+Basics for Electronics & Radio Engineering

Functions

Problems) Plot following functions by hand and using MATLAB with your own
discussions.

1-dimensional functions

sin( x)
f ( x) = 100 , f ( x) = 3x + 5 , f ( x) = 2 x 2 + 10 , f ( x) = ( x − 1)( x − 4)( x − 6) , f ( x) = ,
x
e x + e− x e x − e− x
f ( x) = e x , e − x , f ( x) = cosh( x) = (even function), f ( x) = sinh( x) = (odd
2 2
function), f ( x) = 10 log10 ( x) , f ( x) = ln( x) = log e ( x)
( x − m )2
1 −
Normal distribution f ( x) = e 2 2
where m=mean and  =standard deviation
2
(Plot this with various m and  !)

v(t ) = 10cos(2 t ) , v(t ) = 10cos(20 t ) , f (t ) = 10cos(200 t ) ,

  
f (t ) = cos(2 t − ) , f (t ) = 10 cos(2 t − ) , f (t ) = 10 cos(2 t − ) ,
6 3 2
f (t ) = 10cos(2 t −  ) , f (t ) = 10cos(2 t − 2 ) , f (t ) = 10cos(2 t +  )

sin( x)
f ( x) =
x
sin( x)
1
 2
x


x

sin( x)
sin(3 / 2) −1 2
x = =− = −0.2122
1 3 / 2 3 / 2 3

 2
3 / 2 x
−0.2122

sin( x)
= sinc( x)
x
2-dimensional functions : Plot the following functions by insert several positions (x,y)!

f ( x, y) = 5 , f ( x, y ) = x 2 + y 2 , f ( x, y ) = x 2 + y 2 + 5 , f ( x, y ) = x + 2 y ,
  z  sin( x) sin( y )
v( z, t ) = 5cos(10 t −  z ) = 5cos 10  t −   , f ( x, y ) = 10 
  10   x y

3-dimesional functions : How to plot the following functions?

f ( x, y, z ) = 3 , f ( x, y, z ) = 3x , f ( x, y, z ) = 3 y , f ( x, y, z ) = x 2 + y 2 + z 2 , ,
10 10 10 10
f (  ,  , z ) = , f (  ,  , z ) = cos  , f (r ,  ,  ) = , f (r , ,  ) = sin 
  r r

Derivatives

-ordinary differentiation (상미분) : Let f ( x) be a 1-dimensional (1D) function.

df ( x) f f ( x + x) − f ( x) f ( x + x / 2) − f ( x − x / 2)
= lim = lim or lim
dx x → 0 x x → 0 x x → 0 x

V ( x)

V ( x)(Volts )

V
x

x
dV V
= lim (V / m)
dx x →0 x

dV 3( x + x) − 3x V 3x V
If V ( x) = 3x(Volts) , = lim = lim = 3(V / m) .
dx x→0 x m x→0 x m
The voltage increase in a small distance x (=0.01m for example) is given by
dV
V = x = (3V / m)(0.01m) = 0.03(V ) .
dx
The voltage increase in a small distance x (=0.1m for example) is given by
dV
V = x = (3V / m)(0.1m) = 0.3(V ) .
dx
Examples) When f ( x) = x 2 , then
df ( x) f f ( x + x) − f ( x) ( x + x) 2 − x 2 x 2 + 2 xx + ( x) 2 − x 2
= lim = lim = lim = lim = 2x
dx x →0 x x →0 x x →0 x x → 0 x

When f ( x) = e x , then
df ( x) f f ( x + x) − f ( x) e x +x − e x e x (e x − 1) e x (1 + x − 1)
= lim = lim = lim = lim = lim = ex
dx x → 0 x x → 0 x x → 0 x x → 0 x x → 0 x

df
Prob) Plot the following functions. Obtain the ordinary derivatives of the following
dx
functions using the definitions mentioned above. What is the meaning of them?

f ( x) = 10 , f ( x) = x 2 , x 2 + 3, 2 x + 5 , f ( x) = sin x , f ( x) = ax n , f (t ) = 30cos(2 t )

f (t ) = 30cos(120 t ) , f (t ) = 30 cos(120 t + ) , f (t ) = 30sin(120 t ) , ,
3
e x + e− x
f (t ) = 30sin(120 t +  ) , f (t ) = et , f ( x ) = e 2 x , f ( x) = cosh( x) = ,
2
e x − e− x
f ( x) = sinh( x) =
2

Tip : When f ( x) = sin x ,

df ( x) sin( x + x) − sin( x) sin( x) cos( x) + cos( x) sin( x) − sin( x)
= lim = lim
dx x → 0 x x → 0 x
 (x) 2 
sin( x) 1 + − 1 + cos( x)  x
sin( x)  cos( x) − 1 + cos( x) sin(x)  2 
= lim = lim = cos( x)
x → 0 x x → 0 x
using sin(a  b) = sin a cos b  cos a sin b

-partial differentiation (편미분) : Let f ( x, y) be a 2-dimensional function.

x x
f (x + , y) − f ( x − , y)
f ( x, y ) f ( x + x, y ) − f ( x, y ) 2 2
= lim or lim
x x →0 x x →0 x

y y
f ( x, y + ) − f ( x, y − )
f ( x, y ) f ( x, y + y ) − f ( x, y ) 2 2
= lim or lim
y y → 0 y y →0 y

Prob) Plot the following functions. Obtain the partial derivatives of the functions
V ( x) = x 2 (V)
V ( x, y ) = x 2 + y 2 (V)
V ( x, y, z ) = x 2 + y 2 + z 2 (V)
 100
V (r , ,  ) =
r
with units. What is the meaning of them?
V V V V (r , ,  )  2V  2V  2V  2V  2V  2V
, , , , , , , , ,
x  y z r x 2 y 2 z 2 xy yz z x

- Taylor series

f (t )

f (a + x)

f (a + x)  f (a ) + xf (a )
f (a) x
t
a a+x

x2 x3 xn n
f (a + x) = f (a ) + xf (a ) + f (a ) + f (a ) +  + f (a ) + 
2! 3! n!

x2 x3 xn n
If a = 0 , f ( x) = f (0) + xf (0) + f (0) + f (0) +  + f (0) +  (called a
2! 3! n!
Maclaurin series)
d (sin x)
Examples) sin x  f (0) + xf (0) = 0 + x x = 0 = 0 + x  cos x x = 0 = x
dx
x2 d (cos x) x 2 d 2 (cos x)
cos x  f (0) + xf (0) + f (0) = 1 + x x =0 + x =0
2! dx 2! dx 2
d (cos x) x 2 d 2 (cos x) x2
= 1+ x x =0 + x = 0 = 1 + x  ( − sin x ) x = 0 +  (− cos x) x =0
dx 2! dx 2 2!
x2 x2
= 1 + x  (− sin x) x =0 +  (− cos x) x =0 = 1 −
2! 2

d (e x )
e x  f (0) + xf (0) = 1 + x x =0 = 1 + x  ex x =0 = 1+ x
dx

Use your calculator and check the followings !

sin(0.2) = 0.19867  0.2, sin(0.1) = 0.09983  0.1, sin(0.01) = 0.00999983  0.01

(0.2)2 (0.1) 2
cos(0.2) = 0.980066  1 − = 0.98, cos(0.1) = 0.995004 = 1 − = 0.995,
2 2
(0.01)2
cos(0.01) = 0.9999500004  1 − = 0.99995
2
e0.2 = 1.221403  1 + 0.2 = 1.2, e0.1 = 1.105171  1 + 0.1 = 1.1, e0.01 = 1.01005  1 + 0.01 = 1.01

Integration

Let f ( x) be a 1-dimensional (1D) function. Let f ( x, y) be a 2-dimensional function.

b b b d b d
a
f ( x ) dx or  a
dxf ( x) (area), a c
f ( x, y )dxdy or 
a c
dxdyf ( x, y ) (volume)

f ( x)

f ( x)dx f ( x)

dx

f ( x)

a x b
dx x
f ( x)dx :Area of the small rectangle
b
a
f ( x)dx :Area between a and b

Prob) Plot the integrand, obtain the values of the integrations, and express their
meanings. Finally, you have to check your results with those obtained using any
computer program or MATLAB. For a numerical evaluations using a program, dx must
be replaced by a practically small x .

4 4 3  2
0
(2 x + 3)dx ,  0
(4 x −2 ) dx , 1
( x − 1)( x − 3)dx , 0
10 sin  d 0
10sin  d ,
2 1 2 1 2 2 3
0
sin 2  d ,
2 0
sin 2  d ,
2  0
cos 2  d ,  
−2 − 3
( x 2 + y 2 )dxdy
( x − m )2
 1 −

−
2
e 2 2
dx (=1 for any mean m and standard deviation  )

Classification of signals

Sinusoidal (정현신호, AC) : f (t ) = cos t , cos 2t , sin t .. ( = 2 f ) (periodic)

DC (direct current) signal : f (t ) = 10, 30 ( f = 0,  = 0)
Periodic triangular signal (periodic)
Single triangular signal (non-periodic)
Periodic rectangular signal (periodic)
Single pulse (non-periodic)
Single pulse with width approaching zero and height approaching infinity
→ Delta (or called impulse) function (non-periodic) with area of 1 (See below)

Properties of Delta function

t
  t = x  dt =bdx  a +
−
 (t )dt = 1 , − b
 ( ) dt ⎯⎯⎯⎯⎯
b
→ b −  ( x)dx = b , 
a −
 (t − a)dt = 1 ,

−
f (t ) (t − a )dt = f (a )

( x − m )2
1 −
The normal distribution f ( x) = e 2 (m=mean and  =standard deviation)
2

2
behaves similar to  (t ) when m=0 and  → 0 (  = 0.01, 0.001, …., for example).

Problem) Plot this using MATLAB and discuss the result.

Impulse and impulse response

Time Response
 (t ) : Impulse h(t ) : Impulse Response
0+
0−
 (t )dt = 1 1
 Linear
t System t
 →0

H( f )
1
f f
Y( f )
Frequency Response H ( f ) =
X( f )
 (t )  h(t )
10 (t )  10h(t )
 (t − 3)  h(t − 3)
10 (t − 3)  10h(t − 3)
For example, consider a RC circuit in the low frequency range and obtain the impulse response h(t ) .
R
i (t )
+ +
in (t ) =  (t ) C out (t ) = h(t )
- -
h(t ) : Impulse response
First-order differential equation using
dh ( t )
i ( t ) =C dh(t )
KVL : Ri (t ) + h(t ) =  (t ) ⎯⎯⎯⎯→
dt
RC + h(t ) =  (t ) (1)
dt
dh(t )
Sol) For t>0, RC + h(t ) = 0
dt
Try h(t ) = Ke st , then we get ( RCs + 1) Ke st = 0 .
1
1 − t
s=−  h(t ) = Ke RC (K is still undetermined since we do not know the input
RC
magnitude.)
To determine K, integrate (1) over the time region 0−  t  0+ .

Left side=
0+  dh(t )  0+
0−  dt 
0+
RC + h(t ) dt = RCh(t ) | − + h(t )dt = RC[h(0+ ) − h(0− )]
0−

0

since h (0− ) =0
⎯⎯⎯⎯⎯ → RCh(0+ )
0+
Right side=  −  (t )dt = 1
0
1
 h(0+ ) =
RC
= h(0) :To be understood as an abruptly charged initial electric charge q(0)
1
1 − RC t
Thus, the impulse response is finally given by h(t ) = e (t  0)
RC

If  (t )(area = 1) → h(t ) , what would be the output for an arbitrary input x (t ) ?

x( ) x( )d  h(t −  )

Area : x( )d

x( )

d

t 
Region
of Sourse
For a small input x( )d (red color) at time  , the output is given by x( )d h(t −  ) (See above).
For the whole input x( )(0    t ) , the output is given by
t
y(t ) =  x( )d h(t −  ) = x(0)d h(t − 0) +  + x( )d h(t −  ) +  + x(t )d h(t − t )
0

= x(0)d h(t ) +  + x( )d h(t −  ) +  + x(t )d h(0) (convolution integral)
using superposition.

Dot(내적) and cross product(외적) of vectors

Vectors : Physical quantities having a magnitude and directn
- Sum, Subtraction, Multiplication (Dot and scalar products)

• Dot (scalar) product

A B = AB cos = ( A cos )B = A(B cos ) (scalar)

B cos

 B
A cos

When  = 0 , A B = AB . When  = 90 , A B = 0 . When  = 180 , A B = − AB .

A B  0 A B = 0 A B  0

Different 0 Similar
Orthogonal

A = a1 ax + a2 ay + a3 az , B = b1 ax + b2 ay + b3 az
A A a a + a2 a y + a3 az B B
Unit vector a = = = 1 x , b= =
| A| A a12 + a22 + a32 |B| B

A B = AB cos( ) = (a1 ax + a2 ay + a3 az ) (b1 ax + b2 a y + b3 ) = a1b1 + a2b2 + a3b3 (scalar)

A A =| A || A | cos(0) =| A |2 = A2 = a12 + a22 + a32 ,
A (− A) =| A || A | cos(180) =| A |2 (−1) = − A2 = −(a12 + a22 + a32 )

Scalar projection of A on b
A b = A 1cos( ) = A cos( ) (projection of A on B or projection of A on b )
 A


Scalar projection of Aon b
= A b = A 1 cos  = A cos 
b
Ab

( )
Vector projection of A on b = A b b = A 1 cos  b = A cos  b

If A = a1 ax + a2 ay + a3 az ,
A ax = a1 , A ay = a2 , A az = a3

Ex) A = ax + 2ay − 3az , B = ax + a y , b = (ax + ay ) / 2

ax + a y 3 3 2
projection of A on B = A b = (ax + 2a y − 3az ) = = (check this!)
2 2 2
projection of A on ax = A ax = 1

projection of A on a y = A ay = 2
projection of A on az = A az = −3

2
ax ax = ax = 1 , ay ay = 1 , az az = 1
ax ay = 0 , ay az = 0 , az ax = 0

• Cross (vector) Product

A  B = an AB sin( ) (vector!) ( an is the unit vector following the right hand rule.)

A  B is maximum when  = 90 .

 A  B = an AB sin 

B
an

A

AB sin( ) :Area
B
B sin( )

A
Rotation rule
x  r

y z  z  
Rectangular Cylindrical Spherical

a x  a y = a z , a y  az = ax , a z  a x = a y , a x  a z = − a y

ax  ax = 0 , ay  ay = 0 , az  az = 0

If A = a1 ax + a2 ay + a3 az and B = b1 ax + b2 ay + b3 az ,
A B
= (a1 ax + a2 ay + a3 az )  (b1 ax + b2 ay + b3 az ) = ax (a2b3 − a3b2 ) + a y (a3b1 − a1b3 ) + az (a1b2 − a2b1 )
 ax ay az 
 
= determinant of a matrix  a1 a2 a3  (Do not use this!! Use the above!!)
 b1 b2 b3 

A B
A unit vector normal to a plane made by A and B is given by an = or
| A B |
A B
− .
| A B |
Prob) Given that A = ax + 2ay + 4az and B = ax + a y , obtain the followings and discuss
the meaning of them with a simple sketch..
a (unit vector of A ), a a , b , b b , A B , B A , A  B , B  A , A A , A (− A) , A a , A b ,
B b ( projection of A on B ), B a ( projection of B on A ), a n , Angle between A and
B

• Meaning of A B  C
A B  C becomes a scalar representing the volume made by A , B , and C .
Thus, A B  C = C  A  B = B  C  A (The rotation rule also applies here!)

Prob) For the case of A = 2ax , B = 3ay ,and C = 5az , show that
A B  C = C  A  B = B  C  A .

Prob) For the case of A = 2ax , B = 3ay ,and C = 4ax + 5az , show that
A B  C = C  A  B = B  C  A .

Field(장) : Scalar field and vector field

Let r be a position vector r = xax + ya y + zaz .

• Scalar field(스칼라장) : gives a number (scalar) for each point r = xax + ya y + zaz
in space.

Examples)

Temperature in a room () , T(x,y,z)

Electric potential (V=J/Coul), V(x), V(x,y), V(x,y,z)
Ground profile , h(x,y)

• Gradient(경도) (generalization of derivatives in space)

: The gradient of a scalar field f gives a vector field and is given by

f f f
Gradient f = ax + a y + az (vector field)
x y z
  
If we define Del Operator  = ax + a y + az , a gradient can be written as
x y z
 f f f
Gradient f = ax + a y + az = f (vector field)
x y z
For a function f ( x ) with one variable, the gradient reduces to
f ( x) f ( x) f ( x) f ( x) df ( x)
f ( x ) = ax + ay + az = ax → ax , like a simple derivative!
x y z x dx
dV ( x)
Let V ( x) (Volts) be the electric potential in the x direction, then V ( x) = ax .
dx
If V ( x, y, z) = x2 (V or C)(scalar field )  V ( x) = 2xax (V / m)(vector field ) (What
does this mean?)

9(V ) 4(V ) 1(V ) 0(V ) 1(V ) 4(V ) 9(V )

dV ( x)
−6(V / m) −4(V / m) −2(V / m) 0(V / m) 2(V / m) 4(V / m) 6(V / m)
= 2 x(V / m)
dx

−3 −2 −1 0 1 2 3 x

Problems)

f ( x, y ) = x 2 + y 2 (V or C or m  height ) (scalar field )

   
 f ( x, y ) =  ax + a y + az  ( x 2 + y 2 )
 x y z 
= 2 xax + 2 ya y (V / m, C / m, m / m ) (vector field )
What does this mean?
 y

V = 4a y (V / m)
(0, 2) V = 2 2 ax + 2 2 a y (V / m)
V ( x, y ) = 4(V ) 2 2 a y (V / m)

V = 1(V ) 1
( 2, 2)
2 2 ax (V / m)
0(V / m)

−2 −1 V = 0(V ) 1 V = 4ax (V / m)
(2, 0) x

−1

−2

V V V
Properties of the gradient V ( x, y, z ) = ax + ay + az :
x y z
- V ( x, y, z ) is a vector field. The vector gives you the direction and maximum
rate of change (V/m or C /m).
- The vector at each position is always perpendicular to the equipotential surface
(from low to high potential).

Problem) V ( x, y, z) = x2 + y 2 + z 2 (Volts)  V ( x, y, z) = 2 xax + 2 ya y + 2 zaz (V / m)

(Plot V and V . What does this mean?)

10 10
Problem) V (r ) = (V or C , very hot at the origin)  V (r ) = − 2 ar (V / m or C / m)
r r
(Plot V (r ) and V (r ) . What does this mean?)

Directional derivative : V  a (Projection of V on a : General derivative of V

in the direction of a )

Problem) When V ( x, y, z ) = x 2 + y 2 + z 2 (Volts ) , obtain V , V  ax , V  a y and discuss

their physical meanings.
• Vector field(벡터장) : gives a vector for each point in space.

Examples) electric field, magnetic field, gravitational field, wind

Any given vector field can be represented using rectangular, cylindrical, and spherical
coordinate systems. The choice is a matter of convenience for a given problem.
Divergence and Curl are two important properties of a vector field.

An example of vector field

Representation of vector fields

Examples)
Wind in the sky (m / sec) ,
Electric field (See below), magnetic field, gravitational field, and so on

- using rectangular coordinate system

 A(r ) = A( x, y, z )
az = Ax ax + Ay a y + Az az
A
z (linear combination of orthonomal basis vectors )
ax ( x, y, z )a y ax , a y , az : same at each position
x
r = xax + ya y + zaz
y
y z
ax  a y = az
x
  
= ax + a y + az
x y z

E ( r ) = E ( x , y , z ) = E x ( x, y , z ) a x + E y ( x, y , z ) a y + E z ( x, y , z ) a z
: Ex ( x, y, z ), E y ( x, y, z ), and Ez ( x, y, z ) each is a scalar field.
(convenient for the problems with surface charges)

Example) Draw the following fields.

V ( x) = 3 , V ( x, y, z ) = 3 , V ( x, y, z ) = 3 y , A( x, y, z) = 3ax , A( x, y, z) = xax ,
A( x, y, z) = 2 xay , A( x, y, z) = 3xaz , A( x, y, z) = 3xyzax + 2 x2 yay − z 2 az ,

- using cylindrical coordinate system

 a = az  a A(r ) = A(  ,  , z ) : 0    , 0    2
 az A = A a + A a + Az az
a
z (linear combination of orthonomal basis vectors )
(  , , z )
a
r =  a + zaz az :same at each position,
z a , a :different at different positions
y 
a
 sin 
cos  
 z
x a = cos  ax + sin  a y a  a = az
  
= a + a + az
  z

E(r ) = E(  ,  , z) = E (  ,  , z)a + E (  ,  , z)a + Ez (  ,  , z)az (convenient for the

problems with line charge on the z axis)

Example) Draw the following fields.

10 20 100
T (  ,  , z ) = (C ) , E(  ,  , z) = 5az , E (  ,  , z ) = a , H (  ,  , z ) = a
  2

- using spherical coordinate system

 a (?) = ar  a A(r ) = A(r ,  ,  ): 0  r  , 0    
= Ar ar + A a + A a
z sin 
ar A (linear combination of orthonomal basis vectors )
cos  
( r , ,  )
r a
ar , a , a

:different at different positions
y r
 sin  sin 
sin  cos   sin 
x  
ar = sin  cos  ax + sin  sin  a y + cos  az ar  a = a
  
= ar + a + a
r r  r sin 

E(r, t ) = E(r, ,  , t ) = Er (r, ,  , t )ar + E (r, ,  , t )a + E (r, ,  , t )a : convenient for
the problems with charges at the origin)

Problems) Draw the following fields.

100 10 10 10
V (r , ,  ) = , E (r , ,  ) = ar , E (r , ,  ) = 2 ar , H (r ,  ,  ) = 2 a
r r r r
*** You will get a rough idea if you try to plot this only for several positions!!

Choice of coordinate systems (Very important!!!)

See the following vector representations! How can we find B1 and B2 ? They are just the
projections of A on b1 and b2 , respectively.
 b2 a1 , a2 : orthonormal basis vectors
b1 , b2 : orthonormal basis vectors
a2
45
60 5
a1 b1 A = 1a1 + 2a2 = B1 b1 + B2 b2

Therefore,
3
B1 = A  b1 = (1a1 + 2a2 )  b1 = a1  b1 + 2a2  b1 = cos 45 + 2 cos 45 = 2.12
2
1
B2 = A  b2 = (1a1 + 2a2 )  b2 = a1  b2 + 2a2  b2 = cos135 + 2 cos 45 = − sin 45 + 2 cos 45 = 0.71
2
A = 1a1 + 2a2 = B1b1 + B2 b2 = 2.12b1 + 0.71b2 (Two choices do must not change A )

The choice of the coordinate system is just the matter of convenience! The given vector
field must not change.!
Example of transforming a vector field in rectangular to cylindrical coordinate system :
Do the similar thing as above!

A(r ) = A( x, y, z )
= Ax ( x, y, z )ax + Ay ( x, y, z )a y + Az ( x, y, z )az

Want to express this vector field in a cylindrical coordinate system:

a ay

90 − 
a

 ax


x

- A(r ) = Ax ax + Ay ay + Az az  A a + A a + Az az
- Use the concept of projections!
( )
A = A  a = Ax ax + Ay a y + Az az  a = Ax ax  a + Ay a y  a + Az az  a
= Ax 11 cos  + Ay 11 cos ( 90 −  ) = Ax cos  + Ay sin 
( )
A = A  a = Ax ax + Ay a y + Az az  a = Ax ax  a + Ay a y  a + Az az  a
= Ax 11 cos( + 90) + Ay 11 cos ( ) = Ax ( − sin  ) + Ay cos 
 ( )
Az = A  az = Ax ax + Ay a y + Az az  az = Ax ax  az + Ay a y  az + Az az  az = Az
Above results are summarized:
 A   cos  sin  0   Ax 
 A  =  − sin  cos  0   A 
    y
 Az   0 0 1   Az 
After inserting x =  cos  , y =  sin  , and z = z in
Ax ( x, y, z), Ay ( x, y, z),and Az ( x, y, z) , you finally obtain
A(r ) = A(  ,  , z )
.
= A (  ,  , z )a + A (  ,  , z )a + Az (  ,  , z)az

Example 2.1) Given a point P(-2,6,3) and a vector field A = yax + ( x + z )a y , express P
and A(,, z) in cylindrical coordinates.

z A(1, 0, 0) = 0ax + (1 + 0)a y = a y

A(0, 2, 0) = 2ax + (0 + 0) a y = 2ax
A( x, y, z ) = yax + ( x + z )a y (−2, 6,3) A(0, 0,1) = 0ax + (0 + 1)a y = a y

A(−2, 6,3) = 6ax + (−2 + 3)a y = 6ax + a y

−2
(0, 0,1)

(1, 0, 0) (0, 2, 0) 6 y
ax
x

Let’s examine the top view.

a
a
Top view (−2, 6,3)
ay
 = (−2)2 + 62 = 40 = 6.32 ax 

z 2
tan −1   18.43
6
y
 = 108.43

a

a A(−2, 6,3) = 6ax + 1a y = A a + A a

x

-Cylinder
 = (−2)2 + 62 = 40 = 6.32
 y  6 
 = tan −1   = tan −1   = 108.43 , z = 3
 x  −2 
-Spherical
r = (−2)2 + 62 + 32 = 49 = 7
  6.32 
 = tan −1   = tan −1   = 64.62 , z = 3
 z 3 
 = 108.43

P(-2,6,3)(rectangular)=P(6.32, 108.43 ,3)(cylindrical)=P(7, 64.62 ,

108.43 )(spherical).

A(−2, 6,3) = 6ax + 1a y = A a + A a

 A = A  a = 6ax  a + 1a y  a = 6cos(108.43) + 1cos(18.43) = −0.948
A = A  a = 6ax  a + 1a y  a = 6cos(161.57) + 1cos(108.43) = −6.006
 A ( 6.32,108.43,3) = −0.948a − 6.008a (cylindrical)

A(sphirical) can also be obtained similarly :

A ( 7, 64.62,108.43) = −0.857ar − 0.4066a − 6.008a (spherical)

Vector fields should be the same independent of the coordinate systems : A = 6.083

• Vector calculus

l (m)
Radian angle  = or l = r  (m)
r (m)

 l ( m)
Radian angle  = (no unit !)
r ( m)

r
l

Some examples of position vector :

r = xax + ya y + za z (rectangular), r =  a + zaz (cylinder), r = rar (spherical coordinate)
 y
dxdy
dy
dx
dx x x
ds (m 2 )
V
l
dl (m)
S dv(m3 )

Rectangular

Cylindrical

Spherical
Differential displacement vector (미소 길이 벡터) (m) (See the Sadiku text for more
details!)

  
dl = dxax + dyay + dzaz (rectangular),  = ax + a y + az
x y z
  
dl = d  a +  d a + dzaz (cylindrical) ,  = a + a + az
  z
Take dl =  d and obtain the circumference l (원주) with a radius 1
2 2 2
l= dl = 1d = 1  d = 21 (m) .
 =0  =0  =0

dl = drar + rd a + r sin  d a (m) (spherical coordinate),

  
= ar + a + a
r r  r sin 

Differential surface vector (미소 면적 벡터) (m2)

ds = ax dydz , a y dzdx , az dxdy (rectangular),
ds = a (  d )dz , a dzd  , az d  (  d ) (cylindrical) ➔ Take ds = d  (  d ) and check
the area of a circle with a radius 1
1 2 1 2 1 2
S =  ds =  d  (  d ) =  d   d = 12 (m2 )
 =0  =0  =0  =0  =0  =0

ds = ar (rd )(r sin  d ) , a (r sin  d )(dr ) , a dr (rd ) (spherical coordinate) ➔

Take ds = (rd )(r sin  d ) and check the surface area of a sphere with a radius r
 2  2  2
d = 2 r 2  − cos 0 =4 r 2 (m2 )

S =  ds =  (rd )(r sin  d ) =r 2  sin  d 
 =0 =0  =0 =0  =0  =0

Differential volume (미소 체적) (m3)

dv = dxdydz , d  (  d )dz ( =  d  d dz ), dr (rd )(r sin  d ) ( = r 2 sin  drd d ) or
4 r 2 dr for some special cases.
Prob) Take dv = r 2 sin  drd d and obtain the volume of a sphere with a radius r1
r1  2 r1  2 r1  2
Volume =    dv =    r 2 sin  drd d =  r 2 dr  sin  d  d
r =0 =0 =0 r =0 =0 =0 r =0  =0  =0

r 2 dr  − cos  0  0 = 4 
r1  2 r1 r1
= r 2 dr =  4 r 2 dr
r =0 r =0 r =0

4 3 3
= r1 (m )
3

Need to clearly understand

 l
A dl (circulation around a closed path l , related with curl of A to be defined
later)

 S
A ds (net outward flux of A crossing a closed surface S, related with divergence

of A to be defined later)

 V
v dv (volume integral of the scalar field  v over a volume V)

b
Similar to  a
f ( x ) dx !

dx
A A
V
ds (m 2 ) 2
ds (m ) dv
dl (m) dl (m)
A l S
A
dl (m) A

Ex) A magnetic flux density is given by

B( x, y, z) = 2( x + y + z )ax + 3( x + y)ay + 3xyaz (Wb / m ) . Determine the
2 2 2 2
total
magnetic flux  flowing upward out of the rectangular surface.
 z

2m
y

2m

Sol) B( x, y, z = 0) = 2( x2 + y 2 )ax + 3( x + y)ay + 3xyaz (Wb / m2 ) , ds = dxdyaz ,

2 2
2 2 2 2  x2   y 2 
Magnetic flux  =  B ( x, y, z = 0) ds =   3xydxdy =3 xdx  ydy =3     = 12(Wb)
S 0 0 0 0
 2 0  2 0

Prob) B( x, y, z) is the same as above. When the surface is given by

y=0( 0  x  2 , 0  z  2 ), determine the total magnetic flux(Wb) flowing out of the
surface in the –y direction.

• Divergence 발산) and Curl(회전)

Divergence and Curl are two important properties of a vector field.

• Divergence of a vector field (발산)

Divergence of a vector field F ( flux number / m2 ) or F (#/ m2 ) is defined as

Divergence F = lim
 S
F ds (#)
(Scalar field)
v → 0 v(m3 )
 F :typically given by unit of flux number / m2
z
ds (m 2 )
F F F
F F
F 2
z ds (m )
2
F F ds (m ) S ds (m 2 )
F F
v(m ) 3
F F
ds (m 2 ) x
F F
y
ds (m 2 )
F
( x0 , y0 , z0 )
F y

x
where F  ds : number of flux flowing out of a small area ds ,

 S
F  ds : total number of flux flowing out of a closed surface S,
v : small volume in the surface S

Let the electric flux density(전속밀도) be given by

D( x, y, z) = Dx ( x, y, z)ax + Dy ( x, y, z )a y + Dz ( x, y, z )az ( Coul / m2 ) (Vector
field), then

Divergence D = lim
 S
D ds (Coul )
(Scalar field): Consider the following top view of a
v → 0 v ( m 3 )
cube figure where ds = yzax (for the front surface) and ds =− yzax (for the back
surface).
 y

x
Dx ( x0 − , y0 , z 0 )
x 2
x0 −
2 3 ( x0 , y0 , z0 )
x
x 5
x0 + x
2 Dx ( x0 + , y0 , z 0 )
x 2

Divergence D considering only the front and back surfaces = lim

(
D  yzax + D  −yzax )
v →0 xyz (m ) 3

x x
Dx ( x0 + , y0 , z0 )ax  yzax + Dx ( x0 − , y0 , z0 )ax  (−yzax )
= lim 2 2
v →0 xyz (m ) 3

x x
Dx ( x0 + , y0 , z0 ) − Dx ( x0 − , y0 , z0 )
2 2 D
= lim = x (Coul / m3 ) ( Must understand what
x →0 x x
this means!)
Dx 5 − 3
In the above example for example,  >0. This means the flux in x direction is
x x
coming out of the volume. You need to think of many other situations and get the
meaning of the divergence!

Thus, the divergence of D consideringallof sixsurfaces

발산 D = Div D = lim
 S
D ds (Coul )
=
Dx Dy Dz
+ + (scalar field)
v → 0 v(m ) 3
x y z

  
This result can be simply expressed if we use the Del operator  = ax + a y + az .
x y z
Dx Dy Dz   
+ + = ( ax + a y + az ) ( Dx ax + Dy a y + Dz az ) =  D (Coul / m3 )
x y z x y z

Ax Ay Az

Div A =   A = + + (rectangular)
x y z
1  1 A Az
Div A =   A = (  A ) + + (cylindrical)
    z
1  1  1 A
Div A =  A = 2 (r 2 Ar ) + (sin  A ) + (spherical)
r r r sin   r sin  
Problems)

1) Obtain Div A when A( x, y, z) =10 ax ,

Obtain Div A when A( x, y, z) =10 ax , 5 ay , 3x ax , − 3x ax , 3 y ax ,3 y ay
Ax Ay Az Ax 10 − 10
A( x, y, z ) =10 ax  Div A =   A = + + = = =0
x y z x x

10
2) Obtain Div A when A = 10 a and A = a .
 
1  1 A Az 1  1  10
Div A =   A = (  A ) + + = (  A ) = (10  ) =
    z     
1  1  10 1 
Div A =   A = (  A ) = ( ) = (10) = 0
      

10 10
3) Obtain Div A when A =10 ar , ar , and 2 ar .
r r

Divergence theorem (발산정리)

 (  A) dv = 
V S
A  ds (Why?)

dv(m3 )

  A (#/ m3 ) (  A) dv(#)
closed surface S S
A A
A dv
A
A dv
dv A
dv V V
dv
S

Consider only the above three dv’s for simplicity.

 (   A) dv = 6 + (−1) + (−2) = 3(#) = 
V S
A  ds (understand the divergence
theorem?)

Electric Gauss law

- volume form or integral form

D (Coul / m 2 ) : free charge density or electric flux density

 S
D  ds = Q(Coul ) (This holds for any closed surface in space.)

D
ds D
S
D
Q(Coul )
V

D(Coul / m 2 )
D(Coul / m 2 )
:electric flux density

- point form or differential form from the integral form (Make the closed surface S very
small and divide by dv )

 S
D  ds
(definition of divergence!) =
Q(Coul )
dv dv(m3 )
So,  D =  (Coul / m ) (point form or differential form) This is true for any point
3

in space (This is true in the entire space!)

- Volume form from point-form

Do V
  D =   dv operation.
 

Left =  (  D ) dv ⎯⎯⎯⎯⎯⎯→  D  ds (Coul )

Divergence theorem
V S
Right =   dv = Q(Coul )
V

  S
D  ds(Coul ) = Q(Coul ) (volume form)

Some examples)

Free point charge q (Coul ) at the origin

Dr (Coul / m 2 )

Gauss surface
Dr (Coul / m 2 ) r
q(Coul )
S = 4 r 2

 S
D  ds = Dr  4 r 2 = q (Coul )
q
 Dr = (Coul / m 2 )
4 r 2

in free space. Thus, Dr =  0 Er in free space.

q
Remember that Er =
4 0 r 2
q q
Or Er = = in general.
4 r 2
4 0 r r 2

Line charge with l (Coul / m) on the entire z axis

l (Coul / m)
2
 

D (Coul / m 2 )

L D (Coul / m 2 )
Gauss surface
 S
D  ds = D  2 L = l L(Coul )
l
 D = (Coul / m2 )
2

If the line charge has a finite length as seen below, the above solution cannot be used
since D will look like. The Gauss law always holds true. So,  S
D  ds = Q(Coul ) .
But we need to find the field at specific positions using the superposition!

+
+
+
+
+
+

Surface charge with s (Coul / m2 ) on the xy plane in free space

D(Coul / m 2 )
S (m 2 ) S (m 2 )
ρs (Coul/m 2 ) on the
entire xy plane
Gauss surface
D(Coul / m 2 )

 S
D  ds = 2  DS =  s S (Coul )
s (This is true for both positive and negative  s .)
D= (Coul / m 2 )
2
Surface charge with s (Coul / m2 ) on the PEC (Perfect Electric Conductor) plane

D =  s (Coul / m 2 )
S (m 2 )

 s (Coul / m2 ) on PEC plane

Gauss surface PEC ( E = 0in PEC)
 S
D  ds = DS =  s S (Coul )
 D =  s (Coul / m2 )

Surface charge with s (Coul / m2 ) and −s (Coul / m2 ) in free space (similar to
above!)

−  s (Coul / m 2 )

D =  s (Coul / m 2 )

 s (Coul / m 2 )

Note) The electric flux lines exist only inside the region. They cancel out outside the
region.

• Curl of a vector field (회전)

The curl of a vector field H is defined as

Curl H = an lim l
 H dl ( A) (Vector field)
s → 0 s (m 2 )
(The direction of dl and an must follow the right hand rule! )

z H : given by a unit of A / m

H H H
H H
an dl H

H H
s (m 2 ) H dl
H H H dl H H
dl (m)
H H dl
H
H
H
H
y
dl = −ya y

s = yz zaz
−zaz ( x0 , y0 , z0 ) ( x0 , y0 + y , z0 )
2

ya y
ax
x
Let a vector field H ( x, y, z) = H x ( x, y, z)ax + H y ( x, y, z)a y + H z ( x, y, z)az (A/m) be
given in a specific space.

Curl H in x direction = ax lim

H
l
dl ( A)
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
dl chosen as seen above (right hand rule)
→
s → 0 yz (m 2 )
 y y 
 H z ( x0 , y0 + 2 , z0 ) az zaz + H z ( x0 , y0 − 2 , z0 ) az ( −zaz ) 
 + 
yz
= ax lim  
s → 0  z z 
 H y ( x0 , y0 , z0 + )a y (−ya y ) + H y ( x0 , y0 , z0 − )a y (ya y ) 
+ 2 2 
 yz 
 y y z z 
 H z ( x0 , y0 + , z0 ) − H z ( x0 , y0 − , z0 ) H y ( x0 , y0 , z0 + ) − H y ( x0 , y0 , z0 − ) 
= ax  lim 2 2 − lim 2 2


y →0  y z →0 z 
 

H z H y
= ax ( − ) ( A / m2 ) ( Must understand what this means!)
y z

Then, the curl of

H H y H H z H H x
H = ax ( z − ) + ay ( x − ) + az ( y − ) ( A / m2 ) (considering the
y z z x x y
rotation rule!)
  
= ( ax + a y + az )  ( H x ax + H y a y + H z az ) =   H ( A / m 2 ) (Vector field)
x y z

Note that  ( H )  0.

 H H y  H H z  H y H x
  (  H ) = ( z − )+ ( x − )+ ( − )=0
x y z y z x z x y

Also,   (V )  0.
 V V V      V V V 
  (V ) =    ax + ay + az  = ( ax + a y + az )   ax + ay + az 
 x y z  x y z  x y z 
  2V  2V    2V  2V    2V  2V 
= ax  −  + a y  − +
 z
a − =0
 yz zy   zx xz   xy yx 

Stokes’ theorem (Curl theorem, 회전정리)

 (   A)  ds = 
S l
A  dl (Why?)
Why ?

  A (circulation / m2 ) ds
 A

ds
A A
  A  ds
(circulation around ds )

dl

Advds
A
ds
dv V
A dsdv Open surface S

ds and dl follow the right − hand rule.

Closed path l
dl
Consider only the above three ds ’s for simplicity.

 (   A)  ds = 6 + (−1) + (−2) = 3(circulation) = 

S l
A  dl (understand this?)

Ampere’s law (for static or time-invariant cases)

 H = J ( A / m2 ) for any point in space (point form or integral form).

(  H )  ds = J  ds( A) (Very important!)

Using the superposition principle,
S
( )
   H  ds ⎯⎯⎯⎯⎯→  H  dl =  J  ds = I ( A)
Stokes ' theorem
l S

 H  dl = I ( A) for any closed path l in space (volume form or integral form of

l
Ampere’s Law).
 H
dl
H
H
H H
H dl I ( A)
H

dl
H

Some examples using Ampere’s law) The magnetic fields due to the arbitrary flowing
currents (line, surface, volume currents) can always be obtained using the superposition
principle. But, for the following ideal problems, the magnetic fields can be obtained
more easily using the Ampere’s law.

For a line current I on the entire z axis :

Line current I ( A)
on the entire z axis
H

H
H
H

I
 H  dl = H (2 ) = I ( A)  H = 2 ( A / m)
l

For a surface current J s ( A / m) on the entire infinite xy plane:

H l dl (m) H
J s ( A / m) Surface current J s ( A / m)
on theinfinite xy plane

dl (m) H
L ( m)
Js
 H  dl = 2HL = J L ( A)  H =
l s
2
( A / m) (same magnetic field for all points in space,
independent of the distance from the current source)
For a surface current J s ( A / m) on the infinite PEC plane in (z<0): In PEC,
E = 0, H = 0 .
 H  dl = HL = J L( A)  H = J ( A / m) (same magnetic field for all points in upper
l s s

space, independent of the distance from the current source)

l dl H Air
J s ( A / m) Surface current
on the infinite PEC plane
PEC
PEC H = 0 dl H =0 PEC

L ( m)

For a pair surface current with J s ( A / m) and − J s ( A / m) in free space: Magnetic

fields due to J s and − J s cancel outside and add inside.
− J s ( A / m)
H = J s ( A / m)
J s ( A / m)
Note) The magnetic fields are only inside the region. They cancel out outside the region.
Using the similar Ampere’s law as the above problem, we obtain the same answer.
=➔ H = J s ( A / m) .

Problems)
Plot the given vector field and obtain its divergence and curl.

A( x, y, z) = ax + 2ay
  A = 0(/ m3 )
  A = 0(/ m2 )

A( x, y, z) = 2xax
  A = 2(/ m3 )
  A = 0(/ m2 )
A( x, y, z) = 10x2 ax
  A = 20x(/ m3 )
  A = 0(/ m2 )

A( x, y, z) = 3xay
  A = 0(/ m3 )
  A = 3az (/ m2 )

Problems)
Given that V ( x, y, z ) = 2 xyz + ( x 3 + y 2 + z ) + 5 (V) (scalar field),
100
V (r ) = (Volts) (V) (scalar field),
r
D( x, y, z) = 2 xyzax + 3( x + y)ay + 4 zaz (Coul/ m2 ) (vector field), and
H ( x, y, z) = xyzax + 3( x + y)ay + 5az (A/m) (vector field), obtain the followings and say
what are their meanings.
V |x=1, y=0, z =0 , V |r =5 ,  D |x=0, y =1, z =0 ,  H |x=0, y =0, z =1
Solution )
V ( x, y, z) = (2 yz + 3x2 )ax + (2 xz + 2 y)ay + (2 xy + 1)az (V / m) (vector field),
 V ( x, y, z) |x=1, y =0, z =0 = 3ax + az (meaning: 이 위치에서는 3ax + az 방향으로 갈 때
전위의 최대 변화가 일어나며 그 변화율은 10 V/m 이다.)
Directional derivatives ( V  a ) at this position: V  ax = 3(V / m) , V  az = 1(V / m) ,
3ax + az
V  = 10(V / m) , V  ay = 0(V / m)
10

100
V (r ) = − ar (V / m) and V |r =5 = −4ar (V / m) .
r2
Directional derivatives ( V  a ) at this position: V  ar = −4(V / m) ,
V  (−ar ) = 4(V / m) , V  a = 0(V / m) , V  a = 0(V / m) .

 D( x, y, z) = 2 yz + 3 + 4(Coul / m3 )
  D( x, y, z) |x=0, y =1, z =0 = 7(Coul / m3 ) ( 의미: 이 위치에서는 7 Coul / m3 의 발산이
존재한다)
If dv = 0.1m3 ,  Ddv = 7(Coul / m3 )  (0.1m3 ) = 0.7Coul .
 H z H y H H z H H x
 H ( x, y, z ) = ax ( − ) + ay ( x − ) + az ( y − ) = a y ( xy) + az (3 − xz)( A / m2 )
y z z x x y
  H ( x, y, z) |x=0, y =0, z =1 = 3az ( A / m2 ) (의미: 이 위치에서는 az 방향으로 엄지 손가락을 향할 때
2
나머지 손가락 방향으로 3 A / m 의 회전율이 존재한다.)
az + ax 2
If ds = 0.1az (m2 ),0.1ax (m2 ),0.1a y (m2 ), 0.1 (m ) , then
2
 H  ds = 3az ( A / m2 )  (0.1az )m2 = 0.3( A) ,  H  ds = 3az ( A / m2 )  (0.1ax )m2 = 0( A)
az + ax 2 0.3
 H  ds = 3az ( A / m2 )  (0.1ay )m2 = 0( A) ,  H  ds = 3az ( A / m2 )  (0.1) (m ) = ( A)
2 2

Problems) Given that V ( x, y, z ) = x 2 y + xyz (Volts) ,

D( x, y, z) = 2 xyzax + 3cos( x + y)ay + 4e−2 z az (Coul/ m2 ) and
H ( x, y, z) = 2 xyzax + 3cos( x + y)ay + 4e−2 z az (A/m), obtain the followings and get the
meaning of them.

V ,  D ,   H

Prob) Determine the divergence and curl of the following vector fields.
P( x, y, z) = x2 yzax + xzaz
Q(  ,  , z) =  sin  a +  2 za + z cos  az (see the text for the curl and divergence
written with a cylindrical coordinate system!)

4 kinds of forces in nature

Gravitational force, Electric force, Magnetic force, Nuclear force (Newton or N)

m/s
( 1N = 1Kg  = 1Kg  m / s 2 ) ( Remember F = ma )
s

Gravitational force (between the two masses) and gravitational field (made by a
mass)

m1 m2
Gravitational force between m1 and m2 , F = G aR ( N )
R2
F F
m2 ( Kg )
m1 ( Kg )
F
R g=
( N / Kg )
m2
F m
Gravitational field (force per unit mass) made by m1 , g = = G 12 aR ( N / Kg )
m2 R
For example, g =9.8 aR (N/Kg or m/sec2) on the earth surface ( aR : unit vector pointing
into the center of the earth) (Homework : Check this by googling!)

g aR (unit vector)

M ( Kg ) g
g g g ( N / Kg )

R
F = mg ( N )
g m

g
M
Gravitational field (force per unit mass) made by M , g = G aR ( N / Kg )
R2

Electric force (between the two charges) and electric field (made by a charge)

1 q1q2
Electric force between q1 and q2 , F =  aR ( N )
4 R 2

F F
q2 (Coul )
q1 (Coul )
F
E=
R q2
F 1 q1
Electric field (force per unit charge) made by q1 , E = =  aR ( N / Coul )
q2 4 R 2
 E

E
aR (unit vector)
Q(Coul )
E E E E ( N / Coul )

q
E F = qE ( N )

E
1 Q
Electric field (force per unit charge) made by Q, E =  aR ( N / Coul )
4 R 2

Magnetic force (between the two current elements) and magnetic field (made by a
current element))

Based on observations and experiments, we found that

I1 dl1 R I 2 dl2 I1 dl1 R I 2 dl2 I1 dl1 R

F (attractive) F (repulsive) F
1 I 2 dl2
1 1 R2
 2  2
R R

I1 dl1 R I1 dl1 R
I 2 dl2 I 2 dl2
F=0 F=0
   I dl  a 
Magneticforcebetween I1 dl1 and I2 dl2 , F = I2 dl2   1 1 2 R  = I2 dl2  B (N )
 4 R 
(obtained after many expriments)
F
F
aR (unit vector)
I1 dl1 ( A  m) I 2 dl2 ( A  m)
R B
R
Magneticflux density (force per unit current element) made by I1 dl1 ,
B :magneticflux density  I1 dl1  aR  N Wb 
B = H = = ( see below!)
H :magneticfield 4 R 2  A  m m 2 

d
Recalling that Wb / sec = Volt = J / Coul = ( N  m) / Coul ( = v ) (Faraday’s law), we
dt
N J / m J / A (Coul V ) / (Coul / s) V  s Wb
see that = = 2 = = 2 = 2.
A m A m m m2 m m


aR (unit vector) idl2
 
I dl1 ( A  m)  F = idl2  B

B B B B
R




 I dl1  aR
Magnetic flux density (force per unit current element) made by I dl1 , B =  H =  N / ( A  m) 
4 R 2
H :Magnetic field
0 I dl1  aR
Magnetic flux density B =  H =   N / ( A  m) or Wb / m2 
l 4 R2 

Note again that

F ( gravitational force ) = mg (N),
F ( electric force ) = qE (N),
F ( magnetic force ) = I dl  B ( N ) with the same unit of N.

Superposition principle :  ,

To obtain the field at a specific position due to a line, surface, or volume charges, we
need to add all of the contributions from all sources using the superposition principle.

An example for a line charge with line charge density  l (Coul/m) on l :

line charge density

+ ()
+
l (r ) (Coul / m) E r
 (r )dl  +
+ R = r − r
V (r )
++
l

r
+
+ r
+
( )
l r  dl aR
()
E r =
l 4 0 R 2
= −V r ( ) (Vector field )
( )
l r  dl 
()
V r =
l 4 0 R
( Scalar field )

r  : Source position vector

r : Field position vector
R = r − r  : Distance vector from a source to a field position (vector)
R = R = r − r  : Distance from a source to a field position (scalar)

R r − r
aR = =
R r − r  : unit vector in the direction of R
 0 = permittivity of free space = 8.854 10−12 ( F / m)

Frequently, you will see the following expression:

aR R r − r
= =
R 2 R3 r − r  3

Electric potential (scalar field) generated by a point charge q at the origin

: work to be done to move a 1 Coul of charge from infinity to a specific position at r
 

( ) q q  1 q  N 
r   J
V (r ) =  − E  dl =  E  dl =  ar  drar =  −  =  m = =V 
 r r 4 0 r 2
4 0  r  r 4 0 r  Coul Coul 
     V (r ) q
V (r ) =  ar + a + a  V (r ) = ar = − ar (V / m or N / Coul )
 r r  r sin   r 4 0 r 2
q
 E = −V = ar (V / m or N / Coul )
4 0 r 2

Electric potential at r due to a point charge q at r

q q  J 
V (r ) = =  =V 
4 0 R 4 0 r − r   Coul 

Electric potential and field at r due to a very small charge dq at r 

dq dq aR
V (r ) = = (V ) , E (r ) = ( N / Coul or V / m )
4 0 R 4 0 R 2
V J N m N
(Note that = = = .)
m m  Coul m  Coul Coul

Kinds of a small charge source dq :

l dl  (line charge, Coul),
s ds (surface charge, Coul),
 dv (volume charge, Coul).
For any kind of charge given in the region l(line), S(surface), and V(volume),
dqaR
()
E r =
l , S , V 4 R 2
(Vector field )
0

dq
()
V r =
l, S , V 4 0 R
() ()
( Scalar field )  E r = −V r (easier in many cases)
where dq may be replaced by l dl  , s ds , or  dv (Coul), depending on the kinds
of charges.

Kinds of current sources : I dl (line current, A  m ), J s ds (surface current, A  m ),

Jdv (volume current, A m )
 0 I dl1  aR
B = H =   N / ( A  m) or Wb / m2 
l 4 R2 

Electric potential V and electric field E at r due to electric charge q at the origin

Electric potential V at r (generated by a point charge q at the origin) : defined as the

energy spent to move one Coul of charge from infinity to a position at r.
( )
V (r ) =  − E  dl (Energy you have spent)
r



=  E  dl (Work the field has done)
r

 q q  1 q  N J 
= a  drar =  −  =  m = =V 
r 4 0 r 2 r
4 0  r  r 4 0 r  Coul Coul 

The electric potential difference to move 1 Coul of electric charge from A to B :

q 1 1
VAB = VB − VA =  − 
4 0  rB rA 

Equi − potential surfaces

V = 10(V )
V = J / Coul
V = 20(V )

Work = 10( J )

Work = 20( J )

E ( N / Coul )

V = 4(V )

V = 0(V ) at 
Work = 4( J )
+1(Coul )
 E

q
V= (V ) ( scalar field )
4 r

q(Coul )
E q
E= ar ( N / Coul = V / m)
r 4 r 2
(vector field )

Equi-potentialsurfaces

Ex) Electrostatic Example using superposition principle

Obtain the electric field E (  ,  , z ) , more specifically E (  , 1 ,  2 ) .

Sol) Source position vector : r  = z az (variable between A and B),

 a + zaz (maybe assumed to be a fixed),
Field position vector : r =

Distance vector: R = r − r  =  a + ( z − z )a z ,

1/ 2
Distance: R = R =   + ( z − z ) 
2 2
,
and dl  = dz . Then,
dqaR ( )
l r  dl aR  dz R  dz    a + ( z − z )az 
()
E r =  = l 
B
= = l
l 4 0 R 2 l 4 0 R 2 l 4 R
0
3
4 0 z = A R3
(Integration with respect to z from A to B)

Use “change of variable” for integration with respect to  from1 to 2 .

 
Observing cos  =  R =  / cos  =  sec , z − z =  tan  , and
R
dz  = −  sec2  d (  is now a new variable for the integration)
( z  = A   = 1 , z  = B  =  2 )

l dz    a + ( z − z )az 
()
B
E r =
4 0  z = A R3
(integration with respect to z)

 2 −  sec 2  d   a +  tan  az 
= l
4 0  
= 1  3 sec3 
(integration with respect to  )

l  −d  a + tan  az  l 

4 0   = 4 0   =
= = − d cos   a + tan  az 
2 2

sec 
1 1

l  l
=  −d cos  a + sin  az  = (sin 1 − sin  2 )a + (cos  2 − cos 1 )az 
2

4 0   = 1 4 0   
(Generalsolution for various cases for any ρ, α1 , and α 2 )
if 1 = 2 (no line charge)
⎯⎯⎯⎯⎯⎯⎯ →0
l
if 1 = 90 ,  2 = 0 (half infinite line charge length )
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯ →  a + a z 
4 0   

if  = 90 ,  =−90 (infinite line charge length ) l l

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→
1 2
 2 a = a ( Easily obtained if using Gauss'law)
4 0  2 0 

Ex) Magnetostatic example using superposition principle

Obtain the magnetic field H (  ,  , z ) , more specifically H (  , 1 ,  2 ) .
 I dl   aR I dl   R
H = = ( A / m), B =  H (Wb / m2 )
l 4 R 2 l 4 R3

Sol) Source position vector : r  = z az , Field position vector : r =  a ,

1/ 2
Distance vector: R = r − r  =  a − z az , Distance: R = R =   + ( z ) 
2 2
,

and I dl  = Idzaz . Then,

H =
I dl   R B Idzaz   a − zaz (B I  dza )
l 4 R 3
= z= A 4 R 3
= z= A 4 R3 (integration with respect to z)

Use change of variable : tan  =  z  =  cot  (when z  = A,  = 1 ; when
z
z  = B,  =  2 ),
then dz  = −  cosec  d and R =  cosec .
2

I  dz   2 I  ( −  cosec  ) d
2
I 2
( − sin  ) d
B
H = a    =
4  =1
= a = a
z = A 4 R 3 1 4 3cosec3
(integration with respect to  )
I
= a ( cos  2 − cos 1 ) (Generalsolution for various cases with different ρ, α1 , and α2 )
4
1 = 2 (nosourcecurrent )
⎯⎯⎯⎯⎯⎯⎯ →0
1 =90,  2 = 0 (half infinitecurrent length) I
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯ → a
4
1 =180,  2 = 0 (infinite current length) I
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ a ( A / m)
2

Example) -Text Example 7.3 : Obtain H at P(0,0,h) and plot it as a function of h.

z
dH z
P(0, 0, h)
r R = r − r ' = ha − aa H or B (= 0 H )
z 
a
y
ar ' a
I
I dl = Iad a
I
x magnetic dipole moment m = I  a 2 az ( A  m 2 )
r = zaz , r ' = aa
Sol) R = r − r ' = haz − aa

dH =
I dl  R Iad a  haz − aa
= =
(
Ia hd a + ad az ) ( )
4 R3 3 3
4 ( a + h )
2 2 2
4 ( a + h ) R3
2 2 2

 = 2  = 2 (
Ia hd a + ad az ) ⎯⎯⎯⎯⎯
 =2 
 d  a = 0
H=  dH =  dH =  →
 =0

 =0  =0 3
4 ( a + h )
l
2 2 2

Ia 2 2 az Ia 2 az
3
= 3
4 ( a + h
2 2 2
) 2 ( a 2 + h2 ) 2

if h = 0( at loop center ) Ia 2 az I az I a
⎯⎯⎯⎯⎯⎯→ = ( A / m )(loop current )  (line current )
2a 3 2a 2 a

Electric dipole

z P(r , ,  )

r1
 r2
Q r
d y
−Q  d cos 

x
Q (−Q) Q 1 1  Q  r2 − r1  Q d cos 
V= + =  − =  
4 0 r1 4 0 r2 4 0  r1 r2  4 0  r1r2  4 0 r 2
Define d = d a z . Then, d cos  = d  ar (ar is the unit vector in the direction of P) .
If we define p = Qd (Coul  m) (electric dipole moment),
Q d cos  p  ar
V (r ,  ,  ) = =
4 0 r 2
4 0 r 2
  
Using  = ar + a + a ,
r r  r sin 
     Q d cos 
E (r ,  ,  ) = −V = −  ar + a + a 
 r r  r sin   4 0 r 2
Qd  cos  − sin   Qd cos  Qd sin 
=−  −2 3 ar + a  = ar + a
4 0  r r 3
 2 0 r 3
4 0 r 3

=
Qd
4 0 r
(
3
2 cos  ar + sin  a =
p
)
4 0 r 3
(
2 cos  ar + sin  a )
q Equi-potentialsurface
V 0

E E
V : J / Coul
E −| E −|
V =0

V 0
−q Equi-potentialsurface
See text Fig. 4.21 (b) for more detailed figures!

Magnetic dipole

H or B (= 0 H )
N
a

I
S
magnetic dipole moment m = I  a 2 az ( A  m 2 )

Magnetic field from a loop current (magnetic dipole moment m = I  a2 az ( A  m2 ) )

0 m  ar 0 I  a sin  a
2

Magnetic vector potential A(r , ,  ) = = (Wb / m)

4 r 2 4 r 2
 
m = I  a 2 az ar

0 m
B =  A =
4 r 3
( )
2 cos  ar + sin  a (Wb / m2 )

Compare with electric scalar potential from an electric dipole :

Q d cos  p  ar
V (r ,  ,  ) =
4 0 r 2
=
4 0 r 2
( electric dipole moment p = Qd a )
z

*** ar = sin  cos  ax + sin  sin  ay + cos  az

• Kinds of materials (depending on the binding force between the nuclei and
electrons) : conductors, semi-conductors, dielectric materials, magnetic materials)

Conductors : The binding force between the nuclei and electrons are very weak.
Dielctric materials : The binding force between the nuclei and electrons are very strong.

Free electron

Very weak binding force

Verystrong binding force

Nucleus
Conductor
Dielectric material
(PEC : Ideal conductor without
any friction inside) (wood, glass, etc)

Dielectric materials (Very strong binding force, No current flow, Non-conducting

like wood or glass or plastic materials)
Electric dipole moment p = qd (Coul  m)
N

p i
Polarization P = lim i =1

v
v → 0

P  E  P =  0 e E (e :electricsusceptibility)

electric dipole moment p = qd (Coul  m)

q N

p2
p1 p i

v Polarization P = lim i =1

p7 v
v → 0
p = qd (Coul  m)
= K E =  0  e E (Coul / m 2 )
d E p34
p90
p78 ( 0  e :proportionality constant)
−q p123 p11

 e : electric susceptibility (depending on materials)

 e =0 (free space)
 e =1.2 (wood)
 e =3 (glass)
 e =10-100 (ceramic)

Bound volume charge density and bound surface charge density

P P

P b b
P P

P P

Bound volume charge density b = −  P (Coul / m3 )

 an

an an
P P P
P

an
Bound surface charge density
 sb = P  an (Coul / m2 )

Bound volume charge density b = − P(Coul / m3 )

Bound surface charge density sb = P  an (Coul / m2 )

Definition of permittivity for dielectric materials

( )
In free space, D =  0 E and    0 E =  (Coul / m3 ) (free charge density)

( )
In a dielectric medium,    0 E =  ( free) + b (bound )
where  is free charge density and  b is bound charge density.
( )
  0 E =  −   P

  ( E + P ) = 
0

Define D =  0 E + P , then  D =  (D depends on onlyfree charges!)

( D : free charge density or electric flux density not dependent on medium)

D =  0 E + P =  0 E +  0 e E =  0 (1 + e ) E =  0 r E =  E

Examples)

Assume a free point charge q ( Coul ) at the origin with surrounding dielectric medium
(  =  0 r ). The following direction of polarization occurs when q is positive. The
opposite is true when q is negative.
 

b = −  P = 0(Coul / m3 )
Dielctric medium with  =  0 r

Bound surface charge Q sb

an Pr =  0  e Er (Coul / m 2 )
+q

Dielctric medium with  =  0 r free charge q at the origin

in a sphere with radius b → 0

q
Dr = (Coul / m 2 ) (not dependent on medium, that is, not dependent on  r )
4 r 2

D q
Dr =  0 r Er  Er = r = ( N / Coul ) (dependent on  r )
 0 r 4 0 r r 2
Want to look deeper into the problem.
 −1 q
Polarization Pr =  0 e Er =  0 ( r − 1) Er = r (Coul / m2 )
 r 4 r 2

At the spherical surface at r = b (b : very small, approaching zero), the bound surface
 −1 q  −1 q
charge density sb = P  an = r a  (−ar ) = − r (Coul / m2 )
 r 4 b 2 r
 r 4 b 2

Total bound charge almost at the origin

 −1 q  −1
qb = sb  4 b2 = − r  4 b2 = − r q(Coul )
 r 4 b 2
r
  −1  q
Total charge qt = q( free) + qb (bound ) = q +  − r q  = (Coul ) (decreases by a
 r  r
factor of  r ) (This result is also true for an infinite line charge or surface charge!!!)
Bound volume charge
1  1    −1 q 
density b = −  P = − 2 (r 2 Pr ) = − 2  r  = 0(Coul / m ) (all space looks
3

r r r r   r 4 
 r −1
like a free space if the negative bound surface charge of qb = − q(Coul ) is
r
considered together with the free charge q.

q / r
Electric field Er = ( N / Coul )
4 0 r 2
Problem) Do the same analysis for a free line charge with l (Coul / m) on the entire z axis
surrounded by dielectric medium with  r . Prove that
l
lt = l ( free) + lb (bound ) = (Coul / m) .
r

Problem) Do the same analysis for an infinite surface charge (free) with s (Coul / m2 ) on
a xy plane surrounded by dielectric medium with  r .

Definition of permeability for magnetic materials (in which a spin of current exists)

magnetic dipole moment m = I S ( A  m 2 )

North
Magnet N
S

S (m 2 ) I ( A)

South
A spin of current in an atom S
or molecule exists in magnetic material

m1
m2
v
N

m7 m
lim i =1 i
m34 Magnetization M = =  m H ( A / m)
m90 v → 0 v
m78  m : proportopnality constant
m123 m11
(called magnetic susceptibility)

 m : magnetic susceptibility (depending on materials)

 m = 0 (free space)
 m = 100 - 5000 (ferrite)
 an
I M

an
Jb J sb
I
an
Spin of current J sb

an M = m H an

Volume bound current density Surface bound current density

Jb =   M ( A / m ) 2
J sb = M  an ( A / m)

- Definition of permeability
B B
In free space, B = 0 H (or H = ) and   = J ( A / m2 )
0 0
B
In a magnetic medium,   = J ( free) + J b (bound )
0
where J ( A / m2 ) is the free current density and J b is the bound current density.
B
 = J +  M
0
 B 
 (
− M  =   H = J H depends only on free current! )
 0 
B
Defined H = − M (magnetic field determined by free current only, not dependent on
0
medium), then 0 H = B − 0 M
.
( ) ( )
B = 0 H + M = 0 H +  m H = 0 (1 +  m ) H = 0 r H =  H

Example)
 free line current I ( A)
on the entire z axis
Top view
M  =  m H I
= ( r − 1) H 
H = an
2 J b =   M = 0 ( A / m2 )

H  = 0  r M  =  m H
b  = 0  r : M  =  m H
= ( r − 1) H  I
( r − 1) I (  − 1) I
J sb = M  an = a  (−ar ) = r a z ( A / m)
2 b 2 b
H b : very small (b → 0)
bound current I b = J sb  2 b = ( r − 1) I ( A)

Free current I ( A), Total It = I ( free) + ( r − 1) I (bound ) = r I

I
 H  dl = H (2 ) = I ( A) ( free current )  H = 2 ( A / m) (Not dependent on 
l
r )

I
B =  H  = 0  r H  = 0  r (Wb / m 2 ) (Dependent on  r )
2

Want to look deeper into the problem.

I
Magnitization M  =  m H  = (  r − 1) H  = (  r − 1) ( A / m)
2
On the cylindrical surface with  = b (b: very small),
Bound surface current density
( J sb = M  an = ( r − 1)
I
2 b
( )
a  −a = ( r − 1)
I
2 b
az ( A / m)
Total bound current on the cylindrical surface at  = b is given by
I
I b = J sb  2 baz = ( r − 1) az  2 baz = ( r − 1) I ( A)
2 b
Total current (free + bound) I t = I + ( r − 1) I = r I ( A) (increases by a factor of  r )
Bound volume current density Jb =  M = 0( A / m2 ) (check it!) (So, may be
considered as a free space)
( I )
Magnetic flux density B = 0 r (Wb / m 2 )
2

Problems) Do the same analysis for an infinite surface current with J s = − J 0 ax ( A / m) on a

xy plane surrounded by magnetic medium with  r .

Conductors
Perfect Electric Conductor (PEC): Ideal. No loss, no friction inside, R=0, does not exist.
 (Coul / m2 )
No fields inside. The surface charge and surface current usually given by s
and J s ( A / m) may exist only on the surface of the PEC.

10V
PEC PEC
E ( N / Coul ) = 0 10V 10V
E E
10V
Equi-potential body

Real conductors : copper, silver, gold, etc.

Problems) Investigate the conductivity of real conductors! Which has the largest
conductivity?
1
 ( / m) : conductivity,  (= ) : resistivity (  m) ( depends onthe kinds of metal )

I ( A)

S (m 2 )
V (Volt )  ( / m)
J ( A / m 2 ) E (V / m) l (m), R ()

V
Electric field E = (V / m)
l
V
Current density J =  E = ( A / m2 )
l
 VS
Current I = JS = ( A)
l
V l l
Resistance R = = =  () (Undersatnd the difference of R and  ?)
I S S
I S
Conductance G = =  ( )
V l
Dissipated power density p (W / m3 ) at each point in the conductor
2
V 
p = J ( A / m )  E (V / m) =  E =    (W / m3 )
2 2

l 
S 2
2
V 
Total dissipated power P =  pdv =  E  Sl =    Sl =
2
V = GV 2 = RI 2 (W )
V
l  l
Energy

Examples of energy : Condensed spring, Water stored in a dam,

Potential energy=mgh(J), Food, Electric energy, Magnetic energy, etc

Electric field E and potential V due to a point charge q1

q1 (point charge) R
q1 aR
E= ( N / Coul = V / m) (vector field ) = −V
4 0 R 2
q1
V= (V = J / Coul ) ( scalar field )
4 0 R

Electric Energy We : Electric charges pulled closer against their forces have some
electric energy in J.

q1 (point charge) q2 (point charge)

R
q1 q1q2
Stored electric energyWe = (V = J / Coul )  q2 (Coul ) = (J )
4 0 R 4 0 R

Summary of Electrostatics
q1 aR F
Electric field E = = = −V ( N / Coul = V / m)
4 0 R 2 q2
q1q2 aR
Electric force F = = q2 E ( N )
4 0 R 2
q1 We
Electric potential V = = ( J / Coul = V )
4 0 R q2
q1q2
Electric energy We = = q2V ( J )
4 0 R
 m/s 
 J = N  m, V = J / Coul , N = kg  
 s 

We can express the electric energy in a more systematic form as follows.

q1q2 1 q2 q1  1
We = =  q1 + q2  =  q1V1 + q2V2  .
4 0 R 2  4 0 R 4 0 R  2

Similarly, for an electric charge system with N point charges, we have

 1 1 N
We =  q1V1 + q2V2 +  +qNVN  =  qiVi ( J ) where Vi is the potential at the position of
2 2 i =1
q i due to the other charges.

Similarly, for continuous charges in a volume of V with  (Coul / m3 ) ,

1 N 1 1
We =  qiVi →  (dq)V =  (  dv)V ( J ) (The electric energy is obtained with
2 i =1 2 2
charges and electric potential at the positions of the charges)

We try to obtain another expression expressed with electric fields.

We =
1

2 v
1
( 1
) 1
( )
Vdv =  V  D dv =   V D dv −  D  (V ) dv
2 v 2 v 2 v
( ) ( )
Math:   V D = ( V )  D + V   D ;similar to ( fg ) = f g
 + fg 
Using the divergence theorem and E =−V
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯ →
1
( ) 1
We =  V D  ds (surfaceintegral) +  D  Edv (volumeintegralinside thesurface) ( J )
2 s 2 v
This is true for any choice of a volume v bounded by s. This expression may be
inconvenient because we have to evaluate the two terms.

1 1
If we choose a very large surface with r →  , V D   2 and ds  r 2 , and the first
r r
term goes to 0. Then, using only the second term, the total stored electric energy can be
evaluated by
1 1 1 2 1 2
We =  D  Edv =   E  Edv =   E dv =   E dv =   dv ( J )
2 all space 2 all space all space 2 all space 2 all space e

(obtained by fields)
1 2
where e =  E ( J / m3 ) is the electric energy density (scalar field) .
2

1 1 1 2
Similarly, using the magnetic energy density m = B  H ==  H  H =  H ( J / m3 ) ,
2 2 2
1 2
the stored magnetic energy can also be obtained by Wm = all space m dv = all space  H dv ( J ) .
2

Prob) Find the electric energy We for a two-point charge (Q Coul, each) system
separated by R(m).
 R
q q

q q2
Sol 1) We = ( J / Coul )  q(Coul ) = (J )
4 0 R 4 0 R
1 2 1 1 q q  q2
Sol 2) We =  qiVi =  q1V1 + q2V2  =  q  +q = (J )
2 i =1 2 2  4 0 R 4 0 R  4 0 R

Prob) Find the electric energy We for a three-point- charge (Q Coul, each) system
separated by R(m).

q R q

R R

q
1 3 1 1 2q 2q 2q 
We =  qiVi   q1V1 + q2V2 + q3V3  = q +q +q 
2 i =1 2 2  4 0 R 4 0 R 4 0 R 
Sol)
1 2q 3q 2
= q 3 = ( J ) (three times !!)
2 4 0 R 4 0 R
Note) From Gauss’ law, E and V are the same as the point charge Q case if for the
following cases:

2 (
Coul / m2 ) and potential V =
Q Q
electric flux density D =  0 E = (V = J / Coul )
4 r 4 0 r 2

Problems) Find the electric energy We for a continuous charge on a sphere surface at
r=a(m) in free space. The total charge Q(Coul) is uniformly distributed over the surface.

sphere

a
Q (Coul ) uniformly
distributed on the surface
Sol)
Q D (r ) Q Q
For r  a, Dr (r ) = (Coul / m2 ), Er (r ) = r = ( N / coul or V / m), and V (r ) = (V )
4 r 2
0 4 0 r 2
4 0 r
(same results as the point charge case)
0 D (r )
For r  a, Dr (r ) = = 0(Coul / m2 ), Er (r ) = r = 0( N / coul or V / m),
4 r 2
0
Q a Q a Q
and V (r ) = +  Er (r )dr = +  0  dr = (V )
4 0 a r 4 0 a r 4 0 a

Dr (r )
Er (r ) =
0

Q
4 0 a 2

r
a

V (r )

Q
4 0 a

r
a

Stored electric energy using the first method :

1 1 Q 1 Q 1 Q Q2
We =  (dq)V =  (dq)
4 0 a 2 4 0 a 
=  ( dq ) =   Q = (J )
2 2 2 4 0 a 8 0 a

Stored electric energy using the second method :

Hint: Since the fields depend only on r, we had better use dv = 4 r 2 dr (You understand
what this means?) instead of dv = dr (rd )(r sin  d ) , for quick calculation!

2
1 2    Q  Q2  1
We =   E dv =  0
 2 
 4 r dr =
2
 dr
all space 2 r =a 2
 4 0 r  8 0 r =a r2
r =
Q2  1 Q2
= −
 r = (J )
8 0   r = a 8 0 a

Problems) Find the electric energy We stored in a spherical volume with r=a(m) in free
space. The total free charge Q(Coul) is uniformly distributed in the volume.
 sphere

a
Q (Coul ) uniformly
distributed inside the surface

Sol)
Q D (r ) Q Q
For r  a, Dr (r ) = (Coul / m2 ), Er (r ) = r = ( N / coul or V / m), and V (r ) = (V )
4 r 2
0 4 0 r 2
4 0 r
(same as above cases)
3
r
Q 
For r  a, Dr (r ) =  2 =
a Qr D (r ) Qr
(Coul / m 2 ) , Er (r ) = r = ( N / coul or V / m),
4 r 4 a 3
0 4 0 a 3
Q r2
Q Qr Q Q Q
( a2 − r 2 )
a
and V (r ) = + dr = + a
= +
4 0 a 4 0 a 4 0 a 4 0 a 3 2 4 0 a 8 0 a 3
r 3 r

3Q Qr 2
= − (V )
8 0 a 8 0 a 3
Q 3Q
Note that V (r = a) = (V ) and V (r = 0) = (V ) .
4 0 a 8 0 a
Er (r )

E = −V
Q
4 0 a 2

r
a

V (r )

3Q
8 0 a
Q
4 0 a

r
a

Stored electric energy using the first method (using charges and electric potential)
1 1 1 a Q  3Q Qr 2 
We =  (dq)V =  (  dv)V =   4 r dr  
2
− 3
2 2 2 r = 0 4 a3 / 3  8 0 a 8 0 a 
a
1 3Q 2 a  2 r4  3Q 2  3 r5  3Q 2 4a 3 3Q 2
=
2 8 0 a 4 r =0  3r − a 2  dr = 16 0 a 4  r − 5a 2  = 16 0 a 4  5 = 20 0 a ( J )
r =0
Stored electric energy using the second method (using the electric fields in all space)
2 2
1 2 a   Qr     Q 
We =   =  0 in free space
 E dv ⎯⎯⎯⎯⎯⎯ → 0
   4 r 2
dr + 
0
   4 r 2 dr
 4 0 a   4 0 r 
all space 2 r =0 2 3 r =a 2 2


Q2 a Q2  1 Q2 a5 Q 2  1 Q2 Q2
= r =0 + r = a r 2 =  + − = +
4
r dr dr  r
8 0 a 6 8 0 8 0 a 6 5 8 0   r = a 40 0 a 8 0 a
3Q 2
= ( J ) ( same as above)
20 0 a
6Q 2 5Q 2
= (J )  ( J ) (the result of the previous example for Q only on the sphere) Why ???
40 0 a 40 0 a

-Additional useful observations.

Q
We want to check whether Gauss’ law   D = 0 = (Coul / m 3 ) is satisfied inside the
4 a 3 / 3
sphere and   D =  = 0(Coul / m3 ) outside of the sphere.
1  ( r Dr ) 1  ( D sin  ) 1 D
2

Hint - D = + +
r2 r r sin   r sin  

Sol) Inside the sphere, uniform charge:

3
r
Q 
For r  a, Dr (r ) =  2 =
a Qr
(Coul / m2 )
4 r 4 a3
 Qr 
  r2
D = 2
1  ( r 2
Dr ) = 2

1  4 a 3  1 Q  3r 2
= 2
r r r r r 4 a 3
Q
= =  0 (Coul / m3 ) (given uniform charge)(proven !)
4 a 3 / 3

Outside the sphere, no charge:

Q
For r  a, Dr (r ) = (Coul / m 2 )
4 r 2
 Q 
 r2
1  ( r Dr ) 1  4 r 2 
2

D = 2 = 2 = 0(Coul / m3 ) (proven!)

r r r r

Electric scalar potential V and magnetic vector potential A

Electric scalar potential V

E(son) = −V (dad ) , then  E =  (−V ) = 0 is automatically satisfied.

 
  D =     E =     E =    ( −V ) =
 
 
  V = −   2V = − (Poison's equation)
 
In the locations where  = o
⎯⎯⎯⎯⎯⎯⎯⎯ → 2V = 0 (Laplace equation)
In the rectangular coordinate system,
       
2 ( Laplacian operator ) =   =  ax + a y + az   ax + a y + az 
 x y z   x y z 
2 2 2
= + +
x 2 y 2 z 2

Prob) For the following figure, obtain the voltage V(z) as a function of z.

z
h
PEC Q V0 (V )
0
PEC −Q

0 (V )

Sol) In the region 0  z  h ,  = 0 . So,  2V = −
= 0 . Since the voltage is mainly

 2 2 2  2
dependent on z,  2V =  2 + 2 + 2  V ( z ) = 2 V ( z ) = 0
 x y z  z

 V ( z ) = C (Constant)
z
 V ( z ) = Cz + D
V0
But, we know V ( z = 0) = D = 0(V ) and V ( z = h) = C  h = V0  C = .
h
V0
So, V ( z ) = z.
h
 V ( z)

V0

h z
The electric field is obtained by
    V
E = −V ( z ) = −  ax + a y + az V ( z ) = − 0 az (V / m)
 x y z  h


Solution of Poison’s equation (  2V = − )

()  
The solution V r due to the charge density   r  (Coul / m3 ) in a volume V is given
 
by
   
  r   dv dq  r  
()
V r =  
V 4 R
=   (Solution of Poisson’s equation : We already know
V 4 R

this!)
where R = r − r

Then, E = −V

Magnetic vector potential A

Similar to E = −V , we define the magnetic vector potential A such that

B(daughter ) =  A(mom) . Then,  B =  ( A) = 0 is automatically satisfied.

B
 H = J   = J   B =  J   A =  J

( ) For simplicity , we choose  A = 0
In math,    A =    A −  2 A ⎯⎯⎯⎯⎯⎯⎯⎯⎯ → − 2 A .

Then, −2 A =  J  2 A = − J (Magnetic Poisson’s equation), similar to


 2V = − (Electric Poisson’s equation).

()  
The solution A r due to the current density J  r  in a volume V is given by
 
    
 J  r   dv   r   dv
()
A r =
V
 
4 R
similar to V r =  () V
 
4 R
, where R = r − r .

()  
Note that A r is in the same direction as J  r  and thus is easier to obtain.
 
Then, obtain B =   A .

Real numbers(실수)
2
− 2 − 2
3 Re

−3 −2 −1 0 1 2 3
3x = −2  x = −3 / 2 (can find a realsolution.)
y 2 = 2  y =  2 (can find realsolutions.)
z 2 = −1  z ?(cannot find realsolutions.)

Complex numbers(복소수)
z 2 = −1  z =  −1 =  j (Imaginary numbers)

Imaginary

r
. z=a+jb

Real

z = a + jb = re j = r  = r cos  + jr sin 
| z |= Magnitude( z) = r = a 2 + b2 , Phase( z ) =  = Angle(a, b)

z = re j = (r cos  ) + j (r sin  ) = r (cos  + j sin  )

z * = a − jb = re− j = r (− ) , ( z * )* = z

Re( z ) = Re( z * ) = a , Im( z ) = b (real number!) , Im( z * ) = −b (real number!)

z1 = a1 + jb1 , z2 = a2 + jb2  z1 + z2 = (a1 + a2 ) + j (b1 + b2 )

 z1 − z2 = (a1 − a2 ) + j (b1 − b2 )

j1
z1 = re
1
j1
, z2 = r2e j2  z1z2 = re
1 r2e j2 = rr
1 2e
j (1 +2 )

j1 n
 z1n = (re
1 ) = r1ne jn1
z1 r1e j1 r1 j (1 −2 )
 = = e
z2 r2e j2 r2

z3 = z1 z 2

r1r2

1 +  2
z2
r2
2 r1 z1
1
z3 z3
= z2 = z2
z1 z1
z3 = z1 z2

zz * = re j re − j = r 2 =| z |2 =| z * |2
 z − z * = j 2b (purely
imaginary)

z = a + jb
r (purely
 z + z * = 2a real)
− zz * =| z |2 = r 2
r z * = a − jb

j1
z1z2* = re
1 r2e− j2 = rr
1 2e
j (1 −2 )
1 2 [cos(1 − 2 ) + j sin(1 − 2 )]
= rr

 
j  j 
Ex) z1 = 5 + j 5 3 = 10e 3
= 10 = 1060 , z2 = 1 + j 3 = 2e 6
= 2 = 230
3 6


| z1 |= Magnitude( z1 ) = 10 , Phase( z1 ) =
3
 
−j −j
z1* = 5 − j5 3 = 10e 3
, z1 + z1* = 10, z1z1* = 100 , z1* = 5 − j5 3 = 10e 3

   
−j j −j j
z = 1 − j 3 = 2e
2
* 6
, z z = 10e  2e
1 2
* 3 6
= 20e 6
= 10 3 + j10

Re( z1 ) = Re( z1* ) = 5 , Im( z1 ) = 5 3 , Im( z1* ) = −5 3 (real number!)

  
j j j
z1 z2 = 10e 3 2e 6
= 20e 2
= j 20 , z2 z2 * =| z2 |2 = 22 = 4
 2  5
j j j j
z12 = (10e 3 )2 = 102 e 3 , z15 = (10e 3 )5 = 105 e 3 ,
  
j −j j
z1 z2* = 10e 3 2e 6
= 20e 6


j 
z1 10e 3 j 5 5
= 
= 5e 6 = 3+ j
z2 j 2 2
2e 6
1
  
 j 2
1
j
1
j 30 10
5 + j5 3 = 10e  = 10 e 2
3 2 3
= 10e 6
= +j
  2 2
 3 
j j −j
j = e 2 , , −1 = e j , − j = e 2
=e 2
 2
  j 
( j) =  e 2  = e  j = −1  The solution of z 2 = −1 is z =  j
2

 

Problems) Calculate the followings

j (3 − j 4)*
z1 =
(−1 + j 6)(2 + j )2
1+ j
z2 =
4 − j8
2
1+ j 
z3 = j  3

2− j 
z4 = 630 + j5 − 3 + e j 45
Ans) z1 = 0.1644 − 9.46 , z2 = 0.397654.2
z3 = 0.24 + j 0.32 , z4 = 2.903 + j8.707

j t
*** What is e ? (very important!!) :  = 2 f (rad / s) (angular frequency),
f=frequency (cycles/s=Hz)
j t
: e is a rotating complex number in the counterclockwise sense with magnitude of 1,
f turns in one second and one turn in a period T=1/f(sec).
1
One turn in T = (sec) or f turns in one sec
f
jt
e = e j 2 ft
1
t

If you multiply e jt

to any complex number z (phasor), that complex number (z e )
j t

1
will rotate one turn in T= (sec) or f turns in one sec in a counterclockwise direction.
f

Low-frequency and High-frequency circuits

Propagation constant along the line :

 2 f 2 360
= = = = (rad / m or  / m)
 f  
 :one wavelength (distance the wave travels when thesourceoscillates onecycle)

For a physical length L(m), the electrical length is defined as

 2 L L
L = (rad / m)  L(m) = 2 (rad ) = 360 () . Very important!!!!!
  
Physical     2 3
length L(m)
8 4 2
Electrical    (rad ) 2 (rad ) 4 (rad ) 6 (rad )
length  L (rad ) (rad )
(rad) 4 2

Low-frequency circuit ( L  or  L 1)
L  ( Low frequency circuit )
+
+
 (t ) = V0 cos(t )  (t )  V0 cos(t )
−
−
L
3  105 km / s
For example, when f=60 Hz,  = = 5000 km ,
60 / s
2 2 (rad )
= =  1.25 10−6 ( rad / m)
 5 10 (m)
6


In the low frequency limit where  = 2 f → 0 (DC case),  = →  ( m) ,
f

= → 0(rad / m) and  L = 0(rad )


High-frequency circuit ( L   or  L  1)
 High frequency circuit (roughly L   )

+
 (t ) = V0 cos(t −  L)
+  L 
 (t ) = V0 cos(t ) = V0 cos  (t −

)
 
−  L 
= V0 cos  (t − )  = V0 cos  (t − t ) 
  
−
L ( m)
 L(rad )
L(m) is the physical length.
 L(rad ) is the electrical length (important!!).
L
t = (sec) is the time for the signal to travel the distance of L (m) or

the signal delay time.

Instantaneous and phasor forms, power (Low-frequency circuit)

v(t ) = Vm cos( wt + V ) (instantaneous form)  V = Vme jV (phasor form: physical

quantities expressed by complex numbers)

Im
v(t ) = Vm cos(t + V )
V = Vm e jV
V Vm
V
  3 
−
2 2
2 t Re

v(t ) = Vm cos(t + 0) Im

V = Vm e j 0

 3 
2 2
2 t Re
  Im
v(t ) = Vm cos(t + ) = −Vm sin(t ) j

2 Vm V = Vm e 2


2

  3 
2 2
2 t Re

Im
v(t ) = Vm cos(t +  ) = −Vm cos(t )

  3 
2 2
2 t V = Vm e j = −Vm Re

3 
v(t ) = Vm cos(t + ) = Vm cos(t − ) = Vm sin(t ) Im
2 2

3
2

  3 
2 2
2 t −
 Re
2
3 
j −j
V = Vm e 2
= Vm e 2

How to recover the instantaneous form v (t ) from the phasor form V :

First step : Multiply V with e jwt , then Ve jwt (the red dots in the following figure) will
2 2 1
rotate around the origin of a complex number plane one turn in = = (sec)
 2 f f
and f turns in one second starting from V.
Second : Take the real part of Ve jwt , then it is just the instantaneous form v (t ) (the blue
dots).

v(t ) = Re(Ve jwt )

Im


Ve jt (t = )
2
Ve jt (t = 0, 2 , 4 ,..)

Re

Ve jt (t =  ) 3
Ve jt (t = )
2

V = Vm e jV  v(t ) = Re(Ve jt ) = Vm cos(t + V )

i(t ) = I m cos(wt + I )  I = I me jI

i (t ) = Re( Ie jwt )

i(t)

Circuit size L<< 

i(t)
+ +
v(t) R,L,C
v(t)
- -

Source generated instantaneous power

p(t)=v(t)i(t)(W)

Instantaneous power p(t ) = v(t )i (t ) = Vm I m cos( wt + V ) cos( wt + I )(W ) (function of

time)
 1 T 1 T
Average power P =
T 0 p(t )dt =  Vm I m cos(wt + V ) cos( wt + I )dt
T 0
V I T1 1
= m m   cos(2wt + V + I ) + cos(V − I ) dt = Vm I m cos(V − I )
T 0 2 2
2 2 1
(T = = = sec)
 2 f f
S is the complex power defined by
1 1 1
complex power S = VI * = Vm e jV I m e− jI = Vm I m e j (V −I )
2 2 2
1 1
= Vm I m cos(V − I ) + j Vm I m sin(V − I ) = P + jQ
2 2
[P: Real power(유효전력), Q: Reactive power(무효전력)]

1 1
P = Re( S ) = Vm I m cos(V − I ) , Q = Im( S ) = Vm I m sin(V − I )
2 2

V Vm e jV Vm j (V −I )

Impedance Z = = = e = R + jX
I I me jI Im
(R: Resistance, X: Reactance)

1
V I e j (V −I )
S 2 m m 1
Examine = = I m2 (positive real number)
Z Vm j (V −I ) 2
e
Im
1 S
S and Z are brothers!!! S = I m2 Z or Z =
2 1 2
Im
2
Imaginary
1
Q S = Vm I m e j (V −I )
2

X Vm j (V −I )
Z= e
Im
V −  I

Real
R P

Classifications of the impedance Z

 Imaginary

Purely
Inductive Inductive(R,L)
(L only) (X>0)

Resonant Real

Purely
capacitive Capacitive(R,C)
(C only) (X<0)

Active Passive
(Negative R) (Positive R)

Complex plane for

impedance(Z=R+jX)

I 1
Admittance Y = = = G + jB (G: conductance, B: susceptance)
V Z
1 1 1 1 1
S = VI * = ZII * = | I |2 ( R + jX ) = | I |2 R + j | I |2 X = P + jQ
2 2 2 2 2
1 * 1 1 1 1
S = VI = V (YV )* = | V |2 (G − jB) = | V |2 G − j | V |2 B = P + jQ
2 2 2 2 2


Prob) A voltage source v(t ) = 100cos(t + )(V ) is connected with a load and the
2

current flowing through the load is 1) i (t ) = 5cos( wt + )( A) , 2) i(t ) = 5cos( wt )( A) ,
2
 
3) i(t ) = 5cos( wt +  )( A) , 4) i (t ) = 5cos( wt + )( A) , 5) i (t ) = 5cos( wt − )( A) . For each
6 2
case, plot v (t ) , i(t ) , and plot p(t ) using MATLAB assuming f = 60 Hz (also obtain
average real power P using MATLAB). Besides, for each case, obtain the complex
power S, average real power(평균유효전력) P, average reactive power
(평균무효전력) Q, and load impedance Z(Mark the impedance Z on the complex plane
and discuss whether it is passive or active(resonant, inductive, capacitive).
Prob) Given that V = 100 + j100 (V) and I = 20 2 (A), obtain v (t ) , i(t ) , p(t ) (also
1 T
plot them using MATLAB), real average power P from the definition P =  p(t )dt ,
T 0
complex power S, average real power P from Re(S), reactive power Q from Im(S), and
the impedance Z (Mark the impedance Z on the complex plane and discuss whether it is
passive or active(resonant, inductive, capacitive).

Prob) Given that V = 100(1 − j ) (V) and I = 20 2 (A), obtain v (t ) , i(t ) , p(t ) (also plot
1 T
them using MATLAB), real average power P from the definition P =  p(t )dt ,
T 0
complex power S, average real power P from Re(S), reactive power Q from Im(S), and
the impedance Z (Mark the impedance Z on the complex plane and discuss whether it is
passive or active(resonant, inductive, capacitive).

Prob) Given that V = 100(1 − j ) (V) and I = 20(1 − j ) (A), obtain v (t ) , i(t ) , p(t ) (also
1 T
plot them using MATLAB), real average power P from the definition P =  p(t )dt ,
T 0
complex power S, average real power P from Re(S), reactive power Q from Im(S), and
the impedance Z (Mark the impedance Z on the complex plane and discuss whether it is
passive or active(resonant, inductive, capacitive).

Instantaneous and phasor forms, power (High-frequency or microwave circuit)

Consider the forward traveling voltage and current waves:

v( z, t ) = Vm cos(wt −  z + V )  V ( z) = Vme jV e− j z = V +e− j z

i( z, t ) = Im cos(wt −  z + I )  I ( z) = Ime jI e− j z = I +e− j z

1 1 jV 1
Complex power S = V ( z ) I ( z ) = Vm e
*
e− j z I me − jI e j z = Vm I me j (V −I )
2 2 2
1 1
= Vm I m cos(V − I ) + j Vm I m sin(V − I ) = P + jQ
2 2
[P: Real power(유효전력), Q: Reactive power(무효전력)]

1 1
P = Re( S ) = Vm I m cos(V − I ) , Q = Im( S ) = Vm I m sin(V − I )
2 2

Instantaneous and phasor forms, power (High-frequency or microwave fields)

The phasor E(r ) = Ex (r )ax + Ey (r )ay + Ez (r )az is enough to specify the time-harmonic
fields.
E(r, t ) = Re[E(r )e jwt ]
= ax Re  Ex (r )e jt  + a y Re  E y (r )e jt  + az Re  Ez (r )e jt 
E(r, t ) :Instantaneous form, E(r) : phasor form

Representation of vector fields in instantaneous forms (No j !!!)

E(r, t ) = E( x, y, z, t ) = Ex ( x, y, z, t )ax + Ey ( x, y, z, t )a y + Ez ( x, y, z, t )az : Rectangular
E(r, t ) = E(  ,  , z, t ) = E (  ,  , z, t )a + E (  ,  , z, t )a + Ez (  ,  , z, t )az : Cylindrical
E(r, t ) = E(r, ,  , t ) = Er (r, ,  , t )ar + E (r, ,  , t )a + E (r, ,  , t )a : Spherical
Ex ( x, y, z , t ) , Ey ( x, y, z, t ) , Ez ( x, y, z , t ) , etc are all real scalar fields.

Representation of fields in phasor forms (No t !!!)

E(r ) = E( x, y, z) = Ex ( x, y, z)ax + Ey ( x, y, z)a y + Ez ( x, y, z)az : Rectangular
E(r ) = E(  ,  , z) = E (  ,  , z)a + E (  ,  , z)a + Ez ( ,  , z)az : Cylindrical
E(r ) = E(r, ,  ) = Er (r, ,  )ar + E (r, ,  )a + E (r, ,  )a : Spherical
Ex ( x, y, z ) , Ey ( x, y, z) , Ez ( x, y, z ) , etc are all complex scalar fields.
.
Ex) E( x, y, z) = 10axe− j z (plane wave)
 E ( x, y, z, t ) = Re  E ( x, y, z )e jt  = ax Re 10e − j z e jt  = ax10cos (t −  z )
Ex) E( x, y, z) = 10(ax + j 2ay )e− j  z
 E ( x, y, z, t ) = Re  E ( x, y, z )e jt  = ax Re(10e − j  z e jwt ) + a y Re( j 2e − j  z e jwt )
= ax Re[10e j ( wt −  z ) ] + a y Re[2e j ( wt −  z + 2) ]
= ax cos(wt −  z) + ay 2cos(wt −  z +  2) (Left-handed elliptically polarized wave
propagating in the z direction)

e− j r e − j  r jwt
E (r ,  ,  ) = − j10  E (r ,  ,  , t ) = Re[ E (r ,  ,  )e jwt ] = Re[ − j10 e ]
r r
10 10
= Re[e− j e e ] = cos( wt −  r −  2) (Linearly polarized wave propagating in the
2 − j  r jwt

r r
r direction)

H (  ,  , z, t ) = 20cos(wt −  z +  3)a  H (  ,  , z) = 20e j ( −  z + 3) a = 20e j /3e− j z a

Prob) Instantaneous form → Phasor form

E( z, t ) = E0 cos(t −  z +  )ax → Phasor E( z) = E0e− j z + j ax = E0e− j(  z − ) ax
E( z, t ) = 10cos(6 108 t − 2z +  / 3)ax → E( z) = 10e− j 2 z + j /3 ax = 10e− j( 2 z − /3) ax
     3 
P( x, t ) = 2sin(10t + x − )a y = 2cos  − (10t + x − )  a y = 2cos  − (10t + x)  a y
4 2 4   4 
3  3 
 3  + jx − j + j x− 
= 2cos 10t + x −  a y → P( x) = 2e 4
a y = 2e  4  a y
 4 
A( z, t ) = 5sin(wt − 2z)ax
B( y, t ) = 15e−2 y cos(wt − y)az → B( y) = 15e−2 ye− jy az
C ( x, t ) = 5cos ( wt − x ) ay − 8sin(wt − x)az
10
H (  , z, t ) = cos(3 109 t − 3 )a


Phasor form → Instantaneous form

2
−j r
sin  e 3
sin   2 
H ( r , ,  ) = 10 a → H ( r , ,  , t ) = Re  H ( r , ,  ) e jt  = 10 cos  t − r  a
r r  3 

V = j5e−20 = 5e j 90e−20 = 5e j 70 → v(t ) = Re (Ve jt ) = Re 5e j (t +70)  = 5cos(t + 70)
V ( z ) = 10e j 30e− j z → v( z, t ) = Re 10e j 30e− j z e jt  = 10cos (t −  z + 30 ) (traveling
wave propagating with a phase velocity of
 (rad / sec)
v= (m / sec)
 (rad / m)

H ( z) = 10e− kz ax + j5e− jkz a y

e − j 3r
E (r ,  ,  ) = 10e − j 3 a
r

What is meant by sin(j5) ?

e jx + e − jx e jx − e− jx
cos( x) = sin( x) =
2 2j
−x
e +e
x
e x − e− x e x − e − x x →0 1 + x − (1 − x)
cosh( x) = sinh( x) = tanh( x) = x − x ⎯⎯⎯ → =x
2 2 e +e 2
e jz + e − jz e jz − e− jz sin z
cos( z ) = sin( z ) = tan( z ) =
2 2j cos z
− j ( jy ) −y
e j ( jy )
+e e +e y
cos( jy ) = = = cosh( y )
2 2
e j ( jy ) − e− j ( jy ) e− y − e y e y − e− y
sin( jy) = = =j = j sinh y
2j 2j 2
tan( jy) = j tanh( y)
 e z + e− z e z − e− z e z − e− z
cosh( z ) = sinh( z ) = tanh( z ) = z
2 2 e + e− z
e jy + e − jy
cosh( jy ) = = cos( y )
2
e jy − e− jy e jy − e− jy
sinh( jy) = =j = j sin( y)
2 2j
sin( jy ) j sin( y )
tanh( jy ) = = = j tan( y )
cos( jy ) cos( y )
tan z1 + tan z2
tan( z1 + z2 ) =
1 + tan z1 tan z2
tanh z1 + tanh z2
tanh( z1 + z2 ) =
1 + tanh z1 tanh z2

• Capacitance and inductance

Q
- Capacitance C = (Coul / V = F )
V
i ( A)
A(m 2 ) Q(Coul )


V (Volts )
D, E Q
d (m) Capacitance C = (Coul / V = F )
V

−Q(Coul )
Electric field onlyinside!
V
Electric field E = (V / m)
d
V
Electric flux density D =  E =  0 r (Coul / m2 )
d
V
Surface charge density  s = D =  E =  0 r (Coul / m2 )
d
VA A
Total stored charge Q =  s A =  0 r =  0 r V (Coul )
d d
Q A
Capacitance C = =  0 r (Coul / V = F )
V d
Stored electric energy : Assume that the electric potential at the bottom surface is V0,
then the electric potential at the top surface is V+V0.

1 1 1 1 1
We =  (dq)V = Q(V + V0 ) + (−Q)V0  = QV = (CV )V = CV 2 ( J )
2 2 2 2 2
(Note that We does not change depending on V0.)
We can obtain the electric energy from a field analysis.

1 1
Stored electric energy density we = D  E =  E 2 ( J / m3 )
2 2
Stored electric energy
2
1 1 V  1 A 1 1
We =  we dv =  E 2 Ad ( J ) =    Ad =   V 2 = CV 2 = QV ( J )
V 2 2 d  2 d 2 2

Ex) Consider the following figure : For the given Q, obtain the voltage V and
capacitance C. : Q = CV

V? W −Q(Coul )
 =  0 r
h Q(Coul )
l
W
Front view
r h
Px Px
 r −1
Bound charge Qb = − Q Ex Ex
r
Free charge Q   −1  Q
Total charge Qt = Q(free)+Qb (bound) = Q +  − r Q =
 r  r

Start from the given free charge Q on the bottom surface.

Q
Surface charge density at the bottom  s = (Coul / m2 )
Wl
Q
(Free charge density) Electric flux density Dx =  s = (Coul / m2 ) (not dependent on
Wl
r )
D D Q
Electric field Ex = x = x = ( N / Coul = V / m) (reduced by a factor of
  0 r  0 rWl
 r compared with the case of free space)
hQ
Electric potential difference V = Ex h = (Volts) (reduced by a factor of
 0 rWl
 r compared with the case of free space)

Q  0 rWl
Capacitance C = = (Coul / V = F ) (increased by a factor of  r compared with
V h
the case of free space)
Q
W
C (per unit line length) = V =  0 r ( F / m)
Capacitance per unit line length l h
h
Define g ( geometricalfactor ) =
⎯⎯⎯⎯⎯⎯⎯⎯⎯
W
→  0 r / g ( F / m)
 1 1 hQ hQ2
Electric energy We =  dqV = QV = Q = (J )
2 2  0 rWl 2 0 rWl
( − 1) Q (Coul / m2 )
Polarization Px =  0  e Ex =  0 ( r − 1) Ex = r
 rWl
Bound surface charge density at the bottom
( − 1) Q a = − ( r − 1) Q (Coul / m2 )
( )
sb = an  P = −ax  P = −ax  r ( )
 rWl x
 rWl

Bound surface charge at the bottom Qb =  sbWl = − r

( − 1) Q (Coul )
r
Total surface charge at the bottom Qt = Q( free) + Qb (bound ) = Q +  −  r − 1  Q = Q (Coul )
 r  r

Ex) Consider the following coaxial transmission line : For the given free line charge
Q
density l =  (Coul / m) on the inner (  = a ) and outer surface (  = b ), obtain the
L
voltage V and capacitance per unit length C(F/m).

L ( m)
− l (Coul / m)
lb
l
a l (Coul / m)
Q
l = (Coul / m)
b L r
E

Using Gauss’ law,

Q Q/L 
 S
D  ds = D  2 L = Q(Coul )  D =
2 L
= = l (Coul / m 2 )
2 2
D l
E = = (V / m)
 0 r 2 0 r 
Voltage difference between the two conductors
b l l l b
V =  E  dl =  a  d  a = ln  b
= ln (V )
2 0 r  2 0 r 2 0 r a
l a a

The capacitance per unit length C(F/m) is obtained as

l (Coul / m) l 2 0 r
C ( F / m) = = = ( F / m)
V (Volt ) l b b
ln ln
2 0 r a a
 - Inductance of a solenoid

 (magnetic fluxlinkage) = N (turns)   (magnetic flux) = LI (Wb)

Inductance L =  (magnetic fluxlinkage) = N (magnetic flux) (Wb / A = H )

I I
Magnetic field only inside!

A(m 2 )

i  i
L(m), N turns
 (magneticflux linkage)
Inductance L = (Wb / A = H )
i
B, H

Ampere's closed path l

Use Ampere’s law.

Ni
 H  dl = HL + 0 = Ni( A)  H =
l L
( A / m)

Ni
Magnetic field H = ( A / m)
L
Ni
Magnetic flux density B =  H = 0 r (Wb / m2 )
L
Ni
Magnetic flux  = BA = 0 r A (Wb)
L
N 2i
Magnetic flux linkage  = N  = 0  r A (Wb)
L
d d
(note that the induced voltage v = (Wb / s = V ) (not )
dt dt
 N 2
Inductance L = = 0  r A (Wb / A = H )
i L
1 1 1
Stored magnetic energy Wm =  i = ( Li)i = Li 2 ( J )
2 2 2
We can obtain the electric energy from a field analysis.

1 1
Stored magnetic energy density at each point in the coil wm = B  H =  H 2 ( J / m3 )
2 2
Stored magnetic energy
2
1 1  Ni  1 N2 1
Wm =  wm dv =  H AL( J ) =    AL =  
2
A  i 2 = Li 2 ( J )
V 2 2  L 2 L 2

Ex) Inductance of two parallel plates (Current I flows on the two parallel
conductor plates as shown below). :  = LI

W
I ( A)
 = 0  r
Js = I / W
x ( A / m)
H h M = m H
z l
y W
Front view H M
r h M = m H
Bound current I b = ( r − 1) I H M = ( r − 1) H
Free current I
Total current I t = I (free)+Ib (bound) = I + ( r − 1) I = r I ( A)

I
Surface current density at the bottom J s = ( A / m)
W
I
Surface current density at the top J s = − ( A / m)
W
I
Magnetic field H = J s = ( A / m) ( H is determined only by free current)
W
I
Magnetic flux density B = 0 r H = 0 r (Wb / m2 )
W
  hlI
Magnetic flux  = Bhl = 0 r (Wb)
W
   hl Wb
Inductance L = = 0 r ( = H)
I W A

h
L( per unit line length) = I = 0  r ( H / m)
Inductance per unit line length l W
h
Define g ( geometricalfactor ) =
⎯⎯⎯⎯⎯⎯⎯⎯⎯
W
→  0  r g ( H / m)
1 1   I2
Magnetic energy density wm = B  H = 0  r H 2 = 0 r 2 ( J / m3 )
2 2 2W
Magnetic energy
0  r I 2 0  r hlI 2 1 0  r hl 2 1 2 1
Wm =  wm dv = Whl = = I = LI =  I ( J )
2W 2 2W 2 W 2 2
I
Magnetization M y =  m H y = ( r − 1) ( A / m)
W
Bound surface current density at the bottom
I
( ) I
J sb = M  an = ( r − 1) a y  −ax = ( r − 1) az ( A / m)
W W
Bound surface current at the bottom Ib = ( r −1) I ( A)
Total surface current at the bottom It = I ( free) + Ib (bound ) = I + ( r −1) I = r I ( A)
(This is the reason why the magnetic flux density B is increased by a factor of  r )

Problem) If free currents I (  = a) = I and I (  = b) = − I flow on a coaxial line, show

that the inductance L (H/m) per unit length is given by

Magnetic flux per m l (Wb / m) 2

L ( H / m) = = = 0  r  (Show!!)
Current I ( A) b
ln
a

• Conversion of equations in instantaneous forms to phasor forms

+ vR (t ) -
v (t )
+ L - +
vC (t )-

+ R L C
v(t )
-
i (t )

 R (t ) = Ri(t ) (time-form or instantaneous form)

 Re(VRe jt ) = R Re(Ie jt )
 Re (VR − RI ) e jt  = 0  VR = RI (phasor form)

dx(t )
Consider a time-form equation y (t ) = in general. There are many cases in
dt
engineering and science.

Some examples of y(t) and x(t)

y(t) x(t)
velocity position
acceleration velocity
i(t) q(t)=Cv(t) for a capacitor
v(t)  (t ) =Li(t) for an inductor
 dx(t )
Instantaneous time equations : y (t ) =
dt
dx(t )
If x(t ) = sin t = cos (t − 90) , y (t ) = =  cos t
dt
: y(t) is 90 degrees faster than x(t) and y(t)’s magnitude is  .
x(t ) = sin t

1
t

y (t ) =  cos t =  sin (t + 90 )


t

dx(t )
Instantaneous time equations y (t ) = =➔ Phasor equations Y = j X
dt

Phasor relation (Much simpler!!) : Y = j X

Imaginary
Y = j X

X
Real
 x(t ) = A sin 0t = A cos (0t − 90 )
Examples ) dx(t )
 y (t ) = = A0 cos 0t = A0 sin (0t + 90 )
dt

For an inductor,
d  (t ) di(t )
v(t ) = =L  V = j = j LI
dt dt

For a capacitor,

dq(t ) dv(t )
i(t ) = =C  I = jQ = jCV
dt dt

For a series RLC circuit,

v(t ) = vR (t ) + vL (t ) + vC (t )
 V = VR + VL + VC
1  1 
= RI + j LI + I =  R + j L + I
jC  jC 

Impedance Z = V/I

R L C
Z(w) =V/I R jwL 1/(jwC)
Y(w) =I /V 1/R=G 1/(jwL) jwC
Note that the impedances of L and C are functions of frequency.

Resonant R, L, C circuit
I + VC -

+ R L C
V
-
V 1  1 
Zin = = R + j L + = R + j  L − = R + jX
I jC  C 
(R: Resistance, X: Reactance)
1 1 1 1 2  1 
Complex power S = VI * = ( Zin I ) I * = Zin I = I  R + j   L −
2

2 2 2 2   C 
1 2 1 2 1 2 1  1 2 1 2 1 2 1 
= I R + j 2  I L − I  = I R + j 2  I L − I C
2 4 4  C 2
2
4 4 C 
2 2
 1 2 1 2 1 
= I R + j 2  I L − VC C  = Pl + j 2 (Wm − We )
2

2 4 4 
where Wm = time average magnetic energy stored in the inductor (J) and We = time
average electric energy stored in the capacitor (J).
P + j 2 (Wm − We ) At resonant frequency  =1/ LC , Wm =We
Zin = R + jX = l 2
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ R
I /2
Quality factor
average energy stored W + We 2Wm 2W
Q = = m (dimensionless) ⎯⎯⎯⎯⎯ At resonance
→0 = 0 e
energy loss / sec Pl Pl Pl

1 2
2 I L
 L 1/ (0C ) VL VC
= 0 4 = 0 = = = ( Q : usually 10 ~ 1000 )
1 2 R R V V
I R R R
2
(Ratio of the voltage magnitude applied at L or C and R)

At the resonant angular frequency of  0 ,

VR = RI
0 L
VL = j0 LI = jR I = jRQI = jQVR
R
1
VC = − j I = − j0 LI = −VL = − jQVR
0C
V = VR + VL + VC = VR

Imaginary

VL = jQVR

VR , V Real

VC = − jQVR
Based on the phasor voltages, the instantaneous voltages are :
( )
vR (t ) = Re VR e jt = VR cos (t )
v (t ) = Re ( jQV e ) = Re ( QV e
L R
jt
R
j 90
)
e jt = QVR cos (t + 90 )
vC (t ) = QVR cos (t − 90 )

Investigate Z in at  = 0 +  :  =  − 0 .
 02 
1
1  1  0 = LC
Zin = R + jX = R + j L + = R + j L 1 − 2  ⎯⎯⎯⎯ → = R + j L 1 − 2 
jC   LC    

= R + j L 
 2 − 02
= R + j L 
( + 0 )( − 0 )
 2
2
2  ( − 0 )
if  is near 
⎯⎯⎯⎯⎯ 0
→ R + j L  = R + j 2 L ( − 0 ) ⎯⎯⎯⎯⎯⎯
Define  =  −
0
→ R + j 2 L
 2

2RQ L
= R+ j (using Q = 0 )
0 R
X ( )

0 

Z in ( )

2R
BW
R

1 
0

The maximum current (and power) flows when the impedance becomes smallest (R) at
 = 0 or  = 0 .
 2 
2 RQ Define BW =
0
Zin = R + j ⎯⎯⎯⎯⎯⎯ → R + jRQ  BW
0
When Zin = R + jRQ  BW = 2R , the current decreases as 1/ 2 (power 1/ 2 ).
From this, we obtain Q  BW = 1 or BW = 1/ Q .
Other useful forms of
1    L   0 
Zin ( ) = R + j L + = R + j0 L  − 0  = R + j  − 
jC  0   C  0  
       
= R + jRQ  − 0  = R 1 + jQ  − 0  
 0     0   
L   
: Reactance slope parameter,  − 0  : Frequency variable
C  0  

Poynting’s theorem (Energy (J) conservation, Power conservation (W=J/s)

Poynting vector P(r, t ) = E(r, t )  H (r, t )(W / m2 ) : power density for each point in
space

B H
 E = − = − (V / m 2 ) ………..(1)
t t
D E
 H = J + = J + ( A / m 2 ) …..(2)
t t
H
H (1)  H (  E ) = −  H ………(3)
t
E
E (2)  E (  H ) = J E +  E ……(4)
t
Using the vector identity
 ( A B) = B ( A) − A ( B) ,
H
(3)  ( E  H ) + E (  H ) = −  H ….(5)
t
(4) → (5)
E H
− ( E  H ) = J E +  E+ H
t t
 1 1 
= J E +   E 2 +  H 2  (W / m3 ) …(6) (point form)
t  2 2 
Applying  ( Left = Right )dv operation, we obtain the integral form for Poynting
V
theorem.
 1 2 1 2
−  ( E  H )dv =  J Edv +    E +  H dv(W ) .
V V t  2
V 2 
 ds

V ds

Using the divergence theorem on the left, we get

 1 2 1 2
 S (E  H ) (−ds) = V J Edv + t V  2  E + 2  H dv(W )
or

S P (−ds) = V pdv + t V (m + e )dv(W ) (Integral form)
⎯⎯⎯ Statics
 →0
→  pdv
V
where

P = E  H (W / m2 ) (Poynting vector: power density)

p = J E(W / m3 ) (Ohmic power loss density)
1
m =  H 2 ( J / m3 ) (Magnetic energy density)
2
1 2
e =  E ( J / m3 ) (Electric energy density)
2
(The power flowing into the surface S is equal to the sum of conduction losses and time
change of the electric and magnetic energy)

Ex) Prove the total power flowing into the surface S is I 2 R(W ) in a static case. The
resistive material with radius a in S is characterized by the resistance R.
 Ez
S
R () ds
H
z
I(A)

L(m)

V RI I
Sol) V = RI (V ) , Ez (  = a) = = (V / m) , H (  = a) = ( A / m)
L L 2 a
RI I RI 2
P = E  H = Ez az  H  a = −  a = − a (W / m 2 )
L 2 a 2 a  L
The total power flowing into the the surface S is
RI 2
P =  P (−ds ) = − a (−2 a  La ) = RI 2 (W ) or VI
S 2 a  L

Poynting vector (or power density vector field)

In instantaneous time forms

P(r, t ) = E(r, t )  H (r, t )(W / m2 )

The time average power density vector field is given by

1 T
P(r ) =  E (r , t )  H (r , t )dt (W / m 2 )
T 0
1 *
= Re[ E (r )  H (r )](W / m2 )
2

Complex Poynting vector filed (or complex power density vector field)
1 *
S (r ) = E (r )  H (r ) = P + jQ(W / m2 )
2
1 * 1 *
P(r ) = Re[ S (r )] = Re[ E (r )  H (r )] , Q(r ) = Im[ S (r )] = Im[ E (r )  H (r )]
2 2

Poynting theorem in phasor form

-point form
 E = − j B (V / m2 ) ………..(1)
   
 H = J + j D ( A / m2 ) ⎯⎯⎯⎯→
( left =right )
H = J − j D …..(2)
( )
  
(1)  H :   E  H = − j B  H …(3)

(2)  E : E   H ) = ( J − j D )  E ….(4)
(   

(4) − (3) : E  (  H ) − (  E )  H = J  E + j ( B  H − D  E ) ………(5)

    

Since we already know E  (  H ) − (  E )  H = − ( E  H )

  

using the vector identity  ( A B) = B ( A) − A ( B) ,

(5) becomes
− E  H ( 
) = J  E + j ( B  H − D  E ) .
  

Since the complex Poynting vector (power density) was defined as

1 *
S = E  H (W / m 2 ) , we can write the point-form and volume-form Poynting theorem
2
as follows.

-point (differential) form

1 * 1 * 1 * 
− S = E J + j  B H − D E 
2 2 2 
1  1 * 1   1 1 1 
=  E 2 + j 2  B H − D H  =  E 2 + j 2   H 2 −  E 2  (W / m3 )
2 4 4  2 4 4 

-volume (integral) form

 S (−ds) (total complex power flowing into a closed surface S)
S

1  1 1 
=  E 2 dv (power dissipated or absorbed in V) + j 2    H 2 dv −   E 2 dv 
V 2
 V 4 V 4

= P + jQ (W = J / s)

1
Note that   H 2 dv ( =Wm ) is the time average of the stored magnetic energy in V and
V 4
1 2
  E dv ( = We ) is the time average of the stored electric energy in V.
V 4
 ds

S
- ds

V - ds ds

1
P = Re   S (−ds)  =   E 2 dv
 S  V2
: Average real power(유효전력)(W) flowing into the surface S

 1 1 
Q = Im   S (−ds)  = 2    H 2 dv −   E 2dv  = 2 (Wm − We )
 S   V 4 V 4

: Average reactive power(무효전력)(W) flowing into the surface S

Compare the above field version of Poynting theorem with that of the circuit version
given by

I + VC -

+ R L C
V
-
Complex power
1 1 1 1 2  1 
S = VI * (power flowing into a system) = ( Zin I ) I * = Zin I = I  R + j   L −
2

2 2 2 2   C 
1 2 1 2 1 2 1  1 2 1 2 1 2 1 
= I R + j 2  I L − I  = I R + j 2  I L − I C
2 4 4  C 2
2
4 4 C 
2 2

1 2 1 2 1 
= I R + j 2  I L − VC C  = P + j 2 (Wm −We ) = P + jQ (W = J / s)
2

2 4 4 
where Wm = time average magnetic energy stored in the inductor (J) and We = time
average electric energy stored in the capacitor (J).

Electromotive force (Vemf )

We already know F = I dl  B for conduction current in conductors and
F = qu  B for convection current in space .

Lorentz force under the effect of E and B is given by

F = q(E + u  B) = q(E + Ei )( N ) .
N V
E i = u  B( = ) : Induced field determined by
Coul m
a moving (u) charge in a magnetic field ( B )

Electromotive force (induced voltage)

Vemf =  (u  B) dl (V )
l

The magnetic force experienced by a sliding conductor with length L in the direction of
the current i is

F m = iL  B = − F ext .
( F ext : External force)
In general, F m = i  dl  B = − F ext
l

B - Vemf +
y=a R Y=b y

ds
L

Moving
conductor
i dl
dx

u
x

b
Vemf =  uax  (− Baz ) dya y =  uax  (− Ba z ) dya y
l a
b
Vemf =  uBa y dya y = BLu (V )
a

Faraday’s law of induction

ds = Ldxaz
 ( )
d = B ds = − Baz Ldxaz = − BLdx(Wb)
d dx
Vemf = − = BL = BLu (Wb / sec = V )
dt dt
d
Vemf =− (Wb / s = V )
dt

Faraday’s law of induction (for N-turn coil)

(t)
i(t) Va
+
R Vab(V) N
_ (turns)
Vb
d
Vemf = − N (Wb / s = V )
dt

Define a flux linkage (쇄교자속)  = N (Wb  turns) .

d
Vemf = − (Wb / s = V )
dt
V
i (t ) = emf ( A)
R

Ex) When  (t ) = m sin  t , determine the electromotive force eab (V ) .

 (t ) = N (t )
d
eab (t ) = − = − Nm cos t = − Nm sin ( 90 − t ) = Nm sin (t − 90 ) (V )
dt
( slower than  (t ) by 90)
  (t)

m
 
wt

eab (t)
Nwm

wt

Maxwell’s equation (Faraday’s law)

d
Vemf = − (Wb / s = V )
dt
Vemf =  E dl (V )
l

d d
− = −  B ds(Wb / s = V )
dt dt S
d d
l E dl = − dt S B ds = − dtM (V ) (Integral form)
(폐회로 l 에 대한 E 의 회전량은 개면적 S 를 통과하는 총자속  M (Wb)의 시간에
따른 변화와 같다: 부호 조심!)

d
−  B ds(V ) ⎯⎯⎯ →0
Statics
 =0
dt S

  E dl = 0(V )
l

ds
S
dl
l
Left side =  E dl ⎯⎯⎯
→  ( E ) ds(V )
Stokes
l S
 B
Right side = −  ds(V )
t
S

 B 
S    E = − t  ds (V)
B
  E = − (V / m 2 ) (Differential form or point form)
t

Example) B = Bay (Wb / m2 ) ,  =  0 (rad / s ) , the length of the loop in z direction=2L

(− L  z  L) (m), and the radius of the conductor from z axis=R(m). Determine Vemf .

z
dl

l1

B
ds R
Length=2L
+
l2 y
Vemf
- dl
N turns

u  B =  Ra  Bay = − RB sin taz ( N / Coul = V / m)

Vemf (on l1 with N turns) = N l u  B dl

1
( )
( − RB sin ta ) dza
−L
= N z z = 2 NLRB sin  t (V )
L

Vemf (on l2 with N turns)= Vemf (on l1 with N turns)

 Total Vemf = 4NLRB sin t = NBA sin t (V )
(where A = 4LR is the area of the loop)

ii) B ds = Bds cos(t )(Wb)

 =  B ds = BA cos t (Wb)
S

d
Vemf (with N turns) = − N = NBA sin  t (V )
dt
Self inductance L

 N
Flux linkage  = N = Li  Inductance L = = (Wb / A = H )
i i

d d
vemf = =N (Wb / s = V )
dt dt
(t)
i(t)
+ +
Vemf N
_ V(t) (turns)
_

Ex) Determine the self inductance for L(m).

i(A) i(A)

r H A(m2)

N(turns) in l(m)
Ni
L
H dl = Hl = Ni  H =
l
( A / m)
Ni
B =  H = r 0 (Wb / m2 )
l
Ni
 = BA =  r 0 A(Wb)
l
N 2i
 = N =  r  0 A(Wb  turns )
l
Self inductance for l (m)
 N2A
L= = r 0 (Wb  turns / A = H )
i l
Self inductance for 1(m)
L N2A
L(per a unit length 1m) = =  r  0 2 ( H / m)
l l
Energy in a magnetic field

Magnetic energy density

1 1 1
wm = B H =  H H =  H 2 ( J / m3 )
2 2 2
i(A) i(A)

r H A(m2)

N(turns) in l(m)
Total magnetic energy stored in the inductor
1 1
Wm =  wm dv =   H 2 dv =  H 2 Al
V V 2 2
2
1  Ni  1 N2A 2 1 2
= 0 r   Al = 0  r i = Li ( J )
2  l  2 l 2

Maxwell’s equation from Ampere’s law

In magnetostatics, H
l
dl = I ( A) (Integral form)

ds
S
dl
l
Left side=  H dl =  ( H ) ds( A)
l S

Right side= I =  J ds( A)

S

 (   H = J ) ds( A)
S

  H = J ( A / m2 ) (Differential form or point form)

In a time-varying case,
 H
i(t) H +
+
-
-
+ -
+
D
_ V(t)

D
 H = J + ( A / m2 )
t
D
( A / m 2 ) : Displacement current density (Maxwell added this term)
t
 D 
  H = J + (A/ m )
3

 t 
 ( D )
0= J + ( A / m3 )
t

 J =− ( A / m3 ) (Current continuity law)
t

 D 
S 
   H = J +  ds
t 
Left side =  ( H ) ds ⎯⎯⎯
Stokes
→  H dl ( A)
S l


Right side =  J ds +  D ds ( A)
S t S
 
l H dl = S J ds + t S D ds = I + t E ( A)
(폐회로 l 에 대한 H 의 회전량은 개면적 S 를 통과하는 총전류 I 와 총전속
 E (Coul)의 시간에 따른 변화의 합과 같다)

Electric Gauss’ law

 D =  (Coul / m3 ) (point form)

For an integral form,
(
  D =  dv(Coul )
V
)
.
Left side =    Ddv =  D  ds(Coul )
V s

Right side = 
V
 dv = Q(Coul )
  S
D  ds = Q(Coul )
(폐면적 S 밖으로 나오는 총전속  E 는 부피 V 에 존재하는 총 전하량과 같다)
Magnetic Gauss’ law

 B = 0(Wb / m3 ) (point form)

For an integral form,
(
  B = 0 dv(Wb)
V
)
.
Left side =    Bdv =  B  ds(Wb)
V s

Right side = 
V
0dv = 0(Wb)
  S
B  ds = 0(Wb)
(폐면적 S 밖으로 나오는 총자속  M 는 항상 0(Wb)이다. B 는 항상 회전량만
가지며 발산은 0 이다)

Maxwell’s equations with fictitious magnetic charges (QM )

and magnetic currents ( I M )

Integral form (Directions of dl, ds, an , I , I M follow the right-hand rule.)

 
l E dl = − I M − t S B  ds = − I M − tM (V ) (Faraday)
 
l H dl = I + t S D  ds = I + t E ( A) (Ampere)
 D  ds = Q(Coul ) (Electric Gauss)
S

 S
B  ds = QM (Wb) (Magnetic Gauss)

How to convert the integral-form Ampere’s to a differential form

: Divide the equations with ds ( m 2 ) (small area for the closed path l )
an  H dl an I   D  ds
and multiply with an  l
= + an S

ds ds t ds
D
  H = J + ( A / m2 )
t

How to convert the integral-form Gauss’ to a differential form

: Divide the equations with d ( m3 ) (small volume for the closed surface S)

 S
D  ds
=
Q
  D =  (Coul / m3 )
d d

Differential form
 B
  E = −Jm − (V / m 2 ) (Faraday)
t
D
 H = J + ( A / m 2 ) (Ampere)
t
 D =  (Coul / m3 ) (Electric Gauss)
 B = m (Wb / m3 ) (Magnetic Gauss)

D =  E =  0 r E
B =  H = 0 r H


Current and charge relation (current continuity law) :  J = − ( A / m3 ) (point-form)
t
  
To obtain the volume-form, we do    J = − t  dv , then
V

Left =   Jdv =  J ds = I ( A)
V S

 Q
Right = −
t V
 dv = −
t
(Coul / s = A)

Q
 I =  J ds = − ( A = Coul / s ) (volume − form)
S t


The current continuity law  J = − ( A / m3 ) can also be derived by taking the
t
divergence of Ampere’s law:
 D    D ( ) 
:     H = J +   0 =  J +
t  t
  J = −
t
( A / m3 )


0 = 4 10−7 ( H / m)
0 = 8.854 10−12 (F / m)
In a lossy medium with a conductivity  ,
Conduction current J =  E( A / m2 )

E( N / Coul = V / m) :Electric field(전장,전계),

B(Wb / m2 = Tesla) :magnetic flux density(자속밀도),
J m (V / m2 ) : magnetic current density(자류밀도:실재하지는 않으나 가정할 경우
faraday 는 Ampere 법칙과 식의 형태가 유사해 나중에 Field Equivalence
Theorem 등에 사용함),
H ( A / m) :magnetic field(자장,자계),
D(Coul / m2 ) :electric flux density(전속밀도),
J ( A / m2 ) :electric current density(전류밀도),
 (Coul / m3 ) : Electric charge density(전하밀도),
m (Wb / m3 ) : Magnetic charge density(자하밀도:실재하지는 않으나 사용할 경우
Magnetic Gauss 는 Electric Gauss 법칙과 식의 형태가 유사해 나중에 Field
Equivalence Theorem 등에 사용함)

Boundary conditions: Simple Maxwell’s equations applied at a boundary

Region 1
an 
E1 D1 H1 B1

Region 2
E2 D2 H2 B2 

-BC for Ampere’s law

Region 1
an 
ap
dl = dl at H1
at
h→0
l
Region 2 dl = − dl at H2

  E
 H dl = t  D  ds + I = t + I ( A) ⎯⎯⎯⎯→  H dl = I ( A)
h →0( S →0)
l S  E →0 l

  H dl = H dla + H (−dla ) 1 t 2 t
l

( )
= H1 − H 2 dlat = I = J s dla p ( A)

Using at = a p  an ,
( H − H ) (a
1 2 p ) (
) ( H − H ) dl = a  ( H − H ) a dl
 an dl = a p  an 1 2 n 1 2 p

!!Remember A  B  C (Volume made by A, B, and C ) = C  A  B = B  C  A !!

= J s dla p ( A)
(
 an  H1 − H 2 = J s ( A / m) )
-BC for Faraday’s law (similar to Ampere’s)
(
an  E1 − E2 = − J ms (V / m) )
-BC for Electric Gauss’s law
 D  ds = D1 dsan + D2 (−dsan ) = s ds(Coul )
S

( )
an D1 − D2 ds =  s ds (Coul )

Region 1

an
ds D1

Region 2 D2

ds =dsan

(
 an D1 − D2 =  s (Coul / m 2 ) )
-BC for Magnetic Gauss’s law (similar to Electric Gauss’)
( )
an B1 − B2 = ms (Wb / m 2 )

In summary,
( )
an  E1 − E2 = − J ms (V / m) (Faraday)

a  ( H − H ) = J ( A / m) (Ampere)
n 1 2 s

a ( D − D ) =  (Coul / m ) (Electric Gauss)

n 1 2 s
2

a ( B − B ) =  (Wb / m ) (Magnetic Gauss)

n 1 2 ms
2

Special cases of above boundary conditions (t : tangential, n : normal)

a) Region 1,2 : Dielectric medium

  s = 0 , J s = 0 ,  ms = 0 , and J ms = 0
 ( )
an  E1 − E2 = 0 (Faraday) (scalar form: Et1 = Et 2 , E t continuous at the boundary)
: 1 Et1 = Dt1  Dt 2 =  2 Et 2
( )
an  H1 − H 2 = 0 (Ampere) (scalar form: H t1 = H t 2 , H t continuous at the boundary)
: 1 H t1 = Bt1  Bt 2 = 2 H t 2

( )
an D1 − D2 = 0 (E. Gauss) (scalar form: Dn1 = Dn 2 , Dn continuous at the boundary)
Dn1 Dn 2
: = En1  En 2 =
1 2

( )
an B1 − B2 = 0 (M. Gauss) (scalar form: Bn1 = Bn 2 , Bn continuous at the boundary)
Bn1 Bn 2
: = H n1  H n 2 =
1 2
Region1: 1 , 1
Bn1
Dn1 H t1
E t1 Bn 2
Ht 2
Dn 2
Et 2
Region 2 : 2 ,  2

b) Region 1 : Dielectric medium, Region 2 : PEC (Perfect Electric Conductor)

  ms = 0 , and J ms = 0 and all fields in Region 2 is zero.

an  E1 = 0(V / m) (Faraday)
an  H1 = J s ( A / m) (Ampere)
an D1 = s (Coul / m2 ) (Electric Gauss)
an B1 = 0(Wb / m2 ) (Magnetic Gauss)
** Et = 0(or an  E = 0) (important!)
 Region1: 1 , 1
Bn1 = 0
Dn1 =  s H t1 = J s
E t1 = 0   Bn 2 = 0
+++ s Js
Ht 2 = 0
Et 2 = 0
Dn 2 = 0
Region 2 : PEC ( Perfect ElectricConductor ) ( similar to  2 → )

c) Region 1 : Dielectric medium,

Region 2 : PMC (Perfect Magnetic Conductor)
  s = 0 , and J s = 0 and all fields in Region 2 is zero

an  E1 = − J ms (V / m) (Faraday)
an  H1 = 0( A / m) (Ampere)
an D1 = (Coul / m2 ) (Electric Gauss)
an B1 = ms (Wb / m2 ) (Magnetic Gauss)

** Ht = 0(or an  H = 0) (important!)
Region1: 1 , 1
Bn1 =  ms
Dn1 = 0 H t1 = 0  ms
+ + +B = 0
E t1 = J ms n2
Ht 2 = 0
• • •
J ms Dn 2 = 0
Et 2 = 0
Region 2 : PMC ( Perfect MagneticConductor ) (similar to 2 → )

Radiation in terms of Maxwell’s equations and wave equations in a source-free

medium
 E (V / m)
P(W / m 2 )
H ( A / m)

Source region Source-free region ( J = 0,  = 0)

 + +  E = −
B
= −
H
t t
J  D J =0 E
−−  H = J +
t
⎯⎯⎯ →
t

 H Phasor form
  E = − ⎯⎯⎯⎯→  E = − j H (Faraday)
t
 E Phasor form
 H =  ⎯⎯⎯⎯→  H = j E (Ampere)
t
 H =0
 E=0

   E = ( E ) −  2 E = −  (  H )
t
 E  E
2
= −  ( ) = −  2
t t t

where

2 E = ax2 Ex + ay2 Ey + az2 Ez

       
Laplacian operator 2 =   =  ax + a y + az   ax + a y + az 
 x y z   x y z 
2 2 2
= 2+ 2+ 2
x y z

Substituting  E = 0( source free region), then we obtain

( )
2 E r, t
2
( )
 E r , t − 
t 2
= 0 (Helmholtz wave equation, partial differential equation in
an instantaneous form)


( ) ()
Express this in a phasor form ( E r , t → E r ,
t
→ j )

() ()
 2 E r +  2  E r = 0 (Helmholtz wave equation, ordinary differential equation,
phasor form)
wavenumer k and propagation constant )

 (rad / s)  2 f 2
Wavenumber k = = =   = = (rad / m) is the most
 (m / s) 1/  f 
important parameter for electromagnetic waves,
where  (= 2 f ) is the angular frequency of the source to generate waves
and  is the velocity of the wave determined by a surrounding medium (  and  ).
1
1 1 0 0 3 108 m / s c
Note that  = = = = = .
 0 r  0 r r r r r r r
This means that the velocity of the wave is the fastest in free space and slows down as
r and  r become larger.

For TEM waves, the propagation constant  is the same as the wavenumber k.
 = k =   ( rad / m) and  2 =  2  ( rad / m ) . The Holmholtz wave equation
2

may be written as
() () () () ()
 2 E r +  2  E r = 0   2 E r +  2 E r = 0 or  2 E r + k 2 E r = 0 . ()
Examples about )
f = 1GHz = 10 (cyles / s),  r = r = 1
9

 3 108 m / s
 = = = 0.3(m)
f 109 (cyles / s)
 2 f 2 2 (rad ) 20
 = k = = = = = (rad / m)
 f  0.3(m) 3

Electrical length (Very important!!)

2
For a physical length l ( m ) , electrical length=  l (rad ) = (rad / m)  l (m) .

Example) For f = 3GHz(source) in freespace( r = 1, r = 1) ,
 3  108 m / s
= = = 0.1(m) = 10cm and
f 3  10 cycles / s
9

2 2 (rad )
= = = 20 (rad / m)
 0.1(m)
 l =  / 4 = 0.025m 2 (rad ) 2 (rad )
 l =  / 2(rad ) = = = 20 (rad / m)
 ( m) 0.1(m)
l =  / 2 = 0.05m
 l =  (rad )
l =  = 0.1m
 l = 2 (rad )

E0 E0e − j /2 E0e − j E0e − j 2

 = 0.1m

E0e − j 2
E0e − j E0

E0e − j /2

2 l
•• • Electrical length  l =  l = 2 (rad ) • • •
 
l =  / 8,  / 4,  / 2,  , 2 (m)
 l =  / 4,  / 2,  , 2 , 4 (rad )

Wave Propagations

1) 1D problems
Assume an infinite size surface current source at z=constant plane, say, at z=0, then it
will generate a plain wave propagating in z or –z direction.
Let’s assume that the direction of the electric field is in x direction.
For a planar TEM wave with an electric field Ex ( z , t )
propagating in z direction,
 2 Ex ( z , t )  2 Ex ( z, t )
−  = 0 (partial differential equation)
z 2 t 2


Express this in a phasor form ( Ex ( z , t ) → Ex ( z ) , → j )
t

d 2 Ex ( z )
2
+  2  Ex ( z ) = 0 (ordinary differential equation)
dz

  2 f 2
Using  = = =   = = (rad/m)
vp 1/  f 
(  is called the propagation constant)

d 2 Ex ( z )
 2  =  2 → 2
+  2 Ex ( z ) = 0
dz
Try Ex ( z) = Ce .
sz

( s 2 +  2 )Ce sz = 0  s =  j 
− j z j z − j z j z
E ( z) = C e +C e = E +e + E −e
x 1 2
− j z j z 
 E ( z ) =  E + e + E −e  ax
 

When inserting this to Faraday’s law in a phasor form, we obtain

j
  E = − j H  H =  E

j       + − j z j z 
=  ax + a y + az    E e + E −e  ax
  x y z   
j  − j z j z 
= az   E + e + E −e  ax
 z  
− j z j z 
a y  − j  E + e + j  E −e
j
= 
  
 
 E + e− j  z E −e j  z   + − j  z E −e j  z  Let  =  E + e − j  z E −e j  z 
= ay  −  = ay  E e −  ⎯⎯⎯⎯
→ ay  − 
  /   /          
     
   

E+ − j z E−
H y ( z) = e − e j z
 
where intrinsic impedance
E + e− j z E+  In free space 
 = + − j z = + = ⎯⎯⎯⎯⎯ → 0 = 120 = 377()
H e H  0
Transverse( )
plane
Infinite surface Ex
current

Js with radian (Longitudinal)

frequency w
Pz
Hy

Planar TEM
wave

TEM : Transverse Electromagnetic

When the electric and magnetic fields are all in the transverse plane (횡단면),
we call it a TEM (Transverse ElectroMagnetic) wave.

General solution for a plane wave propagating in an arbitrary direction

E
H
k  r = Constant
r = xax + ya y + zaz
k = kn (rad / m)
E
E P = E  H (W / m 2 )
H

n =

E
n E
H=
 H

Wave vector k = kx ax + k y ay + kz az = kn(rad / m)

Position vector r = xax + ya y + zaz (m)

A plane wave propagating with a wave vector k = k n is given by

(
− j kx x + k y y + kz z ) − jk y y − jk z z
E (r ) = E0e− j k r = E0e = E0e− jkx x e e (V / m) .

( )
E (r , t ) = Re  E0 e− j k r e jt  = E0 cos t − k  r = E0 cos t − (k x x + k y y + k z z ) 
 
(time-form or instantaneous form)
Note that k  r is constant on any transverse planes.

Then, the magnetic field is obtained by

n  E (r )  E+  
H (r ) =  = + =  ( A / m) .
  H  
If H (r ) is known, then the electric field is obtained by E(r ) =  H (r )  n
z
Wave Vector

k z az
k = k x ax + k y a y + k z az

k y ay y
k x ax

x
 (rad / s ) 2 f 2
wavenumber k =| k |= k x2 + k y2 + k z2 = = = (rad / m)
v(m / s ) f 
k x , k y , k z are the propagation constants in x, y, and z direction, respectively.

Prob) A wave vector in free space is given by

 3 1  2 f  3 1 
k = k0 n = k0  ax + a y  =  ax + a y  (rad / m) .
 2 2  c  2 2 
The electric field is oriented in the z directions with E0 .
Obtain the phasors E( x, y, z)(V / m) , H ( x, y, z)( A / m) , and complex power
density S ( x, y, z)(W / m2 ) .
 E

H
2 f  3 1 
k = k x ax + k y a y = k0 n =  ax + a y  (rad / m)
c  2 2 

E where c = 3 108 m / s
y
H

r
r = xax + ya y + zaz
k y ay E
k, P
H
k x ax
E
z
x
H E
r
H

Sol)
r = xax + ya y + zaz
E( x, y, z) = E(r ) = E0e− j k r ax .
 3
2 f 1  3 f f
Here, k  r =
c

2 2
(
ax + a y   xax + ya y + zaz =
c
x+) c
y = Constant
 
on the transverse plane. So, the electric fields at all points in space are given by
 3 f f 
− j  x+ y
c 
E ( x, y, z ) = E (r ) = E0e − j k r az = E0e  c
a z ( phasor ) (V / m) .

The time-form or instantaneous form is

j (  t − k r ) 
E ( x, y, z , t ) = E (r , t ) = az Re  E0e − j k r e jt  = a z Re  E0e
   
⎯⎯⎯⎯⎯⎯
if E0 is positive real
(
→ az E0 cos t − k  r a z )
  3 f  f 
= az E0 cos t −  x+ y   (time − form) (V / m)
  c c  
The magnetic field is given by
 0  3 f f 
n  E (r ) 
=
0  3 1  − j

x+ y
c 
H (r ) = ⎯⎯⎯→ 0 +   c
 a a y 
 E e az
 0  2
x
2 
0

 3 f f 
0  1 3  − j  x+ y
c 
= E0  ax −
c
a y  e ( A / m)
0  2 2 

1 *
S ( x, y , z ) = E ( x, y , z )  H ( x, y , z )
2
 3 f f   3 f f 
1 − j

x+ y
c  0  1 3  + j  x+ y
c 
= E0 e  az  E0  ax −
c c
a y  e
2 0  2 2 
 3 1  E
2 2
E0
=  ax + a y  = 0 n (W / m 2 )
 2 2  2
2 0 
0

2
E
P( x, y, z ) = Re  S ( x, y, z )  = Re  E ( x, y, z )  H ( x, y, z )  = 0 n (W / m 2 )
1 *

2   2

For a TEM plane wave propagating in z direction (most waves on a transmission line),
k x = k y = 0, and kz = k =  .
Ex) When an electric field is given by
− jk z − j z − jkz
E ( x, y, z ) = E0e ax = E0e ax = E0e ax (V / m) , obtain the complex power
density S ( x, y, z)(W / m2 ) . Besides, obtain the total power P (W) crossing across a
surface given by
z
a ( m)

P (W )

b( m)

Besides, obtain the instantaneous form E( x, y, z, t ) .

Sol)
 − j z − j z
n  E ( x, y , z ) az  E0 e ax E0e
H ( x, y , z ) = = = a y ( A / m)
  
+ j z 2
1 * 1 − j z E0 e E0
S ( x, y, z ) = E ( x, y, z )  H ( x, y, z ) = E0e ax  ay = az (W / m 2 )
2 2  2

(mostly purely real since  = is usually real)

2 2
E E
Total power P = 0 az (W / m2 ) abaz (m2 ) = 0 ab (W )
2 2
jt 
Instantaneous form E ( x, y, z , t ) = Re  E ( x, y, z )e 

− j  z jt 
= ax Re  E0 e e  = ax E0 cos (t −  z )
 
  z    z 
= ax E0 cos   t −   = ax E0 cos   t −   = ax E0 cos  ( t − t  ) 
   /     v 
z
Note that t  = (sec) is the time for the wave to arrive at z.
v

Prob) Consider a plane wave propagating in the z direction. Show that

− j z
− j z E0e
E ( x, y, z ) = E0e ax (V / m) and H ( x, y, z ) = a y ( A / m) satisfy phasor-
 /

form Maxwell’s equations, where  = =   (rad / m) .
1/ 
Sol) The proper Maxwell’s equations may be Faraday’s and Ampere’s.

Faraday’s
We will use a phasor integral form.
Differential form :  E = − j H (V / m2 )
Integral form :  E dl = − j 
l S
B  ds(V )
− j z
E0e − j z
B = H =  a y =  E0e a y (Wb / m2 )
 /

 E dl = E a  ( −ha ) + 0 + 0 + E e ( )
ax  hax = E0 h ( e − j l − 1)
− j l
Left = 0 x x 0
l
  E0e− j  z a y  dzdxa y = − j   E0e− j  z dzdx
h l h l
Right = − j  B  ds (V ) = − j   
S x =0 z =0 x =0 z =0
l
l − j z l − j z  1 − j z 
= − j h   E0e dx = − j h  E0  e dx = − j h  E0  e 
z =0 z =0
 − j  z =0
l
 − j z 
 = E0 h ( e − 1) (V )
1
= − j h  E0  e − j l

 − j   z =0

Im

1
Re
− l

e− j l − 1 e− j l

The circulation of the electric field becomes maximum when  l =  and minimum
when  l → 0 .
 l =
⎯⎯⎯ → 2 E0 h (V )
Note that E0 h ( e− j l − 1)  l =0
.
⎯⎯⎯ → 0(V )

2 l
l =  l = 2 (rad )
 
2 l =  / 8,  / 4,  / 2,  , 2 ( m)
l = l =  (rad )
  l =  / 4,  / 2,  , 2 , 4 (rad )

x
dl
ds = dzdxa y
Plane wave E
h dl dl
H
dl z
l
Prob) Show that Ampere’s law  H dl = j 
l S
D  ds ( A) is also satisfied for a properly
chosen closed path l and open surface S as you did above for Faraday’s.

2) 2D problems
For a line current with an infinite length, we may get a cylindrical TEM wave.
Direction of
Power flow ( )
Line current with
infinite length

E
E Transverse
H
H Plane ( )
E
E
H H

Top
view
Cylindrical TEM
wave

e − j
Fields Ez , H 

1
Power density P (W / m2 ) = EH  (The total power from the source crossing the

circles (see the above figure) is the same since the circumference of the cylinder
= 2  : Power conservation law)

For a cylindrical wave propagating in  direction,

Ez  In free space 
 =− = ⎯⎯⎯⎯⎯ → 0 = 120 = 377() (why negative sign?)
H  0

3) 3D problems
For a spherical wave propagating in r direction,
E  In free space 
=  = ⎯⎯⎯⎯⎯ → 0 = 120 = 377()
H  0
 Transverse ( )
plane z

H

E

Short current y
element
with radian
frequency w

x
Spherical
TEM wave

Poynting vector or power density vector P = E(V / m)  H ( A / m) always flows

outwardly from the source.

e− j r
Fields E , H 
r
1
Power density Pr (W / m2 ) = EH  ( The total power from the source is the same
r2
since the surface area of the sphere = 4 r 2 : Power conservation law)

Reflection and Transmission of fields at a boundary

(Power conservation : Pin=Pref+Ptransmitted )

Electric field wave

 T
Ex+1 1 Ex+2 2
1 = = 2 = =
H y+1 1 H y+2 2

E1− 2 − 1 E1+ + E1−

= = T = = 1+ 
E1+ 2 + 1 E1+
z=0
In region 1,
Ex1 ( z) = E1+e− j1z (forward travelling wave) + E1−e j1z (backward travelling wave)(V / m)
E1+ − j 1 z E1−
H y1 ( z ) = e − e j1z ( A / m)
1 1
In region 2,
Ex 2 ( z) = E2+e− j2 z (V / m)
E2+
H y 2 ( z) = e− j2 z ( A / m)
2

At z = 0 , the tangential electric and magnetic fields must be continuous

(check the BCs for the dielectric-dielectric interface).
E1+ + E1− = E2+ (1) ,
E1+ E1− E2+
− = (2)
1 1 2

E1+ : known,
E1− , E2+ : unknown (must be solved!)

E1+ E1− E1+ + E1−

(1) → (2) : − =
1 1 2
 2 ( E1+ − E1− ) = 1 ( E1+ + E1− )
 E1− ( 2 + 1 ) = E1+ ( 2 − 1 )
2 −  +
 E1− = E
2 + 1 1

2 −1 + 22
E2+ = E1+ + E1− = E1+ + E1 = E1+
2 + 1 2 +1

The reflection coefficient at z=0 is defined as

E1− 2 − 1
(0) = + =
E1 2 + 1
2 1 (2 → or 2 → ( PMC ), similar to open )
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯ →1
2 =31 1
⎯⎯⎯ →
2
2 =1 ( matched )
⎯⎯⎯⎯⎯ →0
1
 2 = 1 1
⎯⎯⎯ 3
→−
2
2 1 (2 →0 or  2 → ( PEC ), similar to short )
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯ →−1

and transmission coefficient at the load(z=0) is defined as

 E2+ E1+ + E1− 22
T (0) = + = = 1 + (0) =
E1 E1+
2 + 1
2 1 (2 → or 2 → ( PMC ), similar to open )
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯ → 2 (only at the boundary)
2 =31 3
⎯⎯⎯ →
2
2 =1 ( matched )
⎯⎯⎯⎯⎯→1
1
 2 = 1 1
⎯⎯⎯ 3
→
2
2 1 (2 →0; PEC , short )
⎯⎯⎯⎯⎯⎯⎯⎯ →0

Travelling or Standing Wave ratio (SWR )

Reflection coefficient at a position z : ( z )

Ex1 ( z ) = E1+ e− j 1z (forward travelling wave) + E1− e j 1z (backward travelling wave) (V / m)
E1− e j1z
Relection coefficient at z: ( z ) = + − j1z = (0)e j 2 1z
E1 e
 E − e j 1z 
Ex1 ( z ) = E1+ e− j 1z 1 + +1 − j 1z + − j 1 z
 = E1 e 1 + ( z )
 E1 e 

Magnitude Ex1 ( z ) = E1+ 1 + ( z )

Ex1 ( z ) max = E1+ (1 + ( z ) ) : occurs every  / 2
Ex1 ( z ) min = E1+ (1 − ( z ) ) : occurs every  / 2
Distance between the maximum and the minimum :  / 4

Standing Wave ratio (SWR ) :

Emax 1 + ( z )
1(perfect travelling wave)  SWR= =  (perfect standing wave)
Emin 1 − ( z )

If defining travelling wave ratio(TWR), it would be

E 1 − ( z )
0(perfect standing wave)  TWR= min =  1(perfect travelling wave)
Emax 1 + ( z )

PEC (Perfect Electric Conductor) and PMC(Perfect Electric Conductor)

E+  
= + = or
Intrinsic impedance H  
−j

In the PEC and PMC, E=0 and H=0.

At the PEC boundary, Etan=0 and  = 0 (similar to a  →  (or  → ) medium).

At the PMC boundary, Htan=0 and  =  (similar to a  →  medium).

1.7-8 Plane Wave Reflection and Transmission at Media Interface

Normal incidence

x
Ei

Pi
Region 1 Hi Region 2
( 1 ,  1 ) Et
(2 ,  2 )
Pt
Er
Ht

l Pr Hr
y
z
 2
1 = 1 , 1 =  11 2 = ,  2 =   2 2
1 2
z =0

In region 1,
Ex1 ( z) = Ei e− j1z + Er e j1z (V / m)
Ei Er
H y1 ( z ) = e− j1z − e j1z ( A / m)
1 1
In region 2,
Ex 2 ( z) = Et e− j2 z (V / m)
Et
H y 2 ( z) = e− j2 z ( A / m)
2
i : incident , r : reflected, t : transmitted

At z = 0 , the tangential electric and magnetic fields must be continuous

(check the BCs for the dielectric-dielectric interface).
Ex1 ( z = 0) = Ei + Er = Et (1) ,
Ei Er Et
H y1 ( z = 0) = − = (2)
1 1 2

Ei : known,
Er , Et : unknown

Solving equations (1) and (2), we obtain

 2 −1 22
Er = Ei , Et = E
2 + 1 2 + 1 i

The reflection coefficient at z=0 is defined as

Er 2 −1
(0) = =
Ei 2 + 1
2
2 1 (2 = →; PMC , 2 → , open )
2
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯ →1
2 =31 1
⎯⎯⎯ →
2
2 =1 ( matched )
⎯⎯⎯⎯⎯ →0
1
 2 = 1 1
⎯⎯⎯ 3
→−
2
2
2 1 (2 = →0; PEC ,  2 →, short )
2
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯ →−1
and transmission coefficient at the load(z=0) is defined as

E2+ 22
T (0) = = 1 + (0) =
E1+
2 + 1

In summary,
Er 2 − 1 E E + Er 22
(0) = = , T (0) = t = i = 1 + (0) =
Ei 2 + 1 Ei Ei 2 + 1

Oblique incidence

For normal incidence case, there exist only transverse fields. But, for oblique incidence
case, in addition to the transverse (횡) fields, there may be longitudinal (종) fields. The
transverse electric and magnetic fields are solely responsible for the power flow in the z
direction.

For oblique incidence with an incident angle 1 , the transmitted angle  2 can be
obtained using Snell’s law.
: The phase change at the boundary in medium 1 must be the same as the phase change
at the boundary in medium 2 ➔
k1 sin 1 = k2 sin 2  n1k0 sin 1 = n2 k0 sin 2 (rad / m)  n1 sin 1 = n2 sin 2 (Snell’s law), where
n =  r  r is the refractive index (굴절율) and k is the wavenumber (파수) (rad/m).

Based on the wave impedance

Etransverse  Etransverse 
Z=  H transverse = , Pz ( power in z direction) = Etransverse H transverse  which is newly
H transverse  Z 
defined only using the for the transverse fields responsible for the power transfer in the
z direction, we can write, for any kinds of incidence (TE, TM, TEM),
Ei,trans + Er ,trans = Et ,trans (3) ,
Ei ,trans Er ,trans Et ,trans
− (4) =
Z1 Z1 Z2
Equations (3) and (4) are of the same form as the above (1) and (2) .

So, we easily obtain

Er ,trans Z 2 − Z1 Et ,trans 2Z 2
= , = .
Ei ,trans Z 2 + Z1 Ei ,trans Z 2 + Z1

Horizontal, Perpendicular, s-polarized, TE case

Horizontal , Perdendicular ,
s − polarized , TE case
x

1 Hr 2

1 Er
Et
2
z
Ei
1 Ht
x, y : transverse(횡)
Hi z : longitudinal (종)

i :incident , r : reflected , t :transmitted

Ei = Ei ,trans ( verse ) H i = vector sum of H i ,trans ( verse) and H i ,longi (tudinal ) , and
Ei 1
= = 1 (intrinsic impedance)
Hi 1
Etransverse and H transverse are related with power flow in z (longitudinal) direction
(important!).
Etransverse and H longitudinal are related with power flow in x (transverse) direction.

Ei ,trans Ei  Et 2
Wave impedance Z1 = = = 1  1 , Z 2 = =  2
H i ,trans H i cos 1 cos 1 H t cos  2 cos  2
Er Er , trans Z 2 − Z1 2 / cos  2 − 1 / cos 1 2 cos 1 − 1 cos  2
Reflection coefficient (1 ) = = = = =
Ei Ei , trans Z 2 + Z1 2 / cos  2 + 1 / cos 1 2 cos 1 + 1 cos  2
Et Et , trans 2Z 2 22 / cos  2 22 cos 1
Transmission coefficient T (1 ) = = = = =
Ei Ei , trans Z 2 + Z1 2 / cos  2 + 1 / cos 1 2 cos 1 + 1 cos  2

Vertical, Parallel, p-polarized, TM case

 Vertical , Parallel ,
p − polarized , TM case
x

1 2
Er Et
1 Hr Ht 2
Ei z
1 Hi

Note that H i is already a transverse field, Ei ,trans = Ei cos1 (or Ei = Ei ,trans / cos 1 ), and
Et ,trans = Et cos 2 (or Et = Et ,trans / cos  2 ).

Ei ,trans Ei cos 1 E cos  2

Wave impedance Z1 = = = 1 cos 1  1 , Z 2 = t = 2 cos  2  2
H i ,trans Hi Ht
Er Er , trans / cos 1 Er , trans Z 2 − Z1 2 cos  2 − 1 cos 1
= = = = =
Ei Ei , trans / cos 1 Ei , trans Z 2 + Z1 2 cos  2 + 1 cos 1
Et Et , trans / cos  2 cos 1 Et , trans cos 1 2Z 2 cos 1 22 cos  2 22 cos 1
T= = = = = =
Ei Ei , trans / cos 1 cos  2 Ei , trans cos  2 Z 2 + Z1 cos  2 2 cos  2 + 1 cos 1 2 cos  2 + 1 cos 1

For the TE and TM cases with 1 = 0 (normal incidence), it can be seen that
2 − 1 22
(1 = 0) → and T (1 = 0) → (same as TEM case).
2 + 1 2 + 1

For a specific incident angle 1 (Brewster angle),  = 0 for a vertical polarization case :
Er Er , trans / cos 1 Er , trans Z 2 − Z1 2 cos  2 − 1 cos 1
= = = = =
Ei Ei , trans / cos 1 Ei , trans Z 2 + Z1 2 cos  2 + 1 cos 1

n1 n2
Using n1 sin 1 = n2 sin 2  sin 2 = sin 1  cos2 = 1 − sin 2 2 = 1 − 12 sin 2 1
n2 n2
 n12
 = 0   2 cos  2 − 1 cos 1 = 0   2 1 − sin 2 1 = 1 1 − sin 2 1
n22
0 1  0
 1 − sin 1  = 1 − sin 2 1 
2

2  2   1

 1 2 − 1 sin 1 =  2 −  22 sin 2 1

2 2 2

 (  2 + 1 )(  2 − 1 ) sin 2 1 =  2 ( 2 − 1 )
   1
 1 + 1  sin 2 1 = 1  sin 1 = sin b =
 2  
1+ 1
2
 
 
−1  1 
 Brewster angle b = sin (occurs for any combination of ε1 and ε 2 )
  
 1+ 1 
  2 


Conservation of power density

Poynting vector (power density) P(W / m2 ) = E(V / m)  H ( A / m)
1  + − j z  E1+ e− j z   | E1+ |2
*
1
Time average Pin = Re  Ex1 ( z ) H y1 ( z )  = Re  E1 e
*
  = (W / m2 )
2 2 
  1   21
| E − |2 | E + |2 |  |2
Reflected power density Pref = 1 = 1 (W / m2 )
21 21
| E2+ |2 | E1+ |2 | T |2
Transmitted power density Ptrans = = (W / m2 )
22 22

Conservation of power density: Pin = Pref + Ptrans (W / m2 )

12 + |  |2 + | T |2
 ( Pin when E1 = 1) = ( Pref when E1 = 1) + ( Ptrans when E1+ = 1)
21 21 22

Homework) Prove that the conservation of power density is satisfied for any  1 and
 2 !!!
E1− 2 − 1 22
Hint) Use  = = and T = 1 +  = .
E1 2 + 1
+
2 + 1

Standing Wave Ratio :

Ex1 ( z ) = E1+ e− j1z (forward travelling wave) + E1− e j 1z (backward travelling wave) (V / m)
 E − e j1z 
= E1+ e− j1z 1 + +1 − j1z + − j 1 z
 = E1 e 1 + ( z) = E1+e− j1z 1 + (0)e j 2 1z 
 E1 e 
 More considerations

Total field Ex1 ( z) = E1+e− j1z (travelling wave) + E1−e j1z (travelling wave)(V / m)

- At the PEC boundary (Perfect Electric Conductor) (Etan=0),

+ − − + E1−
Ex1 ( z = 0) = E1 + E1 = 0  E1 = − E1 or (0) = + = −1 (Ptrans=0, Pin=Pref)
E1
Im

E1+
Re
−
E 1

E1+ + E1− = 0 always

Ex1 ( z) = E1+e− j1z + E1−e j1z = E1+e− j1z + (−E1+ )e j1z

= E1+ (e− j1z − e j1z ) = − j 2E1+ sin(1z) (sine type perfect standing wave: zero electric
field at the boundary)

Ex1 ( z )

2 E1+
PEC
Etan = 0
 =0
 →
z
− / 2 − / 4

E1+ e− j1z E1− e j 1z E1+ e− j 1z (− E1+ )e j 1z

H y1 ( z ) = − = −
1 1 1 1
+ +
E 2E
= 1
(e− j1z + e j1z ) = 1
cos( 1 z ) (cosine type perfect standing wave: maximum
1 1
magnetic field at the boundary)
2E +
H y1 ( z = 0) = 1 ( A / m)
1
 2E + 
 J s ( z = 0) = an  H ( z = 0) = −az   1 a y 
 1 
+
2E
= 1 ax ( A / m) (Currents are induced on conductor surface
1
when the wave is incident on it : Think of the principle of receiver antenna made of conductor )

Power density crossing across the boundary P = E  H = 0 (W / m2 ) since E = 0 .

As a result, Pin=Pref.
At the PMC (Perfect Magnetic Conductor) boundary (Htan=0),
E+ E−
H y1 ( z = 0) = 1 − 1 = 0  E1− = E1+ or (0) = 1 (Ptrans=0, Pin=Pref)
1 1

Im

E1−
E1+
Re

E1+ E1−
− = 0 always at PMC
1 1

Ex1 ( z) = E1+e− j1z + E1−e j1z = E1+e− j1z + E1+e j1z

= E1+ (e− j1z + e j1z ) = 2E1+ cos(1z) (cosine type perfect standing wave: maximum
electric field at the boundary)

E1+ e− j1z E1−e j1z E1+ e− j1z E1+ e j 1z

H y1 ( z ) = − = −
1 1 1 1
E1+ j 2 E1+
= (e− j1z − e j1z ) = − sin( 1 z ) (sine type perfect standing wave: zero magnetic
1 1
field at the boundary)
H y1 ( z )
E1+
2
1
PMC
H tan = 0
 →
 →
z
− / 2 − / 4

Power density crossing across the boundary P = E  H = 0 (W / m2 ) since H = 0 .

As a result, Pin=Pref.

When  2 = 1 (matched or no reflection, E1− = 0 ) (Pref=0, Pin=Ptrans),

Ex1 ( z = 0) = E1+ + E1− = E1+ = E2+ , (0) = 0
Im

− E1+
E 1
Re

E1− = 0
Ex1 ( z) = E1+e− j1z + E1−e j1z = E1+e− j1z (perfect travelling wave)
Ex1 ( z )
E1+

z
1 = 2

E1+ e− j1z E1−e j1z E1+ e− j1z

H y1 ( z ) = − = (perfect travelling wave)
1 1 1
As a result, Pin=Ptrans.

Problem) For the case of r1 = 1,  r1 = 1, r 2 = 9,  r 2 = 1, and E1+ = 10(V / m) , analyze the
reflection and transmission phenomena in detail.
0  r1 0 0  r 2 9 0
Sol) 1 = = = 120 () , 2 = = = 3 120 ()
 0 r1 0  0 r 2 0
E1− 2 −1 3 − 1 1 1
(0) = = = = and E1− = E1+ = 5(V / m)
E1 2 + 1 3 + 1 2
+
2
1 3 E+ 3
E2+ = E1+ + E1− = E1+ + E1+ = E1+ = 15(V ) and T (0) = 2+ =
2 2 E1 2
E 1+ 
SWR = max = =3
Emin 1 − 
 Im
j  = 1, SWR = 

 = 1/ 2, SWR = 3

Re
−1 1
 = 0, SWR = 1
perfect travelling wave
−j

2
E1+ 102
Incident power density Pin = = 0.133(W / m2 )
21 2  377
2
E1− 52
Reflected power density Pref = = = 0.033(W / m2 )
21 2  377
2
E2+ 152
Reflected power density Pt = = 0.100(W / m2 ) = Pin − Pref
22 2  3  377

1 1
Velocity of electromagnetic waves : v1 = = = c = 3 108 m / s
1 2 0 r1 0 r1
1 1 c
v2 = = = = 108 m / s
2 2 0 r 2 0 r 2 3

Transmission line theory (for high frequency circuit)

-Kinds of typical transmission lines guiding high frequency waves.

IG

VG ZL

Parallel plate line (A plane wave can be propagated along the line.)

IG

VG ZL

Two-conductor line
 IG
ZL
VG

Coaxial line (You see this line in many places.)

IG

VG ZL

Microstrip line (Convenient to insert RF components on it)

• Definition of transmission line (T/L) characteristic impedance

Voltage wave V + e − j z V + (Volt ) V +
Z c ( or Z 0 ) = = = + = + ()
Current wave I + e − j z I ( A) I
: depends on the cross-sectional shapes of transmission lines

• Parallel-plate transmission lines : maybe the simplest transmission line. Easy to

analyze.

E P
Js W s +

H h E
VG E ZL
P H z
H J an , s
- Js
s
IG
y
Important!! : All of the physical quantities like V, I, E, H, Q, etc have the term
e− j z because they are wave propagation in the direction of z.

**********************************************************************
  (rad / s ) 2 f 2
The wave number k is defined by k = = = (rad / m) .
 (m / s) f 
The propagation constant  along a TEM transmission line is the same as the
wavenumber k .
2
 =k = (rad / m)

Example 1) f=1GHz, free space(  r = 1 , r = 1)
−9 c 3 108 m / s
Period T=1/f= 10 sec ,  = = = 0.3m , and the propagation constant
f 109 / s
2 2 (rad ) 20 10
= = = (rad / m) =  2 (rad / m)
 0.3(m) 3 3

Phasor representation of voltage wave travelling to the right

+ V+
I = ( Z 0 : characteristic impedance)
Z0
+ +
V+  (rad / m), Z 0 () V + e− j l
− −
l ( m)
 l (rad )

V+
I+ = ( Z 0 : characteristic impedance)
Z0
+ +
V + e j l  (rad / m), Z 0 () V+
− −
l ( m)
 l (rad )
l (m) : physical length
 l (rad ) : electrical length
*******************************************************************

For the parallel plate T/L given above, let’s assume that the electric field is given by
E( z) = E0e− j z ax (V / m) .
For a TEM wave propagating in the z direction,
− j z − j z
n  E ( z ) az  E ( z ) az  E0e ax E0e
H ( z) = = = = a y ( A / m)
   
 E  0  r 
where the intrinsic impedance  = = = = 0 r () and
H   0 r r
  2 f 2
propagation constant  = = = = (rad / m)
1/  v f 

Surface current density J s ( z)(on − bottom − surface) = an  H ( z)

( )
(This comes from an  H1 − H 2 = J s and H 2 = 0 outside )
E0e− j z E0e− j z
= ax  ay = a ( A / m)
  z

= − J s ( z)(on − top − surface)

h
Voltage wave V ( z ) =  E dl =  E ( z ) dxa x = E0 he − j  z (V )
l x =0

From Ampere’s law, the current wave

W
I ( z) =  H dl ⎯⎯⎯⎯⎯⎯⎯⎯⎯→  H dl = 
Magnetic fields exist only inside
H ( z ) dya y
l W y =0

W E0e− j z WE0e− j z
= a y dya y = ( A)
y =0  

Characteristic impedance of the transmission line,

− j z
V ( z ) E0 he h  h
Z0 = = − j z
= =g = ( )
I ( z ) WE0 e W  W

h
(g = may be called a geometric factor )
W
( Z 0 is related with the intrinsic impedance  and also with the cross-sectional shape
h
of this specific transmission line)
W

Surface charge density

(
 s ( z )(on − bottom − surface) = an D1 − D2 ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯ )
Electric fields exist only inside , D2 = 0
→ an D1 ( z ) = an D( z )
= ax  E( z) =  E0e− j z (Coul / m2 )
= −  s ( z )(on − top − surface)

Surface charge per unit line length

W W
Qs ( z ) =  s ( z )W = W  E0e− j z =  hE0e − j  z = V ( z )(Coul / m)
h h
Capacitance per unit line length
 Q( z ) W Coul / m
C= = = / g ( = F / m) (true for any T/L)
V ( z) h V

Magnetic flux density passing between the two lines

 E0e− j z
B( z ) =  H ( z ) = a y (Wb / m2 )

Magnetic flux per unit line length
 h WE0e− j z  h
 ( z ) = B( z ) ha y =  = I ( z )(Wb / m)
W  W
Inductance per unit line length
 ( z)  h Wb / m
L= = = g ( = H / m) (true for any T/L)
I ( z) W A

Note that LC =  (L, C : circuit parameters; ,  : field parameters)

1 1
= = (m / s) (true for any T/L)
 LC

V ( z) h  h
Characteristic impedance Z0 = = =g = ()
I ( z) W  W
 g L
or Z0 =  g = g= = () (true for any T/L)
 /g C

Total power along the line using the field approach

( )
2
1  E
Power density (or Poynting vector) along the line p = Re E  H = 0 az (W / m 2 )
2 2

2 2 2 2
E0 E Eh V
Total power P =  p  ds =  az  dxdyaz = 0 Wh(W ) = 0 = (W )
S S 2 2 2
h 2Z c
W

Total power along the line using the circuit approach

1  + + * 1  + V +   V
* + 2

P = Re V ( I )  = Re V  =
 2   Z c   2Z c
(same as the above result!)
2 
 

Coaxial transmission line

Prob) Consider a following infinite-length transmission line. For simplicity, the T/L has
been drawn up to L(m). The coaxial line is assumed to be connected to a voltage source
with frequency f (time-varying case). Then, the electromagnetic waves obtained for the
parallel plate T/L are similarly propagated along the line (in the z direction). So, all of
the physical quantities like V, I, E, H, and Q must have e− j z terms. For simplicity, they
Q
are omitted. If the electric charge per unit length is given by l =  (Coul / m) on the
L
inner (  = a ) and outer surface (  = b ), obtain the voltage V and capacitance per unit
length C(F/m). Then, obtain the inductance per unit length L(H/m).

L ( m)
− l (Coul / m)
lb
l
a l (Coul / m)
Q
l = (Coul / m)
b L r
E

Using Gauss’ law,

Q Q/L 
 S
D  ds = D  2 L = Q(Coul )  D =
2 L
= = l (Coul / m 2 )
2 2
D l
E = = (V / m)
 0 r 2 0 r 
Voltage difference between the two conductors
b l l l b
V =  E  dl =  a  d  a = ln  b
= ln (V )
2 0 r  2 0 r 2 0 r a
l a a

The capacitance per unit length C(F/m) is obtained as

l (Coul / m) l 2 0 r 2
C ( F / m) = = = =  0 r 
V (Volt ) l b b b
ln ln ln
2 0 r a a a
b
ln
Define the geometrical factor g = a
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯ 2
→  0 r / g
Since the coaxial line also propagates a TEM wave,

E E l  0 r l
H = = = = ( A / m)
 0 2 0 r  0 2 0 0 r 
 0 r
  = 2 l l
Current I (  = a) =  H  dl = 
l =0
2 0 0 r 
a   d a =
0 0 r
= − I (  = b) ( A)

The inductance per unit length L(H/m) is obtained as

b
ln
Magnetic flux per m l (Wb / m) Use   g
L ( H / m) = = ⎯⎯⎯→ 0  a (You must derivethis from definition !)
Current I ( A) 2

Characteristic impedance
l b b
ln
V 2 0 r a 0 ln a
Zc = = = =  g ( product of intrinsic impedance and geometrical factor )
I l  0 r 2
0 0 r

Representation of incident (forward) and reflected (backward) voltage waves

l1 l2

V + e jl1 V+ V + e − jl2
V − e − jl1 V− V − e jl2

V = V + e jl1 + V −e− jl1 V = V + +V − V = V + e − jl2 + V − e jl2

V + e j  l1 V − e − j  l1 V+ V− V + e − j  l2 V − e j  l2
I= − I= − I= −
Zc Zc Zc Zc Zc Zc

Consider a transmission line terminated with a load ZL.

Characteristic impedance of a transmission line

V + for parallel − plate T / L with W and h E+h  h
Zc = Z0 = ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ + = =g
I +
H W  W
( g is sometimes called a geometric factor )
 V + jl V − − jl
I ( z = −l ) = I (l ) = e − e
Z0 Z0
V+ V−
I (0) = −
Z0 Z0
V+ +
Z0 =
I+ V ( z = −l ) = V (l ) = V e + jl
+V e − − jl
ZL
=

=
2f 2
= (rad / m)
V (0) = V + + V −
v f  Z 0 -
l
Z=
z =−l 0
z

In phasor forms, V(z) and I(z) are given by

V ( z ) = V + e − j  z + V − e j  z or V ( z = −l ) = V (l ) = V + e j  l + V − e − j  l

V + − j z V − j z V + j l V − − j l
I ( z) = e − e or I ( z = −l ) = I (l ) = e − e
Z0 Z0 Z0 Z0
At z = 0 , V ( z = 0) = V = V + + V − (1) ,
V+ V− V
I ( z = 0) = I = − = (2)
Z0 Z0 Z L

V + : known,
V − , V : unknown

Using equations (1) and (2), we obtain

Z L − Z0 + 2Z L
V− = V ,V= V+
Z L + Z0 Z L + Z0

The reflection coefficient at the load (z=0)is defined as

V − Z L − Z0
(0) = + =
V Z L + Z0
Z L = , I = 0, open ; similar to2 = , H t = 0, PMC Vmax 1 + 1
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯ →1( SWR = = = , perfectlystanding)
Vmin 1 − 1
1
1+
Z L =3 Z 0 3 −1 1 V 2 = 3)
⎯⎯⎯→ = ( SWR = max =
3 +1 2 Vmin 1−
1
2
 Z L = Z0 ( matched ) Z0 − Z0 V 1+ 0
⎯⎯⎯⎯⎯ ⎯ → = 0(SWR = max = = 1, perfectly travelling)
Z0 + Z0 Vmin 1 − 0
1 1
1 − 1 1+ −
Z L = Z0 1 V 2
⎯⎯⎯→ 3 3 = − ( SWR = max = = 3)
1 2 V 1
+1 min
1− −
3 2
Z L =0,V =0, short ; similar to2 =0, Et =0, PEC V 1 + −1
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯ →−1( SWR = max = = , perfectlystanding)
Vmin 1 − −1
jZ 0 − Z 0 2Z 0e j135
Z L → jZ0 ( Inductive load ) j 90 1+ j
⎯⎯⎯⎯⎯⎯⎯→ = = e = j ( SWR = = , perfectlystanding)
jZ 0 + Z 0 2Z 0e j 45 1− j
Z L →− jZ0 ( Capacitive load ) − jZ 0 − Z 0 2Z 0e− j135 − j 90 1+ − j
⎯⎯⎯⎯⎯⎯⎯⎯ → = = e = − j ( SWR = = , perfectlystanding)
− jZ 0 + Z 0 2Z 0e− j 45 1− − j

and transmission coefficient at the load(z=0) is defined as

V 2Z L Z L Z0 ( Z L →;open )
T (0) = −
= = 1 + (0) ⎯⎯⎯⎯⎯⎯⎯ →2
V Z L + Z0
Z L =3 Z0 3
⎯⎯⎯ ⎯ →
2
Z L = Z0 ( matched )
⎯⎯⎯⎯⎯⎯ →1
1
Z = Z 1
⎯⎯⎯⎯ →
L 0
3
2
Z L = Z0 ( Z L →0; short )
⎯⎯⎯⎯⎯⎯⎯ →0

Prob) When V + =10 and Z 0 = 50 , obtain V (l ) and I (l ) , and plot | V (l ) | , phase[V (l )] ,

| I (l ) | , phase[ I (l )] as a function of  l in case of Z L = 50 , Z L =  (open),
and Z L = 0 (short).

Properties of (l )
V (l ) = V + e j  l + V − e − j  l
V − e− j l
Reflection coefficient (l ) = = (0)e − j 2  l
V + e j l
Depending on Z L , (0) is determined and (l ) turns clockwise as l increases as seen
below. (l ) becomes the same every half wavelength
Let’s assume that (0) is given as follows :

j
( l =  )
(0)
(l ) = (0)e − j 2 l

(  l =  / 4)
− (l ) = (0)e− j /2

-1 (l ) 1
( l = 3 / 4)

(  l =  / 2)
= (0)e − j = −(0)

-j


l = , ,...
See that (l ) ⎯⎯⎯⎯→(0) . 2
 l = ,2 ,...

Various expressions of V (l ) and I (l )

V ( z = −l ) = V (l ) = V + e j  l + V − e − j  l = V + e j  l + V + (0)e − j  l
= V + e j  l [1 + (0)e− j 2  l ] = V + e j  l [1 + (l )]
V + j  l V − − j  l V + j  l V + (0) − j  l
I ( z = −l ) = I (l ) = e − e = e − e
Z0 Z0 Z0 Z0
V + j l V + j l
= e [1 − (0)e− j 2  l ] = e [1 − (l )]
Z0 Z0

Powers

Input power
1  + V +   V
* + 2
1
Pin = Re V ( I )  = Re V 
+ + *
 = (W )
2 2   Z 0   2Z 0
 

Reflected power
2 2
V− V+ 
2

Pref = = (W )
2Z 0 2Z 0
Power consumed in the load
1  + V−  
*
− V
+
PL = Re VI  = Re (V + V ) 
1
− 
*

2 2   Z0 Z0  
 
V+ 2
V−
2

1 V − (V + )* − (V − )*V + 
= Re  − +
2  Z0 Z0 Z0 
 
− + * − * +
(Note that V (V ) − (V ) V is purely imaginary.)

2 2 2 2
V+ V− V+ V+ 
2

= − = − = Pin − Pref
2Z 0 2Z 0 2Z 0 2Z 0

Power conservation law:

Pin − Pref = PL or Pin = Pref + PL

Investigation of the magnitude of V (l )

Absolute value of voltage along the line:

V (l ) = V + 1 + (0)e− j 2  l = V + 1 + (l )

(0) (l )
1 + (l )

-1 − (l ) (l ) 1

-j

⎯⎯ ⎯
max
→ V + (1 + (l ) ) = Vmax
⎯⎯→
min
V + (1 − (l ) ) = Vmin
Note that the distance between Vmax and Vmin is a quarter wavelength (  / 4 ).
Observe that as (l ) → 0 (perfect travelling wave), Vmin  Vmax and
as (l ) → 1 (perfect standing wave), Vmin → 0 and Vmax → 2 V + .

Voltage Standing Wave ratio

Vmax V (1 + (l ) ) 1 + (l )
+

1  VSWR = = = 
Vmin V + (1 − (l ) ) 1 − (l )
1(perfect travelling wave)  VSWR  (perfect standing wave)

(l ) may be estimated from the measured VSWR using

VSWR − 1
(l ) = .
VSWR + 1

Similar and different things for the following three cases

Electric field wave Voltage wave in T/L

 T
Ex+1 1 Ex+2 2 Zc  T
1 = = 2 = = Z1  T Z2
H y+1 1 H y+2 2 ZL
V+
Case 1 Case 2 Zc = =g Case 3
I+
g : geometrical factor

Case 1
At the boundary of case 1, we may write:
E1+ + E1− = E2+ (1) , E continuous
E1+ E1− E2+
− = (2) , H continuous
1 1 2
Solving equations (1) and (2) gives
 − 22
E1− = 2 1 E1+ and E2+ = E1+
2 + 1 2 + 1
E1− 2 −1 E2+ E1+ + E1− 22
= = and T = = = 1+  =
E1 2 + 1
+
E1+
E1 +
2 + 1
Power density conservation : Pin=Pref+Ptrans

Case 2
At the boundary of case 2, we may write:
V1+ + V1− = V2+ (1) , V continuous
V1+ V1− V2+
− = (2) , I continuous
Z1 Z1 Z 2
Solving equations (1) and (2) gives
Z − Z1 + 2Z2
V1− = 2 V1 and V2+ = V1+
Z2 + Z1 Z2 + Z1
V1− Z 2 − Z1 V2+ V1+ + V1− 2Z 2
= + = and T = + = +
= 1+  =
V1 Z 2 + Z1 V1 V1 Z 2 + Z1
Power density conservation : Pin=Pref+Ptrans

Case 3
At the boundary of case 3, we may write:
V1+ + V1− = VL (1) , V continuous
V1+ V1− VL
− = (2) , I continuous
Zc Zc Z L
Solving equations (1) and (2) gives
Z − Zc + 2Z L
V1− = L V1 and VL = V1+
Z L + Zc Z L + Zc
V1− Z L − Zc VL V1+ + V1− 2Z L
= +
= and T = +
= +
= 1+  =
V1 Z L + Zc V1 V1 Z L + Zc
Power conservation : Pin=Pref+PL

We see that the three cases are mathematically the same problems.

Input impedance (Generalized impedance defined at a specific position)

I (l ) l

+
V (l ) Z0 ZL
−

Z in (l ) Z in (l = 0) = Z L
  V− 
V + e j l 1 + + e− j 2  l 
V (l ) V + e j l + V − e − j  l  V 
Z in ( z = −l ) = Z in (l ) = = = +
I (l ) V + j  l V − − j  l V j l  V − j 2  l 
−
e − e e 1 − + e 
Z0 Z0 Z0  V 
( Note : Z in becomes Z 0 when V − = 0 becauseof the same definition.)
Z L − Z0 − j 2l
1+ e
= Z0
1 + (0)e − j 2l
= Z0
1 + (l )
= Z0
Z L + Z0
= Z0
( Z L + Z 0 ) e j l + ( Z L − Z 0 ) e− j l
1 − (0)e − j 2  l 1 − (l ) 1−
Z L − Z0 − j 2 l
e ( Z L + Z 0 ) e j l − ( Z L − Z 0 ) e− j l
Z L + Z0
2 Z L cos(  l ) + j 2 Z 0 sin(  l ) Z cos(  l ) + jZ 0 sin(  l )
= Z0 = Z0 L
j 2Z L sin(  l ) + 2Z 0 cos(  l ) Z 0 cos(  l ) + jZ L sin(  l )

Z L + jZ0 tan( l )
= Z0
Z0 + jZ L tan( l )
Note : Zin ( l ) is a periodic function since tan( l ) is a periodic function with a period
of  l =  (rad ) (or l =  / 2(m) ) as you see below.

tan(  l )

0  2 l

V (l ) Z + jZ 0 tan(  l )
Z in (l ) = = Z0 L
I (l ) Z 0 + jZ L tan(  l )
 l → 0, ,2 ,3 ,....
⎯⎯⎯⎯⎯⎯
l = 0,  /2,  , 1.5  ,..
→ Z L ( Z in is a periodic funcrion with a period of l =  / 2 or  l =  )

2
l→Z
⎯⎯⎯  →
2 0
(Load impedance Z L is inverted: small Z L → large Z in and large
l= ZL
4
Z L → small Z in ; short → open and open → short; )

Note : Every quarter wavelength, the input impedance is inverted !!!!

 ZL Z0 ( ZL →;open )
⎯⎯⎯⎯⎯⎯⎯ →− jZ0 cot( l )(when  l   / 2, purely capacitive )
 1 
This is like : − j
C 
ZL =Z0 ( matched )
⎯⎯⎯⎯⎯⎯ → Z0
ZL Z0 ( ZL →0;short )
⎯⎯⎯⎯⎯⎯⎯ → jZ0 tan( l )(when  l   / 2, purelyinductive )
Thisislike: jL
Relation between Z in (l ) and (l )

V (l ) V + e j l [1 + (l )] 1 + (l ) Z (l ) 1 + (l )

Zin ( z = −l ) = Zin (l ) = = + = Z0  in = zin (l ) =
I (l ) V j l 1 − (l ) Z0 1 − (l )
e [1 − (l )]
Z0
V + j l
e [1 − (l )]
I (l ) Z 0 1 − (l ) Y (l ) 1 − (l ) 1
Yin (l ) = = + j l = Y0  in = yin (l ) = =
V (l ) V e [1 + (l )] 1 + (l ) Y0 1 + (l ) zin (l )
Convenient to define a normalized input impedance
Z (l ) 1 + (l )
zin (l ) = in =
Z0 1 − (l )
Z (l ) z (l ) −1
(l ) = in = in
Z0 zin (l ) + 1

1+ 
zin (  ) = is a function of  (complex number) (Similarly, zin () is the complex
1− 
function of  ).

The domain of  is   1 on the complex plane. This is not like a real function f ( x) ,
for example, f ( x) = 3x + 2, x2 + 5 , where x is a real number. You can easily plot f ( x) . But,
1+ 
you may have to plot zin (  ) = as seen below.
1− 
 1+  
zin () =
Im 1−  Im
j j
zin

Re Re
−1 0 1 1

−j −j

Looks terrible!!!

⎯⎯
→ zin = r + jx
 ⎯
⎯
Smith chart (made by Smith in 1939) : on one chart.

 = j  zin = 0 + j

x=1

  0.45e j 63.5  zin = 1 + j

r=0
r=1
 = −1  zin = 0 + j 0  = 0  zin = 1 + j 0  = 1  zin = 
x=0

  0.45e− j 63.5  zin = 1 − j

x=-1

 = − j  zin = 0 − j
Smith chart (Complex number  mapped to complex number zin = r + jx on one chart)
   
Prob) When Z 0 = 50() , obtain (0) , (l = ) , (l = ) , (l = ) , (l =  ) , Z in (0)
8 4 2
  
Z in (l = ) , Z in (l = ) , and Zin (l =  ) in case of Z L =  (open),
) , Z in (l =
8 4 2
Z L = 150() , Z L = 50() , Z L = 50 / 3() , and Z L = 0 (short), respectively, by hand
calculation and using the Smith chart. Besides, show that the power conservation law
(Pin=Pref+PL ) is satisfied assuming V + = 10(V ) .

Prob) When V + = 20(V ) and Z 0 = 50() , obtain (0) , V − , V = V + + V − , VSWR, Pin ,

Pref , and PL , in case of Z L =  (open), Z L = 150() , Z L = 50() , Z L = 50 / 3() ,
and Z L = 0 (short). Check if the power conservation law (Pin=Pref+PL ) is satisfied.

Vector space and function space

A linear combination of vectors a1 , a2 , a3 means c1 a1 + c2 a2 + c3 a3 .

- linearly dependent basis vector set : (ax + ay , 2(ax + ay ), az ) :

not complete in 3D space (Completeness means that any arbitrary vector can be
represented.)

- linearly independent basis vector set : complete in 3D space

(ax + ay , ax − 3ay , ax + 2ay + 3az ) : not orthogonal, not normalized
(ax + ay , ax − ay , 2az ) , (ax , 2a y ,3az ) : Orthogonal, but not normalized, still not
convenient
(ax , ay , az ) : Orthogonal and normalized (orthonormal), very convenient

- Dot product of A and B

A B = AB cos

- Othogonality(직교성) means

ax ay = 0
ay az = 0
az ax = 0

- Normalization means
 2
ax ax = ax = 1  ax = 1
2
ay ay = ay = 1  ay = 1
2
az az = az = 1  az = 1

A linear combination of linearly independent basis vectors can represent any vector in
space. If we use orthonormal basis vectors, it is also very convenient.

A = c1 ax + c2 ay + c3 az
z

c3 = A  az

A
c2 = A  a y
y
c1 = A  a x

x
A ax = c1 (projection of A on ax )
A ay = c2 (projection of A on a y ),
A az = c3 (projection of A on az )

The function space is similar to the vector space.

The function f (t ) can be represented by a linear combination of its basis functions.


f = c11 + c22 + c33 + c44 + ... =  cnn
n =1

Especially, the periodic function f (t ) with a period T can be represented by a linear

combination of cos and sin harmonics.

- Orthogonal basis function set:

(1,cos 0t,cos 20t,cos30t,....) : not complete, can only represent even functions
(1,sin 0t,sin 20t,sin 30t,....) : not complete, can only represent odd function
 (1,cos0t,cos 20t,cos30t,...,sin 0t,sin 20t,sin 30t,...) : complete(can represent
any functions), but not
normalized
2
where the fundamental radian frequency is given by 0 = (rad / s)
T

The dot product of f (t ) and g (t ) :

T
f (t ) g (t ) = f (t ), g (t ) =  f (t ) g (t )dt
0

⎯⎯⎯f ,g
similar
→ large positive value
⎯⎯⎯⎯ f ,g
orthogonal
→0
⎯⎯⎯ f ,g
⎯
different
→ large negative value

f g 0 f g =0 f g 0
A B  0 A B = 0 A B  0

Different 0 Similar
Orthogonal
 A A B=0
A
A B0
B
f (t ) f (t ) B

t t
T T
g (t ) g (t )

t t
T T

f (t ) g (t ) f (t ) g (t )

+ + + + +
t t
T - - T
T
f g = f , g =  f (t ) g (t ) dt  0 T
0 f g = f , g =  f (t )g(t )dt = 0
0
Similar vectors and similar functions Orthogonal vectors and functions

A A B0
B
f (t )

t
T
g (t )

t
T

f (t ) g (t )

t
T
- - -
T
f g = f , g =  f (t)g(t)dt  0
0

Different vectors and functions

- Orthonormal basis function set:
1 2 2 2 2 2 2
, cos 0t , cos 20t , cos30t ,..., sin 0t , sin 20t, sin 30t
T T T T T T T

- Orthogonality means

1 2 1 2 T 2
cos 0t = , cos 0t =  cos 0tdt = 0
T T T T 0 T

2 2 2 2 2 T
cos  0t cos 2 0t = cos  0t , cos 2 0t =  cos  0t cos 2 0tdt = 0
T T T T T 0
2 2 2 2 2 T
cos  0t sin  0t = cos  0t , sin  0t =  cos  0t sin  0tdt = 0
T T T T T 0
2 2 2 2 2 T
sin  0t sin 2 0t = sin  0t , sin 2 0t =  sin  0t sin 2 0tdt = 0
T T T T T 0

- Magnitude of a vector and magnitude of a function : Same definitions

Magnitude of a vector A : A = A  A
T
Magnitude of a function f : f = f  f =  f, f  = 0
f 2 dt

- Normalization means

2
1 1 1 1 1 T 1
= , = = dt = 1
T T T T T 0 T

2
2 2 2 2 2
cos 0t cos 0t = cos 0t , cos 0t = cos 0t
T T T T T
2 T 2 T 1 + cos 20t
=
T 0
cos 2 0tdt =
T 0 2
dt = 1

2
2 2 2 2 2
cos 20t cos 20t = cos 20t , cos 20t = cos 20t
T T T T T
2 T 2 T 1 + cos 40t
T 0 
= cos 2 20tdt = dt = 1
T 0 2
 2
2 2 2 2 2
sin 0t sin 0t = sin 0t , sin 0t = sin 0t
T T T T T
2 T 2 T 1 − cos 20t
=
T 0
cos 2 0tdt =
T 0 2
dt = 1
.
1 2 2
 = 1, cos n 0t = 1 , sin n 0t = 1
T T T

In summary, the orthonormal basis functions have the property that

T
i (t )  j (t ) = i (t ),  j (t ) =  i (t ) j (t )dt = 0 for i  j
0

= 1 for i = j

A linear combination of linearly independent basis functions can represent any periodic
functions. . If we use orthonormal basis functions, it is very convenient.

1 2 2 2 2
f (t ) = a0 + a1 cos 0t + a2 cos 20t + ... + b1 sin 0t + b2 sin 20t + ...
T T T T T

b2 f (t )
2 1
sin 2 0t
b1 T T
a0
2
sin  0t 2
T cos  0t
T
a1
 1 1
f (t ) = a0 ( projection of f (t ) on )
T T
2 2
f (t ) cos 0t = a1 ( projection of f (t ) on cos 0t )
T T
2 2
f (t ) cos 20t = a2 ( projection of f (t ) on cos 20t )
T T

2 2
f (t ) sin 0t = b1 ( projection of f (t ) on sin 0t )
T T
2 2
f (t ) sin 20t = b2 ( projection of f (t ) on sin 20t )
T T

1  2 
2
In summary, f (t ) = a0 +  an cos n0t +  bn sin n0t
T n=1 T n =1 T
(called as Fourier series)

where

1 1
a0 = f (t ) ( projection of f (t ) on )
T T
2 2
an = f (t ) cos n0t ( projection of f (t ) on cos n0t )
T T
2 2
bn = f (t ) sin n0t ( projection of f (t ) on sin n0t )
T T
( a n , bn : all real number)

Instead of sinusoidal basis functions which consist of cos and sin, we may use the
exponential orthonormal basis function set
1 − j 20t 1 − j0t 1 − j 0t 1 j0t 1 j 20t
(..., e , e , e , e , e ,...) .
T T T T T

The complex functions 1 (t ) and 2 (t ) are orthogonal if

T
0
1 (t )2* (t ) dt = 0 .
1 − j0t 1 j0t
For example, e and e are orthogonal since
T T
 *
T 1 − j0t  1 j0t  1 T 1 T
 T
e  e  dt =  e− j 20t dt =  (cos 20t − j sin 20t )dt = 0
 T  T 0 T 0
0

The exponential orthonormal basis function set given above is the normalized one since

2 *
1 jn0t T 1 jn0t  1 jn0t 
e = e  e  dt
T 0 T  T 
1 T j 0t 1 jn0t
T 0
= e dt = 1  e = 1.
T

Now, we can conveniently represent any periodic function with

1 − j 20t 1 − j0t 1 − j 0t 1 j0t 1 j 20t

f (t ) = ... + c−2 e + c−1 e + c0 e + c1 e + c2 e + ...
T T T T T

The price we have to pay for using just one kind of basis functions is that the radian
frequencies must be extended to negative ones to form complex conjugate pairs for the
representation of a real function f (t ) .
1 − j 20t 1 j 20t
Note that the pair, for example, c−2 e and c2 e must be a complex
T T
conjugate pair.  c−2 = (c2 )* to represent a real function.

1 − j 20t
To determine c 2 , we do e operations on both sides of the equation.
T
1 − j 20t 1 j 20t 1 − j 20t T 1
f (t ) e = ...0 + c2 e e + 0... = c2  dt = c2
T T T 0 T

1 − jn0t 1 − jn0t T 1
 cn = f (t ) e = f (t ), e =  f (t ) e− jn0t dt (This should be
T T 0 T
1 jn0t
understood as a projection of f (t ) on the orthonormal basis function e . Note the
T
similarity that the projection of A on the direction of x is obtained by A  ax .

Example A) Periodic function with a period T

 f (t ): periodic function with period T


2
Fundamental angular frequency 0 = 2 f0 = (rad / s)
T
1 − j 20t 1 − j0t 1 − j 0t 1 j0t 1 j 20t
Orthonormal basis function set (..., e , e , e , e , e ,...)
T T T T T
1 − j 20t 1 − j0t 1 − j 0t 1 j0t 1 j 20t 1 jn0t
f (t ) = ... + c−2 e + c−1 e + c0 e + c1 e + c2 e + ... + cn e + ...
T T T T T T

1 jn0t
= c
n =−
n
T
e

T 
1 − jn0t 1 − jn0t 1 1
cn = f (t ) e = f (t ), e =  2T f (t ) e − jn0t dt =  2 A e − jn0t dt
T T − T − T
2 2

 
  sin(n0 ) sin(n0 )
1 1  1 2 = A 1
=  2 A
−
 cos(n0t ) − j sin(n0t )  dt = 2  2 A cos(n0t )dt = 2 A
 
2
2 T 0 T 2 T n0 T n
0
2 2

 
sin(n0 ) sin(n0 )
 
1 jn0t 
1 1 jn0t 
A
f (t ) = c n n = c n
T
e =  A
T n 
2
T
e = 
n =− T

2 e jn0t
n =− n =− n =−
0 n0
2 2

Discrete spectrum
 cn
A 
sin(n )
T A T
T 
n
T

n0
−40 −30 −20 −0 0 0 20 30 40

2 T →
Example B) For a non-periodic function with infinite T ( 0 =
⎯⎯⎯ → 0 ) as you see
T
below, we must have a continuous spectrum since the discrete (see
above) 0 , 20 ,30 ,.. are placed infinitely close.

f (t ) :Non-periodic single-pulse function with

period T →  and pulse width 
f (t )
T →

−
   t
2 2

Since there exists a continuous spectrum in a non-periodic signal, we need followings.

cn → C ( )d
n0 →  and e jn0t → e jt

1 
Then, f (t ) =
2 
−
C ( )d e jt

C ( ) =  f (t )e − jt dt (called a Forier transform or continuous spectrum of a signal )
−

For this non-periodic pulse with width  and height A,

   /2  /2 sin(t )  / 2
C ( ) =  f (t )e − jt dt =  Ae − jt dt = 2 A cos(t )dt = 2 A
− − /2 0  0
     
sin   sin  
= 2A 
2 
= A  2  = A  sinc   
  
  2 
2

Continuous frequency spectrum C ( ) or F ( )

: sometimes called a Forier Transform

A
C ( )

6 4 2 2 4 6

− − −
     