CH 10

10.
Magnetic Field and Boundary Conditions

10.1. Introduction
A static magnetic field can originate from either a constant current or a permanent magnet. This chapter will treat the magnetic
fields of constant currents. (Time-varying magnetic fields, which coexist with time-varying electric fields, will be examined in
Chapter 13.) It will also treat the behavior of the magnetic field strength H and the magnetic flux density B across the interface
of two different material. The treatment uses the static magnetic field as the vehicle to develop boundary conditions, but the
results apply to both static and time-varying magnetic fields.
10.2. Biot-Savart Law

A differential magnetic field strength, dH, results from a differential current element I dl. The field varies inversely with the
distance squared, is independent of the surrounding medium, and has a direction given by the cross product of I dl and aR. This
relationship is known as the Biot-Savart law:
I dl × aR
dH = (A/m)
4πR2
The direction of R must be from the current element to the point at whichdH is to be determined, as shown in Fig. 10-1.
Figure 10-1
Current elements have no separate existence. All elements making up the complete current filament contribute to H and must
be included. The summation leads to the integral form of the Biot-Savart law:
I dl × aR
H=∮
4πR2
© McGraw-Hill Education. All rights reserved. Any use is subject to the Terms of Use, Privacy Notice and copyright information.
A closed line integral is required to ensure that all current elements are included (the contour may close at ∞).
Example
EXAMPLE 1.
An infinitely long, straight, filamentary current I along the z axis in cylindrical coordinates is shown in Fig. 10-2. A point in the
z = 0 plane is selected with no loss in generality. In differential form,
I dz az × (rar − zaz)
dH =
4π(r2 + z2)3/2
I dz raϕ
=
4π(r2 + z2)3/2
Figure 10-2
The variable of integration is z. Since aϕ does not change with z, it may be removed from the integrand before integrating.
H = [∫ ] aϕ =
∞
Ir dz 1
aϕ
−∞ 4π(r2 + z2)3/2 2πr
This important result shows that H is inversely proportional to the radial distance. The direction is seen to be in agreement
with the "right-hand rule" whereby the fingers of the right hand point in the direction of the field when the conductor is
grasped such that the right thumb points in the direction of the current.
Example
EXAMPLE 2.
An infinite current sheet lies in the z = 0 plane with K = Kay , as shown in Fig. 10-3. Find H.
Figure 10-3
The Biot-Savart law and considerations of symmetry show that H has only an x component and is not a function of x or y.
Applying Ampère's law to the square contour 12341, and using the fact that H must be antisymmetric in z,
∮ H ⋅ dl = (H)(2a) + 0 + (H)(2a) + 0 = (K)(2a)

K
or H=
2
Thus, for all z > 0, H = (K/2)ax. More generally, for an arbitrary orientation of the current sheet,
1
H= K × an
2
Observe that H is independent of the distance from the sheet. Furthermore, the directions ofH above and below the sheet
can be found by applying the right-hand rule to a few of the current elements in the sheet.
10.3. Ampère's Law

The line integral of the tangential component of the magnetic field strength around a closed path is equal to the current enclosed
by the path:
∮ H ⋅ dl = Ienc
At first glance one would think that the law is used to determine the currentI by an integration. Instead, the current is usually
known and the law provides a method of finding H. This is quite similar to the use of Gauss's law to find D given the charge
distribution.
In order to utilize Ampère's law to determine H, there must be a considerable degree of symmetry in the problem. Two
conditions must be met:
1. At each point of the closed path H is either tangential or normal to the path.
2. H has the same value at all points of the path where H is tangential.
The Biot-Savart law can be used to aid in selecting a path which meets the above conditions. In most cases a proper path will be
evident.
Example
EXAMPLE 3.
Use Ampère's law to obtain H due to an infinitely long, straight filament of current I.
The Biot-Savart law shows that at each point of the circle in Fig. 10-2 H is tangential and of the same magnitude. Then
∮ H ⋅ dl = H(2πr) = I
so that
1
H= aϕ
2πr
10.4. Relationship of J and H

In view of Ampère's law, the defining equation for (curl H) x (see Section 5.10) may be rewritten as
lim Ix
(curl H) ⋅ ax = Δy Δz →0 ≡ Jx
Δy Δz
where Jx = dIx/dS is the area density of x-directed current. Thus, the x components of curl H and the current density J are equal
at any point. Similarly for the y and z components, so that
∇×H = J
This is one of Maxwell's equations for static fields. If H is known throughout a region, then ∇ × H will produce J for that region.
Example
EXAMPLE 4.
A long, straight conductor cross section with radius a has a magnetic field strength H = (Ir/2πa2 )aϕ within the conductor (r <
a) and H = (I/2πr)aϕ for r > a. Find J in both regions.
Within the conductor,
∂ 1 ∂ Ir2
( ) ar + ( ) az =
Ir I
J = ∇×H = − az
∂z 2πa2 r ∂r 2πa2 πa2
which corresponds to a current of magnitude I in the +z direction which is distributed uniformly over the cross-sectional
area πa2 .
Outside the conductor,
∂ 1 ∂
( ) ar + ( ) az = 0
I I
J = ∇×H = −
∂z 2πr r ∂r 2π
10.5. Magnetic Flux Density B

Like D, the magnetic field strength H depends only on (moving) charges and is independent of the medium. The force field
associated with H is the magnetic flux density B, which is given by
B = μH
where μ = μ0 μr is the permeability of the medium. The unit of B is the tesla,
N
1T = 1
A⋅m
The free-space permeability μ0 has a numerical value of 4π × 10−7 and has the units henries per meter, H/m; μr, the relative
permeability of the medium, is a pure number very near unity, except for a small group of ferromagnetic materials, which will be
treated in Chapter 12.
Magnetic flux, Φ, through a surface is defined as
Φ = ∫ B ⋅ dS
S
The sign on Φ may be positive or negative depending upon the choice of the surface normal indS. The unit of magnetic flux is
the weber, Wb. The various magnetic units are related by
1 T = 1 Wb/m2 1 H = 1 Wb/A
Example
EXAMPLE 5.
Find the flux crossing the portion of the plane ϕ = π/4 defined by 0.01 < r < 0.05 m and 0 < z < 2 m (see Fig. 10-4). A current
filament of 2.50 A along the z axis is in the az direction.
Figure 10-4
μ0I
B = μ0H = aϕ
2πr
dS = dr dz aϕ
2 0.05
μ0I
Φ = ∫ ∫ aϕ ⋅ dr dz aϕ
0 0.01 2πr
2μ0I 0.05
= ln
2π 0.01
= 1.61 × 10− 6 Wb or 1.61 μ Wb
It should be observed that the lines of magnetic flux Φ are closed curves, with no starting point or termination point. This is
in contrast with electric flux Ψ, which originates on positive charge and terminates on negative charge. In Fig. 10-5 all of the
magnetic flux Φ that enters the closed surface must leave the surface. Thus, B fields have no sources or sinks, which is
mathematically expressed by
∇⋅B = 0
(see Section 5.5).
Figure 10-5
10.6. Boundary Relations for Magnetic Fields

When H and B are examined at the interface between two different materials, abrupt changes can be expected, similar to those
noted in E and D at the interface between two different dielectrics (seeSection 8.7).
In Fig. 10-6 an interface is shown separating material 1, with properties σ1 and μr1 , from 2, with σ2 and μr2 . The behavior of B can
be determined by use of a small right circular cylinder positioned across the interface as shown. Since magnetic flux lines are
continuous,
∮ B ⋅ dS = ∫ B1 ⋅ dS1 + ∫ B ⋅ dS + ∫ B2 ⋅ dS2 = 0
end 1 cyl end 2
Now if the two planes are allowed to approach one another, keeping the interface between them, the area of the curved surface
will approach zero, giving
∫ B1 ⋅ dS1 + ∫ B2 ⋅ dS2 = 0
end 1 end 2
or
−Bn1 ∫ dS1 + Bn2 ∫ dS2 = 0

end 1 end 2
from which
Bn1 = Bn2
Figure 10-6
In words, the normal component of B is continuous across an interface. Note that either normal to the interface may be used in
calculating Bn1 and Bn2 .
The variation in H across an interface is obtained by the application of Ampère's law around a closed rectangular path, as
shown in Fig. 10-7. Assuming no current at the interface, and letting the rectangle shrink to zero in the usual way,
0 = ∮ H ⋅ dl → Hℓ1 Δℓ1 − Hℓ2 Δℓ2
whence
Hℓ1 = Hℓ2
Thus, tangential H has the same projection along the two sides of the rectangle. Since the rectangle can be rotated 90° and the
argument repeated, it follows that
Ht1 = Ht2
In words, the tangential component of H is continuous across a current-free interface.
The relation
tan θ 1 μ
= r2
tan θ 2 μr1
between the angles made by H1 and H2 with a current-free interface (see Fig. 10-7) is obtained by analogy with Example 6,
Section 8.7.
Figure 10-7
10.7. Current Sheet at the Boundary

If one material at the interface has a nonzero conductivity, a current may be present. This could be a current throughout the
material; however, of more interest is the case of a current sheet at the interface.
Fig. 10-8 shows a uniform current sheet. In the indicated coordinate system, the current sheet has density K = K0 ay and is
located at the interface x = 0 between regions 1 and 2. The magnetic field H′ produced by this current sheet is given by Example
2, Section 10.2,
1 1 1 1
H′1 = K × an1 = K 0az H′2 = K × an2 = K 0(−az)
2 2 2 2
Thus, H′ has a tangential discontinuity of magnitude |K0 | at the interface. If a second magnetic field,H″, arising from some other
source, is present, its tangential component will be continuous at the interface. The resultant magnetic field,
H = H′ + H′′
will then have a discontinuity of magnitude |K0 | in its tangential component. This is expressed by the vector formula
(H1 − H2) × an12 = K
where an12 is the unit normal from region 1 to region 2. The vector relation, which is independent of the choice of coordinate
system, also holds for a nonuniform current sheet, where K is the value of the current density at the considered point of the
interface.
Figure 10-8
10.8. Summary of Boundary Conditions

For reference purposes, the relationships for E and D across the interface of two dielectrics are shown below along with the
relationships for H and B.
Magnetic Fields Electric Fields

Dn1 = Dn2 (charge-free)
Bn1 = Bn2 {
(D1 − D2) ⋅ an12 = −ρs (with surface charge)
Hr1 = Hr2 (current-free)
{ Et1 = Et2
(H1 − H2) × an12 = K (with current sheet)
tan θ 1 μ tan θ 1 ϵ
= r2 (current-free) = r2 (charge-free)
tan θ 2 μr1 tan θ 2 ϵr1
These relationships were obtained assuming static conditions. However, in Chapter 14 they will be found to apply equally well
to time-variable fields.
10.9. Vector Magnetic Potential A

Electric field intensity E was first obtained from known charge configurations. Later, electric potential V was developed and it
was found that E could be obtained as the negative gradient of V, i.e., E = −∇V. Laplace's equation provided a method of
obtaining V from known potentials on the boundary conductors. Similarly, a vector magnetic potential, A, defined such that
∇×A = B
serves as an intermediate quantity, from which B, and hence H, can be calculated. Note that the definition of A is consistent
with the requirement that ∇ · B = 0. The units of A are Wb/m or T · m.
If the additional condition
∇⋅A = 0
is imposed, then vector magnetic potential A can be determined from the known currents in the region of interest. For the three
standard current configurations the expressions are as follows.
μI dl
current filament: A = ∮
4πR
μK dS
sheet current: A=∫
S 4πR
μJ dv
volume current: A=∫
v 4πR
Here, R is the distance from the current element to the point at which the vector magnetic potential is being calculated. Like the
analogous integral for the electric potential (see Section 6.5), the above expressions for A presuppose a zero level at infinity;
they cannot be applied if the current distribution itself extends to infinity.
Example
EXAMPLE 6.
Investigate the vector magnetic potential for the infinite, straight, current filament I in free space.
In Fig. 10-9 the current filament is along the z axis and the observation point is (x, y, z). The particular current element
I d l = I dℓaz
at ℓ = 0 is shown, where ℓ is the running variable along the z axis. It is clear that the integral
∞
μ0I dℓ
A=∫ a
−∞ 4πR z
does not exist, since, when ℓ is large, R ≈ ℓ. This is a case of a current distribution that extends to infinity.
It is possible, however, to consider the differential vector potential
μ0I dℓ
dA = a
4πR z
and to obtain from it the differential B. Thus, for the particular current element at ℓ = 0,
μ0I dℓ
dA = az
4π(x2 + y2 + z2)1/2
and
[ ay]
μ0I dℓ −y x
dB = ∇ × dA = a x +
4π (x2 + y2 + z3)3/2 (x2 + y2 + z2)3/2
This result agrees with that for dH = (1/μ0 ) dB given by the Biot-Savart law.
For a way of defining A for the infinite current filament, seeProblem 10.20.
Figure 10-9
10.10. Stokes' Theorem

Consider an open surface S whose boundary is a closed curve C. Stokes' theorem states that the integral of the tangential
component of a vector field F around C is equal to the integral of the normal component of curl F over S:
∮ F ⋅ dl = ∫ (∇ × F) ⋅ dS
S
If F is chosen to be the vector magnetic potential A, Stokes' theorem gives
∮ A ⋅ dl = ∫ B ⋅ dS = Φ
S
10.11. SOLVED PROBLEMS

10.1. Find H at the center of a square current loop of sideL.
Choose a Cartesian coordinate system such that the loop is located as shown in Fig. 10-10. By symmetry, each half-side
contributes the same amount to H at the center. For the half-side 0 ≤ x ≤ L/2, y = −L/2, the Biot-Savart law gives for the field
at the origin
(I dx ax) × [−xax + (L/2)ay]

dH =
3/2
4π[x2 + (L/2)2]
I dx (L/2)az
=
3/2
4π [x2 + (L/2)2]
Therefore, the total field at the origin is
L/2 I dx (L/2)az
H = 8∫
0 3/2
4π [x2 + (L/2)2]
2√2 I 2√2 I
= az = a
πL πL n
where an is the unit normal to the plane of the loop as given by the usual right-hand rule.
Figure 10-10
10.2. A current filament of 5.0 A in the ay direction is parallel to the y axis at x = 2 m, z = −2 m (Fig. 10-11). Find H at the
origin.
Figure 10-11
The expression for H due to a straight current filament applies,
I
H= aϕ
2πr
where r = 2
√2 and (use the right-hand rule)
ax + az
aϕ =
√2
Thus,
5.0 ax + az a + az
H= ( ) = (0.281) ( x ) A/m
2π(2√2) √2 √2
10.3. A current sheet, K = 10az A/m, lies in the x = 5 m plane and a second sheet, K = −10az A/m, is at x = − 5m. Find H at
all points.
In Fig. 10-12 it is apparent that at any point between the sheets,K × an = −Kay for each sheet. Then, for −5 < x < 5, H =
10(−ay ) A/m. Elsewhere, H = 0.
Figure 10-12
10.4. A thin cylindrical conductor of radius a, infinite in length, carries a current I. Find H at all points using Ampère's law.
The Biot-Savart law shows that H has only a ϕ component. Furthermore, Hϕ is a function of r only. Proper paths for Ampère's
law are concentric circles. For path 1 shown in Fig. 10-13,
∮ H ⋅ dl = 2πrHϕ = Ienc = 0
and for path 2,
∮ H ⋅ dl = 2πrHϕ = I
Figure 10-13
Thus, for points within the cylindrical conducting shell, H = 0, and for external points, H = (I/2πr)aϕ , the same field as that of
a current filament I along the axis.
10.5. Determine H for a solid cylindrical conductor of radius a, where the current I is uniformly distributed over the cross
section.
Applying Ampère's law to contour 1 in Fig. 10-14,
∮ H ⋅ dI = Ienc
πr2
H(2πr) = I ( )
πa2
Ir
H = aϕ
2πa2
Figure 10-14
For external points, H = (I/2πr) aϕ .
10.6. In the region 0 < r < 0.5 m, in cylindrical coordinates, the current density is
J = 4.5e− 2raz (A/m2)
and J = 0 elsewhere. Use Ampère's law to find H.
Because the current density is symmetrical about the origin, a circular path may be used in Ampère's law, with the enclosed
current given by
∮ J · dS. Thus, for r < 0.5 m,
2π r
Hϕ (2πr) = ∫ ∫ 4.5e− 2rr dr dϕ
0 0
1.125
H = (1 − e− 2r − 2re− 2r) aϕ (A/m)
r
For any r ≥ 0.5 m, the enclosed current is the same, 0.594π A. Then,
0.297
Hϕ (2πr) = 0.594π or H= aϕ (A/m)
r
10.7. Find H on the axis of a circular current loop of radius a. Specialize the result to the center of the loop.
For the point shown in Fig. 10-15,
R = − a + a
R = −aar + haz
(Iadϕ aϕ ) × (−aar + haz) (Iadϕ)(aaz + har)
dH = 3/
=
4π(a2 + h2) 2 4π(a2 + h2)3/2
Figure 10-15
Inspection shows that diametrically opposite current elements produce r components which cancel each other. Then,
2π Ia2 dϕ Ia2
H=∫ az = az
0 4π (a2 + h2)3/2 2 (a2 + h2)3/2
At h = 0, H = (I/2a)az.
10.8. A current sheet, K = 6.0ax A/m, lies in the z = 0 plane and a current filament is located at y = 0, z = 4 m, as shown in
Fig. 10-16. Determine I and its direction if H = 0 at (0, 0, 1.5) m.
Figure 10-16
Due to the current sheet,
1 6.0
H= K × an = (−ay) A/m
2 2
For the field to vanish at (0, 0, 1.5) m, |H|due to the filament must be 3.0 A/m.
1
1
|H| =
2πr
1
3.0 =
2π(2.5)
I = 47.1 A
To cancel the H from the sheet, this current must be in theax direction, as shown in Fig. 10-16.
10.9. Calculate the curl of H in Cartesian coordinates due to a current filament along the z axis with current I in the az
direction.
From Example 1,
1 1 −yax + xay
H= aϕ = ( )
2πr 2π x2 + y2
and so
∣ ax ay az ∣
∣ ∂ ∂ ∂ ∣∣
∣
∇ × H = ∣ ∂x ∂y ∂z ∣
∣ −y x ∣
∣ 0 ∣
∣ x +y
2 2 x + y2
2 ∣
∂ ∂ −y
∇×H = [ ( ) − ( ) ] az
x
∂x x2 + y2 ∂y x2 + y2
= 0
except at x = y = 0. This is consistent with ∇ × H = J.
10.10. A circular conductor of radius r0 = 1 cm has an internal field
104 1
( sin ar − cos ar) aϕ (A/m)
r
H=
r a2 a
where a = π/2r0 . Find the total current in the conductor.
There are two methods: (1) to calculate J = ∇ × H and then integrate; (2) to use Ampère's law. The second is simpler here.
104 4r20 π 2r20

( cos ) r0 dϕ
2π
Ienc = ∮ H ⋅ dl = ∫
π
sin −
r=r0 0 r0 π2 2 π 2
8 × 104r20 8
= = A
π π
10.11. A radial field
2.39 × 106
H= cos ϕ ar A/m
r
exists in free space. Find the magnetic flux Φ crossing the surface defined by −π/4 ≤ ϕ ≤ π/4, 0 ≤ z ≤ 1 m. See Fig. 10-17.
3.00
B = μ0H = cos ϕar (T)
r
1 π/4 3.00
Φ = ∫ ∫ ( cos ϕ) ar ⋅ r dϕ dz ar
0 − π/4 r
= 4.24 Wb
Figure 10-17
Since B is inversely proportional to r (as required by ∇ · B = 0), it makes no difference what radial distance is chosen. The
total flux will be the same.
10.12. In cylindrical coordinates, B = (2.0/r)aϕ (T). Determine the magnetic flux Φ crossing the plane surface defined by 0.5
≤ r ≤ 2.5 m and 0 ≤ z ≤ 2.0 m. See Fig. 10-18.
Φ = ∫ B ⋅ dS
2.0 2.5
2.0
= ∫ ∫ aϕ ⋅ dr dz aϕ
0 0.5 r
2.5
= 4.0 (ln ) = 6.44 W b
0.5
Figure 10-18
10.13. In region 1 of Fig. 10-19, B1 = 1.2ax + 0.8ay + 0.4az (T). Find H2 (i.e., H at z = + 0) and the angles between the field
vectors and a tangent to the interface.
Figure 10-19
Write H1 directly below B1 . Then write those components of H2 and B2 which follow directly from the two rules: B normal is
continuous and H tangential is continuous across a current-free interface.
B1 = 1.2ax + 0.8ay + 0.4az (T)

1
H1 = (8.0ax + 5.33ay + 2.67az)10− 2 (A/m)
μ0
1
H2 = (8.0ax + 5.33ay + 102μ0Hz2az)10− 2 (A/m)
μ0
B2 = Bx2ax + By2ay + 0.4az (T)
Now the remaining terms follow directly:
Bz2 0.4
Bx2 = μ0 μr2Hx2 = 8.0 × 10− 2 (T) By2 = 5.33 × 10− 2 (T) Hz2 = = (A/m)
μ0μr2 μ0
Angle θ1 is 90° − α1 , where α1 is the angle between B1 and the normal, az.
B ⋅a
B1 ⋅ az
cos α1 = = 0.27
|B1 |
whence α1 = 74.5° and θ1 = 15.5°. Similarly, θ2 = 76.5°.
Check: (tan θ1 )/(tan θ2 ) = μr2 /μr1 .
10.14. Region 1, for which μr1 = 3, is defined by x < 0 and region 2, x > 0, has μr2 = 5. Given
H1 = 4.0ax + 3.0ay − 6.0az (A/m)
show that θ2 = 19.7° and H2 = 7.12 A/m.
Proceed as in Problem 10.13.
H1 = 4.0ax + 3.0ay − 6.0az (A/m)

B1 = μ0(12.0ax + 9.0ay − 18.0az) (T)
B2 = μ0(12.0ax + 15.0ay − 30.0az) (T)
H2 = 2.40ax + 3.0ay − 6.0az (A/m)
Now
H2 = √(2.40)2 + (3.0)2 + (−6.0)2 = 7.12 A/m
The angle α2 between H2 and the normal is given by
Hx2
cos α2 = = 0.34 or α2 = 70.3°
H2
Then θ2 = 90° − α2 = 19.7°.
10.15. Region 1, where μr1 = 4, is the side of the plane y + z = 1 containing the origin (see Fig. 10-20). In region 2, μr2 = 6. B1
= 2.0ax + 1.0ay (T). Find B2 and H2 .
Figure 10-20
Choosing the unit normal an = (ay + az)/
√2,
(2.0ax + 1.0ay) ⋅ (ay + az) 1

Bn1 = =
√2 √2
1
Bn1 = ( ) an = 0.5ay + 0.5az = Bn2
√2
Bt1 = B1 − Bn1 = 2.0ax + 0.5ay − 0.5az
1
Ht1 = (0.5ax + 0.125ay − 0.125az) = Ht2
μ0
Bt2 = μ0μr2 Ht2 = 3.0ax 0.75ay − 0.75az
Now the normal and tangential parts of B2 are combined.
B2 = 3.0ax + 1.25ay − 0.25az (T)

1
H2 = (0.50ax + 0.21ay − 0.04az) (A/m)
μ0
10.16. In region 1, defined by z < 0, μr1 = 3 and
1
H1 = (0.2ax + 0.5ay − 1.0az) (A/m)
μ0
Find H2 if it is known that θ2 = 45°.
H 1 ⋅ az
cos α1 = = 0.88 or α1 = 28.3°
|H1|
Then, θ1 = 61.7° and
tan 61.7° μ
= r2 or μr2 = 5.57
tan 45° 3
From the continuity of the normal component of B, μr1 Hz1 = μr2 Hz2 , and so
1 1
(0.2ax + 0.5ay + r1 1.0az) =
μ
H2 = (0.2ax + 0.5ay + 0.54az) (A/m)
μ0 μr2 μ0
10.17. A current sheet, K = 6.5az A/m, at x = 0 separates region 1, x < 0, where H1 = 10ay A/m and region 2, x > 0. Find H2 at
x = +0.
Nothing is said about the permeabilities of the two regions; however, since H1 is entirely tangential, a change in permeability
would have no effect. Since Bn1 = 0, Bn2 = 0 and therefore Hn2 = 0,
(H1 − H2) × an12 = K
(10ay − Hy2ay) × ax = 6.5az
(10 − Hy2)(−az) = 6.5az
Hy2 = 16.5 (A/m)
Thus, H2 = 16.5ay (A/m).
10.18. A current sheet, K = 9.0ay A/m, is located at z = 0, the interface between region 1, z < 0, with μr1 = 4, and region 2, z >
0, μr2 = 3. Given that H2 = 14.5ax + 8.0az (A/m), find H1 .
The current sheet shown in Fig. 10-21 is first examined alone.
1
H′1 = (9.0)ay × (−az) = 4.5(−ax)
2
1
H′2 = (9.0) ay × az = 4.5ax
2
Figure 10-21
From region 1 to region 2, Hx will increase by 9.0 A/m due to the current sheet.
Now the complete H and B fields are examined:
H2 = 14.5ax + 8.0az (A/m)

B2 = μ0(43.5ax + 24.0az) (T)
B1 = μ0(22.0ax + 24.0ax) (T)
H1 = 5.5ax + 6.0az (A/m)
Note that Hx1 must be 9.0 A/m less than Hx2 because of the current sheet. Bx1 is obtained as μ0 μr1 Hx1 .
An alternate method is to apply (H1 − H2 ) × an12 = K:
(Hx1ax + Hy1ay + Hz1az) × az = K + (14.5ax + 8.0az) × az

−Hx1ay + Hy1ax = −5.5ay
from which Hx1 = 5.5 A/m and Hy1 = 0. This method deals exclusively with tangential H; any normal component must be
determined by the previous methods.
10.19. Region 1, z < 0, has μr1 = 1.5, while region 2, z > 0, has μr2 = 5. Near (0, 0, 0),
B1 = 2.40ax + 10.0az (T) B2 = 25.75ax − 17.7ay + 10.0az (T)
If the interface carries a sheet current, what is its density at the origin?
Near the origin,
1 1
H1 = B1 = (1.60ax + 6.67az) (A/m)
μ0μr1 μ0
1
H2 = (5.15ax − 3.54ay + 2.0az) (A/m)
μ0
Then the local value of K is given by
1 5.0 ax + ay
K = (H1 − H2) × an12 = (−3.55ax + 3.54ay + 4.67az) × az = ( ) (A/m)
μ0 μ0 √2
10.20. Obtain the vector magnetic potential A in the region surrounding an infinitely long, straight, filamentary current I.
As shown in Example 6, the direct expression for A as an integral cannot be used. However, the relation
μ0I
∇×A = B = aϕ
2πr
may be treated as a vector differential equation for A. Since B possesses only a ϕ component, only the ϕ component of the
cylindrical curl is needed.
∂Ar ∂Ar μ0I

− =
∂z ∂r 2πr
It is evident that A cannot be a function of z, since the filament is uniform with z. Then
dAz μ0I μ0I

− = or Az = − ln r + C
dr 2πr 2π
The constant of integration permits the location of a zero reference. WithAz = 0 at r = r0 , the expression becomes
μ0I r0
A= (ln ) az
2π r
10.21. Obtain the vector magnetic potential A for the current sheet of Example 2.
For z > 0,
μ0K
∇×A = B = a
2 x
whence
∂Az ∂Ay μ0K

− =
∂y ∂z 2
As A must be independent of x and y,
dAy μ0K μ0K

− = or Ay = − (z − z0)
dz 2 2
Thus, for z > 0,
μ0K μ0
A=− (z − z0) ay = − (z − z0)K
2 2
For z < 0, change the sign of the above expression.
10.22. Using the vector magnetic potential found in Problem 10.21, find the magnetic flux crossing the rectangular area
shown in Fig. 10-22.
Figure 10-22
Let the zero reference be at z0 = 2, so that
μ0
A=− (z − 2)K
2
In the line integral
Φ = ∮ A ⋅ dl
A is perpendicular to the contour on two sides and vanishes on the third (z = 2). Thus,
y=2 2
μ0
Φ=∫ A ⋅ dl = − (1 − 2) ∫ K dy = μ0K
y= 0 2 0
Note how the choice of zero reference simplified the computation. By Stokes' theorem it is ∇ ×A, and not A itself, that
determines Φ; hence the zero reference may be chosen at pleasure.
10.12. SUPPLEMENTARY PROBLEMS

10.23. Show that the magnetic field due to the finite current element shown inFig. 10-23 is given by
1
H= (sin α1 − sin α2) aϕ
4πr
Figure 10-23
10.24. Obtain dH at a general point (r, θ, ϕ) in spherical coordinates, due to a differential current element I dI at the origin
in the positive z direction.
10.25. Currents in the inner and outer conductors of Fig. 10-24 are uniformly distributed. Use Ampère's law to show that
for b ≤ r ≤ c,
c2 − r2
( ) aϕ
I
H=
2πr c2 − b2
Figure 10-24
10.26. Two identical circular current loops of radius r = 3 m and I = 20 A are in parallel planes, separated on their common
axis by 10 m. Find H at a point midway between the two loops.
Schaum's Electromagnetics Supplementary Problem 10-26: Magnetic Field

due to Current Flow
This video illustrates the magnetic field strength from a couple of loops
carrying current.
Sudipa Mitra-Kirtley, Ph.D, Professor, Physics and Optical Engineering, Rose-Hulman

Institute of Technology
2013
10.27. A current filament of 10 A in the +y direction lies along the y axis, and a current sheet, K = 2.0ax A/m, is located at z
= 4 m. Determine H at the point (2, 2, 2) m.
10.28. A cylindrical conductor of radius 10−2 m has an internal magnetic field
r2
H = (4.77 × 104) ( ) aϕ
r
− (A/m)
2 3 × 10− 2
What is the total current in the conductor?
Schaum's Electromagnetics Supplementary Problem 10-28: Magnetic Field

and Current Density
This video illustrates relations between magnetic field strength and electric
current using vector algebra.
Sudipa Mitra-Kirtley, Ph.D, Professor, Physics and Optical Engineering, Rose-Hulman

Institute of Technology
2013
10.29. In cylindrical coordinates, J = 105 (cos2 2r)az in a certain region. Obtain H from this current density and then take the
curl of H and compare with J.
10.30. In Cartesian coordinates, a constant current density, J = J0 ay , exists in the region −a ≤ z ≤ a. See Fig. 10-25. Use
Ampère's law to find H in all regions. Obtain the curl ofH and compare with J.
Figure 10-25
10.31. Compute the total magnetic flux Φ crossing the z = 0 plane in cylindrical coordinates for r ≤ 5 × 10−2 m if
0.2
B= (sin2ϕ)az (T)
r
10.32. Given that
πx − 2y
B = 2.50 (sin ) e az (T)
2
find the total magnetic flux crossing the strip z = 0, y ≥ 0, 0 ≤ x ≤ 2 m.
10.33. A coaxial conductor with an inner conductor of radius a and an outer conductor of inner and outer radii b and c,
respectively, carries current I in the inner conductor. Find the magnetic flux per unit length crossing a plane ϕ = const.
between the conductors.
10.34. Region 1, where μr1 = 5, is on the side of the plane 6x + 4y + 3z = 12 that includes the origin. In region 2, μr2 = 3.
Given
1
H1 = (3.0ax − 0.5ay) (A/m)
μ0
find B2 and θ2 .
10.35. The interface between two different regions is normal to one of the three Cartesian axes. If
B1 = μ0(43.5ax + 24.0az) B2 = μ0(22.0ax + 24.0az)
what is the ratio (tan θ1 )/(tan θ2 )?
10.36. Inside a right circular cylinder, μr1 = 1000. The exterior is free space. If B1 = 2.5aϕ (T) inside the cylinder, determine
B2 just outside.
10.37. In spherical coordinates, region 1 is r < a, region 2 is a < r < b and region 3 is r > b. Regions 1 and 3 are free space,
while μr2 = 500. Given B1 = 0.20ar (T), find H in each region.
10.38. A current sheet, K = (8.0/μ0 )ay (A/m), at x = 0 separates region 1, x < 0 and μr1 = 3, from region 2, x > 0 and μr2 = 1.
Given H1 = (10.0/μ0 )(ay + az) (A/m), find H2 .
10.39. The x = 0 plane contains a current sheet of density K which separates region 1, x < 0 and μr1 = 2, from region 2, x < 0
and μr2 = 7. Given
B1 = 6.0ax + 4.0ay + 10.0az (T) B2 = 6.0ax − 50.96ay + 8.96az (T)
find K.
10.40. One uniform current sheet, K = K0 ay , is at z = b > 2 and another, K = K0 (−ay ), is at z = −b. Find the magnetic flux
crossing the area defined by x = const., −2 ≤ x ≤ 2, 0 ≤ y ≤ L. Assume free space.
10.41. Use the vector magnetic potential from Problem 10.20 to obtain the flux crossing the rectangle ϕ = const., r1 ≤ r ≤ r0 ,
0 ≤ z ≤ L, due to a current filament I on the z axis.
10.42. Given that the vector magnetic potential within a cylindrical conductor of radius a is
μ0Ir2
A=− az
4πa2
find the corresponding H.
10.43. One uniform current sheet, K = K0 (−ay ), is located at x = 0 and another, K = K0 ay , is at x = a. Find the vector magnetic
potential between the sheets.
10.44. Between the current sheets of Problem 10.43 a portion of a z = const. plane is defined by 0 ≤ x ≤ b and 0 ≤ y ≤ a.
Find the flux Φ crossing this portion, both from
∫ B · dS and from
∮ A · dI.
10.13. ANSWERS TO SUPPLEMENTARY PROBLEMS

10.24.
I dℓ sin θ
aϕ
4πr2
10.26. 0.908an A/m
10.27. 0.398ax + 1.0ay − 0.398az A/m
10.28. 5.0 A
10.29
r sin 4r cos 4r 1
H = 105 ( + + − ) aϕ .
4 8 32r 32r
10.30.
⎧
⎪ J0aax z > a
H = ⎨ J0zax −a ≤ z ≤ a
⎩
⎪ −J aa z < −a
0 x
curl H = J
10.31. 3.14 × 10−2 Wb
10.32. 1.59 Wb
10.33.
μ0I b
ln
2π n
10.34. 12.15ax + 0.60ay + 1.58az (T), 56.6°
10.35. 0.506
10.36. 2.5aϕ (mT)
10.37.
0.20 4 × 10− 4 0.20
(A/m), (A/m), (A/m)
μ0 μ0 μ0
10.38.
1
(10.0 ay + 2.0az) (A/m)
μ0
10.39.
1
(3.72ay − 9.28az) (A/m)
μ0
10.40. 4μ0 K0 L
10.41.
μ0IL r0
ln
2π r1
10.42. μ0 H = curl A
10.43. (μ0 K0 x + C)ay
10.44. abμ0 K0

CH 10

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

CH 10

Uploaded by

Copyright:

Available Formats

10.

Magnetic Field and Boundary Conditions

10.2. Biot-Savart Law

∮ H ⋅ dl = (H)(2a) + 0 + (H)(2a) + 0 = (K)(2a)

10.3. Ampère's Law

10.4. Relationship of J and H

Within the conductor,

Outside the conductor,

10.5. Magnetic Flux Density B

where μ = μ0 μr is the permeability of the medium. The unit of B is the tesla,

Magnetic flux, Φ, through a surface is defined as

(see Section 5.5).

10.6. Boundary Relations for Magnetic Fields

−Bn1 ∫ dS1 + Bn2 ∫ dS2 = 0

0 = ∮ H ⋅ dl → Hℓ1 Δℓ1 − Hℓ2 Δℓ2

In words, the tangential component of H is continuous across a current-free interface.

10.7. Current Sheet at the Boundary

(H1 − H2) × an12 = K

10.8. Summary of Boundary Conditions

Magnetic Fields Electric Fields

10.9. Vector Magnetic Potential A

If the additional condition

It is possible, however, to consider the differential vector potential

10.10. Stokes' Theorem

If F is chosen to be the vector magnetic potential A, Stokes' theorem gives

10.11. SOLVED PROBLEMS

(I dx ax) × [−xax + (L/2)ay]

Therefore, the total field at the origin is

The expression for H due to a straight current filament applies,

and for path 2,

For external points, H = (I/2πr) aϕ .

J = 4.5e− 2raz (A/m2)

and J = 0 elsewhere. Use Ampère's law to find H.

For the point shown in Fig. 10-15,

Due to the current sheet,

except at x = y = 0. This is consistent with ∇ × H = J.

10.10. A circular conductor of radius r0 = 1 cm has an internal field

where a = π/2r0 . Find the total current in the conductor.

104 4r20 π 2r20

10.11. A radial field

B1 = 1.2ax + 0.8ay + 0.4az (T)

Now the remaining terms follow directly:

whence α1 = 74.5° and θ1 = 15.5°. Similarly, θ2 = 76.5°.

Check: (tan θ1 )/(tan θ2 ) = μr2 /μr1 .

H1 = 4.0ax + 3.0ay − 6.0az (A/m)

show that θ2 = 19.7° and H2 = 7.12 A/m.

Proceed as in Problem 10.13.

H1 = 4.0ax + 3.0ay − 6.0az (A/m)

H2 = √(2.40)2 + (3.0)2 + (−6.0)2 = 7.12 A/m

The angle α2 between H2 and the normal is given by

Then θ2 = 90° − α2 = 19.7°.

(2.0ax + 1.0ay) ⋅ (ay + az) 1

Now the normal and tangential parts of B2 are combined.

B2 = 3.0ax + 1.25ay − 0.25az (T)

10.16. In region 1, defined by z < 0, μr1 = 3 and

Find H2 if it is known that θ2 = 45°.

Then, θ1 = 61.7° and

Thus, H2 = 16.5ay (A/m).

The current sheet shown in Fig. 10-21 is first examined alone.

Now the complete H and B fields are examined:

H2 = 14.5ax + 8.0az (A/m)

An alternate method is to apply (H1 − H2 ) × an12 = K:

(Hx1ax + Hy1ay + Hz1az) × az = K + (14.5ax + 8.0az) × az