You are on page 1of 67

Electrical Circuits I

Lecture 10

Dr. Abdallah Ramadan Fawzy


E-Mail: Abdallah.gad@must.edu.eg

The figures and text included in slides are borrowed from various books, websites, and
other sources for academic purposes only. The author do not claim any originality.
Chap. 6, Capacitors and Inductors
Introduction
Capacitors
Series and Parallel Capacitors
Inductors
Series and Parallel Inductors
6.1 Introduction
Resistor: a passive element which dissipates energy only
Two important passive linear circuit elements:
1) Capacitor
2) Inductor

Capacitor and inductor can store energy only and they can neither
generate nor dissipate energy.
Michael Faraday (1971-1867)
6.2 Capacitors
A capacitor consists of two conducting plates separated by an insulator (or
dielectric).

εA
C=
d
 =  r 0
 0 = 8.854 10 −12 (F/m)
εA
C=
d
Three factors affecting the value of capacitance:
1. Area: the larger the area, the greater the capacitance.
2. Spacing between the plates: the smaller the spacing, the greater the
capacitance.
3. Material permittivity: the higher the permittivity, the greater the capacitance.
Fig 6.4

(a) Polyester capacitor, (b) Ceramic capacitor, (c) Electrolytic capacitor


Fig 6.5

Variable capacitors
Fig 6.3
Fig 6.2
Charge in Capacitors
The relation between the charge in plates and the voltage across a
capacitor is given below.
q = Cv
q
1F = 1 C/V
Linear
Nonlinear

v
Voltage Limit on a Capacitor
Since q=Cv, the plate charge increases as the voltage increases. The
electric field intensity between two plates increases. If the voltage
across the capacitor is so large that the field intensity is large enough to
break down the insulation of the dielectric, the capacitor is out of work.
Hence, every practical capacitor has a maximum limit on its operating
voltage.
I-V Relation of Capacitor
dq dv
+ i q = Cv, i = =C
C
dt dt
v

-
Physical Meaning + i

C
dv v
i=C
dt -
• when v is a constant voltage, then i=0; a constant
voltage across a capacitor creates no current through
the capacitor, the capacitor in this case is the same as
an open circuit.
• If v is abruptly changed, then the current will have an
infinite value that impossibles practically impossible.
Hence, a capacitor is to have an abrupt change in its
voltage except an infinite current is applied.
Fig 6.7
A capacitor is an open circuit to dc.
The voltage on a capacitor cannot change abruptly.

Abrupt change
dv 1 t +
v(t ) =  idt ( v(−) = 0 )
i
i=C
dt C − v C

1 t
v(t ) =  idt + v(to)
C to
( v(to) = q(to) / C ) -

The charge on a capacitor is an integration of current through the capacitor.


Hence, the memory effect counts.
Energy Storing in Capacitor
dv
p = vi = Cv
dt
t t dv v (t ) 1 2
w = − pdt = C − v dt = C v ( − ) vdv = Cv v (t )
v ( − )
dt 2
1
w(t ) = Cv 2 (t ) ( v(−) = 0) + i
2 C
v

q 2 (t )
w(t ) = -
2C
Example 6.1
(a) Calculate the charge stored on a 3-pF capacitor with 20V across it.
(b) Find the energy stored in the capacitor.
Example 6.1
Solution:
q = Cv,
(a) Since
−12
q = 3  10  20 = 60pC
(b) The energy stored is

1 2 1 −12
w = Cv =  3  10  400 = 600pJ
2 2
Example 6.2
The voltage across a 5- F capacitor is

v(t ) = 10 cos 6000t V


Calculate the current through it.
Solution:
By definition, the current is

dv −6 d
i=C = 5 10 (10 cos 6000t )
dt dt
= −5 10−6  6000 10 sin 6000t = −0.3 sin 6000t A
Example 6.3
Determine the voltage across a 2-F capacitor if the current through it is
−3000t
i (t ) = 6e mA
Assume that the initial capacitor voltage is zero.
Solution:
Since 1 t
v = 0 idt + v(0) and v(0) = 0,
C
v=
1 t −3000t 3  10 −3000t t
3
−6 0
−3
6e dt 10 = e
2  10 − 3000 0

−3000t
= (1 − e )V
Example 6.4
Determine the current through a 200- F capacitor whose voltage is
shown in Fig 6.9.
Example 6.4
Solution:
The voltage waveform can be described mathematically as

 50t V 0  t 1
 100 − 50t V 1 t  3
v(t ) = 
− 200 + 50t V 3t  4

 0 otherwise
Example 6.4
Since i = C dv/dt and C = 200 F, we take the derivative of to obtain

 50 0  t  1  10mA 0  t 1
−6 − 50 1  t  3 − 10mA 1 t  3
i (t ) = 200  10   =
50 3  t  4 10mA 3t  4
 
 0 otherwise  0 otherwise

Thus the current waveform is shown in Fig.6.10.


Example 6.4
Example 6.5
Obtain the energy stored in each capacitor in Fig. 6.12(a) under dc
condition.
Example 6.5
Solution:
Under dc condition, we replace each capacitor with an open circuit. By
current division,
3
i= (6mA) = 2mA
3+ 2+ 4
 v1 = 2000 i = 4 V, v 2 = 4000i = 8 V
1 1 −3
 w1 = C1v1 = (2  10 )(4) = 16mJ
2 2

2 2
1 1 −3
w2 = C2 v2 = (4  10 )(8) = 128mJ
2 2

2 2
Fig 6.14

Ceq = C1 + C2 + C3 + .... + CN
6.3 Series and Parallel Capacitors
i = i1 + i2 + i3 + ... + iN
dv dv dv dv
i = C1 + C2 + C3 + ... + C N
dt dt dt dt
 N
 dv
=   CK  = Ceq
dv
 k =1  dt dt
Ceq = C1 + C2 + C3 + .... + C N
The equivalent capacitance of N parallel-connected capacitors is the sum of
the individual capacitance.
Fig 6.15

1 1 1 1 1
= + + + ... +
Ceq C1 C2 C3 CN
Series Capacitors
v(t ) = v1 (t ) + v2 (t ) + ... + vN (t )

1 t 1 1 1 1 t
Ceq −id = ( C1 + C2 + C3 + ... + CN )−id
q (t ) q (t ) q (t ) q(t )
= + ++
Ceq C1 C2 CN

The equivalent capacitance of series-connected capacitors is the reciprocal of


the sum of the reciprocals of the individual capacitances.

1 1 1 C1C2
= + Ceq =
Ceq C1 C2 C1 + C2
Summary
These results enable us to look the capacitor in this way: 1/C has the
equivalent effect as the resistance. The equivalent capacitor of
capacitors connected in parallel or series can be obtained via this point
of view, so is the Y-△ connection and its transformation
Example 6.6
Find the equivalent capacitance seen between terminals a and b of the
circuit in Fig 6.16.
Example 6.6
Solution:

• 20 − F and 5 − F capacitors are in series:


20  5
 = 4F
20 + 5
• 4 − F capacitor is in parallel with the 6 − F
and 20 − F capacitors:
 4 + 6 + 20 = 30F
• 30 − F capacitor is in series with
the 60 − F capacitor. 30  60
Ceq = F = 20F
30 + 60
Example 6.7
For the circuit in Fig 6.18, find the voltage across each capacitor.
Example 6.7
Example 6.7
Solution:
Two parallel capacitors: 1
 Ceq = 1 1 1 mF = 10mF
+ +
60 30 20

Total charge −3
q = Ceq v = 10  10  30 = 0.3 C

This is the charge on the 20-mF and 30-mF capacitors, because they are
in series with the 30-v source. ( A crude way to see this is to imagine
that charge acts like current, since i = dq/dt)
Example 6.7
q 0.3
v1 = = −3
= 15 V,
C1 20  10
Therefore,
q 0.3
v2 = = −3
= 10 V
C2 30  10

Having determined v1 and v2, we now use KVL to determine v3


by
v3 = 30 − v1 − v2 = 5V
Alternatively, since the 40-mF and 20-mF capacitors
are in parallel, they have the same voltage v3 and
their combined capacitance is 40+20=60mF.
q 0.3
 v3 = = −3
= 5V
60mF 60  10
Joseph Henry (1979-1878)
6.4 Inductors
An inductor is made of a coil of conducting wire

N A
2
L=
l
Fig 6.22 L=
N 2 A
l
 =  r 0
 0 = 4 10 −7 (H/m)

N : number of turns.
l :length.
A:cross − sectional area.
 : permeability of the core
In electromagnetism, permeability is the measure of
magnetization produced in a material in response to
an applied magnetic field...
Fig 6.23

(a) air-core
(b) iron-core
(c) variable iron-core
Flux in Inductors
The relation between the flux in inductor and the current through the
inductor is given below.
 = Li
ψ Linear
1H = 1 Weber/A
Nonlinear

i
Energy Storage Form
An inductor is a passive element designed to store energy in the
magnetic field while a capacitor stores energy in the electric field.
I-V Relation of Inductors
An inductor consists of a coil of
i
conducting wire.
+

d di
v= =L v
L
dt dt
-
Physical Meaning

d di
v= =L
dt dt
When the current through an inductor is a constant, then the voltage across the
inductor is zero, same as a short circuit.
No abrupt change of the current through an inductor is possible except an
infinite voltage across the inductor is applied.
The inductor can be used to generate a high voltage, for example, used as an
igniting element.
Fig 6.25
An inductor are like a short circuit to dc.
The current through an inductor cannot change instantaneously.
1 1 t
di = vdt i =  v(t )dt
L L − +

v
1 t
i =  v(t )dt + i (to)
L

L to -

The inductor has memory.


Energy Stored in an Inductor

 di 
P = vi =  L  i +
 dt 
t t di
w = − pdt = −  L idt  v
L

 dt  -
i (t ) 1 2 1 2
= L i ( − ) i di = Li (t ) − Li (−) i (−) = 0,
2 2
The energy stored in an inductor

1 2
w(t ) = Li (t )
2
Model of a Practical Inductor
Example 6.8
The current through a 0.1-H inductor is i(t) = 10te-5t A. Find the voltage
across the inductor and the energy stored in it.
Solution:

di
Since v = L and L = 0.1H,
dt
d
v = 0.1 (10te −5t ) = e −5t + t (−5)e −5t = e −5t (1 − 5t )V
dt
The energy stored is
1 2 1
w = Li = (0.1)100t 2 e −10t = 5t 2 e −10t J
2 2
Find the current through a 5-H inductor if the voltage across it is

30t 2
, t 0
v(t ) = 
 0, t0

Also find the energy stored within 0 < t < 5s. Assume i(0)=0.
Solution:

1 t
Since i = t v(t )dt + i (t0 ) and L = 5H.
L 0

1 t 2 t3
i = 0 30t dt + 0 = 6  = 2t 3 A
5 3
Example 6.9
The power p = vi = 60t , and the energy stored is then
5

6
5 t 5
w =  pdt = 0 60t dt = 60
5
= 156.25 kJ
60
Alternatively, we can obtain the energy stored using
Eq.(6.13), by writing
1 2 1
w(5) − w(0) = Li (5) − Li (0)
2 2
1
= (5)(2  5 ) − 0 = 156.25 kJ
3 2

2
as obtained before.
Example 6.10
Consider the circuit in Fig 6.27(a).
Under dc conditions, find:
(a) i, vC, and iL.
(b) the energy stored in the capacitor
and inductor.
Example 6.10

(a) Under dc condition : capacitor → open circuit


Solution:
inductor → short circuit
12
i = iL = = 2 A, vc = 5i = 10 V
1+ 5
(b) 1 1
wc = Cvc = (1)(10 ) = 50J,
2
2

2 2
1 2 1
wL = Li = (2)(2 2 ) = 4J
2 2
Inductors in Series

Leq = L1 + L2 + L3 + ... + LN
Inductors in Parallel

1 1 1 1
= + ++
Leq L1 L2 LN
6.5 Series and Parallel Inductors
Applying KVL to the loop,
v = v1 + v2 + v3 + ... + vN
Substituting vk = Lk di/dt results in
di di di di
v = L1 + L2 + L3 + ... + LN
dt dt dt dt
di
= ( L1 + L2 + L3 + ... + LN )
dt
 N
 di
=   LK  = Leq
di
 K =1  dt dt
Leq = L1 + L2 + L3 + ... + LN
Parallel Inductors
i = i1 + i2 + i3 + ... + iN
Using KCL,1 t
ik = t vdt + ik (t0 )
But Lk o

1 t 1 t 1 t
 i = t vdt + i1 (t0 ) + t vdt + is (t0 ) + ... + t vdt + iN (t0 )
Lk 0
L2 0 LN 0

1 1 1 t
=  + + ... +  t vdt + i1 (t0 ) + i2 (t0 ) + ... + iN (t0 )
 L1 L2 LN  0

N 1t N
1 t
=    t vdt +  ik (t0 ) = t vdt + i (t0 )
 k =1 Lk  0
k =1 Leq 0
The inductor in various connection has the
same effect as the resistor. Hence, the Y-Δ
transformation of inductors can be similarly
derived.
Table 6.1
Example 6.11
Find the equivalent inductance of the circuit shown in Fig. 6.31.
Example 6.11
Solution:
Series : 20H, 12H, 10H
→ 42H
7  42
Parallel : = 6H
7 + 42
 Leq = 4 + 6 + 8 = 18H
Practice Problem 6.11
Example 6.12
Find the circuit in Fig. 6.33, −10t
If find :
i (t ) = 4(2 − e )mA.
i2 (0) = −1 mA, (a) i (0)
1

(b) v(t ), v1 (t ), and v2 (t ); (c) i1 (t ) and i2 (t )


Example 6.12
−10 t
(a ) i (t ) = 4(2 − e
Solution:
)mA → i (0) = 4(2 − 1) = 4mA.
 i1 (0) = i (0) − i2 (0) = 4 − (−1) = 5mA
(b) The equivalent inductanceis
Leq = 2 + 4 || 12 = 2 + 3 = 5H
di −10 t −10 t
 v(t ) = Leq = 5(4)(−1)(−10)e mV = 200e mV
dt
di −10 t −10 t
v1 (t ) = 2 = 2(−4)(−10)e mV = 80e mV
dt
−10 t
 v2 (t ) = v(t ) − v1(t ) = 120e mV
Example 6.12
1 t
(c) i = 0 v(t ) dt + i (0) 
L
1 t 120 t −10t
i1 (t ) = 0 v2 dt + i1 (0) =  e dt + 5 mA
4 4 0
−10 t t
= −3e + 5 mA = −3e −10t + 3 + 5 = 8 − 3e −10t mA
0
1 t 120 t −10t
i2 (t ) = 0 v2 dt + i2 (0) = 0
e dt − 1mA
12 12
−10 t t
= −e − 1mA = −e −10t + 1 − 1 = −e −10t mA
0
Note that i1 (t ) + i2 (t ) = i (t )

You might also like