ELEC–E4130 Electromagnetic fields, Autumn 2020

Time-varying fields and Maxwell’s equations

Henrik Wallén
September 21, 2020
Time-varying fields and Maxwell’s equations
Outline for lecture weeks 2–3

Monday, 14 Sep Monday, 21 Sep

ñ The electromagnetic model [1, 7–1] ñ Boundary conditions [7–5]
ñ Potentials [7–4]

Thursday, 17 Sep Thursday, 24 Sep

ñ Time-varying EM fields [7–2, 7–3] ñ Wave equations [7–6]
ñ Time-harmonic fields [7–7]

The contents cover Chapter 7 in the textbook D.K. Cheng, Field and Wave
Electromagnetics, 2nd Ed., with some introductory parts from earlier chapters.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
2 (59)
The electromagnetic model

Cheng, Ch. 1 & 7—1 (+ some bits from Ch. 2 to 6)

The electromagnetic model
Electric charges are the fundamental sources of electromagnetic fields and
forces.

I
+q
q
E = aR µ0 I
4π ε0 R 2 B = aφ
2π r

A charge q at rest or in motion Moving charges produce a current I

produces an electric field. that gives rise to a magnetic field.

The fields are defined using the (physically measurable) electromagnetic force

F = q0 E + q0 u × B (Lorentz’s force equation)

on a small test charge q0 moving with velocity u.

A time-varying electric field give rise to a magnetic field and vice versa
⇒ electromagnetic radiation
Fundamental equations

Maxwell’s equations Lorentz’s force

∂B F=q E+u×B
∇×E=− (Faraday’s law)
∂t
∂D Constitutive relations
∇×H=J+ (Ampère’s law)
∂t
D = ε E, B = µH
∇·D=ρ (Gauss’ law)

∇·B=0 . . . depend on the media and can be

more complicated

Following the textbook’s deductive approach, we take these fundamental

relations as postulates. All results of classical electromagnetics can be derived
starting from these equations.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
5 (59)
In-class exercise 2.1
Derive the electric field of a point charge q in free space
q
E= aR
4π ε0 R 2
starting from Gauss’ law ∇ · D = ρ and the constitutive relation D = ε0 E.
Hint: Deduce the general form of E due to spherical symmetry and apply the
divergence theorem.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
6 (59)
Electromagnetic field quantities

E = Electric field intensity [E] = V/m

D = Electric flux density (displacement) [D] = C/m2 = As/m2

B = Magnetic flux density [B] = T = Wb/m2 = Vs/m2

H = Magnetic field intensity [H] = A/m

In simple media D = ε E and B = µ H. Why bother using four field quantities?

Flux density D and B Field intensity E or H

ñ Surface integral gives the total flux ñ Line integral gives a meaningful
through a surface quantity such as voltage or emf
ñ Divergence of a flux density reveals ñ Curl of a field intensity is well
the sources of the flux lines defined
ñ Divergence theorem ñ Stokes’ theorem

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
7 (59)
Electromagnetic source quantities
Charges In Maxwell’s equations,
these are free charges and
ρ = volume charge density [ρ] = C/m3
free currents.
ρs = surface charge density [ρs ] = C/m2
(Bound charges and
q = point charge [q] = C bound currents are
Q = total charge (in a volume) [Q] = C included in ε and µ.)

Currents J can include both

conduction current
J = current density [J] = A/m2
Js = surface current density [Js ] = A/m J=σE
I = current [I] = A
and convection current
due to other moving
Note that J is a vector (∼flux density) and I is a
charges.
scalar.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
8 (59)
Constitutive relations
In simple conducting media

D = ε E = ε0 εr E, ε = permittivity, εr = relative permittivity ≥ 1

B = µ H = µ0 µr H, µ = permeability, µr = relative permeability ≥ 1
A
J = σ E, σ = conductivity, [σ ] = S/m = Vm

This is a simple macroscopic model for the microscopic EM response of matter.

Beware that tabulated material parameters typically are valid for limited frequency
ranges and temperatures. . .
1 m
Speed of light in vacuum c=√ = 299 792 458
ε0 µ0 s
Vs H
Permeability of free space µ0 = 4π × 10−7 ≈ 1.257 × 10−6
Am m
1 −12 As F
 
Permittivity of free space ε0 = ≈ 8.854 × 10 =
µ0 c 2 Vm m

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
9 (59)
SI units in electromagnetics

Charge and electric flux C = coulomb = As In EM it is convenient

Magnetic flux Wb = weber = Vs to expand
SI units in
V, A, m, s although
Magnetic flux density T = tesla = Wb/m2 = Vs/m2
volt is not one of the
Capacitance F = farad = C/V = As/V fundamental SI units:
Inductance H = henry = Wb/A = Vs/A kg m2
V=
Resistance Ω = ohm = V/A A s3
Conductance S = siemens = 1/Ω = A/V

Power W = watt = VA
Work J = joule = Ws = VAs
Force N = newton = J/m = VAs/m

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
10 (59)
Conservation of charge
The principle of conservation of charge is one very fundamental physical
postulate. For an arbitrary volume V and its surface S, we have

dQ d
I Z
I= J · ds = − =− ρ dv,
S dt dt V

where the direction of ds (and the net current I) is outward.

In-class exercise 2.2: Show that electromagnetic fields always satisfy the above
condition if the fields satisfy Maxwell’s equations.

Hint: As an intermediate step, show that Maxwell’s equations imply the equation
of continuity
∂ρ
∇·J=−
∂t

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
11 (59)
Static fields
If the sources are time invariant, the time derivatives in Maxwell’s equations are
zero and the electromagnetic coupling disappears:
Maxwell’s equations
Electrostatics Electric potential,
∂B ∂
∇×E=− =0 ∇×E=0 Coulomb’s law, . . .
∂t ∂t
∂D ∇·D=ρ Steady currents
∇×H=J+
∂t D = εE ∇×E=0
∇·D=ρ
∇·J=0
∇·B=0
Magnetostatics J=σE

Lorentz’s force ∇×H=J

F=q E+u×B ∇·B=0
B = µH

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
13 (59)
Time-varying electromagnetic fields

Cheng, Ch. 7–2 & 7–3

Dynamic fields

Time-varying fields create electromagnetic coupling

∂B ∂B1 ∂D1
∇×E=− → time-varying (and nonconservative) E1 →
∂t ∂t ∂t
∂D ∂D1 ∂B2
∇×H=J+ → time-varying H2 →
∂t ∂t ∂t
∇·D=ρ → E2 → H3 . . .

∇·B=0

Let us start with a time-varying magnetic field (B1 in Faraday’s law) and assume
that the secondary magnetic field (B2 ) is small enough so that it can be ignored.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
15 (59)
Faraday’s law of electromagnetic induction
ds
Integrate Faraday’s law
∂B S
∇×E=−
∂t C
over a stationary surface S with bounding contour C apply Stokes’ theorem:

∂B d dΦ
Z Z I Z
(∇ × E) · ds = − · ds ⇒ E · d` = − B · ds ⇒ V= −
S S ∂t C dt S dt

I
V= E · d` = induced electromotive force (emf) in the contour C
ZC
Φ= B · ds = magnetic flux through the surface S
S

The right-handed orientation of C and ds in the figure is also important!

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
16 (59)
Induced emf in a stationary loop (= transformer emf)
The magnetic flux density B (toward the viewer) is increasing.

The increasing magnetic flux induces an emf in the contour:

dΦ ∂B
Z
C V= − =− · ds < 0
B dt S ∂t

If the contour C is a conducting loop, the emf causes a

current I to flow.
I
The current gives rise to a secondary magnetic field and
magnetic flux through the loop away from the viewer,
opposing the change in the primary magnetic field.

Lenz’s law:

The induced emf and current opposes the change in magnetic flux.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
17 (59)
Example: An ideal transformer
(1) Faraday’s law gives
i1 Φ
i2 dΦ dΦ
v1 = N1 , v2 = N2
N1 dt dt
v1 v2
N2 v1 N1
⇒ =
v2 N2
µr  1
An alternating current i1 (t) in the (2) The output power is the
primary winding with N1 turns creates a same as the input power
time-varying magnetic flux Φ that
induces an emf in the secondary winding v2 i2 = v1 i1
= secondary voltage v2 .
An ideal transformer (1) has no leakage The direction of the windings
flux and (2) is lossless. are chosen to eliminate minus
signs from the formulas.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
18 (59)
In-class exercise 2.3

i2
A time-varying current i1 in a long straight conductor induce
a current i2 in a nearby rectangular loop. The resistance of
i1 h the loop is R and the dimensions are given in the figure.
Assume that di1 /dt > 0 is given and derive i2 .
d R
Check the direction (sign) of i2 using Lenz’ law.
w

Hint: You can use the expression for the magnetic field from slide 4

µ0 I
B= aφ
2π r
and the self-inductance of the loop can be ignored.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
19 (59)
Motional emf / flux cutting emf
A conductor moves with velocity u in a static magnetic field B:
2
−− The magnetic force Fm = q u × B cause free
electrons to drift until an eletric force balance
d` B the magnetic force.
Fm /q = u × B can be interpreted as an induced
l +q u electic field producing the voltage

Fm Z2
V21 = (u × B) · d` = −uBl
++ 1
1
The induced motional emf or flux cutting emf around a closed circuit C is
I
0
V = (u × B) · d`
C

where the parts of C moving across the magnetic field lines (cutting the magnetic
flux) contribute to the emf.
Different kind of emfs?
1. Transformer emf for stationary loop
∂B
Z
V= − · ds
S ∂t

2. Motional emf for static magnetic field

I
V0 = (u × B) · d`
C

3. For a moving circuit in a time-varying magnetic field, the division between

transformer emf and motional emf depends on the frame of reference, but the
sum is always unique and equal to

dΦ
V0 = −
dt

where the change in magnetic flux depends both on the moving circuit and the
time-varying magnetic field.
ELEC–E4130 Electromagetic fields 2020 / Wallén
Time-varying fields and Maxwell’s equations
22 (59)
Another interpretation of the moving conductor

u dt
In time dt, the conductor sweeps the area

B dA = lu dt,

which can be interpreted as a change in

l u magnetic flux:

dΦ B dA
V0 = − =−
dt dt
Blu dt
=− = −Blu
t dt
t + dt

(The emf is unique and we often have more than one way to calculate it.)

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
23 (59)
Ampère’s law

In Ampère’s law
∂D
∇×H=J+
∂t

the displacement current density ∂D/∂t is needed to

ñ explain how ac current can flow through a capacitor,
ñ fullfill the conservation of charge, and
ñ admit electromagnetic waves as a solution of Maxwell’s equations.

Although the displacement current density ∂D/∂t doesn’t consist of moving

charges, it is a similar source of magnetic fields as the current density J.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
24 (59)
Ampère’s circuital law

ds
Integrate Ampère’s law
∂D S
∇×H=J+
∂t C
over a surface S with bounding contour C apply Stokes’ theorem:

∂D ∂D
Z Z   I Z Z
(∇ × H) · ds = J+ · ds ⇒ H · d` = J · ds + · ds
S S ∂t C S S ∂t

∂D
I Z
⇒ H · d` = I + · ds
C S ∂t

The contour integral of the magnetic field around a contour or circuit C is equal
to the total current (including displacement current) flowing through the circuit.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
25 (59)
In-class exercise 2.4

An ideal voltage source v(t) = V0 cos(ωt) is connected to an ideal parallel-plate

capacitor with capacitance C.

Show that the displacement current in the capacitor is the same as the conduction
current in the wires connected to the capacitor.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
26 (59)
Next week

This is the end of Lecture week 2. We’ll continue with the electromagnetic
boundary conditions next week.

The in-class exercises 2.1–2.4 on the previous slides are due 24 Sept at 14:00.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
27 (59)
Electromagnetic boundary conditions

Cheng, Ch. 7–5

Electromagnetic boundary conditions

ñ In a homogeneous medium, the electromagnetic fields must be differentiable

to satisfy Maxwell’s equations ⇒ all components of E, D, B, H must be
continuous.
ñ At the interface between two homogeneous media, we get certain boundary
conditions from Maxwell’s equations in integral form.

E1 Notation:
ε1 , µ1
E1n Subscript 1, 2 refer to medium 1, 2
an2
E1t Medium 1 Subscript t = tangential component
Medium 2 Subscript n = normal component
an1 E2
ε2 , µ2 an2 = outward unit normal for medium 2
(we take this to be the positive normal direction)

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
29 (59)
Faraday’s law
Apply Faraday’s law of induction on the contour C:

d
I Z
E1t an2
E · d` = − B · ds
C dt S
∆h ∆w 1
2 If ∆w is small and ∆h → 0, the magnetic flux through
C the contour vanishes and we get
E2t I
E · d` = (E2t − E1t ) ∆w = 0 ⇒ E1t = E2t
C

The tangential component of the electric field intensity E is continuous at the

interface.
Equivalently, using rotated tangential components:

an2 × E1 = an2 × E2

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
30 (59)
Ampère’s law
If we do the same procedure for Ampère’s circuital law

H1t an2 ∂D
I Z  
H · d` = J+ · ds
C S ∂t
∆h ∆w 1
Js 2 and include the possible surface current density Js at
C ans the interface, we get
H2t
I
H · d` = (H2t − H1t ) ∆w = Jsn ∆w
C

Using suitable unit vectors (an2 × ans points to the right in the figure)

(H2 − H1 ) · an2 × ans ∆w = (H2 − H1 ) × an2 · ans ∆w = Js · ans ∆w

 
gives the more general expression an2 × H1 − H2 = Js

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
31 (59)
No isolated magnetic charge

Integrate ∇ · B = 0 over a volume V and apply the divergence theorem:

Z I
∇ · B dv = B · ds = 0
V S

Using a small cylinder with area ∆S and

∆S an2
height ∆h → 0 placed on the boundary
∆h between medium 1 (above) and medium 2
(below), we get
I Z Z
B · ds = B1 · ds + B2 · ds = B1 · an2 ∆S − B2 · an2 ∆S = 0
S top bottom

The normal component of the magnetic flux density

⇒ an2 · B1 = an2 · B2 B is continuous at the interface.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
32 (59)
Gauss’ law
I Z
Similarly for Gauss’ law ∇ · D = ρ ⇒ D · ds = ρ dv
S V

∆S an2 Same cylinder with ∆S small and ∆h → 0,

∆h but now we may have a surface charge
ρs density ρs on the boundary, and we get
I Z Z
D · ds = D1 · ds + D2 · ds = D1 · an2 ∆S − D2 · an2 ∆S = ρs ∆S
S top bottom

As in electrostatics, any charge distribution acts as a

  source for the electric flux density D. Without
⇒ an2 · D1 − D2 = ρs surface charges, the normal component of D is
continuous.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
33 (59)
Interface between two lossless media

In lossless media, there are no free charges and the conductivity σ = 0

⇒ ρs = 0, Js = 0

and the boundary or interface conditions simplify to

E1t = E2t
If there are no sources at the boundary,
H1t = H2t the tangential component of the field
B1n = B2n intensities and normal component of the
D1n = D2n flux densities are continuous.

This is typically also the case if we have a boundary between ordinary

(low loss) materials.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
34 (59)
In-class exercise 3.1: Steady electric currents

What are the boundary conditions for the normal and tangential components of
the static current density J at the interface between two conducting media with
conductivities σ1 and σ2 ?

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
35 (59)
PEC boundary conditions

In a perfect electric conductor (PEC) the conductivity σ → ∞ and the

electromagnetic fields must be zero. (Why?)
Setting E2 = D2 = H2 = B2 = 0 in the general conditions and omitting some
unneccessary indices, we get the PEC boundary conditions

ε, µ an × E = 0
E
an × H = Js
at the boundary
an an · D = ρs
ρs H Js
PEC an · B = 0

These conditions are for the fields in the ordinary material, when the distance to
the boundary tend to zero.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
36 (59)
Electromagnetic potentials

Cheng, Ch. 7–4

Electromagnetic potentials (1/3)

Since B is solenoidal, ∇ · B = 0, it can be expressed using the vector magnetic

potential A:
B=∇×A

Substituting this into Faraday’s law gives

∂  ∂A
  
∇×E=− ∇×A ⇒ ∇× E+ = 0.
∂t ∂t

Any conservative (= curl-free) vector field can be expressed as a gradient of a

scalar field. To be consistent with electrostatics, choose

∂A ∂A
E+ = −∇V ⇒ E = −∇V −
∂t ∂t

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
38 (59)
Electromagnetic potentials (2/3)
Combining the potentials with Ampére’s law and the constitutive relations gives
!
1 ∂ ∂A
  
∇× ∇×A =J+ ε −∇V −
µ ∂t ∂t
If ε and µ are constant
∂V ∂2A
 
2
a ∇ (∇ · A) − ∇ A = µ J − µε∇ − µε 2
∂t ∂t

∂2A ∂V
 
2
a ∇ A − µε 2 = −µ J + ∇ ∇ · A + µε
∂t ∂t
A vector field is unique (up to an additive constant) if both its curl and divergence
are specified. At this point we choose the Lorenz condition or Lorenz gauge

∂V
∇ · A + µε =0
∂t
Trivia: Ludvig Lorenz (DK) 6= Hendrik Lorentz (NL), known for the Lorentz force
ELEC–E4130 Electromagetic fields 2020 / Wallén
Time-varying fields and Maxwell’s equations
39 (59)
Electromagnetic potentials (3/3)
From Gauss’ law we similarly get
∂A ∂ ρ
  
∇ · ε −∇V − =ρ a ∇2 V + (∇ · A) = −
∂t ∂t ε
and finally using the Lorenz condition

∂2V ρ
a ∇2 V − µε =−
∂t ε
Thus, assuming constant ε, µ and using the Lorenz condition we have two
separate (uncoupled) inhomogeneous wave equations for the potentials:

∂2A
∇2 A − µε = −µ J
∂t 2
∂2V ρ
∇2 V − µε =−
∂t ε

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
40 (59)
In-class exercise 3.2: Coulomb gauge
Another possible choice for the gauge condition is the so called Coulomb gauge

∇ · A = 0.

Derive the equations that the scalar and vector potentials V and A must satisfy, if
we choose this condition (and still assume that ε and µ are constants).

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
41 (59)
Solution of the wave equations

Cheng, Ch. 7–6

Solution of the scalar wave equation (1/2)

Scalar potential V (R, t) of a time-varying point charge ρ(t) ∆v 0 at the origin?

Except at the source, we have
V (R, t)
1 ∂ 2 ∂V ∂2V
 
R − µε =0
R R 2 ∂R ∂R ∂t 2
1
Substituting V = U (R, t) gives
R

∂2U ∂2U
− µε =0
∂R 2 ∂t 2
ρ(t) ∆v 0 √
This is a 1D wave equation with solutions f (t − R µε) and
√
g(t + R µε) for any twice-differentiable f and g. We choose
U (R, t) = f (t − R/u) since it is a solution traveling away
√
from the source with velocity u = 1/ µε.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
43 (59)
Solution of the scalar wave equation (2/2)
Now we have a general solution for the homogeneous wave equation
f (t − R/u)
V (R, t) =
R
and need to choose a specific function f based on the source.
For a static point charge ρ(t) ∆v 0 = q = constant, we know that
q
V =
4π εR
Comparing these two, we get
ρ(t − R/u)
V = ∆v 0 for the point charge
4π εR
and so the potential due to a charge distribution is
1 ρ(t − R/u)
Z
V = dv 0
4π ε V 0 R
where R is now the distance between dv 0 and the observation point.
ELEC–E4130 Electromagetic fields 2020 / Wallén
Time-varying fields and Maxwell’s equations
44 (59)
Retarded potentials

Since the components of the vector potential A satisfy a similar wave equation as
the scalar potential V , we thus have the solutions

1 ρ(t − R/u)
Z
V = dv 0
4π ε V 0 R
µ J(t − R/u)
Z
A= dv 0
4π V 0 R

where the volume integral is over the whole source region and R is the distance
between dv 0 and the observation point. (This distance is approximately the same
as the spherical coordinate R only for a small source at the origin.)

These are called retarded potentials since the potentials at time t depends on
the sources at an earlier time (t − R/u).

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
45 (59)
Source-free wave equations for the fields
For a non-conducting (σ = 0) source-free medium with constant ε and µ,
Maxwell’s equations reduce to
∂H ∂E
∇ × E = −µ , ∇×H=ε , ∇ · E = 0, ∇·H=0
∂t ∂t
Taking the curl of Faraday’s law, we get
   
∇ × ∇ × E = ∇ ∇ · E − ∇2 E = −∇2 E
∂   ∂2E
= −µ ∇ × H = −µε 2
∂t ∂t
Starting from the curl of Ampère’s law we get exactly the same homogeneous
vector wave equation also for H:

1 ∂2E 1 ∂2H 1
∇2 E − =0 ∇2 H − =0 u= √
u2 ∂t 2 u2 ∂t 2 µε

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
46 (59)
Time-harmonic fields

Cheng, Ch. 7–7

Phasors

A phasor is a complex time-independent quantity that has a one-to-one mapping

to the corresponding time-harmonic signal.
Using a cos(ωt)-reference, we have the definition
h i
v(t) = Re V ejωt

From this definition follows for real constants A, B, φ:

v(t) = A cos(ωt + φ) ←→ V = A ejφ = A φ

v(t) = A cos(ωt) + B sin(ωt) ←→ V = A − j B

Beware that other courses and books may use different definitions of the
mapping above! Another popular choice is to use e−iωt .

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
48 (59)
In-class exercise 3.3: Phasors

Convert the phasor

V1 = jV0 e−αz−jβz
to the corresponding time-harmonic signal v1 (t), and the signal

v2 (t) = A sin(ωt + φ)

to the corresponding phasor V2 . All given constants can be assumed real.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
49 (59)
Derivatives and integrals

Time derivatives and integrals are algebraic operations for phasors:

d d d jωt
h i   h i
v(t) = Re V ejωt = Re V e = Re V jω ejωt
dt dt dt

d 1
Z
⇒ ←→ jω and similarly dt ←→
dt jω

The fundamental usefulness of phasors is that complex arithmetic is easier than

calculus and trigonometric functions.

Mathematically, you could also define phasors

as the Fourier transform of time-harmonic signals.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
50 (59)
Time-harmonic electromagnetic fields

A time-harmonic electric field E(x, y, z, t) can be

represented with a vector phasor E(x, y, z):
h i ∇ × E = −jωB
E(t) = Re E ejωt ∇ × H = J + jωD
Note: We use the same symbol E for the ∇·D=ρ
time-harmonic electric field and its vector phasor ∇·B=0
and typically omit the space-coordinate
arguments.

For the rest of the course we will mostly deal with phasors and Maxwell’s
equations for time-harmonic fields represented by phasors.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
51 (59)
Time-harmonic potentials
The wave equations for the potentials are transformed into Helmholtz’ equations
ρ √
∇2 A + k2 A = −µJ, ∇2 V + k2 V = − , k = ω εµ
σ
and the corresponding retarded potential solution becomes

1 ρ e−jkR
Z
V = dv 0
4π ε V 0 R
µ J e−jkR
Z
A= dv 0
4π V 0 R

For a given time-harmonic source current:

R
∇× ∇×
J A H E

This strategy will be used to calculate antenna radiation. . .

ELEC–E4130 Electromagetic fields 2020 / Wallén
Time-varying fields and Maxwell’s equations
52 (59)
Source-free fields in simple media

The previously derived homogeneous vector wave equation can easily be

converted to homogeneous vector Helmholtz’s equations
√
∇2 E + k2 E = 0, ∇2 H + k2 H = 0, k = ω µε

by simply replacing ∂/∂t ⇒ jω. The same results can, of course be obtained from
the time-harmonic Maxwell’s equations.

In-class exercise 3.4:

Derive the equation ∇2 H + k2 H = 0 starting from Ampère’s law for time-harmonic
fields. Assume a simple medium with constant ε and µ, σ = 0, and no sources.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
53 (59)
Conducting media and complex permittivity

If the simple media (constant ε, µ) is conducting, we have a conduction current

J = σ E, and Ampère’s law can be put in the form
!
σ
∇ × H = J + jωD = σ E + jωεE = jω + ε E = jωεc E,
jω

using the complex permittivity

σ σ
 
0 00
εc = ε − jε = ε − j = ε0 εr − j
ω ωε0

Including the conductance in the complex permittivity eliminates the conduction

current, if we replace ε with εc in all equations.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
55 (59)
Loss tangent, good conductors and insulators

The complex permittivity makes the wave number k in Helmholtz equation

complex:
√ q
kc = ω µεc = ω µ(ε0 − jε00 )
We will later discuss how Im[kc ] is related to losses. One measure for power
loss in the medium is the loss tangent

ε00 σ
tan δc = 0
= .
ε ωε
A medium is a good conductor if σ  ωε and a good insulator if ωε  σ .

The same material may be a good conductor at low frequencies and a good
insulator at high frequencies.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
56 (59)
The electromagnetic spectrum (1/2)

Maxwell’s equations (in time-harmonic form) are exactly the same regardless of
frequency.
ñ All EM waves, including radio waves and visible light, are mathematically the
same.
√
ñ All EM waves propagate with the same velocity u = 1/ µε.

ñ The interaction between EM waves and matter do, however, depend on the
frequency. For instance ultraviolet radiation and X-rays are dangerous, while
infrared radiation (heat) is not.

ñ Material parameters depend on frequency: ε = ε(ω), etc. One prime example

is water:
For radio waves (f < 1 GHz) water has high permittivity εr ≈ 80 and the
conductivity depends heavily on the impurity: σ ≈ 1 mS/m for fresh water
and σ ≈ 4 S/m for seawater.
For visible light, water is transparent (nearly lossless) and εr ≈ 1.8.
The electromagnetic spectrum (2/2)
Frequency and wavelength range of EM waves

Penetrates Earth's
Atmosphere?

Radiation Type Radio Microwave Infrared Visible Ultraviolet X-ray Gamma ray
Wavelength (m) 103 10−2 10−5 0.5×10 −6 10−8 10−10 10−12

Approximate Scale
of Wavelength

Buildings Humans Butterflies Needle Point Protozoans Molecules Atoms Atomic Nuclei

Frequency (Hz)

10 4 10 8 1012 1015 1016 1018 1020

Temperature of
objects at which
this radiation is the
most intense
wavelength emitted 1K 100 K 10,000 K 10,000,000 K
−272 °C −173 °C 9,727 °C ~10,000,000 °C

Image from Wikimedia Commons, CC BY-SA 3.0

Outlook

In the following weeks, we’ll be using time-harmonic fields (vector phasors)

extensively for plane waves, guided waves, and antennas.

The in-class exercises 3.1–3.4 are due 1 Oct at 14:00 at the same time as
Homework 3. Remember to ask for help well before the deadline.

ELEC–E4130 Electromagetic fields 2020 / Wallén

Time-varying fields and Maxwell’s equations
59 (59)