Whites, EE 382

Lecture 1

Page 1 of 7

Lecture 1: Magnetic Circuits.

The first topic of this course is actually a continuation of
magnetostatics from EE 381 last semester. This topic is
magnetic circuit analysis and its a lumped-element method for
solving certain types of magnetic field problems.
Magnetic circuit analysis is not always applicable, but when it is
it can greatly simplify the solution to magnetics problems.
This process is similar to the simplification of electric circuits
using R, L, and C lumped elements to represent the effects of
actual devices with physical dimensions.
To illustrate magnetic circuit analysis, consider the toroid:

2016 Keith W. Whites

Whites, EE 382

Lecture 1

Page 2 of 7

This problem is similar to toroid problems you have encountered

earlier in EE 381. There is one difference though; the turns of
wire exist only over a portion of the toroid.
However, with 0 it can be shown through other means
that the B field will largely stay within the toroid. Here we will
ignore all such flux leakage effects. Then, by Ampres law
H dl NI
c

Employing symmetry, and using dl a rd

NI Wb
w
w
B
a
r
a

2 r m
2
2

(1)

We will assume that this toroid has a small cross section such
that B is nearly uniform over the cross section. Therefore,
NI
B
(2)
B
2 a
inside the toroid, and zero outside.
Now, the magnetic flux is
m B ds BA [Wb]

(3)

where A w2 is the cross-sectional area. Therefore, with (2) in

(3)
NI 2
m
(4)
w [Wb]
2 a

Whites, EE 382

Lecture 1

Page 3 of 7

Equivalent Magnetic Circuit

We now develop the equivalent magnetic circuit for this toroid
using this last equation by defining
V
m m
(5)
R
where
Vm NI F [A t] is the source. Called an mmf or
magnetomotive force with units of Ampre-turns,
The reluctance of the core (units of H-1).
l
R
[H-1]
A
where l 2 a is the mean length (circumference).

(6)

The equivalent magnetic circuit for the toroid can be drawn as

There is a direct analogy between electrical and magnetic

circuits.

Whites, EE 382

Lecture 1

Page 4 of 7

Analogous quantities in electrical and magnetic circuit analysis.

Electrical Circuits
conductivity (S/m)

emf (V)
V
current (A)
I

Magnetic Circuits
permeability (H/m)

mmf (A-t)
Vm
magnetic flux (Wb)
m
l
R
reluctance (H-1)
A

l
A

resistance ()

Ohms law

V IR

KVL (around a loop)

V R I

KCL (at a junction)

Vm m R

m ,i

R j m , j
j

m, j

Example N1.1: Determine the magnetic flux through the air gap
in the geometry shown below. The structure is assumed to have
a square cross section of area 10-6 m2, a core with r = 1,000,
and dimensions l1 = 1 cm, l3 = 3 cm, and l4 = 2 cm.
l1

l3
m,2

10 mA
2000
turns
m,1

m,1 R1

l2

l4

m,2

R3
Rg

0.1 mm
Air gap

Vm

R2

m,3

m,3

1 mm

Notice that we can subdivide this circuit so there are only three
nodes in this problem. These nodes are indicated by the dots in
the left-hand figure. We, of course, could create other equivalent

Whites, EE 382

Lecture 1

Page 5 of 7

circuits with more nodes, but that would just increase the
complexity of the analysis without adding any other usefulness.
So how do we solve such a problem? Can we use Ampres law
H dl NI
c

No, we cannot because there isnt sufficient geometrical

symmetry for us to solve for H using Ampres law.
The assumptions inherent in magnetic circuits allow us to find
an approximate solution, however. A drawback to magnetic
circuit analysis is that generally we cant check our solutions
with a simple analytical formula because there isnt one. Usually
our only recourse to check the accuracy of our magnetic circuit
solutions is to use a computational magnetics tool, which can
compute H everywhere in space.
A distinguishing characteristic of this particular example
problem is the air gap. Well assume
The length of the gap is small with respect to the cross
section, thus
No flux fringing effects:

Whites, EE 382

Lecture 1

Page 6 of 7

Flux fringing

No flux fringing

=
1000

=
1000

Summary of assumptions:
1. No flux leakage: B remains entirely within the magnetic
material and the small air gap.
2. No flux fringing: B remains vertical in the air gap in this
problem.
With no flux fringing, then the air gap can be modeled as
another reluctance in series with R 2 and shown in the
equivalent magnetic circuit above.
Compute the four reluctances using (6):
2l l
R1 1 4 31.83 106 H-1
1000 0 A
l l 0.001 / 2
R2 4 g
15.44 106 H-1
1000 0 A
2l l
R 3 3 4 63.66 106 H-1
1000 0 A
l
R g g 79.58 106 H-1
0 A

Whites, EE 382

Lecture 1

Page 7 of 7

The magnetic flux through the source coil is the mmf divided by
the total reluctance seen by the source:
V
NI
m ,1 m
0.286 106 Wb
R total R1 R 2 R g || R 3
Using flux division (analogous to current division), then
R3
m ,2
m ,1 0.115 106 Wb
R3 R2 R g
A practical application of this problem could be to find B in the
air gap, for example. This can be determined from m ,2 and the
no flux fringing and no flux leakage assumptions.