Professional Documents
Culture Documents
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).
C 3.1 6.023 10 23
Total Charge = 4.646 10 18 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
(d) We know there are = 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
3.1NA
Net charge = 0.05 1.6019 1019 = 234C
64
( )
Solution 1.3 (a) 7.573 1017 1.6019 10 19 = 0.1213C
0.1213
(b) Current = = 121.3A flowing from right to left.
103
(c) Again, use proportionality:
x 1.6019 10 19 10 60
10 A = x= 19 = 3.75 1021
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
(b) By inspection:
t t t
sin() = t sin(t) C
1 1
q(t) = i()d = (1 cos()) d = t
0
0 0
(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.
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
t
(ii) 1 ms t 2 ms, q(t) = 1 2 d = 3 2t C
1
t
(iii) 2 ms t 3 ms, q(t) = 1+ d = 3 + t C
2
t
(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
0.8
0.6
0.4
Charge in micro C
0.2
-0.2
TextEnd
-0.4
-0.6
-0.8
-1
0 1 2 3 4 5 6
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
1
Voltage in V
0.5
TextEnd
0
-0.5
-1
0 1 2 3 4 5 6
Time in ms
(b) PB = VI = 27 = 14 W
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.)
I-source: ( )
Pabsorbed = 3 1 e t = 3 + 3e t watts
(b) P = (16-4I L2 )IL. Differentiate this w.r.t. IL and set to zero: 16 12IL2 = 0. Therefore, IL = 1.155A.
Solution 1.17 (a) Power is the product of the current and voltage. We can compute the product
graphically:
10
8
Power in Watts
4
TextEnd
2
0
0 0.5 1 1.5 2 2.5 3
Time in s
(b)
t t
(
W (t ) = p(t )dt = 10 10e 7 d = 10 ]t0 10
7 e
7 t
0
) 10
[
= 10t + e 7t
7
10
7 ]
0 0
6
Energy in J
4
TextEnd
3
0
0 0.5 1 1.5 2
Time in s
Solution 1.19 (a) t = 100o F, Rate of temp. increase is 2.5 Wh/o F per gallon:
(b) Heater generates P = 12010 = 1200 W. We want 7500 Wh. Therefore, the total number of hours
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
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:
= 0.0032 0.00182 = 1.81 10 3 m2
2 2
area = Rout Rin
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 .
Solution 1.28. (a) Power in a wire: P = I2R. Rearranging, we can express the current as
I= P
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 1.30 (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
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 1.35. (a) Use definition of power and substitute given power:
V2 = P R = 98 2 = 14 V
P 12 768.8
Similarly, I3 = = = 2 A, V4 = P R = 16 4 = 8 V, I5 = = 12.4 A, and
R 3 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 .
0 = Ix2 4I x + 4 = ( Ix 2)
2
Thus, Ix = 2 A.
(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).
2000sin (4t )
t t
W (t ) = p(t )dt = 2000t + 2000 cos( 4t )dt = 2000t + J
4
0 0
6000
5000
4000
Energy in J
3000
TextEnd
2000
1000
0
0 0.5 1 1.5 2 2.5 3
Time in s
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.
(b)
Vout R2 100 10
= = = 5 R1 = 200
Vin R1 R1
(c)
2 Vin2
R2
R12 2 R2
Power gain = = = 500
Vin2 R1
R1
Solution 1.48 I1 = Vin /10 = 0.1 A, VR = 10(Vin /10)R = R; Vout = 5R10 = 50R =50RVin
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 2.8. 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
(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
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 %
(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
iii. And the relative efficiency is = (Vmotor / Rmotor)/ (12 8)100 = 83.3 %
2
(d) What is the largest equivalent resistance of the motor that will draw 30A? R = 12/ 30A = 0.4.
0.04
SOLUTION 2.22. I100 = = 0.02 A. Therefore V300 = 0.02 (100 + 200) = 6 V. By KCL,
100
6
I150 = 0.02 + = 0.04 A. Req, seen by the source, is 300 . Therefore Vs = 0.04 300 = 12 V.
300
P2 = I2
2
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 2.27. 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 2.30. 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.
2R
SOLUTION 2.32. Using voltage division, at t=0 v R = 15 = 10V , and t = 5 s v R = 10V , and at t = 10
3R
the voltage goes back to 0V.
Chap 2 Probs P2 - 7 R. A. DeCarlo, P. M. Lin
10
6
Volts
0
0 5 10 15
time (sec)
Rb Rd
Vb = 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
Rb Rd
For arbitrary Vin , this requires that = or equivalently that Rb Rc = Ra Rd .
Ra + Rb Rc + Rd
SOLUTION 2.35. By current division, for I1 to be 2A then 160 + R = 300 ||600 for an even split. Thus
R = 40 .
1/10
SOLUTION 2.36. By current division, i1 = 0.4A = 0.32 A . Therefore using KVL
1/10 + 1/ 40
v d = 10i1 0.25i1 = 3.12V .
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
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 2.44. (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.
SOLUTION 2.45. Using the formulas for parallel resistances, the circuit of figure 2.45 reduces to
(b) RAB cannot be calculated by series parallel formulas, but RBC can be done.
(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 =
1/18k + 1/ 9k 1 1 36m = 3mA
+
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.
30 30 30
0.15 = + = + 0.05
R 600 R
Hence R = 30/0.1 = 300 .
120
SOLUTION 2.50. (a) First, express the total current as I = . 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 .
260
SOLUTION 2.51. At time 0, all switches are open and Vout = 220 = 190.7V .
260 + 40
Then at t = 5s, switch one closes and
260||260
Vout = 220 = 168.2V .
(260 ||260) + 40
At t = 10s,
130 ||260||260
Vout = 220 = 136.2V .
(130 ||260||260) + 40
Finally at t =15 s,
65||130 ||260||260
Vout = 220 = 98.5V .
(65||130 ||260||260) + 40
200
180
160
Volts
140
120
100
80
0 2 4 6 8 10 12 14 16 18 20
Time (sec)
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
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
P5 = 5i5
2 = 180W
SOLUTION 2.56. From Ohms law I1 = 100m / 200 = 0.5mA. By current division
IRL =
20k
150I1 = 75.257mA, and PRL = 200I 2RL = 1.103W .
20k + 200
SOLUTION 2.58. Observing the following relationship, V1 = Vin , the following nodal equation can written:
Iin = Vin / 3 + Vin / 6 2Vin = 1.5Vin .
18 22 11
V1 = 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
AV1 = A Vs =
6 3 3 9A
I2 = V
6+2 4 4 16 s
Therefore
2 81A2 2 81A2 2
P2 = 2 I2 =2 V = V
256 s 128 s
81A 2 2 11 2
Vs = 10 V
128 288 s
Hence
128 110
A= = 0.7769
81 288
6
SOLUTION 2.60. By voltage division V1 = V = (3/ 4)Vin . By current division, and substituting the
6 + 2 in
3
previous equation I2 = 4V = Vin . Using voltage division and Ohms law, and substituting the previous
3+6 1
equation,
10
Vout = 4.5I2 = 3Vin = 30V
10 + 5
Iout = 4.5I2 / (10 + 5) = 0.3Vin = 3A
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
240 . Solving
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 R. A. DeCarlo, P. M. Lin
240||2.22||2.22
(e) Using voltage division, Vout = 14.7 = 14.186V
(240 ||2.22||2.22) + 0.04
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 ,
40
one sees that for zero current vout = Vs = 40V . The slope of the line is = 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
4000
iout = Is = 5A . From the slope of the line, , Rs = 4000 / 5 = 800 .
5
Chap 2 Probs P2 - 15 R. A. DeCarlo, P. M. Lin
T nI
SOLUTION 2.67. Using the following formula: = 1, solve for T, and get 0.625 hrs, or 37.5 min.
n Cn
1/
SOLUTION 2.68. Using the same equation as before and solving for Cn = nI
T
, with n=10, and
n
T=55/60 hrs, the capacity obtained is 20 Ah.
(b) Following is the graph, and the script used to generate it.
4
Power in Watts
0
0 10 20 30 40 50 60 70
800
600
Current in mA
400
200
0
0 10 20 30 40 50 60 70
Resistance in Ohms
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.
Chap 2 Probs P2 - 16 R. A. DeCarlo, P. M. Lin
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
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
d c b a
R1 R2 R4 R6 R7 R9
Ga=1/R10+1/R11;
Ra=1/Ga;
Rb=Ra+R9+(1/(1/R7+1/R8));
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 R8 R9
e d c b R7 a R11
R1 R5
R3
R2 R10
R6
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
Vx 2V x V x
0.6 =0
100 100 50
Vx 2Vx 2Vx = 60
5Vx = 60
V x = 12V
Solution 3.3
Vx 25Vx Vx Vx 0.2Vx
0.6 + =0
100 100 50 40
3V 25V x 8V x
x + = 0.6
100 100 400
0.6 400
Vx =
80
V x = 3V
Va 1 35m 5m 0.5
V = 525 25 5m 15m 0.5
b
1 35 5 0.5
=
0.5 5 15 0.5
40
=
20
Solution 3.7 (a) Again, the matrix equation can be written by inspection:
G1 + G2 + G4 G4 VB 50G1
=
G4 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.
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
3
(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.
Rearranging, we have:
1 1 V
+ 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)
1/20 k + gm1 + 1/10 k 1/10k 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 3.16
V VB V A VB
1) VA = 3000i x = 3000 A =
9000 3
Equation at node B:
VB VA V V VD
+ B + B = 0 is equivalent to:
9000 6000 18000
2) 2V A + 6VB VD = 0
Equation at node D:
VD VB V
+ D + IS = 0 which can be rewritten as:
18000 9000
3) 3VD VB = 360 Solving the system formed by equations (1), (2) and (3) we obtain:
VA = 9V , VB = 18V , VD = 126V .
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
[ ]
(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.
P2k = I1 I1 R = 0.8W.
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.
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: Vin = 2Iin + 14Iin 10V1
V1 = 2Iin
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 )
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:
It follows that:
P = (I 1 + I2 )(590 + 491.296) = 31993W.
Solution 3.38
(a)
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.
(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:
I2 = 0.1A, I3 = 0.
(c) 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
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
100k 0 I 2 90k 4.8m 50
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.
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
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:
Mesh 3 equation: R1(i3 i1) + i3R4 + R3 (i3 i2 ) =0
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 0 1 I1 Vs1
1 1 0 I2 = Is 3
0 10M 1 v Vs2
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
0 1 1 0 0 v 0.5m
0.5 1 0 0 v 2 2m
0
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
I4
I1
I2 I3
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 3.54 (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:
0.1 0 0.09 1 0 V1 2
0
0.1 0.05 1 0 V2 0
0.09 0.05 0.14 0 1 V3 = 0
0 0 1 0 0 I1 50
2 0 0 I 2 0
1 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:
V A 7V
V 5V
B
VC 10V
I x = 1A
I 1A
y
Ib 2A
Is1 0A
V A 38V
V 20V
B
VC = 18V
Ia 0.0075A
I 0.002A
x
Solution 3.59 Using the appropriate element stamps for each element of the circuit, we obtain the
following system:
0.15 + 0.2 0.15 0.2 0 V A 8 3
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
2
1
Solution 3.62 (a) Because RT (T ) can be approximated by a straight line between (250,0 o C) and
0o C 250 54.4415V
5o C 233 52.4136V
10o C 216 50.2368V
15o C 199 47.8938V
o
20 C 182 45.3650V
25o C 165 42.6273V
30oC 148 39.6537V
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 I1 = 0.0096A , I2 = 0.1731A , I3 = 0.0673A;
6I + 14I = 1
1 3
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 = 3.
Solution of 3.66
(a)
At node A: (VA-VC)/2 + (VA-VB)/2 + (VA-VD)/2 = 14
At node B: (VB-VA)/2 + (VB-VC)/2 + (VB-VD)/2 = 7
At node C: (VC-VA)/2 + (VC-VB)/2 + (VC-VD)/2 + 2VC = 0
At node D: (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
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 I2 28 I2 6.75
2 2 I 14 I 1.75
2 6 3= 3=
2 6 2 0 I 4 14 ; I 4 5.5
2 0 2 4.5 I5 7 I5 1
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
P
Vout = (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 4.4. This is essentially the basic inverting configuration, which is defined as
Vo / Vi = 2k / 1k = 2 .
Thevenin Probs, 7/11/01 - P4.2 - @R.A. Decarlo & P. M. Lin
100
SOLUTION 4.5. (a) By voltage division VL = 1V = 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.
Vout = V1
24
= 0.5Vs
24 + 8
(b) By voltage division,
V1 = Vs
32
= 0.8Vs
32 + 40
Vout = V1
24
= 0.6Vs
24 + 8
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
(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 .
Rf Rf Rf
solving for Vout = V1 + V2 + V .
R1 R2 R3 3
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,
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,
(b) The requirement for G f = 2S , sets the scaling constant to 2uS. So the following parameters are
obtained:
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
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
Ra1 = 1/ (4S) = 0.25, Ra2 = 1/ (2S) = 0.5, Rb1 = 1/ (5S) = 1/ 5, Rb2 = 1/ (4S) = 0.25, R f = 1
Ra1 = 12.5k, Ra2 = 25k, Rb1 = 10k, Rb2 = 12.5k, R f = 50k, Rg = 50k, R = 16.67k
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 - P4.6 - @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
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 R1
relationship, vswitch = vref 1 1 = v . Choose vref = 1.5V , and R1 = 2k and R2 = 1k.
R2 R2 ref
Verifying in SPICE,
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
1
100 100. Thus for the conditions listed in the problem, it will always be less than
( )
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 Rf RL Rf
V = vout + vin 1/ + + 1 = vout
R1 R1 Rin R
A out
Rf
and
Rf
1
vout / vin =
1 + Rout + Rout 1+ f + f
R1 R R
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 4.34. 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,
[ ] v Rf
vin / R1 = vout / R f where is the fraction of vout that appears across Rf. Hence, out = .
vin R1
v Rf
When the slider is at the top = 1 and out = . When the slider is at the bottom, the fraction of vout
vin R1
R f / / R0 R f R0 1 R f R0
appearing across Rf is = = R R = . Hence
R f / / R0 + Rp R f + R0 f 0 R R + R (R + R )
+ Rp f 0 p f 0
R f + R0
1 R f R0 + R p (R f + R0 ) Rp Rp vout Rf Rf Rp Rp
= =1 + + = = + +
R1 R0 R f
. It follows that 1 .
R f R0 R0 Rf vin R1
Therefore the range of achievable voltage gain is
Rf v Rf Rp Rp
out 1 + R + R
R1 vin R1 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.
SOLUTION 4.41. The current through the LED is IL = 10 1 / 3.8k , so for (a) it is 1.32mA and for
R
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 Vo
Iout = V 2
R1R2 R2
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 ,
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.
Rf Rf Rf Rf
Vout = Vo V1 V2 .
Ro R1 R2 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,
E2
R = 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
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,
A
R
out
Vout / Vin =
1 1
R + R
L out
1 1 1
Note that = 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
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
L out
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
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 R2
From the previous question, Vout = 2 (V1 V2 ) = (1+ 2Ra / Rb )(Vs1 Vs2 ) . Thus
R1 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:
(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.
Isnew = 0.4 A.
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:
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
V5 = V3 + V4 = 4.5 I5 = = 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
V9 = V8 + V7 = 40.5 I9 = = 13.5 I10 = I9 + I8 = 20.25 V10 = I10 2 = 40.5
3
Solution 5.4 Label the resistances R1 to R10 progressively from right to left just like in the previous
V1
Iout = 1 V1 = Iout 4 = 4 I2 = =1
4
V
I3 = I1 + I 2 = 2 V3 = I3 4 = 8 V4 = V3 + V1 = 12 I4 = 4 = 4
3
V6
I5 = I4 + I 3 = 6 V5 = I5 4 = 24 V6 = V5 + V4 = 36 I6 = = 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
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;
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);
end;
v(8) = i(8)*r(8);
i(9) = v(9)/r(9);
Is = i(9) + i(8);
I1new Isnew 1
(b) By the proportionality property: old = old I1new = 200mA = 11.77mA
I1 Is 16.9877
By inspection:
Vout_a = 300/90012 = 4V
Vout_b = (300||600)60m = 12
_ 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
VL12V = 0.25 + 0.2 + 0.05 12 =
2
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,
1 3
VL3A = 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
Va_1 = 40V
Finally, when VS1 is active, the left part of the circuit consists only of resistances, so Va_3 = 0. Vout is given
Vout_3 = 90/100VS
Va = 40 15 + 0 = 25V
Finally, P = V2 /R = 22.5W.
Solution 5.10
Due to 3A source:
Due to 1A source:
iout = 2/3A again by current division. So, vout_2 = 4/3V.
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:
Vout_2 = 900/(900+100)20 = 18 V.
We have three resistances in parallel with a resistance equal to 90 . So, Vout_3 = 0.190 = 9 V.
Solution 5.12
Thus, this circuit is identical to the inverting amplifier studied in Chapter 4. So,
Rf
Vout _ 0 = V0
R0
Similarly, when each of the other sources is activated, the circuit will be an inverting amplifier. So,
Rf Rf Rf
Vout _1 = 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.
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
Solution 5.14
(a) When VS2 is deactivated. The circuit looks like two inverting amplifiers in cascade.
(b) Similarly, when VS1 is zero, V1 is zero because the first inverting amplifier has zero input. Thus, the
*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,
s3 R
Vout = Vs3 = Vs3 = 2 V
R
(d) By superposition,
s1 s2 s2
Vout = Vout + Vout + Vout = 2 V
Solution 5.16
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
Vout = aIs1 + bV s2
5 = 0 + b10 b = 0.5
1 = a10 + 0 a = 0.1
So,
Solution 5.18
5a + 10b = 15
2a + 5b = 10
Vout = 5 + 20 = 15V
Again,
Vout = aIs1 + bV s2
Substituting the result of the first measurement into this equation yields:
0 + 10b = 55 b = 5.5
First of all,
7a + 3b = 1
9a + b = 3
We know that the output is going to be a linear combination of the three inputs:
50ma 2b + 5c = 13
0 + 3b + 5c = 2
0 + 2b + 4c = 0
0.05 2 5 a 13 a 100
0 3 5 b = 2 b = 4
0 2 4 c 0 c 2
Solution 5.22
Vout = AIs1 + BV s2 + CV s3
Substituting the measurements into this equation results in a system of three equations and three
30 10 3 2 1 A 11.5 A 150
3
2010 4 2 B = 27 B = 5.5
10103
3 1 C 14 C 4
The coefficient matrix is inverted, and both sides of the nodal equation are multiplied by it to obtain:
I1 = 0.0022Vs1 + 0.6296Is2
I3 = 0.0022Vs1 + -0.3704Is2
Solution 5.25
Invert the coefficient matrix and mujltiply both sides of the equation in the problem by this inverse matrix
to obtain
V1 0 0 0 3 2 1 4 I s1
V 0 0 0
2 3 1 1 4 0
V3 0 0 0 12 3 3 16 0
Ia = 0 0 0 4 1 1 5 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
By voltage divider:
Then, due to I2 :
I2 flows through the 20 ohm resistor in series with 6||3. Thus,
Vout_2 = 2I2
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
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 :
This voltage now divides between Vo1 and the 10 ohm resistance:
The parallel combination results in a 12 ohm resistance in series with the remaining two. Thus, by voltage
division:
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
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
A similar analysis of this resistive network can reveal that vout4 = 6is4 . Thus
(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.
Define two clockwise mesh currents: I1 in the bottom left loop and I2 in the top loop. The bottom right
and
6I2 I2 + (I2 I1 )3 = 0
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 ' +'
vout = I1 + i b = 8 + 26 = 34V
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
where the leftmost current is 0.25 with a resistance of 32. Similarly, the downward current is 0.75 and its
v1 v v
1m + 9m + 1 2 = 0
2k 2k
v2 v1 v v
+ 9m + 2 + 2 5m = 0
2k 9.6k 16k
These two equations can be solved using any method to obtain:
V22
P= = 0.6mW
9.6 10 3
Solution 5.33
Then
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
v1 v v
2 + 1 = 12m
500 500 500
v2 v v
1 + 0.032v1 + 2 = 80m
500 500 125
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
*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.
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
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
Rearranging,
3
va = Vs
7+2
Then,
4 5
vout = Vs v
7 7 a
4
vout = V
7 + 2 s
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, Vout_1 = 0
inspection:
Va_2 = 1/6Vt
Vout_2 = Vt
Va = [1/(6 )]Vs
Vout = Vs
6
Solution 5.40
The first step is to replace the dependent source with a temporary independent source. Then, superposition
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
V1 = 4/9Vs
Va_1 = 16/45 Vs
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
16 16
Va = Vs + Va
45 45
1
? Va = V
(4516 ) s
Vout = Vt = Va
and 16
Vout = Vs
45 16
Solution 5.41
By KVL
Rearranging
1
Vg = Vs
0.1 + 1
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
Vout 20 + 4 2
= = 14
Vin 1 + 0.5 2
V V
out a1 + b0 + b1 = out
Vin Vin
Plugging in the data yields three equations in three unknowns which in matrix form are:
0 1 0 a1 264
154 1 1 b = 154
0
168 1 2 b1 84
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
SOLUTION 6.1. (a) Voc is found by removing RL and doing voltage division.
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
VOC = 6V
40 + 0.1A 60 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.
ISC can then be found using the VOC/RTH relationship as 0.58+20sin(50t) mA.
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
c d
SOLUTION 6.10.
RTH = [(18K ||9K) + 66K]||36K = 24K
Thevenin Probs, 7/24/01 - P6.5 - @R.A. Decarlo & P. M. Lin
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:
Solving we obtain, Vtest = 100itest + 2Vs / 3. From eq. 6.10 RTH = 100 , voc = 2Vs / 3.
C D
2
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
0.9
Power/Vs in mW
0.8
0.7
0.6
0.5
0.4
0.3
0 20 40 60 80 100 120 140 160 180 200
Load resistance in Ohms
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
1 3
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/ 30 + 1/ 20 1/ 20 1/ 30 vtest itest
1/ 20 1/ 30 + 0.4 1/ 20 0.4 2 / 30 v = 0
b
1/ 6 1
5/ 6 vc 0
b c
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
5
Power in Watts
1
0 5 10 15 20 25
Resistance in Ohms
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 .
1/ 5
= 1 8 = 5/ 6V for k = 2/3, and ISC = VOC / RTH = 1/19 A .
Thus solving yields VOC
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.
Vs 300ix
Introduce a test source, vt, and get the following equations: it + ix = ix 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.
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
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 .
1 + 1 V 16 = I
40 120 C 40 test
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
18
Hence, Voc = 18 V and Rth = 15 . Thus i = = 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
+
v +
2
- vs
1
-
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 0 VA 1 0
0 1/ 2k 1/ 6k
1/ 3k VC 1 1m
= i +
0 1/ 6k (1/ 6k + 1/ 15k) 1/ 2k VD 0 s 0
0 1/ 3k 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
SOLUTION 6.27. (a) From the previous problem VCB = 4Vs . Thus by voltage division,
180||162
(b) This changes the voltage division at the output, thus VAB / Vs = 4 = 3.24 .
(180||162) + 20
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,
2.5Vs1 2Vs 2
Isc = = 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
V
V0 + 2I = 0.8V0 + 2 0 = 0.78V0 = 1.56Vs2 3.12Vs1
80
Voc =
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
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:
2 0.4 0.24
10 1 0.2
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
R2
Eo / 2 = Voc
RTH + R2
and RTH = R2 .
(b) Now
Rm
Eo = Voc
Rm + RTH
Rm || R2
Eo / 2 = Voc
(Rm || R2 ) + RTH
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 .
R2 Rm Rs Rm
SOLUTION TO 6.39. Using the relation developed in problem 6.37 RTH = = 5k
Rm + Rs R2
R2 Rm
RTH = = 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 1 1/ RTH 0.1m
0.7 =
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
= 22.5mA. The power is now obtained from P = (isc / 2 ) RL . Thus
2
isc = 60mA
(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
The value for the load resistance is 16, and the power is P = oc / RL = 16W .
V
2
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
180
Voc = 1 18 + 40 = 54 V. Therefore, for maximum power transfer
180 + 20
6 1
GL + 2GL + 3GL = 6GL = =
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
P180 = = 4.05 watt.
180
Since the terminal voltages are the same, the absorbed power is inversely proportional to the resistance.
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:
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
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
P = (Voc / 2 ) / RL = 416.7mW .
2
(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;
RL= 1./((1./R)+1./(R+300));
P=(10.*(RL./(RL+60))).^2./RL;
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 G12 0 VN1 IN1
G21 G22 G23 Vm = Im (*)
0 G32 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
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.
2 1 0 0 V1 2
1 0.5 0.5 0 V 0
2 =
0 0.5 2.5 1 V3 0
0 0 1 2 V4 4
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.
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:
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
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
VOC = 15 10
15V 30
= 14V
15 + 60 30 + 30
(b) Doing the same for the circuit right of terminal A-B. RTH = [(30||15) + 10]||20 = 10 and
VOC = 7.5 20
7.5V 30
= 5V .
15 + 15 30 + 30
SOLUTION TO P6.55. The proof is based on superposition. Let us consider the figure below where N1
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 6.60. 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
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 C-
B; 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
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:
diL (t)
v L (t) = L
dt
The calculations for vL(t) for t = 0s to 5s are summarized in the following table:
wL (t) =
1
2 ( )
0.5iin2 (t) = t 2
wL (t) =
1
2( ) 1
0.5iin2 (t) = (4 2t) 2 = t 2 4t + 4
4
1 t
iL (t) =
0.5 0
d = t 2 in the time interval 0s < t 2s
1 t 1 t
iL (t) =
0.5 0
v L ( )d = iL (2) +
0.5 2
2dt = 4 + 4 t2 = 4t 4
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 R. A. DeCarlo, P. M. Lin
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
It is assumed that i1 (0) = i 2 (0) = 0 A . In the interval 0s < t 1s, vin (t) = 10V . Hence, in this interval
1 t 1 t
i1 (t) =
0.5 0
v in ( )d =
0.5 0
(10)d = 20t
1 t 1 t
i2 (t) =
0.25 0
v in ( )d =
0.25 0
(10)d = 40t
1 t 1 t
i1 (t) = i1(1) +
0.5 1
v in( )d = 20 +
0.5 1
(5)d = 10t 10
1 t 1 t
i2 (t) = i2 (1) +
0.25 1
v in( )d = 40 +
0.25 1
(5)d = 20t 20
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 R. A. DeCarlo, P. M. Lin
It is assumed that i1 (0) = i2 (0) = 0 A. In the interval 0s < t 1s, vin (t) = 10t . Hence, in this interval
1 t 1 t
i1 (t) = i1(0) +
0.5 0
v in( )d =
0.5 0
(10 )d = 10t 2
and
1 t 1 t
i2 (t) = i2 (0) +
0.25 0
v in( )d =
0.25 0
(10 )d = 20t 2
In the interval 1s < t 3s, vin (t) = 10t 20 . Hence, in this interval
1 t 1 t
i1 (t) = i1(1) +
0.5 1
v in( )d = 10 +
0.5 1
(10 20)d = 10t 2 40t + 40
1 t 1 t
i2 (t) = i2 (1) +
0.25 1
v in( )d = 20 +
0.25 1
(10 20)d = 20t 2 80t + 80
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 R. A. DeCarlo, P. M. Lin
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
t
1 t 1
iin (t) = iin (0) +
[(0.75) (1.5)] 103 0
v s ( )d = v s ( )d A
0.5 10 3 0
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:
d
t
0.8 + 0.6
vout (t) = (0.8 + 0.6) 10 3 d
[6i (t)] = 0.5 6 dt v s ( )d = 16.8v s (t) V
dt in
0
1 t
iin (t) =
0.5 10 3 0 s
v ( )d = 2t(A), iout (t) = 6iin (t) = 12t(V )
1 t
iin (t) = iin (1) +
0.5 10 3 1
v s ( )d = 2 + 2 2 1t = 4t 2(A)
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:
SOLUTION 7.7. For 0 t < 2 s, using the inductor v-i relationship, we have
t
1 2 8
iin (t) =
0.5 v in ( )d =
0.25
[ cos(0.25 )]t0 = [1 cos(0.25t)] A
0
16
W0.5 (t) = [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
0.25 in
iin (t) = i0.5 (t) + i0.25 (t) = i0.5 (t) + v ( )d
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
2 ( 2
0.5 i0.5 2
(t) i0.5 )1 2
(2) = 0.5 i0.5
2
64
(t) 2 J
Further
16
i0.25 (t) = cos(0.25t) A
in which case the energy stored over the interval [2,t] is
W0.25 (2,t) =
1
2 (2
0.25 i0.25 2
(t) i0.25 ) 1
(2) = 0.25i 20.25(t) J
2
SOLUTION 7.8. Let the 5mH inductor be L1 and the 20mH inductor be L2 .
For 0 t < 3ms,
t t
1 1
iL1(t) = iL1 (0) + v s ( )d = 3 12cos (500 )d = 4.8sin(500t) mA
L1
0 5 10 0
L&C Probs, 11/15/01 P:7-9 R. A. DeCarlo, P. M. Lin
t t
1 1
iL 2 (t) = i L2 (0) +
L2 v s ( )d =
20 10 3 12cos (500 )d = 1.2sin(500t) mA
0 0
For t 3ms,
For 0 t < 3ms, the energies stored in the inductors are given as follows:
For t 3ms, the energies stored in the inductors are given as follows:
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
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))
iC (t) = C
dt C
d
10 10 3 cos(1000t) = C (sin(1000t))
dt
10 10 3 cos(1000t) = 1000C cos(1000t)
Therefore, C=10F.
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
d(v in (t))
iC1(t) = 2 103 = 102 A
dt
and
d(vin (t))
iC 2 (t) = 6 10 3 = 3 10 2 A
dt
Using KCL, iin (t) can be computed as
The calculations for iC1 (t), iC2 (t) and iin (t) for t = -1s to 6s are summarized in the following table:
Time Interval d/dt (vin (t)), Vs-1 iC1 (t), mA iC2 (t), mA iin (t), mA
-1s< t 0s 5 10 30 40
0s< t 1s 0 0 0 0
1s< t 3s -5 -10 -30 -40
3s< t 4s 20 40 120 160
4s< t 5s 0 0 0 0
5s< t 6s -15 -30 -90 -120
Below are the plots of iC1 (t), iC2 (t) and iin (t).
L&C Probs, 11/15/01 P:7-11 R. A. DeCarlo, P. M. Lin
Applying (7.11), the energies stored in the capacitor over the intervals [0,2] and [2,3] are given by
WC[0,2] =
1
2 [ ]
C vC2 (2) v C2 (0) = 9 J
WC[2,3] =
1
2 [ ]
C vC2 (3) v C2 (2) = 11 J
t t
1 1
vC1 (t) = vC1(0) +
C1 iin ( )d =
0.25 iin ( )d
0 0
t t
1 1
vC 2 (t) = v C2 (0) +
C2 iin ( )d =
0.1 iin ( )d
0 0
t
1 t 1
vC1 (t) =
0.25 o iin ( )d =
0.25
( 2 ) = 4 106 t 2 (V )
0
and
t
1 t 1
vC 2 (t) =
0.1 o iin ( )d =
0.1
( 2 ) = 10 10 6 t 2 (V )
0
1 1 t
(2 103 ) = 8 10 3 t 4(V )
t
vC1 (t) = vC1(1) + iin ( )d = 4 +
0.25 1 0.25 1
and
1 1 t
(2 103 ) = 20 10 3 t 10(V )
t
vC 2 (t) = v C2 (1) + iin ( )d =10 +
0.1 1 0.1 1
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:
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
t t
1 1
vin (t) = v in (0) + iin ( )d = 3 iin ( )d
C
0 0.5 10 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
1
(10 103)d = 20t(V )
t
vin (t) =
0.5 10 3 o
In the interval 1s < t 2s, iin (t) = 5(mA). Hence, in this interval
1
(5 103)d = 20 + (10t + 10) = 10t 10(V )
t
vin (t) = v in (1) +
0.5 10 3 1
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:
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
1
(10 10 3 )d = 10t 2 (V)
t
vin (t) =
0.5 10 3 o
In the interval 1s < t 3s, iin (t) = 10t 20(mA) . Hence, in this interval
1 t 3
vin (t) = v in (1) +
0.5 10 3 o
10 (10 20)d = 10 + 10(10 2 40 ) 1t = 10t 2 40t + 40 (V)
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:
d
iin (t) = C1 (v (t)) = 20 6 1500 cos(1500t) = 0.18cos(1500t) A
dt s
Then, we can find vout by applying (7.6),
t
1
0.5m in
vout (t) = vout (0) + 2i ( )d
0
2 0.18
= 10 + sin(1500t)
0.5m 1500
= 10 + 0.48sin(1500t) V
Part 3 Applying (7.11), the energy stored in the capacitor over the interval [0,t] is given by
WC1[0,t] =
1
2 [ ]
C1 vs 2 (t) v s2 (0) = 360sin2 (1500t) J
L&C Probs, 11/15/01 P:7-16 R. A. DeCarlo, P. M. Lin
t
= vout (0) +
1
0.5m 2 20
d
d
(v s ( ))d
0
t
= vout (0) + 0.08
d
d
(v s ( ))d
0
d
p(t) = v out (t) 2iin (t) = ( vout (0) + 0.08 v s(t)) 2 20 (v (t))
dt s
d
= 40( vout (0) + 0.08 vs (t)) (v (t))
dt s
W
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
1
WC (t) = 0.5m v out 2 (t) = 0.25( vout (0) + 0.08 v s (t)) 2 mJ
2
If we assume vout(0)=0V, then
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.
Part 2 The instantaneous power delivered by the dependent source is given by.
p(t) = 40( v out (0) + 0.08 v s (t)) (v s (t))
d
W
dt
If we assume vout(0)=0V, then
p(t) = 3.2 v s (t)
d
(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 R. A. DeCarlo, P. M. Lin
In the interval 0ms < t 1ms, v s (t) = 2 103 t(V ). In this interval
WC (t) = 6.4t 2J
WC (t) = 6.4( J )
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
t
1
vin (t) = v in (0) +
0.3 103 0.6 103 0
is ( )d
t
1
= (0) +
0.3 10 3 0.6 10 3
60 10 3 sin(100 )d
0
= 3 3cos(100t) V
Therefore,
d
iout (t) = (0.2 103 + 0.8 103) (10v in (t)) = 3sin(100t) V
dt
Below is a sketch of iout(t) vs time.
SOLUTION 7.19.
Part 1 Since Q = CV, the charge that resides on each plate of the capacitor = 10F 100V = 1mC
Part 2 Since V = Q/C, the required voltage = 1mC/5F = 200V
Part 3 Since V = Q/C, the required voltage = 50C/1F = 50V
Part 4 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
Hence
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,
t t t
1 1 1
vC (t) =
C1 is ( )d = vC (0) +
C1 is ( )d =
C1 s
i ( )d
0 0
=
12
3
15 10 500
e [
500 t
0 ] (
= 1.6 1 e 500t V )
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
( )
2
WC1 (0 < t < 2ms) = 0.0192 1 e 500t J
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:
2+ ms
qC1 (2+ ms) + qC 2(2 + ms) = qC1(2 ms) + qC2 (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
C1 15
vC (2+ ms) = vC (2 ms) = 1.0114 = 0.37927 V
C1 + C2 15 + 25
= 0.379 +
12
3
40 10 500
e 500 t
[
0.002 ] (
= 0.379 + 0.6 0.36788 e 500t V )
Hence the energy stored in the two capacitors over the interval [2+ ms, t] is
1 1
WCeq (2+ ms < t) = 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
1
WCeq (t) = 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 P:7-23 R. A. DeCarlo, P. M. Lin
d
v L (t) = L i (t)
dt in
(
= 2.5mH 200te10t + 20e 10t )
= 0.05e 10t (10t + 1)
[ ]0
t
= 20 103 0.1 e 10 0.01e 10
[ ]
= 2 103 te10t 0.1e 10t + 0.1 V
and
[ ]
= 0.05e 10t (10t + 1)+ 2000 te10t 0.1e 10t + 0.1
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
1
WL (t) = L[i (t)]2 = 0.5t 2e 20t J
2 in
[
C[vC (t)] = 2 te10t = 0.1e 10t + 0.1 kJ]
1 2 2
WC (t) =
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
d[is (t)]
v L (t) = 0.25 = 4cos(4 t) (V )
dt
Applying the capacitor v-i relationship it follows that
d[2v 2 (t)]
iC (t) = 0.25 = 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:
d
iS (t) = (10mF + 20mF) v (t)
dt s
= 30 103 (12te 5t 30t 2e 5t )
= 0.36te 5t 0.9t 2e 5t A
Part 2. Applying the v-i relationship for the 20mH inductor, it follows that
d
vout (t) = (20mH ) [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
1
WL (t) = (20mH )[12iS (t)]2
2
( )
2
= 1.44 0.36te5t 0.9t 2e 5t 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:
dvC (t)
iS (t) = (C1 + C2 )
dt
Using the v-i relationship for the capacitor C2 it follows that
dvC (t)
iC 2 (t) = C2
dt
dvC (t)
By replacing , it follows that
dt
L&C Probs, 11/15/01 P:7-26 R. A. DeCarlo, P. M. Lin
C2 2
iC 2 (t) = 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:
diL (t)
9iC 2(t) = (L1 + L2 )
dt
Using the v-I relationship for the inductor L2 it follows that
di2 (t)
vout (t) = v L2 (t) = L2
dt
diL (t)
By replacing , it follows that
dt
L2 2
vout (t) = 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:
t
1
vC 2 (t) =
C2 iC ( )d
t
1
vC1 (t) =
C1 iC ( )d
where iC(t) is the current through the capacitors C1 and C2 . Since C1 and C2 are connected in series:
t
1 1
vin (t) = v C1(t) + vC 2 (t) = +
C1 C2 iC ( )d
t
CC
It follows that iC ( )d = v in (t) 1 2 . Hence
C1 + C2
t
1 1 CC C1
vC 2 (t) =
C2 iC ( )d = 1 2 v in (t) =
C2 C1 + C2
v (t)
C1 + C2 in
C1
where the last equality follows by replacing vC2 (t) with 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.
(d) (e)
(f) (g)
Laeq = 1mH
Lbeq = 1mH + 1mH = 2mH
Lceq = 1mH || 1mH = 0.5mH
Ldeq = 1mH + 1mH + 1mH = 3mH
Leeq = 1mH + (1mH || 1mH) = 1.5mH
f
Leq = 1mH || (1mH + 1mH) = 0.667mH
Lgeq = 1mH || 1mH || 1mH) = 0.333mH
L&C Probs, 11/15/01 P:7-28 R. A. DeCarlo, P. M. Lin
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.
SOLUTION 7.36. Leq = (11mH) || (19.25mH) + 3mH = 10mH. Applying the voltage division formula,
3mH
v L1 (t) = vin (t) = 60tet mV
Leq
and
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).
Equivalently
t
1
vin ( )d = 1 1 iin (t)
0 +
L1 L2
Therefore
1 L1L2
t
1 L Lk
iL k (t) = v in ( )d = 1 k 1 iin (t) = iin (t) for k = 1,2
Lk
0 + L1 + L2
L1 L2
12 + 27 2
iL1(t) = iin (t) = 0.231e t mA
12 + 27 + 130
130 2
iL 2 (t) = iin (t) = 0.769e t mA
12 + 27 + 130
Also,
d
iin (t) = 60tet
2
vin (t) = Leq V
dt
The instantaneous energy stored in the 130-mH is given by
d d
vin (t) = Leq iin (t) = 30mH iin (t)
dt dt
The calculations for vin (t) for t = -2s to 7s are summarized in the following table:
Below are the plots for vin (t), iL1(t) and iL2(t).
L&C Probs, 11/15/01 P:7-32 R. A. DeCarlo, P. M. Lin
18
iL 2 (t) = i (t) = 90cos(300 t) mA
18 + 6 in
Part 2
Leq =1.5mH + 6mH || 18mH = 6mH
d
vin (t) = Leq i (t) = -0.679sin(300 t) V
dt in
Part 3 Instantaneous power delivered by the source is given by
The initial condition for the interval [n, n+1] for n even is:
n
1
Leq in
iin (n) = iin (0) + v ( )d
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
n
vin ( )d = 0
0
for n even. The initial condition for the interval [n, n+1] for n odd is:
n
1 1
iin (t) = iin (n 1) +
Leq vin ( )d =
0.425
(16 ) nn1 = 37.6 A
n1
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 R. A. DeCarlo, P. M. Lin
Part 3 By the current division formula, the current through the 0.9H inductor is given by
0.3
iL 0.9H (t) = i (t)
0.3 + 0.9 in
9.4(t 2n) A for 2n < t 2n + 1
=
9.4(2n + 2 t) A for 2n + 1< t 2n + 2
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
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 = (1mF || 2 mF) + (1.2mF || 2mF || 1.5mF) + (4mF || 2.6667mF || 3.2mF || 1.6mF) = 2.1667mF
SOLUTION 7.44.
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
When the switch is closed, the circuit in figure P7.45 can be represented by the following.
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.
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).
6 ||6
vC 3 (t) = vin (t) = 21(1 e 20t ) V
6|| 6 + 9
0.05 + 0.15
vC1 (t) = 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
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
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).
+ C2 in = (C1 + C2 ) in
dv in (t) dv (t) dv (t)
iin (t) = iC1 (t) + iC 2 (t) = C1
dt dt dt
Equivalently,
dvin (t) 1
= i (t)
dt (C1 + C2 ) in
It follows that
dv in (t) Ck
iCk (t) = Ck = i (t)
dt C1 + C2 in
which is the required current division formula.
L&C Probs, 11/15/01 P:7-39 R. A. DeCarlo, P. M. Lin
0.08
iC1(t) = i (t) = 80e 2t A
0.08 + 0.03 0.06 in
0.03 0.06
iC 2 (t) = i (t) = 20e 2t A
0.08 + 0.03 0.06 in
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
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 7.58. 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:
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:
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. The resultant
schematic is shown below.
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. The resultant
schematic is shown below.
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
Part 3 Since the general relationship between Req and Req * is given by Req = 1/Req *. We can write
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.
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
[ ]
(t) = C2* v *out (t) = L2[iL 2 (t)] = 57.6[cos(250t) 1]2 J
* 1 2 1 2
WC2
2 2
The equivalent inductance of the dual circuit (Leq *) = (1+2) || 1.5 + 2.5 = 3.5H.
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,
dv1(t) 1 dv s(t)
i1 (t) = 0.2 =
dt 15 dt
Finally, since the last capacitor is initially uncharged,
t t
1 2 dv ( ) 2
vout (t) =
0.5 i1 ( )d = s d = v s (t)
15 d 15
0 0
A sketch is given below.
SOLUTION 7.67. Following the method outlined in page 269 of the text, we require
1/200
1
vC (t0 )
C i 0( )d 14
0
With i0 () = 2A, vC(0) = 20 V, it follows that
1 1
20 14 2A s
C 200
C 1.667mF
Therefore, the standard capacitor value of 1.8mF should be chosen for this application.
L&C Probs, 11/15/01 P:7-47 R. A. DeCarlo, P. M. Lin
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
10
6
vc(t) in V
4 TextEnd
0
0 1 2 3 4 5
time in s
10
(b) The solution has the same general form as for (a),iC (t) = iC (0+)e t / = e t = 2e t mA =
5 10 3
vC (t)
.
R
1st Order Circuit Probs 11/26/01 P8-2 R. A. DeCarlo, P. M. Lin
-0.5
ic(t) in mA
-1
TextEnd
-1.5
-2
0 1 2 3 4 5
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.
t /Req C
SOLUTION 8.3. (a) The general solution form is vC (t) = vC (0)e t / = v C (0)e . Using the
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
xlabel('Time in s')
ylabel('Inductor Current')
0.16
0.14
0.12
Inductor Current
0.1
0.08
0.06
TextEnd
0.04
0.02
0
0 0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008 0.009 0.01
Time in s
(b) Here v L (t) = RiL (t) = 22.5e 500t V.
-5
Inductor Voltage in V
-10
-15
TextEnd
-20
-25
0 0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008 0.009 0.01
Time in s
22.5 500t
(c) Using linearity, for iL(0) = 50 mA, then v L (t) = e = 7.5e 500t V and for iL(0) = 250
3
250 500t
mA, then v L (t) = 22.5 e = 37.5e 500t V.
150
SOLUTION 8.6. Since = L / R, we can solve for L = 5 mH. Then solving iL (t) = iL (0)e t / for iL(0)
SOLUTION 8.7. (a) We desire to solve iL (t) = iL (0)e t / for iL(0) and R in = 0.08 /(R + 103 ).
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,
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
70
60
Capacitor Voltage (V)
50
40
30
20 TextEnd
10
0
0 0.2 0.4 0.6 0.8 1
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
40
Capacitor Voltage (V)
30
20
TextEnd
10
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Time in s
(b) With = -4, Rth = 90 , = RthC = 0.0225 s, and vC (t) = 50e 44.44t V.
1st Order Circuit Probs 11/26/01 P8-8 R. A. DeCarlo, P. M. Lin
8000
7000
6000
Capacitor Voltage (V)
5000
4000
3000
2000 TextEnd
1000
0
0 0.02 0.04 0.06 0.08 0.1 0.12
Time in s
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.
(b) Rth = 100 , = 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 .
(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
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.
3
By voltage division, vC (0+ ) = V and = 3RC. Hence
4 s
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
1333
iL (0 + ) = 0.02 = 0.0125 A
800 + 1333
For t > 0, the switch is opened and the inductor sees only the 800 resistor. Hence, = L / R = 25 sec
and
1st Order Circuit Probs 11/26/01 P8-11 R. A. DeCarlo, P. M. Lin
14
12
Inductor current in mA
10
4
TextEnd
2
0
0 20 40 60 80 100
Time in micro seconds
SOLUTION 8.18. (a) For t < 0 the applied voltage is constant and at t = 0, the capacitor is like an open
25k ||6.25k
circuit. By voltage division, vC (0+ ) = 30 = 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,
(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
0
0 2 4 6 8 10 12
Time in milli-secs
54
SOLUTION 8.19. (a) From Ohms law iL (0 + ) = = 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
720
v(t) = 60 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
720
v(t) = 690 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
0.6
0.4
0.3
0.2
TextEnd
0.1
0
0 5 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
80
(a) For t < 0, the capacitor acts as an open circuit. Using voltage division, vC (0+ ) = 80 = 30
80 + 133.3
V. For 0 t 60 ms, the time constant is 1 = Rth1C = 40 ms, and
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
20
15
10
TextEnd
0
0 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
Rth1 = = 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
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
20
15
10
TextEnd
0
0 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
10
7
Capacitor Voltage (V)
3 TextEnd
0
0 2 4 6 8 10 12
time in ms
t1= 12e-6;
t2=18e-6;
t3=21e-6;
L= 0.1;
rth1= 800;
rth2= 8000;
rth3=1600;
rth4= 32000;
tau1= L/rth1
tau2= L/rth2
tau3= L/rth3
tau4= L/rth4
tau1 = 1.2500e-04
tau2 = 1.2500e-05
tau3 = 6.2500e-05
tau4 = 3.1250e-06
% Compute initial inductor currents for each of the four time intervals.
il1= 100e-3;
il2=il1*exp(-t1/tau1)
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,
t
[ ( ) v ()]e
vC to+ RTH C
vC (t) = vC () + 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 R. A. DeCarlo, P. M. Lin
t
2
vC (t) = 10 + [10]e = 10(1 e 0.5t ) V.
t
+ 2
(b) With vin(t) = 0 and vC (0 ) = 5 V, vC (t) = 5e = 5e 0.5t V.
(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
0.4
Capacitor Current (mA)
0.3
0.2
TextEnd
0.1
0
0 2 4 6 8 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,
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
t
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
t
0.75RC
vC (t) = Vs1e . The zero order response, the response with 0V initial condition to a forced
t
0.75RC
voltage, is vC (t) = Vs 2 Vs 2e .
SOLUTION 8.25.
For RL circuits with a forced current, equation 8.19b is used as a general solution:
[ ( ) i ()]e
R th
(t t o)
iL (t) = iL () + iL to+ L
L .
[( ) ]
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
0.08
0.04
0.02
-0.02 TextEnd
-0.04
-0.06
0 5 10 15 20
Time in ms
( )
(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.
15
Inductor Voltage (V)
10
5
TextEnd
0
0 5 10 15 20
Time in ms
(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
iL (t) = 0.5I s2 + 0.5[ Is1 Is 2 ]e . L
(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
0.4Rt
iL (t) = 0.5I s1e 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 8.28.
(a) The Thevenin resistance for this configuration is Rth = 1000/ /1000 = 500 and = RthC = 0.25 s.
Hence vC (t) = vC (0+ )e t = 15e 4t V is the zero-input response.
(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
%Problem 8.29
%RTH= (60||120)+120
%tau=L/RTH
RTH=1/(1/60+1/120)+120;
tau=0.2/RTH;
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
240
Current in mA
220
200
180
160
140
120
0 1 2 3 4 5 6 7
Time in seconds -3
x 10
*S OLUTION 8.30.
50
45
Capacitor voltage (V)
40
35
30
25
20
0 50 100 150 200 250 300 350 400
Time in msec
*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)')
1st Order Circuit Probs 11/26/01 P8-24 R. A. DeCarlo, P. M. Lin
80
70
50
40
30
20
10
0
0 0.2 0.4 0.6 0.8 1
Time in sec
*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
1st Order Circuit Probs 11/26/01 P8-25 R. A. DeCarlo, P. M. Lin
80
70
50
40
30
20
10
0
0 0.2 0.4 0.6 0.8 1
Time in sec
80
70
60
50
40
30
20
10
0
0 0.2 0.4 0.6 0.8 1
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;
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;
plot(t,vct);
grid;
xlabel('time in seconds');
ylabel('Volts');
20
15
10
5
Volts
-5
-10
-15
-20
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04
time in seconds
SOLUTION 8.33.
(a)
%Problem 8.33
%(a)
300
290
280
270
260
Current in mA
250
240
230
220
210
200
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
time in seconds 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
0.7
0.5
0.4
0.3
0.2
0.1
0
0 2 4 6 8 10 12 14 16 18
Normalized time in terms of tau
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
20
Volts
15
10
-5
0 0.5 1 1.5 2 2.5
Time in seconds
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;
1st Order Circuit Probs 11/26/01 P8-29 R. A. DeCarlo, P. M. Lin
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
2
Volts
-1
-2
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
time in seconds -4
x 10
(c)
1st Order Circuit Probs 11/26/01 P8-30 R. A. DeCarlo, P. M. Lin
(d)
SOLUTION 8.37.
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.
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
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
(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;
xlabel('time in seconds');
ylabel('Volts');
250
200
150
Volts
100
50
0
0 0.002 0.004 0.006 0.008 0.01 0.012
time in seconds
SOLUTION 8.43.
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
10
Current in mA
5
0
0 0.5 1 1.5 2 2.5 3 3.5 4
time in seconds 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
1 t
[ ]
t
vout (t) = (0.25RCe0.25 d ) = 0.5 e 0.25 = 0.5e 0.25t
2RC 0 0
SOLUTION 8.45.
SOLUTION 8.46.
These are two integrator in cascade. Using 8.26, the output of the first stage is
1 t
RC 0
vin ( )d = 2[cos(50 )]t0 mV = 2cos(50t) 2 mV. Using the same equation again,
t
1 t 2
RC 0
vout (t) = (2cos(50 )d = 10 sin(50 ) mV = 2cos(50t)mV
50 0
1st Order Circuit Probs 11/26/01 P8-36 R. A. DeCarlo, P. M. Lin
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);
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')
1st Order Circuit Probs 11/26/01 P8-37 R. A. DeCarlo, P. M. Lin
tsat =
8.4730e+00
-
(c) vC(0 ) = 0. Observe saturation at about 6.4 seconds.
+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.
1st Order Circuit Probs 11/26/01 P8-38 R. A. DeCarlo, P. M. Lin
+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
SOLUTION 8.48.
Since the op-amps do not load the first stage of the circuit, we can find its transfer function for the op-
v (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 10G(1+ K)
vout (t) = G(1+ K) 10sin ( )d = (cos( t) 1) , where G must be negative.
0
d(10sin( t))
(b) Using the same reasoning, vout (t) = G(1+ K) = 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.
1 t
vC 2 (t) = v C2 (0 + ) +
C2 0
i( )d
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
*S OLUTION 8.51. (a) After the switch closes, we have the circuit shown below.
t /RC eq v (0 + ) t /RCeq
iR (t) = i R (0+ )e t / = iR (0+ )e = 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
t t t
+1
vC1 (t) = vC1(0 ) + iC1( ) d = 2 iR ( ) d = 2 4 e 4 d = 1+ e 4t V
C1
0 0 0
and
t t t
+1
i ( ) d = 0 + iR ( )d = 4 e 4 d = 1 e 4t V
C2 C 2
vC 2 (t) = v C2 (0 ) +
0 0 0
+
(c) The energy stored at t = 0 for each capacitor is:
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
+
Hence the decrease in stored energy from 0 to is 1 J.
+
(d) Computing the energy dissipated in the resistor over [0 , ) is
t t t
+ 1 1
vC1 (t) = vC1(0 ) + iC1( ) d = 1 iR ( ) d = 1 e 10 d = (9 + e 10t ) V
C1 10
0 0 0
and
t t 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
0 0 0
+
(c) The energy stored at t = 0 for each capacitor is:
and
WC2 (0 + ) = 0.5C2vC2 2 (0+ ) = 31.25mJ
Further at t = ,
2
WC1 () = 0.5C1vC1 () = 405mJ
and
WC2 () = 0.5C2vC2 2 () = 101.25mJ
+
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 P8-42 R. A. DeCarlo, P. M. Lin
(0 + ) 2t /RCeq (0+ ) RCeq
]
vCeq
2 2
v Ceq 2t /RCeq
WR (0,) = R
iR2 ( )d = R e d = e
R
2 R 2 0
0 0
(0+ )
2
Ceqv Ceq
= = 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 .
4 + + 4 (v v s ) +
dv v d (v v s ) = 0 (*)
dt 4 dt 2
Equivalently,
dv 3v dv v
8 = +4 s + s (**)
dt 4 dt 2
dv 3 1
= v+ (***)
dt 32 16
1st Order Circuit Probs 11/26/01 P8-43 R. A. DeCarlo, P. M. Lin
0+ 0+ 0+ 0+
dv 3 dv s
8 d = v d +4 d + 0.5 v s d
d 4 d
0
0 0 0
Since the integral of a finite integrand over an infinitesimal interval is zero, we have equivalently,
0+ 0+
dv dv s
8 d = 0+ 4 d +0
d d
0 0
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.
[ ]
(b) x(t +0 ) = K1e t 0 / + F , and K1 = x(t0+ ) F e t 0 / .
SOLUTION 8.57. (a) From the graph, the initial and final values are 0 and 80 V respectively. That sets
R2
the following constraint, 100 = 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 .
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 = 20ms . Thus
L /(R1 || R2 ) = 20ms, and R2 = 0.25m.
(b) R1 stays the same, L = 20ms(R1 || R2 ) = 20H .
tau1= 20*1;
tau2= 1;
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');
5
Volts
0
0 20 40 60 80 100 120 140
time in seconds
SOLUTION 8.60. When the switch is in position A, = 18.18ms . In position B it is = 99.5us . Using
the elapsed time formula, find ta, when the output voltage reaches 90 V.
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
Rsolar + Rstore [ ]
1 e (Rstore+Rsolar )t /L A .
(c) In this time period the circuit reduces to an Lsotre inductor in series with a Rstore +R 1 resistor.
(d)
Vsolar
iL (T1 ) =
Rsolar + Rstore .
(Rstore+ R1)(t T 1)/L
iL (t) = iT e A
1
(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
(g)
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 R. A. DeCarlo, P. M. Lin
-3 dv s (t)
SOLUTION 8.63. (a) Since RC = 10 s, from equation 8.25, v a (t) = RC = cos(1000t) V.
dt
dv a (t) Rf
Hence vb (t) = RC == 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
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
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
and K = H(0), i.e., the dc gain of the first order low pass system. This leads to
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
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
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)
( K cos())2 + ( Ksin())2 = K 2 = A2 + B2
K sin() B
= tan() =
K cos() A
in which case
B
= tan1
A
with due regard to quadrant.
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.
(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 ,
K1 es2 T K es2 T
= s T implies ln 1 = ln s T = ( s2 s1 )T
K2 e 1 K2 e 1
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.
d 2 vC (t) 1 1
+ vC (t) = V
dt 2 LC LC s
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
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
CVs t V t
iL (t) = sin = s sin A
LC LC L 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+) 1
i'L (0+) = L = =0= B
L L LC
d 1 1 B
i'L (0+) = Acos t + B sin t + XF =
dt LC LC t =0 LC
Hence B = 0. Therefore,
1
iL (t) = Is 1 cos( t)
LC
diL (t) d 1 L 1
vC (t) = vL (t) = L = LIs 1 cos LC t = Is C sin LC t
dt dt
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) .
1 2
(c) Maximum value of vC when all energy is in capacitor must satisfy 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
d 2 vC 1
2 + LC vC = 0
dt
with corresponding response
1 1
vC (t) = A cos t +B sin t
LC LC
vC ( 0 ) = A = 0
and
1 i (0+) iL (0+)
vC ' ( 0 )= B = C =
LC C C
in which case
I L
B= S
2 C
Therefore,
I L 1
v C (t) = s sin t
2 C LC
1
Natural frequency is = 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
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+)
vC ' ( 0 )= B = C = L =0
LC C C
Hence
vC (t) = 20 cos(10,000t ) mV
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
iL (t) = Acos(t) + Bsin(t)
From the initial conditions
iL ( 0 )= A = 1
and
v (0 + )
iL ' (0 + ) = B = C =0
L
iL (t) = cos(80t) A
vC (0) = vC ( 0+) = 5 V
iL (0) = i L ( 0+) = 1
(b)
1 1
vC (t) = Acos t + B sin t
LC LC
where
vC (0+) = 5 = A
1 i (0+) 1
vC ' ( 0+) = B = C = B=
LC C C 2
vC (t) = 5cos(2t) sin(2t) V
2
vC (t) = K cos(2t + )
B
and from equation 9.17b K = A 2 + B2 and = tan1 in which case
A
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
Hence, K = 5.24 and = 17o.
(c)
4
Capacitor Voltage (V)
-2
TextEnd
-4
-6
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Time in s
1
SOLUTION TO PROBLEM 9.15. As before, = = 2 40 in which case L = 0.158
LC
H.
Now,
100
vC (0) = vc (0+) = 25 = 20 mV
125
and
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
vC (0+) = K1 + K2 = 1
i (0+)
vC ' ( 0+) = s1K1 + s2 K2 = K1 4K2 = C =0
C
s2 s1
K1 = = 1.3333 and K2 = = 0.3333 .
s2 s1 s2 s1
Finally,
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')
-0.1
-0.2
Capacitor Voltage in V
-0.3
-0.4
-0.5
-0.6
-0.7
TextEnd
-0.8
-0.9
-1
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
Time in s
(b)
K1 + K2 = 1
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
0.8
0.6
Capacitor Voltage in V
0.4
0.2
-0.2
-0.4
TextEnd
-0.6
-0.8
-1
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
Time in s
v (0+)
iL (0+) + iC (0+) + C = 0.1 + iC (0+) + 0.15 = 0
66.667
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
6
Capacitor Voltage in V
-2 TextEnd
-4
-6
0 0.005 0.01 0.015 0.02 0.025
Time in s
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:
R 1
s2 + s+ = s2 + 5s + 4 = (s + 4)(s + 1) = 0
L LC
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
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 .
dvC 2 4t 2 t
iL (t) = iC (t) = C = 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:
The roots are repeated and: s1, s2 = 10,000. The form of the response is:
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
Hence s1,s2 = 5000 j8660.25. The form of the new solution is:
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,
(c)
2.5
1.5
1
iL(t), in A
0.5 TextEnd
-0.5
-1
0 0.5 1 1.5 2 2.5 3 3.5 4
time in s -3
x 10
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 =
-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
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
v C (0+)
i (0+) iL (0+) +
vC '(0) = 4000K1 1000K 2 = C = R
C C
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')
7
Inductor Current in A
3
TextEnd
2
0
0 0.5 1 1.5 2 2.5 3 3.5 4
Time in s -3
x 10
i (0+ ) iL (0 + ) vC (0 + ) / 2
(b) vC (0) = 10 V and vC '(0 + ) = 800 = C = .
C C
(c)
By simple KCL,
1 dv C v C
iL (t) = = 5e 20t + 150 t e 20t A
80 dt 2
Req 1
s2 + s+ =0
L 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
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
-1
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05
vC (0) = K1 = 5
i (0+) iL (0+)
vC '(0) = s1K1 + K2 = 200K1 + K 2 = C = = 4000
C C
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
10
0
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05
iL (0) = iL (0+) = 0
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
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
Similarly,
iL (t) = e 10t ( A cos(5t) + Bsin(5t)) A
Applying initial conditions,
iL (0) = A = 0
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
SOLUTION TO PROBLEM 9.26. The series RLC circuit has characteristic equation
R 1 10 1
s2 + s+ = s2 + s + = s2 + 2 s + 2 + 2d = 0
L LC L LC
10
From the given response, = 10 = which implies that L = 0.5 H. Further,
2L
1 2 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
(
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) )
1 1 1 1
s2 + s+ = s2 + 4 s + = s2 + 2 s + 2 + 2d = 0
RC LC 10 C LC
1
Since = 2 = 1000, C = 0.1 F. Further,
10 4 C
1 107
= = 2 + d2 = 10 6 + 16 2106
LC L
implies that
(b) From the figure and the above calculations, vC(0) = 10 V and
iL (0) = vC (0) /10 4 CvC' (0) = 0.001 + 5000 0.1 106 = 0.5 10 3 A
2
60 4
= = 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:
1 1
s2 + 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
vC (t) = (K1 + K 2t)e s1t = (K1 + K2 t)e 10 t V
4
From ICs,
vC (0) = K1 = 0
i (0+) iL (0+)
vC '(0) = s1K1 + K2 = K 2 = C = = 2 105
C C
Hence
vC (t) = 2 10 5 te10 t V
4
(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
i (0+) i L (0+)
vC '(0) = 1000A + 9950B = C = = 2 10 5
C C
in which case B = 20.1. Thus
(c)
R = 87/17;si = roots([1 1/(R*C) 1/(L*C)])
si =
-3.6328e+04
-2.7527e+03
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
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
vC (t) = (K1 + K 2t)e 250t V
From IC's,
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
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
(b) Here, we have a series case: the indicated behavior occurs when the resistance
causes the circuit to be critically damped, i.e.,
2
R 4 2
L LC = R 144 = 0
Thus R = 12 and
R0 15
R = 12 = 5 =
1+e t 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
1 1
s2 + 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
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
1 1
s2 + s+ = s2 + 500s + 40000 = 0
RC LC
A = [1 1;-100 -400];
b = [2;0];
K = A\b
K=
2.6667e+00
-6.6667e-01
Hence,
8 100t 2 400t
iL (t) = e e 1 A
3 3
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
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
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
R 1
s2 + 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
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.
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,
L
Rth = 40 + R = 0.5 = 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
v R (0 + ) = 40 = 8 V.
10 + 40
+
To compute the derivative of vR at 0 , consider that
d 40 vC (t)
(v R (t)) = (10 i R (t)) = 10
d d d iC(t)
= 0.2 ( vC (t) ) = 0.2
dt dt dt 50 dt C
Hence
dv R (t) iC (0 + )
= 0.2 = 4000iC (0+ )
dt t =0 + C
But
40
iC (0 + ) = i L ( 0+ ) = 0.8 0.8 = 1.6 A
50
Therefore,
dv R (t)
= 4000iC (0 + ) = 6400 V/s
dt t =0 +
(c) Since the circuit is critically damped the roots of the characteristic equation are
1
s1,2 = = 200
2RthC
According to table 9.2 for t 0,
v R (t) = ( K1 + K2t )e 200t + X F
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
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.
in which case
vC (0) = K1 = 0
Voc
i (0+)
iC (0+) Rthcrit L
vC '(0) = s1K1 + K2 = K 2 = =
C C
Hence
10.8 50
K2 = 6 = 2.16 105
10
Therefore
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
R 1 24 1 5
s2 + s+ = s2 + s+ = s2 + 120s + = 0
L LC 0.2 0.2C C
20
For a critically damped response, 120 2 = . 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,
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,
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
1 1 1 1
s2 + s+ = s2 + 4 s + = s2 + 2 s + 2 + 2d = 0
RC LC 10 C LC
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
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
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
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
[ ( ) ( )]
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
1 1 1 1
s2 + s+ = s2 + 4 s + = s2 + 2 s + 2 + 2d = 0
RC LC 10 C LC
[ ]
vC (t) = e 1000t A cos(103 t) + Bsin(10 3 t) 5 V
vC (0) = A 5 = 5 A = 0
and
i (0+) 0.5
vC '(0) = 1000A + 1000B = 10 3 B = C = 3 = 10 4
C 0.05 10
Hence, B = 10 and
SOLUTION TO PROBLEM 9.44. (a) The circuit is a driven series RLC. Hence
R 1
s2 + s+ = s2 + 2 s + 2 + 2d = 0
L LC
(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 (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
SOLUTION TO PROBLEM 9.45. This circuit is the same series RLC as problem 44. Note
that at t = , vC() = 2.4 V. Hence
SOLUTION TO PROBLEM 9.46. For all three cases, assuming iL is pointing downward,
Lastly, all three cases are for a parallel RLC whose characteristic equation is:
1 1
s2 + 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
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
vC (t) = (K1 + K 2t)e 250t V
From IC's,
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
i (0+)
vC '(0) = 250A + 250B = 250B = C = 7500
C
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
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
dsolve('D2y+20*Dy+75*y=0,y(0)=10,Dy(0)=-100')
ans =
5*exp(-5*t)+5*exp(-15*t)
10
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
For t = 0-, there is no source present nor has there been a non-zero excitation. Hence,
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
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
A = [1 1;-15 -5];
b= [-1; 7.5];
K = A\b
K=
-2.5000e-01
-7.5000e-01
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:
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
(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,
(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 (1+) = A = 0.1
and
v (1+)
iL' (1+) = 5A + 21.794B = 0.5 + 21.794B = L = 30
L
Here B = 1.35. Hence
*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
It follows that
v L (0 + ) = 5 10 1 = 5 V
Hence
50 + K2 = 50
or K2 = 0. Finally
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
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
1
0.8
0.6
0.4
0.2
-0.2
-0.4
-0.6
-0.8
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
(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:
(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
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
dt [
= 10 3 (150 0.75)e 150t + (50 11.25)e 50t
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,
(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.
(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)
dvC (t)
iL (t) = 1.5 iC (t) = 1.5 C = 0.16667e 150t 0.83333e 50t + 1.5 A
dt
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,
(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
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
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
in which case
vC (t) = (20 + 8000t)e 400t + 40 V
dvC/dt = 2 iL vC
For t between 0 and 1, we have a parallel RLC circuit, with Req being the parallel
combination of the two 21.1333 resistors.
iL(0+) = A + 1 = 0 A = 1
and
v (0+)
iL' (0+) = 3A + 25B = 3 + 25B = L =0
L
Hence, for 0 t < 1,
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.
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
1.4
Inductor current in A
1.2
0.8
0.6
TextEnd
0.4
0.2
0
0 1 2 3 4 5 6
Time in s
vC = 0.707v C + 0.707iL
iL = 0.707vC 0.707iL + 0.707v in
At t = 0,
vC(0-) = vC(0+) = A + 0.5 = 0 A = 0.5
B = 0.5
Thus
vc(t) = e-0.707t[0.5 cos(0.707t) 0.5 sin(0.707t)] + 0.5 V
Hence
iC = iin vC/2 iL
vL = vC 12iL
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,
Therefore
vC (t) = 0.28573e 3.5t 2e 2t + 1.7143
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
vC1(0) = 2 K1 + K2 = 2
and
' i (0) 2(v C 2 (0) vC1(0)) 2vC1 (0) 8 8
vC1 (0) = C1 = = = 0 = 3K1 K2
C C C
Hence
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 (1)
dt RbC Ra C
Now we apply the formula of (*) to the second stage of our given op amp circuit to
obtain:
dvout v v
= out 1 (2)
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
= 1 s (3)
dt RC1 R1C1
The equations (2) and (3) form a coupled set of state equations which we can write as
1
1
RC2 vout 1
0
out
v RC2
v = 1 v1 + v
1 0 s
1 1
R C
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
where X f = 10 V by part (b). The problem states that the capacitor voltages are initially
zero. Hence
*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:
1
v1 out = 0
v1 d v
+ 10 4 ( v1 vout ) +
100 dt 100 A
Equivalently
dv1 dvout 100
= v 200v1 (*)
dt dt A out
1 1 v1 200 100 / A v1
0 =
1 vout 100A 100 vout
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
2 + (300 100A ) dt + 10 vout = 0
4
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
2 + 1.02 108 C 2 + 1012 vC 2 = 108 i
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.
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
0.8
0.6
vout in V
0.4 TextEnd
0.2
0
0 0.2 0.4 0.6 0.8 1
Time in s 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
dvC 3 3
= + v 3iL
dt 2 RN C
diL 1
= v 4i L
dt 3 C
d 2vC 3 3 dvC 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
dt 2 RN dt RN C
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 .
i1(t) = Im sin(t + ) A
(R1 R2)/(R1R2C) = 0
Im = 0.01
(a)
0 = 1/[(500)(1)] = 2 k-rad/s
v1(0) = A = 5
2
Voltage v1 in V
-2 TextEnd
-4
-6
-8
0 0.2 0.4 0.6 0.8 1
Time in s -3
x 10
1
0 = = 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:
4
vC (0 ) = 10 = 8 = vC (0 + )
5
10
iL (0 ) = = 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
[ ]
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 m m
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
1 20 10 j4000
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
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
diL diL
vin (t) 0.5 200iL = 0 + 400iL = 2vin ( t)
dt dt
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
40
IL e j (j + 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:
2
dv C dv
vin (t) v C C R = 0 RC C + v C = v in (t )
dt dt
This means that finding vC is enough to be able to obtain the output voltage. The input voltage is
represented by the complex exponential:
and vC (t ) = Vm e j ( 250t + ) . Substituting into the differential equation, dividing by e j250t , and rearranging:
dv s dv
RC + v s = Ris (t ) Ris (t ) RC s = v s
dt dt
Hence
i.e.,
R jRC(100 j200) = R 200RC j100RC = 100 j200
3
Thus RC = 2 s, R = 100 + 200RC = 500 , and C = 4 mF.
We represent the input signal with = 104 as vin (t) = 100e j t e j/2 V and iL (t) = IL e j t e j A.
Substituting these two expressions into the differential equation and dividing out e j t :
R 1 j100 j 1 R
2IL e j + j IL e j + IL e j = IL e j 2+ j = 108
L LC L LC L
108
IL e j = = 0.48 j0.64 = 0.8 126.87o
75 106 + j108
Solving for the magnitude and angle (by hand or using MATLAB):
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
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
1 j j0.02
Vme j 2+ =
LC RC C
Hence
j0.02
Vme j = = 1.28 + j0.96
1 j
C 2+
LC RC
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
SOLUTION 10.9. We define a Gaussian surface encompassing the three bottom nodes. Thus, KCL
dictates that the sum of 4 currents be zero:
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
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
I1 = j C VC = j10 0.1 2 = j2
5I1 j10
I2 = = =5
j L j2
Finally,
Vout = 5 2(I1 + I2 ) = 10(5 + j2) = 50 + j20 = 53.8521.8
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
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.
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: = 200 rad/s, I1 = 0.590 = 0.5 j A, and
I1
Vs2 = 40 . From KVL Vs1 = 3I1 + j LI1 + + Vs2 which leads to:
j C
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
1060 1060
IR = = 0.0260 A, IL = = 0.04 30 A, and
500 j1000 0.25
( )
IC = 1060 j1000 2 106 = 0.02150 A. Thus
11
Vin =-1.0000e+01 - 1.0000e+01i
abs(Vin)
ans =1.4142e+01
angle(Vin)*180/pi
ans =-135
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
IR = = 0.6 90 = j0.6
1/0.03
20 90
IL = = 1 180 = 1
j200 0.1
( )
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:
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:
200 20
Vout = 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:
400 20 90
Vout = 1 Vin = = 14.14 45 V
1 j
400 +
j C
Thus, in the time-domain,
vout (t ) = 14.142cos(250t 45) V
12
Solution 10.24 (a) = 10,000 rad/s and Vin = 100 90 = j100 V. Apply the voltage divider
formula:
100 100
Vout = Vin = V = 80 126.87
1
100 + j L + j C 100 + j100 j0.25 in
By Ohms law:
Vout = 100IR = 1.636.87 V
Therefore
vout (t) = 1.6cos(2500t + 36.87o) V
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 +
C R 2 + 2 L2
. It follows that w must satisfy
C R 2 + 2 L2 R + 2L2
2 R2 1
LR2C = R 2 + 2L2 or equivalently 2
= = . 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
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
1 1
Yin ( j250) = + + jC = 0.1 0.05 j + j250C
R1 R2 + jL
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 L
j L
j C C
Zin ( j ) = R + =R j
1 1
+j L L
j C C
Equating the real and imaginary parts of the given impedance, R = 4 and
L L
= = 2 L = 0.1 H
LC 1/ 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:
1 1
Zin ( j )=5+ j L+ = 5 + j L
j C C
We want the imaginary part to be equal to zero. Thus,
1 2 1
L =0 = = 2500 rad/s
C LC
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:
1 1 1 1
Yin ( ) = + j C+ = + j C
5 j L 5 L
1
Again, the imaginary part is equal to zero when = or = 2500, at which point the admittance is
LC
0.2.
1
Solution 10.31 The input admittance is Yin ( ) = + j C = 0.008+0.004. Equating this to the given
R
admittance at = 500yields R = 125 , C = 0.004 or C = 8 F.
1
Zin = = 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 = r = 15,811 rad/s. Observe that
C LC
Vs
Is = =0
R + ZLC ( j r )
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).
R j RC
Solution 10.35 (a) Using voltage division, Vout = Vin = V which
R C
j 1+ j RC in
RC Vin
Vout = 90o in tan1 ( RC ) and Vout =
1+ 2 R 2C 2
(b) For this part, we need to make sure that tan1( RC) = 45 RC = 1 = 1/ RC .
(c) At this frequency,
1
RC
Vout RC 1
= =
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
1 1
Yeq ( j )= + 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
18
Therefore, vout (t) = 8cos(1000t + 45o) V and iL (t) = 2cos(1000t 45o ) A.
(b) If = 618 rad/s, then
1
Yeq ( ) = 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.
R 2
V1 = Vin = (8 j ) = 5.657 45o
1 24 j +2j
R+ +j L
j C
Converting back to time:
vout (t ) = 5.657cos( 4t 45) V
j
VC = 0.02 25 100 = 2 j = 2 90
10 + j0.08 25 j 0.02 25
2
Vout 1 1
SOLUTION 10.42. = = 0.2 25 = 3 + j 8 .
Vin 3 + j8 + 1 j 4C 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,
400
VR = 50 = 4036.897 V
j
400
3.33 10 3 10 6
and
100VR j C 100VR
Vout = = = 398532.1 V
1 j100 C + 1
100 +
j C
21
Hence, vout (t) = 3985cos(3.33 10 3 t + 32.1) V.
100 j 0.1
Z1 = = 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
SOLUTION 10.46. Here = 40 rad/s and Iin = 0.1200 A. This problem is best solved using
MATLAB.
Y2 = 1/R + 1/(j*w*L)
Y2 = 2.0000e-01 - 2.5000e-01i
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:
20
I1 = = 0.5 + j0.5 = 0.70745
2 + 1 j25 0.02
23
Hence, i1 (t) = 0.707cos(25t + 45o ) A.
1 1 1 1
Then, by inspection, Yeq ( j500) = + + = 1.5 S. Thus,
2 1 j2 j2
24
2.50
VC = = 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,
1 1
Yeq = + + 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:
j3 j3
Vx_2 = Vs2 = 390o = 2.121 135o V
3 j3 3 j3
SOLUTION 10.52. (a) As stated, VL = aVs 1 + bIs 2 . To find a, set Is 2 = 0 and use voltage division:
j30
VL = V = 0.5Vs 1 = aVs 1
j30 + j30 s 1
(b) For this part, Vs 1 = 10 and Is 2 = 0.5 90o . Hence from the formula,
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
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
j
Yth ( j103 ) = 0.25 3 + j1000 0.25 10 3 = 0.25S Zth ( j10 3) = 4
1000 4 10
Then,
Voc = Iin 4 = 845 V
3
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,
4
SOLUTION 10.57. For this problem we short the V-source and compute Zth and use voltage division to
find Vo c . Specifically
1
Zth = = 103 j10
1
jC +
0.1+ jL
and
1
jC
Voc = 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
5
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:
j
100
3.3310 0.2510 6
3
VAB = = 99.65 4.77 V
100 j1200
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
Voc 398532.1
Isc = = = 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:
6
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:
100 ( j1000)
VAB = = 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
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:
j
VAB = j0.01 200 IS I 2IS = (2 + j)IS
200 0.005 S
V
Therefore, Zth = AB = 2 + j .
Is
7
j2I 6I 6
Apply KCL: = Is I I = I . Now, VAB is the voltage across the inductor:
6 j2 s
6 j2
VAB = I = 6Is Zth = 6 .
j2 s
Solution 10.63. = 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
+ gm VAB + j C( V1 VAB ) = 0
10 3
j C(VAB V1 ) + AB = IS
V
j L
Substituting values, these can be cast into the following matrix equation:
8
3 1 +j 3 j V1 10-3 VS
j =
j VAB IS
10
Therefore Zth = 500 and Voc = 6.25 + j6.25 = 8.83945o V. For the Norton equivalent we need
Voc 8.839
Isc = = 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
1
+ j C1 + j C2 Vx j C2VAB = IS
R
1
( j C2 + gm )Vx + j C2 +
V =I
R AB S2
In MATLAB
Thus
Vx 3 95.68 j20 0.21+ j0.955 40 10-6
V = 10 954.5 + 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
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
1
IA = V Isc (1)
Zth s
By inspection of the circuit,
V
IA = IC + I L 1 = jCVs + L 1 = j2Vs jVL 1 (2)
jL
But
VL = 0.25IC + Vs = 0.25 jCVs + Vs = (0.5 j + 1)Vs (3)
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
Because of the addition of Ix , we can write the modified nodal equations more or less by inspection:
1/ jL 0 1 VL j 0 1 VL 1
0
jC 1 VAB = 0 2j 1 VAB = Is
1 (1+ j0.25C) 0 I x 1 1 0.5 j 0 I x 0
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
Multiplying the second row of Ainv times the right-most vector of our equations produces
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
1
Vout = 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 6 = = 9950 rad/s
Vin 1+ j 0.001 1+ 2 10 100
VZ 140 VZ V + 890
+ + Z =0
j15 25 j20 j20
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
YRLC = + 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
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
1 2Ix 1 1 2Ix
0 = IC Ix + = Ix + Ix = 0.3 + j0.1= 0.31618.44 o A
j 1 j j
13
1
j C = j0.0866 L = 3.73 H
L
(c)
I
VC = in =
2030
Yin 0.05 + j0.0866
= 200 30 V vC ( t) = 200cos 20t 30o V ( )
(d)
IL =
20030
j20 3.73 (
= 2.68 120A i L ( t) = 2.68cos 20t 120o A )
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
(
vC (t ) = v A (t) = 0.707cos 1000t 45o V )
(b) IL =
0.707 45
2 + j2 ( )
= 0.25 90 A implies that iL ( t) = 0.25cos 1000t 90o = 0.25sin(1000t) A.
(c) We already determined Voc in part (a). Now, turn off the source to compute the equivalent impedance:
1
Yth = 0.75 j + + 0.25 = 0.5 + j0.5 Zth = 1 j . This is the series connection of a 1 resistor
2 + j2
1
with a capacitor of value C = = 10 3 F. This completes the definition of the Thevenin equivalent.
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
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
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:
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
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 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.
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
[
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
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
VR
IR
(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
25 100
80
20 60
40
Phase in degrees
Magnitude Zin
15 20
10 -20
TextEnd
-40 TextEnd
5 -60
-80
0 -100
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Frequency in rad/s Frequency in rad/s
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):
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.
1 1
Yin = + +j C
R j L
1 V
Zin = = s
Yin Is
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:
23
100
90
80
70
Magnitude Zin 60
50
40
TextEnd
30
20
10
0
0 50 100 150 200 250 300
Frequency in rad/s
plot(w,angle(Zin)*180/pi)
grid
xlabel('Frequency in rad/s')
ylabel('Phase Zin in degrees')
100
80
60
Phase Zin in degrees
40
20
-20
-40
TextEnd
-60
-80
-100
0 50 100 150 200 250 300
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 =
V 1
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
1 0.2 0.2
=
1 2 1 2 R
R2 + 57L R2 + 50L
57C 50C
Hence
2
2 1 R2
R + 57 0.4 = = 25R 2
57 10 3 0.04
R2 + 57x0.4 - 1 2 R2 = 25R 2
57x0.001 0.04
5.2561
From which R = = 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:
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,
o
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
100 j 0.1
VL = I = (50.265 + j50) 0.01 = 0.50265 + j0.5
100 + j0.1 in
V =-10VL = 5.0265 j5
Therefore
vC (t) = 5cos(320t + 179.7o )V
2
Parts (b) and (c). For the SPICE simulation we have the following circuit in B -SPICE:
27
Prb10-87-Small Signal AC-2 Frequency (Hz)
+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:
Vin
Iin = = 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,
28
Vin
SOLUTION 10.89. (a) = 800rad/s and Vin = 1 90V Iin = . No current flows into Op-
106
Amp terminals:
1 j ( j)
Vout = Iin = = 1.25 103 V
j C 800
Thus, vout(t) = 1.25cos(800t) mV.
Vin 1 j ( j)
(b) Again, Iin = A, and Vout = Iin 10 = C = 2.5 nF .
200 10 3 j C C2 105
1
Solution 10.90 (a) = 2 700rad /s and Vin = 10 V Iin = A. Further,
150 103
j j
150 103 C 1 150 103 C o
Vout = j I in = 150 103 j = 0.001590 V
3 3
150 10 C 150 10 C
(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,
Vin
Solution 10.91 (a) Use the virtual short property: Iin = . All of current flows through the
10 3+ j 1C
4 10 4 Vin Vout
feedback path: Vout = 10 Iin = j = 7.07 135o .
3 Vin
10 C
(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
VS + ZRCIR = Vo o = 1 + C = 1 +
VS R 3R j j3 RC + 1
C
(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,
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
Vx Vin
+ j C1Vx + j C2 ( Vx Vout ) = 0
200
Vout
j C2 (0 Vx ) =0
28 103
In MATLAB,
Vout
Hence, = 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.
Vout Zf
=
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
Vout Zf
of resistance, in this case R2 . Thats exactly the requirement of this problem, since we want =
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:
1/ j C
Vop2 = 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
2 10 5 j C2 2 10 6 j C2 2 106
Now, the first op-amp circuit is an inverting amplifier, but its + terminal is at VRR now. Thus,
6
Magnitude
4 TextEnd
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Frequency Hz
(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 P11-1 @ DeCarlo & P. M. Lin
1
1
1 e 2t
SOLUTION 11.1. Using equation 11.3, Pav = (e t 1)2 Rdt = + t 2e t = 0.758 W.
2
0 2 0
SOLUTION 11.2.
V2
(a) From 11.6, Pav = m = 50 mW for a sinusoidal input.
2R
(b) From 11.3,
R
/20 3 /20 2 /10 2
Im R
Pav = (10cos(10t)) 2 dt +
2 2
(10cos(10t)) dt + (10cos(10t)) dt = = 50 mW
2 2
0 /20 3 /20
just as the previous case since the square of the absolute cos(10t) is the same as the square of cos(10t).
(c)
10 3
2 /10 2 /10
[ ]
10R 2
[0.5cos(20t) + 0.125cos(40t) + 0.375]dt
2
Pav = 0.01cos (10t) dt = R
2 2
0 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
W
TextEnd
0
0 0.2 0.4 0.6 0.8 1
0.1
0.05
W
TextEnd
0
0 0.2 0.4 0.6 0.8 1
x 10-5
1
0.5
W
TextEnd
0
0 0.2 0.4 0.6 0.8 1
time in seconds
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
35
Instantaneous Power in W 30
25
20
15 TextEnd
10
0 0.5 1 1.5 2 2.5 3 3.5 4
Time in s
40
35
Instantaneous Power in W
30
25
20
15
10
TextEnd
5
0
0 0.5 1 1.5 2 2.5 3 3.5 4
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
1
square root, Veff = 15.81 V. For fig. b, Veff = (20t )2 dt = 400
3
= 11.55 V
0
2
2 Veff
(b) Pav = Ieff R= 8 = 20 W.
10
2
Veff
(c) Pav = 8 = 10.67 W
10
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.
( )
1 1 t 2
SOLUTION 11.6. (a) Ieff =
2 0
e 1 dt = 0.615 A.
2
(b) Pav = Ieff R = 0.758 W.
(b)
1 1
0 (10cos(2t) + 5cos(4t))2 dt = [62.5t]0 = 62.568
2
Veff =
Hence Veff = 7.91 V.
(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:
2
Veff
1
(
= 100cos 2 (2t) + 25cos 2(4 t) + 25cos 2 (4t /4) + 50cos(4 t)cos(4t /4) dt
0 )
Hence
SOLUTION 11.8. The voltage is V = 500 V and the impedance Zeq = 100 100 j = 141.42 45o
2
V 2 0.3536
. Thus I = = 353.645o mA. Hence, Pav = R Ieff = 100 = 6.2516 W.
Zeq 2
Complex Power Prbs 2/5/00 P11-5 @ DeCarlo & P. M. Lin
1
SOLUTION 11.9. (a) The equivalent load seen by the source, Zeq = = 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
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.
2
(c) 3Is = 300 W
SOLUTION 11.12. (a) By KCL the current through the resistor is 5I. So by KVL,
Vs
I= = 107.33 26.57o A
1000 + j500
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.
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.
1
SOLUTION 11.15. First use voltage division and observe that VL = V = 6060 V. Now the
2 s
VL*
complex power is SL = 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
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
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:
Sz2 = V2*conj(V2*Y2)
Sz2 =
1.0000e+03 + 5.5000e+03i
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;
(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
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
(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
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.
Ss
(b) Is = = 22.675 A
Vs
(c) The total power delivered to the three groups of impedance following V1 is S1 = 41.5 + j22. From
S1 + S 2 + S4
the current obtained in (b) V1 = = 2071.5 V.
Is
S 4 + S2
(d) From V1 , and the total power delivered to Z4 and Z2 , I2 = = 12.385 A. Finally
V1
S2
V2 = = 1805.5 V.
I2
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
1
SOLUTION 11.21. From equation 11.30, we find Q = P 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
SL 4000
IL = = = 33.33 A. The power absorbed by the transmission line is then from
VL 120
SOLUTION 11.24. The capacitor must absorb a reactive power of Qnew Qold = 17.9 kvar. Thus
2 QC
jQC = j17.9 = j CV s , and C = 2 = 0.897 mF.
Vs
1
SOLUTION 11.25. From equation 11.30, Qnew = 86.6 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
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:
1
Q2 = 1440 1 = 1080 var
(0.8)2
As an aside we compute the magnitude of the current without the capacitor attached.
S1,2
Is = = 17.493 A
120
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
2
QC = C Vsource
Hence C = 0.2 mF. Finally
Pave 1800
pf = = =1
S1 + S2 + SC 1800
1
SOLUTION 11.27. (a) From equation 11.30, Qold = 7 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.
Complex Power Prbs 2/5/00 P11-12 @ DeCarlo & P. M. Lin
7 7
(c) Sold = = 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.
(k) Snew
s = Vs
new new *
Ieff ( )
= 228.898.57 = 226.33 + j 34.12 kVA, and the efficiency is 41.5%.
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 11.31. (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 .
2
Voc
(b) The complex power absorbed by the load is, SL = ZL = 750 + j500 VA, and the average
Zth + ZL
power 750 W.
V1 V2
V1 V2 = 4(V1 /8)
Is + Itest = V1(1/8 + j0.25) + V2 (1/16 + j0.5)
Vtest = V2 + 8Itest
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.
2
Voc
(c) SL = ZL = 275.2 + j30.26 VA and the average power is 275.2 W.
Zth + ZL
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 P11-14 @ DeCarlo & P. M. Lin
10 + j20
Iload = 10 = 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 = 600.63 W.
Zth + ZL
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 2 100 2 100
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.
P = 10Iload =
2 (
10 50 2 )2 12.5 103
2 =
1 1 2
(R + 10)2 + L (R + 10) 2 + L
C C
Here, again R = 0 with C chosen to eliminate the reactive term maximizes power transfer. Hence
1
C= 2L = 0.01 F
with Pmax = 125 W.
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
1 7 10 4 j C
ZL ( j ) = j L + 4 = j10 L + 8
10 + j C 10 + 1014 C 2
10 4 C
= 8 + j 107 L 8 = 10 + j1000
10 + 10 C14 2 10 + 10 C
14 2
C = sqrt(1e-4 - 1e-7)/sqrt(1e15)
C = 3.1607e-10
w = 1e7;
L = (1e3 + w*C/(1e-8 + 1e14*C^2))/1e7
L = 1.3161e-04
(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 .
2
V2
P10k =
104
Thus if P10k is maximized, then V2 is maximized.
2
SOLUTION 11.38. (a) From equation 11.4, if we substitute , then the resulting instantaneous
T
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 =
T t0
f (t)dt . Without going into detailed calculation, f 2 (t) will
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,
Feff =
1 to +T 2
T t 0
f (t)dt =
1 t0 +T 2
T t 0 (
Fo + 2F12 cos2 ( 1t + 1) + ... dt )
=
1 t0 +T 2
T t 0 (
Fo + F12 + F12 cos(2 1t + 2 1 ) + ... dt )
1 2
[ ] T
= Fo t + F12 t + F12 sin(2 1 t + 2 1 ) + ... = Fo2 + F12 ...
T 0
dvC
iC (t) = C = 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.
2
QC = IC ,eff VC ,eff = CVC,eff = 0.5 CVm2 = 2 (0.25CVm2) = 2 WC,ave .
Q
Therefore, WC,ave = C .
2
Complex Power Prbs 2/5/00 P11-17 @ DeCarlo & P. M. Lin
diL
v L (t) = L = 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
WL (t) = 0.5LiL2 (t) = 0.5LIm
2
sin2 ( t) = 0.25LIm
2
(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.
2
QL = IL,eff VL,eff = LI L,eff 2
= 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:
Vs j110
Is = = = 10.17 33.69 A
6 j9 6 j9
or iL (t) = 10.17 2 cos(120 t 33.69o ) A. Similarly
15 110
VC = j15 Is = = 152.54 123.69 V
6 j9
Complex Power Prbs 2/5/00 P11-18 @ DeCarlo & P. M. Lin
6 1
or vC (t) = 152.54 2 cos(120 t 123.69o ) V. Further, L = = 15.9 mH and C = = 176.8 F.
15
The instantaneous energy stored in the inductor is
(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
2 110 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
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 @ DeCarlo & P. M. Lin
w = 2*pi*60;
L = sqrt( magZcoil^2 - 2.7^2)/w
L=
4.0725e-02
magZin = 220/magIcoil
magZin = 2.2863e+01
R = Realpart - 2.7
R = 1.4241e+01
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
Z
V2
(
= L (cos( v i ) + cos(2 t + v + i )) + L cos( v i ) + cos(2 t + v + i + 120o )
Z
)
+
VL2
Z (cos( v i ) + cos(2 t + v + i 120o ) )
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:
(2
Ploss = 2 IL R = 2 2R
PL2
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:
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):
( ) ( ) ( )
Vp Vp Vp
Voc = 30o 150o = 1 30o 1 150o = Vp
3 3 3
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
I1 = + ( Z || Z || Z ) = V1 Z Z + Z Z + Z Z V2
Z1 Z1Z1 Z2Z1 1 2 3
Z3
. Do the
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 +
Z31 Z12 Z12
. Now substitute the formulas in the problem
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
SOLUTION 12.6. By Ohms law IA = = = 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
2
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
ICmag = IAmag
ICmag = 8.9443e+00
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
Z 4 + j3
VAN = VsA = 120 = 116.94 V
Z + Zg 4 + j3 + 0.1 + j0.2
120 116.94
Thus the drop is: 100 = 2.55 % .
120
Z V 0.945 120
= AN = = 0.945
Z + Z1 + Zg 120 120
From this equation,
Z
Z + Z1 + Zg = = 5.291.
0.945
So (4.05 + 0.1)2 + (3.15 + 0.2)2 = 27.99 . Solving the quadratic equation results in = 0.7893.
SOLUTION 12.10. Solving the single phase equivalent circuit for phase A,
V p V p120
IA = = 58.87 56.57
10 + j5
V p 120 V p
IB = = 58.87 176.57
10 + j5
V p120 V p 120
IC = = 58.8764.43
10 + j5
Vp
IAB = = 6.57 26.57
30 + j15
By phase inference,
2
(b) The average power is P = 3 30 IAB = 3.88 kW, and the total reactive power is
2
Q = 3 15 IAB = 1.94 kvar.
25 746
SOLUTION 12.12. From the efficiency we know that = 0.85 , so the total power being
Pdeliverd
21941
delivered to the load is 21941 W. Using 12.4, IL = = 63.31 A.
3 VL pf
300 746
SOLUTION 12.13. (a) Using the power efficiency relationship, Ptot = = 239.36 kW.
0.935
239.36 10 3
(b) Using 12.4, IL = = 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
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,
Vp Vp Vp
VAN = 30, VBN = 150, VCN = 90
3 3 3
VAN
IA = = 19.63 56.57
Z
From phase inference
IB = IA 120 = 19.63 176.57
IC = IA 120 = 19.6363.43
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 @ DeCarlo & P. M. Lin
Vp
IA = = 34.08 26.56
(Z / 3)
IB = IA 120 = 34.08 146.56
IC = IA 120 = 34.0893.43
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:
(b) Perform a delta to Y-transformation and use Ohms law and phase inference to obtain in amps:
220 30
IA = = 11.36 56.57
3Z
IB = 11.36 176.57
IC = 11.3663.43
(c) First,
5
pf = cos tan1 = 0.8944
10
2
The reactive power may be found as follows, Q = 3 5IL = 1.936 kvar.
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
Part (c). Compute line voltages VAB , VBC , and VCA . From the voltage phasor diagram of figure 12.2.
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,
V p 30
IA = = 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)
V p 30 230 30
IA = = = 32.84 32.73 A
3 ZYnew 3( 4.044 + j0.1923)
IB = 32.84 152.73 A
IC = 32.8487.27 A
2
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
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
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
( ) (
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);
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
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
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,
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.
VA = 220
VB = 220*exp(-120*pi*i/180);
VC = 220*exp(120*pi*i/180);
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 2
where Vl = 3V p using Y to Delta source transformation. Thus PAB = 2904 W, PBC = 1452 W, and
PCA = 2904 W.
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:
Equivalently
2.2VM = 88 j152.42 or VM = 40 j69.282 V
1.619
(b) Repeating the previous calculations, Req2 = 50 = 0.08095 .
1000
Vapp = 115 2 Req214 = 112.7334 V, which corresponds to 1.97 %. So the wire gauze is appropriate.
(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.
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
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
S OLUTION 1 3 . 1 . Given
i (t) + 16i (t) + 4Bi(t) = v (t) + 8v(t)
1
Prbs Ch 13 March 18, 2002 P13-2 R. A. DeCarlo, P. M. Lin
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
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
2
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
i4 = 2v1 = 2
d 1
( )
v + 2vo = 2(vo + 2vo )
dt o
i5 = i3 + i4 = 2(v o + v o) + 2(vo + 2vo ) = 2vo + 6vo + 2vo
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)
3
Prbs Ch 13 March 18, 2002 P13-4 R. A. DeCarlo, P. M. Lin
S OLUTION 1 3 . 3 . (a) From the given differential equation, the characteristic equation is
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))
2
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 )
S OLUTION 1 3 . 5 .
(a) Let T = 1.
5
Prbs Ch 13 March 18, 2002 P13-6 R. A. DeCarlo, P. M. Lin
1.5
0.5
0
-1 -0.5 0 0.5 1 1.5 2 2.5 3
-0.5
-1
-1.5
-2
-1 -0.5 0 0.5 1 1.5 2 2.5 3
(c)
0.25
0.2
0.15
0.1
0.05
0
-1 -0.5 0 0.5 1 1.5 2 2.5 3
(d)
6
Prbs Ch 13 March 18, 2002 P13-7 R. A. DeCarlo, P. M. Lin
0.8
0.6
0.4
0.2
0
-1 -0.5 0 0.5 1 1.5 2 2.5 3
(e)
0
-1 -0.5 0 0.5 1 1.5 2 2.5 3
r
(f) i (t i)
i =0
(g)
0
-1 -0.5 0 0.5 1 1.5 2 2.5 3
7
Prbs Ch 13 March 18, 2002 P13-8 R. A. DeCarlo, P. M. Lin
S OLUTION 1 3 . 6 .
T2
e st
(a) L[ f1(t)] = F1(s) =
f (t)e
0 1
st
dt ==
T 2 st
T1
e dt =
s
=
s(
1 sT 1
e e sT 2 )
T1
S OLUTION 1 3 . 7 .
st 4t st e (s+ 4)t 5
(a) F(s) = f (t)e dt = 5e e dt = 5 =
0 0 s + 4 s+ 4
0
4.5
3.5
3
f(t)
2.5
TextEnd
1.5
0.5
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
t
8
Prbs Ch 13 March 18, 2002 P13-9 R. A. DeCarlo, P. M. Lin
e (s +4) t 5e (s +4) 4
st 4t st s 5e
(b) F(s) = f (t)e dt = 5e e dt = 5 = =e
0 1 s + 4 s+4 s+4
1
0.1
0.09
0.08
0.07
0.06
f(t)
0.05
TextEnd
0.04
0.03
0.02
0.01
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
t
st 4(t 1) st e (s+ 4)t 5
(c) F(s) = dt = 5e = e s
4
f (t)e e dt = 5e
0 1 s + 4 s+ 4
1
4.5
3.5
3
f(t)
2.5
TextEnd
2
1.5
0.5
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
t
(d) F(s) = f (t)e st dt = 5e 4( t 1) (t)e st dt = 5e 4
0 0
(e) F(s) = f (t)e st dt = 5e 4( t 1) (t 1)e st dt = 5e s
0 0
9
Prbs Ch 13 March 18, 2002 P13-10 R. A. DeCarlo, P. M. Lin
(f)
F(s) =
0
f (t)e st dt = 2
0
[ u(t)u(1 t)]e 0.25t e stdt = 20
1
e 0.25te st dt =
2
s + 0.25 (
1 e (s +0.25) )
= 2
0
e 0.25t e stdt 2
1
e 0.25te st dt =
2
s + 0.25 (
1 e s )
1.8
1.6
1.4
1.2
f(t)
1
TextEnd
0.8
0.6
0.4
0.2
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
t
F(s) = f (t)e st dt =
0
S OLUTION 1 3 . 8 .
A
(a) F(s) = f (t)e st dt = Ae t e stdt = Ae(s )t dt =
0 0 0 s
A
(b) F(s) = f (t)e st dt = Ae (t 1) u(t 1)e st dt = Ae e (s )t dt = e s
0 0 1 s
(c) F(s) = f (t)e st dt = Ae (t 1) (t)e st dt = Ae
0 0
1
e (s )t
(d) F(s) =
0
f (t)e st
dt =
1
0
Ae et st
dt = A =
A
(s ) s (1 e(s ))
0
10
Prbs Ch 13 March 18, 2002 P13-11 R. A. DeCarlo, P. M. Lin
st 1 1 s
L[ f (at)] = f (at)e dt = f (q)e (s/a)q dq = F
a a a
0 0
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
cos( t) =
2
Similarly, subtract the equations and divide by 2j to obtain
e j t e j t
sin( t) =
2j
Note that
[ ]
L ej t =
1
s j [
, L e j t =
1
s+ j ]
(b)
e j t + e j t 0.5 0.5 s
L[cos t ] = L = + = 2
2 s j s+ j s + 2
and
0.5 j 0.5 j
L[sin t] = + = 2
s j s+ j s + 2
11
Prbs Ch 13 March 18, 2002 P13-12 R. A. DeCarlo, P. M. Lin
L[sin( t) u(t)] =
1 1
= 2
s
2 s + 2
+1
Using the time differentiation property,
d
L[ cos( t)u(t)] = L (sin( t)u(t)) = sL[sin( t) u(t)] sin(0) = 2
s
dt s + 2
d s 2 s2
L[ t cos( t)u(t)] = 2 =
ds s + 2 s2 + 2
( )
2
s2 + 2
Hence
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
L[sin( t)] = 2
s + 2
By the damping property
[
L e at sin( t) = ] (s a) 2 + 2
L[g2 (t)] = A
(s a) 2 + 2
d
By the differentiation in the time domain property cos(t) = sin(t) which implies
dt
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
[
L Asin( )e at cos( t) = Asin( ) ] sa
(s a)2 + 2
[
G3(s) = L Aeat sin( t + ) = Acos( ) ] (s a) 2 + 2
+ Asin( )
sa
(s a)2 + 2
A
= ( cos( ) + sin( )(s a))
(s a) 2 + 2
cos(t ) =
d
dt
sin(t) L[cos(t )] = 2
s
s +1
sin 0 = 2
s
s +1
( )
By the frequency scaling property with
s
s 1 s 2
L[cos( t )] =
1 s
F = = = 2
s 2
+ ( )
1
2
( s2 + 2
) s + 2
L[ At cos( t )u(t)] =
A ( 2 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
L[ Acos( t)] = A 2
s
s + 2
By the Frequency-shift property
[
L Aeat cos( t) = A ] s a
(s a) 2 + 2
[
L Aeat cos( t) = A ] s a
(s a) 2 + 2
Hence
[ ]
L Aeat cos( )cos( t) = Acos( )
sa
( s a)2 + 2
It follows that
[ ]
L Aeat cos( t + ) = Acos( )
s a
(s a) +
2 2 [
A sin( )L e at sin( t) ]
s a cos( )(s a) sin( )
= A cos( ) A sin( ) = A
(s a) 2 + 2 (s a) 2 + 2 (s a) 2 + 2
L[cos(t )] = 2
s
s +1
L[cos( t )] = 2
s
s + 2
Using the multiplication-by-t property
d s s2 + 2 2s2 s2 2
L[t cos( t )] = = =
ds s2 + 2
( ) ( )
2 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,
(s a) 2
[ ]
2
at
L te cos( t) =
[( s a) 2
+ 2]2
(b) With
e at + e at
cosh(at) =
2
1 1 1 1 s a + s + a 1 2s s
L[cosh(at)] = + = 2 = 2 = 2
2 s + a s a 2 s a 2 2 s a s a2
2
L[sinh(at )] = 2
a
s a2
So that by the multiplication by t property
d a a(2s)
L[t sinh(at )] =
2as
= =
ds s a
( ) ( )
2 2 2 2
s2 a 2 s2 a 2
16
Prbs Ch 13 March 18, 2002 P13-17 R. A. DeCarlo, P. M. Lin
d s
L[t cosh(at)] = 2 =
( )
s2 a 2 s(2s) s2 + a 2 + 2s2
= =
s2 + a2
ds s a 2
( ) ( ) ( )
2 2 2
s2 a2 s2 a2 s2 a 2
(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 + 2) + 1 s + 3
(s + 2) + 2 s + 4
L[g3 (t)] = G1(s + 2) = 5e 2(s+ 2) = 5e 2(s +2)
(s + 2) + 1 s + 3
(d) Since g4 (t) = 5tf (t) , use the multiplication-by-t and linearity principle to obtain
d 5(s + 2) 5[(s + 1) (s + 2)]
L[g4 (t)] =
5
= =
ds s + 1 (s + 1)2 (s + 1)2
17
Prbs Ch 13 March 18, 2002 P13-18 R. A. DeCarlo, P. M. Lin
e Ts s
(a) L[ Af (t T )u(t T )] = A 2
s +4
d d s
L[ Atf (t)u(t)] = Af (s) = A 2
2
(
s + 4 s(2s)
= A
)
s2 4 ( )
(b)
ds s + 4 = A 2
( ) ( )
ds 2 2
s2 + 4 s +4
(c) Note that the answer is simply a time shift of the function given in (b).
L[ A(t T ) f (t T)u(t T )] = Ae sT ( s2 4 )
(s 2 + 4 )
2
(d) This function is that of part (a) multiplied by t. Hence, by the multiplication by t property,
d e Ts s
L[ Atf (t T )u(t T )] = A 2
(
Ts
e Tse
= A
)
Ts (s 2 + 4) 2s se Ts
ds s + 4
( )
2
s2 + 4
2 3 2
2s (1 Ts ) (s 2
+ 4) Ts + s + 4Ts 4
= AeTs = Ae Ts
( ) ( )
2 2
s2 + 4 s2 + 4
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,
d d (2s) 2 s
L[ Atf (t)u(t)] = A [F (s)] = A 2 2 = A =A
ds s +
( ) ( )
ds 2 2
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).
2 s
L[ A(t T ) f (t T)u(t T )] = AesT
(s 2 + 2 )
2
(d) The answer here uses the multiplication-by-t property applied to the answer of part (a).
Ts 2
d e Ts = A Te (s + 2 ) + 2se Ts
L[ Atf (t T )u(t T )] = A
(
ds s2 + 2 )
(
s2 + 2
2
)
2 2
Ts + 2s + T
= A e Ts
( )
2
s2 + 2
0+ 0+
(2t)dt = 0.5 ( )d = 0.5
0 0
under the transformation = 2t. For the second term, the function peaks at t = 0.5 and
0.5+ 0+
(2t 1)dt = 0.5 ( )d = 0.5
0.5 0
19
Prbs Ch 13 March 18, 2002 P13-20 R. A. DeCarlo, P. M. Lin
L[ f (2t)] =
1
2 (
F (s /2) = 0.5 1+ e 0.5s )
(b)-(ii) For this part,
(
L[ f (2t)] = 0.5L[ (t) + (t 0.5)] = 0.5 1+ e 0.5s )
( )
= 2s2 s F (s) 2sg(0 ) 2g'(0 ) + g(0 ) =
(2s 1)(s + 1)
s2
(b)
L[v(t)] = 2L[ f "(t)] L[ f '(t)] = 2s2F (s) sF(s) 2sf (0 ) 2 f '(0 ) + f (0 )
(2s 1)(s + 1)
( )
= 2s2 s F (s) 2s 2 + 1 =
s2
2s 2 + 1
(c)
t 0
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 s + 1 1 1
= s 3 = (s + 1) 2 4
s s s s
(d)
t F(s) 1
0
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
ds[ ( ) ( )]
ln s2 + 4 ln s2
2s 2s 2s 2 2s2 2s2 8 8
= 2 2= 2 = =
s +4 s s +4 s 2
s s +4 2
ss +4 ( ) ( )
(ii) Using the solution to (a)-(i), by the frequency shift property
[
L te 2t f (t)u(t) = ] 8
=
8
(
(s + 2) ( s + 2) + 4
2
) (
(s + 2) s2 + 4s + 8 )
8
= 3
s + 6s + 16s + 16
(b) If
8
G(s) =
(
s s2 + 4 )
a partial fraction expansion may be employed
8K1 As + B 2 2
G(s) = = + = +
(
s s2 + 4 s ) s +4
2 s s +4
2
Hence,
g(t) = [2cos(2t) 2] u(t)
and
g(t) 2 2
f (t) = = 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 P13-22 R. A. DeCarlo, P. M. Lin
d s + a d 2a
= (ln[s + a] ln[s a]) =
d 1 1
L[tf (t)u(t)] = F(s) = ln = 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,
2a 2a
L[te at f (t)u(t)] = L[e at ( tf (t)u(t))] = G(s + a) = =
(s + a) a
2 2 s(s + 2a)
2a
Part (b): g(t) = L1[G(s)]= L1 2
s a
2
1 1 1 1
= L s + a L s a = e
at
(
e at u(t) )
More specifically,
e at e at
( )
g(t) = e at e at u(t) = 2
2 u(t) = 2sinh(at) u(t)
g(t) 2sinh(at)
Hence f (t) = = .
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
s + b (s + a) ba
[ ln (s + a) ln( s + b)] =
d 1 1
L[tf (t)u(t)] = + = =
ds s + a s + b (s + a)(s + b) (s + a)(s + b)
[ ]
L te at f (t)u(t) =
ba
= 2
b a
(s + 2a)(s + a + b) s + s(3a + b) + 2a(a + b)
(b) If
22
Prbs Ch 13 March 18, 2002 P13-23 R. A. DeCarlo, P. M. Lin
ba
G(s) = L[tf (t)u(t)] =
(s + a)(s + b)
a partial fraction expansion may be employed
ba K K 1 1
G(s) = = 1 + 2 = +
(s + a)(s + b) s + a s + b s + a s + b
( )
Hence, g(t) = e at e bt u(t)
and
g(t) e bt e at
f (t) = =
t 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
t
S OLUTION 1 3 . 2 6 . For 0 t < T1, we see that g(t) = f (q)dq . Thus one presupposes here that
t
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 R. A. DeCarlo, P. M. Lin
F (s) A BeT1s Ce T 2s
G(s) = = +
s s s s
S OLUTION 1 3 . 2 7 .
1 e s 1
(a) f (t) = u(t) + u(t 1) f (s) = +
s s s (
= 1 e s )
(b)
1 e s e 3s 1
f (t) = u(t) + u(t 1) u(t 3) f (s) = +
s s
s s (
= 1+ e s e 3s )
(c)
1 e s 2e 3s 1
f (t) = u(t) + u(t 1) 2u(t 3) F (s) =
s
+
s
s
= 1 + e s 2e 3s
s ( )
(d)
2 e 2s e 3s 1
f (t) = 2u(t) u(t 2) u(t 3) F(s) =
s
s
s
= 2 e 2s e 3s
s ( )
S OLUTION 1 3 . 2 8 .
2 2e s
(a)
s s
2
f (t) = 2r(t) 2r(t 1) F (s) = 2 2 = 2 1 e s
s
( )
2 2e s e 2s
(b)
s s s
1
f (t) = 2r(t) 2r(t 1)+ r(t 2) F (t) = 2 2 + 2 = 2 2 2e s + e 2s
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
s s s s s
2
(
F(s) = 2 2 2 + 2 = 2 1 e s e 2s + e 3s )
24
Prbs Ch 13 March 18, 2002 P13-25 R. A. DeCarlo, P. M. Lin
3 3
S OLUTION 1 3 . 2 9 . (a) Here f (t) = r(t) 3r(t 2) + r(t 4) . Thus,
2 2
3e 2s 3e 4s
3
F(s) = 2 2 +
2s s 2s 2
3
2s
(
= 2 1 2e 2s + e 4s )
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
s s s s
(
= 2 1 e s se 4s )
(b) Here f (t) = 2u(t) r(t 2) + r(t 4) implies
2 e 2s e 4s 2
F(s) =
s s s s
(
2 + 2 = 2 s e 2s e 4s )
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
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
SOLUTION 14.2.
(a)
80
(10 + 0.2s)( )
Vs (s) s s(800 + 16s) 800s + 4000
Zin (s) = = = = 2
I s (s) 10 + 0.2s + 80 s(0.2s + 10s + 80) s + 50s + 400
2
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)
2400s + 12,000 12,000 48,000
K2 = = = 180
(s + 10)(s + 40) s= 20 (10)(20)
2/23/02 page P14.2 R. A. DeCarlo, P. M. Lin
SOLUTION 14.3.
(a)
1 1
Yp(s) = Cs + = 2 10 3 s + = 2 10 3 (s + 50)
R 0.10
Then
500
Z p(s) =
s + 50
and
500 1.25s2 + 62.5s + 500
Zin (s) = 1.25s + =
s + 50 s + 50
and
I (s) 1 s + 50 0.80s + 40
Yin (s) = s = = =
Vs (s) Zin (s) 1.25s2 + 62.5s + 500 s2 + 50s + 400
3.20 3.20 24
Is (s) = +
s + 10 s + 40 (s + 40)2
and for t > 0
is (t) = 3.20e 10t 3.20e 40t + 24te40t A
SOLUTION 14.4.
(a) Find Zin (s) vis Yin (s)
1 1 200Cs + 10LCs 2 + 10 + 20 + Ls
Yin (s) = Cs + + =
Ls + 20 10 10(Ls + 20)
10LCs2 + (200C + L)s + 30
=
10Ls + 200
and
1 10Ls + 200
Zin (s) = =
Yin (s) 10LCs + (200C + L)s + 30
2
K K K
= 300 1 + 2 + 3
s 200 300
It follows that
s + 400 400 1
K1 = = =
(s + 200)(s + 300) s =0 200(300) 150
s + 400 200 1
K2 = = =
s(s + 300) s =200 (200)(100) 100
and
s + 400 100 1
K3 = = =
s(s + 200) s =300 (300)(100) 300
Thus
2/23/02 page P14.4 R. A. DeCarlo, P. M. Lin
1 1 1
150
Vin (s) = 300 100 + 300
s s + 200 s + 300
2 3 1
= +
s s + 200 s + 300
and for t > 0
vin (t) = 2 3e 200t + e 300t V
SOLUTION 14.5.
s + 20
Z(s) =
s + 40
and the network is at rest
(a) If
20
vin (t) = 20u(t) Vin (s) =
s
then
V (s) 20 s + 40 s + 40
Iin (s) = in = = 20
Z(s)
s s + 20 s(s + 20)
s + 40 K K2
Iin (s) = 20 = 20 1 +
s(s + 20) s s + 20
in which case
s + 40 40 s + 40 20
K1 = = = 2, K 2 = = = 1
s + 20 s=0 20 s s =20 20
Thus
2 1
Iin (s) = 20
s s + 20
and
iin (t) = 20(2 e 20t )u(t) A
(b) Note that
20
vin (t) = 20e 40t Vin (s) =
s + 40
Then
V (s) 20 s + 40 20
Iin (s) = in = =
Z(s)
s + 40 s + 20 s + 20
in which case
iin (t) = 20e 20t u(t) A
(c) Note that
20
vin (t) = 20e 20t Vin (s) =
s + 20
Then
2/23/02 page P14.5 R. A. DeCarlo, P. M. Lin
Vin(s) 20 s + 40 s + 40
Iin (s) = = = 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(20) = 20
p'( s) = 1 p'(20) = 1
and
p'(20) p(20) 20
C1 = = 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
Vin 200
det 100
0 100 + 100s + 100
s = 1
I1 = V
100 200 s 100s + 100 in
det100 100 + 100 100 200
s
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
Vin (s) = Iin (s) + I (s) + V2 (s)
s 100 in
Now apply KCL to the right side to obtain
s 100
Iin (s) = V2 (s) + V (s)
100 s 2
Thus
s2 + 104
Vin (s) = Iin (s) + V2 (s)
100s
To find V2 (s) note that
s2 + 104
Iin (s) = V (s)
100s 2
implying that
2/23/02 page P14.7 R. A. DeCarlo, P. M. Lin
100s
V2(s) = 2
s + 104 in
I (s)
Thus
s2 + 10 4 100s (s2 + 104 )2 + 104 s2
Vin (s) = + 2 Iin (s) = Iin (s)
100s s + 104 100s(s2 + 104 )
implying that
V (s) s4 + 3 104 s2 + 108
Zin (s) = in =
Iin (s) 100s(s2 + 104 )
and
1 100s(s2 + 104 )
Yin (s) = = 4 S
Zin (s) s + 3 104 s2 + 108
SOLUTION 14.9.
Three mesh equations for the circuit
SOLUTION 14.10.
1
(a) Yin (s) = Cs + implies a parallel RC circuit with values R and C respectively.
R
2/23/02 page P14.8 R. A. DeCarlo, P. M. Lin
1 1
(b) Yin (s) = = 2s + which is a parallel RC circuit of values 4 and 2 F respectively.
Zin (s) 4
1 1
(c) Zin (s) = 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).
2s + 8 4 1
(d) Zin (s) = = 2+ = 2+
s+2 s+2 1 . Using the results of parts (b) and (c), this circuit is a 2
0.25s +
2
resistor in series with a parallel combination of a 0.25 F capacitor and a 2 resistor.
s+3 s+6 2 2 1 1
(e) Zin (s) = + = 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.
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 page P14.9 R. A. DeCarlo, P. M. Lin
s2 + 1 0.25s2 + 1 1 4 1
(e) Zin (s) = + = 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
(f) Zin (s) = + = s+ + . This circuit is a 1 H inductor in series with a 1 F
s 0.25s + 1
2 s s+ 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
s 0.25s + 1
2 s s+ 1
0.25s
inductor which is in parallel with a series connection of a 1 H inductor and a 0.25 F capacitor.
SOLUTION 14.13.
With L[v out (t)] = Vo(s) and L[v in (t)] = Vi (s) and vout (0 ) = 0,
100 10
sV0 (s) + 25V0 (s) + 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 + +
(s + 20)(s + 5) 3 s + 20 s + 5 (s + 5) 2 (s + 5)3
5s 10 110 110 22
K1 = = = =
(s + 5) s =20 (15)
3 3 3375 675
5s 10
p(s) =
s + 20
(s + 20)(5) (5s = 10) 110
p(s) = = = 110(s + 20) 2
(s + 20) 2 (s + 20) 2
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 R. A. DeCarlo, P. M. Lin
220 220 44
p(5) = 3 = =
(15) 3375 675
Then
p(5) 1575 p(5) 330 p (5) 1 44 22
C3 = = , C2 = = , C1 = = =
01 675 1! 675 2! 2 675 675
and
1 22 22 330 1575
Vout (s) = +
675 5 + 20 s + 5 (s + 5) 2 (s + 5) 3
This yields
1 1575 2 5t
vout (t) = 22e 20t 22e 5t + 330te 5t t e u(t) V
675 2
(b) If vin (t) = u(t) V,
1
Vin (s) =
s
and
5s 10 K K K3
Vout (s) = = 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
5s 10 25 10 35 7
K2 = = = =
s(s + 20) s= 5 (5)(+15) 75 15
and
5s 10 100 10 110 11
K3 = = = =
s(s + 5) s =20 (20)(15) 300 30
Thus
7 11 1
vout (t) = e 5t e 20t u(t) V
15 30 10
By virtue of linearity and time invariance, if vin (t) = [u(t) u(t 0.5)] V ,
7 11 1
vout (t) = e 5t e 20t u(t)
15 30 10
7 11 20(t 0.5) 1
e 5(t 0.5) e u(t - 0.5) V
15 30 10
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
I (s) s + 1 2 2
H(s) = out = s =
Vin (s) s
2
s +1
SOLUTION 14.15.
1
Here vin (t) = tet u(t) V which implies that Vin (s) = . 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
1 1 1 1 s
Vout (s) = + + + 2 2
s + 1 (s + 1)2 (s + 1) 3 s +1 s +1
(a) Hence
V (s) 1 1 1 s 1 2
H(s) = out = + + (s + 1)
Vin (s) s + 1 (s + 1) 2 (s + 1) 3 s + 1
2
1 (s 1)(s + 1)2
= (s + 1)+ 1 +
(s + 1) s2 + 1
Simplifying
s3 + 2s2 + 5s + 2
H(s) =
(s + 1)(s2 + 1)
1 1 s +1
(b) If vin (t) = (1+ t)u(t) V, then Vin (s) = + 2 = 2 . Hence
s s s
s + 2s2 + 5s + 2 5 2
3
4s
Vout (s) = H (s)Vin (s) = = + 2 2
s (s + 1)
2 2 s s s +1
implying that
vout (t) = [5 + 2t 4cos(t)]u(t) V
SOLTUION 14.16.
(a) By a voltage divider (Fig. P14.16a)
Z4 (s)
Vout (s) = V (s)
Z3(s) + Z4 (s) in
and
Z4 (s)
H(s) =
Z3(s) + Z4 (s)
(b) In Fig. P14.16b,
Yin (s) = Y1(s) + Y2 (s)
and
1 1
Vout (s) = 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,
1
Z3 (s) + Z4 (s) 1
Iout (s) = Iin (s) =
Y1 (s) + Y2 (s) +
1 [Y1(s) + Y2 (s)][Z3(s) + Z4 (s)] + 1 Iin (s)
Z3 (s) + Z4 (s)
Hence
Z4 (s)
Vout (s) = Z4 (s)Iout (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
10 s + 103
V 4 1 10
H(s) = 0 = = =
Vi 103 + 1 0.1s + 2 s + 20
4 3
10 s + 10
(a) Vout (s) =
400
s(s + 20)
=
20
20
s s + 20 ( )
v out (t) = 20 20e 20t u(t) V. Plot omitted.
(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
V0(s) = H (s)Vi (s) = vout (t) = 400te 20t u(t) V
(s + 20) 2
40
(e) If vin = 40te20t u(t) V , then Vi (s) = . Hence,
(s + 20)2
400
V0(s) = H (s)Vi (s) = v out (t) = 200t 2e 20t u(t) V
(s + 20) 3
2/23/02 page P14.13 R. A. DeCarlo, P. M. Lin
SOLUTION 14.18.
(a) By voltage division
2s + 1 2s + 1
s s 2s + 1
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
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)
2s + 1 64 128s + 64
Vout (s) = H (s)Vin (s) = =
4s + 3 s2 + 64 (4 s + 3)(s2 + 64)
0.12391s + 31.907 0.12391
=
s2 + 64 s + 0.75
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 R. A. DeCarlo, P. M. Lin
128/1033
ans =
1.2391e-01
4120/1033
ans =
3.9884e+00
(
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 R L 1 1
LCs2 + s + 1 R R
+ Cs + 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
3 s2 1 3s2 K1 C1 C2
IC (s) = H (s)Iin (s) = = = + +
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=
[]
6 4.5 1.5
IC (s) = +
s + 2 s + 1 (s + 1)2
and
9 3
iC (t) = 6e 2t e t + tet u(t) A
2 2
SOLUTION 14.20.
(a) Make a source transformation:
1 250
Vin (s) = Iin (s) = I (s)
Cs s in
By voltage division
10 250 2500
Vout (s) = s Iin (s) = 1 2 I (s)
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 =
Iin (s) s(s + 200s + 10,000)
2
(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) = =
s(s + 100) 2 s s + 100 (s + 100) 2
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
[ ]
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)
Yin = 1 + 1 + 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)
( s + 100) (s + 300)
Iout (s) = H(s) = = 1 - 50 - 150
( s+ 200) (s + 400) s + 200 s + 400
Hence
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 R. A. DeCarlo, P. M. Lin
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
15
10
iout in mA
-5
-10
0 0.01 0.02 0.03 0.04 0.05
time in second
1
YLC (s) = Cs +
Ls
Using voltage division,
2/23/02 page P14.18 R. A. DeCarlo, P. M. Lin
1 1
C1s + C1s +
Ls Ls LC1s2 + 1
Vout (s) = 1 1 Vin (s) = 2 Vin (s) = L(C + C )s2 + 2 Vin (s)
C1s +
Ls
+ C2 s +
Ls
( C1 + C2 )s +
Ls
1 2
Finally
1
s2 +
Vout (s) C1 LC1 s2 + 4 10 6
H(s) = = = 0.2 2
Vin (s) (C1 + C2 ) s2 + 2 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
1
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.
1 R1C1s + 1
Y1(s) = C1s + =
R1 R1
1 R C s +1
Y2(s) = C2 s + = 2 2
R2 R2
R1 R2
Then Z1(s) = and Z2(s) = . By voltage division,
R1C1s + 1 R2C2s + 1
R2 R2
V (s)
R2C2 S + 1 R2C2s + 1 in
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
V (s) R2 (R1C1s + 1)
H(s) = out =
Vin (s) (C1 + C2 )R1R2s + R1 + R2
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 = =
6R2 s + 5R2 s 5 3 s(s + 5 )
6R22 s(s + ) 6R2
6R2
R2 (R1C1s + 1) R2 s + R2
H(s) = =
R1R2C1s + R1R2C2 s + R1 + R2 [ R2 (R1C1) + R1 (R2C2 )]s + R1 + R2
R2 ( s + 1) R2
= =
(R1R2 )( s + 1) R2 + R1
R2
Vout (s) = 10u(t)
R2 + R1
(d) Using H(s) from part (c) with the requirement that R1C1 = R2C2 , then
R2 1
H(s) = =
R1 + R2 10
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
1 RCs + 1
YR (s) = Cs + =
R R
so that
R
ZR (s) =
RCs + 1
Then
R
Vout (s) = ZR (s) Ib (s) =
2000(RCs + 1)
(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
[Vout (s) Vin (s)] Vout (s) + 2sVout (s) = Vin (s) + + 2s Vout (s) = 0
2 2 2
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
2000 5 K1 K K*
VC (s) = = + +
s(s + 10 j20)(s + 10 + j20) s s + 10 j20 s + 10 + j20
2000 5 2000 5
= = = 4.472 + j2236 = Jeij
(10 + j20)( j40) (800 + j400)
where = 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
Here
2/23/02 page P14.22 R. A. DeCarlo, P. M. Lin
A = 4.472, B = 2.236, A2 + B2 = 5
and
B 2.236 1
arc tan = 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
Vx (s) = s + 100
40s 0.5s
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
0.5s 20 20 s 10 10
Vx (s) = =
s + 50 s s + 40 s + 50 s s + 40
and
2.5 2.5 1 1 0.05 0.25 0.2
IL (s) = V x (s) = = + +
s s + 50 s s + 40 s s + 40 s + 50
Hence
[ ]
iL (t) = 0.05 + 0.2e 50t 0.25e 40t u(t) A
20 20
(b) Vin (s) = . 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
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
V (s) V (s)
C1sVin (s) + in = C2sVout (s) out
R1 R2
R1C1s + 1 R2C2s + 1
R Vin (s) = Vout (s)
1 R2
2/23/02 page P14.24 R. A. DeCarlo, P. M. Lin
Then
1 1
s+ s+
V (s) R R C s + 1 RRC R1C1 C R1C1
H(s) = out = 2 1 1 = 1 2 1 = 1
Vin (s) R1 R2C2 s + 1 R1RC2 s + 1 C2 s + 1
R2C2 R2C2
(c) If
H(s) = 5, C2 = 1 F and R2 = 1M
then
C1
= 5 and C1 = 5 F
C2
The bracketed term must cancel and with
R2C2 = 106 (106 ) = 1
Then with C1 = 5 F
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
s +1
H(s) = 5
s+2
with C2 = 1 F
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 = =
+ R2 R2Cs + 1 C(s +
1
)
Cs R2C
and
1 1 1
R + C(s + R C ) Vin (s) = C(s + R C )Vout (s)
1 2 2
1 1
R1C s + + 1Vin (s) = R1C s +
R2C out
V (s)
R2C
and
1 1 1
R1C(s + ) +1 s + +
V (s) R2C R2C R1C
H(s) = out = =
Vin (s) 1 1
R1C(s + ) s+
R2C R1C
1 1
Here + =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
1 250,000 1 250,000
Zin,1 = 25k, Zin,2 = 50k , Z f ,1 = = , and Z f ,2 = =
1 s+5 1 s + 2.5
C s+ C s+
RC RC
250,000 250,000
Ha (s) = s + 2.5 s + 5.0 =
50
25,000 50,000 (s + 2.5)(s + 5)
1
If vin (t) = u(t), then Vin (s) = and
s
50 4 8 4
Vout (s) = = +
s( s + 2.5)(s + 5) s s + 2.5 s + 5
Hence,
[ ]
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
s + 5 10 s
Hb(s) = s + 2.5 250,000
10s
= =
25,000 s + 2.5 s + 5 5
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
V1 (s) s
(c) This subcircuit is again an integrator, with
V1(t) = - 1
V2 (s) s
(d) Applying KCL to the inverting input terminal of the top left op amp, we have
(e) Applying KCL to the inverting input terminal of the bottom op amp, we have
V1(s) = - sV out(s)
H(s) = Vout(s) = G1
Vin (s) 2
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 s 1 1
10 + 40 + 10 10 vC (0 ) = 0 implies 10 + 50 VC (s) = 10 (20) = 2
s
20
Equivalently, (50s + 10)VC (s) = (s + 0.2)VC (s) = 20 or VC (s) = . Therefore,
s + 0.2
vC (t) = 20e 0.2t u(t) V
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
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
2 1
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
1
0 = 0.5Vout 0.2VR1 + (V + 0.8)
5 + 0.4s out
= 0.5Vout + 5 0.2
1
5 + 0.4s
(Vout + 0.8) +
1
(V + 0.8)
5 + 0.4s out
= 0.5Vout +
2
(V + 0.8)
5 + 0.4s out
Therefore
8
Vout =
s + 22.5
and
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
VC (s)
+ CsVC (s) = Cv(0 ) + Iin (s)
R
With R = 50 and C = 0.02 F ,
VC (s)
+ 0.02sVC (s) = 0.02vC (0 ) + Iin (s)
50
or
VC (s) + sVC (s) = vC (0 ) + 50Iin (s)
which is equivalent to
(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 ,
10
(s + 1)VC (s) = 8 + 2 implies VC (s) =
s +1
and
vC (t) = 10e 10t u(t) V
0.2
(b) With vC (0 ) = 1 V and iin (t) = 200e t u(t) mA we have that Iin (s) = . Thus
s +1
v C (0 ) 50 1 10
VC (s) = + Iin (s) = +
(s + 1) (s + 1) (s + 1) (s + 1)2
and
( )
vC (t) = e t + 10tet u(t) V
SOLUTION 14.36.
2/23/02 page P14.30 R. A. DeCarlo, P. M. Lin
H(s) = IL(s) = 1
Iin (s) Ls + 1
(b) The given data in Laplace transforms are:
- -
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
Given iL(0-) = 0.01 A and vin(t) = 2e-200tu(t) V, then Vin(s) = 2/(s + 200) and
2 + 0.02
I L(s) = s + 200 = 0.01s + 3
2s + 200 ( s + 100)( s + 200)
= 0.02 - 0.01
s + 100 s + 200
in which case
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) = 2s
s2 + 40000
and
2/23/02 page P14.32 R. A. DeCarlo, P. M. Lin
2s + 0.02
I L(s) = s2 + 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
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
1
+ Cs
[
CvC (0 ) =
]
s + 0.25
R
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 in s + 0.25 s + 0.25 in s + 0.25
for all inputs and initial conditions.
20
(a) If vin (t) = 20u(t) and vC (0 ) = 10 V, then Vin (s) = and
s
VC (s) =
5
+
10
s(s + 0.25) s + 0.25
=
20
10
s s + 0.25 ( )
v C (t) = 20 10e 0.25t u(t) V
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
5
(b) If vin (t) = 5e 0.25t u(t) V, then Vin (s) = . Hence,
s + 0.25
1.25 10
VC (s) = + v C (t) = (10 + 1.25t)e 0.25t u(t) V
(s + 0.25) 2 s + 0.25
and
0.25s 0.5 0.25 0.0625
IC (s) = =
(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
8
Input and Capacitor voltages, V
3
TextEnd
2
0
0 5 10 15 20
Time in s
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
s = C = 20s + 10 26.6 6.6
VC (s) = =
1 1 1 1 s + 250s + 10
2 4 s + 200 s + 50
Cs + + s2 + s+
R Ls RC LC
Hence
200t 50t
vC(t) = [26.6e 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)
Ls 1 Vin (s) 20000 2 0.6667 2.6667
VC (s) = = = = +
1 1 LC 2 1 1 s 3 + 250s 2 + 104 s s s + 200 s + 50
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
200t 50t
vC(t) = [2 + 27.267e 9.2667e ]u(t) V
(d) By linearity and time-invariance,
200t 50t
vC(t) = [2 + 0.6667e 2.6667e ]u(t)
200(t0.01) 50(t0.01)
+ [4 + 1.3334e 5.3334e ]u(t 0.01) V
2/23/02 page P14.4 R. A. DeCarlo, P. M. Lin
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
( )
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,
1
+1 s +1 4
Iout (s) = 1 s [CvC (0 )] = 4 2 = iout (t) = 4e t u(t) A
s + 2s + 1 s + 1
+1 + s +1
s
(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
4 4 s2 + 2s + 4 8 s+8
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
s2 + 14s 4
Veq (s) =
(
s s2 + 2s + 4 )
By a voltage division,
1
s vC (0 ) 2 s2 + 14s 4 1
Vout (s) = 1 Veq (s) + = +
s
+ 0.5 s (
s + 2 s s2 + 2s + 4 s)
=
(
2s2 + 28s 8 + (s + 2) s2 + 2s + 4 ) = (s2 + 6s + 36)
(
s( s + 2) s2 + 2s + 4 ) (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=
Hence
[ ( )
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
vC1 (0-)/s
I1K(s) = = - 0.25/s = -0.25
R+ 1 1000 + 50/s 1000s + 50
sC 1
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
v (0-)/s - 2/s -2 -810-4
I2.5K (s) = C1 = = =
R1 + 1 2500 + 5000/s 2500s + 5000 s+2
sC 1
Next,
1 -810-4 1 -4
Vout (s) =VC2 (s) = I 2.5K (s) = =
sC 2 + 1 s + 2 0.0002s + 0.0002 (s + 2) (s + 1)
R2
(c) Hence
Vout (s) = -4 =-4 1 - 1
(s + 2) (s + 1) s + 1 s +2
and
( )
vout (t) = 4 e 2t e t u(t) V
(d) SPICE plot omitted.
R2
V Z R2 2s + 1
C R2 C1 s
H(s) = C2 = 2 = =
Vin Z1 + Z2 1 + R1 + R2 2
R1 C1 R2 C2 s + ( R1 C1 +R2 C2 + R2 C1 )s + 1
C1 s R2 C2 s + 1
( )
vC 2 (t) = 7 e 0.25t e 4t u(t) V
15
10
vin and vc2
TextEnd
5
0
0 1 2 3 4 5 6 7 8
Time in s
( )
vC 2 (t) = 7 e 0.25t e 4t u(t) V
(e) By linearity, the answer is the sum of parts (b), (c) and (d).
R2 R2 Vin
IL2 = I =
R2 + L2 s L1 (R1 + L1 s) (R2 + L2 s) + + R2 L2 s
Therefore
H(s) = IL2 = R2 G1
=
Vin (R1 + L1 s) (R2 + L2 s) + + R2 L2 s 2
G1 L1 G2 L2 s + (G 1 L1 + G 2 L2 + G 1 L2 )s + 1
H(s) = 2 = 2
4 7 2 4 7 7
21 s + (21+ + 2 )s + 1 s 2 + 4.25s + 1
7 8 7 8 8
IL2 = H(s)Vin = 2 15 = 30 32 2
s 2 + 4.25 s + 1 s s - s+ 0.25 - s+ 4
Hence
( )
iL 2 (t) = 30 32e 0.25t + e 4t u(t) A
Plot omitted.
7 15
L2 iL2(0-) 8 15( s + 2.25) 8
IL2(s) = = = = + 7
R2 (R1 + L1 s) 7 1.75(0.5 + s) s 2 +4.25s + 1 s + 0.25 s + 4
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).
H2 (s) = IL2 = R2 2
=
VC2 (R1 + L1 s) (R2 + L2 s) + + R2 L2 s 2
s + 4.25s + 1
(c)
(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
SOLUTION 14.47. (a) For this passive circuit, we may write the nodal equations by inspection.
0.8s + 2 + 10 - 10
s s VC = 2Vs1
VR -Is2
- 10 1 + 10
s s
(b) Vs1= 3/s and Is2 = 3/s. We solve for VR by Cramer's rule to obtain
0.8s + 2 + 10 6
s s
- 10 - 3s
VR = s = - 2.4s2 - 6s + 30 = - 4 + 1
0.8s + 2 + 10 - 10 s(0.8s2 + 10s + 30) s + 7.5 s
s s
- 10 1 + 10
s s
and
( )
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 - 10 6 + 2.4
s s VC s
=
VR
- 10 1 + 10 - 3s
s s
0.8s + 2 + 10 6 + 2.4
s s
- 10 - 3s 2
VR = s = - 2.4s +18s + 30 = s1 - 16 + 12
0.8s + 2 + 10 - 10 2
s(0.8s + 10s + 30) s + 7.5 s + 5
s s
- 10 1 + 10
s s
2/23/02 page P14.11 R. A. DeCarlo, P. M. Lin
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:
v (0)
0.5 + 1.25 - 1.25 Vs1 - Cs
s s Is1 =
IL vC(0)
- 1.25 1 + 0.1s + 1.25
s s 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
vC(0)
0.5 + 1.25 - 1.25 Vs1 - 3
s s Is1 = s s
=
IL vC(0)
- 1.25 1 + 0.1s + 1.25 0.3 + 3s
s s s + LiL(0) + I s2
SOLUTION 14.49. (a) Represent the initialized capacitors by their parallel equivalent circuits.
(b) Write two nodals equation by inspection
2.4 + 0.006
0.001s + 0.4 - 0.2 VC1 = 0.2V in + 0.001vC1 (0-) = s
- 0.2 0.001s + 0.4 VC2 0.001v C2 (0-) 0.002
SOLUTION 14.50. (a) Let VC denote the node voltage across the capacitor. By inspection the nodal
equations in matrix form are:
1+ 1 /R + 4s 1 /R VC Vin
1 /R =
1 + 1/ R + 1 / ( s4) Vout Vin / ( 4s)
1 + 1/ R + 4s 1
det
Vout (s) 1 /R 1 / ( 4s) ( 4s + 1)(1 + 1/ R)
H(s) = = =
Vin (s)
det
1+ 1 /R + 4s 1 /R (
4s (1+ 1/ R + 4s)(1+ 4s / R + 4s) 1/ R 2 )
1 /R 1 + 1/ R + 1 / ( s4)
(4s + 1)(1 + 1/ R) 1
= =
(1 + 8s + 16s )(1+ 1 /R) (1+ 4s)
2
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)
4s (1 + 4s)2
Zin (s) = = 2 =1
1 ( + )
1+ + (1+ 4s) 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) =
0.25 a ( )
e e 0.25t u(t) V
For a = 0.25,
2/23/02 page P14.13 R. A. DeCarlo, P. M. Lin
2.5
Vout (s) = implying that vout (t) = 2.5te0.25t u(t) V
(s + 0.25)2
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.
0.5(s+ 8)
Vout (s) = 4 Vin (s) + iL(0-) + 0.5s vC(0-)
s+4 (s+4) 2 (s+4) 2
Vin (s) = 4s - 3 = s + 4
s+1 s( s + 1)
and
4 0.25(s+ 8)
Vout (s) = + + 0.75s
s(s + 1) (s+4) 2 (s+4) 2
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 V
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
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
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 R. A. DeCarlo, P. M. Lin
Vin (s) + 0.5vC1 (0) + 0.25s + 1 vC2 (0) +( 0.125s2 + s +1.5 )vC3 (0)
VC3 (s) =
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
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
I1 0
s
1000
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 R. A. DeCarlo, P. M. Lin
1000
+ 0.016s 0 0
s
1000
det 2 + + 0.016s 2 V1
s
2 7 V1 V2 1000
+ 0.016s (5V1 + 2V2 )
s
Vout = =
1000 6000 7000
+ 0.016s 0 1 + 0.096s 10 + + 0.112s
s s s
1000
det 2 + + 0.016s 2 2
s
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
2G 0 Va gm + G 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
(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 s2 + 1 LC
Vout =
ZLC
Rth + ZLC
Voc = 1 Cs Voc = Rth (5V1 + 2V2 )
+ Ls + Rth s +
2 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
R + Cs Iin 1
det 0 0 1
1 0 z0 (s)
= Iin Iin
V2 = =
1 1 1 1
R + Cs 1 R 1 2 + Cs z0 (s)Cs + Cs + Cs (2 + z0 (s)Cs)
R R R
det 0 Cs 1
1 1 z0 (s)
(c) Here
2/23/02 page P14.20 R. A. DeCarlo, P. M. Lin
Iin
V2 =
1
R
(
+ Cs 2 + LCs 2
)
2
in which case = .
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
For t > 5, the inductor decays with a time constant of 0.1 s. Thus
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
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
(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) =
1 s + 12.5 in 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
- -
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,
5 5 10
VC = = =
1 1 (s + 1) 2 + 2
s 1+ 0.5s + 0.5s2 + s +
0.184s 0.184
- +
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
37.5 1 150
VC (s) = =
s 1 s2 + 4
Cs +
Ls
Hence from table 13.1,
iC (1 ) = C
dvC
dt t =1 [ ]
= 0.25 100e 1 + 300e 2 200e 4 = 0.037378
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:
+
s 4s iL (1 ) + 4s 10.634
VC (s) = e + Cv (1 ) = e s 2 + 4.3507
s + 4
2 s C
s +4 s
Therefore from table 13.1, for t > 1,
Plots omitted.
Therefore for 0 t 1,
( )
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
s 20 s+2
10 +
s
Therefore,
10 4.7281 4.7281 4.7281 s 2.3641 2.3641
Vout (s) = IC (s) + e s = e s + =e +
s s s(s + 2) 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.
0.005s + 0.03 0.01 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 =
0.005s + 0.03 0.01 s + 10s + 16 s + 2 s + 8
det
0.01 0.0025s + 0.01
Therefore for t 0,
40 2t 10 8t
vout (t) = e e V
3 3
1 2s
SOLUTION 14.66. (a) Zin (s) = 2 +
2 = 2 + s2 + 4
.
0.5s +
s
(b) Here, the initial condition is zero and
2/23/02 page P14.25 R. A. DeCarlo, P. M. Lin
2s
2
VC (s) = s + 4 10 = 10
2s s s +s+ 4
2
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,
(d) Therefore
1 s+4
e1.5sVC (s) = s 0.5 0.57237 = 0.57237 s2 + 10s + 16
0.5s + 2 +
s+ 4
(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
s+4 0.19079 0.38158
VC (s) = e 1.5s 0.57237 2 = e 1.5s +
s + 10s + 16 s+2 s + 8
(f) Finally
[ ]
vC (t) = 0.19079e 2(t 1.5) + 0.38158e 8(t 1.5) u(t 1.5) V
R
SOLUTION 14.67. (a) v1(0 ) = v1 (0+ ) = v 2 (0 ) = v2 (0 + ) = 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.
2/23/02 page P14.26 R. A. DeCarlo, P. M. Lin
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
5 + 0.1s 1 175
Iout (s) = = 2
1 1 s s + 275s + 2500
0.057143 + +
0.35s 5 + 0.1s
Use MATLAB to do the partial fraction expansion
r=
0.6831
-0.6831
p=
-265.5869
-9.4131
V1(s) = 4 106 s
30 30 5
= =
1 1 1 s 6s s
6 + 6 +
4 10 s 4 10 s 1 106 s
Therefore, v1(0 + ) = 5 V.
(b) Again use voltage division:
V1(s) = 5 10 6 s
40 80
=
1 1 1 s 17s
+ +
5 106 s 1 10 6 s 2 106 s
Therefore, v1(0 + ) = 4.7059 V.
SOLUTION 14.70. (a) Consider a mesh current I(s) in the usual direction and use the series equivalent
circuit for each capacitor. Thus
SOLUTION 14.71. (a) For 0 < t < 2, the 150 mF capacitor is charged to 25 V. From conservation of
charge,
SOLUTION 14.72. (a) Let the middle node have voltage Va(s). Then writing node equations
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
10
V1(s) = V2 (s) = v1(t) = v 2 (t) = 10 V for 0 < t < 1.
s
Hence with the switches in position B, let us write a single node equation using the parallel equivalent
circuit the initialized capacitors:
11
0.002sV2 (s) 0.002 10 + 0.004sV2 (s) + 0.004 10 = 0.004s
s
Equivalently
0.006sV2 (s) = 0.044 0.02 = 0.024 V2 (s) = 4 / s
SOLUTION 14.74. (a) At t = 0+, the frequency domain equivalent circuit is given below.
By superposition,
1 8 1 1 6
V1 = + 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.
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
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
s s R K1
I(s) = =
1 1 1 1 s + p1
R+ + s + +
C1s C2s RC1 RC2
in which case i(t) = K1e p1t u(t) where
ab 1 1
K1 = and p1 = +
RC1 RC2
.
R
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
1 1
W (0 ) = C1a2 + C2b2
2 2
Also at t = ,
K K
vC1 () = a 1 and vC 2 () = b + 1 .
C1 p1 C2 p1
Hence
2/23/02 page P14.35 R. A. DeCarlo, P. M. Lin
2 2
1 K 1 K
W () = C1 a 1 + C2 b + 1
2 C1 p1 2 C2 p1
(c)
RK12 (a b)2 (a b)2
Ri (t)dt = RK12e 2 p1t dt =
2
= =
2p1 2Rp1 1 1
0 0 2 +
C1 C2
Observe that
2 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
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
lim I(s) = 1 s s = a b = C1C2 (a b)
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 )
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:
(d) (i) Using the series equivalent circuit for C1, we have
1 1 1 /R 1 2t R
I(s) = = i(t) = e u(t) A
2 s s + 2 /R R
R+
s
Thus
VC 2 (s) =
I(s)
s
=
1/ R
s(s + 2 / R)
=
0.5
s
0.5
s + 2 /R ( )
vC 2 (t) = 0.5 1 e 2t R u(t) V
and
VC1(s) =
I(s) 1
s
+ =
1/ R 1 0.5
+ =
s s(s + 2 / R) s s
+
0.5
s + 2 /R (
vC1 (t) = 0 . 5 1+ e 2t R u(t) V )
(ii) The energy dissipated in the resistor is given by
1 4t R e 4t R 1
WR (0,) = R i ( )d = e
2
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
Further, as R 0, i(t) = 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.
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).
Yin (s)
= 0.225s +0.075 = 0.1 + 0.125
s (s+ 0.2)(s + 0.5) s+ 0.2 s + 0.5
Hence
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
Za (s) = 1 = s 2 + 1 = 2s + 1
Ya (s) 0.5s 0.5s
Zb (s) = 1 = s 2 +2 = 0.5s + 1
Yb (s) 2s s
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
1 .02 20 103
Vout (s) = =
1 s(s + 102)
+ 106 s s
98039
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
5
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)
(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
Hence
0.02
5 10 6 + 3
Vout (s) = s = 5s + 20 10
1 s(s + 102)
10 6 s +
9803.9
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
( )
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
Cs i (0-) 1 108
VC (s) = LiL(0-) = L =
1 + Ls + R C s2 + R s + 1 s2 + 125s + 1.25 109
Cs L LC
A plot of vC(t) is given below with the vertical axis in V and the horizontal axis in seconds.
3000
2000
1000
-1000
-2000
-3000
0 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
0
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
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
Case 2: t 0.
t
]
t
f 2(t) f 3 (t) = 8 e 2 d = 4e 2 = 4(1 e 2t )
0
0
In sum,
0, t<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 = 5 (t )4e 2 u( )d =
= 20e 2
u( ) ] =t = 20e2t u(t)
(d) By the definition of the convolution integral
1/25/02 P16-2 R. A. DeCarlo, P. M. Lin
2(t ) 2 2t
f 3(t) f 3(t) = f 3 (t ) f 3( )d = 4e u(t )4e u( )d = 16e 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
(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)
Using the results of parts (a) and (b) the result follows immediately
0, t<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( T1)K2 u(t T2 )d = K1K2 u(t T2 )d
T1
Here observe that u(t T2 ) = 0 for > t T2 . Hence there are two cases to consider:
t T2 < T1 and t T2 T1 .
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,
0, t < T1 + T2
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 u(t + T1 )d
T 2
Here observe that u(t + T1) = 0 for > t + T1. Hence there are two cases to consider:
t + T1 < T2 and t + T1 T2 .
Case 1: t + T1 < T2 . Here u(t + T1) = 0 , since is restricted to the interval [T2 ,): f 3(t) = 0.
Case 2: t + T1 T2 .
t +T 1
f 3(t) = K1K2 d = K1K 2 (t + T1 + T2 )
T 2
In sum,
0, t < T1 T2
f 3(t) =
K1K 2 (t T2 T1), t T1 T2
(c) By definition
a
f 3(t) = f1(t ) f 2( )d = K1u(t )K 2e u( )d = K1K 2 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.
The integrand is nonzero only when t + T1 . Hence, there are two cases to consider: t + T1 < 0
and t + T1 0.
Case 1: t + T1 < 0. Here u(t + T1) = 0 , since is restricted to the interval [0,). Hence
f 3(t) = 0.
Case 2: t + T1 0.
t +T 1 t +T1
e a
f 3(t) = K1K2 e a
d = K1K 2
a
0
KK
a [
= 1 2 1 e a(t +T 1) . ]
0
Therefore
KK
[
f 3(t) = 1 2 1 e a(t +T1 ) u(t + T1 ).
a ]
(e) By definition
a
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
a e a KK
f 3(t) = K1K2 e d = K1K2 = 1 2 , for t 0.
a 0 a
0
Case 2: t > 0.
a e a KK
f 3(t) = K1K2 e d = K1K2 = 1 2 e at , for t > 0.
a a
t t
In sum,
K1K 2
a , t0
f 3(t) = K K
1 2 e at , t > 0
a
Solution 16.3.
(a) By definition
a a(t )
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 t. Hence, there are two cases to consider: t < 0
and t 0.
t
at
f 3(t) = K1K2e d = K1K 2e at t , for t 0.
0
In sum,
Cs
Ic (s) = 1 Iin (s)
Cs +
R
(b) By definition
a=1
K =1
The integrand, u(t ) , is nonzero only when t. Hence, there are two cases to consider: t < 0
and t 0.
(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
0, t<0
f 3(t) = 10t
1000e t, t 0
t
] = 1,
t
For part (ii) the parameters have the following values: vc (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
3
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.
3
]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
(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
t
]1
t
y(t) = 2 e 2( t) d = e 2( t) = 1 e 2(1t ) , for 1 t 3.
1
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 R. A. DeCarlo, P. M. Lin
SOLUTION 16.5.
(a) By definition
f 4 (t) = 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 = (t 2)e 2 u( )d
=e 2
u( ) ] =t 2 = e
2(t 2)
u(t 2)
(c) By definition
2
f 6(t) = 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
]0
t +1
f 6(t) = 2 e 2 d = e 2 = 1 e 2(t +1)
0
It follows that
[ ]
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 R. A. DeCarlo, P. M. Lin
f 7(t) = f1 (t ) f 3( 2)d = (t 2)e 2( 2) u( 2)d =
=e 2( 2)
u( 2) ] =t 2 = e2(t 4)u(t 4) .
SOLUTION 16.6.
(a) By definition
[
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 t
e a e at
y(t ) = K e d = K
a =
a
K
a
Case 2: t > 0
t 0 0
e a K
y(t) = K e a d = K e a d = K =
a a
(b) By definition
a a
y(t) = K u( t)e u()d = K u()e d
t
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 a K
y(t) = K e a d = K =
a 0 a
0
Case 2: t 0
e a K at
y(t) = K e a d = K = e
a t a
t
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 ) ( 4)d =
=e a(t )
u(t ) ] =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 R. A. DeCarlo, P. M. Lin
(b) By definition
u(t
2 2
f 6(t) = f1 (t ) f1( )d = K u(t )u( )d = K )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. Hence
f 6(t) = 0 , for t < 0.
Case 2: 0 t.
t
d
2
f 6(t) = K = K 2t , for 0 t.
0
A picture of f 6(t) , for K = 1, is sketched in the next figure.
(c) By definition
a
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 =
K a t K
a
e
0 ]
= 1 e at , for 0 t.
a ( )
0
1/25/02 P16-11 R. A. DeCarlo, P. M. Lin
(d) By definition
0
Ku(t
a
f8 (t) = f1(t ) f 3( )d = )e u( )d = K e a u(t )d
The integrand, u (t ) , is nonzero only when t . This suggests two regions of consideration:
t 0 and 0 < t .
Case 1: t 0 .
t
f8 (t) = K ea u(t )d =
K a t
a
e
K
a]
= e at , for t 0.
Case 2: 0 < t.
0
f8 (t) = K ea u(t )d =
K a 0
a
e
K
]
= , for 0 < t.
a
In sum,
K at
a e , t 0
f8 (t) = K
, 0< t
a
SOLUTION 16.8.
(a) Using the current division formula
Cs
Ic (s) =
1 Iin (s)
Cs +
R
By Ohms law the Laplace transform of capacitors voltage
1
Vc (s) = I (s)
Cs c
Therefore the transfer function of the circuit
V (s) 1 1
H(s) = c = =
Iin (s) Cs + 1 s+ 4
R
Taking the inverse Laplace transform of H(s) yields the impulse response h(t) = e 4t u(t) .
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
t
]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
1
Vout (s) = Cs
1 Vin (s)
R+
Cs
Therefore the transfer function
1
V (s) 1 Cs
H(s) = out = =
Vin (s) R + 1 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 = [ ]
e (t ) u(t ) u( ) + 2e 2 u( ) d =
0
(t )
= e 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
] = 1, for t < 0.
t
vc (t) = e (t ) d = e t
Case 2: 0 t.
0 t
(t )
vc (t) = e d + 2 e (t + ) d
0
] ]0 ( )
0 t
= e t 2e (t ) = e t 2 e 2t e t = 3e t 2e 2t , for 0 t
vout (t) = h(t )vin ( )d = e (t ) u(t )e a| |d
0
e (t ) u(t )e a d + e (t )
= u(t )e a d
0
0
t t
e e
(a+1) (1a)
=e u(t )d + e 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 (a+1) t
t t e e at
(a+1)
vout (t) = e e d =e = , for t < 0.
a + 1 a +1
Case 2: 0 t.
0 t
t t
e
(a+1) (1a)
vout (t) = e e d +e d =
0
0 t
e (a+1)
t t
t t e t
+e e + e e (1a) d
(1a)
=e d =
a + 1 a +1
0 0
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)
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
The integrand is nonzero only when t. This suggests three regions of consideration: t < 1,
1 t < 1 and 1 t.
Case 1: t < 1. y(t) = 0
Case 2: 1 t < 1.
t
[ ]
y(t) = 2 e 0.2( t ) d = 10 e 0.2(t +1) 1 , for 1 t < 1.
1
Case 3: 1 t.
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)
y(t) = h(t) v(t) = h(t )v( )d
The integrand is nonzero only when t. This suggests three regions of consideration: t < T ,
T t < T and T t.
Case 1: t < T . y(t) = 0
Case 2: T t < T .
t
y(t) = K
K
[ ]
e a(t ) d = 1 e a(t +T ) , for T t < T .
a
T
Case 3: T t.
T
y(t) = K
K
(
e a(t ) d = e at e aT e aT , for T t.
a )
T
In sum,
0, t < T
K
[
y(t) = 1 e a(t +T ) ,
a
]
T t < T
(
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
2
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
t 2
(
f 3(t) = 8 ( 2)d = 8 2 = 4 t 2 4t , for 0 t < 2.
2
)
0 0
Case 3: 2 t.
2
2 2
f 3(t) = 8 ( 2)d = 8 2 = 16, for 2 t.
0 2
0
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.
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 =
2 4
= 4 u(t )d + 4 (4 )u(t )d
0 2
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, is restricted
to the interval [2,4]. Therefore
t
f 3(t) = 4 d = 2t 2 , for 0 t < 2.
0
Case 3: 2 t < 4.
t
2 t 2
f 3(t) = 4 d + 4 (4 )d = 8 + 4 4 = 2t 2 + 16t 16, for 2 t < 4.
0 2 2
2
Case 4: 4 t.
4
2 4 2
f 3(t) = 4 d + 4 (4 )d = 8 + 4 4 = 16, for 4 t.
0 2 2
2
In sum,
0, t<0
2t 2,
0 t < 2
f 3(t) = 2
2t + 16t 16, 2 t < 4
16, 4t
2
Vout s 1 2/3 2/3
H(s) = = = =
Vin 2 s + 2.5s + 1 (s + 0.5) (s + 2)
2
2s + 5 +
s
Hence, the impulse response is
2 2
h(t) = e 0.5tu(t) e 2tu(t)
3 3
(b) By definition
20
h(t) * vin (t) = h(t )vin () d = e 0.5(t )u(t ) e u() d
3
20 2(t )
e u(t ) e u() d
3
Case 1: t 0.
t t t t
20 0.5(t ) 20 2(t ) 20 0.5t 1.5 20 2t 3
vout (t) =
3 e e d
3 e e d =
3 e e d
3 e e d
Case 2: t > 0.
20e 0.5t 1.5 0 20e 2t 3 0 20e 0.5t 20e 2t
vout (t) =
4.5
e [ ]
9
e
= [ ]
4.5
9
SOLUTION 16.13.
(a) The impulse response of the circuit has been computed in problem 16.12
2 0.5t 2
h(t) = 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 =
1/25/02 P16-20 R. A. DeCarlo, P. M. Lin
[e 0.5(t ]
20
e 2(t ) u(t )e | |d
)
=
3
20 0.5t 1.5 t
[ ] [ ]
20 t
= e e e 2t e 3 =
4.5 9
= 4.444e t 2.222e t = 2.222e t , for t 0.
Case 2 : 0 < t.
0 t
[e ] [ ]
20 0.5(t ) 2(t ) 20
vout (t) = e e d + e 0.5(t ) e 2(t ) e d =
3 3
0
0 t
[e ] [ ]
20 0.5t +1.5 2t + 3 20
e 0.5t 0.5 e 2t + d =
3
= e d +
3
0
0 t
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
d=
1 7 14 8
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
( ) ( ) (
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) ]
[
8u(t T )* 2e u(t) 2e 2t u(t) + 4e 4t u(t)
t
]
We denote
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
In order to compute y1(t) we will use the following equation which has been obtained in problem
16.2(e)
K1K 2
a , t0
[
at
]
[K1u(t)] K 2e u(t) = K1K 2 at
e , t>0
a
Therefore,
16, t0
y1(t) = t 2t 4t
16e 8e + 8e , 0 < t
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)
A partial fraction expansion of H(s) can be obtained using the residue command in MATLAB:
r=
-2.5000
2.5000
-5.0000
10.0000
p=
-2
0
0
0
k=
[]
Hence
2.5 2.5 5 10
H(s) = + + 2 + 3
s+2 s s s
Taking the inverse Laplace transform yields the impulse response of the cascade
(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:
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
1.25 1.25 2.5 5 10
Y(s) = + + 2 + 3 + 4
s+2 s s s s
Taking the inverse Laplace transform yields the step response of the cascade
SOLUTION 16.16.
(a) By the voltage division formula
1
Cs 1
Vout (s) = V (s) = Vin (s)
1 i 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
1 1
Vin (s) = +
s (s + 1)2
By the impulse response theorem and the convolution theorem it follows that
1 1 1 1 1
Vout (s) = H (s)Vin (s) = + = +
s + 1 s (s + 1)2 s(s + 1) (s + 1)3
1 1 1
Vout (s) = +
s s + 1 (s + 1)3
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)
64
Vout (s) = H (s)Vin (s) =
(s + 10)(s2 + 16)
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
[
Vin (s) = L 2e 2t sin(4t)u(t) =] 8
(s + 2)2 + 16
Therefore
64
Vout (s) = H (s)Vin (s) =
[(s + 2) 2
]
+ 16 (s + 10)
Using again the technique of example 13.14, page 514, Vout (s) can be written as
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 vin (t) 0 for t < 0. The
time domain convolution method will be used to compute the response vout (t). By the impulse
response thorem
[ ]
vout (t) = h(t) vin (t) = 8e 10t u(t) [u( t)]
0.8, t0
vout (t) = 10t
0.8e , 0<t
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)
(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
s 1 s
Vout (s) = H (s)V (s) = = 4
2s2 + 5s + 2 (s + 2) 2 2s + 13s + 30s2 + 28s + 8
3
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,
0.0741 0.0741 0.1111 0.6667
Vout (s) = + + +
s + 0.5 s+2 (s + 2) 2 (s + 2)3
[
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
0
[0.1667e 0.5(t ]
+ 0.6667e 2(t ) u(t )e 2 d
)
=
The integrand of the previous integral is nonzero only when t. This suggests two regions of
consideration: t < 0 and 0 t.
Case 1: t < 0.
t t
0.5t 2t
e
2.5
vout (t) = 0.1667e d + 0.6667e e4 d =
e 2.5 t e 4 t
= 0.1667e 0.5t + 0.6667e 2t =
2.5 4
= 0.1e 2t , for t < 0.
Case 2: 0 t.
0 0
0.5t 2t
e
2.5
vout (t) = 0.1667e d + 0.6667e e4 d =
e 2.5 0 e 4 0
= 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
s5 10
H(s) = = 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 =
= 10 [ ( ) 10e 5 u( )]cos[10(t )]d =
= 10 ( )cos(t )d 10 10e 5 u( )cos[10(t )]d
Using the sifting property of the delta function and expanding cos(t )it follows that
1/25/02 P16-29 R. A. DeCarlo, P. M. Lin
5
vout (t) = 10cos(t ) 100cos(10t) e cos(10 )d 100sin(10t) e 5 sin(10 )d
0 0
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 2
Vout (s) = H (s)W (s) =
s(s + 2)
w
A partial fraction expansion of Vout (s) is
w 1 1
Vout (s) =
s s+2
Taking the inverse Laplace transform yields
(t) = u(t) e 2t u(t)
w
vout
(b) Since
vin (t) = w(t + T )
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
Vout (s) = H (s)W (s) =
2
s(s + 2) ( 1
1 e 2sT = )
1
s s+2 1 e 2sT = ( )
1 1 1 1 2sT
= e
s s+2 s s + 2
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 )
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 T1 )g() d =
f (t' )g() d
= [ f (t' ) * g(t' )]t' =t T
1
t' =t T1
(b) The steps in this part are similar to those of part (a).
1/25/02 P16-31 R. A. DeCarlo, P. M. Lin
f (t T1) * g(t T2 ) = 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
Since t = t' +T1 + T2 , and t and t' are dummy variables, we have
1 1
SOLUTION 16.23. From table 13.1, G(s) = , H(s) = .
s+2 (s + 2) 2
(a) Define p(t) = f(t 2) = u(t).
1
Then, P(s) = . Consider
s
1 0.5 0.5
P(s)G(s) = =
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
1 0.25 0.25 0.5
P(s)H(s) = 2 = s s+ 2
s(s + 2) (s + 2)2
Hence
1/25/02 P16-32 R. A. DeCarlo, P. M. Lin
( )
p(t) * h(t) = 0.25 0.25e 2t 0.5te2t u(t)
[( ) ]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
SOLUTION 16.25.
Define
w(t) = vin (t 2)
Hence
w(t) = u(t)
and, from table 13.1,
1
W (s) =
s
From table 13.1 we also have that
1 1
H1(s) = and H2 (s) =
s +1 (s + 1)2
>> 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 1 1 1 1
Vout (s) = + + +
s s + 1 (s + 1)2 (s + 1)3
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
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.
0
0 1 2 3 4 5 6 7
(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
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.
(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 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
y(t) = x(t) h(t) = x( )h(t )d = x( )[ u(t ) u(t 4) ]d
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
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
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
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.
v1 h0, 0 t 1
v0 h0,
t 1< < 1
v(t )h( ) =
v0 h1, 1 < t
0, otherwise
Therefore the area beneath v(t )h( ) equals
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
v0 h0 t, 0 t <1
t (v1 h0 v 0 h0 + v 0 h1) v1 h0 + 2 v 0 h0 v0 h1,
1 t < 2
t (v1 h0 + v1 h1 v0 h1 + v0 h2) + 3 v1 h0 2 v1 h1 + 3 v 0 h1 2 v 0 h2, 2t< 3
y( t) =
t (v1 h1 + v1 h2 v0 h2) + 4 v1 h1 3 v1 h2 + 4 v 0 h2, 3t< 4
v1 h2 (5 t), 4t<5
0, 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
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)]
(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
SOLUTION 16.33.
(a) Using the techniques of convolution algebra f 3(t) can be written as
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)
Using the sifting property of the delta function f 3(t) can be computed 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;
(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
SOLUTION 16.34. This problem is solved using the techniques of the convolution algebra with the
graphical method left to the student.
where the superscript (-1) means integration and the superscript (1) means differentiation. By inspection,
40
30
20
10
-10
-20
-30
-40
0 2 4 6 8 10 12 14 16 18 20
40
30
20
10
-10
-20
-30
-40
0 2 4 6 8 10 12 14
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.
(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
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.
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 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
[ ]t 1
1
f 5(t) = (t ) d = 0.3333 3 + 0.5t 2
t 1
= 0.1667t 3 + t 0.6667 for 1 t < 2.
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 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.
[ ] [
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.
In sum,
0.5t 2, 0 t <1
2
t + 3t 1.5, 1 t < 2
f 3(t) =
0.5t + 2t 1.5, 2 t < 3
2
0,
otherwise
SOLUTION 16.37.
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.
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 R. A. DeCarlo, P. M. Lin
4
F2 (s) =
s
By the convolution theorem
F3(s) = F1 (s)F2 (s)
Therefore
8
(
F3(s) = 3 1 e 2s
s
)
Taking the inverse Laplace transform yields
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
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.
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
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)
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 =
0, t<T
t T
= K1K2 u(t T )d = K K
0 1 2 d , Tt
0
0, t<T
= = 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 ),
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)]
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 P16-15 R. A. DeCarlo, P. M. Lin
At t = 1s
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
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 R. A. DeCarlo, P. M. Lin
At t = 0.5s
y(0.5) = 16 V
At t = 1s
y(1) = 65 V
At t = 1.5s
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:
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)):
+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)
At t = 0s
vout (0) = vout ,40 (0) vout ,40 (0.2) = 0 V.
At t = 0.5s
vout (0.5) = vout ,40 (0.5) vout ,40 (0.3) = 12 V.
At t = 1s
vout (1) = vout ,40 (1) vout ,40 (0.8) = 20 V.
At t = 1.5s
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)
(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
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.
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
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.
(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)
( ) (
vout (t) = t 2 + e (t 1) u(t 1) t 3 + e (t 2) u(t 2) )
+( t 4 + e (t 3) ) u(t 3) ( t 5 + e (t 4) ) u(t 4) V.
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) = [1 cos( t)] [u(t) u(t 4)] + 2 [1 + cos( t)] [ u(t 1) u(t 3)] .
SOLUTION 16.46.
(a) By the current division formula
Cs s2
IC (s) = I (s) = 2 + 1 Iin (s)
1 in s
Cs +
Ls
The transfer function of the circuit can be computed as below
(b) The impulse response is computed as the inverse Laplace transform of the transfer function
(c) Assuming zero initial conditions it follows, by the impulse response theorem, that
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
vout (t) = h(t) vin (t) = h (1)(t) vin
(1)
(t)
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
SOLUTION 16.48.
First observe, from figure P16.48(a), that
h(t) = (1 t) [ u(t) u(t 1)] =
1/25/02 P16-24 R. A. DeCarlo, P. M. Lin
[ ]
2 t
[ ]1
t
= 0.5 u(t) 0.5 2 u(t 1) =
0
= (t 0.5t )u(t) (t 0.5t 2 0.5)u(t 1).
2
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)
+ Ls LCs
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 1
= L
LCs 2 + 1 LC s2 + 1 LC LC
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
t
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
(t) = [v out (t) vout (t T )]
T 1
vout
T
where vout (t)is the step response obtained in part (b). Therefore
T 1 1 1 1
vout (t) = 1 cos t u(t) 1 cos t 2 u(t 2 LC )
2 LC LC 2 LC LC
1
=
2
1
1 cos
LC
[
t u(t) u(t 2
LC
LC ]
T
A picture of vout (t), for L = 1H and C = 1F , is sketched in the next figure.
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)]
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).
SOLUTION 16.52.
The overall impulse response of the configuration is
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
where the superscript (1) means differentiation and the superscript (-1) means integration.
We have
h1(1) (t) = u (1)(t) = (t)
and
t t
[ ]0 = 2(1 et )u(t)
t
2e u( )d = 2u(t) e d = 2u(t) e
h4(1) (t) =
0
Therefore
[ ]
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)
h(t) = h1 (t) h3(t) + h2 (t) h3(t) = (t) cos( t)u(t) + (t 1)cos( t)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
where the superscript (-1) means integration and the superscript (1) means differentiation.
We have
t t
1 sin( t)
[cos( t)u(t)] cos( = u(t) cos( [sin( t)]t0
(1)
= )u( )d )d = u(t) = u(t) .
0
Therefore
sin( t) sin( t)
[cos( t)u(t)] u(t) = u(t) (t) = u(t)
SOLUTION 16.54.
(a) The overall impulse response of the configuration in figure P16.53 is
1/25/02 P16-31 R. A. DeCarlo, P. M. Lin
h(t) = h1 (t) h3(t) + h2 (t) h3(t) = (t) cos( t)u(t) + (t 3) cos( t)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
where the superscript (-1) means integration and the superscript (1) means differentiation. We have
t t
1 sin( t)
[cos( t)u(t)] cos( )u( )d = u(t) cos( [sin( t)]t0
(1)
= )d = u(t) = u(t) .
0
Therefore
1/25/02 P16-32 R. A. DeCarlo, P. M. Lin
sin( t) sin( t)
[cos( t)u(t)] u(t) = u(t) (t) = u(t)
SOLUTION 16.55.
(a) The overall impulse response of the configuration in figure P16.55 is
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
where the superscript (-1) means integration and the superscript (1) means differentiation. We have
t t
1 sin( t)
[cos( t)u(t)] cos( = u(t) cos( [sin( t)]t0
(1)
= )u( )d )d = u(t) = u(t) .
0
Therefore
sin( t) sin( t)
[cos( t)u(t)] u(t) = u(t) (t) = u(t)
SOLUTION 16.56.
(a) By definition
h(t) f (t) = h(t ) f ( )d
By making a change of variable
1 = t
we have
h(t) f (t) = h( 1) f (t 1)d 1 = f (t 1 )h( 1 )d 1
By definition
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.
1/25/02 P16-34 R. A. DeCarlo, P. M. Lin
[h(t) f (t)] g(t) = h( ) f (t )d g(t) = h( ) f (t 1 )d g( 1 )d 1 =
= [ h( ) f (t 1 )g( 1 )]d d 1
[h(t) f (t)] g(t) = h( ) f (t 1 )g( 1 )d 1 d
SOLUTION 16.57.
We have
f (t) h(t) = f (t) (t kT )
k=0
L[ f (t)h(t)] = L f (t) (t kT) = = f (kT )e skT
k =0 k =0
SOLUTION 16.58.
(a) By the voltage division formula we have
1
Vout (s) = 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 = 0.5e u(t).
2s + 1 1
s +
2
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.
(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
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.
( ) ( ) ( ) ( )
2 3 4
vout (t) = 0.5e 0.5(t 4) 1+ e 0.5 + e 0.5 + e 0.5 + e 0.5 V.
(e) Using the sum formula for geometric series with = e 0.5 we have
( )
5
1 e 0.5
( ) (
1 + e 0.5 + e ) + (e ) + (e ) = 1 e0.5
0.5 2 0.5 3 0.5 4
1 e 0.5
Therefore
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.
0.5
vout (t) e 0.5(t n) V for n < t < n + 1.
1 e 0.5
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.
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.
1
vout (t) = 0.5e 0.5t V for 0 < t.
1 e 0.5
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
2.6
2.4
2.2
1.8
1.6
1.4
1.2
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
1
The transfer function of the circuit of figure 16.58 is H(s) = 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
0.6
0.5
0.4
0.3
0.2
0.1
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
5/15/01 P17-1 R.A. DeCarlo & P.M. Lin
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 R22
r = (1)
LC L2
Further from example 17,3, at r,
L
Z1( j r ) =
R2C
We require that Z1( j r ) = R1 in which case
L
R1R2 = (2)
C
It is necessary to solve equations (1) and (2) simultaneously for L and C. From (2), L = R1R2 C .
Substituting into the square of (1) yields
1 R22
2r =
R1R2C 2 R12 R22 C2
Hence
1 R1 R2
C=
r R1 R2
It follows that
5/15/01 P17-2 R.A. DeCarlo & P.M. Lin
1 R1 R2 1
L = R1R2 = R2 (R1 R2 )
r R1 R2 r
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 ) +
j r C +
1 r
j r CR1 + 1 2 1 2 R R2
j 1 +1
R1 R2
R1 R2
R1 jR1
R2 R1 R2
= j R2 ( R1 R2 ) + = j R ( R R ) + R jR = R2
R1 R2 2 1 2 2 2
R
1+ 2
R2
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
Interchanging the subscripts of 1 and 2 in case 1 produces the derivation for this case.
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
1 R2 Im
2
CR2 Im
2
cos2 ( 0 t) + 2
wC (t) + wL (t) = 2 sin ( 0t)
2 2 0 L
1 2 2 2 1 1
sin 2 ( 0t) = R2 Im
2
= R ImC cos ( 0t) + 2 C
2 0 LC 2
T T T
w R (0,T ) = R iR2 (t) dt = R iin2 (t) dt = cos
2 2
RIm ( 0t)dt
0 0 0
5/15/01 P17-5 R.A. DeCarlo & P.M. Lin
2 T 2 T 2 2
RIm RIm RIm RIm
2 2
= dt + cos(2 0 t) dt = T =
2 0
0 0
(e) Finally
(b) H1(s) is precisely of the form of equation 17.18 with a single zero at the origin. It follows
1 R 1 1
that 2p = and 2 p = . Hence, from equation 17.19, m = p = , H1( j m ) = ,
LC L LC R
R p 1 L 1 L
B = 2 p = , and Qcir = Qp = = = .
L 2 p LC R R C
(c) With s = j,
2
1
2 LC 1
H2 ( j) = =
(1 LC2 )
2 2 2
1 2 + R + ( RC)2
LC L
2
Instead of maximizing H2 ( j) we minimize its reciprocal, through differentiation. Let
5/15/01 P17-6 R.A. DeCarlo & P.M. Lin
( )
1 2
+ ( RC )2
2
f (C) = 2 = 1 LC
H2 ( j)
Then
(
f '(C) = 2C(R)2 + 2 2 LC 1 2 L = 0 )
implies that
L 1 1 = 1 1
C= 2 =
R (L)2 L 2 R2 L 2 1 + 1
2 2 + 1 Q2
L 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
2
H( j) 2 = K 2
a + j
( p 2 ) + j2 p
= K2 2
a2 + 2
2 2
2
2 2 K f
( p ) + 4 p
2
( )
(a) Differentiating with respect to 2 and setting the derivative equal to zero yields
0=
( )=
d f 2 d
a2 + 2
d 2 d 2 ( 2p 2 )2 + 4 2p 2
5/15/01 P17-7 R.A. DeCarlo & P.M. Lin
= 2
1 ( )(
a 2 + 2 2( 2p 2 ) + 4 2p )
2 2
[ ]
2 2 2 2 2
( p ) + 4 p 2 2 2
( p ) + 4 p
Given that the denominator of the first term is non-zero, this is equivalent to
( )(
0 = ( 2p 2 )2 + 42p 2 a 2 + 2 2( 2p 2 ) + 42p )
( 2) ( )( 2p 2 ) 4(a 2 + 2 )2p
2 4 2 2 2 2 2 2
= + p 2 p + 4 p + 2 a +
( )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
2
( 2p + a2 )2 (2a p )2
Thus, to achieve a real positive solution we obtain,
m = a +
2
( 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
1 s2 + 2 ps + 2p s2 + 2 p (s + a) 2 pa + 2p 2 p s2 2 pa + 2p
= = = +
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
1 2 2 pa + 2p
= Im
( )
2 2 pa + 2p (a j)
Im = Im
H( j) K( j + a) K( 2
+ a 2
)
=
( 2 2 pa + 2p )()
=0
K( 2 + a 2 )
Note that such a frequency may not exist if the quantity under the radical is negative.
Rs
s +
V 1 1 1 L
H(s) = out = = =
Iin Yin(s) 1
+ Cs +
1 C R 1 Rs 1
Ls + Rs Rp s2 + s + s+ + 1
L R pC Rp LC
s + 1.333 103
= 8 105 2
s + 2000s + 1.0006 20
m = a +
2
( 2p + a 2 )2 (2a p )2
gives the exact value of m. In MATLAB,
wm = sqrt(wmsqrd)
wm = 4.001108184444246e+04
j p + 1.333 10 3
8 105 1.333 103
H( j m ) H( j p ) = 8 10 5 = 1 j
2p + j2000 p + 2p 2000 p
400
Rs 1
R + 1 LC
p p 0
Qcir Qp = =
2 p Rs 1 Rs 1
L + R C L + R C
p p
Rs 1
+
1 L RpC Rs 1 1 1 Qcoil + Qcap
= + = + =
Qcir 0 L 0 Rp C 0 Qcoil Qcap Qcoil Qcap
Hence,
5/15/01 P17-10 R.A. DeCarlo & P.M. Lin
QcoilQcap
Qcir
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
+ + Cs Rs s s2 + s+ CRs
R Ls RC LC
Equivalently,
1
0 RC >
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)
1 1 1 2 0 1
j d = j = j 1
2RC 0
2RC LC 2Q 4Q2
Here
5/15/01 P17-11 R.A. DeCarlo & P.M. Lin
0 1
a== d = 0 1
2Q 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
di di di di
L1 1 + M 2 + R1i1 = 10 1 + 3 2 + i1 = 0
dt dt dt dt
and
di2 di di di
L2 + M 1 + R2i2 = 2 2 + 3 1 + i2 = 0
dt dt dt dt
Taking the Laplace transform of these equations yields
10sI1 10i1 (0) + 3sI2 3i2 (0) + I1 = (10s + 1)I1 + 3sI2 10i1(0) 3i2 (0) = 0
and
2sI2 2i2 (0) + 3sI1 3i1(0) + I2 = (2s + 1)I2 + 3sI1 3i1(0) 2i2 (0) = 0
Solving yields
I1 (10s + 1) 3s 1 1 1 2s + 1 3s 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,
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 R. A. DeCarlo & P.M. Lin
W(0,) =
0
(v1i1 + v2i2 ) dt = 0.5L1i12 () + 0.5L2i22 () + Mi1()i2 ()
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.
(R1i1 (t) + R2i2 (t))dt , the actual energy
2 2
Remark: the interested student might computer the integral
0
dissipated in the resistors over [0,), and show that this is 5 J.
(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 P18-3 R. A. DeCarlo & P.M. Lin
(d)
L1 = 4; L2 = 9; M = 3;
k = M/sqrt(L1*L2)
k = 5.0e-01
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
(a)
Z2 = RL*m^2
Z2 = 40000
Req1 = Z2*R/(Z2+R)
Req1 = 8000
Z1 = Req1*n^2
Z1 = 200000
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
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
(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 R. A. DeCarlo & P.M. Lin
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
Isecmag = Ismag*N
Isecmag =
4.2916e+00
(d)
% Our first step is to compute the reflected impedance:
Zrefl = 100*N^2
Zrefl =
40000
Vpar = Vseff*Zpar/Zin
Vpar =
8.7342e+01 + 3.2500e+01i
SOLUTION PROBLEM 18.65. (a) The equivalent circuit accounting for initial conditions is given
below:
(c) Hence
I1' 0.6s 0.1472s 1 V1 15 0.1472 0.1472 V1
'= =
I2 0.1472s 0.1472s V2 s 0.1472 0.6 V2
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
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 (s + 2.208) V1
0 s 2 = s (s + 2.208) 2s + 9 V2
or equivalently
1
V1 2s + 2.208 (s + 2.208) sVs + 2
V = (s + 2.208) 2s + 9 2
2
1 2s + 9 (s + 2.208) sVs + 2
=
3s2 + 18s + 15 (s + 2.208) 2s + 2.208 2
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
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
[ ]
= 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 = (N1)2 ( N2 )2 L2
2 2
= N1N 2 L . It
immediately follows that
L1 : L2 : M = N12 : N22 : (N1N2)
(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
Rb = 200*(1/4)^2
Rb =
1.2500e+01
1 RL s 12.5s 1
Z(s) = 20 + + Las + b b = 1.125s + 20 + +
Cs Lbs + Rb s+
100 Cs
3
Hence,
j12.5 j
12.5 2 1 416.67
Z( j) = j1.125 + 20 + = 20 + 4 + j 1.125 +
j +
100 C 10 C 10 4
+
2 +
2
3 9 9
(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.
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
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
Writing two mesh equations we obtain the following matrix form of the mesh equations:
s + + 1 s ks I1 Iin s
=
ks s ++ 1 s I2 0
s + + 1 s Iin s
det
ks 0 kIin
I2 = =
s + + 1 s ks (s + + 1 s)2 (ks )2
det
ks s + + 1 s
V I s ks ks
H(s) = out = 2 = =
Iin Iin ( s + + 1 s) ( ks) (s + ks + + 1 s)(s ks + + 1 s)
2 2
ks
=
[(1+ k)s 2
][
+ s + 1 (1 k)s 2 + s + 1 ]
5/31/01 Mag Crt Probs P18-11 R. A. DeCarlo & P.M. Lin
(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
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
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 and
f2 = 0.1614 Hz. Finally
Bf = f2 - f1
Bf = 4.5000e-03 Hz
Bfnew = Kf*Bf
Bfnew = 1.2906e+04 Hz
(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 I1 Iin Cs
=
kLs Ls + R + 1 Cs I2 0
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
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
)
LC(1+ k)s + CRs + 1 LC(1 k)s + CRs + 1 (1+ k)s2 + R s + 1 (1 k)s2 + R s + 1
2
L LC L LC
k 20 s
= C
2 2 2 2
(1+ k)s + 0 s + 0 (1 k)s + 0 s + 0
Q Q
jk 30
H( j 0 ) = C
2 (1 k)2 + j 0 + 2
j 20 2
(1+ k) 2+
0 + 0 0 0
Q Q
jk jk
C 0 C 0
= =
j j1 1
k k + k + Q2
2
Q Q
Therefore
1 k
H( j 0 ) =
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 = 2 =0
1 1
dk C 0 dk k + C 0 k 2 + 1
Q2 k + 2
2
Q2
Q
It follows that
2k 2
1 =1
k2 + 2
Q
Hence
1
k2 =
Q2
or k = 1/Q. With this value of k,
1 1/ Q Q
H( j 0 ) max = 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).
Prbs Chap 19, 1/7/02 P19-1 R. A. DeCarlo, P. M. Lin
Zin = Z1 + (1 + gm*Z1)*Z2
Zin = 4.0000e+03
(b)
tau = Zin*C
Prbs Chap 19, 1/7/02 P19-2 R. A. DeCarlo, P. M. Lin
tau = 4.0000e-01
SOLUTION PROBLEM 19.4. (a) First observe that since no current can flow into the secondary
we have
(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
10 2.5 20
Zth (s) = + s + 4 + 4 + 0.25s + 9 = + 2s + 40
s s s
V1 1 a
det 0 0.5 0
0 0 0.5 0.25V1 1
I1 = = = 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
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
(
I3 = W22 W21W11 1
)
W12 Vout + W21W11 1 Iin
0
1 0.5 0.25 1 0.5 Iin
= 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
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:
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
2b 2 2b + 1
= Vout bIin
R
Equivalently
R bR
Vout = Iout + 2 Iin
2b 2 2b + 1 2b 2b + 1
Therefore
R bR
Zth = , Voc = 2 Iin , Isc = bIin
2b2 2b + 1 2b 2b + 1
R
(b) Isc = b45o , Zth = .
2b2 2b + 1
Prbs Chap 19, 1/7/02 P19-6 R. A. DeCarlo, P. M. Lin
1 1 2b2 2b + 1 5
(c) B = 10 = = = = . Hence C = 0.02 F. Further,
ZthC R RC 25C
C
2b 2 2b + 1
2 1 1
0 = 25 = = . Therefore L = 2 H.
LC 0.02L
V V 1 2 100s
(d) H(s) = out = out = 2 2 = =
Iin 0.5I sc 2b 2b + 1 1 1 s2 + 10s + 25
+ + Cs 0.2 + 2s + 0.02s
R Ls
2 s 10
and Iin (s) = 2 . Hence,
2 s + 100
50 2 s(s 10)
Vout (s) =
(s2 + 10s + 25)(s2 + 100)
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
5.6 2 30 2 (5.6s 8) 2
Vout (s) = + +
s+5 (s + 5) 2 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.
I1 Y1 + Y3 Y3 + gm V1 y11 y12 V1
I = Y =
2 3 Y2 + Y3 V2 y 21 y 22 V2
1 1
Zin = =
y11 Y1 + Y3
and since V2 = 0,
KY3
I2 = y 21V1 =
s
I1 Y1 + Y3 Y3 V1 y11 y12 V1
I = Y + g =
2 3 m Y2 + Y3 V2 y 21 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 .
I1 Y1 + Y3 Y3 + gm V1 y11 y12 V1
I = Y =
2 3 Y2 + Y3 V2 y 21 y 22 V2
V1 L1s Ms I1 I1 1 L2 M V1
V = Ms L s I I =
2 2 2
2 L1L2 M 2 s M( ) L1 V2
V1 L1s Ms I1 I1 1 L2 M V1
V = Ms L s I I =
2 2 2
2 L1L2 M 2 s M ( ) L1 V2
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 R. A. DeCarlo, P. M. Lin
I1 1 L2 M V1 1 4 1 V1
I ' = =
( )
2 L1L2 M 2 s M 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
2K a
V1 = 2
s +4
Hence
2G1K a
I1 =
s2 + 4
To compute I2 , we reflect the parallel of G1 and G2 to the secondary of the ideal transformer.
a2
Zsec =
G1 + G2
Therefore,
Prbs Chap 19, 1/7/02 P19-1 0 R. A. DeCarlo, P. M. Lin
V2 2K( G1 + G2 ) a2
I2 = =
Zsec s2 + 4
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
400
( (6 //12) ) =
Zout = 80 + 320/ /16
3
I1 2s + 2 1 2s V1
I 2 = 1 s + 2 s V2
0 2s s 5s V3
I1 2s + 2 1 1 2s V1
I = 1 [2s s]
2 s + 2 5s s V2
SOLUTION PROBLEM 19.17. (a) The ideal transformer yields the constraints
I2 2 1 V2
=
I2 1 2 V2
Substituting the first two equations into the last one, we obtain
I1 2a 2 a V1 y11 y12 V1
I = =
2 a 2 V2 y 21 y 22 V2
(b)
2
y12 y 21 2 a
Yin = y11 = 2a = 1.75a2 S
y 22 + YL 4
1 4
Zin = = 2
Yin 7a
and
Gv = V2 = -y21 = a
V1 y22 + y L1 4
Part (a)
% Parameter specification
Ys = 1e-3;YL = 1e-3;
y11=4e-3; y12 = -0.1e-3;
y21 = 50e-3; y22 = 1e-3;
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
PL = V2^2/1e3
PL = 1.1111e+00
(b) v2 = - y 21 = -2 = - 6.667
v1 y22 +YL 0.2 +0.1
Hence
v2(t) = - 6.667 v1(t) = - 3.333 vs(t) = - 33.33 u(t) V
and
2 2
PL = V2 = 33.33 = 111.11 W
RL 10
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 R. A. DeCarlo, P. M. Lin
1 -3 I1 2 0 V1
=
0 2 I2 -2 1 V2
Therefore
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 1s s+ 4
y11 = 1 = =
V1 4 4
s(s + 4)
and
I 1 (s + 3) s(s + 4)
y 21 = 2 = =
V1 4 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
I1 = y11V1 + y 12V2
from which we obtain
1 - s + 4 5s + 16
s 4 s(s + 4 ) 2
y12 = I1 - y 11V1 = = - (s + 7)
V2 1 4 ( s + 4)
s+7
- (s + 7) - s(s + 4)
2
Yin = y11 - y12y21 = s + 4 - 4 ( s + 4) 4(s + 3) = (s + 4)
y22 +Y L 4 s2 + 23s + 64 +1 5(s+3.2)
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,
The imaginary part is zero when = 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 R. A. DeCarlo, P. M. Lin
(2s +2) 2
Yin = y 11 - y12y21 = 3s+2 -
2
= 2s + 3s + 1 = 2s + 1
y22 +Y L 3s + 3 + s + 1 s+1
V1 = Z in = 1 = 1 = 1
Vs Z in + Z s 1 + Z s Yin 1 + 2(2s + 1) 4s + 3
and
V2 = -y21 = 2s +2 = 0.5
V1 y 22 + y L 3s + 3 + s + 1
Thus,
V2 = V1 V2 = 1 0.5 = 1 = 1
Vs Vs V1 4s + 3 2
8s + 14s + 6 8s+ 6
(c) The impulse reponse is
1 1 0.125
h(t) = L1{H (s)} = L1 0.75t
= L = 0.125e u(t)
8s + 6 s + 0.75
(d) We must compute the complete Laplace transform and invert. Here
Vs(s) = 12.75 2
s2 + 4
and
V2(s) = H(s)V2(s) = 0.125 12.75 2 = 3.1875
s+ 0.75 s + 4 (s + 0.75) (s2 + 4)
2
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
4
y 12 y 21 0.52410
Yin1 = 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
V = = = 217.8
1 stage1 y 22 + YL,stage1 (100 + 500 + 501.8) 10 6
and
V2 y 21 0.075
V = = = 17.92
1 stage2 y 22 + YL,stage2 (185 + 4000) 10 6
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
y 21 1 s 2 8 2 1 1
V2(s) = V1 (s) = = = + +
y 22 + 0.25s 1 s + 0.25s s s(s 4)
2 s s+2 s2
Hence, for t 0,
(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,
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
V1 L1s Ms I1
V = Ms L s I
2 2 2
Hence, the y-parameters are:
L1s Ms
Ms L s
2
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 1
V
reciprocal, z21 (= z12 ) = 2 = aR2 .
I1 I =0
2
z12z21 a 2R22 1
Zin = z11 = R1 + R2 2 = R1 + 2 R2
z22 + ZL a R2 + R2 a +1
2aKR 2a 2KR2
V1 = z12 I2 = 2 2 and V2 = z22I 2 =
s +4 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
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
With port 2 open and I1 injected into port 1, we have Zpr = 20 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 Z3 I1 z11 z12 I1
V = Z + r =
2 3 m Z2 + Z3 I2 z21 z22 I 2
K ( Z1 + Z3 ) K (Z3 + rm )
(b) V1 = z11I1 = and V2 = z21I1 =
s s
Z1 + Z3 Z3 z11 z12
Z + r =
3 m Z2 + Z3 z21 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
(b) Zin = s s = =
s 10(s + 1) s s(2s + 1) 2s(s + 0.5)
+ 10
s
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
This is of the form of equation 17.18 with K = 10. Here according to equation 17.19f,
r = p = 25 = 5 rad/s
p 5
Q = Qp = = =5
2 p 1
10
7
Magnitude of Zin
4
TextEnd
3
0
0 2 4 6 8 10 12 14 16 18 20
Frequency in rad/s
Prbs Chap 19, 1/7/02 P19-2 4 R. A. DeCarlo, P. M. Lin
100
80
60
40
Phase of Zin degrees
20
-20
TextEnd
-40
-60
-80
-100
0 2 4 6 8 10 12 14 16 18 20
Frequency in rad/s
z11 z12 2 3
SOLUTION PROBLEM 19.33. (a) Since Z = = . Assuming that Zin does
z21 z22 30 4
not include Zs , it follows that
z12z21 90
Zin = z11 = 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
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
PZL = v2^2/20
PZL = 6.1728e+02
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
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
( )
h(t) = 106 1012 t e 10 t u(t) V
6
0 1000 1 0 0.001
(d) y = =
0
S
1000 0 0.001
Hence
z z 1.2075 63.75
(a) Zin 2 = z11 12 21 = 62.582 = 1.7778 k
z22 + ZL 1.25 + 0.016
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
L1s Ms
z=
Ms L2s
(b) Utilizing the properties of an ideal transformer,
V V k 2 L2
z11 = 1 = L1s z22 = 2 = L s + (1 k 2 )L2s = L2s
I1 I =0 I2 I =0 L1 1
2 1
V k L2
z12 = 1 = L s = k L1L2 s = Ms
I2 I =0 L1 1
1
Finally,
V k L2
z21 = 2 = L s = k L1L2 s = Ms
I1 I =0 L1 1
2
V V k 2 L1
z22 = 2 = L2 s z11 = 1 = L s + (1 k 2 )L1s = L1s
I2 I =0 I1 I =0 L2 2
1 2
V k L1
z12 = 1 = L s = k L1L2 s = Ms
I2 I =0 L2 2
1
Finally,
V k L1
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 =
Therefore
1 1
m = = 105 rad/s and B = = 2 104 rad/s.
LC ReqC
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
V1 Z1 1 I1 h11 h12 I1
I = 1 Y V = h
2 2 2 21 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 Z1 1 0 1 I1
I = 1 0 V I = Y
2 2 2 2 1 1 0 V2
Thus
V1 1 Z1 1 I1
I = Z Y + 1 1 Y V
2 1 2 2 2
1
V1 Z1 1 I1 Cs 1 I
1 1 1 1 R RCs I1
I = Z Y + 1 1 Y V = 1 =
1 V2 RCs + 1 RCs Cs V2
2 1 2 2 2 + 1 1
RCs 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 2 2 2 2
I2 = I2 n . Hence
V1 1 R RCs I1 V1 1 R RCs I1
I = RCs + 1 RCs Cs V nI = RCs + 1 RCs Cs V n
2 2 2 2
Therefore
V1 1 R RCs n I1
I = RCs + 1 RCs n Cs n 2 V
2 2
Cs R
From table 19.1, if h22 = 0, then the z-parameters exist and if h11 = 0,
n 2(RCs + 1) RCs + 1
the y-parameters exist, i.e., if C 0 and R 0 respectively.
Prbs Chap 19, 1/13/02 P19-1 R. A. DeCarlo, P. M. Lin
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 1 1 1 1
I1 = I1 b . Hence
V1 R 1 I1 V1 b R 1 bI1
= = 1 Cs V
I 2 1 Cs V2 I2 2
Therefore
V1 b2 R b I1
I =
2 b Cs V2
From table 19.1, if h22 = Cs 0, then the z-parameters exist and if h11 = b 2R 0, the y-
parameters 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
2 V2 =0 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
(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
Zin1 = h 11 = 2000
(b) For stage 1, because h12 = 0, the output impedance is unaffected by the source impedance.
Thus,
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
V2 = 1 -h 21 = 1 -50 = - 39.46
stage 1
V1 Z in1 h22 + Y L1 2000 (0.05 + 0.5834) 10-3
For stage 2, the load is 64 . Hence
V2 = 1 -h 21 = 1 51 = 0.7762
stage 2
V1 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
1 1
f 3dB = = = 39.789 Hz
2(Rs + Rin1)C 2(2000 + 2000)106
Zout = Z out2 = Z L = 64 = 1
Yout2
This requires that
V2 = 1 -h 21 = 1 -50 = - 36.74
stage 1
V1 Z in1 h22 + Y L1 2000 (0.05 + 0.63044) 10-3
For stage 2, the load is 64 . Hence
V2 = 1 -h 21 = 1 51 = 0.7762
stage 2
V1 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
(b) For stage 1, because h12 =0, the output impedance is unaffected by the source
impedance, and
V2 = 1 -h 21 = 1 51 = 0.7762
stage 2
V1 Z in2 h22 + Y L2 2000 0.0016 + 1/32
Finally, the overall voltage gain is the product of the three gains calculated above
SOLUTION PROBLEM 19.46. (a) Since the currents through YL and h22 are the same, h22 = YL.
YL I2 Y h
(b) From current division, I2 = h I = L 21 .
YL + h22 21 1 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
h12h21 150
h11 = Zin + = 1000 = 400 .
h22 + YL 0.25
V1 h11 h12 I1
I = h h22 V2
2 21
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
1 1 h h
Yout = = = h22 12 21 = h22
Zout 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.
V h21 h21 h
Gv2 = 2 = 100 = = = 21
V1 (
Zin h22 + YL b 2
) (
3
960 1.25 10 +1.25 10 3 2.4 )
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
V1 h11 h12 I1
I = h h22 V2
2 21
V
(a) From this expression and specification 1, h12 = 1 = 0.01.
V2 I =0
1
Zout = 800 = b2 Z L = 8b 2
Therefore, b = 10.
Now we find Zin. From specification 3 and voltage division,
V1 24 Z in
= = (1)
Vs 25 Zin + 40
Equivalently,
Vs 25 40
= =1 + Zin = 40 24 = 960
V1 24 Zin
0.01h21 h21
Zin = 960 = h11 h11 = 960 + 0.01 (2)
h22 + 1.25 10 3 h22 + 1.25 10 3
Prbs Chap 19, 1/13/02 P19-10 R. A. DeCarlo, P. M. Lin
V2 h21 h21
= 100 = =
V1 (
Zin h22 + YL b2 ) (
960 h22 + 1.25 10 3 )
which implies that
h21
= 960 10 2
(h22 )
(3)
+ 1.25 103
Also
h12h21 0.01h21
Yout = 1.25 103 = h22 = h22
h11 + Zs 1960
Equivalently,
1960 1.25 10 3 = 2.45 = 1960h22 0.01h21 (5)
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=
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 m-
file:
% 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)
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.
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
I C sV + gmV C s + gm
(a) h21 = 2 = =
I1 V =0 1
2
R
(
)
+ C + C s V
(
C + C s +
1
R
)
1
Rx + 1 I1
V1
R
(
+ C + C s
) 1
(b) h11 = = = Rx + 1
I1 V =0
2
I1
R
(
+ C + C s )
1
1
+ C s
V V R C s C s
h12 = 1 = = 1 = =
V2 I =0 V2
1
C s
+
1
1 1
+ C s + C s C + C s +
1
( )
+ C s R R
R
1 1
gmV + + C s V C s C s + gm +
I R R
(d) h22 = 2 = =
V 2 I 1 =0 ( )
C + C s +
1
R
(
C + C s +
1
R
)
V
C s
V1 t11 t12 V2
I = t
1 21 t22 I2
For the given network,V2 = ZL (I 2 ) and
(b) For the output impedance relationship, from the t-parameter relationships
V t22Zs + t12
Zout = 2 =
I2 t21Zs + t11
1 z11 V1 0 z12 V2
0 z I = 1 z I
21 1 22 2
Solving for [V1 I1 ] yields
T
V1 1 Z1 V2
I = 0 1 I
1 2
V1 1 0 V2
I = Y
1 2 1 I2
1 0 1 Z1 1 Z1
Tnew = =
1 0 1 Y2 1 + Z1Y2
Y2
1 1
V1 = nV2 I1 = I 2 = ( I2 )
n n
Therefore
V1 n 0 V2
I = 0 1 n I
1 2
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= =
Y2 1 0 1 n nY2 1n
(b)
n 0 1 + Z1Y2 Z1 n (1+ Z1Y2 ) nZ1
T= =
0 1 n Y2 1 Y2 n 1 n
SOLUTION PROBLEM 19.57: For each 2-port of the form of figure P19.53a, we have that the t-
parameters are given by
Prbs Chap 19, 1/13/02 P19-16 R. A. DeCarlo, P. M. Lin
1+ Z1Y2 Z1
T=
Y2 1
The given network consists of three such sections in cascade whose t-parameters are respectiely,
1 + s2 s 1+ 1 0.5s 1 1 + 0.5s 1
T1 = , T2 = =
1 0.5s 1
, and
s 1 0.5s
1 1 2
1 + 2s 4 s 2s 1+ 2 s 2s
T3 = = 1
1
s 1 s 1
4 4
Observe that
1 3 3 2 1
2 s + 2 s + 2 s + 1 s + s + 1
2
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 s + 4 s + s + 4 s + 4 s + 1 s + 3s + 2s + 3s + 1
4 3 2
T = T1T2T3 =
1 4 3 3 3 2 7
s + s + s + s s3 + 3s2 + s + 1
4 4 4 4
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;
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
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
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)
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:
[
= 3.005 3e t + e t 2 10 5 cos(500t /6) + 0.01sin(500t /6) ]
(b) From the problem statement
cos
6
(s + 1) + 250
3
z21 (s) =+5
s +1 ( s + 1)2 + (500)2
From reciprocity, z12 (s) = z21(s). For steady state analysis, we use phasors to obtain
cos (1+ j500) + 250
3 6
V1 = z12 ( j500)I2 = +5
1+ j500 (1+ j500) 2 + (500) 2
V2 + bI1 (I 2 I 3) = 0 V2 = bI1 + I2 I3
I
1 a V1 3 0 3 1
0 1 V = b 1 1 I 2
2
I3
whose solution is
I I
V1 1 a 3 0 3 1 3 ab a 3 a 1
V = 0 I =
1 b 1 1 2 b 1 1 2
I
2 I3 I3
Hence
V1 3 ab a I1 3 a
V = b +
1 I2 1 3
I
2
2 2
H=
(0.5a + 1) 0.5
Prbs Chap 19, 1/13/02 P19-20 R. A. DeCarlo, P. M. Lin
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
I1 y11 y12 V1
I = y y V
2 21 22 2
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
Therefore
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
I y V I y V
Also, I1 = y11V1 + y12V2 y12 = 1 11 1 = 1 11 1 . Hence
V2 I 2
1
=
(s + 2)
In conclusion
2(s + 1)2 1 2(s + 1)2
y11 y12 s(s + 2)
(s + 2) = 1 1
y y = s + 1
21 22 1 (s + 2) s
s+2
s + 1 1
(s + 2)
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
1 z22 z12 1 s2 + 2 2
Y= =
Z z21
z11 20s2 + 2s + 32 (4s + 2) 20s + 2
where
(20s + 2)(s2 + 2) 2(4 s + 2) 20s3 + 2s2 + 32s
Z = = = 20s2 + 2s + 32
s s
Finally, the h-parameters are given as
(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
Therefore
1.5R 0.5
H=
0.5 0
1 1 h12 1 4 2
Y= =
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 2 4.5 1 4.5 1 12 6
Ymid = =
s 1 4.5 1 27 + 1 4.5 27s 6 7
y 22 1
y y 21 7 6 4.5s
T2 = y y11 = 8 27s 2
21
y 21 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
63a2 27ab
a 2 2.25s
0.5a / b 1 12b2 6ab
H= , Z = s 16 8 , Y =
0.5a / b 4 27b
2 27ab 27b 2 27a2b 2s 6ab 7a2
8 4
I1 g11 g12 V1
V = g
2 21 g12 I2
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
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)
V Zin Ys Ys Ys (g22 + ZL )
G1 = 1 = = = =
Vs Zs + Zin Ys + Yin Y + g g12g21
s 11
( g22 + ZL )( g11 + Ys ) g12g21
g22 + ZL
ZL V g Z
(d) Refer to figure P19.64, where V2 = 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
I1 g11 g12 V1
V = g
2 21 g22 I2
For the given circuit
1 1 1
V2 = (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 1
y y 21 0.1 0.5
YN1 TN 1 = y y11 = 3.82 25.1
21
y 21 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 h11
h h21 1 56 13 11.2 2.6
HN 2 TN 2 = h21
21 1 = 5 2 1 = 0.4 0.2
h21 h21
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:
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
1
Rth
(b) To prove that c = or
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.
1
c = = 25.58 10 6 rad/s
Rth C
Prbs Chap 19, 1/13/02 P19-26 R. A. DeCarlo, P. M. Lin
(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
1 1
= = = 464.45 rad/s
1076.54 2 106
c Rth C
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
with dependent sources and ideal op amps, etc), we have Zth(s) = Rth and Voc(s) =
By voltage division,
Ls K 0s
VL = Voc = R Input(s) (1)
Ls + Rth s + th
L
and
1 K0
RthC
VC = 1 Cs Voc = 1 Input(s) (2)
+ Rth s+
Cs RthC
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)
K0 s
Ouput(s) = K1 + K2 R Input(s) (4)
s + th
L
and for the case of VC(s),
K0
RthC
Ouput(s) = K1 + K2 1 Input(s) (5)
s+
Rth C
For either case, the transfer function H(s) has the form
Ouput(s) K3s + K4
H(s) = = (6)
Input(s) s + c
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
K3s + K4
H() = = K3
s + c s=
K3 s + K 4 K
H(0) = = 4
s + c s=0 c
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
1
ZC (0) = =
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
Therefore
4 2
s +
2 2
H()s + c H(0) 3 s + 9 3 3
H(s) = = =
s + c s+1 s +1
Therefore
4 2 4
H()s + c H(0) 9 s + (s + 1.5)
H(s) = = 3 = 9
s + c s+1 s +1
SOLUTION PROBLEM 19.70. According to problem 19.68, the transfer function of the circuit is
Ks
H(s) =
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
1
= = 255.65 rad/s
1955.8 2 106
c
SOLUTION PROBLEM 19.71. (a) Except for the terminating resistor, let the other element
branches of the circuit be given by
1 1 2s 1
1 = , Y2 = 1 = 2 + 1, and Z3 =
3s1 2s + 4s s
2s
Consider the circuit
2
1
V3 1 1 V
V1 1+ 23 V3
s I
2
I = 4s + 1 3s and =
1 I 3 I 3 0 1 2
1
2s
4s2 + 1
V1
1 4s2 + 5/ 3 (16s2 + 6) V2
I = 2 3s I
1 4s + 1 2
2s 4s + 3
2
(b)
y 21
GV (s) =
y 22 + y L
(c)
5
3s 4s2 +
1 2
3s(4s + 1) t11 3 s(12s + 5)
2
y 21 = = , y =
22 t = = , and YL = 1
t12 16s2 + 6 12 16s2 + 6 16s2 + 6
(d) Hence,
3s(4s2 + 1)
y 21 2 3s(4 s2 + 1)
GV (s) = = 16s2 + 6 =
y 22 + y L s(12s + s) 12s3 + 16s2 + 5s + 6
+1
16s2 + 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
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,
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
Gv = Gv1*Gv2*a
Gv =
-2.8071e+00
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
Psnorm = Vs*Is
Psnorm =
3.4214e-04
VL = Gv*1
VL =
-2.8071e+00
PLnorm = VL^2/8
PLnorm =
9.8496e-01
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
+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
+60.000
+50.000
+40.000
+30.000
+20.000
+10.000
+0.000e+000
MAG(V(IVM2))
SOLUTIONS PROBLEMS CHAPTER 20
USEFUL MATLAB M-FILES FOR USE IN THE SOLUTION TO PROBLEMS IN THIS CHAPTER.
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
SOLUTION 20.1.
R1 + R3 R3
Za =
R3 R2 + R3
R1 + R3 R3
Zb =
R3 R2 + 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 R3
Znew = Za + Zb = 2
R3 R2 + R3
SOLUTION 20.2. For networks a and b, the y-parameters are by inspection
3 2
Ya = Yb = S
2 4
1 4 2
Za = Zb =
8 2 3
1 4 2
Znew = Za + Zb =
4 2 3
s s
Za =
s s
For network b, we have
1 1 1 0.5 0.5
Zb = [Yb ]
1
= =
2 2 4 1 2
s + 0.5 s + 0.5
Znew = Za + Zb =
s +1 s + 2
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
1 3 2 0.6 0.4
Za = =
5 2 3 0.4 0.6
1.6 0.9
Znew = Za + Zb =
0.4 0.4
0.7 0.2
YNa = YNb = S
0.2 0.7
20 0.7 0.2 1 14 4
ZNa = ZNb = =
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
2 14 4
Znew = ZNa * + ZNb =
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 =
0.2 0.7 9 4 14
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:
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
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,
2 14 4
Znew = ZNa * + ZNb =
9 4 14
V1 4 + R1 2 R1 I1
V2 = 2 4 + R2 R2 I 2
0 R1 R2 R1 + R2 I 3
For I3 to be zero, the third equation implies that R1I1 = R2I2. Therefore,
2R
V1 = (4 + R1) I1 + 2 1 I1 = 4 + R1 + 1 I1
R
R2 R2
Similarly
(4 + R2 )R1
V2 = 2I1 + ( 4 + R2 ) 1 I1 + = 2 +
R
I1
R2 R2
Therefore
2R1
4 + R1 +
V1
=
R2
=
(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
However,
V13 I1
V = Z A I
23 2
Therefore
V1 Z1 + Z3 Z3 I1 I1 Z1 + Z3 Z3 I1
V = Z + ZA = + Z A
2 3 Z2 + Z3 I 2 I2 Z 3 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,
8 1
Zb =
1 5
For Nc,
2 1
Zc = Za =
1 2
Therefore
12 3
Z = Za + Zb + Zc =
3 9
1
7 2 3 2
YN = ZN1 = = S
10 3 10 7
s s
YC = S
s s
s s 3 2 s + 3 (s + 2)
Y = YN + YC = + =
s + 7
S
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.
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
It follows that
2.1 0.4
Y = Ya + Yb = S
0.4 1.05
+1.000
+800.000m
+600.000m
+400.000m
+200.000m
+0.000e+000
MAG(V(IVM))
(b)
C0 C1
15.9n 15.9n
R1
500
R R0
1k 1k
IVm
V C
0 31.8n
Prob 20.15-Small Signal AC-11 Frequency (Hz)
+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:
G1 + j C1 0 2 + j10.21 0
Ya = =
0.07143
mS
gm G0 95
1 1 1 1
(
Yb = G f + j C2 )
1 1
= (0.8333 + j1.021) mS
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 y-
parameters:
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 :
1+ 2s s
TLR =
2 1
Therefore
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
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
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
(b) Since the h-parameters of the transistors are given, we can again use MATLAB and the m-
file twoporth defined earlier. Hence:
(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
o
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 z-
parameters.
(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
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
% 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
Using MATLAB,
I1 5 0 0 5 V1
I 0
2 = 6 7 1 V2
I3 0 4 8 4 V3
0 5 2 1 8 V4
(d) This part is similar to part (a) as it does not require the method of matrix partitioning. By
inspection,
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
(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
13 7
YCB = S
7 9
y y
Yin = y11 12 21 = 13 S
y 22 + YL
Hence
1
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
6
Therefore, V2 = 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
IB 1 0 VBE
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
By inspection, with G as the common terminal, S as the input terminal, and D as the output
terminal, we obtain,
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 =
2
S and the bottom 2-port has y-parameters
s 1
2s 2s 1
ybot =
2
S. Since these 2-ports are connected in parallel, the overall 2-port y-
s 1
parameters are
s 2s 1 2s 2s 1 3s 2
y = y top + y bot = + =
2 s 1 2 2 4
S
s 1
Hence
0.2 + j2.5 0.19 j1.85
yGS =
3.34 + j2
S
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 Y1 2 V A
IB = Yind VB = Y2 2 Y2 2 VB
IC VC Y1 2Y2 2Y2 Y1 VC
(b)
Y1 + Y2 + 2 Y2
y AB =
Y2 2 Y2
(c)
Y1 + Y2 + 2 Y1 2
y AC =
Y1 2Y2 Y1
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
IB = Yind VB = Y2 + gm Y2 + Y3 Y3 gm VB
IC VC Y1 gm Y3 Y1 + Y3 + gm VC
(b)
Y1 + Y2 Y2
y AB =
Y2 + gm Y2 + Y3
(c)
Y1 + Y2 Y1
y AC =
Y1 gm Y1 + Y3 + gm
I A VA 8 0 0 8 VA
I W
B = 11 W12 VB = 0 2 + 2s 2s 10 8 VB
IC W 21 W22 VC 0 2s 16 + 2s 16 VC
0 VD 8 2 6 16 VD
where VD is the internal node voltage. Using the method of matrix partitioning,
8 0 0 8
1
= 0 2 + 2s 2s 10 8 [8 2 6]
1
Yind = W11 W12W22 W 21
16
0 2s 16 + 2s 16
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 16 + 2s 1 8 2s 2 10 + 2s
4 3
y AC = S
8 10 + 2s
s + 0.5 s 0 1
0.5
1 [1 1 s]
1
Yind = W11 W12W22 W 21 = s s + 0.5 0
s+2
0 0 0.5s s
s + 0.5 s 0 1 1 s
0.5
= s s + 0.5 0 1 1 s
s+2
0 0 0.5s s s s2
SOLUTION 20.35. (a) Rule 1: Consider the two networks NA (3 external nodes) and NB (4
external nodes) given below:
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:
6 2 4 5 3 2 11 5 6
YindN A = 2 3 1 , YindN B = 3 5 2 , and YindN C = 5 8 3 .
4 1 5 2 2 4 6 3 9
Clearly,
YindN C = YindN A + YindN B
5 3 2
YindN A = 3 5 2 S
2 2 4
I1 5 3 2 V1
I 2 = 3 5 2 V2
0 2 2 4 V3
5 3 1 2 4 4
YindN B = [2 2] =
4
S
3 5 4 2 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 =
0 0 0 0 0
0 0 0 0 0
0 0 7.5000e-01 -5.0000e-01 -2.5000e-01
0 0 -5.0000e-01 6.2500e-01 -1.2500e-01
0 0 -2.5000e-01 -1.2500e-01 3.7500e-01
Part (b)
Yind = Yinda + Yindb
Yind =
W12=[Yind(1:3, 4:5)]
W12 =
-2.5000e-01 0
0 -2.5000e-01
-5.0000e-01 -2.5000e-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
(
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 a1 a a1 a2 2 a1a2 a1 a2 2 a1
0 = p(s) = s3 + 2 s2 + s + 0 = s s2 + + s + 2 = s s2 + + s +
a3 a3 a3 a3 a3 a3 a3 a3 a3
a a
= s2 + 1 s + 2
a3 a3
a1 a
Therefore, the roots are: s = j , 2.
a3 a3
SOLUTION 20.38.
(a) The four 2-port equations arising from the interconnection are:
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:
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.
-5
h21 = h21a h21b = 96 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
SOLUTION 20.40.
(a) The y-parameters for Na are:
1
7 4 1 7 4
y AB = =
33 4 7
S
4 7
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:
1 7 3
y BC =
33 3 6
S
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')
-5
dB Gain
TextEnd
-10
-15
0 0.5 1 1.5 2
Frequency rads/s
(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.
Further
( p)2 2.2207 10
7
H(s) = HNLP (s 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
6
H(s) = HNLP (s K f ) = 2 = 2
s + K f 2s + (K f ) 2 s + 4.4429 103 s + 9.8696 106
0.9
0.8
0.7
Magnitude
0.6
0.5
TextEnd
0.4
0.3
0.2
0.1
0 500 1000 1500
Frequency in Hz
-2
-4
-6
Magnitude in dB
-8
-10
-12
TextEnd
-14
-16
-18
-20
0 500 1000 1500
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
SOLUTION TO 21.5. (a) max is that value of e that places the magnitude response curve through Amax
at = p. Therefore
p
2n
2
Amax = 10log 10 H ( j p) = 10log 10 1+ max = 10log 10 1+ 2max
2
p
( )
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
Amin = 10log10 H ( j s ) = 10log 10 1+ min
2 2 s
p
Therefore
100.1A min 1
2min = 2n
s
p
which is equivalent to the required formula.
p
SOLUTION TO 21.6. The relationship of and c is given by the formula: c = . Further, max
() n
1
in putting the magnitude response curve through the Amax spec produces and min in putting the
c min ,
magnitude response curve through the Amin spec produces c max . Hence, from the solution to problem
5,
p p p
c min = 1n = 2n c max = = 2n s
(max ) 10 0.1Amax 1 (min )1 n 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
s + 2s + 2s + 1
2
(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
0.9
0.8
0.7
0.6
Magnitude
0.5
0.4 TextEnd
0.3
0.2
0.1
0
0 1 2 3 4 5
Normalized Frequency
Notice how the 3rd order filter has a sharper transition to zero.
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) = 1 e
t
) u(t) 1.1547e0.5t sin(0.86603t)u(t)
SOLUTION TO 21.8.
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
0.8
0.6
0.4
Imaginary part
0.2
-0.2
-0.4
-0.6
-0.8
-1
-1 -0.5 0 0.5 1
Real part
znew = []
pnew =
-4.8762e+02 + 8.4458e+02i
-4.8762e+02 - 8.4458e+02i
-9.7524e+02
knew = 9.2753e+08
1
0.9
0.8
0.7
Gain magnitude
0.6
0.5
0.4
0.3
0.2
0.1
0
0 500 1000 1500
Frequency in Hz
-10
-20
Gain in dB
-30
-40
-50
-60
0 500 1000 1500
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
0.7
Gain magnitude
0.6
0.5
0.4
0.3
0.2
0.1
0
0 500 1000 1500
Frequency in Hz
-5
-10
Gain in dB
-15
-20
-25
-30
-35
-40
0 500 1000 1500
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
0.8
0.6
0.4
Imaginary part
0.2
-0.2
-0.4
-0.6
-0.8
-1
-1 -0.5 0 0.5 1
Real part
1
0.9
0.8
0.7
Gain magnitude
0.6
0.5
0.4
0.3
0.2
0.1
0
0 100 200 300 400 500 600
Frequency in Hz
-10
-20
Gain in dB
-30
-40
-50
-60
-70
0 100 200 300 400 500 600
Frequency in Hz
SOLUTION TO 21.11.
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
0.7
Gain magnitude
0.6
0.5
0.4
0.3
0.2
0.1
0
0 100 200 300 400 500 600 700 800
Frequency in Hz
-10
-20
Gain in dB
-30
-40
-50
-60
-70
0 100 200 300 400 500 600 700 800
Frequency in Hz
L2 4L + 4 = (L 2)(L 2) = 0
1 1
Hence L = 2 H is the only solution. Thus C1C2 = 1 C1 = + 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.
1 1
H(s) = Cs + G1 = LC
1 1
Ls + s +
2
s+
Cs + G RC LC
(b) With R = 1 ,
1
LC 1
H(s) = 1 1 = 2
s2 + s+ s + 2s + 1
RC LC
(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)
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
1 1
H(s) = Cs LC
1 = 2 Rs 1
Ls + Rs + s + s+
Cs L LC
(b) With R = 1 ,
1
LC 1
H(s) = R 1 = 2
s2 + s s + s + 2s + 1
L LC
(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.
(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
H(s) =
Vout
= Cs + G = LC
Vin Ls + Rs +
1 1 Rs 1 + Rs RL
Cs + G s + CR + L s +
2
L LC
(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 = =
+ Ls + 1 Cs + G s2 + 1 + 1 s + 1 + 1 Rs
Vin 1
Cs + G R C L 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)
K K
H( j) = 2 = 2
j +1 +1
p p
1 1
2 =
c 2
+1
p
Equivalently
c = p 2 1 = 0.64359 p = 6.4359 10 4 rad/sec.
p t
(b) h(t) = K 2pte u(t) since
K K 2p
H(s) = = .
( s + p )
2 2
s
+ 1
p
Further,
1 K 2p K K K p
H(s) = =
(
s s s+ 2 )
s s + p ( )
2
p s + p
1
L1L2C
=
1 1 L + L2 + C 2
s3 + + s2 + 1 s+
L1 L2 L1L2C L1L2C
1
L1L2C 1
= 3
1 1 L + L2 + C 2 s + 2s + 2s + 1
2
s3 + + s2 + 1 s+
L1 L2 L1L2C L1L2C
2
L1 + L2 +
2 L + L2 + C L1L2 2
yields C = and 1 = = 2; equivalently, L1 + L2 + = 4. Further,
L1L2 L1L2C 2 L1L2
1 1 L1 + L2 2
L + = L L = 2 implies that L1 + L2 = 2L1L2 implies 2L1L2 + L = 4 . This requires that
1 L2 1 2 L1 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.
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.
R L L0
1K 7.96m 7.96m
IVm
V R0
0 C
1K
15.9n
+500.000m
+400.000m
+300.000m
+200.000m
+100.000m
+0.000e+000
MAG(V(IVM))
SOLUTION TO 21.20. (a) From figure P21.19a
Vin V1
= I1, V2 = Vout
Rs
in which case
V1 = z11I1 + z12 0 = Vin Rs I1
Also
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
= =
Vin y22 +
1 y 22 + GL
RL
Consider here that
1 1
Vout z 1 s 2s
= 21 = 2 = =
Vin z11 + 1 s + 2s + 1 1 s 1
s+ + 2 +
s 2 2s
1 s 1
Hence z21 = and z11 = + . This leads to the circuit
2s 2 2s
1
with L = H and C = 2 F.
2
( c) Similarly,
1
y 21 2s
=
y 22 + 1 s 1
+ +1
2 2s
1 s 1
implies y 21 = and y 22 = + .
2s 2s 2s
1
L = 2 H and C = F.
2
SOLUTION TO 21.21.
1
Y R1
H(s) = in = 1
Yout Cs +
R2
SOLUTION TO 21.22.
1 1 1 1
R1 C2s C1R1 R3C2
H(s) = 1 1 = 1 1
Cs + R3 + s+ s+
R2 C2s C1 R2 R3C2
10 1 10
H(s) = =
s + 1 s + 1 (s + 1)2
(c)
Kf = 1e5;
Km = 0.1/(Kf*1e-9)
Km = 1000
Hence, in the final design
(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 w0/Q Num w0 Q KNLP KMA Kf=wp KmR KMB KMC
1 1.41 1.00 1.0000 0.7071 1.00 22507.86 6283.20 10000.00 22507.86 15915.46
KMS
19492.37
RA RB K C1 C2 R1 R2 alpha R3 R4
Design A 0 1 2Q 1/(2Q) 1 1 KNLP/K 1/alpha 1/(1-alpha)
Design B 1 1 2 1 1/Q 1 Q KNLP/K 1/alpha 1/(1-alpha)
Design C 1 1-1/Q 3-1/Q 1 1 1 1 KNLP/K 1/alpha 1/(1-alpha)
Saraga RA RA/3 3-Apr rt(3)Q 1 1/Q 1/rt(3) KNLP/K 1/alpha 1/(1-alpha)
RA RB K C1 C2 R1 R2 alpha R3 R4
Design A 0.0000 1.0000 1.4142 0.7071 1.0000 1.0000 1.0000 1.0000 #DIV/0!
Design B 1.00 1.0000 2.0000 1.0000 1.4142 1.0000 0.7071 0.5000 2.0000 2.0000
Design C 1.00 -0.4142 1.5858 1.0000 1.0000 1.0000 1.0000 0.6306 1.5858 2.7071
Saraga 3.00 1.0000 1.3333 1.2247 1.0000 1.4142 0.5774 0.7500 1.3333 5.6569
w0 scale
Design A 0.0000 1.0000 1.4142 0.7071 1.0000 1.0000 1.0000 1.0000 #DIV/0!
Design B 1.00 1.0000 2.0000 1.0000 1.4142 1.0000 0.7071 0.5000 2.0000 2.0000
Design C 1.00 0.1140 2.1140 1.0000 1.0000 1.0000 1.0000 0.6306 1.5858 2.7071
Saraga 3.00 1.0000 1.3333 1.2247 1.0000 1.4142 0.5774 0.7500 1.3333 5.6569
wp scale
Design A 0.000E+00 1.000E+00 2.251E-04 1.125E-04 1.000E+00 1.000E+00 1.000E+00 1.000E+00 #DIV/0!
Design B 1.00 1.000E+00 2.000E+00 1.592E-04 2.251E-04 1.000E+00 7.071E-01 5.000E-01 2.000E+00 2.000E+00
Design C 1.00 1.140E-01 2.114E+00 1.592E-04 1.592E-04 1.000E+00 1.000E+00 6.306E-01 1.586E+00 2.707E+00
Saraga 3.00 1.000E+00 1.333E+00 1.949E-04 1.592E-04 1.414E+00 5.774E-01 7.500E-01 1.333E+00 5.657E+00
Km scale
Design A 0.000E+00 1.000E+00 1.000E-08 5.000E-09 2.251E+04 2.251E+04 1.000E+00 2.251E+04 #DIV/0!
Design B 10000 1.000E+04 2.000E+00 7.0711E-09 1.000E-08 2.251E+04 1.592E+04 5.000E-01 4.502E+04 4.502E+04
Design C 10000 1.140E+03 2.114E+00 1.000E-08 1.000E-08 1.592E+04 1.592E+04 6.306E-01 2.524E+04 4.308E+04
Saraga 30000 1.000E+04 1.333E+00 1.000E-08 8.165E-09 2.757E+04 1.125E+04 7.500E-01 2.599E+04 1.103E+05
SOLUTION TO 21.24 AND 21.25. In problem 21.8, the transfer function information was computed in
MATLAB as:
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 w0/Q Num w0 Q KNLP KMA Kf=wp KmR KMB KMC
1 1.00 1.00 1.0000 1.0000 1.000 41016.58 975.22 10000.00 20508.29 20508.29
KMS
35521.40
RA RB K C1 C2 R1 R2 alpha R3 R4
Design A 0 1 2Q 1/(2Q) 1 1 KNLP/K 1/alpha 1/(1-alpha)
Design B 1 1 2 1 1/Q 1 Q KNLP/K 1/alpha 1/(1-alpha)
Design C 1 1-1/Q 3-1/Q 1 1 1 1 KNLP/K 1/alpha 1/(1-alpha)
Saraga RA RA/3 3-Apr rt(3)Q 1 1/Q 1/rt(3) KNLP/K 1/alpha 1/(1-alpha)
RA RB K C1 C2 R1 R2 alpha R3 R4
Design A 0.0000 1.0000 2.0000 0.5000 1.0000 1.0000 1.0000 1.0000 #DIV/0!
Design B 1.00 1.0000 2.0000 1.0000 1.0000 1.0000 1.0000 0.5000 2.0000 2.0000
Design C 1.00 0.0000 2.0000 1.0000 1.0000 1.0000 1.0000 0.5000 2.0000 2.0000
Saraga 3.00 1.0000 1.3333 1.7321 1.0000 1.0000 0.5774 0.7500 1.3333 4.0000
w0 scale
Design A 0.0000 1.0000 2.0000 0.5000 1.0000 1.0000 1.0000 1.0000 #DIV/0!
Design B 1.00 1.0000 2.0000 1.0000 1.0000 1.0000 1.0000 0.5000 2.0000 2.0000
Design C 1.00 0.1140 2.1140 1.0000 1.0000 1.0000 1.0000 0.5000 2.0000 2.0000
Saraga 3.00 1.0000 1.3333 1.7321 1.0000 1.0000 0.5774 0.7500 1.3333 4.0000
wp scale
Design A 0.000E+00 1.000E+00 2.051E-03 5.127E-04 1.000E+00 1.000E+00 1.000E+00 1.000E+00 #DIV/0!
Design B 1.00 1.000E+00 2.000E+00 1.025E-03 1.025E-03 1.000E+00 1.000E+00 5.000E-01 2.000E+00 2.000E+00
Design C 1.00 1.140E-01 2.114E+00 1.025E-03 1.025E-03 1.000E+00 1.000E+00 5.000E-01 2.000E+00 2.000E+00
Saraga 3.00 1.000E+00 1.333E+00 1.776E-03 1.025E-03 1.000E+00 5.774E-01 7.500E-01 1.333E+00 4.000E+00
Km scale
Design A 0.000E+00 1.000E+00 5.000E-08 1.250E-08 4.102E+04 4.102E+04 1.000E+00 4.102E+04 #DIV/0!
Design B 10000 1.000E+04 2.000E+00 5.000E-08 5.000E-08 2.051E+04 2.051E+04 5.000E-01 4.102E+04 4.102E+04
Design C 10000 1.140E+03 2.114E+00 5.000E-08 5.000E-08 2.051E+04 2.051E+04 5.000E-01 4.102E+04 4.102E+04
Saraga 30000 1.000E+04 1.333E+00 5.000E-08 2.887E-08 3.552E+04 2.051E+04 7.500E-01 4.736E+04 1.421E+05
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.
KMS
40559.76
RA RB K C1 C2 R1 R2 alpha R3 R4
Design A 0 1 2Q 1/(2Q) 1 1 KNLP/K 1/alpha 1/(1-alpha)
Design B 1 1 2 1 1/Q 1 Q KNLP/K 1/alpha 1/(1-alpha)
Design C 1 1-1/Q 3-1/Q 1 1 1 1 KNLP/K 1/alpha 1/(1-alpha)
Saraga RA RA/3 3-Apr rt(3)Q 1 1/Q 1/rt(3) KNLP/K 1/alpha 1/(1-alpha)
RA RB K C1 C2 R1 R2 alpha R3 R4
Design A 0.0000 1.0000 2.6131 0.3827 1.0000 1.0000 1.0000 1.0000 #DIV/0!
Design B 1.00 1.0000 2.0000 1.0000 0.7654 1.0000 1.3066 0.5000 2.0000 2.0000
Design C 1.00 0.2346 2.2346 1.0000 1.0000 1.0000 1.0000 0.4475 2.2346 1.8100
Saraga 3.00 1.0000 1.3333 2.2630 1.0000 0.7654 0.5774 0.7500 1.3333 3.0615
w0 scale
Design A 0.0000 1.0000 2.6131 0.3827 1.0000 1.0000 1.0000 1.0000 #DIV/0!
Design B 1.00 1.0000 2.0000 1.0000 0.7654 1.0000 1.3066 0.5000 2.0000 2.0000
Design C 1.00 0.1140 2.1140 1.0000 1.0000 1.0000 1.0000 0.4475 2.2346 1.8100
Saraga 3.00 1.0000 1.3333 2.2630 1.0000 0.7654 0.5774 0.7500 1.3333 3.0615
wp scale
Design A 0.000E+00 1.000E+00 4.683E-03 6.859E-04 1.000E+00 1.000E+00 1.000E+00 1.000E+00 #DIV/0!
Design B 1.00 1.000E+00 2.000E+00 1.792E-03 1.372E-03 1.000E+00 1.307E+00 5.000E-01 2.000E+00 2.000E+00
Design C 1.00 1.140E-01 2.114E+00 1.792E-03 1.792E-03 1.000E+00 1.000E+00 4.475E-01 2.235E+00 1.810E+00
Saraga 3.00 1.000E+00 1.333E+00 4.056E-03 1.792E-03 7.654E-01 5.774E-01 7.500E-01 1.333E+00 3.061E+00
Km scale
Design A 0.000E+00 1.000E+00 1.000E-07 1.464E-08 4.683E+04 4.683E+04 1.000E+00 4.683E+04 #DIV/0!
Design B 10000 1.000E+04 2.000E+00 1.000E-07 7.654E-08 1.792E+04 2.342E+04 5.000E-01 3.585E+04 3.585E+04
Design C 10000 1.140E+03 2.114E+00 1.000E-07 1.000E-07 1.792E+04 1.792E+04 4.475E-01 4.005E+04 3.244E+04
Saraga 30000 1.000E+04 1.333E+00 1.000E-07 4.419E-08 3.104E+04 2.342E+04 7.500E-01 5.408E+04 1.242E+05
KMS
17922.81
RA RB K C1 C2 R1 R2 alpha R3 R4
Design A 0 1 2Q 1/(2Q) 1 1 KNLP/K 1/alpha 1/(1-alpha)
Design B 1 1 2 1 1/Q 1 Q KNLP/K 1/alpha 1/(1-alpha)
Design C 1 1-1/Q 3-1/Q 1 1 1 1 KNLP/K 1/alpha 1/(1-alpha)
Saraga RA RA/3 3-Apr rt(3)Q 1 1/Q 1/rt(3) KNLP/K 1/alpha 1/(1-alpha)
RA RB K C1 C2 R1 R2 alpha R3 R4
Design A 0.0000 1.0000 1.0824 0.9239 1.0000 1.0000 1.0000 1.0000 #DIV/0!
Design B 1.00 1.0000 2.0000 1.0000 1.8478 1.0000 0.5412 0.5000 2.0000 2.0000
Design C 1.00 -0.8478 1.1522 1.0000 1.0000 1.0000 1.0000 0.8679 1.1522 7.5703
Saraga 3.00 1.0000 1.3333 0.9374 1.0000 1.8478 0.5774 0.7500 1.3333 7.3912
w0 scale
Design A 0.0000 1.0000 1.0824 0.9239 1.0000 1.0000 1.0000 1.0000 #DIV/0!
Design B 1.00 1.0000 2.0000 1.0000 1.8478 1.0000 0.5412 0.5000 2.0000 2.0000
Design C 1.00 0.1140 2.1140 1.0000 1.0000 1.0000 1.0000 0.8679 1.1522 7.5703
Saraga 3.00 1.0000 1.3333 0.9374 1.0000 1.8478 0.5774 0.7500 1.3333 7.3912
wp scale
Design A 0.000E+00 1.000E+00 1.940E-03 1.656E-03 1.000E+00 1.000E+00 1.000E+00 1.000E+00 #DIV/0!
Design B 1.00 1.000E+00 2.000E+00 1.792E-03 3.312E-03 1.000E+00 5.412E-01 5.000E-01 2.000E+00 2.000E+00
Design C 1.00 1.140E-01 2.114E+00 1.792E-03 1.792E-03 1.000E+00 1.000E+00 8.679E-01 1.152E+00 7.570E+00
Saraga 3.00 1.000E+00 1.333E+00 1.680E-03 1.792E-03 1.848E+00 5.774E-01 7.500E-01 1.333E+00 7.391E+00
Km scale
Design A 0.000E+00 1.000E+00 1.000E-07 8.536E-08 1.940E+04 1.940E+04 1.000E+00 1.940E+04 #DIV/0!
Design B 10000 1.000E+04 2.000E+00 5.412E-08 1.000E-07 3.312E+04 1.792E+04 5.000E-01 6.624E+04 6.624E+04
Design C 10000 1.140E+03 2.114E+00 1.000E-07 1.000E-07 1.792E+04 1.792E+04 8.679E-01 2.065E+04 1.357E+05
Saraga 30000 1.000E+04 1.333E+00 9.374E-08 1.000E-07 3.312E+04 1.035E+04 7.500E-01 2.390E+04 1.325E+05
SOLUTION TO 21.28. For this problem we use the excel spread sheet given below. First we observe that
K K K 02 (= K NLP )
H(s) = H ( 0 s) = = 1
s2 + 0 s + 20
Q
( 0 s) +
2
Q
0
( 0 s) + 0
2 s2 + s + 1
Q
Thus after this type of frequency scaling, the new transfer function is:
0.7943
Hnew (s) = 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 w0/Q Num w0 Q KNLP KMA Kf=wp KmR KMB KMC
0.82306 0.8038 0.65378 0.9072 1.1286 0.7943 1131.42 43982.40 10000.00 501.23 501.23
KMS
979.83
RA RB K C1 C2 R1 R2 alpha R3 R4
Design A 0 1 2Q 1/(2Q) 1 1 KNLP/K 1/alpha 1/(1-alpha)
Design B 1 1 2 1 1/Q 1 Q KNLP/K 1/alpha 1/(1-alpha)
Design C 1 1-1/Q 3-1/Q 1 1 1 1 KNLP/K 1/alpha 1/(1-alpha)
Saraga RA RA/3 3-Apr rt(3)Q 1 1/Q 1/rt(3) KNLP/K 1/alpha 1/(1-alpha)
RA RB K C1 C2 R1 R2 alpha R3 R4
Design A 0.0000 1.0000 2.2573 0.4430 1.0000 1.0000 0.7943 1.2589 4.8621
Design B 1.00 1.0000 2.0000 1.0000 0.8860 1.0000 1.1286 0.3972 2.5179 1.6588
Design C 1.00 0.1140 2.1140 1.0000 1.0000 1.0000 1.0000 0.3758 2.6613 1.6019
Saraga 3.00 1.0000 1.3333 1.9549 1.0000 0.8860 0.5774 0.5957 1.6786 2.1917
w0 scale
Design A 0.0000 1.0000 2.4881 0.4883 1.0000 1.0000 0.7943 1.2589 4.8621
Design B 1.00 1.0000 2.0000 1.1023 0.9766 1.0000 1.1286 0.3972 2.5179 1.6588
Design C 1.00 0.1140 2.1140 1.1023 1.1023 1.0000 1.0000 0.3758 2.6613 1.6019
Saraga 3.00 1.0000 1.3333 2.1548 1.1023 0.8860 0.5774 0.5957 1.6786 2.1917
wp scale
Design A 0.000E+00 1.000E+00 5.657E-05 1.110E-05 1.000E+00 1.000E+00 7.943E-01 1.259E+00 4.862E+00
Design B 1.00 1.000E+00 2.000E+00 2.506E-05 2.220E-05 1.000E+00 1.129E+00 3.972E-01 2.518E+00 1.659E+00
Design C 1.00 1.140E-01 2.114E+00 2.506E-05 2.506E-05 1.000E+00 1.000E+00 3.758E-01 2.661E+00 1.602E+00
Saraga 3.00 1.000E+00 1.333E+00 4.899E-05 2.506E-05 8.860E-01 5.774E-01 5.957E-01 1.679E+00 2.192E+00
Km scale
Design A 0.000E+00 1.000E+00 5.000E-08 9.813E-09 1.131E+03 1.131E+03 7.943E-01 1.424E+03 5.501E+03
Design B 10000 1.000E+04 2.000E+00 5.000E-08 4.430E-08 5.012E+02 5.657E+02 3.972E-01 1.262E+03 8.315E+02
Design C 10000 1.140E+03 2.114E+00 5.000E-08 5.000E-08 5.012E+02 5.012E+02 3.758E-01 1.334E+03 8.029E+02
Saraga 30000 1.000E+04 1.333E+00 5.000E-08 2.558E-08 8.682E+02 5.657E+02 5.957E-01 1.645E+03 2.148E+03
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 = =
j p 2
5.5 2
c + 2 c + 1 j 2
5.5
+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
1 LC
Hcir (s) =
1 1 2
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
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
L C 1 1 s Km
(c) Here Km = 1000. Ls and s. Thus in
s 1 Cs C C C C
s
LKm 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.
p 1
= , Amax = 2 dB, = max = 10 0.12 1 = 0.76478, c = 6 = 1.0935
10 0.1 2
1
Wc=1/(10^0.2-1)^(1/6)
Wc = 1.0935e+00
(b)
wchp = 2*pi*5e3/Wc
wchp = 2.8730e+04
fchp = wchp/(2*pi)
fchp = 4.5725e+03
p
Thus = = 28.73 krad/s.
chp c
1 1
Chp = = 17.404 F, Lhp = = 34.807 H
Llp chp Clp chp
Chp
K m = 174.02, Chp,new = = 100 nF , Lhp,new = K m Lhp = 6.06 mH
Km
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:
1 s2 s2
HNHP (s) = HNLP = 2 = K 2
s s + 2s + 1 s + d(1) s + d(2)
INPUT: d(1) d(2) K K Kf Km KmR
1.414213562 1 1.33334 1 18849.6 12247.44871 30000
A plot of the design without input attenuation is shown below. Notice that as predicted the gain is 4/3.
R
10k
Vin C C0
3.063n 7.503n 15
Vpls
XOpAmp Vminus
R2
10k 15
R0
12.247k IVm
R1
30k
S&K HP-Small Signal AC-3 Frequency (Hz)
+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
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')
0.9
0.8
Magnitude 4th Order Butterworth
0.7
0.6
0.5
0.4
0.3
TextEnd
0.2
0.1
0
0 0.5 1 1.5 2 2.5 3 3.5
Normalized Frequency
The Saraga design parameters are given in the following Excel tables:
INPUT: d(1) d(2) KNLP/NHP w0LP/HP Q
NLP H(s) 0.76537 1 1 1 1.306557613
NHP H(s) 0.76537 1 1 1 1.306557613
K Kf Km KmR alpha
1.33334 18849.6 22630.24168 60000 0.74999625
K Kf Km KmR alpha
1.33334 18849.6 10000 30000 0.74999625
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
-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 1 s
HNHP (s) = HNLP = =
2 (1 0LP )
s 2 2
1 1 2 ( 0LP ) s + (1 0LP)
2
+ 0LP + ( 0LP ) s +
s Q s Q
( 0HP )2 s2
=
s2 + 0HP s + ( 0HP ) 2
Q
The S&K Saraga design for d1 is given by the following excel spreadsheet:
K Kf Km KmR alpha
1.33334 31416 45260.48336 60000 0.74999625
The S&K Saraga design for d2 is given by the following excel spreadsheet:
INPUT: d(1) d(2) KNLP/NHP w0LP/HP Q
NLP H(s) 1.8478 1 1 1 0.541184111
NHP H(s) 1.8478 1 1 1 0.541184111
K Kf Km KmR alpha
1.33334 31416 20000 60000 0.74999625
8
8
H(s) = = L
Ls + 8 s + 8
L
8
Thus, = 2000 2 L = 636 H.
L
For the tweeter,
8 8Cs s
H(s) = 1 = =
+ 8 8Cs + 1 s +
1
Cs 8C
1
Thus, = 2000 2 C = 9.95 F.
8C
1
HNLP (s) = 2
s + 2s + 1
1 1
Thus 2= 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
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
V fb Z Z k 1 1 I fb
Vout,b = V fb + I fb + Z 1 = V fb + I fb 1 + V fb = (2k 1)V fb +
1 k k k k Z1
k 1
If
(2k 1)V fb V fa and I fb I fa (**)
then
1 Vout ,a
Vout,b H a (s)Vin =
k k
Z1
For gain enhancement, k < 1. However, for the (**) to be valid, we require that 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.
CHAPTER 22 PROBLEM SOLUTIONS
[ ]
1 1
1
e ( ln(2)) t e j 2nt dt = ( ln(2)+ j 2n ) t
cn = e dt = e ( ln(2)+ j 2n ) 1
0 0
(ln(2) + j2n)
0.5
=
(ln(2) + j2n)
(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
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..
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) = +
T cos(n 0t)
T n=1
Therefore
2A
f '( t) = 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
A
f (t) = 0.5A sin(n 0t)
n n=1
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
4
f '( t) = 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
4A cos(n 0t)
f (t) = 0.5A +
2 n=1,odd n2
= 1 {
2Asin(n)
[cos(n0 (t+0.5T))-cos(n0 (t-0.5T)]
T n=1 n \
= 1 [
- 4Asin2 (n)
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
f(t) = A + 2A { [
sin(n) 2
] cos(n0 t)}
n=1
2 n
f'(t)= 1 fp (t + 0.5T) - f (t )
T
n=1
= 2A {[
sin(n)
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
f(t) =0.5A + A { [
sin(n)
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 = A {sin(n)cos(n) -n}
2 n2
bn = -A {sin(n)cos(n) -n}
2 n2
1 1
f ''( t) = 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
[cos(n 0 t + 2n ) cos(n 0 t)]
2A
f ''( t) =
(1 )T 2 n=1
4A
= [sin(n 0t + n )sin(n )]
(1 )T 2 n=1
Therefore,
4A sin(n )
f (t) = [ f (t)]ave +
(1 )T 2 n=1 ( n 0) 2
sin(n 0 t + n )
A A sin(n )
= + 2 2 cos(n 0t + (n 0.5) )
2 n=1 n (1 )
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:
YL 1 1 1
H(j) = = = =
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,
2
Pav =204.48 = 4.1812 W
104
(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
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
SOLUTION PROBLEM 22.15. (a) This proof is a special case of the general proof given in the
TO OBTAIN
(b)
voutmin =
1.1235e+01
voutmax =
8.3013e+01
(c)
error1 =
8.9045e+00
error2 =
-1.2048e+00
s
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
V KV KV
con = L1 con con = KVcon 1 e t
s s s +1 ( )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.
Note that = e 0.5T / . Similarly, applying equation 8.19 to the interval [ T/2, T] leads to
1 a0 (1 )KVmax
1 =
b0 (1 )KVmin
(3)
V Vmin V V
= KVmax K max = KVmax K max 0.5Tmin
1 + 1/ 1+ e /
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)))
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
0.2
0.1
vout in V
-0.1
-0.2
-0.3
0 50 100 150 200
time in seconds
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
V22 + V23
H.D. = 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
Using A=1, and = 0.1864in item #2 of table 22.4, equation 22.37 becomes
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
1 T
t
2
RC t ( 1 )
v 0 (t ) = (Vm cos( 1)) e = (Vm cos( 1)) e RC
2 1 +
cos( 1 )e RC = cos( 2 )
The desired proof is complete.
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 -
2 2RC RC
For the diode average and peak currents, use equations 22.49a and 22.52
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
1 - cos( 2 )
ripple factor =0.12%
3 [ 1 +cos( 2 )]
For the diode average and peak currents, use equations 22.49a and 22.52
1 1
Z par ( j ) = 1 C
j C+
RL
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,
1
Zin ( j )
C
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
R 1950
H ( j2 120) 0.02 0.02 = 0.02 = 0.18042 V
( RC ) ( )
2 2
2 120 1950 10 106
Hence, the effective value is 0.18042 2 = 0.12757 V and the ripple factor is
0.12757
100 = 0.42524 %
30
H(s) = Vo = Vin Vo
I I Vin
Under the condition 1/(C) << R and 1/(C) << RL, we have
Vo 1
and 2
Vin 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
Hence, the effective value is 0.012/ 2 =0.0848 V and the ripple factor is
0.0848 100% = 0.2827 %
30
0.5s
0.25s
s s -0.25 s
-jnT 0 s
-0.25 s
-0.5 s
= 1 sin (n ), n = 1,2,...
n 2
b n = 0,
dn=an = 2 sin (n ),
n 2
The Fourier series representation of H(j) is
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)
0.5
-0.5
0 20 40 60 80 100
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