Chapter 1

BASIC PRINCIPLES FOR ELECTRIC

MACHINE ANALYSIS
1.1 INTRODUCTION
There are several basic concepts that must be established before the analysis of elec-
tric machines can begin. The principle of electromechanical energy conversion is
perhaps the cornerstone of machine analysis. This theory allows us to establish an
expression of electromagnetic torque in terms of machine variables, generally the
currents and the displacement of the mechanical system. Other principles that
must be established are (I) the derivation of equivalent circuit representations of
magnetically coupled circuits, (2) the concept of a sinusoidally distributed winding,
(3) the concept of a rotating air-gap magnetomotive force (MMF), and (4) the deri-
vation of winding inductances. The above-mentioned basic principles are presented
in this chapter, concluding with the voltage equations of a 3-phase synchronous
machine and a 3-phase induction machine. It is shown that the equations, which
describe the behavior of alternating-current (ac) machines, contain time-varying
coefficients due to the fact that some of the machine inductances are functions of
the rotor displacement. This establishes an awareness of the complexity of these vol-
tage equations and sets the stage for the change of variables (Chapter 3), which
reduces the complexity of the voltage equations by eliminating the time-dependent
inductances.
1.2 MAGNETICALLY COUPLED CIRCUITS
Magnetically coupled electric circuits are central to the operation of transformers
and electric machines. In the case of transformers, stationary circuits are
1
2 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
+

, l

/'
-; / m2
I (
I ,
<l>ml

+
Figure 1.2-1 Magnetically coupled circuits.
magnetically coupled for the purpose of changing the voltage and current levels. In
the case of electric machines, circuits in relative motion are magnetically coupled
for the purpose of transferring energy between mechanical and electrical systems.
Because magnetically coupled circuits play such an important role in power trans-
mission and conversion, it is important to establish the equations that describe their
behavior and to express these equations in a form convenient for analysis. These
goals may be achieved by starting with two stationary electric circuits that are mag-
netically coupled as shown in Fig. 1.2-1. The two coils consist of turns N
I
and N
2
,
respectively, and they are wound on a common core that is generally a ferromagnetic
material with permeability large relative to that of air. The permeability of free
space, flo, is 4n x 10-
7
Him. The permeability of other materials is expressed as
u == flrflo where fl
r
is the relative permeability. In the case of transformer steel the
relative permeability may be as high 2000 to 4000.
In general, the flux produced by each coil can be separated into two components:
a leakage component denoted with an 1 subscript and a magnetizing component
denoted by an m subscript. Each of these components is depicted by a single stream-
line with the positive direction determined by applying the right-hand rule to the
direction of current flow in the coil. Often, in transformer analysis, i2 is selected
positive out of the top of coil 2, and a dot is placed at that terminal.
The flux linking each coil may be expressed as
<PI == <P/I + <Pml +<Pm2
<P2 == <P/2 + <Pm2 +<Pml
(1.2-1)
( 1.2-2)
The leakage flux <P/I is produced by current flowing in coil 1, and it links only the
turns of coil 1. Likewise, the leakage flux <P/2 is produced by current flowing in coil 2,
and it links only the turns of coil 2. The magnetizing flux <Pml is produced by current
flowing in coil I, and it links all turns of coils 1 and 2. Similarly, the magnetizing
flux 4>m2 is produced by current flowing in coil 2, and it also links all turns of coils 1
and 2. With the selected positive direction of current flow and the manner in which
MAGNETICALLY COUPLED CIRCUITS 3
the coils are wound (Fig. 1.2-1), magnetizing flux produced by positive current in
one coil adds to the magnetizing flux produced by, positive current in the other
coil. In other words, if both currents are actually flowing in the same direction,
the magnetizing fluxes produced by each coil are in the same direction, making
the total magnetizing fluxor the total core flux the sum of the instantaneous magni-
tudes of the individual magnetizing fluxes. If the actual currents are in opposite
directions, the magnetizing fluxes are in opposite directions. In this case, one coil
is said to be magnetizing the core, the other demagnetizing.
Before proceeding, it is appropriate to point out that this is an idealization of the
actual magnetic system. Clearly, all of the leakage flux may not link all the turns of
the coil producingit. Likewise, all of the magnetizingfluxof one coil may not link all
of the turns of the other coil. To acknowledge this practical aspect of the magnetic
system, the number of turns is consideredto be an equivalent number rather than the
actual number. This fact shouldcause us little concern because the inductances of the
electric circuit resulting from the magnetic coupling are generally determined from
tests.
The voltage equations may be expressed in matrix form as
d"J...
v == rt + dt
where r == diag ['1 '2]' a diagonal matrix, and
(f)T== [/1 12]
(1.2-3)
(1.2-4)
wheref represents voltage, current, or fluxlinkage. The resistances '1 and '2 and the
flux linkages AI and A2 are related to coils 1 and 2, respectively. Because it is
assumedthat q,I links the equivalent turns of coil 1and q,2 links the equivalent turns
of coil 2, the flux linkages may be written as
AI == N
I
4> ,
A2 == N24>2
where <PI and <P2 are given by (1.2-1) and (1.2-2), respectively.
Linear Magnetic System
(1.2-5)
(1.2-6)
(1.2-8)
(1.2-9)
(1.2-10)
If saturation is neglected, the system is linear and the fluxes may be expressed as
<I> - Nli
l
(1.2-7)
II - uP
3111
<I> _Nli l
ml - ulJ
;:/lm
<I> _ N2
i
2
12 - 0lJ
dil2
~ _ N2
i2
'J:'m2 ---
~ m
4 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
where ~ l l and f!Jl
l2
are the reluctances of the leakage paths and PAm is the reluctance
of the path of the magnetizing fluxes. The product of N times i (ampere-turns) is the
MMF, which is determined by application of Ampere's law. The reluctance of the
leakage paths is difficult to express and impossible to measure. In fact, a unique
determination of the inductances associated with the leakage flux cannot be made
by tests; instead, it is either calculated or approximated from design considerations.
The reluctance of the magnetizing path of the core shown in Fig. 1.2-1 may be com-
puted with sufficient accuracy from the well-known relationship
f!Jl == _l
J1A
(1.2-11 )
where 1is the mean or equivalent length of the magnetic path, Ais the cross-sectional
area, and Jl is the permeability.
Substituting (1.2-7)-(1.2-10) into (1.2-1) and (1.2-2) yields
(1.2-12)
(1.2-13)
Substituting (1.2-12) and (1.2-13) into (1.2-5) and (1.2-6) yields
(1.2-14)
(1.2-15)
(1.2-16)
( 1.2-17)
When the magnetic system is linear, the flux linkages are generally expressed in
terms of inductances and currents. We see that the coefficients of the first two terms
on the right-hand side of (1.2-14) depend upon the turns of coil 1 and the reluctance
of the magnetic system, independent of the existence of coil 2. An analogous state-
ment may be made regarding (1.2-15). Hence the self-inductances are defined as
N
2
N
2
L
I I
==-;;;!-+ (Jljl ==Lll +Lml
vtll vtm
N
2
N
2
L
22
== (Jlj2 +(Jlj2 == L12 +L
m
2
~ 1 2 ~ m
where L
ll
and LI2 are the leakage inductances and L
ml
and L
m2
are the magnetizing
inductances of coils 1 and 2, respectively. From (1.2-16) and (1.2-17) it follows that
the magnetizing inductances may be related as
(1.2-18)
MAGNETICALLY COUPLED CIRCUITS 5
The mutual inductances are defined as the coefficient of the third term of
(1.2- 14) and (1.2-15).
( 1.2-19)
( 1.2-20)
Obviously, L
I
2 == L21. The mutual inductances may be related to the magnetizing
inductances. In particular,
The flux linkages may now be written as
A== Li
where
(1.2-21)
(1.2-22)
N'lL ] N; ml
L/2 +Lm2
( 1.2-23)
Although the voltage equations with the inductance matrix L incorporated may
be used for purposes of analysis, it is customary to perform a change of variables that
yields the well-known equivalent T circuit of two magnetically coupled coils. To set
the stage for this derivation, let us express the flux linkages from (1.2-22) as
A.I = LlIi, +Lm l (il +Z ~ i2 )
A.2 = L12
i
2 +Lm2 (Z: i
l
+i 2)
(1.2-24)
( 1.2-25)
Now we have two choices. We can use a substitute variable for (N2/ N
I
)i2 or for
(N
J/N2
)i J. Let us consider the first of these choices
( 1.2-26)
whereupon we are using the substitute variable i ~ that, when flowing through coil 1,
produces the same MMF as the actual ;2 flowing through coil 2. This is said to be
referring the current in coil 2 to coil I whereupon coil 1 becomes the reference coil.
On the other hand, if we use the second choice, then
( 1.2-27)
6 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
Here, i ~ is the substitute variable that produces the same MMF when flowing through
coil 2 as iI does when flowing in coil 1. This change of variables is said to refer the
current of coil 1 to coil 2.
We will demonstrate the derivation of the equivalent T circuit by referring the
current of coil 2 to coil 1; thus from (1.2-26) we obtain
(1.2-28)
Power is to be unchanged by this substitution of variables. Therefore,
(1.2-29)
whereupon V2i2 = v ~ i ~ . Flux linkages, which have the units of volt-second, are
related to the substitute flux linkages in the same way as voltages. In particular,
(1.2-30)
If we substitute (1.2-28) into (1.2-24) and (1.2-25) and then multiply (1.2-25)
by NI/N2 to obtain A; and we further substitute (Ni/Nr)Lml for L
m
2 into (1.2-24),
then
where
( )
2
I N,
L/2 = N2 Ln
(1.2-31)
(1.2-32)
(1.2-33)
The voltage equations become
where
(1.2-34)
(1.2-35)
(1.2-36)
MAGNETICALLY COUPLED CIRCUITS 7
Lb
v'
2
Figure 1.2-2 Equivalent circuit with
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ - coil 1selected as the ~ ~ ~ n c e coil.
The above voltage equations suggest the T equivalent circuit shown in Fig. 1.2-2. It
is apparent that this method may be extended to include any number of coils wound
on the same core.
Example lA It is instructive to illustrate the method of deriving an equiva-
lent T circuit from open- and short-circuit measurements. For this purpose let
us assume that when coil 2 of the two-winding transformer shown in Fig. 1.2-1
is open-circuited, the power input to coil 2 is 12 W with an applied voltage is
100 V (rms) at 60 Hz and the current is I A (rms). When coil 2 is short-
circuited, the current flowing in coil 1 is 1 A when the applied voltage is
30 V at 60 Hz. The power during this test is 22 W. If we assume L
II
== Lb,
an approximate equivalent T circuit can be determined from these measure-
ments with coil 1 selected as the reference coil.
The power may be expressed as
(lA-I)
where Vand / are phasors and cP is the phase angle between VI and /1 (power-
factor angle). Solving for cP during the open-circuit test, we have
~ -I PI -I 12 == 83.70
'I' = cos iVdlld = cos 110 x 1
( IA-2)
With VI as the reference phasor and assuming an inductive circuit where i,
lags VI, we obtain
z = ~ I = 11O/rr:. = 12+ 109.3 n
II 1/-83.7 J
(IA-3)
( lA-4)
If we neglect hysteresis (core) losses, which in effect assumes a linear mag-
netic system, then rl == 12O. We also know from the above calculation that
XII + X
ml
== 109.3 O.
For the short-circuit test we will assume that i I == - i ~ because transformers
are designed so that X
ml
I r ~ +jXhl. Hence using (lA-I) again we obtain
22
cP == cos -I --== 42.8
30 x I
8 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
In this case the input impedance is (rl +r ~ ) +j(X
lI
+X;2)' This may be deter-
mined as follows:
30/0 .
Z = 1/ _ 42.80 =22 +J20.40
(IA-5)
Hence, ~ == 10n and, because it is assumed that XII == Xh, both are 10.2 fl.
Therefore X
ml
== 109.3 - 10.2 == 99.1 O. In summary,
'1 == 120 L
ml
== 262.9mH r ~ == IOn
L
II
== 27.1 mH L;2 == 27. I mH
Nonlinear Magnetic System
Although the analysis of transformers and electric machines is generally performed
assuming a linear magnetic system, economics dictate that in the practical design of
these devices some saturation occurs and that heating of the magnetic material exists
due to hysteresis losses. The magnetization characteristics of transformer or
machine materials are given in the form of the magnitude of flux density versus mag-
nitude of field strength (B-H curve) as shown in Fig. 1.2-3. If it is assumed that the
magnetic flux is uniform through most of the core, then B is proportional to q> and H
is proportional to MMF. Hence a plot of flux versus current is of the same shape as
1.6
1.2
l
;> 0.8
~
0.4
200 400 600 800
H,Alm
Figure 1.2-3 B-H curve for typical silicon steel used in transformers.
MAGNETICALLY COUPLED CIRCUITS 9
the B-H curve. A transformer is generally designed so that some saturation occurs
during normal operation. Electric machines are also designed similarly in that a
machine generally operates slightly in the saturated region during normal, rated
operating conditions. Because saturation causes coefficients of the differential equa-
tions describing the behavior of an electromagnetic device to be functions of the coil
currents, a transient analysis is difficult without the aid of a computer. Our purpose
here is not to set forth methods of analyzing nonlinear magnetic systems. This pro-
cedure is quite straightforward for steady-state operation, but it cannot be used when
analyzing the dynamics of electromechanical devices [I]. A method of incorporating
the effects of saturation into a computer representation is of interest.
Computer Simulation of Coupled Circuits
Formulating the voltage equations of stationary coupled windings appropriate for
computer simulation is straightforward and yet this technique is fundamental to
the computer simulation of ac machines. Therefore it is to our advantage to consider
this method here. For this purpose let us first write (1.2-31) and (1.2-32) as
Al == L
II
i
l
+Am (1.2-37)
A; == +Am (1.2-38)
where
Am == Lml(i
l
+i;)
Solving (1.2-37) and (J .2-38) for the currents yields
1
i
l
== - (AI - Am)
L/I
= - Am)
/2
( 1.2-39)
(1.2-40)
(1.2-41 )
If (1.2-40) and (1.2-41) are substituted into the voltage equations (1.2-34) and
(1.2-35) and we solve the resulting equations for flux linkages, the following equa-
tions are obtained:
( 1.2-42)
( 1.2-43)
( 1.2-44)
where
(
I 1 1 )-1
La == -+-+-
L
tnl
L
I I
L
12
(1.2-45)
10 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
We now have the equations expressed with Al and A ~ as state variables. In the com-
puter simulation, (1.2-42) and (1.2-43) are used to solve for AI and A ~ and (1.2-44) is
used to solve for Am. The currents can then be obtained from (1.2-40) and (1.2-41). It
is clear that (1.2-44) could be substituted into (1.2-40)-( 1.2-43) and that Am could be
eliminated from the equations, whereupon it would not appear in the computer simu-
lation. However, we will find Am (the mutual flux linkages) an important variable
when we include the effects of saturation.
If the magnetization characteristics (magnetization curve) of the coupled winding
is known, the effects of saturation of the mutual flux path may be readily incorpo-
rated into the computer simulation. Generally, the magnetization curve can be ade-
quately determined from a test wherein one of the windings is open-circuited
(winding 2, for example), and the input impedance of the other winding (winding
1) is determined from measurements as the applied voltage is increased in magni-
tude from zero to say 150% of the rated value. With information obtained from this
type of test, we can plot Am versus (i I + i ~ ) as shown in Fig. 1.2-4 wherein the slope
of the linear portion of the curve is L;I . From Fig. 1.2-4, it is clear that in the region
of saturation we have
(1.2-46)
Figure 1.2-4 Magnetization curve.
ELECTROMECHANICAL ENERGY CONVERSION 11
Figure 1.2-5 f(A
m)
versusAm
from Fig. 1.2-4.
wheref(A
m)
may be determined from the magnetization curve for each value of Am.
In particular,f(A
m
) is a function of Am as shown in Fig. 1.2-5. Therefore, the effects
of saturation of the mutual flux path may be taken into account by replacing (1.2-39)
with (1.2-46) for Am. Substituting (1.2-40) and (1.2-41) for ; I and i ~ , respectively,
into (1.2-46) yields the following computer equation for Am:
( 1.2-47)
Hence the computer simulation for including saturation involves replacing Am given
by (1.2-44) with (1.2-47) where f (Am) is a generated function of Am determined from
the plot shown in Fig. 1.2-5.
1.3 ELECTROMECHANICAL ENERGY CONVERSION
Although electromechanical devices are used in some manner in a wide variety of
systems, electric machines are by far the most common. It is desirable, however, to
establish methods of analysis that may be applied to all electromechanical devices.
Energy Relationships
Electromechanical systems are comprised of an electrical system, a mechanical sys-
tem, and a means whereby the electrical and mechanical systems can interact. Inter-
action can take place through any and all electromagnetic and electrostatic fields that
are common to both systems, and energy is transferred from one system to the other
as a result of this interaction. Both electrostatic and electromagnetic coupling fields
may exist simultaneously, and the electromechanical system may have any number
of electrical and mechanical systems. However, before considering an involved sys-
tem it is helpful to analyze the electromechanical system in a simplified form. An
electromechanical system with one electrical system, one mechanical system, and
12 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
Figure 1.3-1 Block diagram of an elementary electromechanical system.
one coupling field is depicted in Fig. 1.3-1. Electromagnetic radiation is neglected,
and it is assumed that the electrical system operates at a frequency sufficiently low so
that the electrical system may be considered as a lumped parameter system.
Losses occur in all components of the electromechanical system. Heat loss will
occur in the mechanical system due to friction, and the electrical system will dissi-
pate heat due to the resistance of the current-carrying conductors. Eddy current and
hysteresis losses occur in the ferromagnetic material of all magnetic fields, whereas
dielectric losses occur in all electric fields. If WE is the total energy supplied by the
electrical source and WM is the total energy supplied by the mechanical source, then
the energy distribution could be expressed as
WE == We + WeL +WeS
W
M
== W
m
+ W
mL
+W
mS
(1.3-1 )
(1.3-2)
In (1.3-1), W
eS
is the energy stored in the electric or magnetic fields that are not
coupled with the mechanical system. The energy WeL is the heat losses associated
with the electrical system. These losses occur due to the resistance of the current-
carrying conductors as well as the energy dissipated from these fields in the form of
heat due to hysteresis, eddy currents, and dielectric losses. The energy Weis the energy
transferred to the coupling field by the electrical system. The energies common to the
mechanical system may be defined in a similar manner. In (1.3-2), WmS is the energy
stored in the moving member and compliances of the mechanical system, WmL is the
energy losses of the mechanical system in the form of heat, and Wm is the energy
transferred to the coupling field. It is important to note that with the convention adopt-
ed, the energy supplied by either source is considered positive. Therefore, WE(W
M
)
is negative when energy is supplied to the electrical source (mechanical source).
If WF is defined as the total energy transferred to the coupling field, then
(1.3-3)
where WI is energy stored in the coupling field and WjZ. is the energy dissipated in the
form of heat due to losses within the coupling field (eddy current, hysteresis, or
dielectric losses). The electromechanical system must obey the law of conservation
of energy; thus
which may be written
W
f
+WfL == We + Wm
This energy relationship is shown schematically in Fig. 1.3-2.
(1.3-4)
(1.3-5)
ELECTROMECHANICAL ENERGY CONVERSION 13
+ W", +
+
W ~ s WI WmS
r-- Electricalsystem-+-Couplingfield+ Mechanical system-1
Figure 1.3-2 Energy balance.
The actual process of converting electrical energy to mechanical energy (or vice
versa) is independent of (1) the loss of energy in either the electrical or the mechan-
ical systems (W
eL
and W
mL
) , (2) the energies stored in the electric or magnetic fields
that are not common to both systems (W
eS
) , or (3) the energies stored in the mechan-
ical system (Ws). If the losses of the coupling field are neglected, then the field is
conservative and (1.3-5) becomes
( 1.3-6)
Examples of elementary electromechanical systems are shown in Figs. 1.3-3 and
1.3-4. The system shown in Fig. 1.3-3 has a magnetic coupling field while the elec-
tromechanical system shown in Fig. 1.3-4 employs an electric field as a means of
transferring energy between the electrical and mechanical systems. In these systems,
v is the voltage of the electric source andf is the external mechanical force applied to
the mechanical system. The electromagnetic or electrostatic force is denoted by fee
The resistance of the current-carrying conductors is denoted by r, and I is the induc-
tance of a linear (conservative) electromagnetic system that does not couple the
mechanical system. In the mechanical system, M is the mass of the movable member
while the linear compliance and damper are represented by a spring constant Kand a
v
Figure 1.3-3 Electromechanical system with a magnetic field.
14 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
v
r
M
-q
Xo I
x
f
f ~
D ~
Figure 1.3-4 Electromechanical system with an electric field.
damping coefficient D, respectively. The displacement Xo is the zero force or equili-
brium position of the mechanical system which is the steady-state position of the
mass with Ie and f equal to zero. A series or shunt capacitance may be included
in the electrical system wherein energy would also be stored in an electric field
external to the electromechanical process.
The voltage equation that describes both electrical systems may be written as
(1.3-7)
where ef is the voltage drop across the coupling field. The dynamic behavior of the
translational mechanical system may be expressed by employing Newton's law of
motion. Thus,
d
2x
dx
f == M-+ D-+K(x -xo) - h
dt2 dt e
The total energy supplied by the electric source is
WE = Jvidt
The total energy supplied by the mechanical source is
which may also be expressed as
(1.3-8)
(1.3-9)
(1.3-10)
(1.3-11)
ELECTROMECHANICAL ENERGY CONVERSION 15
Substituting (1.3-7) into (1.3-9) yields
WE = r JPdt +I Jidi +Jeji dt
( 1.3-12)
The first term on the right-hand side of (1.3-12) represents the energy loss due to the
resistance of the conductors (W
eL
) . The second term represents the energy stored in
the linear electromagnetic field external to the coupling field (Wes). Therefore, the
total energy transferred to the coupling field from the electrical system is
(1.3-13)
Similarly, for the mechanical system we have
(1.3-14)
Here, the first and third terms on the right-hand side of ( 1.3-14) represent the energy
stored in the mass and spring, respectively (W
mS
)' The second term is the heat loss
due to friction (WmL). Thus, the total energy transferred to the coupling field from
the mechanical system is
(1.3-15)
It is important to note that a positive force,!e, is assumed to be in the same direction
as a positive displacement, dx. Substituting (1.3-13) and (1.3-15) into the energy bal-
ance relation, (1.3-6), yields
( 1.3-16)
The equations set forth may be readily extended to include an electromechanical
system with any number of electrical and mechanical inputs to any number of cou-
pling fields. Considering the system shown in Fig. 1.3-5, the energy supplied to the
coupling fields may be expressed as
J K
LWej+ LWmk
j=1 k=1
( 1.3-17)
wherein J electrical and K mechanical inputs exist. The total energy supplied to the
coupling field at the electrical inputs is
J J
=
(1.3-18)
16 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
Wet
....
W
m1
,
.....
W
e2
.....
..,
W
m2
I
,
.....
Couplingfields
I
I I
I I
I I
I I
W
u
I
....
I W
mK
~
,
.....
Figure 1.3-5 Multiple electrical and
mechanical inputs.
The total energy supplied to the coupling field from the mechanical inputs is
(1.3-19)
The energy balance equation becomes
In differential form we obtain
J K
dW
f
== L efjij dt - Lhkdx;
j=1 k=1
(1.3-20)
(1.3-21)
Energy in Coupling Fields
Before using (1.3-21) to obtain an expression for the electromagnetic force fe, it is
necessary to derive an expression for the energy stored in the coupling fields. Once
we have an expression for WI, we can take the total derivative to obtain dW
I,
which
can then be substituted into (1.3-21). When expressing the energy in the coupling
fields it is convenient to neglect all losses associated with the electric and magnetic
fields whereupon the fields are assumed to be conservative and the energy stored
therein is a function of the state of the electrical and mechanical variables. Although
the effects of the field losses may be functionally accounted for by appropriately
introducing a resistance in the electric circuit, this refinement is generally not nece-
ssary because the ferromagnetic material is selected and arranged in laminations so
as to minimize the hysteresis and eddy current losses. Moreover, nearly all of the
energy stored in the coupling fields is stored in the air gaps of the electromechanical
device. Because air is a conservative medium, all of the energy stored therein can be
ELECTROMECHANICAL ENERGY CONVERSION 17
returned to the electrical or mechanical systems. Therefore, the assumption of loss-
less coupling fields is not as restrictive as it might first appear.
The energy stored in a conservative field is a function of the state of the system
variables and not the manner in which the variables reached that state. It is conve-
nient to take advantage of this feature when developing a mathematical expression
for the field energy. In particular, it is convenient to fix mathematically the position
of the mechanical systems associated with the coupling fields and then excite the
electrical systems with the displacements of the mechanical systems held fixed.
During the excitation of the electrical systems, Wmk is zero even though electroma-
gnetic or electrostatic forces occur. Therefore, with the displacements held fixed the
energy stored in the coupling fields during the excitation of the electrical systems is
equal to the energy supplied to the coupling fields by the electrical systems. Thus,
with Wmk == 0, the energy supplied from the electrical system may be expressed
from (1.3-20) as
(1.3-22)
It is instructive to consider a singly excited electromagnetic system similar to that
shown in Fig. 1.3-3. In this case, ef == dA/dt and (1.3-22) becomes
(1.3-23)
Here J == 1; however, the subscript is omitted for the sake of brevity. The area to the
left of the A-i relationship (shown in Fig. 1.3-6) for a singly excited electromagnetic
device is the area described by (1.3-23). In Fig. 1.3-6, this area represents the energy
stored in the field at the instant when A== A
a
and i == i
a
The A-i relationship need not
be linear; it need only be single-valued, a property that is characteristic to a conser-
vative or lossless field. Moreover, because the coupling field is conservative, the
energy stored in the field with A== A
a
and i == i
a
is independent of the excursion
of the electrical and mechanical variables before reaching this state.
The area to the right of the A-i curve is called the coenergy and is expressed as
which may also be written as
We = JAdi
( 1.3-24)
(1.3-25)
Although the coenergy has little or no physical significance, we will find it a conve-
nient quantity for expressing the electromagnetic force. It should be clear that for a
linear magnetic system where the A-i plots are straight-line relationships, W
f
== We.
18 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
dA
di
Figure 1.3-6 Stored energy and coenergy in a magnetic field of a singly excited electro-
magnetic device.
The displacement x defines completely the influence of the mechanical system
upon the coupling field; however, because Aand i are related, only one is needed
in addition to x in order to describe the state of the electromechanical system. There-
fore, either Aand x or i and x may beselected as independent variables. If i and x are
selected as independent variables, it is convenient to express the field energy and the
fl ux linkages as
W
f
== Wf{i,x)
A== A(i,x)
( 1.3-26)
( 1.3-27)
With i and x as independent variables we must express dAin terms of di before sub-
stituting into (1.3-23). Thus from (1.3-27)
d
' ( ) == oA(i,x) di oA(i,x) d
It I, x ai I + ax x
(1.3-28)
In the derivation of an expression for the energy stored in the field, dx is set equal to
zero. Hence, in the evaluation of field energy, dl is equal to the first term on the
ELECTROMECHANICAL ENERGY CONVERSION 19
right-hand side of (1.3-28). Substituting into (1.3-23) yields
(1.3-29)
where is the dummy variable of integration. Evaluation of (1.3-29) gives the
energy stored in the field of the singly excited system. The coenergy in terms of i
and x may be evaluated from (1.3-24) as
Wc(i,x) =JA(i,x)di =L
With A. and x as independent variables we have
W
f
== Wf{A.,x)
i == i(A,x)
The field energy may be evaluated from (1.3-23) as
(1.3-30)
(1.3-31)
( 1.3-32)
(1.3-33)
In order to evaluate the coenergy with A. and x as independent variables, we need to
express di in terms of dA.; thus from (1.3-32) we obtain
d
( '1 ) == 8i(A., x) d'1 8i(A., x) d
l /I., X aA /I. + ax x
Because dx == 0 in this evaluation, (1.3-24) becomes
W.(A ) == JA 8i(A,X) dA == Jie dl'
c ,x 8A 0
( 1.3-34)
(1.3-35)
For a linear electromagnetic system the A-i plots are straight-line relationships;
thus for the singly excited system we have
or
A(i,x) = L{x)i
i(A,X) =
(1.3-36)
( 1.3-37)
20 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
Let us evaluate From (1.3-28), with dx == 0
== L(x)di
Hence, from (1.3-29)
(1.3-38)
(1.3-39)
It is left to the reader to show that Wf(A, x), Wc(i, x), and x) are equal to
(1.3-39) for this magnetically linear system.
The field energy is a state function, and the expression describing the field energy
in terms of system variables is valid regardless of the variations in the system vari-
ables. For example, (1.3-39) expresses the field energy regardless of the variations in
L(x) and i. The fixing of the mechanical system so as to obtain an expression for the
field energy is a mathematical convenience and not a restriction upon the result.
In the case of a multiexcited, electromagnetic system, an expression for the field
energy may be obtained by evaluating the following relation with dx, == 0:
( 1.3-40)
Because the coupling fields are considered conservative, (1.3-40) may be evaluated
independently of the order in which the flux linkages or currents are brought to their
final values. To illustrate the evaluation of (1.3-40) for a muItiexcited system, we
will allow the currents to establish their final states one at a time while all other cur-
rents are mathematically fixed in either their final or unexcited state. This procedure
may be illustrated by considering a doubly excited electric system with one mechan-
ical input. An electromechanical system of this type could be constructed by placing
a second coil, supplied from a second electrical system, on either the stationary or
movable member of the system shown in Fig. 1.3-3. In this evaluation it is conve-
nient to use currents and displacement as the independent variables. Hence, for a
doubly excited electric system we have
(1.3-41 )
In this determination of an expression for W
f
, the mechanical displacement is held
constant (dx == 0); thus (1.3-41) becomes
(1.3-42)
ELECTROMECHANICAL ENERGY CONVERSION 21
We will evaluate the energy stored in the field by employing (1.3-42) twice. First we
will mathematically bring the current i I to the desired value while holding i2at zero.
Thus, iJ is the variable of integration and di2 == O. Energy is supplied to the coupling
field from the source connected to coil 1. As the second evaluation of (1.3-42), ii is
brought to its desired current while holding i I at its desired value. Hence, i: is the
variable of integration and di I == O. During this time, energy is supplied from both
sources to the coupling field because i I dAI is nonzero. The total energy stored in the
coupling field is the sum of the two evaluations. Following this two-step procedure
the evaluation of (1.3-42) for the total field energy becomes
(1.3-43)
which should be written as
( 1.3-44)
The first integral on the right-hand side of (1.3-43) or (1.3-44) results from the first
step of the evaluation with iI as the variable of integration and with i
2
== 0 and
di
2
== O. The second integral comes from the second step of the evaluation with
iI == ii, di I == 0 and i
2
as the variable of integration. It is clear that the order of allow-
ing the currents to reach their final state is irrelevant; that is, as our first step, we
could have made i
2
the variable of integration while holding i) at zero idi, == 0)
and then let i I become the variable of integration while holding i: at its final variable.
The results would be the same. It is also clear that for three electrical inputs the eva-
luation procedure would require three steps, one for each current to be brought math-
ematically to its final state.
Let us now evaluate the energy stored in a magnetically linear electromechanical
system with two electrical inputs and one mechanical input. For this let
AI (ii, i
2
,x) == L
I I
(x)il + L
I2(X)i2
A2(ii, ;2,x) == L21 (x)il + L22 (X)i2
With that mechanical displacement held constant (dx == 0), we obtain
d)q(i
l,i2,X)
== LI1(x)di) + L
I2(X)
di2
dA2(i" i2,X) == L,2(X)di, + L
22(X)
di;
( 1.3-45)
( 1.3-46)
( 1.3-47)
( 1.3-48)
22 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
It is clear that the coefficients on the right-hand side of (1.3-47) and (1.3-48) are the
partial derivatives. For example, L
1
l(x) is the partial derivative of Al(i1, i2'x) with
respect to iI. Appropriate substitution into (1.3-44) gives
(1.3-49)
which yields
(1.3-50)
The extension to a linear electromagnetic system with J electrical inputs is straight-
forward whereupon the following expression for the total field energy is obtained:
(1.3-51)
It is left to the reader to show that the equivalent of (1.3-22) for a multiexcited
electrostatic system is
(1.3-52)
Graphical Interpretation of Energy Conversion
Before proceeding to the derivation of expressions for the electromagnetic force, it is
instructive to consider briefly a graphical interpretation of the energy conversion
process. For this purpose let us again refer to the elementary system shown in
Fig. 1.3-3 and let us assume that as the movable member moves from x == X
a
to
x =Xb, where Xb < X
a
, the A-i characteristics are given by Fig. 1.3-7. Let us further
assume that as the member moves from X
a
to Xb the A-i trajectory moves from point
A to point B. It is clear that the exact trajectory from A to B is determined by the
combined dynamics of the electrical and mechanical systems. Now, the area
OACOrepresents the original energy stored in field; area OBDOrepresents the final
energy stored in the field. Therefore, the change in field energy is
~ WI == area OBDO- area OACO
The change in We, denoted as ~ We' is
J
A.B
~ We == idA. =area CABDC
A.A
(1.3-53)
(1.3-54)
ELECTROMECHANICAL ENERGY CONVERSION 23
o
Figure 1.3-7 Graphical representation of electromechanical energy conversion for A-i path
A to B.
We know that
(1.3-55)
Hence,
~ W
m
== area OBDO - area OACO - area CABDC == -area OABO (1.3-56)
The change in Wm, denoted as ~ W m , is negative; energy has been supplied to the
mechanical system from the coupling field part of which came from the energy
stored in the field and part from the electrical system. If the member is now moved
back to X
a
, the A-i trajectory may be as shown in Fig. ].3-8. Hence the ~ Wm is still
area OABO but it is positive, which means that energy was supplied from the
mechanical system to the coupling field, part of which is stored in the field and
part of which is transferred to the electrical system. The net ~ Will for the cycle
from A to B back to A is the shaded area shown in Fig. 1.3-9. Because ~ ~ f is
zero for this cycle
(1.3-57)
For the cycle shown the net ~ We is negative, thus LlW
m
is positive; we have gen-
erator action. If the trajectory had been in the counterclockwise direction, the net
~ W e would have been positive and the net ~ W m would have been negative, which
would represent motor action.
24 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
A
o
Figure1.3-8 Graphical representation of electromechanical energy conversion for A-i path
B toA.
o
Figure1.3-9 Graphical representation of electromechanical energy conversion for A-i path
A to B to A.
ELECTROMECHANICAL ENERGY CONVERSION 25
Electromagnetic and Electrostatic Forces
The energy balance relationships given by (1.3-21) may be arranged as
K J
L fekdx, == L efj ij dt - dWf
k=1 j=1
( 1.3-58)
In order to obtain an expression for feh it is necessary to first express Wf and then
take its total derivative. One is tempted to substitute the integrand of (1.3-22) into
(1.3-58) for the infinitesimal change of field energy. This procedure is, of course,
incorrect because the integrand of (1.3-22) was obtained with all mechanical
displacements held fixed (dXk == 0), where the total differential of the field energy
is required in (1.3-58).
The force or torque in any electromechanical system may be evaluated by
employing (1.3-58). In many respects, one gains a much better understanding of
the energy conversion process of a particular system by starting the derivation of
the force or torque expressions with (1.3-58) rather than selecting a relationship
from a table. However, for the sake of completeness, derivation of the force equa-
tions will be set forth and tabulated for electromechanical systems with K mechan-
ical inputs and J electrical inputs [2].
For an electromagnetic system, (1.3-58) may be written as
K J
L fekdx; == L ij dAj - dW
f
k=1 j=1
With i
j
and Xk selected as independent variables we have
~ f == ~ f ( i l , ,iJ;XI, ,XK)
Aj == Aj (u, ,iJ; X I , , XK )
From (1.3-60) and (1.3-61) we obtain
( 1.3-59)
(1.3-60)
(1.3-61 )
(1.3-62)
(1.3-63)
In (1.3-62) and (1.3-63) and hereafter in this development the functional notation of
(i I , ... , i
J
; XI , ... ,XK) is abbreviated as (i
j
, Xk). The index n is used so as to avoid
confusion with the index j because each d ~ i must be evaluated for changes in
all currents in order to account for mutual coupling between electrical systems.
26 BASIC PRINCIPLES FOR ELECTRtC MACHINE ANALYSIS
[Recall that we did this in (1.3-42) for J == 2.] Substituting (1.3-62) and (1.3-63)
into (1.3-59) yields
(1.3-64)
By gathering terms, we obtain
(1.3-65)
When we equate coefficients, we obtain
( 1.3-66)
(1.3-67)
Although (1.3-67) is of little practical importance, (1.3-66) can be used to eval-
uate the force at the kth mechanical terminal of an electromechanical system with
only magnetic coupling fields and with i
j
and Xk selected as independent variables. A
second force equation with i
j
and Xk as independent variables may be obtained from
(1.3-66) by incorporating the expression for coenergy. For a multiexcited system the
coenergy may be expressed as
J
We == LijAj - WI
j=1
(1.3-68)
Because i
j
and Xk are independent variables, the partial derivative with respect to x is
(1.3-69)
Hence, substituting (1.3-69) into (1.3-66) yields
(1.3-70)
ELECTROMECHANICAL ENERGY CONVERSION 27
Table 1.3-1 Electromagnetic Force at kth Mechanical
Input"
"For rotational systems replace f('k with T
ek
and x, with (h.
It should be recalled that positiveji, and positive dx, are in the same direction. Also,
if the magnetic system is linear, then We == ~ ' .
By a procedure similar to that used above, force equations may be developed for
magnetic coupling with Aj and Xk as independent variables. These relations are given
in Table 1.3-1 without proof. In Table ).3-1 the independent variables to be used are
designated in each equation by the abbreviated functional notation. Although only
translational mechanical systems have been considered, all force relationships
developed herein may be modified for the purpose of evaluating the torque in rota-
tional systems. In particular, when considering a rotational system,!ek is replaced
with the electromagnetic torque Tek, and Xk is replaced with the angular displace-
ment Ok. These substitutions are justified because the change of mechanical energy
in a rotational system is expressed as
(1.3-71)
The force equation for an electromechanical system with electric coupling
fields may be derived by following a procedure similar to that used in the case
of magnetic coupling fields. These relationships are given in Table 1.3-2 without
explanation.
It is instructive to derive the expression for the electromagnetic force of a singly
excited electrical system as shown in Fig. 1.3-3. It is clear that the expressions given
in Table 1.3-1 are valid for magnetically linear or nonlinear systems. If we assume
that the magnetic system is linear, then A(i,x) is expressed by (1.3-36) and Wf(i,x) is
expressed by (1.3-39), which is also equal to the coenergy. Hence, either the first or
second entry of Table 1.3-1 can be used to express fe. In particular,
E (. ) _ aWc(i,x) _ I .2 dL(x)
se ,,X - ax - 2' ~
(1.3-72)
28 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
Table 1.3-2 Electrostatic Force at kth Mechanical
Input"
"For rotational systems replace Jek with Tek and Xk with fh.
With the convention established, a positive electromagnetic force is assumed to
act in the direction of increasing x. Thus with (1.3-15) expressed in differential
form as
dW
m
== -!edx ( 1.3-73)
we see that energy is supplied to the coupling field from the mechanical system when
Ie and dx are opposite in sign, and energy is supplied to the mechanical system from
the coupling field when Ie and dx are the same in sign.
From (1.3-72) it is apparent that when the change of L(x) with respect to x is
negative, Ie is negative. In the electromechanical system shown in Fig. 1.3-3 the
change L(x) with respect to x is always negative; therefore, the electromagnetic force
is in the direction so as to pull the movable member to the stationary member. In
other words, an electromagnetic force is set up so as to maximize the inductance
of the coupling system, or, since reluctance is inversely proportional to the induc-
tance, the force tends to minimize the reluctance. Because Ie is always negative in
the system shown in Fig. 1.3-3, energy is supplied to the coupling field from the
mechanical system (generator action) when dx is positive and is supplied from the
coupling field to the mechanical system (motor action) when dx is negative.
Steady-State and Dynamic Performance
of an Electromechanical System
It is instructive to consider the steady-state and dynamic performance of the elemen-
tary electromagnetic system shown in Fig. 1.3-3. The differential equations that
describe this system are given by (1.3-7) for the electrical system and by (1.3-8)
for the mechanical system. The electromagnetic force Ie is expressed by (1.3-72).
If the applied voltage, v, and the applied mechanical force, I, are constant, all
ELECTROMECHANICAL ENERGY CONVERSION 29
derivatives with respect to time are zero during steady-state operation, and the beha-
vior can be predicted by
v == ri
I == K(x - xo) - Ie
Equation (1.3-75) may be written as
-Ie == I - K(x - xo)
( 1.3-74)
(1.3-75)
( 1.3-76)
The magnetic core of the system in Fig. 1.3-3 is generally constructed of ferro-
magnetic material with a relative permeability on the order of 2000 to 4000. In this
case the inductance L(x) can be adequately approximated by
k
L(x) == -
x
(1.3-77)
In the actual system the inductance will be a large finite value rather than infinity, as
predicted by (1.3-77), when x == O. Nevertheless, (1.3-77) is quite sufficient to illus-
trate the action of the system for x > O. Substituting (1.3-77) into (1.3-72) yields
ki
2
Ie(i,x) == - -22
x
( 1.3-78)
A plot of (1.3-76), with Ie replaced by (1.3-78), is shown in Fig. 1.3-10 for the
following system parameters:
r == Ion
K == 2667N/m
Xo == 3 mm
k == 6.293 x 10-
5
H . m
In Fig. 1.3-10, the plot of the negative of the electromagnetic force is for an applied
voltage of 5 V and a steady-state current of 0.5 A. The straight lines represent the
right-hand side of (1.3-76) with j' == 0 (lower straight line) and I == 4N (upper
straight line). Both lines intersect the -Iecurve at two points. In particular, the upper
line intersects the -Iecurve at 1 and 1'; the lower line intersects at 2 and 2'. Stable
operation occurs at only points 1 and 2. The system will not operate stably at points
l' and 2'. This can be explained by assuming that the system is operating at one of
these points (1' and 2') and then showing that any system disturbance whatsoever
will cause the system to move away from these points. If, for example, x increases
slightly from its value corresponding to point I', the restraining force I - K(x - xo)
is larger in magnitude than -fe, and x will continue to increase until the system
reaches operating point J. If x increases beyond its value corresponding to operating
point I, the restraining force is less than the electromagnetic force. Therefore, the
system will establish steady-state operation at I. If, on the other hand, x decreases
30 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
-Iefor i =O.5A
12
10
8
4
2
o
Kxo
2
x,mm
3
-K(x - xo)
4 5
Figure 1.3-10 Steady-state operation of electromechanical system shown in Fig. 1.3-3.
from point 1', the electromagnetic force is larger than the restraining force. There-
fore, the movable member will move until it comes in contact with the stationary
member (x == 0). It is clear that the restraining force that yields a straight line below
the -Ie curve will not permit stable operation with x > o.
The dynamic behavior of the system during step changes in the source voltage v is
shown in Fig. 1.3-11, and Figs. 1.3-12 and 1.3-13 show dynamic behavior during
step changes in the applied force f. The following system parameters were used
in addition to those given previously:
1==0 M == 0.055 kg D == 4N slm
The computer traces shown in Fig. 1.3-11 depict the dynamic performance of the
example system when the applied voltage is stepped from zero to 5 V and then
back to zero with the applied mechanical force held equal to zero. The system vari-
ables are el' A, i,fe, x, We, WI, and W
m.
The energies are plotted in millijoules (mJ).
Initially the mechanical system is at rest with x == Xo (3 mm). When the source
5
ELECTROMECHANICAL ENERGY CONVERSION 31
L Source voltage decreased
'from 5 to 0 V
t-Source voltage increased
from 0 to 5 V
~ - - - -
005LJ ~ ;, A
---
f.,N .r'..__----J(------
~ IE- 0.02 s
:f
x,mm
We,mJ
: L ~
W/7mJ
:u---
~ [
Wm,rnJ
\--------
\_----
Figure 1.3-11 Dynamic performance of the electromechanical system shown in Fig. 1.3-3
during step changes in the source voltage.
voltage is applied, x decreases; and when steady-state operation is reestablished, x is
approximately 2.5 mm. Energy enters the coupling field via We. The bulk of this
energy is stored in the field (Wf) with a smaller amount transferred to the mechanical
system, some of which is dissipated in the damper during the transient period while
the remainder is stored in the spring. When the applied voltage is removed, the elec-
trical and mechanical systems return to their original states. The change in Wm
32 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
0.5
e"V
0
-0.5
0OO2
f
'\, Wbt 0.01
0
'''---------

l
Force decreased from
4 to zero N
0.02s

i, A

t: N
-:t
-2
x,mm
:t
increased from
zero to 4 N
We,mJ
;[
4[
Wf,mJ
0
Wm,mJ

-4
Figure 1.3-12 Dynamic performance of the electromechanical system shown in Fig. 1.3-3
during step changes in the applied force.
is small, increasing only slightly. Hence, during the transient period there is an
interchange of energy between the spring and mass which is finally dissipated in
the damper. The net change in WI during the application and removal of the applied
voltage is zero; hence the net change in We is positive and equal to the negative of the
net change in W
m
The energy transferred to the mechanical system during this cycle
is dissipated in the damper, because f is fixed at zero, and the mechanical system
returns to its initial rest position with zero energy stored in the spring.
In Fig. 1.3-12, the initial state is that shown in Fig. 1.3-11 with 5 Vapplied to the
electrical system. The mechanical force f is increased from zero to 4 N, whereupon
energy enters the coupling field from the mechanical system. Energy is transferred
ELECTROMECHANICAL ENERGY CONVERSION 33
i
0.016
0.012
~ 0.008
..(
0.004
/
/
/
x = 2.5 mm-"
/
/
/
/ /
/ /
/ ./
/ //
/ ./
/ ./
/ /./
/ ./ x = 4.3 mm
/ ./
/ ./
/ //
/ ./
/ ./
/.//
o
i, A
Figure 1.3-13 Systemresponse shown in Fig. 1.3-3.
from the coupling field to the electrical system, some coming from the mechanical
system and some from the energy originally stored in the magnetic field. Next the
force is stepped back to zero from 4 N. The electrical and mechanical systems return
to their original states. During the cycle a net energy has been transferred from the
mechanical system to the electrical system which is dissipated in the resistance. This
cycle is depicted on the A-i plot shown in Fig. 1.3-13.
Example IB It is helpful to formulate an expression for the electromagnetic
torque of the elementary rotational device shown in Fig. 1B-1. This device
consists of two conductors. Conductor 1 is placed on the stationary member
(stator); conductor 2 is fixed on the rotating member (rotor). The crossed lines
inside a circle indicate that the assumed direction of positive current flow is
into the paper (we are seeing the tail of the arrow), whereas a dot inside a circle
indicates positive current flow is out of the paper (the point of the arrow). The
length of the air gap between the stator and rotor is shown exaggerated relative
to the inside diameter of the stator.
34 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
Stator
Stator
(a) (b)
Figure 1B-1 Elementary rotational electromechanical device. (a) End view; (b) cross-
sectional view.
The voltage equations may be written as
(IB- I)
(lB-2)
where rl and r2 are the resistances of conductor 1 and 2, respectively. The
magnetic system is assumed linear; therefore the flux linkages may be
expressed as
Al = Lllil +L l 2
i
2
A2 == L21 i) +L22
i
Z
(IB-3)
( IB-4)
The self-inductances LII and L22 are constant. Let us assume that the mutual
inductance may be approximated by
(IB-5)
where (Jr is defined in Fig. IB- I. The reader should be able to justify the form
of (lB-5) by considering the mutual coupling between the two conductors as
(Jr varies from 0 to 2n radians.
Because the magnetic system is assumed to be linear, we have
1 1
Wc(il, ii- Or) =2LIl ii +L12i 1
i
2 +2 L 2 2 i ~
Substituting into (1B-6) yields
(IB-6)
(IB-7)
(lB-8)
MACHINE WINDINGS AND AIR-GAP MMF 35
Figure IB-2 Stator and rotor poles for constant currents.
Consider for a moment the form of the torque if i I and ;2 are both constant.
For the positive direction of current shown, the torque is of the form
T, == -Ksin(J, ( IB-9)
where K is a positive constant. We can visualize the production of torque by
considering the interaction of the magnetic poles produced by the current
flowing in the conductors. If both iI and i: are positive, the poles produced
are as shown in Fig. I B-2. One should recall that flux issues from a north
pole. Also, the stator and rotor each must be considered as separate electro-
magnetic systems. Thus, flux produced by the I-I' winding issues from the
north pole of the stator into the air gap. Similarly, the flux produced by the
2-2' winding enters the air gap from the north pole of the rotor. It is left to
the reader to justify the fact that the range of (J, over which stable operation
can occur for the expression of electromagnetic torque given by (I B-9)
is -n/2 ::; (J, ::; n/2.
1.4 MACHINE WINDINGS AND AIR-GAP MMF
For the purpose of discussing winding configurations in rotating machines and the
resulting air-gap MMF as well as the calculations of machine inductances, it is con-
venient to begin with the elementary 2-pole, 3-phase, wye-connected salient-pole
synchronous machine shown in Fig. 1.4-1. Once these concepts are established
for this type of a machine, they may be readily modified to account for all types
of induction machines and easily extended to include the synchronous machine
with short-circuited windings on the rotor (damper windings).
The stator windings of the synchronous machine are embedded in slots around
the inside circumference of the stationary member. In the 2-pole machine, each
phase winding of the 3-phase stator winding is displaced 120
0
with respect to
each other as illustrated in Fig. 1.4-1. The field or fd winding is wound on the rotat-
ing member. The as, bs, cs, andfd axes denote the positive direction of the flux pro-
duced by each of the windings. The as, bs, and cs windings are identical in that each
winding has the same resistance and the same number of turns. When a machine has
three identical stator windings arranged as shown in Fig. 1.4-1, it is often referred to
36 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
bs axis
lj C'
I
Figure 1.4-1 Elementary, 2-pole, 3-phase, wye-connected salient-pole synchronous
machine.
MACHINE WINDINGS AND AIR-GAP MMF 37
as a machine with symmetrical stator windings. We will find that the symmetri-
cal induction machine has identical multiphase stator windings and identical multi-
phase rotor windings. An unsymmetrical induction machine has nonidentical
multiphase stator windings (generally 2-phase) and symmetrical multiphase rotor
windings.
In Fig. 1.4-1, it is assumed that each coil spans 1t radians of the stator for a 2-pole
machine. One side of the coil (coil side) is represented by a indicating that the
assumed positive direction of current is down the length of the stator (into the paper).
The 0 indicates that the assumed positive direction of current is out of the paper.
Each coil contains n; conductors. Therefore, in the case of the as winding, positive
current flows in a conductor of coil aI, then axially down the length of the stator and
back the length of the stator and out at coil side a ~ . This is repeated for n
c
conduc-
tors. The last conductor of the coil aI - a ~ is then placed in the appropriate slot so as
to start coil a2-a; wherein the current flows down the stator via coil side a2 and then
back through a; and so on until a ~ . The bs and cs windings are arranged similarly,
and the last conductors of coil sides a ~ , b ~ , and c ~ are connected together to form the
wye-connected stator. The end turns (looping of the coils) at both ends of the stator
so as to achieve the span of 1t radians are not shown in Fig. 1.4-1. As mentioned, each
coil consists of n
c
conductors, each of which makes up an individual single conduc-
tor coil within the main coil. Thus the number of turns of each winding is determined
by the product of n; and the number of coils or the product of n; and the number of
coil sides carrying current in the same direction. In the case of the fd winding, each
coil (/1-I{, for example) consists of nf conductors. It should be mentioned that in
Example IB, the stator and rotor coils each consisted of only one coil side with
one conductor (n
c
== 1) in each coil side.
One must realize that the winding configuration shown in Fig. 1.4-1 is an over-
simplification of a practical machine. The coil sides of each phase winding are con-
sidered to be distributed uniformly over 60 of the stator circumference. Generally,
the coil sides of each phase are distributed over a larger area, perhaps as much as
120, in which case it is necessary for some of the coil sides of two of the phase
windings to occupy the same slot. In some cases, the coil sides may not be distrib-
uted uniformly over the part of the circumference that it occupies. For example, it
would not be uncommon, in the case of the machine shown in Fig. 1.4-1, to have
more turns in coil sides az and a3than in at and a4. (Similarly for the bs and cs wind-
ings.) We will find that this winding arrangement produces an air-gap MMF which
more closely approximates a sinusoidal air-gap MMF with respect to the angular
displacement about the air gap. Another practical consideration is the so-called
fractional-pitch winding. The windings shown in Fig. 1.4-1 span 1t radians for the
2-pole machine. This is referred to as full-pitch winding. In order to reduce voltage
and current harmonics, the windings are often wound so that they span slightly less
than 1t radians for a 2-pole machine. This is referred to as a fractional-pitch winding.
All of the above-mentioned practical variations from the winding arrangement
shown in Fig. 1.4-1 are very important to the machine designer; however, these fea-
tures are of less importance in machine analysis, where in most cases a simplified
approximation of the winding arrangement is sufficient.
38 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
A salient-pole synchronous machine is selected for consideration because the
analysis of this type of machine may be easily modified to account for other machine
types. However, a salient-pole synchronous machine would seldom be a 2-pole
machine except in the case of small reluctance machines which are of the synchro-
nous class but which do not have a field winding. Generally, 2- and 4-pole machines
are round-rotor machines with the field winding embedded in a solid steel (nonlami-
nated) rotor. Salient-pole machines generally have a large number of poles com-
posed of laminated steel whereupon the field winding is wound around the poles
similar to that shown in Fig. 1.4-1.
For the purposes of deriving an expression for the air-gap MMF, it is convenient
to employ the so-called developed diagram of the cross-sectional view of the
machine shown in Fig. 1.4-1. The developed diagram is shown in Fig. 1.4-2. The
length of the air gap between the stator and rotor is exaggerated in Figs. 1.4-1
and 1.4-2 for clarity. The fact that the air-gap length is small relative to the inside
diameter of the stator permits us to employ the developed diagram for analysis pur-
poses. In order to relate the developed diagram to the cross-sectional view of the
machine, it is helpful to define a displacement to the left of the origin as positive.
The angular displacement along the stator circumference is denoted 4>sand l/Jr along
the rotor circumference. The angular velocity of the rotor is ronand Or is the angular
displacement of the rotor. For a given angular displacement relative to the as axis we
can relate 4Js' l/Jr' and Or as
4Js == 4Jr +Or
(1.4-1 )
Our analysis may be simplified by considering only one of the stator windings at a
time. Figure 1.4-3 is a repeat of Figs. 1.4-1 and 1.4-2 with only the as winding
shown. Due to the high permeability of the stator and rotor steel the magnetic fields
asaxis
fdaxis
cs axis
bs axis
asaxis
Figure 1.4-2 Development of the machine shown in Fig. 1.4-1.
MACHINE WINDINGS AND AIR-GAP MMF 39
oj 02
a4 Qi
as axis
f c
r---T--- d
I I
I I
I I
a' I a' I
~ I .1 I
I
f
I I
L .1___ a
e b
.$
(b)
Figure 1.4-3 The as winding. (a) Direction of H; (b) developed diagram.
essentially exist only in the air gap and tend to be radial in direction due to the short
length of the air gap relative to the inside stator diameter. Therefore, in the air gap
the magnetic field intensity ii and the flux density Bhave a component only in the
d, direction, and the magnitude is a function of the angle <Ps whereupon
( 1.4-2)
and
(1.4-3)
40 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
With the assumed direction of the current las, the magnetic field intensity ii in the
air gap due to the as winding is directed from the rotor to the stator for
-n12 < cPs < nl2 and from the stator to the rotor for nl2 < cPs < ~ n (Fig. 1.4-3).
Ampere's law may now be used to determine the form of the air-gap MMF due to
the as winding. In particular, Ampere's law states that
J
.... ....
H dL == i (1.4-4)
where i is the net current enclosed within the closed path of integration. Let us con-
sider the closed path of integration depicted in Fig. 1.4-3b. Applying Ampere's law
around this closed path denoted as abcda, where the path be is at 4>s == n14, and
neglecting the field intensity within the stator and rotor steel, we can write and eval-
uate (1.4-4) as
J
r
(1t / 4)+g(1t / 4) (1) Jr(O)
H - dL+ H(O)dL == 0
r(1t/4) 4 r(O)+g(O)
(1.4-5)
where r( n14) and r(0) are the radius of the rotor at the respective paths of integra-
tion, and g(nI4) and g(O) are the corresponding air-gap lengths.
The magnetomotive force is defined as the line integral of ii. Therefore, the terms
on the left-hand side of (1.4-5) may be written as MMFs. In particular, (1.4-5) may
be written as
MMFG) +MMF(O) =0
(1.4-6)
In (1.4-6) the MMF includes sign and magnitude; that is, MMF(O) == -Hr(O)g(O).
Let us now consider the closed path aefda where path ef occurs at l/Js == 1'6 n; here
Ampere's law gives
( 1.4-7)
The right-hand side of (1.4-7) is negative in accordance with the right-hand or
corkscrew rule.
We can now start to plot the air-gap MMF due to the as winding. For the lack of a
better guess let us assume that MMF(O) == O. With this assumption, (1.4-6) and
(1.4-7) tell us that the MMF is zero from <Ps == 0 to where our path of integration
encircles the first coil side (a;). If we continue to perform line integrations starting
and ending at a and each time progressing further in the 4>s direction, we will obtain
the plot shown in Fig. 1.4-4. Therein, a step change in MMF is depicted at the center
MACHINE WINDINGS AND AIR-GAP MMF 41
as axis
Q4 Qj a2 aj
as axis
-4nJas
.-----..-----------1.--------- -------Jo
lPs
Figure 1.4-4 Plot of the air-gap MMF due to the as winding with the assumption that
MMF(O) is zero.
of the conductors; actually, there would be a finite slope as the path of integration
passes through the conductors.
There are two items left to be considered. First, Gauss's law states that
(1.4-8)
Hence, no net flux may travel across the air gap because the flux density is assumed
to exist only in the radial direction:
J
27t
o Br ( <Ps)rl d<ps = 0
{1.4-9}
42 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
where 1is the axial length of the air gap of the machine and r is the mean radius of
the air gap. It follows that rl d4Js is the incremental area of an imaginary cylindrical
surface within the air gap of the machine. This fact in itself makes us suspicious of
our assumption that MMF(O) == 0 because the air-gap MMF shown in Fig. 1.4-4 is
unidirectional, which would give rise to a flux in only one direction across the air
gap. The second fact is one that is characteristic of electrical machines. In any prac-
tical electric machine, the air-gap length is either (a) constant as in the case of a
round-rotor machine or (b) a periodic function of displacement about the air gap
as in the case of salient-pole machine. In particular, for a 2-pole machine
(1.4-10)
or
(1.4-11 )
Equations (1.4-8) and (1.4-10) are satisfied if the air-gap MMF has zero average
value and, for a 2-pole machine,
(1.4-12)
Hence, the air-gap MMF wave for the as winding, which is denoted as MMF
as
, is
shown in Fig. 1.4-5. It is clear that the MMF due to the bs winding, MMFbs, is iden-
tical to MMFas but displaced to the left by 120
0
MMFcs is also identical but dis-
placed 240
0
to the left. The significance of the as, bs, and cs axes is now
apparent. These axes are positioned at the center of maximum positive MMF corre-
sponding to positive current into the windings.
asaxis
_------t2ncias
L.- ..L-.......... ---1. ----,,--...I---------...,O
217'
17'
-2ncias
Figure 1.4-5 Air-gap MMF due to as winding.
MACHINE WINDINGS AND AIR-GAP MMF 43
The waveform of the MMF's produced by the stator phase windings shown in
Fig. 1.4-1 may be considered as coarse approximations of sinusoidal functions of
cPr Actually most electric machines, especially large machines, are designed so
that the stator windings produce a relatively good approximation of a sinusoidal
air-gap MMF with respect to cPs so as to minimize the voltage and current harmonics.
In order to establish a truly sinusoidal MMF waveform (often referred to as a space
sinusoid) the winding must also be distributed sinusoidally. Except in cases where
the harmonics due to the winding configuration are of importance, it is typically
assumed that all windings may be approximated as sinusoidally distributed wind-
ings. We will make this same assumption in our analysis.
A sinusoidally distributed as winding and a sinusoidal air-gap MMF
mi
are
depicted in Fig. 1.4-6. The distribution of the as winding may be written as
N
as
== N
p
sin cPs,
Nils == - N
p
sin 4>s,
o~ 4>s <n
n <4>s ~ 2n
( 1.4-13)
(1.4-14)
where N
p
is the maximum turn or conductor density expressed in turns per radian.
If N." represents the number of turns of the equivalent sinusoidally distributed
as axis
Conductors
(a)
as axis
(b)
Figure 1.4-6 Sinusoidal distribution. (a) Equivalent distribution of as winding; (b) MMF
due to equivalent as winding.
44 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
winding, then
(1.4-15)
It is important to note that the number of turns is obtained by integrating N
as
over
nradians rather than 2n radians. Also, N.'i is not the total number of turns of the wind-
ing; instead, it is an equivalent number of turns of a sinusoidally distributed winding
which would give rise to the same fundamental component as the actual winding
distribution. For example, if we represent the air-gap MMF wave given in Fig.
1.4-5 as a sinusoidal air-gap MMF as given in Fig. 1.4-6, it can be readily shown
by Fourier analysis that N,/2 is equal to 2.43n
c
It is clear that the influence of design
features that have not been considered in the analysis such as fractional-pitch wind-
ings are accounted for in N.<;. (See problem at end of chapter.) Information regarding
machine windings may be found in reference 3.
Current flowing out of the paper is assumed negative so as to comply with the
convention established earlier when applying Ampere's law. The circles, which
are labeled as and as' and positioned at the point of maximum tum density in
Fig. 1.4-6b, will be used hereafter to signify schematically a sinusoidally distributed
winding. It is interesting to note in passing that some authors choose to incorporate
the direction of current flow in the expression for the turns whereupon the turns for
the as winding are expressed as
N, . A,.
N
as
== - 2 sme,
(1.4-16)
The air-gap MMF waveform is also shown in Fig. 1.4-6b. This waveform is read-
ily established by applying Ampere's law. The MMF waveform of the equivalent as
winding is
It follows that
N
s
( 21t)
MMFbs == 21bsCOS 4Js -"3
N
s
( 21t)
MMFcs == 21csCOS 4Js +"3
(1.4-17)
(1.4-18)
(1.4-19)
Let us now express the total air-gap MMF produced by the stator currents. This
can be obtained by adding the individual MMFs, given by (1.4-17)-( 1.4-19):
N
s
[. . ( 21t). ( 2n)]
MMFs == 2 las COS 4J
s
+IbsCOS 4Js -"3 +IcsCOS 4J
s
+"3
(1.4-20)
MACHINE WINDINGS AND AIR-GAP MMF 45
For balanced, steady-state conditions the stator currents may be expressed as
las == v2Jscos [Wet + Oei(O)]
hs = Vir,cos [wet - 2
3
1 +Oei(O)]
I
es
= Vilscos [Wet +2; +Oe;(O)]
(1.4-21 )
(1.4-22)
(1.4-23)
where 0ei (0) is the phase angle at time zero. Substituting the currents into (1.4-20)
and using the trigonometric relations given in Appendix A yields
(1.4-24 )
If the argument is set equal to a constant while the derivative is taken with respect to
time, we see that the above expression describes a sinusoidal air-gap MMF wave
with respect to 4>s' which rotates about the stator at an angular velocity of We in
the counterclockwise direction and which may be thought of as a rotating magnetic
pole pair. If, for example, the phase angle 0ei (0) is zero, then at the instant t == 0 the
rotating air-gap MMF is positioned in the as axis with the north pole at 4>s == 180
0
and the south pole at 4>s == o. (The north pole is by definition the stator pole from
which the flux issues into the air gap.)
The rotating air-gap MMF of a P-pole machine can be determined by considering
a 4-pole machine. The arrangement of the windings is shown in Fig. 1.4-7. Each
phase winding consists of two series connected windings, which will be considered
as sinusoidally distributed windings. The air-gap MMF established by each phase is
now a sinusoidal function of 24>s for a 4-pole machine or, in general, {P/2)4>5 where
P is the number of poles. In particular,
n., P
MMFas == p las COS "2
1J
s
N
s
. (P 27t)
MMF
b
== -lb'COS -,I,.. __
J r :" 2 0/.\ 3
u., (P 2n)
MMFcs == p'csCOS 2"4>s +3
(1.4-25)
(1.4-26)
( 1.4-27)
where tv... is the total equivalent turns per phase. With balanced steady-state stator
currents as given previously, the air-gap MMF becomes
( 1.4-28)
Here we see that the MMF produced by balanced steady-state stator currents rotates
about the air gap in the counterclockwise direction at a velocity of (2/P)w
e
. It may
46 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
csl axis
bsl axis
ftl2 axis
as2axis
bs2axis
cs2axis
+
-----.
csl V
cs
cs'!
cs2
a'2
bsl
bs'l
bs2
bs'2 Vbs
os'2
as2
as'l
:
+---
as! 2
+
+
+
Figure 1.4-7 Winding arrangement of an elementary 4-pole, 3-phase, wye-connected,
salient-pole synchronous machine.
at first appear necessary to modify extensively the analysis of a 2-pole machine in
order to analyze a P-pole machine. Fortunately, we will find that the modification
amounts to a simple change of variables.
We can now start to see the mechanism by which torque is produced in a synchro-
nous machine. The stator windings are arranged so that with balanced steady-state
currents flowing in these windings, an air-gap MMF is produced which rotates about
the air gap as a set of magnetic poles at an angular velocity corresponding to the
frequency of the stator currents and the number of poles. During steady-state opera-
tion the voltage applied to the field winding is constant. The resulting constant field
(lC-3)
(1C-2)
(1 C-I)
WINDING INDUCTANCES AND VOLTAGE EQUATIONS 47
current produces a set of magnetic poles, stationary with respect to the rotor. If the
rotor rotates at the same speed, or in synchronism with the rotating MMF established
by the stator currents, torque is produced due to the interaction of these poles. It fol-
lows that a steady-state average torque is produced only when the rotor and the stator
air-gap MMFs are rotating in synchronism, thus the name synchronous machine.
Example 1C When the stator currents of a multi phase electric machine,
which is equipped with symmetrical stator windings, are unbalanced in ampli-
tude and/or in phase, the total air-gap MMF consists of two oppositely rotating
MMFs. This can be demonstrated by assuming that during steady-state opera-
tion, the 3-phase stator currents are unbalanced in amplitude. In particular,
las == J2lacosWef
h" = J2hcos (Wet - 2
3
1t )
t., = J2l
e
cos ( wet + 2
3
1t )
where la, I
h
, and I; are unequal constants. Substituting into (1.4-20) yields
(1 C-4)
The first term is an air-gap MMF rotating in the counterclockwise direction at
We. The last two terms are air-gap MMFs (which may be combined into one)
that rotate in the clockwise direction at We. Note that when the steady-state
currents are balanced, la, li, and I; are equal whereupon the last two terms
of (1C-4) disappear.
1.5 WINDING INDUCTANCES AND VOLTAGE EQUATIONS
Once the concepts of the sinusoidally distributed winding and the sinusoidal air-gap
MMF have been established, the next step is to determine the self- and mutual induc-
tances of the machine windings. As in the previous section, it is advantageous to use
the elementary 2-pole, 3-phase synchronous machine to develop these inductance
relationships. This development may be readily modified to account for additional
windings (damper windings) placed on the rotor of the synchronous machine or for a
synchronous machine with a uniform air gap (round rotor). Also, we will show that
these inductance relationships may be easily altered to describe the winding induc-
tances of an induction machine.
48 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
Synchronous Machine
Figure 1.5-1 is Fig. 1.4-1 redrawn with the windings portrayed as sinusoidally
distributed windings. In a magnetically linear system the self-inductance of a
winding is the ratio of the flux linked by a winding to the current flowing in the wind-
ing with all other winding currents zero. Mutual inductance is the ratio of flux linked
by one winding due to current flowing in a second winding with all other winding
currents zero including the winding for which the flux linkages are being deter-
mined. For this analysis it is assumed that the air-gap length may be approximated
bs axis
as'
+
Figure 1.5-1 Elementary 2-pole, 3-phase, wye-connected, salient-pole synchronous
machine.
WINDING INDUCTANCES AND VOLTAGE EQUATIONS 49
as (Fig. 1.4-2)
(1.5-1 )
or
( 1.5-2)
where the minimum air-gap length is ((XI + (X2)-1 and the maximum is ((XI - (X2)-1.
Recall from (1.4-5) and (1.4-6) that MMF is defined as the line integral of fIe
Thus, from (1.4-3)
MMF
B, == 110--
g
(1.5-3)
or, because it is convenient to express the stator MMF in terms ofMMF(<ps)' we can
write
(1.5-4)
The air-gap flux density due to current in the as winding (i
as
) with all other currents
zero may be obtained by substituting (1.4-17) and (1.5-2) into (1.5-4). In particular,
the flux density with all currents zero except i
as
may be expressed as
(1.5-5)
Similarly, the flux density with all currents zero except ibs is
(1.5-6)
With all currents zero except i., we have
(1.5-7)
In the case of salient-pole synchronous machines the field winding is generally
uniformly distributed and the poles are shaped to approximate a sinusoidal distribu-
tion of air-gap flux due to current flowing in the field winding. In the case of
round-rotor synchronous machines the field winding is arranged to approximate
more closely a sinusoidal distribution [4-6]. In any event it is sufficient for our pur-
poses to assume as a first approximation that the field winding is sinusoidally dis-
tributed with N
f
equivalent turns. Later we will assume that the damper windings
( 1.5-8)
50 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
(additional rotor windings) may also be approximated by sinusoidally distributed
windings. Thus, the air-gap MMF due to current ltd flowing in the fd winding
may be expressed from Fig. 1.4-2 as
MMFfd = - tv; i
fd
sin4>,
Hence the air-gap flux density due to ltd with all other currents zero may be
expressed from (1.5-3) as
B,(4>,) = -J1otv; ifdsin 4>,(IXI - 1X2 cos 24>,) (1.5-9)
In the determination of self-inductance it is necessary to compute the flux linking
a winding due to its own current. To determine mutual inductance it is necessary to
compute the flux linking one winding due to current flowing in another winding.
Let us consider the flux linkages of a single turn of a stator winding which spans n
radians and which is located at an angle <Ps. In this case the flux is determined by
performing a surface integral over the open surface of the single tum. In particular,
<P( 4>s' 0,) = f' +n B , ( ~ , 0,)rl dt;
4>.\.
(1.5-10)
where ep is the flux linking a single turn oriented <Ps from the as axis, I is the axial
length of the air gap of the machine, r is the radius to the mean of the air gap (essen-
tially to the inside circumference of the stator), and ~ is a dummy variable of inte-
gration. In order to obtain the flux linkages of an entire winding, the flux linked by
each tum must be summed. Because the windings are considered to be sinusoidally
distributed and the magnetic system is assumed to be linear, this summation may be
accomplished by integrating over all coil sides carrying current in the same direc-
tion. Hence, computation of the flux linkages of an entire winding involves a double
integral. As an example, let us determine the total flux linkages of the as winding due
to current flowing only in the as winding. Here
J J J
4>.s +1C
A.as = L1sias + Nas(4)s)<P(4>s,O,) d4> =LI.,ias + Nas(4)s) cP., B , ( ~ , O,)ri d ~ d4>s
(1.5-11 )
In (1.5-11), Lis is the stator leakage inductance due primarily to leakage flux at the
end turns. Generally, this inductance accounts for 5% to 10% of the maximum self-
inductance. Substituting (1.4-14) (with N
p
replaced by N,/2) and (1.5-5) for
B
r
(~ , Or) into (1.5-11) yields
(1.5-12)
WINDING INDUCTANCES AND VOLTAGE EQUATIONS 51
The interval of integration is taken from n to 2n so as to comply with the convention
that positive flux linkages are obtained in the direction of the positive as axis by cir-
culation of the assumed positive current in the clockwise direction about the coil
(right-hand rule). The self-inductance of the as winding is obtained by dividing
(1.5-12) by ius. Thus
(1.5-13)
The mutual inductance between the as and bs windings may be determined by
first computing the flux linking the as winding due to current flowing only in the
bs winding. In this case it is assumed that the magnetic coupling that might occur
at the end turns of the windings may be neglected. Thus
(1.5-14)
Substituting (1.4-14) and (1.5-6) into (1.5-14) yields
Therefore, the mutual inductance between the as and bs windings is obtained by
evaluating (1.5-15) and dividing the result by ibs. This gives
(1.5-16)
The mutual inductance between the as andfd windings is determined by substi-
tuting (1.5-9), expressed in terms of </Js - er into (1.5-14). Thus
Evaluating and dividing by i
f d
yields
( 1.5-18)
The self-inductance of the field winding may be obtained by first evaluating the
flux linking thefd winding with all currents equal to zero except ifd. Thus, with thefd
winding considered as sinusoidally distributed and the air-gap flux density expressed
52 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
by (1.5-9) we can write
from which
(1.5-20)
(1.5-21 )
(1.5-24)
(1.5-23)
(1.5-22)
where Llfd is the leakage inductance of the field winding.
The remaining self- and mutual inductances may be calculated using the same
procedure. We can express these inductances compactly by defining
LA = 21tJl.
o
rllXI
LB= (i )2
1t
Jl.o
rlIX2
LSfd = (i)1tJl.orl( 1X 1+
Lmfd = (i) 21tJl.
o
rl ( IXI +
The machine inductances may now be expressed as
L
asas
== Lis +LA - LB cos 20r
Lbsbs = Lis +LA - LBCOS2(Or _
Leses =Lis +LA - LBCOS2(Or +2;)
Lf dfd == Llfd +Lmfd
Lashs = - LBCOS2(Or
Lases =
1
L
bscs
== - 2LA - LBcos2(Or +n)
L
asf d
== L
Sf d
sin Or
Lbsfd = L.ifdsin (Or - 2
3
1t )
L
esfd
= L.ifd sin (Or +2
3
1t )
(1.5-25)
(1.5-26)
(1.5-27)
(1.5-28)
(1.5-29)
(1.5-30)
(1.5-31)
(1.5-32)
(1.5-33)
(1.5-34)
WINDING INDUCTANCES AND VOLTAGE EQUATIONS 53
In later chapters we will consider a practical synchronous machine equipped
with short-circuited rotor windings (damper windings). The expressions for the
inductances of these additional windings can be readily ascertained from the work
presented here. Also, high-speed synchronous machines used with steam turbines
are round rotor devices. The 20,. variation is not present in the inductances of a
uniform air-gap machine. Therefore, the winding inductances may be determined
from the above relationships by simply setting (X2 == 0 in (1.5-22)-(1.5-24). It is clear
that with (X2 == 0, L
B
== O.
The voltage equations for the elementary synchronous machine shown in
Fig. 1.5-1 are
dAliS
Vas == rslas +
dAbs
vs. == r.\'bs +
dAcs
Vcs == r,\"CS +
. dAfd
Vfd == r.fdYd +
( 1.5-35)
(1.5-36)
(1.5-37)
( 1.5-38)
where r, is the resistance of the stator winding and r.fd the resistance of the field
winding. The flux linkages are expressed as

Abs == Lbsas;as +Lbsbsibs +Lbscsics +
Acs == Lcsasias + Lcsbsibs +Lcsc.\ics +
== L.fdas;as +L.fdbs;bs +L.fdcs;cs +Lfdfdifd
( 1.5-39)
( 1.5-40)
(1.5-41 )
(1.5-42)
One starts to see the complexity of the voltage equations due to the fact that some of
the machine inductances are functions of Orand therefore a function of the rotor
speed W
r
Hence, the coefficients of the voltage equations are time-varying except
when the rotor is stalled. Moreover, rotor speed is a function of the electromagnetic
torque, which we will find to be a product of machine currents. Clearly, the solution
of the voltage equations is very involved. In fact, a computer is required in order to
solve for the transient response of the combined electrical and mechanical systems.
In Chapter 3 we will see that a change of variables allows us to eliminate the time-
varying inductances and thereby markedly reduce the complexity of the voltage
equations. However, the resulting differential equations are still nonlinear, requiring
a computer to solve for the electromechanical transient behavior.
Induction Machine
The winding arrangement of a 2-pole, 3-phase, wye-connected symmetrical
induction machine is shown in Fig. 1.5-2. The stator windings are identical with
54 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
bs axis
br axis
craxis
i"s
~
+
r,
V
cs
+
r,
;a
Vb$
-----+
+
~ r ~ _
a r ~ __--11
~ s 9,
~ - - - + - + - - - ~ L_
Figure 1.5-2 Two-pole, 3-phase, wye-connected symmetrical induction machine.
equivalent turns N, and resistance r., The rotor windings, which may be wound or
forged as a squirrel cage winding, can be approximated as identical windings with
equivalent turns N, and resistance Yr. The air gap of an induction machine is uniform
and it is assumed that the stator and rotor windings may be approximated as sinu-
soidally distributed windings.
In nearly all applications the induction machine is operated as a motor with the
stator windings connected to a balanced 3-phase source and the rotor windings short-
circuited. The principle of operation in this mode is quite easily deduced [4-6]. With
balanced 3-phase currents flowing in the stator windings, a rotating air-gap MMF is
established, as in the case of the synchronous machine, which rotates about the air
gap at a speed determined by the frequency of the stator currents and the number
WINDING INDUCTANCES AND VOLTAGE EQUATIONS 55
of poles. If the rotor speed is different from the speed of this rotating MMF, balanced
3-phase currents will be induced (thus the name induction) in the short-circuited
rotor windings. The frequency of the rotor currents corresponds to the difference
in the speed of the rotating MMF due to the stator currents and the speed of the rotor.
The induced rotor currents will, in turn, produce an air-gap MMF that rotates relative
to the rotor at a speed corresponding to the frequency of the rotor currents. The speed
of the rotor air-gap MMF superimposed upon the rotor speed is the same speed as
that of the air-gap MMF established by the currents flowing in the stator windings.
These two air-gap MMFs rotating in unison may be thought of as two synchronously
rotating sets of magnetic poles. Torque is produced due to an interaction of the two
magnetic systems. It is clear, however, that torque is not produced when the rotor is
running in synchronism with the air-gap MMF due to the stator currents because in
this case currents are not induced in the short-circuited rotor windings.
The winding inductances of the induction machine may be expressed from the
inductance relationships given for the salient-pole synchronous machine. In the
case of the induction machine the air gap is uniform. Thus, 20
r
variations in the
self- and mutual inductances do not occur. This variation may be eliminated by set-
ting CX2 == 0 in the inductance relationship given for the salient-pole synchronous
machine. All stator self-inductances are equal; that is, L
asas
== Lbs
bs
== L
cscs
with
( 1.5-43)
where L
ms
is the stator magnetizing inductance that corresponds to LA in (1.5-25). It
may be expressed as
L
_ (N.fI) 2nfJori
ms - 2 g
( 1.5-44)
where g is the length of the uniform air gap. Likewise all stator-to-stator mutual
inductances are the same. For example,
1
LlIsbs == - 2Lms
which corresponds to (1.5-29) with L
B
== O.
It follows that the rotor self-inductances are equal with
L
arar
== L; +L
mr
where the rotor magnetizing inductance may be expressed as
L _(N
r
) 2nfJori
mr - 2 g
( 1.5-45)
(1.5-46)
( 1.5-47)
56 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
The rotor-to-rotor mutual inductances are
1
Larbr == - 2" Lmr
(1.5-48)
Expressions for the mutual inductances between stator and rotor windings may be
written by noting the form of the mutual inductances between the field and stator
windings of the synchronous machine given by (1.5-32)-(1.5-34). Here we see
that L
asar,
Lbsb" and Leser are equal with
L
asar
== i; cos Or
Also, Lasb" Lbse" and L
esar
are equal with
Finally, Lase" L
bsa
" and L
esbr
are equal with
Laser = i;cos (()r - 2
3
11: )
where
(1.5-49)
(1.5-50)
(1.5-51)
(1.5-52)
The voltage equations for the induction machine shown in Fig. 1.5-2 are
dAas
Vas == rslas +--;[t"
dA.bs
Vbs == rs1bs +--;[t"
dAes
Ves == 'sles +--;It
dAar
Var == 'rlar +--;It
dAbr
Vbr == 'r1br +--;It
dAcr
Ver == 'rier +--;[t"
(1.5-53)
(1.5-54)
(1.5-55)
(1.5-56)
(1.5-57)
(1.5-58)
where 's is the resistance of each stator phase winding and r, is the resistance of each
rotor phase winding. The flux linkages may be written following the form of A
as
:
( 1.5-59)
WINDING INDUCTANCES AND VOLTAGE EQUATIONS 57
Here again we see the complexity of the voltage equations due to the time-vary-
ing mutual inductances between stator and rotor circuits (circuits in relative motion).
We will see in Chapter 3 that a change of variables eliminates the time-varying
inductances resulting in voltage equations which are still nonlinear but much more
manageable.
Example ID The winding inductances of a P-pole machine may be deter-
mined by considering the elementary 4-pole, 3-phase synchronous machine
shown in Fig. 1.4-7. Let us determine the self-inductance of the as winding.
The air-gap flux density due to current only in the as winding may be
expressed from (1.5-4); in particular,
(ID-l )
(10-2)
where for a P-pole machine, MMF
as
is given by (1.4-25) and Orm is defined in
Fig. 1.4-7. From (1.5-2) and Fig. 1.4-7 we obtain
1
g(cP, - (Jnn) = ( / ) (cP (J)
. (X, - (X2cos P 2 2 s - rm
Substituting into (10-1) yields
Following (1.5-11) for a P-pole machine we have
p J4TC/P f
4J
,+2TC /P
Aas == Lisias -"2 Nas(cPs) s.u, Orm)rl di; dcPs
2TC!P 4>\
where
(10-3)
(10-4)
2n 4n
- < ~ <-
P - o/s - p
(ID-5)
The double integral is multiplied by P/2 to account for the flux linkages of the
complete as winding. Evaluating (ID-4) and dividing by ills yields
N; [ (12 (P)]
Lasas == LI.'i +2Pn/lorl (x, - 2 cos2 2" Orm
If, in the above relation, we substitute
P
Or == "2 (Jrm
( 10-6)
( ID-7)
then (ID-3) is of the same form as for the 2-pole machine, (1.5-13). Moreover,
if we evaluate all of the machine inductances for the P-pole machine and make
this same substitution, we will find that the winding inductances of a P-pole
58 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
machine are of the same form as those of a 2-pole machine. Also, co,is defined
as
p
W, == ZWrm
(ID-8)
and the substitute variables lJ, and co, are referred to as the electrical angular
displacement and the electrical angular velocity of the rotor, respectively. This
connotation stems from the fact that (J,( w,) refers a rotor displacement (velo-
city) to an electrical displacement (velocity). It is clear that, at synchronous
speed, os, is equal to the angular velocity of the electrical system connected
to the stator windings regardless of the number of poles.
Because the winding inductances of a P-pole machine are identical in form
to those of a 2-pole machine if (J, and co, are defined by (10-7) and (10-8),
respectively, all machines can be treated as if they were 2-pole devices as far as
the voltage equations are concerned. We will find that it is necessary to multi-
ply the torque equation of a 2-pole machine by P/2 in order to express the
electromagnetic torque of a P-pole machine correctly.
REFERENCES
[1] C. M. Ong, Dynamic Simulation ofElectric Machinery, Prentice-Hall PTR, Upper Saddle
River, NJ, 1998.
[2] D. C. White and H. H. Woodson, Electromechanical Energy Conversion, John Wiley and
Sons, New York, 1959.
[3] M. S. Say, Alternating Current Machines, AHalsted Press Book, John Wiley and Sons, New
York, 1976.
[4] A. E. Fitzgerald, C. Kingsley, Jr., and S. D. Umans, Electric Machinery, 5th ed., McGraw-
Hill, New York, 1990.
[5] S. J. Chapman, Electric Machinery Fundamentals, 3rd ed., McGraw-Hili, New York, 1999.
[6] G. McPherson and R. D. Laramore, An Introduction to Electrical Machines and
Transformers, 2nd ed., John Wiley and Sons, New York, 1990.
PROBLEMS
1 A two-winding, iron-core transformer is shown in Fig. I.P-I. N, =50 turns, N
2
= 100
turns, and IlR = 4000. Calculate Lmt, L
m
2, and LI2. The resistance of coils I and 2 is 2 n
and 4n, respectively. If the voltage applied to coil I is VI = 10cos400t, calculate the
voltage appearing at the terminals of coil 2 when coil 2 is open-circuited.
2 Repeat Problem 1if the iron core has an air gap of 0.2 cm in length and is cut through the
complete cross section. Assume that fringing does not occur in the air gap, that is, the
effective cross-sectional area of the air gap is 25 crrr'.
3 Two coupled coils have the following parameters
L
II
= 100mH
L
22
= 25mH
N, = 1000turns
L/1 =O.IL"
rl = 10 n
ri = 2.5 n
N
2
=500tums
L
II
= O.IL"
PROBLEMS 59
~ - - - - 30em- - - - ~
30em
- l _ ~ __
5cm
Figure 1.P-l Two-winding, iron-core transformer.
(a) Develop a Tequivalent circuit with coil I as the reference coil. (b) Repeat with coil 2
as the reference coil. Determine the input impedance of the coupled circuits if the applied
frequency to coil I is 60 Hz with coil 2 (c) open-circuited; (d) short-circuited. Repeat (d)
with the current flowing in the magnetizing reactance neglected.
4 A constant 10 V is suddenly applied to coil I of the coupled circuits given in Problem 3.
Coil 2 is short-circuited. Calculate the transient and steady-state current flowing in each
coil.
5 Consider the coils given in Problem 3. Assume that the coils are wound so that, with the
assigned positive direction of currents i I and 12, the mutual inductance is negative. (a) If
the coils are connected in series so that 11 == ;2, calculate the self-inductance of the series
combination. (b) Repeat (a) with the coils connected in series so that 11 == -12.
6 A third coil is wound on the ferromagnetic core shown in Fig. 1.2-1. The resistance is rs
and the leakage and magnetizing inductances are L
f3
and L
m3
, respectively. The coil is
wound so that positive current (13) produces 4>m3 in the opposite direction as 4>m1 and
<P
m2
. Derive the T equivalent circuit for this three-winding transformer. Actually one
should be able to develop the equivalent circuit without derivation.
7 Use E and 1/p to denote summation and integration, respectively. Draw a time-domain
block diagram for two coupled windings with saturation.
8 A resistor and an inductor are connected as shown in Fig. I.P-2 with R == 15 nand
L == 250 mll. Determine the energy stored in the inductor W
eS
and the energy dissipated
by the resistor W
eL
for; > 0 if 1(0) == lOA.
9 Consider the spring-mass-damper system shown in Fig. I.P-3. At t == 0, x(O) == Xo (rest
position) and dx/dt== 1.5m/s. M==0.8kg, D== 10Ns/m, and K== 120Nm. For
t > 0, determine the energy stored in the spring W
111SI
, the kinetic energy of the mass
W
mS
2, and the energy dissipated by the damper W",L.
60 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
L
Figure I.P-2 R-L circuit.
R
10 ExpressW/(i,x) and We(i,x) for (a) A(i,X) = i
3
/
2x3
; (b) A(i,X) = xi
2
+kicosx.
11 The energy stored in the coupling field of a magnetically linear system with two
electrical inputs may be expressed as
Express B
II
, B
12
, and B
22
in terms of inductances L
I
J, L
12
, and L22.
12 An electromechanical system has two electrical inputs. The flux linkages may be
expressed as
AI (ii, ;2,X) = x 2 ; ~ +Xi2
A2(i
l,i2,X)
= x 2 i ~ +xil
Express Wj(il, i2,X) and We (it, i2,X).
13 Express fe(i, x) for the electromechanical systems described by the relations given in
Problem 10.
14 Express Ie (i I, i
2
, x) for the electromechanical system given in Problem 12.
15 Refer to Fig. 1.3-7. As the system moves from X
a
to Xb the A-i trajectory moves from A
to B. Does the voltage v increase or decrease? Does the applied force I increase or
decrease? Explain.
16 Refer to Fig. 1.3-11. Following the system transients due to the application of the source
voltage (v == 5V), the system assumes steady-state operation. During this steady-state
operation calculate W
eS
, WI, We, and W
mS

17 Refer to Fig. 1.3-12. Repeat Problem 16 for steady-state operation following the
application off = 4N.
18 Refer to Fig. 1.3-13. Identify the area corresponding to ~ W m when (a) x moves from 2.5
mm to 4.3 mm, and (b) x moves from 4.3 mm to 2.5 mm.
M
Figure I.P-3 Spring-mass-damper system.
bs axis
PROBLEMS 61
as axis
Figure I.P-4 Stator winding arrangement of a 2-pole, 2-phase machine.
19 Assume the steady-state currents flowing in the conductors of the device shown in
Fig. IB-1 are
I, == I.fiCOS[WI t +0,(0)]
12 == I, cos [W2t +O
2
(0)]
where 0,(0) and O
2(0)
are the time-zero displacements of It and 1
2
, respectively. Assume
also that during steady-state operation the rotor speed is constant; thus
0, == w,t +0,(0)
where Or(O) is the rotor displacement at time zero. Determine the rotor speeds at which
the device produces a nonzero average torque during steady-state operation if
(a) w, == W2 == 0; (b) WI == W2 i-0; (c) W2 == O. Express the instantaneous torque at
each of these rotor speeds.
20 The developed diagram shown in Fig. I.P-4 shows the arrangement of the stator windings
of a 2-pole, 2-phase machine. Each coil side has n; conductors, and i
as
flows in the
as winding and ibs flows in the bs winding. Draw the air-gap MMF (a) due to ias(MMF
as)
and (b) due to ibs(MMF
bs
) '
21 Assume that each coil side of the winding shown in Fig. I.P-4 contains n; conductors.
The windings are to be described as sinusoidally distributed windings with N
p
the
maximum turns density and N
s
the number of equivalent turns in each winding, which is
a function of nco Express (a) Nos and Nbs" (b) MMF
ll S
and MMF
h
.\"" and (c) the total air-gap
MMF (MMF
s
) produced by the stator windings.
22 A fractional-pitch stator winding (a phase) is shown in developed form in Fig. l.P-5.
Each coil side contains n; conductors. The winding is to be described as a sinusoidally
distributed winding. Express the number of equivalent turns N, in terms of nco
as axis
1r
~ 1r/6 ~
Oa2
. aj a.
11'/2
Figure I.P-5 Stator winding arrangement for fractional-pitch winding.
62 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS
23 Consider the 2-phase stator windings described in Problem 21. (a) With las = ViIs
cos[wet +8
es(O)]
determine IbJ in order to obtain a constant amplitude MMF (MMF
s
)
which rotates clockwise around the air gap of the machine. Repeat (a) with (b)
las = -J2,''iCOswet, (c) laJ = /2lssinw
et,
and (d)las = -/2lssin [wet +8
es(O)].
24 A single-phase source is connected between terminals as and bs of the 4-pole, 3-phase
machine shown in Fig. 1.4-7. The cs terminal is open-circuited. The current into the as
winding is las = J2,/ cos wet. Express the total air-gap MMF.
25 An elementary 2-pole, 2-phase, salient-pole synchronous machine is shown in Fig. I.P-6.
The winding inductances may be expressed as
L
asas
=L,s +LA - L
B
cos 28r
Lbsbs = l-, +LA +LB cos 2(}r
Lasbs = - LB sin Z8r
Lfdfd = Llfd +Lmfd
Lasfd = LSfd sin 8r
Lbsld = - Lsld cos ()r
taxis
+
Figure 1.P-6 Elementary 2-pole, 2-phase, salient-pole synchronous machine.
PROBLEMS 63
Verify these relationships and give expressions for the coefficients LA, L
B
, L",fd, and L.ifd
similar in form to those given for a 3-phase machine. Modify these inductance relation-
ships so that they will describe a 2-phase, uniform air-gap synchronous machine.
26 Write the voltage equations for the elementary 2-pole, 2-phase, salient-pole synchronous
machine shown in Fig. I.P-6 and derive the expression for T
e
(ias , ibs, ~ f d , Or).
27 An elementary 4-pole, 2-phase, salient-pole synchronous machine is shown in Fig. 1.P-7.
Use this machine as a guide to derive expressions for the winding inductances of a P-pole
synchronous machine. Show that these inductances are of the same form as those given in
Problem 25 if {P/2)Orm is replaced by Or.
bsl axis
fd2 axis
bs'l
as2axis
as'2
as!
bs'2
fdl axis
bs2 axis
8b.
\ ~ b . \
.-
8'
\ ~ ' \ . \
Vas r,
~ d
Vfd
+
+
Vbs +
Figure 1.P-7 Elementary 4-pole, 2-phase, salient-pole synchronous machine.
64 BASIC PRINCIPLES FOR ELECTRIC MACHINE ANALYSIS


8,
__-
as axis
as
r,
+
N,
Vas
N
s
N
s

Va'
+
+
r,
Vb'
Figure I.P-8 Elementary 2-pole, 2-phase symmetrical induction machine.
28 Derive an expression for the electromagnetic torque, Te(ias, ibs, i
f d
, Or), for a P-pole,
2-phase, salient-pole synchronous machine. This expression should be identical in form
to that obtained in Problem 26 multiplied by P/2.
29 A reluctance machine has no field winding on the rotor. Modify the inductance
relationships given in Problem 25 so as to describe the winding inductances of a 2-pole,
2-phase, reluctance machine. Write the voltage equations and derive an expression for
T,(ias , ibs, Or).
30 An elementary 2-pole, 2-phase, symmetrical induction machine is shown in Fig. I.P-8.
Derive expressions for the winding inductances. If you have worked Problem 25, you
may modify those results accordingly.
31 Write the voltage equations for the induction machine shown in Fig. 1.P-8 and derive an
expression for the electromagnetic torque Te(ias, ibs, ianibnOr).
32 An elementary 4-pole, 2-phase, symmetrical induction machine is shown in Fig. 1.P-9.
Use this machine as a guide to derive expressions for the winding inductance of a P-pole
induction machine. Show that these inductances are of the same form as those given in
Problem 30 if {P/2)Orm is replaced by Or.
PROBLEMS 65
brl axis
as2 axis
ar2axis
bsl axis
bs2 axis
br2 axis
r,
r,
Figure 1.P-9 Elementary 4-pole, 2-phasesymmetrical induction machine.