PROBLEM SOLUTIONS CHAPTER 1.

Solution 1.1. (a) Charge on one electron: -1.601910-19 C. This means that charge on 1013 electrons
is: -1.601910-6 C. Net charge on sphere is: 1.601910-6 C (POSITIVE).
Solution 1.2. (a) 1 atom -4.64610-18 C. By proportionality, 64g NA atoms .
3.1NA
3.1g ? atoms 3.1g
atoms .
64
Total Charge = 4.646 10 18

C
3.1 6.023 10 23

atoms = 1.3554 10 5 C
atom
64

(b) Total charge per atom is -4.64610-18 C. Total charge per electron is 1.601910-19 C. Therefore,
there are 29 electrons per atom of copper.
Q
Q 1.36 105
(c) 0.91 A 0.91 C/s. i =
t =
=
= 1.49 105 sec .
t
i
0.91
3.1NA
= 2.9174 10 22 atoms in the penny. Removing 1 electron from
64
3.1NA
3.1NA
0.05
atoms means removing 0.05
electrons. Therefore,
64
64

(d) We know there are

Net charge = 0.05

3.1NA
1.6019 1019 = 234C
64

Solution 1.3 (a) 7.573 1017 1.6019 10 19 = 0.1213C

0.1213
= 121.3A flowing from right to left.
103
(c) Again, use proportionality:
(b) Current =

x 1.6019 10 19
10 60
10 A =
x=
= 3.75 1021
19
60sec
1.6019 10
dq
(d) i( t ) =
= 1 e 5t A. This is an exponential evolution with an initial value of 0, a final value of 1,
dt
and a time-constant of 1/5 (signal reaches ~63% of its final value in one time-constant).

i(t)
1

0.2

time in sec

(e) Current is the slope of the charge waveform. Therefore, by inspection:

Solution 1.4 (a) 6.0231023(-1.601910-19) = 9.65104 C.

(b) Current flows from right to left (opposite electrons), and:
I=

9.65 10 4
= 9.65 107 A
3
10

(c) Using proportionality:

5A =
(d) i( t ) =

x 1.6019 1019
60sec

x=

5 60
21
19 = 1.87 10
1.6019 10

dq
= 1+ 0.5 cos(t ) i (1sec ) = 1 1.57 = 0.57A . Current flows from left to right.
dt

Solution 1.5 (a) i( t ) = 1 4e 2t + 3e 3t t 0. Then

t

q(t ) = i(t )dt = 1 4e

0

+ 3e

)d =

]t

4 e
0

= t 4 0.5e 2

d + 3 e 3 d
0

]0 + 3[0.333e3 ]0 = t + 2e2t e3t 1

t

(b) By inspection:

(c) q(t ) = 120cos(120t ) . Hence

i( t ) =

dq
= 120 120sin (120t ) = 14400sin(120t ) A
dt

Solution 1.6. (a) i(t) = 1 cos(t) A. Hence

t

q(t) = i()d = (1 cos()) d = t

t
1
1
sin() = t sin(t) C

(b) Charge is integral of current. Graphically, the charge at time t is the area under the current curve up to
time t: (note the quadratic nature between 2 and 4 seconds)

Solution 1.7
Again, Q is the running area under the current curve. Between 0 and 3 seconds, current decreases linearly
until zero. So, Qtot = 7.5 C. From 0 to 6: Qtot = 7.5 + Q3_6 = 7.5 -1/10.5 + -1/10.5 + -11 = 5.5 C,
where the curve from 3 to 6 was divided into two triangular sections and one rectangular one.
Solution 1.8 Charge is the area under the current curve. Thus, Q = 0.1*4 0.1*2 = 0.2 C.
Solution 1.9 Calculate the change in energy for the electron: E = Q V = 3.21810-15.
Equate this to kinetic energy:
3.218 10 15 =

1 2
mv v = 8.4 10 7 m / s
2

where the mass of an electron, 9.106610-31 has been substituted.

3

Solution 1.10 P = VI. Hence I = P/V = 210 /120 = 16.6667 A

PROBLEM Solution 1.11 (a) It is necessary to integrate the i(t) curve to obtain q(t). We do this
interval by interval:
t

(i) 0 t 1 ms, q(t) = 0 + d = t C

0

(ii) 1 ms t 2 ms, q(t) = 1 2 d = 3 2t C

1

(iii) 2 ms t 3 ms, q(t) = 1+ d = 3 + t C

2

(iv) 3 ms t 5 ms, q(t) = 0 + (8 2 )d = 8t t 2 15 C

3

(v) 5 ms t, q(t) = 0 C

1
0.8
0.6

Charge in micro C

0.4
0.2
0
-0.2
TextEnd
-0.4
-0.6
-0.8
-1

3
Time in ms

(b) Voltage is the ratio of the power and current curve. In this case, the division can be done graphically
by inspection. Note that the ratio of a quadratic function and a linear function is a linear function:

2
1.5

Voltage in V

1
0.5
TextEnd

0
-0.5
-1

3
Time in ms

Solution 1.12 (a) VA = P/I = 20/4 = 5 V

(b) PB = VI = 27 = 14 W
(c) VC = P/I = -3W/3A = -1V
(d) ID = P/V = -27W/3V = -9A
(e) IE = P/V = 2/1 = 2A
(f) PF = VI = -45 = -20W
In all of the above, note that the direction of the current flow relative to the polarity of the voltage across a
device determines whether power is delivered or absorbed. Power is absorbed when current flows from the
positive terminal of the device to the negative one.
Solution 1.13 (a) By inspection: Circuit Element (CE) 1 absorbs 5W, and CE 2 absorbs 6W.
(b) Compute power absorbed by all elements including independent sources:
I3A :
-15
CE1: -5
V3V : -12
CE2: +6
V5V : 10
I2A :
16
----------

Sum: 0 (Verifies conservation of power.)

Solution 1.14 (a) Compute power absorbed:
I5A :
-85
CE1: 98
V3V : 33
CE2: 16
V7V : -42
I2A :
-20
------------Sum: 0
(b) Add all terms:

( )
V-source:
Pabsorbed = 2(3e t 1) = 6e t + 2 watts
CE1: Pabsorbed = 3e t 3(1 e t ) = 9e t 9e2t watts
CE2: Pabsorbed = (3e t 1)(3e t 1) = 9e 2t 6e t + 1 watts
I-source:

Pabsorbed = 3 1 e t = 3 + 3e t watts

Simple algebraic manipulation of the the sum of all the above terms reveals that the result is zero.
Solution 1.15 (a) When IL = 1, P = VLIL = (16-4)1 = 12 W. When IL = 2, P = VLIL = (16-16)1 = 0.
(b) P = (16-4I L2 )IL. Differentiate this w.r.t. IL and set to zero: 16 12IL2 = 0. Therefore, IL = 1.155A.
Solution 1.16 (a) When IL = 2, P = (16-4)2 = 24W. When IL = 3, P = (16 - 9)3 = 21 W.
(b) Maximum occurs in the interval from 0 to 4: P = (16 - IL2 ) IL
Differentiate w.r.t. IL and set to zero: 16 3IL2 = 0.
Therefore, IL = 2.31 A.
Solution 1.17 (a) Power is the product of the current and voltage. We can compute the product
graphically:

10

Power in Watts

8
6
4
TextEnd
2
0

0.5

1.5
Time in s

2.5

(b)
t

10
10
7 t
W (t ) = p(t )dt = 10 10e 7 d = 10 ]t0 10
= 10t + e 7t
7 e
0
7
7

This can be used as an aid to plot the work function:

9
8
7

Energy in J

6
5
4
TextEnd
3
2
1
0

0.5

Solution 1.18 (a) Since, i( t ) = 115 23t mA ,

1
Time in s

1.5

23t 2
3
q( 7) = i( ) d = 115t
10 = 0.2415 C
2 0
7

(b) Energy is the integral of power:

7

E = p(t )dt = v( t ) i(t )dt = 25 i(t )dt

= 25 0.2415 = 6.0375 C
Solution 1.19 (a) t = 100o F, Rate of temp. increase is 2.5 Wh/o F per gallon:
Energy = 2.5Wh/o F/gallon100 o F30gallons = 7500 Wh = 2.7107 J.
(b) Heater generates P = 12010 = 1200 W. We want 7500 Wh. Therefore, the total number of hours
needed is 7500Wh/1200W = 6.25 h.
Solution 1.20 First compute the change in temperature required, in o F:
t = 80-25 = 55o C = 559/5 o F= 99 o F
Next, compute the energy spent every hour, which means on 40 gallons of water:
E = 2.5 Wh/o F/gallon99o F40gallons = 9900 Wh
Since the heater is not 100 % efficient, we spend more energy than is actually needed to heat the water:
E_spent = 9900 Wh/0.9 = 11000 Wh
So, far, this was the energy spent every hour. Over six hours, the total energy spent is:
E6h = 110006 = 66,000 Wh
Finally, the total energy spent per month is Em = 66,00030 = 1980 kWh
and the bill is 1980 kWh0.14$/kWh = $277.2

Solution 1.21
Energy = 120 W 6 h = 720 Wh = 0.72 kWh
Therefore, cost per day = 0.72 kWh 8 = 5.76 cents, and cost per month is 5.7631 = $1.785.

Solution 1.22

We need to compute the difference between the inner diameter of the tube and the outer one in order to get
the cross-sectional area:
2
2
area = Rout
Rin
= 0.0032 0.00182 = 1.81 10 3 m2

Then, R = 1.710-5(12/1.81) = 11.3 m.

Solution 1.23 L = 20 m, W = 0.015m, H = 0.001 m. Thus, A = WH, and R = 5.1

copper

L/A

= 0.116 .
Solution 1.24. (a) 500 ft, 20 gauge wire: 10.35 /1000 ft from table 1.3. This implies that
R = 5.175 .
(b) 55 ft, 20 gauge, nickel wire:
10.35
R = 5.1
55 = 2.9
1000
(c) Rtot = 2.9 + 5.175 = 8.08 .
Solution 1.25. R(T ) = R(20 )[1 + (T 20)] . Substituting at T = 10 yields:
21 = R(20)[1+ 0.0039( 30)] or R( 20) = 23.78
Evaluating at T = +10 yields,
R(10) = 23.78 + 23.78 0.0039 ( 10 ) or R(10) = 22.85

Solution 1.26. For tungsten, we know that = 0.0045. Therefore:

R(150) = R (20)[1 + (T 20)] = 200[1 + 0.0045(150 20)]
= 317
Rate of change of resistance is (317-200)/(150-20) = 0.9 /o C.

Solution 1.27. Plug numbers directly into the same formula as problem 1.26:
0.0022 = 0.002 + 0.0020.0039(T-20)
Rearrange to obtain: T = 45.64o C.
Solution 1.28. (a) Power in a wire: P = I2R. Rearranging, we can express the current as
I=

R.

Substitute given P and R to obtain I = 0.707 mA.

(b) Use the same formula for current obtained above to get 50 A.
Solution 1.29. Use formula for power: P = V2/R. Rearranging, R = V2/P = 96 .

Solution 1.30 (a) I = V/R = 12 A, out of the positive terminal of the battery.
(b) Up through the resistor.
(c) Absorbed power by resistor: P = V2 /R = 14.4 W. Same power is delivered by source.
(d) From table 1.2 and 1.3, 1000 feet of 18 AWG aluminum wire has resistance:
R1000ft = 1.6*6.51
R1000ft = 1.0416e+01
By proportionality, 1000 0.1 = L 10.416 . Hence,
L = 100/10.416
L = 9.6006e+00 meters.
Solution 1.31 (a) V = 10 V.
(b) P = V2 /R, which means that R = V2 /P = 100/25 = 4 .
(c) I = V/R = 10/4 = 2.5 A. Current flow is downwards through resistor.
(d) Up through resistor.
(e) P = V2/R10 = 100/10 = 10 W. Hence, I10 = V/R10 = 1 A. Without applying material from a future
chapter, a legitimate way to obtain Isource is to apply conservation of power first and then compute Isource
from the power formula. Hence, Psource = 10 + 25 = 35 watts. Using material from a later chapter, in
particular KCL, we may conclude that, Isrc = 2.5 + 1 = 3.5 A. Thus, Psource = VIsource = 103.5 = 35 W.
This approach indicates that power is conserved.

Solution 1.32 (a) From 0 to 1 s, i(t) = 10-3t. Thus, i 2 R = 106 t 2 R is the power absorbed during this
interval. Integrating this expression for the power from 0 to 1 s gives us the total energy used:
1

10 6 t 3 R
10 6
= 5000
= 0.001667 J.
3
3
0
Finally, we need to multiply this by 2 to account for the interval from 1 to 2 seconds. Thus, the total energy
spent is 3.33 mJ.
(b) The same charge that got transported in one direction during the interval from 0 to 1 is being
transported back in the interval from 1 to 2 (by symmetry). Therefore, total charge transfer is zero.
Solution 1.33. (a) 60 W + 120 W = 180 W.
(b) P = IV I = P/V = 180/12 = 15 A.
(c) P = Energy/Time Time = 1.2 MJ/180 W = 6.67104 sec = 1.85 h.

Solution 1.34. P = I 2R. Therefore, 325 = 25(5+4+2R). Solving for R, yields R = 2 .

Solution 1.35. (a) Use definition of power and substitute given power:
V2 = P R = 98 2 = 14 V
Similarly, I3 =

P
12
=
= 2 A, V4 =
R
3

P R = 16 4 = 8 V, I5 =

768.8
= 12.4 A, and
5

V6 = 486 6 = 54 V.
(b) Ptot = P dissipated = 98 + 12 + 16 + 768.8 + 486 = 1380.8 W.
V
(c) Vin = V2 + V6 = 68 V. Iin + I3 = I5 + I4 and I4 = 4 . Thus, Iin = I5 + I4 I3 = 12.4 A.
4
Solution 1.36. (a) Sources are the top, right-most, and bottom left. The reason is that current flows out
of the positive terminal of the device.
(b) The 32/16 element is a 2 resistor. The 54.5/18.167 element is a 3 resistor. The 13/2.167 element
is a 6 resistor. The 93/2.833 element is a 32.827 resistor. The 24/5 element is a 4.8 resistor.

Solution 1.37. Power: 12 = Ix 2 R, which means that R = 12/Ix 2 . Now, analyze the loop: 16 = Ix (R+4).
12
Substitute the value of R into this expression: 16 = I x 2 + 4Ix . Hence: Ix2 4I x + 3 = 0 .
Ix
This equation has two solutions: one is at Ix = 1 A or R = 12 . The other is at Ix = 3 A or R = 4/3 .
Solution 1.38. (a) Conservation of power:
16I x = 4I 2x + 12 + 10 6
Hence
2
0 = Ix2 4I x + 4 = ( Ix 2)

Thus, Ix = 2 A.
(b) 32Ix = 4I 2x + 28 Ix2 8Ix + 7 = ( Ix 7)(I x 1) = 0 . Hence, Ix = 7A or Ix = 1A .
Solution 1.39. (a)
(i) AA:
I = 36/12 = 3A
BB:
I = 24/12 = 2A
CC:
I = 14.4/12 = 1.2A
(ii) Sum = 6.2A
(iii) P = VI = 6.212 = 74.4W. This is equal to the sum of the powers absorbed by the bulbs.
(iv) R = V/I
AA: R = 12/3 = 4
BB: R = 12/2 = 6
CC: R = 12/1.2 = 10
(b) Each AA bulb draws 3 A. Thus, up to five bulbs can be connected without blowing the fuse (53=15).
So, 6 or more would blow the fuse.
(c) Similar analysis suggests that 13 or more bulbs would blow the fuse. Intuitively, the bulbs draw less
current, so more of them can be used.
Solution 1.40. (a)
p(t) = i 2 (t )R = [ 20cos(2t )] 10 = 4000
2

1 + cos(4t )
= 2000(1 + cos(4t )) W
2

(b)
t

W (t ) = p(t )dt = 2000t + 2000 cos( 4t )dt = 2000t +

2000sin (4t )
J
4

6000
5000

Energy in J

4000
3000
TextEnd
2000

1000
0

0.5

1.5
Time in s

2.5

Solution. 1.41. When the switch is closed, a constant current of 5/10000 = 0.5 mA flows through the
circuit. When the switch is open, no current flows. So, 50% of the time, a 0.5 mA current flows, and the
other 50% no current flows. The average current is therefore 0.25 mA.

Solution. 1.42 When the switch is at A, the current is 5/5000 = 1 mA. When the switch is at B, the
current is 5/10000 = 0.5 mA. Now, the switch is at position A 20% of the time (1ms out of a 5ms period,
after which the events repeat). So, the average current is 0.2*1 mA + 0.8*0.5 mA = 0.6 mA.
2

Solution. 1.43 The current in the load resistor is 2 A. So, the power is 2 RL = 8 W.
Solution 1.44. Vin = IinR1 Iout = Vin /R2 = IinR1/R2.
Solution 1.45 (a) I1 = Vin /R1 . Hence, Vout = Vin R2 /R1 .
(b)
Vout R2 100 10
=
=
= 5 R1 = 200
Vin
R1
R1
(c)
Power gain =

2 Vin2
R2
R12
Vin2
R1

2 R2
= 500
R1

Solution 1.46 (a) V1 = 200 mA 5 = 1 V implies V2 = 0.88 = 6.4 V. Hence

Vout = 56.4 = 32 V and Iout = 32/64 = 0.5 A.
(b) Current Gain = 0.5/0.2 = 2.5.
(c) Power values for the 5, 8, and 64 resistors are, respectively, P5 = 0.2 W, P8 = 5.12 W, P64 = 16 W.
Solution 1.47 (a) I1 = 5A I2 = 35/3 = 5 A Iout = 25 A, Vout = 50 V.
(b) Voltage Gain = 5.
(c) Pin = 552 = 50 W, P1 = 553 = 75W, P2 = 2550 = 1250W.

Solution 1.48 I1 = Vin /10 = 0.1 A, VR = 10(Vin /10)R = R; Vout = 5R10 = 50R =50RVin
Vout/Vin = 50R. If we want Vout/Vin to be 150, R has to be 3 .

SOLUTIONS CHAPTER 2
SOLUTION 2.1. Using KCL at the center node of each circuit:
(a) I3 = I2 I1 = 1 2 = 3A
(b) I3 = I1 + I2 I4 = 2 1 0.5 = 0.5A
SOLUTION 2.2. KCL at the bottom node gives I1 = 7 8 = 15A , and at the right node
I4 = 6 8 = 14 . From these, KCL at the top node gives I3 = I 4 5 = 19A, and finally at the central
node gives I2 = 6 + I3 7 = 20 A,
SOLUTION 2.3. Use a gaussian surface on the top triangle. Performing KCL around this surface yields
1A 2A + 3A + 4 A 5A = I = 1A.
SOLUTION 2.4. Use a gaussian surface around the bottom rectangle. KCL yields
I1 = 2A + 10A + 3A = 15A .
SOLUTION 2.5. Using KVL, V1 = 55V 15V + 105V 100V 30V = 15V .
SOLUTION 2.6. Using KVL, Vx = 5V 1V 1V 1V + 1V 1V = 2V .
SOLUTION 2.7. Using KVL once again.
v1 = 7 + 6 + 5 = 18V
v2 = 6 + 7 8 = 5V
v3 = 5 6 = 11V
v 4 = 8 7 = 1V
SOLUTION 2.8. KVL is used to find the voltage across each current source, and KCL to find the current
through each voltage source.
I3V = 6A 7A = 1A
I4V = I3V + 8A = 7A
I5V = 8A 6A = 14A
V7 A = 4V + 3V = 7V
V8A = 4V + 5V = 1V
V6 A = V8 A 3V = 2V

Chap 2 Probs

P2 - 2

R. A. DeCarlo, P. M. Lin

SOLUTION 2.9. Using the same method as before, the current and voltages are found through and across
each sources.
I5V = 9 + 8 7 = 10 P = 50W
I4V = 6 I5V = 16 P = 64W
I2V = 6 7 = 1A P = 2W
I3V = I2V 9 = 8A P = 24W
V8A = 4 5 = 1V P = 8W
V9 A = 3 + V8 A = 2V P = 18W
V7 A = 2 V9 A = 0 P = 0W
V6 A = 5 V7A = 5 P = 30W
Summing all the power give 0W, hence conservation of power.
SOLUTION 2.10. Doing KVL around the right loop does not balance out. Changing 8V to 5V would fix
this.
SOLUTION 2.11. Using KVL to determine the voltages, and KCL to determine the currents:
Vy = 8V
Vx = Vy 4 = 4V
Ia = 4 A
Iy = 4 14 + 2Ia = 2A
Ix = Ia Iy = 6A
SOLUTION 2.12. First Vin = I2 8 = 24V . Then I1 = Vin / 3 = 8A and I3 = 12A I1 I2 = 1A.
Therefore RL = Vin / I3 = 24 P = I3 Vin = 24W
SOLUTION 2.13. (a) First, from current division, get

1/ 3
12 / 3
I1 =
(12 aI1 ) I1 =
.
1/ 3 + 1/ 6 + 1/ RL
(1 + a ) / 3 + 1/ 6 + 1/ RL
(b) Using the previous equation and solving for 1/ RL = (12/ 3I1 ) 1/ 6 (1+ a ) / 3 = 0.5S or RL = 2 .

1/ RL
The power P = I32 RL =
(12 aI1 ) RL = 18W
1/ 3 + 1/ 6 + 1/ RL

SOLUTION 2.14. For the power delivered by the source to be 60W, the voltage across it should be
V = P / 2A = 30V . Therefore the current through the 20 must be I20 = 30 / 20 = 1.5A , and by KCL the
current through IR L = 2 I20 = 0.5A . From this, RL = V / I RL = 60 .

Chap 2 Probs

P2 - 3

R. A. DeCarlo, P. M. Lin

SOLUTION 2.15. Writing KVL around the loop 25V 4I 15V 5I I = 0 I = 1A, and
P5 = I 2 R5 = 5W
SOLUTION 2.16. The total power supplied by the source is P = 50V 0.5A = 25W . The power absorbed
by the resistor is P60 = ( 0.5A ) 60 = 15W . Therefore by conservation of power, the power absorbed by
2

X is 10W.
SOLUTION 2.17. (a) As this loop is open, no current flows through it, so IR is 0A. The output voltage is
VOUT = 2V + 3V 2V = 1V by KVL.
(b) Writing out the KVL equation around the loop 3 2 I R R 2 I R 2R I R R = 0 1 = I R 4R.
Therefore IR = 1/ 4R and VOUT = IR R = 1/ 4V .
SOLUTION 2.18. Writing out KVL around the loop 60 30I 30 20 + 60 40I = 0 I = 1A . From
ohm's law R = V / I = 30.
SOLUTION 2.19. (a) Using Ohms law Iin = V2 / ( 20 + 12 ) = 0.75A, and V1 = 12 Iin = 9V . To find R, write
KCL and get VR = 30 V2 = 6V . Therefore using Ohms law again, R = VR / IIN = 8 .
(b) Writing KVL around the loop, 30 = aV1 + Iin R + Iin 20 + V1 , and substituting Iin = V1 / 12,
V1 = 30 / [( R + 32) /12 + a] is obtained. Next substitute back V1 = 12Iin and solve for
30

R=
a 12 32 = 40
12I IN

SOLUTION 2.20. (a) i. Using R = Vxy / Ibat the value of each resistors starting with the top one are 2.7,
0.6, and 0.25. Using the same relationship, the resistance for the motor is 1.25.
ii. Using P = Vxy2 / R the power dissipated by each resistor is 16.875W, 3.75W, 1.5625W, and for the
motor 7.8125W.
iii. The relative efficiency is = 7.8125/ (122.5) 100 = 26 %
(b) i. Performing voltage division across each resistor
VAB = 0
VBC = 12 RBC / (RBC + RCD + Rmotor ) = 3.43V
VCD = 12 RCD / (RBC + RCD + Rmotor ) = 1.43V
Vmotor = 12 Rmotor / (RBC + RCD + Rmotor ) = 7.14V

Chap 2 Probs

P2 - 4

R. A. DeCarlo, P. M. Lin

ii. Ibat = 12 / (RBC + RCD + Rmotor ) = 5.71A

2
iii. The relative efficiency is = (Vmotor
/ Rmotor)/ (12 5.71)100 = 59.5 %

(c) i. Repeating the steps from (b), the voltages across the first two resistance are 0, then across the other
and the motor 2V, and 10V
ii. Ibat = 12 / (RCD + Rmotor) = 8A
2
iii. And the relative efficiency is = (Vmotor
/ Rmotor)/ (12 8)100 = 83.3 %

(d) What is the largest equivalent resistance of the motor that will draw 30A? R = 12/ 30A = 0.4.
SOLUTION 2.21. (a) Observe that i = IO , thus v = ki3 = kI03 .
3
(b) Using KVL and previous equation, v x = (R1 + R2 )IO + VO + kIO
.
4
(c) The power is = IO v x = (R1 + R2 )IO2 + VO IO + kIO

SOLUTION 2.22. I100 =

I150 = 0.02 +

0.04
= 0.02 A. Therefore V300 = 0.02 (100 + 200) = 6 V. By KCL,
100

6
= 0.04 A. Req, seen by the source, is 300 . Therefore Vs = 0.04 300 = 12 V.
300

SOLUTION 2.23. Using KCL IR = 5 3 = 2A , and KVL VR = 10 + 6 = 16V . Thus

R = VR / IR = 16 / 2 = 8.
SOLUTION 2.24. Using KCL, KVL, along with Ohms law,
I5 = 6 7 = 1A
I4V = 8 I5 = 9A
I2 = 8 + 7 = 15A
V6 A = 4 + 5I5 = 1V
V8A = 4 + 2I2 = 26V
V7 A = V8 A 5I5 = 31V
Now, the power delivered or absorbed by each element is calculated:
P6A = I6A V6 A = 6W
P7A = I7A V7 A = 217W
P8A = I8A V8A = 208W
P4V = I4V 4 = 36W
2
P5 = I5
5 = 5W
2
P2 = I2
2 = 450W

Chap 2 Probs

P2 - 5

R. A. DeCarlo, P. M. Lin

Note that for passive elements, when the power is positive it is absorbed, while for independent sources it is
generated when the power is positive.
SOLUTION 2.25. Note that I1 = 6A . Thus by KCL
I3 = 6 0.5I1 = 3A
I2 = 2 + 0.2I1 = 3.2A
I4 = 8 0.3I1 = 6.2A
And finally using KVL
V2 = 8A 1+ 4I 4 + 3I3 = 41.8V
V1 = 2I2 3I3 = 2.6V
SOLUTION 2.26. (a) Using KCL,
I4 = 5 4 = 1A
I3 = I 4 2 = 1A
I2 = 3 2 = 1A
I1 = I2 5 = 6A
(b) Using KVL and Ohms law,
V1 = 3I1 = 18V
V2 = 12 I 2 = 12V
V3 = 10 I 3 = 10V
V4 = 4 I 4 = 4V

(c)
P3 A = 3 A(V2 V1 ) = 90W
P2 A = 2(V1 V2 V3 ) = 40W
P4 A = 4( V3 V4 ) = 24W
P5 A = 5(V4 + V3 V1 ) = 60W

Chap 2 Probs

P2 - 6

R. A. DeCarlo, P. M. Lin

SOLUTION 2.27. Write KVL around the outside loop, 40 = 500Ix + (400 + 200)i . And write KCL
equation i = Ix 2I x . Solving yields Ix = 0.4A . The dependent source delivers 2I x (600i) = 192W , and
the independent 40I x = 16W . Finally the resistors absorb 500Ix2 + 400i 2 + 200i 2 = 176W verifying the
conservation of energy since the source generate 192W-16W=176W.
SOLUTION 2.28. By voltage division V2 =

[((90 ||180) + 60)||40]

60

V = 1/14 Vs .
[((90||180) + 60)||40] + 160 60 + (90||180) s

Therefore Vs = 14V2 = 280V .

SOLUTION 2.29. By voltage division
v x = 9V

18 + 3
= 7V
(18 + 3) + 6

SOLUTION 2.30. By voltage division we get the following two equations in order to solve for the two
unknowns.
R1
V2 = V1
R1 + R2
R1 + R2
V1 = 100V
R1 + R2 + 60
Solving yields R1 = 40 , and R2 = 100 .
SOLUTION 2.31. Dividing 1400 in four gives 350. If we only need 1/4 and 2/4, the resistor string can be
made of three resistances: 350, 350, and 700.

SOLUTION 2.32. Using voltage division, at t=0 v R = 15

the voltage goes back to 0V.

2R
= 10V , and t = 5 s v R = 10V , and at t = 10
3R

Chap 2 Probs

P2 - 7

R. A. DeCarlo, P. M. Lin

10
9
8
7

Volts

6
5
4
3
2
1
0

10

15

time (sec)

SOLUTION 2.33. By voltage division

Vb =

Rb
Rd
Vin and Vd =
V
Ra + Rb
Rc + Rd in

By KVL, if Vout = 0, then

Rb
Rd
0 = Vout = Vb Vd =

V
Ra + Rb Rc + Rd in

For arbitrary Vin , this requires that

Rb
Rd
=
or equivalently that Rb Rc = Ra Rd .
Ra + Rb Rc + Rd

SOLUTION 2.34. First Geq = 1m + 1.5m + 2m + 3m = 7.5mS . By current division

I2 = 100mA

1.5m
= 20mA, P = 100mA I2 /1.5mS = 1.33W .
Geq

SOLUTION 2.35. By current division, for I1 to be 2A then 160 + R = 300 ||600 for an even split. Thus
R = 40 .
SOLUTION 2.36. By current division, i1 = 0.4A

1/10
= 0.32 A . Therefore using KVL
1/10 + 1/ 40

v d = 10i1 0.25i1 = 3.12V .

SOLUTION 2.37. (a) Req = (8k ||2k) + (9k ||1k) = 2.5k

Chap 2 Probs

P2 - 8

R. A. DeCarlo, P. M. Lin

(b) Req = 2k ||[(2 k ||2k) + (2k ||2 k)] = 1k

SOLUTION 2.38. (a) Req = 2 + 15 + 10 + 10 + 40 + 30 + 20 + 8 = 135 .
(b) Four of the resistors are shorted, thus Req = 2 + 15 + 10 + 8 = 35 .
(c) Lumping the series resistance together Req = 8 + [50 ||(50||25)] + 2 = 22.5
SOLUTION 2.39. (a) Req = [2R + (4R ||4 R)]||[2 R + (4 R ||4 R)] = 2R
(b) Req = 2R ||2 R + (4R ||4 R ||4 R ||4 R) = 2R
SOLUTION 2.40. (a) First Req = 150 + [375||(250 + 500)] = 400 . Next Iin = 14V / Req = 35mA. The
power delivered by the source is then = 14Iin = 0.49W .
(b) Req = 150 + [375||(250 + 500)||1k] = 350, and Iin = 14 / Req = 40mA . The power delivered by the
source is = 14Iin = 0.56W .
As the equivalent resistance decreases, more of it gets dissipated by it.

SOLUTION 2.41. Req1 > Req2. Without going into a detailed analysis using methods of Chapter 3, we
present the following intuitive argument. First note that the points a and b represent points on an
unbalanced bridge circuit meaning that the voltage between a and b would not be zero. Also note that when
two resistors are placed in parallel, the equivalent resistance becomes smaller than either resistance. The
addition of the resistor R in circuit 2 essentially creates an internal parallel resistance resulting in an Req2
lower than Req1.
SOLUTION 2.42. Req1 = Req2. As was the case in the previous problem, this is a balanced bridge circuit.
Hence no voltage appears between a and b making the additional resistor irrelevent.
SOLUTION 2.43. (a) Rin = [(20||20) + 10]||(1/ 0.12)||(1/ 0.08) = 4.
(b) Rin = 6R ||[( R || R ||0.5 R) + 0.75R + (2R ||2 R)] = 1.5R
SOLUTION 2.44. (a) The infinite resistance are essentially open circuits, thus
Req = 1 + 2 + 3 + 4 + 2 + 4 + 3 + 2 + 1 = 22
(b) 0 resistances are short circuits. Labeling one branch x and the other y, it can be seen that the circuit is a
set of 3 resistor strings in parallel to each other between x and y, then added in series to the two 1 Ohm
resistor. Thus Req = [(2 + 3)||(4 + 2 + 4)||(2 + 3)]+ 1 + 1 = 4 .

Chap 2 Probs

P2 - 9

R. A. DeCarlo, P. M. Lin

(c) Writing out Req = 1 + [Rx ||(2 + 3 + 4 + 2 + 4)] + 3 + 2 + 1 = 7 + [Rx ||15] , and solving for Rx = 3.75.
(d) No, it requires methods to be covered in the next chapter.

SOLUTION 2.45. Using the formulas for parallel resistances, the circuit of figure 2.45 reduces to

(a) RAC = [((2 + 1) / /6 ) + 8] / /10 = 5

(b) RAB cannot be calculated by series parallel formulas, but RBC can be done.
RBC = [((8 + 10) / /6 ) + 2] / /1 = 0.86667
SOLUTION 2.46. (a) Req = 300 + (R ||5.6k), thus R = 800.
(b) Req = R + (R ||1.2K) , the following quadratic equation must be solved R2 + 1.4k R 1.2M = 0 . This
yields R = 600 .
(c) Req = 500 + 300 + (800||400 || R). Solving for R yields 800 .
SOLUTION 2.47.
(a) Using the fact that the resistance seen into terminal a-b is the same as that seen in terminal c-d, we can
obtain the following relationship. Req = R + R || Req . This produces a quadratic equation whose solution is
Req = 1.618R .
(b) Using the previous argument Req = 5 + 10 || (5 + Req ) . Solving for Req = 11.18 .
1

1/ 18k

6k + (9k ||18k)
SOLUTION 2.48. By current division Ix =

1
1 36m = 3mA
1/18k + 1/ 9k
+

6k + (9k ||18k) 4k

Chap 2 Probs

P2 - 10

R. A. DeCarlo, P. M. Lin

SOLUTION 2.49. The 500 resistor has no effect on the current entering the circuit to its right.
0.15 =

30 30 30
+
=
+ 0.05
R 600 R

Hence R = 30/0.1 = 300 .

SOLUTION 2.50. (a) First, express the total current as I =

120
. Next, find RL2 that
0.5 + (20 ||30||40 || RL2 )

will cause I to be 15A. Thus RL2 = 40 or less will cause the fuse to blow as this will cause the current to
be 15A or more.
(b) Repeating the previous procedure, RL2 = 20 .
(c) RL2 = 120 .
SOLUTION 2.51. At time 0, all switches are open and Vout =

260
220 = 190.7V .
260 + 40

Then at t = 5s, switch one closes and

Vout =

260||260
220 = 168.2V .
(260 ||260) + 40

At t = 10s,
Vout =

130 ||260||260
220 = 136.2V .
(130 ||260||260) + 40

Finally at t =15 s,
Vout =

65||130 ||260||260
220 = 98.5V .
(65||130 ||260||260) + 40

200

180

Volts

160

140

120

100

80

10
12
Time (sec)

14

16

18

20

Chap 2 Probs

P2 - 11

R. A. DeCarlo, P. M. Lin

SOLUTION 2.52. (a) Lumping the two sources together and the resistors into an equivalent resistor gives
9cos(2t) 3cos(2t)
i1 (t) =
= 0.24cos(2t)mA .
7k + 9k + 8k + (2k ||3k ||6 k)
1/ 6k

(b) By current division i2 (t) =

i (t) = 40cos(2t) A .
1/ 2k + 1/ 3k + 1/ 6k 1

SOLUTION 2.53. (a) Starting with,

Req1 = 5||(10 + 10) = 4
Req 2 = 10||(6 + Req1 ) = 5
Req 3 = 5 + Req2 = 10
(b) Using the values R just obtained,
R
Va = 100 eq 2 = 50V
Req3
Vb = Va

Req1
= 20V
6 + Req1

10
= 10V
10 + 10
(c) Finally,
100
Iin =
= 10A
Req3
Vc = Vb

Id =

Va
= 5A .
6 + Req1

Ie =

Vb
= 1A
10 + 10

SOLUTION 2.54. (a) Circuit a: Using voltage division,

300||(20 + 30 + 50)
30
= 33.75sin(377t)V , and Ohms law
vout (t) = vin (t)

[300 ||(20 + 30 + 50)] + 5 30 + 20 + 50

iout(t) = vout(t)/ 30 = 1.125sin(377t )A . The instantaneous power is then
P(t) = iout (t) vout (t) = 37.969sin 2 (377t)W .
Circuit b: By current division

1/ (20 + 30 + 50)

iout(t) = iin (t)

= 60sin(377t)A , and from Ohms law
1/ (20 + 30 + 50) + 1/ 300 + 1/ (50 + 100)
vout (t) = 50 iout(t) = 3000sin(377t)V . The instantaneous power is P(t) = 180sin 2 (377t)kW .
(b) No, since the current source forces the amount of current in the circuit.

Chap 2 Probs

P2 - 12

R. A. DeCarlo, P. M. Lin

SOLUTION 2.55. (a) Noting that i2 = v1 /10 = 6A, then we can write KCL at the top left node,
isource = i2 + v1 / 6 + (v1 5i2 )/ 5 = 22A. Thus P = 60 22 = 1.32kW .
(b) First, determine the current through each resistor:
i2 = 60 /10 = 6A

60

i2.5 =
= 12A
2.5 + (5||5)
i5 = 1/ 2 i2.5 = 6A
Then calculate the power absorbed by each resistor:
P10 = 10i22 = 360W
2
P2.5 = 2.5i2.5
= 360W
2 = 180W
P5 = 5i5

SOLUTION 2.56. From Ohms law I1 = 100m / 200 = 0.5mA. By current division
20k
IRL =
150I1 = 75.257mA, and PRL = 200I 2RL = 1.103W .
20k + 200
2
SOLUTION 2.57. First, using voltage division, Vx = Vs
= (2 / 3)Vs . Then using KCL and the
2 + 1
previous equation, Is = (Vs / 3) Vx = (1/ 3)Vs . Finally using Ohms law Req = Vs / Is = 3.
SOLUTION 2.58. Observing the following relationship, V1 = Vin , the following nodal equation can written:
Iin = Vin / 3 + Vin / 6 2Vin = 1.5Vin .

SOLUTION 2.59. Step 1. From voltage division

V1 =

18
22
11
Vs = 0.75Vs and Vin =
Vs = Vs
18 + 4 + 2
24
12

Hence
V2
11 11 2 11 2
Pin = in =
V =
V
22 22 144 s
288 s
Step 2. For the load, by current division

Chap 2 Probs

P2 - 13

I2 =

R. A. DeCarlo, P. M. Lin

6
3
3
9A
AV1 = A Vs =
V
6+2
4
4
16 s

Therefore
81A2 2 81A2 2
V =
V
256 s
128 s

2
P2 = 2 I2
=2

Step 3. P2 = 10 Pin implies that

81A 2 2
11 2
Vs = 10
V
128
288 s
Hence
A=

SOLUTION 2.60. By voltage division V1 =

previous equation I2 =

128 110
= 0.7769
81 288

6
V = (3/ 4)Vin . By current division, and substituting the
6 + 2 in

3
4V = Vin . Using voltage division and Ohms law, and substituting the previous
3+6 1

equation,
10
= 3Vin = 30V
10 + 5
Iout = 4.5I2 / (10 + 5) = 0.3Vin = 3A

Vout = 4.5I2

Finally, from the previous equations | Vout / Vin |= 3 .

V 0.04ibat
SOLUTION 2.61. Writing out KCL when the switch is closed, ibat = 150A + bat
. Solving
240
gives ibat = 150.02A and Vout 6V . When the switch is open Vout = Vbat

240
12V . Therefore,
240 + 0.04

the reason for the radio stopping is insufficient supply voltage.

SOLUTION 2.62. (a) Using the following relationship P = V 2 / R, the resistance of each headlight on low
beam is R = V 2 / P = 4.11 .
(b) Using the same relationship R = 2.22 .
240
(c) By voltage division, Vout = 14.7
= 14.698V .
240 + 0.04
240 ||4.11||4.11
(d) Using voltage division, Vout = 14.7
= 14.417V
(240||4.11||4.11) + 0.04

Chap 2 Probs

P2 - 14

(e) Using voltage division, Vout = 14.7

R. A. DeCarlo, P. M. Lin

240||2.22||2.22
= 14.186V
(240 ||2.22||2.22) + 0.04

SOLUTION 2.63. By voltage division

11.96 =

15
180
12 =
15 + R0
15 + R0

Therefore
R0 =

180 15 11.96
= 0.050167
11.96

SOLUTION 2.64. (a) Using KVL Vt = 102 0.0580 = 98 V.

(b) Using KVL Vt = 102 + 0.05 50 = 104.5 V.
(c) P = Vt 50 = 5.225. kW

SOLUTION 2.65. Minimum load means the minimum load resistance that the system can handle.
MaxPwr = 0.8*50e6
MaxPwr = 40000000
Vs = 750e3;
Iline = MaxPwr/Vs
Iline = 5.3333e+01
Rmin = Vs/Iline
Rmin = 1.4062e+04, i.e., Rmin = 14.062 k.
SOLUTION 2.66. (a) Using the following general form for a non-ideal voltage source: vout = Rsiout + Vs ,
one sees that for zero current vout = Vs = 40V . The slope of the line is

40
= Rs = 0.04 , thus
1000

Rs = 0.04 .
60
(b) This curve represents a resistors I-v characteristic, thus the slope 1 = R = 45 .
1
3
1
(c) The general form for a non-ideal current source is iout = Vout + I s . When the voltage is zero,
Rs
iout = Is = 5A . From the slope of the line,

4000
, Rs = 4000 / 5 = 800 .
5

Chap 2 Probs

P2 - 15

SOLUTION 2.67. Using the following formula:

R. A. DeCarlo, P. M. Lin

T nI
= 1, solve for T, and get 0.625 hrs, or 37.5 min.
n Cn

T
SOLUTION 2.68. Using the same equation as before and solving for Cn = nI
n
T=55/60 hrs, the capacity obtained is 20 Ah.
SOLUTION 2.69. C20 = 50 Ah
(a) In eq. 9, solving for I with n=20, and T=10, I=4.2A
(b) Calculate the capacity for n=10 and T=10, this yields 42 Ah.
SOLUTION 2.70. (a) Using a sequence of voltage division,
48
= 48mV
50
195
V2 = 50V1
= 2.34V
200
Vload = 2.5V2 = 5.85V
V1 = 50mV

2
And the power is RRL = Vload
/ RL = 2.278W .

(b) Following is the graph, and the script used to generate it.

Power in Watts

5
4
3
2
1
0

10

20

10

20

30

40

50

60

70

50

60

70

Current in mA

800
600
400
200
0

%Script for Question 70 in chapter 2

RL=8:1:64;
V2=2.34;
IL=2.5*V2 ./ RL;
%Note the use of the ".*" which means that the division
%is performed for each value of RL.

30
40
Resistance in Ohms

1/

, with n=10, and

Chap 2 Probs

P2 - 16

PL=RL .* (IL .^ 2);

%Plot the Power versus RL
subplot(2,1,1);
plot(RL,PL);
ylabel('Power in Watts');
%Plot the Current versus IL
subplot(2,1,2);
plot(RL,1000.*IL);
xlabel('Resistance in Ohms');
ylabel('Current in mA');
%The use of subplot lets you subdivide the graphing
%window in two halfs.

SOLUTION 2.71. (a) Using the following script:

%Script for problem 2.71
R1=15; R2=4; R3=9; R4=2; R5=8;
R6=18;
Ra= R4+R5;
Ga= 1/Ra;
Gb= Ga+1/R1;
Rb= 1/Gb;
Rc= 1/(1/R6+1/R3)+Rb;
Gc= 1/Rc;
Geq= Gc+1/4;
Req= 1/Geq;
Irc= 20*Gc/Geq;
Vrb= Irc*Rb;
Vout= Irc*(Ga/Gb)*8;
Req
Vout

So (a) Req = 3 , and (b) Vout = 24V

SOLUTION 2.72. Using the following script:
%Script for problem 2.72
R1=1e3; R2=2.2e3; R3=2e3; R4=5e3; R5=3e3; R6=R5;
R7=3.2e3; R8=1.2e3; R9=1.6e3;

R. A. DeCarlo, P. M. Lin

Chap 2 Probs

P2 - 17

R. A. DeCarlo, P. M. Lin

Ga=1/R7+1/(R8+R9);
Ra=1/Ga;
Gb=Ga+1/R6;
Rb=1/Gb;
Gc=1/R4+1/(R5+Rb);
Rc=1/Gc;
Gd=1/R2+1/(R3+Rc);
Rd=1/Gd;
Geq=1/R1+Gd;
Req=1/Geq
%Going through the same step to find Vout
Id=200e-3*(Gd/(Geq));
Ic=Id*((1/(R3+Rc))/Gd);
Ib=Ic*((1/(R5+Rb))/Gc);
Ia=Ib*Ga/Gb;
Iout=Ia*((1/(R8+R9))/Ga)
Vout=Iout*R9

The following values are obtained:

Req =
5.9121e+02
Iout =
5.5431e-03
Vout =
8.8689e+00

d
R1

R2

SOLUTION 2.73. Using the following script:

%Script for problem 2.73
R1=20; R2=40; R3=60; R4=30; R5=10; R6=135;
R7=150; R8=300; R9=130; R10=200; R11=50;
Ga=1/R10+1/R11;
Ra=1/Ga;
Rb=Ra+R9+(1/(1/R7+1/R8));

R4

R6

R7

R9

Chap 2 Probs

P2 - 18

R. A. DeCarlo, P. M. Lin

Gb=1/Rb;
Gc=Gb+1/R6;
Rc=1/Gc;
Rd=Rc+R5+(1/(1/R3+1/R4));
Gd=1/Rd;
Ge=Gd+1/R2;
Re=1/Ge;
Rin=R1+Re
Ie=10/Rin;
Id=Ie*Gd/Ge;
I1=Id*(1/R6)/Gc
Ib=Id*Gb/Gc;
Vout=Ib*Ra

The following values are obtained:

Rin = 50
Vout = 0.667V
I1 = 33.3mA

R4
e d

R5

R1

c b

R8
R7

R9
R11

R3
R2

R6

R10

SOLUTION 2.74. An identical procedure to the one followed in the previous problem will yield the
following values:
Rin = 50.53
Iout = 133.8mA

PROBLEM SOLUTIONS CHAPTER 3.

Solution 3.1. Select the bottom node as the reference node, and write a node equation at the positive
terminal of the V1 resistor:
V1 V0 V1 V1 4V0
+
+
=0
3R
6R
6R
2V1 2V0 + V1 + V1 4V0 = 0
4V1 = 6V0
V1 = 1.5V0

Solution 3.2 Write a node equation at the top node:

Vx 2V x V x

=0
100 100 50
Vx 2Vx 2Vx = 60
5Vx = 60
V x = 12V
0.6

Solution 3.3
Vx 25Vx Vx Vx 0.2Vx
+

=0
100 100
50
40
3V
25V x 8V x
x +

= 0.6
100
100
400
0.6 400
Vx =
80
V x = 3V
0.6

Solution 3.4 (a)

It is evident from the figure that Vc = 20. We need to write two equations in Va and Vb and put them in
matrix form. In this case, we can write the matrix equation by inspection. Note that the resistors are
identified by conductance values.
15m 5m Va 0.5
5m 35m V = 0.5

(b) Solve the matrix equation by inverting the left-most matrix:

Va
35m 5m 0.5
1
V = 525 25 5m 15m 0.5
b

1 35 5 0.5
=
0.5 5 15 0.5
40
=
20
(c) Vx = Vab = Va Vb = 20V , Vda = -Va = -40, Vdb = -Vb = -20.
(d) Pi = 0.540 = 20W, Pv = 20(20-20) = 0. P diss = 404010m + 20205m + 20205m = 20 W.
Power delivered equals dissipated power.

Solution 3.6 Write two nodal equations:

Vs1 V1
V V2
= Is 3 + 1
3000
6000
V2
V V2 V1 V2
= s2
+
30000
12000
6000
Rewrite equations as:
2Vs1 2V1 = 6000Is 3 + V1 V2
2V2 = 5Vs2 5V2 + 10V1 10V2
Cast into a matrix equation
3 1 V1 6000I s3 2Vs1
10 17 V =

5Vs 2

Solving the matrix equation yields:

V1 181.46
V = 124.39
2

Power absorbed by the 6k resistor is (V1 -V2 )2 /R = 0.5429W .

Similarly, Ps1 = (Vs1 -V1 )/3000Vs1 = 4.7W, Ps2 = (Vs 2 V2 ) /12000 Vs 2 = 0.32W
Ps3 = Is 3 (Vs2 V1) = 1.21W

Solution 3.7 (a) Again, the matrix equation can be written by inspection:

G1 + G2 + G4

G4

G4
VB 50G1
=
G3 + G4 + Gs VC 50G3

(b) Substituting the values of conductances and inverting the above matrix equation yields:
VB = 34.0132V
VC = 33.6842V
(c) Power delivered is 80.7566W . Using the Principle of Conservation of Power:
Pdel = P1 + P2 + P3 + P4 + P5
or,
V VB VA VC
Pdel = 50 A
+
= 80.7566W
20
20
(d) In this part, we take the above matrix equation and solve it for each value of Gs. If we do this, we can
get a feel for the behavior of VB and VC w.r.t. changes in Gs. The following plot is the voltage difference
between the two nodes as a function of Gs, and hence as a function of temperature.

As can be seen, in this figure, the voltage difference between B and C does not change linearly with Rs.
Since this resistance itself changes linearly with temperature, this means that VB-VC does not change
proportionally to temperature.

Solution 3.8 The answer is:

G1 + G2
G
1

G2

G1
G1 + G3 + G4 + G7
G4
G7

G2
G4
G4 + G5 + G6 + G2
G6

0
V1 Is1
V 0
G7
2 =

V3 0
G6

G6 + G8 + G7 V4 I s2

Solution 3.9 Write the matrix equation by inspection:

Vs1
4 /100 1/100 1/100 V1 100

Vs1
1/100 4 /100 1/100 V2 = 100
1/100 1/100 4 /100 V3 I s2

Solving the equation in MATLAB, we get: V1 = 7, V2 = 7, V3 = 11, and P = 0.6.
3
V V V V2
3
P = Vs1 S1 1 + S1
=
10

+
10 10 = 0.6W
100
100

Solution 3.10 (a)

Nodal equation for A:
VA Vs1 VA VB VA VC
+
+
=0
10
10
10
(b) At node B:
VB VB V A
+
Is2 = 0
10
10
(c) At node C:
V VA
Is2 Is 3 + C
=0
10
(d) Manipulate algebraically to cast as the following matrix equation:
3 1 1 VA
Vs1

1 2 0 VB = 10I s2

1 0 1 VC 10(I s3 I s2 )

VA 13 13
1
VB = 13 3
VC 43 1

(e) P = (-10-13.333)/10(-10) = 23.33W.

Solution 3.12
We are required to write the equations in matrix form. First, write a node equation at VA and Vout:
V
V Vout
VA 5 + A + A
=0
5
10
Vout V A
V
7.5VA + out = 0
10
10
Now group the coefficients for VA and Vout, and write the matrix equation:
1+ 1 / 5 + 1/10 1/10 V A 5
=
1/10 + 7.5
2/10 Vout 0

VA 1
V = 37
out

where the matrix inversion was performed in MATLAB. The ratio of the output voltage to the input voltage
is -37/5.

Solution 3.13 (a) Nodes A and B are already labeled:

(V A 9)0.1 + (V A 0.5VB )0.2 + (V A VB )0.3 = 0

(VB V A )0.3 + (VB 9)0.5 + 0.4VB = 0
This can be rearranged Into:

(0.1+ 0.2 + 0.3)VA 0.3VB (0.2 0.5)VB = 0.9

0.3VA + (0.3 + 0.5 + 0.4 )VB = 4.5
The matrix equation can now be easily obtained:
0.6 (0.5)(0.2) 0.3 V A 0.9
0.3
V = 4.5
1.2

B
V A 4.8
V = 4.95
B

(b) Iin = -(VA-Vs)0.1 - (VB-Vs)0.5 = 2.445A

(c) Ps = VS Iin = 22.005W, Pdep = 0.5 Vout (0.5Vout V A ) 0.2 = 1.151W .

(d) P = V2 /R = (4.95)(4.95)0.4 = 9.801W.

Solution 3.14 (a) We write node equations at VA and VB:

VA
V VB
+ gm1VA + A
=0
20k
10k
VB VA
V
gm2 (V A VB ) + B + I s2 gm1V A = 0
10k
2.5k

I s1 +

Rearranging, we have:
1
V
1
+ gm1 +
V A B = I s1
20k
10k
10k
1
1

1
gm2 gm1 V A +
+ gm2 +
V = I s2
10k

10k
2.5k B
(b)
1/10k
1/20 k + gm1 + 1/10 k
V A Is1
1/10k gm gm 1/10 k + gm + 1/2.5k V = I

2
1
2
B s 2
(c) The above matrix is inserted into MATLAB, with all the values substituted, to obtain:
V A 9.722
V = 5.972
B

(d) Vo = VA-VB = 3.75V

(e) P1 = VAIs1 = 0.0292W, Pgm1 = -Vo gm1 VA = -0.008W, Pgm2 = VBgm2 Vo = 0.0112W, P2 = -VBIs2 = 0.0119W.

Solution 3.15 I1 = 0.4. Write nodal equations at A and B

VA V A VB
+
+ 0.03(V A VB ) = 0.4
100
20
V
VB VA VB VB 80 B 40
+
+
=0
20
40
40
Rearranging and casting into matrix form:

1/20 0.03 VA 0.4

1/100 + 1/20 + 0.03
=

1/20
1/20 + 1/40 1/40 VB 0

VA 40
V = 40
B
It is obvious then, that Vx , the voltage between A and B, is zero.

Solution 3.16
V VB V A VB
VA = 3000i x = 3000 A
=
9000
3
Equation at node B:
1)

2)

VB VA
V
V VD
+ B + B
= 0 is equivalent to:
9000
6000
18000
2V A + 6VB VD = 0

Equation at node D:

3)

VD VB
V
+ D + IS = 0 which can be rewritten as:
18000
9000
3VD VB = 360 Solving the system formed by equations (1), (2) and (3) we obtain:
VA = 9V , VB = 18V , VD = 126V .

Solution 3.17 (a) Choose E as the reference node

VA = 2ix
At node B
6 = (VB-VC)/3 + (VB-VA)/2
Or
6 = 5/6VB 1/3VC VA/2
At node C,
VC = 2iy
At node D,
VD = -12V
iy = (VD-VA)/2 = VD/2 ix
From here on, the solution involves algebraic manipulations to solve the system of equations. MATLAB or
hand analysis can be performed to obtain:
VA = 48V, VB = 12V, VC = -60V.
(b) P6A = 6(12 = 72W

I12V = 30-8 = 22A ( P12V = 264W.

P2i x = 2ix (iz iY ) = 2 24 (+18 + 30) = 2304W
v vD
P2i y = 2iy (i x + c
) = 2 (30) (24 8) = 1920W
6
(c) P3 = i x (VB VC ) = 1728W = 1728W
P6 = 688 = 384W
P2 y = 2iy iy = 1800W
P2 = (VB-2ix ) (VB-2ix )/2 = 648W.
(c) VD = -12V, VA = 2ix , VC = 2iz.
Substitute the above VA and VC into the node equation for node B:
iz = (VB-2ix )/2 = VB/2 ix and
ix = (VB-2iz)/3 = VB/3 2/3iz
Substitute iz into ix to obtain: ix = 0, VA = 0. Then, VB can be deduced to be 12.
Finally, VC = 12V.
Now, compute the powers.
P6A = 72 W , P3 = 0, P6 = (VC-VD)2 /6 = 96W, P2 y = 72W, P2 z = 72W.
V VD VC VD
P12V = 12 ( A
+
) = 12 (6 + 4) = 120W,
2
6
P2ix = 0W , P2iy = 12 4 = 48W

Solution 3.18 The three node equations at A, C, and D are:

0.8 0.3 = 0.015VA + 0.02V A 0.02VC
Vc = 440
0.8 + 2.5 = 0.005VD + 0.025VC 0.025VD
As can be seen, these really reduce to only two equations in two unknowns. These can be solved rather
easily either by hand or by MATLAB to obtain: VA = 220V, VD = 310V . Note that all of these voltages
are already referenced to node B (i.e. VA = VAB, etc).

Solution 3.19 (a) Supernode is BC (50 V source).

(b) Only one node equation needs to be written:

VB VA VB VC VA VC

+
+

+
=0
90 90 10 10 10 90
with the constraint that VC VB = 50.
(c) The constraint equation can be substituted into the B node equation to obtain
VB = 125V. Thus, VC = 175V, and ix = (VC-VA)/10 = -12.5A.
(d) (VA-VB)/90 = 1.94A P300 = 300(1.94+12.5) = 4332W.
VC/90 = 1.94 P50 = -528W.
V
P50 = 50 (i x + C ) = 528W
90

Solution 3.20 (a) VB = VA 440 and VC = VA 460.

(b) Supernode is one including A, B, and C.
(c) (VC 40)0.15 + 0.05VB + 0.25V A 25 + 0.2(V A 40) = 0
This can easily be rearranged to get VA = 200V.
(d) power Ps = V A I = 200 25 = 5000W = 5KW

Solution 3.22 (a) VC = Vs2 = 6V

(b) Ix = 0.01VA
(c) Supernode at A,B, encompassing the controlled source. So, we have one equation:
Is1 = 0.01V A + 0.0125VB + 0.1VB 0.1VC
V
(d) Substitute the constraint equations, VA VB = 20Ix = 0.2VA , (equivalently: VA = B ) into the above
0.8
V
equation: VA = B
0.8
V
Is1 = 0.01 B + 0.0125VB + 0.1VB 0.1VC
0.8
VB = 6.4V
V A = 8V
(e) Ix = 0.08A.
(f) P0.0125 = VB2 / R = 0.512W
(g) P = I s1 V A = 0.2 8 = 1.6W

Solution 3.23 (a) VC = Vs2 = 50V.

(b) ix = VA/100.
(c) Supernode A,B:
Is1 = 0.01V A + 0.05VB + 0.05VB 0.05Vs2 + 0.09V A 0.09Vs2
VA = VB + 300i x = VB + 3V A
(d) Solving the above two equations yields: VA = -90V, VB = 180V.
V
90
(e) i x = A =
= 0.9A
100 100
(f) (VB-VC) (VB-VC)/R = 845W.
(g) PS1 = I S1 VA = 2 (90) = 180W

Solution 3.24 (a) VB VC = 3Vx = 3VB ; VC = 2VB

(b) Supernode at B and C, encompassing controlled source.
(c)
V
V
V
V
V
V
Is2 = ( B A ) + B + C + ( C A ); VC = 2VB; 10I s2 = 2(VB + V A )
10 10
10 10
10 10
equivalently: 2V A + 2VB = 10
(d)
(0.1V A 0.1Vs1 ) + (0.1VA 0.1VB ) + (0.1V A 0.1VC ) = 0; 0.3VA + 0.1VB = 0.1VS1 ; equivalently:
3V A VB = 10
(e) Again, any method can be used to simplify and solve the system of two equations. The solution is:
VA = -2.5V, VB = -2.5V.

Solution 3.25 (a) VA = Vs1 = 16V.

(b) Supernode at C and D, encompassing controlled voltage source.
(c)

Is2 = (0.75mVD 0.75mVB ) + (1mVC 1mV A )

(d)
(e)

VC = 4VB + V D
0.75mVD + (0.75mVB 0.75mVD ) + (0.25mVB 0.25mVA ) = 0
or 1mVB 0.25mV A = 0

(f) We now have three equations in VB, VC, and VD. These can be solved using any method. By inspection,
we can immediately deduce VB from VA using the last equation: VB = 4V.
The remaining two equations can be solved to obtain: VC = 20V and VD = 4V.

Solution 3.26 (a) The supernode is the combination of A, C, and the controlled voltage source.
(b) Write node equations starting at the supernode:
(G2VA G2Vin ) + (G3VA ) + (G4V A G4VB ) + (G6VC ) + (G5VC G5VB ) = 0
(G2 + G3 + G4 )V A + (G4 G5 )VB + (G6 + G5)VC = G2Vin
and
(2G6VC ) + (G4VB G4VA ) + (G5VB G5VC ) + (G1VB G1Vin ) = 0
(G4 )VA + (G1 + G4 + G5)VB + (2G6 G5 )VC = G1Vin
and
VA VC = 3V x , VA VC = 3(Vin VA ), 4V A VC = 3Vin
In matrix form:
0.8 0.5 0.2 V A 6

0.4 0.6 0.1 VB = 6
4
0 1 VC 180
(c) The above system of equations can be solved to obtain: VA = 38.75V, VB = 40V, VC = -25V.
(d) Iin = (G7Vin )+ (G 2 Vin G2 VA) + (G1 Vin G1 VB) = 5A.
Req = 12 and P = 300W.
P = Vin Iin = 60 5 = 300W
(e) Iout = VCG6 = -2.5A P = 62.5W.

Solution 3.27 (a) Supernode is A,B encompassing controlled voltage source.

(b)
(VA Vs1 ) + 0.4V B + (0.2V B 0.2VC) = 0
VA + 0.6V B 0.2VC = Vs1
(c)

VA VB = IB = 0.4 VB VA = 1.4VB.

(d)
(e)

Is2 = 0.2VB + (0.2V C 0.2VB) = 0.2VC.

In matrix form:

1 0.6 0.2 V A 8

1 1.4 0 VB = 0
0 0 0.2 VC 2
The solution is: VC = 10V,V A = 7V ,VB = 5V .

i = (VA-Vs1 )1S = -1 Pccvs = (VA VB ) (V A Vs1 ) 1S = 2W

(f)

Pvccs = (Vs1 -VC)(0.2VB) = -2W.

Solution 3.28 (a) Supernode at A,C, CCVS.

(b) Node equation at supernode:

Constraint:

Is + 0.25mVA = G 1 VA + (G2 VA G2 VB) + (G5 VC) + (G4 VC G4VB )

Is = (G 1 + G 2 0.25m)VA + (G2 G4 )VB + (G4 +G 5 )VC
VA VC = 104 ix = 104 G3 VB
0 = VA 104 G3 VB VC

At node B:
G3 VB + (G2 VB G2 VA) + (G4 VB G4 VC) = 0

G2 VA + (G3 +G 2 +G 4 ) G4 VC = 0

(c) Matrix equation:

G1 + G2 0.25m
G2 G4

1
10 4 G3

G2
G3 + G2 + G4

G4 + G5 VA I s

1 VB = 0 ;
G4 VC 0

0
1
1

1000 1000 1000

0.2 1.1
0.8

V A 2

VB = 0
VC 0

(d) Substitute the values of conductances and solve the above matrix equation in MATLAB to obtain:
VA = -38V, VB = -20V, VC = -18V.

(e) P = V I = (104 G3VB ) I s + VA G1 + G2 (V A VB) = (20) [2m 1.9m 3.6m] = 0.15W

Solution 3.29
Loop equation:

Vin = 2kI1 + 500(I1 + 20m)

Vin = 2500I1 + 10 I1 = 20mA.

Pvin = 20m60 = 1.2W.

PI = 20m(500I1 + 50020m) = 0.4W.
P2k = I1 I1 R = 0.8W.
P500 = (I1 + 20m) 2 R = 0.8W.
total power absorbed by resistors: PR = 0.8 + 0.8 = 1.6
total power delivered by sources: Ps =1.2 + 0.4 W=1.6W

Conservation of power is verified.

Solution 3.30
Loop equation:

100(I1 0.5) + 200I1 + 500 (I1 + 20m) = 0

I1 = 0.05A.
P0.5A = I V100; V100 = 100 (0.5 0.05); where V100 is the voltage on the 100 resistor.
P0.5A = 0.5(0.5100 0.05100) = 22.5W.
P20m = 20m(I1 +20m)500 = 0.7W.

Solution 3.31
Loop equation:

3.3 = 50I1 + (50m + I1 )100 + (I1 30m)40 + (I 1 50m)60

I1 = 0.01A.
The power delivered by the independent voltage source:
P = I 1 3.3 = 0.033W.

Solution 3.32
Loop equation:

50 = 300I1 + (I1 0.4I1 )500

50 = (300 + 500 200)I1 I1 = 0.0833A.

Power absorbed by the 500 resistor.
P500 = (I1 0.4I1 )2 500 = 1.25W.

Solution 3.33
Loop equation:

1000(I1 Is) + 4000I1 + 5000(I1 gmVx ) = 0 10000I1 2Vx = 50

and
Vx = 1000(I s I1) 1000I1 + VX = 50 .
Solve the above two equations in I1 and Vx to obtain: I1 = 12.5mA, Vx = 37.5V.
Thus, Req = Vx /Is = 750_,
P = I vccs Vvccs = gm V x 5000 (gmVx I1 )=0.1875W

Solution 3.34
Loop equation:
V1 = 2Iin

Vin = 2Iin + 14Iin 10V1

After replacing V1 in the loop equation we obtain:

Vin = 4Iin
V
R1eg = in = 4
Iin

Solution 3.35
Loop equation:

Vs = 500I1 + 100(I1 + 0.5) + 400(I1 0.001Vx ) + 100(I1 + 0.005Vy )

Vx = 500I1 ,Vy = 400I1 - 4000.001Vx = 400I1 200I1 = 200I1

After replacing Vx and Vy in the loop equation we obtain:
Vs 50 = 1000I1 I1 = 0.1A
Vy = 200I1 = 20V P400ohm = Vy2 /400= 1W
Req = Vs/I1 = 150/0.1 = 1500.

Solution 3.36
Select clockwise loop current I1 in the left loop. Select anti clockwise loop current I2 in the right loop.
The two mesh equations are:
12 = I1 +10(I1 + I2 )
and
10(I2 + I1 ) + 2I2 + 12 = 0
The two simultaneous equations can be solved easily to obtain: I1 = 0.75A, I2 = +0.375A.
P10ohm = (I1 + I 2 ) 2 /10 = 0.127W.
Battery 1 supplies more current. (I1 > I2 )

Solution 3.37
(a) The equation for the left loop is:
660 = I1 R + 1.296(I1 + I 2 )+ 590 + I1 R
The equation for the right loop is:
660 = (0.3 R) I2 + 1.296 (I1 + I2 ) + 590 + (0.3 R) I2

Simplifying the two equations:

70 = 1.596I1 + 1.296I2
70 = 1.296I1 + 1.596I2
The solution of these two equations is: I1 = I2 = 24.2A.
(b) I1 + I2 = 48.4 , voltage across locomotive = 590 + 48.41.296 power = 31592W.
(c) Because the locomotive is 1/3 distance from either station it follows that
R = 1/30.3 = 0.1 . The two equations become:
70 = I1 (2R + 1.296) + 1.296I2
70 = 1.296I1 + (1.296 + 0.6 2R)I2
The solution of these two equations is: I1 = 32.64A, I2 = 16.32A.
Current in locomotive motor I1 + I2 = 48.96A.
Voltage across locomotive

590 + (I1 + I 2 ) 1.296

It follows that:

P = (I 1 + I2 )(590 + 491.296) = 31993W.

Solution 3.38
(a)

(b) The three loop equations are:

660 590 = 0.1I1 + 1.296 (I1 I2 ) + 0.1I1
0 = 1.296 ( I2 I1) + 0.2I2 + 1.296 ( I2 I3)
70 = 1.296( I3 I 2) + 0.2I3
These three equations can be solved using any method to obtain:

I1 = 46.8A, I2 = 0, I3 = 46.8A.
(c) Motor currents are 46.8A each.
(d) Ps = VI = 66046.8 = 30.9kW. Each source supplies 30.9kW.

Solution 3.39 (a) Define three meshes with three mesh currents. The first, I1 , is a clockwise current
around the first mesh. The second, I2 , is a clockwise current around the middle loop of the circuit
(through the 10mS, 5ms, and 5ms conductances). The third, I3 , is a counterclockwise current through the
right-most loop containing the voltage source.

*current names shown above.

(b)

I1 = 0.5A

I
I
I
I
I
( 2 1 ) + 2 + ( 2 + 3 ) = 0.
10m 10m 5m 5m 5m
I
I
I
( 3 + 3 ) + 2 = 20
25m 5m
5m
These are two equations in two unknown currents. After grouping the terms, it can be verified that:
(c)

I2 = 0.1A, I3 = 0.
Vx = 20V
Vad = (0.5 I2 )/10m = 40V
Vbd = 20V

(d)

P0.5 = Va 0.5 = 400.5 = 20W

P20V = 0W
Presistors = 2 I2 2 /5m + (0.5 I2 )2 /10m = 20W

The conservation of power is verified.

Solution 3.40 (a) We can either write down the equations or evaluate the matrix by inspection:

90 (I 2 4.8m) + 10kI2 + 50 = 0
90kI3 + 10 (I 3 4.8m) = 50
OR
0 I 2 90k 4.8m 50
100k
=
0
100k I 3 10k 4.8m + 50

(b) The solution of the above equation is: I2 = 3.82mA, I3 = 0.98mA.

(c) Current source: P = 4.8m[(4.8m I2 )90k + (4.8m I 3 )10k] = 0.61W.
Voltage source: P = 50(I3 I2 ) = 142mW.

Solution 3.41 (a) By inspection:

112k 90k 10k I1 180

0 I 2 = 60
90k 100k
10k
0
100k I3 60
(b) Using MATLAB:
I1 = 4.4mA, I2 = 3.36mA, I3 = 1.04mA
ix = I2 = 3.36mA
(c) P180 = 1804.4m = 0.792W, P60 = 60( I3 I 2 ) = -0.139W.

Solution 3.42 The matrix equation is:

8 6 3 I1 14

6 8 2 I2 = 0
3 2 4 I 3 6
whose solution is: I1 = 4A, I2 = 3A, I3 = 0 A
v = ( I1 + I3 + I2 ) 2 = 2V

Solution 3.43 (a) First, note that two mesh currents are needed. Two clockwise currents are defined: I1
in the middle loop, and I2 in the right-most loop:
Middle loop equation:
100I1 100Is1 + 20Ix + 80I1 80I2 = 0, where Ix = Is1 I1
and

Right-most loop equation:

80I2 80I1 + 10I2 + V s2 = 0
These can easily be cast into the following matrix equation:
160 80 I1 16
80 90 I = 6

2
(b) The solution of this equation is: I1 = 0.12A, I2 = 0.04A.
(c) VA = 100(Is1 I1 ) = 8V and VB = 80x(I1 I2 ) = 6.4V .
(d) Ps1 = Is1 VA = 1.6W.
(e) P0.0125S = (I1 I2 ) 2 /0.0125 = 0.512W.

Solution 3.44 (a) Create two clockwise mesh currents in the top loop (I 2 ) and the bottom-left loop (I1) .
The bottom-right loop has an independent current source. Writing the loop equations:
Vs1 = 200 (I1 I2 ) + 200 (I1 + Is2 )
200 (I2 I1 ) + 100I2 + 300Ix + 200 (I2 + Is2 ) = 0, where Ix = I1 I2
(b) Solving, we get: I1 = -0.1A, I2 = -0.7A, Ix = 0.6A.
(c) VB = (I1 + Is2 )200 = 130V.
(d) Pvs1 = I1 Vs1 = 25W, Pis2 = (VB + (Is2 + I2 )200)Is2 = 105W, P300ix = (I 2 )(300Ix ) = 126W.

Solution 3.45 (a) Create two clockwise mesh currents in the top loop (I1) and the middle loop (I 2 ) (all
resistor loop):
Top loop equation:
0.5vx = 500 (I1 - I2 ,) +500I1 where vx = 500I1
and
Middle loop equation:
600 (I2 Is1 ) + 500 (I2 I1) + 900 (I2 + Is2 ) = 0
(b) Solving, we get: I1 = 0.015A, I2 = 0.0375A, vx = -7.5V.
(c) Pis1 = Is1 [0.5vx + (Is1 I2 )600] = 109.7W
P0.5vx = 0.5vx (I1 Is1 ) = 1.63W
Pis2 = (I2 + Is2 )900Is2 = 53.2W

Solution 3.46 Write the mesh equations in terms of Rs and then substitute the values from the matrix:
Mesh 1 equation:
v1 = R1 (i1 i3) + R2 (i1 i2 )-25i2
From this equation, and the first row of the matrix equation, we can deduce that
R1 = 5
and
R2 + 25 = 40 R2 = 15 .
Similarly:
R1(i3 i1) + i3R4 + R3 (i3 i2 ) =0

Mesh 3 equation:
From which we can deduce:
R3 = 25

and

R4 = 5 .

Solution 3.47
Modified loop 1 equation:
Vs1 = 3MI 1 + v + 2MI1
Constraint equation:
I1 + I2 = Is3
Modified loop 2 equation:
v = 2MI2 + V s2 + 8MI2
Or in matrix form:
5M

1
0
Solving: I1 = -1.1

0
1
10M

1 I1 Vs1

0 I2 = Is 3
1 v Vs2

A, I2 = -0.95

A.

The power Ps3 = Is3 xv= 7.58 W

Solution 3.50 I2 = 2A, I3 = 7A

Loop 1 equation:
Vs = 3I1 + 3(I1 I 2 ) + 6(I1 + I3 I 2 ) + 2v y + 2(I1 + I 3) = 14I1 + 2v y 74 14I1 + 2VY = 88
vy = 3(I1 I2 )= 3I1 6

Solving the above system, we obtain: I1 = 5A vy = 9, v can be found from the loop 3 equation
v + (I1 + I3 I 2 ).6 + 2v y + 2(I3 + I1) = 0 . Solving, we obtain v = 10V
Finally, Pvs = VsI1 = 70W.

Solution 3.51
I1

I3
I2

Mesh 1 equation:
9kI1 + 3k (I1 I4 ) + 6k (I1 I3 ) + 12k (I1 I2 ) = 0
where we have used the fact that Ix = I1 I4 (and I4 = 4mA)
Mesh 3 equation:
6k (I3 I1 ) v2 +2kI3 + v = 0
Mesh 2 equation:
2.4kI2 + 12k (I2 I1 ) v = 0
Constraint equations:
I2 I3 = 0.5mA
I3 I4 = 0.5Ix = 0.5I1 0.51I 4 = 0.51I1 + 2mA ; 0.51I1 I3 = 2mA
The above five equations need to be put into matrix form:
30k 12k 6k 0 0 I1 12
6k
0
8k 1 1 I2 0

12k 14.4k 0
1 0 I 3 = 0

1
1 0 0 v 0.5m
0
0.5
0
1 0 0 v 2 2m

The solution is:

 I1 0.002A
I 0.0025A
2

I 3 = 0.003A

v 12V
v 24V
2

I x = I1 I4 = 2mA
Power delivered by 0.5mA current source: P1 = 0.5mxv = 6mW
Power delivered by the dependent current source: P2 = 0.5I x v 2 = 1m (24) = 24mW
Power delivered by 4mA current source: P3 = 4m (3k I x V2 ) = 120mW

Solution 3.52

I5
I6

I1

I4

I2

Write the following equations:

Mesh 2 equation:
100 (I2 I1 ) + 150 (I2 I4 I6 ) + v = 0
Mesh 4 equation:
3vx + 30 (I4 I3 ) + 150 (I4 - I2 + I6 ) = 0
Mesh 5 equation:
800I5 3vx + 10( I5 I1 + I 6) =0
Mesh 6 equation:
250I 6 + 150(I6 + I 4 I2 ) + 10(I 6 + I5 I1) = 0
Constraint equation:

I3

Is2 = I 2 + I3
Substituting the values of I1 = I s1 = 1.15 and I3 = I s3 = 0.95 , and noting that
vx = (I1 I6 I5 )10
we can write:
150 1 I2 100I1
250 150 0
150 180 30 120 0 I 30I 30I
1

4 3
0

0
840
40 0 I 5 =
40I1

10
410 0 I6 10I1
150 150

0
0
0
0 v I3 I S2

I2 0.65
I 0.45
4

I5 = 0.05

I6 0.1
v 35

Finally, vx = 10V and v is as given above.

Solution 3.53 (a)

Replace the voltage source by current sources:
At node 1
is1 = (V2 V1) /12k
At node 2
(V2 V1) /12k + (V2 V3) /90k + (V2 V4 ) /10k = 0
At node 3
is2 = (V3 V2) /90k + V3 /10k
At node 4
is2 + (V4 V2 ) /10k + V 4 /90k = 0
V1 = 180
V4 V3 = 60
(b) In matrix form:

1/12k
0
0
1 0 V1 0
1/12k
1/12k 1/12k + 1/90 k + 1/10k
1/90k
1/10k
0 0 V2 0

0
1/90 k
1/90k + 1/10 k
0
0 1 V3 0

=
1/10k
0
1/90 k + 1/10k 0 1 V4 0
0
1
0
0
0
0 0 is1 180

0
0
1
1
0 0 is 2 60
The solution from MATLAB is
180.0000
127.2000
33.6000
93.6000
-0.0044
0.00232
(c) Power delivered by S1 is:
Ps1 = Vs1 (I s1 ) = 0.792W
Power delivered by S2 is:
Ps2 = Vs 2 (I s2 ) = 0.139W

Solution 3.54 (a) Replace the 100 ohm resistor, the controlled voltage source, and vs2 by current source.
Then write the node equations:
Is1 = ix + I1 + 0.09 (V 1 V3)
I1 = V2 /20 + 0.05 (V2 V3)
I2 = 0.05 (V3 V2 ) + 0.09 (V 3 V1)
V3 = Vs2
V1 V2 = 300ix = 300V1 /100
In matrix form:
0 0.09 1 0 V1 2
0.1
0
0.1 0.05 1 0 V2 0

0.09 0.05 0.14 0 1 V3 = 0

0
1
0 0 I1 50
0
2
1
0
0 0 I 2 0

(b) Using MATLAB to solve the above system the solution is:
V1 = -90.0000V
V2 = 180.0000V

V3 = 50.0000V
I1 = 15.5000A
I2 = -6.1000A
(c) Power delivered by the current source is
Ps1 = Is1 V1 = 180W
Power delivered by the voltage source is:
Ps2 = Vs 2 (I 2 ) = 305W

Solution 3.55 Modify the circuit so that it looks like the following:

The modified node equations are:

Is1 = - I y + I x
Ib = Iy + 0.2 (V C VB )
Is2 = 0.2 (VC VB) + Ix
The equations describing the constitutive relationships of elements in the original network are:
VA VB = Ib
I x = 0.2Vb
Ib = 0.4Vb
I y = 8 VA
These can be cast into a matrix equation and solved easily to obtain the same result as previously arrived
at.
In matrix form:

0
0

1
0

0
0.2
0.2
1
0.2
0.4
0

0
0.2
0.2
0
0
0
0

1
0
1
0
1
0
0

1
1
0
0
0
0
1

0
1
0
1
0
1
0

1
0
0

0
0
0

V A 0
V 0
B
VC 2

I x = 0
I 0
y
Ib 0

Is1 8

V A 7V
V 5V
B

VC 10V

I x = 1A
I 1A
y

Ib 2A

Is1 0A

Solution 3.56 Modify the circuit as follows:

The modified node equations are:

At node A:

Is = G 1 VA + G2(VA VB ) + Ia

At node B:

G2(VA VB ) = Ix + G4 (VB VC )

At node C:

0.25mVA + Ia = G4 (VC VB ) + G 5 VC

The equations describing the constitutive relationships of elements in the original network are:
Ix = G 3 VB
VA VC = 104 Ix

These can be cast into a matrix equation that can be solved in MATLAB.
In Matrix form:
0
1
0
0.25m 0.2m
0.2m 1m 0.8m 0
1

0.25m 0.8m 1m 1 0

0.1m
0
0
1
0
1
0
1 0 104

V A 2m
V 0
B
VC = 0

Ia 0
I 0
x

The solution is:

V A 38V
V 20V
B

VC = 18V

Ia 0.0075A
I 0.002A
x

We observe that we have obtained the same results as in problem 3.28.

Solution 3.57 Replace dependent source by i35 (from 3 to 5). Also, replace voltage source by i10 (from 1
to 0). Now, write the modified noted equations. The reference node is O:VO = 0V :
At node 1:

i10 = (V6 V1 ) + (V 2 V1)

At node 2:

2 = (V2 V1 ) + (V 2 V3)

At node 3:

i35 = (V4 V3 ) + (V 2 V3)

At node 4:
At node 5:
At node 6:
Constraints:

2 = 2 + (V 4 V3 ) + V 4
i35 = (V5 V6 ) + V 5
2 = (V5 V6 ) V6

V3 V5 = 15vx = 15V4
V1 = 5
The following matrix equation is obtained:

2
1

0
0

0
1

1 0
0
0 1 1 0 V1 0
2 1 0
0 0
0 0 V 2 2
1 2 1
0 0
0 1 V 3 0

0 1 2
0 0
0 0 V 4 0
=
0 0
0
2 1 0 1 V 5 0
0 0
0
1 2 0 0 V 6 2

0 1 15 1 0
0 0 i10 0
0 0
0
0 0
0 0 i35 5

The solution of this equation is obtained from MATLAB:

V1 = 5.0000V
V2 = 3.3571V
V3 = 0.2857V
V4 = -0.1429V
V5 = 1.8571V
V6 = -0.0714V
i10 = -6.7143A
i35 = 3.7857A
The power delivered by the dependent voltage source connected between nodes 3 and 5:
P35 = 15v x (i35 ) = 15 V4 (i35) = 8.115W
The power delivered by the current source connected between nodes 2 and 4:
P24 = (2A) (V2 V4 ) = 7W
The power delivered by the current source connected between nodes 4 and 6:
P46 = (2A) (V4 V6 ) = 0.143W
The power delivered by the voltage source connected between nodes 1 and 0:
P10 = 5V (I12 + I16 ) = 5x [(V1 V2 ) + (V1 V6 )] = 33.57W

Solution 3.59 Using the appropriate element stamps for each element of the circuit, we obtain the
following system:
0.15
0.2
0 V A 8 3
0.15 + 0.2
0.15
0.15 + 0.05
0
1 VB 3

=
0.2
0
0.25 + 0.2 1 VC 25

0
1
1
0 I x 440

Solution 3.60

0
1/20 k + 1/10k + gm1
1/10k g g

m1
m2

1/10k
V1 Is1
=
gm2 + 1/10k + 1/2.5 k V2 I s2

The solution is the same as that of problem 3.14.

Solution 3.62 (a) Because RT (T ) can be approximated by a straight line between (250,0 o C) and
(80,50 o C) it follows that:
RT (T )= 3.4T + 250
(b) For T = 25o C , RT = 165
(c) The voltage across the RT + RL series combination can be obtained from voltage division:
RT + RL
VT ,L =
12 = 4.7857V
RT + 2RL + R
This is the same as the voltage across Rx because the meter is at zero deflection. Thus,
Rx
12 = 4.7857. It follows that Rx = 165.84.
Rx + R
(d) We first denote the nodes:
A - the node common to R,R x and the voltmeter;
B - the node common to R,R and the voltage source;
C - the node common to RL ,RT and RL ;
D - the node common to Rx ,RL and the voltage source.
The reference node is D:VD = 0. It follows that VB = 12V .
We also have: vout = V A VC
The node equations are:

VA VC V A 12 V A
+
+
=0
Rm
R
Rx
Equivalently: VA (RRx + Rm Rx + Rm R) - VC RRx = 12Rm R x
At node A:

VA 4199.86 VC 41.46 = 19900.8

At node C:

(1)

VC 12
VC
V VA
+
+ C
=0
R + RL RT + RL
Rm

Equivalently: (VC 12) 0.004 +

VC
+ (VC V A ) 10 4 = 0
RT + 2.5

VC (0.004 RT + 0.01 + 1+ 0.00025 + RT 104 )

V A (RT 10 4 + 0.00025) = 0.48 (RT + 2.5)
The last equation can be rewritten as:
VC (0.0041RT + 1.01025) - VA (RT 10 4 + 0.00025) = 0.48(RT + 2.5)
From (1) and (2), we obtain:
(0.0041RT + 1.01025)

(2)

19900.8 + VA 4199.86

41.46

VA (RT 10 4 + 0.00025) = 0.48(RT + 2.5)

Equivalently: VA (RT 0.415 + 102.337) = 2.448 RT + 486.12
It follows that VA =

2.448 RT + 486.12
0.415 RT + 102.337

From the equation at node A:

V 12 VA
2.448 RT + 486.12
vout = V A VC = Rm A
+
=
(100.3) + 480

R
Rx
0.415 RT + 102.337
At T = 0o C:

RT = 250. It follows that vout = 54.4415V

At T = 50o c:

RT = 80. It follows that vout = 80V

(e): The formula has been derived at part d):

T

RT

vout

0o C

250

54.4415V

5o C

233

52.4136V

10o C

216

50.2368V

15o C

199

47.8938V

20 C

182

45.3650V

25o C

165

42.6273V

30oC

39.6537V

148

Solution 3.63
Place a source Vin between C and D, and calculate the current drawn from the source as below:

Loop 1 equation:
I1R1 + (I1 I2 )R2 + (I1 I 3)R3 = 0
Equivalently:
I1(R1 + R2 + R3 ) I 2R2 I3R3 = 0
Loop 2 equation:
(I 2 I1)R2 + I2 R4 1 = 0
Equivalently:
I1R2 + I2 (R2 + R4 ) = 1
Loop 3 equation:
1 + I 3R5 + (I3 I1)R3 = 0
Equivalently:
R3I1 + I 3(R3 + R5 ) = 1
We obtain the following system of equations:
30I1 4I2 6I3 = 0

4I1 + 6I2 = 1
6I + 14I = 1

1
3

I1 = 0.0096A , I2 = 0.1731A , I3 = 0.0673A;

Iin = I 2 I 3 = 0.2404 A
V
1
Reg,CD = in =
= 4.16
Iin 0.2404

Solution 3.64
The node equation at node A is:
VA G1 + (V A VB )G2 + (VA VC )G3 = 0
Equivalently:
(G1 + G2 + G3)VA VB G2 VC G3 = 0
The supernode is identified by a Gaussian surface enclosing the controlled voltage source. The supernode
equation is:
G2(VA + VB ) 6 + G4VC + G3 (VC VA ) = 0
Equivalently, we have:
V A (G2 + G3) + G2VB + VC (G3 + G4 ) = 6
One way of obtaining the solution to the problem is:
We multiply the above two equations by 30.
30(G2 + G3)VA + 30G2VB + 30(G3 + G4 )VC = 180
and
30(G1 + G2 + G3 )V A 30G2VB 30G3VC = 0
By equating the coefficient of the above two equations with the coefficients of the first and second given
equations, we obtain:
30G2 = 30 G2 = 0.1S
30G3 = 2 G3 = 0.067S
30(G1 + G2 + G3 ) = 11 G1 = 0.2S
30(G3 + G4 ) = 32 G4 = 0.87S
can be obtained as follows:
VC VB = V X = (VC VA )
Equivalently:
VA VB + (1 )VC = 0
By comparing with the third given equation

Solution of 3.66
(a)
At node A:
At node B:
At node C:
At node D:

= 3.

(VA-VC)/2 + (VA-VB)/2 + (VA-VD)/2 = 14

(VB-VA)/2 + (VB-VC)/2 + (VB-VD)/2 = 7
(VC-VA)/2 + (VC-VB)/2 + (VC-VD)/2 + 2VC = 0
(VD-VA)/2 + (VD-VB)/2 + (VD-VC)/2 + 0.5VD = 0

These can be solved in MATLAB to obtain:

22.0000
18.5000
7.5000
12.0000
(b) Mesh analysis would result in the same voltages

The loops and their current loops are:

A,C,14A:I1

A,C,D:I2

A,B,D:I3

B,C,D:I 4

C,D,reference node:I 5

B,D,7 A:I6
I1 = 14A,I6 = 7A
Loop ACDequation: 2(I 2 I1) + 2(I 2 I 3) + 2(I 2 + I4 + I5) = 0
Loop BCD equation: 2(I 3 + I 6 + I4 ) + 2(I2 + I 4 + I 5) + 2I 4 = 0
Loop ABD equation: 2(I 3 + I 4 + I 6 ) + 2(I3 I 2 ) + 2I3 = 0
Loop CDref node equation: 2(I 5 + I 2 + I4 ) + 2(I5 I 6 ) + 0.5(I5 + I1) = 0
In matrix form:

6
2

2
2
6
0

2
6
2
2

2
2
0

4.5

I2 28
I 14
3=

I 4 14 ;

I5 7

I2 6.75
I 1.75
3=

I 4 5.5

I5 1

VA = 2(I1 I 2 ) + 0.5(I1 + I5 ) = 22V

VB = 2(I 4 + I 3 + I 6 ) + 2(I6 I5 ) = 18.5V
VC = 0.5(I1 + I5 ) = 7.5V
VD = 2(I 6 I 5) = 12V
(c) Mesh analysis requires more work.
(d) The removal of the top resistor will result in more node equations than loop equations. The addition of
a resistor between node A and reference node will result in more loop equations than node equations.

Thevenin Probs, 7/11/01

- P4.1 -

@R.A. Decarlo & P. M. Lin

PROBLEM SOLUTIONS CHAPTER 4

SOLUTION 4.1. First, find Vout / Vs for each circuit. Then solve for R knowing
Vout =

P 10 = 14.142V .

(a) Writing KCL at the inverting terminal, 1 / 1 k(v vs ) = 1/ R(Vout v ) Vout / Vs = R / 1k , since
the inverting terminal is a virtual short. Solving for R = Vout 1k / Vs = 2.828k .
(b) Writing KCL at the inverting terminal, Vs /1.5k = (Vout Vs )/ R Vout / Vs = R / 1.5k + 1, solving
for R = 1.5k(Vout / Vs 1) = 2.743k.
(c) From (a) Vout / Vs = 12k / R , thus R = 12k Vs / Vout = 4.243k.
(d) This is the same circuit as (b) except the output voltage is taken across two resistors. Thus
Vout =

P
(10 + 6) = 22.627V . Using the general form from (b), R = 400(Vout / Vs 1) = 1.410k
10

SOLUTION 4.2. (a) First, find the voltage at the non-inverting terminal as v + = 1/ 2 Vs . Then write KCL
at the inverting terminal, and make use of the virtual short property,
(Vs / 2)/10 k = (Vout Vs / 2) / 30k Vout / Vs = 30k(1/ 20k + 1/ 60k) = 2 .
(b) Relating the output of the amplifier to the output of the circuit, Vout = Vamp (500 / 800) . Then writing
KCL at the inverting terminal, Vs / 400 = (Vamp Vs ) /1.2 k Vamp / Vs = 1.2k / 400 + 1 = 4. Therefore
Vout / Vs = (Vamp / Vs ) (Vout / Vamp ) = 2.5.
(c) Note that since no current goes into the non-inverting terminal of the op-amp, the voltage at that node is
Vs. KCL at the inverting terminal, Vs / 4k = (Vout + Vs )/ 20k Vout / Vs = 6.
SOLUTION 4.3. Write KCL for both terminals,
(V Vi ) / 1 k = (Vo V ) / 2k
V /1k = (Vo V )/ 3k
Solving and doing the appropriate substitutions, Vo / Vi = 8 .
SOLUTION 4.4. This is essentially the basic inverting configuration, which is defined as
Vo / Vi = 2k / 1k = 2 .

Thevenin Probs, 7/11/01

SOLUTION 4.5. (a) By voltage division VL = 1V

- P4.2 -

@R.A. Decarlo & P. M. Lin

100
= 0.5V . Using Ohms law
200

1
Is = I L =
= 5mA .
100 + 100
(b) No current flows in the input terminal of an ideal op-amp, thus Is = 0A and VL = 1V . From Ohms
law Ia = I L = VL /100 = 10mA.
SOLUTION 4.6. (a) Using voltage division,
32||(8 + 24) 2
V1 = Vs
= V
[32||(8 + 24)] + 8 3 s
24
Vout = V1
= 0.5Vs
24 + 8
(b) By voltage division,
32
V1 = Vs
= 0.8Vs
32 + 40
24
Vout = V1
= 0.6Vs
24 + 8
32
(c) Using voltage division, V1 = Vs
= 0.8Vs , as no current enters the non-inverting terminal of the
32 + 8
24
op-amp. Due to the virtual short property, Vout = V1
= 0.6V1 . This is indeed the same results as
24 + 8
(b), which should be expected because of the isolation provided by the ideal buffers.
SOLUTION 4.7. Write KCL at the inverting terminal,
Vs1 /1k Vs 2 / 2k = Vout / 4k Vout = 4Vs1 2Vs2 = 40mV .
SOLUTION 4.8. (a) The voltage at the non-inverting terminal is V+ = 3 / 2V , KCL at the inverting
terminal gives (1.5 2.5) /10k = (Vout 1.5)/ 30k Vout = 1.5V . The power is
2
P = Vout
/ 500 = 4.5mW .

(b) The voltage at the non-inverting terminal is 3V this time, thus KCL
2
(3 2.5)/ 10k = (Vout 3) / 30k Vout = 4.5V . The power is P = Vout
/ 500 = 40.5mW .

SOLUTION 4.9. (a) Define the point between the two op-amp as Vint . Observe that the first op-amp is in
the basic inverting configuration, and the second the non-inverting configuration. By inspection,

Thevenin Probs, 7/11/01

- P4.3 -

@R.A. Decarlo & P. M. Lin

Vint / Vs = R1 / 1k
Vout / Vint = (1+ R2 /1k )
Cascading the two stages, (Vint / Vs )(Vout / Vint ) = Vout / Vs = R1 /1k(1+ R2 /1k) . Solving for
R1 = 20 1k / (1+ R2 /1k) = 5k. The power absorbed is P = (20 0.5)2 / 8 = 12.5W .
(b) Using the same equations as (a), solve for R2 = (20 1k / 2k 1)1k = 9k .
(c) Rewriting the equation obtained in (a), R12 + 1kR1 20M = 0, and solving the quadratic equation yields
R1 = R2 = 4k .
SOLUTION 4.10. This is a cascade of two non-inverting configuration op-amp of the form
Vo / Vs = (1+ 10k /10k ) for each. Therefore 2 2 = 4.
SOLUTION 4.11. This system is made up of a non-inverting stage with a gain of 1+10k/10k, a voltage
divider of gain 8k/(2k+8k), and a second non-inverting stage of gain 1+10k/10k. The product of all three
yields Vout / Vin = (2)(0.8)(2) = 3.2.
SOLUTION 4.12. (a) By inspection, the gain of the first stage is 1. Then write KCL for the second stage
Vs1 / 2R Vs2 / R = Vout / 2R Vout = Vs1 2Vs2 = 10V .
(b) The first stage gain is 0.5, thus Vout = 2R(0.5Vs1) / 2R 2R(Vs 2 )/ 0.5R = 7.5V , using the same
procedure as in (a).
SOLUTION 4.13. (a) This is a cascade of a summing amplifier with the following transfer characteristic,
Vo = 4Vs1 2Vs2 , and an inverting stage of gain 1.5. Thus Vout = 1.5(4Vs1 + 2Vs2 ) = 2.25V .
(b) Notice that the only difference is the gain of the inverting stage, which is now 2. Therefore
Vout = 2(4Vs1 + 2Vs2 ) = 3V .
SOLUTION 4.14. This circuit is a cascade of two summing amplifier where the output of the first is an
input of the second stage. The transfer function of the first stage is Vo = 2RVs1 / 2R 2RVs2 / R , which
is substituted in the transfer function of the second stage to obtain
Vout = R[2RVs1 / 2R 2RVs2 / R]/ R RVs3 / R = Vs1 + 2Vs2 Vs3 = 2V .
SOLUTION 4.15. Writing KCL at the inverting node, V1 / R1 V2 / R2 V3 / R3 = Vout / R f , and
Rf
Rf
Rf
solving for Vout = V1 +
V2 +
V .
R2
R3 3
R1

Thevenin Probs, 7/11/01

- P4.4 -

@R.A. Decarlo & P. M. Lin

SOLUTION 4.16. Referring figure P4.15, the value of the resistance must satisfy the following
R1 = R2 = R3 = 3R
constraints:
Rf = R
These will yield the inverted average. If polarity is a concern, a second inverting stage should be added
with a unity gain, i.e. both Rs equal.
SOLUTION 4.17. Using the topology of 4.12 the following parameters are chosen,
Ga1 = 3, Ga2 = 5, Gb1 = 2, Gb2 = 4
For the time being assume G f = 1. Now we calculate = (1+ 3 + 5) (2 + 4) = 3, this sets Gg = 3.
(a) The requirement for G f = 10S sets the scaling factor K = 10 / 1 = 10 . This then yields the
following set of parameters,
Ga1 = 30S, Ga2 = 50 S, Gb1 = 20S, Gb2 = 40S, G f = 10S, Gg = 30S
(b) The requirement for G f = 2S , sets the scaling constant to 2uS. So the following parameters are
obtained:
Ga1 = 6S, Ga2 = 10 S, Gb1 = 4 S, Gb2 = 8S
Furthermore for Gg = 12S , G = 6S in order to make the incident conductance equal at both terminal.
(c) Using the starting values from (a), one could choose a scaling constant of 5 S. This will yield the
following resistances:
Ra1 = 66.67k, Ra2 = 40k, Rb1 = 100k, Rb2 = 50k, R f = 200k, Rg = 66.67k
These are all reasonable values for circuit implementation.
SOLUTION 4.18. (a) Choosing the following initial values:
Ga1 = 3S, Ga2 = 5S, Gb1 = 11S, Gb2 = 4S, G f = 1S

Thevenin Probs, 7/11/01

- P4.5 -

@R.A. Decarlo & P. M. Lin

then calculate = (1+ 3 + 5) (11+ 4) = 6 . Thus Gg = 1S, and G = 1 + 6 = 7S . Scaling everything by

1S, yield this final set of parameters, which meet the requirements.
Ga1 = 3S, Ga2 = 5S, Gb1 = 11S, Gb2 = 4S, G f = 1S, Gg = 1S, G = 7uS
(b) The set of parameters remains unchanged, except for G which now becomes 6uS in order to
maintain the equal termination conductance requirement due to Gg = 0S.
(c) Scale the initial parameters of (a) by 5uS, and get the following set of resistances:
Ra1 = 66.67k, Ra2 = 40k, Rb1 = 18.18k, Rb2 = 50k, R f = 200k, Rg = 200k, R = 28.57k

SOLUTION 4.19. (a) Choosing the following initial set of parameters:

Ra1 = 1/ (4S) = 0.25, Ra2 = 1/ (2S) = 0.5, Rb1 = 1/ (5S) = 1/ 5, Rb2 = 1/ (4S) = 0.25, R f = 1
and = (1+ 4 + 2) (5 + 4) = 2 , thus choose Rg = 1/ (1S) = 1 and R = 1/ (1+ 2) = 1/ 3 . To meet
the R f = 50k requirement, all the parameters must be scaled by 50k, which gives
Ra1 = 12.5k, Ra2 = 25k, Rb1 = 10k, Rb2 = 12.5k, R f = 50k, Rg = 50k, R = 16.67k
(b) Same as (a) with a 100k scaling constant:
Ra1 = 25k, Ra2 = 50k, Rb1 = 20k, Rb2 = 25k, R f = 100k, Rg = 100k, R = 33.33k
SOLUTION 4.20. (a) When the op-amp is in its active region vout / vs = 5 . Thus it will operate in its
active region when 3 vs 3, and will saturate at 15V when vs 3 , and at 15V when vs 3. SPICE
yield the following plot:

Thevenin Probs, 7/11/01

(b) Using SPICE the following plot is obtained:

- P4.6 -

@R.A. Decarlo & P. M. Lin

Thevenin Probs, 7/11/01

- P4.7 -

@R.A. Decarlo & P. M. Lin

SOLUTION 4.21. The first stage is in a summing configuration, thus its output is, assuming it's in the
active region of operation, -15 V which means it is just about to saturate. The second stage is in the
inverting configuration with a gain of -1.5, which means that the overall output will be saturated at 15V.
v + 1.5
SOLUTION 4.22. When vin 80k in
> 0, or vin > 6 the output of the comparator saturates at
100k
15 V, when it is vin < 6 , it will saturate at 15 V. The following plot is obtained from SPICE.

v + 20
SOLUTION 4.23. When vin 10k in
> 0 , or vin > 2 the output of the comparator will be
110k
saturated at -15V. Otherwise when it is < 2V the output saturates at 15V. In SPICE:

Thevenin Probs, 7/11/01

- P4.8 -

@R.A. Decarlo & P. M. Lin

SOLUTION 4.24. Based on the same reasoning as the previous questions,

R + R2
R1
The output will be +Vsat, when vin < vref 1 1
v , and Vsat for
=

R2
R2 ref

R + R2
R1
vin > vref 1 1
v .
=

R2
R2 ref
SOLUTION 4.25. Using the previously derived relationship, and the topology of figure P4.24, set
vref = 1.5V , and R1 = 2k and R2 = 3k . Set the power supplies to the Op-amp to +/ 10V to satisfy
the Vsat requirement. Also the input to the inverting and non-inverting terminal are reversed for fig. P4.24.
Verifying in SPICE we obtain the following,

Thevenin Probs, 7/11/01

- P4.9 -

@R.A. Decarlo & P. M. Lin

SOLUTION 4.26. The design that fulfills the requirement is the same as for P4.25, with the input to the op
amp reversed. The following is obtained from SPICE,

Thevenin Probs, 7/11/01

- P4.10 -

@R.A. Decarlo & P. M. Lin

SOLUTION 4.27. First, for the comparator to give +Vsat for the lower voltages, the inputs to the op amp
in the topology of P4.24 must be interchanged. Then the components are chosen to satisfy the following

R + R2
relationship, vswitch = vref 1 1
=

R2
Verifying in SPICE,

R1
v . Choose vref = 1.5V , and R1 = 2k and R2 = 1k.
R2 ref

Thevenin Probs, 7/11/01

- P4.11 -

@R.A. Decarlo & P. M. Lin

SOLUTION 4.28. Write KCL at the inverting terminal, noting that the no current flows into it:
(V vin )/ R = (vout V )/ R . Use the following relationship vout = A(V+ V ) = AV . Solving
A
using the previous two equations yields vout / vin =
.
A+2
SOLUTION 4.29. (a) By inspection the voltage gain for the ideal case is 1. When A=1000, the gain
becomes 0.998, thus 0.2%.
ARf
(b) Repeating the method of P4.28, and setting vout / vin =
to 1 and solving for
R f + R1 + AR1
R f = 10.417k.
(c) Solving the previous equation when the gain is 1, R f / R1 = (A + 1)/ ( A 1).
SOLUTION 4.30. (a)The first part was obtained in P4.29. Rearranging the equation yields

Rf
1

.
vout / vin =
R1 1 + 1+ R f / R1 / A

Thevenin Probs, 7/11/01

- P4.12 -

@R.A. Decarlo & P. M. Lin

(b) The error is caused by (1+ R f / R1) / A in the denominator, and may be defined, in percent, as

100. Thus for the conditions listed in the problem, it will always be less than
100
1+ 1+ R f / R1 / A

2.05%. With A = 10000 it will be less than 0.21%.

SOLUTION 4.31. (a) Substituting the non-ideal model, and writing KCL at the inverting terminal,
(V vin )/ R1 + V / Rin = (vout V ) / R f is obtained. Now observe the following dependencies,
iout = vout / RL , and vout = AV (iout + (vout V )/ R f )Rout . Using these three equations, substitute
the second into the third and then solve for vout / vin using the last two. This yields
1 Rout Rout

Rf
Rf
RL
Rf
Rf

V = vout + vin
1/
+
+ 1 = vout
R
R1 R1 Rin

A out

Rf

and

Rf
vout / vin =
R1

R
R
1 + Rout + Rout 1+ f + f

Rf
RL
R1 Rin
1+

R

A out

Rf

A gain of 9.988
(b) For an ideal op-amp the gain is R f / R1 =-10.
(c) The error is about 0.1175%.
SOLUTION 4.32. The gain is 9.883, and the error 1.16%
SOLUTION 4.33. This derivation was performed in P4.31.

Thevenin Probs, 7/11/01

- P4.13 -

@R.A. Decarlo & P. M. Lin

SOLUTION 4.34. Assume that the sliding contact is at the bottom of Rp . Then, writing KCL at the
inverting terminal yields vin / Ro = (vout vin )/ R p . This implies vout / vin = 1 + Rp / Ro . When the
Rp
slider is at the top, it is evident that vout = vin . Therefore 1 vout / vin 1 +
.
Ro
SOLUTION 4.35. Writing KCL at inverting input, and making use of voltage division,

Rf
v
vin / R1 = vout / R f where is the fraction of vout that appears across Rf. Hence, out =
.
vin
R1
Rf
v
When the slider is at the top = 1 and out =
. When the slider is at the bottom, the fraction of vout
vin
R1
appearing across Rf is =

R f / / R0
R f R0
R f R0
1
=
R R
=
. Hence
R f / / R0 + Rp R f + R0
f 0
R
R
+
R
(R
+
R
)
f
0
p
f
0
+ Rp
R f + R0

1 R f R0 + R p (R f + R0 )
=
=1 +

R f R0

Rp
R0

Rp
Rf

. It follows that

Rf
Rf
Rp Rp
vout
=
=
1
+
+
.
vin
R1
R1
R0 R f

Therefore the range of achievable voltage gain is

Rf

Rf
v
out
R1
vin
R1

Rp Rp
1 + R + R

0
f

SOLUTION 4.36. Using the basic non-inverting configuration of figure 4.10 characterized by
Rf
Rf

vout / vin = 1+
, i.e., = 1+
.

R1

R1
SOLUTION 4.37. At first glance, one might use two inverting configurations, figure 4.5, in cascade.
However, such would not have infinite input resistance. To circumvent this problem we add a buffer
amplifier as per figure 4.7 at the front end of a cascade of two inverting configurations. The resulting
R f1 R f 2
overall gain is =

. Indeed, such a configuration can achieve theoretically any gain greater

R11 R12
than zero.

Thevenin Probs, 7/11/01

- P4.14 -

@R.A. Decarlo & P. M. Lin

SOLUTION 4.38. Using a single inverting amplifier configuration, figure 4.5, preceded by a buffer stage
Rf
of figure 4.7. The gain is =
.
R1
SOLUTION 4.39. By KVL for figure P4.39a, Vo = i1R f . Thus to achieve Vo = i1rm in figure P4.39b,
we set R f = rm .
SOLUTION 4.40. Writing KCL at the inverting node of the ideal op amp yields IL = Vi / Ra , which is
indeed independent of the load resistor which has no effect on the load current.
R
SOLUTION 4.41. The current through the LED is IL = 10 1 / 3.8k , so for (a) it is 1.32mA and for
10k
(b) 2.11mA.
SOLUTION 4.42. Applying KCL at the inverting terminal, IL = vin / R1 . Again, ideally, RL does not
affect IL.
SOLUTION 4.43. (a) Defining a temporary voltage Vo at the output of the op-amp, we can write KCL at
the inverting and non-inverting terminal:
(V 2)/ 1k = (Vo V ) / 2k
V /100 + (V Vo ) / 200 = Iout
Substituting the first equation into the second and simplifying causes Vo to drop out and Iout = 20mA .
(b) The answer remains the same as the value of the load resistance was not used for finding the load
current.
SOLUTION 4.44. Using the same approach as for the previous question, but with resistor labels instead,
the following equations are obtained from KCL:
R V + R1Vo
V = 2 s
R2 + R1
R + R1
Iout = V 2

R1R2
Substituting the first into the second yields Iout = Vs / R1 .

Vo
R2

Thevenin Probs, 7/11/01

- P4.15 -

@R.A. Decarlo & P. M. Lin

SOLUTION 4.45. (a) Vs = 5 V, (b) Iout = 10mA sets R1 = Vs / Iout = 500 . (c) From KVL and Ohm's
law, Is = (Vs + RL Iout ) / R1 . We require Is < 0.5 mA. This means that in the worst case, RL = 500 ,
(Vs + RL Iout ) 5 + 500 0.01
=
= 40 < .
0.25
0.5 103 R1
(d) From KVL and Ohm's,
Vo = RL Iout (Iout + (RL Iout )/ R1)R2 20 V
Hence
R2

20 RL Iout
20 5
=
= 750
(Iout + (RL Iout ) / R1 ) 0.01 + 5/ 500

Hence one design is to pick R2 = 750 and = 40 which impliesR2 = 30 k.

SOLUTION 4.46. The exact same design as P4.45 can be used with the isolation buffer of figure 4.7
placed at the input of it in order to provide the infinite input resistance needed by P4.46b.
SOLUTION 4.47. The general expression for this summing circuit is

Vout =

Rf
Ro

Vo

Rf
R1

V1

Rf
R2

V2

Rf
R3

(a) Using the expression above | Vout |=| 1 0 0 8 | E = 9E .

(b) | Vout |=| 0 2 4 0 | E = 6E .
(c) It has to be a linear combination of 8, 4, 2, 1, thus [1 1 0 1] would yield 13E.
(d) With the same approach, [0 1 1 1].
SOLUTION 4.48. For this implementation we add an extra R-2R branch along with an extra summing
input to the op amp. From the theory developed in Example 4.9 the total resistance seen by the source is
E2
2R. For the total power supplied by the source to be less than 0.02 W, we require R
= 2.5 k.
2( 0.02)

Thevenin Probs, 7/11/01

- P4.16 -

@R.A. Decarlo & P. M. Lin

SOLUTION 4.49. The same steps as in the previous questions are repeated. Because the resistance seen
by the source is unchanged no matter how many branches are added to the R-2R network,
R

E2
= 5 k.
2( 0.01)

SOLUTION 4.50. (a) If the input is 3vmax / 8, then the first comparator will give Vsat, keeping S2 down.
The next comparator will output +Vsat, causing S1 to go up. After subtraction, the input to the last
comparator is vmax / 8 yielding +Vsat at its output since its input is slightly above the reference input
level. Thus the logic output values are [0 1 1].
(b) Putting in 6v max / 8, will cause +Vsat and S2 to go up. The input to the second comparator will be
2vmax / 8, which will cause +Vsat and S1 to go up. The input to the last comparator will be 0, thus it will
output Vsat. The corresponding logic output is [1 1 0].
SOLUTION 4.51. Simply add a subtractor and switch to the last comparator, followed by an additional

comparator. The reference level to the new (additional) comparator will be vmax
/16 , and its output will be

the new least significant bit.

SOLUTION 4.52. (a) Writing the node equation for figure P4.52c,
Vout (Vout V1) A(0 V1) Vout
+
=
10k
Rout
R'L
which implies that

Vout / V1 =

1
A
10k R
out

1
1
1
R' + 10k + R
out
L

For figure P4.52a, the corresponding node equation is

Vout A(0 Vin ) Vout
=
RL
Rout
which leads to

Thevenin Probs, 7/11/01

- P4.17 -

@R.A. Decarlo & P. M. Lin

A
R
out

Vout / Vin =
1
1
R + R
L
out
Note that

1
1
1
= 1 +
, which when substituted into the later equation make both of them
RL RL 10k

approximately the same since the 1/10k term in the numerator of Vout / V1 has a negligible contribution.
(b) Writing the node equation for figure P4.52d, yields
Vout (Vout V2 ) A(0 V1 ) Vout
+
=
10k
Rout
R'L
Hence

V2
100|| Rin
V1 = V2

(100 || Rin ) + 10k 101

1
A
10k 101R
out

Solving produces Vout / V2 =

. Note that as in (a) the 1/10k term in the numerator is
1
1
1
R1 + 10k + R
out
L
V
V
negligible; after eliminating this negligible term, one sees that out is 101 time smaller than out .
V2
Vin
SOLUTION 4.53. (a) Using the equation just derived, after substituting in the values, the gain is 980.392
(b) From the previous equation, Vout / V1 = 980.382 ; write KCL at the non-inverting terminal to obtain,
Vin V1
V
V Vout
= 1 + 1
; substitute Vout = 980.382V1 ; solve for V1 / Vin , and then multiply both gains
10k
Rin
10k
to obtain (Vout / V1)(V1 / Vin ) = Vout / Vin = 0.9979 .
(c) They only differ by about 0.01%, thus they are very similar.
R
R
SOLUTION 4.54. Writing out the transfer equation, Vout = 2 Vs2 2 Vs1 , thus R2 / R1 = 4 . Using
R1
R1
R2 = 100k , R1 = 25k . As expected SPICE shows to noticeable difference in outputs when the source
resistances are varied.
SOLUTION 4.55. Due to the ideal nature of the op-amp, the voltage VRb = Vs2 Vs1 . By KVL

Thevenin Probs, 7/11/01

- P4.18 -

@R.A. Decarlo & P. M. Lin

V2 = Vs2 + Ra (Vs2 Vs1 )/ Rb

V1 = Vs1 Ra (Vs 2 Vs1) / Rb
Next, V1 V2 = (Vs1 Vs 2 )(1+ 2Ra / Rb ) .
R
R
SOLUTION 4.56. (a) Noticing that the final stage is a summing op-amp in which Vout = 2 V1 2 V2 .
R1
R1
R
From the previous question, Vout = 2 (V1 V2 ) =
R1

R2
(1+ 2Ra / Rb )(Vs1 Vs2 ) . Thus
R1

R
= 2 (1+ 2Ra / Rb ). The gain can be varied by adjusting the single resistance Rb .
R1
(b) Picking the set of values below will satisfy the requirement:
R2 = 100k, R1 = 100k, Ra = 20k, Rb = 10k .
(c) Doing the SPICE simulation using the parameters from (b) yield 5 V at the output for
Vs1 Vs2 = 2 1 V. Setting Rb arbitrarily to 20 k, the output now becomes 3 V, which agrees with the
relationships developed earlier.

PROBLEM SOLUTIONS CHAPTER 5.

Solution 5.1. (a) Vs = 10 V, P = 20 W and P = Vs Is implies Is = 2 A.

(b) Rin = Vs /Is = 10/2 = 5

(c) By the linearity/proportionality property:

Vsnew Isnew
2 Isnew
=
which
implies
=
implies
10
2
Vsold
Isold

Isnew = 0.4 A.
(d) P new = Vsnew Inew
= 2 0.4 = 0.8 watts. Observe that
s
P new 0.8 Vsnew 2
=

=
20 Vsold 10
P old
It follows that the proportionality property does not hold for power calculations.

Solution 5.2 First note that the ratio IR/VS is constant. With the given values of voltage and current, this
ratio is:
IR/VS = 0.25/25 = 0.01
Power dissipated in the resistor is
P = I R2 R = 2.5 IR2 = 2.5/R = 0.25 IR = 0.5
Since IR is always 0.01VS, it follows that VS = 50V.

Solution 5.3 Label the resistances R1 , R2 , and so on in the manner shown in Example 5.11. In this
problem, we have R1 to R10 (the last being the 2 Ohm resistance at the voltage source). First, assume that
V1 (the voltage across R1 ) is 1V. Then evaluate the rest of the currents and voltages until you deduce the

resulting VS. It should be noted that the equivalent resistance looking into R3 , R5 , R7 , and R9 is always
2.
V
V1 = 1 I1 = 1 = 0.25 I 2 = 0.25 V2 = I2 2 = 0.5 V
4
V
V3 = V1 + V2 = 1.5 I3 = 3 = 0.5 I4 = I3 + I 2 = 0.75 V4 = I4 4 = 3
3
V5 = V3 + V4 = 4.5 I5 =

V5
= 1.5 I6 = I5 + I 4 = 2.25 V6 = I5 4 = 9
3

V
V7 = V6 + V5 = 13.5 I7 = 7 = 4.5 I8 = I7 + I6 = 6.75 V8 = I8 4 = 27
3
V9 = V8 + V7 = 40.5 I9 =

V9
= 13.5 I10 = I9 + I8 = 20.25 V10 = I10 2 = 40.5
3

VS = V9 + V10 = 40.5 + 40.5 = 81 V

Thus, an 81 V input produces a 1 V output Vout = (1/81)VS = 2 V.

Solution 5.4 Label the resistances R1 to R10 progressively from right to left just like in the previous
problem. Then, assume Iout = 1 and proceed as follows:
Iout = 1 V1 = Iout 4 = 4 I2 =

V1
=1
4

V
I3 = I1 + I 2 = 2 V3 = I3 4 = 8 V4 = V3 + V1 = 12 I4 = 4 = 4
3
I5 = I4 + I 3 = 6 V5 = I5 4 = 24 V6 = V5 + V4 = 36 I6 =

V6
= 12
3

V
I7 = I6 + I5 = 18 V7 = I7 4 = 72 V8 = V7 + V6 = 108 I8 = 8 = 36
3
V
I9 = I8 + I 7 = 54 V9 = I9 4 = 216 V10 = V9 + V8 = 324 I10 = 10 = 108
3

IS = I10 + I9 = 162
Iout/IS = 1/162 Iout = (1/162)40.5 = 0.25A.

Solution 5.5 (a) MATLAB code given in problem.

(b) Subsitute to obtain V1 = 10V.
(c) Req = VS/IS = 11.6667.
(d) First, define r1 = 1:0.25:10;
then create an outermost loop around the code of part (a) as: for j=1:length(r1)
then, in the statement defining R, do R = [R1(j), R2, R3, R4, R5, R6, R7, R8];
Finally, replace the last statement with Vs(j) = V(n) + V(n-1); end;
The following is the resulting plot:

Solution 5.6 (a)

The following code can be used:
n = 9;

v = zeros(n,1);
i = zeros(n,1);
r = [r1 r2 r3 r4 r5 r6 r7 r8 r9];
i(1) = 1;
v(1) = i(1)*r(1);
i(2) = i(1);
for k=2:2:n-2
v(k) = r(k)*i(k);
v(k+1) = v(k)+v(k-1);
i(k+1) = v(k+1)/r(k+1);
i(k+2) = i(k+1) + i(k);
end;
v(8) = i(8)*r(8);
v(9) = v(8) + v(7);
i(9) = v(9)/r(9);
Is = i(9) + i(8);
It follows that Is = 16.9877A .
(b) By the proportionality property:
(c) Req = v(9)/Is = 38.15.

Solution 5.7 Va = 12V, iB = 60m

By inspection:
Vout_a = 300/90012 = 4V
Vout_b = (300||600)60m = 12

I1new Isnew
1
= old I1new =
200mA = 11.77mA
old
16.9877
I1
Is

Vout = 4 + 12 = 16V.

*S OLUTION 5.8. Part 1: Set the 3 A current source to zero. This generates an open circuit in place of
the current source eliminating the effect of the series 0.1 S resistor. The equivalent circuit is:

By voltage division,
1
2
0.25 + 0.2 + 0.05
VL12V =

12
=
12 = 2 V
1
1
2
+
10
+
0.25 + 0.2 + 0.05 0.1

Part 2: Set the 12 V source to zero. This generates a short circuit in place of the voltage source which
shorts out the effect of the 0.5 S resistor. The equivalent circuit is:

Note that the 0.1 S resistor in series with the 3 A source is redundant to the calculation of VL. Hence, by
Ohm's law,
VL3A =

1
3
3=
=5 V
0.25 + 0.2 + 0.05 + 0.1
0.6

Therefore by superposition,
VL = VL12V + VL3A = 2 + 5 = 7 V

Solution 5.9 Replace the dependent source by an independent voltage source VS:

VS

In the following analysis, we have to always compute Va because It defines the constraint on Vs. So, when
only the 88V source Is active, VA Is the result of voltage division between the 60||30 resistor and the
120||30 resistor. So,
Va_1 = 40V
And, since deactivated VS, Vout_1 = 0.
Now, due to the 55V source, we have

Now, the 120 and 60 resistors are in parallel, and the same can be said about the 30 and 30 resistors.
Thus, another voltage divider gives:
Va_2 = 15V and Vout_2 = 0V.
Finally, when VS1 is active, the left part of the circuit consists only of resistances, so Va_3 = 0. Vout is given
by another divider formula:
Vout_3 = 90/100VS
Now add all contributions:
Va = 40 15 + 0 = 25V
Vout = 0 + 0 + 0.9VS, where VS = 2Va.
Vout = 0.9225 = 45V.
Finally, P = V2 /R = 22.5W.

Solution 5.10
Due to 3A source:

iout = 1A by current division between the two paths. So, vout_1 = 2 V.

Due to 1A source:

iout = 2/3A again by current division. So, vout_2 = 4/3V.

Due to 1vV source:

iout = 14/6 = 7/3A (by Ohms Law). So, vout_3 = 14/3V.

Finally, vout = 6/3 + 4/3 + 14/3 = 8V, and the power delivered by the source is 81 = 8 W.

Solution 5.11
Due to 22 V source:

Req = 900||225 = 180. Now, by voltage divider:

Vout_1 = 0.522 = 11 V.
Due to the 20 V source:

Req = 180||225 = 100 . So, again, by voltage division:

Vout_2 = 900/(900+100)20 = 18 V.
Finally, due to current source:

We have three resistances in parallel with a resistance equal to 90 . So, Vout_3 = 0.190 = 9 V.

Vout = 11 + 18 + 9 = 38V and P = 3838/900 = 1.6 W.

Solution 5.12
Find contribution to Vout :

First, note that no current flows through R1 - R3 because of the virtual ground property of the op-amp.
Thus, this circuit is identical to the inverting amplifier studied in Chapter 4. So,
Vout _ 0 =

Rf
R0

V0

Similarly, when each of the other sources is activated, the circuit will be an inverting amplifier. So,
Vout _1 =

Rf
Rf
Rf
V1, Vout _2 =
V2, Vout _3 =
V
R1
R2
R3 3
V
V V
V
Vout = R f 0 + 1 + 2 + 3
R0 R1 R2 R3

Solution 5.13 Due to the 4 V source, the circuit looks like an inverting amplifier:

So, Vout_1 = -30/10(-4) = 12 (from the results of Chapter 4).

Again, note here that no current flows through the two resistances connected to the + terminal of the op
amp. Since no current flows through them, then no voltage develops across them. So, the + terminal can be
assumed to be connected to ground, and this is why we say that the circuit looks like that of the inverting
amplifier.

Now, due to the 6 V source:

The voltage at the + terminal is established by a resistive divider between the two 10K resistors. So, this
voltage is 3V. Thus, the voltage at the negative terminal is also 3V. We can now use KVL on the inverting
side of the op-amp to get:
Vout_2 = 3 + 30k0.3m = 12V

So, Vout = 12 + 12 = 24V, and P = (24)(24)/500 = 1.15W.

Solution 5.14
(a) When VS2 is deactivated. The circuit looks like two inverting amplifiers in cascade.
Thus, by inspection, V1 = 2VS1 and Vout_1 = 3V1 = 6VS1 = 3V.
(b) Similarly, when VS1 is zero, V1 is zero because the first inverting amplifier has zero input. Thus, the
circuit consists just of the second inverting amplifier:

Vout_2 = 3/2VS2 = 4.5

(c)

Vout = 3 4.5 = 1.5V.

*S OLUTION 5.15. For Vs1 and Vs2 , the situation reduces to the analysis of two inverting amplifiers in
cascade. For Vs3 , the situation is simply a single stage inverting amplifier. Note that because of the
virtual ground at the inverting terminal of the op amp, when Vs1 and Vs2 are zero, they have no
contribution to the input of the second stage.
(a) With Vs2 and Vs3 set to zero,
s1 R 2R
Vout
=
V = Vs1 = 5 V
R 2R s1

(b) With Vs1 and Vs3 set to zero,

s2 R 2R
Vout
=
V = 2Vs 2 = 2 (2.5) = 5 V
R R s2
(c) With Vs1 and Vs2 set to zero,

R
s3
Vout
= Vs3 = Vs3 = 2 V
R
(d) By superposition,
s1
s2
s2
Vout = Vout
+ Vout
+ Vout
= 2 V

Solution 5.16
If the op-amp were ideal, we would get:
Vout = 4VS1 2VS2 = 26
This is clearly beyond the linear range of operation of the op-amp. In other words, the amplifier responds
in a non-linear manner to this level of input. Hence, superposition, which relies on linearity, cannot be
used.

Solution 5.17
We know by linearity that
Vout = aIs1 + bV s2
Substitute the first measurement to obtain:
5 = 0 + b10 b = 0.5
Now, take the second measurement:
1 = a10 + 0 a = 0.1
So,
Vout = 0.1Is1 + 0.5V s2
At 20A, 20V: Vout = 12V

Solution 5.18
Again

Vout = aIs1 + bV s2

Substitute the two measurements to obtain:

5a + 10b = 15
2a + 5b = 10
These two simultaneous equations in a and b can easily be solved to obtain:
a=5

and

Therefore, at 1A and 5V,

Vout = 5 + 20 = 15V

Solution 5.19 (a)

Again,
Vout = aIs1 + bV s2

b=4

Substituting the result of the first measurement into this equation yields:
a4 cos(2t) + 0 = 2cos(2t) a = 0.5
Now, substitute the second measurement:
0 + 10b = 55 b = 5.5
Therefore, for the given input current and voltage:
Vout = cos(2t) 55cos(2t) = 56cos(2t) V
(b)

Vout = 2cos(5t) + 110cos(5t) = 112cos(5t) V

Solution 5.20 (a)

First of all,
Iload = aVa + bIb
Now, substitute the two measurements into this equation:
7a + 3b = 1
9a + b = 3
Solving these two equations for the unknowns a and b, we get
a = 0.4
(b)

and

b = 0.6

Iload = 0.415 0.69 = 0.6A

Solution 5.21 (a)

We know that the output is going to be a linear combination of the three inputs:
Vout = aIs1 + bV s2 + cVs3
Now, substitute the three measurements into this relationship:
50ma 2b + 5c = 13
0 + 3b + 5c = 2

0 + 2b + 4c = 0

These equations can be written in matrix form and solved as follows:

0.05 2 5 a 13
a 100

3 5 b = 2 b = 4
0
0
c 2
2 4 c 0
(b) Substitute the given values of the input sources to obtain:
Vout = 1001 440 + 210 = 40V

Solution 5.22
Vout = AIs1 + BV s2 + CV s3
Substituting the measurements into this equation results in a system of three equations and three
unknowns. This system can be written in matrix form by inspection:
30 10 3

3
2010
10103

2 1 A 11.5
A 150

4 2 B = 27 B = 5.5
C 4
3 1 C 14

This can be solved to obtain:

Solution 5.23 (a)

The coefficient matrix is inverted, and both sides of the nodal equation are multiplied by it to obtain:
0.233 0.02Vs1 0.00125Vs2
Va 43.0108
V = 43.0108 1.4337

Vs 2
b

Expanding the first row of the above equation gives:

Va = 0.8602Vs1 0.0538Vs2 + 0.233V s2
This is exactly in the form required, where A = 0.8602 and B = 0.1792

(b) For this part, we expand the second row of the equation:
Vb = 0.8602Vs1 0.0538Vs2 1.4337Vs2
Again, this is in the desired form, where A = 0.8602 and B = 1.4875
(c)

Vab = Va Vb = 0 + 1.667 V s2

Thus, A = 0 and B = 1.667

Solution 5.24 (a)

Again, we invert the coefficient matrix to obtain

0.0022
0.6296
0.0741 Vs1
I1 0.0022 0.0022
I 0.0022 0.0022
0.0022 0.7037
0.2593 0
2
I 3 = 0.0022 0.0022
0.0022 0.3704 0.0741 0


v1 0.2222 0.7778 0.2222 137.037 92.5926 I s2
v 0.3333 0.3333 0.6667 11.1111 77.7778 0
2

Expanding the first row of this equation:

I1 = 0.0022Vs1 + 0.6296Is2
So, A = 0.0022, B = 0.6296
(b) Similarly, expanding the third row:
I3 = 0.0022Vs1 + -0.3704Is2
So, A = 0.0022, B = -0.3704
(c) By the same procedure, A = -0.3333 and B = -11.111

Solution 5.25
Invert the coefficient matrix and mujltiply both sides of the equation in the problem by this inverse matrix
to obtain

3
2
1
4 I s1
V1 0 0 0
V 0 0 0
3
1
1
4 0
2
V3 0 0 0
12
3
3
16 0


4
1
1
5 0
Ia = 0 0 0
I 0 1 0 16.6
9.2 5.2 21.8 Vs2
b
Ic 1 1 0 42.1 16.2 11.7 53.8 0


Id 1 1 1 26.1 12.2 7.7 32.8 0

The second to last row can be expanded to get Ic = Is1 16.2Vs2 A = 1, B = 16.2

Solution 3.26 (a)

First compute the response due to V1 :

By voltage divider:
Vout_1 = 6/(2+6+1)V1 = 2/3V1

Then, due to I2 :

I2 flows through the 20 ohm resistor in series with 6||3. Thus,

Vout_2 = 2I2
Therefore,

Vout = 2/3V1 2I2

(b) Vout = 8cos(10t) 4

*S OLUTION 5.27. (a) By linearity Vout = A Vs1 + B I s2 .

To find A, let Is2 = 0. The circuit becomes a ladder network as follows.

A
Let Vout
= 1 V. Then

I1 = 1/420 + 1/70
I1 = 1.6667e-02
V400 = I1*40 + 1
V400 = 1.6667e+00
I400 = V400/400
I400 = 4.1667e-03
I2 = I1 + I400
I2 = 2.0833e-02

Vs1 = I2*20 + V400

Vs1 = 2.0833e+00
A = 1/Vs1
A = 4.8000e-01
To find B, let Vs1 = 0. The circuit becomes a ladder network as follows.

B
Again assume that Vout
= 1 V. Then

I1 = 1/420 + 1/70
I1 = 1.6667e-02
V400 = I1*40 + 1
V400 = 1.6667e+00
I400 = V400/400
I400 = 4.1667e-03
I20 = V400/20
I20 = 8.3333e-02
Is2 = I20 + I400 + I1
Is2 = 1.0417e-01
B = 1/Is2
B = 9.6000e+00
Hence by linearity Vout = 0.48 Vs1 + 9.6 Is 2 .
(b)
Vout = A*20 + B* 0.5
Vout = 1.4400e+01
(c) Doubling resistances does not change voltage ratios hence A is the same. However, the doubling also
doubles the voltage to current ratio. Hence, B is doubled. It follows that if all resistances are
doubled, then
Vout = 0.48 Vs1 + 19.2 I s2

Solution 5.28 (a) (b)

We solve parts (a) and (b) at the same time. First, we find the responses to Vs1 :

Equivalent resistance across Vo2 : 60||30 = 20

Now, by voltage divider:

Vo2_1 = 20/140Vs1

This voltage now divides between Vo1 and the 10 ohm resistance:
Vo1_1 = 20/30Vo2_1 = 2/31/7Vs1 = 2/21Vs1
Now, compute the responses due to Is2 :

Equivalent resistance across Vo2 : 60||120 = 40

By voltage division:

Vo2_2 = 40/50Vo1_2

Where

Vo1_2 = Is2 (50||20) = 14.286Is2

Now, add the contributions:

Vo1 = 0.0952Vs1 + 14.286Is2
Vo2 = 0.1429Vs1 + 11.4288Is2

Solution 5.29 (a)

Find contribution due to v1 :

The parallel combination results in a 12 ohm resistance in series with the remaining two. Thus, by voltage
division:
vout1 = 24/48v1 = 0.5v1
Now, due to v2

The equivalent resistance across v in this figure is 36||(24+12) = 18, which means that by voltage divider:
v = 0.5v2
Similarly, by another voltage division application
vout2 = 24 / 36 v= 0.660.5v2 = 0.333v2

Now, due to i3

Define Req1 = 24 + 36||18 = 36 . This resistance is in parallel with the 12 ohm resistance to introduce
an equivalent of 9 . The total voltage that develops across this 9 ohm resistance is 9is3 . This voltage
divides between vout and the 36||18 resistance:
vout3 = 24/36v1 = -924/36is3 = -6is3
Finally, due to i4:

A similar analysis of this resistive network can reveal that vout4 = 6is4 . Thus

vout = 0.5v1 + 0.333v2 6is3 + 6i s4

(b) For this part, note that scaling resistance values does not affect voltage ratios. This can be evident from
the application of any voltage divider formula. On the other hand, scaling resistances does affect current-

to-voltage or voltage-to-current ratios. This is by definition of a resistance! So, in the above equation,
doubling the resistances does not affect the first two terms, but doubles the second two terms.

Solution 5.30 (a)

Define two clockwise mesh currents: I1 in the bottom left loop and I2 in the top loop. The bottom right
loop has a current source, so it will not be considered:

Va (I1 -I2 )3 I2 (I1 ib ) = 0

and
6I2 I2 + (I2 I1 )3 = 0
Solving these two equations for the two currents gives:
I1 = 8 A

and

I2 = 3 A

The power delivered by the dependent source is: P = i x I ix where I ix is the current leaving the ' +'
terminal of the dependent voltage: P = i x (I2 I1) = I 2 (I2 I1) = 15W
vout = I1 + i b = 8 + 26 = 34V
(b) Now, we express vout = Ava + Bi b
Turning off ib , we still have two loops, in which we can define the same mesh currents as above to obtain:
va 3(I1 I2 ) I2 I1 = 0
I2 3(I2 I1 ) 6I2 = 0
Solving for the two currents, we get I1 = 4/13va which sets up vout across the 1 ohm resistor:
Vout1 = 4/13va
Now, turn off the voltage source:

Write a node equation at vout:

vout vout vout vout 6
+
+
ib = 0
6
1
3
9
? vout 2 = ib
13

So,

vout = 4/13va + 9/13i b

(c) Substitute into the above equation: vout = 35V

Solution 5.31 By inspection:

where the leftmost current is 0.25 with a resistance of 32. Similarly, the downward current is 0.75 and its
resistance is 32. This reduces to:

Solution 5.32
The circuit can be transformed as follows:

Write two nodal equations at 1 and 2:

v1
v v
9m + 1 2 = 0
2k
2k
v2 v1
v
v
+ 9m + 2 + 2 5m = 0
2k
9.6k 16k

1m +

These two equations can be solved using any method to obtain:

v1 = 11.2V

and

v2 = 2.4V

The power absorbed by the 9.6k resistor is:

P=

Solution 5.33

Then

V22
= 0.6mW
9.6 10 3

Output voltage is 5V
(b) P = 1.25W
(c) For a given resistance, doubling the voltage increases the current by two times. So, the current is
new
old
doubled. It follows that Vout
= 2 Vout
= 10V

Solution 5.34

This circuit is easy enough to solve by inspection. Vs = 28V.

Solution 5.35

Now, write two node equations:

v1
v
v
2 + 1 = 12m
500 500 500
v2
v
v
1 + 0.032v1 + 2 = 80m
500 500
125

These two equations can be solved to get v1 = 2.8V, v2 = -0.4V.

Solution 5.36

Now, we can write the node equations:

v1 v2 v1 100
v
+
+ 0.2 + 0.4 + 1 0.95 = 0
50 50 25 25
20
v2 100 v2 v1
v

+ + 2 + 1 0.1 0.2 = 0
100 100 50 50 100

Solving these two equations yields:

v1 = 25V

and

v2 = 20V

*S OLUTION 5.37. After a source transformation on the 30 V independent source and one on the 9Vx
dependent source we obtain the circuit below.

Transforming the two dependent voltage sources and combining yields the following circuit.

Writing a single node equation for Vx yields

V
V
V
7.5 = x + x + x = 1.25Vx
4
2
2
Hence, Vx = 6 V.

Solution 5.38
Replace the dependent source with a temporary independent source. When doing the analysis, always
compute Va in order to keep track of the constraint on the dependent source.

When Vt is not active, vout is obtained from a voltage divider between the 60||30 combination and the 15
ohm resistor:
vout1 = 4/7Vs
Similarly

va1 = 3/7Vs

Now, compute the response due to the temporary source.

Straightforward voltage division also applies here to get:

va2 = -2/7Vt
and

vout2 = -5/7Vt

So,

vout = 4/7Vs 5/7Vt

va = 3/7Vs 2/7Vt,

where Vt = va va = 3/7Vs 2/7va

Rearranging,
va =

3
Vs
7+2

Then,
4
5
Vs
v
7
7 a
4
vout =
V
7 + 2 s
vout =

Solution 5.39 Replace the dependent source by a temporary independent source:

When Vt is shorted, the result is a ladder network. The input resistance looking each of the vertical
branches is R. Label these vertical branches V1 , Va, V2 from left to right. It follows that
V1 = R/3RVs
_
Also,

Va_1 = R/2R R/3RVs = 1/6Vs

Vout_1 = 0

Now, short VS and turn on Vt.

Again, the result is a ladder network (note the symmetry in the above figure). Thus, we can write by
inspection:
Va_2 = 1/6Vt
Vout_2 = Vt

Adding the contributions:

Vout = 0 + Vt = Vt = Va
and

where we have substituted the constraint on Vt

Va = 1/6Vs + 1/6Vt = 1/6Vs + 1/6 Va

Va = [1/(6 )]Vs

Substitute this Va into the expression for Vout:

Vout =

Vs

Solution 5.40
The first step is to replace the dependent source with a temporary independent source. Then, superposition
can proceed as usual.

Now, lets short the temporary source, Vt. Again, this network is a ladder network, like the one in the
previous problem. However, now, the equivalent resistance looking into each of the vertical branches (from
left to right) is different. Now, it is
Req = 20R||(R+4R) = 20R5R/25R = 4R

Now, again, define the voltages across these three vertical branches (from left to right) as V1 , Va, V2 . It
follows by voltage division that
V1 = 4/9Vs

Va_1 = 4/5 4/9 Vs

Va_1 = 16/45 Vs

It should be noted that Vout = 0.

Now, short the input source and find the response due to Vt. Again, in this case, the circuit is identical to
the case when Vs was active (note the symmetry in the above figure). Therefore,
Va_2 = 16/45 Vt
Vout_2 = Vt

Adding the two contributions:

16
16
Vs +
Va
45
45
1
V
? Va =
(4516 ) s

Va =

Vout = Vt = Va

and

Vout =

16
Vs
45 16

Solution 5.41
First, replace the controlled source by a temporary independent source:

Now, do a source transformation on the Vt source:

The equivalent resistance seen by the current source is 99. Therefore:

Vout = -99/100Vt = -0.99Vt
Vt = Vg

Now,

and

IL = -0.99Vt/990k where IL is the current through R2

VR2 = ILR2 = (100k/990k)(-0.99Vt) = -0.1Vt

By KVL
Vg = VR2 + V s = -0.1Vt + Vs = -0.1Vg + V s
Rearranging
Vg =

1
0.1 + 1

Vs

Finally, and substituting:

Vout
1
= 0.99
Vs
1 + 0.1
Vout
= 8.91
Vs

Solution 5.42
Vout b0 + b1 x b0 + b1R
=
=
Vin
a0 + a1 x 1+ a1R

At R = 0
Vout b0
= = 20
Vin
1
? b0 = 20

At infinite R
Vout b1
= =8
Vin
a1
? b1 = 8a1

Finally, at R = 10
20 + 80a1
= 10
1 + 10a1
? a1 = 0.5
? b1 = 4

Substituting these values for R = 2 yields

Vout 20 + 4 2
=
= 14
Vin 1 + 0.5 2

SOLUTION 5.43. From the chapter

Vout b0 + b1 b0 + b1
=
=
Vin
a0 + a1
1+ a1
Equivalently,
V
V
out a1 + b0 + b1 = out
Vin
Vin
Plugging in the data yields three equations in three unknowns which in matrix form are:
1 0 a1 264
0
154 1 1 b = 154

168 1 2 b1 84
A=[0
A=
0
-154
-168

1 0;-154 1 1;-168 1 2]
1

0
1
1

1
2

y = [264 154 84]'

y=
264
154
84
Coefs = A\y
Coefs =
2.8571e-01
2.6400e+02
-6.6000e+01
Therefore, a1 = 0.28571, b0 = 264, and b1 = 66. Making a1 = 1, will yield a different set of answers.
When m = , Vout /Vin = b1 /a1 = 231.

Thevenin Probs, 7/24/01

- P6.1 -

@R.A. Decarlo & P. M. Lin

PROBLEM SOLUTIONS CHAPTER 6

SOLUTION 6.1. (a) Voc is found by removing RL and doing voltage division.
VOC = 63V

(600 || 700)
600

= 28V
(600 || 700) + 300 600 + 100

RTH is found by setting the source to zero and by calculating the equivalent resistance seen looking back
between the A and B terminal.
RTH = [(300 || 600) + 100] || 600 = 200

(b) Using
P = RL I L

the power for each resistance may be found by substituting the appropriate RL in the following equation.
2

VOC
P =
RL
RTH + RL

For 50 , 200 , and 800 , the power obtained is 627.2 mW, 980.0 mW, and 627.2 mW respectively.
The use of Thevenin equivalent does reduce the effort in obtaining the answer.
SOLUTION 6.2. (a) To find RTH, open circuit the current source and short-circuit the voltage source. The
resulting resistance seen from terminal A-B is 1 k. Using superposition, the contribution of the current
and voltage source at the open circuit output may be summed as 30 V (2 k/4 k) + 10 mA (2 k/4 k) (2 k).
VOC is then 25 V and ISC = VOC/RTH is 25 mA.
(b) Following is a plot of
2

VOC
P =
RL
RTH + RL

for RL from 100 to 4 k .

Thevenin Probs, 7/24/01

- P6.2 -

@R.A. Decarlo & P. M. Lin

SOLUTION 6.3. (a) Turning off the two sources

RTH = (60 + 60)||40 = 30 ,
and using superposition
VOC = 6V

40
60
+ 0.1A
40 = 3 V.

40 + 60 + 60
40 + 60 + 60

ISC is the obtained as 100 mA.

(b) Using
2

VOC
P =
RL
RTH + RL

a load of 90 will absorb 56.25 mW.

(c) It absorbs 75 mW; hence the 30 resistor absorbs more power.
SOLUTION 6.4. As both resistor divider ration are the same (3/6), the voltage at A and B is the same
resulting in a VOC of 0 V.
RTH = (3K ||6 K) + (9K ||18 K) = 8 k
The relation VOC/ISC cannot be used in this situation.
SOLUTION 6.5. Using superposition

20 103
5 103
VOC = 20

10

+ 20 20sin(50t) = 14 + 400sin(50t) V,
20 103 + 5 103
20 103 + 5 103
and

Thevenin Probs, 7/24/01

- P6.3 -

@R.A. Decarlo & P. M. Lin

RTH = (20k ||5k) + 20k = 24 k.

ISC can then be found using the VOC/RTH relationship as 0.58+20sin(50t) mA.
SOLUTION 6.6. Once again using superposition
6K
4K

VOC = 72V
+ 54mA
6 K + 54mA 2 K = 204V ,
6 K + 8K + 4 K
4 K + 8 K + 5K

and
RTH = [( 4 K + 8 K ) || 6 K ] + 4 K + 2 K = 10 K

SOLUTION 6.7. Using source transformation, (a) is obtained from the original circuit. Then combining
in series the resistors and voltage sources, and retransforming them (b) is obtained. Finally adding the two
currents and transforming back the circuit to its Thevenin form (c) is obtained.

From (c),
Voc = 60V
RTH = 6k
I L = 60 /(6k + 6k ) = 5mA
P = I L RL = 0.15W
2

Thevenin Probs, 7/24/01

- P6.4 -

@R.A. Decarlo & P. M. Lin

SOLUTION 6.8. First, each source in series with 2R, can be replaced by an up going current source of
V x / 2 R in parallel with 2R. Then starting from the left, the two 2R in parallel are combined and then

retransformed to a voltage source of Vo / 2 in series with 2R once added to the series resistance. Repeating
the previous steps,
Vo / 2 + 2 R ? Vo / 4 R|| 2 R|| 2 R|| V1 / 2 R
Vo / 4 R + V1 / 2 R|| R ? Vo / 4 + V1 / 2 + R + R
Vo / 4 + V1 / 2 +2 R ? Vo / 8 R + V1 / 4 R|| 2 R|| 2 R|| V2 / 2 R
Vo / 8 R + V1 / 4 R + V2 / 2 R|| R ? Vo / 8 + V1 / 4 + V2 / 2 + R + R
Vo / 8 + V1 / 4 + V2 / 2 +2 R ? Vo / 16 R + V1 / 8 R + V2 / 4 R|| 2 R|| 2 R|| V3 / 2 R
Vo / 16 R + V1 / 8 R + V2 / 4 R + V3 / 2 R|| R ? Vo / 16 + V1 / 8 + V2 / 4 + V3 / 2 + R + R

Thus Voc = Vo / 16 + V1 / 8 + V2 / 4 + V3 / 2 and RTH = 2 R .

SOLUTION 6.9. First we find RTH
RTH = 2K + [(6K ||3K ) + (15K ||10 K )]||24 K = 8K
Use nodal analysis to solve for VOC. At node a
1mA = 1/ 6K(Va Vc ) + 1/ 3K(Va Vd ) + 1/ 24K(Va ) or
(1/ 6K + 1/ 3K + 1/ 24K)Va 1/ 6K(Vc ) 1/ 3K(Vd ) = 1m
doing the same for the supernode, and the equation inside the supernode yields the following two
equations:
(1/ 6K + 1/ 3K)Va (1/ 6K + 1/15 K)Vc (1/ 3K + 1/10 K)Vd = 0, and
Vc Vd = 30
using these three equations, it is now possible to solve for the three unknowns VA, VC, and VD. Using
MATLAB they are respectively 4.5V, 22.8750V, and -7.1250V. VOC being VA, 4.5V.

SOLUTION 6.10.
RTH = [(18K ||9K) + 66K]||36K = 24K

Thevenin Probs, 7/24/01

- P6.5 -

@R.A. Decarlo & P. M. Lin

By superposition, noting that all resistances are in k-ohms,

VOC =

66+(9||18)
(66 + 36)||9
36
36
36 2.5 +

18
30 = 52 V
66+(9||18)+36
[(66 + 36)||9] + 18 36 + 66
36 + (9||18) + 66

When a 2 k is connected, the current IL becomes VOC/(RTH+2k) = 2mA; thus the power absorbed is 8
mW.
SOLUTION 6.11. (a) Introduce a test voltage source at the output, and write the nodal equations in matrix
form:
1/ 50
1/100
(1/100 + 1/ 50)
Vtest itest

1/ 100
1/100
(1/ 100 + 1/ 200 + 1/100) VC = 0

VD Vs / 100
1/ 50
(1/ 50 + 1/100 + 1/100)
1/100
Solving we obtain, Vtest = 100itest + 2Vs / 3. From eq. 6.10 RTH = 100 , voc = 2Vs / 3.

voc
(b) To obtained the power the following equation is used, P =
RL .
RTH + RL

Thevenin Probs, 7/24/01

- P6.6 -

@R.A. Decarlo & P. M. Lin

1.2
1.1
1

Power/Vs in mW

0.9
0.8
0.7
0.6
0.5
0.4
0.3

20

40

60

80
100
120
Load resistance in Ohms

140

160

180

200

SOLUTION 6.12.
VOC = 0, as no independent sources are present. Writing the following nodal equation where vx is the
voltage across both ports,
2.5 +
ix = ((v x iz ) ix ) + (v x i z )/ 2, RTH can be found as v x / ix =
,
1.5
SOLUTION 6.13.
Defining vi and ii as the voltage across and current into the input ports, writing the nodal equation at the
input: ii + gmv x = 1/ 200K(vi vx ) . We can also get the following equation v x = 200K(gmv x + ii ) . Using

200K gm
1
the previous two equations we can solve for RTH = vi / ii = 200K 1+
+
.
1 + 200K gm 1 + 200K gm
This yields a gm of 10 S.
SOLUTION 6.14.
ISC is null as no independent source are present. To find RTH vi and ii are defined as the voltage across and
current into the input ports. Writing the nodal equation we get:
ii = Vx / 1.8K + (1/ 200)(Vx 3 / 4Vx ) , and vi = Vx 300 ii . Solving RTH = vi / ii = 600 .
SOLUTION 6.15.
First, write out the equation around loop 1:

Thevenin Probs, 7/24/01

- P6.7 -

@R.A. Decarlo & P. M. Lin

Vx = i1 1200 i2 100 i3 300 . Then substituting the following relationships,

i1 = Ix , i2 = 2I x , and i3 = 0.01Vx + i2 = 0.01Vx + 2Ix , and solving for Vx / Ix = RTH = 100 . VOC = 0 as no
independent sources are present.

SOLUTION 6.16.
Introduce a test source at the output terminals, and write out the nodal equations in matrix for the top node,
and the supernode comprised of the current controlled voltage source (ccvs). vb is the node left of ccvs, and
vc the node to the right.
1/ 20
1/ 30 vtest itest
1/ 30 + 1/ 20
1/ 20 1/ 30 + 0.4 1/ 20 0.4 2 / 30 v = 0

1/ 6
1
5/ 6 vc 0
Solving, we obtain vtest = 90itest , thus RTH = 90 .

SOLUTION 6.17.
(a) Turn off independent source. Introduce a test source and write loop equation:
v x = 6ix 4i1 + 10ix . Note that i1 = ix . Now solve for v x / ix = RTH = 12 .

Thevenin Probs, 7/24/01

- P6.8 -

@R.A. Decarlo & P. M. Lin

Short the input and write the loop and nodal equations:
3A = i1 + ISC
10i1 = 4i1 + 6ISC
Solving yields ISC = 1.5A, and VOC = ISC RTH = 18V .
2

VOC
(b) In MATLAB the following plot is generated: P =
RL for 1 RL 24.
RTH + RL
7

Power in Watts

10
15
Resistance in Ohms

20

25

Maximum power is absorbed by 12 load.

SOLUTION 6.18.
To find thevenin resistance, introduce a test source and write the following equations:
v s = 20i1 + 40i1 + 40i1 = 100i1 , and is = vs /100 + i1 . Solving for v s / is = RTH = 50 .
Next, use the following nodal equation; 0.2A = i1 + ISC , and loop equation 20i1 + 40i1 = 40ISC . Solving
using these two equations yields ISC = 0.12A , and consequently VOC = RTH ISC = 6V .
SOLUTION 6.19.
(a) Introduce a test source, vs, and get the following two equations: v s 15is = Vx , and
1

kVx + Vx / 3 + Vx / 5 + is = 0. Solving yields v s / is = RTH = 15 +

, or 95/6 for k = 2/3.
k + 8 /15
Next, write the following nodal equation kVx + Vx / 3 + (Vx 1) / 5 = 0 and observe that VOC = 1V Vx .

Thus solving yields VOC

1/ 5

= 1
8 = 5/ 6V for k = 2/3, and ISC = VOC / RTH = 1/19 A .
k +

15

Thevenin Probs, 7/24/01

- P6.9 -

@R.A. Decarlo & P. M. Lin

(b) Solving the previously obtained equation for k when VOC = 0, yields k = -1/3, and consequently
RTH = 20 .
SOLUTION 6.20.
Introduce a test source, vt, and get the following equations: it + ix = ix

Vs 300ix
Vs = 300it + 300ix ,
300

and vt = Vs 300ix + 100it . Solving yields vt / it = RTH = 400 .

Observe how no current flows through the 300 resistor in parallel with the dependent source. Thus VOC
will always be 0 V and is independent of VS .
SOLUTION 6.21. (a) For this part, consider the modified circuit below.

Step 1: Applying KCL to node A, we have

1
V
(VA 100) + A + 0.1 Is = 0
400
800
Multiplying through by 800 yields
3VA = 800Is + 120
Step 2. Computing Vs , we have
2000
Vs = VA 2000i x = VA
(100 VA ) = 6VA 500
400
Hence
800
Vs = 6
Is + 40 500 = 1600Is 260 = RthIs + Voc
3
Therefore Rth = 1.6 k and Voc = 260 V.
(b) By linearity, Voc = 130 V.

Thevenin Probs, 7/24/01

- P6.10 -

@R.A. Decarlo & P. M. Lin

SOLUTION 6.22.
Introduce a test source, vt,, and get the following two equations: vt = (30m V1 / 100)400 V1 = 12 5V1 ,
and it = 0.06V1 + 30m V1 / 100 = 30m + 0.05V1 . Solving obtain vt = 100it + 15. Thus RTH = 100 ,
and ISC = VOC / RTH = 150mA .
SOLUTION 6.23. Insert Itest as per text. Hence
1 + 1 V 16 = I
test
40 120 C 40
Solving, we obtain
VC = 30Itest + 12
By KVL,
60
Vtest = VC + 30 ( 2i1 + Itest ) = VC +
(16 VC ) + 30Itest = 0.5VC + 30Itest + 24
40
Substituting for VC yields
Vtest = 15Itest + 18
Hence, Voc = 18 V and Rth = 15 . Thus i =

18
= 0.2 A. Further, P75 = 75(0.2)2 = 3 W.
75 + 15

SOLUTION 6.24.
(a) and (b)

Thevenin Probs, 7/24/01

- P6.11 -

@R.A. Decarlo & P. M. Lin

+
2

+
vs

(c) Writing KVL around loop 1, loop 2, and finally relating v to v x ,

vs = 10(0.02 I1) + v 20I1
v = 5I2 15(I2 0.02)
3vx = I1 I2 = 3[10(0.02 I1)] = 0.6 30I1
(d) Rewriting these in matrix form,
30 0 1 I1 1
0.2
0 20 1 I = 0 v + 0.3

2 s

31 1 0 v 0
0.6
(e), (f), and (g). Using MATLAB,
A=[-30 0 1;0 20 1; 31 -1 0];
b1 = [1 0 0]';
b2 = [-.2 .3 .6]';
I1 = inv(A)*b1
I1 =
-1.5385e-03
-4.7692e-02
9.5385e-01
I2 = inv(A)*b2
I2 =
1.9231e-02
-3.8462e-03
3.7692e-01
Rth=-1/I1(1)
Rth = 650
Note: Isc = I2(1) = 0.01923 A
Voc= -Rth*I2(1)

Thevenin Probs, 7/24/01

- P6.12 -

@R.A. Decarlo & P. M. Lin

Voc =
1.2500e+01
(h)

SOLUTION 6.25.
(a)
(b) Write four nodal equations,
is = (VA VC ) / 2k
is + 1m = (VC VD )/ 6k + (VC VE )/ 3k
VE / 2k = (VC VD ) / 6k VD /15k
VE / 2k = VE /10k + (VE VC )/ 3k
(c)
1/ 2k 1/ 2k
0
1/ 2k

0
1/ 6k

1/ 3k
0

0
0 VA 1
0

1m
1/ 6k
1/ 3k VC 1
=
i +
(1/ 6k + 1/ 15k) 1/ 2k VD 0 s 0


0
1/ 15k VE 0
0

(d) Solving in MATLAB VA = 5.818kis + 3.8182 , thus

RTH = 5.818k
VOC = 3.8182V
(e) This only changes VOC = 38188m = 30.54V .
SOLUTION 6.26. For this circuit, no current flows though the 20 resistor. Therefore VAB = VCB .
Further, from the examples in the chapter, VCB = 4Vs . Hence, Voc = VAB = VCB = 4Vs . Also, shorting
V
4V
V
V
terminals A and B, yields Isc = CB = s = s . It follows that Rth = oc = 20 . Note that the
20
20
5
Isc
Thevenin equivalent to the left of C-B is a voltage source of value 4Vs . Therefore the Thevenin equivalent
to the left of A-B is 20 in series with 4Vs .
SOLUTION 6.27. (a) From the previous problem VCB = 4Vs . Thus by voltage division,
VOC = VAB = VCB

180
= 3.6Vs . Next find ISC = VCB / 20 = Vs / 5, and then RTH = VOC / ISC = 18 .
180 + 20

Thevenin Probs, 7/24/01

- P6.13 -

@R.A. Decarlo & P. M. Lin

180||162
(b) This changes the voltage division at the output, thus VAB / Vs = 4
= 3.24 .
(180||162) + 20
SOLUTION 6.28. (a) Writing the following two KCL equations,
vtest / 4k = (Vo vtest ) /12 k
Itest = (vtest Vo ) /15k
where Vo is the voltage at the output of the op-amp. Doing the appropriate substitution get
vtest = 15k / 3 Itest , thus
RTH = 5k
VOC = 0V
Since no independent source exist right of A-B
(b) Applying Ohms law Is = 1/ (10k 5k )Vs = 0.2mVs
SOLUTION 6.29. (a) Adding a test source at terminal A-B, and noting that the voltage at the output of the
op-amp is Vo = 5 / 2Vs . Write KCL at terminal A,
vtest / 900 = (Vo v test )/ 100 + itest
vtest = 90itest 2.25Vs
Where one sees by inspection that
RTH = 90
VOC = 2.25Vs
ISC = VOC / RTH = 0.025Vs
(b) Noticing the virtual short to ground provided by the ideal op-amp, RTH = 20k , and VOC = 0V since no
independent sources are present right of the input terminal.
SOLUTION 6.30. The output voltage of the ideal op amp is 2.5Vs1 2Vs2 which drives a voltage divider
circuit. Hence
Voc = VAB = 0.9(2.5Vs1 2Vs2 ) = 2.25Vs1 1.8Vs 2
Further,
Isc =

2.5Vs1 2Vs 2
= 0.0225Vs1 0.018Vs 2
100

Finally,
V
0.9(2.5V 2V )
Rth = oc = (2.5V s1 2V s)2 = 90
Isc
s1
s2
100

Thevenin Probs, 7/24/01

- P6.14 -

@R.A. Decarlo & P. M. Lin

Equivalently if one sets Vs1 and Vs2 to zero, then the ouput terminal of the op amp goes to ground. Hence
Rth = 100 / /900 = 90
SOLUTION 6.31. Define the node at the output of the op-amp as Vo . Note how the circuit left of this
node is a general summing circuit as per text. Thus, Vo = 2Vs2 4Vs1 . Hence we simply replace the op
amp circuit to the left of the 20 resistor by an ideal voltage source of value V0 . Hence

Voc =

80
V
V0 + 2I = 0.8V0 + 2 0 = 0.78V0 = 1.56Vs2 3.12Vs1
80 + 20
100

Alternately, one can introduce a test source at the output terminal and write out a set of equations using
KVL,
Vo = 2Vs2 4Vs1 = 20I 80(I itest )
vtest = 2I + 80(itest I)
Solving yields vtest = 17.6itest + 0.78(2Vs2 4Vs1 ) , and by inspection,
RTH = 17.6
VOC = 0.78(2Vs2 4Vs1 ) = 56Vs2 3.12Vs1

SOLUTION 6.32. Upon inspection when the op-amp is in active range, for inputs from 3V to 3V,
RTH = 1k
VOC = 0V
However when the input exceeds 3V, the output of the op-amp will saturate at 15V, and
Vs = 15V + 6kI s . Therefore from eq. 6.10,
RTH = 6k
VOC = 15V
When the input is less than 3V, Vs = 15 + 6kI s , thus
RTH = 6k
VOC = 15V
SOLUTION 6.33. (a) The op-amp configuration has a gain of 4. So when the input is between 3V and
3V it is operating in its active region, thus

Thevenin Probs, 7/24/01

- P6.15 -

@R.A. Decarlo & P. M. Lin

RTH = 4k
VOC = 0V
When the input is greater than 3V the output saturates at 12V and
Vs = 20kI s 12
RTH = 20k
VOC = 12V
When the input is less than 3V the output saturates at 12V,
Vs = 20kI s + 12
RTH = 20k
VOC = 12V
(b) When the input is in the active range,
RTH = 16k
VOC = 4Vs
When it is greater than 3V,
RTH = 0k
VOC = 12V
And when less than 3V,
RTH = 0k
VOC = 12V
The last two obtained using figure 6.28.
The maximum power is when the output is in saturation, P = (VOC ) / 28k = 6mW .
2

SOLUTION 6.34. From the table the following two equations can be written:
6 = 2RTH + v oc
12 = 8RTH + voc
Putting in matrix form and solving,
1 1 RTH 6
8 1 v = 12

oc
1/ 6 1/ 6 6 RTH
4 / 3 1/ 3 12 = v

oc
Thus RTH = 1 k since current was in mA, and voc = 4 V.
SOLUTION 6.35. (a) From Ohms law, IAB = VAB / RL . Thus 0.2uA and 0.1uA.
(b) Note how using this topology VAB = voc I AB RTH , thus

Thevenin Probs, 7/24/01

- P6.16 -

@R.A. Decarlo & P. M. Lin

1 0.2u voc 0.4

1 0.1u R = 1

TH
2 0.4 voc
1
10M 10M 1 = R

TH
Thus, voc = 1.6V , and RTH = 6M .
SOLUTION 6.36. (a) Although the text describes finding Rth from a measurement or calculation of both
Voc and Isc , measurement of Isc is often impractical. Hence the procedure outlined in this problem
provides a more practical means of determining the Thevenin equivalent.
Since the internal meter resistance is 10 M,
V
V
IAB (A) = AB + AB
10
RL
Hence, the completed table is:
RL(M )

vAB (V)

IAB (A)

0.4

0.24

10

0.2

(b) The terminal relationship assuming a Thevenin equivalent is given by

VAB = Voc RthI AB
In matrix form with the data from the table
1 0.24 Voc 0.4
1 0.2 R = 1

th
Hence Voc = 4 V and Rth = 15 M. Note that since we used A and V, Rth is in M.
SOLUTION 6.37. (a) For this scenario, the circuit is essentially a voltage source with a resistance R in
series with the circuit under test, in parallel with a voltmeter measuring the voltage division between the

Thevenin Probs, 7/24/01

- P6.17 -

@R.A. Decarlo & P. M. Lin

later two. Therefore if replacing R = 0 by R = R2 causes the voltage measured by the voltage meter to
RTH
drop by half, then by voltage division, Eo / 2 = Eo
, and RTH = R2 .
R2 + RTH
(b) Using the same reasoning and voltage division,
RTH
Eo = Vs
RTH + Rs
RTH
Eo / 2 = Vs
RTH + Rs + R2
Therefore RTH + Rs = R2 or equivalently RTH = R2 Rs .
(c) Again by voltage division,
RTH || Rm
Eo = Vs
(RTH || Rm ) + Rs
RTH || Rm
Eo / 2 = Vs
(RTH || Rm ) + Rs + R2
RTH Rm
R R Rs Rm
Therefore (RTH || Rm ) + Rs = R2 or
= R2 Rs . Solving for RTH = 2 m
.
RTH + Rm
Rm + Rs R2
SOLUTION TO 6.38. (a) The voltmeter measures the voltage division between the two resistors involved
thus,
Eo = Voc
Eo / 2 = Voc

R2
RTH + R2

and RTH = R2 .
(b) Now
Rm
Rm + RTH
Rm || R2
Eo / 2 = Voc
(Rm || R2 ) + RTH
Eo = Voc

From the division of the former by the later R2 = RTH || Rm . And from the former
Voc = (Rm + RTH )Eo / Rm = (1+ RTH / Rm )Eo .
SOLUTION TO 6.39. Using the relation developed in problem 6.37 RTH =

SOLUTION TO 6.40. Using the equation developed in problem 6.38,

R2 Rm Rs Rm
= 5k
Rm + Rs R2

Thevenin Probs, 7/24/01

RTH =

- P6.18 -

@R.A. Decarlo & P. M. Lin

R2 Rm
= 4M
Rm R2

Voc = (1+ RTH / Rm )E o = 20V

SOLUTION TO 6.41. Writing the line equation in the following general form,
i = vab / RTH isc
i = 2v ab 4
Thus RTH = 0.5 , and isc = 4A .
SOLUTION TO 6.42. (a) In this range the curve appears to be varying linearly between (0V,0mA) and
(0.2V, 0.1mA) pair, thus writing the line equation i = (0.1m / 0.2)v yields RTH = 2k Voc = 0V .
(b) Writing the line equation of the following form, i = v / RTH isc , into a matrix equation,
0.2
0.7

1 1/ RTH 0.1m
=
1 isc 10.1m

and solving gives

RTH = 50
isc = 3.9mA
voc = 0.195V
(c) First make a guess as to which region of the curve, N will operate in. Assuming that it will operate in
the A-B region, then by KVL i(t) = (vs (t) + vb v oc )/ (R + RTH ) = (50sin(1000t) 0.095) / 550. It can be
seen that this guess is wrong as the range of i(t) is not in the appropriate region. Assuming the 0-A region,
by KVL i(t) = (vs (t) + vb v oc )/ (R + RTH ) = (50sin(1000t)m + 0.1)/ 2500 = 0.02sin(1000t) + 0.04mA.
Note that the highest current is 0.06mA, thus still in the appropriate region of operation.
(d) Repeating the procedure above and guessing the region A-B, by KVL
i(t) = (vs (t) + vb v oc )/ (R + RTH ) = (200sin(1000t)m + 305m) /100 = 2sin(1000t) + 3.05mA. The
maximum and minimum current are 5.05 and 1.05 mA respectively, thus the assumption made was correct.
SOLUTION TO 6.43. (a) By the power transfer theorem RL = RTH . For circuit (a) RL = 80 ||240 = 60,
and circuit (b) RL = (900||180) + 50 = 200.
(b) Finding isc for each circuit: (a) isc = 36 / (80 + 240||60) 240 / 300 = 225mA and (b)

900 ||180

2
isc = 60mA
= 22.5mA. The power is now obtained from P = (isc / 2 ) RL . Thus
(900||180) + 50 + 200
759mW for (a) and 25.3mW for (b).

Thevenin Probs, 7/24/01

- P6.19 -

@R.A. Decarlo & P. M. Lin

SOLUTION TO 6.44. Finding the Thevenin equivalent,

RTH = [(30||15 ) + 10 ] ||80 = 16
Voc = 32V
2

V
The value for the load resistance is 16, and the power is P = oc / RL = 16W .
2
SOLUTION TO 6.45. (a) Find the Thevenin equivalent,
RTH = 12k ||6k + 8k = 12k = RL
Voc = 8k(2mA) + 24

6
= 24V
6 + 12

The power is P = (Voc / 2 ) / RL = 12mW .

2

(b) The maximum power will be transferred to the load, when the value of its load is closest to 12k. Thus
2

Voc
the power is P =
10k = 11.9mW .
RTH + 10k
2

Voc
(c) Same reasoning as (b) the power is P =
15k = 11.9mW .
RTH + 15k
SOLUTION TO 6.46. (a) The Thevenin equivalent to the left of RL has Rth = 12 + 20 / /180 = 30 and
Voc = 1 18 +

180
40 = 54 V. Therefore, for maximum power transfer
180 + 20
GL + 2GL + 3GL = 6GL =

6
1
=
RL 30

Hence RL = 180 .
(b) For this part, let VL denote the voltage (top to bottom) across the load. With RL = 180 , then the
parallel combination equals Rth and hence VL = 27 V. It follows that
(27)2
= 4.05 watt.
180
Since the terminal voltages are the same, the absorbed power is inversely proportional to the resistance.
P180 =

Hence P90 = 8.1 watt, and P60 = 12.15 watt.

Thevenin Probs, 7/24/01

- P6.20 -

@R.A. Decarlo & P. M. Lin

SOLUTION 6.47. To find the Thevenin equivalent introduce a test voltage source at the output, and write
KCL at the two node in the circuit. By inspection the following matrix expression is obtained:
1/ 400
(1/ 200 + 1/ 400)
VA 0.1
=

1/ 400
0.0015 + 1 / 1 k + 1/ 400 Vtest Itest

Solving Vtest = 240Itest + 8. And RTH = 240 , Voc = 8 V. By voltage division, the voltage across the load
resistor is 4V, and the power delivered to it is 66.7mW.

VA

Vtest

SOLUTION 6.48. Using KCL get ix = 1mA, which independent of what is connected to the output. Thus
isc = 10ix = 10mA
RTH = 3k
The power is then P = (10mA / 2) 3k = 75mW .
2

SOLUTION 6.49. Performing a source transformation on (a), and combining the elements will simplify to
one current source going up of 2/3A in parallel with a 10 resistor. This is essentially the Norton
equivalent of the circuit,
isc = 2/ 3A
RTH = 10
For (b), combine the voltage sources and resistor in series. The circuit obtained is one voltage source of
5V in series with a 45 resistor. This is the Thevenin equivalent,
voc = 5V
RTH = 45
(a) The value of the load resistor is simply the thevenin resistance obtained above.
(b) Using Ohms law for (a) VL = isc (10 ||10) = 10/ 3V , and voltage division for (b) VL = v oc (1/ 2) = 2.5V
(c) Using the following formula, P = VL2 / RL , (a) absorbs 1.1W and (b) 139mW, thus (a) absorbs more
power.

Thevenin Probs, 7/24/01

- P6.21 -

@R.A. Decarlo & P. M. Lin

SOLUTION 6.50. (a) Note that the circuit left of the terminal is already in its Thevenin form. The load
RL = R ||( R + 300)
R2 + R(300 2RL ) 300RL = 0
Solving, R = 71.6 . By voltage division, the voltage across the load is 5V. The power absorbed is
2
P = (Voc / 2 ) / RL = 416.7mW .
(b) The following script can be used to plot the power absorbed by the load versus R:
%Script for problem 6.50b
R=0:2:400;

%Calculate Load resistance

RL= 1./((1./R)+1./(R+300));

%Calculate the power

P=(10.*(RL./(RL+60))).^2./RL;

%Plot the power versus R

plot(R,P);
ylabel('Power in Watts');
xlabel('Resistance in Ohms')

SOLUTION 6.51. First, find the Thevenin equivalent by writing out the transfer equation v ab = 200i + 40.
Thus RTH = 200 Voc = 40V . The maximum power will then be P = (Voc / 2) 2 / 200 = 2W .
SOLUTION 6.52. The assumption that all controlling voltages or currents for dependent sources within
Ni are assumed to be in Ni , implies that the nodal equation matrix of figure P6.52a has the partitioned
form:
G11

G21
0

G12
G22
G32

0 VN1 IN1

G23 Vm = Im
G33 VN 2 IN 2

(*)

where VN1 is the vector of UNKNOWN and INDEPENDENT node voltages internal to N1 and VN2 is
the vector of UNKNOWN and INDEPENDENT node voltages internal to N2 . The right side of the

Thevenin Probs, 7/24/01

- P6.22 -

@R.A. Decarlo & P. M. Lin

equation consists of (effective) currents injected into the appropriate node. However, IN1 depends only on
sources in N1 and IN2 depends only on sources in N2 .
At this point we must presume that the matrix equation (*) has a unique solution, i.e., the determinant
of the coefficient matrix is non-zero. Hence we can calculate VN1, Vm, and VN2 uniquely. As such, by
considering the first row of (*), we can assert that VN1 satisfies
G11VN1 = I N1 G12 Vm

(**)

Note that we are not claiming that we can solve for VN1 from this equation.
Replacing N2 by a voltage source of value Vm results in the network of figure P6.52b. For this
network, the nodal equations are
G11VN1 = I N1 G12 Vm

(***)

where Gij is the same as in (*). Clearly, this is the same as equation (**). Again we presume there is a
unique solution to this equation, i.e., the determinant of G11 is non-zero. If so, we can solve for VN1
uniquely and the result is the same as that obtained by solving (*).
This theorem can be extended to RLCM networks (to be studied in later chapters) or even
nonlinear networks under appropriate conditions.
To emphasize the subtlety of this result and the need for unique solvability in each network,
consider the following circuit.

The resulting nodal equations are:

Thevenin Probs, 7/24/01

- P6.23 -

@R.A. Decarlo & P. M. Lin

1
0
0 V1 2
2
1 0.5 0.5 0 V 0

2 =
0 0.5 2.5 1 V3 0

0
1
2 V4 4
0
There exists a unique solution and from MATLAB, we find
Vm = V3 = 2 V
To apply the voltage source substitution, we replace N2 by a voltage source of 2 V and obtain the
following circuit.

The nodal equations here are

2 1 V1 2
1 0.5 V = 1

2
Observe that the coefficient matrix has a zero determinant. Thus there is either no solution or many
solutions, i.e., no unique solution. This demonstrates that unique solvability of the larger network does not
imply the unique solvability of the smaller derived network.

SOLUTION 6.53. The assumption that all controlling voltages or currents for dependent sources within
Ni are assumed to be in Ni , implies that the loop equation matrix of figure P6.53a has the partitioned form:
R11

R21
0

R12
R22
R32

0 I N1 EN1

R23 Im = Em (*)
R33 I N 2 EN 2

Thevenin Probs, 7/24/01

- P6.24 -

@R.A. Decarlo & P. M. Lin

where IN1 is the vector of unknown and independent loop currents internal to N1 and IN2 is the vector of
unknown and independent loop currents internal to N2 and Im a single independent loop current common
to N1 and N2 . The right side of the equation represents the net contribution of voltage sources present in
the appropriate loop. However, EN1 depends only on N1 and EN2 depends only on N2 .
At this point we must presume that the matrix equation (*) has a unique solution, i.e., the determinant
of the coefficient matrix is non-zero. Hence we can calculate IN1, Im, and IN2 uniquely. As such, by
considering the first row of (*), we can assert that IN1 satisfies
R11IN1 = EN1 R12 Im

(**)

Note that we are not claiming that we can solve for IN1 from this equation.
Replacing N2 by a current source of value Im results in the network of figure P6.53b. For this
network, the loop equations are
R11IN1 = EN1 R12 Im

(***)

where Rij is the same as in (*). Clearly, this is the same as equation (**). Again we presume there is a
unique solution to this equation, i.e., the determinant of R11 is non-zero. If so, we can solve for IN1
uniquely and the result is the same as that obtained by solving (*). For some subtlety in the proof refer to
the solution of 6.52.
This theorem can be extended to RLCM networks (to be studied in later chapters) or even
nonlinear networks under appropriate conditions.
SOLUTION 6.54. (a) The thevenin equivalent to the left of terminal A-B is
RTH = [(30||60) + 20]||10 = 8 , and using Ohms law along with voltage division
15V 30
VOC = 15 10
= 14V
15 + 60 30 + 30
(b) Doing the same for the circuit right of terminal A-B. RTH = [(30||15) + 10]||20 = 10 and
7.5V 30
VOC = 7.5 20
= 5V .
15 + 15 30 + 30
10
8
(c) Using superposition, VAB = 14 + 5 = 10V .
18
18
(d) (e) (vCB 15) / 30 + vCB / 60 = (10 vCB )/ 20 Hence, vCB = 10V .

Thevenin Probs, 7/24/01

- P6.25 -

@R.A. Decarlo & P. M. Lin

SOLUTION TO P6.55. The proof is based on superposition. Let us consider the figure below where N1
and N2 are differently named but identical networks.

We first compute the contribution to ia from the independent sources in N1 with those of N2 deactivated.
Let this current be i1a . The contribution to ia from the independent sources in N2 with those of N1
deactivated is ia2 . But because N1 and N2 are identical, i1a = ia2. Hence by superposition
ia = ia1 + ia2 = 0. By the current source substitution theorem we can replace the lines by current sources of
value 0-amp. This defines an open circuit and the connecting line can be replaced by an open circuit.
From the given network we also note by KVL that Vx + Vy = 0 which implies that Vx = Vy . On the other
hand, since the networks are identical, Vx = Vy . Thus we conclude that Vx = Vy = 0. Thus we can replace
Vx and Vy by a voltage source of 0-volt (voltage source substitution theorem) which is the definition of a
short circuit.
SOLUTION 6.56. Label the potential between each line starting from the top as Vx1 , Vx2 on the left, and Vy1
and Vy2 on the right. Now by superposition and linearity notice that
Vx1 = Vy1
Vx2 = Vy2
because the independent source is negative on the right side. Additionally, from KVL,
Vx1 = Vy1
Vx2 = Vy2
The only way all these condition can be met, is if all the voltages are 0 V, or short circuited.

Thevenin Probs, 7/24/01

- P6.26 -

@R.A. Decarlo & P. M. Lin

SOLUTION 6.57. (a) Using the results from P6.55, no current flows between the two halves. So the right
hand side circuit may be analyzed as if it was stand alone. By voltage division then,
3+2
v a =
18 = 15V .
3 + 2 + 1
(b) From the results of P6.56, all the lines crossing the symmetry line are shorted together. Consequently,
6 ||3
by voltage division, v a =
(18) = 12V .
(6 ||3) + 1
SOLUTION 6.58. Note how this circuit is the same as in P6.57: it is just redrawn with the neighboring
resistors added in parallel or in series. Using superposition, we can solve for va when the sources [vs1 vs2 ]
are [18 18], and then [18 18]. By linearity adding the two contribution will be equivalent to solving for
[36 0] directly, since adding the source contributions [18 18]+[18 18] =[36 0]. The contributions of
15V12V were obtained in P5.57; thus v a = 3V .
SOLUTION 6.59. Yes since

[ 45

27] = [36 36] + [ 9 9] = 2[18 18] + 1/ 2[18 18] = 2(15) + (12) / 2 = 24V .

SOLUTION 6.60. For this proof we attach an arbitrary network N to each of the networks N1 and N2 in
figure P6.60 as shown below.

N may have internal independent sources, but we consider N1 and N2 external excitations to N and
we assume no violation of KVL in the attachment. Choose node 3 as a reference node. Then
V13 = Va Vc and V23 = Vb Vc for both figures. Hence N1 and N2 provide the identical external

Thevenin Probs, 7/24/01

- P6.27 -

@R.A. Decarlo & P. M. Lin

excitations to N and hence all currents and voltages in N remain unchanged.

The extension of this result to 4 external nodes is shown in the figure below. The verification is
the same as above. The extension of course to n-terminals is clear.

SOLUTION 6.61. Using the E-shift theorem, remove the 9 V source from each branch and add it to the
4V source, and notice by inspection that VOC = 4 + 9 = 5 V.
SOLUTION 6.62. (a) Writing a KCL equation for each node in N2:
I1 = Ia
I2 = Ia Ia = 0
I3 = I a I a = 0
I4 = Ia
Do the same for N1:
I1 = Ia
I2 = 0
I3 = 0
I4 = Ia
This shows that the two have identical outcomes.
(b)

Thevenin Probs, 7/24/01

- P6.28 -

@R.A. Decarlo & P. M. Lin

SOLUTION 6.63. Using the I-shift theorem, this circuit is essentially a 2A source in parallel with a
series combination of resistors, and a 5A source. Thus isc = 5 2 = 3A .
SOLUTION 6.64. (a) This can be done by inspection. An equal source is connected between A-C and CB; thus by the I-shift theorem, it is equivalent to the same source just connected between A-B.
(b) In figure 6.64c the VCCS is replaced by a resistor using the Ohms law relationship
100 = V1 / (0.01V1 ).
100||900
(c) (d) By voltage division V1 = Vs
, and by Ohms law
(100 ||900) + 10
100||900
Vout = Vs
0.01( 20k ||5k ) = 36Vs
(100 ||900) + 10
Vout / Vs = 36

L&C Probs, 11/15/01

P:7-1

R. A. DeCarlo, P. M. Lin

PROBLEM SOLUTIONS CHAPTER 7

SOLUTION 7.1. Given the coil has 48 turns and 12 turns/cm, we know that the length of the coil is 4 cm.
Since the length of the coil is greater than 0.4 times its diameter, the formula given in the question can be
used:
L=

4 105 (0.02)2 ( 48)2

= 18.81 H
18 (0.02) + 40 (0.04)

SOLUTION 7.2. Part 1. Applying (7.1)

The voltage vL(t) can be computed using the inductor v-i relationship:
v L (t) = L

diL (t)
dt

The calculations for vL(t) for t = 0s to 5s are summarized in the following table:
Time Interval
0s< t 1s
1s< t 3s
3s< t 4s
4s< t 5s

d/dt (iin (t))

2 As-1
-2 As-1
2 As-1
0 As-1

vL(t)
1V
-1V
1V
0V

Below is the plot of VL vs time.

Part 2. Applying (7.4)

wL (t) =
In the time interval 0s < t 1s, iin (t) = 2t . Thus
wL (t) =

1
(0.5 iin2 (t))
2

1
0.5iin2 (t) = t 2
2

In the time interval 1s < t 3s, iin (t) = 4 2t . Thus

L&C Probs, 11/15/01

P:7-2

wL (t) =

R. A. DeCarlo, P. M. Lin

1
1
0.5iin2 (t) = (4 2t) 2 = t 2 4t + 4
2
4

In a similar way, wL (t) can be computed in the remaining intervals.

The calculations for wL(t) for t = 0s to 5s are summarized in the following table:
Time Interval
0s< t 1s
1s< t 3s
3s< t 4s
4s< t 5s

iin (t)
2t
4-2t
-8+2t
0

wL(t)
t2
2
t -4t+4
t2 -8t+16
0

Below is the plot of wL vs t.

SOLUTION 7.3. Applying (7.2)

1 t
v ( )d
L 0 L
It is assumed that iL(0) = 0 A. Using the preceding formula and the fact that vin (t) = t in the time interval
0s < t 2s, the current is
iL (t) = iL (0) +

iL (t) =

1 t
d = t 2 in the time interval 0s < t 2s

0.5 0

In the time interval 2s < t 3s vin (t) = 2 . Hence

iL (t) =

1 t
1 t
v L ( )d = iL (2) +
2dt = 4 + 4 t2 = 4t 4

0.5 0
0.5 2

In a similar way iL (t) can be computed in the remaining intervals.

The calculations for iL(t) for t = 0s to 6s are summarized in the following table:

L&C Probs, 11/15/01

P:7-3

Time Interval
0s< t 2s
2s< t 3s
3s< t 5s
5s< t 6s

R. A. DeCarlo, P. M. Lin

vin (t), V
t
2
-2
0

iL(t), A
t2
4t-4
20-4t
0

Below is the plot of iL vs t.

SOLUTION 7.4. Part 1. Using (7.1), we have

d
vin (t) = 0.2 10 3 [is (t)] = 0.2 103 1000cos(1000t) = 0.2cos(1000t)mV
dt
For the 2mH inductor
t

1
1
iout (t) = iout (0) + 10v in ( )d =
3 2cos(1000 )d = sin(1000t)mA
L
2

10
0
0
Below is a sketch of iout vs t.

Part 2. Instantaneous power delivered by the dependent source is given by

pL (t) = v L (t) iL (t) = 10vin (t) iout (t) = 2cos(1000t) sin(1000t) = sin(2000t) W

L&C Probs, 11/15/01

P:7-4

R. A. DeCarlo, P. M. Lin

Part 3. The energy stored in the 2mH inductor is given by

1
1
WL (t) = LiL 2 (t) = Liout 2 (t) = sin2 (1000t)nJ
2
2
Below is a sketch of WL vs t

SOLUTION 7.5. Part 1 For the excitation in Figure P7.5b,

t

1
i1 (t) = i1(0) +
v ( )d ,
0.5 in
0

1
i2 (t) = i2 (0) +
v ( )d = 2i1(t)
0.25 in
0

It is assumed that i1 (0) = i 2 (0) = 0 A . In the interval 0s < t 1s, vin (t) = 10V . Hence, in this interval
i1 (t) =
i2 (t) =

1 t
1 t
v in ( )d =
(10)d = 20t

0.5 0
0.5 0

1 t
1 t
v in ( )d =
(10)d = 40t

0.25 0
0.25 0

In the interval 1s < t 2s, vin (t) = 5V . Hence, in this interval,

i1 (t) = i1(1) +

i2 (t) = i2 (1) +

1 t
1 t
v in( )d = 20 +
(5)d = 10t 10

0.5 1
0.5 1

1 t
1 t
v in( )d = 40 +
(5)d = 20t 20

0.25 1
0.25 1

In a similar fashion i1 (t) and i2 (t) can be computed in other intervals.

The calculations for i1 (t) and i2 (t) for t = 0s to 5s are summarized in the following table:

L&C Probs, 11/15/01

P:7-5

Time Interval
0s< t 1s
1s< t 2s
2s< t 3s
3s< t 4s
4s< t 5s

vin (t), V
-10
-5
0
5
10

R. A. DeCarlo, P. M. Lin

i1 (t), A
-20t
-10-10t
-30
-60+10t
100+20t

i2 (t), A
-40t
-20-20t
-60
-120+20t
200+40t

Below are the plots of i1 vs t and i2 vs t.

Part 2 For the excitation in Figure P7.5c,

t

1
i1 (t) = i1(0) +
v ( )d ,
0.5 in
0

1
i2 (t) = i2 (0) +
v ( )d = 2i1(t)
0.25 in
0

It is assumed that i1 (0) = i2 (0) = 0 A. In the interval 0s < t 1s, vin (t) = 10t . Hence, in this interval
i1 (t) = i1(0) +

1 t
1 t
v in( )d =
(10 )d = 10t 2

0.5 0
0.5 0

and
i2 (t) = i2 (0) +

1 t
1 t
v in( )d =
(10 )d = 20t 2

0.25 0
0.25 0

In the interval 1s < t 3s, vin (t) = 10t 20 . Hence, in this interval
i1 (t) = i1(1) +
i2 (t) = i2 (1) +

1 t
1 t
v in( )d = 10 +
(10 20)d = 10t 2 40t + 40

0.5 1
0.5 1

1 t
1 t
v in( )d = 20 +
(10 20)d = 20t 2 80t + 80

0.25 1
0.25 1

The calculations for i1 (t) and i2 (t) for t = 0s to 3s are summarized in the following table:

L&C Probs, 11/15/01

P:7-6

Time Interval
0s< t 1s
1s< t 3s

R. A. DeCarlo, P. M. Lin

vin (t), V
10t
10t-20

i1 (t), A
10t2
10t2 -40t+40

i2 (t), A
20t2
20t2 -80t+80

Below are the plots of i1 vs t and i2 vs t.

SOLUTION 7.6. For the circuit in Figure P7.6, the parallel combination of the 0.75mH and 1.5mH
inductors can be replaced by an inductor with the inductance of (0.75||1.5)mH. The v-i relationship for
this inductor is
1

1
iin (t) = iin (0) +
v
(
)d
=

s
v s ( )d A
0.5 10 3 0
[(0.75) (1.5)] 103 0
t

The series combination of the 0.8mH and 0.6mH can be replaced by an (0.8mH + 0.6mH) inductor. The
v-i relationship for this inductor gives:
t

0.8 + 0.6
d
3 d
vout (t) = (0.8 + 0.6) 10
[6i (t)] = 0.5 6 dt v s ( )d = 16.8v s (t) V
dt in
0

Below is a plot of vout(t) vs t.

L&C Probs, 11/15/01

P:7-7

Assume iL(0) =0 for all inductors.

R. A. DeCarlo, P. M. Lin

6
iout (t) = 6iin (t) =
vs ( )d = 12 10 3 vs ( )d A

3
0.5 10 0
0
In the interval 0s < t 1s, v s (t) = 1mV . Hence,
iin (t) =

t
1
v ( )d = 2t(A), iout (t) = 6iin (t) = 12t(V )

3
0 s
0.5 10

The power p(t) on the same interval is computed as:

p(t) = 6iin (t) vout (t) = (12t) (16.8 10 3 ) = 0.202t(W )
In the interval 1s < t 3s, v s (t) = 2mV . Hence,
iin (t) = iin (1) +

and

t
1
v s ( )d = 2 + 2 2 1t = 4t 2(A)

3
1
0.5 10

iout (t) = 6iin (t) = 24t 12(A)

p(t) = 6iin (t) vout (t) = (24t 12) (16.8 2 103) = 0.806t 0.403(W )

In a similar fashion, iout(t) and p(t) can be computed for the remaining intervals.
The calculations for iout(t) and the instantaneous power delivered by the dependent source, p(t), for t = 0s
to 6s are summarized in the following table:
Time Interval
0s< t 1s

vs(t), V
110-3

vout(t), V
16.810-3

iout(t), A
12t

p(t), W
0.202t

1s< t 3s

210-3

33.610-3

-12+24t

0.806t-0.403

3s< t 5s
5s< t 6s

-210-3
0

-33.610-3
0

132-24t

0.806t-4.435

12

L&C Probs, 11/15/01

P:7-8

R. A. DeCarlo, P. M. Lin

SOLUTION 7.7. For 0 t < 2 s, using the inductor v-i relationship, we have
t

1
2
8
iin (t) =
v in ( )d =
[ cos(0.25 )]t0 = [1 cos(0.25t)] A

0.5
0.25

The associated energy stored as a function of t for this time interval is

W0.5 (t) =

16
[1 cos(0.25t)]2 J
2

The energy for the second inductor remains zero over this interval.
For 2 s t, we have
t

1
iin (t) = i0.5 (t) + i0.25 (t) = i0.5 (t) +
v ( )d
0.25 in
2

8
4
8
16
[1 cos(0.25t)] +
[ cos(0.25 )]t2 = [1 cos(0.25t)] cos(0.25t)

0.25

8 24
=
cos(0.25t) A

Here the current
8
i0.5 (t) = [1 cos(0.25t)] A

in which case the energy stored over the interval [2,t] is

W0.5 (2,t) =

1
1 2
64
2
2
0.5 i0.5
(t) i0.5
(2) = 0.5 i0.5
(t) 2 J
2
2

Further
i0.25 (t) =

16
cos(0.25t) A

in which case the energy stored over the interval [2,t] is

W0.25 (2,t) =

1
1
2
2
0.25 i0.25
(t) i0.25
(2) = 0.25i 20.25(t) J
2
2

SOLUTION 7.8. Let the 5mH inductor be L1 and the 20mH inductor be L2 .
For 0 t < 3ms,
t

1
1
iL1(t) = iL1 (0) + v s ( )d =
3 12cos (500 )d = 4.8sin(500t) mA
L1
5

10
0
0

L&C Probs, 11/15/01

P:7-9

For t 3ms,

R. A. DeCarlo, P. M. Lin

1
1
iL 2 (t) = i L2 (0) +
v s ( )d =

3 12cos (500 )d = 1.2sin(500t) mA

L2
20

10
0
0

iL1(t) = 4.8sin(500t) mA as L1 is still subjected to the same voltage.

on the other hand,
iL 2 (t) = i L2 (3 ms) +

1
L2

3 ms

1
v L2 ( )d = iL 2 (3 ms) +
L

0d

3 ms

= iL 2 (3 ms)
= 1.2sin(500 3 103) = 1.197mA
For 0 t < 3ms, the energies stored in the inductors are given as follows:
1
2
L1[iL1 (t)] = 57.6sin 2 (500t) nJ,
2
1
2
WL2 (t) = L2 [iL2 (t)] = 14.4sin 2 (500t) nJ
2
WL1 (t) =

For t 3ms, the energies stored in the inductors are given as follows:
1
2
L1[iL1 (t)] = 57.6sin 2 (500t) nJ,
2
1
2
WL2 (t) = L2 [iL2 (t)] = 14.328 nJ
2
WL1 (t) =

SOLUTION 7.9. Given the dielectric parameters and the dimensions of the capacitor, the capacitance of the
paper capacitor is given by
A
0.04 0.8
C = r 0 = 3 8.854 1012
= 8.5nF
d
104
SOLUTION 7.10. . Part 1. Applying the capacitor v-i relationship:
iC (t) = C

d
(v (t)) = 1 100 1000 ( sin(1000t)) = 0.1sin(1000t) A
dt C

Part 2. Applying the capacitor v-i relationship:

L&C Probs, 11/15/01

P:7-10

R. A. DeCarlo, P. M. Lin

d
(v (t))
dt C
d
10 10 3 cos(1000t) = C (sin(1000t))
dt
iC (t) = C

10 10 3 cos(1000t) = 1000C cos(1000t)

Therefore, C=10F.

SOLUTION 7.11. Applying the capacitor v-i relationship for C1 and C2 :

iC1(t) = (2mF)

d
(v (t)),
dt in

iC 2 (t) = (6mF)

d
(v (t)) = 3iC1(t),
dt in

In the interval 1s < t 0s, vin (t) = 5t + 5. Hence, in this time interval
iC1(t) = 2 103

d(v in (t))
= 102 A
dt

and
iC 2 (t) = 6 10 3

d(vin (t))
= 3 10 2 A
dt

Using KCL, iin (t) can be computed as

iin (t) = iC1 (t) + iC 2 (t) = 4 10 2 A
In the interval 0s < t 1s, vin (t) = 5. Thus
iC1(t) = iC2 (t) = 0 A
Using KCL, iin (t) = iC1 (t) + iC 2 (t) = 0 A .
In a similar fashion iC1 (t), iC2 (t) and iin (t) can be computed for the remaining intervals.

The calculations for iC1 (t), iC2 (t) and iin (t) for t = -1s to 6s are summarized in the following table:
Time Interval
-1s< t 0s
0s< t 1s
1s< t 3s
3s< t 4s
4s< t 5s
5s< t 6s

d/dt (vin (t)), Vs-1

5
0
-5
20
0
-15

Below are the plots of iC1 (t), iC2 (t) and iin (t).

iC1 (t), mA
10
0
-10
40
0
-30

iC2 (t), mA
30
0
-30
120
0
-90

iin (t), mA
40
0
-40
160
0
-120

L&C Probs, 11/15/01

P:7-11

R. A. DeCarlo, P. M. Lin

SOLUTION 7.12. Applying (7.6)

2

1
1
vC (2) = vC (0) + iC ( )d = 4V + d = 5V
C
C
0

vC (3) = vC (2) +

1
1
iC ( )d = vC (2) + 2d = 5V + 1V = 6V

C
C

Applying (7.11), the energies stored in the capacitor over the intervals [0,2] and [2,3] are given by

WC[0,2] =

1
C vC2 (2) v C2 (0) = 9 J
2

WC[2,3] =

1
C vC2 (3) v C2 (2) = 11 J
2

SOLUTION 7.13. Part 1 Applying (7.6)

L&C Probs, 11/15/01

P:7-12

R. A. DeCarlo, P. M. Lin

1
1
vC1 (t) = vC1(0) +
iin ( )d =

C1
0.25
0

vC 2 (t) = v C2 (0) +

1
1
iin ( )d =

C2
0.1
0

iin (

)d

iin (

)d

0
t

In the interval 0s < t 1s, iin (t) = 2t . Hence, in this interval

vC1 (t) =

1
0.25

iin ( )d =

t
1
( 2 ) = 4 106 t 2 (V )
0.25
0

1
0.1

iin ( )d =

t
1
( 2 ) = 10 10 6 t 2 (V )
0.1
0

and
vC 2 (t) =

In the interval 1s < t 3s, iin (t) = 2 103 (A). Thus

vC1 (t) = vC1(1) +

t
t
1
1
iin ( )d = 4 +
(2 103 ) = 8 10 3 t 4(V )
1
0.25
0.25
1

vC 2 (t) = v C2 (1) +

t
t
1
1
iin ( )d =10 +
(2 103 ) = 20 10 3 t 10(V )
1
0.1
0.1
1

and

In a similar fashion, vC1 (t) and vC2 (t) can be computed for the remaining intervals.
The calculations for vC1 (t) and vC2 (t) for t = 0s to 8ms are summarized in the following table:
Time Interval
0s< t 1ms

iin (t), A
2t

vC1 (t), V
4106 t2

vC2 (t), V
10106 t2

1ms< t 3ms

210-3

-4 + 8103 t

-10 + 20103 t

3ms< t 5ms
5ms< t 8ms

-210-3
0

44 - 8103 t
4

110 - 20103 t
10

Below are the plots of vC1 (t) and vC2 (t).

L&C Probs, 11/15/01

P:7-13

R. A. DeCarlo, P. M. Lin

Part 2 Applying (7.12),

WC1 (t) =

1
C v 2 (t) ,
2 1 C1

WC 2 (t) =

1
C v 2 (t)
2 2 C2

The expressions for WC1 (t) and WC2 (t) for t = 0ms to 8ms are listed in the following table:
Time Interval

W C1 (t), J

W C2 (t), J

0s< t 1ms

0.125 (4 10 t )

0.05 (10106 t2 )2

1ms< t 3ms

0.125 (-4 + 8103 t)2

0.05 (-10 + 20103 t)2

3ms< t 5ms

0.125 (44 8103 t)2

0.05 (110 20103 t)2

5ms< t 8ms

0.125 (4)2

0.05 (10)2

6 2 2

Below are the plots of WC1 (t) and WC2 (t).

Part 3 Since the current iC1 (t) stays constant at 0A after t = 5ms,
vC1 () = v C1 (5) = 4V
vC 2 () = vC 2 (5) = 10V
SOLUTION 7.14. Part 1 Applying (7.6),

L&C Probs, 11/15/01

P:7-14

R. A. DeCarlo, P. M. Lin

1
1
vin (t) = v in (0) + iin ( )d =
3 iin ( )d
C
0.5

10
0
0
In this part, we use the current excitation signal described in Figure P7.14b. In the interval 0s < t 1s,
iin (t) = 10(mA). Thus, in this interval
vin (t) =

t
1
(10 103)d = 20t(V )

3
o
0.5 10

In the interval 1s < t 2s, iin (t) = 5(mA). Hence, in this interval
vin (t) = v in (1) +

t
1
(5 103)d = 20 + (10t + 10) = 10t 10(V )

3
1
0.5 10

In a similar fashion, vin (t) can be computed for the remaining intervals. The calculations for vin (t) for t = 0s
to 5s are summarized in the following table:
Time Interval
0s< t 1s
1s< t 2s
2s< t 3s
3s< t 4s
4s< t 5s

iin (t), mA
-10
-5
0
5
10

vin (t), V
-20t
-10 -10t
-30
-60+10t
-100+20t

Below is the plot of vin (t) vs. time.

Part 2. In this part, we use the current excitation signal described in Figure P7.14c. In the interval 0s < t
1s, iin (t) = 10(mA) . Thus, in this interval
vin (t) =

t
1
(10 10 3 )d = 10t 2 (V)

3
o
0.5 10

In the interval 1s < t 3s, iin (t) = 10t 20(mA) . Hence, in this interval
vin (t) = v in (1) +

t 3
1
10 (10 20)d = 10 + 10(10 2 40 ) 1t = 10t 2 40t + 40 (V)

3
o
0.5 10

L&C Probs, 11/15/01

P:7-15

R. A. DeCarlo, P. M. Lin

The calculations for vin (t) for t = 0s to 3s are summarized in the following table:
Time Interval
0s< t 1s
1s< t 3s

iin (t), mA
10t
10t-20

vin (t), V
10t2
10t2 - 40t + 40

Below is the plot of vin (t) vs. time.

SOLUTION 7.15. Part 1 Using the capacitor v-i relationship,

iin (t) = C1

d
(v (t)) = 20 6 1500 cos(1500t) = 0.18cos(1500t) A
dt s

Then, we can find vout by applying (7.6),

t

vout (t) = vout (0) +

1
2i ( )d
0.5m in
0

= 10 +

2 0.18
sin(1500t)
0.5m 1500

= 10 + 0.48sin(1500t)

Part 2 The instantaneous power delivered by the independent source is given by

p(t) = 2iin (t) vout (t) = 0.36cos(1500t) [10+ 0.48sin(1500t)]

Part 3 Applying (7.11), the energy stored in the capacitor over the interval [0,t] is given by
WC1[0,t] =

1
C1 vs 2 (t) v s2 (0) = 360sin2 (1500t) J
2

L&C Probs, 11/15/01

P:7-16

R. A. DeCarlo, P. M. Lin

SOLUTION 7.16. Part 1 From the solution of problem 7.15,

t

1
vout (t) = vout (0) +
2i ( )d
0.5m in
0
t

1
d
= vout (0) +
2 20
(v s ( ))d

0.5m
d
0

= vout (0) + 0.08

d
(v s ( ))d
d

= vout (0) + 0.08v s( )

If we assume vout(0) = 0V, then
vout (t) = 0.08 v s ( )
The following is a plot of vout(t) vs time.

Part 2 The instantaneous power delivered by the dependent source is given by

p(t) = v out (t) 2iin (t) = ( vout (0) + 0.08 v s(t)) 2 20
= 40( vout (0) + 0.08 vs (t))
If we assume vout(0) = 0V, then
p(t) = 3.2 v s (t)

d
(v (t))
dt s
d
(v (t))
dt s

d
(v (t))
dt s

The calculations for d/dt (vS(t)) and p(t) for t = 0s to 6ms are summarized in the following table:

L&C Probs, 11/15/01

P:7-17

R. A. DeCarlo, P. M. Lin

Time Interval
0ms< t 1ms

d/dt (vS(t)), Vs-1

3
2 10

1ms< t 3ms

-2 10

12.8 (t-2)

3ms< t 4ms
4ms< t 6ms

12.8 (t-4)
0

p(t), W
12.8 t

2 10
0

The following is a plot of p(t) vs time.

Part 3 The energy stored in the 0.5-mF capacitor is given by

WC (t) =

1
0.5m v out 2 (t) = 0.25( vout (0) + 0.08 v s (t)) 2
2

If we assume vout(0)=0V, then

WC (t) = 1.6(v s (t)) 2

In the interval 0s < t 1ms, v s (t) = 2 103 t(V ). Thus, in this interval
WC (t) = 6.4t 2J
In the interval 1s < t 3ms, v s (t) = 2 103 (2 t)(V ). Hence, in this interval
WC (t) = 6.4 (2 t) 2J
In a similar way WC(t) can be computed for the interval 3ms < t 4ms.
The following is a plot of W C(t) vs time.

mJ

L&C Probs, 11/15/01

P:7-18

R. A. DeCarlo, P. M. Lin

SOLUTION 7.17. Part 1 From the solution of problem 7.16,

vout (t) = v out (0) + 0.08 v s ( )
If we assume vout(0)=0V, then
vout (t) = 0.08 v s ( )
The following is a plot of vout(t) vs time.

Part 2 The instantaneous power delivered by the dependent source is given by.
d
p(t) = 40( v out (0) + 0.08 v s (t)) (v s (t))
W
dt
If we assume vout(0)=0V, then
d
p(t) = 3.2 v s (t)
(v (t)) W
dt s
The calculations for d/dt (vS(t)) and p(t) for t = 0s to 6ms are summarized in the following table:

L&C Probs, 11/15/01

P:7-19

Time Interval
0ms< t 1ms
1ms< t 2ms
2ms< t 4ms
4ms< t 5ms
5ms< t 6ms

R. A. DeCarlo, P. M. Lin

d/dt (vS(t)), Vs-1

2 103
0
-2 103
0
2 103

p(t), W
12.8 t
0
12.8 (t-3)
0
12.8 (t-6)

The following is a plot of p(t) vs time.

Part 3 The energy stored in the 0.5-mF capacitor is given by

WC (t) = 0.25( v out (0) + 0.08 v s (t))2
If we assume vout(0) = 0V, then

WC (t) = 1.6(v s (t)) 2

In the interval 0ms < t 1ms, v s (t) = 2 103 t(V ). In this interval
WC (t) = 6.4t 2J
In the interval 1ms < t 2ms, v s (t) = 2(V ). In this interval
WC (t) = 6.4( J )
In a similar fashion WC(t) can be computed for the remaining intervals.
The following is a plot of W C(t) vs time.

mJ

L&C Probs, 11/15/01

P:7-20

R. A. DeCarlo, P. M. Lin

SOLUTION 7.18. Part 1 Applying the capacitor v-i relationship for the equivalent capacitor of the series
combination of 0.3 mF and 0.6 mF capacitors
vin (t) = v in (0) +

= (0) +

is (

0.3 103 0.6 103 0

t

1
0.3 10 3

)d

0.6 10 3

60 10 3 sin(100 )d
0

= 3 3cos(100t) V

Therefore,
iout (t) = (0.2 103 + 0.8 103)

d
(10v in (t)) = 3sin(100t)
dt

Below is a sketch of iout(t) vs time.

Part 2 The instantaneous power delivered by the dependent source is given by

p(t) = iout (t) 10v in (t) = 90sin(100t) [1 cos(100t)] W

L&C Probs, 11/15/01

P:7-21

R. A. DeCarlo, P. M. Lin

Part 3 The instantaneous energy stored in the 0.8mF capacitor is given by

WC (t) =

1
1
2
CvC 2 (t) = (0.8mF)[10vin (t)] = 0.361
[ cos(100t)]2 J
2
2

SOLUTION 7.19.
Part 1
Part 2
Part 3
Part 4

Since Q = CV, the charge that resides on each plate of the capacitor = 10F 100V = 1mC
Since V = Q/C, the required voltage = 1mC/5F = 200V
Since V = Q/C, the required voltage = 50C/1F = 50V
The energy required = 0.5 10F (100V)2 = 0.05J

SOLUTION 7.20. When 0s t <2s, vC(t) = 25V. Conservation of charge requires that
q1(2 ) + q2 (2 ) = q1 (2+ ) + q2 (2 + ). Since q2 (2 ) = 0 C it follows that

25 V 150 mF = vC(2+)(150 mF + 100 mF)

Hence
vC (2+ ) =

25V 150mF + 0V 100mF

= 15 V
150mF + 100mF

Thus, for t > 2s, vC(t) = 15 V.

SOLUTION 7.21. For this solution consider the figure below in which C1 and C2 are labeled.

There are two cases to consider: (i) t < 2 ms and (ii) t > 2 ms.
Case 1. t < 2 ms. Here, since the current source is zero for t < 0, and C1 is uncharged at t = 0,
1
vC (t) =
C1
=

1
1
is ( )d = vC (0) +
is ( )d =
i ( )d

C1
C1 s

12
500 t
e
= 1.6 1 e 500t V
3
0
15 10 500

L&C Probs, 11/15/01

P:7-22

R. A. DeCarlo, P. M. Lin

Note that vC (2 ms) = 1.0114 V. Hence the energy stored over [0, 2 ms] is

WC1 (0 < t < 2ms) = 0.0192 1 e 500t

and

WC2 (0 < t < 2ms) = 0

Case 2. 2 ms < t. At t = 2 ms the switch closes, forcing a discontinuity in the capacitor voltages. To
+
calculate the capacitor voltages at 2 ms, we use conservation of charge. Here, the relevant equation is:
qC1 (2+ ms) + qC 2(2 + ms) = qC1(2 ms) + qC2 (2 ms) +

2+ ms

is ( )d

2 ms

Note that since vC 2 (2 ms) = 0, qC 2 (2 ms) = C2vC 2 (2 ms) = 0 and the integral of the bounded
continuous function is (t) over an infinitesimal interval is zero, this equation reduces to
qC1 (2+ ms) + qC 2(2 + ms) = qC1(2 ms)
or equivalently, since for t > 2 ms, vC1 (t) = vC 2 (t) = vC (t) ,
C1vC (2 + ms) + C2vC (2+ ms) = C1vC (2 ms)

Therefore

vC (2+ ms) =

C1
15
vC (2 ms) =
1.0114 = 0.37927 V
C1 + C2
15 + 25

and it follows that

1
vC (t) = vC (2 ms) +
Ceq
+

= 0.379 +

2ms

1
is ( ) d = 0.379 +
C1 + C2

is ( )d

2ms

12
500 t
e
= 0.379 + 0.6 0.36788 e 500t V
3
0.002
40 10 500

Hence the energy stored in the two capacitors over the interval [2+ ms, t] is
WCeq (2+ ms < t) =

1
1
Ceq vC2 (t) Ceqv C2 (2 + ms) J
2
2

whereas the instantaneous stored energy, i.e., the energy stored over (-,t > 2 ms] is given by
WCeq (t) =
-

1
Ceq vC2 (t)
2

What happens between 2 ms and 2 ms is beyond the scope of the material in this chapter. Please
refer to problem 51 in chapter 8 for an explanation.

L&C Probs, 11/15/01

SOLUTION 7.22.

P:7-23

R. A. DeCarlo, P. M. Lin

Part 1 Applying the inductor v-i relationship,

v L (t) = L

d
i (t)
dt in

= 2.5mH 200te10t + 20e 10t

= 0.05e 10t (10t + 1)

Applying the capacitor v-i relationship, we have:
t

1
vC (t) = vC (0) + iin ( )d
C
0

20
te10
1m
0

]0
t

= 20 103 0.1 e 10 0.01e 10

= 2 103 te10t 0.1e 10t + 0.1 V

and
vin (t) = v L (t) + vC (t)

= 0.05e 10t (10t + 1)+ 2000 te10t 0.1e 10t + 0.1

= 2000.5 te10t 199.95 e 10t + 200(V)
The sketches of vL(t), vC(t) and vin (t) are shown below.

L&C Probs, 11/15/01

P:7-24

R. A. DeCarlo, P. M. Lin

Part 2 The energy stored in the inductor is given by

WL (t) =

1
L[i (t)]2 = 0.5t 2e 20t
2 in

The sketch of WL(t) vs. time is shown below.

Part 3 The energy stored in the capacitor is given by
WC (t) =

2
1
2
C[vC (t)] = 2 te10t = 0.1e 10t + 0.1 kJ
2

The sketch of WC(t) vs time is shown below.

L&C Probs, 11/15/01

P:7-25

R. A. DeCarlo, P. M. Lin

SOLUTION 7.23. Applying the inductor v-i relationship we have that

v L (t) = 0.25

d[is (t)]
= 4cos(4 t) (V )
dt

Applying the capacitor v-i relationship it follows that

iC (t) = 0.25

d[2v 2 (t)]
= 0.25 2 4 4 [ sin(4t)] = 8sin(4 t) A
dt

SOLUTION 7.24.
Part 1. By KCL is (t) = iC1(t) + iC 2 (t). Applying the v-i relationship for capacitor C1 and C2 we have:
iS (t) = (10mF + 20mF)

d
v (t)
dt s

= 30 103 (12te 5t 30t 2e 5t )

= 0.36te 5t 0.9t 2e 5t

Part 2. Applying the v-i relationship for the 20mH inductor, it follows that
vout (t) = (20mH )

d
[12is (t)]
dt

= 0.24 4.5t 2e 5t 3.6te5t + 0.36e 5t

= 1.08t 2e 5t 0.864te5t + 0.0864e 5t (V )

Part 3 The energy stored in the 20mH inductor for t > 0 is given by
WL (t) =

1
(20mH )[12iS (t)]2
2

= 1.44 0.36te5t 0.9t 2e 5t

= 1.44 t 2e 10t (0.36 0.9t)2 J

SOLUTION 7.25. We denote by vC(t) the voltage across the capacitors C1 and C2 and by iL(t) the current
through the inductors L1 and L2 . The equivalent capacitance of the parallel combination of C1 and C2 is
(C1 + C2 ) and thus:
iS (t) = (C1 + C2 )

dvC (t)
dt

Using the v-i relationship for the capacitor C2 it follows that

iC 2 (t) = C2
By replacing

dvC (t)
, it follows that
dt

dvC (t)
dt

L&C Probs, 11/15/01

P:7-26

iC 2 (t) =

R. A. DeCarlo, P. M. Lin

C2
2
i S (t) = is (t)
C1 + C2
3

The equivalent inductance of the series combination of L1 and L2 is (L1 + L 2 ) and thus:
9iC 2(t) = (L1 + L2 )

diL (t)
dt

Using the v-I relationship for the inductor L2 it follows that

vout (t) = v L2 (t) = L2
By replacing

di2 (t)
dt

diL (t)
, it follows that
dt
vout (t) =

L2
2
9iC 2 (t) = 9iC 2(t) = 6iC 2 (t) = 4is (t)
L1 + L2
3

SOLUTION 7.26. Using the v-i relationship for the capacitor we can write:
1
vC 2 (t) =
C2
1
vC1 (t) =
C1

iC ( )d

iC ( )d

where iC(t) is the current through the capacitors C1 and C2 . Since C1 and C2 are connected in series:
1
1
vin (t) = v C1(t) + vC 2 (t) = +
C1 C2
t

It follows that

iC ( )d

iC (

)d

CC
= v in (t) 1 2 . Hence
C1 + C2

1
vC 2 (t) =
C2

iC ( )d =

1
CC
C1
1 2 v in (t) =
v (t)
C2 C1 + C2
C1 + C2 in

Since L1 and L2 are combined in parallel it follows that

1
1
vout (t) = +
L1 L2
=

d(AvC 2(t))
dt

L1L2
dv (t)
LL
C1 dvin (t)
A C2 = 1 2 A
L1 + L2
dt
L1 + L2
C1 + C2 dt

L&C Probs, 11/15/01

P:7-27

where the last equality follows by replacing vC2 (t) with

R. A. DeCarlo, P. M. Lin

C1
v (t)
C1 + C2 in

SOLUTION 7.27. Observe first that the 0.3H and 0.6H parallel inductances combine to make a 0.2H
inductance. Also the 0.4H and 1.2H parallel inductances combine to make a 0.3H inductance. Finally, the
0.2H and 0.3H inductances are combined in series and the equivalent inductance is 0.5H. Shortly, all the
above steps can be written as: Leq = (0.3H || 0.6H) + (0.4H || 1.2H) = 0.5H.
SOLUTION 7.28. Observe first that the 1mH and 5mH inductors combine to make a 6mH inductance.
This inductance combines in parallel with the 3mH inductance to make a 2mH inductance. The next step
is to combine in series the 2mH inductance with the 10mH inductance. The equivalent inductance is
12mH. This inductance is combined in parallel with the 36mH inductance and the result is 9mH. Finally,
the 9mH inductance is combined in series with the 4mH inductance and the result is 13mH. Shortly, all
the above steps cam be written as: Leq = [(5mH + 1mH) || 3mH + 10mH] || 36mH + 4mH = 13mH.
SOLUTION 7.29. Step 1. The parallel combination of the 0.6mH and 1.2mH inductors is equivalent to a
0.4mH inductor.
Step 2. The series combination of the 2.4mH and 0.4mH inductors is equivalent to a 2.8mH inductor.
Step 3. The parallel combination of the 2.8mH and 7mH inductors is equivalent to a 2mH inductor.
SOLUTION 7.30. The three inductors can be arranged in the seven fashions as shown below.

(a)

(b)

(c)

(d)

(e)

(f)
Laeq
Lbeq
Lceq
Ldeq
Leeq
f
Leq
Lgeq

= 1mH
= 1mH + 1mH = 2mH
= 1mH || 1mH = 0.5mH
= 1mH + 1mH + 1mH = 3mH
= 1mH + (1mH || 1mH) = 1.5mH
= 1mH || (1mH + 1mH) = 0.667mH
= 1mH || 1mH || 1mH) = 0.333mH

(g)

L&C Probs, 11/15/01

P:7-28

R. A. DeCarlo, P. M. Lin

SOLUTION 7.31. For Fig. P7.31a, Leq = 2 + 15 + 10 + 10 + 40 +30 + 20 + 8 = 135mH

The circuit in Fig. P7.31b is equivalent to the following circuit.

Therefore, Leq = 2 + 15 + 10 + 8 = 35mH

The circuit in Fig. P7.31c is equivalent to the following circuit.

Therefore, Leq = 2 + [(15+10) || (10+40) || (20+30)] + 8 = 22.5mH

SOLUTION 7.32. When the switch is open, the circuit in Fig. P7.32 can be rearranged as the following.

Therefore, Leq = 8L || 8L = 4L
When the switch is closed, the circuit in Fig. P7.32 can be rearranged as the following.

Therefore, Leq = 4L || 4L + 4L || 4L = 4L

L&C Probs, 11/15/01

P:7-29

R. A. DeCarlo, P. M. Lin

SOLUTION 7.33. L eq1 =Leq2 Without going into a detailed analysis, we present the following intuitive
argument. Note that the points a and b represent points on a balanced bridge circuit meaning that the
voltage between a and b would be zero. Therefore, no current will flow through the additional inductance
L. Therefore the presence of L does not affect the equivalent inductance value.

SOLUTION 7.34. L eq1 >Leq2 Without going into a detailed analysis, we present the following intuitive
argument. Note that the points a and b represent points on an unbalanced bridge circuit, meaning that the
voltage between a and b would not be zero. Also note that when two inductors are placed in parallel, the
equivalent inductance becomes smaller than either inductance. The addition of the inductor L in circuit 2
essentially creates an internal parallel inductance resulting in an Leq2 lower than Leq1 .

SOLUTION 7.35. First we add an iin label to the circuit as shown below.

From KVL and the derivative definition of the capacitor

di (t)
di (t)
di (t)
vin (t) = v L1(t) + vL 2 (t) = L1 in + L2 in = ( L1 + L2 ) in
dt
dt
dt
Equivalently,
diin (t)
1
=
v (t)
dt
( L1 + L2 ) in
It follows that
v L k (t) = Lk

diin (t)
Lk
=
v (t)
dt
L1 + L2 in

which is the required voltage division formula.

SOLUTION 7.36. Leq = (11mH) || (19.25mH) + 3mH = 10mH. Applying the voltage division formula,
v L1 (t) = vin (t)
and

3mH
= 60tet mV
Leq

v L2 (t) = v in (t) vL1(t) = 140tet mV

L&C Probs, 11/15/01

P:7-30

R. A. DeCarlo, P. M. Lin

SOLUTION 7.37. First consider the circuit below which contains the additional label of vin (t).

From KCL and the integral definition of the inductor,

t

1
1
iin (t) = iL1 (t) + iL 2 (t) = v in ( )d +
v ( )d
L1
L2 in
Equivalently

1
vin ( )d = 1 1 iin (t)
+
0
L1 L2
t

Therefore
1

L1L2
L

Lk
1
iL k (t) =
v in ( )d = 1 k 1 iin (t) =
iin (t) for k = 1,2

Lk
+

L1 + L2
0
L1 L2

SOLUTION 7.38. In the circuit illustrated in Fig. 7.38,

Leq = (12mH + 27mH) || (130mH) = 30mH
Applying the current division formula,
2
12 + 27
iin (t) = 0.231e t mA
12 + 27 + 130
2
130
iL 2 (t) =
iin (t) = 0.769e t mA
12 + 27 + 130

iL1(t) =

Also,
vin (t) = Leq

2
d
iin (t) = 60tet
dt

The instantaneous energy stored in the 130-mH is given by

WL1 (t) =

2
1
2
L1[iL1 (t)] = 0.8e 2t nJ
2

L&C Probs, 11/15/01

P:7-31

R. A. DeCarlo, P. M. Lin

SOLUTION 7.39. Using the inductor v-i relationship,

vin (t) = Leq

d
d
iin (t) = 30mH iin (t)
dt
dt

The calculations for vin (t) for t = -2s to 7s are summarized in the following table:
Time Interval
-2s< t <=0s
0s< t <=2s
2s< t <=3s
3s< t <=4s
4s< t <=6s
6s< t <=7s

d/dt (iin (t)), mAs-1

100
-200
0
200
100
-200

From the solutions to problem 7.38,

iL1(t) = 0.231 iin (t)
iL 2 (t) = 0.769 iin (t)
Below are the plots for vin (t), iL1(t) and iL2(t).

vin (t), mV
3
-6
0
6
3
-6

L&C Probs, 11/15/01

P:7-32

R. A. DeCarlo, P. M. Lin

SOLUTION 7.40. Part 1 Using the current division formula,

iL 2 (t) =

18
i (t) = 90cos(300 t) mA
18 + 6 in

Part 2
Leq =1.5mH + 6mH || 18mH = 6mH
vin (t) = Leq

d
i (t) = -0.679sin(300 t) V
dt in

Part 3 Instantaneous power delivered by the source is given by

p(t) = v in (t) i in(t) = [-0.679sin(300 t)] [120cos(300 t)] mW
= -40.7sin(600 t) mW
Below is a plot of p(t)

SOLUTION 7.41. Part 1 Leq = 0.3 || 0.9 + 0.4 || 0.4 = 0.425 H

For computing iin (t) we need to apply
1
iin (t) = iin (t0 ) +
Leq

v in (

)d

t0

to each interval [0, 1], [1, 2], , [n, n+1],

The initial condition for the interval [n, n+1] for n even is:
n

1
iin (n) = iin (0) +
v ( )d
Leq in
0

We assume that iin (0) = 0 and it follows that iin (n) = 0 for n even, since

L&C Probs, 11/15/01

P:7-33

R. A. DeCarlo, P. M. Lin

vin (

)d = 0

for n even. The initial condition for the interval [n, n+1] for n odd is:
1
iin (t) = iin (n 1) +
Leq

n1

vin ( )d =

1
(16 ) nn1 = 37.6 A
0.425

For the interval [n, n+1] with n even:

t

1
1
iin (t) = iin (n) +
v in ( )d = 0 +
16 tn = 37.6( t n) A

Leq
0.425
n

For the interval [n, n+1] with n odd:

t

1
1
iin (t) = iin (n) +
v in ( )d = 37.6 +
(16 ) tn = 37.6( n + 1 t) A

Leq
0.425
n

For the interval [n, n+1] with n even:

t

1
1
iin (t) = iin (n) +
v in ( )d = 0 +
(16 ) tn = 37.6(t n) A

Leq
0.425
n

Hence,

37.6(t 2n), 2n < t 2n + 1

iin (t) =
37.6(2n + 2 t), 2n + 1< t 2n + 2

where n is a non-negative integer. Below is a sketch for the input current vs. time.

Part 2 The total energy stored in the set of four inductors is given by

L&C Probs, 11/15/01

P:7-34

1
2
Leq [iin (t)]
2
300(t 2n)2 J
=
300(2n + 2 t) 2 J

R. A. DeCarlo, P. M. Lin

W (t) =

for 2n < t 2n + 1
for 2n + 1< t 2n + 2

where n is a non-negative integer.

Below is a sketch of W(t) vs. t.

Part 3 By the current division formula, the current through the 0.9H inductor is given by
0.3
i (t)
0.3 + 0.9 in
9.4(t 2n) A
=
9.4(2n + 2 t) A

iL 0.9H (t) =

where n is a non-negative integer.

Below is a sketch of the iL0.9H vs. time.

for 2n < t 2n + 1
for 2n + 1< t 2n + 2

L&C Probs, 11/15/01

P:7-35

R. A. DeCarlo, P. M. Lin

SOLUTION 7.42.. a)
Step 1. The parallel combination of the 1F and 2F capacitances is equivalent to a 3F capacitance.
Step 2. The series capacitors 1.5F and 3F combine to make a 1F capacitance.
Step 3. Finally, the 2F capacitance is combined in parallel with the 1F capacitance that was the result of
Step 2 to make a 3F capacitance.
Shortly, the above steps can be written as:
Ceq = [(1F + 2F) || 1.5F] + 2F] = 3F
b) Proceeding in a similar fashion as in part a:
Ceq = 30mF + [9mF || (9mF + 18mF) || 5.4mF] = 33mF
SOLUTION 7.43. a)
Step 1. Combine the parallel capacitances of 1F and 2F to make a 3F capacitance.
Step 2. Combine the series capacitances of 1.5F and 3F to make 1F capacitance.
Step 3. Combine the parallel capacitances of 2.5F and 1F (the result of Step 2) to make a 3.5F
capacitance.
Shortly, the above steps can be written in a condensed form as:
Ceq = 2.5F + [1.5F || (1F + 2F) = 3.5F
b) Proceeding in a similar fashion as in part a:
Ceq = (1mF || 2 mF) + (1.2mF || 2mF || 1.5mF) + (4mF || 2.6667mF || 3.2mF || 1.6mF) = 2.1667mF

SOLUTION 7.44.
Part 1 Ceq = ((( 30 || 60 + 40 ) || 30 + 40 ) || 30 + 40 ) || 30 + 40 = 60mF
Part 2 The value of vin (t) is given by
t

1
1
vin (t) = v in (0) + iin ( )d = 0 +
200sin(20 )d = 166.7 166.7cos(20t) mV
C
0.06

SOLUTION 7.45. When the switch is open, the circuit in figure P7.45 can be represented by the
following.

L&C Probs, 11/15/01

P:7-36

R. A. DeCarlo, P. M. Lin

Therefore, Ceq = 6.4C

When the switch is closed, the circuit in figure P7.45 can be represented by the following.

Therefore, Ceq = ( 4 + 4 ) || ( 16+ 16 ) = 6.4C

SOLUTION 7.46. C eq1 <Ceq2 Without going into a detailed analysis, we present the following intuitive
argument. Note that the points a and b represent points on an unbalanced bridge circuit, meaning that the
voltage between a and b would not be zero. Also note that when two capacitors are placed in parallel, the
equivalent capacitance becomes bigger than either capacitance. The addition of the capacitor C in circuit 2
essentially creates an internal parallel capacitance resulting in a Ceq2 higher than Ceq1 .

SOLUTION 7.47. The three capacitors can be arranged in the seven fashions as shown below.

For configuration (a), Ceq = 1 || 1 || 1 = 0.3333F

For configuration (b), Ceq = 1 || (1 + 1) = 0.6667F
For configuration (c), Ceq = 1 || 1 + 1 = 1.5F
For configuration (d), Ceq = 1 + 1 + 1 = 3F
For configuration (e), Ceq = 0 + 1 + 1 = 2F
For configuration (f), Ceq = 1 || 1 + 0 = 0.5F
For configuration (g), Ceq = 1 + 0 + 0 = 1F

L&C Probs, 11/15/01

P:7-37

R. A. DeCarlo, P. M. Lin

SOLUTION 7.48. As Q = CV for any capacitor, C=Q/V. The equivalent capacitance of the network is
given by
1C 1C 1C
Ceq =
||
||
= 0.1 F
2V 3V 5V

SOLUTION 7.49. First consider the circuit below which contains the additional label of iin (t).

From KVL and the integral definition of the capacitor,

t

1
1
vin (t) = v C1(t) + vC 2 (t) =
iin ( )d +
i ( )d

C1
C2 in
Equivalently,
t

iin (
0

1
)d = 1
1 v in (t)
+
C1 C2

Therefore,
1
Ck

C1C2
1
Ck
vCk (t) =
iin ( )d = 1
v
(t)
=
v in (t)

in
1
Ck
C
+
C
1
2
+
0
C1 C2
t

Hence,
C2
v (t)
C1 + C2 in
C1
vC 2 (t) =
v (t)
C1 + C2 in
which is the required voltage division formula.
vC1 (t) =

SOLUTION 7.50. By the voltage division formula,

vC 3 (t) =

6 ||6
vin (t) = 21(1 e 20t ) V
6|| 6 + 9

SOLUTION 7.51. Part 1 By the voltage division formula,

L&C Probs, 11/15/01

P:7-38

R. A. DeCarlo, P. M. Lin

0.05 + 0.15
v (t) = 6.667sin(120 t) V
0.05 + 0.15 + 0.1 in
vC 2 (t) = v in (t) vC1(t) = 3.333sin(120 t) V
vC1 (t) =

Part 2 Let the 0.1F capacitor be C1 and the 0.15F capacitor be C2 . The energies stored in the 0.1F
and the 0.15F capacitors are given by
WC1[0,t] =

1
C v 2 (t) vC12 (0) = 2.222sin 2 (120 t) J
2 1 C1

1
WC2 [0,t] = C2 vC 22 (t) v C 22 (0) = 0.8333sin 2 (120 t) J
2
SOLUTION 7.52. Part 1 As both the terminals of 0.08F are tied to the voltage source vin (t),
vC1 (t) = vin (t) = 100e 2t V
By the voltage division formula,
vC 2 (t) =

0.06
v in (t) = 66.67e 2t V
0.03 + 0.06

Part 2 Let the 0.03-F capacitor be C2 . The energy stored in the 0.03-F capacitor over the interval [0,t] is
given by

1
WC2 [0,t] = C2 vC 22 (t) v C 22 (0) = 66.67 e 4t 1 J
2

SOLUTION 7.53. First consider the circuit below which contains the additional label of vin (t).

From KCL and the derivative definition of the capacitor

iin (t) = iC1 (t) + iC 2 (t) = C1

dv in (t)
dv (t)
dv (t)
+ C2 in = (C1 + C2 ) in
dt
dt
dt

Equivalently,
dvin (t)
1
=
i (t)
dt
(C1 + C2 ) in
It follows that
iCk (t) = Ck
which is the required current division formula.

dv in (t)
Ck
=
i (t)
dt
C1 + C2 in

L&C Probs, 11/15/01

P:7-39

R. A. DeCarlo, P. M. Lin

SOLUTION 7.54. Part 1 By the current division formula,

iC1(t) =

0.08
i (t) = 80e 2t A
0.08 + 0.03 0.06 in

iC 2 (t) =

0.03 0.06
i (t) = 20e 2t A
0.08 + 0.03 0.06 in

Part 2 Let the 0.03-F capacitor be C2.

1
vC 2 (t) = v C2 (0) +
iC 2 ( )d = 333.3(1 e 2t ) V

C2
0

The energy stored in the 0.03-F capacitor over the interval [0,t] is given by

1
WC2 (0,t) = C2 vC 22 (t) v C 22 (0) = 1666.5(1 e 2t ) 2 J
2
SOLUTION 7.55. Part 1 By the current division formula,
2
iC 2 (t) =
i (t) = 20sin(250t) mA
1+2 s
Part 2 By the voltage division formula,
vout (t) =

2
9ic2 (t) = 0.12sin(250t) V
1+ 2

Part 3 The current in the 2mH inductor is given by

t

1
iL 2 (t) = i L2 (0) +
v ( )d = 0.24[cos(250t) 1] A
L2 out
0
The energy stored in the 2-mH inductor is given by
1
WL2 (t) = L2iL 22 (t) = 57.6[cos(250t) 1]2
J
2

SOLUTION 7.56. Part 1 The dual network N* is shown below.

Part 2 Since the equivalent conductance seen by Is* is 500S, the equivalent resistance seen by Is* is equal
to 0.002.

L&C Probs, 11/15/01

P:7-40

R. A. DeCarlo, P. M. Lin

SOLUTION 7.57. Part 1 The dual network N* is shown below.

Part 2 S*eq = 350S I*out = 8A.

SOLUTION 7.58. First we draw the graph, given below, of the circuit in figure P7.58. We construct the
graph associated with dual network graph for N* from the graph of N. The dual network graph is given
by the dashed lines.

This circuit and its dual have the branch characteristics given in the following table:
ORIGINAL NETWORK
V1 = 40 V
V2 = 500 I2
I3 = 2 I2
V4 = 400I4
V5 = 200 I5

DUAL NETWORK
*
I1 = 40 A
*
*
I2 = 500 V2
*
*
V3 = 2 V2
*
*
I4 = 400V4
*
*
I5 = 200 V5

Finally replace the branches by the elements given in the table above. This produces the dual network
below.

SOLUTION 7.59. The dual for the circuit in figure P7.59 is shown below.

L&C Probs, 11/15/01

P:7-41

R. A. DeCarlo, P. M. Lin

SOLUTION 7.60. First we draw the graph, given below, of the circuit in figure P7.60.

This circuit and its dual have the branch characteristics given in the following table:
ORIGINAL NETWORK
V1 = 5 I1

DUAL NETWORK
I1* = 5V1*

I2 = 8 A

V2* = 8 V

I3 = 6 A

V3* = 6 V

V4 = 4 V

I4* = 4 A

V5 = 2 I5

I5* = 2V5*

I6 = 7 A

V6* = 7 V

Now we construct the graph associated with dual network graph for N* from the graph of N. The dual
network graph is given by the dashed lines.

L&C Probs, 11/15/01

P:7-42

R. A. DeCarlo, P. M. Lin

The graph of the dual network then is pulled out and flipped vertically to produce the graph topology of the
dual network, N*.

Finally replace the branches by the elements given in the table above. This produces the dual network
below.

SOLUTION 7.61. Part 1 We first redraw the circuit to eliminate branch crossing.
schematic is shown below.

The resultant

L&C Probs, 11/15/01

P:7-43

R. A. DeCarlo, P. M. Lin

Then we draw the graph, given below, of the above circuit. We construct the graph associated with dual
network graph for N* from the graph of N. The dual network graph is given by the dashed lines.

The resultant dual network given by the above graph is shown below.

Part 2 The equivalent resistance Req * seen by the current source in the dual circuit is equal to 1/Req = 1/
.

SOLUTION 7.62. Part 1 We first redraw the circuit to eliminate branch crossing.
schematic is shown below.

The resultant

L&C Probs, 11/15/01

P:7-44

R. A. DeCarlo, P. M. Lin

Then we draw the graph, given below, of the above circuit. We construct the graph associated with dual
network graph for N* from the graph of N. The dual network graph is given by the dashed lines.

The resultant dual network given by the above graph is shown below.

Part 2 If we label the conductances in the dual network by their corresponding resistance values, we have
the circuit below.

L&C Probs, 11/15/01

P:7-45

R. A. DeCarlo, P. M. Lin

By comparing to the original network, we conclude that Req * = Req .

Part 3 Since the general relationship between Req and Req * is given by Req = 1/Req *. We can write
Req = 1/Req * =1/R eq
Solving the above equation, we know that Req = 1 .
Part 4 To retain the special properties of parts (b) and (c), we require one resistance value must be a
reciprocal of the other one.

SOLUTION 7.63. The dual network is shown below.

SOLUTION 7.64. The dual network is shown below.

From the answer of problem 7.55,

iC2 (t) = 20 sin (250t) mA, and vout(t) = 0.12 sin (250t) V
*
Therefore, vC2 *(t) = 20 sin (250t) mV and iout
(t) = 0.12 sin (250t) A

The energy stored in the 2mF capacitor is given by

2 1
1
2
*
WC2
(t) = C2* v *out (t) = L2[iL 2 (t)] = 57.6[cos(250t) 1]2 J
2
2

SOLUTION 7.65. The dual network is shown below.

L&C Probs, 11/15/01

P:7-46

R. A. DeCarlo, P. M. Lin

The equivalent inductance of the dual circuit (Leq *) = (1+2) || 1.5 + 2.5 = 3.5H.

SOLUTION 7.66. By voltage division

1
0.4

0.2
1
v1(t) = 1
1 v s (t) = 0.2 + 0.4 v s (t) = 3 vs (t)
+
0.4 0.2
Note that the 0.4 F resistor in parallel with the dependent voltage source has no effect on i1 (t) and hence is
redundant. By the definition of a capacitor,
i1 (t) = 0.2

dv1(t) 1 dv s(t)
=
dt
15
dt

Finally, since the last capacitor is initially uncharged,

t

1
2 dv ( )
2
vout (t) =
i1 ( )d = s d = v s (t)

0.5
15
d
15
A sketch is given below.

SOLUTION 7.67. Following the method outlined in page 269 of the text, we require
1
vC (t0 )
C

1/200

i 0( )d 14

With i0 () = 2A, vC(0) = 20 V, it follows that

1
1
2A
s
C
200
C 1.667mF
Therefore, the standard capacitor value of 1.8mF should be chosen for this application.
20 14

L&C Probs, 11/15/01

P:7-47

R. A. DeCarlo, P. M. Lin

SOLUTION 7.68. From the relationships given in Problem 7.67, we have

1
d
d
t
v L (t) = [ L(t)i L (t)] = I 0 [ L(t)] = I 0L1 4 2
dt
dt
T1 T1
Below is a plot of vL(t) vs t.

From the plot, we notice that when t=T1 , the value of vL(t) changes from positive to negative. This means
that vL(t) will change sign when a car is inside the loop. Therefore, we can make a circuit to monitor the
voltage vL(t) and whenever a negative voltage is detected, the left turn signal should be initiated during the
next traffic light change.

1st Order Circuit Probs 11/26/01

P8-1

R. A. DeCarlo, P. M. Lin

SOLUTIONS PROBLEMS CHAPTER 8

SOLUTION 8.1. (a) By KCL, C

dvC (t)
v (t)
dv (t) v (t)
= C
or C + C = 0 . Using 8.12
dt
R
dt
RC

vC (t) = vC (0)e t / = 10e t V where

= RC = 1s . Plotting this from 0 to 5 sec

t = 0:.05:5;
vc = 10*exp(-t);
plot(t,vc)
grid
xlabel('time in s')
ylabel('vc(t) in V')
10

vc(t) in V

TextEnd

time in s

(b) The solution has the same general form as for (a),iC (t) = iC (0+)e t / =
vC (t)
.
R

10
e t = 2e t mA =
3
5 10

1st Order Circuit Probs 11/26/01

P8-2

R. A. DeCarlo, P. M. Lin

ic(t) in mA

-0.5

-1
TextEnd
-1.5

-2

time in s

(c) By linearity, if vC(0) is cut in half, all resulting responses are cut in half. If vC(0) is doubled, then all
resulting responses are doubled. Alternately, one view this as a simple change of the initial condition
with the same conclusion reached from linearity.

SOLUTION 8.2. (a) From inspection of the general form, 8.12, 0.1/ = 1= 0.1/ RC C = 0.1/R = 5
F.
(b) Since = RC = 0.1, vC (t) = 10e 10t V.
SOLUTION 8.3. (a) The general solution form is vC (t) = vC (0)e t / = v C (0)e

t /Req C

. Using the

v (0.001) 18.394
v (0)e (0.001)/
given data, take the following ratio, C
=
= 2.7183 = C
= e 0.001/ . Hence,
(0.002)/
vC (0.002) 6.7668
v C (0)e
K = 18.394/6.7668
K =2.7183
tau = 1e-3/log(K)
tau =0.0010
C = 5e-6;
Req = tau/C
Req =200.0008
% Req = R*4e3/(R+4e3)

1st Order Circuit Probs 11/26/01

P8-3

R. A. DeCarlo, P. M. Lin

R = Req*4e3/(4e3-Req)
R =210.5272
vC0 = 6.7668/exp(-0.002/tau)
vC0= 49.9999 V
(b)
% vC (t) = 50e 1000t V
t = 0:tau/100:5*tau;
vc = vC0*exp(-t/tau);
plot(t,vc)
grid
xlabel('Time in ms')
ylabel('vC(t) in V')

SOLUTION 8.4. After one time constant the stored voltage, 8 V, decays to 8/e = 2.943 V. From the
graph, the time at which the output voltage is 2.94 V is approximately 0.19 s. Thus

= 0.19 s, and R =

/C = 190 .
SOLUTION 8.5. (a) The general form of the inductor current is iL (t) = iL (0)e t / = 0.15e t / A where
= L / R = 2 10 3 s. Plotting iL (t) = 0.15e 500t A from 0 to 10 msec yields:
t = 0:.01e-3:10e-3;
iL = 0.15*exp(-t/2e-3);
plot(t,iL)
grid

1st Order Circuit Probs 11/26/01

P8-4

R. A. DeCarlo, P. M. Lin

xlabel('Time in s')
ylabel('Inductor Current')
0.16
0.14
Inductor Current

0.12
0.1
0.08
0.06
TextEnd

0.04
0.02
0

0.001 0.002

0.003 0.004

0.005 0.006 0.007

Time in s

0.008 0.009

0.01

0.005 0.006 0.007

Time in s

0.008 0.009

0.01

(b) Here v L (t) = RiL (t) = 22.5e 500t V.

0

Inductor Voltage in V

-5
-10
-15
TextEnd

-20
-25

0.001 0.002

0.003 0.004

(c) Using linearity, for iL(0) = 50 mA, then v L (t) =

mA, then v L (t) = 22.5

SOLUTION 8.6. Since

22.5 500t
e
= 7.5e 500t V and for iL(0) = 250
3

250 500t
e
= 37.5e 500t V.
150

= L / R, we can solve for L = 5 mH. Then solving iL (t) = iL (0)e t / for iL(0)

at t = 4 sec yields iL(0) =15 mA.

1st Order Circuit Probs 11/26/01

P8-5

SOLUTION 8.7. (a) We desire to solve iL (t) = iL (0)e t /

i (0.05 ms) 9.197
Using the following ratio, L
=
=
iL (0.15 ms) 1.2447

R. A. DeCarlo, P. M. Lin

for iL(0) and R in

= 0.08 /(R + 103 ).

iL (0)e (0.05m)/
0.110 3 /
=
e
= 7.3889. Hence
iL (0)e (0.15m)/

K = 9.197/1.2447
K =7.3889
tau = 0.1e-3/log(K)
tau =5.0000e-05
L = 0.08;
Req = L/tau
Req =1.6000e+03
R = Req - 1000
R = 599.9862
iL0 = 9.197e-3/exp(-0.05e-3/tau)
iL0 = 0.0250 A
(b)

= 80m /(R + 1000) = 50 sec , iL (t) = 0.025e t /5010

A.

SOLUTION 8.8. By Ohms law, v R (0+) = (4k ||16k)iL (0+) = 32 V. The time constant
= L /(4 k ||6 k) = 25 sec , i.e.,
Req = (4e3*16e3/(4e3+16e3))
Req =3200
L = 0.08;

1st Order Circuit Probs 11/26/01

P8-6

R. A. DeCarlo, P. M. Lin

tau = L/Req
tau =2.5000e-05
Using the general equation, v R (t) = v R (0+)e t / = 32e t /25 V. Equivalently,
4
v R (t) = Req iL (t) = Req 0.01e 410 t V.

SOLUTION 8.9. (a) Note that the Thevenin resistance seen by the capacitor is Rth:
R1 = 360+60*120/(60+120)
R1 =400
Rth = 400*1200/1600
Rth =300
Hence, vC (t) = vC (0)e t / = 80e t /0.15 V where

= RTH C = 300 0.5 103 = 0.15 s.

80

Capacitor Voltage (V)

70
60
50
40
30
TextEnd

20
10
0

0.2

0.4

0.6

0.8

Time in s

(b) Here iC (0+) = v C (0 + ) / RTH = 0.2667 . Therefore iC (t) = iC (0+)e t / = 0.2667e t /0.15 A.
v (t)
Equivalently, iC (t) = C = 0.2667e t /0.15 A.
Rth

60||120

(c) By voltage division, v R (0+) = vC (0)

= 8 V; for t > 0 v R (t) = 8e t /0.15 V.

(60||120) + 160 + 200

1st Order Circuit Probs 11/26/01

P8-7

R. A. DeCarlo, P. M. Lin

SOLUTION 8.10. First, find the Thevenin equivalent seen at the left of the inductor. Introducing a test
source in place of the inductor we obtain the following KCL equation at that node.

v
itest = v test /1k + vtest 200 test 200 . Let vtest = 1 V. Then
1k

itest = 1e-3 + (1 - 200*1e-3)/200

itest =0.0050
Rth = 1/itest
Rth =200
Thus Rth = 200 ,

= L / Rth = 0.25 ms, and iL (t) = 0.025e 4000t A. Next from 8.13b find

v L (t) = (RTH iL (0))e 4000t = 5e 4000t V, and from Ohms law i x (t) = 5e 4000t mA.

SOLUTION 8.11. For all parts it is necessary to find the Thevenin equivalent resistance seen by the
capacitor. To this end we apply an external test current to the remainder of the circuit to obtain:
v test = R1i test + R2 (itest + itest )
Thus
Rth = v test / itest = R1 + R2 (1+ ) = 120 + 70(1+ )
(a) With = 4, Rth = 470 ,

= RthC = 0.1175 s, and vC (t) = vC (0)e t / = 50e 8.51t V.

50

Capacitor Voltage (V)

40

30

20
TextEnd

10

0.1

(b) With = -4, Rth = 90 ,

0.2

0.3
0.4
Time in s

0.5

0.6

= RthC = 0.0225 s, and vC (t) = 50e 44.44t V.

0.7

1st Order Circuit Probs 11/26/01

P8-8

R. A. DeCarlo, P. M. Lin

8000

Capacitor Voltage (V)

7000
6000
5000
4000
3000
2000

TextEnd

1000
0

0.02

0.04

0.06
Time in s

0.08

0.1

0.12

Note how this is not a stable design as V increases exponentially without bound.
(c) From the general equation developed at the beginning, Rth = 120 + 70(1+ ) > 0 requires that
> 2.7143.

SOLUTION 8.12. Find the Thevenin resistance left of the inductor. Forcing a test current source into
the output node,
v test = R1i test + R2 (itest R1itest ) = (100 + 50(1 100))itest = (150 5000 )i test
and
Rth = v test / itest = 150 5000
(a) Using the above equation, Rth = 350 and

= L / Rth = 7.1429 10-5 s. Hence,

iL (t) = 0.1e14000t A, an unbounded response due to the presence of the negative equivalent resistance.
= L / Rth = 2.5 104 s, v R (0+) = R2 (1 R1)i L (0) = 0 , but more importantly
v R2 (t) = R2 (1 R1)i L (t) = 0 iL (t) = 0 .

(b) Rth = 100 ,

(c) R1 + R2 (1 R1) = (150 5000) > 0 implies < 0.03.

SOLUTION 8.13. Over a long period of time the inductor L, is seen as a short circuit. Thus at time 0-,
the current through the inductor is, by current division, Is /2. As such, iL (t) = 0.5I s e Rt /2L A. A sketch
will reveal an exponentially decreasing current from an initial 0.5Is A.

1st Order Circuit Probs 11/26/01

P8-9

R. A. DeCarlo, P. M. Lin

SOLUTION 8.14. This is similar to problem 8.13. Here the current turns off at time zero instead of a
switch opening. By current division iL (0 + ) = Is /2 . The difference between this problem and problem
8.13 is that the Thevenin resistance seen by the inductor is different. Here, RTH = 2R ||0.5 R = 0.4R . So
for t > 0, iL = (Is /2)e R th t /L = 0.5I se Rt /2.5L A. A sketch of this function plotted with respect to this
new time constant will be identical to the one in problem 8.13.
Define the time constant of problem 8.13 as old . The slower decay in the plot below represents
the fall of the inductor current for problem 8.14 relative to that of problem 8.13 which is the faster
decaying curve in the plot below.

1st Order Circuit Probs 11/26/01

P8-10

R. A. DeCarlo, P. M. Lin

SOLUTION 8.15. Over a long period of constant applied voltage, a capacitor looks like an open circuit.
By voltage division, vC (0+ ) =

3
V and = 3RC. Hence
4 s
vC (t) = vC (0+)e t / = 0.75Vse t /(3RC ) V

SOLUTION 8.16. Same problem as 8.15 for t < 0. For t > 0 the effective resistance changes.
Rth = 3R / /R = 0.75R and = 0.75RC. Thus, vC (t) = 0.75Vse t /(0.75RC) V. Same behavior as in the
previous problem, except for a faster decay than in problem 8.15 due to a smaller effective resistance.
Note how the decreased resistance affects the RC and RL circuit differently.

SOLUTION 8.17. For t < 0 the inductor looks like a short circuit. Let R1 = 1333/ /800 = 500 . The
12
current supplied by the source is Is =
= 0.02 A. By current division,
100 + 500
iL (0 + ) = 0.02

1333
= 0.0125 A
800 + 1333

For t > 0, the switch is opened and the inductor sees only the 800 resistor. Hence,
and

= L / R = 25 sec

1st Order Circuit Probs 11/26/01

P8-11

R. A. DeCarlo, P. M. Lin

iL (t) = 12.5e 40000t mA

14

Inductor current in mA

12
10
8
6
4
TextEnd
2
0

20

40
60
Time in micro seconds

80

100

SOLUTION 8.18. (a) For t < 0 the applied voltage is constant and at t = 0, the capacitor is like an open
circuit. By voltage division, vC (0+ ) = 30

25k ||6.25k
= 25 V. For 0 < t < 1 ms, the source is off
(25k ||6.25k) + 1k

and the capacitor discharges through three resistors in parallel; thus

= RthC = 833.33 0.6 106 = 0.5 ms and vC (t) = 25e 2000t V.
(b) From continuity vC (0.001)= 25e 2000(0.001) = 3.383. For t > 1ms, the capacitor keeps on
discharging through only one resistance, the 25 k resistor; thus the new time constant is new = 15 ms,
and vC (t) = 3.383e (t 0.001)/0.015 V.
(c) PlotvC (t) = 25e 2000t[ u(t) u(t 0.001)] + 3.383e (t 0.001)/0.015 u(t 0.001) V
t = 0:.01:12;
vc = 25*exp(-2*t) .*(u(t)-u(t-1))+3.3834*exp(-(t-1)/15) .*u(t-1);
plot(t,vc)
grid
xlabel('Time in milli-secs')
ylabel('Capacitor Voltage in V')

1st Order Circuit Probs 11/26/01

P8-12

R. A. DeCarlo, P. M. Lin

25

Capacitor Voltage in V

20

15

10
TextEnd
5

6
Time in milli-secs

SOLUTION 8.19. (a) From Ohms law iL (0 + ) =

10

12

54
= 0.6 A. For t > 0, the Thevenin resistance
60 + 30

seen by the inductor is Rth = (60 + 30) || 720 = 80 and = L/Rth = 1/160 s. Thus
iL (t) = iL (0+ )e t / = 0.6e 160t A. From Ohms law and current division
v(t) = 60

720
i L (t) = 32e 160t V
90 + 720

(b) From continuity property, iL (0.01) = 0.6 160(0.01) = 121.14 mA. For t > 10 ms, the Thevenin
resistance seen by the inductor is Rth = (690 + 30) || 720 = 360 and new = L/Rth = 1/720 s. Hence,
iL (t) = 121.14e 720(t 0.01) mA for t > 10 ms. From Ohms law and current division
v(t) = 690

720
iL (t) = 41.793e 720(t 0.01)u(t 0.01) V
720 + 720

Therefore
iL (t) = 0.6e 160t [u(t) u(t 0.01)] + 0.12114e 720(t 0.01)u(t 0.01) A

1st Order Circuit Probs 11/26/01

P8-13

R. A. DeCarlo, P. M. Lin

0.7

Inductor Current in A

0.6
0.5
0.4
0.3
0.2

TextEnd

0.1
0

10

15

Time in milli-seconds

SOLUTION 8.20. For both circuits we first compute the Thevenin resistance seen to the right of the
capacitor for 0 t 60 ms. If we excite the circuit to the right of the capacitor over this time interval,
v
(1 0.25)v test
1
then itest = test +
= 0.0125v test . Let Rth1 =
= 80 .
200
100
0.0125
(a) For t < 0, the capacitor acts as an open circuit. Using voltage division, vC (0+ ) =

80
80 = 30
80 + 133.3

V. For 0 t 60 ms, the time constant is 1 = Rth1C = 40 ms, and

vC (t) = vC (0+ )e t

= 30e t (Rth1C ) = 30e 25t V

From continuity, vC (60+ ms) = 6.694 V. The new Thevenin resistance is Rth2 = 200 . Thus for t >
60 ms, the time constant is 2 = 200C = 100 ms, and vC (t) = 6.694e 10(t 0.06) V. The resulting
capacitor voltage is plotted below.

1st Order Circuit Probs 11/26/01

P8-14

R. A. DeCarlo, P. M. Lin

30

Capacitor Voltage (V)

25

20

15

10

TextEnd

50

100

150

200

250

time in ms

(b) It is the same circuit as above for t < 0; thus vC (0+ ) = 30 V. However the Thevenin resistances seen
by the capacitor are different because there is no switch to disconnect the independent voltage source and
its series resistance. First for 0 t 60 ms, the Thevenin resistance to the right remains as
1
= 80 . However, for 0 t 60 ms, the Thevenin resistance seen by the capacitor
0.0125
changes to Rth3 = Rth1 //133.3 = 50 . Then new time constant is 3 = Rth 3C = 25 ms and for 0 t
Rth1 =

60 ms
vC (t) = 30e 40t V
From continuity, vC (60+ ms) = 2.72 V. The new Thevenin resistance is Rth4 = 200/ /133.3 = 80 .
Thus for t > 60 ms, the time constant is 4 = 80C = 40 ms, and vC (t) = 2.72e 25(t 0.06) V. The
resulting capacitor voltage is plotted below.

1st Order Circuit Probs 11/26/01

P8-15

R. A. DeCarlo, P. M. Lin

30

Capacitor Voltage (V)

25

20

15

10

TextEnd

50

100

150

200

250

time in ms

(c) For t < 60 ms, the voltage decays faster in (b) due to the smaller time constant. Similarly, for t > 60
ms.
SOLUTION 8.21. Following, are the switching times with the time constants associated with them.
t = 0 Rth = 20 k = 20 ms
t = 5 ms Rth = 4 k = 4 ms
t = 7.5 ms Rth = 800 = 0.8 ms
It follows that with t in ms,
vC (t) = 10e 0.050t [u(t) u(t 5)] + 7.788e 0.250( t 5)[ u(t 5) u(t 7.5)] + 4.169e 1.25(t 7.5)u(t 7.5) V

1st Order Circuit Probs 11/26/01

P8-16

R. A. DeCarlo, P. M. Lin

10
9

Capacitor Voltage (V)

8
7
6
5
4
TextEnd

3
2
1
0

6
time in ms

*S OLUTION TO 8.22. This solution is done in MATLAB.

% Define switching times, inductance, and Thevenin equivalent resistances.
t1= 12e-6;
t2=18e-6;
t3=21e-6;
L= 0.1;
rth1= 800;
rth2= 8000;
rth3=1600;
rth4= 32000;
% Define time constants for each of the four time intervals.
tau1= L/rth1
tau2= L/rth2
tau3= L/rth3
tau4= L/rth4
tau1 =
tau2 =
tau3 =
tau4 =

1.2500e-04
1.2500e-05
6.2500e-05
3.1250e-06

% Compute initial inductor currents for each of the four time intervals.
il1= 100e-3;
il2=il1*exp(-t1/tau1)

10

12

1st Order Circuit Probs 11/26/01

P8-17

R. A. DeCarlo, P. M. Lin

il3=il2*exp(-(t2-t1)/tau2)
il4=il3*exp(-(t3-t2)/tau3)
il2 = 9.0846e-02
il3 = 5.6214e-02
il4 = 5.3580e-02
% Determine inductor currents for each of the four time intervals. Plot.
t = 0:0.5e-7:36e-6;
seg1= il1*exp(-t/tau1) .*(ustep(t)-ustep(t-t1));
seg2 =il2*exp(-(t-t1)/tau2) .*(ustep(t-t1)-ustep(t-t2));
seg3=il3*exp(-(t-t2)/tau3) .*(ustep(t-t2)-ustep(t-t3));
seg4= il4*exp(-(t-t3)/tau4) .* ustep(t-t3);
iL=seg1 + seg2 +seg3 + seg4;
plot(t,iL)
grid

SOLUTION 8.23.
For circuits with a forced voltage, equation 8.19c is used as a general solution,
vC (t) = vC () +

[ ( ) v ()]e
vC to+

RTH C

(a) At time zero the voltage is 0 V. As time approaches infinity, the capacitor looks like an open with
voltage 10 V. The Thevenin resistance is 10 k. Thus for t > 0.

1st Order Circuit Probs 11/26/01

P8-18

vC (t) = 10 + [10]e
+

(b) With vin(t) = 0 and vC (0 ) = 5 V, vC (t) = 5e

R. A. DeCarlo, P. M. Lin

t

2

t

2

= 10(1 e 0.5t ) V.
= 5e 0.5t V.

Part (a)

Part (b)

(c) From linearity, vC (t) = 10(1 e 0.5t ) + 5e 0.5t = 10 5e 0.5t V. Using Ohms law,
v (t) v (t)
iC (t) = in 4 C 4 . Thus iC (t) = 1 1 0.5e 0.5t = 0.5e 0.5t mA.
10
10

0.5

Capacitor Current (mA)

0.4

0.3

0.2
TextEnd

0.1

10

time in s

(d) This is the same as (a), under the condition that the input is 1.5 times larger. Hence by linearity,
vC (t) = 1.5 10(1 e 0.5t ) = 15(1 e 0.5t ) V.
(e) By linearity, ANSWER = 2x(ANSWER to (b)) + 3x(ANSWER to (a)):

1st Order Circuit Probs 11/26/01

P8-19

R. A. DeCarlo, P. M. Lin

vC (t) = 2 5e 0.5t + 3 10(1 e 0.5t ) = 15(1 e 0.5t ) = 30 40e 0.5t V

SOLUTION 8.24.
(a) At t = 0-, the capacitor looks like an open circuit; therefore, by voltage division and the continuity
3R
3R
property, vC (0 ) = vC (0+ ) =
Vs1 = 0.75Vs1 . Similarly, at t = , vC () =
V = 0.75Vs2 . The
4R
4R s2
circuit time constant is = (3R / /R)C = 0.75RC . Hence
vC (t) = 0.75Vs 2 + 0.75[Vs1 Vs 2 ]e

0.75RC

(b) A sketch will show an exponentially varying voltage from 0.75Vs1 converging to 0.75Vs2 with the
computed time constant.
(c) The response to the initial condition when the inputs are set to zero, zero-input response, is
vC (t) = Vs1e

0.75RC

. The zero order response, the response with 0V initial condition to a forced

voltage, is vC (t) = Vs 2 Vs 2e

0.75RC

SOLUTION 8.25.
For RL circuits with a forced current, equation 8.19b is used as a general solution:
iL (t) = iL () +

[ ( ) i ()]e
iL to+

Since Rth = R = 100 and L = 0.4 H, we have

R th
(t t o)
L
.

= 4 ms and

[( )

iL (t) = iL () + iL to+ iL () e 250(t to )

(a) Here, iL (0) = 0 and as time approach infinity, the inductor becomes a short and
iL () = 10 100 = 0.1 A. Thus iL (t) = 0.11 e 250t A.

(b) Here iL (0) = 50 mA and because the input is zero, iL () = 0 . Thus, iL (t) = 0.05e 250t A. Plots
for parts (a) and (b) appear below.

1st Order Circuit Probs 11/26/01

P8-20

R. A. DeCarlo, P. M. Lin

0.1

Inductor Current (A)

0.08
0.06
0.04
0.02
0
TextEnd

-0.02
-0.04
-0.06

10
Time in ms

15

20

(c) By linearity iL (t) = 0.11 e 250t 0.05e 250t = 0.1 0.15e 250t A. Further, by KVL and Ohm's
law, v L (t) = vin (t) 100iL (t) implies v L (t) = 10 10 + 15e 250t = 15e 250t V.

Inductor Voltage (V)

15

10

5
TextEnd

10
Time in ms

15

20

(d) Observe that the new initial condition is 0.5 times the old one and that the new input voltage is 1.5
times the old one. Therefore, by linearity,

iL (t) = 1.5 0.11 e 250t + (0.5) 0.05e 250t = 0.15 0.125e 250t
and thus

v L (t) = 15 15 + 12.5e 250t = 12.5e 250t V

The plot is similar to part (c) with initial point 12.5 instead of 15.
SOLUTION 8.26. For this problem Rth = 2R / /0.5R = 0.4R in which case

= L Rth = L 0.4R .

1st Order Circuit Probs 11/26/01

P8-21

R. A. DeCarlo, P. M. Lin

(a) At t = 0, the inductor looks like a short circuit. Hence by current division, iL (0 ) = iL (0+ ) = 0.5Is1 .
A similar argument yields iL () = 0.5Is 2 . Using the general form of the solution,

0.4Rt
L

iL (t) = 0.5I s2 + 0.5[ Is1 Is 2 ]e

.
(b) A sketch will show an exponentially varying current from 0.5Is1 A converging to 0.5Is2 .
(c) The response to the initial condition when the inputs are set to zero, zero-input response, is
iL (t) = 0.5I s1e

0.4Rt
L

. The zero state response, the response with no initial condition, to the input
0.4Rt

Is2 u(t), is iL (t) = I s2 1 e L .

SOLUTION 8.27. For this problem, vC (0) = 20 10e 0.4t

)t =0 = 10 V and

vC () = 10Is = 20 = lim 20 10e 0.4t . Hence Is = 2 A. Further,

t
0.4t

dvC (t)
= 4Ce
= 0.4e 0.4t which implies that C = 0.1 F. Since
dt
= 1/0.4 = 2.5 = (10+ R)C = 0.1(10 + R) , it follows that R = 15 .

iC (t) = C

SOLUTION 8.28.
(a) The Thevenin resistance for this configuration is Rth = 1000/ /1000 = 500 and
Hence vC (t) = vC (0+ )e t = 15e 4t V is the zero-input response.

= RthC = 0.25 s.

(b) Using a source transformation and voltage division, vC () = 3 V. Thus vC (t) = 3 1 e 4t V.

(c) Here vC () = 4 V, thus vC (t) = 4 1 e 4t V.

(d) This is the superposition of parts (b) and (c), i.e., vC (t) = 7 1 e 4t

(e) The complete response is the superposition of parts (d) and (a), i.e., vC (t) = 7 + 8e 4t V.
(f) From linearity,
vC (t) = 0.5 7 1 e 4t + 2 15e 4t = 3.5 + 26.5e 4t

SOLUTION 8.29. Solution done in MATLAB

%Problem 8.29
%RTH= (60||120)+120
%tau=L/RTH
RTH=1/(1/60+1/120)+120;
tau=0.2/RTH;
%Using superposition at t<0
il01=36/120/2;
il02=(72/180)*(60/180);
il0= il01+il02;

1st Order Circuit Probs 11/26/01

P8-22

R. A. DeCarlo, P. M. Lin

% At t>0 36volts is off thus

ilinf=il02;
t=0:5*tau/1000:5*tau;
ilt=ilinf+(il0-ilinf)*exp(-t/tau);
plot(t,1000*ilt);
xlabel('Time in seconds');
ylabel('Current in mA');
300
280
260

Current in mA

240
220

200
180
160
140
120

3
4
Time in seconds

*S OLUTION 8.30.
% Rth = 120 + 120\\60 = 160 kohm
% tau = Rth*C
tau = 160e3*0.5e-6
tau =
0.0800
% Initial capacitor voltage is computed by
% voltage division and superposition
vc0 = 24+24
vc0 =
48
% At t = , capacitor looks like an open circuit. Hence
vcinf=24;
t = 0:1e-3:5*tau;
vct = vcinf+(vc0-vcinf)*exp(-t/tau);
plot(t*1000,vct)
grid
xlabel('Time in msec')
ylabel('Capacitor voltage (V)')

7
x 10

-3

1st Order Circuit Probs 11/26/01

P8-23

R. A. DeCarlo, P. M. Lin

50

Capacitor voltage (V)

45
40
35
30
25
20

50

100

150

200
250
Time in msec

300

*S OLUTION 8.31A
vc0 =0;
% Consider 0 t 0.5
vcinf = 50;
% Rth = 600\\300 = 200 ohms
% tau1 = Rth*C
tau1 = 300e-6*200
vc0 = 0;
t = 0:5e-3:1;
vct = (vcinf+(vc0-vcinf)*exp(-t/tau1)) .* (ustep(t)-ustep(t-.5));

% Consider 0.5 t 1
tau2 = tau1
vc5 = (vcinf+(vc0-vcinf)*exp(-.5/tau1))
vcinf2 = 80;
vct2 = (vcinf2+(vc5-vcinf2)*exp(-(t-0.5)/tau1)) .* ustep(t-0.5);
vca = vct+vct2;
plot(t,vca)
grid
xlabel('Time in sec')
ylabel('Capacitor Voltage (V)')

350

400

1st Order Circuit Probs 11/26/01

P8-24

R. A. DeCarlo, P. M. Lin

80

Capacitor Voltage (V)

70
60
50
40
30
20
10
0

0.2

0.4

Time in sec

0.6

*S OLUTION 8.31B
vc0 =0;
% for 0 t 0.5
vcinf = 75;
tau1 = 300e-6*300
vc0 = 0;
t = 0:5e-3:1;
vct = (vcinf+(vc0-vcinf)*exp(-t/tau1)) .* (ustep(t)-ustep(t-.5));
% for 0.5 t 1
tau2 = 300e-6*200
vc5 = (vcinf+(vc0-vcinf)*exp(-.5/tau1))
vcinf2 = 80;
vct2 = (vcinf2+(vc5-vcinf2)*exp(-(t-0.5)/tau1)) .* ustep(t-0.5);
vcb = vct+vct2;
plot(t,vcb)
grid
xlabel('Time in sec')
ylabel('Capacitor Voltage (V)')
pause
plot(t,vca,t,vcb,'b')
grid

0.8

1st Order Circuit Probs 11/26/01

P8-25

R. A. DeCarlo, P. M. Lin

80

Capacitor Voltage (V)

70
60
50
40
30
20
10
0

0.2

0.4
0.6
Time in sec

0.8

0.4

0.8

Comparison of the two responses.

80
70
60
50
40
30
20
10
0

SOLUTION 8.32.
%Problem 8.32
%Consider t < 0
%vin=-20V, thus
vc0=(8/10)*(-20);
%For 0<t<20ms
RTH=1/(1/2e3+1/8e3);
tau1=5e-6*RTH;

0.2

0.6

1st Order Circuit Probs 11/26/01

P8-26

R. A. DeCarlo, P. M. Lin

vcinf=(8/10)*20;
t=20e-3;
vc20ms=vcinf+(vc0-vcinf)*exp(-t/tau1);
%For 20ms<t
RTH=1/(1/2e3+1/8e3+1/1.6e3);
tau2=5e-6*RTH;
%By superposition
vcinf2=20*0.5+20*0.4;
t=0:(40e-3)/1000:40e-3;
vct= (ustep(t)-ustep(t-20e-3)).*(vcinf+(vc0-vcinf).*exp(-t./tau1)) ...
+ ustep(t-20e-3).*(vcinf2+(vc20ms-vcinf2).*exp(-(t-0.02)./tau2));
plot(t,vct);
grid;
xlabel('time in seconds');
ylabel('Volts');
20

15

10

Volts

-5

-10

-15

-20

0.005

0.01

SOLUTION 8.33.
(a)
%Problem 8.33
%(a)
%at t<0 only one source is contributing thus,
il0=24/60*0.5;
%For t>0
RTH=60+1/(1/30+1/60);
tau=16e-3/RTH;
%As t goes to infinity, by superposition,
ilinf=24/(60)*0.5+24/80*1/3;
t=0:5*tau/1000:5*tau;
ilt=ilinf+(il0-ilinf).*exp(-t/tau);
figure(1);
plot(t,1000.*ilt);
grid;
xlabel('time in seconds');
ylabel('Current in mA');

0.015

0.02
0.025
time in seconds

0.03

0.035

0.04

1st Order Circuit Probs 11/26/01

P8-27

R. A. DeCarlo, P. M. Lin

300
290
280

Current in mA

270
260
250
240
230
220
210
200

0.1

0.2

0.3

0.4
0.5
0.6
time in seconds

0.7

0.8

0.9

1
x 10-3

(b) Using 8.23, The time constant remains the same as in (a). In order to find the inductor voltage, we
must do so indirectly by solving for v L (t) = V A 60IL . At t=0+,
iL (0 + ) = 200mA
60
30
VA = 24
+ 24
iL (0+ )(60|| 30) = 20V
90
90
v L (0+ ) = x(t0+ ) = V A 60iL (0+ ) = 8V

For t > 0, X e = v L () = 0V , Thus v L (t) = 8e 5000tV .

(c) By linearity,
iL (t) = 600 200e 5000t mA
v L (t) = 16e 5000tV
VA

SOLUTION 8.34.
(a) Since the voltage has been constant for a long time, the capacitor acts as an open circuit. Thus by
voltage division and continuity, vC (0 ) = vC (0+ ) = 0.75Vo .
(b) RTH = 6R ||18R || 3R = 1.8R .
(c) For that period of time the switch is closed, vC (t) = 0.9Vo 0.15Voe t /1.8RC .
(d) Using the previous equation, and by continuity, vC (T ) = vC (T + ) = 0.9Vo 0.15Voe T /1.8RC .
(e) The time constant remains the same as the only difference is the source turning off.
(f) For t > T, vC (t) = vC (T + )e (t T )/1.8RC .
(g)

1st Order Circuit Probs 11/26/01

P8-28

R. A. DeCarlo, P. M. Lin

0.9
0.8

Normalized Volts in terms of Vo

0.7
0.6
0.5
0.4
0.3
0.2
0.1
0

6
8
10
12
Normalized time in terms of tau

14

16

18

SOLUTION 8.35.
(a) For t < 0, The switch is closed, the current source off, and the voltage source has been providing a
constant voltage for a prolonged period of time. Thus, vC (0 ) = 50uV .
(b) Since voltage in continuous across a capacitor, vC (0+ ) = v C (0 ) = 50uV .
(c) The thevenin resistance seen by the capacitor is RTH = 200||200 = 100 , thus = RTH C = 0.2s .
(d) As t goes to infinity the capacitor voltage goes to 16V, thus vC (t) = 16 + (50u 16)e 5tV .
(e) Again using the continuous property of a capacitor, vC (0.5+ ) = v C (0.5 ) = 14.687V .
(f) For t > 0.5s, the switch is open, thus RTH = 200 . = RTH C = 0.4s .
(g) As t goes to infinity, the capacitor voltage goes to 32V. vC (t) = 32 + (14.687 32)e 2.5(t 0.5)V .
(h)
35

30

25

Volts

20

15

10

-5

SOLUTION 8.36.
(a) vC (0+ ) = 5V
(b) Doing so in matlab.
%Problem 8.36b
%Initial condition
vc0=-5;
%From 0<t<80us
RTH=300e3;
tau=RTH*(1/3)*1e-9;

0.5

1
1.5
Time in seconds

2.5

1st Order Circuit Probs 11/26/01

P8-29

R. A. DeCarlo, P. M. Lin

%So the initial condition on Vout

vout0=(10+5)/300e3*60e3-5; %-2V
voutinf1=10;
t=80e-6;
vout80us= 10+(vout0-10)*exp(-t/tau);
%For t > 80us
%tau stays the same
voutinf2=-5;
t=0:160e-6/1000:160e-6;
vout= (ustep(t)-ustep(t-80e-6)).*(10+(vout0-10).*exp(-t./tau)) ...
+ustep(t-80e-6).*(-5+(vout80us+5).*exp(-(t-80e-6)./tau));
plot(t,vout);
xlabel('time in seconds');
ylabel('Volts');
grid;
%Notice that Vout is vin-vct times a constant 60/300 plus vct
%Thus Vout=vin(60/300)+vct(1-60/300)
5

Volts

-1

-2

(c)

0.2

0.4

0.6

0.8
1
time in seconds

1.2

1.4

1.6
-4
x 10

1st Order Circuit Probs 11/26/01

(d)

SOLUTION 8.37.

P8-30

R. A. DeCarlo, P. M. Lin

1st Order Circuit Probs 11/26/01

P8-31

R. A. DeCarlo, P. M. Lin

If the source voltage has been 10 V for a long time then the switch is open and vC (0+ ) = 10V . The
time constant with the switch open is = 5us . At t > 0, the input voltage changes to 20V. It then follows
that vC () = 20V , and
vC (t) = 20 30e t /5uV .
Using the elapsed time formula, 8.24, we wish to find when the switch closes.
10 20
ta 0 = 5us ln
= 2.03us . At that time the input voltage is still 20V, and the switch closes.
0 20
The time constant is now = 98ns , and vC () = 0.39V . Note that because the voltage converges to a
value greater than zero, this time interval will be from 2.03us to 5us when the input changes back to
10V, thus
vC (t) = 0.39 0.39e t /98nV .
vC (5us) = 0.39V . At t > 5us, the voltage changes to 10 V, so
vC (t) = 0.2 + (0.39 + 0.2)e t /98nV .
Using the elapsed formula, we get tb = 0.1us for the voltage to go down to 0V again and cause the switch
to open again. At this point the time constant becomes the original value again and
vC (t) = 10 + 10e t /5uV
for t > (0.1+5) us.
SOLUTION 8.38.
(a) Introduce a test current source at the output and write KVL,
3
itest = v test /200 + (v test 6V ) /200 + (6V v test ) / 400. Solving for itest = v test
0.015 . This
400
implies the following,
RTH = 400 /3
iSC = 15mA
vOC = RTH iSC = 2V
(b) Using the general form, vC (t) = 2 8e 15000t .
SOLUTION 8.39.
(a) Introduce a test voltage and solve for KVL,
v test = 5kitest + 101(v test + 1 40itest ) 1 + 40itest
v test = 10itest 1
vOC = 1V
RTH = 10
(b) The complete response is vC (t) = 1 + e t V. Note that the voltage goes to infinity as t goes to infinity
because of the negative time constant.
SOLUTION 8.40.
Compute the thevenin equivalent seen by the inductor at t > 0. Using KCL write,
itest = 100 itest + (v test 100itest + 25) /50 . Then one obtains the following,

1st Order Circuit Probs 11/26/01

P8-32

R. A. DeCarlo, P. M. Lin

itest = v test /125 + 1 / 5 A

RTH = 125
iSC = 1 / 5 A
vOC = 25V
A t < 0, the applied voltage has been 25V for a long time. Using the previously obtained thevenin
equivalent and linearity, iL (0 ) = 1/5 A . iL () = i SC = 1/5 A , So iL (t) = 0.2 + 0.4e 6250t A .
SOLUTION 8.41.
(a) Introducing a test source and using KCL,
v 1.5v s v test
itest = test
+
40
100
v
1.5v s v test
0.2 = s +
40
40
v
Solving for the test source current in terms of the test voltage, itest = test 0.12 A. Thus the thevenin
50
equivalent is,
Rth = 50
voc = i sc * Rth = 6V
(b) vC (t) = 6 6e 200t V.
(c) From linearity all the currents increase by the same ratio, thus vC (t) = 15 15e 200t V.
(d)

(e)

1st Order Circuit Probs 11/26/01

P8-33

R. A. DeCarlo, P. M. Lin

SOLUTION 8.42.
%Problem 8.42
C= 1e-6;
vc0=0;
%For 0 < t < 5ms
Rth= 20e3;
tau1= Rth*C;
vcinf1= 50e-3*Rth;
vc5ms= vcinf1+(vc0-vcinf1)*exp(-5e-3/tau1);
%For 5ms < t < 7.5ms
Rth= 4e3;
tau2= Rth*C;
vcinf2= 50e-3*Rth;
vc75ms= vcinf2+(vc5ms-vcinf2)*exp(-(7.5e-3-5e-3)/tau2);
%For t > 7.5ms
Rth=800;
tau3= Rth*C;
vcinf3= 50e-3*Rth;
t=0:12e-3/1000:12e-3;
vct= (ustep(t)-ustep(t-5e-3)).*(vcinf1+(vc0-vcinf1).*exp(-t/tau1)) ...
+ (ustep(t-5e-3)-ustep(t-7.5e-3)).*(vcinf2+(vc5ms-vcinf2).*exp(-(t-5e-3)/tau2)) ...
+ (ustep(t-7.5e-3)).*(vcinf3+(vc75ms-vcinf3).*exp(-(t-7.5e-3)/tau3));
plot(t,vct);
grid;

1st Order Circuit Probs 11/26/01

P8-34

R. A. DeCarlo, P. M. Lin

xlabel('time in seconds');
ylabel('Volts');
250

200

Volts

150

100

50

0.002

0.004

0.006
time in seconds

0.008

0.01

0.012

SOLUTION 8.43.
(a) From the thevenin resistance
(b) From the thevenin resistance
(c) From the thevenin resistance
(d) From the thevenin resistance
(e) 0 mA.
(f) In MATLAB:
%Problem 8.43f

RTH = 800 , 1 = 125 s .

RTH = 8k , 2 = 12.5 s .
RTH = 1.6k , 3 = 62.5 s .
RTH = 32k, 4 = 3.125 s .

tau1= 125e-6;
tau2= 12.5e-6;
tau3= 62.5e-6;
tau4= 3.125e-6;
vs= 100;
il0= 0;
ilinf1= vs/800;
ilinf2= vs/8e3;
ilinf3= vs/1.6e3;
ilinf4= vs/32e3;
il12us= ilinf1+(il0-ilinf1)*exp(-12e-6/tau1);
il18us= ilinf2+(il12us-ilinf2)*exp(-(18e-6-12e-6)/tau2);
il21us= ilinf3+(il18us-ilinf3)*exp(-(21e-6-18e-6)/tau3);
t= 0:36e-6/1000:36e-6;
ilt= (ustep(t)-ustep(t-12e-6)).*(ilinf1+(il0-ilinf1).*exp(-t/tau1))+ ...
(ustep(t-12e-6)-ustep(t-18e-6)).*(ilinf2+(il12us-ilinf2).*exp(-(t-12e-6)/tau2))+ ...
(ustep(t-18e-6)-ustep(t-21e-6)).*(ilinf3+(il18us-ilinf3).*exp(-(t-18e-6)/tau3))+ ...
(ustep(t-21e-6)).*(ilinf4+(il21us-ilinf4).*exp(-(t-21e-6)/tau4));
plot(t,1000*ilt);
grid;
xlabel('time in seconds');
ylabel('Current in mA');

1st Order Circuit Probs 11/26/01

P8-35

R. A. DeCarlo, P. M. Lin

15

Current in mA

10

0.5

1.5

2
2.5
time in seconds

3.5

4
x 10-5

SOLUTION 8.44.
The first stage is a differentiator, and from 8.25, the output of the first op-amp is
dv (t)
= RC in = 0.25RCe0.25t . The second stage is an integrator and using 8.26,
dt
t
1 t
vout (t) =
(0.25RCe0.25 d ) = 0.5 e 0.25
= 0.5e 0.25t

2RC 0
0

SOLUTION 8.45.
(a) First note the following relationships,
1 t
vout (t) = vC 2 (t) =
i ( )d
C2 0 C 2
vin (t) = v C1(t)
dv (t)
v (t)
dv (t)
iC 2 (t) = v in (t) / R C1 C1 = in C1 in
dt
R
dt
Doing the appropriate substitution, and solving,
1 t
C
C
vout (t) =
v in ( )d 1 vin (t) + 1 vin (0).

C2R 0
C2
C2
4
C
(b)
e 0.25t 1 + 1 1 e 0.25t V.
C2 R
C2
1
C
(c)
sin( t) + 1 (1 cos( t)) V.
C2R
C2

SOLUTION 8.46.
These are two integrator in cascade. Using 8.26, the output of the first stage is
1 t

vin ( )d = 2[cos(50 )]t0 mV = 2cos(50t) 2 mV. Using the same equation again,

0
RC
t
1 t
2

vout (t) =
(2cos(50 )d = 10 sin(50 ) mV = 2cos(50t)mV
RC 0
50
0

1st Order Circuit Probs 11/26/01

P8-36

*S OLUTION 8.47. The following solution is done in MATLAB

c= 1e-6;
rf= 10e6;
rs=1e6;
tau=c*rf;
vgain= -10/1;
vofinal= -3*vgain;
% Part (a)
voinit= 0;
t= 0: tau/100: tau;
vout= vofinal + (voinit -vofinal).*exp(-t/tau);
% Time at which output voltage reaches saturation is tsat
tsat= tau*log((0 - vofinal)/(15- vofinal))
plot (t,vout)
grid
xlabel('Time in secs')
ylabel('Output Voltage in volts')
tsat =
6.9315e+00

% Part (b)
voinit=-5;
vout= vofinal + (voinit -vofinal).*exp(-t/tau);
tsat= tau*log((-5 - vofinal)/(15- vofinal))
plot (t,vout)
grid
xlabel('Time in secs')

R. A. DeCarlo, P. M. Lin

1st Order Circuit Probs 11/26/01

P8-37

R. A. DeCarlo, P. M. Lin

ylabel('Output Voltage in volts')

tsat =
8.4730e+00

(c) vC(0 ) = 0. Observe saturation at about 6.4 seconds.

Leaky Integrator-Transient-7
+0.000e+000

(V)

+2.000

+4.000

+6.000

+15.000

+10.000

+5.000

+0.000e+000
V(IVM)

vC(0 ) = -5 V. Observe that the time of saturation is 8.198 seconds.

Time (s)
+8.000

+10.000

1st Order Circuit Probs 11/26/01

P8-38

R. A. DeCarlo, P. M. Lin

Prb Sol 8.47-Transient-16

+0.000e+000

(V)

+2.000

+4.000

Time (s)
+6.000

+8.000

+10.000

+15.000

+10.000

+5.000

+0.000e+000

-5.000

V(IVM)

Note that in both cases the time of saturation is much lower than in the MATLAB computations which
assume an ideal op amp. In the Burr Brown model used by the SPICE simulation of this circuit, the
input resistance is 2 M which is comparable with the external input resistance. Hence the assumption
of an infinite input resistance is not valid for the SPICE simulation and causes the discrepancy in the
time of saturation. However, if the external input resistance is changed to 10 k and the feedback
resistance to 100 k with a corresponding change in the capacitor to 100 F, one obtains results
comparable to the MATLAB computations.

SOLUTION 8.48.
Since the op-amps do not load the first stage of the circuit, we can find its transfer function for the opv (t)
amp stage as out
= (1+ K) .
v + (t)
(a) From the problem statement, we know that the overall function, is a scaled integrator. As the op-amp
stage only provides gain it is logical to assume that the R-C stage will perform the integration of the input
times some constant, G. With this in mind we have
t

vout (t) = G(1+ K) 10sin (

0

)d =

10G(1+ K)

(b) Using the same reasoning, vout (t) = G(1+ K)

(cos( t) 1) , where G must be negative.

d(10sin( t))
= 10G(1+ K) cos( t)where G is
dt

positive.
For low frequency (a) yields a big output, while (b) a small one. For high frequency the reverse happens.

1st Order Circuit Probs 11/26/01

P8-39

R. A. DeCarlo, P. M. Lin

SOLUTION 8.49.
For (b) the integral i-v relationship is
v(t) = vCeq (0 + ) +

1 t
C + C2 t
i( )d = vC1 (0+ ) + v C 2 (0+ ) + 1
i( )d . Repeating the same for (a),

Ceq 0
C1C2 0

1 t
i( )d
C1 0
1 t
vC 2 (t) = v C2 (0 + ) +
i( )d
C2 0
vC1 (t) = vC1(0 + ) +

By KVL the two capacitor voltage can be added together, thus give the same relationship as for (b).
SOLUTION 8.50.
First calculate Ceq = C1 ||C2 = 0.2F . Then find the initial voltage
vCeq (0 + ) = vC1 (0+ ) + v C 2 (0+ ) = 30V . The final voltage will be 12V, and the time constant is
= Req Ceq = 0.4s . Thus vout (t) = 12 + (30 12)e 2.5tV .
*S OLUTION 8.51. (a) After the switch closes, we have the circuit shown below.

From Chapter 7, Ceq = 0.5 F and vC (0+ ) = v C1(0+ ) v C 2 (0+ ) = 2 0 = 2 V. Hence,

iR (t) = i R (0+ )e t / = iR (0+ )e

t /RC eq

v (0 + ) t /RCeq
= C
e
= 4e 4t u(t) A
R

(b) For this part we apply the integral definition of the capacitor. Specifically,

1st Order Circuit Probs 11/26/01

P8-40

R. A. DeCarlo, P. M. Lin

1
vC1 (t) = vC1(0 ) +
iC1( ) d = 2 iR ( ) d = 2 4 e 4 d = 1+ e 4t V

C1
+

and
t

1
vC 2 (t) = v C2 (0 ) +
i ( ) d = 0 + iR ( )d = 4 e 4 d = 1 e 4t V
C2 C 2
+

(c) The energy stored at t = 0 for each capacitor is:

2
WC1 (0+ ) = 0.5C1vC1
(0 + ) = 2 J

and
WC2 (0 + ) = 0.5C2vC2 2 (0+ ) = 0 J
Further at t = ,
2
WC1 () = 0.5C1vC1
() = 0.5 J

and
WC2 () = 0.5C2vC2 2 () = 0.5 J
Computing total instantaneous stored energies, we have
WCtot (0 + ) = 2 J and WCtot () = 1 J
+

Hence the decrease in stored energy from 0 to is 1 J.

+

(d) Computing the energy dissipated in the resistor over [0 , ) is

WR (0,) = R

iR2 (

vC2 (0+ ) 2t /RCeq

vC2 (0+ ) RCeq
2t /RCeq
)d = R
e
d
=

R
2
0
R2 0

Ceqv C2 (0 + )
=
=1J
2
(e) From the expressions developed in part (d), the dissipated energy is independent of the value of R.
R only affects the rate at which energy is dissipated. Clearly, the energy stored at 0 is 2 J while
the energy dissipated over [0,) is 1 J and the remaining energy at t = is 1 J. Hence
conservation of energy is verified.

1st Order Circuit Probs 11/26/01

P8-41

R. A. DeCarlo, P. M. Lin

SOLUTION 8.52. (a) Using the relations developed in P8.49,

Ceq = 0.2F
vCeq = 1 0.5 = 0.5V
vCeq (0+ ) t /RC eq
iR (t) = i R (0 )e
= iR (0 )e
=
e
= e 10t u(t)
R
(b) For this part we apply the integral definition of the capacitor. Specifically,
+

t /

t /RC eq

1
1
vC1 (t) = vC1(0 ) +
iC1( ) d = 1 iR ( ) d = 1 e 10 d = (9 + e 10t ) V

C1
10
+

and
t

1
1
vC 2 (t) = v C2 (0 ) +
iC 2 ( ) d = 0.5 + 4 iR ( )d = 0.5 + 4 e 10 d = (9 4e 10t ) V

C2
10
+

(c) The energy stored at t = 0 for each capacitor is:

2
WC1 (0+ ) = 0.5C1vC1
(0 + ) = 0.5 J

and
WC2 (0 + ) = 0.5C2vC2 2 (0+ ) = 31.25mJ
Further at t = ,
2
WC1 () = 0.5C1vC1
() = 405mJ

and
WC2 () = 0.5C2vC2 2 () = 101.25mJ
Computing total instantaneous stored energies, we have
WCtot (0 + ) = 531.25mJ

and WCtot () = 506.25mJ

Hence the decrease in stored energy from 0 to is 25 mJ.

+

(d) Computing the energy dissipated in the resistor over [0 , ) is

1st Order Circuit Probs 11/26/01

WR (0,) = R

iR2 (

P8-42

R. A. DeCarlo, P. M. Lin

2
2
vCeq
(0 + ) 2t /RCeq
v Ceq
(0+ ) RCeq
2t /RCeq
)d = R
e
d
=

2

R
2
0
R
0

2
Ceqv Ceq
(0+ )
=
= 25mJ
2
(e) From the expressions developed in part (d), the dissipated energy is independent of the value of R.
R only affects the rate at which energy is dissipated. Clearly, the energy stored at 0 is 531.25 mJ
while the energy dissipated over [0,) is 506.25 mJ and the remaining energy at t = is 25 mJ.
Hence conservation of energy is verified.

SOLUTION 8.53. As all the switches are open initially, the initial current through the inductors is 0A.
For 0 < t < 50 ms, iL (t) = 54.54 54.54e 20000t mV. At t > 50 ms, the equivalent inductance is 10 mH,
the initial current through the 110 mH inductance is 54.54 mA, and through the 11 mH inductance 0 A.
So assuming the current splits equally between the two branches in steady state,
iL1 = 27.27 + (54.54 27.27)e 220000t
iL 2 = 27.27 27.27e 220000t
SOLUTION 8.54. (a) Charges will distribute in order to achieve equal voltage by KVL. Since q=CV,
v R (0 ) = 0V , due to equal capacitance the charges will distribute half and half, v R (0 + ) = 0.5V .
(b) The equivalent capacitance is 2 F, thus v R (t) = 0.5e 0.5t .
*S OLUTION 8.55. (a) Writing a node equation at v we have for all t,

dv v
d
(v v s ) = 0
+ + 4 (v v s ) +
dt 4
dt
2

(*)

Equivalently,
8

dv
3v
dv
v
= +4 s + s
dt
4
dt
2

(**)

Grouping terms and dividing by 8 yields when t > 0,

dv
3
1
= v+
dt
32
16

(***)

1st Order Circuit Probs 11/26/01

P8-43

Notice that vs = 1 for t > 0 and for t > 0,

R. A. DeCarlo, P. M. Lin

dvs (t) du(t)

=
= 0.
dt
dt

(b) By inspection v(0 ) = 0; both capacitors are uncharged at 0 . Recall from part (a) that KCL at the
node for v yields (*) which is equivalent to (**). Since conservation of charge follows by
integrating (*) or equivalently integrating (**) we have
0+

8
0

dv
3
d =
d
4

0+

0+

v d

+4

0+

dv s
d + 0.5 v s d
d

Since the integral of a finite integrand over an infinitesimal interval is zero, we have equivalently,
0+

dv
d = 0+ 4
d

0+

dv s
d +0
d

Evaluating these integrals we obtain

) (

8 v(0+ ) v(0 ) = 4 v s (0+ ) v s (0 ) = 4u(0+ ) = 4

-

(c) Since v(0 ) = 0, v(0 ) = 0.5 V. Since v satisfies (***), i.e.,

dv
3
1
1
= v + v + F (***)
dt
32
16
from equation 8.17,

2 1

v(t) = F + v(0+ ) F e t / = e 3t /32 u(t) V

3 6
-

Using v(0 ) would have led to an incorrect answer.

SOLUTION 8.56. (a)
x(t) = K1e t / + K 2
dx(t)
K
= 1 e t /
dt
K
1
doing the substitution, 1 e t / = (K1e t / + K 2 ) + F . In order to satisfy the equality, K 2 = F .

1st Order Circuit Probs 11/26/01

P8-44

[
]
+ K 2 = [ x(t0+ ) F ]e (t t )/ + F

R. A. DeCarlo, P. M. Lin

(b) x(t +0 ) = K1e t 0 / + F , and K1 = x(t0+ ) F e t 0 / .

(c) x(t) = K1e t /

SOLUTION 8.57. (a) From the graph, the initial and final values are 0 and 80 V respectively. That sets
the following constraint, 100

R2
= 80V . From vC ( ) = 80 80e 1 = 50.57V . Looking at the Graph
R1 + R2

= 5ms . Thus R1 ||R2 C = 5ms . Solving, R1 = 6250 and R2 = 25k .

(b) Using the same equalities, R1 = 0.25R2 = 2k , and C =

2k || 8K

= 3.125uF .

SOLUTION 8.58. (a) From the graph, the initial and final values are 0 and 100 mA respectively. Thus
R1 = 200 /100m = 2k. From iL ( ) = 100 100e 1mA = 63.21mA , the graph shows a
L /(R1 || R2 ) = 20ms, and R2 = 0.25m.
(b) R1 stays the same, L = 20ms(R1 || R2 ) = 20H .
SOLUTION 8.59. This question is done in matlab
%Problem 8.59
tau1= 20*1;
tau2= 1;
%For 0 < t < ta
vo= 0;
vinfa= 10;
%Using the elapsed time formula,
ta=tau1*log((0-10)/(9-10));
%For ta < t < tb
vinfb=0;
tb=tau2*log((9-0)/(1-0));
%For tb < t < tc
vinfc=vinfa;
tc=tau1*log((1-10)/(9-10));

= 20ms . Thus

1st Order Circuit Probs 11/26/01

P8-45

R. A. DeCarlo, P. M. Lin

%Next switching is just a repeat of ta < t < tb

t1=ta;
t2=ta+tb;
t3=t2+tc;
t4=t3+tb;
t5=t4+tc;
t=0:t5/1000:t5-1/1000;
vt= (ustep(t)-ustep(t-t1)).*(10-10.*exp(-t/tau1))+ ...
(ustep(t-t1)-ustep(t-t2)).*(9.*exp(-(t-t1)/tau2))+ ...
(ustep(t-t2)-ustep(t-t3)).*(10-9.*exp(-(t-t2)/tau1))+ ...
(ustep(t-t3)-ustep(t-t4)).*(9.*exp(-(t-t3)/tau2))+ ...
(ustep(t-t4)-ustep(t-t5)).*(10-9.*exp(-(t-t4)/tau1));
Frequency= 1/(tb+tc)
plot(t,vt);
grid;
xlabel('time in seconds');
ylabel('Volts');
(b) The frequency is 0.0217 Hz.
9
8
7

Volts

6
5
4
3
2
1
0

20

40

60
80
time in seconds

100

120

140

SOLUTION 8.60. When the switch is in position A, = 18.18ms . In position B it is

the elapsed time formula, find ta, when the output voltage reaches 90 V.

= 99.5us . Using

1st Order Circuit Probs 11/26/01

P8-46

R. A. DeCarlo, P. M. Lin

60 136.36
ta = 18.18msln
= 9.07ms . At this point the switch goes to B, and the elapsed time until the
90 136.36
90 59.45
voltage reaches 60 V is tb = 99.5usln
= 0.4ms . Adding both time, F = 105.6Hz .
60 59.45
SOLUTION 8.61. (a) The circuit can be rearranged in a series of one Vsolar V voltage source, one Lstore
inductor, and one Rstore +R solar resistor.
(b) iL (t) =

Vsolar
1 e (Rstore+Rsolar )t /L A .
Rsolar + Rstore

(c) In this time period the circuit reduces to an Lsotre inductor in series with a Rstore +R 1 resistor.
(d)
Vsolar
Rsolar + Rstore

iL (T1 ) =

iL (t) = iT e

(Rstore+ R1)(t T 1)/L

.
A

(e) The two elements in series are an Lstore inductor and a resistor Req = Rstore + (R1 || R2 ).
(f)
iL (T2 ) = iT e (Rstore+ R1)(T 2 T 1)/Lsotre A
1

iL (t) = iT e (Re q)(t T 2)/Lstore A

2

(g)

V 2 solar
PVsolar = Vsolar iL (t) =
1 e (Rstore+Rsolar)t /L W
Rsolar + Rstore
PRsolar = Rsolar iL2 (t)W
PRstore = RstoreiL2 (t)W
di (t)
PLstore = LstoreiL (t) L W
dt
1
(h) WL (0,t) = LiL2 (t)J
2
SOLUTION 8.62. The light turns off when the current through it goes down to 0.5 mA. This
corresponds to ib = 10uA , and a voltage across the capacitor of vC = ib(R1 + 2k) + 0.5 = 0.7V . The time
constant of this circuit is = (R1 + 2k) || 5k 1000uF = 4s. The final voltage across the capacitor is by
1.5 0.1
voltage division, 0.1V. Thus using the elapsed time formula t1 = 4ln
= 3.39s .
0.7 0.1

1st Order Circuit Probs 11/26/01

P8-47

-3

SOLUTION 8.63. (a) Since RC = 10 s, from equation 8.25, v a (t) = RC

R. A. DeCarlo, P. M. Lin

dv s (t)
= cos(1000t) V.
dt

Rf
dv a (t)
== sin(1000t) V, and vout (t) =
v (t) = sin(1000t) = v s(t) V.
dt
R b
(b) With the switch moved to position B, there is no source in the circuit. But the output at the
switching instant is sin(1000t) V which coincides with vs(t). Hence, the input to the first amp remains
Hence vb (t) = RC

the same and the circuit continues to produce vout (t) = sin(1000t) V, i.e., the circuit becomes an
oscillator.

*S OLUTION 8.64. Before attacking the problem proper, consider driving an ideal unity gain integrator
with the square wave of figure P8.64b. If we start the integration when the square wave goes positive,
then we have a triangular waveform as follows:

On the other hand, if we start the integration when the square wave goes negative, we get the following
waveform

One concludes that without some further physical assumptions, there is no unique solution to this
problem.
Physically speaking all capacitors have a leakage resistance. Hence, in modeling the capacitor we
put a very large resistance in parallel with an ideal C, producing a nearly ideal leaky integrator circuit.
The leaky integrator circuit has a first order response. Hence over time, when the circuit reaches steady
state, the dc level of the resulting output waveform will be proportional to the average level of

1st Order Circuit Probs 11/26/01

P8-48

R. A. DeCarlo, P. M. Lin

the square wave which is zero in this case.

Comments: Actually the proportionality constant in the above statement is the overall dc gain of the
integrator-inverter. See the formulas given in P22.16. Adding up the two formulas, we have
Output(t)max + Output(t) min = K (Vmax + V min )
and K = H(0), i.e., the dc gain of the first order low pass system. This leads to
Average of output = (dc gain) (average of input levels)
See the analysis in example 8.7 and later an exact analysis is given in problems 22.15 and 22.16. In
other words, one would expect that the output of our (leaky) integrator in steady state to be given by the
waveform below.

Now we can start to solve the problem. The first part is to design a (leaky) integrator circuit to
produce a triangular waveform of value 2 V peak-to-peak. For this we consider the following figure
which consists of the leaky integrator followed by an inverter.

1st Order Circuit Probs 11/26/01

P8-49

R. A. DeCarlo, P. M. Lin

To handle this analysis, recall that i q/t in which case q = Cv. Hence, to have a peak-to-peak
voltage at v2 (t) and v3 (t) of 2 V, we require that
i t 9 50 10 6
v = in =

=2
C
R1
C
Hence R1 C = 2.2510-4. If we choose R1 = 10 k, then C = 22.5 nF. At this point the waveform of
v3 (t) is given below.

In order to complete the design, we must raise the portion of the curve with positive slope by 1 V and
lower the portion with negative slope by 1 V. This can be done by adding one-ninth of vin (t) to v3 (t).
This can be done by using the following circuit. In this circuit, there is a voltage-divider at the non-

1st Order Circuit Probs 11/26/01

P8-50

R. A. DeCarlo, P. M. Lin

inverting terminal of the second op amp. Here V+ equals one-eighteenth of vin (t). However the gain of
the non-inverting portion is 2; therefore one-ninth of the input is added to v3 (t) as desired.

CHAPTER 9 PROBLEM SOLUTIONS

SOLUTION TO PROBLEM 9.1. If we can compute expressions for K and q that are real,
then these quantities exist by construction. Consider that A, B, K and q must satisfy the
following relationship:
K cos(t + ) = ( K cos()) cos(t) + (K sin())sin(t) Acos(t) + Bsin(t)
Therefore K cos() = A and K sin() = B . Consequently,

( K cos())2 + ( Ksin())2 = K 2 = A2 + B2
in which case K =

A 2 + B2 . Further,
K sin()
B
= tan() =
K cos()
A

in which case
= tan1

B
A

with due regard to quadrant.

SOLUTION TO PROBLEM 9.2. For the inductor,

2
1
C 1
CV02
1
2
1
WL (t) = 2 L iL (t) = LVo
sin
t =
sin2
t

LC
2
L
LC
2
and for the capacitor,
2

CV 2
1
1
WC (t) = 12 C vC2 (t) = 12 C Vo cos
t = 0 cos2
t .
LC
LC
2

Hence,
CV02 2 1
CV02
2
2
2 1
1
1
WC + WL = 2 C vC (t) + 2 L iL (t) =
sin
t + cos
t =
LC
LC
2
2
SOLUTION TO PROBLEM 9.3. Since x(t) = (K1 + K2 t)e t ,
x' (t) = K1e t + K2e t t K2e t
and
x' ' t)( = 2 K1e t K
2e t K
2e t + 2 t K2 e t
Substituting into the differential equation, we have
2 K1e t 2 K2e t + 2 t K 2e t

+ 2 K1e t + K2 e t t K2 e t + 2 K1e t + K2tet = 0

This means that the solution form satisfies the differential equation.

SOLUTION TO PROBLEM 9.4.

(a) Suppose x(T ) = 0 at some T. Then K1e s1T = K2e s2 T . Since e siT > 0 whenever si
is real and T is finite, K1 & K2 must have opposite signs.
(b) For this we solve for T and show there can only be one solution. Since
K1e s1T = K2e s2 T and e siT > 0 ,
es2 T
K
K1
es2 T
= s T implies ln 1 = ln s T = ( s2 s1 )T
K2 e 1
K2
e 1
Hence the unique solution is given by
K
T = ln 1
K2

(s2 s1 )

provided s2 s1 which is the case for distinct roots.

SOLUTION TO PROBLEM 9.5. Suppose x(T ) = 0 at some T >0. This is true if and only if
(*)
K1e s1T = K2Te s1T
s1T
Since e > 0 and T > 0, (*) is true if and only if K1 = K2T which is true if and only if
K1 & K2 have opposite signs.
SOLUTION 9.6. (a) Denote one period of oscillation by T. Then by definition
9950T = 2. Hence, T = 0.63148 ms. The time constant of decay is 1 ms. Therefore,
NT = N 0 . 6 3 1 4 8 =. 1Hence N = 1.5836 cycles.
2 1
(b) Here observe that the time constant of decay is 1/ s. Hence NT = N
= . One
d

concludes that N = d .
2
SOLUTION 9.7. The differential equation for the capacitor voltage is
d 2 vC (t)
1
1
+
vC (t) =
V
2
LC
LC s
dt
1
For t > 0, the characteristic equation is s2 +
= 0 . Hence from table 9.2, the solution
LC
for either the inductor current or capacitor voltage has the general form

vC (t) = Acos( d t) + Bsin( d t) + X F

1
V
1
. From table 9.2, X F = s
= Vs . Further,

LC
LC
LC
vC (0 + ) = A + X F = A + Vs = 0
Hence A = Vs . Also,
where d =

v'C (0 + ) = d B =
Hence B = 0. Therefore

1
i (0 + )
B= L
=0
LC
C

t
vC (t) = Vs cos( d t) + Vs = Vs 1 cos
V
LC

To obtain the expression for iL(t) (= iC(t)), we can either repeat the above derivation or
differentiate and multiply by C. We choose the latter. Therefore
iL (t) =

CVs
V
t
t
sin
= s sin
A
LC
L
LC LC
C

SOLUTION TO PROBLEM 9.8. Essentially this is example 9.7, case 1, with literals and R =
. Clearly the circuit is a driven parallel LC circuit having characteristic equation
1
j
j

s2 +
= s +
s
= (s + j d )(s j d ) = 0
LC
LC
LC
Thus we obtain
1
1
iL (t) = Acos( d t) + Bsin( d t) + X F = Acos
t +Bsin
t +X F
LC
LC
Here X F = I S is the value of the current when the inductor is shorted and the capacitor is
open. Applying the initial conditions,
iL (0+) = A + Is = 0 A = Is
Further
v (0+) vC (0+)
i'L (0+) = L
=
=0=
L
L

1
B
LC

because capacitor voltage is continuous and because

i'L (0+) =

1
1
Acos
t + B sin
t + XF
=

dt
LC
LC
t =0

Hence B = 0. Therefore,

B
LC

iL (t) = Is 1 cos(
t)
LC

Rather than repeat the above derivation,

vC (t) = vL (t) = L

diL (t)
d
= LIs
dt
dt

L 1
1
1 cos LC t = Is C sin LC t

SOLUTION TO PROBLEM 9.9. Observe that the circuits of figures (a) and (b) are dual
circuits. Hence the numerical value of vout(t) and iout(t) are the same for the same
excitation. Since the circuit is linear, when the excitation is doubled, the response is
doubled (given zero initial conditions) by linearity. Therefore, iout (t) = 2g(t) .

SOLUTION TO PROBLEM 9.10.

(a) The initial conditions are:

vc (0) = 0 = vc ( 0+)
iL (0) = 10 / 0.5 = 20 = i L ( 0+)

(b)
1
L i 2L ( 0 ) = 0.5 J
2
WC (0) = 0

WL ( 0 )=

(c) Maximum value of vC when all energy is in capacitor must satisfy

1
2
C vC,max
= 0.5
2

or equivalently vC,max = 1000 V.

(d) From the text development, the parallel LC circuit has the differential equation,
d 2 vC
1
2 + LC vC = 0
dt
The solution form is:
1
1
vC (t) = A cos
t +B sin
t
LC
LC
The initial conditions are:

vC ( 0 ) = A = 0

and
vC ' ( 0+) =

1
i (0+)
i ( 0+)
B= C
= L
LC
C
C

Hence
B = 1000
Thus

 1
vC (t) = 1000 sin
t V
LC

SOLUTION TO PROBLEM 9.11.

The switch has been closed for a long time which means that the inductor acts like a short
and the capacitor like an open. Hence, at t = 0 , vL = 0 and iC = 0. Hence Is divides
equally between the resistors, i.e.,
iL (0) = i L ( 0+) = Is / 2 and vC (0) = vC ( 0+) = 0
For t 0, the differential equation is
d 2 vC
1
2 + LC vC = 0
dt
with corresponding response
1
1
vC (t) = A cos
t +B sin
t
LC
LC
Evaluating at the initial conditions,
vC ( 0 ) = A = 0
and
vC ' ( 0 )= B

1
i (0+) iL (0+)
= C
=
LC
C
C

in which case
I
B= S
2

L
C

Therefore,
I
v C (t) = s
2

L 1
sin
t
C LC

SOLUTION TO PROBLEM 9.12.

Natural frequency is

1
= 5000 2 . Hence,
LC

C = 100e-9;
L = 1/((10e3 *pi)^2 *C)
L=
1.0132e-02

By voltage divider,

vC (0) = vC ( 0+) = 20 mV

Current through L,

iL (0) = i L ( 0+) = 0

Voltage across capacitor satisfies,

1
1
vC (t) = Acos
t + B sin
t
LC
LC
Using the ICs,
vC (0+) = A = 20 mV
and
1
i (0+)
i ( 0+)
= C
= L
=0
LC
C
C

vC ' ( 0 )= B
Hence

vC (t) = 20 cos(10,000t ) mV

SOLUTION TO PROBLEM 9.13. By definition

=

1
= 2 40
LC

in which case
C = 0.1e-3;
w = 2*pi*40;
L = 1/(w^2 *C)
L=
1.5831e-01 (rad/s).
Observe that

iL (0) = i L ( 0+) = 1A

From the given circuit, the capacitor is never connected to a source. Therefore,
vC (0) = vC ( 0+) = 0. Also, since
d 2 iL
1
2 + LC iL = 0
dt
it follows that
From the initial conditions
and

iL (t) = Acos(t) + Bsin(t)

iL ( 0 )= A = 1

v (0 + )
iL ' (0 + ) = B = C
=0
L
iL (t) = cos(80t) A

SOLUTION TO PROBLEM 9.14.

(a) From the continuity property, the capacitor voltage and inductor current remain the
same,
vC (0) = vC ( 0+) = 5 V
iL (0) = i L ( 0+) = 1
However, iC (0) = 0 but iC (0+ ) = iL (0+) = 1 A and v L (0) = 0 but v L (0+) = 5 V.
These values change to maintain satisfaction of KVL and KCL.
(b)
1
1
vC (t) = Acos
t + B sin
t
LC
LC
where

vC (0+) = 5 = A
1
i (0+)
1

B=
vC ' ( 0+) = B
= C
=
LC
C
C
2
vC (t) = 5cos(2t)

sin(2t) V
2

Alternately, from equation 9.4,

vC (t) = K cos(2t + )
and from equation 9.17b K =
A = 5; B = -pi/2;
K = sqrt(A^2 + B^2)
K=
5.2409e+00
theta = atan2(-B,A)
theta =
3.0440e-01
thetadeg = theta*180/pi
thetadeg =
1.7441e+01

A 2 + B2 and = tan1

B
in which case
A

Hence, K = 5.24 and = 17o.

(c)
6

Capacitor Voltage (V)

-2

TextEnd

-4

-6

0.2

0.4

0.6

0.8

1
1.2
Time in s

SOLUTION TO PROBLEM 9.15. As before, =

1.4

1.6

1.8

1
= 2 40 in which case L = 0.158
LC

H.
Now,
vC (0) = vc (0+) =
and

100
25 = 20 mV
125

iL (0) = i L ( 0+) = 10 mA

The solution form is:

iL (t) = Acos(80t) + Bsin(80t)

where

iL (0+) = A = 10 mA

and
v ( 0+) vC (0+)
iL ' (0+) = L
=
= 80B B = 0.50265 mA
L
L

SOLUTION TO PROBLEM 9.16.

(a) Note: vC (0) = vC ( 0+) = 1 V and iL ( 0 )= 5 A. The characteristic equation is:

s2 +

1
1
s+
= s2 + 5s + 4 = (s + 1)(s + 4 )= 0
RC
LC

As stated, the circuit is overdamped. Hence,

vC (t) = K1e s1t + K2 es2 t = K1e t + K2e 4t
vC (0+) = K1 + K2 = 1
i (0+)
vC ' ( 0+) = s1K1 + s2 K2 = K1 4K2 = C
=0
C
K1 =
Finally,

s2
s1
= 1.3333 and K2 =
= 0.3333 .
s2 s1
s2 s1
vC (t) = 1.3333e t + 0.3333e4t V

Using MATLAB to plot:

t = 0:.02:4.5;
vc = -1.33333*exp(-t) + 0.33333*exp(-4*t);
plot(t,vc)
grid
ylabel('Capacitor Voltage in V')
xlabel('Time in s')
0
-0.1

Capacitor Voltage in V

-0.2
-0.3
-0.4
-0.5
-0.6
-0.7

TextEnd

-0.8
-0.9
-1

(b)

0.5

1.5

2
2.5
Time in s

K1 + K2 = 1

3.5

4.5

i (0+) 10
vC ' ( 0+) = s1K1 + s2 K2 = K1 4K2 = C
=
= 10
C
C
Hence,
A = [1 1;-1 -4]
A=
1 1
-1 -4
b = [1;-10]
b=
1
-10
K = A\b
K=
-2
3
yielding

vC (t) = 2e t + 3e 4t V
1

0.8

Capacitor Voltage in V

0.6
0.4
0.2
0
-0.2
-0.4

TextEnd

-0.6
-0.8
-1

0.5

1.5

2
2.5
Time in s

Obviously, there is only one zero crossing.

SOLUTION TO PROBLEM 9.17.

(a) First,
100
30 = 10 V
300
iL (0) = i L ( 0+) = 0.1 A

vC (0) = vC ( 0+) =

3.5

4.5

Clearly, iC (0) = 0 and v L (0) = 0 . However,

v (0+)
iL (0+) + iC (0+) + C
= 0.1 + iC (0+) + 0.15 = 0
66.667
Hence, iC (0+ ) = 0.25 A. Further, v L (0+) = 10 V.
(b) The characteristic equation is:
s2 +

1
1
s+
=0
RC
LC

From MATLAB,
R = 66.667;C = 25e-6; L = 0.5;
b = 1/(R*C)
b = 6.0000e+02
c = 1/(L*C)
c=
80000
si = roots([1 b c])
si = -3.9999e+02
-2.0000e+02
We take the roots to be: s1 = 400 and s2 = 200 .
(c) Overdamped response implies,
vC (t) = K1e s1t + K2 es2 t = K1e 400t + K2e 200t
(d)

vC (0+) = K1 + K2 = 10

i (0+)
vC ' ( 0+) = s1K1 + s2 K2 = 400K1 200K2 = C
= 10 4
C
A = [1 1;-400 -200];
b = [10;-1e4];
K = A\b
K=
40
-30
Finally,

vC (t) = 40e 400t 30e200t V

(e)
t=0:0.01e-3:25e-3;
vc = 40*exp(-400*t) - 30*exp(-200*t);
plot(t,vc)
grid
ylabel('Capacitor Voltage in V')
xlabel('Time in s')

10

Capacitor Voltage in V

8
6
4
2
0
TextEnd

-2
-4
-6

0.005

0.01

0.015

0.02

0.025

Time in s

SOLUTION TO PROBLEM 9.18.

(a)

iL (0) = i L ( 0+) = 0
vC (0) = vC (0+) = 5 V

At t = 0+, the circuit is a series RLC with R = 12.5 , L = 2.5 H, and C=0.1 F. The
resulting characteristic polynomial is:

s2 +

R
1
s+
= s2 + 5s + 4 = (s + 4)(s + 1) = 0
L
LC

Hence, s1,s2 = 4,1 and the form of the response is:

vC (t) = K1e 4t + K 2e t
At t = 0+,

vC (0+) = K1 + K2 = 5

and
i (0+) iL (0+)
vC '(0) = 4K1 K2 = C
=
=0
C
C
Solve for K1 and K2 we obtain:
A = [1 1; -4 -1];
b = [5;0];

K = A\b
K=
-1.6667e+00
6.6667e+00
Hence,

vC (t) = 1.66667e 4t + 6.66667e t V

(b)

iL (0+) = i L (0) = 0
Since, the stable (passive) circuit contains no source for t > 0, all initial energy is
absorbed by the resistor. Hence limt iL (t) = 0 , i.e., iL () = 0 .
iL (t) = iC (t) = C

dvC 2 4t 2 t
= e
e A
dt
3
3

SOLUTION TO PROBLEM 9.19. This circuit is a series RLC in which case the
R
1
characteristic equation is always: s2 + s +
=0
L
LC
(a) For this time period, R = 2 k in which case the characteristic equation is found and
solved in MATLAB as follows:
R = 2e3; L = 0.1; C = 0.1e-6;
b = R/L
b=
20000
c = 1/(L*C)
c=
100000000
s12 = roots([1 b c])
s12 =
-10000
-10000

The roots are repeated and: s1, s2 = 10,000. The form of the response is:
iL (t) = (K1 + K2t)e 10000t
To find K1 and K2:
iL (0+) = K1 = 2.5
and

v (0+) vC (0+) RiL (0+) 6 2000 2.5

iL '(0+) = 25000 + K 2 = L
=
=
= 49940
L
L
0.1
K2 = 24940
Thus

iL (t) = (2.5 24940t)e 10000t

0 t 0.1m sec

(b) After the switch closes, R = 1 k,

iL (0.1 10 3) = (2.5 24940 0.1 10 3 )e 1 = 0.0022073
and

iL '(10 3) = 25000e 1 24940e 1 + 24940e 1 = 9197

The new characteristic equation is computed as follows:

R = 1e3; L = 0.1; C = 0.1e-6;b = R/L
b=
10000
c = 1/(L*C)
c = 100000000
s12new = roots([1 b c])
s12new =
-5.0000e+03 + 8.6603e+03i
-5.0000e+03 - 8.6603e+03i
Hence s1,s2 = 5000 j8660.25. The form of the new solution is:
iL (t') = e 5000t' [ Acos( t') + Bsin( t') ] A
where t'= t 0.1 10 3 and

= 8660.25 rad/sec. Observe that

iL (t'= 0) = 0.0022073 = A

and

iL '( t'= 0) = 5000A + 8660.25B = 9197

From MATLAB,
w = imag(s12new(1));
sig = real(s12new(1));
B = (-9197 -sig*2.2073e-3)/w
B=
-1.0607e+00
Hence for t > 0.1 ms,
iL (t) = e 5000(t 0.1ms) [0.0022073cos( (t 0.1ms)) 1.0607sin ( (t 0.1ms))] A
(c)

2.5
2

iL(t), in A

1.5
1
0.5

TextEnd

0
-0.5
-1

0.5

1.5

2
time in s

2.5

3.5

K1 = 2.5;
K2 = -24940;
t = 0:0.01e-3:4e-3;
A = 0.0022073;
B = (-9197 -sig*2.2073e-3)/w
B = -1.0607e+00
iL = (K1 + K2*t) .* exp(-10000*t) .* (u(t)-u(t-1e-4)) ...
+ exp(-5000*(t - 1e-4)) .* (A*cos(w*(t - 1e-4))+B*sin(w*(t - 1e-4))) ...
.* u(t -1e-4);
plot(t,iL)
grid
iL = (K1 + K2*t) .* exp(-10000*t) .* (u(t)-u(t-1e-4))
(d)
K1 = 2.5;
K2 = -24940;
t = 0.1e-3;
iL = (K1 + K2*t) .* exp(-10000*t)
iL = 2.2073e-03
% The energy stored in the inductor over [0,0.1ms] is in J:
WL = 0.5*0.1*(iL^2 - 2.5^2)
WL =
-3.1250e-01
% The energy stored in the capacitor over [0,0.1ms] first
% requires computation of vL and then vC.
vL = 0.1*(K2*exp(-10000*t) - 10000*(K1 + K2*t) .* exp(-10000*t))
vL =
-9.1970e+02
vC = vL + 2000*iL
vC =

4
x 10

-3

-9.1528e+02
% The energy stored in the capacitor over [0,0.1ms] is in J:
WC = 0.5*0.1e-6*(vC^2 - 6^2)
WC =
4.1885e-02
% To compute energy dissipated in resistor, we make
% use of conservation of energy: WR + WC + WL = 0
WR = -WL - WC
WR =
2.7061e-01

SOLUTION TO PROBLEM 9.20. For this problem, iL (0) = 8 A, vC (0) = 20 V. For the
1
1
parallel RLC, the characteristic equation s2 +
s+
= 0 is solved as follows:
RC
LC
R = 20; C = 0.01e-3; L = 25e-3;
si = roots([1 1/(R*C) 1/(L*C)])
si =
-4.0000e+03
-1.0000e+03
s1 = si(1); s2 = si(2);
Hence

vC (t) = K1e 4000t + K2e 1000t

To compute the constants,

vC (0) = K1 + K 2 = 20
and
iL (0+) +
i (0+)
vC '(0) = 4000K1 1000K 2 = C
=
C
C
iL0 = 8; vC0 = 20;
vCprime = -(iL0 +vC0/R)/C
vCprime =
-9.0000e+05
A = [1 1;-4000 -1000];
b = [20;vCprime];
K = A\b
K=
2.9333e+02
-2.7333e+02

v C (0+)
R

Hence,

vC (t) = 293.33e 4000t 273.33e 1000t V

Also,
v (t)
iL (t) = C CvC' (t) = 2.933e 4000t + 10.933e 1000t A
R
where
K1 = K(1); K2 = K(2);
KR1 = K1/R; KR2 = K2/R;
KCp1 = -4000*K1*C; KCp2 = -1000*K2*C;
KL1 = -KR1-KCp1
KL1 =
-2.9333e+00
KL2 = -KR2 - KCp2
KL2 =
1.0933e+01
t = 0:0.01e-3:4e-3;
iL = KL1*exp(-4000*t) + KL2*exp(-1000*t);
plot(t,iL)
grid
xlabel('Time in s')
ylabel('Inductor Current in A')

9
8

Inductor Current in A

7
6
5
4
3

TextEnd

2
1
0

0.5

1.5

2
Time in s

2.5

3.5

SOLUTION TO PROBLEM 9.21.

(a) The circuit is critically damped and the characteristic polynomial is:

4
x 10

-3

s2 +

1
1
s+
= (s + 20)2 = s2 + 40s + 400 = 0
RC
LC

R = 2; C = 1/80;
L = 1/(400*C)
L=
2.0000e-01
i (0+ ) iL (0 + ) vC (0 + ) / 2
(b) vC (0) = 10 V and vC '(0 + ) = 800 = C
.
=
C
C
vC0 = 10; vCp0 = -20*10 -600
vCp0 =
-800
iL0 = C*800-vC0/2
iL0 =
5
Hence: iL(0-) = iL(0+) = 5 A.
(c)
By simple KCL,
iL (t) =

1 dv C v C

= 5e 20t + 150 t e 20t A

80 dt
2

SOLUTION TO PROBLEM 9.22.

(a) The characteristic equation for the series RLC
s2 +

Req
L

s+

1
=0
LC

For critically damped response, want (Req / L)2 4 / (LC ) = 0 . Solving yields Req = 4
. Hence,
5R
Req =
5+R
implies that R = 20 .
(b) Solving for the resulting roots implies that
si = roots([1 Req/L 1/(L*C)])
si =
-50
-50
s1 = 50;
Hence

vC (t) = (K1 + K 2t)e 50t

From the initial conditions, vC (0) = K1 = 5 and
i (0+) iL (0+)
5
vC '(0) = 250 + K2 = C
=
=
= 500
C
C
0.01
Hence, K2 = 250 and
vC (t) = (5 250t)e 50t V
Set to zero and solve for t = 0.02. See MATLAB plot.
5

-1

0.005

0.01

0.015

0.02

0.025

0.03

0.035

SOLUTION TO PROBLEM 9.23.

(a) The circuit is a series RLC. Hence
R = 40; C = 0.25e-3; L = 0.1;
si = roots([1 R/L 1/(L*C)])
si =
-200
-200
s1 = si(1);
Since the circuit is critically damped,
iL (t) = (K1 + K 2t)e 200t A
Using initial conditions,

0.04

0.045

0.05

iL (0) = K1 = 1
v (0+) vC (0+) 40i L (0+)
iL '(0) = s1K1 + K2 = 200K1 + K 2 = L
=
= 350
L
L
Hence

K 2 = 150

Thus

iL (t) = (1 150t)e 200t A

Now using MATLAB,

R/L
ans = 400
1/(L*C)
ans =
40000
y = dsolve('D2y + 400*Dy + 40000*y = 0, y(0) = 1,Dy(0) = -350')
y=
exp(-200*t)-150*exp(-200*t)*t
This answer coincides with the analytical solution.
(b) As above, the form of the solution is
vC (t) = (K1 + K 2t)e 50t
Applying initial conditions,
vC (0) = K1 = 5
i (0+) iL (0+)
vC '(0) = s1K1 + K2 = 200K1 + K 2 = C
=
= 4000
C
C
Hence, K 2 = 5000 and

vC (t) = (5 + 5000t)e 200t V

In MATLAB,
y = dsolve('D2y + 400*Dy + 40000*y = 0, y(0) = 5,Dy(0) = 4000')
y=
5*exp(-200*t)+5000*exp(-200*t)*t
which verifies the analytical solution. Plotting we obtain,

12

10

0.005

0.01

0.015

0.02

0.025

0.03

0.035

0.04

Observe, there is no zero crossing as per problem 9.5.

SOLUTION TO PROBLEM 9.24.

Source has been on for a long time and turns off at t = 0. Hence
iL (0) = iL (0+) = 0
vC (0) = vC (0+) = 0.1 40 = 4 V
At t 0, we have a series RLC circuit:
R = 40; C = 4e-3; L = 2;
si = roots([1 R/L 1/(L*C)])
si =
-1.0000e+01 + 5.0000e+00i
-1.0000e+01 - 5.0000e+00i
wd = imag(si(1))
wd =
5
sig = -real(si(1))
sig =
10
Hence

0.045

0.05

vC (t) = e 10t ( Acos(5t) + B sin(5t))

Applying initial conditions,
vC (0) = A = 4
i (0+) iL (0+)
vC '(0) = 10A + 5B = C
=
=0
C
C
Hence, B = 8. The final form is:
vC (t) = e 10t (4 c o s ( 5t) + 8sin(5 t)) V
Similarly,

iL (t) = e 10t ( A cos(5t) + Bsin(5t)) A

Applying initial conditions,
iL (0) = A = 0
v (0+) v C (0+) 40iL (0+)
iL '(0) = 10A + 5B = L
=
= 2
L
L
Thus

iL (t) = 0.4e 10t sin(5t) A

SOLUTION TO PROBLEM 9.25.

-

At t = 0 , the capacitor is an open circuit and the inductor is a short circuit. The
resulting circuit is a simple resistive network. The first step in the solution is to solve this
network for the initial conditions on the capacitor and inductor. Specifically, solve for
the capacitor voltage (i.e. the voltage across the series connection of the 6 resistor and
the independent voltage source) and the inductor current (i.e. the current flowing through
the 4 resistor).
Verify that iL (0) = iL (0+) = 1 A and vC (0) =v C (0+) = 12 V.
When the switch opens, the branch containing the independent voltage source is
eliminated. So, we end up with a series RLC circuit. The equivalent resistance is
Req = 4 + 24 / / 4 8= 20 .
R = 20; C = 0.01; L = 2;si = roots([1 R/L 1/(L*C)])
si =
-5.0000e+00 + 5.0000e+00i
-5.0000e+00 - 5.0000e+00i

sig = -real(si(1))
sig =
5
wd = imag(si(1))
wd =
5

iL (t) = e 5t [ A cos(5t) + Bsin(5t)] A

Applying ICs,
iL (0) = A = 1
v (0+) 8
iL '(0) = 5A + 5B = L
=
= 4
L
2
Hence B = 0.2. It follows that
1
iL (t) = e5t cos(5t) + sin(5t) A

SOLUTION TO PROBLEM 9.26. The series RLC circuit has characteristic equation
s2 +

R
1
10
1
s+
= s2 + s +
= s2 + 2 s + 2 + 2d = 0
L
LC
L
LC

From the given response,

= 10 =

10
which implies that L = 0.5 H. Further,
2L

2
1
2
= = 2 + d2 = 10 2 + 10 3 = 400
LC C

Hence, C = 5 mF.
v (0)
Now, from given response, iL (0) = 0 and iL '(0) = 500 3 = L
= 2v L (0) .
L
Hence
v L (0) = 250 3 V. In addition, vC (0) = 10iL (0) v L (0) = 250 3 V and
vC '(0) = iL (0) / C = 0
To find the capacitor voltage we have

vC (t) = e 10t Acos(10 3t) + B sin(10 3t)

It follows that vC (0) = 250 3 = A and

vC '(0) = 0 = 2500 3 + 10 3 B
B = 250. Therefore

vC (t) = e 10t 250 3 cos(10 3t) + 250sin(10 3t)

SOLUTION TO PROBLEM 9.27. (a) From the given information,

vC (t) = Ae t cos( t + ) V
2
where = 2 f = 2 / T =
= 4000 rad/s. f = 2 kHz.
0.5 10 3
The time constant of the exponential term is 1/ which is 2 ms from the figure.
Hence = 500 . The characteristic equation of the parallel RLC is:
s2 +
Since

1
1
1
1
s+
= s2 + 4 s +
= s2 + 2 s + 2 + 2d = 0
RC
LC
LC
10 C

1
= 2 = 1000, C = 0.1 F. Further,
10 4 C
1
107
=
= 2 + d2 = 10 6 + 16 2106
LC
L

implies that
L = 1e7/(1e6 + 16*pi^2 *1e6) or L = 63 mH.
(b) From the figure and the above calculations, vC(0) = 10 V and
vC (t) = 10e 500t cos(4000 t) V,
Hence

vC '(t) = 5000e 500t cos(4000 t) 40000 e 500t sin(4000 t)

implying that vC' (0) = 5000 . Thus
iL (0) = vC (0) /10 4 CvC' (0) = 0.001 + 5000 0.1 106 = 0.5 10 3 A

SOLUTION TO PROBLEM 9.28. As given vC (t) = Ae3t cos(4t + ) and R = 10 . The

characteristic polynomial of the parallel RLC circuit is:
1
1
1
1
s2 +
s+
= s2 +
s+
= s2 + 2 s + 2 + 2d = 0
RC
LC
10C
LC

Hence 1/(RC) = 6 implies C = 1/60 F. Further

4 60
36
L
= 4 . Hence L = 2.4
2

H.
To change R to obtain a cricially damped circuit,
2
4
60
=
= 100
R
LC
Hence R2 = 36 or R = 6 . It follows that 2 = 10 or = 5. The form of the response is:

vC (t) = (K1 + K 2t)e 5t

SOLUTION TO PROBLEM 9.29. For all cases, vC(0-)= vC(0+) = 0 and iL(0-)= iL(0+) =
20/20 = 1 A. Further for all cases the circuit is a parallel RLC with characteristic
equation:
s2 +

1
1
s+
= s2 + 2 s + 2 + d2 = 0
RC
LC

(a)
L = 2e-3; C = 5e-6;
c = 1/(L*C);
R = 10;
b = 1/(R*C);
si = roots([1 b c])
si =
-10000
-10000
% Solution is cricitally damped.
Thus
4
vC (t) = (K1 + K 2t)e s1t = (K1 + K2 t)e 10 t V

From ICs,

vC (0) = K1 = 0
i (0+) iL (0+)
vC '(0) = s1K1 + K2 = K 2 = C
=
= 2 105
C
C

Hence
4
vC (t) = 2 10 5 te10 t V

(b)
R = 100;
si = roots([1 1/(R*C) 1/(L*C)])

si =
-1.0000e+03 + 9.9499e+03i
-1.0000e+03 - 9.9499e+03i
vC (t) = e 1000t ( Acos(9950t) + B sin(9950t)) V
From ICs.

vC (0+) = A = 0

i (0+) i L (0+)
vC '(0) = 1000A + 9950B = C
=
= 2 10 5
C
C
in which case B = 20.1. Thus
vC (t) = 20.1e 1000t sin(9950t) V
(c)
R = 87/17;si = roots([1 1/(R*C) 1/(L*C)])
si =
-3.6328e+04
-2.7527e+03
We define s1,s2 = 2753, 36327. Thus
vC (t) = K1e s1t + K 2e s2t = K1e 2753t + K 2e 36327t V
From the IC's

vC (0) = K1 + K 2 = 0
vC '(0) = s1K1 + s2K 2 = 2 105

A = [1 1;si(2) si(1)];
b = [0; -2e5];
K = A\b
K=
-5.9568e+00
5.9568e+00
Therefore

vC (t) = 5.9568 e 2753t e 36327t V

SOLUTION TO PROBLEM 9.30. For all cases, vC(0-) = vC(0+) = 0 and iL(0-) = iL(0+) =
10/100 = 0.1 A. Further for all cases the circuit is a parallel RLC with characteristic
equation:
1
1
s2 +
s+
= s2 + 2 s + 2 + d2 = 0
RC
LC

(a)
R = 50; C = 0.04e-3; L = 0.625;
si = roots([1 1/(R*C) 1/(L*C)])
si =
-400
-100
Define the two roots as:
s1 = 100
s2 = 400
Hence,

vC (t) = K1e 100t + K2e 400t V

From, the initial conditions

vC(0+) = 0 = K1 + K2
and
vC '(0) = s1K1 + s2K 2 = 100K1 400K2 =

iL (0+)
= 2500
C

A = [1 1;si(2) si(1)];
b = [0;-2500];
K = A\b
K=
-8.3333e+00
8.3333e+00
Therefore

vC (t) = 8.3333 e 100t e 400t V

(b)
L = 0.4;
si = roots([1 1/(R*C) 1/(L*C)])
si =
-2.5000e+02
-2.5000e+02
Thus
From IC's,

vC (t) = (K1 + K 2t)e 250t V

vC (0) = K1 = 0
i (0+) iL (0+)
vC '(0) = s1K1 + K2 = K 2 = C
=
= 2500
C
C

Therefore

vC (t) = 2500te250t V

(c)
L = 0.2;
si = roots([1 1/(R*C) 1/(L*C)])
si =
-2.5000e+02 + 2.5000e+02i
-2.5000e+02 - 2.5000e+02i
vC (t) = e 250t [A cos(250t) + Bsin(250t)] V
From ICs.

vC (0+) = A = 0

i (0+) i L (0+)
vC '(0) = 250A + 250B = 250B = C
=
= 2500
C
C
in which case B = 10. Thus
vC (t) = 10e 250t sin(250t) V

SOLUTION TO PROBLEM 9.31.

(a) The indicated behavior occurs when the resistance causes the circuit to be critically
damped, i.e.,
2
4 4 2
1

=
16 = 0
RC
LC R
Thus R = 1 and
R = 1 = R0 + e t 5 = 0.8 + e t 5
So
t = 5+log(1-0.8)
t=
3.3906e+00 (years)
(b) Here, we have a series case: the indicated behavior occurs when the resistance
causes the circuit to be critically damped, i.e.,
2
4
R
2
L LC = R 144 = 0
Thus R = 12 and
R0
15
R = 12 =
=
t
5
1+e
1 + e t 5
t = 5 + log((15-12)/12)
t=
3.6137e+00 (years)

SOLUTION TO PROBLEM 9.32. Step 1: Since the step functions are 0 from t = up to t =
0-,
vC(0-) = vC(0+) = 0, iL(0-) = iL(0+) = 0
Step 2: At t = 0+, we have
v (0+)
iC (0+) = 2 C
iL (0+) = 2 A
20
and

v L (0+) = v C (0+) 5iL (0+) 50 = 50 V

SOLUTION TO PROBLEM 9.33. Since the step function is 0 from t = up to t = 0-,

vC(0-) = vC(0+) = 0, iL(0-) = iL(0+) =0
Since the circuit is a parallel RLC
s2 +

1
1
s+
= s2 + 100s + 1600 = 0
RC
LC

R = 4; L = 0.25; C = 2.5e-3;
si = roots([1 1/(R*C) 1/(L*C)])
si =
-80
-20
Hence s1,s2 = -20, -80. The general form is:
vC (t) = K1e 80t + K2e 20t + X f V
When the capacitor is open and the inductor is a short, Xf = 0. Thus,
vC (t) = K1e 80t + K2e 20t V
From the ICs
vC(0+) = 0 = K1 + K2
and

i (0+) iL (0+) v C (0+) 8 + (20 vC (0+)) 8

vC '(0) = 80K1 20K 2 = C
=
C
C
2.5
=
= 1000
2.5 10 3

A = [1 1;-80 -20];
b = [0; 1000];
K = A\b
K=

-1.6667e+01
1.6667e+01
Hence,

vC (t) = 16.667 e 80t e 20t V

SOLUTION TO PROBLEM 9.34. From the continuity property and the fact that at t = 0-,
the capacitor looks like an open and the inductor looks like a short at t = 0-,
iL(0-) = iL(0+) = 1A
vC(0-) = vC(0+) = 0
Since the circuit is a parallel RLC
s2 +

1
1
s+
= s2 + 500s + 40000 = 0
RC
LC

R = 40; C = 0.05e-3;L = 0.5;

si = roots([1 1/(R*C) 1/(L*C)])
si =
-400
-100
Define the two roots as:
s1 = 100
s2 = 400
Hence, the general form is:
iL (t) = K1e 100t + K2e 400t + X f
When the capacitor is open and the inductor is a short, Xf = 1 A. Thus,
iL (t) = K1e 100t + K2e 400t 1
From, the initial conditions

iL (0+) = K1 + K2 1 = 1

and
v (0+) vC (0+)
iL '(0) = 100K1 400K2 = L
=
=0
L
L
A = [1 1;-100 -400];
b = [2;0];
K = A\b
K=
2.6667e+00

-6.6667e-01
Hence,
iL (t) =

8 100t 2 400t
e
e
1 A
3
3

SOLUTION TO PROBLEM 9.35.

(a) From the problem specs and the continuity property,
iL(0-) = iL(0+) = 0.008 A
vC(0-) = vC(0+) = 2
At t = , the inductor looks like a short and the capacitor looks like an open; hence iL()
= 0 and vC() = 4000.006 = 2.4 V. The circuit is a series RLC with characteristic
polynomial
R
1
s2 + s +
=0
L
LC
R = 400; C = 6.6667e-6; L = 0.2;
si = roots([1 R/L 1/(L*C)])
s1 = si(1); s2 = si(2);
si =
-1.5000e+03
-5.0000e+02
Hence

vC (t) = K1e 1500t + K2e 500t + 2.4 V

Using initial conditions

vC(0) = K1 + K2 + 2.4 = 2
iC (0+) iL (0+)
8 10 3
vC '(0) = 1500K1 500K 2 =
=
=
= 1200
C
C
6.6667 10 6
A = [1 1;-1500 -500];
b = [2-2.4;1200]
b=
-4.0000e-01
1.2000e+03
K = A\b
K=
-1.0000e+00
6.0000e-01

Hence

vC (t) = e 1500t + 0.6e 500t + 2.4 V

(b) vL(t) is going to have the same form as vC(t) above except that vL() = 0 since the
inductor is a short at t = . Alternately however, we have
diL (t)
diC (t)
d 2vC (t)
v L (t) = L
=L
= LC
= 3e 1500t + 0.2e 500t V
2
dt
dt
dt
SOLUTION TO PROBLEM 9.36.
At t = 0, inductor is a short circuit and the capacitor is an open circuit. Since the current
source is 0 at t = 0-, and the continuity property,
iL(0-) = iL(0+) = 1 A
vC(0-) = vC(0+) = 65 V
For positive time, we have a series RLC circuit with characteristic equation
s2 +

R
1
s+
=0
L
LC

R = 65; C = 0.1e-3; L = 0.1;

si = roots([1 R/L 1/(L*C)])
s1 = si(1); s2 = si(2);
si =
-4.0000e+02
-2.5000e+02
At t = , vC() = 0.6*65 = 39 V. Hence
vC (t) = K1e 400t + K2e 250t + 39 V
Using initial conditions,
vC(0) = K1 + K2 + 39 = 65
i (0+) iL (0+)
1
vC '(0) = 400K1 250K 2 = C
=
= 4 = 10 4
C
C
10
A = [1 1;s1 s2];
b = [65-39; -1e4];K=A\b
K=
2.3333e+01
2.6667e+00

Hence,

vC (t) = 23.333e 400t + 2.6667e 250t + 39 V

SOLUTION 9.37.
(a) Rth = 200//50 + R = (40 + R). The characteristic equation is:
2
1
1
s+
s +
=0

RthC
LC
Critically damped means that both roots are the same, so the discriminant is zero, i.e.
2

1
4
=0

RthC
LC
Equivalently,
Rth = 40 + R = 0.5

L
= 50
C

Thus R = 10 .
(b)
Short the inductor and open the capacitor. Because the capacitor is in parallel
+
with the shorted inductor at t = 0-, vC(0 ) = vC(0 ) = 0. The Thevenin equivalent
resistance seen by the LC-parallel combination is Rth = 50 from part (a). A simple
+

calculation indicates that Voc = 0.8*50 = 40 V. Therefore, iL(0 ) = iL(0 ) = 40/50 = 0.8
+
A. To find vR(0 ) we use the following equivalent circuit:

Hence,
10
= 8 V.
10 + 40
+
To compute the derivative of vR at 0 , consider that
v R (0 + ) = 40

d
d
d 40 vC (t)
d
iC(t)
(v R (t)) = (10 i R (t)) = 10
=
0.2
v
(t)
=
0.2
(
)
C

dt
dt
dt
50
dt
C

Hence

dv R (t)
iC (0 + )
=
0.2
= 4000iC (0+ )

+
dt t =0
C

But
iC (0 + ) =

40
i L ( 0+ ) = 0.8 0.8 = 1.6 A
50

Therefore,

(c)

dv R (t)
= 4000iC (0 + ) = 6400 V/s

+
dt t =0
Since the circuit is critically damped the roots of the characteristic equation are
s1,2 =

1
= 200
2RthC

According to table 9.2 for t 0,

v R (t) = ( K1 + K2t )e 200t + X F
It follows from the circuit and this equation that
X F = v R () = 8 V
K1 = v R (0 + ) XF = 0
dv R (t)
= 200K1 + K 2 = K2 = 6400
dt t =0 +
Therefore

v R (t) = 6400te200t 8 u(t) V

A plot of the waveform is given below

SOLUTION TO PROBLEM 9.38. For this problem we first compute the Thevenin
equivalent of the circuit to the left of the capacitor for t > 0. Consider

Now observe
V 12 Vs kv1 Vs 12 Vs k (12 Vs )
Is = I1 + I2 = s
+
=
+
500
500
500
500
It follows that
500
1 + k
Vs =
Is +
12 = Rth Is + Voc

2+ k
2 + k
The parallel LC is now driven by this Thevenin equivalent.
L = 0.01; C = 1e-6;
% Critical damping means (1/(Rth*C))2 - 1/(L*C) = 0
x = sqrt(4/(L*C));
Rthcrit = 1/(C*x)
Rthcrit =
50
kcrit = (500 - 2*50)/50
kcrit =
8
% For parallel circuit, larger Rth means less damping
% Hence, smaller Rth means overdamped. Smaller Rth
% means larger k. Therefore k > 8 is the ranger for
% overdamped response.
For the critically damped response we have R = Rth = 50 ; hence
R = 50; C = 1e-6; L = 0.01;
si = roots([1 1/(R*C) 1/(L*C)])
si =

-10000
-10000
in which case
vC (t) = (K1 + K 2t)e 50t + X f
At t = , vC () = v L () = 0 in which case X f = 0. From the initial conditions,
vC (0) = K1 = 0
Voc
i (0+)
iC (0+) Rthcrit L
vC '(0) = s1K1 + K2 = K 2 =
=
C
C
Hence
K2 =
Therefore

10.8 50
= 2.16 105
6
10

vC (t) = 2.16 105 e 50t V

SOLUTION TO PROBLEM 9.39. (a) The series RLC leads to a characteristic equation of
the form
s2 +

R
1
24
1
5
s+
= s2 +
s+
= s2 + 120s + = 0
L
LC
0.2
0.2C
C

For a critically damped response, 120 2 =

20
. Hence, C = 1.3889 mF.
C

(b)
C = 20/120^2
C=
1.3889e-03
L = 0.2; R = 24;
si = roots([1 R/L 1/(L*C)])
si =
-60
-60
Hence
iL (t) = (K1 + K2t)e 60t + 0.4 A
where iL() = 0.4 because at t = , the capacitor looks like an open and the inductor like
a short. Hence all current from the source flows through the inductor.

Using the initial conditions,

iL(0) = K1 + 0.4 = 0 K1 = 0.4
v (0+) vC (0+) 24(i L (0+) 0.4)
iL '(0) = s1K1 + K2 = 60 0.4 + K 2 = L
=
= 48
L
L
Hence, K2 = 24 and

iL (t) = (0.4 + 24t)e 60t + 0.4 A

SOLUTION TO PROBLEM 9.40. This problem differs from 39 in the initial condition
calculation. Specifically,
iL(0-) = iL(0+) = 0.4
vC(0-) = vC(0+) = 0
Again

iL (t) = (K1 + K2t)e 60t + 0.4 A

and
iL(0+) = K1 + 0.4 = 0.4 K1 = 0.8
v (0+) vC (0+) 24 (iL (0+) 0.4)
iL '(0) = 60 0.8 + K 2 = L
=
= 96
L
L
Hence,
K2 = 48
and

iL (t) = (0.8 + 48t)e 60t + 0.4 A

SOLUTION TO PROBLEM 9.41.

(a) At 0-, the capacitor is an open circuit and inductor is a short circuit. So,
vC(0-) = vC(0+) = 5 V
-3
-3
iL(0-) = iL(0+) = 510 410 = 1 mA
Now, at 0+, replace the capacitor by a 5 V voltage source and the inductor by a 1 mA
current source. Also, the original independent current source is turned off. Solve the
resulting circuit to obtain.
-3
-3
iC + 510 110 = 0
iC(0+) = 4 mA

vL(0+) = 0
(b) Since the circuit is a parallel RLC, the characteristic polynomial is
1
1
1
1
s+
= s2 + 4 s +
= s2 + 2 s + 2 + 2d = 0
RC
LC
LC
10 C
R = 1e3; C = 0.5e-6; L = 0.184;
si = roots([1 1/(R*C) 1/(L*C)])
si =
-1.0000e+03 + 3.1416e+03i
-1.0000e+03 - 3.1416e+03i
s2 +

wd = imag(si(1))
wd =
3.1416e+03
sig = -real(si(1))
sig =
1000
-3

Also, at t = , iL() = 510 A. Therefore

iL (t) = e 1000t Acos( 103 t) + Bsin( 10 3 t) + 5 10 3 A

Using initial conditions
iL (0) = A + 5 103 = 10 3 A = 4 10 3
and
v (0+)
iL '(0) = 1000A + 103 B = 4 + 103 B = L
=0
L
Hence,
B = -4e-3/pi
B=
-1.2732e-03
Finally,

iL (t) = e 1000t 4 c o s
( 10 3 t) + 1.2732sin( 103 t) + 5 mA

SOLUTION TO PROBLEM 9.42. The response here coincides with that of problem 41 up to
time t = 2s. At this point we need the new initial conditions on the circuit at t = 2+.
However, at t = 2, e 1000t = 0 for all practical purposes. Hence, iL (2) = 5 103 A.
Differentiating the expression for iL(t) and evaluating at t = 2 yields zero by inspection.

This follows because Ke 2000 = 0 for K in the range of 1 to 10 . This can also be seen
from the circuit because at t = 2 s, the capacitor has charged to 5 V, making
v (2+)
iL' (2+) = L
= 0.
L
4

To find steady state current, solve the circuit with the new current source value and with
the capacitor and inductor as open and short circuits, respectively:
iL () = 5 10 3 + 4 10 3 = 9 103

[ (

)]

iL (t) = e 1000(t 2) A cos 103 (t 2) + B sin 10 3(t 2) + 9 10 3 A

Using the new initial conditions
iL (2+) = A + 9 10 3 = 5 103 A = 4 103
and
v (0+)
iL '(2+) = 1000A + 10 3 B = 4 + 10 3 B = L
=0
L
-3

Hence B = 1.273210 and for t 2s,

[ (

)]

iL (t) = e 1000(t 2) 4cos 10 3(t 2) + 1.2732sin 103 (t 2) + 9 mA

SOLUTION TO PROBLEM 9.43. At t = 0-, iL(0-) = 10/20 = 0.5 A and vC(0-) = 5 V by the
usual considerations. At t = 0+, we have a parallel RLC circuit. Hence
s2 +

1
1
1
1
s+
= s2 + 4 s +
= s2 + 2 s + 2 + 2d = 0
RC
LC
LC
10 C

R = 10; C = 0.05e-3; L = 0.01;

si = roots([1 1/(R*C) 1/(L*C)])
si =
-1.0000e+03 + 1.0000e+03i
-1.0000e+03 - 1.0000e+03i
wd = imag(si(1))
wd =
1.0000e+03
sig = -real(si(1))
sig =
1000

Also, at t = , vC() = 5 V. Therefore

vC (t) = e 1000t A cos(103 t) + Bsin(10 3 t) 5 V

Using initial conditions
vC (0) = A 5 = 5 A = 0
and

i (0+)
0.5
vC '(0) = 1000A + 1000B = 10 3 B = C
=
= 10 4
3
C
0.05 10

Hence, B = 10 and
vC (t) = 10e 1000t sin(103 t) 5 V
Further

v L (t) = 5 v C (t) = 10e 1000t sin(103 t) V

SOLUTION TO PROBLEM 9.44. (a) The circuit is a driven series RLC. Hence
s2 +

R
1
s+
= s2 + 2 s + 2 + 2d = 0
L
LC

R = 400; C = 0.5e-6; L = 0.2;

si = roots([1 R/L 1/(L*C)])
si =
-1.0000e+03 + 3.0000e+03i
-1.0000e+03 - 3.0000e+03i
wd = imag(si(1))
sig = -real(si(1))
wd =
3000
sig =
1000
Also, at t = , vC() = 4000.006 = 2.4 V. Therefore
vC (t) = e 1000t [A cos(3000t) + B sin(3000t)] + 2.4 V
Using initial conditions
vC (0) = A + 2.4 = 2 A = 0.4
and

i (0+) i L (0+)
vC '(0) = 1000A + 3000B = 400 + 3000B = C
=
= 1.6 104
C
C
Hence, B = 5.2 and
vC (t) = e 1000t [0.4cos(3000t ) + 5.2sin(3000t )] + 2.4 V
(b) Consistent with underdamped circuit behavior and because the inductor behaves as a
short and the capacitor as an open at t = (iL() = 0),
iL (t) = e 1000t [ Acos(3000t) + Bsin(3000t)] V
Using the initial conditions,
iL (0+) = A = 0.008 A
Further,

v (0+) 400(0.006 0.008) 2

iL '(0) = 1000A + 3000B = 8 + 3000B = L
=
= 14
L
0.2

Hence, B = 0.002 and

iL (t) = e 1000t [0.008cos(3000t) 0.002sin(3000 t)] V
Finally

v L (t) = v R (t) vC (t) = 400[0.006 i L (t)] vC (t) = 2.4 400i L (t) v C (t)

which implies that

v L (t) = e 1000t [3.6cos(3000t) 13.2sin(3000t )] V

SOLUTION TO PROBLEM 9.45. This circuit is the same series RLC as problem 44. Note
that at t = , vC() = 2.4 V. Hence
vC (t) = e 1000t [A cos(3000t) + B sin(3000t)] 2.4 V
Now, the initial conditions are:
vC(0-) = 2.4 = vC(0+) V, iL(0-) = 0 = iL(0+)
Thus

vC (0) = A 2.4 = 2 . 4 A = 4.8

Further,
i (0+)
vC '(0) = 1000A + 3000B = 4800 + 3000B = C
=0
C

Hence, B = 1.6 and

vC (t) = e 1000t [4.8cos(3000t) + 1.6sin(3000 t)] 2.4 V

SOLUTION TO PROBLEM 9.46. For all three cases, assuming iL is pointing downward,
At t = ,

vC (0) = vC (0+) = 0 and iL (0) = iL (0+) = 0.1 A

vC () = 0 and iL () = 0.2 A

Further
iC (0+) = iL (0+) +

10 vC (0+)
= 0.1 0.2 = 0.3 A
50

Lastly, all three cases are for a parallel RLC whose characteristic equation is:
s2 +

1
1
s+
=0
RC
LC

(a)
R = 50; C = 0.04e-3; L = 0.625;
si = roots([1 1/(R*C) 1/(L*C)])
s1 = si(1); s2 = si(2);
si =
-400
-100
Hence,

vC (t) = K1e 100t + K2e 400t V

From, the initial conditions

vC(0+) = 0 = K1 + K2
and

i (0+)
0.3
vC '(0) = s1K1 + s2K 2 = 100K1 400K2 = C
=
= 7500
C
0.04 10 3

A = [1 1;si(2) si(1)];
b = [0;-7500];
K = A\b
K=
-25
25
Therefore

vC (t) = 25 e 100t e 400t V

(b)
R = 50; C = 0.04e-3; L = 0.4;
si = roots([1 1/(R*C) 1/(L*C)])
s1 = si(1); s2 = si(2);
si =
-2.5000e+02
-2.5000e+02
Thus
From IC's,

vC (t) = (K1 + K 2t)e 250t V

vC (0) = K1 = 0
i (0+)
vC '(0) = s1K1 + K2 = K 2 = C
= 7500
C

Therefore

vC (t) = 7500te250t V

(c)
L = 0.2;
si = roots([1 1/(R*C) 1/(L*C)])
si =
-2.5000e+02 + 2.5000e+02i
-2.5000e+02 - 2.5000e+02i
vC (t) = e 250t [A cos(250t) + Bsin(250t)] V
From ICs.

vC (0+) = A = 0
i (0+)
vC '(0) = 250A + 250B = 250B = C
= 7500
C

in which case B = 30. Thus

vC (t) = 30e 250t sin(250t) V

SOLUTION TO PROBLEM 9.47.

At t = 0-, he capacitor is open and the inductor is a short. This together with the
continuity property implies vC(0-) = vC(0+) = 10 V and iL(0-) = iL(0+) = 1 A by
inspection.

Now, for t = 0+, vin = 0, replace capacitor and inductor with a voltage source and a
current source, respectively (values are those of the initial conditions). Solve for initial
capacitor current and initial inductor voltage to obtain:
vL(0+) = -10 V
iC(0+) = iL(0+) iR1 iR2 = 1 A
Notice that the resulting circuit is an undriven parallel RLC circuit with Req = 10//10 = 5
.
R = 5; C = 0.01; L = 4/3;
si = roots([1 1/(R*C) 1/(L*C)])
s1 = si(1); s2 = si(2);
si =
-15
-5
Hence,

vC (t) = K1e 15t + K 2e 5t V

From, the initial conditions

vC(0+) = 10 = K1 + K2
and

i (0+)
1
vC '(0) = s1K1 + s2K 2 = 15K1 5K 2 = C
=
= 100
C
0.01

A = [1 1;-15 -5];
b= [10; -100];
K = A\b
K=
5.0000e+00
5.0000e+00
vC (t) = 5e 15t + 5e 5t V
Verify with dsolve function in matlab: dsolve('D2y+20*Dy+75*y=0,y(0)=10,Dy(0)=100')
dsolve('D2y+20*Dy+75*y=0,y(0)=10,Dy(0)=-100')
ans =
5*exp(-5*t)+5*exp(-15*t)

10

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

SOLUTION TO PROBLEM 9.48.

For t = 0-, there is no source present nor has there been a non-zero excitation. Hence,
vC(0-) = vC(0+) = 0 and iL(0-) = iL(0+) = 0
At t = 0+, replace capacitor and inductor by 0-valued voltage and current sources to
obtain:
vL(0+) = 10 V, iC(0+) = 1 A
For t > 0, we have a driven parallel RLC circuit with vC() = 10 V. Thus
R = 5; C = 0.01; L = 4/3;
si = roots([1 1/(R*C) 1/(L*C)])
s1 = si(1); s2 = si(2);
si =
-15
-5
and

vC (t) = K1e 15t + K 2e 5t + 10 V

From, the initial conditions

vC(0+) = 0 = K1 + K2 +10
and

i (0+)
1
vC '(0) = s1K1 + s2K 2 = 15K1 5K 2 = C
=
= 100
C
0.01
A = [1 1;-15 -5];
b= [-10; 100];
K = A\b
K=
-5.0000e+00
-5.0000e+00
in which case

vC (t) = 5e 15t 5e 5t + 10 V

To compute iL, note that iL() = 1 A,

iL (t) = K1e 15t + K 2e 5t + 1 A
From, the initial conditions
iL(0+) = 0 = K1 + K2 +1
and
v (0+)
10
iL '(0) = s1K1 + s2K2 = 15K1 5K2 = L
=
= 7.5
L
4/3
A = [1 1;-15 -5];
b= [-1; 7.5];
K = A\b
K=
-2.5000e-01
-7.5000e-01
iL (t) = 0.25e 15t 0.75e 5t + 1 A

SOLUTION TO PROBLEM 9.49.

Input to this circuit is a superposition of the inputs in problems 9.47 and 9.48. So, the
output of the circuit here is a superposition of the output of the circuit in problems 9.47
and 9.48:
vC (t) = 10e 15t + 10e 5t 10 V
For t > 0, by linearity this is the difference of the zero-input circuit response (i.e., due to
the IC's as per problem 47) and the zero-state (zero ICs) as per problem 48.

SOLUTION TO PROBLEM 9.50.

(a)
vC(0-) = 60 V, iL(0-) = 0.1 A
(b) By continuity property,
vC(0+) = 60, iL(0+) = 0.1
(c)
Replace capacitor by voltage source of value 60 V and inductor by current source of
value 0.1 A.
vL(0+) + vC(0+) = 60
vL(0+) = 120
and
iC(0+) + iR1(0+) 1 iL(0+) = 0
iC(0+) = 1
(d) Req = 120//600 = 100 .
R = 100; C = 1e-3; L = 2;
si = roots([1 1/(R*C) 1/(L*C)])
si =
-5.0000e+00 + 2.1794e+01i
-5.0000e+00 - 2.1794e+01i
sig = -real(si(1)); wd = imag(si(1));
(e) Note that if the excitation of 60 V had remained forever, then iL() would be 0.1 A.
Therefore for the interval 0 < t < 1,
iL (t) = e 5t [ A cos(21.794t) + B sin(21.794t)] + 0.1 A
(f)
iL(0+) = A + 0.1 = 0.1
A = 0.2
v (0+)
iL' (0+) = 5A + 21.794B = 1 + 21.794B = L
= 60
L
Hence, B = 2.707 and
iL (t) = e 5t [0.2cos(21.794t) + 2.707sin(21.794t)] + 0.1 A
(g) For t > 1, the forcing function is zero iL() = 0. Thus,
iL (t) = e 5(t 1) [ A cos(21.794(t 1) ) + Bsin(21.794(t 1) )] A

From part (f) of the exp(-5t) term, we can guess that iL(1) approximates 0.1 and also that
vC(1) approximates its steady state value of 60 V. These may be off by percent or two,
but are good enough for our engineering calculations. It follows that vL(1+)
approximates 60 V.
iL (1+) = A = 0.1
and
v (1+)
iL' (1+) = 5A + 21.794B = 0.5 + 21.794B = L
= 30
L
Here B = 1.35. Hence
iL (t) = e 5(t 1) [0.1cos(21.794(t 1) ) 1.35sin (21.794(t 1) )] A
*SOLUTION 9.51. To find the initial conditions, use the following equivalent circuit at t =

0 .

By inspection iL (0 + ) = i L ( 0 ) = 1 A and vC (0 + ) = v C( 0 ) = 5 V.
To find the characteristic roots, set independent source to zero which means open
circuit the independent current source in figure P9.51. This leaves a series RLC with Rth
= 10 . Hence
R
1
s2 + th s +
= s2 + 100s + 2.5 10 4 = 0
L
LC
Using MATLAB, we find
Rth = 10;C = 0.4e-3; L = 0.1;
s12=roots([1 Rth/L 1/(L*C)])
s12 =
-50
-50
Since for t > 0, the source is off, we use table 9.1, case 3 to obtain
iL (t) = ( K1 + K2t )e 50t A

It follows that 1 = iL (0 + ) = K1 and

di L +
1
(0 ) = 50K1 + K2 = 50 + K2 = v L (0 + ) = 10v L (0 + )
dt
L
+
+
To find v L (0 ) we consider the equivalent circuit valid at 0 :

It follows that
Hence
or K2 = 0. Finally

v L (0 + ) = 5 10 1 = 5 V
50 + K2 = 50
iL (t) = e50t u(t) A

SOLUTION TO PROBLEM 9.52. To find the initial conditions, use the following equivalent

circuit at t = 0 .

By inspection iL (0 + ) = i L ( 0 ) = 1 A and vC (0 + ) = v C( 0 ) = 5 V.
To find the characteristic roots, set independent source to zero which means open
circuit the independent current source in figure P9.51. This leaves a series RLC with Rth
= 10 . Hence
R
1
s2 + th s +
= s2 + 20s + 5 10 3 = 0
L
LC
Using MATLAB, we find
Rth = 10;C = 0.4e-3; L = 0.5;
s12=roots([1 Rth/L 1/(L*C)])
s12 =
-1.0000e+01 + 7.0000e+01i
-1.0000e+01 - 7.0000e+01i
Since for t > 0, the source is off, we use table 9.1, case 2 to obtain
iL (t) = e 10t [ Acos(70t) + Bsin(70t)] A
It follows that 1 = iL (0 + ) = A and
diL +
1
(0 ) = 10A + 70B = 10 + 70B = v L (0+ ) = 2v L (0+ )
dt
L
+
+
To find v L (0 ) we consider the equivalent circuit valid at 0 :

It follows that

v L (0 + ) = 5 10 1 = 5 V
Hence 10 + 70B = 10 or B = 0. Finally
iL (t) = e 10t cos(70t)u(t) A

0.8

0.6

0.4

0.2

-0.2

-0.4

-0.6

-0.8

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

SOLUTION TO PROBLEM 9.53. (a) vC (0) = vC (0+) = 150 103 80 = 12 V.

(b) iL(0-) = iL(0+) = 150 mA
(c) For these values at 0+, the independent current source becomes an open circuit.
Replace the inductor and capacitor with current and voltage sources to represent the
initial conditions. Solve the resulting simple circuit to obtain:
vL(0+) = 6 = 12 120*0.15 V, iC(0+) = iL(0+) = 150 mA
(d)
Rth = 120;C = 2/9 * 1e-3; L = 0.6;
s12=roots([1 Rth/L 1/(L*C)])
s12 =
-1.5000e+02
-5.0000e+01
(e)

vC (t) = K1e 150t + K2e 50t V

From, the initial conditions

vC(0+) = 12 = K1 + K2
and

i (0+)
vC '(0) = s1K1 + s2K 2 = 150K1 50K 2 = C
= 675
C
A = [1 1;-150 -50];
b= [12; -675];
K = A\b
K=
7.5000e-01
1.1250e+01
in which case

vC (t) = 0.75e 150t + 11.25e 50t V

(f)
iL (t) = iC (t) = C

dvC (t) 2
= 10 3 (150 0.75)e 150t + (50 11.25)e 50t
dt
9

= 0.025e 150t + 0.125e 50t A

SOLUTION TO PROBLEM 9.54. (a) At t = 0-, the independent current source is off, the
inductor is a short circuit, and the capacitor is an open circuit. By voltage division,
vC(0-) = vC(0+) = (80/100)50 = 40 V
(b)
iL(0-) = iL(0+) = 50/100 = 0.5 A
(c) At t = 0+, we have an independent current source. Also, we replace the inductor with
a current source and the capacitor with a voltage source to represent the initial conditions.
1.5 iL(0+) = iC(0+)
iC(0+) = 1 A
Further,
vC(0+) + 40 iC(0+) = 80 iL(0+) + vL(0+)
vL(0+) = 40 V
(d)
Rth = 120;C = 2/9 * 1e-3; L = 0.6;
s12=roots([1 Rth/L 1/(L*C)])
s12 =
-1.5000e+02
-5.0000e+01
(e) In steady state, the capacitor is open and the inductor is a short in which case, Xf = 80
* 1.5 = 120 V.
vC (t) = K1e 150t + K2e 50t + 120 V

From, the initial conditions

vC(0+) = 40 = K1 + K2 + 120
and
i (0+)
vC '(0+) = 150K1 50K2 = C
= 4.5 103
C
A = [1 1;-150 -50];
b= [-80; 4.5e3];
K = A\b
K=
-5
-75
in which case

vC (t) = 5e 150t 75e 50t + 120 V

(f)
iL (t) = 1.5 iC (t) = 1.5 C

dvC (t)
= 0.16667e 150t 0.83333e 50t + 1.5 A
dt

SOLUTION TO PROBLEM 9.55.

(a) At t = 0-, we replace the inductor by a short circuit and the capacitor by an open
circuit; hence
iL(0-) = iL(0+) = 1 A
and
vC(0-) = vC(0+) = (40)(1) 20 = 20 V
(b) At t = 0+, we replace the capacitor by a voltage source of value 20 V and the
inductor by a current source of value 1 A. Since the inductor current is 1 A and the
independent current source outputs 1 A, no current flows through the branch containing
the capacitor. Therefore,
iC(0+) = 0
Also, because of the zero current in the branch containing the capacitor, no voltage drop
occurs across the resistance in series with the capacitor. Therefore, the voltage across the
independent current source is vC(0+). Therefore,
vL(0+) = vC(0+) 40*iL(0+) = 20 V
(c) At steady state (large t), the capacitor becomes an open circuit and the inductor
becomes a short circuit. By inspection,
vC() = 40 V

(d)
Rth = 80;C = 1/15 * 1e-3; L = 0.1;
s12=roots([1 Rth/L 1/(L*C)])
s12 =
-5.0000e+02
-3.0000e+02
vC (t) = K1e 500t + K2e 300t + 40 V
(e) From, the initial conditions
vC(0+) = 20 = K1 + K2 + 40
and
i (0+)
vC '(0+) = 500K1 300K 2 = C
=0
C
A = [1 1;-500 -300];
b= [-20; 0];
K = A\b
K=
3.0000e+01
-5.0000e+01
in which case

vC (t) = 30e 500t 50e 300t + 40 V

SOLUTION TO PROBLEM 9.56.

(a) At t = 0-, we replace the inductor by a short circuit and the capacitor by an open
circuit; hence
iL(0-) = iL(0+) = 1 A
and
vC(0-) = vC(0+) = (40)(1) 20 = 20 V
(b) At t = 0+, we replace the capacitor by a voltage source of value 20 V and the
inductor by a current source of value 1 A. Since the inductor current is 1 A and the
independent current source outputs 1 A, no current flows through the branch containing
the capacitor. Therefore,
iC(0+) = 0
Also, because of the zero current in the branch containing the capacitor, no voltage drop
occurs across the resistance in series with the capacitor. Therefore, the voltage across the
independent current source is vC(0+). Therefore,

vL(0+) = vC(0+) 40*iL(0+) = 20 V

(c) At steady state (large t), the capacitor becomes an open circuit and the inductor
becomes a short circuit. By inspection,
vC() = 40 V
(d)
Rth = 80;C = 62.5e-6; L = 0.1;
s12=roots([1 Rth/L 1/(L*C)])
s12 =
-400
-400
vC (t) = (K1 + K 2t )e 400t + 40 V
(e) From, the initial conditions
vC(0+) = 20 = K1 + 40 K1 = 20
and
i (0+)
vC '(0+) = 400K1 + K 2 = 8000 + K 2 = C
=0
C
in which case

vC (t) = (20 + 8000t)e 400t + 40 V

SOLUTION TO PROBLEM 9.57.

Step 1:
dvC/dt = iC/C = 2iC
diL/dt = vL/L = 0.5vL
Step 2: iR = vC/2. From KCL, iL iR ic = 0. Therefore, ic = iL vc/2.
Step 3: Similarly, vin vL vc = 0, which implies vL = vin vC. Hence,
dvC/dt = 2 iL vC
diL/dt = 0.5 vin 0.5 vC
Step 4. Eliminate terms in vC (see equation 9.47 in text) to obtain:
d2iL/dt2 + diL/dt + iL = 0.5 dvin/dt + 0.5 vin

SOLUTION TO PROBLEM 9.58.

(a) At t = 0-,
vC(0-) = vC(0+) = 0
iL(0-) = iL(0+) = 0
For t between 0 and 1, we have a parallel RLC circuit, with Req being the parallel
combination of the two 21.1333 resistors.
R = 21.1333/2; C = 15.7729e-3; L = 0.1;
si = roots([1 1/(R*C) 1/(L*C)])
si =
-3.0000e+00 + 2.5000e+01i
-3.0000e+00 - 2.5000e+01i
If there were no further switchings, then Xf = 1 A. Hence,
iL (t) = e 3t [ A cos(25t) + Bsin(25t)] + 1 A
Applying the initial conditions,
iL(0+) = A + 1 = 0 A = 1
and
Hence, for 0 t < 1,

v (0+)
iL' (0+) = 3A + 25B = 3 + 25B = L
=0
L
iL (t) = e 3t [cos(25t) + 0.12sin(25 t)] + 1 A

Now, for the next interval, we need initial conditions. These are obtained from the above
equation for iL(t) at t = 1.
iL(1) = e-3[-cos(25) 3/25 sin(25)] + 1 = 0.9514
and

iL' (1) = 3e 3[cos(25) + 0.12sin(25)] e 3[25sin(25) + 0.12 25cos(25)] = 0.1671

The circuit is still a parallel RLC circuit, but now there is no source and R = 1.268:
R = 1.268; C = 15.7729e-3; L = 0.1;
si = roots([1 1/(R*C) 1/(L*C)])
si =
-2.5000e+01 + 3.0002e+00i
-2.5000e+01 - 3.0002e+00i
Hence,

iL (t) = e 25(t 1)[ A cos(3(t 1)) + B sin(3(t 1))]

Using initial conditions

A = 0.9514
0.1671 = 25A+ 3B B = 7.8726
Thus, for 1 t < 2
iL (t) = e 25(t 1) [0.9514cos(3(t 1)) + 7.8726sin(3( t 1)) ] A
(b)
In period between 1 and 2 seconds, the response has a time constant of 1/25. So, when t =
2, 25 time constants would have passed from the time the switch is turned (t = 1). This
means that the L and C currents and voltages would have settled almost identically to
their values at 0. A similar argument can be made for the other cycle. Thus the overall
response effectively becomes a periodic response equal to the response over 0 t < 2
that reflects the periodicity of the switching.
(c)
1.8
1.6

Inductor current in A

1.4
1.2
1
0.8
0.6

TextEnd

0.4
0.2
0

3
Time in s

SOLUTION TO PROBLEM 9.59.

(a)
dvc/dt = ic/C = 0.707ic
diL/dt = vL/L = 0.707vL
To find expressions for iC and vL we use the following figure.

From this resistive circuit, iC = vC/1 + iL and vL = vin 1iL vC

vC = 0.707v C + 0.707iL
iL = 0.707vC 0.707iL + 0.707v in
Using equation 9.47,
d 2vC
dv
+ 1.414 C + v C = 0.5vin
2
dt
dt
si = roots([1 sqrt(2) 1])
si =
-7.0711e-01 + 7.0711e-01i
-7.0711e-01 - 7.0711e-01i
(b) At steady state (large t), vC = 0.5
vC(t) = e-0.707t[A cos(0.707t) + B sin(0.707t)] + 0.5 V
At t = 0,
vC(0-) = vC(0+) = A + 0.5 = 0 A = 0.5
i (0) iL (0) vC (0)
vC' (0) = C
=
= 0 = 0.707A + 0.707B
C
C
B = 0.5
Thus
vc(t) = e-0.707t[0.5 cos(0.707t) 0.5 sin(0.707t)] + 0.5 V

SOLUTION TO PROBLEM 9.60.

(a)
dvC
= 3iC
dt
diL v L
=
dt
3

To eliminate iC and vL, consider

Hence
iC = iin vC/2 iL
vL = vC 12iL
vC = 1.5v C 3iL + 3iin
1
iL = v C 4iL
3
Using equation 9.47,
d 2vC
dv C
'
+
5.5
+ 7v C = 3iin
+ 12iin
2
dt
dt
si = roots([1 5.5 7])
si =
-3.5000e+00
-2.0000e+00
Note: iin(t) = 0 for t>0.
At t = 0-, current source is off, inductor is a short circuit, and capacitor is an open circuit.
vc(0-) = vc(0+) = 0
iL(0-) = iL(0+) = 0
At t = 0+, the current source is on. Replace the inductor and capacitor by current and
voltage sources to represent the initial conditions. Hence, iC(0+) = 1 A. At t = ,
vC () = 1 (12 / / 2) = 1.7143. Thus,
vC (t) = K1e 3.5t + K2e 2t + 1.7143 V
To find the constants,

vC (0) = K1 + K 2 + 1.7143 = 0

and
i (0)
vC' (0) = C
= 3 = 3.5K1 2K 2
C

A = [1 1;-3.5 -2];
b = [-1.7143;3];
K = A\b
K=
2.8573e-01
-2.0000e+00
Therefore

vC (t) = 0.28573e 3.5t 2e 2t + 1.7143

(b) From equation 9.47b,

d 2iL
di
+ 5.5 L + 7vC = iin
2
dt
dt

Since iL () = 0.14286 A,

iL (t) = K1e 3.5t + K2e 2t + 0.14286 A

Using the initial conditions,

iL (0) = K1 + K 2 + 0.14286 = 0
and

v (0) vC (0) 12iL (0)

iL' (0) = L
=
= 0 = 3.5K1 2K2
L
L

A = [1 1;-3.5 -2];
b = [-1/7; 0];
K = A\b
K=
1.9048e-01
-3.3333e-01
Therefore,

iL (t) = 0.19048e 3.5t 0.3333e 2t + 0.14286 A

SOLUTION TO PROBLEM 9.61.

(a)
dvC1/dt = 0.5iC1
dvC2/dt = 0.5iC2
Writing a node equation,
iC1 + vC1/0.5 + (vC1 vC2)/0.5 = 0
iC1 = 4vC1 + 2vC2
By symmetry,
iC2 = 4vC2 + 2vC1
Hence

vC1 = 2v C1 + vC 2
vC 2 = v C1 2v C2
From equation 9.47a,
d 2vC
dv C
+
4
+ 3vC = 0
dt
dt 2
(b)
si = roots([1 4 3])
si =
-3
-1
Hence

vC (t) = K1e 3t + K 2e t

(c)
vC1(0) = 2 K1 + K2 = 2
and

Hence

i (0) 2(v C 2 (0) vC1(0)) 2vC1 (0) 8 8

'
vC1
(0) = C1
=
=
= 0 = 3K1 K2
C
C
C
vC (t) = e 3t + 3e t V

*SOLUTION P9.62. (a) For this problem we need to define a voltage at the output of the
first op amp as shown below.

Also, let us relate the input and output voltages for an arbitrary leaky integrator as shown
below.

We now write a node equation at the inverting terminal of the op amp. Here
va vb
dv
+
+C b = 0
Ra Rb
dt
Equivalently
dv b
v
v
= b a
dt
RbC Ra C

(1)

Now we apply the formula of (*) to the second stage of our given op amp circuit to
obtain:
dvout
v
v
(2)
= out 1
dt
RC2 RC2
where R = 1 M and C2 is to be determined.
Applying the formula of (*) to the first stage we obtain:
dv1
v
v
(3)
= 1 s
dt
RC1 R1C1
The equations (2) and (3) form a coupled set of state equations which we can write as
1
1

v
out
RC2
RC2 vout 1
v
v =
1 v1 +
s
1 0

R
C
1 1

RC1
Using equation 9.47a of text we can write down the characteristic equation as
1

1
1
1
1
s2 +
+
= s +
s +
s +
=0
RC2 RC1
RC1 RC2
RC2
RC1
We require that the natural frequencies be 4 and 12 in which case

1
1
= 4,
= 12
RC2
RC1
From MATLAB
R = 1e6;
C2 = 1/(R*4)
C2 = 2.5000e-07
C1 = 1/(R*12)
C1 = 8.3333e-08
(b) For this part, the overall dc gain must be 10. The dc gain of the second stage is 1.
106
The dc gain of the first stage must be 10 =
; hence R1 = 100 k.
R1
(c) Since the roots are distinct and real, for t > 0
vout (t) = K1e 4t + K 2e 12t + X f = K1e 4t + K2e 12t + 10 V
where X f = 10 V by part (b). The problem states that the capacitor voltages are initially
zero. Hence
0 = vC2 ( 0 ) = vout (0) = K1 + K2 + 10
Equivalently

K1 + K2 = 10

(1)

Also,
dvout
1
(0) =
i ( 0 ) = 4K1 12K2 = 0
dt
C2 C2

(2)

because iC2 ( 0 )= 0 . This is so because vC1( 0 )= 0 = v1( 0 ) means no current flows

through the 1 M input resistor to stage 2. This fact and the fact that vC2 (0) = 0, means
that no current flows through C2.
From equation 2, K1 = 3 K2. Substituting into equation (1) yields K2 = 5 and
hence K1 = 15. Finally,

vout (t) = 15e 4t + 5e 12t + 10 V

*SOLUTION 9.63. This problem requires the characteristic equation in terms of A. For
this we may set vin = 0 and the circuit becomes the one given below. Note the new label
v1.

The first step is to write a node equation at v1:

v1
d
1
v
+ 10 4 ( v1 vout ) +
v1 out = 0

100
dt
100
A
Equivalently
dv1 dvout 100

=
v 200v1
dt
dt
A out

(*)

Now we write a node equation at vg = vout/A. Here,

10 4 dvout
1 vout
+
v1 = 0

A
dt
100 A
Equivalently,
dvout
= 100v out + 100 Av1 (**)
dt
Let us put (*) and (**) in matrix form to obtain:
1
0

1 v1 200 100 / A v1
=
1 vout 100A 100 vout

We can now solve this to obtain the state equations

v1 1 1 200 100 / A v1 200 + 100A 100 + 100 /
v = 0 1 100A 100 v = 100 A
100
out

out

A v1
v
out

Compare these equations with equation 9.37 and use the formula of 9.47b to obtain the
following second order differential equation in vout:
d 2 vout
dvout
4
2 + (300 100A ) dt + 10 vout = 0
dt
The discriminant of this characteristic equation is plotted below for 0 < A < 3. For values
of A > 3, the circuit is unstable. A negative value of the discriminant indicates
underdamped (1 < A < 3) and a positive value overdamped (0 < A < 1). For A = 1, we
have critical damping.

Solution 9.64 We can write two state equations as follows:

(i) From the definition of a capacitor,
dvC1
= 10 7 iC1
dt
dvC 2
= 109 iC 2
dt
(ii) From KVL and Ohm's law
dvC1
= 10 6 (vi vC1 v C 2 ) = 10 6 vC1 10 6 vC 2 + 106 v i
dt
dvC 2
= 108 (vi vC1 v C 2 0.01vC 2 ) = 108 v C1 1.01 108 v C2 + 108 vi
dt
Casting these two equations into a second order differential equation, as described in the
text:
d 2vC 2
dv
dv
+ 1.02 108 C 2 + 1012 vC 2 = 108 i
2
dt
dt
dt
The characteristic equation for this differential equation has real roots:
si = roots([1 1.02e8 1e12])
si =
-1.0199e+08
-9.8049e+03

Since the capacitors become open circuits, vC2() = 0 and vC1() = vi.
8
3
vC 2 ( t) = K1e 1.0210 t + K2e 9.810 t V

Applying IC's:
vC 2 (0) = K1 + K2 = 0
Also,
i (0+)
0.1
vC' 2 (0) = C 2
= 9 = 1.02 108 K1 9.8 10 3K 2
C2
10
Thus

vC' 2 (0) = 1 = 1.02K1 9.8 105K 2

b = [0; 1];
A = [1 1; -1.02 -9.8e-5];
K = A\b
K=
-9.8049e-01
9.8049e-01
8
3
vC 2 ( t) = 0.9805e 1.0210 t + 0.9805e 9.80510 t V

t = 0:1/(abs(s1)*100):100/abs(s1);vc = -0.9805*exp(s1*t) + 0.9805*exp(s2*t);

plot(t,vc)
grid
xlabel('Time in s')
ylabel('vout in V')
1

vout in V

0.8

0.6

TextEnd

0.4

0.2

0.2

0.4

0.6
Time in s

0.8

1
x 10-6

Solution 9.65. First, derive the differential equation by writing state equations:
dvC
= 3iC
dt
diL v L
=
dt
3
Now, assume that the capacitor is a voltage source and the inductor is a current source,
and write by KCL
v
v
iC = C + C iL
2
RN
And by KVL:
v L = vC 12i L
Substitute into the differential equations:
dvC 3
3
= +
v 3iL
dt 2 RN C
diL 1
= v 4i L
dt 3 C
Using equation 9.47 we obtain
dvC 3

d 2vC 3
3
3

4
+

+
(
4
)
+
1

vC = 0

dt 2
2 RN
dt
2 RN

or equivalently
d 2vC
3 dvC
3
+ 5.5
+ 7
v =0

RN dt
RN C
dt
For constant amplitude oscillations, the middle term should be zero, which means that RN
= 3/5.5 = 0.54545 . Thus the negative resistance is RN = 0.54545 .
SOLUTION 9.66. The problem data is
i1(t) = Im sin(t + ) A
R1 = 500 + 100(Im 0.01), R2 = 500

Suppose it starts out with exponentially growing amplitude. R1 will increase with
increasing amplitude. This changes the location of the roots until equilibrium is reached
where the roots and the amplitude are stable. This is achieved when the roots of the
characteristic equation describing the output voltage are purely imaginary, i.e.,
(R1 R2)/(R1R2C) = 0
R1 = R2 = 500 = 500 + 100(Im 0.01)
Im = 0.01
(a)

0 = 1/[(500)(1)] = 2 k-rad/s

(b) Amplitude of i1 is 0.01. Thus i1 = 0.01 sin(0t) A. Let v1 = Vm cos(0t + ) V.

Then
dv1/dt = 0Vmsin(0t + ) = (0.01/C) sin(0t)
Vm = 0.01/(0C) = 5 V and = rad
Now, v2 + v1 + i1R1 3v2 = 0 v2 = v1/2 + i1R1/2. Finally,
vout = 3v 2 = 1.5v1 + 1.5R1i1 = 7.5cos( 0t + ) + 7.5sin( 0 t) V
Hence the amplitude of vout(t) is: 7.5 2 V.

SOLUTION 9.67. (a)

n

1
= 10 4 rad/s
RC

(b) From equation 9.47a,

d 2 v1
1
2 + 2 2 v1 = 0
dt
R C
Hence,
1
1
v1( t) = A cos
t + Bsin
t V
RC
RC
Using the value of v1(0) we have,
v1(0) = A = 5
To compute the second initial condition,

i (0+)
i (0+)
v1'(0 + ) = C1
= R1 6 = 104 B
C
0.1 10
But,
iR1 (0+) =

3v 2 (0+) v 2 (0+) v1(0+) 2v2 (0+) v1(0+)

=
= 0.005 Amps
R1
103

Hence
v1'(0

5 10 3
+) =
= 104 B
6
0.1 10

Hence B = 5.
v1( t) = 5cos (10,000t ) 5sin (10,000 t) V

Voltage v1 in V

0
TextEnd

-2

-4

-6

-8

0.2

0.4

0.6
Time in s

0.8

1
x 10

-3

*SOLUTION 9.68. For this problem, R2 = 10 k should be Rf = 10 k and R1 = 1 k

should be R2 = 1 k.
(a) For sustained sinusoidal oscillation, R1 = R2 = 1 k. From equation 9.59,
0 =

1
= 10 4 rad/s or 1.5915 kHz
R1C

(b) From figure P9.68, to obtain an R1 = 1 k, IR1,peak = 0.2 mA. Therefore,

dv
iR1(t) = 0.2sin( 0t + ) mA for appropriate . Since C 1 = iC1 = iR1 = 0.2sin( 0t + )
dt
mA, we know that v1 has the following form:
v1(t) = V1m cos( 0t + )
In which case
dv
d
C 1 = C (V1m cos( 0 t + )) = CV1m 0 sin( 0 t + + ) = 0.2sin( 0t + ) mA
dt
dt
Therefore CV1m 0 = 0.2 10 3 . It follows that V1m = 0.2 volts. Here V1m is the peak
value of the sinusoid. However, the op amp peak output voltage with respect to ground,
as shown in problem 66, is 1.5 2V1m . Also, for such a small amplitude, we expect the
output waveform to be quite close to sinusoidal. By choosing a different lamp (R1) with
a different characteristic, we can obtain larger peak output voltages.

SOLUTION 9.69. (a) Note that the capacitor is like an open circuit and the inductor is
like a short circuit at t=0-. Thus, we can obtain the capacitor voltage by voltage division:
vC (0 ) = 10

4
= 8 = vC (0 + )
5

Similarly, the inductor current is obtained by applying Ohms Law:

iL (0 ) =

10
= 2 = iL (0 + )
5

(b) Here, we note that the new initial conditions are just 2.5 times the values that we just
obtained in part (a). This can be achieved by simply changing the input voltage source,
from 10 to 25 V.

PROBLEM SOLUTIONS CHAPTER 10

SOLUTION 10.1 Using KCL, we can write
C

dvC v C
+
= iin ( t)
dt
R

Dividing by C:
dvC
1
i (t )
+
v C = in
dt
RC
C
We know that iin (t ) = 20sin (400t ) mA, which can be represented by a complex exponential,

iin (t ) = Re 20e j 400t e j/2 mA. For convenience we will simply let iin (t ) = 20e j 400te j/2 mA, knowing
that we must take the real part to complete our solution. The output voltage will also be reparesented as a
complex exponential:
v (t ) = V e j ( 400t + ) = V e j 400t e j
C

Substituting this expression into the differential equation and canceling e j400t :
V
20 103e j/2
j400Vme j + m e j =
RC
C
Thus
3
j4000
1
20 10
Vme j
+ j 400 =
Vm e j =
= 3.714 111.8o
RC
C
1000 + j400

where the values for R = 100 and C = 5 mF were substituted in. Thus,
Vm = 3.714
400
= 0 tan1
= 111.8
1000
Taking into account a 90o phase shift we obtain
vC (t ) = 3.714cos( 400t 111.8) = 3.714sin ( 400t 21.8) V
and
iout ( t) = 18.57sin (400t 21.8) mA

SOLUTION 10.2 From KCL and component definitions:

1

v
0.1 diL
diL
iin (t ) L iL = 0
+ iL = iin (t )
+ 250iL = 250iin (t)
25
25 dt
dt
We represent the input signal by the complex exponential: iin (t ) = 0.2e j 250t A and the unknown current
can be represented as iL (t) = IL e j ( 250t + ) .
Substituting this into the differential equation and canceling e j250t :
j250IL e j + 250IL e j = 50
Thus
IL e j

( j250 + 250) = 50

50
IL e j =
= 0.14142 45o
250 + j250

and
IL = 0.141,

= 45 iL (t) = 0.141cos(250t 45) A

SOLUTION 10.3. Construct differential equation by KVL and device definitions:

diL
diL
200iL = 0
+ 400iL = 2vin ( t)
dt
dt

vin (t) 0.5

We represent vin (t) as the complex exponential function, vin (t) = 20e j 400t V. The current in the inductor
has the form: iL = IL e j( 400t + ) . Substituting into the differential equation and canceling e j400t :
j400IL e j + 400IL e j = 40
Thus
IL e j

(j

40
+ 400) = 40 IL e j =
= 0.070711 45o
400 + j 400

and
IL = 0.0707,

= 45, i L (t) = 70.7cos (400t 45) mA

Hence,
vout (t ) = 14.14cos (400t 45) V

SOLUTION 10.4. Construct differential equation using KVL and device definitions:

vin (t) v C C

dv C
dv
R = 0 RC C + v C = v in (t )
dt
dt

The output voltage is defined as:

vout (t ) = vin ( t) v C (t )
This means that finding vC is enough to be able to obtain the output voltage. The input voltage is
represented by the complex exponential:
vin (t) = 20e j 250t e j/2 V
and vC (t ) = Vm e j ( 250t + ) . Substituting into the differential equation, dividing by e j250t , and rearranging:
j250RCVC e j + VC e j = j20 VC e j ( j250RC + 1) = j20
j20
VC e j =
= 14.142 135o
1+ j
Now
Vout e j = j20 VC e j

Voute j = 10 j10 = 14.142 45o

Thus, in the time-domain,

vout (t ) = 14.142cos(250t 45) V

SOLUTION 10.5. The circuit is identical to that of problem 10.1. Thus,

RC

dv s
dv
+ v s = Ris (t ) Ris (t ) RC s = v s
dt
dt

Moreover, the complex exponential solution is given by

v s (t) = Vm e j = 223.6e j 63.43 V
o

Hence
R jRC223.6e j 63.43 = 223.6e j 63.43 = 100 j200
o

i.e.,
R jRC(100 j200) = R 200RC j100RC = 100 j200

Thus RC = 2 s, R = 100 + 200RC = 500 , and C = 4 mF.

SOLUTION 10.6. First use KVL

di
1
vin v L vc vout = 0 vin L L
dt C

i Ldt RiL = 0

Differentiating this equation, rearranging, and dividing by L,

d 2iL R diL
1
1 dvin
+
+
i
=
L
L dt LC
L dt
dt 2
= 104 as vin (t) = 100e j t e j/2 V and iL (t) = IL e j t e j A.

We represent the input signal with

Substituting these two expressions into the differential equation and dividing out e j t :
R
1
j100 j
2IL e j + j
IL e j +
IL e j =
L
LC
L
IL e j =

108
75 106

j108

R
1
IL e j
2+ j
= 108
LC
L

= 0.48 j0.64 = 0.8 126.87o

Solving for the magnitude and angle (by hand or using MATLAB):
iL ( t) = 0.8cos (10,000t 126.87) = 0.8sin (10,000t 36.87) A
and
vout (t ) = 80sin (10,000t 36.87) V

SOLUTION 10.7. Using standard reference directions, from KCL and component definitions,
t

v
1
dv
iin = iR + iL + iC = out + iL (0) + vout ( ) d + C out
R
L
dt
0

Taking a second derivative and dividing by C yields

d 2vout
1 dvout
1
1 diin
+
+
v
=
out
RC dt
LC
C dt
dt 2
4

We now let = 2500 rad/s and represent iin (t) by the real part of the complex exponential 0.02e j t .
Further we represent vout (t) as the real part of the complex exponential Vme j( t + ) = Vm e j te j .
Substituting these expressions into the differential equation and taking the indicated derivatives yields
V
1
0.02
Vm ( j ) 2 e j te j + m j e j te j +
Vm e j te j =
j ej t
RC
LC
C
Observe that e j t cancels out on both sides of this equation producing
j j0.02
1
Vme j
2+
=
LC
RC
C
Hence
Vme j =

j0.02
= 1.28 + j0.96
1
j

C
2+
LC
RC

This was obtained using MATLAB as follows:

w = 2500;
L = 40e-3;
C = 1e-6;
R = 100;
Vout = j*0.02*w/(1/L - C*w^2 +j*w/R)
Vout = 1.2800e+00 + 9.6000e-01i
magVout = abs(Vout)
magVout = 1.6000e+00
angVout = angle(Vout)*180/pi
angVout = 3.6870e+01
Therefore
vout (t) = 1.6cos(2500t + 36.87o) V

SOLUTION 10.8. (a) From KCL: 315 545 I = 0 I = 315 545. In MATLAB,
Ibar = 3*exp(j*15*pi/180) - 5*exp(j*45*pi/180)
5

Ibar =-6.3776e-01 - 2.7591e+00i

abs(Ibar)
ans =2.8318e+00
angle(Ibar)*180/pi
ans =-1.0302e+02
Therefore,
i( t) = Re {Ibar} = 2.83cos( t 103) A
(b) From KCL,
Ibar = (1 + 2*j) - (-2 + j*6)
Ibar =
3.0000e+00 - 4.0000e+00i
abs(Ibar)
ans = 5
angle(Ibar)*180/pi
ans =-5.3130e+01
Therefore i( t) = 5cos (50t 53) A.

SOLUTION 10.9. We define a Gaussian surface encompassing the three bottom nodes. Thus, KCL
dictates that the sum of 4 currents be zero:
Ibar = -(-2-j*8) + (3 +j*12) + 10
Ibar =
1.5000e+01 + 2.0000e+01i
abs(Ibar)
ans =
25
angle(Ibar)*180/pi
ans =
5.3130e+01
Therefore i( t) = 25cos (1000t + 53.13) A.

SOLUTION 10.10. First represent the time-domain functions as phasors:

6

V1 = 20, V2 = 2 2 45
Then, by KVL
V1 = 2; V2 = 2*sqrt(2)*exp(-j*pi/4);
VL = V1 - V2
VL =
0 + 2.0000e+00i
Therefore, VL = 290 V and v L (t ) = 2cos( t + 90) = 2sin( t) V

SOLUTION 10.11. Apply KVL by simply following the loop defined by the independent voltage sources:
Vx = 4 j 2 j 1 1+ (1 j) (1+ j) = 2

V
SOLUTION 10.12. First note that Iout = R . VR = j10 + (5 j5) 10 VR = 5 j15 V.
R
5 j15
Thus, Iout =
= 1 j 3 = 3.16 108.4 iout (t) = 3.16cos (500 t 108.4) A.
5

SOLUTION 10.13. Using a Gaussian surface,

Iy = (2 + j3) (1+ j2) + (1 j5) = 2 j10 = 10.2 101.3o A.
Therefore, i( t) = 10.2cos (2000 t 101.3) A. Now applying KVL,
Vx = (2 + j2) + (2 + j 3) (1 4 j) = 3 + 9 j
Vx=2+2*j+2+j*3-1+4*j
Vx =
3.0000e+00 + 9.0000e+00i
abs(Vx)
ans =
9.4868e+00
angle(Vx)*180/pi
7

ans =
7.1565e+01
Thus, v x ( t) = 9.487cos (2000 t + 71.6) V.

SOLUTION 10.14. (a) At

= 1000 rad/s,
YC ( j ) = j C = j4.7 10 3 C = 1.496 F

= 50 , YC ( j ) = j C = j2.3499 104 S ZC (j ) = j4255 .

(b) At = 1000 rad/s, ZL = j L = j18.85 L = 6 mH. Since impedance is proportional to
So, at

frequency, multiplying the frequency by 20 means the impedance is multiplied by 20. Thus, at
= 20 (103 ) rad/s,
ZL = j18.85 20 = j 377 YL = j2.652 mS

SOLUTION 10.15. The input voltagephasor as 20. By inspection:

I1 = j C VC = j10 0.1 2 = j2
From the definition of a dependent V-source and an inductor:

I2 =

5I1
j10
=
=5
j L
j2

Finally,
Vout = 5 2(I1 + I2 ) = 10(5 + j2) = 50 + j20 = 53.8521.8
vout (t ) = 53.85cos(10t + 21.5) V

SOLUTION 10.16. Using KCL and the definition of a resistor:

V10 = 10(60 390) = 10(6 3j) = 60 j 30 V
Thus,
Z( j 0 ) =

60 j 30
= 10 j20
j3
8

And, the combination of this Z( j 0 ) with the 10 resistance, at this frequency, is 10 j5.
Z=-10-j*20; R = 10;
ZZ = R*Z/(R+Z)
ZZ =1.0000e+01 - 5.0000e+00i
C = 1/(-5*j*j*2000*pi)
C =3.1831e-05
This is equivalent to a 10 resistor in series with a 31.83 F capacitor.

SOLUTION 10.17.
w = 2*pi*60;
VL = 3 +12*exp(-j*30*pi/180) + 6 -12*exp(j*30*pi/180)
VL =9.0000e+00 - 1.2000e+01i
ZL = j*w/60
ZL =0 + 6.2832e+00i
IL = VL/ZL
IL =-1.9099e+00 - 1.4324e+00i
abs(IL)
ans =2.3873e+00
angle(IL)*180/pi
ans =-1.4313e+02
Therefore, iL (t) = 2.387cos(120t 143.1o ) A.

Solution 10.18 (a)

First represent the inputs with their phasors: Is1 = 1030 = 8.66 + j5 and Vs2 = 1000.
1
The admittance of the RC combination is: Y1 = + j C = 0.1+ j0.1. Using superposition and noting that
R
a 0-volt V-source is a short circuit and a 0-amp current source is an open circuit,
Ix = Y1Vs2 Is1 = 1.3397 + j5 = 5.176475o A
Therefore,
i x (t) = 5.176cos (100t + 75)
(b) Let i x1 be the contribution to i x generated by the current source and i x2 the contribution generated by
the voltage source. Then i x1 (t) = 10cos(50t + 30o ) A, and i x2 (t) = 14.142cos(100t + 45o ) A since
9

Vs2 = 100;
w = 100;
R = 10; C = 1e-3;
Y1 = 1/R + j*w*C
Y1 =1.0000e-01 + 1.0000e-01i
Ix2 = Y1*Vs2
Ix2 =1.0000e+01 + 1.0000e+01i
abs(Ix2)
ans =1.4142e+01
angle(Ix2)*180/pi
ans =45
Therefore i x (t) = 10cos(50t + 30o ) 14.142cos(100t + 45o ) A.

Solution 10.19 First of all, write out the given values:

Vs2 = 40 . From KVL Vs1 = 3I1 + j LI1 +

= 200 rad/s, I1 = 0.590 = 0.5 j A, and

I1
+ Vs2 which leads to:
j C

w = 200; I1 = 0.5j; Vs2 = 4;

Vs1 = 3*I1 + j*w*0.04*I1 + I1/(j*w*1e-3) +Vs2
Vs1 =2.5000e+00 + 1.5000e+00i
abs(Vs1)
ans =
2.9155e+00
angle(Vs1)*180/pi
ans =
3.0964e+01
Therefore v s1 (t) = 2.9155cos(200t + 30.964 o ) V.

Solution 10.20 The phasor for the input can be written as

= 1000 rad/s and Vin = 1060 = 5 + j8.66

V. The currents can be obtained easily by applying Ohms law for phasors:

10

IR =

1060
1060
= 0.0260 A, IL =
= 0.04 30 A, and
500
j1000 0.25

IC = 1060 j1000 2 106 = 0.02150 A. Thus

Iin = 0.0260 + 0.04 30 + 0.02150 = 0.028315 A
and
iin (t ) = 0.0283cos(1000t + 15) A
using the following MATLAB code:
Vin = 10*exp(j*60*pi/180)
Vin =5.0000e+00 + 8.6603e+00i
R = 500; L = 0.25; C = 2e-6;
w = 1e3;
IR = Vin/500
IR =1.0000e-02 + 1.7321e-02i
IL = Vin/(j*w*L)
IL =3.4641e-02 - 2.0000e-02i
IC = j*w*C*Vin
IC =-1.7321e-02 + 1.0000e-02i
Iin = IR + IL + IC
Iin =2.7321e-02 + 7.3205e-03i
abs(Iin)
ans =2.8284e-02
angle(Iin)*180/pi
ans =1.5000e+01

Solution 10.21 In MATLAB

Iin = -100*j*1e-3; R = 100;
L = 0.04; C = 2e-6; w = 2500;
VR = R*Iin
VR =0 - 1.0000e+01i
VL = j*w*L*Iin
VL =10
VC = Iin/(j*w*C)
VC =-20
Vin = VR + VL + VC
11

Vin =-1.0000e+01 - 1.0000e+01i

abs(Vin)
ans =1.4142e+01
angle(Vin)*180/pi
ans =-135
Therefore, vin (t) = 14.14cos(2500t 135) V.
Solution 10.22 (a) Here, i1 (t) = 0.6cos (200t) A and vout (t ) = 20sin (200t) = 20cos (200t 90) V.
For = 200 rad/s, the phasors are by inspection: I1 = 0.60 A, Vout = 20 90 V .
(b) Write down the resistor, inductor, and capacitor current phasors, given Vout :
20 90
= 0.6 90 = j0.6
1/0.03
20 90
IL =
= 1 180 = 1
j200 0.1
IR =

IC = 20 90 j200 0.4 10 3 = 1.60

Now, by KCL
I2 = IR + IL + IC I1 = j0.6
where we have substituted the above values of branch currents. The time-domain function is:
i2 ( t) = 0.6cos (200t 90) = 0.6sin (200t ) A

Solution 10.23 (a)

= 400 rad/s and Vin = 200 V. We can easily use the voltage divider formula for

phasors and substitute values to obtain:

Vout =

200
20
Vin =
= 14.14 45
200 + j L
1+ j

in which casevout (t ) = 14.14cos (400t 45) V

(b)

= 250 rad/s and Vin = 20 90 = j20 V. Again, we can easily use voltage division:
Vout =

400

1 Vin =
400 +
j C

20 90
= 14.14 45 V
1 j

Thus, in the time-domain,

vout (t ) = 14.142cos(250t 45) V
12

= 10,000 rad/s and Vin = 100 90 = j100 V. Apply the voltage divider

Solution 10.24 (a)

formula:
Vout =

100
100
Vin =
V = 80 126.87
1
100 + j100 j0.25 in
100 + j L + j C

The steady-state response is thus,

vout (t ) = 80sin (10,000t 36.87) V
The phasor method provides for a much easier way of obtaining the steady-state response.
(b) Here,

= 2500 rad/s and Iin = 0.020 A. Now, apply current division:

IR =

0.01
0.01
I
=
I = 0.01636.86
in
0.01 + j0.0025 j0.01 in
0.01 + j C + 1 j L

By Ohms law:
Vout = 100IR = 1.636.87 V
Therefore
vout (t) = 1.6cos(2500t + 36.87o) V

Solution 10.25 (a) It is easier to find the admittance first:

Yin ( j100) =

+ j C = j0.2 + j12.5 = j12.3

j L
Zin ( j100) = j0.0813
(b) Yin ( j100) =

1
1
=
+ j100C . Hence, in MATLAB,
25 j j100 0.05

C = (1/(25*j) - 1/(j*w*L) )/(j*100)

C = 1.6000e-03

Solution 10.26 (a)

Zin ( j ) =

1
j C

R j L j
j RL
=
+
= 9.975 0.0072
R+ j L
C R+ j L

R = 10; L = 0.1; C = 1e-3; w = 2e3;

Zin = -j/(w*C) + j*w*R*L/(R + j*w*L)
Zin = 9.9751e+00 - 1.2469e-03i
13

abs(Zin)
ans = 9.9751e+00
angle(Zin)*180/pi
ans = -7.1620e-03
(b) As the frequency increases, the capacitor becomes a short circuit and the inductor becomes an open
circuit. Thus, the impedance approaches R. Analytically,
1
R
lim Zin ( j ) = lim
+ lim R
=R
j C
+1
j L

j j RL(R j L)
LR 2
1
R 2 L2
(c) Zin ( j ) =
+
= j 2

+
. It follows that w must satisfy
C
C R 2 + 2 L2
R 2 + 2 L2
R + 2L2

LR2C = R 2 + 2L2 or equivalently

R2
1
=
. For the given component
LR2C L2 LC1 L

CR 2

L
values 1
= 0; hence there is no finite value of frequency for which the impedance is real.
CR 2

SOLUTION 10.27. We note that the input admittance is given by:

Yin ( j ) =

1
100 j 0.1
6
+ j 106 = 4
2 + j 10
100 + j 0.1
10 + 0.01
100
0.1
6 j
= 4
+
j

10
10 + 0.01 2
10 4 + 0.01 2

Thus, must satisfy 10 5 = 10 4 + 0.01 2 or equivalently,

= 107 10 6 = 3000 rad/s.

SOLUTION 10.28. (a) As usual we will use MATLAB.

R1 = 20; R2 = 10;
L = 0.04; C = 0.6e-3;
w = 250;
14

Yin = 1/R1 + 1/(R2 + j*w*L) + j*w*C

Yin = 1.0000e-01 + 1.0000e-01i
Hence
Yin ( j250) = 0.1+ j0.1 S
(b) For this part we observe that
Yin ( j250) =

1
1
+
+ jC = 0.1 0.05 j + j250C
R1 R2 + jL

For this to be real, the imaginary part must be zero, i.e.,

0.05
1
C=
=
= 0.2 mF
250 5000

Solution 10.29 (a) We can derive an expression for the input impedance by noting that it is the series
combination of the resistance and the inductor/capacitor pair connected in parallel. Thus,
1

j L

L
C

j C
=R j
1
1
+j L
L
j C
C

Zin ( j ) = R +

Equating the real and imaginary parts of the given impedance, R = 4 and
L
LC 1/

L
= 2 L = 0.1 H
2L 0.25

(b) At zero inductance, the above reactance is zero. Also, at L = 0.125, the denominator of the above
reactance is zero, which means that the reactance is infinite.

Solution 10.30 (a) First derive an expression for the input impedance as a function of frequency:
Zin ( j

)=5+ j

L+

1
1

= 5 + j L
j C
C

We want the imaginary part to be equal to zero. Thus,

L

1
=0
C

LC

= 2500 rad/s

15

where we have substituted the values of L and C.

The magnitude of the impedance is minimum when the imaginary part is zero, and Zin (j2500) = 5.
(b) Derive an expression for the admittance:
Yin ( ) =

1
1
1
1

+ j C+
= + j C
5
j L 5
L

Again, the imaginary part is equal to zero when

1
or
LC

= 2500, at which point the admittance is

0.2.

Solution 10.31 The input admittance is Yin ( ) =

admittance at

1
+ j C = 0.008+0.004. Equating this to the given
R

= 500yields R = 125 , C = 0.004 or C = 8 F.

Solution 10.32 Derive an expression for the admittance at

Yin ( j ) =

1
j L

= 1000:

0.25 j C
j0.125
= j0.5 +
= 0.2 j0.4 S
0.25 + j C
0.25 + j0.5

Note that this is equivalent to a 0.2 S conductance (i.e. 5 resistance) in parallel with a 2.5 mH inductance
(at the given frequency!). Now, the impedance is:

Zin =

1
= 1+ j2
0.2 j0.4

This is equivalent to a resistance of 1 in series with a 2 mH inductance (at the given frequency).

Solution 10.33 The current is zero when the input impedance of the parallel combination of inductor and
capacitor is infinite. The latter is given by:
1
L
j L
j C
C
ZLC ( j ) = 1
= j
1
+j L
L
j C
C
1
1
The magnitude of this is infinite when r L
= 0 r2 =

C
LC
Vs
Is =
=0
R + ZLC ( j r )

= 15,811 rad/s. Observe that

16

Hence is (t) = 0 at r = 15,811 rad/s. At this frequency, the voltage across the LC tank is equal to the input
voltage (since there is no drop across the resistor).

Solution 10.34 (a) By inspection:

2 j 0.4
j
=
2 + j 0.4 2 +
1
1
Yin = +
= 0.5
R j L
Zin =

0.8
j 0.4
2.5
j

(b) Use the current divider formula, and substitute the given frequency, to obtain:
IL =

R
2 2
2=
= 1 45
R+ j L
2 + j2

and iL (t) = cos(5t 45o ) A.

Solution 10.35 (a) Using voltage division, Vout =

R
j RC
Vin =
V which
j
1+ j RC in
R C

Vout = 90o in tan1 ( RC ) and Vout =

(b) For this part, we need to make sure that tan1( RC) = 45

RC Vin
1+ 2 R 2C 2

RC = 1

= 1/ RC .

(c) At this frequency,

1
RC
Vout
1
RC
=
=
Vin
1
2
1+ 2 2 R 2C 2
R C

1
V
j C
1
1
Solution 10.36 Here, = 1/ RC and C = 1
=
=
. Therefore
Vin
1
+
j
RC
1
+
j
+R
j C
1
VC =
V = 0.707Vm 45
1 + j in
The time-domain function is
1

vC (t ) = 0.707Vm cos
t 45 V
RC
17

Solution 10.37 (a) The magnitude of the capacitor voltage is 10/14.14 = 0.707 times the magnitude of the
input signal. We just showed in the above problem that
VC
1
VC
1
=

=
Vin 1+ j10RC
Vin
1 + 100R 2C 2
And we also showed that the ratio is 0.707 when the frequency is 1/RC. So, C = 1/(10R) = 0.01 F.
(b) Again, from the results of the previous problem, the angle is 45 degrees.

Solution 10.38

= 1000 rad/s and Iin = 245 A. The equivalent admittance is

Yeq ( j

)=

1
1

+ j C
= 0.25 + j (0.25 0.25) = 0.25 S
R
L

In MATLAB
Yeq = 0.25;
Iin = 2*exp(j*pi/4)
Iin =
1.4142e+00 + 1.4142e+00i
Vout = Iin/Yeq
Vout =
5.6569e+00 + 5.6569e+00i
abs(Vout)
ans = 8
angle(Vout)*180/pi
ans = 4.5000e+01
% Using Current Division
IL = (1/(j*1000*4e-3)/Yeq)*Iin
IL =
1.4142e+00 - 1.4142e+00i
abs(IL)
ans = 2
angle(IL)*180/pi
ans = -45
18

Therefore, vout (t) = 8cos(1000t + 45o) V and iL (t) = 2cos(1000t 45o ) A.

(b) If

= 618 rad/s, then

Yeq ( ) =

1
j0.405 + j0.1545 = 0.25 j0.25 = 0.3536 45
4

and
Vout = Iin/Yeq
Vout =
-3.4779e-04 + 5.6565e+00i
abs(Vout)
ans =
5.6565e+00
angle(Vout)*180/pi
ans =
9.0004e+01
IL = Iin*(1/(j*618*4e-3))/Yeq
IL =
2.2882e+00 + 1.4069e-04i
abs(IL)
ans =
2.2882e+00
angle(IL)*180/pi
ans =
3.5228e-03
Therefore, vout (t) = 5.657cos(1000t + 90o ) = 5.657sin(1000t) V and iL (t) = 2.288cos(1000t) A.

Solution 10.39 Write the input phasor:

= 1000 rad/s and Iin = 0.01 260A.

w =1000;
Iin = 0.01*sqrt(2)*exp(j*60*pi/180)
Iin =
7.0711e-03 + 1.2247e-02i
Yeq = 1/500 +1/(j*w*0.25) +j*w*2e-6
Yeq =
2.0000e-03 - 2.0000e-03i
Vin = Iin/Yeq
19

Vin =
-1.2941e+00 + 4.8296e+00i
abs(Vin)
ans =
5
angle(Vin)*180/pi
ans =
1.0500e+02
IR = Iin*(1/500)/Yeq
IR =
-2.5882e-03 + 9.6593e-03i
abs(IR)
ans =
1.0000e-02
angle(IR)*180/pi
ans =
1.0500e+02
IL = Iin*(1/(j*w*0.25))/Yeq
IL =
1.9319e-02 + 5.1764e-03i
abs(IL)
ans =
2.0000e-02
angle(IL)*180/pi
ans =
1.5000e+01
IC = Iin*j*w*2e-6/Yeq
IC =
-9.6593e-03 - 2.5882e-03i
abs(IC)
ans =
1.0000e-02
angle(IC)*180/pi
ans =
-1.6500e+02

20

Therefore, vin (t) = 5cos (1000t 105) V, iC ( t) = 0.01cos(1000t 165) A, iL ( t) = 0.02cos (1000t + 15)
A, and iR (t) = 0.01cos(1000t + 105) A.

SOLUTION 10.40.

V1 =

= 4 . Using voltage division

R
R+

1
j C

+j L

Vin =

2
(8 j ) = 5.657 45o
24 j +2j

Converting back to time:

vout (t ) = 5.657cos( 4t 45) V

SOLUTION 10.41.

= 25 rad/s, Vs = 100 V. By voltage division,

j

VC =

0.02 25
100 = 2 j = 2 90
10 + j0.08 25 j 0.02 25

Thus, vC (t ) = 2cos(25t 90) V.

2
Vout
1
1

SOLUTION 10.42.
=
= 0.2 25 = 3 + j 8
.
Vin
4C
3 + j8 + 1 j 4C

1 2
1

Thus 8
= 25 9 C =
= 0.0625 F.

4C
16

SOLUTION 10.43. Here,

= 3.33 103 rad/s and Vin = 500 V. Using phasors,

VR =

400
400

50 = 4036.897 V

3.33 10 3 10 6

and
Vout =

100VR j C
100VR
=
= 398532.1 V
1
j100
C
+
1
100 +
j C
21

Hence, vout (t) = 3985cos(3.33 10 3 t + 32.1) V.

SOLUTION 10.44. Here,

Z1 =

= 104 rad/s, Iin = 0.010 A.

100 j 0.1
= 1+ j1000 VL = ZL Iin = 0.01 ZL = 0.01 + j10
100 + j 0.1

Now, in MATLAB
w=1e4; R = 100;L = 0.1; C = 0.1e-6;
Z1 = R*j*w*L/(R+j*w*L)
Z1 = 9.9010e+01 + 9.9010e+00i
Iin = 0.01;
VL = Z1*Iin
VL =9.9010e-01 + 9.9010e-02i
Z2 = 1/(1/R + j*w*C)
Z2 = 9.9010e+01 - 9.9010e+00i
VC = Z2*VL
VC = 9.9010e+01 + 1.7764e-15i
abs(VC)
ans = 9.9010e+01
angle(VC)*180/pi
ans = 1.0280e-15
Thus, vC (t ) = 99cos(10000t) V.

SOLUTION 10.45. Here = 40 rad/s and Vin = 1200. This problem is best done in MATLAB using
parallel impedance computation, voltage division, and Ohm's law for phasors:
R1 = 500; R2 = 80;
C = 0.1e-3; L = 2;
w = 40; Vin = 120;
Z1 = R1/(j*w*C)/(R1 + 1/(j*w*C))
Z1 = 1.0000e+02 - 2.0000e+02i
Z2 = R2*j*w*L/(R2 + j*w*L)
Z2 = 4.0000e+01 + 4.0000e+01i
Use voltage division
VC = Z1*Vin/(Z1+Z2)
VC = 1.2212e+02 - 3.1858e+01i
abs(VC)
ans = 1.2621e+02
angle(VC)*180/pi
ans = -1.4621e+01
22

 Use voltage division and Ohm's law for inductors

VL = Z2*Vin/(Z1+Z2)
VL =-2.1239e+00 + 3.1858e+01i
IL = VL/(j*w*L)
IL =3.9823e-01 + 2.6549e-02i
abs(IL)
ans = 3.9911e-01
angle(IL)*180/pi
ans = 3.8141e+00
Therefore, vC (t) = 126.21cos(40t 14.621o ) V and iL (t) = 0.399cos(40t + 3.814) A.

SOLUTION 10.46. Here

MATLAB.

= 40 rad/s and Iin = 0.1200 A. This problem is best solved using

w = 40; Iin = 0.12;R = 5;

C = 0.004; L = 0.1;
Y1 = 1/R + j*w*C
Y1 = 2.0000e-01 + 1.6000e-01i
Y2 = 1/R + 1/(j*w*L)
Y2 = 2.0000e-01 - 2.5000e-01i
% USING CURRENT DIVISION
IL = Iin*Y2/(Y1 + Y2)
IL = 7.3171e-02 - 5.8537e-02i
abs(IL)
ans = 9.3704e-02
angle(IL)*180/pi
ans = -3.8660e+01
% AGAIN USING CURRENT DIVISION
IC = Iin*Y1/(Y1 + Y2)
IC = 4.6829e-02 + 5.8537e-02i
% USING OHM'S LAW FOR CAPACITORS
VC = IC/(j*w*C)
VC = 3.6585e-01 - 2.9268e-01i
abs(VC)
ans = 4.6852e-01
angle(VC)*180/pi
ans = -3.8660e+01

SOLUTION 10.47 Here,

= 25 rad/s and IS = 20 A. Now perform a source transformation. The

combination of current source in parallel with resistor is changed into a voltage source in series with the
same resistor. The voltage source value is: VS = 1 IS = 20 V. Apply Ohms law to obtain:
I1 =

20
= 0.5 + j0.5 = 0.70745
2 + 1 j25 0.02
23

Hence, i1 (t) = 0.707cos(25t + 45o ) A.

SOLUTION 10.48 Apply a source transformation to obtain:

The impedance of the inductor branch is ZL ( j 4) = 2 + j4 0.5 = 2 + j2 and

YL ( j 4) =

1
= 0.25 j0.25 S. Now, the total admittance seen by the source:
ZL ( j4)
Yeq ( j 4) =

Thus, by inspection, VC =

1
+ j0.25 + YL ( j4) = 0.75 S
2

60
= 80 V and IC = 0.2590 80 = 290 A. Therefore
0.75

iC (t) = 2cos(4t + 90o ) = 2sin(4 t) A.

SOLUTION 10.49. Apply a source transformation to obtain:

Then, by inspection, Yeq ( j500) =

1 1 1
1
+ +

= 1.5 S. Thus,
2 1 j2 j2

24

VC =

2.50
= 1.6670 V = VL IL = 1.667/ j2 = 0.833 90 A
1.5

In the time-domain:
vC (t ) = 1.667cos(500t) V and iL ( t) = 0.833cos(500t 90) = 0.833sin(500t ) A.

25

PROBLEM SOLUTIONS CHAPTER 10

SOLUTION 10.50. The input voltage phasor is

= 2000 rad/s and VS = 200 V. Now, do a source

transformation on the phasor circuit:

where
V
200
IS = S =
= 1 90 = j A
ZL j2000 10 10 3
Now,
Yeq =

1
1
+
+ j2000 50 10 6 = 0.070745 S
20 j20

and
I
1 90
VC = S =
= 14.14 135 V
Yeq 0.070745

SOLUTION 10.51. Use superposition. First, find response to current source using circuit below:

3 ( j 3)
Vx_1 = Is1ZRC = Is1
= 20 2.121 45 = 4.242 45
3 j3
Now, find the response due to the voltage source using the following circuit:
1

The voltage across the inductor is the same as the input source, and this voltage divides between the series
combination of capacitor and resistor:
Vx_2 =

j3
j3
Vs2 =
390o = 2.121 135o V
3 j3
3 j3

Combining the two contributions implies that:

Vx = 4.242 45o + 2.121 135o = 4.74 71.6o V

SOLUTION 10.52. (a) As stated, VL = aVs 1 + bIs 2 . To find a, set Is 2 = 0 and use voltage division:
VL =

j30
V = 0.5Vs 1 = aVs 1
j30 + j30 s 1

To find b, set Vs 1 = 0 and use parallel impedance and Ohm's law:

VL = ( j30/ / j30)Is 2 = j15Is 2 = bIs 2
Hence
VL = 0.5Vs 1 +j15Is 2
(b) For this part, Vs 1 = 10 and Is 2 = 0.5 90o . Hence from the formula,
VL = 0.5Vs1 + j15Is 2 = 0.5 10 + j15 0.5( j) = 5 + 7.5 = 2.5 V
2

Therefore vL(t) = 2.5 cos(100t) V.

SOLUTION 10.53. In this problem, we can make use of the linearity property for phasors. Specifically,
from the given information, we can write
V1 2045
V
590
=
= 245 = a and 1 =
= 0.545 = b
Iin 100
Vin 1045
Hence,
V1 = aIin + bVin
Substituting the new values of input current and voltage, we obtain:
V1 = 100 + 1045 = 18.4822.5 V

SOLUTION 10.54. (a) For Vout to be zero, we want

RC

j L
L
=

R
R
=
C
L
j
RL + j L
C
RC
C
(b)

Substituting RC C = 2 s and RL = 3 into the above expression gives: L = RCCRL = 6 H.

(c) The bridge circuit can be represented by an impedance Zbridge(j). The voltage that appears across the
bridge, say Vbridge, is obtained by voltage division. Hence, by the voltage substitution theorem, the problem
may be solved as in part (a) with this new source voltage Vbridge appearing across Zbridge(j).

SOLUTION 10.55 The input phasor is: = 1000 rad/s and Iin = 245 A, assuming peak value. First
compute
Yth ( j103 ) = 0.25

j
+ j1000 0.25 10 3 = 0.25S Zth ( j10 3) = 4
3
1000 4 10

Then,
Voc = Iin 4 = 845 V

in which case voc (t) = 8cos(10 3 t + 45o ) V. The final equivalent is a voltage source (having value Voc ) in
series with a resistance of 4 .

SOLUTION 10.56. (a) First note that the frequency is given in Hz, so, = 1281.77rad/s and Iin = 100
A. Then, in MATLAB,
R = 0.25; L = 1.17e-3; C = 520e-6;
w = 2*pi*204;
Yin = j*w*C +1/(R + j*w*L)
Yin =
1.0815e-01 + 1.7737e-02i
Zin = 1/Yin
Zin =
9.0039e+00 - 1.4766e+00i
abs(Zin)
ans =
9.1242e+00
angle(Zin)*180/pi
ans =
-9.3134e+00
Therefore Zth = 9.1242 9.313 . Finally, Voc = ZthIin = 91.2 9.313 V.
(b) Now, the circuit looks like the following:

Simple voltage division can yield:

VL =

ZL
V = 46.2190o v L ( t) = 46.2cos(1281.77t) V
ZL + ZTH oc

SOLUTION 10.57. For this problem we short the V-source and compute Zth and use voltage division to
find Vo c . Specifically
Zth =

= 103 j10
1
jC +
0.1+ jL

and
Voc =

1
jC

2 = 200 j V
1
+ 0.1 + jL
jC

where
w=1000;
L = 0.01;
C = 0.1e-3;
Zth = 1/(j*w*C + 1/(0.1 + j*w*L))
Zth = 1.0000e+03 - 1.0000e+01i
Voc = 2*(1/(j*w*C))/(0.1 + j*w*L + 1/(j*w*C))
Voc =
0 - 2.0000e+02i
To compute the load voltage, define
Zload = 103 + j10 2 = 103 + j10
Again using voltage division and MATLAB we have
Zload = 1e3 + j*10;
Vload = Voc*Zload/(Zload + Zth)
Vload =
1.0000e+00 - 1.0000e+02i
magVload = abs(Vload)
magVload =
1.0000e+02 (volts)

SOLUTION 10.58. Here,

= 3.33 103 rad/s and Vin = 500 V. Here we note that we already found

Voc in Problem 10.43. Thus, Voc = 398532.1 V. In order to find Zth, we introduce a fictitious 1 A
current source at the A and B terminals:

Now, we note that the VR voltage phasor is zero. Thus, the dependent source has zero volts across it. This
way, the temporary current source sees the parallel combination of a resistor and a capacitor:

VAB =

100

j
3.3310 0.2510 6
3

100 j1200

= 99.65 4.77 V

Thus, Zth = 99.65 4.77 . The Thevenin equivalent consists of Voc in series with
Zth = 99.65 4.77 = 99.3 j8.2865 . Thus the Norton equivalent is the parallel combination of Zth
and the current source with value

Isc =

Voc
398532.1
=
= 39.9936.91o A.
Zth 99.65 4.77

SOLUTION 10.59.
= 10,000 rad/s and Vin = 100 V. Again, we have already found Voc in Problem
10.44: Voc = 990 V. Now, to compute the impedance, we introduce the temporary current source of 1
A:

Again, the inductor has no voltage across it. So, the dependent source generates no current. Hence, the
independent source sees the parallel combination of a resistor and a capacitor:

VAB =

100 ( j1000)
= 99.5 5.71 V
100 j1000

Thus Zth ( j104 ) = 99.5 5.71 . So, the Thevenin equivalent is the series combination of the Voc
source and the above Zth ( j104 ) = 99.5 5.71 .

SOLUTION 10.60. Inject a current source at terminals A and B. Then, write a KCL equation at node A:
IR = IS I1 VR = IS I1 VC = VAB = VR + I1 = IS I1 + I1 = IS
Since the voltage across the current source is equal to its current, the equivalent impedance across this
current source is 1 .

SOLUTION 10.61. Inject a current source as usual. Then, write Ohms law for phasors for the equivalent
series RLC circuit. Note that the controlling current for the dependent source is the input current:

VAB = j0.01 200 IS

j
I 2IS = (2 + j)IS
200 0.005 S

V
Therefore, Zth = AB = 2 + j .
Is

SOLUTION 10.62. Inject a current source Is :

Apply KCL:

j2I 6I
6
= Is I I =
I . Now, VAB is the voltage across the inductor:
6
j2 s
VAB =

Solution 10.63.

6 j2
I = 6Is Zth = 6 .
j2 s

= 2000 rad/s and VS = 250 V. Now, inject a current source, and express VAB as a

function of this current source and Voc :

Now, write two nodal equations at A and the top of the dependent current source:
V1 VS
10 3

+ gm VAB + j C( V1 VAB ) = 0

V
j C(VAB V1 ) + AB = IS
j L
Substituting values, these can be cast into the following matrix equation:

j 3 j V1 10-3 VS
=

j
j VAB IS

10

3 1 +

This can be solved in MATLAB to obtain:

V1
2.5 j2.5 5 + j10 10-3 VS
=
100

V
2.5 + j2.5

5
AB

IS
Thus
VAB = (0.25 + j0.25)Vs + 500Is = 500Is + 6.25 + j6.25
Therefore Zth = 500 and Voc = 6.25 + j6.25 = 8.83945o V. For the Norton equivalent we need
Isc =

Voc 8.839
=
45o = 0.0176845o A.
Zth
500

SOLUTION 10.64. Inject an upward current source, IS2 , at VAB. Then, write the following two nodal
equations at Vx and VA: let R1 = R2 = R, then
Vx
+ j C1Vx + j C2 (Vx VAB ) = IS
R
V
j C2( VAB Vx ) + gm Vx + AB = IS2
R
which after grouping terms becomes
1

+ j C1 + j C2 Vx j C2VAB = IS
R

( j

C2 + gm )Vx + j C2 +
V =I

R AB S2

In MATLAB
R = 100e3; C1 = 1e-9; C2 = 1e-10;
gm = 0.1e-3; w = 1e3;
Nodal = [1/R+j*w*C1+j*w*C2 -j*w*C2;-j*w*C2+gm j*w*C2+1/R]
Nodal =
1.0000e-05 + 1.1000e-06i
0 - 1.0000e-07i
9

1.0000e-04 - 1.0000e-07i 1.0000e-05 + 1.0000e-07i

Nodalinv = inv(Nodal)
Nodalinv =
9.5680e+04 - 2.0070e+04i 2.1024e+02 + 9.5470e+02i
-9.5449e+05 + 2.1120e+05i 9.7783e+04 - 1.0523e+04i
Thus
0.21+ j0.955 40 10-6
Vx
3 95.68 j20
=
10

V
954.5 + j211.2 97.78 j10.52
AB

IS2
Therefore
VAB = 38.18 + j8.448 + (97.8 j10.5) 103 IS2
from which we identify
Voc = 38.18 + j8.448 = 39.1167.5o V
and Zth = (97.8 j10.5) 103 = 98.35 103 6.14 o .

SOLUTION 10.65. We solve this problem by the method illustrated in example 6.3 where a fictitious
source is applied and the repsonse is calculated. One can either apply a voltage source or a current source
(see figure 6.10 a and b). Generally speaking, neither choice can be claimed as superior to the other. But
for a specific circuit, one choice can lead to a much simpler solution than the other. To illustrate this point,
we solve the problem with both choices below. Note that although the first method is much simpler than
the second, it lacks the generality. If one more resistor were inserted into the circuit, the simplicity of
solution of solution may disappear totally, whereas the second method will proceed with very few changes.
Method 1. For this solution we apply an arbitrary voltage source, labeled in phasor form as Vs as
indicated in the circuit below.

For this circuit we will compute an equation of the form ofequation 6.11:

10

IA =

1
V Isc
Zth s

(1)

By inspection of the circuit,

V
IA = IC + I L 1 = jCVs + L 1 = j2Vs jVL 1
jL
But

VL = 0.25IC + Vs = 0.25 jCVs + Vs = (0.5 j + 1)Vs

Substituting (3) into (2) produces

IA = (0.5 + j )Vs 1

(2)

(3)

(4)

By comparing (4) with (1), we obtain the answers for the Norton equivlent circuit: Is c = 1 A and Zth =1/(0.5 + j) =
0.4 j0.8 .
Method 2: For this method we apply an arbitrary current source, labeled in phasor form as Is as indicated
in the circuit below.

Notice that we have added a current label Ix as we plan to use modified nodal analysis method to obtain the
desired answer. For this we will compute an equation of the form
VAB = Zth Is + Voc
Because of the addition of Ix , we can write the modified nodal equations more or less by inspection:
1/ jL
0

0
1 VL j
0
1 VL 1

jC
1 VAB = 0
2j
1 VAB = Is
(1+ j0.25C) 0 I x 1 1 0.5 j 0 I x 0

where from MATLAB,

11

w=2000;
L = 0.5e-3;
C = 1e-3;
Y11 = 1/(j*w*L)
Y11 =
0 - 1.0000e+00i
Y22 = j*w*C
Y22 =
0 + 2.0000e+00i
Y32 = -(1+j*0.25*w*C)
Y32 = -1.0000e+00 - 5.0000e-01i
Solving the equations in MATLAB produces,
A = [-j 0 1;0 2*j -1; 1 -1-0.5*j 0];
Ainv = inv(A)
Ainv =
8.0000e-01 - 6.0000e-01i 8.0000e-01 - 6.0000e-01i 1.6000e+00 + 8.0000e-01i
4.0000e-01 - 8.0000e-01i 4.0000e-01 - 8.0000e-01i 8.0000e-01 + 4.0000e-01i
1.6000e+00 + 8.0000e-01i 6.0000e-01 + 8.0000e-01i -8.0000e-01 + 1.6000e+00i
Multiplying the second row of Ainv times the right-most vector of our equations produces
VAB = (0.4 j 0.8)Is + (0.4 0.8j )
This implies that Zth = 0.4 j0.8 and Vo c = 0.4j0.8 V. For the Norton equivalent we need
V
Isc = oc = 1 A
Zth

SOLUTION 10.66. (a)

= 1000 rad/s and VS = 2045 V. By voltage division:

Vout =

1
2045 = 14.140 V IL = 14.140 A
1+ j

(b) At dc, Vout = Vin. We want the frequency at which Vout = 0.1Vin . Thus, we want:

12

Vout
1
1
1
=
= 0.1
=

2
6
Vin
1+ j 0.001
100
1+
10

= 9950 rad/s

SOLUTION 10.67. Z = 25 j20 . By KCL:

VZ 140
VZ
V + 890
+
+ Z
=0
j15
25 j20
j20
This equation simply needs to be manipulated in order to obtain:
VZ = 41.35 73.46

SOLUTION 10.68. In this problem, we note that the impedance, jX, is in series with the parallel RLC
circuit to the right. Thus, all we need to do is to find an expression for the equivalent impedance of the
parallel RLC circuit:
1
j

+ j0.025 ZRLC = 19.2 + j14.4

30 20
Now, the total impedance seen by the source is 19.2 + j14.4 +jX. Therefore, for this to be real, the
YRLC =

unknown reactance has to be 14.4 . Also, the input current now is I = 96/19.2 = 5 A. Hence
i(t) = 5cos(10t) A.

1
SOLUTION 10.69. First, compute the current through the series RC section: IC =
A.
1 j
Now, by KCL, we can write
0 = IC Ix +

SOLUTION 10.70.
(a) Since Yin =

1 2Ix
1
1 2Ix
=
Ix +
j
1 j
j

Ix = 0.3 + j0.1= 0.31618.44 o A

= 20 rad/s and Yin = 0.05 + j0.0866 S.

1
j

+ j C , equating the real parts of the above two expressions implies that R =
R
L

1/0.05 = 20 .
(b) Similarly, equating the imaginary parts and substituting, we obtain:
13

j C
= j0.0866 L = 3.73 H
L
(c)

I
2030
VC = in =
= 200 30 V vC ( t) = 200cos 20t 30o V
Yin 0.05 + j0.0866
(d)
IL =

20030
= 2.68 120A i L ( t) = 2.68cos 20t 120o A
j20 3.73

1
(e) Voc is just the voltage VC, which was obtained in (c), and Zth =
= 10 60 .
Y
in

(f) Zth = 5 j8.66 . This is equivalent to a series combination of a 5 resistance and a 5.77 mF
capacitance at the given frequency = 20 rad/s.

SOLUTION 10.71. (a) Vs = 2 + j0V,

= 1000 rad/s. By KCL

VA
VA
V Vs
+
+ A
= 0 VA = 0.5 j0.5 = 0.707 45 V
j1.33 2 + j2
4
The time-domain expression is:

vC (t ) = v A (t) = 0.707cos 1000t 45o V

(b) IL =

0.707 45
= 0.25 90 A implies that iL ( t) = 0.25cos 1000t 90o = 0.25sin(1000t) A.
2 + j2

(c) We already determined Voc in part (a). Now, turn off the source to compute the equivalent impedance:
Yth = 0.75 j +

1
+ 0.25 = 0.5 + j0.5 Zth = 1 j . This is the series connection of a 1 resistor
2 + j2

with a capacitor of value C =

= 10 3 F. This completes the definition of the Thevenin equivalent.

SOLUTION 10.72. In MATLAB,

R = 5; L = 1e-3; C = 20e-6;Vs1 = 5; Is2 = 0.5*j;
G = 1/R; w = 10e3;
YL = 1/(j*w*L)
YL =
0 - 1.0000e-01i
14

YC = j*w*C
YC =
0 + 2.0000e-01i
Hence
G(VC Vs1 ) + YC VC + YL (VC Vs1 2VR ) = Is2
Substituting for VR,
G(VC Vs1 ) + YC VC + YL (VC Vs1 2(Vs1 VC )) = Is2
Therefore

(G + YC + 3YL )VC = (G + 3YL )Vs1 + Is2

Again, using MATLAB,
a = G + YC + 3*YL
a=
2.0000e-01 - 1.0000e-01i
b = (G+3*YL)*Vs1 + Is2
b=
1.0000e+00 - 1.0000e+00i
VC = b/a
VC =
6.0000e+00 - 2.0000e+00i
abs(VC)
ans =
6.3246e+00
angle(VC)*180/pi
ans =
-1.8435e+01
Hence, vC (t) = 6.325cos(104 t 18.44o ) V.

SOLUTION 10.73. Denote by vC1 , the node voltage of the 2.5 mF capacitor . Note that at = 800 rad/s,
o

Vs = 200 V. From this we write a set of nodal equations by inspection after observing the following
from MATLAB:

15

w = 800;
C1 = 2.5e-3;
L = 1.25e-3;
Y1 = 0.5+j*w*C1-j/(L*w)
Y1 = 5.0000e-01 + 1.0000e+00i
Yoff=j/(L*w)
Yoff =
0 + 1.0000e+00i
Y2 = 0.25+j*w*3.75e-3 - j*w*1.25e-3
Y2 = 2.5000e-01 + 2.0000e+00i
This information leads to the following matrix nodal equation:
j
0.5 + j
VC1 0.5Vs
=
j
0.25 + j2 Vout 0.25Vs

To solve these equations we again use MATLAB:

A = [Y1, Yoff;Yoff, Y2]
A=
5.0000e-01 + 1.0000e+00i
0 + 1.0000e+00i
0 + 1.0000e+00i 2.5000e-01 + 2.0000e+00i
b = [0.5*20;0.25*20]
b=
10
5
Vnodes = inv(A)*b
Vnodes =
7.1141e+00 - 6.9799e+00i
-3.6242e+00 + 5.3691e-01i
magVnodes = abs(Vnodes)
magVnodes =
9.9664e+00
16

3.6637e+00
angVnodes = angle(Vnodes)*180/pi
angVnodes =
-4.4454e+01
1.7157e+02
Therefore
vout (t) = 3.664cos(800t + 171.57o ) V

SOLUTION 10.74. For this problem we use loop analysis with loops indicated in the figure below.

Since there are no controlled sources, we can write down the loop equations by inspection:
180 + 103.92
10 + 9 j 6 + 5j I A 120 120 120o
=
6 + 5 j 12 + 9 j I =

C 120120 o 120 120o j207.85

The solution of this equation is done in MATLAB as follows:
b1=120-120*exp(-j*2*pi/3)
b1 = 1.8000e+02 + 1.0392e+02i
b2=120*exp(j*2*pi/3)-120*exp(-j*2*pi/3)
b2 =
0 + 2.0785e+02i
17

A = [10+j*9, 6+j*5;6+j*5,12+j*9]
A=
1.0000e+01 + 9.0000e+00i 6.0000e+00 + 5.0000e+00i
6.0000e+00 + 5.0000e+00i 1.2000e+01 + 9.0000e+00i
I=inv(A)*[b1;b2]
I=
1.4472e+01 - 1.3469e+01i
4.2926e-01 + 1.7703e+01i
% Please note that using the commands I=inv(A)*[b1,b2]'
% will lead to the wrong answer because a conjugate is
% inserted along with the transpose.
magI = abs(I)
magI =
1.9770e+01
1.7708e+01
angleI = angle(I)*180/pi
angleI =
-4.2944e+01
8.8611e+01
IB = -I(1)-I(2)
IB = 1.8288e+00 + 1.7234e+01i
magIB = abs(IB)
magIB = 1.7331e+01
angleIB = angle(IB)*180/pi
angleIB = 8.3943e+01
Changing the sign on each source amounts to multiplying its value by "1". This means that all
o

magnitudes remain the same, but there is a 180 phase shift for each current, i.e., add 180 to each current
angle.
18

SOLUTION P10.75. For this problem we have both a transient component to the response and a steady
state component. The steady state component is computed in the usual way because the circuit is stable,
i.e., the time constant is positive. Once the steady state part is computed, we use initial conditions to obtain
the coefficient B in the response.
Part 1: Compute steady state response. For this we use MATLAB,
R = 0.5; L = 0.866;
Vs = 10;
w = 1;
Zin = R + j*w*L
Zin = 5.0000e-01 + 8.6600e-01i
IL = Vs/Zin
IL = 5.0002e+00 - 8.6604e+00i
magIL = abs(IL)
magIL = 1.0000e+01
angIL = angle(IL)*180/pi
angIL = -5.9999e+01
Hence
iL (t) = 10cos(t 60o ) + Be0.577t A
Part 2: From the initial conditions we have

iL (0) = 1 = 10cos(t 60 o ) + Be0.577t

]t =0 = 10cos(60o ) + B = 5 + B

Hence B = 4. It follows that

iL (t) = 10cos(t 60o ) 4e 0.577t A

SOLUTION 10.76. KCL dictates that:

IL + IR + IC = IS IL = IS IR IC
We can perform this sum graphically as follows:

19

j2

-IC
IL=-j
-IR

SOLUTION 10.77. (a) Note that VS = VR + VL, but that VL leads VR by 90 degrees. Similarly, IS = IR +
IC, but IC leads VS by 90 degrees. Also note that the inductor current is also IR, and the capacitor voltage is
V S.

IS

IC
VS
VL
IR

VR

(b) Using graph paper to construct the phasor diagram to scale, we find the difference between the phase
angles of Is and Vs is zero.

Solution 10.78. First note that VC, the capacitor voltage, will lag IS by 90 degrees. Now, VC plus the
unknown element voltage should result in a vector that runs diagonally between the two vectors. From the
following illustration, it follows that the unknown voltage should have the same phase as the input current:

IS
VS
VC

VC

Vu
20

This means that the unknown element is a resistor. The 45o phase difference implies that VC = Vu or
Is
1
1
= Is R . Therefore R =
= 3
= 1000 .
C
C 10 10 6

SOLUTION 10.79. The student can construct the phasor diagram using graph paper. The diagram is going
to look like that in the problem statement, except that the proper lengths and angles will be used.

SOLUTION 10.80. As the frequency approaches infinity, the capacitor shorts and the inductor opens. So,
the output voltage is zero. As the frequency approaches zero, the capacitor opens, but the inductor shorts,
so the output is also zero. A plot of the complete response is shown below. (Note that the magnitude
response at 10rad/s is infinite):
L = 0.04; C = 0.25;
w = 0: 30/300:30;
% Vout = Zin * Iin
Zin = j*w*L ./(j*w*L*j.*w*C + 1);
plot(w, abs(Zin))
grid
ylabel('Magnitude Zin')
xlabel('Frequency in rad/s')
plot(w,angle(Zin)*180/pi)
grid
xlabel('Phase in degrees')
ylabel('Phase in degrees')
xlabel('Frequency in rad/s')

21

100

25

80
60

Phase in degrees

Magnitude Zin

20

15

10
TextEnd

40
20
0
-20
TextEnd

-40
-60

-80
0

10
15
20
Frequency in rad/s

25

30

-100

10
15
20
Frequency in rad/s

25

30

SOLUTION 10.81. At infinite frequency, the resistor current is zero (because the inductor opens). So, the
output voltage is zero. At DC, the inductor is short, and the output voltage is equal to the input voltage. The
plot of the frequency response is shown below (a logarithmic x-axis is used):

SOLUTION 10.82. The circuit inside the black box is

22

At DC, the capacitor is an open circuit. Thus, the voltage across the resistor is 1mAR. But we know that
this voltage is 1 from the graph. This means that R = 1 k. Now, in general for the above diagram:
V
R
=
I 1+ j RC
The magnitude of this function is R/sqrt(2) when

= 1/ RC . Substituting the frequency from the graph

(1000 rad/s), we get C = 1 F.

Solution 10.83. The admittance of the parallel RLC circuit is:

1
1
+
+j C
R j L
1
V
Zin =
= s
Yin Is

Yin =

The function we want to find the frequency response for is nothing but the input admittance of the circuit.
Using MATLAB, the following plot can be obtained:
R = 100; L = 0.1; C = 1e-3;
w = 0:0.5:300;
w = 0.01:0.5:300;
Yin = 1/R + 1. ./(j*w*L) + j*w*C;
Zin = 1 ./Yin;
plot(w,abs(Zin))
grid
xlabel('Frequency in rad/s')
ylabel('Magnitude Zin')

23

100
90
80
Magnitude Zin

70
60
50
40
TextEnd

30
20
10
0

50

100
150
200
Frequency in rad/s

250

300

250

300

plot(w,angle(Zin)*180/pi)
grid
xlabel('Frequency in rad/s')
ylabel('Phase Zin in degrees')

100
80

Phase Zin in degrees

60
40
20
0
-20
-40

TextEnd

-60
-80
-100

50

100
150
200
Frequency in rad/s

24

SOLUTION 10.84. The circuit inside the box is a series RLC circuit. It cannot be a parallel RLC, because
as per problem 83, the admittance of a parallel RLC does not approach zero as w approaches infinity. Thus,
I
1
= Yin =
1
V
R+
+j L
j C
The resonance frequency is 50 rad/s and is determined by 1/sqrt(LC). Given L = 0.4 H, C = 1 mF.
To obtain R, we make use of the fact that, from the given graph at = 57 rad/s, the current
magnitude is approximately 0.2 times the peak magnitude. Therefore
1
R2

1 2

+ 57L

57C

0.2

R2

1 2

+ 50L

50C

0.2
R

Hence
2

R2
R + 57 0.4
= 25R 2
=
3

0.04
57 10
2

R2 + 57x0.4 -

From which R =

1
R2 = 25R 2
2
57x0.001
0.04

5.2561
= 1.0728 .
24

SOLUTION 10.85. Create three mesh currents in the three planar loops. All currents are clockwise: I1 in
the voltage source loop, I2 in the top bridge loop, and I3 in the bottom one. The three mesh equations are:
V I1R1 R2 (I1 I2 ) R3 (I1 I3 ) = 0
R2(I2 I1) +

1
I + Rmeter (I2 I3 ) = 0
j C1 2
1
R3(I3 I1 ) + Rmeter (I3 I2 ) +
I =0
j C2 3
The plots of the magnitude and phase of VB VC = Rmeter (I3 I2 ) are shown in the text.
25

SOLUTION 10.86. We can see all ranges by plotting on a logarithmic scale:

Note that the output will decay when we start to reach the bandwidth of the op-amp. In other words, the
inverting amplifier says that the output is 1 times the input (provided the op amp works properly). Once
the op amps gain starts dropping, the output voltage also decays with it.

SOLUTION 10.87. Correction: Change the 0.01 F capacitor to 1 F. (a) For this part consider the
diagram below,

From the problem statement, = 320 rad/s, and Iin = 0.010 A. Observe that the 50 k resistor input
to the inverting op amp terminal is in parallel with the 100 resistor because of the virtual ground at the op
amp terminals. However, for all practical purposes, this has no effect on the 100 resistor, hence from
Ohm's law
26

VL =

100 j 0.1
I = (50.265 + j50) 0.01 = 0.50265 + j0.5
100 + j0.1 in

From the inverting op amp configuration,

V =-10VL = 5.0265 j5
From voltage division,
VC =

103
o

3
9 V = (0.49735 j0.5) (5.0265 j5) = 5179.7 V
10 + j10

Therefore
vC (t) = 5cos(320t + 179.7o )V

Parts (b) and (c). For the SPICE simulation we have the following circuit in B -SPICE:

which leads to the response below

27

Prb10-87-Small Signal AC-2

+1.000

(V)

+10.000

+100.000

Frequency
+1.000k

(Hz)

+10.000k

+500.000m

+400.000m

+300.000m

+200.000m

+100.000m

+0.000e+000

MAG(V(IVM))

The magnitude at 160 Hz is 0.499 for a 1 mA current input. Thus a 10 mA input current should lead to
4.99 V by linearity which approximates the 5 V computed analytically in part (a). Hence with a 15 V
saturation limit, the input magnitude may increase by a factor of 3 to 30 mA.

SOLUTION 10.88. (a)

= 400 rad/s and Vin = 103 90 V. By the virtual short property:

Iin =

Vin
= j CVin
1/ j C

All this current flows through the 1 M resistor: Vin = j C106 Vin = 0.4 180. Thus,
vout(t) = 0.4cos(400t) V
(b)

= 200 rad/s, Vin = 1030 V, and Vout = 103 90 V. Again,

Vout = j10 3 = j C 4 105 Vin = j80 10 3C C = 12.5 nF

28

SOLUTION 10.89. (a)

= 800rad/s and Vin = 1 90V Iin =

Vin
. No current flows into Op106

Amp terminals:
Vout = Iin

1
j ( j)
=
= 1.25 103 V
j C
800

Thus, vout(t) = 1.25cos(800t) mV.

(b) Again, Iin =

Vin
1
j ( j)
A, and Vout = Iin
10 =
C = 2.5 nF .
3
j C
200 10
C2 105

Solution 10.90 (a)

= 2 700rad /s and Vin = 10 V Iin =

j

1
A. Further,
150 103
j

150 103 C
150 103 C
1
o
Vout =
j I in = 150 103
j = 0.001590 V
3
3
150 10 C
150 10 C
Thus, vout(t) = 1.5cos(2700t + 90 o) mV.
(b)

(c) The output lags the DC response by 45 degrees (note that at DC, the amplifier is inverting, or has a
phase of 180 degrees). Now, the frequency response is really determined by the RC circuit in the
feedback path of the op-amp. The first resistance at the input simply converts the input into a current that

29

drives this RC circuit. It can be shown that a 45 degree phase shift occurs in an RC circuit when the
frequency is 1/RfC (directly from the results of an analysis on an RC circuit). So,
RfC = 1/2000 C = 0.016 nF (We know Rf = 10 M.)
7

Also, at this frequency, the response is 0.707DC response. The DC response is 10 /R. So, the DC gain
is 14.14. Thus, R = 7.07 k.

Solution 10.91 (a) Use the virtual short property: Iin =

4

feedback path: Vout = 10 Iin =

10 4 Vin
3

10

j
C

Vin
. All of current flows through the
+ j 1C

10 3

Vout
= 7.07 135o .
Vin

(b)

(c) The spice result looks the same at low frequencies. However, at high frequencies, the response falls
back to zero as the op-amp non-ideal frequency response starts to affect the behavior of the circuit.

V
SOLUTION 10.92. (a) The negative terminal of the op-amp is at Vs. This implies IR = S . By KVL,
R

30

 j 3R

V
1
3
C
VS + ZRCIR = Vo o = 1 +
=
1
+
VS
R 3R j
j3 RC + 1
C
The MATLAB plot for the given values is:

(c) The spice result looks pretty much the same, especially since the cut-off frequency of this circuit is
much lower than the frequency at which the op-amp ceases to operate as an ideal op-amp.

SOLUTION 10.93. To compute the gain as a function of we observe that by the properties of an ideal op
amp,
Vout
2 105 + j0.5 106 20 + j0.5
Gain =
=
=
Vin
100 + j 0.1
104 + j0.1 10 6
In MATLAB
G1 = 1/50e3;
G2 = 1/10e3;
C1 = 0.5e-6;
C2 = 0.1e-6;
w = logspace(-1,5,1500);
Y1 = G1 +j*w*C1;
Y2 = G2 + j*w*C2;
H = Y1 ./ Y2;
semilogx(w,abs(H))
grid

31

SOLUTION 10.94. (a) The equations are:

V1 Vin V1 V2 V1 V2
+
+
=0
1000
1000
1/ jwC f
V2 V1
V2
+
=0
1000
1/ j C2
V2 = Vout
Substituting values and solving for V1 and V2 ,
V1 = 0.8535 82o Vin
V2 = 0.723 114.7o Vin
The second one is the relation that we are looking for.
(b) At 100 Hz,
V1 = 1.00 3.68o Vin
V2 = 1.00 7.3o Vin
At 3000 Hz,
V1 = 0.15 102o Vin
V2 = 0.0728 164o Vin
32

(c) The response is that of a low-pass filter, as predicted from the results of part (b) above.

SOLUTION 10.95. (a) The two nodal equations:

Vx Vin
+ j C1Vx + j C2 ( Vx Vout ) = 0
200
Vout
j C2 (0 Vx )
=0
28 103
In MATLAB,
w = 2*pi*1.34e3; C = 0.05e-6; R1 = 200; R2 = 28e3;
A = [1/R1+j*w*C+j*w*C -j*w*C;-j*w*C -1/R2]
A=
5.0000e-03 + 8.4195e-04i
0 - 4.2097e-04i
0 - 4.2097e-04i -3.5714e-05
b = [1/R1; 0];
V = A\b
V=
2.6664e-01 - 5.9266e+00i
-6.9859e+01 - 3.1429e+00i
Vout = V(2)
Vout =
-6.9859e+01 - 3.1429e+00i
abs(Vout)
ans =
6.9929e+01
angle(Vout)*180/pi
ans =
-1.7742e+02

Hence,

Vout
= 69.93 177.4 o .
Vin

When the capacitors are shorts, the output is shorted to the virtual ground input, at 0 V. Similarly, when
they are opens, the virtual ground makes sure that vout is zero, since there is no drop across the feedback
resistor.
33

(b) The band-pass response can be computed using any SPICE program.

SOLUTION 10.96. First, note the input-output relationship:

Zf
Vout
=
Vin
R1
where Zf is the impedance of the parallel RLC circuit. We have studied this circuit extensively earlier, and
we have shown that at resonance, the impedance of this circuit is going to be real and equal to the value
of resistance, in this case R2 . Thats exactly the requirement of this problem, since we want

Zf
Vout
=
Vin
R1

to simply be equal to R2 /R1 . It remains to note that this resonance occurs at a frequency = 1/sqrt(LC).

SOLUTION 10.97. (a) First, analyze the feedback amplifier circuit. The output of this op-amp circuit is:

Vop2 =

1/ j C
Vout
2 10 6

Also, by voltage division, the voltage at the resistive voltage divider (+ terminal of first op-amp):
2 103
1
0.01
VRR =

Vout =
Vout
5
6
2 10
j C2 2 10
j C2 2 106
Now, the first op-amp circuit is an inverting amplifier, but its + terminal is at VRR now. Thus,
Vout = 10(Vin VRR ) + VRR
Substituting the above VRR means that:
Vout =

10
1.1 Vin
1+
j

Since the input voltage has unity magnitude and zero phase, the above expression gives the required
magnitude and phase of the output voltage.
(b)
34

10
9
8

Magnitude

7
6
5
TextEnd

4
3
2
1
0

0.2

0.4

0.6

0.8
1
1.2
Frequency Hz

1.4

1.6

1.8

(c) As can be seen, the response to zero frequency (i.e. DC) is zero. Also, the circuit goes back very quickly
(less than 2 Hz) to provide the required operation, which is to achieve a gain of 10.

35

Complex Power Prbs 2/5/00

CHAPTER 11.

P11-1

@ DeCarlo & P. M. Lin

PROBLEM SOLUTIONS
1

1
e 2t

1
SOLUTION 11.1. Using equation 11.3, Pav = (e t 1)2 Rdt =
+ t 2e t = 0.758 W.
2
2

SOLUTION 11.2.
V2
(a) From 11.6, Pav = m = 50 mW for a sinusoidal input.
2R
(b) From 11.3,
R
Pav = (10cos(10t)) 2 dt +
2
0
/20

3 /20

2
(10cos(10t))
dt
=

3 /20

2 /10

(10cos(10t)) dt +

/20

2
Im
R
= 50 mW
2

just as the previous case since the square of the absolute cos(10t) is the same as the square of cos(10t).
(c)
10R
Pav =
2

2 /10

10 3
0.01cos (10t) dt =
R
2
2

2 /10

[0.5cos(20t) + 0.125cos(40t) + 0.375]dt

0

2 /10
1 0.5
0.125

=
sin(20t) +
sin(40t) + 0.375t
= 37.5 mW
2 20
40
0

(d)
t=0:1/1000:1;
R=1e3;
pta= (0.01*cos(10.*t)).^2.*R;
ptb= (0.01*abs(cos(10.*t))).^2.*R;
ptc= (0.01*cos(10.*t)).^4.*R;
subplot(3,1,1);
plot(t,pta);
grid
ylabel('W');
subplot(3,1,2);
plot(t,ptb);
grid
ylabel('W');
subplot(3,1,3);
plot(t,ptc);
grid
ylabel('W');
xlabel('time in seconds');

Complex Power Prbs 2/5/00

P11-2

@ DeCarlo & P. M. Lin

0.1

0.05
TextEnd

0.2

0.4

0.6

0.8

0.2

0.4

0.6

0.8

0.2

0.4
0.6
time in seconds

0.8

0.1

0.05
TextEnd

0
x 10-5

0.5
TextEnd

1 2 v 2 (t)
1
SOLUTION 11.3. (a) For figure a, the period is 2, and Pav =
dt =
(400 + 100) = 25 W.
2 0 R
2R
1
1 (20t )2
400 t 3
In figure b, the period is 1, and Pav =
dt =
= 13.3 W.
0 R
R 3
0
(b)

Complex Power Prbs 2/5/00

P11-3

@ DeCarlo & P. M. Lin

40
Instantaneous Power in W

35
30
25
20
TextEnd

15
10

0.5

1.5

2
Time in s

2.5

3.5

2.5

3.5

40
Instantaneous Power in W

35
30
25
20
15
10
TextEnd

5
0

0.5

1.5

2
Time in s

SOLUTION 11.4. (a) For (a), looking the definition for the effective voltage, one sees graphically that
the integral over one period, 2, of the squared waveform, is 500. Dividing by the period, and taking the
1

square root, Veff = 15.81 V. For fig. b, Veff =

(20t )2 dt =
0

Veff
2
(b) Pav = Ieff
R=
8 = 20 W.
10
2

Veff
(c) Pav =
8 = 10.67 W
10

400

1
= 11.55 V
3

Complex Power Prbs 2/5/00

P11-4

@ DeCarlo & P. M. Lin

SOLUTION 11.5. (a) This can be done graphically quite easily. The period of fig a, is 9s. The total
75
area of one period of the squared waveform is 75. This yields Ieff =
= 2.89 A. In fig. b, the area
9
25
over one period is 25 which yields Ieff =
= 2.89 A.
3
60 2

(b) Using current division, Pav = Ieff

30 = 111.36 W.
90
(c) The same result as (b) is obtained since the effective current is the same.

2
1 1 t
e 1 dt = 0.615 A.

2 0

SOLUTION 11.6. (a) Ieff =

2
(b) Pav = Ieff
R = 0.758 W.

SOLUTION 11.7. (a)

2 /20
20 2 /20
20
2

2
2
Veff =
(10 + 2cos(20t)) dt = 100t + 2t + sin(40t) + 2sin(20t)
= 102.01
2 0
2
40
0
Hence Veff = 10.1 V.
(b)
2
Veff
=

Hence Veff = 7.91 V.

1
0 (10cos(2t) + 5cos(4t))2 dt = [62.5t]0 = 62.568

(c)
Without going into detailed calculation, note the following fact about v 32(t) . Only the product terms that
have the same frequency will produce a non-zero result when integrated. Thus the integral reduces
to the following:
1
2
Veff
= 100cos 2 (2t) + 25cos 2(4 t) + 25cos 2 (4t /4) + 50cos(4 t)cos(4t /4) dt
0

Hence
Veff =

[50 + 12.5 + 12.5 + 25cos( / 4)] = 9.627 V

SOLUTION 11.8. The voltage is V = 500 V and the impedance Zeq = 100 100 j = 141.42 45o
2

. Thus I =

V
0.3536
2
= 353.645o mA. Hence, Pav = R Ieff = 100
= 6.2516 W.

Zeq
2

Complex Power Prbs 2/5/00

P11-5

SOLUTION 11.9. (a) The equivalent load seen by the source, Zeq =

@ DeCarlo & P. M. Lin

1
= 2 36.87o . Thus
j C + 1/ R

VL = IZeq = 10 36.87o V and v L (t) = 10cos(30t 36.87o) V.

V I
10 5
(b) Pav = m m cos( v i ) =
cos(36.87o ) = 20 W, and from 11.4
2
2
pL (t) = 25cos(36.87o ) + 25cos(60t 36.87o ) W
SOLUTION 11.10. (a) First find the impedance Zeq = 3 + j4 = 553.13o , then
V
Is = s = 10 143.13o A. The magnitude is 10 A rms or 14.14 A peak-value.
Zeq
(b) Pav = 10 50cos(53.13o ) = 300 W.
(c) 3Is

= 300 W

SOLUTION 11.11. (a) By KVL, Vs = 10IL + j100IL + 9 10IL and

Vs
1000
1
IL =
=
=
45o A.
100 + j100 141.4245o
2
(b) PavVs = Vs IL cos(45o ) = 50 W and Pav9V1 = 910IL IL cos(180o ) = 45 W.
SOLUTION 11.12. (a) By KCL the current through the resistor is 5I. So by KVL,
Vs = j1000I j500I + 200 5I and hence
I=

Vs
= 107.33 26.57o A
1000 + j500

The power absorbed by the resistor is, Pav = 200 5I 5I = 57.6 W.

(b) PavVs = Vs I cos(26.57o ) = 11.52 W and Pav4I = 4I 200 5I cos(0) = 46.08 W.
SOLUTION 11.13. First find the input impedance, Zeq = 2 j8 + 6 + j2 = 8 j6 . Then calculate the
100
current IL =
= 8 + j6 A. The complex power is then computed:
8 j6
ZL = 6 + j*2;
IL = 8 + j*6;
VL = ZL*IL
VL =
3.6000e+01 + 5.2000e+01i
SL = VL * conj(IL)
SL =
6.0000e+02 + 2.0000e+02i

Complex Power Prbs 2/5/00

P11-6

@ DeCarlo & P. M. Lin

abs(SL)
ans =
6.3246e+02
Thus the apparent power is 790 VA, the average power 600 W, the reactive power 200 var, and the
apparent power is 632.46 VA.
SOLUTION 11.14. (a) Using MATLAB
Vseff = 100*exp(j*pi/6)
Vseff = 8.6603e+01 + 5.0000e+01i
ZL = 350 +j*1*300;
Zin = 50 + ZL
Zin = 4.0000e+02 + 3.0000e+02i
ILeff = Vseff/Zin
ILeff = 1.9856e-01 - 2.3923e-02i
ILpk = sqrt(2)*abs(ILeff)
ILpk = 2.8284e-01
ILang = angle(ILeff)*180/pi
ILang = -6.8699e+00
SL = ZL*ILeff*conj(ILeff)
SL = 1.4000e+01 + 1.2000e+01i
Therefore, iL (t) = 2 (0.2) cos(300t 6.87o ) = 0.2828cos(300t 6.87o) A. SL = 14 + j12 VA, and the
average power is 14 W.

SOLUTION 11.15. First use voltage division and observe that VL =

complex power is SL = VL

1
V = 6060 V. Now the
2 s

VL*
= 36053.13 = 216 + j288 VA. Thus the average power is 216
(6 + j8)*

W.
*S OLUTION 11.16. (a) To find V2 we write a node equation. First we note that Yi = 1/Zi is the
corresponding admittance. Hence
Y1(V2 Va ) + Y2 V2 + Y3( V2 Vb ) = 0
Hence

Y V + Y3Vb
V2 = 1 a
= 100 + j50 = 111.826.57o V
Y1 + Y2 + Y3

obtained using MATLAB as follows

Z1 = 0.1+j*0.1; Z2 = 0.4+j*2.2;
Z3 = 0.2 + j*0.2; Va = 104 + j*50; Vb = 106 + j*48;
Y1 = 1/Z1

Complex Power Prbs 2/5/00

P11-7

@ DeCarlo & P. M. Lin

Y1 = 5.0000e+00 - 5.0000e+00i
Y2 = 1/Z2
Y2 = 8.0000e-02 - 4.4000e-01i
Y3 = 1/Z3
Y3 = 2.5000e+00 - 2.5000e+00i
V2 = (Y1*Va + Y3*Vb)/(Y1 + Y2 + Y3)
V2 = 1.0000e+02 + 5.0000e+01i
magV2 = abs(V2)
magV2 = 1.1180e+02
angleV2 = angle(V2)*180/pi
angleV2 = 2.6565e+01
(b) Again working strictly in MATLAB we have the following complex powers of the loads and the two
sources:
Sz1 = (V2 - Va)*conj((V2-Va)*Y1)
Sz1 =
8.0000e+01 + 8.0000e+01i
Sz2 = V2*conj(V2*Y2)
Sz2 =
1.0000e+03 + 5.5000e+03i
Sz3 = (V2 - Vb)*conj((V2-Vb)*Y3)
Sz3 =
1.0000e+02 + 1.0000e+02i
Sva = Va*conj((Va - V2)*Y1)
Sva =
1.0800e+03 + 3.0800e+03i
Svb = Vb*conj((Vb - V2)*Y3)
Svb =
1.0000e+02 + 2.6000e+03i
(c) To verify conservation of power observe that:
TotSrsPwr = Sva + Svb
TotSrsPwr =
1.1800e+03 + 5.6800e+03i
TotLdPwer = Sz1 + Sz2 + Sz3
TotLdPwer =
1.1800e+03 + 5.6800e+03i
which provides the desired verification.
SOLUTION 11.17. Use MATLAB and refer to the following figure:

Complex Power Prbs 2/5/00

P11-8

Z1

Z4
Z2

Z3

%(a)
%Bundle the impedances as per the following figure and
%obtain the following.
Z1=2+2*j;
Y1=1/Z1;
Y2=2+0.5*j;
Z2=1/Y2;
ZL3=4*j;
YL3=1/ZL3;
Y3=4+0.25*j;
Z3=1/Y3;
Z4=4+4*j;
Y4=1/Z4;
V1=10+2*j;
V2=12+2*j;
%Write out KCL for node 1 and 2
%(Va-V1)*Y1=V1*Y2+(V1-V2)*YL3
%(Vb-V2)*Y4=V2*Y3+(V2-V1)*YL3
Va= (V1*Y2+(V1-V2)*YL3)/Y1+V1
Vb= (V2*Y3+(V2-V1)*YL3)/Y4+V2
Va = 29.0000 +59.0000i
Vb = 1.6000e+02+ 2.3400e+02i
(b)
Sr3=( (Va-V1)*Y1*2)*conj((Va-V1)*Y1)
Sl1= ((Va-V1)*Y1*(2*j))*conj((Va-V1)*Y1)
Sc1= V1*conj(V1*0.5*j)
Sr1= V1*conj(V1/0.5)
Sl3= (V1-V2)*conj((V1-V2)*YL3)
Sc2= V2*conj(V2*0.25*j)
Sr2= V2*conj(V2/0.25)
Sl2= (V2-Vb)*Y4*4*j*conj((V2-Vb)*Y4)
Sr4= (V2-Vb)*Y4*4*conj((V2-Vb)*Y4)
SVA = Va * conj((Va-V1)*Y1)
SVB = Vb * conj((Vb-V2)*Y4)
Sr3 = 9.0250e+02
Sl1 = 0 + 9.0250e+02i
Sc1 = 0 - 5.2000e+01i

@ DeCarlo & P. M. Lin

Complex Power Prbs 2/5/00

P11-9

@ DeCarlo & P. M. Lin

Sr1 = 208
Sl3 =

0 + 1.0000e+00i

Sc2 = 0 - 3.7000e+01i
Sr2 = 592
Sl2 =

0 + 9.4660e+03i

Sr4 = 9466
SVA = 1.1115e+03 + 8.4550e+02i
SVB = 1.0057e+04 + 9.4350e+03i
(c) Take the real part of each of the complex power found in (b). The only components with non-zero
average power will be the resistors which have 208 W, 592 W, 902.5 W, and 9466 W average power
respectively.
%(d)
Total_passive=Sr1+Sr2+Sr3+Sr4+Sl1+Sl2+Sl3+Sc1+Sc2
Total_active= Va*conj((Va-V1)*Y1)+Vb*conj((Vb-V2)*Y4)
Total_passive =
1.1168e+04+ 1.0280e+04i
Total_active = 1.1168e+04+ 1.0280e+04i
which verifies the conservation of power.
SOLUTION 11.18. (a) From conservation of energy, the complex power is the sum of the complex
power absorbed by every circuit elements. Thus Ss = 1240 + j145 VA, and the apparent power is 1248.4
VA. The average power is 1240 W.
S
(b) From S s = Vs I*s , Is = s = 5.428 A.
Vs
SOLUTION 11.19. (a) The complex power delivered by the source is the sum of the complex power
consumed by the circuit elements. Thus Ss = 44 + j28 kVA.
(b) Is =

Ss
= 22.675 A
Vs

(c) The total power delivered to the three groups of impedance following V1 is S1 = 41.5 + j22. From
the current obtained in (b) V1 =

S1 + S 2 + S4
= 2071.5 V.
Is

(d) From V1 , and the total power delivered to Z4 and Z2 , I2 =

V2 =

S2
= 1805.5 V.
I2

S 4 + S2
= 12.385 A. Finally
V1

Complex Power Prbs 2/5/00

P11-10

@ DeCarlo & P. M. Lin

*
44 + j28
SOLUTION 11.20. (a) Is =
= 22.675 (32.47 0) = 22.675 32.47 A.
2.3
S + S2 + S4
(b) V1 = 1
= 2071.5 4.54 V
I*s

S + S2
(c) Similarly as before I2 = 1
= 12.385 37.61 A, and V2 = Z2I2 = 1805.5 11.05 V.
V1

SOLUTION 11.21. From equation 11.30, we find Q = P

1
1 = 455.61 var; thus
pf 2

S = 1000 + j455.61 = 109924.5 VA.

SOLUTION 11.22. (a) S1 = 1000 W, and S2 = 800 + j600 VA. Thus the total power delivered by the
S*
source is Stot = 1800 + j600 = 1897.3718.43 VA, and Is = tot
= 15.81 18.43 = 15 j5 A rms.
Vs*
S
(b) V1 = *1 = 63.25 18.43 V.
Is
S
(c) V2 = *2 = 63.2518.43 V.
Is
P
3000
SOLUTION 11.23. Using 11.25, find SL = ave =
= 4000 VA, and the load current
pf
0.75
IL =

SL
VL

4000
= 33.33 A. The power absorbed by the transmission line is then from
120

Pline = R IL = 0.2 ( 33.33) 2 = 222.2 W.

2

SOLUTION 11.24. The capacitor must absorb a reactive power of Qnew Qold = 17.9 kvar. Thus
QC
2
jQC = j17.9 = j CV s , and C =
2 = 0.897 mF.
Vs
1
1 = 31.43 VA. Thus the reactive
(0.94)2
QC
power absorbed by the capacitor is 18.57 var. Hence C =
2 = 3.714 F.
Vs
SOLUTION 11.25. From equation 11.30, Qnew = 86.6

Complex Power Prbs 2/5/00

P11-11

@ DeCarlo & P. M. Lin

SOLUTION 11.26. Device 1 has a complex power of S1 = P 1 + jQ1 = 360 + j 0 VA. Recall equation
P
1
11.29, pf = ave , and equation 11.30 Qnew = Pave
1
S
(pf ) 2
where with a lagging pf, Q > 0, and with a leading pf for Q < 0. Using equation 11.30, we have for
device 2:
Q2 = 1440
S2 = 1440 + j 1080 VA.

1
1 = 1080 var
(0.8)2

S1,2 = S1 + S2 = 1800 + j1080 VA

As an aside we compute the magnitude of the current without the capacitor attached.
Is =

S1,2
120

= 17.493 A

The capacitor is used to achieve a lower source current with the same average power. The first step is to
find the desired QC. Here
Hence

Stot = S1 + S2 + SC = 1800 + j1080 + jQC

Stot 1800 + j1080 + jQC
=
= 15
120
120

In MATLAB we have:
QC = sqrt((15*120)^2 - 1800^2) - 1080
QC =
-1080
From the formula on page 451 of the text,
QC = C Vsource

Hence C = 0.2 mF. Finally

pf =

Pave
1800
=
=1
S1 + S2 + SC 1800

1
1 = 8.1839 kvar. Therefore, the
(0.65)2
power absorbed without the capacitor bank is Sold = 7 + j8.1839 kVA. When the bank is added we
1
want Qnew = 7
1 = 5.25 kvar, and henceSnew = 7 + j5.25 kVA. Thus the reactance that must
(0.8)2
QC
be absorbed by the bank is 2.934 kvar, and Ceq =
2 = 0.13511 mF.
Vs
(b) As was just determined 5.25 kVA.
SOLUTION 11.27. (a) From equation 11.30, Qold = 7

Complex Power Prbs 2/5/00

(c) Sold =

P11-12

@ DeCarlo & P. M. Lin

7
7
= 10.77 kVA. Snew =
= 8.75 kVA. The kVA saving is: 10.77 8.75 = 2.02
0.65
0.8

kVA.
Savings = 20*2.02*12
Savings =
4.8480e+02
i.e., $484.80.
SOLUTION 11.28. (a) The apparent power is simply 94kW/0.78=120.51 kVA.
(b) Sm = 120.5138.74o = 94 + j 75.41 kVA.
(c) 75.41 kvar.
S*
(d) Ieff = m
* = 523.96 38.74 A.
Veff
(e) By KVL, Vs = RlineIeff + Veff = 516.08 j229.52 V.
(f) Ss = VsI*eff = 295.9414.76 kVA.
120.51cos(38.74)
(g) The efficiency is =
100 = 32.96% . Note that the line resistance of 0.7 is much
295.94cos(14.76)
to large for practical usage. This value is chosen for pedagogical reasons.
(h) With a power factor of 0.94, Snew
m = 94 + j 34.12 kVA. The average power of the motor must be kept
the same. The reactance that must be provided by the capacitor is, Qnew Qold = QC = 41.29 kvar. The
2

Veff
j42.29k
1
Veff
proper capacitor current will be IC =
,
and
Z
=
=
=
. Solving for
C
*
j C
IC
Veff
j 42.29 103
C=

42.29 10 3
Veff

= 2.12 mF.
*

(i)

Inew
eff

Snew
= m = 434.79 19.95 A
Veff

o
(j) By KVL, Vs = RlineInew
eff + Veff = 516.1 j103.84 = 526.43 11.38 V.

new new *
(k) Snew
=
V
Ieff
= 228.898.57 = 226.33 + j 34.12 kVA, and the efficiency is 41.5%.
s
s

SOLUTION 11.29. (a) The Thevenin equivalent seen at the output is, Zth = 5 j /(0.1 ). For
VOC
*
maximum power transfer, ZL = Zth
= 5 + j . Note that VOC = 50 V rms, and that Ieff =
=5 A
Zth + ZL
rms. Thus SL = ZL Ieff I*eff = 125 + j25 VA, and the average power is 125 W.
SOLUTION 11.30. First find the following Thevenin equivalent,
10 2
Zth = 2 + j 4 , Voc =
V.
3
By the maximum transfer property, RL = 2 , and C = 1/(4 ) = 0.25 mF.

Complex Power Prbs 2/5/00

P11-13

@ DeCarlo & P. M. Lin

SOLUTION 11.31. (a) The thevenin equivalent left of the load is by KCL,
Itest + Is = VR ( j0.001+ 0.001) and Vtest = 3VR + j500Itest V. Substituting for VR ,
Vtest = Itest[1500 j1000] + Is[1500 j1500] V, and Zth = 1500 j1000 with
Voc = Is 2121.3 45 V. The value of the load for maximum power transfer is then,
ZL = 1500 + j1000 .
Voc
(b) The complex power absorbed by the load is, SL = ZL
Zth + ZL
power 750 W.

= 750 + j500 VA, and the average

SOLUTION 11.32. Consider

V1

V2

(a) Using nodal analysis get the following equations:

V1 V2 = 4(V1 /8)
Is + Itest = V1(1/8 + j0.25) + V2 (1/16 + j0.5)
Vtest = V2 + 8Itest
Using MATLAB we get the following expression,
Vtest = (8.2847- j0.9110)Itest + (0.2847- j0.9110)Is
From this expression Voc = 28.47 j91.1 V, and Zth = 8.2847- j0.9110 . The phasor equivalent of
this circuit is a Voc source in series with a 8.28 resistor and a 7.32 mF capacitor.
(b) ZL = 8.2847+ j0.9110 , which is a 8.28 Ohm resistor in series with a 6.07mH inductor. Same
configuration as before.
Voc
(c) SL = ZL
Zth + ZL

= 275.2 + j30.26 VA and the average power is 275.2 W.

SOLUTION 11.33. (a) The Norton equivalent may be found by inspection as Zth = 10 + j20 and
Isc = 10 A. Thus for maximum power transfer, ZL = 20 j20 . This is a 10 resistor and a 0.005 F
2

I Z
capacitor in series. The maximum power is SL = ZL sc th
= 1250 j2500 VA. The maximum
Zth + ZL
average power is 1250 W.
(b) If R is set to 20 , the closest that can be achieved to maximum power transfer is ZL = 20 j20 , or
C equal to 0.005 F. With ZL as above, by current division

Complex Power Prbs 2/5/00

Iload = 10

P11-14

@ DeCarlo & P. M. Lin

10 + j20
= 3.333 j6.6667 = 7.4563.434o A
(10 + j20) + (20 j20)
2

The maximum average power then 20 Iload = 1111 W.

2

I Z
(c) Using 11.38, RL = 31.62 . Pav = RL sc th
Zth + ZL

= 600.63 W.

SOLUTION 11.34. (a) From Thevenin Zth = 19.2 j14.4 , and from 11.38, set RL = 24 .
2
Vs( j 40)
Voc
Voc =
= 80 36.87. The maximum power is Pav =
R = 74.07 W.
30 j40
Zth + RL L
RL
(b) The voltage is V = Voc
, from this relationship, one sees that as the load resistance increase
Zth + RL
to infinity the output voltage goes to Voc , which is the maximum output voltage.
SOLUTION 11.35. Correction: the inductor symbol in the load should be a resistor. Since the source
resistance is variable, example 6.21 serves as a reference suggesting that R = 0 is the answer. To see this
consider that
2 100
2 100
2
P = 2Iload =
2 = (R + 2) 2 + (2 2) 2
1

(R + 2)2 + L

C
Hence, decreasing R produces increasing power and the maximum power is transferred when R = 0 with
Pmax = 50 W assuming that the source voltage is given in rms V.
SOLUTION 11.36. As per problem 35,
2

P = 10Iload =

10 50
(R + 10)2 +

)2

2 =

12.5 103
(R + 10) 2 +

1 2

Here, again R = 0 with C chosen to eliminate the reactive term maximizes power transfer. Hence
C=
with Pmax = 125 W.

1
2L

= 0.01 F

SOLUTION 11.37. (a) By the maximum power transfer theorem, P1 is maximized when ZL is chosen
as the conjugate of Zsource, i.e.
ZL = 10 + j1000
(b) To find the appropriate values of L and C observe that

Complex Power Prbs 2/5/00

P11-15

@ DeCarlo & P. M. Lin

1
10 4 j C
7
ZL ( j ) = j L + 4
= j10 L + 8
10 + j C
10 + 1014 C 2
10 4

= 8
+ j 107 L 8
= 10 + j1000
14
2
14
2

10 + 10 C
10 + 10 C
Equating real parts leads to:
%10^-4 = 10^-7 + 10^15*C^2

C = sqrt(1e-4 - 1e-7)/sqrt(1e15)
C = 3.1607e-10
Thus C = 0.31607 nF.
Equating imaginary parts using the above value of C leads to:
w = 1e7;
L = (1e3 + w*C/(1e-8 + 1e14*C^2))/1e7
L = 1.3161e-04
Thus L = 0.1316 mH.
(c) In part (b), L and C are chosen to maximize P1 , the power delivered to ZL. Since L and C consume
no average power, this maximum power is transferred to the 10 k fixed resistor with the
computed values of L and C. Thus ZL is the same as in part (b) or ZL = 10 + j1000 .
Since we know ZL,
Pmax =

(0.1) 2
= 0.25 mW
4 10

This is the average power consumed by the 10 k resistor. Therefore

V2 = 0.25 103 10 4 = 2.5
2

It follows that V2 = 1.5811 V. Power to the 10 k fixed resistor is

P10k =

V2

104

Thus if P10k is maximized, then V2 is maximized.

SOLUTION 11.38. (a) From equation 11.4, if we substitute

, then the resulting instantaneous

V I
V I
4
power will be p(t) = m m cos( v i ) + m m cos(
t + v + i ) W where it is clearly seen that the
2
2
T
fundamental period will now be halved. Note that by the same argument the fundamental frequency of
the instantaneous power is double that of the voltage and current.

Complex Power Prbs 2/5/00

P11-16

@ DeCarlo & P. M. Lin

(b) As a sinusoid, the fundamental period is 2/10, any integer multiple of this period will also be
periodic.
(c) This is the same as (b) with an offset of 1 V added.
1 to +T 2
SOLUTION 11.39. First, Feff =
f (t)dt . Without going into detailed calculation, f 2 (t) will

T t0
give a summation of two types of products, a product of each element with themselves, and products of
each element with the other element. In the later case, we know that two cosines multiplied with one
another and integrated over one period will yield zero if their angular frequency are different. As for the
former case the integral will yields the result we are looking for. For example look at the first two terms,
1 to +T 2
f (t)dt =
T t 0

Feff =

1 t0 +T 2
Fo + 2F12 cos2 ( 1t + 1) + ... dt
T t 0

1 t0 +T 2
Fo + F12 + F12 cos(2 1t + 2 1 ) + ... dt
T t 0

T
1 2
Fo t + F12 t + F12 sin(2 1 t + 2 1 ) + ... = Fo2 + F12 ...
0
T

SOLUTION 11.40. (a) We are given that vC (t) = Vm sin( t) V. Hence

iC (t) = C

dvC
= CVm cos( t) A
dt

It follows that
p(t) = v C (t)iC (t) = CVm2 sin( t)cos( t) = 0.5 CVm2 sin(2 t) Watts
Clearly, p(t) has a peak value of 0.5 CVm2 and the integral of the sign over one period is zero implying
that the average value of p(t) is zero.
(b) Here
WC (t) = 0.5CvC2 (t) = 0.5CVm2 sin2 ( t) = 0.25CVm2 (1 cos(2 t)) J
Here the peak value occurs when cos(2t) = 1 in which case the peak value is 0.5CVm2 . Further, the
average value of cos(2t) over one period, T = /, is zero whereas the average of a constant over the
same period is simply the constant. Hence, WC,ave = 0.25CVm2 J.
(c) From example 11.6,
2
QC = IC ,eff VC ,eff = CVC,eff
= 0.5 CVm2 = 2

Q
Therefore, WC,ave = C .
2

(0.25CVm2) = 2

WC,ave .

Complex Power Prbs 2/5/00

P11-17

@ DeCarlo & P. M. Lin

SOLUTION 11.41. (a) We are given that iL (t) = I m sin( t) A. Hence

v L (t) = L

diL
= LIm cos( t) V
dt

It follows that
2
2
p(t) = v L (t)iL (t) = LIm
sin( t)cos( t) = 0.5 LIm
sin(2 t) watts
2
Clearly, p(t) has a peak value of 0.5 LIm
and the integral of the sign over one period is zero implying
that the average value of p(t) is zero.

(b) Here
2
2
WL (t) = 0.5LiL2 (t) = 0.5LIm
sin2 ( t) = 0.25LIm
(1 cos(2 t)) J
2
Here the peak value occurs when cos(2t) = 1 in which case the peak value is 0.5LI m
. Further, the
average value of cos(2t) over one period, T = /, is zero whereas the average of a constant over the
2
same period is simply the constant. Hence, WL,ave = 0.25LIm
J.

(c) As an extension to example 11.5,

2
2
QL = IL,eff VL,eff = LI L,eff
= 0.5 LI m
=2

(0.25CI m2 ) = 2

WL,ave .

Q
Therefore, WL,ave = L .
2
SOLUTION 11.42. (a) The complex power absorbed by the load is, SL = VI* = ZII* = Z I 2 . Now note
that the average power is the real part of the complex power. Also note that a complex number
multiplied by its complex conjugate will yield a real value. Therefore the real part of Z I 2 is just the
real part of Z, R, multiplied by I 2 , Pav = R I 2 . With the same reasoning, the reactance is the
imaginary part of Z, X, multiplied by I 2 , Q = X I 2 .
(b) The complex power absorbed by the load is SL = VI* = V(YV)* = Y V 2 . The same reasoning as in
(a) holds thus the real part of the admittance times V 2 , yields Pav = G V 2 . Using the imaginary
part, Q = B V 2 .
SOLUTION 11.43. (a) The equivalent resistance seen by the source is Req = 6 j9 . So the current
delivered by the source is:
Is =

Vs
j110
=
= 10.17 33.69 A
6 j9 6 j9

or iL (t) = 10.17 2 cos(120 t 33.69o ) A. Similarly

VC = j15 Is =

15 110
= 152.54 123.69 V
6 j9

Complex Power Prbs 2/5/00

P11-18

or vC (t) = 152.54 2 cos(120 t 123.69o ) V. Further, L =

@ DeCarlo & P. M. Lin

= 15.9 mH and C =

1
= 176.8 F.
15

The instantaneous energy stored in the inductor is

WL (t) = 0.5LiL2 (t) = 1.646cos 2 (120 t 33.69o ) J
and the instantaneous energy stored in the capacitor is
WC (t) = 0.5CvC2 (t) = 4.115cos 2 (120 t 123.69o ) J
The source voltage is zero when t = 0. Therefore

and

WL (0) = 1.646cos 2 (33.69o ) = 1.1395 J

WC (0) = 4.115cos 2 (123.69o ) = 1.266 J

(b) and (c) Observe that iL(t) and vC(t) are 90o out of phase. When one is zero, the other has a peak
value. Therefore whenWC = 0 implies vC(t0 ) = 0 for appropriate t0 ; hence WL (t 0 ) = 1.646 J.
Similarly, when WL = 0, say at t0 , then WC (t 0 ) = 4.115 J.
SOLUTION 11.44. In order to solve this problem, we want to express the power in terms or R's and L's
in both circuits. First, looking at the circuit with just the coil and the 110 V source: I = V / Zcoil ,
Zcoil = R + j L , and
110 2
2
Pcoil = I R = 2
R = 300 watts
(*)
R + 2L2
Next, looking at the circuit when a resistance is added in series with the coil, I = V /(8 + Zcoil ),
Zcoil = R + j L ,
220 2
Pcoil = I 2 R =
R = 300 watts
(**)
(8 + R)2 + 2L2
To find R, solve equation (*) for R 2 + 2 L2 and substitute into equation (**) to obtain
R = 300*64/(220^2 - 300*16 -110^2)
R = 6.0952e-01
Substituting R into equation (*) yields L = 13.05 mH.
SOLUTION 11.45. The average power consumed by the 2.7 resistor is 250 watts. This allows us to
compute the magnitude of | Icoil |. We know that | Vcoil | is 150 Vrms. Thus we can compute
the magnitude of the coil impedance and hence L as follows:
magIcoil = sqrt(250/2.7)
magIcoil = 9.6225e+00
magZcoil = 150/magIcoil
magZcoil = 1.5588e+01

Complex Power Prbs 2/5/00

P11-19

% magZcoil^2 = 2.7^2 + (w*L)^2

w = 2*pi*60;
L = sqrt( magZcoil^2 - 2.7^2)/w
L=
4.0725e-02
% Magnitude of impedance seen by 220 V source is:

magZin = 220/magIcoil
magZin = 2.2863e+01
% magZin^2 = (R + 2.7)^2 + (w*L)^2

Realpart = sqrt(magZin^2 - (w*L)^2)

Realpart = 1.6941e+01
R = Realpart - 2.7
R = 1.4241e+01

@ DeCarlo & P. M. Lin

9/26/01

P12-1

@ DeCarlo & P. M. Lin

CHAPTER 12. PROBLEM SOLUTIONS

SOLUTION P12.1. By conservation of energy, the instantaneous power consumed by each load when
summed together is equivalent to the total power consumed by the three phase load. Thus writing out
ptot (t) = pAB (t) + pBC (t) + pCA (t) = v AB (t)i AB (t) + vBC (t)iBC (t) + vCA (t)iCA (t)
V 2
V 2
= VL 2 cos( t + v ) L
cos( t + i ) + VL 2 cos( t + v 120o ) L
cos( t + i 120o )
Z
Z
V 2
+ VL 2 cos( t + v + 120o ) L
cos( t + i + 120o )
Z

V2
V2
= L (cos( v i ) + cos(2 t + v + i )) + L cos( v i ) + cos(2 t + v + i + 120o )
Z
Z

VL2
+
cos( v i ) + cos(2 t + v + i 120o )
Z

3VL2
3VL2
=
cos( v i ) =
pf
Z
Z

SOLUTION P12.2. To justify the point of this problem we equate the following two equations:
(i) For the 3 phase system:
3 PL2 ' PL2 '
= 3 IL R = 3 2 R = 2 R
9 VL
VL
(ii) For the single phase system:
PL2
2
Ploss = 2 IL R = 2 2R
VL
It follows that R' = 2R. Since both systems have the same distance of transmission and the resistance of
a wire is inversely proportional to the cross sectional area, the condition R' = 2R implies that the cross
section A' of each wire in the three-phase system need only be half of the area A of the wire in the single
phase system. But there are two wires in the single phase system and three wires in the three-phase
system. Therefore the ratio of the materials used is:
'
Ploss

2 '

material in 3 phase system 3A ' 3 1

=
= = 75%
material in 1 phase system 2A 2 2

SOLUTION P12.3. For row 1 of table 12.1, the impedance in (a) seen between each pair of terminal is
Z
Zik = Z || 2Z = 2 . In (b) the impedance seen between any two terminals is
3
Z Z
Z
Z jk =
+
=2
.
3
3
3

9/26/01

P12-2

@ DeCarlo & P. M. Lin

In row 2, the impedance between any two terminals is Z jk = ZY + ZY = 2ZY for (c), and
Zik = 3ZY || 6ZY = 2ZY for (d).

SOLUTION P12.4. Consider the -Y relationship of the figures below (row 3 of table 12.1):

Let us consider the terminal pair (1,2):

Z
(i) For the Y-connected case, Zth = 2 and
3
Voc =

Vp
3

30o

Vp
3

150o =

Vp

1 30o 1 150o = Vp
3

(ii) For the D-connected case,

2(Z ) 2 2
= Z
3Z
3
Now note that there is no load connected to the -configuration. Applying KVL to the indicated loop
implies that:
0 = 3Z Iloop + V p + V p 120o + V p + 120o = 3ZIloop
Zth = Z / /(2Z ) =

Hence Iloop = 0 . Finally, Voc = V12 = ZIloop + V p = V p . Therefore, looking into terminals 1-2,
both the -configuration and the Y-configuration have the same Thevenin equivalent. For terminal pairs
(1-3) and (2-3), the proof is virtually the same. Hence this establishes the equivalence in row 3 of table
12.1.
To establish the equivalence in row 4 of table 12.1, we do all the same computations with the
slightly different labeling to obtain the same result, i.e., the circuits are equivalent.

V V
SOLUTION 12.5. For (a) first note the following relationship VN = 1 + 2 ( Z1 || Z2 || Z3) , which is
Z1 Z2
obtained by KCL at the center node, N, with node 3 as the reference node.. Write out KCL at terminal 1,

V1 V1
V2
Z3 + Z2
Z3
I1 =

+
( Z || Z || Z ) = V1 Z Z + Z Z + Z Z V2
. Do the
Z1 Z1Z1 Z2Z1 1 2 3
2 3
1 3
1 2
Z1Z2 + Z2 Z3 + Z3 Z1

9/26/01

P12-3

@ DeCarlo & P. M. Lin

1
1 V2
same for terminal 1 in (b), I1 = V1
+

. Now substitute the formulas in the problem

Z31 Z12 Z12

Z2 + Z3
Z3
statement into the later equation, I1 = V1
V2
, which is
Z1Z2 + Z2Z3 + Z3Z1
Z1Z2 + Z2 Z3 + Z3 Z1
the same as the equation for (a).
Using the same method, the same result is obtained for node 2. So the fact that the substitution
of the equivalence in (b) yields the same equation as in (a) proves that the equivalences are accurate.

Vp 2000
=
= 17.89 26.57 A. By the same relationship
Z
10 + j5
IB = 17.89 146.57 A, and IC = 17.8993.43 A. The neutral line current is the sum of the other
three and is zero. The power of each phase is the same, as they have the same load, and current
SOLUTION 12.6. By Ohms law IA =

magnitude. Using the current at terminal A, the total power is, P = 3R IA = 9602 W.

SOLUTION 12.7. For this balanced Y-Y connection,we follow the method of example 12.6. Using
MATLAB:
Z = 20 + j*10'
Z = 2.0000e+01 + 1.0000e+01i
VAN = 200;
IA = VAN/Z
IA = 8.0000e+00 - 4.0000e+00i
IAmag = abs(IA)
IAmag = 8.9443e+00
IAangle = angle(IA)*180/pi
IAangle = -2.6565e+01
% Use phase inference to obtain:
IBmag = IAmag
IBmag = 8.9443e+00
IBangle = IAangle - 120
IBangle = -1.4657e+02
ICmag = IAmag
ICmag = 8.9443e+00
ICangle = IAangle + 120
ICangle = 9.3435e+01
Ptotave = 3*real(Z)*IAmag^2
Ptotave = 4.8000e+03

9/26/01

P12-4

@ DeCarlo & P. M. Lin

SOLUTION 12.8. Using the same approach as in the example 12.6 we compute
VAN = VsA

Thus the drop is:

Z
4 + j3
= 120
= 116.94 V
Z + Zg
4 + j3 + 0.1 + j0.2

120 116.94
100 = 2.55 % .
120

SOLUTION 12.9. A 5.5 % drop corresponds to VAN = 113.4 V. This means

V
Z
0.945 120
= AN =
= 0.945
120
120
Z + Z1 + Zg
From this equation,
Z + Z1 + Zg =

Z
= 5.291.
0.945

So (4.05 + 0.1)2 + (3.15 + 0.2)2 = 27.99 . Solving the quadratic equation results in

SOLUTION 12.10. Solving the single phase equivalent circuit for phase A,
IA =

V p V p120
= 58.87 56.57
10 + j5

Using phase inference,

IB = IA 120 = 58.87 176.57
IC = IA 120 = 58.8764.43
Alternately, using much more work,
IB =

V p 120 V p

= 58.87 176.57
10 + j5
V p120 V p 120
IC =
= 58.8764.43
10 + j5
The total power, using the previously derived relationship, is

where

P = 3VL IL pf = 3 380 58.87 pf = 34.655 kW

5
pf = cos tan1 = 0.8944

10

= 0.7893.

9/26/01

P12-5

@ DeCarlo & P. M. Lin

SOLUTION 12.11. (a) By KVL and Ohm's law,

IAB =

Vp
= 6.57 26.57
30 + j15

By phase inference,
IBC = IAB 120 = 6.57 146.57
ICA = IAB 120 = 6.5793.43
Alternately,
IBC =
ICA =

V p 120
30 + j15
V p120
30 + j15

= 6.57 146.57

= 6.5793.43

To compute the line currents we have in Amps:

IA = IAB ICA = 11.36 56.57
IB = IBC IAB = 11.36 176.57
IC = ICA IBC = 11.3663.43
(b) The average power is P = 3 30 IAB

= 3.88 kW, and the total reactive power is

Q = 3 15 IAB = 1.94 kvar.

25 746
= 0.85 , so the total power being
Pdeliverd
21941
= 63.31 A.
3 VL pf

SOLUTION 12.12. From the efficiency we know that

delivered to the load is 21941 W. Using 12.4, IL =

SOLUTION 12.13. (a) Using the power efficiency relationship, Ptot =

(b) Using 12.4, IL =

300 746
= 239.36 kW.
0.935

239.36 10 3
= 68.28 A.
3 VL pf

SOLUTION 12.14. The magnitude of the power delivered to each delta connected load is 79.787 kW.
We can now perform the analysis on a single phase. Using equation 11.30, we obtain

9/26/01

P12-6

@ DeCarlo & P. M. Lin

2
1
Qold = 79787
1 = 43064 vars
0.88

and
2
1
Qnew = 79787
1 = 26225 vars
0.95

Hence
Sold = 79787 + j 43064
Snew = 79787 + j26225
So the reactive power to be supplied by the capacitor is QC = 16839 = 43064 26225. The
capacitance
QC
C=
2 = 8.44 uF.
Vl

SOLUTION 12.15. Replacing the delta configuration of the source by its Y-equivalent, the following
voltages (in V) are obtained,
VAN =

Vp
3

30, VBN =

Vp
3

150, VCN =

Vp
3

90

From the single phase equivalent,

IA =
From phase inference

VAN
= 19.63 56.57
Z

IB = IA 120 = 19.63 176.57

IC = IA 120 = 19.6363.43

The power factor of the load is

5
pf = cos tan1 = 0.8944

10
Hence, the total power in the balanced system is
P = 3VL IL pf = 3 380 19.63 0.8944 = 11.556 kW.

SOLUTION 12.16. First replace the delta connected load by it Y-equivalent. Then analyzing a single
phase and using phase inference, we obtain in amps,

9/26/01

P12-7

IA =

@ DeCarlo & P. M. Lin

Vp

= 34.08 26.56
(Z / 3)
IB = IA 120 = 34.08 146.56
IC = IA 120 = 34.0893.43
The power factor of the load is

5
pf = cos tan1 = 0.8944

10
The total power is P = 3VL IL pf = 3127 34.08 pf = 11.61 kW. The voltage across each load is
3 V p = VL = 219.97 V.

SOLUTION 12.17. (a) Referring to figure 12.2, the following relationships may be pointed out:
VAB = VAN + VNB = VAN VBN
VBC = VBN + VNC = VBN VCN
VCA = VCN + VNA = VCN VAN
(b) Perform a delta to Y-transformation and use Ohms law and phase inference to obtain in amps:
220 30
= 11.36 56.57
3Z
IB = 11.36 176.57
IC = 11.3663.43
IA =

(c) First,

5
pf = cos tan1 = 0.8944

10
The average power, Pav = 3VL IL pf = 3 220 I A 0.8944 = 3.87 kW.
2

The reactive power may be found as follows, Q = 3 5IL = 1.936 kvar.

SOLUTION 12.18. Part (a) Apply Y transformation to the source.

Apply the transformation of row 3, Table 12.1 to the source. The result is the circuit shown in figure
12.19 of example 12.8. Therefore we may use many of the calculated values in example 12.9. In
particular, the line currents are
IA = 19.49/-68.9o A, IB = 19.49/-188.9o A, and

IC = 19.49/51.09o A

Part (b). Compute line-to-neutral voltage. From figure 12.21 and by Ohm's law

9/26/01

P12-8

@ DeCarlo & P. M. Lin

VAN = IA ZY = 19.49/-68.9o 536.87o = 97.4532.03o V

By inference on phase angles
VBN = 97.45152.03o V and VCN = 97.4587.97o V
Part (c). Compute line voltages VAB , VBC , and VCA . From the voltage phasor diagram of figure 12.2.
VAB = 3 VAN e j 30 = 168.8e j 2.03

VBC = VABe j120 = 168.8e j122.03

VCA = VABe j120 = 168.8e j117.97

By inference on phase angles,

o

Part (d). Compute the total power. From equation 12.4

Ptotal = 3 VL IL pf = 3 168.8 19.49 0.8 = 4558 W

SOLUTION 12.19. Perform a delta to Y transformation on the source. This yields

Vp
380
V1N =
30o =
30o V. Perform a delta to Y-transformation on the load. This yields a
3
3
Z
new impedance Znew = = 4 + j 3 . Now analyzing a single phase and finding the others by
3
inference yields, in amps,
V1N
IA =
= 42.18 67.97
Z1 + Znew
IB = 42.18 187.97
IC = 42.1852.03
2

The average power delivered to the load is Pav = 3 IA 4 = 21.35 kW.

SOLUTION 12.20. Performing a delta-to-Y-transformation on the source and the load, the new source
magnitudes are multiplied by 1 3 , and the loads are divided by 3. Observe that Zg = 0.15 + j0.45 ,
Z1 = 0.1 + j0.2 , and Z = 12 + j9 . Using Ohms law on phase one, and then inference on the
phase angles for the other currents, yields in amps,
IA =

V p 30
= 41.14 68.91
Zg
Z
3
+ Z1 +
3
3

IB = 41.14 188.91
IC = 41.1451.09

9/26/01

P12-9

@ DeCarlo & P. M. Lin

12 2
The total power may be calculated as follows, Pav = 3 I L = 12 (41.14)2 = 20.310 kW.
3

SOLUTION 12.21. Perform delta-to-Y-transformation on the delta load. Because of the property of the
neutral to be ground in balance circuit, the two loads in the same phase combine in parallel. Thus the new
impedance seen in one branch is
Z
(5 + j5) (5 j 3)
ZYnew = ZY || =
= 4.044 + j0.1923
3
(5 + j5) + (5 j3)
Now looking at a single phase and inferring for the others,
IA =

V p 30

230 30
= 32.84 32.73 A
3 ZYnew
3( 4.044 + j0.1923)
IB = 32.84 152.73 A
IC = 32.8487.27 A
=

The complex power is S = 3 ZYnew IL = 13.07 + j0.615 kVA.

SOLUTION 12.22. For this problem we do not need to use any 3-phase knowledge if we are clever.
We will provide a clever solution here. VDN = VDB VNB . Using B as the reference we write a node
equation at N to obtain:
V
V
V
V
V VAB VNB VCB
0 = NB + NA + NC = NB + NB
+
Z1
Z1
Z1
Z1
Z1
Z1
Therefore
3VNB = VAB + VCB = 115 115 120o = 172.5 + j99.59
From voltage division
VDB =

Z3
2 + j4
VAB =
115 = 57.5 + j57.5 V
Z3 + Z2
2 + j4 + 4 j2

Now using MATLAB we have

115-115*exp(-j*2*pi/3)
ans = 1.7250e+02 + 9.9593e+01i
VNB =ans/3
VNB = 5.7500e+01 + 3.3198e+01i
VDB = 115*(2+j*4)/(6+j*2)
VDB = 5.7500e+01 + 5.7500e+01i
VDN = VDB - VNB
VDN = 7.1054e-15 + 2.4302e+01i
magVDN = abs(VDN)
magVDN = 2.4302e+01
angVDN = angle(VDN)*180/pi
angVDN = 9.0000e+01

9/26/01

P12-10

@ DeCarlo & P. M. Lin

SOLUTION P12.23. First we write the following two loop equations,

VsA VsA 120 = IA Zg + Z1 + ZA + ( IC + IA) Z + Z1 + Zg

VsA120 VsA 120 = (IA + 2IC ) Zg + Z1 + Z

and solve in MATLAB:
(a)
%Problem 12.23a
Zg=0.05+0.15i;
Z=4+3i;
Z1=0.1+0.2i;
ZA=5+2i;
VA= 120;
VB= 120*exp(-120*i*pi/180);
VC= 120*exp(120*i*pi/180);
%Write mesh equations in matrix form and solve:
% i.e. [IA;IC] = A - 1*C
A=[ 2*(Z1+Zg)+ZA+Z Z+Z1+Zg; Zg+Z1+Z 2*(Zg+Z1+Z)];
Ainv= A^(-1);
C=[VA-VB;VC-VB];
B=Ainv*C;
IA= B(1);
IC= B(2);
IB= -IA-IC;
IA
abs(IA)
180*angle(IA)/pi
IB
abs(IB)
180*angle(IB)/pi
IC
abs(IC)
180*angle(IC)/pi
%And By Ohm's law,
VAN= IA*ZA
abs(VAN)
180*angle(VAN)/pi
VBN= IB*Z
abs(VBN)
180*angle(VBN)/pi
VCN= IC*Z
abs(VCN)
180*angle(VCN)/pi
IA = 21.7629.12 A
IB = 23.88155.60 A
IC = 20.64 82.39 A

9/26/01

P12-11

@ DeCarlo & P. M. Lin

VAN = 117.20-7.32 V
VBN = 119.4118.73 V
VCN = 103.2119.26 V
(b)
% From phasor diagram
VAB = VAN - VBN
abs(VAB)
180*angle(VAB)/pi
VBC = VBN - VCN
abs(VBC)
180*angle(VBC)/pi
VCA = VCN-VAN
abs(VCA)
180*angle(VCA)/pi
VAB = 195.5 27.34
VBC = 194.9 92.04
VCA = 196.99 147.8
(c) Calculating the average power delivered to each load using the general formula
Ptot = R I

= 5 (21.76) 2 + 4 (23.88)2 + 4 (20.64) 2 = 6.353 kW

SOLUTION P12.24. Doing this problem in MATLAB,

%Problem 12.24
ZDg=0.15+0.45i;
ZD=12+9i;
Z1=0.1+0.2i;
ZDA=15+6i;
%Apply the delta to Y transformation to the load as follows,
ZA=(ZD*ZDA)/(2*ZD+ZDA);
ZB=(ZDA*ZD)/(2*ZD+ZDA);
ZC=(ZD*ZD)/(2*ZD+ZDA);
%Apply delta to Y trans. to the source,
V1= 180/sqrt(3)*exp(-30*i*pi/180);
V2= 180/sqrt(3)*exp(-150*i*pi/180);
V3= 180/sqrt(3)*exp(90*i*pi/180);
Zg= ZDg/3;
%Write nodal equations in matrix form and solve
%i.e. [IA;IC]=A-1*C
A=[ 2*(Z1+Zg)+ZA+ZB ZB+Z1+Zg; Zg+Z1+ZB 2*(Zg+Z1)+ZC+ZB];
Ainv= A^(-1);
C=[V1-V2;V3-V2];
B=Ainv*C;
IA= B(1);
IC= B(2);
IB= -IA-IC;
IA
abs(IA)

9/26/01

P12-12

@ DeCarlo & P. M. Lin

180*angle(IA)/pi
IB
abs(IB)
180*angle(IB)/pi
IC
abs(IC)
180*angle(IC)/pi
IA = 17.4 63.1 A
IB = 20.1 178.9 A
IC = 19.5 51 A
(b)
% By Ohm's law,
VAN= IA*ZA;
VBN= IB*ZB;
VCN= IC*ZC;
%From phasor diagram
VAB=VAN - VBN
abs(VAB)
180*angle(VAB)/pi
VBC = VBN-VCN
abs(VBC)
180*angle(VBC)/pi
VCA =VCN - VAN
abs(VCA)
180*angle(VCA)/pi
VAB = 170.40 2.5 V
VBC = 168.58 122.4 V
VCA = 169.75 118.1 V
(c)
% by Ohm's law
IAB=VAB/ZDA
abs(IAB)
180*angle(IAB)/pi
IBC=VBC/ZD
abs(IBC)
180*angle(IBC)/pi
ICA=VCA/ZD
abs(ICA)
180*angle(ICA)/pi
o

IAB = 10.55 24.3 A

IBC = 11.24 159.3 A
ICA = 11.32 81.2 A
(d)
Ptot = R I

= 15 (10.55) 2 + 12 (11.24) 2 + 12 (11.32) 2 = 4.721 kW

9/26/01

P12-13

@ DeCarlo & P. M. Lin

SOLUTION P12.25. The average power that the load draws is P = 440 40 pf = 8.8 kW. From this we
obtain the old and new complex powers:
Sold = 8.8 + j15.242 kVA and Snew = 8.8 + j15.242 + jQC kVA
2

Here the reactance supplied by the capacitor is QC = CV l = 7299 var. So the new power factor is

Q
pf = cos tan1 new = 0.742 . From S = Vl Il = 11.855 kVA, we solve for

8800
Il = IA = IB = 26.94 A.

SOLUTION P12.26. Using Ohms law and the power factor of the load,
IC = 440 C90 = 16.59 90 A, IA = 4060 A, and IB = IC IA = 26.94 137.9 A.

SOLUTION P12.27. By simply applying Ohms law to each load and then KCL at node N, we get the
following:
V
V
V
IA = A = 4.4 A, IB = B = 4.4 120 A, IC = C = 0.88120 A, and
50
50
250
IN = IA + IB + IC = 3.52 60 A.
Finally,
PA = 50 (4.4) 2 = 968 W, PB = 50 (4.4 )2 = 968 W, PC = 250 (0.88) 2 = 193.6 W.

SOLUTION P12.28. Using MATLAB,

VA = 220
VB = 220*exp(-120*pi*i/180);
VC = 220*exp(120*pi*i/180);
%Use KVL to solve for the different currents
IA = (VA - VB)/50 + (VA - VC)/50
abs(IA)
180*angle(IA)/pi
IB = (VB - VC)/100 + (VB - VA)/50
abs(IB)
180*angle(IB)/pi
IC = (VC - VB)/100 + (VC - VA)/50
abs(IC)
180*angle(IC)/pi
IA = 13.2 A
IB = 10.1 130.9 A

9/26/01

P12-14

@ DeCarlo & P. M. Lin

IC = 10.1 130.9 A
2

V
The total power delivered to each resistor may be calculated using the following relationship P = l ,
R
2

where Vl = 3V p
PCA = 2904 W.

using Y to Delta source transformation. Thus PAB = 2904 W, PBC = 1452 W, and

SOLUTION P12.29. Consider the circuit of figure 12.29 with two additional node labeling:

Choose N as the reference node and apply KCL to node M. This yields the following node equation:
o
o
0.02(VM 220) + 0.02 VM 220e j120 + 0.004 VM 220e j120 = 0

Equivalently

2.2VM = 88 j152.42 or VM = 40 j69.282 V

The remaining calculations proceed in MATLAB as follows:

% Compute node voltage VM
X = +220 +220*exp(-j*2*pi/3) + 0.2*220*exp(j*2*pi/3)
X = 8.8000e+01 - 1.5242e+02i
VM = X/2.2
VM = 4.0000e+01 - 6.9282e+01i
% Compute line currents
IA = 0.02*(220 - VM)
IA = 3.6000e+00 + 1.3856e+00i
magIA = abs(IA)
magIA = 3.8575e+00
angIA = angle(IA)*180/pi
angIA = 2.1052e+01
IB = 0.02*(220*exp(-j*2*pi/3) - VM)
IB = -3.0000e+00 - 2.4249e+00i
magIB = abs(IB)
magIB = 3.8575e+00
angIB = angle(IB)*180/pi
angIB = -1.4105e+02
IC = 0.004*(220*exp(j*2*pi/3) - VM)
IC = -6.0000e-01 + 1.0392e+00i
magIC = abs(IC)

9/26/01

P12-15

@ DeCarlo & P. M. Lin

magIC = 1.2000e+00
angIC = angle(IC)*180/pi
angIC = 1.2000e+02
% Compute total power
Ptot = 50*magIA^2 + 50*magIB^2 + 250*magIC^2
Ptot = 1.8480e+03

SOLUTION P12.30. (a) The equivalent resistance of a 40 ft size #14 wire is

2.575
Req1 = 40
= 0.103 . The voltage at the light and appliance is then
1000
Vapp = 115 2 Req110 = 112.94 V, which is equivalent to a 1.8% drop, thus the wire gauze is
appropriate.
1.619
= 0.08095 .
1000
Vapp = 115 2 Req214 = 112.7334 V, which corresponds to 1.97 %. So the wire gauze is appropriate.

(b) Repeating the previous calculations, Req2 = 50

(c) Under normal operating condition the current in the live wire should equal the current in the neutral
wire; thus no current should be in the ground wire.
(d) Note the following relationship, VAB = VAN + VNB = VAN VBN . This means that
VAN = 115 V and VBN = 115180 V.

SOLUTION P12.31. Assuming a very large resistance for the person who touches the prong. For prong
A, the voltage is approximately V = 115 V. For prong N, the voltage is approximately V = 0. For the
ground prong G, no current flows through the wire, so the voltage should be ground or zero.

SOLUTION P12.32. For an approximate analysis, we use the circuit models below given the following
assumptions:
(i) The resistance from body to ground is very large (possibly due to rubber shoes);
(ii) The resistances of all connecting wires are negligible.
(iii) The motor winding is represented by R1 in series with R2 . Here we can further estimate
that. R1 + R2 = 115/3 = 38.3 , and R1 = R2 = 19.15 .
(iv) The trigger switch of the drill has been depressed.
(a) The point P is connected to the metal case as shown below. It is obvious that Vcase = Vp = 0.

9/26/01

P12-16

@ DeCarlo & P. M. Lin

(b) The point M is connected to the metal case as shown below. The current through the hot wire is
115/19.15 = 6 A. If the fuse capacity is smaller than 6 A, then it will blow and Vcase = VM = 0. On the
other hand , if the fuse capacity is greater then 6 A, then R2 is nearly shorted and Vcase = VM = 0.

(c) The point Q is connected to the metal case as shown below.

9/26/01

P12-17

@ DeCarlo & P. M. Lin

Both R1 and R2 are nearly shorted. The line current is so large that it will blow the fuse. Hence Vcase
= VQ = 0.
CONCLUSION: In all three cases, the person touching the metal case of the defective appliance will
experience zero or a very low voltage. The circuit is safe.

SOLUTION P12.33. For an approximate analysis, we use the circuit models below given the following
assumptions:
(i) The resistance from body to ground is very large (possibly due to rubber shoes);
(ii) The resistances of all connecting wires are negligible.
(iii) The motor winding is represented by R1 in series with R2 . Here we can further estimate
that. R1 + R2 = 115/3 = 38.3 , and R1 = R2 = 19.15 .
(iv) The trigger switch of the drill has been depressed.
(a) The point P is connected to the metal case as shown below. It is obvious Vcase = Vp = 0.

(b) The point M is connected to the metal case as shown below. Simple voltage divider action leads to

9/26/01

P12-18

@ DeCarlo & P. M. Lin

Vcase = VM = 0.5115 = 57.5 V

(c) The point Q is connected to the metal case as shown below.

It is obvious that Vcase = VQ = 115 V.

CONCLUSION: In all three cases, the fuse will not blow. For cases 2 and 3, the voltage appearing on
the metal case may cause serious injury to the person.

Prbs Ch 13

March 18, 2002

P13-1

R. A. DeCarlo, P. M. Lin

PROBLEM SOLUTIONS CHAPTER 13

S OLUTION 1 3 . 1 . Given
i (t) + 16i (t) + 4Bi(t) = v (t) + 8v(t)
(a) with v(t) = v (t) = 0 and i1 (0) + i2 (0) = 0
(i) the characteristic equation is
s2 + 16s + 48 = 0
(ii) the characteristic equation has factors
(s + 4) (s + 12) = 0
and hence
s1,s2 = 4,12
(iii) Equivalent circuit at t = 0

(iv) Here by KCL

i1 (0) = i 2 (0) = 6A
At point (1)
i1 (0) + i2(0) = i2 (0)
6A + i2(0) = 6A
i2(0) = 0
and
v L1 = v R2 (0) = 2i 2 (0) = 0V
Then by KVL at t = 0
v L1 (0) v 6 (0) + v L 2 (0) = 0 v L 2 = 36V
di2
= 36V
dt
i (0) = i2 (0) = 36A / s
L2

(v) from part ii

i(t) = Ae 4t + Be12t
and then
i (t) = 4 Ae4t 12Be 12t
Then using the initial conditions
A+ B=6
4A 12B = 36
1

Prbs Ch 13

March 18, 2002

P13-2

R. A. DeCarlo, P. M. Lin

solving yields B = 1.5 and A = 4.5. Then

i(t) = 4.5e 4t + 1.5e 12t A
(b)

If v(t) = 12V and i(0) = i (0) = 0 , then v (t) = 0 and

i (t) + 16i (t) + 48i(t) = 8v(t)

with
i(t) = Ae 4t + Be12t
i(t) = 4Ae 45 12Be12t
i (t) = 16Ae 4t + 144Be12t
Then at t = when v() = 12V
i() = C, i () = 0 and i () = 0
Thus
48C = 96

C=2

Then
i(t) = Ae 4t + Be12t + 2
and
i(0) = 0 = A + B + 2
i (0) = 0 = 4 A 12B
Multiply the first of these by 4 yields
4 A + 4B = 8
4A 12B = 0
Solving yields B = 1 and A = -3. Thus for t > 0
i(t) = 3e 4t + e 12t + 2 A
(c) If v(t) is as in fig. 13.1(b) then
2t 0 < t 2
v(t) =
t2
0
The value of C in part (b) will change and at t = 2s, a new set of initial conditions will be required
(obtainable from the solution at t = 2s) and these would be used in the decay portion described by
i (t) + 16i (t) + 48i(t) = 0 t 2s

Prbs Ch 13

March 18, 2002

P13-3

R. A. DeCarlo, P. M. Lin

S OLUTION 1 3 . 2 .
(a) Use the figure with the currents i1 through i5 designated in the circuit below.
Then work from v 0 to vin using repeated applications of KVL, KCL and the elemental equations:
i1 = 2vo
i2 = 2vo

i3 = i1 + i2 = 2(v o + v o)

with
v 2 = vo

v1 = 0.5i3 + v 2 = 0.5[2(vo + vo ) + vo ] = vo + 2v o

then

d 1
v + 2vo = 2(vo + 2vo )
dt o
i5 = i3 + i4 = 2(v o + v o) + 2(vo + 2vo ) = 2vo + 6vo + 2vo

i4 = 2v1 = 2

Finally
vin = 0.5i5 + v1 = 0.5(2vo + 6vo + 2vo ) + v o + 2v o = vo + 4v o + 3v o
Hence
vout (t) + 4vout (t) + 3v out (t) = vin (t)
(b) Note from part (a) that
vout (t) = v 2 (t)
and
v1(t) = vout (t) + v out (t)
Hence
vout (o) = v 2 (o) = 1V
vout (o) = v1(o) vout (o) = 7 1 = 6V
(c) The characteristic equation
s2 + 4s + 3 = 0
has factors

(s + 1)( s + 3) = 0
and roots
s1,v 2 = 1,3
3

Prbs Ch 13

March 18, 2002

P13-4

R. A. DeCarlo, P. M. Lin

Thus, because of the input, vin (t) = 6V

vout (t) = Aet + Be 3t + C
vout (t) = Aet 3Be3t
vout (t) = Aet + 9Be 3t
and at t =
vout () + 4vout () + 3v out () = 6V
0

0
+ 3C
C = 2V

= 6V

and
vout (t) = Aet + Be 3t + 2V

S OLUTION 1 3 . 3 . (a) From the given differential equation, the characteristic equation is
s3 + 14s2 + 52s + 24 = (s + 6)(s2 + 8s + 4) = 0
Therefore the roots a = 6, b = 4 + 2 3 = 0.5359,c = 4 2 3 = 7.4641.
(b)

(i) v(0) = v C3 (0) = 6 V, as given.

(ii) To compute v'(0) we write a nodal equation at node 3. In particular

(v vC1 ) + (v v C2 ) + 0.5v'+v = 0 which implies that

v'(t) = 2vC1 (t) + 2vC 2 (t) 6v(t)

(*)

Hence v'(0) = 2vC1 (0) + 2vC 2 (0) 6v(0) = 24 + 18 36 = 6 V/s.

(iii) To compute v''(0) we first differentiate equation (*). This yields
v''(t) = 2v'C1 (t) + 2v'C 2 (t) 6(2vC1 (t) + 2vC 2 (t) 6v(t))

(**)

To express v'C1 (t) and v'C 2 (t) in terms of the node voltages we write node equations at nodes 1and
2 respectively. At node 1
'
(vC1 (t) vC 2 (t)) + (vC1 (t) v(t)) + 0.5vC1
(t) = i(t)

and at node 2
(vC 2 (t) vC1 (t)) + (vC 2 (t) v(t)) + 0.5v 'C 2 (t) = 0
Hence
v'C1 (t) = 4v C1(t) + 2v C 2 (t)) + 2v(t) + 2i(t)
4

Prbs Ch 13

March 18, 2002

P13-5

R. A. DeCarlo, P. M. Lin

and
vC' 2 (t) = 2vC1 (t)) 4vC 2 (t) + 2v(t)
Substituting these two equations into (**) yields the desired result when t is set to 0. However, this
quantity has no direct physical meaning.
v''(t) = 2(4vC1(t) + 2vC 2 (t)) + 2v(t) + 2i(t)) + 2(2vC1 (t)) 4vC 2 (t) + 2v(t))
6(2vC1 (t) + 2vC 2 (t) 6v(t))

Hence, v''(0) = 2(48 + 18 + 12 + 2i(0)) + 2(24 36 + 12) 6(24 + 18 36) = 72 + 4i(0) V/s .
Finally, using the characteristic roots found in part (a) and assuming a constant input, the form
of the solution is
v(t) = Ae6t + Be0.5359t + Ce 7.4641t + D
Following the methods of example 13.2,
A = 0, B = (7.3301 1.0774i(0)), C = (1.3301 + 0.0774i(0)), and D = i(0).
(c)

(i) Not proportional to a single voltage but it is proportional to iC3 (0).

(ii) Much more complex.

(iii) No.
(iv) No. This is why we use the Laplace transform approach.

S OLUTION 1 3 . 4 .
(a) f (q + T0 )]q=t = f (t + T0 )
(b) e 5q cos(0.5q + 0.25 )

]q=2t = e10t cos(t + 0.25)

S OLUTION 1 3 . 5 .
(a) Let T = 1.

Prbs Ch 13

March 18, 2002

P13-6

R. A. DeCarlo, P. M. Lin

1.5

0.5

0
-1

-0.5

0.5

1.5

2.5

(b) Again let T = 1.

0

-0.5

-1

-1.5

-2
-1

-0.5

0.5

1.5

2.5

-0.5

0.5

1.5

2.5

(c)
0.25
0.2
0.15
0.1
0.05
0
-1

(d)

Prbs Ch 13

March 18, 2002

P13-7

R. A. DeCarlo, P. M. Lin

1
0.8
0.6
0.4
0.2
0
-1

-0.5

0.5

1.5

2.5

(e)
7
6
5
4
3
2
1
0
-1

(f)

-0.5

0.5

1.5

-0.5

0.5

1.5

2.5

(t i)

i =0

(g)
5
4
3

2
1

0
-1

2.5

(h) Pulses of height 1 and width T.

7

Prbs Ch 13

March 18, 2002

P13-8

R. A. DeCarlo, P. M. Lin

S OLUTION 1 3 . 6 .
(a) L[ f1(t)] = F1(s) =

f (t)e
0 1

st

dt ==

T 2 st
e dt =
T1

T2

e st

T1

1 sT 1
e
e sT 2
s

(b) f 2(t) = f1(t) . Hence, the answer is the same as in (a).

(c) L[ f 3(t)] = F3(s) =
(d) F4 (s) =
(e) F5(s) =

f 3 (t)e st dt = 2

f 4 (t)e st dt = 2
f 5(t)e st dt =

(t)cos(4 t 0.25)e st dt = 2

2
= 2
2

(t T )cos(4t 0.25)e st dt = 2cos(4 T 0.25)e sT

(t) (t T )]e st dt = 1 e sT

S OLUTION 1 3 . 7 .
0

f (t)e

st

dt =

5e

4t st

e (s+ 4)t
dt = 5

s + 4

5
s+ 4

1.4

1.6

5
4.5
4
3.5
3

f(t)

(a) F(s) =

2.5
TextEnd
2
1.5
1
0.5
0

0.2

0.4

0.6

0.8

1
t

1.2

1.8

Prbs Ch 13

March 18, 2002

(b) F(s) =

f (t)e

st

P13-9

dt = 5e
1

4t st

R. A. DeCarlo, P. M. Lin

e (s +4) t
dt = 5

s + 4

4
5e (s +4)
s 5e
=
=e
s+4
s+4

0.1
0.09
0.08
0.07

f(t)

0.06
0.05
TextEnd
0.04
0.03
0.02
0.01
0

(c) F(s) =

f (t)e

st

0.2

0.4

dt = 5e

0.6

4(t 1) st

0.8

1
t

1.2

1.4

1.6

e (s+ 4)t
dt = 5e

s + 4
4

1.8

= e s

5
s+ 4

5
4.5
4
3.5

f(t)

3
2.5
TextEnd
2
1.5
1
0.5
0

(d) F(s) =
(e) F(s) =

0.2

f (t)e st dt =

f (t)e st dt =

0.4

0.6

0.8

1
t

1.2

1.4

1.6

1.8

5e 4( t 1) (t)e st dt = 5e 4

5e 4( t 1) (t 1)e st dt = 5e s

Prbs Ch 13

March 18, 2002

P13-10

R. A. DeCarlo, P. M. Lin

(f)
F(s) =

f (t)e st dt = 2

= 2

[ u(t)u(1 t)]e 0.25t e stdt = 20

1

e 0.25t e stdt 2

e 0.25te st dt =

e 0.25te st dt =

2
1 e (s +0.25)
s + 0.25

2
1 e s
s + 0.25

2
1.8
1.6
1.4

f(t)

1.2
1
TextEnd
0.8
0.6
0.4
0.2
0

F(s) =

0.2

0.4

0.6

0.8

1
t

1.2

1.4

1.6

1.8

f (t)e st dt =

S OLUTION 1 3 . 8 .
(a) F(s) =

f (t)e st dt =

f (t)e st dt =

f (t)e st dt =

st

(b) F(s) =
(c) F(s) =

(d) F(s) =

0
0

f (t)e

dt =

Ae t e stdt = Ae(s )t dt =
0

A
s

Ae (t 1) u(t 1)e st dt = Ae e (s )t dt = e s
1

A
s

Ae (t 1) (t)e st dt = Ae
t st

Ae e

1
e (s )t
A
dt = A
=
(s ) s
0

(1 e(s ))
10

Prbs Ch 13

March 18, 2002

P13-11

R. A. DeCarlo, P. M. Lin

S OLUTION 1 3 . 9 . Consider the following in which q = at and t = q/a.

L[ f (at)] =

f (at)e

st

1
dt =
a

f (q)e (s/a)q dq =

1 s
F
a a

S OLUTION 1 3 . 1 0 .
(a)
e j t = cos t + j sin t
e j t = cos t j sin t
Add these equations and divide by 2 to obtain
e j t + e j t
2
Similarly, subtract the equations and divide by 2j to obtain
cos( t) =

sin( t) =

e j t e j t
2j

Note that

[ ]

L ej t =

1
1
, L e j t =
s j
s+ j

(b)
e j t + e j t
0.5
0.5
s
L[cos t ] = L
=
+
= 2
2
s+ j
s + 2

s j
and
L[sin t] =

0.5 j
0.5 j
+
= 2
s j
s+ j
s + 2

S OLUTION 1 3 . 1 1 . From the time frequency scaling property,

11

Prbs Ch 13

March 18, 2002

P13-12

L[sin( t) u(t)] =

R. A. DeCarlo, P. M. Lin

1
2

s
+1

= 2
s + 2

Using the time differentiation property,

s
d

L[ cos( t)u(t)] = L (sin( t)u(t)) = sL[sin( t) u(t)] sin(0) = 2

dt

s + 2
Finally using the multiplication by t property

L[ t cos( t)u(t)] =

d
s
2 s2
=

2
ds s + 2 s2 + 2
s2 + 2

Hence

( s2 + 2 ) 2
2
(s 2 + 2 )

2
2
2 s
L[sin( t)u(t) t cos( t)u(t)] = 2

=
2
s + 2
s2 + 2

s2

2 3

(s 2 + 2 )

S OLUTION 13.12. (a)

1
g1 (t) = At sin( t), f (t) = sin(t)u(t), F (s) = 2
s +1
By the frequency scaling property

2
1
1
= 1
L[ f ( t)] =
2

2 + 2 = s2 + 2
s
s

+ 1

By the multiplication by t property

2 s
2 s
L[t sin( t)u(t)] = 2
=

2
2
ds s + 2
s2 + 2 s2 + 2

By the linearity property

L[ At sin( t)] = A

2 s

(s 2 + 2 )

12

Prbs Ch 13

March 18, 2002

P13-13

R. A. DeCarlo, P. M. Lin

(b)
1
g2 (t) = Ae at sin( t)u(t), f (t) = sin(t)u(t), F(s) = 2
s +1
By the frequency scaling property,
L[sin( t)] = 2
s + 2
By the damping property

L e at sin( t) =

(s a) 2 +

and by the linearity property

L[g2 (t)] = A

(s a) 2 +

(c) Before beginning, note that

sin( t + ) = cos( )sin( t) + sin( )cos( t)
Here
g3 (t) = Aeat sin( t + )u(t) = cos( )Aeat sin( t)u(t) + sin( )Aeat cos( t)u(t)
1
f (t) = sin(t)u(t), F(s) = 2
s +1
From the linearity property and part (b),
G3(s) = Acos( )

(s a) +
2

+ L{sin( )Aeat cos( t)u(t)}

d
sin(t) which implies
dt
s
s
L[cos(t)] = s{L[sin(t)]} f (0) = 2
0= 2
s +1
s +1

By the differentiation in the time domain property cos(t) =

By the frequency scaling property

s
2
1
1 s
L[cos t ] =
=

2
2 s2 +

+ 1
s

( )

=
2
s2 + 2

Then by the linearity property

13

Prbs Ch 13

March 18, 2002

P13-14

R. A. DeCarlo, P. M. Lin

s
L[ Asin( )cos( t)] = A sin( ) 2
s + 2
Then by the frequency-shift property

sa
(s a)2 + 2

L Asin( )e at cos( t) = Asin( )

G3(s) = L Aeat sin( t + ) = Acos( )

=

(s a) 2 +

sa
+
Asin(
)
(s a)2 + 2
(s a) 2 + 2

( cos( ) + sin( )(s a))

S OLUTION 13.13. (a) We are given that

1
f ( t) = sin( t) F(s) = 2
s +1
And must find the transform of
g1 (t) = At cos( )tu(t)
By the time differentation property
cos(t ) =

d
sin(t)
dt

( )

s
s
L[cos(t )] = 2
sin 0 = 2
s +1
s +1

By the frequency scaling property with

L[cos( t )] =

s 1

F =
=
2
s
+
1

( )

s 2
s2

s
= 2
s + 2

Using the multiplication-by-t property

d s

s2 + 2 s(2s)
s2 2
L[t cos( t )] = 2
=
2 = 2
2
ds s + 2
s2 + 2
s + 2

Finally, by the linearity property

L[ At cos( t )u(t)] =

s2

(s 2 + )

2 2

14

Prbs Ch 13

March 18, 2002

P13-15

R. A. DeCarlo, P. M. Lin

(b) Let g2 (t) = Ae at cos( t )u(t) . Recall from part (a) that
s
L[ Acos( t)] = A 2
s + 2
By the Frequency-shift property

L Aeat cos( t) = A

s a

(s a) 2 +

(c) From Trig identities,

cos( t + ) = cos( t)cos( ) sin( t)sin( )
Recall from part (b) that

L Aeat cos( t) = A

s a

(s a) 2 +

Hence

s a

L Aeat cos( )cos( t) = Acos( )

sa

( s a)2 +

It follows that

L Aeat cos( t + ) = Acos( )

= A cos( )

(s a) +
2

A sin( )L e at sin( t)

cos( )(s a) sin( )

A
sin(
)
=
A

(s a) 2 + 2
(s a) 2 + 2

(s a) 2 + 2

s a

S OLUTION 1 3 . 1 4 . (a) We are given that

s
L[cos(t )] = 2
s +1
which implies by the frequency scaling property that
s
L[cos( t )] = 2
s + 2
Using the multiplication-by-t property
L[t cos( t )] =

d s

s2 + 2 2s2
s2 2
=

=
2
2
ds s2 + 2
s2 + 2
s2 + 2

15

Prbs Ch 13

March 18, 2002

P13-16

R. A. DeCarlo, P. M. Lin

(b) If g(t) = teat cos( t), then using part (a) and frequency shift property,

at

L te cos( t) =

(s a) 2

[( s a)

2
+ 2

S OLUTION 1 3 . 1 5 . (a) With

e at e at
sinh(at) =
2
L[sinh(at )] =

1 1
1 1 s + a ( s a) 1 2a
a

=
= 2
= 2

2
2
2
2 s a s + a 2 s a
2 s a s a2

(b) With
e at + e at
cosh(at) =
2
L[cosh(at)] =

1 1
1 1 s a + s + a 1 2s
s
+
= 2
= 2
= 2

2
2
2 s + a s a 2 s a
2 s a s a2

S OLUTION 1 3 . 1 6 . (a) From Problem 15

a
L[sinh(at )] = 2
s a2
So that by the multiplication by t property
L[t sinh(at )] =

d a
a(2s)
2as
=
=

2
2
2
2
ds s a
s2 a 2
s2 a 2

) (

(b) From Problem 15

s
L[cosh(at)] = 2
s a2
So that by the multiplication-by-t property

16

Prbs Ch 13

March 18, 2002

P13-17

R. A. DeCarlo, P. M. Lin

s2 a 2 s(2s) s2 + a 2 + 2s2
d s
s2 + a2
L[t cosh(at)] = 2
=
=
=
2
2
2
ds s a 2
s2 a2
s2 a2
s2 a 2

S OLUTION 13.17. Here

s+2
s +1
Since g1 (t) = 5 f (t 2), use the time shift and linearity properties to obtain
F(s) =

(a)

s + 2
L[g1 (t)] = 5e 2s
s + 1
(b)

Since g2 (t) = 5e 2t f (t), use the frequency shift and linearity properties,
( s + 2) + 2
s + 4
L[g2 (t)] = 5
= 5

s + 3
( s + 2) + 1

(c) From part (a),

s + 2
L[5 f (t 2)] = 5e 2s
s + 1
Therefore, since g3 (t) = 5e 2t f (t 2) , by the frequency shift property
(s + 2) + 2
s + 4
L[g3 (t)] = G1(s + 2) = 5e 2(s+ 2)
= 5e 2(s +2)

s + 3
(s + 2) + 1
(d)

Since g4 (t) = 5tf (t) , use the multiplication-by-t and linearity principle to obtain
L[g4 (t)] =

d 5(s + 2)
5[(s + 1) (s + 2)]
5
=
=

ds s + 1
(s + 1)2
(s + 1)2

S OLUTION 1 3 . 1 8 . In all parts

s
L[ f (t)u(t)] = F (s) = 2
s +4

17

Prbs Ch 13

(a)

(b)
(c)

March 18, 2002

P13-18

R. A. DeCarlo, P. M. Lin

e Ts s
L[ Af (t T )u(t T )] = A 2
s +4

L[ A(t T ) f (t T)u(t T )] = Ae

(d)

(
(

)
)

s2 4
d
d s
s + 4 s(2s)
L[ Atf (t)u(t)] = Af (s) = A 2
= A
= A 2
2
2
ds
ds s + 4

s2 + 4

s +4

Note that the answer is simply a time shift of the function given in (b).

sT

( s2 4 )
2
(s 2 + 4 )

This function is that of part (a) multiplied by t. Hence, by the multiplication by t property,

Ts
Ts (s 2 + 4) 2s se Ts
d e Ts s
e Tse

L[ Atf (t T )u(t T )] = A 2
= A

2
ds s + 4

s2 + 4

3 2

2s

(
1
Ts
)
(s
+
4)

Ts
+
s
+
4Ts

= AeTs
= Ae Ts

2
2

s2 + 4

s2 + 4

S OLUTION 1 3 . 1 9 . In all parts

F(s) = L[ f (t)u(t)] = 2
s + 2
e Ts

(a) By the time shift property, L[ Af (t T )u(t T )] = A 2

s + 2

Using the multiplication-by-t property,

L[ Atf (t)u(t)] = A

d
d

(2s)
2 s
[F (s)] = A 2 2 = A
=A
2
2
ds
ds s +

s2 + 2
s2 + 2

18

Prbs Ch 13

March 18, 2002

P13-19

R. A. DeCarlo, P. M. Lin

The answer here is an application of the time shift property to the answer of part (b).
L[ A(t T ) f (t T)u(t T )] = AesT

2 s

(s 2 + 2 )

(d) The answer here uses the multiplication-by-t property applied to the answer of part (a).

d e Ts
L[ Atf (t T )u(t T )] = A
ds s2 + 2

= A

Ts 2

Te (s + 2 ) + 2se Ts

s2 + 2

2
2

Ts
+
2s
+
T

= A e Ts

2
s2 + 2

Solution 13.20. (a)

f (2t) = (2t) + (2t 1)
For the first term on the right, the peak occurs at t = 0 and
0+

0+

(2t)dt = 0.5

( )d = 0.5

under the transformation = 2t. For the second term, the function peaks at t = 0.5 and
0.5+

0+

(2t 1)dt = 0.5

0.5

( )d = 0.5

under the transformation = 2t 1. Hence

f (2t) = (2t) + (2t 1) = 0.5[ (t) + (t 0.5)]
Therefore, a = 0.5 = b.
19

Prbs Ch 13

March 18, 2002

P13-20

R. A. DeCarlo, P. M. Lin

(b) (i) Here F(s) = L[ f (t)] = 1 + e s . By the time scaling property

L[ f (2t)] =

1
F (s /2) = 0.5 1+ e 0.5s
2

(b)-(ii) For this part,

L[ f (2t)] = 0.5L[ (t) + (t 0.5)] = 0.5 1+ e 0.5s

S OLUTION 13.21. (a)

L[v(t)] = 2L[g"(t)] L[g'(t)] = 2s2F (s) 2sg(0 ) 2g'(0 ) sF(s) + g(0 )

(2s 1)(s + 1)
s2

= 2s2 s F (s) 2sg(0 ) 2g'(0 ) + g(0 ) =

(b)

L[v(t)] = 2L[ f "(t)] L[ f '(t)] = 2s2F (s) sF(s) 2sf (0 ) 2 f '(0 ) + f (0 )

= 2s2 s F (s) 2s 2 + 1 =

(2s 1)(s + 1)
s2

2s 2 + 1

(c)

0
t

G(s) 1
1

L[v(t)] = L[ g'( t)] L g(q)dq = sG(s) g(0 )

g(q)dq = s G(s)
s
s
s

1
1
1 s + 1
= s 3 = (s + 1) 2 4
s s
s
s
(d)

0
t

F(s) 1

L[v(t)] = L[ f '( t)] L f (q)dq = sF(s) f (0 )

f (q)dq
s
s

1
1
1
= (s + 1) 2 4 1
s
s
s

assuming > 0. The expression is ill-defined if 0.

20

Prbs Ch 13

March 18, 2002

P13-21

R. A. DeCarlo, P. M. Lin

S OLUTION 1 3 . 2 2 .
(a) (i) If
s2 + 4
F(s) = L[ f (t)u(t)] = ln 2 = ln s2 + 4 ln s2
s

) ( )

Then by the multiplication-by-t property

L[tf (t)u(t)] = +

[ (

) ( )]

d
ln s2 + 4 ln s2
ds

2s
2s
2s
2 2s2 2s2 8
8
= 2
2= 2
=
=
2
2
s +4 s
s +4 s
s s +4
ss +4

(ii)

Using the solution to (a)-(i), by the frequency shift property

L te 2t f (t)u(t) =

(s + 2) ( s + 2) + 4
2

8
= 3
s + 6s + 16s + 16

(s + 2) s2 + 4s + 8

(b) If
G(s) =

s s2 + 4

a partial fraction expansion may be employed

G(s) =

K1 As + B 2
2
=
+
=
+
2
2
s
s s +4
s +4
s s2 + 4

Hence,
g(t) = [2cos(2t) 2] u(t)
and
f (t) =

g(t) 2 2
= cos(2t)
t
t t

S OLUTION 13.23: Part (a)-(i): From table 13.2, the multiplication by t property implies that
21

Prbs Ch 13

March 18, 2002

L[tf (t)u(t)] =

P13-22

R. A. DeCarlo, P. M. Lin

d
d s + a d
1
1
2a
F(s) = ln
= (ln[s + a] ln[s a]) =

= 2

ds
ds s a ds
s + a s a s a2

2a
Part (a)-(ii): Let us make use of the answer to part (a)-1. Let G(s) = L[tf (t)u(t)]= 2
.
s a2
Then by the frequency shift property in table 13.2,
L[te at f (t)u(t)] = L[e at ( tf (t)u(t))] = G(s + a) =

2a
2a
=
2
2
s(s + 2a)
(s + a) a

2a
1 1
1 1
at
Part (b): g(t) = L1[G(s)]= L1 2
e at u(t)
= L s + a L s a = e
2
s a
More specifically,
e at e at
g(t) = e at e at u(t) = 2
u(t) = 2sinh(at) u(t)
2

Hence f (t) =

g(t) 2sinh(at)
=
.
t
t

S OLUTION 1 3 . 2 4 .
(a)-(i) If
s + a
F(s) = L[ f (t)u(t)] = ln
= ln(s + a) ln(s + b)
s + b
Then by the multiplication-by-t property
L[tf (t)u(t)] = +

d
1
1
s + b (s + a)
ba
ln (s + a) ln( s + b)] =

=
=
[
ds
s + a s + b (s + a)(s + b) (s + a)(s + b)

(ii) By the frequency-shift property

L te at f (t)u(t) =

ba
b a
= 2
(s + 2a)(s + a + b) s + s(3a + b) + 2a(a + b)

(b) If

22

Prbs Ch 13

March 18, 2002

P13-23

G(s) = L[tf (t)u(t)] =

R. A. DeCarlo, P. M. Lin

ba
(s + a)(s + b)

a partial fraction expansion may be employed

G(s) =

ba
K
K
1
1
= 1 + 2 =
+
(s + a)(s + b) s + a s + b s + a s + b

Hence, g(t) = e at e bt u(t)

and
f (t) =

g(t) e bt e at
=

t
t
t

d
S OLUTION 1 3 . 2 5 . The relationship is f (t) = g(t) or equivalently, g(t) =
dt

f (q)dq .

Now we have that f (t) = 6 (t) 12 (t 2) + 6 (t 4) . Therefore,

F(s) = 6 12e 2s + 6e 4s
From the time integration property,
F (s) 6 12e 2s 6e 4s
G(s) =
=
+
s
s
s
s

S OLUTION 1 3 . 2 6 . For 0 t < T1, we see that g(t) =

f (q)dq . Thus one presupposes here that

d
the relationship is f (t) = g(t) or equivalently, g(t) =
dt

f (q)dq . As such E = A B and D = A

B + C.
Further, f (t) = A (t) B (t T1 ) + C (t T2 ) which implies that
F(s) = A BeT 1s + Ce T 2 s
Thus
23

Prbs Ch 13

March 18, 2002

P13-24

G(s) =

R. A. DeCarlo, P. M. Lin

F (s) A BeT1s Ce T 2s
=
+
s
s
s
s

S OLUTION 1 3 . 2 7 .
(a) f (t) = u(t) + u(t 1)

1 e s 1
f (s) = +
= 1 e s
s
s
s

(b)
f (t) = u(t) + u(t 1) u(t 3)

1 e s e 3s 1
f (s) = +

= 1+ e s e 3s
s
s
s
s

(c)

f (t) = u(t) + u(t 1) 2u(t 3) F (s) =

1 e s 2e 3s 1
+

= 1 + e s 2e 3s
s
s
s
s

(d)
f (t) = 2u(t) u(t 2) u(t 3) F(s) =

2 e 2s e 3s 1

= 2 e 2s e 3s
s
s
s
s

S OLUTION 1 3 . 2 8 .
(a)

2 2e s
2
f (t) = 2r(t) 2r(t 1) F (s) = 2 2 = 2 1 e s
s
s
s

(b)

2 2e s e 2s
1
f (t) = 2r(t) 2r(t 1)+ r(t 2) F (t) = 2 2 + 2 = 2 2 2e s + e 2s
s
s
s
s

)
(

(c) f (t) = 2r(t) 2r(t 1) 2r(t 2) + 2r(t 3). It follows that

2 2e s 2e 2s 2e 3s
2
F(s) = 2 2 2 + 2 = 2 1 e s e 2s + e 3s
s
s
s
s
s

24

Prbs Ch 13

March 18, 2002

P13-25

S OLUTION 1 3 . 2 9 . (a) Here f (t) =

R. A. DeCarlo, P. M. Lin

3
3
r(t) 3r(t 2) + r(t 4) . Thus,
2
2

3
3e 2s 3e 4s
3
F(s) = 2 2 +
= 2 1 2e 2s + e 4s
2
2s
s
2s
2s

V
2V
V
(b) Here f (t) = o r(t) o r(t T ) + o r(t 2T ). Thus
T
T
T
V 1 2e T e 2T
F(s) = o 2 2 + 2
T s
s
s
(c) Here f (t) = 2r(t 1) 4r(t 2) + 4r(t 4) 2r(t 5)

1
F(s) = 2 2e s 4e 2s + 4e 4s 2e 5s
s

S OLUTION 1 3 . 3 0 .
(a) Here f (t) = 2r(t) 2r(t 1) 2u(t 4) implies
2 2e s 2e 4s 2
F(s) = 2 2
= 2 1 e s se 4s
s
s
s
s

(b) Here f (t) = 2u(t) r(t 2) + r(t 4) implies

F(s) =

2 e 2s e 4s 2
2 + 2 = 2 s e 2s e 4s
s
s
s
s

S OLUTION 13.31. (a) Here f (t) = 2u(t) r(t) + 2r(t 2) 2r(t 4) 2u(t 4)
2 1 2e 2s 2e 4s 2e 4s
Thus F(s) = 2 + 2 2
.
s s
s
s
s
1 1 e 2s e 2s
(b) f (t) = u(t) + r(t) r(t 2) u(t 2). Hence F(s) =
+ 2
.
s s2
s
s

25

Prbs Ch 13

March 18, 2002

P13-26

R. A. DeCarlo, P. M. Lin

(c) f (t) = 2r(t) 2r(t 1) 2u(t 1) + 2u(t 2) 2r(t 2) + 2r(t 3).

2 2e s 2e s 2e 2s 2e 2s 2e 3s
Hence F(s) = 2 2
+
2 + 2 .
s
s
s
s
s
s

26

2/23/02

page P14.1

R. A. DeCarlo, P. M. Lin

PROBLEM SOLUTIONS CHAPTER 14

S OLUTION 1 4 . 1 .
(a)
1
)
2
2
2
Cs = Cs(RLCs + R) = RLC(s + 1 LC) = R(s + 1 LC)
Z(s) =
1
Cs(RCs + LCs2 + 1) LC(s2 + R s + 1 ) s2 + R s + 1
R + Ls +
Cs
L
LC
L
LC
R(Ls +

(b)
1
)
LCs
Ls
Cs = R +
Z(s) = R +
=
R
+
1
C(LCs2 + 1)
LCs2 + 1
Ls +
Cs
1
1
2
RLCs2 + Ls + R RLC(s + RC s + LC )
=
=
1
LCs2 + 1
LC(s2 +
)
LC
(Ls)(

Hence,
1
1

R s2 +
s+

RC
LC
Z(s) =
1
s2 +
LC

SOLUTION 14.2.
(a)
80
)
Vs (s)
s(800 + 16s)
800s + 4000
s
Zin (s) =
=
=
= 2
2
80
I s (s) 10 + 0.2s +
s(0.2s + 10s + 80) s + 50s + 400
s
(b) If is (t) = 3e 20t u(t) A then
3
Is (s) =
s + 20
and
800s + 4000
3
2400s + 12,000
Vs(s) = Zin (s)Is (s) = 2

=
s + 50s + 400 s + 20 (s + 10)(s + 20)(s + 40)
K1
K2
K3
=
+
+
s + 10 s + 20 s + 40
Here
2400s + 12,000
+12,000 24,000
K1 =
=
= 40
(s + 20)(s + 40) s =10
(10)(30)
(10 + 0.2s)(

K2 =

2400s + 12,000
12,000 48,000
=
= 180
(s + 10)(s + 40) s= 20
(10)(20)

2/23/02

page P14.2

K3 =

and for t > 0,

2400s + 12,000
12,000 96,000
=
= 140
(s + 10)(s + 20) s =40
(30)(20)
180
40
140
Vs(s) =

s + 20 s + 10 s + 40

R. A. DeCarlo, P. M. Lin

v s (t) = 180e 20t 40e 10t 140e 40t V

SOLUTION 14.3.
(a)
Yp(s) = Cs +

1
1
= 2 10 3 s +
= 2 10 3 (s + 50)
R
0.10

Then
Z p(s) =
and
Zin (s) = 1.25s +
and

500
s + 50

500
1.25s2 + 62.5s + 500
=
s + 50
s + 50

I (s)
1
s + 50
0.80s + 40
Yin (s) = s
=
=
=
Vs (s) Zin (s) 1.25s2 + 62.5s + 500 s2 + 50s + 400

With v s (t) = 90e 40t u(t) , then

Vs(s) =
and

90
s + 40

90
0.80s + 40

s + 40 (s + 10)(s + 40)
72s + 3600
K1
C1
C2
=
=
+
+
2
s + 10 s + 40 (s + 40)2
(s + 10)(s + 40)

Is (s) = Vs (s)Yin (s) =

Here
K1 =
and with

72s + 3600
3600 720 2880
=
=
= 3.20
2
900
(s + 40) s =10
(30) 2

72s + 3600
720
p(40) =
= 24
s + 10
30
(s + 10)(72) (72s + 3600)
p(s) =
(s + 10) 2
30(72) (2880 + 3600) 2160 720
p(40) =
=
= 3.20
900
(30)2
p(s) =

Then
C2 =

p(40)
p(40)
= 24 , C1 =
= 3.20 ,
0!
1!

2/23/02

page P14.3

Is (s) =
and for t > 0

R. A. DeCarlo, P. M. Lin

3.20
3.20
24

+
s + 10 s + 40 (s + 40)2

is (t) = 3.20e 10t 3.20e 40t + 24te40t A

SOLUTION 14.4.
(a) Find Zin (s) vis Yin (s)
Yin (s) = Cs +

1
1 200Cs + 10LCs 2 + 10 + 20 + Ls
+ =
Ls + 20 10
10(Ls + 20)

10LCs2 + (200C + L)s + 30

=
10Ls + 200

and

Zin (s) =
With C = 103 F and L = 0.05 H
Zin (s) =
(b) If is (t) = 0.3u(t) A, then

1
10Ls + 200
=

2
Yin (s) 10LCs + (200C + L)s + 30

0.50s + 200
1000s + 4 10 5
=

0.0005s2 + 0.25s + 30 s2 + 500s + 60,000

Is (s) =

0.30
s

and

s + 400

s + 400

Vin (s) = Zin (s)Is (s) = 300 2

= 300

s(s + 500s + 60,000)

s(s + 2000)(s + 300)
K
K
K
= 300 1 + 2 + 3
s 200 300
It follows that
K1 =
K2 =

s + 400
200
1
=
=
s(s + 300) s =200 (200)(100)
100

K3 =

s + 400
100
1
=
=
s(s + 200) s =300 (300)(100) 300

and

Thus

s + 400
400
1
=
=
(s + 200)(s + 300) s =0 200(300) 150

2/23/02

page P14.4

R. A. DeCarlo, P. M. Lin

1
1
1
150

Vin (s) = 300

100 + 300
s + 200 s + 300
s

2
3
1
=
+
s s + 200 s + 300

and for t > 0

vin (t) = 2 3e 200t + e 300t V

SOLUTION 14.5.
Z(s) =
and the network is at rest
(a) If

s + 20
s + 40
20
s

vin (t) = 20u(t) Vin (s) =

then

V (s) 20 s + 40
s + 40
Iin (s) = in =
= 20

Z(s)
s s + 20
s(s + 20)
Using a partial function expansion
s + 40
K2
K
Iin (s) = 20
= 20 1 +

s + 20
s
s(s + 20)
in which case
K1 =
Thus

s + 40
40
s + 40
20
=
= 2, K 2 =
=
= 1
s + 20 s=0 20
s s =20 20
1
2
Iin (s) = 20
s s + 20

and
(b) Note that
Then

in which case
(c) Note that

iin (t) = 20(2 e 20t )u(t) A

20
vin (t) = 20e 40t Vin (s) =
s + 40
V (s)
20 s + 40
20
Iin (s) = in =
=

Z(s)
s + 40 s + 20
s + 20
iin (t) = 20e 20t u(t) A
vin (t) = 20e 20t Vin (s) =

Then

20
s + 20

2/23/02

page P14.5

Iin (s) =

R. A. DeCarlo, P. M. Lin

s + 40
Vin(s)
20 s + 40
=
=
20

Z (s)
s + 20 s + 20
(s + 20)2

Using a partial fraction expansion

s + 40
C1
C2
Iin (s) = 20
=
20
+

(s + 20)2
s + 20 (s + 20)2
Here

p(s) = s + 40
p'( s) = 1

p(20) = 20
p'(20) = 1

and
C1 =

p'(20)
p(20) 20
= 1, C2 =
=
= 20
1!
0!
1

in which case
1
20
Iin (s) = 20
+

s + 20 (s + 20)2
so that

iin (t) = 20e 20t + 400te 20t u(t) A

SOLUTION 14.6.
(a) Apply an arbitrary Iin (s) to the upper terminal of Fig. P14.6a. Assuming branch currents Ia(s) and
Ib (s), it follows by KCL that
Iin (s) = I a (s) + Ib (s) = 0.020sVa (s) + 0.005s Va(s) sVa (s)
= (0.020 + 0.005 0.015)sVa (s) = 0.010sVa (s)
Hence,
V (s)
1
100
Zin (s) = a
=
=

Iin (s) 0.010s

s
(b) Similarly apply an arbitrary Iin (s) to Fig P14.6b to obtain, in the s-domain, by KCL
1
V (s)
1 30

Iin (s) = 10sVin (s) + Vin (s) + 0.10sVin (s) in = 10s +

+
V (s)
50
4
50 4s in

2000s2 + 4s + 1500
=
Vin (s)
200s

Hence
I (s) 10s2 + 0.02s + 7.50
Yin (s) = in
=
Vin (s)
s
(c) Here we apply Vin (s) to the input terminals of figure P14.6c. By KCL
400
1
200

Vin (s) = 10Iin (s) + 0.2sIin (s) +
Iin (s) Iin (s) = 10 + 0.2s +
I

s
2
s in

in which case

2/23/02

page P14.6

R. A. DeCarlo, P. M. Lin

V (s) 0.2s2 + 10s + 200

Zin (s) = in
=

Iin (s)
s

S OLUTION 1 4 . 7 . Writing two loop equations we obtain:

Vin (s) = 100I1(s) + 200I 2 (s)

and

100

100I1(s) + 100 +
I (s) = 0
s 2
In matrix form (dropping the s-dependence)
200 I1 Vin
100
100 =

100 100 + s I2 0
Using Cramer's rule,
200
Vin
det
100
0 100 +
100s + 100
1

s
=
I1 =

V
200
100
s
100s + 100
in
100
200
det100 100 + 100

s
Hence

V
s 1
Zin (s) = in = 100
I1
s +1

SOLUTION 14.8.
Working in the s-domain, apply KVL to the left side of the circuit to obtain
100
s
Iin (s) +
I (s) + V2 (s)
s
100 in

Vin (s) =
Now apply KCL to the right side to obtain

Iin (s) =

s
100
V2 (s) +
V (s)
100
s 2

Thus
s2 + 104
Vin (s) =
Iin (s) + V2 (s)
100s
To find V2 (s) note that
Iin (s) =
implying that

s2 + 104
V (s)
100s 2

2/23/02

page P14.7

R. A. DeCarlo, P. M. Lin

100s
V2(s) = 2
I (s)
s + 104 in
Thus
s2 + 10 4
100s
(s2 + 104 )2 + 104 s2
Vin (s) =
+ 2
Iin (s) =
Iin (s)
s + 104
100s(s2 + 104 )
100s
implying that
V (s) s4 + 3 104 s2 + 108
Zin (s) = in
=

Iin (s)
100s(s2 + 104 )
and
Yin (s) =

1
100s(s2 + 104 )
= 4
S
Zin (s) s + 3 104 s2 + 108

SOLUTION 14.9.
Three mesh equations for the circuit
R + Z2 (s)
Z2 (s)
R I1 (s) Vin (s)

2R + Z2 (s)
R I 2 (s) = 0
Z2 (s)
R
R
2R + Z1 (s) I 3(s) 0
Solve for I1(s) via Cramers rule
Vin
Z2 (s)
R

det 0 2R + Z2 (s)
R

0
Vin 3R 2 + 2R( Z1 + Z2 ) + Z1Z2
R
2R + Z1(s)
I1 =
=
R + Z2 (s)
Z2 (s)
R

d(s)

det Z2 (s)
2R + Z2 (s)
R

R
R
2R + Z1 (s)
where
d(s) = (R + Z2 ) 3R 2 + 2R( Z1 + Z2 ) + Z1Z2 + Z2 Z2(2R + Z1 ) R 2 2R 2Z2 + 2R 3

) (

= R 3 + 2R 2 Z1 + 2R 2 Z2 + 3RZ1Z2
Under the condition that Z1 (s)Z2 (s) = R 2 , we have
Vin R 3 + 2R 2 Z1 + 2R 2 Z2 + 3RZ1Z2 4R 3 + 2R 2 (Z1 + Z2 )
Zin (s) =
=
=
=R
I1
3R 2 + 2R( Z1 + Z2 ) + Z1Z2
4R 2 + 2R (Z1 + Z2 )

SOLUTION 14.10.
1
(a) Yin (s) = Cs +
implies a parallel RC circuit with values R and C respectively.
R

2/23/02

(b) Yin (s) =

page P14.8

R. A. DeCarlo, P. M. Lin

1
1
= 2s + which is a parallel RC circuit of values 4 and 2 F respectively.
Zin (s)
4

1
1
=1 +
. Using the result of part (b), this circuit is a 1 resistor in series
2s + 0.25
Yb(s)
with the parallel RC of part (b).
(c) Zin (s) = 1 +

2s + 8
4
1
= 2+
= 2+
1 . Using the results of parts (b) and (c), this circuit is a 2
s+2
s+2
0.25s +
2
resistor in series with a parallel combination of a 0.25 F capacitor and a 2 resistor.
(d) Zin (s) =

s+3 s+6
2
2
1
1
+
= 2+
+
= 2+
+
. Using the above results, this
s +1 s + 4
s +1 s + 4
0.5s + 1 2 0.5s + 1 0.5
circuit is a 2 resistor in series with a parallel combination of a 0.5 F capacitor and a 2 resistor which is
in series with another parallel combination of a 0.5 F capacitor and a 0.5 resistor.
(e) Zin (s) =

SOLUTION 14.11.
(a) Clearly this is an inductor of value L in series with a resistor of value R.
(b) Inverting the admittance we have Zin (s) of the form of part (a). Hence the circuit is a 0.5 H inductor
in series with a 10 resistor.
1
1
(c) Yin (s) = 0.2 +
= 0.2 +
. Using the result of part (b), the circuit is 0.2 S resistor in
0.5s + 10
Zb(s)
parallel with a series connection of a 0.5 H inductor and a 10 resistor.
10s + 50
40
1
(d) Yin (s) =
= 10 +
= 10 +
. This is similar to part (c). Hence the circuit is a
s +1
s +1
0.025s + 0.025
10 S resistor in parallel with a series connection of a 25 mH inductor and a 0.025 resistor.
s+3 s+6
2
2
1
1
(e) Yin (s) =
+
= 2+
+
= 2+
+
. Hence, the circuit is a 2 S
s +1 s + 4
s +1 s + 4
0.5s + 0.5 0.5s + 2
resistor in parallel with the series connection of a 0.5 H inductor and a 0.5 resistor which in turn is in
parallel with a 0.5 H inductor and 2 resistor.

SOLTUION 14.12.
1
(a) Zin (s) = Ls +
represents a series connection of an inductance L and a capacitance C.
Cs
1
(b) Yin (s) = Cs +
represents a parallel connection of an inductance L and a capacitance C.
Ls
0.125s2 + 1
1
(c) Zin (s) =
= 0.5s +
which is a series connection of a 0.5 H inductor and a 0.25 F
0.25s
0.25s
capacitor.
0.125s2 + 1
1
(d) Yin (s) =
= 0.5s +
which is a parallel connection of a 0.5 F capacitor and 0.25 H
0.25s
0.25s
inductor.

2/23/02

(e) Zin (s) =

page P14.9

R. A. DeCarlo, P. M. Lin

s2 + 1 0.25s2 + 1
1 4
1
+
= 2s + + = 2s +
which is a 2 H inductor in series with a 0.2 F
s
0.25s
s s
0.2s

capacitor.
s2 + 1
0.25s
1
1
+
= s+ +
. This circuit is a 1 H inductor in series with a 1 F
2
s
s s+ 1
0.25s + 1
0.25s
capacitor which is in series with a parallel connection of a 1 F capacitor and a 0.25 H inductor.
s2 + 1
0.25s
1
1
(g) Yin (s) =
+
= s+ +
. This circuit is a 1 F capacitor in parallel with a 1 H
2
s
s s+ 1
0.25s + 1
0.25s
inductor which is in parallel with a series connection of a 1 H inductor and a 0.25 F capacitor.
(f) Zin (s) =

SOLUTION 14.13.
With L[v out (t)] = Vo(s) and L[v in (t)] = Vi (s) and vout (0 ) = 0,
sV0 (s) + 25V0 (s) +

100
10
V0 (s) = 5Vi (s) Vi (s)
s
s

which implies that

s2 + 25s + 100
5s 10

V0 (s) =
Vi (s)
s
s

The transfer function is

V (s)
5s 10
5s 10
H(s) = 0 = 2
=
Vi (s) s + 25s + 100 (s + 5)(s + 20)
5t
(a) If vin (t) = te u(t) V, then
1
Vi (s) =
(s + 5)2
and
5s 10
K1
C
C2
C3
Vout (s) =
=
+ 1 +
+
3
2
s + 20 s + 5 (s + 5)
(s + 20)(s + 5)
(s + 5)3
5s 10
110
110
22
K1 =
=
=
=
3
3
3375 675
(s + 5) s =20 (15)
5s 10
p(s) =
s + 20
(s + 20)(5) (5s = 10)
110
p(s) =
=
= 110(s + 20) 2
2
2
(s + 20)
(s + 20)
220
p(s) =
(s + 20)3
25 10 35
7
1575
p(5) =
=
= =
15
15
3
675
110
110 330
p(5) =
=
=
(15)2 225 675

2/23/02

page P14.10

p(5) =
Then
C3 =

R. A. DeCarlo, P. M. Lin

220
220
44
=
=
3
3375
675
(15)

p(5)
1575
p(5) 330
p (5) 1 44
22
=
, C2 =
=
, C1 =
=
=

01
675
1!
675
2!
2 675
675

and
Vout (s) =

1 22
22
330
1575

675 5 + 20 s + 5 (s + 5) 2 (s + 5) 3

This yields
vout (t) =
(b) If vin (t) = u(t) V,

1
1575 2 5t
22e 20t 22e 5t + 330te 5t
t e u(t) V

675
2

Vin (s) =

and

1
s

5s 10
K
K
K3
= 1+ 2 +
s(s + 5)(s + 20)
s
s + 5 s + 20
5s = 10
10
1
K1 =
=
=
(s + 5)(s + 20) s=0 100
10

Vout (s) =

K2 =

5s 10
25 10
35 7
=
=
=
s(s + 20) s= 5 (5)(+15) 75 15

and
K3 =
Thus

5s 10
100 10 110 11
=
=
=
s(s + 5) s =20 (20)(15)
300
30

11
1
7
vout (t) = e 5t e 20t u(t) V
30
10
15
By virtue of linearity and time invariance, if vin (t) = [u(t) u(t 0.5)] V ,
11
1
7
vout (t) = e 5t e 20t u(t)
30
10
15
11 20(t 0.5) 1
7
e 5(t 0.5)
e
u(t - 0.5) V
30
10
15

SOLUTION 14.14.
Here vin (t) = cos(t) u(t) V and iout (t) = 2sin(t)u(t) A, in which case

2/23/02

page P14.11

R. A. DeCarlo, P. M. Lin

2
2

I (s) s + 1 2
H(s) = out
= s =
Vin (s)
s
2
s +1

SOLUTION 14.15.
1
. Further,
(s + 1)2
vout (t) = (1+ t 0.5t 2 )e t u(t) + sin(t)u(t) cos(t)u(t) V in which case

Here vin (t) = tet u(t) V which implies that Vin (s) =

Vout (s) =

1
1
1
1
s
+
+
+ 2
2
2
3
s + 1 (s + 1)
(s + 1)
s +1 s +1

(a) Hence

Simplifying

V (s) 1
1
1
s 1
2
H(s) = out
=
+
+

(s + 1)
2
3
2
Vin (s) s + 1 (s + 1)
(s + 1)
s + 1
1
(s 1)(s + 1)2
= (s + 1)+ 1 +

(s + 1)
s2 + 1
s3 + 2s2 + 5s + 2
H(s) =
(s + 1)(s2 + 1)
1 1
s +1
+ 2 = 2 . Hence
s s
s
3
s + 2s2 + 5s + 2 5 2
4s
Vout (s) = H (s)Vin (s) =
= + 2 2
2
2
s s
s (s + 1)
s +1

(b) If vin (t) = (1+ t)u(t) V, then Vin (s) =

implying that

vout (t) = [5 + 2t 4cos(t)]u(t) V

SOLTUION 14.16.
(a) By a voltage divider (Fig. P14.16a)
Vout (s) =

Z4 (s)
V (s)
Z3(s) + Z4 (s) in

and
H(s) =
(b) In Fig. P14.16b,

Z4 (s)
Z3(s) + Z4 (s)

Yin (s) = Y1(s) + Y2 (s)

and
Vout (s) =

1
1
Iin (s) =
I (s)
Yin (s)
Y1 (s) + Y2 (s) in

2/23/02

page P14.12

R. A. DeCarlo, P. M. Lin

Hence
V (s)
1
H(s) = out
=
Iin (s) Y1 (s) + Y2 (s)
(c) By current division,

Iout (s) =

1
Z3 (s) + Z4 (s)
Y1 (s) + Y2 (s) +

1
Z3 (s) + Z4 (s)

Iin (s) =

1
[Y1(s) + Y2 (s)][Z3(s) + Z4 (s)] + 1 Iin (s)

Hence
Vout (s) = Z4 (s)Iout (s) =

Z4 (s)
[Y1(s) + Y2 (s)][Z3 (s) + Z4 (s)] + 1 Iin (s) .

and
V
Z4 (s)
H(s) = out =
Iin
[Y1(s) + Y2 (s)] [Z 3(s) + Z4 (s)] + 1

V
SOLUTION 14.17. With Vin (s) = Vi and Vout (s) = V0 , H(s) = 0 . By voltage division,
Vi
1
4
V
1
10
10 s + 103
H(s) = 0 =
=
=
1
Vi 103 +
0.1s + 2 s + 20
4
3
10 s + 10
400
20
20
(a) Vout (s) =
=

v out (t) = 20 20e 20t u(t) V. Plot omitted.

s(s + 20)
s s + 20
(b) If vin (t) = 40[ u(t) u(t 0.2)]V , then by linearity and time invariance

vout (t) = 20(1 e 20t )u(t) 20 1 e 20(t 0.2) u(t 0.2) V

(c) If vin (t) = 40[ u(t) + u(t 0.2)]V , then by linearity and time invariance

vout (t) = 20(1 e 20t )u(t) + 201 e 20(t 0.2) u(t 0.2) V
40
(d) If vin (t) = 40e 20t u(t) V , Vi (s) =
. Hence,
(s + 20)
400
vout (t) = 400te 20t u(t) V
2
(s + 20)
40
(e) If vin = 40te20t u(t) V , then Vi (s) =
. Hence,
(s + 20)2
V0(s) = H (s)Vi (s) =

V0(s) = H (s)Vi (s) =

400
(s + 20) 3

v out (t) = 200t 2e 20t u(t) V

2/23/02

page P14.13

R. A. DeCarlo, P. M. Lin

SOLUTION 14.18.
(a) By voltage division

2s + 1
2s + 1
2s + 1
s
s
Vout (s) =
2 2s + 1 Vin (s) = 2s + 2 + 2s + 1 Vin (s) = 4s + 3 Vin (s)
2+ +
s
s
s

Hence

V (s) 2s + 1
H(s) = out
=
Vin (s) 4s + 3

(b) With vin (t) = 8u(t) then

2s + 1 8 16s + 8
Vout (s) = H (s)Vin (s) =
=
4s + 3 s s(4s + 3)

Using MATLAB
n = [16 8]; d = [4 3 0];
[r,p,k] = residue(n,d)
r=
1.3333e+00
2.6667e+00
p=
-7.5000e-01
0
k = []
Then
(c) If vin (t) = 8sin(2 t)u(t) , then

8 4

vout (t) = + e 0.75t u(t) V

3 3
2s + 1 16
Vout (s) = H (s)Vin (s) =

4s + 3 s2 + 4

Using MATLAB,
ilaplace( (32*s+16)/((4*s+3)*(s^2+4)) )
ans =
-32/73*exp(-3/4*t)+32/73*cos(2*t)+280/73*sin(2*t)
Hence,

vout (t) = 0.43836e 0.75t + 0.43836cos(2t) + 3.8356sin(2t) u(t) V

(d) With vin (t) = 8sin(8 t)u(t)
128s + 64
2s + 1 64
Vout (s) = H (s)Vin (s) =
=

4s + 3 s2 + 64 (4 s + 3)(s2 + 64)
0.12391s + 31.907 0.12391
=

s + 0.75
s2 + 64
Using MATLAB
ilaplace((128*s + 64)/((4*s+3)*(s^2+64)))
ans =
-128/1033*exp(-3/4*t)+128/1033*cos(8*t)+4120/1033*sin(8*t)

2/23/02

page P14.14

128/1033
ans =
1.2391e-01
4120/1033
ans =
3.9884e+00

R. A. DeCarlo, P. M. Lin

vout (t) = 0.12391cos(8t) + 3.9884sin(8t) 0.12391e 0.75t u(t) V

SOLUTION 14.19.
V (s)
With a source transformation Iin (s) = in .
R
(a) By current division,
Cs
Vin (s)
LCs 2
Vin (s)
s2
Vin (s)
IC (s) = 1
=
=
1
L
1
1
R
R
+ Cs +
LCs2 + s + 1 R
s2 +
s+
R
Ls
R
RC
LC
Here
IC (s) 1
s2
H(s) =
=
Vin (s) R s2 + 1 s + 1
RC
LC
2
(b) With R = , C = 0.5F and L = 1H ,
3

3
s2
H(s) = 2

2 s + 3s + 2
1
If vin (t) = e t u(t) V, then Vin (s) =
. Hence
s +1
IC (s) = H (s)Iin (s) =

1
3
s2
3s2
K1
C1
C2
=
=
+
+

2 s2 + 3s + 2 s + 1 2(s + 1)2 (s + 2) s + 2 s + 1 (s + 1)2

Using MATLAB,
n = [3 0 0]; d = conv([2 4],[1 2 1]);
[r,p,k] = residue(n,d)
r=
6.0000e+00
-4.5000e+00
1.5000e+00
p=
-2.0000e+00
-1.0000e+00
-1.0000e+00
k=
[]

2/23/02

page P14.15

R. A. DeCarlo, P. M. Lin

6
4.5
1.5

+
s + 2 s + 1 (s + 1)2

IC (s) =
and

9
3

iC (t) = 6e 2t e t + tet u(t) A

2
2

SOLUTION 14.20.
(a) Make a source transformation:
Vin (s) =

1
250
Iin (s) =
I (s)
Cs
s in

By voltage division

10
2500

250

Vout (s) =
I (s)
s Iin (s) = 1 2
250 250 1
in

+
+ s + 10
s s + 10s + 500
s
s
20
20
50,000
=
Iin (s)
s s2 + 200s + 10,000

and
V (s)
50,000
H(s) = out
=
2
Iin (s)
s(s + 200s + 10,000)
(b) If iin (t) = (t) implies Iin (s) = 1. Using MATLAB
n = 50e3; d = [1 200 10e3 0];
[r,p,k] = residue(n,d)
r=
-5
-500
5
p=
-100
-100
0
k=
[]
Hence
50,000
5
5
500
Vout (s) =
=

2
s s + 100 (s + 100) 2
s(s + 100)
and
vout (t) = 5 5e 100t 500te 100t u(t) V
This is the impulse response
0.1
(c) If Lin (t) = 100u(t) mA so that Iin (s) =
. Therefore
s
5000
Vout (s) = 2
s (s + 100) 2
In MATLAB,

2/23/02

page P14.16

R. A. DeCarlo, P. M. Lin

n = 5000; d = conv([1 0 0],[1 200 1e4])

d=
1
200
10000
0
0
[r,p,k] = residue(n,d)
r=
1.0000e-02
5.0000e-01
-1.0000e-02
5.0000e-01
p=
-100
-100
0
0
k=
[]
Hence
0.01 0.5
0.01
0.5
Vout (s) =
+ 2 +
+
s
s + 100 (s + 100) 2
s
and

vout (t) = 0.01e 100t + 0.5te 100t 0.01 + 0.5t u(t)

(d) By superposition and time invariance, if
iin (t) = 100[ u(t) + u(t 1)] mA
then the result of part (c) can be adjusted to

vout (t) = 0.01e 100t + 0.5te 100t 0.01 + 0.5t u(t)

0.01e 100(t 1) + 0.5te100(t 1) 0.01 + 0.5(t 1) u(t 1) V

SOLUTION 14.21. For this problem change the 20 mH inductor to one of 0.3 H.
(a)
( s + 200) (s + 400)
1
1
Yin = 1 +
+
=
15 0.3s + 90 0.1s + 10 15( s+ 100) (s + 300)
and
( s + 100) (s + 300)
H(s) = Iout = 1/15 =
Iin
Yin
( s+ 200) (s + 400)
(b) If iin(t) = (t), then Iin(s) = 1 and
Iout (s) = H(s) =
Hence

( s + 100) (s + 300)
= 1 - 50
- 150
( s+ 200) (s + 400)
s + 200 s + 400

iout(t) = (t) + (- 50 e-200t - 150 e-400 t) u(t) A

(c) We first find the response to iin(t) = 16u(t) mA. Here Iin(s) = 0.016/s and

2/23/02

page P14.17

Iout (s) = H(s)Iin (s) =

R. A. DeCarlo, P. M. Lin

0.016( s + 100) (s + 300)

0.004
0.006
= 0.006
s + s + 200 + s + 400
s( s+ 200) (s + 400)

Hence

iout(t) = (6 + 4 e-200t + 6 e-400 t)u(t) mA

By linearity and time invariance, the response to iin(t) = 16[ut) u(t 0.01)] mA is

iout(t) = (6 + 4 e-200t + 6 e-400 t)u(t) - (6 + 4 e-200(t - 0.01) + 6 e-400 (t - 0.01) )u(t - 0.01) mA
A plot of iout(t) using MATLAB is given below.
t= 0: 0.0005: 0.05;
f1= (6 + 4*exp(-200*t) + 6*exp(-400*t)).*u(t);
f2= (6 + 4*exp(-200*(t-0.01)) + 6*exp(-400*(t-0.01))).*u(t-0.01);
iout= f1 - f2;
plot(t, iout)
grid
ylabel('iout in mA')
xlabel(' time in second')
20

iout in mA

15
10
5
0
-5
-10

0.01

0.02
0.03
time in second

0.04

REMARK : Notice that the resistor current is not continuous.

S OLUTION 1 4 . 2 2 . (a) First observe that the admittance of a parallel LC is

YLC (s) = Cs +
Using voltage division,

1
Ls

0.05

2/23/02

page P14.18

C1s +

1
Ls

C1s +

R. A. DeCarlo, P. M. Lin

1
Ls

LC1s2 + 1
Vout (s) =
1
1 Vin (s) =
2 Vin (s) = L(C + C )s2 + 2 Vin (s)
1
2
C1s +
+ C2 s +
C1 + C2 )s +
(
Ls
Ls
Ls
Finally
1
Vout (s)
C1
LC1
s2 + 4 10 6
H(s) =
=
= 0.2 2
2
Vin (s) (C1 + C2 ) s2 +
s + 1.6 106
L(C1 + C2 )
(b) Using MATLAB,
syms s t
ilaplace(0.2*(s^2+4e6)/(s^2+1.6e6))
ans =
1/5*Dirac(t)+120*10^(1/2)*sin(400*10^(1/2)*t)
120*10^(1/2)
ans = 3.7947e+02
s2 +

s2 + 4 106
h(t) = 0.2L 2
= 0.2 (t) + 379.47sin(1264.9t)u(t) V
s + 1.6 10 6

SOLUTION 14.23.

Then Z1(s) =

Y1(s) = C1s +

1 R1C1s + 1
=
R1
R1

Y2(s) = C2 s +

1
R C s +1
= 2 2
R2
R2

R1
R2
and Z2(s) =
. By voltage division,
R1C1s + 1
R2C2s + 1

R2
R2

V (s)
R2C2s + 1 in
R2C2 S + 1
Vout (s) =
Vin (s) = R R C s + R + R R C s + R
R1
R2
1 2 2
1
1 2 1
2
+
R1C1s + 1 R2C1s + 1
(R1C1s + 1)(R2C2 s + 1)
R2 (R1C1s + 1)
=
V (s)
(C1 + C2 )R1R2s + R1 + R2 in

Thus the transfer function is:

V (s)
R2 (R1C1s + 1)
H(s) = out
=
Vin (s) (C1 + C2 )R1R2s + R1 + R2
(b) If C1 = 0.5 F , C2 = 1.0 F and vin (t) = 10 u(t) V, then

2/23/02

page P14.19

R. A. DeCarlo, P. M. Lin

0.5R1R2s + R2 10
Vout (s) = H (s)Vin (s) =
1.5R1R2s + R1 + R2 s
R
Moreover, with 1 = 4 so that R1 = 4R2
R2

1
1

20R22 (s +
)
s+

2R22 s + R2 10
2R2
10
2R2
Vout (s) = 2
=
=

5 3 s(s + 5 )
6R2 s + 5R2 s
6R22 s(s +
)

6R2
6R2

The partial fraction expansion is

s+

10
2R2 10 K1
K2
Vout (s) =
=
+

5
3
3 s
5
s+
s s +

6R2
6R2

Observe that
K2
5
s+
6R2

L1

K 2e

5
t
6R2

and that it is required that

Thus

5
5
=
6R2
3

R2 = 0.5 , R1 = 4R1 = 2

and

10
s + 1 10 0.6
0.4 2 4 / 3
Vout (s) =
=
+
5 = s +
5
5 3 s
3
s
+
s
+
s
s
+

3
3
3
Thus,
5

t
4
vout (t) = 2 + e 3 u(t)

(c) If R1C1 = R2C2 = , then the transfer function is

H(s) =
=

R2 (R1C1s + 1)
R2 s + R2
=
R1R2C1s + R1R2C2 s + R1 + R2 [ R2 (R1C1) + R1 (R2C2 )]s + R1 + R2
R2 ( s + 1)
R2
=
(R1R2 )( s + 1) R2 + R1

The zero-state response is

2/23/02

page P14.20

Vout (s) =

R. A. DeCarlo, P. M. Lin

R2
10u(t)
R2 + R1

(d) Using H(s) from part (c) with the requirement that R1C1 = R2C2 , then
H(s) =
With R2 = 106 , then 10R2 = R1 + R2

R2
1
=
R1 + R2 10

R1 = 9R2 = 9 M . Since C2 = 5 10 12 F, then

RC
C1 = 2 2 = 0.556 10 12 F
R1

SOLUTION 14.24.
V (s)
V (s)
(a) H(s) = out . Here Ib(s) = in . The parallel admittance at the right is
Vin (s)
2000
YR (s) = Cs +

1 RCs + 1
=
R
R

so that
ZR (s) =
Then

R
RCs + 1

R
Vout (s) = ZR (s) Ib (s) =
2000(RCs + 1)
(b) With V1(s) = Vin (s) Vout (s) , then

1
1
Cs(Vout (s) Vin (s)) + Vout (s) + [Vout (s) 3Vin (s) + 3Vout (s)] = 0
s
2

Hence
3
1

Cs
+
V
(s)
=
Cs
+
+ 2 V (s)
in

2
s out
and

V (s) s(Cs + 1.5)

s(s + 0.75)
H(s) = out
= 2
= 2
Vin (s) Cs + 2s + 1 s + s + 0.5

(c) Transform the current source iin (t) into a voltage source. In the s-domain with Iin (s) = 2Vin (s)
1
1
Here ZC (s) =
=
which implies YC (s) = 2s . A single node equation yields
Cs 2s
2
2
2
2 2

Vout (s) Vin (s)] Vout (s) + 2sVout (s) = Vin (s) + + 2s Vout (s) = 0
[
3
3
3
3 3

2/23/02

page P14.21

R. A. DeCarlo, P. M. Lin

2
V (s)
Vin (s) = 2sVout (s) implies Vout (s) = in
3
3s
I (s)
But Vin (s) = in
in which case
2
I (s)
Vout (s) = in
6s
and
V (s) 1
H(s) = out
=
Iin (s)
6s

SOLUTION 14.25. Here in the s-domain

V (s) VC (s)
Iin (s) = in
0.5s + 10

and with a node at VC (s)

Iin (s) +
Hence,

sVC (s) 1
sVC (s) 1
V (s) VC (s)
+ Iin (s) = 0 implies
= Iin (s) = in
500
2
50
2
s + 20
s2 + 20s + 500

VC (s) = Vin (s)

500

and the transfer function is

V (s)
500
H(s) = C
= 2
Vin (s) s + 20s + 500
4 5
With vin (s) = 4 5 u(t) V which implies Vin (s) =
, and
s
2000 5
K1
K
K*
VC (s) =
=
+
+
s(s + 10 j20)(s + 10 + j20)
s s + 10 j20 s + 10 + j20
where K * designates the complex conjugate of K
2000 5
2000 5
K1 = 2
=
= 8.944
500
s + 20s + 500 s =0
K=

2000 5
2000 5
=
= 4.472 + j2236 = Jeij
(10 + j20)( j40) (800 + j400)
= 153.44 o . Then with
K * = 4.472 j2.236 = 5e j
8.944
A + jB
A jB
VC (s) =
+
+
s
s + 10 + j20 s + 10 j20
=

where

Here

2000 5
s(s + 10 + j20) s= (10 j 20)

2/23/02

page P14.22

A = 4.472, B = 2.236,

R. A. DeCarlo, P. M. Lin

A2 + B2 = 5

and
arc tan

B
2.236
1
= arc tan
= arc tan = 153.44 o
A
4.472
2

With the help of Table 13.1.

vC (t) = 8.944 + 10e 10t cos(20t + 153.44 o ) u(t) V

SOLUTION 14.26. In the s-domain we first find Vx (s) in terms of Vin (s) via voltage division:
Z p (s)
Vx (s) =
V (s)
40 + Z p (s) in
where
(0.40s)(40)
40s
Z p(s) =
=
0.40s + 40 s + 100
Hence
40s
40s
0.5s
Vx (s) = s + 100
40s Vin (s) = 80s + 4000 = s + 50 Vin (s)
40 +
s + 100
and
V (s) 2.5
IL (s) = x =
V (s)
0.4s
s x
Then from the right hand side by another voltage division
10
10s
0.25s
Vout (s) = 1000
0.25V x (s) =
0.25V x (s) =
Vx (s)
10s
+
1000
s
+
100
+ 10
s
20
20
(a) If vin (s) = 20(1 e 40t )u(t), then Vin (s) =

. Hence
s s + 40
20 s 10
10
0.5s 20
Vx (s) =

s + 50 s s + 40 s + 50 s s + 40
and
IL (s) =
Hence

2.5
1 0.05 0.25
0.2
2.5 1
V x (s) =

=
+
+

s
s + 50 s s + 40
s
s + 40 s + 50

iL (t) = 0.05 + 0.2e 50t 0.25e 40t u(t) A

20
20

. From, part (a), it was found that

s s + 40
0.25s
0.25s
0.5s
s
s
2.5
2.5
Vout (s) =
V x (s) =

Vin (s) =

s + 100
s + 100 s + 50
s + 100 s + 50 s
s + 40

(b) Vin (s) =

2/23/02

page P14.23

R. A. DeCarlo, P. M. Lin

100s
20/ 3
10
10 /3
=
+
+
(s + 40)( s + 50)(s + 100) (s + 40) ( s + 50) ( s + 100)

Thus,
10
20

Vout (t) = 10e 50t e 100t e 40t u(t) V

3
3

SOLUTION 14.27. In both parts (a) and (b), the op-ampis ideal. It will not draw current and the
virtual groundprincipal requires that
v + = v = 0
(a) For a note at the inverting terminal with mode voltage v1(t) = 0 , KCL gives in the s-domain

Vin (s)
V (s)
V (s)
1
= CsVout (s) out
implies in
= Cs + Vout (s)
R1
R2
R1

R2
in which case
R Cs + 1
Vin (s)
= 2
V (s)
R1
R2 out
Then,

V (s)
R
1
1
1
H(s) = out
= 2
=

Vin (s)
R1 R2Cs + 1
R1C s + 1

R2C
To make
H(s) =
make
1
= 20 and
R1C
If

20
s+4
1
1
=
R2C 4

C = 1 F = 106 F
1
1
106
= 4 or R2 =
=
= 250 k
R2C
4C
4

and
1
2
106
= 20 or R1 =
=
= 50 k
R1C
20C 20
(b) With Fig. 14.27b in the s-domain and v1(t) at the inverting terminal, KCL gives
V (s)
V (s)
C1sVin (s) + in = C2sVout (s) out
R1
R2
R1C1s + 1
R2C2s + 1
R
Vin (s) =
Vout (s)
R2
1

2/23/02

page P14.24

R. A. DeCarlo, P. M. Lin

Then
V (s)
R R C s + 1
H(s) = out
= 2 1 1
Vin (s)
R1 R2C2 s + 1
(c) If
then

1
1

s+
s+

RRC
R1C1
C
R1C1
= 1 2 1
= 1
1
R1RC2 s +
C2 s + 1

R2C2
R2C2

H(s) = 5, C2 = 1 F and R2 = 1M
C1
= 5 and C1 = 5 F
C2

The bracketed term must cancel and with

Then with C1 = 5 F

R2C2 = 106 (106 ) = 1

R1C1 = 1
1
1
R1 =
=
= 200k
C1 5 106

(d) Using H(s) in part (b)

1

s+
C
R1C1
H(s) = 1

C2 s + 1

R2C2
to obtain
H(s) = 5
with C2 = 1 F

s +1
s+2

C1 = 5C2 = 5 10 6 F (5 F)
1
1 106
= 1 or R1 =
=
= 200k
R1C1
C1
5
1
1
106
= 2 or R2 =
=
= 500k
R2C2
2C2
2

SOLUTION 14.28. Here, the op-amp will not draw current at the non-inverting terminal and the
principal of the virtual ground demand that
v1 = v 2 = Vin (s)
For Fig. 14.28 in the s-domain with a node V1(s)
taken at the inverting terminal
Vin (s) Vin (s) Vout (s)
+
=0
R1
Z p (s)
Here

2/23/02

page P14.25

R. A. DeCarlo, P. M. Lin

R2
Cs

R2
1
Z p(s) = 1
=
=
1
+ R2 R2Cs + 1 C(s +
)
Cs
R2C
and
1
1
1
R + C(s + R C ) Vin (s) = C(s + R C )Vout (s)
1
2
2

1
1
+ 1Vin (s) = R1C s +
V (s)
R1C s +

R2C

R2C out

and
V (s)
H(s) = out
=
Vin (s)
Here

1
) +1 s +
R2C
=
1
R1C(s +
)
R2C

R1C(s +

1
1
+
R2C R1C
1
s+
R1C

1
1
+
=4
R2C R1C

and
1
=2
R1C
1 10 6
If C = 1 F then R1 =
=
= 500 k
2C
2
1
and
= 4 2 = 2 implies R2 = 500 k .
R2C

SOLUTION 14.29. For the non-inverting configuration in the s-domain, each of the two op-amps in
cascade have a transfer function
Z f (s)
H(s) =
Zin (s)
Then for the two op-amps
Z f 1(s) Z f 2 (s) Z f ,1(s)Z f ,2 (s)
H0 (s) =
=

Zin1 (s) Zin,2 Zin,1(s)Zin,2 (s)
For Fig. P14.29a in the s-domain
Zin,1 = 25k, Zin,2 = 50k , Z f ,1 =

Hence for Fig. P14.29a

1
1

C s+

RC

250,000
, and Z f ,2 =
s+5

1
1

C s+

RC

250,000
s + 2.5

2/23/02

page P14.26

R. A. DeCarlo, P. M. Lin

250,000 250,000
50

Ha (s) = s + 2.5 s + 5.0 =

25,000 50,000 (s + 2.5)(s + 5)

If vin (t) = u(t), then Vin (s) =

1
and
s
Vout (s) =

Hence,

50
4
8
4
=
+
s( s + 2.5)(s + 5) s s + 2.5 s + 5

vout (t) = 4 8e 2.5t + 4e 5t u(t) V

(b) For Fig. P14.29b in the s-domain, Zin,1 , Z f 1 , and Z f 2 are in part (a). However,
1 250,000
Zin,2 =
=
Cs
s
Thus
250,000 250,000
10s

s + 5 10 s
Hb(s) = s + 2.5 250,000
=
=

5
s + 2.5 s + 5
25,000

s + (s + 5)

s
2
1
With Vin (s) = ,
s

10s
1
20
4
4
Vout (s) =
=
=

(s + 2.5)(s + 5) s (s + 2.5)(s + 5) s + 2.5 s + 5

and
vout (t) = 4e 2.5t 4e 5t u(t) V

SOLUTION 14.30.
(a) -(b). The subcircuit is an integrator, with

Vout(t) = - 1
s
V1 (s)
(c) This subcircuit is again an integrator, with

V1(t) = - 1
s
V2 (s)
(d) Applying KCL to the inverting input terminal of the top left op amp, we have
or

G2Vout(s) + V3(s) + V 2(s) = 0

V2(s) = - G 2Vout(s) - V3(s)

2/23/02

page P14.27

R. A. DeCarlo, P. M. Lin

(e) Applying KCL to the inverting input terminal of the bottom op amp, we have
or

G3V1(s) + G1 Vin(s) + V 3(s) = 0

V3(s) = - G 3V1(s) - G1 Vin(s)

(f) The results of parts (b), (c) and (d) do not involve Vin. Therefore, , we can solve for V1, V2 and V3
in terms of Vout from these threes equations:

V1(s) = - sV out(s)
and

V2(s) = - sV 1(s) = s 2Vout(s)

V3(s) = - V 2(s) - G2Vout(s) = - s2 + G 2 Vout(s)

Substituting these relationships into the result of part (e), we obtain

- sG3Vout(s) + G1 Vin(s) - s2 + G 2 Vout(s) = 0

Therefore

G1
H(s) = Vout(s) =
2
Vin (s)
s + G 3s + G2
SOLUTION 14.31. Use the parallel equivalent circuit model for the capacitor with the standard
directions for voltage and current as given in figure 14.16. For the single node with vC (0 ) = 20 V,
VC (s) VC (s)
1
1
1
s

10 + 40 + 10 10 vC (0 ) = 0 implies 10 + 50 VC (s) = 10 (20) = 2

s
Equivalently, (50s + 10)VC (s) = (s + 0.2)VC (s) = 20 or VC (s) =
vC (t) = 20e 0.2t u(t) V

20
. Therefore,
s + 0.2

SOLUTION 14.32. Using the equivalent model for the inductor in figure 14.19, we can compute the
total admittance as
5
1
1
5 1 2s + 40 s + 20
Y(s) =
+
+ =
+ =
=
2s 40 10 2s 8
16s
8s
Using current division,

2/23/02

page P14.28

R. A. DeCarlo, P. M. Lin

0.1
i (0)
0.8iL (0 )
1.6
Iout (s) = s + 20 L
=
=
s
s + 20
s + 20
8s
Thus

iout (t) = 1.6e 20t u(t) A

SOLUTION 14.33. Using the equivalent model for the inductor in figure 14.19 and for the capacitor
using figure 14.16, we may combine the current sources to form an equivalent source (with C = 0.1 F) to
obtain
1
i (0 )
1
Ieq (s) = vC (0 ) L
= 0.2
10
s
s
Note that
1
2
s +
1
LCs 2 + 1

LC
Y(s) = Cs +
=
= C

LS
Ls
s

1
With C = 0.1 F and L = 0.4 H,
= 25 and
LC
1 s
10s
Z(s) = 2
= 2

C s + 25
s + 25
Thus
10s s 5 2(s 5)
2s
10
VC (s) = Z(s)Ieq (s) = 2
= 2
= 2
2

s + 25 5s
s + 25 s + 25 s + 25
and
vC (t) = (2cos(5t) 2sin(5 t)) u(t) V

SOLUTION 14.34. Consider the equivalent circuit below:

Writing a single node equation we have,

2/23/02

page P14.29

0 = 0.5Vout 0.2VR1 +

= 0.5Vout + 5 0.2

1
(V + 0.8)
5 + 0.4s out
1
1
Vout + 0.8) +
(
(V + 0.8)
5 + 0.4s
5 + 0.4s out

= 0.5Vout +
Therefore

2
(V + 0.8)
5 + 0.4s out

Vout =
and

R. A. DeCarlo, P. M. Lin

8
s + 22.5

vout (t) = 8e 22.5t u(t) V

SOLUTION 14.35. Redraw the circuit in the s-domain and use an equivalent circuit for the capacitor
(figure 14.16) that accounts for the initial condition. By KCL

With R = 50 and C = 0.02 F ,

VC (s)
+ CsVC (s) = Cv(0 ) + Iin (s)
R
VC (s)
+ 0.02sVC (s) = 0.02vC (0 ) + Iin (s)
50

or
which is equivalent to

VC (s) + sVC (s) = vC (0 ) + 50Iin (s)

(s + 1)VC (s) = vC (0 ) + 50Iin (s)

(a) With vC (0 ) = 8 V and iin (t) = 40 (t) mA so that Iin (s) = 0.04 ,
(s + 1)VC (s) = 8 + 2 implies VC (s) =
and

10
s +1

vC (t) = 10e 10t u(t) V

(b) With vC (0 ) = 1 V and iin (t) = 200e t u(t) mA we have that Iin (s) =

0.2
. Thus
s +1

v C (0 )
50
1
10
VC (s) =
+
Iin (s) =
+
(s + 1) (s + 1)
(s + 1) (s + 1)2
and

SOLUTION 14.36.

vC (t) = e t + 10tet u(t) V

2/23/02

page P14.30

R. A. DeCarlo, P. M. Lin

(a) By current division

H(s) = IL(s) = 1
Iin (s) Ls + 1
(b) The given data in Laplace transforms are:

Iin(s) = I0 and IL (s) = 15 - 3 + 3 = 75

s2
s2 s s+ 5 s2 s+ 5
Under the assumption of zero initial inductor current,

I0
I0/L
75
IL (s) = H(s) Iin (s) = H(s) = IL(s) =
=
=
2
2
2
Iin (s)
Ls + 1 s
s + 1/L s
s s+ 5
Equating coefficients, we obtain the answers
L = 1/5 = 0.2 H and I 0 = 75L =15 A
(c) The s-domain equivalent is shown below.

Applying KVL to the right mesh, we have

LsIL(s) - L iL(0-) + 1 [IL(s) - 1] = 0

Solving for IL(s) from this equation, and equating it to 10/(s+ 5), we have
IL(s) = L i L(0 ) + 1 = i L(0 ) + 1/L = 10
Ls+1
s + 1/L
s+5

from which L= 0.2 H and iL (0-) = 5 A.

SOLUTION 14.37.
(a) By inspection,

2/23/02

page P14.31

R. A. DeCarlo, P. M. Lin

1
H(s) = IL(s) =
= 0.5
Vin (s) 2s + 200 s + 100
Given vin(t) = 2u(t) V, then Vin(s) = 2/s , and

I L(s) =

1
= 0.01 - 0.01
s
s s + 100
s + 100

in which case

i L(t) = 0.01 ( 1 - e-100t) A

Plots are omitted.

(b) By linearity and time invariance,

i L(t) = 0.01 ( 1 - e-100t)u(t) - 0.01( 1 - e-100(t- 0.05))u(t - 0.05) A

(c) Correction: (a) should be (c). With nonzero initial inducto current, the s-domain equivalent
becomes:

Given iL(0-) = 0.01 A and vin(t) = 2e-200tu(t) V, then Vin(s) = 2/(s + 200) and

2
+ 0.02
s
+
200
0.01s + 3
I L(s) =
=
2s + 200
( s + 100)( s + 200)
=
in which case

0.02 - 0.01
s + 100 s + 200

iL( t) = 0.02e-100t - 0.01e-200t u(t) A

(d) Correction . (b) should be (d).
Given iL(0-) = 0.01 A and vin(t)= 2cos(200t)u(t) V, then
Vin (s) =
and

2s
s2 + 40000

2/23/02

page P14.32

I L(s) =

s2

R. A. DeCarlo, P. M. Lin

2s
+ 0.02
+ 40000
0.01s2 + s + 400
=
2s + 200
( s + 100)( s2 + 40000)

We use MATLAB to do the partial fraction expansion.

n= [ 0.01 1 400];
d= conv([ 1 100], [ 1 0 40000]);
[ r p k ] = residue (n,d)
r=
0.0010 - 0.0020i
0.0010 + 0.0020i
0.0080
p=
1.0e+02 *
0.0000 + 2.0000i
0.0000 - 2.0000i
-1.0000
From the above MATLAB output,

I L(s) =

0.008 + 0.001-j0.002 + 0.001+j0.002 = 0.008 + 0.002s + 0.8

s + 100
s - j200
s + j200
s + 100
s2 + 2002

From table 13.1, item 18,

iL( t) = 0.008e-100t + 0.002 cos(200t) + 0.004 sin(200t) u(t) A

2/23/02

page P14.1

R. A. DeCarlo, P. M. Lin

SOLUTION 14.38. In the s-domain, we break the response up into the part due to the initial condition
and the part due to the source with the initial condition set to zero. The transfer function with the initial
condition set to zero is
V (s)
1 Cs
1 RC
0.25
H(s) = C
=
=
=
Vin (s) R + 1 Cs s + 1 RC s + 0.25
Using the parallel equivalent circuit for the charged capacitor while setting the source voltage to zero,
the capacitor voltage due only the initial condition is:
vC (0 )
VC,IC (s) = 1
CvC (0 ) =
s + 0.25
+ Cs
R
1

Hence,

0.25
vC (0 )
VC (s) =
V (s) +
s + 0.25 in
s + 0.25

and

V (s) VC (s)
0.25
0.05v C (0 )
0.05s
0.05vC (0 )

IC (s) = in
= 0.05 1
V
(s)

=
V
(s)

20
s + 0.25
s + 0.25 in
s + 0.25
s + 0.25 in
for all inputs and initial conditions.
20
(a)
If vin (t) = 20u(t) and vC (0 ) = 10 V, then Vin (s) =
and
s
5
10
20
10
VC (s) =
+
=

v C (t) = 20 10e 0.25t u(t) V

s(s + 0.25) s + 0.25
s s + 0.25
and
1
0.5
0.5
IC (s) =

=
iC (t) = 0.5e 0.25t u(t) A
s + 0.25 s + 0.25 s + 0.25

(b) If vin (t) = 5e 0.25t u(t) V, then Vin (s) =

VC (s) =

5
. Hence,
s + 0.25

1.25
10
+
2
s + 0.25
(s + 0.25)

and
IC (s) =

v C (t) = (10 + 1.25t)e 0.25t u(t) V

0.25s
0.5
0.25
0.0625

(s + 0.25) 2 s + 0.25 s + 0.25 (s + 0.25) 2

Hence
iC (t) = (0.25 + 0.0625t)e 0.25t u(t) A

2/23/02

page P14.2

R. A. DeCarlo, P. M. Lin

10
9

Input and Capacitor voltages, V

8
7
6
5
4
3
TextEnd
2
1
0

10
Time in s

15

20

SOLUTION 14.39. The figure which accounts for the initial conditions is given below.

(a) For the zero-input response, the above circuit reduces to a parallel RLC driven by two current
sources.
Hence VC(s) equals the total current divided by the total admittance, i.e.,

2/23/02

page P14.3

R. A. DeCarlo, P. M. Lin

i (0 )
i (0 )
Cv C (0 ) + L
svC (0 ) + L
20s + 10
26.6
6.6
s =
C =
VC (s) =
=

2
4
1
1
1
1
s + 200 s + 50
s + 250s + 10
Cs + +
s2 +
s+
R Ls
RC
LC
Hence
vC(t) = [26.6e

200t

50t

6.6e

]u(t) V

(b) For the zero-state response, the current sources disappear. Executing a source transformation on
the remaining voltage source, we obtain a current, I(s) = Vin(s)/(Ls), driving a parallel RLC circuit.
Hence, the zero input response is
Vin (s)
Vin (s)
20000
2 0.6667 2.6667
1
Ls
VC (s) =
=
=
=
+

3 + 250s 2 + 104 s
1 1 LC 2
1
1
s
s
+
200
s + 50
s
Cs + +
s +
s+
R Ls
RC
LC
Hence

200t

50t

vC(t) = [2 + 0.6667e
2.6667e
]u(t) V
(c) By superposition, the complete response is the sum of the answers to (a) and (b). Hence
vC(t) = [2 + 27.267e
(d) By linearity and time-invariance,
vC(t) = [2 + 0.6667e

200t

50t

2.6667e

50t

9.2667e

]u(t) V

]u(t)

200(t0.01)

+ [4 + 1.3334e

200t

50(t0.01)

5.3334e

]u(t 0.01) V

2/23/02

page P14.4

R. A. DeCarlo, P. M. Lin

SOLUTION 14.40. f (t) = Lin (t) = e 2t u(t)A [part (c)]

L (0 )
for the zero-input response, f (t) = Lin (t) = 2
[part (d)]
s
for the zero-input response, f (t) = Lin (t) = CvC (0)A [part (e)]

1
s2 + 2s + 1 (s + 1)2
(a) Yin (s) = 1 + + s + 1=
=
s
s
s
Y1 (s)
s +1
s(s + 1)
s
(b) Iout (s) =
Iin (s) = 2
=
=
Iin (s)
Yin (s)
s + 2s + 1 (s + 1)2 s + 1
s
and
I (s)
s
H(s) = out
=
Iin (s) s + 1
1
(c) If iin (t) = e 2t u(t) A, then Iin (s) =
, then
s+2
s
1
2
Iout (s) = H (s)Iin (s) =
=
+
(s + 1)(s + 2) s + 1 s + 2
which implies that the zero-state response is

iout (t) = 2e 2t e t u(t) A

(d) If iL (0 ) = 2 A, vC (0 ) = 0, and iin (t) = 0. Using the parallel equivalent circuit for the inductor,
figure 14.19, we have
i(0 ) s 2
2
Iout (s) = H (s)
=
=
iout (t) = 2e t u(t) A

s
s +1
s
s +1
(e) Use the parallel equivalent circuit for the capacitor, figure 14.16, to obtain by current division,

Iout (s) = 1
s

1
+1
s

s +1
4
[CvC (0 )] = 4 2
=
iout (t) = 4e t u(t) A
s
+
1
s + 2s + 1
+1 + s +1

(f) By superposition, the complete response is the sum of the answers to parts (c), (d) and (e).

SOLUTION 14.41. With vin (t) = 4u(t)V and vC (0 ) = 1 V, a single node equation at the front half of
the circuit yields with CvC (0 ) = 1 1= 1:

2/23/02

page P14.5

R. A. DeCarlo, P. M. Lin

s2 + 2s + 4
4
8
s+8
4

2 + 2 + + s VC1(s) 1= 0
VC1 (s) = + 1 =

5
s
s
s
s

or

s+8
VC1(s) = 2
s + 2s + 4

For the rear-half, represent the capacitor by a series equivalent circuit. Thus we can obtain an equivalent
voltage source with value:
V (0 )
2(s + 8)
1 2s(s + 8) (s2 + 2s + 4)
Veq (s) = 2VC1 (s) C
= 2
=
s
s + 2s + 4 s
s s2 + 2s + 4

or equivalently
Veq (s) =

s2 + 14s 4

s s2 + 2s + 4

By a voltage division,
1
s

vC (0 )
2
s2 + 14s 4 1
Vout (s) = 1
Veq (s) +
=

+
s
s + 2 s s2 + 2s + 4
s
+ 0.5
s
=

2s2 + 28s 8 + (s + 2) s2 + 2s + 4

s( s + 2) s2 + 2s + 4

) = (s2 + 6s + 36)
(s + 2)(s2 + 2s + 4)

Using MATLAB
num = [1 6 36];
den = conv([1 2],[1 2 4])
den =
1 4 8 8
[r,p,k] = residue(num,den)
r=
7.0000e+00
-3.0000e+00 - 2.8868e+00i
-3.0000e+00 + 2.8868e+00i
p=
-2.0000e+00
-1.0000e+00 + 1.7321e+00i
-1.0000e+00 - 1.7321e+00i
k=
However, it would appear easier here to use ilaplace:
syms t s
ilaplace((s^2+6*s+36)/((s+2)*(s^2+2*s+4)))
ans =
7*exp(-2*t)-6*exp(-t)*cos(3^(1/2)*t)+10/3*exp(-t)*3^(1/2)*sin(3^(1/2)*t)

2/23/02

Hence

page P14.6

R. A. DeCarlo, P. M. Lin

vout (t) = 7e 2t + e t 3.334 3 sin( 3t 6cos( 3t) u(t) V

SOLUTION 14.42. Using the series equivalent circuit (figure 14.17) for C1, we have
I1K(s) =

Next, since vC 2 (0 ) = 0 , we have

vC1 (0-)/s
= - 0.25/s = -0.25
1
1000 + 50/s 1000s + 50
R+
sC 1

I (s) 0.25 500

0 / 1 2 5
VC 2 (s) = 1k =
=
C2s
s(1000s + 50) s(s + 0.05)

Finally,
Vout (s) = -VC2 (s) =

0.125 = 2.5( 1 1
s s + 0.05 )
s(s + 0.05)

and
vout (t) = 2.5( 1 - e -0.05t )u(t) V

SOLUTION 14.43. (a) It is preferable to use the series equivalent circuit (figure 14.17) for C1, and the
parallel equivalent circuit (figure 14.16) for C2.
(b) The current through the 2.5 k resistor is given by
-810-4
v (0-)/s
- 2/s
-2
I2.5K (s) = C1
=
=
=
2500 + 5000/s 2500s + 5000
s+2
R1 + 1
sC 1
Next,
-810-4
1
1
-4
Vout (s) =VC2 (s) = I 2.5K (s)
=

=
1
s
+
2
0.0002s
+
0.0002
(s
+
2)
(s + 1)
sC 2 +
R2
(c) Hence
Vout (s) =
and

-4
=-4 1 - 1
(s + 2) (s + 1)
s + 1 s +2

vout (t) = 4 e 2t e t u(t) V

(d) SPICE plot omitted.

SOLUTION 14.44. (a) From voltage division,

2/23/02

page P14.7

R. A. DeCarlo, P. M. Lin

R2
V
Z
R
C
R2 C1 s
2
2 2s + 1
H(s) = C2 =
=
=
2
Vin Z1 + Z2
R2
1 + R1 +
R1 C1 R2 C2 s + ( R1 C1 +R2 C2 + R2 C1 )s + 1
C1 s
R2 C2 s + 1
(b) If vin(t) = 15u(t) V, then Vin (s) = 15/s and
VC2 = H(s)Vin =

s2

1.75s
7
15
=
- 7
s
s+
0.25
s+ 4
+ 4.25 s + 1

Hence

vC 2 (t) = 7 e 0.25t e 4t u(t) V

15

vin and vc2

10

TextEnd
5

4
Time in s

(c) Using the series equivalent for C1, we have

VC2 = H(s)

vC1 (0-)
1.75s
=
15
s
2
s + 4.25 s + 1 s

which is the same as result in part (b). Therefore

vC 2 (t) = 7 e 0.25t e 4t u(t) V

(d) Using the parallel equivalent for C2, we have

2/23/02

page P14.8

R. A. DeCarlo, P. M. Lin

1
1
1
=2 15
=
+ 14
2s + 4 + 1
7
2s + 4 + s
s + 0.25
s+4
7
7 2 + 1/s
7
7 2s + 1

VC2 (s) = C 2 vC2 (0-)

Hence

vC 2 (t) = e 0.25t + 14e 4t u(t) V

(e) By linearity, the answer is the sum of parts (b), (c) and (d).

SOLUTION 14.45. (a) Z1 = R1 + L1 s and Z2 =

IL1 =

R2 L2 s . From Ohm's law

R2 + L2 s

(R2 + L2 s)Vin
Vin =
Vin
=
Z1 + Z2 R + L s + R2 L2 s
(R1 + L1 s) (R2 + L2 s) + + R2 L2 s
1
1
R2 + L2 s

Using current division, we have

IL2 =

R2
R2 Vin
I =
R2 + L2 s L1 (R1 + L1 s) (R2 + L2 s) + + R2 L2 s

Therefore
R2
G1
H(s) = IL2 =
=
2
Vin
(R1 + L1 s) (R2 + L2 s) + + R2 L2 s
G1 L1 G2 L2 s + (G 1 L1 + G 2 L2 + G 1 L2 )s + 1
With the given element values,
H(s) =

2
2
=
2 + 4.25s + 1
4
7
4
7
7
2
s
21 s + (21+ + 2 )s + 1
7 8
7 8
8

(b) If vin(t) = 15u(t) V, then Vin (s) = 15/s and

IL2 = H(s)Vin =
Hence

2
32
2
15 = 30
s - s+ 0.25 - s+ 4
s 2 + 4.25 s + 1 s

iL 2 (t) = 30 32e 0.25t + e 4t u(t) A

Plot omitted.
(c) Using the series equivalent for L1, we have
IL2 = H(s) L 1 iL1(0-) =
Therefore

2
8
115 =
- 8
s + 0.25 s +4
s 2 + 4.25 s + 1

iL 2 (t) = 8 e 0.25t e 4t u(t) A

2/23/02

page P14.9

R. A. DeCarlo, P. M. Lin

(d) Using the parallel equivalent for L2, we have

7 15
L2 iL2(0-)
15( s + 2.25)
8
8
IL2(s) =
=
=
=
+ 7
2 +4.25s + 1
R2 (R1 + L1 s)
1.75(0.5 + s)
s + 0.25 s + 4
s
7
L2 s +
s+
R2 + (R1 + L1 s)
8
1.75 + (0.5 + s)
Hence
iL2(t) = (8e-0.25t + 7 e-4t ) u(t) A
(e) By linearity, the answer is the sum of parts (b), (c) and (d).

SOLUTION 14.46. (a) Using the result of problem 14.44(a)

VC2 =
R2 C1 s
2Iin
R1 C1 R2 C2 s2 + ( R1 C1 +R2 C2 + R2 C1 )s + 1
Therefore
H1 (s) = VC2 =
Iin

1.75s
s2 + 4.25 s + 1

3.5s
s2 + 4.25 s + 1

(b) Using the result of problem 14.45(a),

R2
2
H2 (s) = IL2 =
=
2
VC2 (R1 + L1 s) (R2 + L2 s) + + R2 L2 s
s + 4.25s + 1
(c)
3.5s
2
7s
H(s) = IL2 =H 1 (s)H 2 (s) =

=
Iin
s2 + 4.25s + 1 s 2 + 4.25s + 1 (s2 + 4.25s + 1)2
(d) We first represent the initialized capacitor by the series equivalent (figure 14.17), and then apply a
source transformation. From this circuit, by utilizing the expression derived in part (c), we have
IL2 =sIL2 H(s) =H (s)H (s) =
7s
1
2
vC1 (0) 7.5
(s2 + 4.25s + 1)2
2s
Therefore
IL2(s) =

(s2

52.5
= 1.9911 + 3.7333 - 1.9911 + 3.7333
2
s+4
+ 4.25s + 1)
( s + 4) 2 s + 0.25 ( s + 0.25)2

and
iL2(t) = [(1.9911 + 3.7333t) e-4t +(- 1.9911 + 3.7333t) e-0.25t ] u(t) A
Note: the book answer for part (d) should be divided by 2.
(e) Since Iin(s) = 15/s, we have

2/23/02

page P14.10

IL2(s) = H(s)Iin (s) =

(s2

R. A. DeCarlo, P. M. Lin

105
= 3.9822 + 7.4666 - 3.9822 + 7.4666
2
s+4
+ 4.25s + 1)
( s + 4) 2 s + 0.25 ( s + 0.25)2

and
iL2(t) = [(3.9822 + 7.4666t) e-4t +(- 3.9822 + 7.4666t) e-0.25t ] u(t) A

SOLUTION 14.47. (a) For this passive circuit, we may write the nodal equations by inspection.
0.8s + 2 + 10
s
- 10
s

- 10
s

VC
VR

1 + 10
s

= 2Vs1
-Is2

(b) Vs1= 3/s and Is2 = 3/s. We solve for VR by Cramer's rule to obtain
0.8s + 2 + 10
s
- 10
s
VR =
0.8s + 2 + 10
s

6
s
- 3s
- 10
s

- 10
s
and

- 2.4s2 - 6s + 30 = - 4 + 1
s(0.8s2 + 10s + 30) s + 7.5 s

1 + 10
s

v R (t) = 1 4e 7.5t u(t) V

(c) We represent the initialized capacitor by the parallel equivalent circuit ( figures 14.16) In this case
the nodal equations becomes
0.8s + 2 + 10
s
- 10
s

- 10
s
1 + 10
s

VC
VR

6 + 2.4
s
=
- 3s

Solve for VR by Cramer's rule to obtain

0.8s + 2 + 10
s
- 10
s
VR =
0.8s + 2 + 10
s
- 10
s

6 + 2.4
s
- 3s
- 10
s
1 + 10
s

2
= - 2.4s +18s + 30 = s1 - 16 + 12
2
s + 7.5 s + 5
s(0.8s + 10s + 30)

2/23/02

page P14.11

R. A. DeCarlo, P. M. Lin

vR(t) = (1 - 16e-7.5t + 12e- 5t ) u(t) V

SOLUTION 14.48. (a) After performing the suggested source transformation, and representing the
initialized capacitor and inductor by their series equivalent circuits, we can write two mesh equations by
inspection:
0.5 + 1.25
s
- 1.25
s

- 1.25
s
1 + 0.1s + 1.25
s

v (0)
Vs1 - Cs
Is1 =
IL
vC(0)
s + LiL(0) + I s2

(b) With Vs1= 3/s, Is2 = 3/s, vC(0) = 0, and iL(0) = 3 A, the above mesh equation becomes
0.5 + 1.25
s
- 1.25
s

- 1.25
s
1 + 0.1s + 1.25
s

vC(0)
s

3
s
Is1 =
=
IL
vC(0)
0.3 + 3s
s + LiL(0) + I s2
Vs1 -

Solve for IL(s) by Cramer's rule to obtain

IL(s) =

0.15s 2 +1.875s + 7.5

= 2
+ -3 + 4s
3
2
s
+
7.5
s+5
0.05s + 0.625s + 1.875s

Therefore
iL(t) = (4 + 2e-7.5t -3 e-5t ) u(t) A

SOLUTION 14.49. (a) Represent the initialized capacitors by their parallel equivalent circuits.
(b) Write two nodals equation by inspection

0.001s + 0.4
- 0.2
- 0.2
0.001s + 0.4

VC1 = 0.2V in + 0.001vC1 (0-) =

VC2
0.001v C2 (0-)

2.4 + 0.006
s
0.002

(c) Solve for VC2(s) by Cramer's rule to obtain

0.001s + 0.4 2.4/s + 0.006
2s 2 + 2103 s + 48104
- 0.2
0.002
0
-2
VC2 =
=
=
+
+ + s4
0.001s + 0.4
- 0.2
2
4
s
+
600
s
+
200
2
s(s + 810 s + 1210 )
- 0.2
0.001s + 0.4
(d)

2/23/02

page P14.12

R. A. DeCarlo, P. M. Lin

vC2 (t) = (4 - 2e-200t ) u(t) V

SOLUTION 14.50. (a) Let VC denote the node voltage across the capacitor. By inspection the nodal
equations in matrix form are:
1 /R
1+ 1 /R + 4s
VC Vin
=
1 /R
1 + 1/ R + 1 / ( s4) Vout Vin / ( 4s)

(b) By Cramer's rule,

1
1 + 1/ R + 4s
det
1 /R
1 / ( 4s)
Vout (s)
( 4s + 1)(1 + 1/ R)

H(s) =
=
=
1+
1
/R
+
4s
1
/R
Vin (s)

4s (1+ 1/ R + 4s)(1+ 4s / R + 4s) 1/ R 2

det
1 + 1/ R + 1 / ( s4)
1 /R

(4s + 1)(1 + 1/ R)
1
=
2
(1 + 8s + 16s )(1+ 1 /R) (1+ 4s)

Clearly, R does not affect the transfer function. The question is why? Note that the circuit can be
redrawn as a balanced Wheatstone bridge circuit in which there is no voltage across R and no current
through R. Hence R has no effect on the transfer function and on the impedance at the input. Hence R
can be removed in the analysis of the circuit. In this case, the transfer function follows trivially by
voltage division.
(c) In view of the answer to (b), the impedance can be calculated with R removed. Hence
1

1+
(1+ 4s)

(1 + 4s)2
4s
Zin (s) =
=
2 =1
1

(
1
+
4s
)
1+
+ (1+ 4s)

4s
Hence, the input impedance is a constant resistance and the network is called a constant resistance
network.
(d)
The input is vin(t) = 10eatu(t) V and R = 5 . Find vout(t) for t 0 for the three cases, a = 0,
0.5, 0.25.
(d) From part (b), for s 0.25,
0.25
10
2 . 5 / ( 0 . 25 a) 2 . 5 / ( 0 . 25 a)
Vout (s) =

(s + 0.25) s + a
s+ a
(s + 0.25)
which leads to
2.5 at
vout (t) =
e e 0.25t u(t) V

0.25 a
For a = 0.25,

2/23/02

page P14.13

Vout (s) =

2.5
(s + 0.25)2

R. A. DeCarlo, P. M. Lin

vout (t) = 2.5te0.25t u(t) V

implying that

SOLUTION 14.51.
(a) This bridged-T circuit was analyzed in problem 14.9. Here R = 1 , Z1(s) = 0.25s and Z2(s) = 4/s.
Since the condition Z1(s) Z2(s) = R2 is met, we have Zin(s)= 1.
(b) The s-domain equivalent circuit accounting for initial conditions is given below.

(c) Two nodal equations at Vc and Vout are:

VC - Vin + VC - Vout + 0.25sVC = 0.25 vC (0-)

and

Vout - VC +

Vout - Vin
+ Vout = iL(0 )
s
0.25s

Writing these in matrix form, we have

0.25s + 2
-1

-1

VC

4 +2
s

Vout

Solving for Vout by Cramer's rule yields

Vin + 0.25vC (0-)

=
4Vin + iL(0-)
s
s

2/23/02

page P14.14

R. A. DeCarlo, P. M. Lin

0.5(s+ 8)
Vout (s) = 4 Vin (s) +
iL(0-) + 0.5s vC(0-)
2
s+4
(s+4)
(s+4) 2
(d) Given vin(t) = 4u(t) - 3e-t u(t) V, then
Vin (s) = 4s - 3 = s + 4
s+1 s( s + 1)
and
Vout (s) =

0.25(s+ 8)
4
+
+ 0.75s
s(s + 1)
(s+4) 2
(s+4) 2

Taking the inverse Laplace transform, we obtain, for t 0,

vout(t) = (4 - 4 e-t) + (0.25 e-4t + te-4t) + (0.75 e-4t - 3 te-4t) = 4 - 4 e-t + e -4t - 2te-4t

SOLUTION 14.52. A supernode is defined by drawing a curve to enclose the controlled voltage source.
One node within the supernode has voltage Vout and the other has voltage V1 that is equal to
V1 = -2I1 - Vout = -2 Vin - VC = - Vin + VC - Vout
2
Next, we write nodal equations at VC and the supernode: At node VC
0.5 (VC - Vin ) + 0.5sVC + 0.5
s (VC- Vout ) = 0
At the supernode
1 (V - V ) + 1 Vout + - Vin + VC + Vout = 0
2s out C
2
1
In matrix form, the nodal equations are:
0.5(s + 1 + 1/s)
1 - 0.5/s

-0.5/s
1.5 +0.5/s

VC = 0.5 V in
Vout
Vin

Solving by Cramer's rule yields

2( s 2 + 1.5)
H(s) = Vout = 0.5s + 0.75/s =
Vin
0.75s + 1+ 1.5/s
3s 2 + 4s + 6

SOLUTION 14.53. A supernode is defined by drawing a curve to enclose the controlled voltage source.
One node within the supernode has voltage Vout and the other has voltage V1 which is equal to

2/23/02

page P14.15

R. A. DeCarlo, P. M. Lin

V1 = -2I1 - Vout = -2 Vin - VC = - Vin + VC - Vout

2
Next, write nodal equations at VC and the supernode: At node VC
s (VC- Vout ) = 0
0.5 (VC - Vin ) + (0.25s + 4s )V +
C
s2 + 16
At the supernode
s (Vout - VC) + 1 Vout + - Vin + VC + Vout = 0
2
1
s 2 + 16
In matrix form, the nodal equation are:
0.25s + 0.5 +s/( s2 + 16)

-s/(s2 + 16)

1 - s/(s 2 + 16)

1.5 + s/(s2 + 16)

VC = 0.5 V in
Vout
Vin

Solving by Cramer's rule yields

0.25s 3 +5.5s
H(s) = Vout =
Vin
0.375s 3 + s 2 + 9s + 12

SOLUTION 14.54. Write nodal equations at V1 and V2:

0.5 (V1 - Vin ) + 0.125( V1 + 0.2V 2 ) + 0.1s(V1 - V2 ) = 0
and
0.1s(V2 - V1 ) + 1 V2 + V2 - 5V1 = 0
4
1
In matrix form, the nodal equations are:
(0.1s + 0.625)
(-0.1s - 5)

(-0.1s + 0.025)
(0.1s + 1.25)

VC = 0.5 V in
Vout
0

Solving by Cramer's rule yields

0.05s + 2.5
H(s) = Iout = V2 =
Vin Vin - 0.31s + 0.9062

SOLUTION 14.55.
(a) Simply replace each capacitor by the parallel form circuit model given in figure 14.16.
(b) For this passive circuit, we can write the nodal equation by inspection.

2/23/02

page P14.16

0.5s +2
-1
0
-1
0.5s +2
-1
0
-1
0.5s +2

VC1
VC2
VC3

R. A. DeCarlo, P. M. Lin

Vin + 0.5vC1 (0)

=
0.5v C2 (0)
0.5v C3 (0)

Solving for VC3 by Cramer's rule yields

VC3 (s) =

Vin (s) + 0.5vC1 (0) + 0.25s + 1 vC2 (0) +( 0.125s2 + s +1.5 )vC3 (0)
0.125s 3 + 1.5s2 + 5s + 4

(c) Substituting Vin(s) = 12/s, vC1(0) = 0, vC2(0)= 6, and vC3(0) = 2 into the above expression, we
obtain
VC3 (s) =

12/s + 6 0.25s + 1 + 2( 0.125s2 + s +1.5 )

3
2
= 0.25s + 3.5s + 9s +12
3
2
3
0.125s + 1.5s + 5s + 4
s 0.125s + 1.5s2 + 5s + 4

Now use MATLAB to do the partial fraction expansion.

n= [ 0.25 3.5 9 12];
d= [ 0.125 1.5 5 4 0];
[ r p k ] = residue (n,d)
r=
-2.5000
4.0000
-2.5000
3.0000
p=
-6.8284
-4.0000
-1.1716
0
From the MATLAB output, we have, for t 0,

vC3(t) = 3 - 2.5 e-6.828t + 4e-4t - 2.5e-1.1716t V

SOLUTION 14.56. For this problem we utilize loop analysis with loops as indicated below.

2/23/02

page P14.17

R. A. DeCarlo, P. M. Lin

In doing the following loop analysis, note that we will use gmVout = I3 and that due to our judicious
choice of loops
1

Vout = + Ls I1
Cs
or equivalently,
1

0 = + Ls I1 Vout
Cs
For loop 1,
1

V1 = 2R +
+ Ls I1 + 2R I2 + RgmVout
Cs
For loop 2,
V1 V2 = 2R I1 + 7R I2 + 5RgmVout
In Matrix form
1000

+ 0.016s
0 1

s
1000
I1 0
2 +
+ 0.016s 2 2 I 2 = V1
s

2
7 10 Vout V1 V2

By Cramer's rule

2/23/02

page P14.18

1000
+ 0.016s
0

s
1000
det 2 +
+ 0.016s 2
s

2
7

Vout =
1000
+ 0.016s

s
1000
+ 0.016s
det 2 +
s

R. A. DeCarlo, P. M. Lin

V1

V1 V2
1000

+ 0.016s (5V1 + 2V2 )

=
6000
7000

0 1
+ 0.096s 10 +
+ 0.112s

s
s

2 2

7 10

Hence
1000

+ 0.016s (5V1 + 2V2 )

s

s2 + 62500
Vout =
=
2 + 625s + 62500 (5V1 + 2V2 )
1000
s
10 +
+ 0.016s
s
The answers to (a) and (b) are clear at this point.
(c) Using MATLAB
n = [21 0 21*62500];
d = [1 625 62500 0];
[r,p,k] = residue(n,d)
r=
35
-35
21
p=
-500
-125
0
k = []
Hence
vout(t) = 21 35e 125t + 35e 500t u(t ) V.

SOLUTION 14.57. (a) Replace the LC combination by a 1 V source after setting V1 and V2 to zero.
We need to compute the current leaving the 1 V source which will be 1/Rth. Let the left node be denoted
by Va and the right node by Vb. Also let G = 1/R. The nodal equations are by inspection
0 Va gm + G
2G
Va
( gm + G) 2G

0 1.25G V = g + 0.25G V =

b m

b ( gm + 0.25G) 1.25G

2/23/02

page P14.19

R. A. DeCarlo, P. M. Lin

Thus, the current leaving the 1 V source is

I1V = G(1 Va ) + 0.25G(1 Vb ) = G(1 (gm + G) 2G) + 0.25G(1 (gm + 0.25G) 1.25G)
= 0.5G 0.5gm + 0.25G + 0.2gm 0.05G = 0.7G 0.3gm
Substituting G = 1/R we obtain
1
1
R
1
Rth =
=
=
=
= 10
I1V 0.7G 0.3gm 0.7 0.3gm R 0.7 0.3 2
(b) Replace the LC combination by a short circuit and compute Isc. This makes the controlled source
zero. By inspection
V
V
Isc = 1 + 2
2R 5R
Thus
V
V
Voc = RthI sc = Rth 1 + 2 = 10(0.5V1 + 0.2V2 )
2R 5R
(c) By voltage division
1
+ Ls
ZLC
s2 + 1 LC
Vout =
Voc = 1 Cs
Voc =
(5V1 + 2V2 )
Rth
Rth + ZLC
2
+ Ls + Rth
s +
s + 1 LC
Cs
L
s2 + 62500
= 2
(5V1 + 2V2 )
s + 625s + 62500

SOLUTION 14.58. (a) The last equation is the constraint equation for the controlled floating voltage
source. Hence, we have
V1 V2 z0 (s)I 0 = 0
(b) By Cramer's rule,
1

1
R + Cs Iin

det
0
0
1

1
0 z0 (s)

Iin
Iin

=
V2 =
=
1
1
1
1

1 2 + Cs z0 (s)Cs + Cs + Cs (2 + z0 (s)Cs)
R + Cs 1 R
R
R
R

det
0
Cs
1

1
1 z0 (s)

(c) Here

2/23/02

page P14.20

V2 =

in which case

Iin
1

+ Cs 2 + LCs 2
R

R. A. DeCarlo, P. M. Lin

2
.
LC

SOLUTION 14.59. (a) Since the switch has been at position A for a very long time, the inductor looks
like a short and iL(5-) = iL(5+) = 10/4 = 2.5 A. For t > 5, the switch moves to position B and the
inductor current decays according to
iL (t) = iL (5+ )e (t 5)/ = 2.5e (t 5)/0.1 = 2.5e 10(t 5) A
(b) Note that iL(0-) = iL(0+) = 0. Hence
1
0.1 50
50 12.5
62.5
50
IL (s) =
Vin (s) =

+
10s + 4
s + 0.4 s s + 0.5
s
s + 0.4 s + 0.5
0.4 t
0.5t
Hence for 0 t 5s, iL (t) = 12.5 62.5e
A. Here iL(5 ) = iL(5+) = 8.1458 A.
+ 50e
For t > 5, the inductor decays with a time constant of 0.1 s. Thus
iL (t) = 8.1458e 10(t 5)

SOLUTION 14.60. (a) Since the switch has been at position A for a very long time, the capacitor
looks like an open and vC(5-) = vC(5+) = 40 V. For t > 5, the switch moves to position B and the
capacitor voltage decays according to
vC (t) = vC (5+ )e (t 5)/ = 40e (t 5)/2 = 40e 0.5(t 5) V
(b) Note that vC(0-) = vC(0+) = 0. Hence
1 /Cs
12.5 50
50
7812.5
VC (s) =
Vin (s) =

=
1 /Cs + 40
s + 12.5 s s + 12.5 s( s + 12.5)2
In MATLAB,
syms t s
ilaplace(7.8125e3/(s*(s+12.5)^2))
ans =
50-625*t*exp(-25/2*t)-50*exp(-25/2*t)
Hence for 0 t 5s, vC (t) = 50 625te12.5t 50e 12.5t V. Here vC(5-) = vC(5+) = 50 V.
For t > 5, the capacitor voltage decays with a time constant of 0.08 s. Thus
vC (t) = 50e 12.5(t 5) V

2/23/02

page P14.21

R. A. DeCarlo, P. M. Lin

SOLUTION 14.61. (a) Since the switch has been closed for a very long time, the capacitor looks like
an open and vC(5-) = vC(5+) = 32 V. For t > 5, the switch opens and the capacitor voltage decays
according to
vC (t) = vC (5+ )e (t 5)/ = 32e (t 5)/0.4 = 32e 2.5(t 5) V
(b) Note that vC(0-) = vC(0+) = 0 and vout = vC. Hence for 0 t 5,
1
Vin (s)
10
10 50
50
VC (s) = 1
=
V
(s)
=

in
1
s + 12.5
s + 12.5 s s + 12.5
+
+ 0.002s 50
50 200
6250
=
s(s + 12.5)2
In MATLAB,
syms t s
ilaplace(6250/(s*(s+12.5)^2))
ans =
40-500*t*exp(-25/2*t)-40*exp(-25/2*t)
Hence for 0 t 5s, vC (t) = 40 500te12.5t 40e 12.5t V. Here vC(5-) = vC(5+) = 40 V.
For t > 5, the capacitor voltage decays with a time constant of 0.4 s. Thus
vC (t) = 40e 2.5(t 5) V

SOLUTION 14.62. (a) At 0-, vC( 0 ) = 0 and iL(0 ) = 50/10 = 5 A.

(b) For this part, consider the equivalent circuit below.

By inspection,
VC =

From table 13.1,

5
1

s 1+ 0.5s +

0.184s

5
0.5s2 + s +

1
0.184

10

(s + 1) 2 + 2

2/23/02

page P14.22

R. A. DeCarlo, P. M. Lin

vC (t) = 3.1831e t sin(t)u(t) V

(c) Using MATLAB,

SOLUTION 14.63. (a) Since the switch has been closed for a long time, iL(1 ) = iL(1 ) = 30/0.8 = 37.5 A
-

and vC(1 ) = vC(1 ) = 0. Represent the initialized inductor by its parallel equivalent circuit. Then
VC (s) =

37.5
1
150

=
1
s
s2 + 4
Cs +
Ls

Hence from table 13.1,

vC (t') = 75sin(2 t') V

vC (t) = 75sin(2( t 1 ) ) V for t > 1s.

(b) All initial conditions at t = 0 are zero. For 0 t 1s,

1
Vin (s)
5s
150s 1
1
VC (s) = 1
=
V
(s)
=

in
2
2
1 0.8
s + 5s + 4
s + 5s + 4 s s + 2
+ 0.25s +
0.8
s

Hence, for 0 t 1s,

300s 1 100 150

50
= 2

=
s +1 s + 2 s + 4
s + 5s + 4 s(s + 2)

vC (t) = 100e t 150e 2t + 50e 4t V

Here, vC(1-) = vC(1+) = 17.403 V. Note, vin(1-) = 25.94 V. Next,

2/23/02

page P14.23

iC (1 ) = C

R. A. DeCarlo, P. M. Lin

dvC
= 0.25 100e 1 + 300e 2 200e 4 = 0.037378

dt t =1

Thus in MATLAB
vin1 = 30*(1 - exp(-2))
vin1 = 2.5940e+01
vc1 = 17.403
vc1 = 1.7403e+01
ic1 = 0.25*(-100*exp(-1) + 300*exp(-2)-200*exp(-4))
ic1 = 3.7378e-02
iL1 = (vin1 - vc1)/0.8 - ic1
iL1 = 1.0634e+01
Therefore, iL (1 ) = iL (1+ ) = 10.634 A. For t 1, we use the parallel equivalent circuit for both the
inductor and the capacitor:
+
4s 10.634

s 4s iL (1 )
+
VC (s) = e
+
Cv
(1
) = e s 2
+ 4.3507

C
2

s
s
s + 4

s +4
Therefore from table 13.1, for t > 1,
vC (t) = 21.268sin (2(t 1)) + 17.403cos (2(t 1)) V
Plots omitted.

SOLUTION 14.64. Here vC(0) = 0 for both capacitors.

Part 1: 0 t 1s.
10
20
s 20 20 =
Vout (s) =
10 s s + 2 s(s + 0.5)(s + 2)
20 +
s
From MATLAB
num = 20;
den = [1 2.5 1 0];
[r,p,k] = residue(num,den)
r=
6.6667e+00
-2.6667e+01
2.0000e+01
p=
-2.0000e+00
-5.0000e-01
0
k = []

2/23/02

Therefore for 0 t 1,

page P14.24

R. A. DeCarlo, P. M. Lin

vout (t) = 20 26.667e 0.5t + 6.667e 2t (u(t) u(t 1) )

Part 2. 1 t. Here the initial condition for the right-most capacitor is vout (1 ) = vout (1 + ) = 4.7281 V.
As above, the left-most capacitor has zero value at t = 1s. Let us use the series equivalent circuit for the
right capacitor. Then,
4.7281
1
s 0.47281
IC (s) = e s

=
e
20
s
s+2
10 +
s
Therefore,
4.7281
4.7281 4.7281 s 2.3641 2.3641
10
Vout (s) = IC (s) + e s
= e s
+
=e
+

s
s
s + 2
s
s
s(s + 2)
and for t 1,
vout (t) = 2.3641+ 2.3641e 2(t 1) u(t 1) V

SOLUTION 14.65. Assume the switch has been in position A for a long time. Both capacitors behave as
open circuits and both capacitors have initial voltages at t = 0 of 10 V. For t 0, use the parallel
equivalent circuits for both capacitors and write nodal equations. Let the left capacitor have voltage
VCa.
0.01
0.005s + 0.03
VCa 0.005 10 0.05
=
=

0.01
0.0025s + 0.01 Vout 0.0025 10 0.025

By Cramer's rule,
0.005s + 0.03 0.05
det
0.01
0.025
s + 10
4 0 /3 10 / 3

Vout (s) =
= 10 2
=

0.01
0.005s + 0.03

s + 10s + 16 s + 2 s + 8
det
0.01
0.0025s + 0.01

Therefore for t 0,
vout (t) =

2s
.
2 = 2 + s2 + 4
0.5s +
s
(b) Here, the initial condition is zero and
SOLUTION 14.66. (a) Zin (s) = 2 +

40 2t 10 8t
V
e
e
3
3

2/23/02

page P14.25

R. A. DeCarlo, P. M. Lin

2s
VC (s) =

10
s + 4 10 =
2
2s
s s +s+ 4
2+ 2
s +4

syms t s
ilaplace(10/(s^2+s+4))
ans =
4/3*exp(-1/2*t)*15^(1/2)*sin(1/2*15^(1/2)*t)
Hence using MATLAB above or table 13.1 we have for 0 t 1.5s,
vC (t) = 5.164e 0.5t sin(1.9365t) V
and

vC (1.5 ) = vC (1.5 + ) = 0.57237 V

(c) Use the parallel equivalent circuit for the left capacitor. The right capacitor has a zero initial voltage
at t = 1.5. Hence, we do not use an equivalent circuit for the right capacitor.
(d) Therefore
e1.5sVC (s) =

s+4
s 0.5 0.57237 = 0.57237 s2 + 10s + 16
0.5s + 2 +
s+ 4
1

(e) In MATLAB
[r,p,k] = residue(0.57237*[1 4],[1 10 16])
r =V
3.8158e-01
1.9079e-01
p=
-8
-2
k=
[]
Hence

(f) Finally

s+4
0.19079 0.38158
VC (s) = e 1.5s 0.57237 2
= e 1.5s
+
s + 8
s+2
s + 10s + 16

vC (t) = 0.19079e 2(t 1.5) + 0.38158e 8(t 1.5) u(t 1.5) V

R
16 = 8 V.
2R
(b) For 0 t 1, v1(t) = 8 V and v 2 (t) = 8e t /RC = 8e 0.6931t V.
(c) v1(1 ) = 8 V and v 2 (1 ) = 8e 0.6931 = 4 V.
SOLUTION 14.67. (a) v1(0 ) = v1 (0+ ) = v 2 (0 ) = v2 (0 + ) =

2/23/02

page P14.26

R. A. DeCarlo, P. M. Lin

(d) From KVL, v1(1+ ) = v 2 (1 + ). From conservation of charge,

1 v1(0 ) + 1 v 2 (0 ) = 12 = 2 v1 (0+ ) = 2 v 2 (0+ ) . Therefore v1(1+ ) = v 2 (1 + ) = 6 V.
(e) This response represents a decay with time constant = 2R = 2.8854 s. Hence
v1(t) = v 2 (t) = 6e 0.34657( t 1)u(t 1) V
It follows that

v1(3) = v 2 (3) = 6e 0.346572 = 3 V

(f) Both capacitor voltages change abruptly at t = 1.

SOLUTION 14.68. Label the current down through the first inductor as i1(t).
(a) i1 (0 ) = i1(0+ ) = 1 A a n diout (0 ) = iout (0+ ) = 0 .
(b) For 0 t we use a parallel equivalent for the first inductor. By current division
1
1
175
5 + 0.1s
Iout (s) =

= 2
1
1
s
s + 275s + 2500
0.057143 +
+
0.35s 5 + 0.1s
Use MATLAB to do the partial fraction expansion
num = -175; den = [ 1 275 2500]; [ r, p, k] = residue (num, den)
r=
0.6831
-0.6831
p=
-265.5869
-9.4131
From the MATLAB output
Iout (s) = Therefore ,

0.68313 + 0.68313
s+ 9.4131
s+ 265.59

iout (t) = 0.68313 e 265.59t e 9.413t u(t) A

SOLUTION 14.69. (a) Here we use voltage division:

2/23/02

page P14.27

R. A. DeCarlo, P. M. Lin

1
V1(s) =

1
6

Therefore, v1(0 + ) = 5 V.
(b) Again use voltage division:

4 10

4 106 s
1
4 10

1
1 106 s

30 30 5
=
=
s
6s s

1
V1(s) =

1
5 106 s

Therefore, v1(0 + ) = 4.7059 V.

5 10 6 s
1
1 10 6 s

40 80
=
s 17s

2 106 s

SOLUTION 14.70. (a) Consider a mesh current I(s) in the usual direction and use the series equivalent
circuit for each capacitor. Thus
I(s) =

1
4 10 6 s

1
1
4 106 s

1
1 10 6 s

3 0.3 0.9 0.6 1.2

6
6
s s s s = 1.5 10 = 0.8 10

Therefore for t > 0,

1
0.3 0.8 10 6 0.3 0.5
V1(s) =
I(s) +
=
+
=
v1(t) = 0.5 V
s
s
s
4 106 s
4 106 s
Similarly for t > 0
0.8 10 6 0.9 1.1
V2(s) =
+
=
v 2 (t) = 1.1 V
s
s
4 106 s
and
0.8 106 0.6 1.4
V3(s) =
+
=
v 3(t) = 1.4 V
s
s
1 106 s
(b) Consider a mesh current I(s) in the usual direction and use the series equivalent circuit for each
capacitor. Thus
I(s) =

1
5 10 6 s

1
1
1 106 s

1
2 10 6 s

4 0.3 0.9 0.6 2.2

6
6
s s s s = 1.7 10 = 1.2941 10

Therefore for t > 0,

1
0.3 1.2941 10 6 0.3 0.55882
V1(s) =
I(s) +
=
+
=
s
s
s
5 106 s
5 10 6 s
Similarly for t > 0

v1(t) = 0.55882 V

2/23/02

page P14.28

V2(s) =

1.2941 10 6 0.9 2.1941

+
=
v 2 (t) = 2.1941 V
s
s
1 10 6 s

V3(s) =

1.2941 106 0.6 1.2471

+
=
v 3 (t) = 1.2471 V
s
s
2 106 s

and

R. A. DeCarlo, P. M. Lin

SOLUTION 14.71. (a) For 0 < t < 2, the 150 mF capacitor is charged to 25 V. From conservation of
charge,
0.15 25 = 0.15vC (2+ ) + 0.1vC (2 + ) = 0.25vC (2+ )
Therefore
vC (2+ ) =

0.15 25
= 15 V
0.25

This voltage remains constant for t > 2 s.

(b) For 0 < t < 2, the 150 mF capacitor is charged to 25 V. From conservation of charge,
0.15 25 + 0.1 10 = 0.15v C (2 + ) + 0.1vC (2+ ) = 0.25v C (2+ )
Therefore
vC (2+ ) =

0.15 25 + 0.1 10
= 19 V
0.25

This voltage remains constant for t > 2 s.

SOLUTION 14.72. (a) Let the middle node have voltage Va(s). Then writing node equations
8s 2s Va 1.144
Va 1 6 2 1.144 s 0.156 / s
2s 6s V = 0 V = 44 2 8 0 = 0.052 / s V

out

out

Thus for t > 0, vout (t) = 0.052 V.

(b) Again define Va(s) as the middle node. Then
3
0.286 6 0.286 0.132
Va (s) = 1 10s

=
=
1
3
s
13
s
s
+
+
10s 4s 10s
Hence for t > 0,
1
0.132 0.132 0.044
Vout (s) = 1 4s 1
=
=
v out (t) = 0.044 V
s
3s
s
+
2s 4s

2/23/02

page P14.29

R. A. DeCarlo, P. M. Lin

SOLUTION 14.73. With switches in position A, the equivalent capacitance to the right of v2 is 4 mF.
Therefore at t = 0+, by voltage division
V1(s) = V2 (s) =

10
s

v1(t) = v 2 (t) = 10 V for 0 < t < 1.

Hence with the switches in position B, let us write a single node equation using the parallel equivalent
circuit the initialized capacitors:
0.002sV2 (s) 0.002 10 + 0.004sV2 (s) + 0.004 10 = 0.004s
Equivalently

11
s

0.006sV2 (s) = 0.044 0.02 = 0.024 V2 (s) = 4 / s

Hence, for t > 1 s, v 2 (t) = 4 V and v1(t) = 7 V.

SOLUTION 14.74. (a) At t = 0+, the frequency domain equivalent circuit is given below.

To compute Ceq, we observe

1
Ceq = 1+ 1 1 1 = 2 mF
+ +
3 2 6
By voltage division
Ceq s
20 10
0.002s
20 10
and V2 =

=
Ceqs + 0.002s s
s
Ceq s + 0.002s s
s
Hence, for 0 < t <1, v1(t) = v2(t) = 10 V.
(b) When the switch moves to B, the pertinent part of the equivalent frequency domain circuit is given
below.
V1 =

2/23/02

page P14.30

R. A. DeCarlo, P. M. Lin

By superposition,
1
8
1
1
6
+
0.02
0.01=
1 + 2 s 0.003s
0.003s
s
Hence, for 1 < t < 2, v1(t) = 6 V and by KVL, v2(t) = 2 V.
V1 =

SOLUTION 14.75. When the switch in position A, the 2 F capacitor is charged to 2 V. Hence, the
charge on the top plate is C*vC = 4 C. When the switch is moved to position B, due to the virtual
ground, the 2 F capacitor voltage is zero meaning it cannot retain any charge. Hence, assuming an
ideal op amp, all charge moves to the 1 F capacitor with 4 C on the left plate. Hence, vout = (4
C)/1 F = 4 V.

SOLUTION 14.76. (a) vout(t) = 0 for 0 < t < 1 ms. Every time the switch moves to position A, the
capacitor, C, charges to 8 V. When the switch moves to position B, because of the virtual ground, all
charges moves to the kC capacitor. Hence with k = 1, vout(t) = 8 V for 1 ms < t < 3 ms. For 3 ms < t <
5 ms, vout(t) = 16 V. Repeating the pattern implies that for 5 ms < t < 7 ms, vout(t) = 24 V, etc. See
for example figure 14.51.
(b) With k = 0.5, the voltages computed in part (a) double.
(c) With k = 2, the voltages computed in part (a) are halved.

SOLUTION 14.77.
(a) For this part consider the circuit below.

2/23/02

page P14.31

R. A. DeCarlo, P. M. Lin

For the normalized design C = 1 F, G1 = 0.5 S, G2 = 2 S, and G3 = 1.5 S. After magnitude scaling with
Km = 106, then C = 1 F , R1 = 2 M, R2 = 5 0 0
k , R3 = 6 6 6 . 7 .
k
(b) For this part, consider the circuit below.

For the normalized design C = 1 F, G1 = 1 S, G3 = 2 S, G2 = 0.5 S, and G4 = 2.5 S. After magnitude

scaling with Km = 106, then C = 1 F , R1 = 1 M, R3 = 5 0 0
k , R2 = 2 M, R4 = 4 0 0
k.
(c) Now consider the circuit below.

2/23/02

page P14.32

R. A. DeCarlo, P. M. Lin

For the normalized design C = 1 F, G1 = 0.25 S, G3 = 0.5 S, G2 = 0.75 S, G4 = 1 S, and G5 = 1 S. After

magnitude scaling with Km = 106, then C = 1 F , R1 = 4 M, R3 = 2 M, R2 = 4 / 3 M,
R4 = 1 M, R5 = 1 M.
(d) Finally, we consider the following circuit.

For the normalized design C = 1 F, G1 = 2 S, G3 = 1.5 S, G6 = 1 S, G2 = 0.5 S, G4 = 2 S, and G5 = 2 S.

After magnitude scaling with Km = 106, then C = 1 F , R1 = 0 . 5 M
, R3 = 2 /3 M ,
R6 = 1 M, R2 = 2 M , R4 = 0 . 5 M
, R5 = 0 . 5 M
.

2/23/02

page P14.33

R. A. DeCarlo, P. M. Lin

SOLUTION 14.78. With V1 = Vout a prototype design is given by the topology below.

Using MATLAB
Km = 1e7;
Gin = 1; G1 = 1.405; G2 = 0.402; G3 = 0.942;
DG = 1.865;
Rinnew = Km/Gin
Rinnew =
10000000
R1new = Km/G1
R1new =
7.1174e+06
R2new = Km/G2
R2new =
2.4876e+07
R3new = Km/G3
R3new =
1.0616e+07
DRnew = Km/DG
DRnew =
5.3619e+06
Later, when we study frequency scaling, Km will be smaller and the filter will have a cutoff frequency in
a more reasonable range.

SOLUTION TO 14.79. Note corrections to problem statement. W(0+) should be W() in part (b) and in
part (c) one should calculate W(0-) W(). The frequency domain equivalent circuit is given by the
figure below.

2/23/02

page P14.34

R. A. DeCarlo, P. M. Lin

(a) To find the required time functions, we first find their Laplace equivalents.
a b
ab

K1
s s
R
I(s) =
=

1
1
1
1 s + p1
R+
+
s
+
+

C1s C2s
RC1 RC2

in which case i(t) = K1e p1t u(t) where

K1 =

1
ab
1
and p1 =
.
+
R
RC1 RC2

Further,
K1

I(s) a a
C1
K 1
K
1
VC1(s) =
+ =
= a 1 + 1
C1s
s s s(s + p1)
C1 p1 s C1 p1 (s + p1)
in which case

K
K
vC1 (t) = a 1 + 1 e p1t u(t)
C1 p1 C1 p1

Also, by symmetry,
K1

I(s) b b
C2
K 1
K
1
VC 2 (s) =
+ = +
= b + 1 1
C2s s s s(s + p1)
C2 p1 s C2 p1 ( s + p1 )
in which case

K
K
vC 2 (t) = b + 1 1 e p1t u(t)
C2 p1 C2 p1

(b) The total energy stored in the capacitors at time 0- is

W (0 ) =
Also at t = ,

Hence

1
1
C1a2 + C2b2
2
2

K
K
vC1 () = a 1 and vC 2 () = b + 1 .

C1 p1

C2 p1

2/23/02

page P14.35

R. A. DeCarlo, P. M. Lin

1
K
1
K
W () = C1 a 1 + C2 b + 1
2
C1 p1
2
C2 p1
(c)

Ri
0

(t)dt =

RK12e 2 p1t dt =

RK12 (a b)2
(a b)2
=
=
2p1
2Rp1
1
1
2 +

C1 C2

Observe that
2

1
1
1
K
1
K
W (0 ) W () = C1a 2 + C2b 2 C1 a 1 C2 b + 1
2
2
2
C1 p1
2
C2 p1

aK1 bK1 1 K12

1 K12
(a b)2 1 (a b)2 (a b)2
(a b)2

=
=
p1
p1 2 C1 p12 2 C2 p12
Rp1
2
Rp1
2Rp1
1
1
2
+
C1 C2

This indicates that the total energy lost between 0- and infinity is the energy dissipated in the resistor
and the result is independent of the value of R.
(d) When R 0,
a b

s
s = a b = C1C2 (a b)
lim I(s) = 1
1
1
1
(C1 +C2 )
R 0
+
+
C1s C2s C1 C2
Therefore
C1C2
i(t) =
(a b) (t)
(C1 +C2 )
Further
C2
a
C1
C2
VC1(s) =
(a b) + =
a+
b
s(C1 +C2 )
s s(C1 +C2 )
s(C1 +C2 )
and
C1
b
C1
C2
VC 2 (s) =
(a b) + =
a+
b
s(C1 +C2 )
s s(C1 +C2 )
s(C1 +C2 )
Therefore
C1

C2
vC1 (t) = vC 2 (t) =
a+
b u(t)
(C1 +C2 )
(C1 +C2 )

SOLUTION TO 14.80. (a) From conservation of charge

C1vC1(0 ) + C2vC 2 (0 ) = C1v C1(0+ ) + C2vC 2 (0 + ) = (C1 + C2 )v C1(0 + ) = (C1 + C2 )v C 2 (0+ )
Therefore

2/23/02

page P14.36

R. A. DeCarlo, P. M. Lin

C v (0 ) + C2v C 2 (0 )
vC1 (0+ ) = v C2 (0 + ) = 1 C1
(C1 + C2 )
(b) Inserting values into out answer for part (a) yields
C1
vC1 (0+ ) = v C2 (0 + ) =
= 0.5 V
(C1 + C2 )
and the voltage remains the same for t > 0.
(c) Before the switch is closed, the energy in C2 is zero and the energy in C1 is the total stored energy:
2
Wtot (0 ) = WC1(0 ) = 0.5C1v C1
(0 ) = 0.5 J

After the switch is closed,

2
Wtot (0 + ) = WC1(0 + ) + WC 2 (0+ ) = 0.5C1vC1
(0 + ) + 0.5C2vC2 2 (0 + ) = 0.25 J
(d) (i) Using the series equivalent circuit for C1, we have
I(s) =

1
1 /R
1 2t R

i(t)
=
e
u(t) A
2 s s + 2 /R
R
R+
s

Thus
VC 2 (s) =
and
VC1(s) =

I(s)
1/ R
0.5
0.5
=
=

vC 2 (t) = 0.5 1 e 2t R u(t) V

s
s(s + 2 / R)
s
s + 2 /R

I(s) 1
1/ R
1 0.5
0.5
+ =
+ =
+
vC1 (t) = 0 . 5 1+ e 2t R u(t) V
s
s s(s + 2 / R) s
s
s + 2 /R

(ii) The energy dissipated in the resistor is given by

1 4t R
e 4t R
1
WR (0,) = R i ( )d = e
d =
= J
R
4
4
0
0
0

(iii) For all R,

1 2t R
e 2t R
1
Area under i(t) = i( )d = e
d =
=
R
2
2
0
0
0

1 2t R
e
u(t) has a decay that becomes infinitely fast and its magnitude (1/R)
R
approaches . Thus we have infinite height, zero-width, but a finite area of 0.5. Thus as R 0,
i(t) 0.5 (t) A. (We have avoided a more rigorous explanation as the above argument is more
plausible to sophomores.). As R 0, the exponential terms in the expressions for vC1 (t) and vC 2 (t)
have infinitely fast decays and hence disappear from the expressions yielding the stated result.
Further, as R 0, i(t) =

2/23/02

page P14.37

R. A. DeCarlo, P. M. Lin

SOLUTION TO 14.81.
(a)
Zin (s) =

2s + 4.5
1
=
+ 1
= Za (s) + Z b (s)
(s+ 0.5)(s + 4)
s+ 0.5
s+4
Ya (s) =

1 = s+ 0.5
Za (s)

Yb (s) =

1 =s+4
Zb (s)

From the above expressions, the RC circuit consists of a series connection of (a 1 farad capacitor in
parallel with a 2 resistor) and (a 1 F capacitor in parallel with a 0.25 resistor).
(b)
Yin (s) =

12s + 440
10 +
2
=
= Ya (s) + Yb (s)
(s+ 120)(s + 20)
s+ 120
s + 20
Za (s) =

1 = 0.1s + 12
Ya (s)

Zb (s) =

1 = 0.5s + 10
Yb (s)

From the above expressions, the RL circuit consists of a parallel connection of (0.1 H inductor in series
with a 12 resistor) and (a 0.5 H inductor in series with a 10 resistor).
(c) Following the hint, we have
Yin (s)
= 0.225s +0.075 = 0.1 + 0.125
s
(s+ 0.2)(s + 0.5)
s+ 0.2
s + 0.5
Hence
Yin (s) =

0.1s + 0.125s = Ya (s) + Yb (s)

s+ 0.2
s + 0.5

Za (s) =

1 = s + 0.2 = 10 + 1
Ya (s)
0.1s
0.5s

Zb (s) =

1 = s + 0.5 = 8 + 1
Yb (s)
0.125s
0.25s

From the above expressions, we see that each term in Yin (s) represents a series RC circuit. The RC
circuit for Yin(s) consists of a parallel connection of (a 0.5 F capacitor in series with a 10 resistor)
and (a 0.25 F capacitor in series with a 8 resistor).
(d) CORRECTION: for part (d),change the second term to 2s/(s2 + 2).

2/23/02

page P14.38

R. A. DeCarlo, P. M. Lin

Yin (s) = 0.5s + 2s = Ya (s) + Yb (s)

s2 + 1
s2 + 2
Za (s) =
Zb (s) =

1 = s 2 + 1 = 2s + 1
Ya (s)
0.5s
0.5s
1 = s 2 +2 = 0.5s + 1
s
Yb (s)
2s

From the above expressions, we see that each term in Yin (s) represents a series LC circuit. The LC
circuit for Yin(s) consists of a parallel connection of (a 0.5 F capacitor in series with a 2 H inductor)
and (a 1 F capacitor in series with a 0.5 H inductor).

SOLUTION 14.82. CORRECTIONS TO PROBLEM STATEMENT: (i) v0(t), should read vout(t)
and (ii) there should be a connection from the circuit inside the shaded box to the bottom line or
reference node.
(a) (i) 0 t < 1 ms. Since the capacitor voltage is initially zero and the switch is in position (a), a
simple source transformation yields a Norton equivalent (seen by the capacitor) consisting of a 20 mA
current source in parallel with 9.8039 k resistor. Hence
Vout (s) =

1
1
+ 106 s
98039

.02 20 103
=
s
s(s + 102)

Using MATLAB
n = .02*1e6;
p1 = 1e6/Rth
p1 = 102
d = [1 p1 0];
[r,p,k] = residue(n,d)
r=
-1.9608e+02
1.9608e+02
p=
-102
0
k = []

Hence, for 0 t < 1 ms, vout (t) = 196.08 1 e 102t u(t) V. It follows that vout (1 ms) = 19.014 V.
(ii) 1 ms t < 1.05 ms. The frequency domain equivalent circuit is given below.

2/23/02

page P14.39

R. A. DeCarlo, P. M. Lin

Writing a single node equation yields

200
1
Vout
Vout
s +
s + 10 6 sV
19.014 106 =
out
10000
10
1.2 10 5
Equivalently 19.014 +
= (s + 100100)Vout or
s

19.014s + 1.2 10
Vout =
s(s + 100100)

Therefore, vout (t') = 1.1988 + 17.815e 100100t' u(t') , and for 1 ms t < 1.05 ms,
vout (t) = vout (t 0.001) in which case

vout (t) = 1.1988 + 17.815e 100100(t 0.001) u(t 0.001)

Again using MATLAB,

(b) Part 1: for t > 0 up to t1 which denotes the time when vout(t) reaches 80 V, i.e., the capacitor is
charging. The frequency domain equivalent circuit is

2/23/02

page P14.40

R. A. DeCarlo, P. M. Lin

Using our knowledge of part (a), this circuit simplifies to

Hence
0.02
3
s = 5s + 20 10
Vout (s) =
1
s(s + 102)
10 6 s +
9803.9
5 10 6 +

and from MATLAB

syms s t
ilaplace((5*s+20e3)/(s^2+102*s))
ans =
10000/51-9745/51*exp(-102*t)
in which case
vout (t) = 196.08 191.08e 102t u(t) V

From this expression,

vout (t1 ) = 80 = 196.08 191.08e 102t1 V
and
t1 = log((80-10000/51)/(-9745/51))/(-102)
t1 = 4.8864e-03
This part of the problem considers t1 t < t2, i.e., the capacitor is discharging where vout(t2) = 5.
The equivalent frequency domain circuit is given below which is a slight modification of the circuit of
(a)-(ii):

2/23/02

page P14.41

R. A. DeCarlo, P. M. Lin

Making use of our knowledge of part (a)-(ii), we have

80s + 1.2 105

Vout =
s(s + 100100)

in which case vout (t') = 1.1988 + 78.801e 100100t ' u(t') V, and

vout (t) = 1.1988 + 78.801e 100100(t t1) u(t t1)

Here t'2 = 3.0286 10 5 s and t2 = t1 + t'2 = 4.9167 ms where t'2 is the duration of the discharge cycle.
As a final point, note that the frequency of the sawtooth is 1/t2 = 203.39 Hz. Finally a plot is given
below.

2/23/02

page P14.42

R. A. DeCarlo, P. M. Lin

SOLUTION 14.83. CORRECTION: In example 14.10, page 560, delete the four minus signs in the
equation for VC(s) and one more for vC(t).
We use MATLAB instead of SPICE to solve this problem. Applying voltage division to the circuit of
figure P14.83, we have
1
i (0-)
Cs
1
108
VC (s) =
LiL(0-) = L
=
1 + Ls + R
C s2 + R s + 1
s2 + 125s + 1.25 109
Cs
L
LC
From table 13.1, item 18
vC (t) = 2828e-62.5t sin(35,355t) u(t) V
A plot of vC(t) is given below with the vertical axis in V and the horizontal axis in seconds.
3000
2000
1000
0
-1000
-2000
-3000

0.2

0.4

0.6

0.8

1
x10-3

The waveform for the first few cycles is essentially the same as the example 14.10. Thus for the firs few
cycles, the lossless circuit of example 14.10 is a good approximation to the more accurate circuit model
of this problem. The effect of the presence of 100 resistance is a slow decay (with respect to ms
intervals) of the peak values.

1/25/02

P16-1

R. A. DeCarlo, P. M. Lin

PROBLEM SOLUTIONS
Solution 16.1.
(a) By the definition of the convolution integral

f 2(t) f 2 (t) =

f 2 (t ) f 2 ( )d =

2u(t

)2u( )d = 4 u(t )d

The integrand, u(t ) , is nonzero only when t. This suggests that there are two regions of
consideration: t < 0 and t 0.
Case 1: t < 0. Here u(t ) = 0 since is restricted to the interval [0,). Hence
f 2(t) f 2 (t) = 0 , for t < 0.

Case 2: t 0.

f 2(t) f 2 (t) = 4
In sum,

u(t )d = 4 d = 4t , for t 0.
0

0, t < 0
f 2(t) f 2 (t) =
4t, t 0

(b) By the definition of the convolution integral

f 2(t) f 3 (t) =

f 2 (t ) f 3( )d =

2u(t

)4e

u( )d = 8 e 2 u(t )d
0

The integrand, u(t ) , is nonzero only when t. This suggests that there are two cases to
consider: t < 0 and t 0.
Case 1: t < 0. Here u(t ) = 0 since is restricted to the interval [0,). Hence
f 2(t) f 3 (t) = 0, for t < 0.
Case 2: t 0.
t

f 2(t) f 3 (t) = 8 e 2 d = 4e 2
= 4(1 e 2t )
0
0

In sum,

t<0
0,
f 2(t) f 2 (t) =
2t
4(1 e ), t 0

(c) By the definition of the convolution integral and the sifting property of the delta function

f1(t) f 3(t) =

f1 (t ) f 3( )d =

= 20e

u( )

(t )4e 2 u( )d =

] =t = 20e2t u(t)

(d) By the definition of the convolution integral

1/25/02

P16-2

f 3(t) f 3(t) =

f 3 (t ) f 3( )d =

4e

2(t )

R. A. DeCarlo, P. M. Lin

u(t )4e

u( )d = 16e

2t

u(t

)d

The integrand, u(t ) , is nonzero only when t. This suggests that there are two cases to
consider: t < 0 and t 0.
Case 1: t < 0. Here u(t ) = 0 since is restricted to the interval [0,). Hence
f 3(t) f 3(t) = 0 , for t < 0.

Case 2: t 0.

f 3(t) f 3(t) = 16e

2t

= 16te 2t

In sum,

t<0
0,
f 3(t) f 3(t) =
2t
16te , t 0

(e) By the definition of the convolution integral and the sifting property of the delta function

f1(t + 2) f 2 (t + 4) =

f1(t + 2 ) f 2 ( + 4)d =

5 (t + 2 )2u( + 4)d =

= 10u( + 4)] =t +2 = 10u(t + 6)

(f) By the distributive property of convolution

f 2(t) [ f 2 (t) + f 3(t)] = f 2 (t) f 2 (t) + f 2 (t) f 3 (t)
Using the results of parts (a) and (b) the result follows immediately
t<0
0,
f 2(t) [ f 2 (t) + f 3(t)] =
2t
4(1+ t e ), t 0

Solution 16.2.
(a) By definition

f 3(t) =

f1( ) f 2(t )d =

K1u(

Here observe that u(t T2 ) = 0 for

t T2 < T1 and t T2 T1 .

T1)K2 u(t T2 )d = K1K2 u(t T2 )d

> t T2 .

T1

Hence there are two cases to consider:

Case 1: t T2 < T1 . Here u(t T2 ) = 0 , since is restricted to the interval [T1,) .

f 3(t) = 0
Case 2: t T2 T1 .

1/25/02

P16-3

R. A. DeCarlo, P. M. Lin

t T 2

f 3(t) = K1K2

d = K1K2 (t T2 T1 )

T1

In sum,

t < T1 + T2
0,
f 3(t) =
K1K 2 (t T2 T1), t T1 + T2

(b) By definition

f 3(t) =

f1(t ) f 2( )d =

K1u(t

+ T1 )K 2u( + T2 )d = K1K2

Here observe that u(t + T1) = 0 for

t + T1 < T2 and t + T1 T2 .

> t + T1.

Case 1: t + T1 < T2 . Here u(t + T1) = 0 , since

Case 2: t + T1 T2 .

u(t + T1 )d

Hence there are two cases to consider:

is restricted to the interval [T2 ,): f 3(t) = 0.

t +T 1

f 3(t) = K1K2
In sum,

T 2

T 2

d = K1K 2 (t + T1 + T2 )

t < T1 T2
0,
f 3(t) =
K1K 2 (t T2 T1), t T1 T2

(c) By definition

f 3(t) =

f1(t ) f 2( )d =

The integrand is nonzero only when

K1u(t

)K 2e

u( )d = K1K 2 e a u(t )d
0

t. Hence, there are two cases to consider: t < 0 and t 0.

Case 1: t < 0. f 3(t) = 0, for t < 0.

Case 2: t 0.
t

f 3(t) = K1K2 e
0

e a
d = K1K2
a

KK
= 1 2 (1 e at ).
a
0

Therefore
KK
f 3(t) = 1 2 (1 e at )u(t). for t 0.
a
(d) By definition

f 3(t) =

f1(t ) f 2( )d =

K1u(t

+ T1 )K 2e

u( )d = K1K 2 e a u(t + T1)d

0

1/25/02

P16-4

R. A. DeCarlo, P. M. Lin

t + T1 . Hence, there are two cases to consider: t + T1 < 0

The integrand is nonzero only when

and t + T1 0.

Case 1: t + T1 < 0. Here u(t + T1) = 0 , since

f 3(t) = 0.
Case 2: t + T1 0.

t +T 1

f 3(t) = K1K2

is restricted to the interval [0,).

e a
d = K1K 2
a

Therefore

t +T1

Hence

KK
= 1 2 1 e a(t +T 1) .
a

KK
f 3(t) = 1 2 1 e a(t +T1 ) u(t + T1 ).
a
(e) By definition

f 3(t) =

f1(t ) f 2( )d =

K1u(t +

)K2e

u( )d = K1K2 e a u(t + )d
0

The integrand is nonzero only when t. Hence, there are two cases to consider: t 0 and t > 0.
Case 1: t 0. Here u( t + ) = 1, since 0. Hence

f 3(t) = K1K2 e

Case 2: t > 0.

f 3(t) = K1K2 e

e a
d = K1K2
a

KK
= 1 2 , for t 0.
a
0

e a
KK
d = K1K2
= 1 2 e at , for t > 0.
a
a
t

In sum,
K1K 2
t0
a ,
f 3(t) = K K
1 2 e at , t > 0
a

Solution 16.3.
(a) By definition

f 3(t) =

f1( ) f 2(t )d =

K1e

u( )K 2e

u(t )d

= K1K2 e at u(t )d
0

The integrand, u(t ) , is nonzero only when

and t 0.
Case 1: t < 0. Here u(t ) = 0 , since
Case 2: t 0.

a(t )

t. Hence, there are two cases to consider: t < 0

0. Hence f 3(t) = 0, for t < 0.

1/25/02

P16-5

f 3(t) = K1K2e

at

R. A. DeCarlo, P. M. Lin

= K1K 2e at t , for t 0.

In sum,
Ic (s) =
(b) By definition

Cs

1 Iin (s)
Cs +
R
a=1
K =1

The integrand, u(t ) , is nonzero only when

and t 0.
Case 1: t < 0. Here u(t ) = 0 , since
Case 2: t 0.
f 3(t) = K1K2e

bt

t. Hence, there are two cases to consider: t < 0

0. Hence f 3(t) = 0, for t < 0.

t
= 1 e (ba)t 1
b a

e (ba) d
0

Therefore, for t 0,

K1K 2e bt tu(t)

f 3(t) = K1K 2 at
e bt
b a e

if

a= b

if

a b

if a = b
if a b

(c) By replacing K1 = 50, K 2 = 20 and a = 10 in the formula of f 3(t) in part (a) the answer for
part (i) is easily obtained as
t<0
0,
f 3(t) =
10t
t, t 0
1000e
t

For part (ii) the parameters have the following values: vc (t) =

] = 1,

t
e (t ) d = e t

K 2 = 0.2, a = 10 and b = 0.2. Using these values in the formula developed in part (b) for f 3(t) the
answer follows immediately
f 3(t) = 0.102 e 0.2t e 10t u(t)

SOLUTION 16.4.
(a) Using the impulse response theorem and the definition of the convolution integral the response
of the system, y(t), can be computed as follows

y(t) =

h(t )v( )d =

2e 2(t ) u(t )[u(

1) u( 3)]d

Observing that u( 1) u( 3) is nonzero only when 1 < 3 yields

1/25/02

P16-6

R. A. DeCarlo, P. M. Lin

y(t) = 2 e 2( t) u(t )d
1

The integrand in the above equation is nonzero only when t. This suggests three regions of
consideration: t < 1, 1 t 3, and 3 < t.
Case 1: t < 1. Here u(t ) = 0 , since is restricted to the interval [1,3]. Hence y(t) = 0, for t < 1.
Case 2: 1 t 3.
t

]1

t
y(t) = 2 e 2( t) d = e 2( t) = 1 e 2(1t ) , for 1 t 3
1

Case 3: 3 < t.

]1

3
y(t) = 2 e 2( t) d = e 2( t) = e 2(1t ) (e 4 1), for 3 < t
1

In sum,
0,
t<1

2(1t )
y(t) = 1 e
,
1 t 3
e 2(1t ) (e 4 1), 3 < t

A picture of y(t) is sketched in the next figure.

(b) Using the impulse response theorem and the definition of the convolution integral the response
of the system, y(t), can be computed as follows

y(t) =

h(t )v( )d =

2e 2(t ) u(t

)u(2 t + )[ u( 1) u( 3)]d

Observing that u( 1) u( 3) is nonzero only when 1 < 3 yields

3

y(t) = 2 e 2( t) u(t )u(2 t + )d

1

The integrand in the above equation is nonzero only when t 2 t. This suggests four regions
of consideration: t < 1, 1 t 3, 3 < t 5and 5 < t.
Case 1: t < 1. Here u(t ) = 0 , since is restricted to the interval [1,3]. Hence y(t) = 0, for t < 1.
Case 2: 1 t 3. Here u(t )u(2 t + ) is nonzero only when 1 t. Therefore,

1/25/02

P16-7

R. A. DeCarlo, P. M. Lin

]1

t
y(t) = 2 e 2( t) d = e 2( t) = 1 e 2(1t ) , for 1 t 3.
1

Case 3: 3 < t 5. Here u(t )u(2 t + ) is nonzero only when t 2 3. Therefore,

3

y(t) = 2

t 2

]t 2 = e2(3t) e4 , for 3 < t 5.

3
e 2( t) d = e 2( t)

Case 4: 5 < t. Here u(t )u(2 t + ) = 0 , since is restricted to the interval [1,3]. Therefore,
y(t) = 0, for 5 < t.
A picture of y(t) is sketched in the next figure.

(c) By the impulse response theorem, the zero-state response of the circuit y(t) is
y(t) = h(t) v(t)
Using the definition of the convolution integral and the sifting property of delta function it follows
that
0, t < 0
2, 0 t < 1

2, 1 t < 2
y(t) = 2h(t) 2h(t 1) + h(t 2) =
0, 2 t < 3
1,
3 t < 4

0, 4 t
Using the waveform of h(t) given in figure P16.4, y(t) is sketched in the next picture.

1/25/02

P16-8

SOLUTION 16.5.
(a) By definition

f 4 (t) =

R. A. DeCarlo, P. M. Lin

f1(t ) f 2( )d =

(t 2) 2 u( + 1)d

By the sifting property of delta function it follows that f 4 (t) = 2u( + 1)] =t 2 = 2u(t 1).
(b) By the definition of convolution and the sifting property of delta function

f 5(t) =

f1(t ) f 3( )d =

=e

u( )

(c) By definition

f 6(t) =

(t 2)e 2 u( )d

2(t 2)

] =t 2 = e

u(t 2)

f 2 (t ) f 3( )d =

2u(t

+ 1)e

u( )d = 2 e 2 u(t + 1)d
0

The integrand in the above equation is nonzero only when t + 1. This suggests two regions of
consideration: t < 1and 1 t .
Case 1: t < 1. Here u(t + 1)= 0, since 0 . Hence f 6(t) = 0 , for t < 1.
Case 2: 1 t . Here u(t + 1) is nonzero only when t + 1. Therefore,
t +1

f 6(t) = 2

e 2 d = e 2

It follows that

]0

t +1

= 1 e 2(t +1)

f 6(t) = 1 e 2(t +1) u(t + 1)

(d) By the definition of convolution and the sifting property of delta function we have

1/25/02

P16-9

f 7(t) =

f1 (t ) f 3( 2)d =
=e

2( 2)

SOLUTION 16.6.
(a) By definition

u( 2)

R. A. DeCarlo, P. M. Lin

(t 2)e 2( 2) u( 2)d =

] =t 2 = e2(t 4)u(t 4) .

f 6(t) = 1 e 2(t +1) u(t + 1)

Here observe that u(t ) = 0 for > t. Hence, there are two cases to consider: t 0 and t > 0.
Case 1: t 0

t
e a
e at
a
y(t ) = K e d = K
K
=
a
a

Case 2: t > 0

y(t) = K

e a d = K

(b) By definition
y(t) = K

e a d = K

u( t)e

e a
K
=
a a

u()d = K u()e

Here observe that u( t) = 0 for < t; hence the lower limit of integration is t. Also, because of the
presence of u() in the integrand, there are two cases to consider: t < 0 and t 0.
Case 1: t < 0

e
y(t) = K e a d = K
a
0

Case 2: t 0

y(t) = K e a d = K
t

K
=
a
0

e a
K at
= e
a t
a

SOLUTION 16.7.
(a) Using the definition of the convolution integral and the sifting property of delta function,
f 5(t)can be computed as below

f 5(t) =

f 2 (t ) f 4 ( )d =
=e

a(t )

u(t )

e a(t ) u(t ) ( 4)d =

] =4 = ea(t 4)u(t 4)

A picture of f 5(t), for a = 1, is sketched in the next figure.

1/25/02

P16-10

(b) By definition

f 6(t) =

R. A. DeCarlo, P. M. Lin

f1 (t ) f1( )d = K

u(t )u( )d = K

u(t

)d

Since u(t ) is nonzero only when t, there are two regions of consideration: t < 0and 0 t.
Case 1: t < 0. Here u(t ) = 0 , since 0. Hence
f 6(t) = 0 , for t < 0.
Case 2: 0 t.
t

f 6(t) = K

= K 2t , for 0 t.

A picture of f 6(t) , for K = 1, is sketched in the next figure.

(c) By definition
f 7(t) =

f1 (t ) f 2 ( )d =

Ku(t

)e

u( )d = K e a u(t )d
0

Since u(t ) is nonzero only when t, there are two regions of consideration: t < 0and 0 t.
Case 1: t < 0. Here u(t ) = 0 , since 0. Therefore, f 7(t) = 0 , for t < 0.
Case 2: 0 t.
t

f 7(t) = K e a d =
0

K a t K
e
= 1 e at , for 0 t.
a
0
a

1/25/02

P16-11

R. A. DeCarlo, P. M. Lin

A picture of f 7(t) , for t < 0 and a = 1, is sketched in the next figure.

(d) By definition
f8 (t) =

f1(t ) f 3( )d =

Ku(t

Case 2: 0 < t.

ea

u(t )d =

ea

u( )d = K

e a u(t )d

t . This suggests two regions of consideration:

K a t
K
e
= e at , for t 0.
a

f8 (t) = K

)e

The integrand, u (t ) , is nonzero only when

t 0 and 0 < t .
Case 1: t 0 .
f8 (t) = K

u(t )d =

K a 0
K
e
= , for 0 < t.
a

In sum,
K at
a e , t 0
f8 (t) = K
,
0< t
a
A picture of f8 (t), for K = 1 and a = 1, is sketched in the next figure.

1/25/02

P16-12

R. A. DeCarlo, P. M. Lin

SOLUTION 16.8.
(a) Using the current division formula
Ic (s) =

Cs

1 Iin (s)
R
By Ohms law the Laplace transform of capacitors voltage
1
Vc (s) =
I (s)
Cs c
Therefore the transfer function of the circuit
Cs +

V (s)
1
1
H(s) = c
=
=
1
Iin (s) Cs +
s+ 4
R
Taking the inverse Laplace transform of H(s) yields the impulse response h(t) = e 4t u(t) .
(b) By the impulse response theorem

vc (t) = iin (t) h(t) =

iin (t )h( )d =

3e (t ) u(t )e 4 u( )d =

= 3 e (t + 3 ) u(t )d
0

The integrand is nonzero only when t. Therefore there are two regions of consideration:
t < 0and 0 t.
Case 1: t < 0. Here u(t ) = 0 , since 0 . Hence vc (t) = 0 , for t < 0.
Case 2: 0 t.

1/25/02

P16-13

R. A. DeCarlo, P. M. Lin

]0

t
vc (t) = 3 e (t + 3 ) d = e (t + 3 ) = e t e 4t , for 0 t.
0

In sum,

vc (t) = e t e 4t u(t)V.

SOLUTION 16.9.
(a) By voltage division formula
Vout (s) =

1
Cs
R+

1 Vin (s)
Cs

Therefore the transfer function

1
Cs

V (s)
1
H(s) = out
=
=
1
Vin (s) R +
s +1
Cs
Taking the inverse Laplace transform of H(s) yields yields the impulse response h(t) = e t u(t).
By the impulse response theorem and the convolution definition

vout (t) =

h(t )vin ( )d =
0

(t )

e (t ) u(t ) u( ) + 2e 2 u( ) d =

u(t )d + 2 e (t + ) u(t )d
0

For both integrals the integrand is nonzero only when t. This suggests two regions of
consideration: t < 0and 0 t.
Case 1: t < 0. Here the second integral is zero since, for this integral, is restricted to [0,).
t

vc (t) =
Case 2: 0 t.

] = 1, for t < 0.

t
e (t ) d = e t

vc (t) =

]0

(t )

d + 2 e (t + ) d
0

0
t
= e t
2e (t ) = e t 2 e 2t e t = 3e t 2e 2t , for 0 t

(c) By the impulse response theorem and the definition of convolution

1/25/02

P16-14

vout (t) =
0

=e

h(t )vin ( )d =

R. A. DeCarlo, P. M. Lin

e (t ) u(t )e a| |d

(t )

e (t ) u(t )e a d + e

(a+1)

u(t )d + e

u(t )e a d

(1a)

u(t )d

For both integrals the integrand is nonzero only when t. This suggests two regions of
consideration: t < 0and 0 t.
Case 1: t < 0. Here the second integral is zero since, for this integral, is restricted to [0,).
vout (t) = e
Case 2: 0 t.

(a+1)

d =e

t e

vout (t) = e

(a+1) t

a + 1

(a+1)

d +e

e at
=
, for t < 0.
a +1

(1a)

d =

0
t

0
t
t
e (a+1)
e
t
(1a)
t
=e
+e e
d =
+ e e (1a) d
a + 1
a +1
t

Here observe that a + 1 is nonzero since a > 0.

For computing the second integral, in case 2, we need to distinguish two subcases: a = 1
and a 1.
t
e t t
if a = 1
t
(1a)
e e
d = 1
at
t
if a 1
1 a e e
0
Therefore, for 0 t,
e t
+ e t t
if a = 1

a
+
1
vout (t) = t
e + 1 e at e t if a 1
a + 1 1 a

SOLUTION 16.10.
(a) The impulse response is obtained by taking the inverse Laplace transform of the transfer
function
h(t) = 2e 0.2t u(t)
By the impulse response theorem the response y(t)equals

1/25/02

P16-15

R. A. DeCarlo, P. M. Lin

y(t) = h(t) v(t) =

h(t )v( )d

Substituting v(t) = u(t + 1) u(t 1) in the above integral yields

y(t) =

h(t )[u( + 1) u( 1)]d

Here observe that u( + 1) u( 1) is nonzero only when 1 1. Hence

1

y(t) =

h(t

The integrand is nonzero only when

1 t < 1 and 1 t.
Case 1: t < 1. y(t) = 0
Case 2: 1 t < 1.
t

)d = 2 e 0.2( t ) u(t )d
1

t. This suggests three regions of consideration: t < 1,

y(t) = 2 e 0.2( t ) d = 10 e 0.2(t +1) 1 , for 1 t < 1.

Case 3: 1 t.

1
1

y(t) = 2 e 0.2( t ) d = 10e 0.2t e 0.2 e 0.2 , for 1 t.

1

(b) The transfer function of the leaky integrator (see equation 14.14 in the textbook) is given by
1
R1
H(s) =
1
Cs +
R2

where R2 is the leakage resistance of the capacitor C and R1 is the resistance connected at the
inverting input of the op amp. Equating the two expressions of H(s) we obtain that
1

R1
2
=
1
s + 0.2
Cs +
R2
Matching the coefficients and taking into account that the smallest resistor is 10k the following
values are obtained: R1 = 10k, R2 = 100k and C = 5105 F .
(c) The impulse response is obtained by taking the inverse Laplace transform of the transfer
function
h(t) = Ke at u(t)
By the impulse response theorem the response y(t)equals

1/25/02

P16-16

R. A. DeCarlo, P. M. Lin

y(t) = h(t) v(t) =

h(t )v( )d

Substituting v(t) = u(t + T ) u(t T ) in the above integral yields

y(t) =

h(t )[u( + T ) u( T )]d

Here observe that u( + T ) u( T ) is nonzero only when T T . Hence

T

y(t) =

h(t

)d = K

y(t) = K
Case 3: T t.

In sum,

u(t )d

K
e a(t ) d = 1 e a(t +T ) , for T t < T .
a

y(t) = K

t. This suggests three regions of consideration: t < T ,

The integrand is nonzero only when

T t < T and T t.
Case 1: t < T . y(t) = 0
Case 2: T t < T .
t

e a(t

K
e a(t ) d = e at e aT e aT , for T t.
a

0,
t < T
K
y(t) = 1 e a(t +T ) ,
T t < T
a
K e at e aT e aT , T t
a

SOLUTION 16.11.
(a) First observe, from figure P16.11(a), that
f 2(t) = (2t + 4)[u(t) u(t 2)]
By definition

f 3(t) =

f1(t ) f 2( )d =

4u(t )(2 + 4)[ u( ) u( 2)]d = 8 ( 2)u(t )d

0

The integrand is nonzero only when t. This suggests three regions of consideration: t < 0,
0 t < 2 and 2 t.
Case 1: t < 0. Here u(t ) = 0 due to the fact that is restricted to the interval [0,2]. Hence
f 3(t) = 0, for t < 0.

1/25/02

P16-17

R. A. DeCarlo, P. M. Lin

Case 2: 0 t < 2.
t
2

f 3(t) = 8 ( 2)d = 8 2 = 4 t 2 4t , for 0 t < 2.

2

0
0
t

Case 3: 2 t.

2
2

f 3(t) = 8 ( 2)d = 8 2 = 16, for 2 t.

2

0
0
2

In sum,
0,
t<0

2
f 3(t) = 4 t 4t , 0 t < 2
16,
2 t

A picture of f 3(t) is sketched in the next figure.

(b) First observe, from figure P16.11(b), that

f 2(t) = t[u(t) u(t 2)] + (4 t)[u(t 2) u(t 4)]

By definition

f 3(t) =

f1(t ) f 2( )d

By replacing f1(t) and f 2(t) with their expressions we have

f 3(t) =

4u(t ){ [u( ) u( 2)] + (4 )[u( 2) u( 4)]}d =

4u(t ) [u( ) u( 2)]d +

4u(t )(4 )[u( 2) u( 4) ]d =

= 4 u(t )d + 4 (4 )u(t )d

1/25/02

P16-18

R. A. DeCarlo, P. M. Lin

The integrands are nonzero only when t. This suggests four regions of consideration: t < 0,
0 t < 2, 2 t < 4, and 4 t.
Case 1: t < 0. Here u(t ) = 0 due to the fact that is restricted to the interval [0,4]. Hence
f 3(t) = 0, for t < 0.
Case 2: 0 t < 2. here observe that the second integral is zero since, for this integral,
to the interval [2,4]. Therefore

is restricted

f 3(t) = 4 d = 2t 2 , for 0 t < 2.

0

Case 3: 2 t < 4.

f 3(t) = 4 d + 4 (4 )d = 8 + 4 4 = 2t 2 + 16t 16, for 2 t < 4.

2

0
2
2
Case 4: 4 t.
2

f 3(t) = 4 d + 4 (4 )d = 8 + 4 4 = 16, for 4 t.

2

0
2
2
2

In sum,
0,
2t 2,

f 3(t) = 2
2t + 16t 16,
16,
A picture of f 3(t) is sketched in the next figure.

SOLUTION 16.12. (a) By voltage division,

t<0
0 t < 2
2 t < 4
4t

1/25/02

P16-19

R. A. DeCarlo, P. M. Lin

2
Vout
1
2/3
2/3
s
H(s) =
=
=
=

2
2 s + 2.5s + 1 (s + 0.5) (s + 2)
Vin
2s + 5 +
s
Hence, the impulse response is
2
2
h(t) = e 0.5tu(t) e 2tu(t)
3
3
(b) By definition
h(t) * vin (t) =

h(t )vin () d =

20
3

e 0.5(t )u(t ) e u() d

20

2(t )

u(t ) e u() d

Case 1: t 0.

20
vout (t) =
3

20
e d
3

0.5(t )

20
e d =
3

2(t )

0.5t 1.5

20
d
3

20e 0.5t 1.5 t

20e2t 3 t
=
e

e
= 4.444e t 2.222et = 2.222e t

4.5
9

[ ]

[ ]

Case 2: t > 0.
vout (t) =

20e 0.5t 1.5 0

20e 2t 3 0
20e 0.5t 20e 2t
e

e
=

4.5
9
4.5
9

[ ]

[ ]

SOLUTION 16.13.
(a) The impulse response of the circuit has been computed in problem 16.12
h(t) =

2 0.5t
2
e
(t) e 2t u(t)
3
3

By the impulse response theorem and the convolution definition

vout (t) = h(t) vin (t) =

h(t )vin ( )d =

2t 3

e d

1/25/02

P16-20

20
=
3

[e 0.5(t

R. A. DeCarlo, P. M. Lin

e 2(t ) u(t )e | |d

The integrand is nonzero only when

t. Hence
t

[e 0.5(t

20
vout (t) =
3

e 2(t ) e | |d

The existence of the function e | | under the integral suggests two regions of consideration: t 0
and 0 < t.
Case 1: t 0.
t

[e 0.5(t

20
vout (t) =
3
20 0.5t
=
e
3

e 2(t ) e d =

20
d e 2t
3

1.5

[ ]

e3 d =

[ ]

t
20 0.5t 1.5 t
20
e
e
e 2t e 3
=
4.5

= 4.444e t 2.222e t = 2.222e t , for t 0.

=
Case 2 : 0 < t.
20
vout (t) =
3
20
=
3

[e

[e

0.5(t )

0.5t +1.5

2(t )

2t + 3

20
e d +
e 0.5(t ) e 2(t ) e d =

20
d +
e 0.5t 0.5 e 2t + d =
3
0

20 e 0.5t +1.5
e 2t + 3
20 e 0.5t 0.5
=

+
e 2t + =
3
1.5
3
3 0.5
0

= 17.778e 0.5t 20e t + 4.444e 2t , for 0 < t.

In sum,
2.222e t ,
0 t
vout (t) =
0.5t
t
2t
17.778e
20e + 4.444e , 0 < t

SOLUTION 16.14.
(a) The impulse response of the circuit is obtained by taking the inverse Laplace transform of H(s)
h(t) = 2e t 2e 2t + 4e 4t u(t)

(b) The result follows from the following MATLAB code:

1/25/02

P16-21

R. A. DeCarlo, P. M. Lin

>> p = [-1,-2,-4];
>> r = [2,-2,4];
>> k = 0;
>> [n,d] = residue(r,p,k)
n=
4

14

16

14

d=
1

Therefore,
4s2 + 14s + 16
H(s) = 3
s + 7s2 + 14s + 8
(c) By the impulse response theorem

y(t) = u(t) h(t) = u(t)* 2e t u(t) 2e 2t u(t) + 4e 4t u(t)

Using the distributive property of convolution we have

y(t) = u(t) 2e t u(t) + u(t) 2e 2t u(t) + u(t) 4e 4t u(t)

In problem P16.2(c) the following equation has been obtained

[K1u(t)] [K 2e at u(t)] = K1aK2 (1 e at )u(t)

Using the above equation y(t)is immediately obtained

y(t) = 2 1 e t u(t) 1 e 2t u(t) + 1 e 4t u(t) =

= 2 2e t + e 2t e 4t u(t)
(d) By the impulse response theorem

y(t) = f (t) h(t) = [8u( t) 8u( t T )] * 2e t u(t) 2e 2t u(t) + 4e 4t u(t)

Using the distributive property of convolution we have

y(t) = 8u( t) 2e t u(t) 2e 2t u(t) + 4e 4t u(t)

We denote

8u(t T )* 2e u(t) 2e 2t u(t) + 4e 4t u(t)

1/25/02

P16-22

R. A. DeCarlo, P. M. Lin

y1(t) = 8u(t) 2e t u(t) 2e 2t u(t) + 4e 4t u(t)

By the time invariance property it follows that

y(t) = y1(t) y1(t T )
In order to compute y1(t) we will use the following equation which has been obtained in problem
16.2(e)
K1K 2
t0
a ,
at
[K1u(t)] K 2e u(t) = K1K 2 at

e , t>0
a
Therefore,
t0
16,
y1(t) = t
2t
4t
+ 8e , 0 < t
16e 8e

The zero-state response to the input f (t) can now be computed

0,
t0
t
2t
4t
y(t) = 16e 8e + 8e
16,
0< tT
16 e t e (t T ) 8 e 2t e 2(t T ) + 8 e 4t e 4(t T ) , T < t

] [

] [

SOLUTION 16.15.
(a) Using the convolution theorem the transfer function of the cascade is
H(s) = L [h(t)] = L[ h1(t) h2 (t) h3(t)] = L[ h1(t)] L[ h2(t)] L[ h3(t)] = H1 (s) H 2 (s) H 3(s)
From table 13.1
1
s
10
H2 (s) =
s+2
2
H3(s) = 2
s
H1(s) =

Therefore,
20
H(s) = 3
s (s + 2)
A partial fraction expansion of H(s) can be obtained using the residue command in MATLAB:
>> num = [20];
>> den = [1 2 0 0 0];
>> [r,p,k] = residue(num,den)

1/25/02

P16-23

R. A. DeCarlo, P. M. Lin

r=
-2.5000
2.5000
-5.0000
10.0000
p=
-2
0
0
0
k=
[]
Hence
H(s) =

2.5 2.5 5 10
+
+ 2 + 3
s+2
s
s
s

Taking the inverse Laplace transform yields the impulse response of the cascade
h(t) = 2.5e 2t u(t) + 2.5u(t) 5tu(t) + 5t 2 u(t)
(b) By the impulse response theorem and the convolution theorem, the Laplace transform of the
step response of the cascade equals
20
1
20
Y(s) = H (s)U (s) = 3
= 4
s (s + 2) s s (s + 2)
A partial fraction expansion of H(s) can be obtained using the residue command in MATLAB:
>> num = [20];
>> den = [1 2 0 0 0 0];
>> [r,p,k] = residue(num,den)
r=
1.2500
-1.2500
2.5000
-5.0000
10.0000
p=
-2
0
0

1/25/02

P16-24

R. A. DeCarlo, P. M. Lin

0
0
k=
[]
Hence
Y(s) =

1.25 1.25 2.5 5 10

+
+ 2 + 3 + 4
s+2
s
s
s
s

Taking the inverse Laplace transform yields the step response of the cascade
y(t) = 1.25e 2t u(t) 1.25u(t) + 2.5tu(t) 2.5t 2 u(t) + 1.667t 3u(t) .

SOLUTION 16.16.
(a) By the voltage division formula
Vout (s) =

1
Cs

1
V
(s)
=
Vin (s)
i
1
CRs
+
1
R+
Cs

Therefore, the transfer function of the circuit is

V (s)
1
1
H(s) = out
=
=
Vin (s) CRs + 1 s + 1
Taking the inverse Laplace transform yields the impulse response h(t) = e t u(t).
(b) From table 13.1 the Laplace transform of vin (t) is
Vin (s) =

1
1
+
s (s + 1)2

By the impulse response theorem and the convolution theorem it follows that
Vout (s) = H (s)Vin (s) =

1 1
1
1
1
=
+
+

s + 1 s (s + 1)2 s(s + 1) (s + 1)3

A partial fraction expansion of Vout (s) is

1
1
1
Vout (s) =
+
s s + 1 (s + 1)3
Taking the inverse Laplace transform yields

1/25/02

P16-25

R. A. DeCarlo, P. M. Lin

vout (t) = u(t) e t u(t) 0.5t 2e t u(t).

Using the time domain convolution method vout (t) can be computed as follows

vout (t) =

h(t )vin ( )d =

e (t ) u(t ) u( ) + e u( ) d

From the experience earned by computing convolution integrals we know that the computation of the
above integral requires more computational work than the Laplace transform method. More
computations imply, of course, more sources of errors.
From the solution of this problem we have seen that, in the case of the Laplace transform method,
the computational burden consists in computing Laplace and inverse Laplace transforms. For a
large class of functions these transforms can be found in tables(for example table 13.1). The only
computation that we did, in the solution of this problem, was the partial fraction expansion of
Vout (s) .
(c) In this case the (unilateral) transform method cannot be used because vin (t) 0 for t < 0.

SOLUTION 16.17.
(a) The impulse response can be obtained by taking the inverse Laplace transform of H(s) .
Therefore
h(t) = 8e 10t u(t)
(b) From table 13.1
8
Vin (s) = 2
s + 16
By the impulse response theorem and the convolution theorem it follows that
Vout (s) = H (s)Vin (s) =

64
(s + 10)(s2 + 16)

Vout (s) can be further written as

Vout (s) =

0.5517s + 5.517 0.5517

s
4
0.5517
+
= 0.5517 2
+ 1.379 2
+
2
s + 10
s + 16
s + 16
s + 16 s + 10

The above expansion of Vout (s) can be obtained by using the technique of example 13.14, page
514. Taking the inverse Laplace transform yields
vout (t) = 0.5517cos(4t)u(t) + 1.379sin(4 t)u(t) + 0.5517e 10t u(t).

1/25/02

P16-26

(c) In this case

R. A. DeCarlo, P. M. Lin

8
(s + 2)2 + 16

Vin (s) = L 2e 2t sin(4t)u(t) =

Therefore
Vout (s) = H (s)Vin (s) =

[(s + 2)

64
2

+ 16 (s + 10)

Using again the technique of example 13.14, page 514, Vout (s) can be written as
Vout (s) =

0.8s + 4.8
0.8
s+2
4
0.8
+
= 0.8
+ 1.6
+
2
2
2
(s + 2) + 16 s + 10
(s + 2) + 16
(s + 2) + 16 s + 10

Taking the inverse Laplace transform yields

vout (t) = 0.8e 2t cos(4t)u(t) + 1.6e 2t sin(4 t)u(t) + 0.8e 10t u(t) V.
In this context the Laplace transform method is faster than the time domain convolution. This is
due to the fact that vin (t) has a relatively complicated expression and the convolution integral will
require more computational work than the Laplace transform method.
(d) In this context the Laplace transform method cannot be used because
time domain convolution method will be used to compute the response
response thorem
vout (t) = h(t) vin (t) = 8e 10t u(t) [u( t)]

vin (t) 0 for t < 0. The

vout (t). By the impulse

Using the result of problem 16.2, part (e), it follows that

t0
0.8,
vout (t) =
10t
, 0<t
0.8e

SOLUTION 16.18.
(a) Replacing R1, R2 , C1 and C2 with their values the transfer function can be obtained
s
H(s) = 2
s +5s + 2
The only zero of H(s) is 0 and the poles of H(s) are 0.5 and 2. A partial fraction expansion of
H(s) is:
0.167 0.667
H(s) =
+
s + 0.5 s + 2
The impulse response can be obtained by taking the inverse Laplace transform of H(s)
h(t) = 0.167e 0.5t + 0.667e 2t

1/25/02

P16-27

R. A. DeCarlo, P. M. Lin

(b) vout (t) will be computed using the Laplace transform method. This approach is valid because
h(t) and vin (t) are zero for t < 0.
From table 13.1 the Lapace transform of vin (t) is
1
Vin (s) =
(s + 2)2
By the impulse response theorem and the convolution theorem it follows that
Vout (s) = H (s)V (s) =

2s2

s
1
s
= 4
2
3
+ 5s + 2 (s + 2)
2s + 13s + 30s2 + 28s + 8

A partial fraction expansion of Vout (s) can be obtained using the residue command in MATLAB:
>> a = [1 0];
>> b = [2 13 30 28 8];
>> [r,p,k] = residue(a,b)
r=
0.0741
0.1111
0.6667
-0.0741
p=
-2.0000
-2.0000
-2.0000
-0.5000
k=
[]
Therefore,
Vout (s) =

0.0741 0.0741 0.1111 0.6667

+
+
+
s + 0.5
s+2
(s + 2) 2 (s + 2)3

Taking the inverse Laplace transform yields

vout (t) = 0.0741e 0.5t + 0.0741e 2t + 0.1111te2t + 0.3333t 2e 2t u(t) V.

One would prefer the time domain convolution method to compute vout (t), but the computations
may require more work relatively to the Laplace transform method.
(c) In this part vin (t) 0 for t < 0. Therefore the time domain convolution method will be used to
compute vout (t).
By the impulse response theorem

vout (t) = h(t) vin (t) =

[0.1667e 0.5(t

+ 0.6667e 2(t ) u(t )e 2 u( )d =

1/25/02

P16-28

[0.1667e 0.5(t

R. A. DeCarlo, P. M. Lin

+ 0.6667e 2(t ) u(t )e 2 d

t. This suggests two regions of

The integrand of the previous integral is nonzero only when

consideration: t < 0 and 0 t.
Case 1: t < 0.
vout (t) = 0.1667e

Case 2: 0 t.

0.5t

2.5

d + 0.6667e

2.5 t

2t

e4 d =

4 t

e
e
= 0.1667e 0.5t
+ 0.6667e 2t
=
2.5
4
= 0.1e 2t , for t < 0.
vout (t) = 0.1667e

0.5t

2.5

d + 0.6667e

2.5 0

2t

e4 d =

4 0

e
e
= 0.1667e 0.5t
+ 0.6667e 2t
=
2.5
4
= 0.1667e 0.5t + 0.6667e 2t , for 0 t.

SOLUTION 16.19.
Replacing Rand C with their values
H(s) =

s5
10
= 1
s+5
s+5

The zero-state response vout (t) will be computed using the time domain convolution method
because vin (t) 0 for t < 0.
The impulse response of the circuit is
h(t) = (t) 10e 5t u(t)
By the impulse response theorem

vout (t) = h(t) vin (t) =

h( )vin (t )d =

( ) 10e 5 u( )]cos[10(t
[

= 10

= 10

( )cos(t )d 10

10e 5 u(

)]d =

)cos[10(t )]d

Using the sifting property of the delta function and expanding cos(t )it follows that

1/25/02

P16-29

vout (t) = 10cos(t ) 100cos(10t) e

0

R. A. DeCarlo, P. M. Lin

cos(10 )d 100sin(10t) e 5 sin(10 )d

0

Using the definition of the Laplace transform we observe that

e 5
0

and

e 5
0

Therefore

cos(10 )d = L [cos(10t)u(t)]s=5 = 2
= 0.04

s + 100 s =5
10
sin(10 )d = L[sin(10t)u(t)]s=5 = 2
= 0.08

s + 100 s =5
vout (t) = 10cos(10t) 4cos(10t) 8sin(10t)

= 6cos(t) 8sin(10t) = 10cos 10t + tan1 ( 43 )

Notice that vout (t) and vin (t) have the same frequency and magnitude.

SOLUTION 16.20.
(a) First notice that vin (t T) = u(t). Therefore w(t) = u(t) and
1
W (s) =
s
(b) Using the properties of the Laplace transform it follows that
w
Vout
(s) = H (s)W (s) =

2
s(s + 2)

w
A partial fraction expansion of Vout
(s) is

1
1
w
Vout
(s) =
s s+2
Taking the inverse Laplace transform yields
w
vout
(t) = u(t) e 2t u(t)
(b) Since
vin (t) = w(t + T )
it follows, by the time invariance property, that

Therefore,

v
w
vout
(t) = vout
(t + T ).
v
vout
(t) = u(t + T ) e 2(t +T ) u(t + T ) V.

SOLUTION 16.21.
(a) First observe from figure P16.21 that
vin (t) = u(t + T ) u(t T )
From the definition of w(t) it follows that

1/25/02

P16-30

R. A. DeCarlo, P. M. Lin

w(t) = vin (t T ) = u(t) u(t 2T )

Therefore

1 1 2sT
e
s s
(a) By the impulse response theorem and the convolution theorem it follows that
W (s) =

2
1
1
1 e 2sT =
1 e 2sT =

s(s + 2)
s s+2
1
1
1 2sT
1
=

e
s s+2
s s + 2

w
Vout
(s) = H (s)W (s) =

Taking the inverse Laplace transform and using the time shift property of the Laplace transform
yields
w
vout
(t) = 1 e 2t u(t) 1 e 2(t 2T ) u(t 2T )
Because
vin (t) = w(t + T )

it follows, by the time invariance property, that

v
w
vout
(t) = vout
(t + T ) =

= 1 e 2(t +T ) u(t + T ) 1 e 2(t T ) u(t T ) V.

SOLUTION 16.22. (a) The use of t = t + T1 in the problem statement means replace t by t + T1 .
However, strictly speaking we should have used a statement of the form t = t' +T1 which is done in
the proof below. By definition of the convolution and the property of commutivity,

f (t T1) * g(t) =

f
(t

)g()
d
=
f
(t'
)g()
d
= [ f (t' ) * g(t' )]t' =t T
1

t' =t T1

Observe that t = t' +T1 . Hence

[ f (t T1 )* g(t)]t =t' +T1 = f (t' ) * g(t' )

Realizing that t and t' are simply dummy variables, we immediately obtain the result. From a
systems perspective, this corresponds to the property of time-invariance where a shift of an input
function by T1 yields a corresponding shift of the output function by T1 .
(b) The steps in this part are similar to those of part (a).

1/25/02

P16-31

f (t T1) * g(t T2 ) =

R. A. DeCarlo, P. M. Lin

f (t T1 )g( T2 ) d

Let = T2 in which case = + T2 and d = d . Hence

f (t T1) * g(t T2 ) = f (t T1 T2 )g()d = f (t' )g( )d

t' =t T1 T2

= [ f (t' )* g(t') ]t' =t T T

1

Since t = t' +T1 + T2 , and t and t' are dummy variables, we have

[ f (t T1 )* g(t T2 )]t =t' +T1 +T2 =

f (t' )* g(t')

and the result follows.

SOLUTION 16.23. From table 13.1, G(s) =

1
1
, H(s) =
.
s+2
(s + 2) 2

(a) Define p(t) = f(t 2) = u(t).

1
Then, P(s) = . Consider
s
P(s)G(s) =

1
0.5 0.5
=

s(s + 2)
s
s+2

Hence
p(t) * g(t) = 0.5u(t) 0.5e 2tu(t)
From problem 16.22 part (a),

f (t)* g(t) = [ p(t) * g(t)]t =t +2 = 0.5u(t) 0.5e 2tu(t)

]t =t +2 = [0.5 0.5e2(t +2) ]u(t + 2)

(b) Define p(t) = f(t 2) = u(t).

1
Then, P(s) = . Consider
s
P(s)H(s) =
Hence

1
0.25 0.25
0.5
2 = s s+ 2
s(s + 2)
(s + 2)2

1/25/02

P16-32

R. A. DeCarlo, P. M. Lin

p(t) * h(t) = 0.25 0.25e 2t 0.5te2t u(t)

From problem 16.22 part (a),

[(

) ]t =t +2

f (t)* h(t) = [ p(t)* h(t)]t =t +2 = 0.25 0.25e 2t 0.5te2t u(t)

= 0.25 0.25e 2(t +2) 0.5(t + 2)e 2(t +2) u(t + 2)

SOLUTION 16.24.
(a) The pictures of f (t) and g(t) are sketched in the next figures

(b) Using the convolution theorem it follows that

L[ f (t 2 ) g(t )] = L[ f (t 2 )] L [g(t )]
From table 13.1
1
L[ f (t 2 )] = L[u(t)] =
s
1
L[ g(t )] = L[sin(t)u(t)] = 2
s +1
Therefore
1
1
s
L[ f (t 2 ) g(t )] =
= 2
2
s(s + 1) s s + 1
Taking the inverse Laplace transform yields
f (t 2 ) g(t ) = u(t) cos(t)u(t) = [1 cos(t)]u(t)
Using the property give in problem 16.22(b) it follows that
f (t) g(t) = [ f (t 2 ) g(t )]t =t +2 + = [1 cos(t + 3 )]u(t + 3 )

1/25/02

P16-33

SOLUTION 16.25.
Define

R. A. DeCarlo, P. M. Lin

w(t) = vin (t 2)

Hence

w(t) = u(t)

and, from table 13.1,

W (s) =

1
s

From table 13.1 we also have that

H1(s) =
The impulse response of the cascade is

1
1
and H2 (s) =
s +1
(s + 1)2
h(t) = h1 (t) h2 (t)

Hence the transfer function of the cascade is

1
(s + 1)3
w
We denote by vout
(t)the zero state response due to the input w(t). Hence,
1
w
Vout
(s) = H (s)W (s) =
s(s + 1)3
w
A partial fraction expansion of Vout
(s) is obtained using the residue command in MATLAB:
H(s) = H1 (s)H 2 (s) =

>> a = [1];
>> b = [1 3 3 1 0];
>> [r,p,k] = residue(a,b)
r=
-1.0000
-1.0000
-1.0000
1.0000
p=
-1.0000
-1.0000
-1.0000
0
k=
[]
Therefore
w
Vout
(s) =

1 1
1
1
+
+
+
s s + 1 (s + 1)2 (s + 1)3

Taking the inverse Laplace transform yields

1/25/02

P16-34

R. A. DeCarlo, P. M. Lin

w
vout
(t) = 1 e t tet 0.5t 2e t u(t) V

Due to the fact that

vin (t) = w(t + 2)

the time invariance property implies that
w
vout (t) = vout
(t + 2) =

= 1 e (t + 2) (t + 2)e (t +2) 0.5(t + 2) 2 e (t + 2) u(t + 2) V.

SOLUTION 16.26.
(a) Using the sifting property of the delta function it follows that
f 4 (t) = [ (t) + (t 4)] f 2 (t) = f 2 (t) + f 2 (t 4)
The right-hand side of the above equation interprets as a graphical sum of shifted pictures of f 2(t) .
A picture of f 4 (t) is sketched in the next figure.
4
2
0

(b) In order to compute the area beneath f 2(t ) f 2 ( ) four regions will be considered: t < 0,
0 t < 1, 1 t < 2 and 2 t.
Step 1: t < 0. In this case f 2(t ) f 2 ( ) = 0 for all . Therefore
f 2(t) f 2 (t) = 0 for t < 0.
Step 2: 0 t < 1. In this case f 2(t ) f 2 ( ) = 16 for 0 t and is zero elsewhere. The area
beneath f 2(t ) f 2 ( ) equals 16t . Therefore
f 2(t) f 2 (t) = 16t for 0 t < 1.
Step 3: 1 t < 2. In this case f 2(t ) f 2 ( ) = 16 for t 1 < 1 and is zero elsewhere. Hence
the area beneath f 2(t ) f 2 ( )equals
f 2(t) f 2 (t) = 16(2 t) for 1 t < 2.
Step 4: 2 t. In this case f 2(t ) f 2 ( ) = 0 for all . Therefore
f 2(t) f 2 (t) = 0 for2 t.
A picture of f 5(t)is sketched in the next figure.

1/25/02

P16-35

R. A. DeCarlo, P. M. Lin

(c) In order to compute the area beneath f 2(t ) f 3( ) five regions will be considered: t < 0,
0 t < 1, 1 t < 2, 2 t < 3 and 3 t.
Step 1: t < 0. In this case f 2(t ) f 3( ) = 0 for all . Therefore f 2(t) f 3 (t) = 0 for t < 0.
Step 2 : 0 t < 1. In this case f 2(t ) f 3( ) = 8 for 0 t and is zero elsewhere. Therefore
the area beneath f 2(t ) f 3( ) equals
Step 3: 1 t < 2. Here

f 2(t) f 3 ( ) = 8t for 0 t < 1.

8, t 1 < < 1

f 2(t ) f 3( ) = 24, 1 < t

0, otherwise

Hence, the area beneath f 2(t ) f 3( ) equals

f 2(t) f 3 ( ) = 81
[ (t 1)] + 24(t 1) = 8(2t 1) for 1 t < 2.
Step 4: 2 t < 3. In this case f 2(t ) f 3( ) = 24 for t 1 < < 2 and is zero otherwise. Hence,
the area beneath f 2(t ) f 3( ) equals
f 2(t) f 3 (t) = 24[2 (t 1)] = 24(3 t) for 2 t < 3.
Step 5: 3 t. Here f 2(t ) f 3( ) = 0 for all . Therefore f 2(t) f 3 (t) = 0 for t < 0.
A picture of f 6(t) is sketched in the next figure.

1/25/02

P16-36

R. A. DeCarlo, P. M. Lin

SOLUTION 16.27.
By the impulse response theorem, it follows that the response is
y(t) = h(t) f (t) =
= h(t) [ (t) (t 1)]
Using the distributive property of convolution and the sifting property of delta function y(t) can be
written as
y(t) = h(t) h(t 1)
The right-hand side of the above equation interprets as a graphical sum of (shifted) pictures of h(t).
The pictures of h(t), h(t 1) and y(t) are sketched in the next figures.

SOLUTION 16.28.
(a) From the picture of f (t) and h(t) in figure P16.28 we observe that, in order to compute the
area beneath h(t ) f ( ) , we need to consider four cases: t < 0, 0 t < 4, 4 t < 8 and 8 t.
Step 1: t < 0. Here h(t ) f ( ) = 0 for all . Therefore the area beneath h(t ) f ( ) equals zero
and
h(t) f (t) = 0 for t < 0.
Step 2: 0 t < 4. In this case h(t ) f ( ) = 1 for 0 t and is zero otherwise. Hence the area
beneath h(t ) f ( ) equals
h(t) f (t) = t for 0 t < 4.
Step 3: 4 t < 8. In this case
1, t 4 < 4

h(t ) f ( ) = 2, 4 t
0, otherwise

Therefore the area beneath h(t ) f ( ) equals

h(t) f (t) = [ 4 (t 4)] + 2(t 4) = t for 4 t < 8.

1/25/02

P16-37

R. A. DeCarlo, P. M. Lin

Step 4: 8 t. Here h(t ) f ( ) = 2 for t 4 < t and is zero otherwise. Hence

h(t) f (t) = 2[ t (t 4) = 8] for 8 t.
A picture of y(t) is sketched in the next figure.

(b) The impulse response is

h(t) = u(t) u(t 4)

By the impulse response theorem

y(t) = x(t) h(t) =

x( )h(t )d =

x( )[ u(t ) u(t 4) ]d

Here observe that u(t ) u(t 4) is nonzero only when t < t 4.

Therefore
t 4

y(t) =

x( )d

which interprets as the running area under x(t) over the interval [t 4,t].

SOLUTION 16.29.
The response, y(t), is obtained as indicated in the statement of the problem, by using the following
MATLAB code:
>> tstep = 1;
>> vin = [1];
>> h = [0, 2, 3, 1, 1];
>> y = tstep*conv(vin, h);
>> y = [0 y 0];
>> t = 0:tstep:tstep*(length(vin)+length(h));
>> plot(t,y)
>> grid
The response is plotted in the next figure.

1/25/02

P16-38

R. A. DeCarlo, P. M. Lin

SOLUTION 16.30.
A picture of vin (t) sketched in the next figure.

In order to plot the response, y(t), the MATLAB code of problem 16.29 will be used with only one
modification. Namely
vin = [1, 1, 2, 2]
as it can be observed from the picture of vin (t) with the time step tstep = 1.
Therefore the MATLAB code is:
>> tstep = 1;
>> vin = [1, 1, 2, 2];
>> h = [0, 2, 3, 1, 1];
>> y = tstep*conv(vin,h);
>> y = [0 y 0];
>> t = 0:tstep:tstep*(length(vin)+length(h));
>> plot(t,y)
>> grid
The response is plotted in the next figure.

1/25/02

P16-39

R. A. DeCarlo, P. M. Lin

1/25/02

P16-1

R. A. DeCarlo, P. M. Lin

SOLUTION 16.31.
In order to compute the area beneath v(t )h( ) seven regions will be considered: t < 0, 0 t < 1,
1 t < 2, 2 t < 3, 3 t < 4, 4 t < 5 and 5 t.
Step 1: t < 0. For t in this region v(t )h( ) = 0 for all . Hence
y(t) = v(t) h(t) = 0 for t < 0.
Step 2: 0 t < 1. In this case v(t )h( ) = v0 h0 for 0 t and is zero otherwise. Therefore the
area beneath v(t )h( ) equals
y(t) = v(t) h(t) = v0 h0 t for 0 t < 1.
Step 3: 1 t < 2. For t in this region we have
v1 h0,
v0 h0,

v(t )h( ) =
v0 h1,
0,
Therefore the area beneath v(t )h( ) equals

0 t 1
t 1< < 1
1 < t
otherwise

y(t) = v(t) h(t) = v1 h0 [(t 1) 0] + v0 h0 [1 (t 1)] + v0 h1 (t 1)=

= t (v1 h0 v0 h0 + v0 h1) v1 h0 + 2 v0 h0 v0 h1, for 1 t < 2
Step 4: 2 t < 3. In this case

Hence, for 2 t < 3,

v1 h0,
v1 h1,

v(t )h( ) = v0 h1,

v0 h2,

0,

t 2< <1
1 t 1
t 1< < 2
2 < t
otherwise

y(t) = v(t) h(t) =

= v1 h0 [1 (t 2)] + v1 h1 [(t 1) 1] + v0 h1 [2 (t 1)] + v0 h2 (t 2) =
= t (v1 h0 + v1 h1 v0 h1+ v0 h2) + 3 v1 h0 2 v1 h1 + 3 v0 h1 2 v0 h2
Step 5: 3 t < 4. In this case
v1 h1, t 2 < < 2
v1 h2, 2 t 1

v(t )h( ) =
v0 h2, t 1< < 3
0,
otherwise
Hence, for 3 t < 4,
y(t) = v(t) h(t) =
= v1 h1 [2 (t 2)] + v1 h2 [(t 1) 2] + v0 h2 [ 3 (t 1)] =
= t (v1 h1 + v1 h2 v0 h2) + 4 v1 h1 3 v1 h2 + 4 v0 h2
Step 6: 4 t < 5. In this case v(t )h( ) = v1 h2 for t 2 < < 3 and is zero otherwise. Therefore
y(t) = v(t) h(t) = v1 h2 [ 3 (t 2)] = v1 h2 (5 t) for 4 t < 5.

1/25/02

P16-2

R. A. DeCarlo, P. M. Lin

Step 7: 5 t. For t in this region v(t )h( ) = 0 for all . Hence

y(t) = v(t) h(t) = 0 for 5 t.
In sum,
v0 h0 t,
t (v1 h0 v 0 h0 + v 0 h1) v1 h0 + 2 v 0 h0 v0 h1,

t (v1 h0 + v1 h1 v0 h1 + v0 h2) + 3 v1 h0 2 v1 h1 + 3 v 0 h1 2 v 0 h2,

y( t) =
t (v1 h1 + v1 h2 v0 h2) + 4 v1 h1 3 v1 h2 + 4 v 0 h2,
v1 h2 (5 t),

0,

0 t <1
1 t < 2
2t< 3
3t< 4
4t<5
otherwise

Hence,
y1 = y(1) = v0 h0 = 6
y 2 = y(2) = v0 h1+ v1 h0 = 8
y 3 = y(3) = v0 h2 + v1 h1 = 6
y 4 = y(4) = v1 h2 = 4
(b) Using the expressions of p(x) and q(x) it follows that
p(x) q(x) = x 3 (v0 h0) + x 2 (v0 h1+ v1 h0) + x (v0 h0 + v1 h1)+ v1 h2
We observe that the coefficients of p(x) q(x) are exactly y1 , y 2 , y 3 and y 4 , respectively. Therefore
r(x) = p(x) q(x) .

SOLUTION 16.32.
(a) This part will be solved using the techniques of convolution algebra. Therefore we can write f 3(t) as
f 3(t) = f1(1) (t) f 2(1)(t)
Where the superscript (-1) means integration and the superscript (1) means differentiation. From figure
P16.32 we observe that
f1(t) = 4[ u(t) u(t 4) ]
Hence
f1(1) (t) = 4[ tu(t) (t 4)u(t 4) ] =
= 4[r(t) r(t 4)]
By inspection, from the same figure, we have
f 2(1)(t) = 4[ (t) 2 (t 2) + 2 (t 4) 2 (t 6) + (t 8)]
Using the sifting property of the delta function f 3(t) can be computed as follows
f 3(t) = {4[ r(t) r(t 4) ]} {4[ (t) 2 (t 2) + 2 (t 4) 2 (t 6) + (t 8)]} =
= 16[r(t) 2r(t 2) + r(t 4) r(t 8) + 2r(t 10) r(t 12)]
A picture of f 3(t) is sketched in the next figure.

1/25/02

P16-3

R. A. DeCarlo, P. M. Lin

(b) Using the techniques of problem 16.31 and considering the time step tstep = 2, the polynomials p(x) ,
q(x) and r(x) can be associated with the functions f1(t), f 2(t) and
f 3(t), respectively, as below:
p(x) = 4x + 4
q(x) = 4 x 3 4 x 2 + 4 x 4
r(x) = 32x 4 32
We need to verify that the equality

p(x) q(x) tstep = r(x)

holds. The equality indeed holds because

p(x) q(x) tstep = 32(x + 1)(x 3 x 2 + x 1) = 32(x 4 1)= r(x).
The results obtained in part (a) and part (b) coincide.

SOLUTION 16.33.
(a) Using the techniques of convolution algebra f 3(t) can be written as
f 3(t) = f1(1) (t) f 2(1)(t)
Where the superscript (-1) means integration and the superscript (1) means differentiation. From figure
P16.33 we observe that
f1(t) = 2[ u(t + 1) u(t 4)]
Therefore
f1(1) (t) = 4[(t + 1)u(t + 1) (t 4)u(t 4)] =
= 4[r(t + 1) r(t 4)] = g(t)
By inspection, from the same figure, we have
f 2(1)(t) = 4 (t) 8 (t 2) + 6 (t 4) 2 (t 6)
Using the sifting property of the delta function f 3(t) can be computed as follows
f 3(t) = 4g(t) 8g(t 2) + 6g(t 4) 2g(t 6)

1/25/02

P16-4

R. A. DeCarlo, P. M. Lin

This is plotted in MATLAB as follows:

>> t = -2:0.01:14;
>> g = 2*(t+1).*u(t+1)-2*(t-4).*u(t-4);
>> g1 = 4*g;
>> g2 = -8*( 2*(t-1).*u(t-1)-2*(t-6).*u(t-6) );
>> g3 = 6*( 2*(t-3).*u(t-3)-2*(t-8).*u(t-8) );
>> g4 = -2*( 2*(t-5).*u(t-5)-2*(t-10).*u(t-10) );
>> f3 = g1+g2+g3+g4;
>> plot(t,f3);
>> grid;
A picture of f 3(t) is sketched in the next figure.

(b) To account for the fact that f1(t)is nonzero for negative t the following formula (see problem 16.22)
f1(t) f 2 (t) = [ f1(t 1) f 2 (t)]t =t +1
will be used to compute f 3(t). Using a slightly modified version of the code of problem 16.31, we have
>> f1 = [2, 2, 2, 2, 2];
>> f2 = [4, 4, -4, -4, 2, 2];
>> T = 1;
>> tstep = T;
>> f3 = tstep*conv(f1,f2);
>> f3 = [0 f3 0];
>> t = -1:tstep:tstep*(length(f1)+length(f2))-1;
>> plot(t,f3)
>> grid

1/25/02

P16-5

R. A. DeCarlo, P. M. Lin

The results of parts (a) and (b) coincide.

SOLUTION 16.34. This problem is solved using the techniques of the convolution algebra with the
graphical method left to the student.
f3 (t) = f1(t) * f2 (t) = [ f1(t)]

(1)

* [ f2 (t)]

(1)

where the superscript (-1) means integration and the superscript (1) means differentiation. By inspection,

[ f1(t )](1) = 4tu(t) 4(t 6)u(t 6) = g(t)

and

[ f2 (t)](1) = 4(t) 8(t 2) + 8(t 6) 4(t 8)

Hence the response say y(t) satisfies
f3 (t) = 4g(t) 8g(t 2) + 8g(t 6) 4g(t 8)
This is plotted in MATLAB as follows:
t=0:.05:20;
g = 4*t .* u(t) - 4*(t-6) .*u(t-6);
g1=4*g;
g2 = -8*(4*(t-2) .* u(t-2) - 4*(t-8) .*u(t-8));
g3 = 8*(4*(t-6) .* u(t-6) - 4*(t-12) .*u(t-12));
g4 = -4*(4*(t-8) .* u(t-8) - 4*(t-14) .*u(t-14));
f3 = g1+g2+g3+g4;
plot(t,f3)
grid

1/25/02

P16-6

R. A. DeCarlo, P. M. Lin

40
30
20
10
0
-10
-20
-30
-40

(b) Using the code of problem 31, we have

f1 = [4 4 4];
f2 = [4 -4 -4 4];
T = 2;
tstep = T;
f3 = [0 conv(f1,f2)*tstep 0];
t = 0: tstep : tstep* (length(f1) + length(f2));
plot(t,f3)
grid

10

12

14

16

18

20

1/25/02

P16-7

R. A. DeCarlo, P. M. Lin

40
30
20
10
0
-10
-20
-30
-40

10

12

14

The results of parts (a) and (b) coincide.

SOLUTION 16.35.
(a) In order to compute the area beneath f1(t ) f1 ( ) four regions will be considered: t < 0, 0 t < 1,
1 t < 2, and 2 t.
Step 1: t < 0. Here f1(t ) f1 ( ) = 0 for all . Hence
f 3(t) = f1 (t) f1(t) = 0 for t < 0.
Step 2: 0 t < 1. In this case f1(t ) f1 ( ) = 1 for 0 t and is zero otherwise. Therefore the area
beneath f1(t ) f1 ( ) equals
f 3(t) = t for 0 t < 1.
Step 3: 1 t < 2. In this case f1(t ) f1 ( ) = 1 for t 1 < < 1 and is zero otherwise. Therefore the
area beneath f1(t ) f1 ( ) equals
f 3(t) = 1 (t 1)= 2 t for 1 t < 2.
Step 4: 2 t. Here f1(t ) f1 ( ) = 0 for all . Hence
In sum,

f 3(t) = f1 (t) f1(t) = 0 for2 t.

t,
0 t <1

f 3(t) = 2 t, 1 t < 2
0,
otherwise

1/25/02

P16-8

R. A. DeCarlo, P. M. Lin

(b) In order to compute the area beneath f1(t ) f 2 ( ) four regions will be considered: t < 0, 0 t < 1,
1 t < 2, and 2 t.
Step 1: t < 0. Here f1(t ) f 2 ( ) = 0 for all . Hence
f 4 (t) = f1 (t) f 2 (t) = 0 for t < 0.
Step 2: 0 t < 1. In this case f1(t ) f 2 ( ) = for 0 t and is zero otherwise. Therefore the area
beneath f1(t ) f 2 ( ) equals
f 4 (t) = 0.5t 2 for 0 t < 1.
Step 3: 1 t < 2. For t in this region f1(t ) f 2 ( ) = for t 1 < < 1 and is zero otherwise.
Therefore the area beneath f1(t ) f 2 ( ) equals
f 4 (t) = 0.5 0.5t 2 for 1 t < 2.
Step 4: 2 t. Here f1(t ) f 2 ( ) = 0 for all . Hence
f 4 (t) = f1 (t) f1(t) = 0 for2 t.
In sum,
0.5t 2,
0 t <1

2
f 4 (t) = 0.5 0.5t , 1 t < 2
0,
otherwise

(c) In order to compute the area beneath f 2(t ) f 2 ( ) four regions will be considered: t < 0, 0 t < 1,
1 t < 2, and 2 t.
Step 1: t < 0. Here f 2(t ) f 2 ( ) = 0 for all . Hence
f 5(t) = f 2 (t) f 2 (t) = 0 for t < 0.
Step 2: 0 t < 1. In this case f 2(t ) f 2 ( ) = (t ) for 0 t and is zero otherwise. Therefore
the area beneath f 2(t ) f 2 ( ) equals
t

]0

t
f 5(t) = (t ) d = 0.3333 3 + 0.5t 2 =0.1667t 3 for 0 t < 1.
0

Step 3: 1 t < 2. For t in this region f 2(t ) f 2 ( ) = (t ) for t 1 < < 1 and is zero otherwise.
Therefore the area beneath f 2(t ) f 2 ( ) equals
1

f 5(t) =

t 1

]t 1

1
(t ) d = 0.3333 3 + 0.5t 2

= 0.1667t 3 + t 0.6667 for 1 t < 2.

Step 4: 2 t. Here f 2(t ) f 2 ( ) = 0 for all . Hence

1/25/02

In sum,

P16-9

R. A. DeCarlo, P. M. Lin

f 5(t) = f 2 (t) f 2 (t) = 0 for2 t.

0.1667t 3,
0 t <1

3
f 5(t) = 0.1667t + t 0.6667, 1 t < 2
0,
otherwise

SOLUTION 16.36.
In order to compute the area beneath f1(t ) f 2 ( ) five regions will be considered: t < 0, 0 t < 1,
1 t < 2, 2 t < 3and 3 t.
Step 1: t < 0. Here f1(t ) f 2 ( ) = 0 for all . Hence
f 3(t) = f1 (t) f 2 (t) = 0 for t < 0.
Step 2: 0 t < 1. In this case f1(t ) f 2 ( ) = for 0 t and is zero otherwise. Therefore the area
beneath f1(t ) f 2 ( ) equals
f 3(t) = 0.5t 2 for 0 t < 1.
Step 3: 1 t < 2. For t in this region

,
t 1< < 1

f1(t ) f 2 ( ) = 2 , 1 t
0,
otherwise

Therefore the area beneath f1(t ) f 2 ( ) equals

] [

f 3(t) = 0.5 0.5(t 1)2 + 0.5 0.5(2 t)2

2

= t + 3t 1.5 for 1 t < 2.

Step 4: 2 t < 3. For t in this region f1(t ) f 2 ( ) = 2 for all t 1 < < 2. Hence
f 3(t) = f1 (t) f 2 (t) = 0.5 0.5(2 t) 2 = 0.5t 2 + 2t 1.5 for 2 t < 3.
Step 5: 3 t. Here f1(t ) f 2 ( ) = 0 for all . Hence
f 3(t) = f1 (t) f 2 (t) = 0 for 3 t.
In sum,
0.5t 2,
2
t + 3t 1.5,
f 3(t) =
2
0.5t + 2t 1.5,
0,

SOLUTION 16.37.

0 t <1
1 t < 2
2 t < 3
otherwise

1/25/02

P16-10

R. A. DeCarlo, P. M. Lin

(a) In order to compute the area beneath f1( ) f 2 (t ) three regions will be considered: t < 0,
0 t < 2, and 2 t.
Step 1: t < 0. Here f1( ) f 2 (t ) = 0 for all . Hence
f 3(t) = f1 (t) f 2 (t) = 0 for t < 0.
Step 2: 0 t < 2. In this case f1( ) f 2 (t ) = 8 for 0 t and is zero otherwise. Therefore the
area beneath f1( ) f 2 (t ) equals
f 3(t) = 0.5(t 8t) = 4t 2 for 0 t < 2.
Step 3: 2 t. For t in this region

8 , 0 < 2

f1( ) f 2 (t ) = 16, 2 < t

0, otherwise

Therefore the area beneath f1( ) f 2 (t ) equals

In sum,

f 3(t) = 16 + 16(t 2) = 16(t 1) for 2 t.

4t 2,
0 t < 2

f 3(t) = 16(t 1), 2 t

0,
otherwise

A picture of f 3(t) is sketched in the next figure.

(b) From figure P16.37 we observe that

f1(t) = 2tu(t) 2(t 2)u(t 2)
From table 13.1 and the time shift property of the Laplace transform it follows that

2
F1(s) = 2 1 e 2s
s

1/25/02

P16-11

F2 (s) =
By the convolution theorem

R. A. DeCarlo, P. M. Lin

4
s

F3(s) = F1 (s)F2 (s)

Therefore

8
F3(s) = 3 1 e 2s
s
Taking the inverse Laplace transform yields

f 3(t) = 4t 2 u(t) 4(t 2) 2 u(t 2)

The results of parts (a) and (b) coincide.

SOLUTION 16.38.
(a) In order to compute the area beneath f1(t ) f 2 ( ) six regions will be considered: t < 0, 0 t < 2,
2 t < 6, 6 t < 8, 8 t < 10, and 10 t.
Step 1: t < 0. Here f1(t ) f 2 ( ) = 0 for all . Hence
f 3(t) = f1 (t) f 2 (t) = 0 for t < 0.
Step 2: 0 t < 2. In this case f1(t ) f 2 ( ) = 8(t ) for 0 t and is zero otherwise. Therefore
the area beneath f1( ) f 2 (t ) equals
f 3(t) = 0.5(t 8t) = 4t 2 for 0 t < 2.
Step 3:2 t < 6. For t in this region

16,
0 < t 2

f1(t ) f 2 ( ) = 8(t ), t 2 < t

0,
otherwise

Therefore the area beneath f1(t ) f 2 ( ) equals

f 3(t) = 16 + 16(t 2) = 16(t 1) for 2 t < 6.
Step 4: 6 t < 8. For t in this region

16,
8(t ),

f1(t ) f 2 ( ) =
8(t ),
0,

0 < t 2
t 2 < 6
6 <t
otherwise

Therefore, for 6 t < 8, the area beneath f1(t ) f 2 ( ) equals

8(t 6)2 8(t 6) 2

f 3(t) = 16(t 2) + 16

2
2

1/25/02

P16-12

R. A. DeCarlo, P. M. Lin

= 8(t 6)2 + 16t 16 = 8t 2 + 112t 304

Step 5: 8 t < 10. For t in this region

0 <6
16,
16,
6 < t 2

f1(t ) f 2 ( ) =
8(t ), t 2 < 8
0,
otherwise

Therefore, for 8 t < 10, the area beneath f1(t ) f 2 ( ) equals

8(t 8) 2
f 3(t) = 96 16(t 8) 16
= 4t 2 80t + 464
2

Step 6: 10 t. For t in this region

Therefore,
In sum,

16, 0 < 6

f1(t ) f 2 ( ) = 16, 6 < 8

0,
otherwise

f 3(t) = 96 32 = 64 for 10 t.

0,
4t 2,

16t 16,
f 3(t) = 2
8t + 112t 304,
4t 2 80t + 464,

64,
A picture of f 3(t) is sketched in the next figure.

(b) From figure P16.38 we observe that

t<0
0 t < 2
2 t < 6
6 t < 8
8 t < 10
10 t

1/25/02

P16-13

R. A. DeCarlo, P. M. Lin

f 2(t) = 4[u(t) u(t 6)] 4[ u(t 6)u (t 8)]

= 4u(t) 8u(t 6) + 4u(t 8)
From table 13.1 and the time shift property of the Laplace transform it follows that
4 8
4
F2 (s) = e 6s + e 8s
s s
s
From problem 16.37 we have that
2
F1(s) = 2 1 e 2s
s
By the convolution theorem
F3(s) = F1 (s)F2 (s)
Therefore
8
8
16
24
8
F3(s) = 3 3 e 2s 3 e 6s + 3 e 8s 3 e 10s
s
s
s
s
s

Taking the inverse Laplace transform yields

f 3(t) = 4t 2 u(t) 4(t 2) 2 u(t 2) 8(t 6)2 u(t 6)
+12(t 8)2 u(t 8) 4(t 10)2 u(t 10)
The results of parts (a) and (b) coincide.

SOLUTION 16.39.
p(t) q(t) will be computed using the techniques of convolution algebra. Therefore we can write
p(t) q(t) = p (1)(t) q(1) (t)
where the superscript (1)means integration and the superscript (1)means differentiation. By inspection
p(t) = (t + 4)[ u(t + 4) u(t)] + ( t + 4)[u(t) u(t 4)] + 4[u(t 4) u(t 8)] =
= (t + 4)u(t + 4) + (2t)u(t) + tu(t 4) 4u(t 8)
Therefore,
p(1) (t) = 0.5(t + 4)2 u(t + 4) t 2u(t) + (0.5t 2 8)u(t 4) (4t 32)u(t 8)
By inspection we also have
q (1) (t) = 4 (t)
By the sifting property of the delta function it follows that
p(t) q(t) = 4 p (1) (t) =
= 2(t + 4) 2 u(t + 4) 4t 2u(t) + 2(t 2 16)u(t 4) 16(t 8)u(t 8).

SOLUTION 16.40.
(a) First observe that

1/25/02

P16-14

R. A. DeCarlo, P. M. Lin

h(t) = 0.1u(t 0.1) + 0.2u(t 0.2) + 0.2u(t 0.3) + 0.2u(t 0.4) + 0.2u(t 0.5) +
+0.1u(t 0.6) 0.1u(t 1) 0.2u(t 1.3) 0.2u(t 1.5)
0.2u(t 1.7) 0.2u(t 2) 0.1u(t 2.2)
Due to the fact that h(t) is a linear combination of terms of the type Ku(t T ), the convolution of h(t)
and vin (t) reduces to a linear combination of terms of the following type: [K1u(t)] [K 2u(t T )] . Using
the definition of the convolution, the previous convolution product is computed below

[K1u(t)] [K 2u(t T )] = K1K 2

u( )u(t T )d =

t<T
0,

t T

= K1K2 u(t T )d = K K
1 2 d , Tt
0

0
t<T
0,
=
= K1K2 (t T )u(t T )
K1K2 (t T ), T t
Therefore vout (t) is a linear combination of functions of type K1K 2(t T )u(t T ),
vout (t) = h(t) vin (t) =
= 10(t 0.1)u(t 0.1)+ 20(t 0.2)u(t 0.2) + 20(t 0.3)u(t 0.3) + 20(t 0.4)u(t 0.4) +
+20(t 0.5)u(t 0.5) + 10(t 0.6)u(t 0.6) 10(t 1)u(t 1) 20(t 1.3)u(t 1.3)
20(t 1.5)u(t 1.5) 20(t 1.7)u(t 1.7) 20(t 2)u(t 2) 10(t 2.2)u(t 2.2)
At t = 0s

vout (0) = 0 V.

At t = 0.5s

vout (0.5) = 16 V.

At t = 1s

vout (1) = 65 V.

At t = 1.5s

vout (1.5)= 106 V.

(b) In this case vout (t) will be computed using the techniques of the convolution algebra. Hence we have
(1)
vout (t) = vin (t) h(t) = v in
(t) h (1) (t)

= 50t 2 u(t) [0.1 (t 0.1) + 0.2 (t 0.2) + 0.2 (t 0.3) + 0.2 (t 0.4) +
+0.2 (t 0.5) + 0.1 (t 0.6) 0.1 (t 1) 0.2 (t 1.3)
0.2 (t 1.5) 0.2 (t 1.7) 0.2 (t 2) 0.1 (t 2.2)]

Using the sifting property of delta function it follows that

vout (t ) = 5(t 0.1) 2 u (t 0.1) + 10(t 0.2) 2 u (t 0.2) + 10(t 0.3) 2 u (t 0.3) +

+10(t 0.4) 2 u(t 0.4) + 10(t 0.5)2 u(t 0.5) + 5(t 0.6) 2 u(t 0.6)
5(t 1)2 u(t 1) 10(t 1.3)2 u(t 1.3) 10(t 1.5)2 u(t 1.5)
10(t 1.7) 2 u(t 1.7) 10(t 2)2 u(t 2) 5(t 2.2)2 u(t 2.2) V.

1/25/02

At t = 1s

P16-15

R. A. DeCarlo, P. M. Lin

vout (1) = 22.25 V.

(c) From the expression of vout (t)obtained in part (a) we observe that vout (t)does not change after
t = 2.2s . Therefore it is sufficient to compute vout (t)for t 2.2s . Hence vin (t) can be considered to be
equal to
vin (t) = 100[u(t) u(t 2.2)] V.
Using the code of problem 16.31 we have
>> vin = 100*ones(1,22);
>> h = [0, 0.1, 0.3, 0.5, 0.7, 0.9, 1, 1, 1, 1, 0.9, 0.9, 0.9, 0.7, 0.7, 0.5, 0.5, 0.3, 0.3,
0.3, 0.1, 0.1];
>> T = 0.1;
>> tstep = T;
>> y = tstep*conv(vin,h);
>> y = [0 y 0];
>> t = 0:tstep:tstep*(length(h)+length(vin));
% After t = 2.2s vout(t) does not change
>> t = t(1:length(h)+1);
>> y = y(1:length(h)+1);
>> plot(t,y)
>> grid
A picture of y(t) is sketched in the next figure.

Using the previous MATLAB code we the values of y(t) at the specified instants of time are:
At t = 0s
y(0) = 0 V

1/25/02

P16-16

At t = 0.5s
At t = 1s
At t = 1.5s

R. A. DeCarlo, P. M. Lin

y(0.5) = 16 V
y(1) = 65 V
y(1.5)= 106 V

The results of parts (a) and (c) coincide.

SOLUTION 16.41.
Using the MATLAB code of problem 16.31 we have:
>> vin = [1];
>> h = [9, -6, 3, -2];
>> T = 1;
>> tstep = T;
>> y = tstep*conv(vin,h);
>> y = [0 y 0];
>> t = 0:tstep:tstep*(length(h)+length(vin));
>> plot(t,y)
>> grid
The breakpoints in y(t) of the above figure are [9, -6, 3 2] as expected because the polynomial
associated with vin (t) is the constant 1 and the polynomial associated with h(t) is the polynomial
9x 3 6x 2 + 3x 2, as it can be observed from figure P16.41.

SOLUTION 16.42.
(a) Let vout ,40 (t) denote the response that has been obtained in problem 16.40, part (a), to the input
100u(t) .
The expression of vout ,40 (t) is (see problem 16.40, part (a)):
vout ,40 (t) = h(t) [100u(t)] =
= 10(t 0.1)u(t 0.1)+ 20(t 0.2)u(t 0.2) + 20(t 0.3)u(t 0.3) + 20(t 0.4)u(t 0.4) +

1/25/02

P16-17

R. A. DeCarlo, P. M. Lin

+20(t 0.5)u(t 0.5) + 10(t 0.6)u(t 0.6) 10(t 1)u(t 1) 20(t 1.3)u(t 1.3)
20(t 1.5)u(t 1.5) 20(t 1.7)u(t 1.7) 20(t 2)u(t 2) 10(t 2.2)u(t 2.2)
Using the distributive property of the convolution product and the time invariance property it follows that
vout (t) = h(t) vin (t) = h(t) [100u(t) 100u(t 0.2)] =
= h(t) [100u(t)] h(t) [100u(t 0.2)] =
= vout,40 (t) vout,40 (t 0.2)
Using the above expression of vout ,40 (t) we have:
vout ,40 (0) = 0 V and vout ,40 (0.2) = 0V,
vout ,40 (0.5) = 16 V and vout ,40 (0.3) = 4 V,
vout ,40 (1) = 65 V and vout ,40 (0.8) = 45V,
vout ,40 (1.5) = 106 V and vout ,40 (1.3)= 92V.
At t = 0s
At t = 0.5s
At t = 1s
At t = 1.5s

vout (0) = vout ,40 (0) vout ,40 (0.2) = 0 V.

vout (0.5) = vout ,40 (0.5) vout ,40 (0.3) = 12 V.
vout (1) = vout ,40 (1) vout ,40 (0.8) = 20 V.
vout (1.5)= vout ,40 (1.5) vout ,40 (1.3) = 14 V.

(b) In this case vout (t) will be computed using the techniques of convolution algebra.
We have
(1)
vout (t) = vin (t) h(t) = v in
(t) h (1) (t)
where the superscript (-1) means integration and the superscript (1) means differentiation.
From figure P16.42 observe that
vin (t) = 100t[u(t) u(t 0.5)] + 100(1 t)[u(t 0.5) u(t 1)] =
= 100tu(t) + 100(1 2t)u(t 0.5) + 100(t 1)u(t 1)
Therefore
vin(1) (t) = 50t 2u(t) + (100t 2 + 100t 25)u(t 0.5) + (50t 2 100t + 50)u(t 1) = g(t)
By the sifting property of the delta function we have
(1)
vout (t) = vin
(t) h (1)(t) = g(t) h (1) (t)

= g(t) [0.1 (t 0.1) + 0.2 (t 0.2) + 0.2 (t 0.3) + 0.2 (t 0.4) +

+0.2 (t 0.5) + 0.1 (t 0.6) 0.1 (t 1) 0.2 (t 1.3)
0.2 (t 1.5) 0.2 (t 1.7) 0.2 (t 2) 0.1 (t 2.2)] =
= 0.1g(t 0.1)+ 0.2g(t 0.2) + 0.2g(t 0.3) + 0.2g(t 0.4) +
+0.2g(t 0.5) + 0.1g(t 0.6) 0.1g(t 1) 0.2g(t 1.3)

1/25/02

P16-18

R. A. DeCarlo, P. M. Lin

0.2g(t 1.5) 0.2g(t 1.7) 0.2g(t 2) 0.1g(t 2.2)

The values ofvout (t) at the specified instants of time can be computed using MATLAB. The results are:
vout (0) = 0V
vout (0.5) = 2.2V
vout (1) = 17.85V
vout (1.5)= 23.3V.

SOLUTION 16.43.
Using the techniques of convolution algebra we have
vout (t) = h(t) vin (t) = h (1) (t) vin(1) (t)
where the superscript (-1) means integration and the superscript (1) means differentiation.
We have
h (1) (t) = 2(1 e 2t )u(t)
and
vin(1) (t) = (t) (t 1)
Using the sifting property of the delta function it follows that
vout (t) = 2(1 e 2t )u(t) [ (t) (t 1)] =

2t

= 2(1 e )u(t) 2(1 e 2(t 1) )u(t 1)

A picture of vout (t)is sketched in the next figure.

SOLUTION 16.44.
(a) From table 13.1 it follows that the Laplace transform of vin (t) is
1
1
1
s2 + 2
Vin (s) =

=
s s + 1 s + 2 s(s + 1)(s + 2)

1/25/02

P16-19

R. A. DeCarlo, P. M. Lin

And the Laplace transform of vout (t) is

1
2
1
1
s2 + 2
Vout (s) =

+
=
s s + 1 (s + 1)2 s + 2 s(s + 1)2(s + 2)
Therefore the transfer function of the circuit is
V (s)
1
H(s) = out
=
Vin (s) s + 1
A simple RC circuit that represents this transfer function is a series RC circuit with R = 1 and C = 1F .
vout (t)is represented by the capacitor voltage and vin (t) is the source voltage.
(b) The impulse response of the circuit is
1 t
h(t) = L1[ H (s)] = L1
= e u(t).
s + 1
(c) Assuming zero initial conditions we have

(1)
vout (t) = vin (t) h(t) = v in
(t) h (1) (t) = (t) (1 e t )u(t) = (1 e t )u(t) V.

(d) Using the techniques of convolution algebra the zero-state response can be computed as
(2)
vout (t) = vin (t) h(t) = v in
(t) h (2) (t)
where the superscript (2) means double differentiation and the superscript (-2) means double integration.
First, from figure P16.44, observe that
vin(1) (t) = [ u(t 1) u(t 2)] + [ u(t 3) u(t 4) ]
Therefore
vin(2)(t) = (t 1) (t 2) + (t 3) (t 4)
h (2) (t) is computed as the integral of h (1) (t) .
t

(2)

(t) =

(1)

( )d =

(1 e

)u( )d =

(1 e

)d = (t + e t 1)u(t)

The zero-state response can now be computed

(2)
vout (t) = vin
(t) h (2) (t) =

= [ (t 1) (t 2) + (t 3) (t 4) ] (t + e t 1)u(t)
By the sifting property of the delta function it follows that

(
)
+( t 4 + e (t 3) ) u(t 3) ( t 5 + e (t 4) ) u(t 4) V.

vout (t) = t 2 + e (t 1) u(t 1) t 3 + e (t 2) u(t 2)

1/25/02

P16-20

R. A. DeCarlo, P. M. Lin

SOLUTION 16.45.
Using the techniques of the convolution algebra we have
y(t) = f (t) g(t) = f (2) (t) g(2) (t)
where
t

(1)

(t) =

cos(

)d = sin( t)u(t)

and
t

(2)

(t) =

sin(

)d = [1 cos( t)]u(t)

Differentiating f (t) twice leads to

f (2) (t) = (t) + 2 (t 1) 2 (t 3)+ (t 4)
Therefore
y(t) = f (2) (t) g(2) (t) =
= [ (t) + 2 (t 1) 2 (t 3) + (t 4)] {[1 cos( t)]u(t)}
Using the sifting property of the delta function it follows that
y(t) = [1 cos( t)]u(t) + 2{1 cos[ (t 1)]}u(t 1)
2{1 cos[ (t 3)]}u(t 3) + {1 cos[ (t 4)]}u(t 4)
Simplifying the expression of y(t) yields
y(t) = [1 cos( t)] [u(t) u(t 4)] + 2 [1 + cos( t)] [ u(t 1) u(t 3)] .
A picture of y(t) is sketched in the next figure.

1/25/02

P16-21

R. A. DeCarlo, P. M. Lin

SOLUTION 16.46.
(a) By the current division formula
IC (s) =

s2

I
(s)
=
2 + 1 Iin (s)
1 in
s
Cs +
Ls
Cs

The transfer function of the circuit can be computed as below

V (s) IC (s) ZC (s)
s2 1
s
H(s) = C
=
= 2
= 2
Iin (s)
Iin (s)
s +1 s s +1
(b) The impulse response is computed as the inverse Laplace transform of the transfer function
h(t) = L1[ H (s)] = cos(t)u(t).
(c) Assuming zero initial conditions it follows, by the impulse response theorem, that
vout (t) = iin (t) h(t)
Using the techniques of the convolution algebra we have
vout (t) = iin(2) (t) h (2) (t)
By inspection

(1)
iin
(t) = [u(t) u(t 2 )] + [ u(t 4 ) u(t 6 )]

Therefore
(2)
iin
(t) = (t) (t 2 ) + (t 4 ) (t 6 )

And
t

(1)

(t) =

cos(

)d = sin(t)u(t)

Hence
t

(2)

(t) =

sin(

)d = [1 cos(t)]u(t)

Using the sifting property of the delta function we have

vout (t) = [ (t) (t 2 ) + (t 4 ) (t 6 )] {[1 cos(t)]u(t)}
= [1 cos(t)]u(t) [1 cos(t 2 )]u(t 2 )
+[1 cos(t 4 )]u(t 4 ) [1 cos(t 6 )]u(t 6 )
= [1 cos(t)] [u(t) u(t 2 ) + u(t 4 ) u(t 6 )]

1/25/02

P16-22

R. A. DeCarlo, P. M. Lin

V.
(d) A picture of vout (t) is sketched in the next figure.

SOLUTION 16.47.
(a) The step response, vout (t), is computed using the convolution algebra techniques.
We have
(1)
vout (t) = h(t) vin (t) = h (1)(t) vin
(t)
From figure P16.47 observe that
h(t) = 2u(t) u(t 1) 2u(t 2) u(t 3)+ u(t 5) + 2u(t 6) + u(t 7) 2u(t 8)
Differentiating we have
h (1) (t) = 2 (t) (t 1) 2 (t 2) (t 3) + (t 5) + 2 (t 6) + (t 7) 2 (t 8)
Since
by integration it follows that

vin (t) = u(t),

vin(1) (t) = tu(t).

Using the sifting property of the delta function it follows that

(1)
vout (t) = h (1) (t) vin
(t)
= [2 (t) (t 1) 2 (t 2) (t 3) + (t 5) + 2 (t 6) + (t 7) 2 (t 8)] [tu(t)]
= 2tu(t) (t 1)u(t 1) 2(t 2)u(t 2) (t 3)u(t 3) +
+(t 5)u(t 5) + 2(t 6)u(t 6) + (t 7)u(t 7) 2(t 8)u(t 8)V.

A picture of the step response is sketched in the next figure.

1/25/02

P16-23

R. A. DeCarlo, P. M. Lin

(b) Using the convolution algebra techniques we have

(1)
vout (t) = h(t) vin (t) = h (1)(t) vin
(t)

For computing vin(1) (t) we have

t
e d , t<0
t

e t , t < 0
vin(1) (t) = e u( )d =
=

0
1, 0 t

e d , 0 t

Using the sifting property of the delta function it follows that

(1)
vout (t) = h (1) (t) vin
(t) =

(1)
= [2 (t) (t 1) 2 (t 2) (t 3) + (t 5) + 2 (t 6) + (t 7) 2 (t 8)] vin
(t) =

Using the

(1)
(1)
(1)
(1)
= 2v in
(t) vin
(t 1) 2vin
(t 2) vin
(t 3) +
(1)
(1)
(1)
+v in
(t 5) + 2vin(1) (t 6) + vin
(t 7) 2v in
(t 8)
(1)
expression of vin (t) computed above it follows that

vout (7.5) = 0.7869 V,

vout (6.5) = 1.1603V,
vout (5.5) = 0.2720V,
vout (0.5) = 0.8848V.

SOLUTION 16.48.
First observe, from figure P16.48(a), that
h(t) = (1 t) [ u(t) u(t 1)] =

V.

1/25/02

P16-24

R. A. DeCarlo, P. M. Lin

= (1 t)u(t) (1 t)u(t 1).

Using the convolution algebra techniques it follows that the step response y(t) is
y(t) = v(t) h(t) = v (1) (t) h (1)(t) =
= (t) h (1) (t) = h (1) (t)
where
t

(1)

(t) =

0.5

(1 )u( )d

2 t
0
2

u(t)

(1

)u( 1)d =

]1

t
0.5 2 u(t 1) =

= (t 0.5t )u(t) (t 0.5t 2 0.5)u(t 1).

Hence the step response is
y(t) = (t 0.5t 2 )u(t) (t 0.5t 2 0.5)u(t 1).
Let y v (t) denote the zero-state response to the input v(t).
From figure P16.48(b) we observe that
v(t) = u(t) + u(t 1) 2u(t 2) .
Using the distributive property of the convolution it follows that
y v (t) = h(t) v(t) =
= h(t) [ u(t) + u(t 1) 2u(t 2)] =
= h(t) u(t) + h(t) u(t 1) 2 h(t) u(t 2).
Due to the fact that
y(t) = h(t) u(t)
by the linearity and time invariance properties it follows that
y v (t) = y(t) + y(t 1) 2y(t 2) =
= (t 0.5t 2 )u(t) + (t 1)u(t 1)+
+(1.5t 2 8t + 12)u(t 2) + (0.5t 2 3t + 4)u(t 3).

SOLUTION 16.49.
(a) By the voltage division formula it follows that
1
1
Vout (s) = 1 Cs Vin (s) =
2 + 1 Vin (s)
LCs
+ Ls
Cs
Therefore the transfer function is
V (s)
1
H(s) = out
=
Vin (s) LCs 2 + 1
Taking the inverse Laplace transform yields
1

1
1
LC = 1 sin 1 t u(t) .
h(t) = L1
=
L

LC
LC LC
LCs 2 + 1
s2 + 1

LC
(b) The step response is computed as the convolution of the impulse response and the step function.
vout (t) = h(t) u(t)
Using the techniques of the convolution algebra it follows that

1/25/02

P16-25

R. A. DeCarlo, P. M. Lin

vout (t) = h (1) (t) u(1) (t)

where the superscript (-1) means integration and the superscript (1) means differentiation.
Taking the integral of h(t) we have
t

(1)

(t) =

1
1
sin
u( )d =
LC LC

1
1
= u(t) cos
= 1 cos
t u(t) .
LC 0
LC

Therefore
vout (t) = h (1) (t) u(1) (t) =

1
= 1 cos
t u(t) (t) =
LC

1
= 1 cos
t u(t) V.
LC

T
(c) We denote by vout
(t) the output to the rectangular pulse in figure P16.49(b).
Observe, from figure P16.49(b), that
1
vin (t) = [u(t) u(t T )] V.
T
By linearity and time invariance it follows that
1
T
vout
(t) = [v out (t) vout (t T )]
T
where vout (t)is the step response obtained in part (b). Therefore
T
vout
(t) =

1
1
t u(t)
1 cos
LC
LC
2 LC

1
2

1
t u(t) u(t 2
1 cos
LC
LC

1
2

t 2 u(t 2
1 cos

LC

T
A picture of vout
(t), for L = 1H and C = 1F , is sketched in the next figure.

LC

LC )

1/25/02

P16-26

R. A. DeCarlo, P. M. Lin

SOLUTION 16.50.
The impulse response of the configuration in the figure P16.50 is
h(t) = h1 (t) [h2 (t) + h3(t)] h4 (t)
Due to the fact that h4 (t) = 2 (t) , the sifting property of the delta function can be applied and it follows
that
h(t) = 2 h1 (t) [h2 (t) + h3(t)]
Using the techniques of the convolution algebra we can further write
h(t) = 2 h1(1) (t) [h2 (t) + h3(t)]
h(t) =

2 h1(1) (t)

We have

h3(1) (t) =

2e 2 u( )d = u(t) e 2

h2(1) (t) + h3(1)(t)

h1(1) (t) = u (1)(t) = (t)

t

h2(1) (t)

(1)

8e 4 u( )d = u(t) 2e 4

]0 = (1 e2t )u(t)
t

]0 = 2(1 e4t )u(t)

t

Substituting the expressions of h1(1) (t) , h2(1) (t) and h3(1) (t) in the expression of h(t), and using the
sifting property of the delta function we have

h(t) = 2 1 e 2t u(t) + 4 1 e 4t u(t).

A picture of h(t) is sketched in the next figure.

1/25/02

P16-27

R. A. DeCarlo, P. M. Lin

SOLUTION 16.51.
Observe first that h2(t) , h3(t) and h4 (t) have the same expressions as in the problem 16.50. In problem
16.50 the following convolution has been computed

u(t) [ h2 (t) + h3(t)] h4 (t) = 2 1 e 2t u(t) + 4 1 e 4t u(t)

Using the time shift property of the convolution (see problem 16.22, part (a)) it follows that

u(t 2) [ h2 (t) + h3(t)] h4 (t) = 2 1 e 2(t 2) u(t 2) + 4 1 e 4( t 2) u(t 2)

Using the above expressions and the distributive property of the convolution it follows that the overall
impulse response can be computed as below
h(t) = h1 (t) [h2 (t) + h3(t)] h4 (t) = [u(t) u(t 2)] [h2 (t) + h3 (t)] h4 (t)
= u(t) [h2 (t) + h3 (t)] h4 (t) u(t 2) [ h2 (t) + h3(t)] h4 (t) =

= 21 e 2t u(t) + 4 1 e 4t u(t) 2 1 e 2(t 2) u(t 2) + 4 1 e 4(t 2) u(t 2).

A picture of h(t) is sketched in the next figure.

1/25/02

P16-28

R. A. DeCarlo, P. M. Lin

SOLUTION 16.52.
The overall impulse response of the configuration is
h(t) = h1 (t) [h2 (t) + h3(t)] h4 (t)
Using the distributive property of the convolution we have
h(t) = h1 (t) h2 (t) h4 (t) + h1(t) h3 (t) h4 (t)
Replacing the expressions for h2(t) and h3(t), and using the sifting property of the delta function it
follows that
h(t) = 2 h1 (t) (t) h4 (t) 2 h1 (t) (t 2) h4 (t)
= 2 h1(t) h4 (t) 2 [ h1(t) h4 (t)]t =t 2
Using the techniques of the convolution algebra it follows that
h1(t) h4 (t) = h1(1)(t) h4(1) (t)
where the superscript (1) means differentiation and the superscript (-1) means integration.
We have
h1(1) (t) = u (1)(t) = (t)
and
t

h4(1) (t)
Therefore

2e u( )d = 2u(t) e d = 2u(t) e
0

]0 = 2(1 et )u(t)
t

h1(t) h4 (t) = (t) 2(1 e t )u(t) = 2(1 e t )u(t)

1/25/02

P16-29

R. A. DeCarlo, P. M. Lin

Replacing the above expression into the expression of h(t) it follows that

h(t) = 4(1 e t )u(t) 4 (1 e t )u(t)

]t =t 2 = 4(1 et )u(t) 41[ e(t 2)]u(t 2).

A picture of h(t) is sketched in the next picture.

SOLUTION 16.53.
(a) The overall impulse response of the configuration in figure P16.53 is
h(t) = [ h1(t) + h2 (t)] h3 (t)
By the distributive property of convolution it follows that
h(t) = h1 (t) h3(t) + h2 (t) h3(t) = (t) cos( t)u(t) + (t 1)cos( t)u(t)
By the sifting property of the delta function it follows that
h(t) = cos( t)u(t) + cos[ (t 1)]u(t 1).
A picture of h(t) is sketched in the next figure.

1/25/02

P16-30

R. A. DeCarlo, P. M. Lin

(b) The response y(t) is computed as

y(t) = h(t) u(t) = {cos( t)u(t) + cos[ (t 1)]u(t 1)} u(t)
Using the distributive property and the time shift property of convolution(see problem 16.22, part (a) ) we
have
y(t) = [cos( t)u(t)] u(t) + {cos[ (t 1)]u(t 1)} u(t) =
= [cos( t)u(t)] u(t) + {[cos( t)u(t)] u(t)}t =t 1
Using the techniques of convolution algebra, the convolution cos( t)u(t) u(t) is computed as

[cos(

t)u(t)] u(t) = [cos( t)u(t)](1) u (1)(t)

where the superscript (-1) means integration and the superscript (1) means differentiation.
We have

[cos(

t)u(t)]

(1)

cos(

)u( )d

Therefore

[cos(

)d = u(t)

[sin( t)]t0

sin( t)

sin( t)
sin( t)

t)u(t)] u(t) =
u(t) (t) =
u(t)

Hence the step response is

y(t) =

= u(t) cos(

sin( t)

u(t) +

sin[ (t 1)]

u(t 1) =

sin( t)

[ u(t) u(t 1)] .

SOLUTION 16.54.
(a) The overall impulse response of the configuration in figure P16.53 is

u(t) .

1/25/02

P16-31

R. A. DeCarlo, P. M. Lin

h(t) = [ h1(t) + h2 (t)] h3 (t)

By the distributive property of convolution it follows that
h(t) = h1 (t) h3(t) + h2 (t) h3(t) = (t) cos( t)u(t) + (t 3) cos( t)u(t)
By the sifting property of the delta function it follows that
h(t) = cos( t)u(t) + cos[ (t 3)]u(t 3).
A picture of h(t) is sketched in the next figure.

(b) The response y(t) is computed as

y(t) = h(t) u(t) = {cos( t)u(t) + cos[ (t 3)]u(t 3)} u(t)
Using the distributive property and the time shift property of convolution(see problem 16.22, part (a) we
have
y(t) = [cos( t)u(t)] u(t) + {cos[ (t 3)]u(t 3)} u(t)
= [cos( t)u(t)] u(t) + {[cos( t)u(t)] u(t)}t =t 3
Using the techniques of convolution algebra, the convolution cos( t)u(t) u(t) is computed as

[cos(

t)u(t)] u(t) = [cos( t)u(t)](1) u (1)(t)

where the superscript (-1) means integration and the superscript (1) means differentiation. We have

[cos(

t)u(t)]

(1)

cos(

Therefore

)u( )d = u(t) cos(

0

)d = u(t)

[sin( t)]t0

sin( t)

u(t) .

1/25/02

P16-32

[cos(

R. A. DeCarlo, P. M. Lin

sin( t)
sin( t)

t)u(t)] u(t) =
u(t) (t) =
u(t)

Hence the step response is

y(t) =

sin( t)

u(t) +

sin[ (t 3)]

u(t 3) =

sin( t)

[ u(t) u(t 3)] .

SOLUTION 16.55.
(a) The overall impulse response of the configuration in figure P16.55 is
h(t) = [ h1(t) + h2 (t) + h3(t) + h4 (t)] h5(t)
By the distributive property of convolution it follows that
h(t) = h1 (t) h5(t) + h2 (t) h5(t) + h3 (t) h5 (t) + h4 (t) h5(t) =
= (t) cos( t)u(t) + (t 1)cos( t)u(t)
(t 3) cos( t)u(t) (t 4) cos( t)u(t)
By the sifting property of the delta function it follows that
h(t) = cos( t)u(t) + cos[ (t 1)]u(t 1)
cos[ (t 3)]u(t 3) cos[ (t 4) ]u(t 4) .
A picture of h(t) is sketched in the next figure.

(b) The response y(t) is computed as

y(t) = h(t) u(t) = {cos( t)u(t) + cos[ (t 1)]u(t 1)
cos[ (t 3)]u(t 3) cos[ (t 4) ]u(t 4)} u(t)

1/25/02

P16-33

R. A. DeCarlo, P. M. Lin

Using the distributive property and the time shift property of convolution (see problem 16.22, part (a)) we
have
y(t) = [cos( t)u(t)] u(t) + {cos[ (t 1)]u(t 1)} u(t)
{cos[ (t 3)]u(t 3)} u(t) {cos[ (t 4)]u(t 4)} u(t) =
= [cos( t)u(t)] u(t) + {[cos( t)u(t)] u(t)}t =t 1
{[cos( t)u(t)] u(t)}t =t 3 {[cos( t)u(t)] u(t)}t =t 4

Using the techniques of convolution algebra, the convolution cos( t)u(t) u(t) is computed as

[cos(

t)u(t)] u(t) = [cos( t)u(t)](1) u (1)(t)

where the superscript (-1) means integration and the superscript (1) means differentiation. We have

[cos(

t)u(t)]

(1)

cos(

)u( )d

= u(t) cos(

)d = u(t)

[sin( t)]t0

sin( t)

u(t) .

Therefore

[cos(

sin( t)
sin( t)

t)u(t)] u(t) =
u(t) (t) =
u(t)

Hence the step response is

y(t) =

sin( t)

u(t) +

sin[ (t 1)]
=

sin( t)

u(t 1)

u(t 3)

sin[ (t 4)]

[ u(t) u(t 1)+ u(t 3) u(t 4)] .

SOLUTION 16.56.
(a) By definition

h(t) f (t) =

we have

h(t ) f ( )d

By making a change of variable

= t

h(t) f (t) =
By definition

sin[ (t 3)]

h( 1) f (t 1)d 1 =

f (t 1 )h( 1 )d 1

f (t) h(t) =

f (t )h( )d

From the above expressions we observe that

h(t) f (t) = f (t) h(t)
because and 1 are only variables of integration.

u(t 4)

1/25/02

P16-34

R. A. DeCarlo, P. M. Lin

(b) Using the definition of convolution we have

[h(t) f (t)] g(t) = h( ) f (t )d g(t) = h(

= [ h( ) f (t 1 )g( 1 )]d

) f (t 1 )d g( 1 )d 1 =

d 1

Changing the order of integration we have

[h(t) f (t)] g(t) = h(

) f (t 1 )g( 1 )d 1 d

By the definition of the convolution we have

f (t) g(t) =

f (t )g( )d

Therefore

[ f (t) g(t)]t =t
Hence

[h(t) f (t)] g(t) =

f (t )g( )d

{h( ) [ f (t) g(t)]t =t }d

By the definition of the convolution product we have

h(t) [ f (t) g(t)] =
Therefore

{h( ) [ f (t) g(t)]t =t }d

[h(t) f (t)] g(t) = h(t) [ f (t) g(t)] .

Thus the associative property of convolution is proved.

SOLUTION 16.57.
We have
f (t) h(t) =

f (t) (t kT )

k=0

For some nonnegative k , the Laplace transform of f (t) (t kT) is

L[ f (t) (t kT)] = f (kT)e skT

1/25/02

P16-35

R. A. DeCarlo, P. M. Lin

by the sifting property of the delta function. Therefore we have

L[ f (t)h(t)] = L f (t) (t kT) = = f (kT )e skT

k =0

k =0

Using the notation z = e sT we have

L[ f (t)h(t)] =

f (kT)z k .

k =0

SOLUTION 16.58.
(a) By the voltage division formula we have
Vout (s) =

1
Cs

1 Vin (s)
R+
Cs

Therefore the transfer function of the circuit is

1
Vout (s)
1
1
H(s) =
= Cs1 =
=
.
Vin (s) R +
CRs + 1 2s + 1
Cs
Taking the inverse Laplace transform yields
1
1
-1 2
0.5t
h(t) = L-1
=
=
L
u(t).

= 0.5e

1
2s + 1
s +

2
(b) By the impulse response theorem it follows that
vout (t) = h(t) vin (t) =

= 0.5e 0.5t u(t) [ (t) + (t 1) + (t 2) + K]

Using the sifting property of the delta function it follows that

vout (t) = 0.5e 0.5t u(t) + 0.5e 0.5(t 1) u(t 1) + 0.5e 0.5(t 2) u(t 2) + K V.
Therefore for 0 < t < 1

vout (t) = 0.5e 0.5t V

because only u(t) is nonzero for 0 < t < 1.

(c) From the expression of vout (t)obtained in the part (b) it follows that

1/25/02

P16-36

R. A. DeCarlo, P. M. Lin

vout (t) = 0.5e 0.5t + 0.5e 0.5(t 1) V, for 1 < t < 2

because only u(t) and u(t 1) are nonzero for 1 < t < 2.
(d) For t in the interval (4,5) , only u(t) , u(t 1), u(t 2) , u(t 3) and u(t 4) are nonzero. Therefore
vout (t) = 0.5e 0.5t + 0.5e 0.5(t 1) + 0.5e 0.5(t 2) + 0.5e 0.5(t 3) + 0.5e 0.5(t 4) V.
The above expression can be written as

) (

) (

) (

2
3

vout (t) = 0.5e 0.5(t 4) 1+ e 0.5 + e 0.5 + e 0.5 + e 0.5

(e) Using the sum formula for geometric series with

) (

Therefore

1 e 0.5

) + (e ) + (e ) = 1 e0.5
0.5 5
1
e
(
) V.
(t) = 0.5e 0.5(t 4)

1 + e 0.5 + e

vout

= e 0.5 we have

0.5 2

0.5 3

0.5 4

V.

1 e 0.5

vout (t) = 8.6189 e 0.5t V.

(f) Using the expression of vout (t)obtained in part (b) it follows that, for n < t < n + 1,
vout (t) = 0.5e 0.5t + 0.5e 0.5(t 1) + 0.5e 0.5(t 2) + K + 0.5e 0.5(t n) =
2
n

= 0.5e 0.5(t n) 1+ e 0.5 + e 0.5 + K + e 0.5 V.

) (

Using the sum formula for geometric series we have

(g) For large n, e 0.5

n+1

1 e 0.5
0.5(t n)
vout (t) = 0.5e

1 e 0.5
n+1

V.

0. Therefore, for large n we have

vout (t) e 0.5(t n)

0.5
V for n < t < n + 1.
1 e 0.5

A picture of vout (t), for large n, is sketched in the next figure.

1/25/02

P16-37

R. A. DeCarlo, P. M. Lin

SOLUTION 16.59.
By the impulse response theorem we have

vout (t) = h(t) vin (t) = 0.5e 0.5t u(t) [ (t) + (t + 1)+ (t + 2) + K] V.
Using the sifting property of the delta function it follows that
vout (t) = 0.5e 0.5t u(t) + 0.5e 0.5(t +1)u(t + 1) + 0.5e 0.5(t + 2) u(t + 2) + K V.
For 0 < t we have

vout (t) = 0.5e 0.5t + 0.5e 0.5(t +1) + 0.5e 0.5(t + 2) + K =

2

= 0.5e 0.5t 1+ e 0.5 + e 0.5 + K V.

) (

Using the sum formula for geometric series( for n = ) we have

vout (t) = 0.5e 0.5t

1
V for 0 < t.
1 e 0.5

Simplifying the expression of vout (t) it follows that

vout (t) = 1.2707 e 0.5t V for 0 < t.

SOLUTION 16.60. First we plot for reader convenience vin (t) and its staircase approximation.

1/25/02

P16-38

R. A. DeCarlo, P. M. Lin

vin and its Staircase Approximation

2.8
2.6
2.4
2.2
2
1.8
1.6
1.4
1.2
1
0

0.1

0.2

0.3

0.4

0.5

0.6

The transfer function of the circuit of figure 16.58 is H(s) =

0.7

0.8

0.9

1
in which case
2s + 1

h(t) = 0.5e 0.5t u(t). Because we only want the output for 0 t 2, we only need h(t) for 0 t 2 s.
Hence we need to generate staircase approximations to both vin (t) and h(t) as follows:
t = 0:0.05:2;
vin = exp(t .^2) .* (u(t) - u(t - 1));
h = 0.5*exp( - 0.5*t) .* (u(t) - u(t - 2));
T = 0.05;
tstep = T;
y = [0 conv(vin,h)*tstep 0];
t = 0:tstep:tstep*(length(vin)+length(h));
% For plotting through time 2 s we set
t=t(1:41);
y = y(1:41);
plot(t,y)
grid

1/25/02

P16-39

R. A. DeCarlo, P. M. Lin

Convolution of vin and h

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.2

0.4

0.6

0.8

1.2

1.4

1.6

1.8

5/15/01

P17-1

R.A. DeCarlo & P.M. Lin

CHAPTER 17 P ROBLEM SOLUTIONS

SOLUTION PROBLEM 17.11. Case 1: suppose R1 > R2 . From example 17.3, page 696, if L
and C are connected as indicated in part (a), then Z1 can be made real and larger than R2 . This
means we can solve the problem at least for Z1 . Specifically, consider the figure below

From example 17.3, at a specified frequency, r, for which Z1 is real, then L, C, and R2 must
satisfy,
1

LC

r =

R22
L2

(1)

Further from example 17,3, at r,

Z1( j r ) =

L
R2C

We require that Z1( j r ) = R1 in which case

L
(2)
C
It is necessary to solve equations (1) and (2) simultaneously for L and C. From (2), L = R1R2 C .
R1R2 =

Substituting into the square of (1) yields

2r =

1
R22

R1R2C 2 R12 R22 C2

Hence
C=
It follows that

1
r R1

R1 R2
R2

5/15/01

P17-2

1
L = R1R2
r R1

R1 R2
1
=

R2 r

R.A. DeCarlo & P.M. Lin

R2 (R1 R2 )

Observe that since R1 > R2 , both C and L are real, i.e., exist. Please note that this connection would
not result in real values of C and L had R1 < R2 . If we can now show that Z2 = R 2 , then parts (a)
and (b) are valid for this case.
By direct computation
1
R1
R1
Z2 (j r ) = j r L +
=
j
L
+
=
j
R
(
R

R
)
+
r
2
1
2
1
R R2
j r CR1 + 1
j r C +
j 1
+1
R1
R2

= j R2 ( R1 R2 ) +

R1 R2
R2
R1 R2
=
j
R
(
R

R
)
+
R

jR
= R2
2
1
2
2
2
R1 R2
R
2
1+
R2

R1 jR1

Thus, (a) and (b) are true for the case R1 > R2 .
We can also arrive at the conclusion that Z2 = R2 using maximum power transfer concepts.
Since Z1 is constructed so that Z1 = R1 , we have set up the conditions for maximum power transfer
of a V-source in series with R1 to the "load" Z 1 . Since the LC coupling network is lossless,
whatever average power is received by the network to the right of R1 , will be dissipated by R2 .
Therefore maximum power is transferred to the load R2 . Looking back from R2 , it must be that R2
sees a Thevenin resistance Z2 = R2 since it is known that there is a non-zero R1 .
Case 2, R 1 < R 2 . Now consider the configuration

5/15/01

P17-3

R.A. DeCarlo & P.M. Lin

Interchanging the subscripts of 1 and 2 in case 1 produces the derivation for this case.
Combining cases 1 and 2 using the text notation produces
C=

1
Rlarge

Rlarge Rsmall
Rsmall

L=

SOLUTION PROBLEM 17.16.

(a)
C = 0.1e-6;
R = 510;
wr = 2*pi*1.56e3;
%From equation 17.4 in example 17.3 which analyzes
%this particular circuit,
%L must satisfy
% wr^2*L^2 - L/C + R^2 = 0.
% Therefore

L = roots([wr^2 -1/C R^2])

L=
5.3133e-02
5.0952e-02
(b) From HW problem 17.9,

( r )2
Hence, in MATLAB,
Linv = C*(wr^2 + 1/(R^2*C^2))
Linv =
4.8054e+01
L = 1/Linv
L = 2.0810e-02

1
R
R
Rsmall
small large

1
1
2 2
LC R C

5/15/01

P17-4

R.A. DeCarlo & P.M. Lin

SOLUTION PROBLEM 17.22. (Correction: change RL in figure P17.22 to R.) We solve this
problem using phasors rather than the Laplace transform approach as it is simpler. Consider

As such, since by definition 0 = 1

LC , the resonant frequency,

VC = VL = Zin ( j 0 )Iin = RIin = RIm

Hence
vC (t) = vL (t) = RI m cos( 0t) and iL (t) =

RIm
sin( 0 t).
0 L

(a) By direct computation

wC (t) =

1 2
1
2
CvC (t) = CR2 Im
cos2 ( 0 t)
2
2

(b) By direct computation

w L (t) =

2
1 2
R2 Im
LiL (t) =
sin 2 ( 0t)
2
2
20 L

(c) Also by direct computation, since 20 = 1 LC ,

wC (t) + wL (t) =
=

2
1
R2 Im
2
2
CR2 Im
cos2 ( 0 t) +
2 sin ( 0t)
2
2 0 L

1
1 2 2 2
1
2
R ImC cos ( 0t) + 2
sin 2 ( 0t) = R2 Im
C
2

0 LC
2

(d) The energy dissipated in the resistance in one period is

T

w R (0,T ) = R

iR2 (t) dt =

iin2 (t) dt =

2
RIm

cos
0

( 0t)dt

5/15/01

P17-5

R.A. DeCarlo & P.M. Lin

2
2
2
2
RIm
RIm
RIm
RIm
dt
+
cos(2
t)
dt
=
T
=
0
2
2
2
0
0
0

(e) Finally

maximum energy stored

0.5R2 I 2mC
= 2
= 0 RC = Q
total energy lost per period
RI 2m
0

by equation 17.13.

SOLUTION PROBLEM 17.27.

1
s
IL
1
1
L
(a) H1(s) =
=
=
=
R
1
Vs Zin (s) R + Ls + 1
s2 + s +
Cs
L
LC
s
1
1

1
1
1 VC
L Cs
LC
H2 (s) = H1(s)
= IL =
=
=
R
1
R
1
2
2
Cs Vs
Cs
Vs s + s +
s + s+
L
LC
L
LC
(b) H1(s) is precisely of the form of equation 17.18 with a single zero at the origin. It follows
that 2p =

1
R
and 2 p = . Hence, from equation 17.19, m = p =
LC
L

B = 2 p =

p
R
, and Qcir = Qp =
=
L
2 p

1
1
, H1( j m ) = ,
LC
R

1
L 1 L
=
.
LC R R C

(c) With s = j,
2
1
LC

H2 ( j) =

1 2 + R
LC

L
2

(1 LC2 )

+ ( RC)2

Instead of maximizing H2 ( j) we minimize its reciprocal, through differentiation. Let

5/15/01

P17-6

f (C) =

1
H2 ( j)

R.A. DeCarlo & P.M. Lin

2
2 = 1 LC

+ ( RC )2

Then

f '(C) = 2C(R)2 + 2 2 LC 1 2 L = 0
implies that

L
1
1
1
= 1

C= 2
=
L 2 1 + 1
R (L)2 L 2 R2
2 2 + 1
Q2

coil

If the coil has high Q, then

1
C
L 2
in which case
V
L
H2 ( j) = C
= Qcoil
Vs
R
Therefore
VC max Qcoil Vs

SOLUTION PROBLEM 17.31. Here

a + j
H( j) 2 = K 2
( p 2 ) + j2 p

= K2

( )

a2 + 2
2
2
2
2 2
2 2 K f
( p ) + 4 p

(a) Differentiating with respect to 2 and setting the derivative equal to zero yields

0=

( )=

d f 2
d 2

d
a2 + 2

d 2 ( 2p 2 )2 + 4 2p 2

5/15/01

P17-7

R.A. DeCarlo & P.M. Lin

)(

a 2 + 2 2( 2p 2 ) + 4 2p
1
= 2
2 2
2 2
2
2 2
2 2 2
( p ) + 4 p
( p ) + 4 p

Given that the denominator of the first term is non-zero, this is equivalent to

)(

2 2

2 2

0 = ( 2p 2 )2 + 42p 2 a 2 + 2 2( 2p 2 ) + 42p

( 2)

+ p 2 p + 4 p + 2 a +

)( 2p 2 ) 4(a 2 + 2 )2p

( )2 + 4p 2 2p 2 + 4 2p 2 + 2a 2 2p 2a2 2 + 22p 2 2( 2 )2 4a2 2p 4 2p 2

= 2

( )2 2a 2 2 + 4p + 2a2 2p 4a 2 2p

= 2

Hence

( 2 )2 + 2a 2 2 + a 4 = 4p + 2a2 2p 4a2 2p + a 4
where we have added a 4 to produce perfect squares, i.e.,

( 2 + a2 )2 = ( 2p + a2 )2 (2a p )2
This implies that
2

= a

( 2p + a2 )2 (2a p )2

Thus, to achieve a real positive solution we obtain,

m = a +

( 2p + a 2 )2 (2a p )2

(b) Here, as in part (a), the arithmetic is simpler if we deal with 1/H(s) rather than H(s).
Specifically
s2 + 2 ps + 2p s2 + 2 p (s + a) 2 pa + 2p 2 p s2 2 pa + 2p
1
=
=
=
+
H(s)
K(s + a)
K(s + a)
K
K(s + a)

5/15/01

P17-8

R.A. DeCarlo & P.M. Lin

The problem asks for a non-zero value of . For zero phase shift of H(j), Im[H(j)] = 0
Im[H(j)] > 0. This necessarily requires that the imaginary part of the above expression at s = j
must be zero. Thus

2 2 pa + 2p (a j)
2 2 pa + 2p
1

= Im
Im
= Im
2
2

H(
j)
K(
j
+
a)

K(
+
a
)

2 2 pa + 2p )()
(
=
=0
K( 2 + a 2 )

The solutions to this are dc, i.e., = 0, and

(for zero phase shift) = 2p 2 pa
Note that such a frequency may not exist if the quantity under the radical is negative.

SOLUTION PROBLEM 17.40. (a) Compute the transfer function:

Rs
s
+
V
1
1
1
L
H(s) = out =
=
=
1
1
Iin
Yin(s)
C
R
1
1 Rs
+ Cs +
s2 + s +
s+
+ 1
Ls + Rs
Rp

L R pC Rp
LC
s + 1.333 103
= 8 105 2
s + 2000s + 1.0006 20
where 20 = 1.6 10 9 and 2p = 1.0006 20 1.60089 109 and p = 4.001 10 4 rad/sec.
p
R
(b) Qp =
= 20 which is high Qp . Further a = s = 1.333 103 << p = 40.011109 103 .
2 p
L
(c) Equation 17.28,

5/15/01

P17-9

m = a +

R.A. DeCarlo & P.M. Lin

( 2p + a 2 )2 (2a p )2

gives the exact value of m. In MATLAB,

wmsqrd = -a^2 + sqrt((wp^2+a^2)^2 - (twosig*a)^2)
wmsqrd = 1.600886670362673e+09
wm = sqrt(wmsqrd)
wm = 4.001108184444246e+04
Clearly, m approximates p . Finally,
H( j m ) H( j p ) = 8 10 5

j p + 1.333 10 3

8 105
1.333 103
=
1
j
2000
p
2p + j2000 p + 2p

400
(d) From equation 17.30,
B 2 p = 2000 rad/sec, 1 m 0.5B = 39.011 103 rad/sec, and
2 m + 0.5B = 41.011 103 rad/sec.
(e) From equation 17.30,

Qcir Qp =

Qcir
Hence,

Rs
1
+
L RpC
0

p
2 p

Rs
1
R + 1 LC
p

Rs
1
L + R C

Rs
1
L + R C

Qcoil + Qcap
Rs
1
1
1
+
=
+
=
L 0 Rp C 0 Qcoil Qcap
Qcoil Qcap

5/15/01

P17-10

Qcir

R.A. DeCarlo & P.M. Lin

QcoilQcap
Qcoil + Qcap

SOLUTION PROBLEM 17.65. Here we consider the equivalent circuit valid for t > 0. Note that
-

iL(0 ) = iL(0 ) = E/Rs and vC(0 ) = vC(0 ) = 0. Hence

It follows that
1
E
1
E
VC (s) = 1 1

1
1
CRs
+
+ Cs Rs s s2 +
s+
R Ls
RC
LC
Complex roots occur when
2
2
1 4 = 1 4 2 < 0
0
RC
LC RC

Equivalently,
0 RC >

1
2

as was to be shown. Further, since vC(0 ) = 0, the general form of the capacitor voltage for constant
excitation is (as per chapter 10)
vC (t) = et ( Acos( d t) + Bsin( d t)) = Bet sin( dt)
From the characteristic equation, the complex roots are
1
1 1 2 0
1
j d =
j

j
1
0
2RC
LC
2RC
2Q
4Q2
Here

5/15/01

P17-11

a==

0
2Q

d = 0 1

R.A. DeCarlo & P.M. Lin

1
4Q2

Letting B = Vm, for the appropriate value of Vm, we obtain the desired result.
1 2Q 2Q
(b) If Q is large, d 0 . The Vm e at will drop to 1/e of its peak value in t = =

a 0 d
2
seconds. The period of oscillation of the damped sinusoid is
. Therefore the number of cycles
d
contained in this interval is
2Q
d Q
2 =
d

5/31/01 Mag Crt Probs

P18-1

R. A. DeCarlo & P.M. Lin

CHAPTER 18 P ROBLEM SOLUTIONS

SOLUTION PROBLEM 18.34. There is a correction to this problem: set M = 3 H.

(a) The stored energy at t = 0 is:
W(0) = 0.5L1i12 (0) + 0.5L2i22 (0) + Mi1 (0)i2 (0) = 8 J
L1 = 10; L2 = 2; M = 3; i10 = 1; i20 = -3;
W0 = 0.5*L1*i10^2 + 0.5*L2*i20^2 + M*i10*i20
W0 = 5
(b) Writing two differential mesh equations we obtain
di
di
di
di
L1 1 + M 2 + R1i1 = 10 1 + 3 2 + i1 = 0
dt
dt
dt
dt
and
di2
di
di
di
+ M 1 + R2i2 = 2 2 + 3 1 + i2 = 0
dt
dt
dt
dt

L2

Taking the Laplace transform of these equations yields

and

10sI1 10i1 (0) + 3sI2 3i2 (0) + I1 = (10s + 1)I1 + 3sI2 10i1(0) 3i2 (0) = 0
2sI2 2i2 (0) + 3sI1 3i1(0) + I2 = (2s + 1)I2 + 3sI1 3i1(0) 2i2 (0) = 0

Putting these equations in matrix form yields

3s I1 10i1(0) + 3i2 (0) 1
(10s + 1)
=
=
3s
(2s + 1) I2 3i1(0) + 2i2 (0) 3

Solving yields
3s 1 1
3s 1
I1 (10s + 1)
1
2s + 1
=
I = 3s

(2s + 1) 3 11s2 + 12s + 1 3s 10s + 1 3

2
=

1
11s + 1
1 1
=

(11s + 1)(s + 1) 33s 3 s + 1 3

Therefore, by inspection,
i1(t) = e tu(t) A and i2 (t) = 3e tu(t) A
Remark: normally, i1(t) and i2(t) would have two exponential terms present. Because of the special
choice of initial conditions, a pole cancelled out.
(c) From equation 18.24 with the lower limit changed to zero and the upper limit changed to , we have

5/31/01 Mag Crt Probs

P18-2

W(0,) =

R. A. DeCarlo & P.M. Lin

(v1i1 + v2i2 ) dt = 0.5L1i12 () + 0.5L2i22 () + Mi1()i2 ()

0.5L1i12 (0) + 0.5L2i22 (0) + Mi1(0)i2 (0)

From part (b) all currents at t = are zero, hence

W(0,) = 0.5L1i12 (0) + 0.5L2i22 (0) + Mi1(0)i2 (0) = 5 J
The result of part (a) indicates that the initial store energy is 5 J. The result of part (c) indicates that the
energy returned to the circuit is also 5 J, i.e., the total energy accumulated in the inductors over [0,) is
5 J. Hence 5 J is dissipated in the resistors.

Remark: the interested student might computer the integral

(R1i1 (t) + R2i2 (t))dt , the actual energy

2

dissipated in the resistors over [0,), and show that this is 5 J.

SOLUTION PROBLEM 18.35.

(a)
L1 = 4; L2 = 9; M = 3;
I1 = 2; I2 = -3;
W = 0.5*L1*I1^2 + 0.5*L2*I2^2 - M*I1*I2
W=
6.6500e+01
(b)
K = 0.5*L1*I1^2
K= 8
% Minimize (over I2) K + 0.5*9*I2^2 - 3*2*I2
% Take Derivative and set to zero; then solve for I2.
% Derivative is: 9*I2 6 = 0
% The result is I2 = 2/3 A.
I2 = 2/3;
Wmin = 0.5*L1*I1^2 + 0.5*L2*I2^2 - M*I1*I2
Wmin = 6
(c)

5/31/01 Mag Crt Probs

(d)
L1 = 4; L2 = 9; M = 3;
k = M/sqrt(L1*L2)
k = 5.0e-01

SOLUTION PROBLEM 18.37.

k = 0.5;
L1 = 9; L2 = 4;L3 = 1;
M = 0.5*sqrt(L1*L2)
M= 3
Lcpld = L1 + L2 + 2*M
Lcpld = 19
Leq = Lcpld + L3
Leq = 20
Imax = 2;
Wmax = 0.5*Leq*Imax^2
Wmax = 40 J

SOLUTION PROBLEM 18.41.

RL = 100; Rs = 300e3; R = 10e3;
m = 20; n = 5;
(a)
Z2 = RL*m^2
Z2 =
40000
Req1 = Z2*R/(Z2+R)
Req1 = 8000
Z1 = Req1*n^2
Z1 = 200000

P18-3

R. A. DeCarlo & P.M. Lin

5/31/01 Mag Crt Probs

P18-4

R. A. DeCarlo & P.M. Lin

(b)
%Gv1 = v1/vin
%v1 = [Z1/(Rs + Z1)]vin
Gv1 = Z1/(Rs+Z1)
Gv1 = 4.0000e-01
% Gv2 = v2/vin
% Gv2 = v2/v1 * v1/vin = (1/n) * G1
Gv2 = G1/n
Gv2 = 8.0000e-02
% Gv3 = v3/vin
Gv3 = -Gv2/m
Gv3 = -4.0000e-03
(c)
% Gi2 = i2/iin
Gi2 = n*R/(R+Z2)
Gi2 = 1
% Gi3 = i3/iin = i3/i2 * i2/iin = m*Gi2
Gi3 = -m*Gi2
Gi3 = -20

SOLUTION PROBLEM 18.55. (a) The parameters in the circuit of figure P18.55b are given by
equations in figure 18.22b. Specifically, since k = M L1 L2 = 0.16 3.5 0.008 = 0.95618
M = 0.16; L1 = 3.5; L2 = 0.008;
k=M/sqrt(L1*L2)
k=
9.5618e-01
La = (1 - k^2)*L1
La =
3.0000e-01
Lb = k^2 * L1
Lb =
3.2000e+00
N =M/L2
N=
20
(b)
R = 500;
w = 2*pi*60;
Vseff = 110;
Zin = R + j*La*w + j*Lb*w
Zin =
5.0000e+02 + 1.3195e+03i
Iseff = Vseff/Zin
Iseff =
2.7624e-02 - 7.2899e-02i
Pave = R*abs(Iseff)^2

5/31/01 Mag Crt Probs

P18-5

Pave =
3.0387e+00
(c)
Zin2 = R + j*La*w
Zin2 =
5.0000e+02 + 1.1310e+02i
Iseff2 = Vseff/Zin2
Iseff2 =
2.0929e-01 - 4.7341e-02i
Is2mag = abs(Iseff2)
Is2mag =
2.1458e-01
% The current in the secondary is (in A):
Isecmag = Ismag*N
Isecmag =
4.2916e+00
(d)
% Our first step is to compute the reflected impedance:
Zrefl = 100*N^2
Zrefl =
40000
% We now compute the impedance of the parallel combination
% of Lb and Zrefl denoted Zpar
Zpar = 1/(1/Zrefl + 1/(j*w*Lb))
Zpar =
3.6350e+01 + 1.2053e+03i
% We now compute the input impedance:
Zin = R+j*w*La + Zpar
Zin =
5.3635e+02 + 1.3184e+03i
% Now we compute the voltage across the primary of the
% ideal transformer, by voltage division:
Vpar = Vseff*Zpar/Zin
Vpar =
8.7342e+01 + 3.2500e+01i
% Now we compute the voltage across the load:
Vload = Vpar/N
Vload =
4.3671e+00 + 1.6250e+00i
Vloadmag = abs(Vload)
Vloadmag =
4.6596e+00
Iloadmag = Vloadmag/100
Iloadmag =
4.6596e-02

R. A. DeCarlo & P.M. Lin

5/31/01 Mag Crt Probs

P18-6

R. A. DeCarlo & P.M. Lin

SOLUTION PROBLEM 18.65. (a) The equivalent circuit accounting for initial conditions is given
below:

(b) From the definition of coupled inductors

0.1472s I1'
V1 0.6s
=
V 0.1472s 0.1472s '
2
I2
(c) Hence

I1' 0.6s
0.1472s 1 V1 15 0.1472
'=
=
I2 0.1472s 0.1472s V2 s 0.1472

0.1472 V1
0.6 V2

(d) Writing nodal equations we obtain,

Vs 2 1 V1 I1 2 1 V1 1 i1 (0 ) I1'
+
0 = 1 2 V + I = 1 2 V + s

2 2
2
i2 (0 ) I'2
(e) Now we substitute our result of part (c):
Vs 2 1 V1 1 2 1 2.208 2.208 V1
0 = 1 2 V + s 2 + s 2.208
9 V2

2

which simplifies to

Vs 1 2 1 2s + 2.208
0 s 2 = s (s + 2.208)

(s + 2.208) V1
2s + 9 V2

or equivalently
2s + 2.208 (s + 2.208) V1 sVs + 2
=
(s + 2.208)
2s + 9 V2 2

(f) Solving these equations we obtain:

5/31/01 Mag Crt Probs

P18-7

R. A. DeCarlo & P.M. Lin

1
V1 2s + 2.208 (s + 2.208) sVs + 2
V = (s + 2.208)
2s + 9 2
2

(s + 2.208) sVs + 2
1
2s + 9

3s2 + 18s + 15 (s + 2.208) 2s + 2.208 2

(g) If vs (t) = 10u(t) V, then

(s + 2.208) 12
V1
1
2s + 9
1
26s + 112.42
V = 2
(s + 2.208) 2s + 2.208 2 = 2

2 3s + 18s + 15
3s + 18s + 15 16s + 30.912
From MATLAB
[r,p,k]=residue([16 30.912], [3 18 15])
r=
4.0907e+00
1.2427e+00
p=
-5
-1
k=
[]
Therefore,

v2 (t) = 4.0907e 5t + 1.2427e t u(t) V

SOLUTION PROBLEM 18.67. The solution to this problem is based upon the following: (i) Leq = L1 +
L2 + 2M for series aiding connection (see example 18.4) and (ii) k = M L1 L2 (a definition), and (iii) k
= 1 (an assumption).
(a) Given L1 = L2 = L and k = 1, Leq = L1 + L2 + 2M = L + L + 2k L2 = 4L . Hence, when the
number of turns is doubled, the inductance is quadrupled.
(b) For this part, let us first consider L2 which has 2N turns. We can view L2 as two coils of N turns
each connected in series aiding with coupling coefficient k = 1. Hence, according to part (a), the
inductance of L2 is four times that of L1 which only has N turns. Hence,
Leq = L1 + L2 + 2M = L + 4L + 2k 4L2 = 9L
2

Observe that the coil has 3N turns yielding an inductance of 9L = 3 L.

(c) Suppose coil 1 and coil 2 consist of one turn each. Here the total number of turns is 2N where N =
1 turn. Suppose further that L1 = L. From part (a), Leq = 4L = (2)2 L . Now suppose coil one consists
of one turn and coil 2 consists of M turns. We assume here as an induction hypothesis that

5/31/01 Mag Crt Probs

P18-8

R. A. DeCarlo & P.M. Lin

Leq = ( M + 1)2 L
We must show that if coil 2 has (M+1)N turns then,
Leq = ( M + 2)2 L
Our first step is to compute the equivalent inductance of coil 2. However, coil 2 consists of a single turn
coupled to an M-turn coil, which by the induction hypothesis means that
L2 = ( M + 1)2 L
Thus coil 1 in a series aiding connection with L2 leads to
Leq = L1 + L2 + 2M = L + (M + 1)2 L + 2 (M + 1)2 L2 = L + ( M + 1)2 L + 2( M + 1) L

= L ( M + 1)2 + 2( M + 1) + 1 = L[( M + 1) + 1] = ( M + 2)2 L

2

Given this relationship, if coil 1 consists of N1 turns, and one turn has an inductance L, then
L1 = ( N1 ) L . Similarly, L2 = ( N2 ) L , and M = k L1 L2 = L1L2 =
immediately follows that
L1 : L2 : M = N12 : N22 : (N1N2)
2

(N1)2 ( N2 )2 L2

= N1N 2 L . It

SOLUTION PROBLEM 18.70. (a) k = M / L1L2

M = 1.5; L1 = 1.5; L2 = 6;
k = M/sqrt(L1*L2)
k = 5.0000e-01
(b) For this part and the remaining parts consider the following equivalent circuit where the coupled
coils have been replaced by the model of figure 18.22(b).

La = (1 - k^2)*L1
La =
1.1250e+00
Lb = k^2 * L1
Lb =
3.7500e-01

5/31/01 Mag Crt Probs

P18-9

R. A. DeCarlo & P.M. Lin

% The turns ratio is M:L2, i.e.,

1.5/6
ans = 2.5000e-01
% Therefore the turns ratio is 1:4. It follows that
Rb = 200*(1/4)^2
Rb =
1.2500e+01
To compute Z(s) we have,
Z(s) = 20 +

1
RL s
12.5s
1
+ Las + b b = 1.125s + 20 +
+
100 Cs
Cs
Lbs + Rb
s+
3

Hence,

j12.5
j
1
416.67
Z( j) = j1.125 + 20 +

=
20
+
+
j
1.125

4
100 C
C
10
10 4
2

2
j +
+

9
9
12.5 2

(c) For this part we need to make the imaginary part of Z(jw) real. To this end:
K1 = 12.5*100/3
K1 = 4.1667e+02
w = 1333;
K2 = 1.125*w + K1*w/(w^2 + 1e4/9)
K2 = 1.4999e+03
C = 1/(K2*w)
C = 5.0015e-07
Hence, we take C = 5 F.
(d) At resonance, we have
12.5 2r
Z( j r ) = Z( j1333) = 20 +
= 32.5
10 4
2
r +
9
and
Zb (j r ) =

j12.5 r
j16,662
=
100 33.33 + j1333 = 12.492 + j0.31238
jr +
3

By voltage division
Vout ( j r ) Vout Vb 4 Zb ( j r )
=

=
= 1.5379 + j0.038456
Vs ( j r )
Vb
Vs 1 Z( j r )
Zb = j*12.5*w/(j*w + 100/3)
Zb =
1.2492e+01 + 3.1238e-01i
Zwr=20 + 12.5*w^2/(w^2 + 1e4/9)
Zwr =

5/31/01 Mag Crt Probs

P18-10

R. A. DeCarlo & P.M. Lin

3.2492e+01
Gv = 4*Zb/Zwr
Gv =
1.5379e+00 + 3.8456e-02i
MagGv = abs(Gv)
MagGv =
1.5384e+00
AngGv = angle(Gv)*180/pi
AngGv =
1.4325e+00

SOLUTION PROBLEM 18.71.

(a) Following the hint we apply a source transformation to obtain

Writing two mesh equations we obtain the following matrix form of the mesh equations:
ks
s + + 1 s

ks
s ++ 1

I1 Iin s
=
s I2 0

Solving for I2 yields

s + + 1 s Iin s
det
ks
0
kIin

I2 =
=
ks
s + + 1 s
(s + + 1 s)2 (ks )2
det
ks
s + + 1 s

To find H(s) we have

V
I s
ks
ks
H(s) = out = 2 =
=
2
2
Iin
Iin
( s + + 1 s) ( ks) (s + ks + + 1 s)(s ks + + 1 s)
=

[(1+ k)s

ks
2

][

+ s + 1 (1 k)s 2 + s + 1

5/31/01 Mag Crt Probs

P18-11

(b) R = 0.02 , b = R = 0.02 , Q = 50, and k = 0.01, 0.02, and 0.04.

beta = 0.02; k1 = 0.01; k2 = 0.02; k3 = 0.04;
p11 = roots([(1+k1) beta 1]);
p12 = roots([(1-k1) beta 1]);
p1 = [p11;p12]
p1 =
-9.9010e-03 + 9.9499e-01i
-9.9010e-03 - 9.9499e-01i
-1.0101e-02 + 1.0050e+00i
-1.0101e-02 - 1.0050e+00i
p21 = roots([(1+k2) beta 1]);
p22 = roots([(1-k2) beta 1]);
p2 = [p21;p22]
p2 =
-9.8039e-03 + 9.9010e-01i
-9.8039e-03 - 9.9010e-01i
-1.0204e-02 + 1.0101e+00i
-1.0204e-02 - 1.0101e+00i
p31 = roots([(1+k3) beta 1]);
p32 = roots([(1-k3) beta 1]);
p3 = [p31;p32]
p3 =
-9.6154e-03 + 9.8053e-01i
-9.6154e-03 - 9.8053e-01i
-1.0417e-02 + 1.0206e+00i
-1.0417e-02 - 1.0206e+00i
(c)
f = 0.14:.0001:.18;
n1 = [k1/(1-k1)^2 0];
n2 = [k2/(1-k2)^2 0];
n3 = [k3/(1-k3)^2 0];
h1 = freqs(n1, poly(p1), 2*pi*f);
h2 = freqs(n2, poly(p2), 2*pi*f);
h3 = freqs(n3, poly(p3), 2*pi*f);
plot(f, abs(h1), f, abs(h2), f, abs(h3))
grid

R. A. DeCarlo & P.M. Lin

5/31/01 Mag Crt Probs

P18-12

R. A. DeCarlo & P.M. Lin

(d) An inspection of [f, abs(h2)] (i.e., a tabulation of the values) in part (c) indicates that fpeak = 0.15865
Hz and Hpeak = 26.148 . The frequency scale factor Kf is defined according to:
Kf = 455e3/fpeak
Kf = 2.8679e+06
Further,
Km = Kf*2.35e-3
Km = 6.7397e+03
Lnew = Km*1/Kf
Lnew =
2.3500e-03
Cnew = 1/(Km*Kf)
Cnew =
5.1736e-11
Rnew = Km*0.02
Rnew =
1.3479e+02
The 3 dB down value of h2 is Hpeak/sqrt(2). Hence

5/31/01 Mag Crt Probs

P18-13

R. A. DeCarlo & P.M. Lin

Hmax = max(abs(h2))
Hmax =
2.6148e+01
H3db = Hmax/sqrt(2)
H3db =
1.8490e+01
Again, inspecting the tabulated values indicates that the 3 dB frequencies are: f1 = 0.1569 Hz
f2 = 0.1614 Hz. Finally
Bf = f2 - f1
Bf = 4.5000e-03 Hz
Bfnew = Kf*Bf
Bfnew = 1.2906e+04 Hz

and

(e) For this part we redo part (a) with R, L, and C as literals.
Writing two mesh equations we obtain the following matrix form of the mesh equations:
Ls + R + 1 Cs

kLs

kLs
I1 Iin Cs
=
Ls + R + 1 Cs I2 0

Solving for I2 yields

Ls + R + 1 Cs Iin Cs
det
kLs
0
kL C

I2 =
=
I
Ls
+
R
+
1
Cs
kLs

( Ls + R + 1 Cs)2 (kLs)2 in
det
kLs
Ls + R + 1 Cs

To find H(s) we have

V
I Cs
kL C 2s
kL C2 s
H(s) = out = 2
=
=
Iin
Iin
( Ls + R + 1 Cs)2 (kLs)2 ( Ls + kLs + R + 1 Cs)( Ls kLs + R + 1 Cs)
ks
kLs
LC2
=
=
2
2
LC(1+ k)s + CRs + 1 LC(1 k)s + CRs + 1 (1+ k)s2 + R s + 1 (1 k)s2 + R s + 1

L
LC
L
LC

)(

k 20 s

2
2
2
2
(1+ k)s + 0 s + 0 (1 k)s + 0 s + 0

Evaluating this expression at s = j0 , yields

5/31/01 Mag Crt Probs

P18-14

H( j 0 ) =

R. A. DeCarlo & P.M. Lin

jk 30

2+
2 (1 k)2 + j 0 + 2
(1+
k)
+
0
0
0
0

Q
Q

j 20

jk

jk
C 0
C 0
=
=
1
2

j
j1
k k + k + Q2

Q
Q
Therefore
H( j 0 ) =

1
k

C 0 k 2 + 1
Q2

(f) To solve this part we differentiate and set equal to zero as follows:

d H (j 0 )

1
d
k
1
1
2k 2
=

2
=
=0
2
1
1
dk
C 0 dk k +
C 0 k 2 +

Q2

Q2 k + 2
Q

It follows that
2k 2
k2

1 =1
+ 2
Q

Hence
k2 =

1
Q2

or k = 1/Q. With this value of k,

1
1/ Q
Q
2 =
C 0
2C0
Q2
At 0 , the magnitude of the transfer function increases with increasing k, reaching a peak at k = 1/Q and
then decreases with a further increase in k as born out in the plots of part (c).
H( j 0 ) max =

Prbs Chap 19, 1/7/02

P19-1

CHAPTER 19 PROBLEM SOLUTIONS

SOLUTION PROBLEM 19.1. Refer to figure 19.3.
Vs = 100; ZL = 20; Rs = 1e3;
beta = 149;
Zinbox = (beta + 1)*ZL
Zinbox =
3000
% By voltage division
V1 = Vs*Zinbox/(Zinbox + Rs)
V1 = 75
% To obtain the power delivered by the source
I1 = Vs/(Zinbox +Rs)
I1 = 2.5000e-02
Psource = I1*Vs
Psource = 2.5000e+00

SOLUTION PROBLEM 19.2. Refer to figure 19.4.

Vs = 100; Z1 = 30e3; Rs = 50; beta = 149;
Zboxin = Z1/(beta+1)
Zboxin = 200
V1 = Vs*Zboxin/(Zboxin + Rs)
V1 = 80
Psource = Vs^2/(Rs + Zboxin)
Psource = 40
SOLUTION PROBLEM 19.3. Refer to figure 19.5.
(a) C = 0.1e-3; vc0 = 10; Z1 = 300; Z2 = 1e3;
Z3 = 1e3; gm = 9e-3;

Zin = Z1 + (1 + gm*Z1)*Z2
Zin = 4.0000e+03
(b)
tau = Zin*C

R. A. DeCarlo, P. M. Lin

Prbs Chap 19, 1/7/02

P19-2

R. A. DeCarlo, P. M. Lin

tau = 4.0000e-01
Hence, vC (t) = vC (0)e t / = 10e 2.5t V.

SOLUTION PROBLEM 19.4. (a) First observe that since no current can flow into the secondary
we have
Voc = aV pri = aRI in = 800 Vrms
Now
Zth =

1
+ a2 R = 640 j 360
j C

(b) ZL = (Zth ) = 640 + j360 .

Voc = 800; Rth = 640;
Pmax = Voc^2/(4*Rth)
Pmax = 250
*

(c) By inspection the circuit is a 640 resistor in series with a 3.6 H inductor.

SOLUTION PROBLEM 19.5. Because the output is open circuited, no current flows into the
secondary of the transformer, hence
voc = v sec + sin(3t)u(t) = 2v pri + sin(3t)u(t) = [cos(3t) + sin(3t)]u(t).
Additionally
Zth (s) =

10
2.5
20
+ s + 4 + 4
+ 0.25s + 9 =
+ 2s + 40
s
s
s

SOLUTION PROBLEM 19.6. Using Cramer's rule,

Prbs Chap 19, 1/7/02

P19-3

I1 =

V1

det 0
0

R. A. DeCarlo, P. M. Lin

1 a

0.5 0
0 0.5 0.25V1
1
=
=
V
1.5a
1.5a
6a 1

Therefore Rin = 6a .
2
V1,eff
100
To compute the average power, V1,eff = 10 V. Hence Pave =
=
watts.
Rin
6a

SOLUTION PROBLEM 19.7. As per the hint, we write loop equations as follows:
Vout s + 1 1 0 Iout

10 s = 1 6 2 I2
40 s 0
2 4 I3
Using equation 19.6,

0.2 0.1 1
0.2 0.1 10 s
Vout = s + 1 [1 0]
Iout + [1 0]

0.1 0.3 0
0.1 0.3 40 s
2
= (s + 0.8)Iout +
s
Therefore
Vout = (s + 0.8)(Iout )

2
= Zth ( Iout ) + Voc
s

2
i.e., Zth = s + 0.8, and Voc = .
s

SOLUTION PROBLEM 19.8. (a) Let the node voltages from left to right be V1 , V2 , and Vout .
Also inject a current I3 into node 3. Writing nodal equations by inspection we have:

Prbs Chap 19, 1/7/02

P19-4

R. A. DeCarlo, P. M. Lin

Iin 1.5
1
0.25
V1

2
0.5
0 = 1
V2
I 3 0.25 0.5 0.0625s + 0.75 Vout
Using equation 19.11, we have

1
1 Iin
I3 = W22 W21W11
W12 Vout + W21W11
0

0.5 0.25
0.5 Iin
1
1
= 0.0625s + 0.75 [0.25 0.5]
Vout [0.25 0.5]

0.5 0.75 0.5

0.5 0.75 0

Thus
I3 = (0.0625s + 0.375)Vout 0.5Iin
1
16
Therefore Isc = I 3]V =0 = 0.5Iin . Further Zth (s) =
=
.
out
0.0625s + 0.375 s + 6
16I sc 8Iin
(b) Vout (s) = Zth (s)I sc =
=
. By inspection, the impulse response is
s+6 s+6
vout ,imp (t) = 8e 6t u(t) V. Further, from MATLAB
n = 8; d = [1 6 0];
[r,p,k] = residue(n,d)
r=
-1.3334e+00
1.3334e+00
p=
-6
0
k=
[]
Hence the step response is:
vout ,step (t) =

4
1 e 6t u(t) V
3

SOLUTION PROBLEM 19.9. (a) Consider the following figure:

Prbs Chap 19, 1/7/02

P19-5

R. A. DeCarlo, P. M. Lin

Let Iout enter the output terminal and I1 and I2 be the currents entering the primary and secondary
of the transformer respectively. It follows that

Iin =

bVout
(b 1)Vout (2b 1)Vout
+ I1 +
=
+ I1
R
R
R

which implies that

I1 = Iin

(2b 1)Vout
R

Further,
Iout =

(1 b)Vout
(1 b)Vout
+ I2 =
bI1
R
R

Therefore
Iout =

(1 b)Vout
(1 b)Vout
(1 b)Vout
(2b 1)Vout

+ I2 =
bI1 =
b Iin

R
R
R
R
=

2b 2 2b + 1
Vout bIin
R

Equivalently
Vout =

2b 2

R
bR
Iout + 2
Iin
2b + 1
2b 2b + 1

Therefore
Zth =
(b) Isc = b45o , Zth =

2b2

2b2

R
bR
, Voc = 2
Iin , Isc = bIin
2b + 1
2b 2b + 1

R
.
2b + 1

Prbs Chap 19, 1/7/02

(c) B = 10 =

P19-6

1
ZthC

1
R

2b 2 2b + 1

R. A. DeCarlo, P. M. Lin

2b2 2b + 1
5
=
. Hence C = 0.02 F. Further,
RC
25C

1
1
=
. Therefore L = 2 H.
LC 0.02L
V
V
1
2
100s
(d) H(s) = out = out = 2 2
=
=
1
Iin
0.5I sc
2b 2b + 1 1
s2 + 10s + 25
+
+ Cs 0.2 + 2s + 0.02s
R
Ls
2
0

= 25 =

2
s 10
and Iin (s) =
2
. Hence,
2
s + 100
Vout (s) =

(s2

50 2 s(s 10)
+ 10s + 25)(s2 + 100)

To compute vout (t) in MATLAB,

%partial fraction expansion of Vout/2 = = n(s)/d(s)
n =50*[ 1 -10 0];
d = conv([ 1 0 100], [ 1 10 25]);
[r ,p, k] = residue(n,d)
%numerator polynomial of combined complex pole terms
num = [1 -p(1)]*r(2) +[1 -p(2)]*r(1)
Output from MATLAB
r=
2.8000 + 0.4000i
2.8000 - 0.4000i
-5.6000
30.0000
p=
-0.0000 +10.0000i
-0.0000 -10.0000i
-5.0000
-5.0000
num =
5.6000 -8.0000
Laplace transform of ouput
Vout (s) =

5.6 2
30 2
(5.6s 8) 2
+
+
2
s+5
(s + 5)
s2 + 100

Prbs Chap 19, 1/7/02

P19-7

R. A. DeCarlo, P. M. Lin

Taking the inverse Laplace transform using table 13.1 on page 515:

vout (t) = 30 2te 5t 5.6 2e 5t + 5.6 2 cos(10t) 0.8 2 sin(10t) u(t) V

The steady state part consists of the cosine and sine terms only. Since the parallel RLC acts like
a band pass circuit and the peak value occurs at 5 rad/s, one expects the magnitude of the steady
state output to be much smaller at 100 rad/s.

SOLUTION PROBLEM 19.10. (a) Write two nodal equations by inspection:

I1 Y1 + Y3 Y3 + gm V1 y11
=
I = Y
Y2 + Y3 V2 y 21
2
3

y12 V1
y 22 V2

(b) When port-2 is shorted, y11 is the input admittance. Therefore

Zin =

1
1
=
y11 Y1 + Y3

and since V2 = 0,
I2 = y 21V1 =

KY3
s

SOLUTION PROBLEM 19.11. (a) By inspection

Y3 V1 y11
I1 Y1 + Y3
I = Y + g
=
2 3
m Y2 + Y3 V2 y 21

y12 V1
y 22 V2

Clearly, Y3 = y12 . Then, Y1 = y11 Y3 = y11 + y12 and Y2 = y22 Y3 = y 22 + y12 . Finally,
gm = y 21 + Y3 = y 21 y12 .
(b) Recall that

Prbs Chap 19, 1/7/02

P19-8

R. A. DeCarlo, P. M. Lin

I1 Y1 + Y3 Y3 + gm V1 y11
=
I = Y
Y2 + Y3 V2 y 21
2
3

y12 V1
y 22 V2

Clearly, Y3 = y 21 . Then, Y1 = y11 Y3 = y11 + y 21 and Y2 = y22 Y3 = y 22 + y 21 . Finally,

gm = y12 + Y3 = y12 y 21 .

SOLUTION PROBLEM 19.12. (a) By definition of coupled inductors

V1 L1s Ms I1
I1
L2
1
V = Ms L s I I =

2
2 2
2 L1L2 M 2 s M

M V1
L1 V2

Hence, the y-parameters are:

L2

L1L2 M 2 s M
1

M
L1

(b) By definition of coupled inductors

V1 L1s Ms I1
I1
L2
1
V = Ms L s I I =

2
2 2
2 L1L2 M 2 s M

M V1
L1 V2

Hence, the y-parameters are:

L2

L1L2 M 2 s M
1

M
L1

If the coupling coefficient is 1, L1L2 = M 2 and the y-parameters do not exist since the
determinant of the z-parameter matrix is zero.

SOLUTION PROBLEM 19.13. Let I2' denote the current entering the dotted terminal of the
secondary of the coupled inductors. Then using the result of problem 12a,

Prbs Chap 19, 1/7/02

P19-9

I1
L2
1
I ' =

2 L1L2 M 2 s M

R. A. DeCarlo, P. M. Lin

M V1 1 4 1 V1
=
L1 V2 6s 1 1 V2

1
1
8
2
7
1
From the given circuit I2 = I2' + 2I1 = V1 + V2 + V1 V2 = V1 V2 . Therefore
6s
6s
6s
6s
6s
6s
the y-parameter matrix is:
1 4 1
6s 7 1

SOLUTION PROBLEM 19.14.

I
(a) By definition and the properties of the ideal transformer y11 = 1
= G1 and
V1 V =0
2
I
G +G
y22 = 2
= 1 2 2 . Additionally, since the circuit is obviously reciprocal,
V2 V =0
a
1
I
G
y12 (= y21) = 1
= 1.
V2 V =0
a
1
(b) V 2 reflected to the primary side, denoted by V1 , is
2K a
V1 = 2
s +4
Hence
I1 =

2G1K a
s2 + 4

To compute I2 , we reflect the parallel of G1 and G2 to the secondary of the ideal transformer.
Hence the impedance in parallel with V 2 , denoted Zsec, is
a2
Zsec =
G1 + G2
Therefore,

Prbs Chap 19, 1/7/02

P19-1 0

I2 =

R. A. DeCarlo, P. M. Lin

2K( G1 + G2 ) a2
V2
=
Zsec
s2 + 4

SOLUTION PROBLEM 19.15. (a) By definition

I
1
1
y11 = 1
=
= S

V1 V =0 Zin V =0 9
2
2
where
1

Zin = 6 + 12/ / (320/ /80) = 9

16
Similarly, by definition
I
1
3
y 22 = 2
=
=
S

V2 V =0 Zout V =0 400
1
1
where
Zout = 80 + 320/ /16
( (6 //12) ) =

400

SOLUTION PROBLEM 19.16. Write nodal equations:

I1 2s + 2

I 2 = 1
0 2s

1 2s V1

s + 2 s V2
s
5s V3

Using the matrix partitioning method, we obtain the 2-port y parameters

V1
I1 2s + 2 1 1 2s
[2s s]
I = 1

s + 2 5s s
V2
2
1.2s + 2 0.4s 1 V1 y11
=
=
0.4s 1 0.8s + 2 V2 y 21

y12 V1
y 22 V2

Prbs Chap 19, 1/7/02

P19-1 1

R. A. DeCarlo, P. M. Lin

SOLUTION PROBLEM 19.17. (a) The ideal transformer yields the constraints
I1 = aI2 and V2 = aV1
The three resistors have nodal equations
I2 2 1 V2
=

I2 1 2 V2
Substituting the first two equations into the last one, we obtain
I1 2a 2
I =
2 a

a V1 y11
=
2 V2 y 21

y12 V1
y 22 V2

(b)
2
y12 y 21
2 a
Yin = y11
= 2a
= 1.75a2 S
y 22 + YL
4

Zin =

1
4
= 2
Yin 7a

and

Gv = V2 = -y21 = a
V1 y22 + y L1 4

SOLUTION PROBLEM 19.18. This problem is solved in MATLAB.

Part (a)
% Parameter specification
Ys = 1e-3;YL = 1e-3;
y11=4e-3; y12 = -0.1e-3;
y21 = 50e-3; y22 = 1e-3;
% Calculation of input admittance and impedance
Yin = y11 - y12*y21/(y22 + YL)
Yin = 6.5000e-03
Zin = 1/Yin
Zin = 1.5385e+02
% Calculation of output admittance

Prbs Chap 19, 1/7/02

P19-1 2

R. A. DeCarlo, P. M. Lin

Yout = y22 - y12*y21/(y11 + Ys)

Yout = 2.0000e-03
Part (b)
% Calculation of voltage gain
Gv = (Ys/(Ys + Yin))*(-y21/(y22 + YL))
Gv = -3.3333e+00
Part (c)
V2 = Gv*10
V2 = -3.3333e+01
Therefore, v2 (t) = 33.333u(t) V. Finally,
PL = V2^2/1e3
PL = 1.1111e+00

SOLUTION PROBLEM 19.19.

(a)

V1 = Z in
= Z in
= 0.5 Z in = 10 or 0.1 S
Vs Z in + Z s Z in + 10

Now

Yin = y 11 - y12y21 = y11 - 0.022 = 0.1

y22 +Y L
0.2 + 0.1
Solving for y11 yields y11 = 0.2333 S.
(b)

v2 = - y 21 =
-2
v1 y22 +YL
0.2 +0.1

= - 6.667

Hence
and

v2(t) = - 6.667 v1(t) = - 3.333 vs(t) = - 33.33 u(t) V

2
2
PL = V2 = 33.33 = 111.11 W

RL

10

SOLUTION PROBLEM 19.20.

(a) Writing a node equation at port 1 and mesh equation at port 2, we obtain by inspection

I1 = 2V 1 + 3I 2
V2 = 2V 1 + 2I 2
Rearranging in matrix form , we have

Prbs Chap 19, 1/7/02

P19-1 3

1 -3 I1
0 2

Therefore

I1
I2
(b) Yin = y 11 -

I2

-1

1.5

V1

-1

0.5

V2

R. A. DeCarlo, P. M. Lin

2 0

V1

-2 1

V2

y11
y21

y12
y22

V1
V2

y12y21 = -1 + 1.51 = 1 S
y22 +Y L
0.5 + 0.25

(c) Here we compute Y out seen looking into port-2, i.e.,

Yout = y 22 - y12y21 = 0.5 + 1.51 = -1 S

y11 +Y s
-1 + 0
From current division,

-1
i2(t) = Yout i s(t) =
5u(t) = 6.667u(t) A
Yout +YL
-1 + 0.25
Finally from Ohm's law

v1(t) = - 0.5 3i2(t) = - 0.5 36.667u(t) = - 10u(t) V

REMARK: Because the current source sees a negative resistance, the circuit is unstable as it
stands.

SOLUTION PROBLEM 19.21. (a) With port 2 shorted, the Laplace transform of the given data
are:
1
1
1
4
1
I1(s) = , V1(s) =
=
, I 2 (s) =
s
s s + 4 s(s + 4)
s+ 3
Hence
I
y11 = 1 =
V1
and

1s
s+ 4
=
4
4
s(s + 4)

I
1 (s + 3) s(s + 4)
y 21 = 2 =
=
4
V1
4(s + 3)
s(s + 4)

Next, with port-2 terminated in a 1- reistor, the Laplace transform of the given data are:

Prbs Chap 19, 1/7/02

P19-1 4

R. A. DeCarlo, P. M. Lin

Z L=YL = 1
I 1(s) = 1/s,
V1(s) = 1 - 1 + 1 = 5s + 16
s s+4 (s+4)2
s(s + 4) 2
I2 = - 1
s +7
Now

I 2 = - V2 = - V 2 = y21V1
ZL
y22 +YL
Solving for y22, and using y21 expression found earlier, we obtain

y22 = y 21V1/I 2 - Y L= - s(s + 4) 5s + 16 s + 7 - - 1

4 (s + 3) s(s + 4) 2 -1
2
= s + 23s + 64
4(s + 3)(s+ 4)
Finally, to compute y12, we use the defining equation for y-parameters, i.e.,
from which we obtain

I1 = y11V1 + y 12V2

1 - s + 4 5s + 16
s
4 s(s + 4 ) 2
y12 = I1 - y 11V1 =
= - (s + 7)
V2
1
4 ( s + 4)
s+7
(b) Given YL =1 S, the input admittance is

Yin = y11 - y12y21

y22 +Y L

- (s + 7) - s(s + 4)
2
= s + 4 - 4 ( s + 4) 4(s + 3) = (s + 4)
4
5(s+3.2)
s2 + 23s + 64 +1
4(s + 3)(s+ 4)

and

Zin = 1 = 5(s+3.2)
Yin
(s + 4)2

Prbs Chap 19, 1/7/02

P19-1 5

R. A. DeCarlo, P. M. Lin

(c) For this part, we use phasors to do the sinusoidal steady state analysis: = 10 rad/s and I 1 =
1. Also,

V 1 = Z = 5(j10+3.2) = 0.1973 - j0.4072

in
I1
(j10 + 4)2
and

V 2 = -y21
V1
y22 +YL

- j10(j10 + 4)
4(j10 + 3)
=
= -0.2873+ j0.1787
-100 + 230j + 64 +1
4(j + 3)(j+ 4)

Thus

V 2 = V2 V 1
I1
V 1 I1

= (-0.2873+ j0.1787)(0.1973 - j0.4072)

from which we obtain in Ohms,

V2
I1

= -0.2873+ j0.1787 0.1973 - j0.4072 = 0.8192

SOLUTION PROBLEM 19.22. Looking into port-1, the admittance is:

and

22
Yport 1 (s) = y11 - y12y21 = 0 +
= 40
s
y22 +Y L
0 + 0.1s
Zport 1 (s) = s
40

Therefore

Hence

2
2
Zin (s) = 1000 + s = s + 200
s
40
40s

Zin ( j ) =
The imaginary part is zero when

4 10 4 2
4 10 4 2
=j
j40
40

= 200 rad/s. Hence, the resonant frequency is 200 rad/s.

SOLUTION PROBLEM 19.23. (a) By writing nodal equations for the boxed 2-port, we have by
inspection (note passive circuit in which y21 = y12 ):

Prbs Chap 19, 1/7/02

P19-1 6

I1

I2

3s + 2

-2s -2

V1

-2s -2

3s + 3

V2

R. A. DeCarlo, P. M. Lin

y11
y21

V1

y12
y22

V2

(b) In Ohms, ZL = s+1; Zs = 2. Therefore,

2
(2s +2) 2
Yin = y 11 - y12y21 = 3s+2 = 2s + 3s + 1 = 2s + 1
y22 +Y L
3s + 3 + s + 1
s+1

and

V1 = Z in
1
1
=
=
= 1
Vs Z in + Z s 1 + Z s Yin
1 + 2(2s + 1)
4s + 3
-y21 =
V2 =
2s +2
= 0.5
y
+
y
22
L
V1
3s + 3 + s + 1

Thus,

V2 = V1 V2 = 1 0.5 =
1
= 1
2
Vs Vs V1 4s + 3
8s + 14s + 6 8s+ 6
(c) The impulse reponse is
1
1 0.125
0.75t
h(t) = L1{H (s)} = L1
u(t)
= L
= 0.125e
8s
+
6
s
+
0.75

For the step response,

H (s)
1 0.125 1
0.75t
v 2 (t) = L1
u(t) V
= L
= 1 e
s
s(s
+
0.75)
6

(d) We must compute the complete Laplace transform and invert. Here

Vs(s) = 12.75 2
s2 + 4

and

3.1875
V2(s) = H(s)V2(s) = 0.125 12.75 2 =
2
s+ 0.75
s + 4 (s + 0.75) (s2 + 4)
We use MATLAB to compute the partial fraction expansion
n=3.1875;
d= conv([1 0.75], [ 1 0 4]);
[r p k ] = residue (n,d)
r=
-0.3493 - 0.1310i
-0.3493 + 0.1310i
0.6986
p=
-0.0000 + 2.0000i

Prbs Chap 19, 1/7/02

P19-1 7

R. A. DeCarlo, P. M. Lin

-0.0000 - 2.0000i
-0.7500
Hence, after combining the two complex terms, we obtain

V2(s) = 0.6986 + - 0.6986s + 0.524

s + 0.75
s2 + 4
From Table 13.1, the steady state response is

v2,ss (t) = [-0.6986cos(2t) + 0.262sin(2t)]u(t) V

and the transient response is

v2,tran(t) = 0.6986e-0.75t u(t) V

SOLUTION PROBLEM 19.24. (a) Using the y-parameters of stage 2,

Yin2 = y 11 - y12y21 = 500 + 0.175000 = 501.8 S

y22 +Y L
185 +4000
The load for stage 1 is the parallel combination of Zin2 and the 2 k resistor. Hence, using the yparameters of stage 1, we obtain

Yin1

4
y
y
0.52410
12
21
= y 11 = 2000 +
= 2108.9 S
y22 +Y L
100 +501.8 + 500

(b) We compute the following voltage gains:

V1 = Z in1
1
1
=
=
= 0.9499
Vs Z in1 + Z s 1 + Z s Yin1
1 + 252108.910-6
V2
y 21
0.24
=
=
= 217.8
V
y 22 + YL,stage1 (100 + 500 + 501.8) 10 6
1 stage1
and
V2
y 21
0.075
=
=
= 17.92
V
y 22 + YL,stage2 (185 + 4000) 10 6
1 stage2
Finally Gv is the product of the three gains calculated above

Gv = 0.9499(- 217.8) (-17.92) = 3708.2

Prbs Chap 19, 1/7/02

P19-1 8

R. A. DeCarlo, P. M. Lin

SOLUTION PROBLEM 19.25. (a) With the switch in position A, the load to the 2-port is YL =
Cs = 0.25s. Hence,
V2
y 21
y 21
=
=
V1 y22 + YL y 22 + 0.25s
(b) From the given data, V1 (s) = 2/s and hence,
V2(s) =

y 21
1 s
2
8
2
1
1
V1 (s) =

=
=
+
+
2
y 22 + 0.25s
1 s + 0.25s s
s s+2 s2
s(s 4)

Hence, for t 0,

v2(t) = ( -2 + e-2t + e 2t) u(t) V

(c) The circuit is not stable in the time interval 0 to 1 s, because the transfer function has a pole
in the right half plane.
(d) v2(1-) = -2 + e-2 + e 2 = 5.524 V
(e) Replace the charged capacitor by the parallel combination of an admittance of 0.25s and a
current source of value 0.258.524 (in accordance with figure 14.16
(f) Z1 (s) = b2 1 = 4 .
(g) For t 1s, the capacitor is discharging through a 4- equivalent resistance, with a time
constant 0.254 = 1 s, and an initial voltage v2 (1-) = 5.542 V.
Hence
v 2 (t) = 5.524e (t 1)u(t 1)V
and by the ideal transformer voltage ratio property,
v 3(t) = 2.762e (t 1)u(t 1)V

SOLUTION PROBLEM 19.26. (a) By definition of coupled inductors

V1 L1s Ms I1
V = Ms L s I
2
2 2
Hence, the z-parameters are:

Prbs Chap 19, 1/7/02

P19-1 9

R. A. DeCarlo, P. M. Lin

L1s Ms
Ms L s

2
(b) By definition of coupled inductors
V1 L1s Ms I1
V = Ms L s I
2
2 2
Hence, the y-parameters are:
L1s Ms
Ms L s

2
The z-parameters exist independent of the values of M, L1 , and L2 .

SOLUTION PROBLEM 19.27. (a) By definition and the properties of the ideal transformer
V
V
z11 = 1
= R1 + R2 and z22 = 2
= a2 R2 . Additionally, since the circuit is obviously
I1 I =0
I2 I =0
2

V
reciprocal, z21 (= z12 ) = 2
= aR2 .
I1 I =0
2
(b) The input impedance is given by the formula
z12z21
a 2R22
1
Zin = z11
= R1 + R2 2
= R1 + 2
R2
z22 + ZL
a R2 + R2
a +1
(c) If port-1 is open circuited, I1 = 0. Hence,

2aKR
V1 = z12 I2 = 2 2
s +4

and V2 = z22I 2 =

2a 2KR2
s2 + 4

SOLUTION PROBLEM 19.28. (a) For this part consider the figure below:

Prbs Chap 19, 1/7/02

P19-2 0

R. A. DeCarlo, P. M. Lin

With port 2 open, it follows that Zpr = 320/16 = 20 and

V
20 12
z11 = 1
=6 +
=13.5
I1 I =0
20 + 12
2
With port 1 open and I2 injected into port 2, we have Zsec = 12*16 = 192 and
Vpr
V
V
I (320 / /192 )
z12 = 1
=
= sec
= 2
= 30

I2 I = 0 I2 I =0
4I2 I =0
4I2

I
=0
1
1
1
1
With port 2 open and I1 injected into port 1, we have Zpr = 20 and
4Vpr
V
V
4I1 (12 //20 )
z21 = 2
= sec
=
=
= 30

I1 I =0
I1 I =0
I1 I =0
I1

I2 =0
2
2
2
With port 1 open, it follows that Zsec = 12*16 = 192 and
V
320 192
z22 = 2
= 80 +
= 200
I2 I =0
320 + 192
1
300
2000
(b) V1 (s) = z12 I2 (s) = 2
and V2 (s) = z 22 I2 (s) = 2
.
s +4
s +4

Prbs Chap 19, 1/7/02

P19-2 1

R. A. DeCarlo, P. M. Lin

SOLUTION PROBLEM 19.29. (a) Writing two mesh equations we have by inspection,
V1 Z1 + Z3
V = Z + r
2 3 m
(b) V1 = z11I1 =

Z3 I1 z11
=
Z2 + Z3 I2 z21

z12 I1
z22 I 2

K ( Z1 + Z3 )
K (Z3 + rm )
and V2 = z21I1 =
s
s

SOLUTION PROBLEM 19.30. From the result of problem 29,

Z1 + Z3
Z + r
3 m

Z3 z11
=
Z2 + Z3 z21

z12
z22

Therefore, Z3 = z12 , Z1 = z11 Z3 = z11 z12 , Z2 = z22 Z3 = z22 z12 , and

rm = z21 Z3 = z21 z12 .

SOLUTION PROBLEM 19.31. (a) The z-parameters can be computed by inspection (first write
the z-parameters of the passive part of the network, i.e., with the dependent source ignored; then
add the effect of the dependent source to the resulting equations.) As such, using loop equations,
10
10
5 + s 10 + s 10 0.5s + 1 s + 1
Z = 10

10 =
s + 1

10 + s 1
s
s
10 10
(s + 1) 5s + 10 10(s + 1) (5s + 10)(2s + 1) 10(s + 1)
5s + 10
s
(b) Zin =
s
=

=
s
10(s + 1)

s
s(2s + 1)
2s(s + 0.5)
+ 10

s
5s2 + 7.5s 5s + 7.5
=
=
s(s + 0.5)
s + 0.5
V (s) 10 s + 0.5 0.6667 1.3334
(c) I1(s) = 1
=
=
+
. Hence
Zin (s)
s 5s + 7.5
s
s + 1.5

Prbs Chap 19, 1/7/02

P19-2 2

R. A. DeCarlo, P. M. Lin

2 4

i1 (t) = + e 1.5t u(t) A

3 3

SOLUTION PROBLEM 19.32. (a) To find the resonant frequency we first find
Zin (s) = z11

z12 z21
z z
10z11
10
10
10s
= z11 11 11 =
=
=
= 2
2
10
z22 + Z L
z11 + 10 z11 +10 1 +
s + 25 s + s + 25
z11 1 +
s

This is of the form of equation 17.18 with K = 10. Here according to equation 17.19f,
r = p = 25 = 5 rad/s
(b) To find Q we use equation 17.19e, i.e.,

Q = Qp =

p
2 p

5
=5
1

(c) Using MATLAB we obtain the frequency response plots below:

Prbs Chap 19, 1/7/02

P19-2 3

R. A. DeCarlo, P. M. Lin

10
9
8

Magnitude of Zin

7
6
5
4
TextEnd
3
2
1
0

8
10
12
Frequency in rad/s

14

16

18

20

Prbs Chap 19, 1/7/02

P19-2 4

R. A. DeCarlo, P. M. Lin

100
80
60

Phase of Zin degrees

40
20
0
-20
TextEnd
-40
-60
-80
-100

8
10
12
Frequency in rad/s

z11
SOLUTION PROBLEM 19.33. (a) Since Z =
z21
not include Zs , it follows that
Zin = z11

14

16

18

20

z12 2 3
=
. Assuming that Zin does
z22 30 4

z12z21
90
= 2+
= 5.75
z22 + ZL
24

Assuming that Zout does not include the parallel connection of Z L, then
z z
90
Zout = z22 12 21 = 4 +
= 34
z11 + Zs
3
(b) From equation 19.27
ZL z21
GV = Gv2Gv1 =

.
z22 + ZL Zin + ZS

Prbs Chap 19, 1/7/02

P19-2 5

R. A. DeCarlo, P. M. Lin

Thus in MATLAB
Gv = (20/(4 + 20))*(30/(5.75+1))
Gv = 3.7037e+00
% Therefore
v2 = Gv*30
v2 = 1.1111e+02
Hence, v 2 (t) = 111.11u(t) V. The power absorbed by ZL is therefore (in watts):
PZL = v2^2/20
PZL = 6.1728e+02

SOLUTION PROBLEM 19.34. (a) Before determining b, it is necessary to compute the

impedance seen at the secondary of the 2-port. Here,
z z
2 3R02
Zout = z22 12 21 = 33 2 R0
= 33 2R0 2R0 = 32 2 R0
z11 + Zs
R0 + R0
Thus b must be chosen so that the impedance reflected to the secondary of the transformer is 2
, i.e.,
32 2R0
2=
b = 4 R0
b2
(b) First do a source transformation on the front end of the two port to obtain
v s (t) = R0 2 cos(2t) = 32 2 cos(2t) V
Therefore, Vs,eff = 32 V. Also note that the impedance looking into the primary of the two port
is
z z
2 3R02
Zin = z11 12 21 = R0
= 31.015
z22 + ZL
33 2 R0 + 32 2R0
Hence
ZL z21
GV = Gv2Gv1 =
= 0.5
z22 + ZL Zin + ZS
Consider the following MATLAB calculations:
a=2; R0 = 16;
Zin = a*R0-2*a^3 * R0^2/(65*a^2 *R0)

Prbs Chap 19, 1/7/02

P19-2 6

R. A. DeCarlo, P. M. Lin

Zin = 3.1015e+01
ZL = 32*a^2 * R0
ZL =
2048
Zs = a*R0
Zs = 32
z22 = 33*a^2*R0
z22 =
2112
z21 = 2*a*R0
z21 = 64
Gv = (ZL/(z22+ZL))*(z21/(Zin+Zs))
Gv = 5.0000e-01
V1eff = 32;
V2eff = Gv*V1eff
V2eff = 16
b = 4*a*sqrt(R0)
b = 32
VLoadeff=V2eff/b
VLoadeff = 5.0000e-01
Pmax = VLoadeff^2/2
Pmax = 1.2500e-01
Hence, max power transferred to the load is 125 mW.

SOLUTION PROBLEM 19.35. (a) Using the usual formula

Z(s) = z11

z12 z21
1000 (1000)
=0
= 0.01s
z22 + ZL
0 + 108 s

Yes, the input to the 2-port looks like a 0.01 H inductor.

(b) Under the given conditions, the circuit reduces to parallel RLC with Req = 50 k, L = 0.01
H, and C = 100 pF. Therefore,
(c) For this circuit

1
1
= 106 rad/s and B =
= 2 105 rad/s.
R
C
LC
eq

Prbs Chap 19, 1/7/02

P19-2 7

R. A. DeCarlo, P. M. Lin

1
Vin
sC
Vin
106 s
V1(s) = 1

=
1
1
1 R1 = s2 + 2 10 6 s + 1012 Vin
+
+ Cs R1 s2 +
s+
Req Ls
ReqC
LC
For the impulse response, Vin = 1 and from MATLAB
n = [1e6 0];d = [1 2e6 1e12];
[r,p,k] = residue(n,d)
r=
1.0000e+06
-1.0000e+12
p=
-1000000
-1000000
k=
[]
Therefore, the impulse response is:

6
h(t) = 106 1012 t e 10 t u(t) V

1000 1 0
0.001
0
(d) y =
=
S

0
0
1000
0.001

(e) V2(s) =

y 21
0.001
106 s
1011
V1(s) =

=
y 22 + YL
0 + 108 s s2 + 2 10 6 s + 1012 (s + 10 6 )2

Hence
6
v 2 (t) = 1011 te10 t u(t) V.

SOLUTION PROBLEM 19.36.

Prbs Chap 19, 1/7/02

P19-2 8

R. A. DeCarlo, P. M. Lin

z z
1.2075 63.75
(a) Zin 2 = z11 12 21 = 62.582
= 1.7778 k
z22 + ZL
1.25 + 0.016
Because z12 for stage 1 is zero,
z12z21
Zin = z11
= z = 2 k
z22 + Zin 2 / /2 11
V
Zin
(b) Gv1 = 1 =
= 0.96386 . Let ZL1 = Zin 2 //2 = 0.94118 k. Then
Vs Zin + 75
V
ZL1
z
V
ZL
z
Gv2 = 2 =
21 = 22.472. Finally Gv3 = out =
21 = 0.45319.
V1 ZL1 + z22 Zin
V2
ZL + z22 Zin 2
Thus
Gv = Gv1Gv 2Gv 3 = 9.816

SOLUTION PROBLEM 19.37. (a) By inspection via mesh standard equations

L1s Ms
z=

Ms L2s
(b) Utilizing the properties of an ideal transformer,
V
z11 = 1
= L1s
I1 I =0
2

V
k 2 L2
z22 = 2
=
L s + (1 k 2 )L2s = L2s
I2 I =0
L1 1
1

k L2
V
z12 = 1
=
L s = k L1L2 s = Ms
I2 I =0
L1 1
1
Finally,
k L2
V
z21 = 2
=
L s = k L1L2 s = Ms
I1 I =0
L1 1
2
(c) Utilizing the properties of an ideal transformer,

Prbs Chap 19, 1/7/02

P19-2 9

V
z22 = 2
= L2 s
I2 I =0
1

R. A. DeCarlo, P. M. Lin

V
k 2 L1
z11 = 1
=
L s + (1 k 2 )L1s = L1s
I1 I =0
L2 2
2

k L1
V
z12 = 1
=
L s = k L1L2 s = Ms
I2 I =0
L2 2
1
Finally,
k L1
V
z21 = 2
=
L s = k L1L2 s = Ms
I1 I =0
L2 2
2
102
(d) For this circuit k = 1 and the turns ratio a =
= 0.1. Under this condition the given
1
circuit reduces to a current source of value Is = Vin 5000 driving a parallel RLC with R =
5000 , L = L1 = 0.01 H, the capacitance reflected to the primary of value C = 108 F.
Therefore
m

1
1
= 105 rad/s and B =
= 2 104 rad/s.
ReqC
LC

Finally, at w = wm, the circuit is resonant and Vin appears across the primary of the transformer.
V
This voltage is then stepped up by a factor of 10. Therefore out
= 10.
Vin max

SOLUTION PROBLEM 19.38. (a) By inspection

V1 Z1 1 I1 h11
I = 1 Y V = h
2
2 2 21

h12 I1
h22 V2

(b) By inspection
V1 = V2 Z1I 2 V1 + Z1I2 = V2
and
I1 = Y2V1 I2

Y2V1 I 2 = I1

Prbs Chap 19, 1/7/02

P19-3 0

R. A. DeCarlo, P. M. Lin

In matrix form,
V1 0 1 I1
V1 1
I = 1 0 V I = Y
2
2
2 2

Z1 1 0 1 I1
1 1 0 V2

Thus
V1
Z1 1 I1
1
I = Z Y + 1 1 Y V
2
1 2

2 2

SOLUTION PROBLEM 19.39. (a) From problem 19.38b we have

V1
Z1 1 I1
1
I = Z Y + 1 1 Y V =
2
1 2

2 2

1
Cs
1

1
+ 1 1
RCs

1 I
RCs I1
1 R
1
=

1 V2 RCs + 1 RCs Cs V2
R

(b) This part is a cascade of part (a) and an ideal transformer. Label the voltage and current at
the port 2 of N as V and I . From the properties of the ideal transformer, V = nV and
1

I2 = I2 n . Hence
RCs I1
RCs I1
V1
V1
R
1 R
1
I = RCs + 1 RCs Cs V nI = RCs + 1 RCs Cs V n
2

2
2

2
Therefore
RCs n I1
V1
1 R
I = RCs + 1 RCs n Cs n 2 V
2

From table 19.1, if h22 =

Cs
n 2(RCs + 1)

0, then the z-parameters exist and if h11 =

the y-parameters exist, i.e., if C 0 and R 0 respectively.

R
0,
RCs + 1

Prbs Chap 19, 1/13/02

P19-1

R. A. DeCarlo, P. M. Lin

CHAPTER 19 PROBLEM SOLUTIONS

SOLUTION PROBLEM 19.40. (a) Let Z1 = R and Z2 = 1/Cs or Y2 = Cs. From problem 38,
Z1 1 R 1
h=
=

1 Y2 1 Cs
(b) This part is a cascade of an ideal transformer and part (a). Label the voltage and current at
the port 1 of N as V and I . From the properties of the ideal transformer, V = bV and
1

I1 = I1 b . Hence
V1 R 1 I1
V1
=

I 2 1 Cs V2
I2

b R 1 bI1
= 1 Cs V

2

Therefore
V1 b2 R b I1

I =
Cs V2
2 b
From table 19.1, if h22 = Cs 0, then the z-parameters exist and if h11 = b 2R 0, the yparameters exist, i.e., if C 0 and R 0 (assuming reasonably that b 0) respectively.

SOLUTION PROBLEM 19.41. For this solution we apply the definition of h-parameters: by
inspection
V
1
h11 = 1
=
I1 V =0 2 + 2s
2
I
2V 2sV1
(2 2s)V1
1 s
h21 = 2
= 1
=
=
I1 V =0
I1
(2 + 2s)V1 V =0 s + 1
V2 =0
2
2
1
s +1
When I1 = 0, then I2 = 2V1 + 2V1 = 4V1 and V2 = 0.5(2V1) + (2V1 ) =
V.
2s
s 1

Prbs Chap 19, 1/13/02

P19-2

R. A. DeCarlo, P. M. Lin

V
s
I
4V
4s
Therefore, h12 = 1
=
and h22 = 2
= 1
=
.
V2 I =0 s + 1
V2 I =0 V2 I =0 s + 1
1
1
1

SOLUTION PROBLEM 19.42. (a) In MATLAB

h11 = 250; h12 = 0.025; h21 = 12.5; h22 = 2.25e-3;
Zs = 1e3; ZL = 500;
YL = 1/ZL
YL = 2.0000e-03
Zin = h11 - h12*h21/(h22 + (1/ZL))
Zin = 1.7647e+02
Yout = h22 - h12*h21/(h11 + Zs)
Yout = 2.0000e-03
Zout = 1/Yout
Zout = 500
(b)
% Gv1 = V1/Vs
Gv1 = Zin/(Zin + Zs)
Gv1 = 1.5000e-01
% Gv2 = V2/V1
Gv2 = -h21/(Zin*(h22 + YL))
Gv2 = -1.6667e+01
Gv = Gv1*Gv2
Gv = -2.5000e+00
(c) Given the above, the Thevenin equivalent seen by the capacitor is Voc = 2.5Vin and
Rth = 500 .
In MATLAB
Zth = ZL*Zout/(ZL + Zout)
Zth = 250
Vin = 10;
Voc = -2.5*Vin;
w = 400;

Prbs Chap 19, 1/13/02

P19-3

R. A. DeCarlo, P. M. Lin

Zc = 1/(j*w*10e-6)
Zc =
0 - 2.5000e+02i
Vc = Voc*Zc/(Zth + Zc)
Vc = -1.2500e+01 + 1.2500e+01iV2mag = abs(Vc)
V2mag = abs(Vc)
V2mag =
1.7678e+01
V2ang = angle(Vc)*180/pi
V2ang =
135
From above,
v 2 (t) = 17.678 2 cos(400t + 135o ) V
Therefore
Pave = V2mag^2/500
Pave =
6.2500e-01

SOLUTION PROBLEM 19.43.

(a) Using the h-parameters of stage 2

Zin2 = h 11 - h12h21 = 1000+ 0.96651 = 4000

h22 +Y L
0.0008 + 1/64
The load for stage 1 is the parallel combination of Zin2 and the 3 k resistance. However,
because h12 = 0, the input impedance is unaffected by the load, and hence for stage 1,

Zin1 = h 11 = 2000
(b) For stage 1, because h12 = 0, the output impedance is unaffected by the source impedance.
Thus,

Yout1 = h 22 = 0.05 10-3 S,

Zout1 =1/Y out1 = 20 103

Prbs Chap 19, 1/13/02

P19-4

R. A. DeCarlo, P. M. Lin

For stage 2, the source impedance is the parallel combination of Zout1 and the 3 k resistance.
Thus

Zs2 = 200003000 =2608.7

20000 + 3000

and

0.96651
Yout2 = h 22 - h12h21 = 0.0008+
= 0.0145 S
h11 +Z s2
1000 + 2608.7
Z out2 =1/Yout2 = 69.19
(c)

V1 = Z in1
2000
=
= 0.5
Vs Z in1 + Z s 2000 + 2000
The load of stage 1 is the parallel combination of Zin2 and Zm. Thus

YL1 = Y in2 + 1/3000 = 5.834 10-4 S

Hence

ZL1 = 1/ YL1 = 1714

and

V2
V1

stage 1

-50
= 1 -h 21 = 1
= - 39.46
Z in1 h22 + Y L1 2000 (0.05 + 0.5834) 10-3

For stage 2, the load is 64 . Hence

V2
V1

stage 2

51
= 1 -h 21 = 1
= 0.7762
Z in2 h22 + Y L2 4000 (0.8 + 1000/64) 10-3

Finally, the overall voltage gain is the product of the three gains calculated above

Vout = 0.5 (- 39.46) (0.7762) = - 15.32

Vs
(d) The input circuit consists of a series connection of Vs, Zs, C and Zin1. The remainder of the
circuit is resistive and has no effect on the frequency response. The magnitude response is of the
high pass type with
f 3dB =

1
1
=
= 39.789 Hz
2(Rs + Rin1)C 2(2000 + 2000)106

SOLUTION PROBLEM 19.44. To meet the required matching, we must have

Prbs Chap 19, 1/13/02

P19-5

Zout = Z out2 = Z L = 64 =
This requires that

R. A. DeCarlo, P. M. Lin

1
Yout2

Yout2 = h 22 - h12h21 = 0.0008+ 0.96651 = 1

h11 +Z s2
1000 +Z s2
64
Solving for Zs2, we obtain Zs2 = 2323.2 . Now for stage 2, Z s2 is the parallel combination of
Zout1 = 20 k and Zm:

Zs2 = 20000Z m = 2323.2

20000 +Z m
from which Zm = 2628.5 . With this new value of Zm, we repeat the calculations of problem
19.43 to obtain Zin = 2000 and Vout/Vs = -14 .26. Details follow.
Using the h parameters of stage 2

Zin2 = h 11 - h12h21 = 1000+ 0.96651 = 4000

h22 +Y L
0.0008 + 1/64
For stage 1, because h12 = 0, the input impedance is not affected by the load and

Zin1 = h 11 = 2000 . The voltage gains of the various stages are:

V1 = Z in1
2000
=
= 0.5
Vs Z in1 + Z s 2000 + 2000
The load of stage 1 is the parallel combination of Zin2 and Zm. Thus

YL1 = Y in2 + 1/2628.5 = 6.3044 10-4 S

Hence

V2
V1

stage 1

-50
= 1 -h 21 = 1
= - 36.74
Z in1 h22 + Y L1 2000 (0.05 + 0.63044) 10-3

For stage 2, the load is 64 . Hence

V2
V1

stage 2

51
= 1 -h 21 = 1
= 0.7762
Z in2 h22 + Y L2 4000 (0.8 + 1000/64) 10-3

Finally, the overall voltage gain is the product of the three gains calculated above

Vout = 0.5 (- 36.74) (0.7762) = - 14.26

Vs

Prbs Chap 19, 1/13/02

P19-6

R. A. DeCarlo, P. M. Lin

SOLUTION PROBLEM 19.45.

(a) Using the h-parameters of stage 2

Zin2 = h 11 - h12h21 = 500 + 0.96651

= 2000
h22 +Y L
0.0016 + 1/32
The load for stage 1 is the parallel combination of Zin2 and the 1.5 k resistance. However,
because h12 = 0, the input impedance is unaffected by the load, and Zin1 = h 11 = 1000 .
(b)
For stage 1, because h12 =0, the output impedance is unaffected by the source
impedance, and

Yout1 = h 22 = 0.1 10-3 S,

Zout1 =1/Y out1 =104
For stage 2, the source impedance is the parallel combination of Zout1 and the 1.5 k resistor
10 4 1.5 10 3
Zs2 = 4
= 1304.3
10 + 1.5 10 3

and

Yout2 = h 22 - h12h21 = 0.0016+ 0.96651 = 0.0289 S

h11 +Z s2
500 + 1304.3
Z out2 =1/Yout2 = 34.6
(c)

V1 = Z in1
1000
=
= 0.5
Vs Z in1 + Z s 1000 + 1000
The load of stage 1 is the parallel combination of Zin2 and Zm. Thus

YL1 = Y in2 + 1/1500 = 1.1667 10-3 S

Hence

ZL1 = 1/ YL1 = 857.1

V2
V1

stage 1

-50
= 1 -h 21 = 1
= - 39.47
Z in1 h22 + Y L1 1000 (0.1 + 1.1667) 10-3

For stage 2, the load is 32 . Hence

Prbs Chap 19, 1/13/02

V2
V1

P19-7

stage 2

R. A. DeCarlo, P. M. Lin

51
= 1 -h 21 = 1
= 0.7762
Z in2 h22 + Y L2 2000 0.0016 + 1/32

Finally, the overall voltage gain is the product of the three gains calculated above

Vout = 0.5 (- 39.47) (0.7762) = - 15.32

Vs
(d) The input circuit consists of a series connection of Vs, Zs, C and Zin1. The remainder of the
circuit is resistive and has no effect on the frequency response. The magnitude response is of the
highpass type with
1
1
f 3dB =
=
= 79.58 Hz
2(Rs + Rin1)C 2(1000 + 1000)106

SOLUTION PROBLEM 19.46. (a) Since the currents through YL and h22 are the same, h22 = YL.
(b) From current division, I2 =

YL
h I
YL + h22 21 1

I2
Y h
= L 21 .
I1 YL + h22

I
Y h
(c) 150 = 2 = L 21 = 0.5h21 h21 = 300.
I1 YL + h22
V
1
(d) h12 = 1
=
= 0.5 .
V2 I =0 2
1
I1
Zs
9 10 3
(e)
=
=
= 0.9 Zin = 1000 . Given this quantity,
Is Zs + Zin 9 10 3 + Zin
h11 = Zin +

h12h21
150
= 1000
= 400 .
h22 + YL
0.25

SOLUTION PROBLEM 19.47. Recall that

V1 h11
I = h
2 21

h12 I1
h22 V2

Prbs Chap 19, 1/13/02

P19-8

R. A. DeCarlo, P. M. Lin

V
(a) From this expression and specification 1, h12 = 1
=0.
V2 I =0
1
(b) From the formula for Y out (equation 19.50), specification 2, and the result of part (a), we
have
Yout =

1
Zout

1
h h
= h22 12 21 = h22
800
h11 + Z s

1
Hence, h22 =
= 1.25 mS.
800
For maximum power transfer from amplifier to the load,
Zout = 800 = b2 Z L = 8b 2
Therefore, b = 10.
(c) and (d) Observe that
Zin = h11

h12 h21
= h11
h22 + YL b 2

From specification 3 and voltage division,

V1 24
Z in
=
=
Vs 25 Zin + 40
Equivalently,
Vs 25
40
=
= 1+
V1 24
Zin
Hence h11 = Z in = 40 24 = 960 .
(e) From equation 19.51,
V
h21
h21
h
Gv2 = 2 = 100 =
=
= 21
2
3
3
V1
2.4
Zin h22 + YL b
960 1.25 10 +1.25 10

Hence, h21 = 240 .

Prbs Chap 19, 1/13/02

P19-9

R. A. DeCarlo, P. M. Lin

V2
(f) The power delivered to the load is PL = 2 and the power delivered to the amplifier is
800
V2 V2
Pamp = 1 = 1 . Therefore the power gain is
Zin 960
2

PL
96 V2
4
=
= 1.2 10
Pamp 80 V1

SOLUTION PROBLEM 19.48. Recall that

V1 h11
I = h
2 21

h12 I1
h22 V2

V
(a) From this expression and specification 1, h12 = 1
= 0.01.
V2 I =0
1
(b) For maximum power transfer from amplifier to the load,
Zout = 800 = b2 Z L = 8b 2
Therefore, b = 10.
Now we find Zin. From specification 3 and voltage division,
V1 24
Z in
=
=
Vs 25 Zin + 40

(1)

Equivalently,
Vs 25
40
=
=1 +
V1 24
Zin

Zin = 40 24 = 960

Using the formula for Zin we have the following equation

Zin = 960 = h11

0.01h21
h22 + 1.25 10 3

h11 = 960 + 0.01

h21
h22 + 1.25 10 3

(2)

Prbs Chap 19, 1/13/02

P19-10

R. A. DeCarlo, P. M. Lin

But from the given specs,

V2
h21
h21
= 100 =
=
V1
Zin h22 + YL b2
960 h22 + 1.25 10 3

which implies that

(h22

h21

+ 1.25 103

= 960 10 2

(3)

Substituting (3) into (2) allows us to solve for h11:

h11 = 960 + 0.01 960 102 = 1920
Now let us rewrite equation (3) as:
h21 960 10 2 h22 = 960 102 1.25 10 3 = 120

(4)

Also
Yout = 1.25 103 = h22

h12h21
0.01h21
= h22
h11 + Zs
1960

Equivalently,
1960 1.25 10 3 = 2.45 = 1960h22 0.01h21
Solving equations (4) and (5) simultaneously in MATLAB yields
A = [1 -960e2;-0.01 1960]
A=
1.0000e+00 -9.6000e+04
-1.0000e-02 1.9600e+03
b = [120; 2.45]
b=
1.2000e+02
2.4500e+00
x = A\b
x=

(5)

Prbs Chap 19, 1/13/02

P19-11

R. A. DeCarlo, P. M. Lin

4.7040e+02
3.6500e-03
h21 = x(1); h22 = x(2);
h = [1920 0.01;h21 h22]
h=
1.9200e+03 1.0000e-02
4.7040e+02 3.6500e-03
We can verify these results as well as compute the overall amplifier gain using the following mfile:
% two-port analysis in terms of h-parameters
function [zin, zout] =twoport(h, zL, zs)
['twoport analysis using h-parameters']
h11= h(1,1); h12=h(1,2); h21=h(2,1); h22=h(2,2);
zin = h11 - h12*h21/(h22+ 1/zL)
yout= h22 - h12*h21/(h11+zs);
zout= 1/yout
v1tovs= zin/(zin+zs)
v2tov1= -h21/(zin*(h22+1/zL))
v2tovs= v1tovs*v2tov1

twoporth(h,ZL,Zs)
ans =
twoport analysis using h-parameters
zin =
960
zout =
8.0000e+02
v1tovs =
9.6000e-01
v2tov1 =
-100
v2tovs =
-96

Prbs Chap 19, 1/13/02

P19-12

R. A. DeCarlo, P. M. Lin

Hence the overall voltage gain is VL/Vs = 96/10 = 96 because of the transformer.
Finally to compute power gains,
Vs = 1; Vin = 24/25;
VL = -9.6;
Pin = Vin^2/960
Pin =
9.6000e-04
Pload = VL^2/8
Pload =
1.1520e+01
Pgain = Pload/Pin
Pgain =
12000

SOLUTION PROBLEM 19.49.

C sV + gmV
C s + gm
I
(a) h21 = 2
=
=
1
I1 V =0 1

C + C s +
2
+ C + C s V

R
R

1
Rx + 1
I1

+ C + C s

V1
1
(b) h11 =
=
= Rx + 1
I1 V =0
I1
+ C + C s
2
R

(c) Under the condition that I1 = 0, V1 = V. Using voltage division from V2 to V:

Prbs Chap 19, 1/13/02

P19-13

R. A. DeCarlo, P. M. Lin

1
1

+ C s
C s
C s
R

V
V
h12 = 1
= = 1
=
=
1
1
1
V2 I =0 V2
+
+ C s + C s
C + C s +
1
1
C s
R
+ C s R
R

I
(d) h22 = 2
V

1
gmV +
+ C s V C s C s + gm +

R
=
=
1
1
I 1 =0
C + C s +
C + C s +
R
R
V
C s

SOLUTION PROBLEM 19.50: (a) Recall t-parameter relationship:

V1 t11
I = t
1 21

t12 V2
t22 I2

For the given network,V2 = ZL (I 2 ) and

V1 = t11V2 + t12 (I 2 ) = t11ZL ( I2 ) + t12 ( I2 ) = (t11ZL + t12 )(I 2 )
Further,
I1 = t21V2 + t22 ( I2 ) = (t21ZL + t22 )(I 2 )
Hence,
V
(t Z + t )(I 2 )
Zin = 1 = 11 L 12
=
I1 (t21ZL + t22 )(I 2 )

t11ZL + t12
t21ZL + t 22

(b) For the output impedance relationship, from the t-parameter relationships
V1 = Zs( I1 ) = t11V2 + t12 (I2 ) = Zs (t21V2 + t 22 ( I2 ))
Grouping V2 and I2 terms together on separate sides of the equation implies
Thus

(t 21Zs + t11)V2 = (t22 Zs + t12 )I 2

Prbs Chap 19, 1/13/02

P19-14

V
Zout = 2 =
I2

R. A. DeCarlo, P. M. Lin

t22Zs + t12
t21Zs + t11

SOLUTION PROBLEM 19.51:

45. From the z-parameter relationships
V1 = z11I1 + z12 I2

and

V1 z11I1 = z12 I2 = z12 ( I2 )

V2 = z21I1 + z22I 2 z21I1 = V2 + z22( I2 )

These two equations in matrix form are:

Solving for [V1

I1 ] yields

1 z11 V1 0 z12 V2
0 z I = 1 z I

21 1
22 2

V1
1 z21
I = z 0
1
21

z11 0 z12 V2 1 z11

=
1 1 z22 I2 z21 1

z V2
z22 I2

SOLUTION PROBLEM 19.52: For figure 19.52a, by inspection

V1 1 Z1 V2
I = 0 1 I
1
2
Therefore T is as indicated in the problem.
For figure 19.52b, by inspection
V1 1
I = Y
1 2

0 V2
1 I2

Therefore T is as indicated in the problem.

SOLUTION PROBLEM 19.53: Here we use the results of problem 19.52:

(a) For figure (a)
1
Tnew =
0
(b) For figure (b)

Z1 1
1 Y2

0 1+ Z1Y2
=
1 Y2

Z1
1

Prbs Chap 19, 1/13/02

P19-15

1
Tnew =
Y2

R. A. DeCarlo, P. M. Lin

0 1 Z1 1
Z1
=

1 0 1 Y2 1 + Z1Y2

SOLUTION PROBLEM 19.54: By the properties of an ideal transformer,

V1 = nV2

I1 =

1
1
I 2 = ( I2 )
n
n

Therefore
V1 n 0 V2
I = 0 1 n I
1
2
with the t-parameters given by the 2x2 matrix.

SOLUTION PROBLEM 19.55: This problem uses the results of the previous two problems.
(a)
1+ Z1Y2 Z1 n 0 n (1+ Z1Y2 ) Z1 n
T=
=

1 0 1 n nY2
1n
Y2
(b)
n 0 1 + Z1Y2
T=

0 1 n Y2

Z1 n (1+ Z1Y2 ) nZ1

=

1 Y2 n
1 n

SOLUTION PROBLEM 19.56:

(a)
0.25 0 8 4 2 1
T=
=
4 2 5 8 20
0
(b)
1 0 8
T=

1 R 1 2
(c)
1+ Z1Y2
T=
Y2

4 8
4
=

5 2 + 8 R 5 + 4 R

Z1 8 4 1+ 0.25s 0.5s 8 4 8 + 3s 4 + 3.5s

=
=
1 2 5 0.5
1 2 5 6
7

SOLUTION PROBLEM 19.57: For each 2-port of the form of figure P19.53a, we have that the tparameters are given by

Prbs Chap 19, 1/13/02

P19-16

R. A. DeCarlo, P. M. Lin

1+ Z1Y2 Z1
T=
1
Y2
The given network consists of three such sections in cascade whose t-parameters are respectiely,
1 + s2
T1 =
s

1+ 1 0.5s 1 1 + 0.5s 1
s
, T2 =
=
, and
1 0.5s
1
1
0.5s
1

1 2

1 + 2s 4 s 2s 1+ 2 s 2s
T3 =
= 1

s
1
s
1

4
4

Observe that
1 3 3 2 1

2
2 s + 2 s + 2 s + 1 s + s + 1
T1T2 =

1 2 3

s + s
s +1

2
2

and the overall t-parameters are

1 5 3 4

3 9 2 3
4
3
2
4 s + 4 s + s + 4 s + 4 s + 1 s + 3s + 2s + 3s + 1
T = T1T2T3 =

1 4 3 3 3 2 7

s + s + s + s
s3 + 3s2 + s + 1

4
4
4
4

SOLUTION PROBLEM 19.58: This poblem is done primarily in MATLAB.

Part (a)
% The following code solves part (a) of the problem.
% Parameter Specification
t11= 0.895+j*0.022;
t22= t11;
t12= 40 + j*180;
% t21= (t11*t22 -1)/t12;
% The above formula follows because it is a reciprocal network.
% The actual value is specified.
t21= -2.6175e-05+j*1.1023e-03;
t=[t11 t12; t21 t22]
% Part (a) calculations
vr= 115200
ir = 361
vs= t11*vr +t12*ir
magvs=abs(vs)
angvs= angle(vs)*180/pi

Prbs Chap 19, 1/13/02

P19-17

is=t21*vr+t22*ir
magis= abs(is)
angis= angle(is)*180/pi
pscomp=vs*conj(is)
ps=real(pscomp)
pr=real(vr*conj(ir))
eff= pr/ps
pf= ps/abs(pscomp)
ploss= ps- pr
The MATLAB output is as follows:
T=
8.9500e-01 + 2.2000e-02i 4.0000e+01 + 1.8000e+02i
-2.6175e-05 + 1.1023e-03i 8.9500e-01 + 2.2000e-02i
vr = 115200
ir = 361
vs = 1.1754e+05 + 6.7514e+04i
magvs = 1.3555e+05
angvs = 2.9872e+01
is = 3.2008e+02 + 1.3493e+02i
magis = 3.4736e+02
angis = 2.2857e+01
pscomp = 4.6733e+07 + 5.7501e+06i
ps = 4.6733e+07
pr = 41587200
eff = 8.8989e-01
pf = 9.9252e-01
ploss = 5.1458e+06
Part (b)
% The following code solves part (b) of the problem.
zL=500;
zin= (t11*zL + t12)/( t21*zL + t22)
yin = 1/zin
vsnew=134000;
iin= yin*vsnew
psnew= vsnew^2*real(yin)
m= inv(t)
v2=m(1,1)*vsnew +m(1,2) *iin
magv2=abs(v2)
iload= m(2,1) *vsnew + m(2,2)* iin
magild = abs(iload)
% Check value of rload
rload=abs(v2)/abs(iload)
The MATLAB output for part (b) is:

R. A. DeCarlo, P. M. Lin

Prbs Chap 19, 1/13/02

P19-18

R. A. DeCarlo, P. M. Lin

zin = 4.8759e+02 - 1.0031e+02i

yin = 1.9676e-03 + 4.0478e-04i
iin = 2.6366e+02 + 5.4241e+01i
psnew = 3.5331e+07
% m is the inverse of T-matrix.
m=
8.9500e-01 + 2.2000e-02i -4.0000e+01 - 1.8000e+02i
2.6175e-05 - 1.1023e-03i 8.9500e-01 + 2.2000e-02i
v2 = 1.1915e+05 4.6681e+04i
magv2 = 1.2796e+05
iload = 2.3829e+02 - 9.3362e+01i
magild = 2.5593e+02
rload = 5.0000e+02

SOLUTION PROBLEM 19.59. From the given information, the circuit is linear and reciprocal.
(a) Here i2 (t) is the integral of i1 (t). Therefore, the new v1(t) is the integral of the old v 2 (t).
The result for t 0 is:
v1(t) = 3.005 3e t + e t[0.00865sin(500t) 0.005cos(500t)]

= 3.005 3e t + e t 2 10 5 cos(500t /6) + 0.01sin(500t /6)

(b) From the problem statement

cos
(s + 1) + 250
6

3
+5
s +1
( s + 1)2 + (500)2
From reciprocity, z12 (s) = z21(s). For steady state analysis, we use phasors to obtain
z21 (s) =

V1 = z12 ( j500)I2 =

cos
(1+ j500) + 250
6

3
+5
1+ j500
(1+ j500) 2 + (500) 2

v1ss (t) = 2.505cos(500t 30.15o ) V

SOLUTION PROBLEM 19.60. Writing loop equations we have:

(i) For the left loop,
V1 aV2 3(I1 + I3 ) = 0 V1 aV2 = 3I1 + 3I3
(ii) For the right loop,

Prbs Chap 19, 1/13/02

P19-19

R. A. DeCarlo, P. M. Lin

V2 + bI1 (I 2 I 3) = 0 V2 = bI1 + I2 I3
(iii) For the middle loop,
bI1 + (I3 I 2 ) + I3 + 3(I1 + I3 ) = 0 0 = (b + 3)I1 I 2 + 5I 3
Writing the first two equations in matrix form yields
I
1 a V1 3 0 3 1
0 1 V = b 1 1 I 2

2

I3

whose solution is
V1 1
V = 0
2
Hence

I
a 3 0 3 1 3 ab
I =
1 b 1 1 2 b
I3

I
a 3 a 1
I
1 1 2
I3

V1 3 ab a I1 3 a
+
I
V = b
1 I2 1 3
2

From the third equation

I3 = [0.2b 0.6

I1
0.2]
I 2

Thus
V1 3 ab
V = b
2

a 3 a
+
[0.2b 0.6
1 1

I1
0.2]
I2

V1 0.6b 0.8ab + 0.6a + 1.2 0.6 + 0.8a I1

V =
0.8b + 0.6
0.8 I2
2
For reciprocity,

z12 = z21 a = b

SOLUTION PROBLEM 19.61. (a) and (b) together: Observe that

and

V1 0.5V1 I1 = V2 I1 = 0.5V1 V2 V1 = 2I1 + 2V2

The h-parameters are

I1 + I2 = 0.5(V2 aI1 ) I 2 = (0.5a + 1)I1 + 0.5V2

2
2

H=

(0.5a + 1) 0.5

Prbs Chap 19, 1/13/02

P19-20

Reciprocity requires that

Thus

R. A. DeCarlo, P. M. Lin

h12 = h21 2 = 0.5a + 1 a = 2

2
2
H=

2 0.5

SOLUTION PROBLEM 19.62. There are 2 corrections in the problem statement concerning the
second set of expressions:
(1) v1(t) = 2e-t - 1.5e -1.5t V
(2) i2(t) = 0.5 e-1.5t A
For both parts, recall, the y-parameters:
I1 y11 y12 V1
I = y y V
2 21 22 2
(a) Part-1: From the first set of given data (V2 = 0)
1
0.5
0.5
0.5
0.5(s + 2)
I1 = , I 2 =
, V1 =
+
=
2
s
s +1
s + 1 (s + 1)
(s + 1)2
Hence,
I
2(s + 1)2
I
s +1
y11 = 1
=
, y 21 = 2
=
V1 V =0 s(s + 2)
V1 V =0 s + 2
2
2
(a) Part-2: From the second set of given data (ZL = 1 , I1 = 1 s , etc.), we have
I2 =

0.5
2
1.5
0.5(s + 3)
, V1 =

=
s + 1.5
s + 1 s + 1.5 (s + 1)(s + 1.5)

For a terminated 2-port,

I2 =
Therefore

V2
y 21
y 21
= V2 =
V1 =
V
ZL
y 22 + YL
y 22 + 1 1

V
s +1
0.5(s + 3)
s + 1.5
(s + 3)
1
y 22 = y 21 1 1 =

1=
1 =
I2
s + 2 (s + 1)(s + 1.5)
0.5
(s + 2)
s+2

Also, I1 = y11V1 + y12V2

y12 =

I y V
I y V
y12 = 1 11 1 = 1 11 1 . Hence
V2
I 2

s + 1.5 1 2(s + 1)2 0.5(s + 3)

(2s + 3)(s + 2) 2(s + 1)(s + 3)

=
+

0.5
s(s + 2)
s(s + 2)
s s(s + 2) (s + 1)(s + 1.5)

Prbs Chap 19, 1/13/02

P19-21

R. A. DeCarlo, P. M. Lin

1
(s + 2)

In conclusion
2(s + 1)2
y11 y12 s(s + 2)
y y = s + 1
21 22
s+2

1
2(s + 1)2

(s + 2) = 1
s
1 (s + 2)
s
+
1

(s + 2)

(b) This is a straightforward application of the conversion table 19.1.

SOLUTION PROBLEM 19.63. (a) Consider figure (a). Write two mesh equations:
2
2
2

V1 = 16I1 + (I1 + I 2 ) + 4I1 = 20 + I1 + I2

s
s
s
2
2
2

V2 = sI2 + (I1 + I 2 ) + 4I1 = 4 + I1 + s + I2

s
s
s
Therefore

1 20s + 2 2
Z=
s 4s + 2 s2 + 2

Taking the inverse yields the y-parameters

Y=

1 z22
Z z21

s2 + 2
z12
1
2
=

z11 20s2 + 2s + 32 (4s + 2) 20s + 2

where
Z =

(20s + 2)(s2 + 2) 2(4 s + 2) 20s3 + 2s2 + 32s

=
= 20s2 + 2s + 32
s
s

Finally, the h-parameters are given as

1 y11
H=
z21 z22

z12 z22
1 20s2 + 2s + 32
=

1 z22 s2 + 2 (4 s + 2)

(b) Now consider figure (b). z11 is V1 when I2 = 0. But if I2 = 0, then because of the ideal
transformer I1 = 0, meaning that the ratio is not defined. Hence the z-parameters do not exist.
To find the h-parameters, observe that because of the ideal transformer, I2 = 0.5I1,
I1 = 2I2 , and V pri = 0.5Vsec . Writing a mesh equation at the right mesh first we obtain
V2 = RI 2 + Vsec + (I1 + I 2 )R = Vsec

Prbs Chap 19, 1/13/02

P19-22

R. A. DeCarlo, P. M. Lin

Now writing a mesh equation on the left we have

V1 = RI1 + V pri + (I1 + I 2 )R = 1.5RI1 + V pri = 1.5RI1 + 0.5V2
Therefore
1.5R 0.5
H=

0.5 0
To obtain the y-parameters we use table 19.1:
Y=

1 1 h12 1 4 2
=
h11 h21 h 6R 2 1

Note: the det[Y] = 0 implying again that the z-parameters do not exist.
(c) For this network we consider it as a cascade (left to right) of an ideal transformer, the middle
network of a transformer and an inductor across the top, and finally another ideal transformer.
The t-parameters of these two ports are respectively:
a 0
A B
1 b 0
T1 =
, T2 =
, T3 =

0 1 a
C D
0 b
To find T2 , we replace the mutually coupled inductors by the pi-equivalent circuit of figure

18.25c where L1 = 4 H, L2 = 9 H, M = k L1L2 = 3 H, and = 27. Thus Lleft =

= 4.5
L2 M

H, Lright =
= 27 H, and Ltop =
= 9 H. Notice that the external 9 H inductor is in
L1 M
M
parallel with Ltop leading to Lpar = 4.5 H. The y-matrix of this new pi-network is by inspection:
1 4.5
1 2 4.5
1 12 6
Ymid =
=

s 1 4.5 1 27 + 1 4.5 27s 6 7

To compute the t-parameters we have
y 22
y
21
T2 = y

y 21

1
y 21 7 6 4.5s
y11 = 8 27s 2
y 21

Therefore
a 0 7 6 4.5s 1 b 0 7a 6b 4.5abs
T = T1T2T3 =

0 1 a 8 27s 2 0 b 8 27abs 2b a
From table 19.1, we obtain

Prbs Chap 19, 1/13/02

P19-23

63a2

0.5a / b
H=
, Z = s 16
2
27ab
0.5a / b 4 27b

8
a 2 2.25s

R. A. DeCarlo, P. M. Lin

27ab

12b2 6ab
1
8 , Y =

27b 2
27a2b 2s 6ab 7a2

SOLUTION PROBLEM 19.64.

(a) The defining equation for the g-parameters is:
I1 g11
V = g
2 21

g12 V1
g12 I2

Because of the load impedance Z L, we have V2 = ZL I 2 . Hence substituting for V2 in the

second equation yields
g21V1
V2 = g21V1 + g22 I2 = ZL I 2 I 2 =
g22 + ZL
Substituting this equation into I1 = g11V1 + g12 I2 we obtain
g V
g g
I1 = g11V1 g12 21 1 = g11 12 21 V1
g22 + ZL
g22 + ZL
Therefore,
g g
Yin = g11 12 21
g22 + ZL
(b) Because of the source impedance Zs , we have V1 = Zs I1 or I1 = YsV1 . Hence substituting
for V1 in the first equation yields
g12 I2
I1 = g11V1 + g12 I2 = YsV1 V1 =
g11 + Ys
Substituting this equation into V2 = g21V1 + g22 I2 we obtain
g I

g g
g g
V2 = g21 12 2 + g22I 2 = g22 12 21 I2 Zout = g22 12 21
g11 + Ys

g11 + Ys
g11 + Ys
(c)

Ys (g22 + ZL )
V
Zin
Ys
Ys
G1 = 1 =
=
=
=
Vs Zs + Zin Ys + Yin Y + g g12g21
( g22 + ZL )( g11 + Ys ) g12g21
s
11
g22 + ZL

(d) Refer to figure P19.64, where V2 =

ZL
V
g Z
g21V1 G2 = 2 = 21 L .
g22 + ZL
V1 g22 + ZL

Prbs Chap 19, 1/13/02

P19-24

R. A. DeCarlo, P. M. Lin

Ys (g22 + ZL )
g Z
g21Ys ZL
(e) Gv = G1G2 =
21 L =
(g22 + ZL )( g11 + Ys ) g12 g21 g22 + ZL (g22 + ZL )(g11 + Ys ) g12g21

SOLUTION PROBLEM 19.65. Recall

I1 g11
V = g
2 21

g12 V1
g22 I2

For the given circuit

V2 =

1
1
1
(I2 4V2 ) + V1 V2 = V1 + I 2
4
2
8

Also
V2 = V1

1
(I sV1 )
4 1

This implies that

1
1
1
I1 = (4 + s)V1 4V2 = (4 + s)V1 4 V1 + I2 = (2 + s)V1 I2
2
8
2
Thus the g-parameter matrix is:
1

2 + s 2
G= 1
1

2
8

SOLUTION PROBLEM 19.66. We first convert the y-parameters to t-parameters using table 19.1:
y 22
y
21
YN1 TN 1 = y

y 21

1
y 21 0.1 0.5
y11 = 3.82 25.1
y 21

where y = 50.2 0.2 2 1.2 = 7.64 . Now we convert the h-parameters to t-parameters using
table 19.1:
h
h
HN 2 TN 2 = h21
21
h21

h11
h21 1 56 13 11.2 2.6
1 = 5 2 1 = 0.4 0.2
h21

where h = 13 2 + 6 5 = 56 . To obtain the cascaded t-parameters we compute

Prbs Chap 19, 1/13/02

P19-25

Tcas = TN1TN 2 =

R. A. DeCarlo, P. M. Lin

1.8 1.32
0.36
1 6.6
=

5 264.12 74.76 52.824 14.952

Thus,
603
1
t21ZL + t22
5 = 40.2 S
Yin =
=
= 15
Zin t11ZL + t12
5
To obtain the voltage gain, we first convert the t-parameters back to y-parameters (table
19.1) and then use the derived voltage gain formula:
1.32
Tcas =
52.824
Hence,

0.36
2
41.533
Ycas =

S
14.952
2.7778 3.6667

V
y 21
2.7778
Gv = L =
=
= 0.66667
V1 y 22 + YL 3.6667 + 0.5

Alternately, one could consider the load as a 2-port, compute its t-parameters, construct
1
the overall t-parameters as a cascade of three networks, and then use Gv =
.
t11

SOLUTION PROBLEM 19.67.

(a)

Since only one C or L , weh have in general: H(s) =

have that a = 0. Therefore,

H(s) =

b
c

K
=
d s+ c
s+
c

Rth
where Rth is the Thevenin resistance seen by the energy
Rth C
L
storage element, we refer the reader to problem 19.69 which provides a general derivation with
H() arbitrary; hence this problem is the special case of H() = 0.
(b) To prove that

as + b
a
. Since H() = 0 = , we
cs + d
c

or

(c) Here Rth = 2000/ /(50 + 1000/ /200)) = 195.49 . Hence

c

1
Rth C

= 25.58 10 6 rad/s

Prbs Chap 19, 1/13/02

P19-26

R. A. DeCarlo, P. M. Lin

SOLUTION PROBLEM 19.68.

(a) Because there is only one energy storage element, it follows that the most general form of
as + b
b
the transfer function is: H(s) =
. Since H(0) = 0 = , we have b = 0. Further, since the
cs + d
d
transfer function is high pass, c 0, and
a
s
c
as
Ks
H(s) =
=
=
cs + d
d s + c
s+
c
1
R
(b) To prove that c =
or th where Rth is the Thevenin resistance seen by the energy
Rth C
L
storage element, we refer the reader to problem 19.69 which provides a general derivation with
H(0) arbitrary; hence this problem is the special case of H(0) = 0.
(c) Let us apply the result of example 19.1 to that part of the circuit to the right of the 1 k
resistor and call the associated resistance Zin. Here Z1 = , Z2 = 2 k, ZL = 100 , and beta =
50. Hence
Zin = (beta + 1) ZL = 5.1 k

Thus
Hence,

Rth = 200 + 1000 //(2000 + 5100) = 1076.5

1
Rth C

1
= 464.45 rad/s
1076.54 2 106

SOLUTION PROBLEM 19.69. In this problem we assume (i) a single input single output system
and that linear circuit seen by the energy storage element has a Thevenin equivalent or a
Norton equivalent. For simplicity we will presume the existence of a Thevenin equivalent.
(a) The Thevenin equivalent seen by the dynamic element L or C consists of Zth(s) in series with
Voc(s). Since the remainder network seen by L or C is non-dynamic (resistive, resistive
with dependent sources and ideal op amps, etc), we have Zth(s) = Rth and Voc(s) =
KoInput(s), R th and Ko being real constants.

Prbs Chap 19, 1/13/02

P19-27

R. A. DeCarlo, P. M. Lin

By voltage division,
VL =

Ls
K 0s
Voc =
R Input(s)
Ls + Rth
s + th
L

(1)

and
K0
1
RthC
VC = 1 Cs Voc =
1 Input(s)
+ Rth
s+
Cs
RthC

(2)

After VL(s) or V C(s) has been determined, we can find the Laplace transform of any other output
(voltage or current) using the voltage source substitution theorem (chapter 6) and linearity
(chapter 5):
Ouput(s) = K1 Input(s) + K 2 (VL (s)

or

VC (s))

(3)

For the case of VL(s),

K0 s

Ouput(s) = K1 + K2
R Input(s)

s + th

(4)

and for the case of VC(s),

K0

RthC
Ouput(s) = K1 + K2
1 Input(s)

s+

Rth C

(5)

For either case, the transfer function H(s) has the form
H(s) =

Ouput(s) K3s + K4
=
Input(s)
s + c

(6)

Prbs Chap 19, 1/13/02

P19-28

R. A. DeCarlo, P. M. Lin

R
1
where c = th or
with K3 and K4 real constants.
L
RthC
It remains to give K3 and K4 some physical interpretations. In (6), let s = , we have
K3s + K4
= K3

s + c s=

H() =

On the other hand, letting s = 0 in (6) produces

H(0) =

K3 s + K 4
K
= 4

s + c s=0 c

Substituting K3 and K4 into (6), we obtain the desired result

H(s) =

Ouput(s) H ()s + c H(0)

=
Input(s)
s + c

(b) When s = , the impedance of C is

ZC () =
and the impedance of L is

(7)

1
=0
Cs s=

ZL () = Ls]s= =

Therefore in calculating H(), we may replace C by a short circuit and L by an open circuit. On
the other hand, when s = 0, the impedance of C is
ZC (0) =

1
=
Cs s=0

and the iimpedance of L is

ZL (0) = Ls ]s=0 = 0
Therefore in calculating H(0), we may replace C by an open circuit and L by a short circuit.

Prbs Chap 19, 1/13/02

P19-29

R. A. DeCarlo, P. M. Lin

(c) For figure P19.69a, by inspection

Rth = 3//(2 + 4) = 3//6 = 2
c = 1/(Rth C) = 1/(2 0.5) = 1
H() = 4/(2 + 4) = 2/3
H(0) = 4/(2 + 3 + 4) = 4/9
Therefore
2
2
4 2
s +
H()s + c H(0) 3 s + 9 3
3
H(s) =
=
=
s + c
s+1
s +1
For figure P19.69b
Rth = 3//(2 + 4) = 3//6 = 2
c = Rth/ L = 2/2 =1
H() = 4/(2 + 4) = 2/3
H(0) = 4/(2 + 3 + 4) = 4/9
Therefore
4
2 4
s
+
(s + 1.5)
H()s + c H(0) 9
3
9
H(s) =
=
=
s + c
s+1
s +1

SOLUTION PROBLEM 19.70. According to problem 19.68, the transfer function of the circuit is

H(s) =

Ks
s+ c

1
where c =
and Rth is the Thevenin resistance seen by the storage element C. To find Rth
Rth C
we make use of figure 19.4 and the associated formula. The details are in the MATLAB code
below:
R1 = 200*1e3/1200
R1 =
1.6667e+02
Z1 = R1+2e3
Z1 =
2.1667e+03
Z3 = 100;

Prbs Chap 19, 1/13/02

P19-30

R. A. DeCarlo, P. M. Lin

beta = -50;
Zout = Z1/(1+beta)
Zout =
-4.4218e+01
Rth = 2000 + Zout
Rth =
1.9558e+03
wc = 1/(Rth*2e-6)
wc =
2.5565e+02
Hence
c

1
= 255.65 rad/s
1955.8 2 106

SOLUTION PROBLEM 19.71. (a) Except for the terminating resistor, let the other element
branches of the circuit be given by
1 =

1
1
2s
1
, Y2 =
= 2 + 1, and Z3 =
1
3s1
s
4s
2s +
2s

Consider the circuit

Here from problem 19.53a and from 19.52 a we obtain

Prbs Chap 19, 1/13/02

P19-31

V1 1+ 23
I = 4s + 1
1
2s
4s2 + 1

R. A. DeCarlo, P. M. Lin

1
V3 1
V3
3s and =
I 3
I 3 0
1

1 V
2
s I
1 2

which implies that

(16s2 + 6) V2

V1
1 4s2 + 5/ 3
I = 2

1 4s + 1
2s

I
3s
2
2
4s + 3

(b)
GV (s) =

y 21
y 22 + y L

(c)
5

3s 4s2 +
2

1
3s(4s + 1)
t11
3 s(12s + 5)
y 21 =
=
,
y
=
=
=
, and YL = 1
22 t
t12
16s2 + 6
16s2 + 6
16s2 + 6
12
2

(d) Hence,
3s(4s2 + 1)
GV (s) =

2
y 21
= 16s2 + 6
y 22 + y L s(12s + s)

16s2 + 6

=
+1

3s(4 s2 + 1)
12s3 + 16s2 + 5s + 6

SOLUTION PROBLEM 19.72. For part (a), treating each capacitor as a short circuit yields the
equivalent circuit below.

Prbs Chap 19, 1/13/02

P19-32

R. A. DeCarlo, P. M. Lin

h11T = 4.2e3; h12T = 0; h21T = 150; h22T = 0.1e-3;

hT = [h11T, h12T;h21T h22T];
yT = htoy(hT)
yT =
2.3810e-04
0
3.5714e-02 1.0000e-04
By inspection, the y11 parameter of the overall two-port consists of the sum of y11T plus the
conductances of the two front end resistors. Also, the y22 parameter of the overall two port
is y22T plus the conductance of the 4.7k resistor. Hence,
y = yT + [1/1e4+1/1e5 0;0 1/4700]
y=
3.4810e-04
0
3.5714e-02 3.1277e-04
Hence, the overall h-parameters are:
h = ytoh(y)
h=
2.8728e+03
0
1.0260e+02 3.1277e-04
(b)
yout = h(2,2) - (h(1,2)*h(2,1)/(h(1,1)+50))
yout =

Prbs Chap 19, 1/13/02

P19-33

R. A. DeCarlo, P. M. Lin

3.1277e-04
Zout = 1/yout
Zout =
3.1973e+03
a = sqrt(8/Zout)
a=
5.0021e-02
(c) To compute the gain we first need Zin . From equation 19.49, since h12 = 0, Zin = h11.
Zin = h(1,1)
Zin =
2.8728e+03
From equation 19.51,
Gv2 = -h(2,1)/(Zin*(h(2,2)+h(2,2)))
Gv2 =
-5.7094e+01
From equation 19.52,
Gv1 = Zin/(Zin + 50)
Gv1 =
9.8289e-01
To compute VL/Vs we use:
Gv = Gv1*Gv2*a
Gv =
-2.8071e+00
To compute VL/V1 we use:
Gvv = Gv2*a
Gvv =

Prbs Chap 19, 1/13/02

P19-34

R. A. DeCarlo, P. M. Lin

-2.8559e+00
(d) For these calculations, we assume Vs is normalized to 1 V. Since we are computing gains,
we may do this without loss of generality.
Vs = 1;
Is = 1/(50 + Zin)
Is =
3.4214e-04
Now we compute the normalized power delivered by the voltage source.
Psnorm = Vs*Is
Psnorm =
3.4214e-04
Next we compute the normalized power absorbed by the load.
VL = Gv*1
VL =
-2.8071e+00
PLnorm = VL^2/8
PLnorm =
9.8496e-01
Next, the power gain from source to load is:
GpLs = PLnorm/Psnorm
GpLs =
2.8788e+03
Further, we compute the power gain from input to the two port to the load as follows:
V1 = Vs*Zin/(Zin + 50)
V1 =

Prbs Chap 19, 1/13/02

P19-35

R. A. DeCarlo, P. M. Lin

9.8289e-01
P1 = V1*Is
P1 =
3.3629e-04
GpL1 = PLnorm/P1
GpL1 =
2.9289e+03
(e) SPICE Simulation Because the frequency response is flat for freqency above 800 Hz, we
only plotted up to 1.6k Hz. The circuit diagram reflects the load back to the primary of the
ideal transformer. In general, this is not possible. Hence for a SPICE simulation, it is
necessary to use one of the models given in Figure 18.15 consisting of two controlled
sources. For this example, this is not necessary. Note however that the actual output
voltage is 0.05 times the values on the graph given below. This simulation assumes a 1 V
source voltage and the parameter of GVCCS is 0.035714. Notice that in this problem

Prbs Chap 19, 1/13/02

P19-36

R. A. DeCarlo, P. M. Lin

Prb 19.72-Small Signal AC-3

+200.000

+400.000

+600.000

+800.000

Frequency
+1.000k

+1.200k

+1.400k

(Hz)

+1.600k

+60.000

+50.000

+40.000

+30.000

+20.000

+10.000

+0.000e+000

MAG(V(IVM2))

(f) For this part, we change 100 F to 10 F. The resulting plot shows degradation of the low
end frequency response.
Prb 19.72-Small Signal AC-4
+200.000
+60.000

+50.000

+40.000

+30.000

+20.000

+10.000

+0.000e+000

MAG(V(IVM2))

+400.000

+600.000

+800.000

Frequency
+1.000k

+1.200k

+1.400k

(Hz)

+1.600k

SOLUTIONS PROBLEMS CHAPTER 20

USEFUL MATLAB M-FILES FOR USE IN THE SOLUTION TO PROBLEMS IN THIS CHAPTER.
Program 1: converts y-parameters to t-parameters
% convert y parameters to t parameters
function [t,t11,t12,t21,t22] = ytot(y)
y11=y(1,1);
y12=y(1,2);
y21=y(2,1);
y22=y(2,2);
deltay= y11*y22-y12*y21;
t11=-y22/y21;
t12 = -1/y21;
t21= -deltay/y21;
t22= -y11/y21;
t= [ t11 t12; t21 t22];
Program 2: Computes Zin, Zout, and gains using two port h-parameters.
% two-port analysis in terms of h-parameters
function [zin, zout] =twoport(h, zL, zs)
['twoport analysis using h-parameters']
h11= h(1,1); h12=h(1,2); h21=h(2,1); h22=h(2,2);
zin = h11 - h12*h21/(h22+ 1/zL)
yout= h22 - h12*h21/(h11+zs);
zout= 1/yout
v1tovs= zin/(zin+zs)
v2tov1= -h21/(zin*(h22+1/zL))
v2tovs= v1tovs*v2tov1
Program 3: Computes Zin, Zout, and gains using two port y-parameters.
% two-port analysis in terms of y-parameters
function [zin, zout] =twoporty(y, zL, zs)
['twoport analysis using y-parameters']
y11= y(1,1); y12=y(1,2); y21=y(2,1); y22=y(2,2);
yin = y11 - y12*y21/(y22+ 1/zL)
zin= 1/yin
yout= y22 - y12*y21/(y11+1/zs)
zout= 1/yout
v1tovs= zin/(zin+zs)
v2tov1= -y21/(y22+1/zL)
v2tovs= v1tovs*v2tov1
Program 4: Computes Zin, Zout, and gains using two port t-parameters.
% two-port analysis in terms of t-parameters
function [zin, zout] =twoportt(t, zL, zs)

['analysis of terminated twoport using t-parameters']

t11= t(1,1); t12=t(1,2); t21=t(2,1); t22=t(2,2);
zin= (t11*zL + t12)/(t21*zL + t22)
zout= (t22*zs + t12)/(t21*zs + t11)
v2tov1= zL/(t11*zL + t12)
v1tovs= zin/(zin+zs)
v2tovs= v2tov1*v1tovs
Program 5: converts z-parameters to t-parameters
%converting z to t paramters (same sormulas as
%converting t to z parameters)
function [t,t11,t12,t21,t22] = ztot(z)
z11=z(1,1); z12=z(1,2); z21=z(2,1); z22=z(2,2);
deltaz= z11*z22 - z12*z21;
t11= z11/z21;
t12= deltaz/z21;
t21= 1/z21;
t22= z22/z21;
t= [ t11 t12; t21 t22];

SOLUTION 20.1.
For network a, the z-parameters are by inspection
R1 + R3
Za =
R3

R3
R2 + R3

Similarly, for network b, the z-parameters are the same:

R3
R1 + R3
Zb =
R2 + R3
R3
The interconnection of networks a and b conforms to figure 20.2b, which is a series
interconnection. Hence, the new overall z-parameters are
R1 + R3
Znew = Za + Zb = 2
R3

R3
R2 + R3

SOLUTION 20.2. For networks a and b, the y-parameters are by inspection

3 2
Ya = Yb =
S
2 4
Hence, their z-parameters are the inverse of the y-parameter matrix:
1 4
Za = Zb =
8 2

The interconnection of networks a and b conforms to figure 20.2b, which is a series

interconnection. Hence, the new overall z-parameters are
Znew = Za + Zb =

1 4 2

4 2 3

SOLUTION 20.3. For network a consisting of the single inductor,

s s
Za =

s s
For network b, we have
Zb = [Yb ]

1 1
2 2

1 0.5 0.5

=
4 1
2

The interconnection of networks a and b conforms to figure 20.2b, which is a series

interconnection. Hence, the new overall z-parameters are
s + 0.5 s + 0.5

Znew = Za + Zb =
s + 2
s +1
SOLUTION 20.4. For network a consisting of the single capacitor,
1 1 1

Za =
s 1 1
The interconnection of networks a and b conforms to figure 20.2b, which is a series
interconnection. Hence, the new overall z-parameters are
1 1 + 0.5s 1+ 0.5s

Znew = Za + Zb =
s 1+ s
1 + 2s

SOLUTION 20.5. For network a, the y-parameters are by inspection

3 2
Ya =
S
2 3
Hence, their z-parameters are the inverse of the y-parameter matrix:
Za =

1 3 2 0.6 0.4

=
5 2 3 0.4 0.6

The interconnection of networks a and b conforms to figure 20.2b, which is a series

interconnection. Hence, the new overall z-parameters are
1.6
Znew = Za + Zb =
0.4

0.9

0.4

SOLUTION 20.6. For networks Na and Nb, the y-parameters are:

0.7 0.2
YNa = YNb =
S
0.2 0.7
Hence, their z-parameters are the inverse of the y-parameter matrix:
ZNa = ZNb =

20 0.7 0.2 1 14 4

=
9 0.2 0.7 9 4 14

The network Na* has the same z-parameters as Na and continues to act as a two when series
interconnected to another 2-port because of the transformer. Hence, the interconnection of
networks Na* and Nb forms a valid series interconnection in which cas
Znew = ZNa * + ZNb =

2 14 4

9 4 14

SOLUTION 20.7. For network Nb, the y-parameters are the same as in problem 20.6, i.e.,
0.7 0.2
1 14 4

YNb =
S and ZNb =

9 4 14
0.2 0.7

For network Na, consider the purely resistive part without the transformer. The y-parameters of
this part are half the y-parameters of Nb, i.e.,
0.35 0.1
YR =
S
0.1 0.35
In MATLAB we use the m-file which converts y-parameters to t-parameters:
y = [0.35 -0.1;-0.1 0.35];
t = ytot(y)
t=
3.5000e+00 1.0000e+01
1.1250e+00 3.5000e+00

From figure 19.34, the transformer has to parameters with a = 2 given by

1 a 0 0.5 0
ttrans =
=

0 a 0 2
ttrans = [0.5 0;0 2];
tNa = ttrans*t
tNa =
1.7500e+00 5.0000e+00
2.2500e+00 7.0000e+00

zNa = ttoz(tNa)
zNa =
7.7778e-01 4.4444e-01
4.4444e-01 3.1111e+00

zNb = [14 4;4 14]/9;

znew = zNa + zNb
znew =
2.3333e+00 8.8889e-01
8.8889e-01 4.6667e+00

SOLUTION 20.8. This problem is identical numerically to problem 20.6. Here however the
isolation transformer is on the right hand side which makes no difference to the interconnection.
Therefore,
Znew = ZNa * + ZNb =
SOLUTION 20.9. Figure 20.4 is

2 14 4

9 4 14

Using the values in figure 20.3, we obtain the following mesh equation matrix
V1 4 + R1

V2 = 2
0 R1

2
4 + R2
R2

R1 I1

R2 I 2
R1 + R2 I 3

For I3 to be zero, the third equation implies that R1I1 = R2I2. Therefore,

R
2R
V1 = (4 + R1) I1 + 2 1 I1 = 4 + R1 + 1 I1
R2

R2
Similarly

R
(4 + R2 )R1
V2 = 2I1 + ( 4 + R2 ) 1 I1 + = 2 +
I1
R2

R2
Therefore

2R1
4 + R1 +

R2
V1
(4R2 + R1R2 + 2R1) = 42 = 7
=
=
V2
( 4 + R2 )R1 (2R2 + R1R2 + 4R1) 48 8
2 +

R2

as was to be shown.
V
7
7
Now suppose, 1 = or equivalently V1 = V2 . With specific values
V2 8
8
V1
7 8 10 2 6 I1



V2 = V2 1 = 2 7 3 I 2
0
0 6 3 9 I3
In MATLAB
Z = [10 2 -6;2 7 3;-6 3 9];
b = [7/8 1 0]';
I = inv(Z)*b
I=
6.2500e-02
1.2500e-01

2.7756e-17
Thus
I1
0.0625

I 2 = V2 0.124
I3
0
SOLUTION 20.10
(a) Write two loop equations assuming I1 and I2 are the usual port currents. Here
V1 = Z1I1 + V13 + Z3( I1 + I2 )
and

V2 = Z2 I2 + V23 + Z3(I1 + I 2 )

which in matrix form are

V1 Z1 + Z3
V = Z
2
3

Z3 I1 V13
+
Z2 + Z3 I 2 V23

However,
V13
I1
V = Z A I
23
2
Therefore
V1 Z1 + Z3
V = Z
2
3

Z3 I1
I1 Z1 + Z3
+ ZA =

Z2 + Z3 I 2
I2 Z 3

Z3
I1
+ Z A

Z2 + Z3
I 2

(b) The procedure of part (a) is repeated to produce the same result.
REMARK: this problem states that the two networks are equivalent two ports. Thus the
configurations can be used interchangeably.

SOLUTION 20.11. Because of the isolation transformers, the overall Z-parameters are the sum of
the three component Z-parameters. For Na,
2 1
Za =

1 2
For Nb,

For Nc,

8 1
Zb =

1 5
2 1
Zc = Za =

1 2

Therefore
12 3
Z = Za + Zb + Zc =

3 9
SOLUTION 20.12. (a) Given the Z-parameters of N, then
YN =

ZN1

1
7 2
3 2
=
=

S
10 3
10 7

The 1-F capacitor considered as a two port has y-parameters

s s
YC =
S
s s
Therefore, the parallel connection of the two ports has y-parameters
(s + 2)
s s 3 2 s + 3
S
Y = YN + YC =
+
=

s + 7
s s 10 7 (s + 10)
(b) For this part, the same reasoning applies with s replaced by 1/s.
SOLUTION 20.13. The isolation transformer allows for the valid parallel connection of N *a and
Nb in the sense that the overall y-parameters are the sum of the individual y-parameters. Further,
because the ideal transformer is 1:1 with the standard dot locations, the y-parameters of N *a are
those of N a . Further, the y-parameters of Nb are the same as those of N a as the circuits are
simple vertical flips of each other. Therefore
0.7 0.2
Ya* = Ya =
= Yb
0.2 0.7
Hence, the overall y-parameters are:
0.7 0.2 1.4 0.4
Y = 2
=
S
0.2 0.7 0.4 1.4
True-False: FALSE because the connection does not conform to figure 20.2a.

SOLUTION 20.14. From the previous example, the y-parameters of Nb are

 0.7 0.2
Yb =
S
0.2 0.7
The resistance values of the resistive part of Na are twice those of Nb. Hence, the y-parameters
of the resistive part are half those of Nb, i.e.,
I1'
V ' 0.35 0.1 V1'
= Ya,R 1 =
S
I 2
V2 0.1 0.35 V2
We obtain the y-parameters of Na by considering the effect of the transformer on these yparameters. Observe that
I1 = 2I1' and V1 = 0.5V1'
Hence
I1 0.7 0.2 V1' 0.7 0.2 2V1 1.4 0.2 V1
I = 0.1 0.35 = 0.1 0.35 V = 0.2 0.35 V
2
V2
2
2
Thus

It follows that

SOLUTION 20.15. (a)

1.4 0.2
Ya =
S
0.2 0.35
2.1 0.4
Y = Ya + Yb =
S
0.4 1.05

Prob 20.15-Small Signal AC-10

+100.000m

+200.000m

Frequency

+300.000m

+1.000

+800.000m

+600.000m

+400.000m

+200.000m

+0.000e+000

MAG(V(IVM))

(b)

C0
15.9n

C1
15.9n

R1
500

R
1k

R0
1k

V
0

IVm
C
31.8n

+400.000m

(Hz)

+500.000m

Prob 20.15-Small Signal AC-11

+10.000k

+20.000k

Frequency
+30.000k

+1.000

+800.000m

+600.000m

+400.000m

+200.000m

+0.000e+000

MAG(V(IVM))

SOLUTION 20.16. (a) For network Na, the y-parameters by inspection are:
0
G1 + j C1 0 2 + j10.21
mS
Ya =
=

G0
95
0.07143
gm
For network Nb, the y-parameters by inspection are:
1 1
1 1
Yb = G f + j C2
= (0.8333 + j1.021)

mS
1 1
1 1

Therefore

2.8333 + j11.23 0.8333 j1.021

Y = Ya + Yb =
mS
94.167 j1.021 0.9048 + j1.021

(b), (c), and (d). Here we use the MATLAB m-file for two port analysis in terms of yparameters:

(Hz)

zs = 50; zL = 50;
twoporty(Y, zL, zs)
ans =
twoport analysis using y-parameters
yin =
6.8501e-03 + 1.5594e-02i
zin =
2.3614e+01 - 5.3756e+01i
yout =
5.3617e-03 + 3.0023e-03i
zout =
1.4199e+02 - 7.9506e+01i
v1tovs =
5.5701e-01 - 3.2349e-01i
v2tov1 =
-4.4915e+00 + 2.6821e-01i
v2tovs =
-2.4150e+00 + 1.6023e+00i

SOLUTION 20.17. The t-parameters of the LR circuit follow from problem 19.53 with Z1 = Ls =
s and Z2 = 0.5 :

Therefore

1+ 2s s
TLR =
1
2
5 + 2s 2 + s
Tcascade = T * TLR =

5 + 2s 2 + s

SOLUTION 20.18. This problem can be solved in many ways. Here we emphasize the cascade
nature of the two ports.
% The y-parameters of Nb are:
Yb = [8 2;20 6];
% The z-parameters of Na are:
Za = [0.75 -0.25;-2.5 1];
% The t-parameters of Na are:
Ta = ztot(Za)
Ta =
-3.0000e-01 -5.0000e-02
-4.0000e-01 -4.0000e-01
% The t-parameters of Nb are:
Tb = ytot(Yb)
Tb =

-3.0000e-01 -5.0000e-02
-4.0000e-01 -4.0000e-01
% The t-parameters of the cascaded two port are:

Tab = Ta*Tb
Tab =
1.1000e-01 3.5000e-02
2.8000e-01 1.8000e-01
% Doing a t-parameter analysis we obtain:
twoportt(Tab,0.25,0.5)
ans =
analysis of terminated twoport using t-parameters
zin =
2.5000e-01
zout =
5.0000e-01
v2tov1 =
4
v1tovs =
3.3333e-01
v2tovs =
1.3333e+00
ans =
2.5000e-01

SOLUTION 20.19. Use the h-parameter analysis m-file.

H1 = [1e3 0.001;50 6e-5]; H2 = [1e3 0.99;-5 8e-4];
zL = 100; zs = 2e3; zm = 10e3;
% For part (a), the common collector stage:
twoporth(H2,zL,0)
ans =
twoport analysis using h-parameters
zin =
1.4583e+03
zout =
1.7391e+02
v1tovs =
1
v2tov1 =
3.1746e-01
v2tovs =
3.1746e-01

Remark: in the above, v2 is Vout and v1 is vm; zout has no signifigance.

% We now compute the load on the first stage.
zin2 = 1.4583e+03;
zLm = zm*zin2/(zin2 + zm)
zLm =
1.2727e+03
% zLm is the load impedance on stage 1.
% For part (a), the common emitter stage:
twoporth(H1,zLm,zs)
ans =
twoport analysis using h-parameters
zin =
9.4088e+02
zout =
2.3077e+04
v1tovs =
3.1993e-01
v2tov1 =
-6.2835e+01
v2tovs =
-2.0103e+01
Remark: in the above, v2 is vm and v1 is Vin; zout is the output impedance of stage 1.
Conclusion: the input impedance to stage 1 is 940.88 and the input impedance to stage 2 is
1.4583 k.
(b) From the above output and remark, Vm/Vin = -62.835. Further Vout/Vm = 0.31746.
(c) From above Vm/Vs = -20.103. Therefore Vout/Vs = Vout/Vm * Vm/Vs = -6.38.
(d)
zout1 = 2.3077e+04;
zs2 = zm*zout1/(zm + zout1)
zs2 =
6.9768e+03
twoporth(H2,zL,zs2)
ans =
twoport analysis using h-parameters
zin =
1.4583e+03
zout =
7.0395e+02
v1tovs =

1.7289e-01
v2tov1 =
3.1746e-01
v2tovs =
5.4885e-02
Conclusion: Zout of amplifier is 704 .
SOLUTION 20.20
(a) Using MATLAB
na = 1.1514;
nb = 3.4012;
Zlprime = nb^2*75 + j*1042.94
ZLprime =
8.6761e+02 + 1.0429e+03i
Zsprime = 75/na^2 + j*30
Zsprime =
5.6573e+01 + 3.0000e+01i
(b) Since the h-parameters of the transistors are given, we can again use MATLAB and the mfile twoporth defined earlier. Hence:
h = [ 60-j*50 0.01; -j*2 0.0005+j*0.0004];
[Zin, Zout] = twoporth(h,ZLprime, Zsprime)
Zin =
5.6569e+01 - 3.0000e+01i
Zout =
8.6763e+02 - 1.0429e+03i
(c) Observe that Zin = 5.6569e+01 - 3.0000e+01i and Zsprime = 5.6573e+01 + 3.0000e+01i,
which are clearly conjugates of each other. Further, Zout = 8.6763e+02 - 1.0429e+03i
and ZLprime = 8.6761e+02 + 1.0429e+03i, which are also conjugates of each other.
Hence maximum power is transferred into and out of the transistor.
(d) For this part, we change all cascaded two ports to t-parameters. Specifically,
t0 = [1 75;0 1];
t1 = [na 0; 0 1/na];
t2 = [ 1 j*30; 0 1];
t3 = htot(h);
t4 = [ 1 j*1042.9; 0 1];
t5 = [nb 0; 0 1/nb];
t6 = [1 0;1/75 1];
t = t0*t1*t2*t3*t4*t5*t6

t=
2.2176e-01 - 2.5959e-01i 8.3160e+00 - 9.7351e+00i
1.4785e-03 - 1.7307e-03i 6.6577e-02 - 7.4415e-02i
gain = 1/t(1,1)
gain = 1.9024e+00 + 2.2270e+00i
gainmag = abs(gain)
gainmag = 2.9289e+00
gainangle = angle(gain)*180/pi
gainangle = 4.9494e+01
In this case, Vout/Vs = gain = 1.9024e+00 + 2.2270e+00i = 2.928949.494

SOLUTION 20.21.
(a)
Y2N = [1 0;20.1 0]*1e-3;
Y10k = [1 -1;-1 1]*1e-4;
Yshade = Y2N + Y10k
Yshade =
1.1000e-03 -1.0000e-04
2.0000e-02 1.0000e-04

(b) This is a series connection of two 2-ports. Hence we first convert the answer of part (a) to zparameters.
Y2N = [1 0;20.1 0]*1e-3;
Y10k = [1 -1;-1 1]*1e-4;
Yshade = Y2N + Y10k
Yshade =
1.1000e-03 -1.0000e-04
2.0000e-02 1.0000e-04
Zshade = inv(Yshade)
Zshade =
4.7393e+01 4.7393e+01
-9.4787e+03 5.2133e+02
Z1k = [1 1;1 1]*1e3;
Zdashed = Zshade + Z1k
Zdashed =
1.0474e+03 1.0474e+03
-8.4787e+03 1.5213e+03

(c)
twoportz(Zdashed,1e12,1e3)

ans =
twoport analysis using z-parameters
zin =
1.0474e+03
zout =
5.8588e+03
v1tovs =
5.1157e-01
v2tov1 =
-8.0950e+00
v2tovs =
-4.1412e+00
Conclusion: Vout/Vs = -4.1412.

SOLUTION 20.22.
Z = [3 1;5 2]*1e3;
Y = inv(Z)
Y=
2.0000e-03 -1.0000e-03
-5.0000e-03 3.0000e-03
% Consider the parallel connection of Y with the 1 k resistor
Y1 = Y + [1 -1;-1 1]*1e-3
Y1 =
3.0000e-03 -2.0000e-03
-6.0000e-03 4.0000e-03

% Now consider Y/Z in parallel with 1 k connected between B and C

Y2 = Y + [0 0;0 1]*1e-3
Y2 =
2.0000e-03 -1.0000e-03
-5.0000e-03 4.0000e-03

% This combination is in series with another 1 k resistor

% Hence we need to compute Z2 first
Z2 = inv(Y2)
Z2 =
1.3333e+03 3.3333e+02
1.6667e+03 6.6667e+02
% Now we compute the series combo of Z2 and the 1 k resistor
Z3 = Z2 + [1 1;1 1]*1e3
Z3 =
2.3333e+03 1.3333e+03
2.6667e+03 1.6667e+03


% We now convert Y1 and Z3 to t-parameters and then multiply
% together to obtain the overall t-parameters

T1 = ytot(Y1)
T1 =
6.6667e-01 1.6667e+02
0 5.0000e-01
T3 = ztot(Z3)
T3 =
8.7500e-01 1.2500e+02
3.7500e-04 6.2500e-01
T = T1*T3
T=
6.4583e-01 1.8750e+02
1.8750e-04 3.1250e-01
Y = ttoy(T)
Y=
1.6667e-03 -8.8889e-04
-5.3333e-03 3.4444e-03

SOLUTION 20.23. Refer to figure 20.13 for all problems.

(a) By inspection,
I1 6
4
2 V1


I 2 = 4 4 + 5s 5s V2
I3 2 5s 2 + 5s V3
The 3x3 coefficient matrix is the desired Yind(s).
(b) Writing the nodal equation by inspection yields
0
0.25 0.25 V1
I1 0.5
I 0
1
0.5 0.5 V2
2 =
I3 0.25 0.5 1.25
0.5 V3


0 0.25 0.5 0.5 1.25 V4
Using MATLAB,
W11=[0.5, 0 -0.25;0 1 -0.5;-0.25 -0.5 1.25]
W11 =
5.0000e-01
0 -2.5000e-01
0 1.0000e+00 -5.0000e-01
-2.5000e-01 -5.0000e-01 1.2500e+00
W12 = [-0.25 -0.5 -0.5]'

W12 =
-2.5000e-01
-5.0000e-01
-5.0000e-01
W21 = [-0.25 -0.5 -0.5]
W21 =
-2.5000e-01 -5.0000e-01 -5.0000e-01
W22 = 1.25
W22 =
1.2500e+00
Yind = W11 - W12*inv(W22)*W21
Yind =
4.5000e-01 -1.0000e-01 -3.5000e-01
-1.0000e-01 8.0000e-01 -7.0000e-01
-3.5000e-01 -7.0000e-01 1.0500e+00

(c) Again, writing the nodal equation by inspection yields

0
0 5 V1
I1 5
I 0
6 7 1 V2
2 =
I3 0 4 8 4 V3


0 5 2 1 8 V4
Again, using MATLAB,
W11 = [5 0 0;0 6 -7;0 -4 8]
W11 =
5 0 0
0 6 -7
0 -4 8
W12 = [-5 1 -4]'
W12 =
-5
1
-4
W21 = [-5 -2 -1]
W21 =
-5 -2 -1
W22 = 8;
Yind = W11 - W12*inv(W22)*W21
Yind =
1.8750e+00 -1.2500e+00 -6.2500e-01
6.2500e-01 6.2500e+00 -6.8750e+00
-2.5000e+00 -5.0000e+00 7.5000e+00

(d) This part is similar to part (a) as it does not require the method of matrix partitioning. By
inspection,
IG (CGD + CGS )s CGD s
CGS s VG


CGD s
gm VD
I D = CGD s + gm
I S CGS s gm
0
CGS s + gm VS
The 3x3 coefficient matrix is the desired Yind(s).
SOLUTION 20.24. (a) With regard to the given information, the associated indefinite admittance
matrix is the coefficient matrix in the following nodal equation given reference to figure 20.13:
IG 0
0
0 VG


I D = gm 0 gm VD
I S gm 0 gm VS
We use property 5 to compute the remaining answers:
(a)-(a): YGD is as given.
gm 0
(a)-(b): YSD =
S
gm 0
0
0
(a)-(c): YGS =
S
gm gm
0 gm
(a)-(d): YDG =
S
0 0
0 gm
(a)-(e): YDS =
S
0 gm
gm gm
(a)-(e): YSG =
S
0
0
(b) Transmission from port-1 to port-2 occurs when the 2-1 entry of the 2-port y-matrix is
nonzero. Hence, the following all have the desired transmission: YGD , YSD , YGS .
SOLUTION 20.25. Using the zero-sum properties of the rows and columns, we have by
inspection:
40 2
42 40
2 42

Yind = ?
? 50 = 30 20 50 S
? 22 92 70 22 92

 8 ? ? 8 2 6

SOLUTION 20.26. (a) Yind = 2 9 7 = 2 9 7 S, where the second equality arises

6 ? ? 6 7 13
because a purely
resistive network has a symmetric indefinite admittance matrix.
(b) Construct a common ground 2-port with input terminal C, common terminal A, and output
terminal B:
13 7
YCB =
S
7 9
Therefore, since the output terminal B is shorted, YL = , i.e.,
y y
Yin = y11 12 21 = 13 S
y 22 + YL
Hence
Zin =

13

(c) Construct a common ground 2-port with input terminal A, common terminal B, and output
terminal C:
8 6
YAC =
S
6 13
Hence, the required voltage gain is
V
y 21
y 21 6
GV = 2 =
=
=
V1 y 22 + YL
y 22 13
Therefore, V2 =

6
V.
13

SOLUTION 20.27.
(a) Using the zero-sum property of the rows and columns of an indefinite admittance matrix we
can write down by inspection (in mS)
IB 1
1
0 VB


IE = 100 100.1 0.1 VE
IC 99
99.1 0.1 VC

(b) The y-parameters (also in mS) of the common emitter configuration are easily computed as
0 VBE
IB 1
I = 99 0.1 V
C
CE
Hence in MATLAB
Y = [1 0;99 0.1];
Z = inv(Y)
Z=
1.0000e+00
0
-9.9000e+02 1.0000e+01
where Z is in k. It follows that VCE = 990I B + 10IC where IB and IC are in mA and VCE is
in volts.

SOLUTION 20.28. Expanding the given y-parameter matrix into an indefinite admittance matrix
yields
IG
VG 0.2 + j2.5 0.01 j0.65 0.19 j1.85 VG



0.05 + j0.8 3.15 j0.15 VD
I D = Yind VD = 3.1 j0.65
I S
VS 3.3 j1.85 0.04 j0.15
3.34 + j2 VS
By inspection, with G as the common terminal, S as the input terminal, and D as the output
terminal, we obtain,
0.04 j0.15
3.34 + j2
Ynew =
mS
3.15 j0.15 0.05 + j0.8
SOLUTION 20.29. Here we use the zero-sum properties of the columns and rows to complete the
indefinite admittance matrix:
s
?
? s 2s 1 3s + 1

Yind = s 1
2
? = s 1
2
s 1 S
1
2s 1 2s 1
2s 1
2s

 s 2s 1
In figure (b), the top 2-port has y top =
S and the bottom 2-port has y-parameters
2
s 1
2s 2s 1
S. Since these 2-ports are connected in parallel, the overall 2-port yybot =
2
s 1
parameters are
s 2s 1 2s 2s 1 3s 2
S
y = y top + y bot =
+
=
2 s 1
2 2 4
s 1
SOLUTION 20.30. From problem 28,
IG
VG 0.2 + j2.5 0.01 j0.65 0.19 j1.85 VG



0.05 + j0.8 3.15 j0.15 VD
I D = Yind VD = 3.1 j0.65
I S
VS 3.3 j1.85 0.04 j0.15
3.34 + j2 VS
Hence

0.2 + j2.5 0.19 j1.85

S
yGS =
3.34 + j2
3.3 j1.85

SOLUTION 20.31. (a) Here, by inspection we can compute the indefinite admittance matrix as
the coefficient matrix of the following nodal equations:
I A
V A Y1 + Y2 + 2 Y2


Y2
IB = Yind VB = Y2 2
IC
VC Y1 2Y2 2Y2
(b)

(c)

Y1 2 V A

2 VB
Y1 VC

Y1 + Y2 + 2 Y2
y AB =
Y2
Y2 2
Y1 + Y2 + 2 Y1 2
y AC =
Y1
Y1 2Y2

SOLUTION 20.32. (a) Here, by inspection we can compute the indefinite admittance matrix as
the coefficient matrix of the following nodal equations:

I A
V A Y1 + Y2
Y2
Y1
VA



Y3 gm VB
IB = Yind VB = Y2 + gm Y2 + Y3
IC
VC Y1 gm
Y3
Y1 + Y3 + gm VC
(b)

Y2
Y1 + Y2
y AB =

Y2 + gm Y2 + Y3

(c)

Y1
Y1 + Y2

y AC =

Y1 gm Y1 + Y3 + gm

SOLUTION 20.33. (a) Here the nodal equation matrix is:

0
0
8 VA
I A
VA 8
I W

B = 11 W12 VB = 0 2 + 2s 2s 10 8 VB
IC W 21 W22 VC 0
2s
16 + 2s 16 VC



2
6
16 VD
0
VD 8
where VD is the internal node voltage. Using the method of matrix partitioning,
1
Yind = W11 W12W22
W 21

8
0
0

= 0 2 + 2s 2s 10 8 [8 2 6]
16
0 2s
16
16 + 2s

8
0
0 0.5
4
1
3

= 0 2 + 2s 2s 10 0.5 [8 2 6] = 4
3 + 2s 2s 7 S
0 2s
8 2s 2 10 + 2s
16 + 2s 1
(b) When C is grounded,
4
y AB =
4
(c) When B is grounded

1
1 3 + 2s 1
S and zAB =

3 + 2s
8s + 16 4
4
3
4
y AC =
S
8 10 + 2s

SOLUTION 20.34. (a) Writing the usual node equations we have,

s
0
0.5 VA
I A
VA s + 0.5
I W

s + 0.5
0
0.5 VB
B = 11 W12 VB = s
IC W 21 W22 VC 0
0
0.5s
0.5s VC



0.5 0.5s 0.5s + 1 VD
0
VD 0.5
Using the method of matrix partitioning,
1
Yind = W11 W12W22
W 21

s + 0.5
s
0
1

0.5
= s
s + 0.5 0
1 [1 1 s]
s+2
0
s
0
0.5s

s + 0.5
s
0
1 1 s

0.5

= s
s + 0.5
0
1 1 s

s+2
0
s s s2
0
0.5s
s2 + 2.5s + 0.5
1 2
=
s 2s 0.5
s+2
0.5s

s2 2s 0.5 0.5s

s2 + 2.5s + 0.5 0.5s S

0.5s
s

(b) Here
y AB =

1 s2 + 2.5s + 0.5 s2 2s 0.5

S
s + 2 s2 2s 0.5 s2 + 2.5s + 0.5

SOLUTION 20.35. (a) Rule 1: Consider the two networks NA (3 external nodes) and NB (4
external nodes) given below:

The two indefinite admittance matrices are

G1 + G2

YindN A = G2
G1

G2
G2 + G3
G3

G1 + G2
G1
G

2
G3 and YindN B =
G1

G1 + G3
0

G2
G1
G2 + G3
G3
G3
G1 + G3
0
0

0
0
0

Observe that YindN B can be obtained from YindN A by adding a column of zeros and a row of
zeros to form a 4x4 matrix.
(b) Rule 2: Consider two networks NA and NB and a third network NC which combines
elements of NA and NB as given below:

The corresponding indefinite admittance matrices are:

6 2 4
5 3

YindN A = 2 3 1 , YindN B = 3 5
4 1 5
2 2
Clearly,
YindN C = YindN A
(c) Rule 3: Consider the 3-terminal network

2
11 5 6

2 , and YindN C = 5 8 3 .
6 3 9
4
+ YindN B

with indefinite admittance matrix

5 3 2

YindN A = 3 5 2 S
2 2 4
If we move node 3 inside to form a 2-terminal network and labeled as NB,

then from nodal analysis we have

I1 5 3 2 V1


I 2 = 3 5 2 V2
0 2 2 4 V3
Using the method of matrix partitioning,
5 3 1 2
4
YindN B =
[2 2] =

3 5 4 2
4

4
S
4

This computation is the one given by the formula in the problem. To see that this is correct, we
observe that the internal simplification of NB leads to the following:

SOLUTION 20.36.
Part (a)
Yinda= [1/2 -1/4 0 -1/4 0; -1/4 1/2 0 0 -1/4; ...
0 0 0 0 0; -1/4 0 0 1/2 -1/4; 0 -1/4 0 -1/4 1/2]
Yinda =
5.0000e-01 -2.5000e-01
-2.5000e-01 5.0000e-01
0
0
-2.5000e-01
0
0
-2.5000e-01

0 -2.5000e-01
0
0
0
-2.5000e-01
0
0
0
0 5.0000e-01 -2.5000e-01
0 -2.5000e-01
5.0000e-01

Yindb= [0 0 0 0 0; 0 0 0 0 0; 0 0 3/4 -1/2 -1/4; ...

0 0 -1/2 5/8 -1/8; 0 0 -1/4 -1/8 3/8]
Yindb =
0
0
0
0
0

0
0
0
0
0
0
0
0
0 7.5000e-01 -5.0000e-01 -2.5000e-01
0 -5.0000e-01 6.2500e-01 -1.2500e-01
0 -2.5000e-01 -1.2500e-01 3.7500e-01

Part (b)
Yind = Yinda + Yindb
Yind =
5.0000e-01
-2.5000e-01
0
-2.5000e-01
0

-2.5000e-01
0
5.0000e-01
0
0
7.5000e-01
0
-5.0000e-01
-2.5000e-01 -2.5000e-01

-2.5000e-01
0
-5.0000e-01
1.1250e+00
-3.7500e-01

0
-2.5000e-01
-2.5000e-01
-3.7500e-01
8.7500e-01

% To suppress nodes 4 and 5 we use the partitioned matrix formula as follows:

W11= [Yind(1:3, 1:3)]
W11 =

5.0000e-01 -2.5000e-01
-2.5000e-01 5.0000e-01
0
0 7.5000e-01

0
0

W12=[Yind(1:3, 4:5)]
W12 =
-2.5000e-01
0
0 -2.5000e-01
-5.0000e-01 -2.5000e-01
W21= [Yind(4:5, 1:3)]
W21 =
-2.5000e-01
0 -5.0000e-01
0 -2.5000e-01 -2.5000e-01
W22= [Yind(4:5, 4:5)]
W22 =
1.1250e+00 -3.7500e-01
-3.7500e-01 8.7500e-01
Yind123 = W11 - W12*inv(W22)*W21
Yind123 =
4.3519e-01 -2.7778e-01 -1.5741e-01
-2.7778e-01 4.1667e-01 -1.3889e-01
-1.5741e-01 -1.3889e-01 2.9630e-01
(d) For the required Y-matrix we delete row and column 3 to obtain
Ysc = Yind123(1:2,1:2)
Ysc =
4.3519e-01 -2.7778e-01
-2.7778e-01 4.1667e-01
Zoc = inv(Ysc)
Zoc =
4.0000e+00 2.6667e+00
2.6667e+00 4.1778e+00

SOLUTION 20.37. Since complex roots must occur in conjugate pairs, we will only check j0.

0 = p( j 0 ) = ja3 30 a2 20 + ja1 0 + a0 = a0 a2 20 + j a1 0 a3 03
Both real and imaginary parts must be zero, i.e.,

a0 a2 02 = 0 and a1 0 a3 30 = 0 a1 a3 02 = 0

From the first equation, a2 20 = a0 . The second equation above must be true for arbitrary 0
which implies that a1 = a3 02 . Equivalently a1a2 = a3a2 02 = a3a0 . Conclusion: this condition
leads to imaginary complex roots.
(b) Given the above condition, what are the resulting imaginary roots of the polynomial? Since
the polynomial is cubic, we can assume a3 0. In this case,
a
0 = p(s) = s3 + 2 s2 +
a3

a1
a
s + 0 = s s2 +
a3
a3

a1
+
a3

a2 2 a1a2
s + 2 = s s2 +
a3

a3

a1 a2 2 a1
+
s +
a3 a3
a3

a
a
= s2 + 1 s + 2

a3
a3
Therefore, the roots are: s = j

a1
a
, 2.
a3
a3

SOLUTION 20.38.
(a) The four 2-port equations arising from the interconnection are:
V1 = V1a V1b

I1 = I1a = I1b

V2 = V2a = V2b

I2 = I2a + I2b

Thus
V1 = V1a V1b = (h11a I1a + h12aV2a ) (h11b I1b + h12bV2b ) = ( h11a + h11b ) I1 + ( h12a h12b )V2
and
I2 = I2a + I2b = ( h21a I1a + h22aV2a ) + (h21b I1b + h22bV2b ) = ( h21a h21b )I1 + (h22a + h22b )V2
This proves that the series-parallel connection has the required h-parameters.
(b) The four 2-port equations arising from the interconnection are:
V2 = V2a V2b

I2 = I2a = I 2b

V1 = V1a = V1b

I1 = I1a + I1b

Thus,
I1 = I1a + I1b = (g11aV1a + g12a I2a ) + (g11bV1b + g12b I2b ) = ( g11a + g11b )V1 + (g12a g12b ) I2
and
V2 = V2a V2b = ( g21aV1a + g22a I 2a ) (g21bV1b + g22b I2b ) = ( g21a g21b )V1 + ( g22a + g22b ) I2
This proves that the parallel-series connection has the required g-parameters.

SOLUTION 20.39.
(a) Refer here to Na in figure P20.39b. With reference to figure 19.28b, h11 = 11 k, h12 = 0,
5

-5

h21 = 95.9, and h22 = 1/10 = 10 S. Similarly, by inspection with reference to equation 19.33,
-5

h11 = 90||10 = 9 k, h12 = 0.1 (reverse voltage division), h21 = 0.1, and h22 = 10 S.
(b) By problem 20.38, part (a),
h11 = h11a + h11b = 20 k
h21 = h21a h21b = 96

h12 = h12a h12b = 0.1

-5

h22 = h22a + h22b = 210 S

V
h h
I
h h
(c) Recall from chapter 19 that Zin = 1 = h11 12 21 and Yout = 2 = h22 12 21 in
I1
h22 + YL
V2
h11 + Zs
which case Zout is the reciprocal. Using our MATLAB script, we have
h = [20e3 -0.1; 96 0.02e-3 ];
zL= 1e8;
zs= 5e3;
twoporth(h,zL,zs)
ans =
twoport analysis using h-parameters
zin =
4.9976e+05
zout =
2.4752e+03
v1tovs =
9.9009e-01
v2tov1 =
-9.5998e+00
v2tovs =
-9.5047e+00
REMARK: We have used the following m-file code for "twoporth":
% two-port analysis in terms of h-parameters
function [zin, zout] =twoport(h, zL, zs)
['twoport analysis using h-parameters']
h11= h(1,1); h12=h(1,2); h21=h(2,1); h22=h(2,2);
zin = h11 - h12*h21/(h22+ 1/zL)
yout= h22 - h12*h21/(h11+zs);

zout= 1/yout
v1tovs= zin/(zin+zs)
v2tov1= -h21/(zin*(h22+1/zL))
v2tovs= v1tovs*v2tov1

SOLUTION 20.40.
(a) The y-parameters for Na are:

1
7 4
1 7 4
S
y AB =
=

33 4 7
4 7

Hence the indefinite admittance matrix for Na is:

7 4 3
1

YindN a =
4 7 3 S

33
3 3 6
Let us consider the associate 2-port with port A grounded and B as the new port 1 input. The
problem is then solved by computing the input impedance with port 2 open circuited. Hence, the
new y-parameters are:
y BC =

1 7 3
S
33 3 6

y y
1
9 5.5 1
Thus Yin = y11 12 21 =
7

=
= S. Hence, Zin = 6 is the unique reading.
y 22
33
6 33 6
(b) The answer is not unique as demonstrated in part (c).
(c) The following two networks have the given Z-parameters, but the meter reading for N1 is 4
but for N2 it is 2 .

SOLUTIONS CHAPTER 21 PROBLEMS

SOLUTION TO 21.1.
(a) Low pass
(b) High pass

SOLUTION TO 21.2.

SOLUTION TO 21.3.
(a)
n = 0.65378;
d = [1 0.80381643 0.82306043];
w = 0:0.005:2;
h = freqs(n,d,w);
plot(w, 20*log10(abs(h)))
grid
xlabel('Frequency rads/s')
ylabel('dB Gain')

dB Gain

-5

TextEnd

-10

-15

(b) poles = roots(d)

poles =
-4.0191e-01 + 8.1335e-01i
-4.0191e-01 - 8.1335e-01i
(c)
% Poles of new transfer function
wp = 2*pi*750'
wp =
4.7124e+03
wp = 2*pi*750;
polesnew = poles*wp
polesnew =
-1.8939e+03 + 3.8328e+03i
-1.8939e+03 - 3.8328e+03i
% All zeros remain at infinity.

0.5

1
Frequency rads/s

1.5

Further
H(s) = HNLP (s

7
( p)2
2.2207 10
)
=
=
p
s2 + 0.80381643 p s + 0.82306043( p )2 s2 + 3.7879 103 s + 1.8277 10 7

1
SOLUTION TO 21.4. (a) The 2nd order normalized LP transfer function is HNLP (s) = 2
. This
s + 2s + 1
must be frequency scaled by Kf = 1000. Hence,
(K f )2

9.8696 10
H(s) = HNLP (s K f ) = 2
= 2
2
s + K f 2s + (K f )
s + 4.4429 103 s + 9.8696 106
(b) Using MATLAB,
n = (1000*pi)^2;
d = [1 sqrt(2)*pi*1e3 (1000*pi)^2];
w = 0:1:2*pi*1500;
h = freqs(n,d,w);
plot(w/(2*pi),abs(h))
grid
xlabel('Frequency in Hz')
ylabel('Magnitude')
plot(w/(2*pi),20*log10(abs(h)))
grid
xlabel('Frequency in Hz')
ylabel('Magnitude in dB')

1
0.9
0.8

Magnitude

0.7
0.6
0.5
TextEnd
0.4
0.3
0.2
0.1

500

1000

1500

Frequency in Hz

0
-2
-4

Magnitude in dB

-6
-8
-10
-12
TextEnd

-14
-16
-18
-20

500

1000
Frequency in Hz

(c)
n
n = 9.8696e+06
d
d = 1.0000e+00 4.4429e+03 9.8696e+06
w = j*2000*pi;
mag = abs(n/(w^2 + d(2)*w + d(3)))
mag = 2.4254e-01

1500

SOLUTION TO 21.5. (a) max is that value of e that places the magnitude response curve through Amax
at = p. Therefore
2n

p
2
Amax = 10log 10 H ( j p) = 10log 10 1+ max = 10log 10 1+ 2max

Therefore 2max = 10 0.1Amax 1 which upon a square root yields the final answer.
(b) Similarly, min puts the magnitude response curve through the Amin spec. Hence
2n

2
s
Amin = 10log10 H ( j s ) = 10log 10 1+ min

Therefore
2min =

100.1A min 1
s

p

2n

which is equivalent to the required formula.

SOLUTION TO 21.6. The relationship of and c is given by the formula:

c min

p
1n

(max )

= 2n

10 0.1Amax 1

p
1
() n

. Further, max

and min in putting the

c max . Hence, from the solution to problem

in putting the magnitude response curve through the Amax spec produces
magnitude response curve through the Amin spec produces
5,

c max

(min )1 n

c min ,

= 2n

10 0.1Amin 1

SOLUTION TO 21.7. (a) From above material, the second order Butterworth NLP transfer function is
1
HNLP2 (s) = 2
s + 2s + 1
and from tables, the third order is

1
HNLP3 (s) = 3
2
s + 2s + 2s + 1
(b)
n1 = 1; d1 = [1 sqrt(2) 1];
n2 = 1; d2 = [1 2 2 1];
w = 0:.01:5;
h1 = freqs(n1,d1,w);
h2 = freqs(n2,d2,w);
plot(w,abs(h1))
grid
xlabel('Normalized Frequency')
ylabel('Magnitude')
hold
plot(w,abs(h2),'r')
hold off

1
0.9
0.8

Magnitude

0.7
0.6
0.5
0.4

TextEnd

0.3
0.2
0.1
0

2
3
Normalized Frequency

Notice how the 3rd order filter has a sharper transition to zero.
(c) % The simplest way to obtain the step response is as follows:

syms s t
StepResp1 = ilaplace(1/(s^3 + sqrt(2)*s^2 + s))
StepResp1 =
1-exp(-1/2*2^(1/2)*t)*cos(1/2*2^(1/2)*t)-exp(-1/2*2^(1/2)*t)*sin(1/2*2^(1/2)*t)
StepResp2 = ilaplace(1/(s^4 + 2*s^3 + 2*s^2 + s))

StepResp2 =
1-exp(-t)-2/3*exp(-1/2*t)*3^(1/2)*sin(1/2*3^(1/2)*t)

Thus the step response of the second order Butterworth normalized LP filter is:
v(t) = u(t) e 0.70711t [cos(0.70711t) sin(0.70711t)]u(t)
and that of the third order Butterworth normalized LP filter is:

v(t) = 1 e

) u(t) 1.1547e0.5t sin(0.86603t)u(t)

SOLUTION TO 21.8.
fp = 100; fs = 1200;Amax = 0.3; Amin = 35;
n = buttord(fp,fs,Amax,Amin,'s')
emax = sqrt(10^(0.1*Amax) - 1)
emin = sqrt(10^(0.1*Amin) - 1)/(fs/fp)^n
fcmin = fp/((10^(0.1*Amax)-1)^(1/(2*n)))
fcmax = fs/((10^(0.1*Amin)-1)^(1/(2*n)))
wcmin = 2*pi*fcmin
wcmax = 2*pi*fcmax
wc = wcmin;
fc = fcmin;
[z,p,k] = buttap(n)
% Numerators are each 1. Denominators are the polynomials
d1 = poly(p(1:2))
d2 = poly(p(3))
zplane(p)
grid
pause
znew = z*wc
pnew = p*wc
knew = k*wc^n
f = 0:fc/50:1.2*fs;
h = freqs(knew*poly(znew),poly(pnew),2*pi*f);
plot(f,abs(h))
grid
xlabel('Frequency in Hz')
ylabel('Gain magnitude')
pause
plot(f,20*log10(abs(h)))
xlabel('Frequency in Hz')
ylabel('Gain in dB')

grid
n= 3
emax = 2.6743e-01
emin = 3.2538e-02
fcmin = 1.5521e+02
fcmax = 3.1324e+02
wcmin = 9.7524e+02
wcmax = 1.9681e+03
z = []
p=
-5.0000e-01 + 8.6603e-01i
-5.0000e-01 - 8.6603e-01i
-1.0000e+00
k= 1
% Numerators are each 1. Denominators are the polynomials
d1 = 1.0000e+00 1.0000e+00 1.0000e+00
d2 = 1 1

1
0.8
0.6

Imaginary part

0.4
0.2
0
-0.2
-0.4
-0.6
-0.8
-1
-1

znew = []
pnew =
-4.8762e+02 + 8.4458e+02i
-4.8762e+02 - 8.4458e+02i
-9.7524e+02
knew = 9.2753e+08

-0.5

0
Real part

0.5

1
0.9
0.8

Gain magnitude

0.7
0.6
0.5
0.4
0.3
0.2
0.1
0

500

1000

1500

1000

1500

Frequency in Hz

-10

Gain in dB

-20

-30

-40

-50

-60

500
Frequency in Hz

SOLUTION TO 21.9.

fp = 100; fs = 1200;Amax = 0.3; Amin = 35;

n = buttord(fp,fs,Amax,Amin,'s');
emax = sqrt(10^(0.1*Amax) - 1);
emin = sqrt(10^(0.1*Amin) - 1)/(fs/fp)^n;
fcmin = fp/((10^(0.1*Amax)-1)^(1/(2*n)));
fcmax = fs/((10^(0.1*Amin)-1)^(1/(2*n)));
wcmin = 2*pi*fcmin;
wcmax = 2*pi*fcmax;
[z,p,k] = buttap(n);
wc = wcmax;
fc = fcmax;
znew = z*wc
pnew = p*wc
knew = k*wc^n
f = 0:fc/50:1.2*fs;
h = freqs(knew*poly(znew),poly(pnew),2*pi*f);
plot(f,abs(h))
grid
xlabel('Frequency in Hz')
ylabel('Gain magnitude')
pause
plot(f,20*log10(abs(h)))
xlabel('Frequency in Hz')
ylabel('Gain in dB')
grid
znew = []
pnew =
-9.8406e+02 + 1.7044e+03i
-9.8406e+02 - 1.7044e+03i
-1.9681e+03
knew = 7.6235e+09

1
0.9
0.8

Gain magnitude

0.7
0.6
0.5
0.4
0.3
0.2
0.1
0

500

1000

1500

1000

1500

Frequency in Hz

5
0
-5

Gain in dB

-10
-15
-20
-25
-30
-35
-40

500
Frequency in Hz

SOLUTION TO 21.10.

fp = 75; fs = 450;Amax = 1; Amin = 45;

n = buttord(fp,fs,Amax,Amin,'s')
emax = sqrt(10^(0.1*Amax) - 1)
emin = sqrt(10^(0.1*Amin) - 1)/(fs/fp)^n
fcmin = fp/((10^(0.1*Amax)-1)^(1/(2*n)))
fcmax = fs/((10^(0.1*Amin)-1)^(1/(2*n)))
wcmin = 2*pi*fcmin
wcmax = 2*pi*fcmax
[z,p,k] = buttap(n)
d1 = poly(p(1:2))
d2 = poly(p(3:4))
zplane(p)
grid
pause
wc = wcmin;
fc = fcmin;
znew = z*wc
pnew = p*wc
knew = k*wc^n
W = 0:0.01:fs/fp;
h = freqs(k*poly(z),poly(p),W);
plot(W*wc/(2*pi),abs(h))
grid
xlabel('Frequency in Hz')
ylabel('Gain magnitude')
pause
plot(W*wc/(2*pi),20*log10(abs(h)))
xlabel('Frequency in Hz')
ylabel('Gain in dB')
grid
n= 4
emax = 5.0885e-01
emin = 1.3721e-01
fcmin = 8.8800e+01
fcmax = 1.2323e+02
wcmin = 5.5795e+02
wcmax = 7.7427e+02
z = []
p=
-3.8268e-01 + 9.2388e-01i
-3.8268e-01 - 9.2388e-01i
-9.2388e-01 + 3.8268e-01i
-9.2388e-01 - 3.8268e-01i
k= 1

d1 = 1.0000e+00 7.6537e-01 1.0000e+00

d2 = 1.0000e+00 1.8478e+00 1.0000e+00
znew = []
pnew =
-2.1352e+02 + 5.1548e+02i
-2.1352e+02 - 5.1548e+02i
-5.1548e+02 + 2.1352e+02i
-5.1548e+02 - 2.1352e+02i
knew = 9.6912e+10

1
0.8
0.6
Imaginary part

0.4
0.2
0
-0.2
-0.4
-0.6
-0.8
-1
-1

-0.5

0
Real part

0.5

1
0.9
0.8

Gain magnitude

0.7
0.6
0.5
0.4
0.3
0.2
0.1
0

100

200

300
400
Frequency in Hz

500

600

100

200

300
400
Frequency in Hz

500

600

0
-10

Gain in dB

-20
-30
-40
-50
-60
-70

SOLUTION TO 21.11.
fp = 75; fs = 450;Amax = 1; Amin = 45;

n = buttord(fp,fs,Amax,Amin,'s');
% The order mfile may not be available in the student edition.
emax = sqrt(10^(0.1*Amax) - 1);
emin = sqrt(10^(0.1*Amin) - 1)/(fs/fp)^n;
fcmin = fp/((10^(0.1*Amax)-1)^(1/(2*n)));
fcmax = fs/((10^(0.1*Amin)-1)^(1/(2*n)));
wcmin = 2*pi*fcmin;
wcmax = 2*pi*fcmax;
[z,p,k] = buttap(n);
wc = wcmax;
fc = fcmax
znew = z*wc
pnew = p*wc
knew = k*wc^n
W = 0:0.01:fs/fp;
h = freqs(k*poly(z),poly(p),W);
plot(W*wc/(2*pi),abs(h))
grid
xlabel('Frequency in Hz')
ylabel('Gain magnitude')
pause
plot(W*wc/(2*pi),20*log10(abs(h)))
xlabel('Frequency in Hz')
ylabel('Gain in dB')
grid
fc = 1.2323e+02
znew = []
pnew =
-2.9630e+02 + 7.1533e+02i
-2.9630e+02 - 7.1533e+02i
-7.1533e+02 + 2.9630e+02i
-7.1533e+02 - 2.9630e+02i
knew = 3.5940e+11

1
0.9
0.8

Gain magnitude

0.7
0.6
0.5
0.4
0.3
0.2
0.1
0

100

200

300
400
500
Frequency in Hz

600

700

800

100

200

300
400
500
Frequency in Hz

600

700

800

0
-10

Gain in dB

-20
-30
-40
-50
-60
-70

SOLUTION TO 21.12. Here we require that

C + C2 2 C1 + C2 + L
2
s3 + 2s2 + 2s + 1= s3 + 1
s +
s+
LC1C2
LC1C2
C1C2
Thus

C + C2 L LC1 + LC2
LC1C2 = 2 C1 + C2 + L = 4 and 2 = 1
=
L
2
C1C2

in which case, LC1 + LC2 = 4 . Therefore, L(C1 + C2 + L) = 4L = 4 + L2 . Equivalently

L2 4L + 4 = (L 2)(L 2) = 0
Hence L = 2 H is the only solution. Thus C1C2 = 1 C1 =

1
1

+ C2 + 2 = 4 or equivalently
C2
C2

C22 2C2 + 1 = (C2 1)(C2 1) = 0 which implies that C1 = C2 = 1 F is the only solution, as was to be
shown.

SOLUTION TO 21.13. (a) By voltage division

(b) With R = 1 ,

1
1
LC
H(s) = Cs + G
=
1
1
1
2
Ls +
s +
s+
RC
LC
Cs + G
1
1
LC
H(s) =
= 2
1
1
s + 2s + 1
s2 +
s+
RC
LC

requires that C = 1/ 2 F and L =

(c) fc = 1000;
wc = 2*pi*fc
wc = 6.2832e+03
Kf = wc;
Km = 1000;
C = 1/sqrt(2);
L = sqrt(2);
Cnew = C/(Kf*Km)
Cnew = 1.1254e-07
Lnew = L*Km/Kf
Lnew = 2.2508e-01
(d)

2 H.

Km = C/(wc*1e-6)
Km = 1.1254e+02
Kf = wc;
Rnew = Km
Rnew = 1.1254e+02
Lnew = L*Km/wc
Lnew =
2.5330e-02

SOLUTION TO 21.14. (a) By voltage division

(b) With R = 1 ,

1
1
Cs
LC
H(s) =
1 = 2 Rs
1
Ls + Rs +
s +
s+
L
LC
Cs
1
1
LC
H(s) =
=
2
R
1
s + 2s + 1
s2 + s s +
L
LC

requires that L = 1/ 2 F and C =

(c)
L = 1/sqrt(2); C = 1/L;
Km = 10;
Kf = 2*pi*500;
Rnew = 10;
Cnew = C/(Km*Kf)
Cnew = 4.5016e-05
Lnew = Km*L/Kf
Lnew = 2.2508e-03
(d)
Km = C/(1e-6*Kf)
Km = 4.5016e+02
Lnew = L*Km/Kf
Lnew = 1.0132e-01
Cnew = C/(Km*Kf)
Cnew = 1.0000e-06

SOLUTION TO 21.15.

2 H.

(a) Since 1 = 2/(LC), L = 2/C. Since (1/L + 1/C) = (C/2 + 1/C) = sqrt(2), we have that C is a root of
2
the quadratic 0.5C sqrt(2)C + 1 = 0. Hence
v = [0.5 -sqrt(2) 1];
r = roots(v)
r=
1.4142e+00
1.4142e+00
C = r(1)
C = 1.4142e+00
L = 2/C
L = 1.4142e+00
(b)
Km = 1e3;
Kf = 2*pi*3500;
Cnew = C/(Km*Kf)
Cnew = 6.4308e-08
Lnew = L*Km/Kf
Lnew = 6.4308e-02
(c)
Km = C/(Kf*10e-9)
Km = 6.4308e+03
Cnew = C/(Km*Kf)
Cnew = 1.0000e-08
Lnew = L*Km/Kf
Lnew = 4.1356e-01
Rs = Km
Rs = 6.4308e+03
RL = Rs
RL = 6.4308e+03

SOLUTION TO 21.16.
(a) Let G = RL. Then by voltage division
1
1
Vout
Cs + G
LC
H(s) =
=
=
1
Vin

1
R
1 + Rs RL
2
s
Ls + Rs +
Cs + G s + CR + L s +
LC
L

(b) Since 1 = 1.25/(LC), L = 1.25/C. Since (Rs/L + 1/RLC) = (2C/1.25 + 1/8C) = sqrt(2), we have that
2

C is a root of the quadratic (16/1.25)C 8sqrt(2)C + 1 = 0. Hence

C = roots([16/1.25 -8*sqrt(2) 1])
C=
7.8427e-01
9.9615e-02
L = 1.25 ./C
L=
1.5938e+00
1.2548e+01
(c)
Km = 1e3;
Kf = 2*pi*5e3
Kf = 3.1416e+04
Cnew = C/(Km*Kf)
Cnew =
2.4964e-08
3.1708e-09
Lnew = L*Km/Kf
Lnew =
5.0734e-02
3.9942e-01

SOLUTION TO 21.17.
(a) Define G = 1/Rs, execute two source transformations, and apply voltage division to obtain
1
V
1
G
Rs LC
H(s) = out =

=
1
Vin
+ Ls + 1 Cs + G s2 + 1 + 1 s + 1 + 1 Rs
R C L
Cs + G
LC
s
2

(b) Since L = 1.5/C, the values of C are the roots of the quadratic, (2/1.5)C 2sqrt(2)C + 1 = 0.
Hence
C = roots([(2/1.5) -2*sqrt(2) 1])
C=
1.6730e+00
4.4829e-01
L = 1.5 ./C
L=
8.9658e-01
3.3461e+00
Km = 2e3;
Kf = 2*pi*5e3;
Cnew = C/(Km*Kf)

Cnew =
2.6627e-08
7.1347e-09
Lnew = Km*L/Kf
Lnew =
5.7078e-02
2.1302e-01
Rsnew = 2*Km
Rsnew =
4000

SOLUTION TO 21.18.
(a)
H( j) =

K

j
+1
p

K
2

+1
p

The 3 dB down frequency, c, occurs when

2

c
+1
p

1
2

Equivalently
c = p
(b) h(t) = K 2pte

p t

2 1 = 0.64359 p = 6.4359 10 4 rad/sec.

u(t) since
H(s) =

s
+
1

K 2p

( s + p )

Further,
K 2p
K p
1
K
K
H(s) =
=

2
s s s+ 2
s s + p
s + p
p

Hence, the step response is

K 1 e

p t

pte

p t

u(t)

SOLUTION TO 21.19.
(a) Using voltage division,
1
V
V
H(s) = C out =
Vin VC

Cs +
1 + L1s +

1
L2 s + 1
1
Cs +

1
L2s + 1

1
L2 s + 1

1
L1L2C
=
1
1
2
L + L2 + C
s3 + + s2 + 1
s+
L1L2C
L1 L2
L1L2C
(b) Matching coefficients in
1
L1L2C
1
= 3
2

1
1
2
L + L2 + C
s + 2s + 2s + 1
s3 + + s2 + 1
s+
L1L2C
L1 L2
L1L2C

yields C =

2
L + L2 + C
and 1
=
L1L2
L1L2C

L1 + L2 +
2

2
L1L2

= 2; equivalently, L1 + L2 +

2
= 4. Further,
L1L2

1
1 L1 + L2
2
L + = L L = 2 implies that L1 + L2 = 2L1L2 implies 2L1L2 + L = 4 . This requires that
L2
L1 2
1
1 2
L1L2 = 1 and from earlier equations that L1 + L2 = 2L1L2 = 2 which forces L1 = L2 = 1 H and C = 2 F.
The idea is to match the denominator coefficients and thus the dc gain is 0.5 instead of the desired 1. A
transformer or some amplifier device is needed to increase the gain to 1.

(c) Using MATLAB:

Km = 1000;
Kf = 2*pi*20e3;
Lnew = Km/Kf
Lnew =
7.9577e-03
Cnew = 2/(Km*Kf)
Cnew =
1.5915e-08


Hence, L1new = L2new = 7.96 mH and C = 15.9 nF.
(d) SPICE simulation

R
1K

L
7.96m

L0
7.96m

V
0

C
15.9n

IVm
R0
1K

Prob 21.19-Small Signal AC-1

+10.000k

+500.000m

+400.000m

+300.000m

+200.000m

+100.000m

+0.000e+000
MAG(V(IVM))

+20.000k

+30.000k

Frequency
+40.000k

(Hz)

SOLUTION TO 21.20. (a) From figure P21.19a

Vin V1
= I1,
Rs
in which case

V2 = Vout

V1 = z11I1 + z12 0 = Vin Rs I1

Also
V2 = z21I1 + z22 0 = Vout
This implies that Vin = (z11 + Rs )I1 and Vout = z21I1 . Finally we conclude that
Vout
z21
=
Vin
z11 + Rs
(b) Now from figure P21.20b, we have V1 = Vin and V2 = Vout = RL I2 . This implies that
V
I2 = out = y21V1 + y 22V2
RL
Thus

Vout
y 21
y21
=
=
1
Vin
y 22 + GL
y22 +
RL

Consider here that

1
1
Vout
z
1
2s
s
= 21 = 2
=
=
1
Vin
z11 + 1 s + 2s + 1 1
s
s+ + 2
+

s
2s
2
Hence z21 =

1
s
1
and z11 =
+
. This leads to the circuit
2
2s
2s

with L =
( c)

1
H and C = 2 F.
2

Similarly,
1
y 21
2s
=
y 22 + 1 s
1

+1
+
2
2s

implies y 21 =

1
and y 22 =
2s

s
1
.
+
2s
2s

This yields the same circuit as above with

L = 2 H and C =
(d)

Here K m = 1000 and K f = 5000rad / s

(d-i)

For (b),

1
F.
2

1
10 3
1
L=

=
= 0.1414 H
3
5 2
2 5 10
and
C= 2

1
5106

= 0.2828 F

(d-ii) For (c)

L = 0.2828 H and C = 0.1414 F.

SOLUTION TO 21.21.
(a) From earlier developments
1
R1

Y
H(s) = in =
1
Yout
Cs +
R2
(b) Let C = 1 F, and R1 = R2 = 1 .
(c) Kf = 2*pi*3500

Kf = 2.1991e+04
Km = 1/(Kf*1e-9)
Km = 4.5473e+04
In the final design, R1 = R2 = 45.5 k.

SOLUTION TO 21.22.
(a) From problem 21 and voltage division,
1
R1

1
C2s

1
1
C1R1
R3C2
H(s) =
1
1 =
1
1
Cs +
R3 +
s+
s+
R2
C2s
C1 R2
R3C2
(b) By inspection, let C1 = 0.1 F, R1 = 1 , R2 = 10 , R3 = 10 , and C2 = 0.1 F, in which case
H(s) =

10
1
10

=
s + 1 s + 1 (s + 1)2

(c)
Kf = 1e5;
Km = 0.1/(Kf*1e-9)
Km =
1000
Hence, in the final design
C1 = 1 nF, R1 = 1 k, R2 = 10 k, R3 = 10 k, and C2 = 1 nF.
(d) Cascade the circuit of figure P21.22 with another op amp section. For the first part of the design,
again set p = 1 and use the same values as in part (c). The extra op amp section has the same values as
the first section. As such, final values are the same as in part (c).

1
SOLUTION TO 21.23. The 2nd order NLP Butterworth transfer function is: HNLP (s) = 2
. The
s + 2s + 1
design parameters and steps are detailed in the excel spread sheet below. An additional design called
design C is also listed. For input attenuation, the resistor R1 is replaced by the voltage divider R3-R4
combination.

w0^2
1

w0/Q
1.41

Num
1.00

w0
1.0000

Q
0.7071

KNLP
1.00

KMA
22507.86

Kf=wp
6283.20

KmR
10000.00

KMB
22507.86

KMC
15915.46

Design A
Design B
Design C
Saraga

RA

1
1
RA

RB
0
1
1-1/Q
RA/3

K
1
2
3-1/Q
3-Apr

C1
2Q
1
1
rt(3)Q

C2
1/(2Q)
1/Q
1
1

R1
1
1
1
1/Q

R2
1
Q
1
1/rt(3)

alpha
KNLP/K
KNLP/K
KNLP/K
KNLP/K

R3
1/alpha
1/alpha
1/alpha
1/alpha

R4
1/(1-alpha)
1/(1-alpha)
1/(1-alpha)
1/(1-alpha)

Design A
Design B
Design C
Saraga

RA

1.00
1.00
3.00

RB
0.0000
1.0000
-0.4142
1.0000

K
1.0000
2.0000
1.5858
1.3333

C1
1.4142
1.0000
1.0000
1.2247

C2
0.7071
1.4142
1.0000
1.0000

R1
1.0000
1.0000
1.0000
1.4142

R2
1.0000
0.7071
1.0000
0.5774

alpha
1.0000
0.5000
0.6306
0.7500

R3
1.0000
2.0000
1.5858
1.3333

R4
#DIV/0!
2.0000
2.7071
5.6569

w0 scale
Design A
Design B
Design C
Saraga

1.00
1.00
3.00

0.0000
1.0000
0.1140
1.0000

1.0000
2.0000
2.1140
1.3333

1.4142
1.0000
1.0000
1.2247

0.7071
1.4142
1.0000
1.0000

1.0000
1.0000
1.0000
1.4142

1.0000
0.7071
1.0000
0.5774

1.0000
0.5000
0.6306
0.7500

1.0000
2.0000
1.5858
1.3333

#DIV/0!
2.0000
2.7071
5.6569

wp scale
Design A
Design B
Design C
Saraga

1.00
1.00
3.00

0.000E+00
1.000E+00
1.140E-01
1.000E+00

1.000E+00
2.000E+00
2.114E+00
1.333E+00

2.251E-04
1.592E-04
1.592E-04
1.949E-04

1.125E-04
2.251E-04
1.592E-04
1.592E-04

1.000E+00
1.000E+00
1.000E+00
1.414E+00

1.000E+00 1.000E+00 1.000E+00

#DIV/0!
7.071E-01 5.000E-01 2.000E+00 2.000E+00
1.000E+00 6.306E-01 1.586E+00 2.707E+00
5.774E-01 7.500E-01 1.333E+00 5.657E+00

Km scale
Design A
Design B
Design C
Saraga

10000
10000
30000

0.000E+00
1.000E+04
1.140E+03
1.000E+04

1.000E+00
2.000E+00
2.114E+00
1.333E+00

1.000E-08
7.0711E-09
1.000E-08
1.000E-08

5.000E-09
1.000E-08
1.000E-08
8.165E-09

2.251E+04
2.251E+04
1.592E+04
2.757E+04

2.251E+04 1.000E+00 2.251E+04

#DIV/0!
1.592E+04 5.000E-01 4.502E+04 4.502E+04
1.592E+04 6.306E-01 2.524E+04 4.308E+04
1.125E+04 7.500E-01 2.599E+04 1.103E+05

KMS
19492.37

SOLUTION TO 21.24 AND 21.25. In problem 21.8, the transfer function information was computed in
MATLAB as:
% Numerators are each 1. Denominators are the polynomials
d1 = 1.0000e+00 1.0000e+00 1.0000e+00
d2 = 1 1
Further we know from MATLAB that
fcmin = 1.5521e+02
The Saraga design and Design A for d1, the second order section of each filter, are given by the excel
spread sheet below, as well as two alternate designs labeled B and C.

w0^2
1

w0/Q
1.00

Num
1.00

w0
1.0000

Q
1.0000

KNLP
1.000

KMA
41016.58

Kf=wp
975.22

KmR
10000.00

KMB
20508.29

KMC
20508.29

Design A
Design B
Design C
Saraga

RA

1
1
RA

RB
0
1
1-1/Q
RA/3

K
1
2
3-1/Q
3-Apr

C1
2Q
1
1
rt(3)Q

C2
1/(2Q)
1/Q
1
1

R1
1
1
1
1/Q

R2
1
Q
1
1/rt(3)

alpha
KNLP/K
KNLP/K
KNLP/K
KNLP/K

R3
1/alpha
1/alpha
1/alpha
1/alpha

R4
1/(1-alpha)
1/(1-alpha)
1/(1-alpha)
1/(1-alpha)

Design A
Design B
Design C
Saraga

RA

1.00
1.00
3.00

RB
0.0000
1.0000
0.0000
1.0000

K
1.0000
2.0000
2.0000
1.3333

C1
2.0000
1.0000
1.0000
1.7321

C2
0.5000
1.0000
1.0000
1.0000

R1
1.0000
1.0000
1.0000
1.0000

R2
1.0000
1.0000
1.0000
0.5774

alpha
1.0000
0.5000
0.5000
0.7500

R3
1.0000
2.0000
2.0000
1.3333

R4
#DIV/0!
2.0000
2.0000
4.0000

w0 scale
Design A
Design B
Design C
Saraga

1.00
1.00
3.00

0.0000
1.0000
0.1140
1.0000

1.0000
2.0000
2.1140
1.3333

2.0000
1.0000
1.0000
1.7321

0.5000
1.0000
1.0000
1.0000

1.0000
1.0000
1.0000
1.0000

1.0000
1.0000
1.0000
0.5774

1.0000
0.5000
0.5000
0.7500

1.0000
2.0000
2.0000
1.3333

#DIV/0!
2.0000
2.0000
4.0000

wp scale
Design A
Design B
Design C
Saraga

1.00
1.00
3.00

0.000E+00
1.000E+00
1.140E-01
1.000E+00

1.000E+00
2.000E+00
2.114E+00
1.333E+00

2.051E-03
1.025E-03
1.025E-03
1.776E-03

5.127E-04
1.025E-03
1.025E-03
1.025E-03

1.000E+00
1.000E+00
1.000E+00
1.000E+00

1.000E+00 1.000E+00 1.000E+00

#DIV/0!
1.000E+00 5.000E-01 2.000E+00 2.000E+00
1.000E+00 5.000E-01 2.000E+00 2.000E+00
5.774E-01 7.500E-01 1.333E+00 4.000E+00

Km scale
Design A
Design B
Design C
Saraga

10000
10000
30000

0.000E+00
1.000E+04
1.140E+03
1.000E+04

1.000E+00
2.000E+00
2.114E+00
1.333E+00

5.000E-08
5.000E-08
5.000E-08
5.000E-08

1.250E-08
5.000E-08
5.000E-08
2.887E-08

4.102E+04
2.051E+04
2.051E+04
3.552E+04

4.102E+04 1.000E+00 4.102E+04

#DIV/0!
2.051E+04 5.000E-01 4.102E+04 4.102E+04
2.051E+04 5.000E-01 4.102E+04 4.102E+04
2.051E+04 7.500E-01 4.736E+04 1.421E+05

KMS
35521.40

The first order (leaky integrator) section is common to both problems. This section consists of an input
resistor (conductance) R1 (G1) connected to the inverting terminal with a parallel R2-C combination
G1
feeding back from the output. The transfer function is: H(s) =
. For the normalized design we
Cs + G2
set G1 = G2 = 1 S (R1 = R2 = 1 ) and C = 1 F. This design can be scaled independently of the S&K 2nd
order section. Hence we set Cnew = 50 nF. Thus Km = 20,508.29. Hence R1 = R2 = 20,508.29 .

SOLUTION TO 21.26 AND 21.27. The relevent data from the solution of problem 21.10 is:
k= 1
d1 = 1.0000e+00 7.6537e-01 1.0000e+00
d2 = 1.0000e+00 1.8478e+00 1.0000e+00
fcmin = 8.8800e+01
wcmin = 5.5795e+02

In providing the designs, we set forth all the possible S&K designs using two excel spreadsheets, one for
each second order section.
The designs for denominator d1 with numerator equal to 1 are:
w0^2
1

0.7654

w0/Q

Num
1.00000

w0
1.0000

Q
1.3066

KNLP
1.0000

KMA
46834.37

Kf=wp
557.95

KmR
10000.00

KMB
17922.81

KMC
17922.81

Design A
Design B
Design C
Saraga

RA

1
1
RA

RB
0
1
1-1/Q
RA/3

K
1
2
3-1/Q
3-Apr

C1
2Q
1
1
rt(3)Q

C2
1/(2Q)
1/Q
1
1

R1
1
1
1
1/Q

R2
1
Q
1
1/rt(3)

alpha
KNLP/K
KNLP/K
KNLP/K
KNLP/K

R3
1/alpha
1/alpha
1/alpha
1/alpha

R4
1/(1-alpha)
1/(1-alpha)
1/(1-alpha)
1/(1-alpha)

Design A
Design B
Design C
Saraga

RA

1.00
1.00
3.00

RB
0.0000
1.0000
0.2346
1.0000

K
1.0000
2.0000
2.2346
1.3333

C1
2.6131
1.0000
1.0000
2.2630

C2
0.3827
0.7654
1.0000
1.0000

R1
1.0000
1.0000
1.0000
0.7654

R2
1.0000
1.3066
1.0000
0.5774

alpha
1.0000
0.5000
0.4475
0.7500

R3
1.0000
2.0000
2.2346
1.3333

R4
#DIV/0!
2.0000
1.8100
3.0615

w0 scale
Design A
Design B
Design C
Saraga

1.00
1.00
3.00

0.0000
1.0000
0.1140
1.0000

1.0000
2.0000
2.1140
1.3333

2.6131
1.0000
1.0000
2.2630

0.3827
0.7654
1.0000
1.0000

1.0000
1.0000
1.0000
0.7654

1.0000
1.3066
1.0000
0.5774

1.0000
0.5000
0.4475
0.7500

1.0000
2.0000
2.2346
1.3333

#DIV/0!
2.0000
1.8100
3.0615

wp scale
Design A
Design B
Design C
Saraga

1.00
1.00
3.00

0.000E+00
1.000E+00
1.140E-01
1.000E+00

1.000E+00
2.000E+00
2.114E+00
1.333E+00

4.683E-03
1.792E-03
1.792E-03
4.056E-03

6.859E-04
1.372E-03
1.792E-03
1.792E-03

1.000E+00
1.000E+00
1.000E+00
7.654E-01

1.000E+00 1.000E+00 1.000E+00

#DIV/0!
1.307E+00 5.000E-01 2.000E+00 2.000E+00
1.000E+00 4.475E-01 2.235E+00 1.810E+00
5.774E-01 7.500E-01 1.333E+00 3.061E+00

Km scale
Design A
Design B
Design C
Saraga

10000
10000
30000

0.000E+00
1.000E+04
1.140E+03
1.000E+04

1.000E+00
2.000E+00
2.114E+00
1.333E+00

1.000E-07
1.000E-07
1.000E-07
1.000E-07

1.464E-08
7.654E-08
1.000E-07
4.419E-08

4.683E+04
1.792E+04
1.792E+04
3.104E+04

4.683E+04 1.000E+00 4.683E+04

#DIV/0!
2.342E+04 5.000E-01 3.585E+04 3.585E+04
1.792E+04 4.475E-01 4.005E+04 3.244E+04
2.342E+04 7.500E-01 5.408E+04 1.242E+05

KMS
40559.76

The designs for denominator d2 with numerator equal to 1 are:

w0^2
1

1.8478

w0/Q

Num
1.00

w0
1.00

Q
0.54

KNLP
1.00

KMA
19399.08

Kf=wp
557.95

KmR
10000.00

KMB
33117.77

KMC
17922.81

Design A
Design B
Design C
Saraga

RA

1
1
RA

RB
0
1
1-1/Q
RA/3

K
1
2
3-1/Q
3-Apr

C1
2Q
1
1
rt(3)Q

C2
1/(2Q)
1/Q
1
1

R1
1
1
1
1/Q

R2
1
Q
1
1/rt(3)

alpha
KNLP/K
KNLP/K
KNLP/K
KNLP/K

R3
1/alpha
1/alpha
1/alpha
1/alpha

R4
1/(1-alpha)
1/(1-alpha)
1/(1-alpha)
1/(1-alpha)

Design A
Design B
Design C
Saraga

RA

1.00
1.00
3.00

RB
0.0000
1.0000
-0.8478
1.0000

K
1.0000
2.0000
1.1522
1.3333

C1
1.0824
1.0000
1.0000
0.9374

C2
0.9239
1.8478
1.0000
1.0000

R1
1.0000
1.0000
1.0000
1.8478

R2
1.0000
0.5412
1.0000
0.5774

alpha
1.0000
0.5000
0.8679
0.7500

R3
1.0000
2.0000
1.1522
1.3333

R4
#DIV/0!
2.0000
7.5703
7.3912

w0 scale
Design A
Design B
Design C
Saraga

1.00
1.00
3.00

0.0000
1.0000
0.1140
1.0000

1.0000
2.0000
2.1140
1.3333

1.0824
1.0000
1.0000
0.9374

0.9239
1.8478
1.0000
1.0000

1.0000
1.0000
1.0000
1.8478

1.0000
0.5412
1.0000
0.5774

1.0000
0.5000
0.8679
0.7500

1.0000
2.0000
1.1522
1.3333

#DIV/0!
2.0000
7.5703
7.3912

wp scale
Design A
Design B
Design C
Saraga

1.00
1.00
3.00

0.000E+00
1.000E+00
1.140E-01
1.000E+00

1.000E+00
2.000E+00
2.114E+00
1.333E+00

1.940E-03
1.792E-03
1.792E-03
1.680E-03

1.656E-03
3.312E-03
1.792E-03
1.792E-03

1.000E+00
1.000E+00
1.000E+00
1.848E+00

1.000E+00 1.000E+00 1.000E+00

#DIV/0!
5.412E-01 5.000E-01 2.000E+00 2.000E+00
1.000E+00 8.679E-01 1.152E+00 7.570E+00
5.774E-01 7.500E-01 1.333E+00 7.391E+00

Km scale
Design A
Design B
Design C
Saraga

10000
10000
30000

0.000E+00
1.000E+04
1.140E+03
1.000E+04

1.000E+00
2.000E+00
2.114E+00
1.333E+00

1.000E-07
5.412E-08
1.000E-07
9.374E-08

8.536E-08
1.000E-07
1.000E-07
1.000E-07

1.940E+04
3.312E+04
1.792E+04
3.312E+04

1.940E+04 1.000E+00 1.940E+04

#DIV/0!
1.792E+04 5.000E-01 6.624E+04 6.624E+04
1.792E+04 8.679E-01 2.065E+04 1.357E+05
1.035E+04 7.500E-01 2.390E+04 1.325E+05

KMS
17922.81

SOLUTION TO 21.28. For this problem we use the excel spread sheet given below. First we observe that
H(s) =

K
K 02 (= K NLP )
H ( 0 s) =
=
1
2
0
2
s2 + 0 s + 20
s2 + s + 1
(
(
0 s) +
0 s) + 0
Q
Q
Q
K

Thus after this type of frequency scaling, the new transfer function is:
Hnew (s) =

0.7943
1
s2 +
s +1
1.1286

The dc gain is of course 0.7943 and the modification of the circuit to achieve the correct dc gain is given
in the spread sheet below via R3 and R4 which constitute a voltage divider that replaces R1.

w0^2
0.82306

0.8038

w0/Q

Num
0.65378

w0
0.9072

Q
1.1286

KNLP
0.7943

KMA
1131.42

Kf=wp
43982.40

KmR
10000.00

KMB
501.23

KMC
501.23

Design A
Design B
Design C
Saraga

RA

1
1
RA

RB
0
1
1-1/Q
RA/3

K
1
2
3-1/Q
3-Apr

C1
2Q
1
1
rt(3)Q

C2
1/(2Q)
1/Q
1
1

R1
1
1
1
1/Q

R2
1
Q
1
1/rt(3)

alpha
KNLP/K
KNLP/K
KNLP/K
KNLP/K

R3
1/alpha
1/alpha
1/alpha
1/alpha

R4
1/(1-alpha)
1/(1-alpha)
1/(1-alpha)
1/(1-alpha)

Design A
Design B
Design C
Saraga

RA

1.00
1.00
3.00

RB
0.0000
1.0000
0.1140
1.0000

K
1.0000
2.0000
2.1140
1.3333

C1
2.2573
1.0000
1.0000
1.9549

C2
0.4430
0.8860
1.0000
1.0000

R1
1.0000
1.0000
1.0000
0.8860

R2
1.0000
1.1286
1.0000
0.5774

alpha
0.7943
0.3972
0.3758
0.5957

R3
1.2589
2.5179
2.6613
1.6786

R4
4.8621
1.6588
1.6019
2.1917

w0 scale
Design A
Design B
Design C
Saraga

1.00
1.00
3.00

0.0000
1.0000
0.1140
1.0000

1.0000
2.0000
2.1140
1.3333

2.4881
1.1023
1.1023
2.1548

0.4883
0.9766
1.1023
1.1023

1.0000
1.0000
1.0000
0.8860

1.0000
1.1286
1.0000
0.5774

0.7943
0.3972
0.3758
0.5957

1.2589
2.5179
2.6613
1.6786

4.8621
1.6588
1.6019
2.1917

wp scale
Design A
Design B
Design C
Saraga

1.00
1.00
3.00

0.000E+00
1.000E+00
1.140E-01
1.000E+00

1.000E+00
2.000E+00
2.114E+00
1.333E+00

5.657E-05
2.506E-05
2.506E-05
4.899E-05

1.110E-05
2.220E-05
2.506E-05
2.506E-05

1.000E+00
1.000E+00
1.000E+00
8.860E-01

1.000E+00
1.129E+00
1.000E+00
5.774E-01

7.943E-01
3.972E-01
3.758E-01
5.957E-01

1.259E+00
2.518E+00
2.661E+00
1.679E+00

4.862E+00
1.659E+00
1.602E+00
2.192E+00

Km scale
Design A
Design B
Design C
Saraga

10000
10000
30000

0.000E+00
1.000E+04
1.140E+03
1.000E+04

1.000E+00
2.000E+00
2.114E+00
1.333E+00

5.000E-08
5.000E-08
5.000E-08
5.000E-08

9.813E-09
4.430E-08
5.000E-08
2.558E-08

1.131E+03
5.012E+02
5.012E+02
8.682E+02

1.131E+03
5.657E+02
5.012E+02
5.657E+02

7.943E-01
3.972E-01
3.758E-01
5.957E-01

1.424E+03
1.262E+03
1.334E+03
1.645E+03

5.501E+03
8.315E+02
8.029E+02
2.148E+03

KMS
979.83

1

SOLUTION TO 21.29. (a) HHP (s) = H NLP c =
. At s = j p ,
s 2
c
c
+ 2
+1
s
s

1
1
HHP ( j p ) = HNLP c =
=
2
j p
5.5 2
5.5
c + 2 c + 1
j 2
+1
j
j
7
7
p
p
Thus in MATLAB,
Magfp = 1/abs(1 - (5.5/7)^2 -j*sqrt(2)*(5.5/7))
Magfp = 8.5091e-01
Attenfp = -20*log10(Magfp)

Attenfp = 1.4023e+00
Magfs = 1/abs(1 - (5.5/1)^2 -j*sqrt(2)*(5.5/1))
Magfs = 3.3040e-02
Attenfs = -20*log10(Magfs)
Attenfs = 2.9619e+01
Thus the attenuation at fp is 1.4023 dB and that at fs is 29.619 dB.
(b) From problem 21.15, the transfer function is
1 LC
2
1 1
s2 +
+
s+
C L
LC
and the values of L and C realizing the 2nd order Butterworth NLP transfer function can be computed
according to
Hcir (s) =

1 1
2
1 C
+ = 2,
=1
+ = 2 C 2 2 2C + 2 = 0
C L
LC
C 2
(C 2)2 = 0 C = 2 F L = 2 H
(c) Here Km = 1000. Ls

L C

1
1
LKm C

and
s

1
s
Km

s. Thus in
Cs
C C
C C

MATLAB,
wc = 2*pi*5.5e3
wc = 3.4558e+04
Km = 1000;
C = sqrt(2); L = sqrt(2);
Lhp = Km/(C*wc)
Lhp = 2.0462e-02
Chp = 1/(Km*wc*L)
Chp = 2.0462e-08
Therefore, the resistors take on values of 1 k, the inductor is changed to a capacitor of value of Chp =
20.46 nF and the capacitor is changed to an inductor of value Lhp = 20.46 mH.

SOLUTION TO 21.30. (a)

=

, Amax = 2 dB,

= max = 10 0.12 1 = 0.76478, c = 6

10

0.1 2

Wc=1/(10^0.2-1)^(1/6)
Wc = 1.0935e+00

= 1.0935
1

(b)
wchp = 2*pi*5e3/Wc
wchp = 2.8730e+04
fchp = wchp/(2*pi)
fchp = 4.5725e+03
Thus

chp

= 28.73 krad/s.

( c ) (i) NLP HP transformation: scale by wchp.

Chp =

1
= 17.404 F, Lhp =
= 34.807 H
Llp chp
Clp chp

(ii) Magnitude scale to obtain proper value of capacitors.

Clp = 1;
Llp = 2;
Chp = 1/(Llp*wchp)
Chp = 1.7404e-05
Lhp = 1/(Clp*wchp)
Lhp = 3.4807e-05
Cnewhp = 100e-9;
Km = Chp/Cnewhp
Km = 1.7404e+02
Lnewhp = Km*Lhp
Lnewhp = 6.0578e-03
K m = 174.02, Chp,new =

Chp
Km

= 100 nF , Lhp,new = K m Lhp = 6.06 mH

SOLUTION TO 21.31.
1
The 2nd order NLP Butterworth transfer function is: HNLP (s) = 2
. Using the transformation
s + 2s + 1
s to 1/s, we obtain the NHP Butterworth transfer function:
s2
s2
1
HNHP (s) = HNLP = 2
= K 2
s
s + 2s + 1
s + d(1) s + d(2)

INPUT:

d(1)
1.414213562

d(2)
1

K
1.33334

NHP Params

w0
1

Q
0.70710678

alpha
0.74999625

NHP Crt Pars

HP Crt Params

C1 = Q
C2 = rt(3)
0.707106781 1.73205081
C1new=C1/(Km*Kf)
3.06293E-09

C2new
7.5026E-09

R1 = 1
1
R1new=R1*Km
12247.44871

K
1

Kf
18849.6

R2 =1/(rt(3)Q)
0.816496581

R2new=R2*Km
10000

Rnew=R*KmR
30000

Km
12247.44871

KmR
30000

R/3
0.333333333
Rnew/3
10000

A plot of the design without input attenuation is shown below. Notice that as predicted the gain is 4/3.

R
10k

Vin

C
3.063n

C0
7.503n

15
Vpls

Vminus
15

XOpAmp
R2
10k
R0
12.247k

IVm
R1
30k

S&K HP-Small Signal AC-3

+100.000

(V)

+316.228

+1.000k

Frequency
+3.162k

(Hz)

+10.000k

+1.000

+500.000m

+0.000e+000

MAG(V(IVM))

Input attenuation requires that we replace C1 with a series combination of capacitors in which C1 = C3
+ C4 and (1/C3)/(1/C4 + 1/C3) = alpha. Here then, C1 = C3 + C4 and alpha = C4/(C3 + C4) = C4/C1.
Thus C4 = alpha*C1 and C3 = (1 alpha)*C1. Thus
Input
Attenuation

C3 = (1 - alpha)C1
7.65744E-10

C4 = alpha*C1
2.29719E-09

SOLUTION TO 21.32. The fourth order Butterworth NLP transfer function can be obtained from tables or
from MATLAB as follows:
[z,p,k] = buttap(4)
z=
[]
p=
-3.8268e-01 + 9.2388e-01i
-3.8268e-01 - 9.2388e-01i
-9.2388e-01 + 3.8268e-01i
-9.2388e-01 - 3.8268e-01i
k=
1
% Second Order Sections

n1 = 1;
d1 = poly([p(1),conj(p(1))])
d1 =
1.0000e+00 7.6537e-01 1.0000e+00
n2 = 1;
d2 = poly([p(3),conj(p(3))])
d2 =
1.0000e+00 1.8478e+00 1.0000e+00

Thus,
1
1
HNLP (s) = 2
2
s + 0.76537s + 1 s + 1.8478s + 1
having frequency response
w = 0:0.01:3.5;
h = freqs(k*poly(z),poly(p),w);
plot(w,abs(h))
grid
xlabel('Normalized Frequency')
ylabel('Magnitude 4th Order Butterworth')
1
0.9

Magnitude 4th Order Butterworth

0.8
0.7
0.6
0.5
0.4
0.3
TextEnd

0.2
0.1
0

0.5

1.5
2
Normalized Frequency

2.5

The Saraga design parameters are given in the following Excel tables:

3.5

INPUT:
NLP H(s)
NHP H(s)

d(1)
0.76537
0.76537

d(2)
1
1

K
1.33334

Kf
18849.6

Km
22630.24168

KmR
60000

alpha
0.74999625

NHP Crt Pars

C1 = Q
1.306557613

C2 = rt(3)
1.732050808

R1 = 1
1

R2 =1/(rt(3)Q)
0.441886576

R
1

R/3
0.333333333

Scale by w0HP

1.306557613

1.732050808

0.441886576

0.333333333

HP Crt Params

C1new=C1/(Km*Kf)
3.06293E-09

C2new
4.0604E-09

R1new=R1*Km
22630.24168

R2new=R2*Km
10000

Rnew=R*KmR
60000

Rnew/3
20000

w0LP/HP
1
1

Q
0.541184111
0.541184111

Input Attenuation

C3 = (1 - alpha)C1
7.65744E-10

KNLP/NHP
1
1

w0LP/HP
1
1

Q
1.306557613
1.306557613

C4 = alpha*C1
2.29719E-09

INPUT:
NLP H(s)
NHP H(s)

d(1)
1.8478
1.8478

d(2)
1
1

K
1.33334

Kf
18849.6

Km
10000

KmR
30000

alpha
0.74999625

NHP Crt Pars

C1 = Q
0.541184111

C2 = rt(3)
1.732050808

R1 = 1
1

R2 =1/(rt(3)Q)
1.066827827

R
1

R/3
0.333333333

Scale by w0HP

0.541184111

1.732050808

1.066827827

0.333333333

HP Crt Params

C1new=C1/(Km*Kf)
2.87106E-09

C2new
9.18879E-09

R1new=R1*Km
10000

R2new=R2*Km
10668.27827

Rnew=R*KmR
30000

Rnew/3
10000

Input Attenuation

C3 = (1 - alpha)C1
7.17777E-10

C4 = alpha*C1
2.15329E-09

SOLUTION TO 21.33. Using MATLAB,

fp = 5e3; fs = 1.5e3;
wp = 2*pi*fp; ws = 2*pi*fs;
Amax = 3; Amin = 40;
n = buttord(wp,ws,Amax,Amin,'s')
n =4
[z,p,k] = buttap(n)
z =[]
p=
-3.8268e-01 + 9.2388e-01i
-3.8268e-01 - 9.2388e-01i
-9.2388e-01 + 3.8268e-01i

KNLP/NHP
1
1

-9.2388e-01 - 3.8268e-01i
k= 1
d1 = real(poly([p(1),p(2)]))
d1 =
1.0000e+00 7.6537e-01 1.0000e+00
d2 = real(poly([p(3),p(4)]))
d2 =
1.0000e+00 1.8478e+00 1.0000e+00
In general,
2
1
1
s
1
HNHP (s) = HNLP =
=

2
2
s
2
2 (1 0LP )
1
1
2 ( 0LP )
0LP
s +
s + (1 0LP)
+
+ ( 0LP )
Q
s
Q s

( 0HP )2 s2
s2 + 0HP s + ( 0HP ) 2
Q

The S&K Saraga design for d1 is given by the following excel spreadsheet:
INPUT:
NLP H(s)
NHP H(s)

d(1)
0.76537
0.76537

d(2)
1
1

K
1.33334

Kf
31416

Km
45260.48336

KmR
60000

alpha
0.74999625

NHP Crt Pars

C1 = Q
1.306557613

C2 = rt(3)
1.732050808

R1 = 1
1

R2 =1/(rt(3)Q)
0.441886576

R
1

R/3
0.333333333

Scale by w0HP

1.306557613

1.732050808

0.441886576

0.333333333

HP Crt Params

C1new=C1/(Km*Kf)
9.18879E-10

C2new
1.21812E-09

R1new=R1*Km
45260.48336

R2new=R2*Km
20000

Rnew=R*KmR
60000

Rnew/3
20000

Input Attenuation

C3 = (1 - alpha)C1
2.29723E-10

KNLP/NHP
1
1

w0LP/HP
1
1

C4 = alpha*C1
6.89156E-10

The S&K Saraga design for d2 is given by the following excel spreadsheet:

Q
1.306557613
1.306557613

INPUT:
NLP H(s)
NHP H(s)

d(1)
1.8478
1.8478

d(2)
1
1

K
1.33334

Kf
31416

Km
20000

KmR
60000

alpha
0.74999625

NHP Crt Pars

C1 = Q
0.541184111

C2 = rt(3)
1.732050808

R1 = 1
1

R2 =1/(rt(3)Q)
1.066827827

R
1

R/3
0.333333333

Scale by w0HP

0.541184111

1.732050808

1.066827827

0.333333333

HP Crt Params

C1new=C1/(Km*Kf)
8.61319E-10

C2new
2.75664E-09

R1new=R1*Km
20000

R2new=R2*Km
21336.55655

Rnew=R*KmR
60000

Rnew/3
20000

Input Attenuation

C3 = (1 - alpha)C1
2.15333E-10

KNLP/NHP
1
1

w0LP/HP
1
1

C4 = alpha*C1
6.45986E-10

This completes the design.

SOLUTION TO 21.34. For the woofer,

8
8
H(s) =
= L
Ls + 8 s + 8
L
Thus,

8
= 2000 2 L = 636 H.
L
For the tweeter,
8
8Cs
s
H(s) = 1
=
=
1
+ 8 8Cs + 1 s +
Cs
8C

Thus,

1
= 2000 2 C = 9.95 F.
8C

SOLUTION TO 21.35. For the woofer,

1
HNLP (s) = 2
s + 2s + 1
The transfer function of the following circuit

Q
0.541184111
0.541184111

is
1
LC
H(s) =
1
1
2
s + s+
LC
C
1
1
C = 0.70711 F and since
= 1, L = 2 = 1.4142 H. Frequency scaling the
C
LC
0.70711
element values by Kf = 4000 and magnitude scaling by Km = 8 yields C =
= 7.0337 F and
Km K f
1.4142K m
L=
= 0.90032 mH:
Kf
Thus

2=

For the tweeter we first realize the NLP Butterworth transfer function as above to obtain as
above

with Thus C = 0.70711 F and L = 2 = 1.4142 H. We now apply the frequency transformation
s

C to each element (capacitors become inductors and inductors become capacitors according to
s
figure 21.24) and we obtain the HP circuit topology

where Cnew = 7.0337 F and Lnew = 0.90032 mH.

SOLUTION TO 21.36. Consider figure (a). Let the current entering the RC network from Z1 be dentoted
by Ifa. Let the voltage from this point to ground be denoted Vfa. Then
Vout,a = V fa + I faZ1 = H a (s)Vin
For figure (b) with a similar denotation of voltage and current, we have

V
I fb
Z
k 1
1

fb Z
Vout,b = V fb + I fb + Z 1 = V fb + I fb 1 +
V fb = (2k 1)V fb +

k
k
k
Z1
1 k

k 1
If

(2k 1)V fb V fa and I fb I fa

(**)

then
Vout,b

Vout ,a
1
H a (s)Vin =
k
k

Z1
be large relative
k 1
to what it sees in the RC network. Hence, in general, k must be close to 1. Thus only small gain
enhancements are possible. For such a potentially sensitive approach to gain enhancement, it might be
better simply to add another op amp stage as op amps are comparatively inexpensive.
For gain enhancement, k < 1. However, for the (**) to be valid, we require that

CHAPTER 22 PROBLEM SOLUTIONS

SOLUTION TO PROBLEM 22.1
(a) For figure P22.1a, T0 = 2 and 0 = . Let t0 = -1 in equation 22.5b. Then f(t) = (t) and
1

c n = 0.5 (t)e jnt dt = 0.5

for all n

From equation 22.6, an = 1 and bn = 0 for all n. Finally from equation 22.2
f (t) = 0.5 +

cos(nt)

n=1

(b) For figure P22.1b, T0 = 2 and 0 = . Let t0 = 0 in equation 22.5b. Then f(t) = - (t -1) and
1

c n = 0.5 (t 1)e jnt dt = 0.5e jn

1

From equation 22.6, bn = 0 for all n, and

an = -1 for n even
an = 1 for n odd
Finally from equation 22.2
f (t) = 0.5 + cos(t) cos(2t) + cos(3t) cos(4t) + K

SOLUTION TO PROBLEM 22.2

(a)
T0 = 1 and 0 = 2 . Let t0 = 0 in equation 22.5b. Then
f (t) = e ( ln(2)) t
and from equation 22.5b
1

cn =

e ( ln(2)) t e j 2nt dt =

e
0

0.5
(ln(2) + j2n)

(b) Using the above result for cn

c0 =0.5 = 0.7213 ,
ln2
c1 = 0.5
= 0.7213e -j1.4609
ln2+ j2
c2 = 0.5
= 0.397e -j1.516
ln2+ j4

( ln(2)+ j 2n ) t

dt =

1
e ( ln(2)+ j 2n ) 1
(ln(2) + j2n)

From equation 22.6

d0 = c0 = 0.7213
d1 =2 c1 = 0.158 ,
1 =-1.461x180/ =-83.7o
d2 =2 c2 = 0.0795
2 =-1.516x180/ =-86.84 o
Thus, f(t) in the form of equation 22.3 is
f(t) = 0.7213 + 0.158cos(2t - 83.7o) + 0.0795os(2t - 86.84o)

SOLUTION TO PROBLEM 22.3.

(a)

T0 = 1 and 0 = 2 . Let t0 = 0 in equation 22.5b. Then

f (t) = e ( ln(2)) t [u(t) u(t 0.5)]

and from equation 22.5b

0.5

cn =

e ( ln(2)) t e j 2nt dt =

e ( ln(2)+ j 2n) t dt = (ln(2) + j2n) [e 0.5( ln(2)+ j 2n) 1]

0.5

(1)n
2
=
(ln(2) + j2n)
1

(b) Using the above result for cn, and MATLAB to evaluate the numerical result,
n= 0;
c0= (1- (-1)^n/sqrt(2))/(log(2) + j*2*n*pi
c0
4.2256e-01
n=1;
c1= (1- (-1)^n/sqrt(2))/(log(2) + j*2*n*pi)
c1 =
2.9612e-02- 2.6843e-01i
abs(c1)
ans =
2.7006e-01
degreec1=angle(c1)*180/pi
degreec1 =
-8.3705e+01
n=2;
c2= (1- (-1)^n/sqrt(2))/(log(2) + j*2*n*pi)
c2 =
1.2817e-03- 2.3237e-02i

abs(c2)
ans =
2.3272e-02
degreec2= angle(c2)*180/pi
degreec2 =
-8.6843e+01
From equation 22.6 and equation 22.3
f(t) = 0.4226 + 0.54cos(2t - 83.7o) + 0.04654os(2t - 86.84o)

SOLUTION PROBLEM 22.4. (a) f(t) = cos(4t) sin(2t) = 0.5[ sin(6t) - sin (2t)] .
The fundamental angular frequency of f(t) is 0= 2 rad/s. The given f(t) can be expressed as
f(t) = -0.5 sin( 0t) + 0.5sin(30t) . Observe that b1= -0.5, b3= 0.5 and all other ai and bi are zero.
From equation 22.4 , d1= 0.5 /-90o and d3 = 0.5 /90o. From equations 22.6a and 22.6b.
c1 = 0.25j and c3 = -0.25j. All other cn are zero for n positive
(b) f(t) = sin2 (4t) cos2 (8t)= 0.5[1 - cos(8t)]x0.5 1+ cos(16t)
= 0.25 [ 1 - cos(8t) + cos(16t) - cos(8t) cos(16t)]
= 0.25 - 0.375 cos(8t) + 0.25cos(16t) - 0.125cos(24t)
The fundamental angular frequency of f(t) is 0= 8 rad/s. The given f(t) can be expressed as
f(t) = 0.25 -0.375 cos( 0t) + 0.25cos(20t) - 0.125sicos(30t). Observe that a0= 0.5, a1=-0.375, a2=
0.25 , a3= -0.125 and all other ai and bi are zero.
Fom equation 22.4,
d0= 0.25, d1= 0.375 /180o , d2 = 0.25 /0o, and d3 = 0.125 /180o,
From equations 22.6a and 22.6b.
c0 = 0.25, c1 = -0.375 and c2 = 0.25, c3 = -0.125. All other cn are zero for n positive.
(c)
f(t) = 2 + 1.5 sin(500t)- 2cos(2000t)]cos(10 6 t)
=2cos(106 t) + 0.75sin(1000500t) + 0.75sin(999500t)
- cos(1002000t) - cos(998000t)
The fundamental angular frequency of f(t) is 0= 500 rad/s. The given f(t) can be expressed as
f(t) =2cos( 20000t) + 0.75sin(20010t) + 0.75sin(19990t) - cos(20040t) - cos(19960t)
Observe that: a1996 = -1, b1999= 0.75, a2000= 2, b2001= 0.75 , a2004= -1, and all other ai and bi are
zero. From equation 22.4 , d1996 = 1 /180o , d1999 = 0.75 /-90o, d2000 = 2 /0o ,
d2001 = 0.75/-90o, d2004 = 1 /180o , and all other di are zero.
From equations 22.6a and 22.6b.
c1996 = -0.5 , c1999 = -j0.375, c2000 = 1 ,

c2001 = -j0.375o, dc2004 =-0.5 , and all other cn are zero for n positive..

SOLUTION PROBLEM 22.5

By inspesction, the derivative of f(t) is

A
A
f '( t) = A (t nT ) = f (t)
T n=
T
where f(t) is shown in figure 22/7, with its Fourier series given by equation 22.20b, i.e.

A 2A
f (t) = +
cos(n 0t)
T
T n=1
Therefore
f '( t) =

2A
cos(n 0 t)
T n=1

The dc component is the average value of f(t) and is given by 0.5A. Other terms in the Fourier series of
f(t) are obtained by integrating the cosine terms in the above expression. The result is

f (t) = 0.5A

A
sin(n 0t)
n n=1

SOLUTION TO PROBLEM 22.6

Denote by f5(t) the waveform of figure P22.5, with A= 0.5 and T =1. Then by inspection
f (t) = f prob5 (t) + 0.5
Substituting the result of problem 22.5 into the above equation, we have
0.5
f (t) = 0.75 +
sin(n 0 t)
n n=1

SOLUTION PROBLEM 22.7

Consider the square wave g(t) shown in figure 22.4 with its Fourier series given by equation 22.13. By
inspection, the derivative of f(t) is
f '( t) =

4
g(t) f'(t) = - 4 g(t)
T
T

Substituting equation 22.13 into the above expression, we have

8A sin(n 0t)
f '( t) =

T n=1,odd
n
The dc component is the average value of f(t) and is given by 0.5A. Other terms in the Fourier series of
f(t) are obtained by integrating the sine terms in the above expression. The result is

f (t) = 0.5A +

4A cos(n 0t)

2 n=1,odd
n2

SOLUTION PROBLEM 22.8

The method used below is simpler than that suggested in the hint.
We first sketch the waveform of f'(t) and observe that it may be expressed as the sum of two periodic rectangular
pulse trains:
f'(t)= 1 [ fp (t + 0.5T) +fp (t - 0.5T)]
T
where fp(t) is sketched in figure 22.8. Using equation 22.23, we have

2Asin(n)
f'(t)= 1 { [A +
cos(n0 (t+0.5T))]
T
n
n=1

-[A +

n=1

2Asin(n)
cos(n0 (t-0.5T))]}
n

2Asin(n)
= 1 {
[cos(n0 (t+0.5T))-cos(n0 (t-0.5T)]
T n=1
n
\

- 4Asin2 (n)
= 1 [
sin(n0 t)]
T n=1
n
The dc component is the average value of f(t) and is given by A. Other terms in the Fourier series of
f(t) are obtained by integrating the sine terms in the above expression. The result is

sin(n) 2
f(t) = A + 2A { [
] cos(n0 t)}
n
2
n=1

SOLUTION TO PROBLEM 22.9

We first sketch the waveform of f'(t) and observe that it may be expressed as the sum of a periodic rectangular
pulse train and a periodic impulse train:
f'(t)= 1 fp (t + 0.5T) - f (t )
T
where fp(t) is sketched in figure 22.8, and f(t) in figure 22.7.
Using equations 22.23 and 22.20b, we have

2Asin(n)
f'(t)= 1 [A +
cos(n0 (t+0.5T))
T
n
n=1

- [A + 2A
T T

cos(n0 t)]

n=1

sin(n)
= 2A {[
cos(n0 (t+0.5T)) -cos(n0 t)]}
T n=1 n
The dc component is the average value of f(t) and is given by 0.5A. Other terms in the Fourier series
of f(t) are obtained by integrating the sine terms in the above expression. The result is

sin(n)
f(t) =0.5A + A { [
sin(n0 (t+0.5T)) -sin(n0 t)]}
n n=1
n
It remains to rewrite the expression in the form of equation 22.2.
To this end, let b = sin(n)/(n) and re-write the terms within [ ] as follows:

bsin(n)cos(n0 t) + {bcos(n) - 1}sin(n0 t)

Hence, f(t) in the form of equation 22.2 has the coefficients, for n=1,2...
an =

A sin2 (n)
2 n2

bn =

dn =

a2n + b2n =

A {sin(n)cos(n) -n}
2 n2

A
sin4 (n) + {sin(n)cos(n) -n}2
2 n2

A
sin2 (n) +(n)[n - sin(2n)]
2 n2
n
n = tan-1 (-b
a )
=

The result is item 6 of table 22.4

SOLUTION PROBLEM 22.10.

Observe that the present f(t) can be derived from that of problem 22.9 by (a) replacing t by -t; and (b)
replacing by . Thus the Fourier series for f(t) is:

f(t) =0.5A +[ an cos(n0 t) + bnsin(n0 t) ]

n=1

where
an =

A sin2 (n)
2
n2

bn =

dn =

a2n + b2n =

-A {sin(n)cos(n) -n}
2 n2

A
sin4 (n) + {sin(n)cos(n) -n}2
2 2
n

A
sin2 (n) +(n)[n - sin(2n)]
2 2
n
n
n = tan-1 (-b
an )
=

SOLUTION PROBLEM 22.11. Following the hint, we have the second derivative of f(t) given by
f ''( t) =

1
1
f (t + T)
f (t)
(1 )T
(1 )T

where fd(t) is given in example 22.5. Notice that we have focused on the part of the waveform over
[T, (1)T]. By making use of equation 22.20b, we obtain

2A
f ''( t) =
[cos(n 0 t + 2n ) cos(n 0 t)]
(1 )T 2 n=1

4A
[sin(n 0t + n )sin(n )]
(1 )T 2 n=1

Therefore,

4A
sin(n )

sin(n
t
+
n
)

(1 )T 2 n=1 ( n 0) 2

A A sin(n )
= + 2 2
cos(n 0t + (n 0.5) )
2 n=1 n (1 )

f (t) = [ f (t)]ave +

Letting T = 1 and = 0.25 we obtain,

16A sin n

4
A

f (t) = +
cos
(
n2t
+
(0.25n

0.5)
)
2 n=1 3n 2 2

Therefore, d0 = 0.5A , and

dn =

16Asin n
4

.
3n 2 2
It follows that d1 = 0.38211A and d2 = 0.13509A.

SOLUTION TO PROBLEM 22.12.

Denote by fp(t, ) the period rectangular waveform of figure 22.5, with A=1.
Then we can express the present f(t), with T= 4, as the sum of 3 terms:
f(t) = 3fp (t-0.125T, 0.25) + 4 fp (t - 0.5T, 0.5) -2
From equation 22.14b, and equation 22.12c, for n= 1,2,...
cn = 3 sin(0.25n)e-jn0x0.125T + 4 sin(0.5n)e-jn0x0.5T
n
n
= 3 sin(0.25n)e-j0.25n + 4 sin(0.5n)e-jn
n
n
The numerical values of the first few Fourier series coefficients are:
c0 = averge value of f(t) = 1 (1 + 4 - 2) = 0.75
4
c1 = 3 sin(0.25)e-j0.25 + 4 sin(0.5)e-j = -0.7958 -j 0.4775

3
4
-j0.5

c2 = sin(0.5)e
+ sin()e-j2 =-j0.4775
2
2

Solution Problem 22.13

(a) For sinusoidal steady analysis, the transfer function is
H(j) =

YL
1
1
1
=
=
=
YL +Y C + Y R 1 +ZLYC + ZLYR (1 - 2 LC) + jL (1 - 410 -5 2 ) + j10--3
R

The transfer function evaluated at various input frequencies are listed below.
H(0) = 1
H(j377) = 0.2128 /-175.4o
H(j3x377) = 0.0199/-178.7o
H(j5x377) = 0.0071/-179.2o
Using equation 15.7 and superposition, we obtain the steady state output voltage (in V):
vout (t) = 200 + 200 20.2128cos (377t -175.4o ) +60 20.0199cos (3377t +30o -178.7 o )
+80 20.0071cos (5377t +50o -17.2 o )

= 200 + 242.55cos (377t -175.4o ) + 21.196cos (3377t -148.7 o )

+ 20.5668cos (5377t -129.2 o )
(b) From equation derived in P11.39,
Vout,eff= 2002 +42.552 + 1.1962 +0.56682 = 204.48 V
The average power absorbed by the 10 k resistor is
Pav =204.48
104

= 4.1812 W

Solution Problem 22.14

One correction in the problem statement: in the angle expression, 5000 should be 10000.
(a) Using the identity cos(x)cos(y)= 0.5cos(x+y) + 0.5cos(x-y), we have
vin (t) = 0.1cos (998,000t) +0.2cos (999,000t) + 2cos (1,000,000t)
+ 0.2cos (1,001,000t) + 0.1cos (1,002,000t)

(b) The transfer function has a constant magnitude of 10, and a phase shift proportional to the deviation
from c. at = c + 2m, the phase shift is -9 degrees. From these facts, we can write directly
vout (t) = cos (998,000t +9 o ) +2cos (999,000t + 4.5o ) + 20cos (1,000,000t)
+ 2cos (1,001,000 - 4. 5o ) + cos (1,002,000t - 9o )

(c) Using the identity

x+y
x+y
cos(x)cos(y) = 2 cos(
)cos(
)
2
2
we can group the terms in vout(t) and re-write it as

vout(t) = 2 cos( 2 mt - 9o)cos( ct) +4cos( mt - 4.5o)cos( ct) + 20cos( ct)

= [20 + 4cos( mt - 4.5o) +2 cos(2 m - 9o)]cos( ct) = g(t)cos( ct)
Thus

g(t) = 20 + 4cos( mt - 4.5 o) +2cos(2 mt - 90)

With td = 78.54 s, then mtd = 1000 x 78.54x10-6 = 0.07854 rad, or 4.5 degrees. We have

10 f(t - td) = 20[1 + 0.2cos( m(t - td) +0.1cos( 2 m(t - td)]

= 20[1 + 0.2cos( mt - 4.5o) + 0.1 cos( 2 mt - 9o)] = g(t)

SOLUTION PROBLEM 22.15. (a) This proof is a special case of the general proof given in the
solution to Problem 22.16. See the solution to problem 22.16, below.

FOR THE REMAINING PARTS , WE USE THE FOLLOWING MATLAB CODE :

% chapter 22, problem 15.
%part (b).
Vmax= 30*pi;
Vmin=0;
T=4;R = 1; C = 1;
voutmin= Vmin+(Vmax - Vmin)/(1+ exp(0.5*T/(R*C)))
voutmax = Vmax - (Vmax -Vmin)/(1+ exp(0.5*T/(R*C)))
t1= 0: 0.05: 2;
vseg1= Vmax+(voutmin- Vmax)*exp((-t1/(R*C)));
t2= 2:0.05: 4;
vseg2= Vmin+(voutmax- Vmin)*exp(-(t2-T/2)/(R*C));
t= [ t1 t2];
v= [vseg1 vseg2];
plot(t,v)
grid
xlabel('t in seconds')
ylabel('Steady State Output Voltage in V')

%part (c)
error1= 100*(12.235- voutmin)/voutmin
error2= 100*(82.013 - voutmax)/voutmax
TO OBTAIN

(b)
voutmin =
1.1235e+01
voutmax =
8.3013e+01
(c)
error1 =
8.9045e+00
error2 =
-1.2048e+00

SOLUTION PROBLEM 22.16. CORRECTION: In the problem statement,

should be s .

For simplicity, let us consider the case when the transfer function is a voltage ratio, i.e., H(s) = Vout./Vin.
If a constant input vin(t) = Vcon is applied to the stable network, then the vout() = KV con ,

independent of the initial conditions. This is because the zero-input response for a stable network
approaches zero as t approaches infinite. To see this observe that the zero-state response is given by
K
vout (t) = L1
s + 1

KV
V
KV
con = L1 con con = KVcon 1 e t
s
s +1
s

)u(t)

from which vout() = KV con as asserted.

For the remainder of our proof we make use of the fact that in steady state, vout(t+T) = vout(t)
with t = 0 in our case. Specifically, after the first order network has reached steady state, the vout(t)
waveform will be periodic as shown in figure P22.15c, where the time reference has been chosen so that
vout,min occurs at t = 0.
Recall equation 8.19

x(t) = x() + x(t +0 ) x() e (t t0 )

Applying this equation to the interval [0, T/2], we have

b0 = KVmax + (a0 KVmax ) e 0.5T / KVmax + (a0 KVmax )
Note that

(1)

= e 0.5T / . Similarly, applying equation 8.19 to the interval [ T/2, T] leads to

a0 = KVmin + (b0 KVmn )

(2)

Equations (1) and (2) can be written as a single matrix equation

1 a0 (1 )KVmax
=
b0 (1 )KVmin

(3)

Solving equation (3) by matrix inverse (or Cramer's rule) results in

a0
b =
0

1
2 1 1

[v out (t)]min = a0 = K
= KVmin +

1 (1 )KVmax
K
(1 )KV = + 1 1

min

1 Vmax
K Vmax + Vmin
V = + 1 V

min
max + Vmin

Vmax + Vmin
V
+ Vmin + Vmin Vmin
= K max
+1
+1

Vmax Vmin
V
Vmin
V
V
= KVmin + K max
= KVmin + K max 0.5T min
/
+1
1+ 1/
1+ e

and

[v out (t)]max = b0 = K Vmax ++ 1Vmin = K Vmax +

Vmin + Vmax Vmax

+1

V
Vmin
V
V
= KVmax K max
= KVmax K max 0.5Tmin
/
1 + 1/
1+ e
This complete completes the derivation of the desired formulas.

SOLUTION TO PROBLEM 22.17

There is one correction in the problem statement. 1-kHz should be 5.1mHz.
(a) The transfer function is, according to equation 4.3, and using the given element values R1 = 10 k,
Rf = 50 k and C = 20 mF:
Rf
1
V
Z
Y
R
R
1 =1
-5
H(s) = out = - f = - 1 = =
Vin
Z1
Yf
1
1
+
R
Cs
1
+
1000s
f
Cs +
Rf
(b) Here we have We have K= -5, t = 1000 s, Vmax = 1 V, Vmin = -1 V and T = 1/f = 1/0.0051=
196.15 s. In using the equations derived in problem 22.16, we note that the subscripts min and max
should be switched in this case because K is negative. The following MATLAB codes perform the
needed numerical calculations.
f= 5.1e-3;
Cf=20e-3;
Rf=50e3;
Rs= 10e3;
T=1/f;
K= -Rf/Rs;
tau=Rf*Cf;
['part (b)']
Vmax=1;
Vmin=-1;
voutmax= K*( Vmin + (Vmax-Vmin)/(1+exp(0.5*T/tau)))
voutmin= K*( Vmax - (Vmax-Vmin)/(1+exp(0.5*T/tau)))
answers from MATLAB are:
voutmax =
0.2449
voutmin =
-0.2449
(c)
['part (c)']

t1= 0:0.005*T:0.5*T;
v1= 5 + (-0.245 -5).*exp(-(t1/tau)) ;
t2= 0.5*T:0.005*T:T;
v2= -5 + (0.245 +5).*exp(-(t2-0.5*T)/tau) ;
t=[t1,t2];
v=[v1,v2];
plot (t,v)
xlabel('time in seconds')
ylabel('vout in V')
grid

0.3

vout in V

0.2
0.1
0
-0.1
-0.2
-0.3

50

100
time in seconds

150

200

SOLUTION PROBLEM 22.18.

vi (t) = cos (t) = cos(6000t)
vo (t) = 10v i + v2i + v3i
= 10cos(t) + cos2 (t) + cos3 (t)
= 10cos(t) +0.5 +0.5cos(2t) + 0.75cos(t) + 0.25cos(3t)
= 0.5 + 10.75 cos(t) + 0.5cos(2t) 0.25cos(3t)
The effective values of various components of the output are:
dc
0.5

fundamental
10.75/2

2nd harmonic
0.5/2

The total harmonic distortion is

0.559 100% = 5.2%
10.75

3rd harmonic
0.25/2

total harmonic
0.559/2

and the average power at the fundamental frequency is

2
0.5 10.75 = 0.5778 W
100

SOLUTION TO PROBLEM 22.19.

vo (t) = V0 + V1 cos(t) + V2 cos(2t) + V3 cos(3t)
We use equation 22.30 and the values of vo(t) read from the oscilloscope
to compute Vk. The total harmonic distortion is given by

H.D. =

V22 + V23
100 %
V1

MATLAb codes:
vo0= 10; vo60= 5.2; vo120= -4.6; vo180= -9.6;
V0= (vo0+2*vo60+2*vo120 +vo180)/6
V1= (vo0+ vo60 - vo120 -vo180)/3
V2= (vo0 - vo60 - vo120 +vo180)/3
V3= (vo0-2*vo60+2*vo120 -vo180)/6
% total harmonic distortion
HD= 100*sqrt( V2^2 +V3^2)/V1
The following answers are obtained frm MATLAB output.
V0 = 0.2667 V
V1 = 9.8000 V
V2 = -0.0667
V3 = 0 V
HD = 0.6803 (percent)

SOLUTION PROBLEM 22.20.

We shall follow the solution given in example 22.12, and only indicate the needed changes below. There are two
corrections in example 22.12: (1) 20cos(T) should be 20cos(0.5T) and (2) 1.842cos(t) should be 1.842
cos(3t).
The new input is
vi (t) = 0.9 cos(6000t)
The positive peak of the output sine wave is clipped for T.

0.9 20cos(0.5T) = 15 = cos-1 (15/18) = 0.1864

Equation 22.35 becomes

Equations 2.36 becomes

Using A=1, and = 0.1864in item #2 of table 22.4, equation 22.37 becomes

and equation 22.38 becomes

Equation 22.39 becomes

from which the harmonic distortions are:

third order
fifth order

1.076/16.57 = 6.595%
0.55/16.57 = 3.319%

SOLUTION PROBLEM 22.21. Corrections: (1) cos( 1 )e

cos( 1

)e

2 1 +

RC

2 1 +

RC

= cos( 2 ) should be

= cos( 2 ) and (2) on page 943, equation 22.46 , 2 should be 2, i.e., equation

22.46 should be cos( 1

)e

2 1 +

RC

The proof is similar to that given on page 943 for the half-wave rectifier case, except for some
minor changes described below.
For the case of a full-wave rectifier, the output voltage waveform is a modification of figure 22.17
as shown below.

The exponential decay of vo(t) starts with the value Vmcos(1) at = 1 (instead of 1 2, as in the
half-wave rectifier case). Therefore, in equation 22.45, change T to T/2 and 2 to . In other words

v 0 (t ) = (Vm cos( 1))

1 T

t
2

RC

= (Vm cos( 1)) e

In equation 22.46, make correction (2) above, and change 2 to , i.e.,

cos( 1

)e

2 1 +

RC

= cos( 2 )

t ( 1
RC

The desired proof is complete.

SOLUTION PROBLEM P22.22

Refer to figure 22.17. Assuming 1 = 0, we have
v0 (t) = Vmexp( -t ) = Vm 1 - ( t ) + ( t )2 - ( t )3 + ...
RC
RC
RC
RC
Assuming 2 = 0, we compute the average value of vo(t) over the time interval [ 0, T]. For the case
RC >> T, we can approximate vo(t) over this interval by keeping only the first two terms of the infinite
series. Thus
v0 (t) Vm 1 - t
RC
which indicates that the plot of vo vs. t over the interval [ 0 T] is approximately a straight line.
Therefore the average of the vo(t) over [ 0 T] is equal to vo(T/2). Thus, for the case RC >> T, or
equivalently RC >> 2,

Vdc v o T = Vm 1 - T = Vm 1 -
2RC
RC
2

SOLUTION PROBLEM 22.23

Given values are: C = 20e-6 F, R =100k, Vm = 20 V; f = 60 Hz. From equation 22.48b
Vdc (1 -

)V = (1 - 1 )Vm = (1 1
)20 = 19.916 V
RC m
2fRC
5
26010 2010-6

To calculate the ripple factor, we first calculate 2 from equation 22.47,

and then use the result in equation 22.50.
2 = cos-1 ( 2 ) = 0.1289 rad
RC

ripple factor

1 - cos( 2 )

=0.2406%

3 [ 1 +cos( 2 )]
For the diode average and peak currents, use equations 22.49a and 22.52
Idc = Vdc = 19.917 = 0.199 10-3 A
R
100,000
id,peak VmCsin( 2 ) = 19.4 10-3 A

SOLUTION PROBLEM 22.24

Equations 22.44 - 22.52 are derived for a half-wave rectifier. For a full-wave rectifier, some of these equations will
be modified slightly as given below. The difference arises from changing T to T/2.
The new equation for 2 is derived in problem 22.21, and repeated below.
The given values are: C = 20e-6 F, R =100k, Vm = 20 V; f = 60 Hz.
A modification of equation 22.48b gives
Vdc = (1 -

)V = (1 - T )Vm = (1 1
)20 = 19.96 V
2RC m
4RC
5
46010 2010-6

To calculate the ripple factor, we first calculate 2 from equation derived in problem 22.21, and then use the result
in equation 22.50.

2 cos-1 (e- RC ) =5.23 degrees

ripple factor

1 - cos( 2 )

=0.12%

3 [ 1 +cos( 2 )]
For the diode average and peak currents, use equations 22.49a and 22.52
Idc = Vdc = 19.96 = 0.1996 10-3 A
R
100,000
id,peak VmCsin( 2 ) = 13.75 10-3 A
_

SOLUTION PROBLEM 22.25. CORRECTIONS: (1) 195 should be 1950 . (2) 100 F should
be 10 F.
Since this problem only requires an estimate of the answer, we can use reasonable approximations to simply
the solution. Let H(s) = Vo (s) I(s) be the transfer function for the linear circuit to the right of the diodes. Then,
Z par (s)
V (s) V (s)
H(s) = in o
= Zin (s)
I(s) Vin (s)
R + Z par (s)
where Z par (s) is the impedance of the parallel C-RL.
The first step is to find the magnitude H ( j ) . As long as 1/RL << C and 1/(C) << R, the
following approximations are valid:
Z par ( j ) =

1
j C+
RL

1
C

which means that the parallel impedance is essentially that of the capacitor, and
Z par ( j )
Z par ( j )
1

R + Z par ( j )
R
RC
and again since 1/(C) << R,
Zin ( j )

1
C

Using these approximations,

Z par ( j )
1
1
R
H ( j ) = Zin ( j )

=
R + Z par ( j )
C
RC ( RC) 2
It is given that Idc = 0.01 A. From the short pulse property, the input current i(t) consists of very short pulses at
120 Hz and all ac components of i(t) have peak magnitudes approximately equal to twice the average dc value.
Hence the peak magnitudes are 0.02 A. Therefore the magnitude of 120-Hz component of the output voltage is
H ( j2 120) 0.02

( RC )

0.02 =

1950
2 120 1950 10 106

0.02 = 0.18042 V

Hence, the effective value is 0.18042

2 = 0.12757 V and the ripple factor is

0.12757
100 = 0.42524 %
30

SOLUTION PROBLEM 22.26

Since this problem only requires an estimate of the answer, we can use reasonable approximations
to simply the solution. Let H(s) = Vo (s) I(s) be the transfer function for the linear circuit to the right of
the diodes. Then
H(s) = Vo = Vin Vo
I
I Vin
Under the condition 1/(C) << R and 1/(C) << RL, we have
Zin (s) ZC(s) = 1
Cs
and

Vo 1
Vin RC

Using these approximations,

H(j) Zin Vout = 1 ( 1 )2 = R( 1 )3
Vin C RC
RC
For a full-wave rectifier, the fundamental frequency is 120 Hz, and = 240 rad/s. It is given that Idc
= 0.01 A. From the short pulse property, the input current i(t) consists of very short pulses at 120 Hz
and all ac components of i(t) have peak magnitudes approximately equal to twice the average dc value.
Hence the peak magnitudes are 0.02 A. Therefore the peak magnitude of 120-Hz component of the
output voltage is

1
H(j120) x0.02 = R( 1 )3 0.02= 975
RC
1209751610-6
Hence, the effective value is 0.012/ 2 =0.0848 V and the ripple factor is

) 0.02 = 0.012 V
3

0.0848 100% = 0.2827 %

30

SOLUTION PROBLEM 22.27

Using H(j) of figure P22.27 and the given formula for cn , we have
0.5s

0.25s

H(j) e -jnT0 d = 1
s

cn = 1
s
-0.5 s

e-jnT0 d =
-0.25 s

s
1
[ e-jnT0 ]0.25
-0.25 s
-jnT 0 s

= 1 sin (n ), n = 1,2,...
n
2
For n = 0, we have c0= d0 = a0 /2 = [Average value of H(j) ]= 0.5 .
For n = 1,2,..., from equation 22.6

b n = 0,
dn=an = 2 sin (n ),
n
2
The Fourier series representation of H(j) is

H(j) = 0.5 + 2 cos(T0) - - 1 cos(3T0) + - 1 cos(5T0) - - 1 cos(7T0 + ...

3
3
7
A plot of H(j) vs. curve using the first 11 terms (n=0,1,...11) of the Fourier
series is given below together with the MATLAB codes.

1.5

H(jw)

1
0.5
0
-0.5

20

40
60
w in radians

%chapter 2, problem 27.

ws= 100;
T0=2*pi/ws;
w= 0: ws/200:ws;
d0w = 0.5;
d1w = (2/pi)*cos(T0*w);
d3w= -(2/pi/3)*cos(3*T0*w);
d5w= (2/pi/5)*cos(5*T0*w);
d7w= -(2/pi/7)*cos(7*T0*w);
d9w= +(2/pi/9)*cos(9*T0*w);
d11w= -(2/pi/10)*cos(11*T0*w);
H =d0w +d1w + d3w +d5w +d7w + d9w +d11w;
plot(w,H)
xlabel(' w in radians')
ylabel('H(jw)')
grid

80

100