly
source
voltage
resistor and must not
connect a
in Fig. 10.2,
the corresponding op-ump cuit iis usually
circuit
For the circuit given 0 v, and summing the
teminals.
Using the fact
10.3.
that V,
" from which
R
Drill Exercise 10.2
w
For the noninverting amplfier givenin Fig 10.4a, suppose that R, -1 ka, R, =9ka
and v, 1 V. (a) Find v, y, and i2. (b) Find the power absorbed by each resisun, the
0V
independent voltage source, and the op amp.
Fig. 10.3 Op-amp circuit-an inverting amplifier. circuits can be quite useful in some situations in which there
Notice how simple the analysis of the op-amp circuit in Fig. 10.3 is when we use Ooerational-amplifier
the fact that v, = 0 V. Although this result was originally deduced from Eq. 10.4, the than a single input.
is more
combination of infinite gain and feedback constrains the voltage applicd to the op-amp Example 10.3
(between terminals v, and v,) to be 0V. In other words, we must have that v, = V2.
in terms of the input voltages v, and v, for
Let us dectermine the output voltage v,
Example 10.2 the op-amp circuit shown in Fig.
10.5. Since v =0 V, then v, =0 V. By KCL, at the
Consider the op-amp circuit with feedback in Fig. 10.4a. Again, the inputs ofte inverting input,
amplifier draw no current, and so in applying KCL at node vj, we have
from which we get
=0 R
R R
w
R
V V2
Example 10.4
operational ampliñer, the gain and the input resistance are large, but
al
For a Amore practical model of an actual op amp is to have resistace R, (called a
Let us find the output voltage v, of the op-amp circuit with two inni not inn esistance of the op amp) connected series with the dependent voltage
in
Fig. 10.6. inputs is shown in the ouatassuming that an op amp 1s ideal often yields simple analysis with very a
Using the fact that v = V a v , by KCL at the inverting input source.
accurate r e s u l t s .
An actual operational ampliier can be described by the block diagram shown in Fig
RR 7 The first stage of an op amp 1s a dilterential amplifier. Since the input resistance
fa simple differential ampliier isnot necessarily high modified versions of the basic
from which
derential amplifier are used, or FETs are used in place of BJTs. Typically, what results
RV+ Rav=(R, + R)v
icaninput resistance of 1 MQ or more. Following the first stage is a high-gain amplifier
Applying KCL at the noninverting input, we get (10.10) thtsupplies additional gain. The overall gain of an op amp is typically 100,000 (100
dB) or more. After the high-gain amplifier comes a buffer, which is usually an emitter
R follower, and a level shifter. The level shifter is used so that the output voltage is zero
when the input is zero. The last stage of the op amp, the driver, is a large-signal (power)
R
amplifier with a low output resistance-typically 100 2. The driver supplies the output
R
' current and voltage.
Vio Buffer
2 Differential High-gain and
Driver
amplifier amplifier level
20 shifter
input e2 I k3
- Qn
op-amp circuit.
10.9 Practical
Fig. 14
30 pF
24 We can get a more accurate result by not
20 which requires a much
assurming that the opamp is ideal. Doing this,
+Vcc +Vcc greater analysis effort (see Drill Exercise
39 k 10.5). yields
v=100.95v,
which is not significantly different from the
14 the same when rounded otf to three
simpie approach taken above, and is even
significiant digits.
K 15 Q16
50 k2
Q17 Drill Exercise 10.5
For the circuit
2
given in Fig. 10.9, suppose that the ampliñer has
M2, output resistance R,= 75 Q (which is connected in sernesinput
resistance R =
op amp typrcauy When both inputs of a ditferential amplifer have the same input vy =V2= om Called a
of R,=75 Q. of R 2 M2, and an =
oulput resisa o-node input, the output is v,= A(-v)= Avm- V=OV.Thisproperty
These constant
voltages supply power to the actual terential amplitier is of great practical importance. Because of environmental
analyzed. components forming the op amp and
are
ignored wlhen the op-in nological factors, quite often a desired signal is subject to interference. Such
CSTgnals, referred to as noise, often appear inthe common mode.
538 Electronics
perfectly syinmetrical, Cona. erational Amplifiers 539
In actual practice, a
differential amplifier is not
is applied, the
output will
cquently,
the smaller the
be nonzero-the
ter
t h e
can be made zero."
Fora
7 a , a d e
Vis
Cot & 141 op amp, the input offset
o u l p u lv o l t
called
The value
the input offset
the gain Acu for a comm voltage
amp.
a c o m m o n - m o d e input
when differential-input signal
to
node the
op
a nonzero
amp
with
output offset voltage can is typically 1.0 mV.
ratio (CMRR) of the amplifo op et
The ratio ofthe gain 4, for a eof an
be modeled
By
detinilnon, a n
positive
circuít
CMRR 4p (10.13)
amp
0 V. This is
= = nul." These can be used
14cu 0V
V, V2
when
done
1s cxpressed in decihele. ag is tapped) as shown in g a potentiometer
so the larger the CMRR the better.
Nomally, the CMRR an
resistor that
adjustable resistor
orshort)-typically io kM-is
Fig. t10.106. The
adjusted so that v,=0V when v,=potentiometer
v=0V.
CMRRJB 20 log10
AcM (10.14) r"pot"
Since the
(a)
stage of an op amp is a diferential amplifier, an op amp's iputs are
ases of
of two
two BJ
BJTs. For an op amp to function properly, the input BJTs
Furthermore, the absolute value of the differential gain is the bases
vpically "
sed. Thus, there must be a dc path between each base and the
ypcabe
appropriately
biased.
4pl=5x10-3 must
reference (ground). call
Reca that an ac source acts as a short circuit to dc.) Consequently,
75
Ix10-3 eral.there are input (base) bias currents, albeit small currents, in the rwo input
in genera
Thus
pITs The average value of these currents, when the output is zero due to an applied
CMRR-44 37,500 i s called the input bias current of the op amp. For a 741 op amp, the input bias
ACM AcM 2x10 arent is typically 80 nA. the
The absolute value of the difference berween
offset current, and for 741
two input
thethis current
and
hias currents is known
as input a op amp
nA.
CMRRaB 20 logio(37,500) = 91.5 dB is typically 20
Slew Rate
Drill Exercise 10.6 of the output voltage
The slew rate of an op amp is the maximum rate of change
The two input terminals of an of
when the input is a large-signal step voltage. It is expressed in volts per microseconds
op-amp are connected to voltage signals of
745uV and 740uV respectively. The gain of the strengh (Vlus).
and its CMRR is 80 dB. Calculate the op-amp in differential mode is 5x 10
output voltage. SR d v/us
ANSWER 2.500037125 dt max
change in response to
Offset Voltages and Currents rapidly the output of an op-amp
Slew rate indicates how
can
(see
10.4) is set
Fig.
to zero,
what results is
oftset
lts is the op-amp
op-amp ocireuit
voltage for this circuit
show mverinlinFigag AVo
Let us
determine the output
R2 10Ah
R
Ri
R
op amp with
op amp with 2
zero offset
non-zero offset
(a)
First let us find output voltage v, that is the result ot the op amp's input
voltage V offset
As indicated in Fig. 10.10, let us model the nonideal opane
Fig 10.11a by an ideal op amp and a voltage source Vas as shown in Fig. 10.1.1
alternative model is given in Problem 10.26 at the end of this chapter) Ignorine reduced the effects of the input bias curents.
the Addition of Ra
have that ia. 10.13
effect of the base currents, by KCL, we ig 0A
Case 1. iBI # 0A and ig =
Therefore, the
=0A. so i, ig.
=
Since v2-Rziz2=0 V,
resulting output voltage v Ri, Rzig1
=
is =
Solving this equation for v, and using the fact that y = Va We get
#0A
Case 2. ip1 0A ign
=
and the output
Thus we have that
then i v/R =-Rzig/R. =
(10.15) Since
v -R3iz2= V1,
R voltage v,2 IS
Next let us determine the effect of the input bias currents iggo1 and ig ag *yRs +y- -Ra:-1R/a
on the output voltage v To do this, we will model the op amp given in Fig. 10.11 as R the input bias
currents is
due to
shown in Fig. 10.12. Since, in essence, v = 0 V, there is no current through R,. Ths two cases, the output voltage
Combining the
means that the current through R, is igj and (10.16)
Y'otYo=Rzis1 - + R i
V,Raie in value, v, can
be made small by making
This output voltage, however, can be reduced by connecting a resistor g t0 n typically close
Since
i and ig are
(10.17)
noninverling input as shown in Fig. 10.13. We procecd by using the principie
superposition.
R
542 Electronics
that
we casily
determine
Operational Amplifiers 543
rom this cquation,
RR - R, |R
input Output
RR,+ R elfects of the input hbias
V
Amplifier
In other words.
to
minimize the
then from Eq.
currents on
sclect R, =RR If wedo so.
10.17, the the ouu
wtma
voltage is
'n)=R),
Rz'81Rzip: =R;00p1
where is the input
curent
ofset of the op amp. yOutput input
in series with
if a resistor R, 1S placed ne voltage
Notc that ceven Thus, from Eq. 10.15 Fecdback network
10.1source
valid.
Fig. 10.11. Eq.
10.15 will still be and Eq.
voltage v that
results from both
the input offset voltage and
8, the
offset currenm o Series-paralle. feedback amplifier.
(when R; is used) 14
uit shown in Fig. 10.14 is an
op amp is of t shou.
circuit
example of a leedback
feedback amplífiers take.amplifer.
overall
ut of
the ack network is connected in series
feedback
In
particular.
with the amplifier
ce the
the
of the 1feedback network is connected in
and 1, cach be positive or negative, the worst-case
case the input paraliel with the amplifer
Since
has an absolute value of
can
output offscl nd of series-parallel feedback (also called
voltage-series
volag . then this tvpe of feedback offers the most benefits for
hck.n voltage-amplifcation
chall see), the other forms of feedback will be relegated to the
pplications (as
(see Problems 10.36 through 10.44) at the end of this chapter.
(10.19
section
is v
V Since the feedback voltage v,is subtracted from the overal input
and input bias currents of 100 nA and 200 nA. Determine R tage of I my smp
this is an example
of negative feedback.
he worse-case voltage Vn ofthe (overall) feedback amplifier, we use the fact that
the gain Ap v/vn
=
10.3 FEEDBACK
Vo
When a portion of the output of a circuit or system is returned to the input, the result is
known as feedback. The consequences of feedback are both useful and important. In
this section, we study some of them.
Series-Parallel Feedback R2
Let us begin by considering the situation
depicted in Fig. 10.14. Shown is an amplifier
that produces an output voltage v, across a load
R. The output voltage is also applied to
the input of some feedback network
(circuit), and the output voltage v,of the feedback
network is returned to the input portion of the
amplifier. If the amplifier's voltage gain R
is vv A, then the
=
R+R (1020)
from which
R (10.21)
of
routine application nodal analvsie
result is obtained by a
(Ofcourse, the same
which we define as B v./
B of the feedback network,
=
B= R (10.22)
RtR Ap
in0V
-RoF
R2
R Rop
(see
Fig. 10.1l). Since v,= Bv, resistauo
feedback network, then where B R,/(R,+R,) is the gain or
v, A(Vinv)= Avin-Av= AVm -ABv,
Operatlonal Amplfiers 545
(10.23)
(10.24)
endent
AYmmes ndepei ot the op-ammP guin. In the limit as A-, this
which is ly. 10.21, as it should be.
Example 10.8
With the use of q. 10.23, we may now demonstrate how negative feedhack produces
a more stable amplitier gain.
l'or the fecedbuck amplifier given in Fig. 10.15, let 4
200.000, R, , R , 0 2 R I k&2, and R, 100k2. We then have that
T000
0.0099
RtR, 1000 +100,000
200,000
A 100.964
"1+(0.0099)220,000)
of the feedback amplifier. The
which is an inerease of only 0.005 per cent in the gain
is a reduction of voltage gain from
price that we pay for this more stable operation
where nmore gain is necded, further
around 200,000 to about 100. Thus in applications
slages of amplification would be required.
Output
and
Efectetof Feedback n lInput
Is investigate wohat fthe
hapyens to the input resistance R ofithe feedback
cplacing the
pli
is finite, R fier
the op amp
Nent. let us inves resistance R,, ot is
given in Fig. 10.15 when the
input thal ils
oulput
ce
resistance
R, = * We
still
assunning However, sino
op amp with itsmodel (but Ky " i .
definilion,
Fig. 10.17. By
shown in
inuit
then
n R (10.25)
into Eq. 10.25.
10.25 yields
have that V/A. Substituting this
From the fact that = h,, we
Ri A (10.26)
Av
R2
R
Fig. 10.17 Circuit for determining the input resistance of the feedback amplifier
For the (typical) case that R> Ri, from
Eq. 10.23, we have the approximate formula
yA1 +AB). Substituting this into Eq. 10.26 results in
R (1 +AB)R
(10.27)
Thus we see that the input resistance Rip of the feedback
input resistance R, of the op amp and the factor (1
amplifier is the product of the
+AB). This,
negative feedback can be used to increase the input resistance oftherefore,
shows how
Now let us consider the
an
amplifier.
output resistance R, of the op amp in Fig. 10.15, and see
how it affects the
output resistacne Rar of the feedback
we replace the
op amp with its model and set the
amplifier. To determine R,F
circuit is shown in Fig. 10.18, where input voltage to zero. The resulting
a typical situation. Ro=v,Ji,
Let
Then, by voltage division, we the
us assume that
R >> R, which is
get approximation
R v.= Bv
R,+R
Operational Amplifiers 547
A V,*Av,V-v
, R R R
R
y. R /(R +R,)V,_ AB
R2 R
R+R
R R+R
Ro
W
RT
o R
R2 12
WW
10.18 Circuit for determining the output resistance of the feedback amplifier.
oher words, Rr is equal to the parallel combination of R/(1 +AB) and R, +R.
hat 1S,
R, and
1000
B= R = 0.0099
R +R 1000+100,000
From Eq. 10.27, the input resistance of the feedback amplifier is
548 Electronics 10s_.
Operational Amplifiers 549
x
3960M 1
R (1 +ABR,
=
[| + (200,000)(0.0099)](2
resistance
cdback aamplificr
ofthe feedback is A Ap
while from Eq. 10.28, the output
75 = 0.038 Q
14 AB 1+ AB)yo 1+
OHF
RoF Rg
AB 1+(200,000)(0.0099)
A B is the
1 + AB)
+ dback-amplifier
gain for low and middle
ApA(I
ain), gain),
(1 +ABXo s the
aand HF= frequencies
upper cutoffi frequency
Drill Exercise 10.9
where dc
ier. T
feedbackamplifier.
the Therefore,
h as was indicated in Eq. for the
caled
in 10.23, when placed in the
ppose that-
given in kig. 10.15, suppose that the shown Fig. 10.15, the de gain A of the
op amp is divided
For the feedback amplifier 50 kO
op amp Kowever, the upper cuto frequency oy (and, hence, by
100 S2, R, = I kS2, and R2
=
As mentioned earlier,
10.44 at the end of this chapter. gain-bandwid
product f ofan op amp is just the product of its dc gain A and
through frequency (bandwidt S# given in given in hertz. That is,
Theuppercutof.
Frequency Response
to this point, we have assumed that S=Al (1031)
discussion of operational amplifiers prior values otA 200,000
andfr=1 MHz, we obtain
=
to a sinple low-pass
filter (e-g, see Fig. 5.8). BR+R 1000+100,000
response that is essentially equivalent
amp, the gain, referred to as the (bandwidth) is
Specifically, for such an internally compensatedus denote this function by A. Then A
op
upper cutoff frequency
Let and the resulting
is a function of frequency.
open-loop gain, (1+AB)0=[1 +(200,000)%0.0099)(10m)
has the form HF
=6.22 x 10* rad/s
(10.29)
A
1+jo/0 Or
where B- R/R +
R). Simplifying Eq. 10.30, we get Other
ucqual words, the gain-bandwidth product
the op amnp
to the gain-bandwidth product of
Ap
(1+ AB)+j(1+AB)|1+j
(1+ AB)0H
550 Electronics
PROB
LEMS
SOLVED
ADDITIONAL
2F
-tVR
Fig. SP 10.1 Thévenin's theorem applied to an op-
amp circuit. Vc V=0 V
Solution: We have already seen (Eq. 10.6) that Vo
R (a)
To find the output resistance Ro. we set v, to zero and
take the ratio R, = v,i, as depicted in Fig. SP10.1(a).
'()=-dvt)
By KCL dt
- 0
i0
R R2
2 R2
OV
(b)
Fig. SP 10.2 (a) Differentiator circuit, and (b) its
output voltage.
Solution: Because the inverting input is ata
of 0 V, then vc=v. Thus potesti
Henr
Fig. SP 10.1(b) Thévenin-equivalent circuit.
Finally, by KVL we have that v, =-
VR.
from which dv
dt
tmut voltage v,)is the derivative Operational Amplifiers 551
) (multiplied by the constant -1). Solutlon: Voltage
a circuit a differentiator. gain with feedback
A
om
he output voltage , in terms of
t h e
t
1000
333.33
.
850
= 314.8
Integrator
circuit, 1+850x0.002
GP 10.3 Percentage change in overall gain will be
Sincev, =0 V then i =2v. Since the input
ArAp 100
Won:
als ofthe
op amp
draw no current, hen
ic=i =
AF
Mus
333.33-314.8 100
eed-52»0)a = 5.6%
333.33
RC
R
the output voltageis the integral of the input voltage -
muliplied by the constant-1). For this reason such a R
rouit is known as an integrator. This type of circuit M-
extremely useful because it is the backbone of the
R
malog computer.
104 The CMRR of a differential amplifier is
dB. If its gain in differential mode is 1200 then R
alculate its gain in common mode.
Solution: -i
Ap
CMRRaB 20 10810A
Aem Fig. SP 10.6
1200
55 20 loS10 Solution: Using the concept of virtual ground at
Acm
node A andB
Am2.13
W. A
negative feedback of B 0.002 is applied
=
dt
Also n -ti Or the op-amp circuit shown in
(a) v, and
R (b) the resistance Fig. P10.1, find
hn
vi,
R R R 4
As 52
W
3Q
0- v1)+ +R, 2
2
20
2 F gP10.3
nD ccircuit
i r c u i t shown in Fig. P104 is
0p-aimp
1ne
a s a ative-impedance converter. For this
negative-
SUMMARY 10.4
nd(b)
(b) the resistance vji
known
find (a) V,
and
Fig. P10.7
fabricated 5. An op amp typicaly has a differential R
Direct-coupled (dc) amplifiers are easily circuit,
1.
on IC chips. single-ended output. erential input and 10.8 For the
and (b) the
op-amp circuit in Fig. P10.8, find (a)
V
2. Differential amplifiers are dc amplifiers that are 6. The industry standard is the 741 op a resistance vi,
amn
typically the first stages in operational amplifiers (op 7. Negative feedback can be used t
amps). amplifier stability. It also can be
3. Differential amplifiers and, hence, op amps, have
used
voltage ampifier's input and output che aprove 20
the ability to reject common-mode signals cost is a reduction of gain.
(noise). .The W W
This ability is expressed in terms of the common-mode 8. The
gain-bandwidth
is a constant.
product of afeedback R
rejection ratio (CMRR).
4. An ideal operational amplifier has infinite gain,
amplifier
amr
) 52
infinite input resistance, and zero output resistance.
PROBLEMS
R
10.1 For the circuit shown in Fig. P10.1, find v, when
the ideal amplifier (a) is an op Fig. P10.8
amp, and (6) has finite Fig. P10.4
gain A.
10.9 For the op-ump circuit shown in Fig. P10.9, find
0.5 For the op-amp circuit shown in Fig. P10.5, find
the resistance vji,. Vo
(a) , and (5)
32
W
19 29
Fig. P10.2 49 R
Fig. P10.1
10.3 For the op-amp circuit shown in
Fig. P10.3, find
10.2 For the op-amp circuit
shown in Fig. P10.2, find (a) o, and (6) i
(a) v, and (b) i, R
Fig. P10.5
For the op-amp circuit shown in Fig. P10.5,
0.6
mlerchange the 1-2 and 2-Q resistors, and nna (a) o
Fig. P10.9
and (b) the
resistance vi