12-Oct-21

Linear Transformations
Lecture 3

Linear Transformation
• A wide variety of rotating machines are polyphase a.c. machines
constructed in a different manner than the primitive machine.
• In such cases, the primitive machine can also be employed in the analysis,
provided the rotating polyphase winding on the rotor and the stationary
polyphase winding on the stator can be represented by the d-q axes coils of
the primitive machine.
• The process of replacing one set of variables by another related set of
variables is called winding transformation or merely, transformation.
• The transformation from old to new set of variables is carried out merely for
making the calculations simpler and less laborious.
• Linear transformation means that the transformation from old to new set of
variables or vice-versa, is governed by linear equations .
2

1
 12-Oct-21

Transformation matrix
• The equations expressing old variables in terms of new variables or
vice versa are called transformation equations.
• These equations, when written in matrix, have the following general
form :
• [New variables] = [Transformation matrix] [Old variables]
• [Old variables] = [Transformation matrix] [New variables]
• The transformation matrix may, therefore, be defined as a matrix ,
containing the coefficients relating the old and new variables.

• Linear transformations in electrical machines are usually carried out

for the purpose of obtaining new equations which are fewer in number
or are more easily solved.
• For example, a 3-phase machine requires three voltage equations
whereas its generalized model requires only two voltage equations.
• Which can be solved more easily as compared to three voltage
equations.
• Further, the circuit equations for a 3-phase machine are more
complicated because of the magnetic coupling amongst the three-
phase windings.
• In machine analysis, generally rotor currents are transformed and the
necessary condition for doing this is that the transformed system of
currents must produce an m.m.f. identical with that of the original
system of currents, so far as magnitude and direction are concerned.
4

2
 12-Oct-21

Transformation from a Displaced Brush-Axis

• If a commutator machine has
brushes displaced from d or q-
axes,
• A transformation is necessary from
the displaced brush axis to d-q
axes.

• Suppose one set of brushes, denoted by AA, makes an angle  with the
d-axis as shown in previous Figure (a). The armature establishes an
m.m.f. Fa along its brush axis.
• This m.m.f. Fa can be resolved along d-axis as Fd and along q-axis as
Fq,
• where Fd=Fa cos() and Fq=Fasin()
• Assuming equal of turns in both armature and d-q coils,
• we get Nid = Niacos()  id = iacos()
• and iq = iasin()
• In addition to brush pair AA, consider another set of brushes BB
making an angle  with q-axis as shown in next Figure.

3
 12-Oct-21

• The resolution of m.m.fs Fa and Fb

along d-q axes gives
• Fd=Facos() - Fbsin()
• Fq=Fasin() + Fbcos()
• Assuming equal number of turns as
above, we get
• id = iacos() - ibsin()
• iq = iasin() + ibcos()

Transformation from Three phase to Two phase

• A symmetrical 2-pole, 3-phase winding on the rotor is represented by
three coils A, B, C each of N effective turns and mutually displaced by
120o. Maximum values of m.m.fs. Fa, Fb, Fc are shown along their
respective phase-axes. The combined effect of these three m,m.fs
results in a constant magnitude m.m.f., which rotates at a constant
angular velocity depending on the poles and frequency.

4
 12-Oct-21

• 𝑖𝑎 = 𝐼𝑚𝑐𝑜𝑠 𝑤𝑡
• 𝑖𝑏 = 𝐼𝑚𝑐𝑜𝑠 𝑤𝑡 −
• 𝑖𝑐 = 𝐼𝑚𝑐𝑜𝑠 𝑤𝑡 −
• 𝐹3∅ =
• In two phase current
• 𝑖𝛼 = 𝐼𝑚𝑐𝑜𝑠 𝑤𝑡
• 𝑖𝛽 = 𝐼𝑚𝑐𝑜𝑠 𝑤𝑡 − = 𝐼𝑚𝑠𝑖𝑛 𝑤𝑡
• F2=ImN
9

The m.m.f.s of three-phase and two-phase systems can be rendered

equal in magnitude by making any one of the following changes:
(i) by changing magnitude of the two-phase currents,
(ii) by changing number of turns of the two phase windings,
(iii) by changing both the magnitude of currents and number of turns.
• (i) If the effective number of turns per phase in case of two-phase
winding is N (i.e. the winding factors are same for both three- and
two-phase windings) then for equal m.m.fs, the magnitude of the
current in the two phases, must be 3/2 times the magnitude of the
three-phase currents.
• 𝑖𝛼𝑁 = 𝑁 𝑖𝑎𝑐𝑜𝑠 0 + 𝑖𝑏𝑐𝑜𝑠 120 + 𝑖𝑐𝑐𝑜𝑠 240
• 𝑖𝛼 = 𝑖𝑎 − 𝑖𝑏 + 𝑖𝑐
• 𝑖𝑁 = 𝑁 𝑖𝑎𝑠𝑖𝑛 0 + 𝑖𝑏𝑠𝑖𝑛 120 + 𝑖𝑐𝑠𝑖𝑛 240
• 𝑖 = 0 + 𝑖𝑏 − 𝑖𝑐
10

5
 12-Oct-21

• For balanced system

• ia+ib+ic=0 & 𝑖𝛼 = 𝑖𝑎 − 𝑖𝑏 + 𝑖𝑐
•  𝑖𝛼 = 𝑖𝑎
• Thus the magnitude of the two-phase currents is 3/2 times the magnitude of
the three-phase currents.
• With the conditions giving above, equal m.m.fs have been attained.
• The power per phase of the two-phase system (3/2VI) is thus 3/2 time the
power per phase (VI) of the three phase system.
• Note that the total power of the two-phase system (= 2*3/2VI) and three-
phase system (=3 VI) is the same.
• Thus the invariance of power has been attained.
• The only disadvantage is that the transformation of current and voltage will
differ of the presence of factor 3/2 in the current transformation.
• As factor 3/2 appears in current transformation and not in voltage
transformation, the per phase parameters of the two-phase and three-phase
induction machines will not be the same.
11

• (ii) If the effective number of per-phase turns of the two-phase

winding is made 3/2 times that of the three phase winding, then for
equal m.m.fs the ,magnitude of the currents in the two phase and three
phase systems must be equal i.e. i = ia. This can be proved as is done
in the previous case.
• With these conditions, the per phase voltage of two phase machine
will be 3/2 times (3/2 V) the per phase voltage (V) of the three phase
system.
• The power per phase in two-phase system =3/2 VI
• Total power in the two-phase system =3 VI
• Also total power in the three-phase system = 3 VI
• Here again the invariance of power has been obtained, but, as earlier,
the transformation of current and voltage will differ because of the
factor 3/2 in the voltage transformation.
12

6
 12-Oct-21

• (iii) Here both the magnitude of currents and number of turns of the
two-phase system are changed to obtain identical transformation for
voltage and current.
• Let the number of per-phase turns in the two-phase winding be made
times the per-phase turns of the three-phase winding.

• Then for equal m.m.fs., 𝑖𝛼 = 𝑖𝑎. This can be proved by resolving

the 3-phase m.m.fs along the -axis as shown below :

• 𝑖𝛼 𝑁 = 𝑁 𝑖 𝑎 − 𝑖 𝑏 − 𝑖𝑐

• 𝑖𝛼 = 𝑖 𝑎 − 𝑖𝑏 − 𝑖 𝑐 = 𝑖𝑎

13

• The phase voltage and current of the two-phase system are times
those of the three-phase system.
• This fact results in identical transformations for both the voltage and
current.
• 𝑖𝛼 = 𝑖𝑎 − 𝑖𝑏 − 𝑖𝑐

• 𝑖𝛽 = 0+ 𝑖𝑏 − 𝑖𝑐

• Since the transformations for voltage and current are identical,

impedance per phase of the two- and three-phase systems is the same.

14

7
 12-Oct-21

• Example 1: A three-phase induction motor has the following per

phase parameters referred to stator :
• Stator resistance — 0.30 
• Rotor resistance — 0.45 
• Stator and rotor leakage reactance- 2.1  each
• Magnetising reactance — 30.00 
• Find out the parameters of an equivalent 2-phase induction motor if its
per phase turns are :
• (a) Same as that of the 3-phase induction motor.
• (b) 3/2 times that of the 3-phase induction motor.
• (c) times that of the 3-phase induction motor.

15

• Sol. (a) Let

• 𝑖𝑎 = 𝐼𝑚𝑐𝑜𝑠 𝑤𝑡 − ∅
• 𝑖𝑏 = 𝐼𝑚𝑐𝑜𝑠 𝑤𝑡 − 120𝑜 − ∅
• 𝑖𝑐 = 𝐼𝑚𝑐𝑜𝑠 𝑤𝑡 − 240𝑜 − ∅
• Since per phase turns in both 2-phase and 3-phase induction motors are the same, for
equal m.m.f.s, the magnitude of currents in the 2-phase motor must be times the
magnitude of currents in the 3-phase motor.
• 𝑖𝛼 = 𝑖𝑎 = 𝐼𝑚𝑐𝑜𝑠 𝑤𝑡 − ∅ ,

• 𝑖𝛽 = 𝑖𝑏 − 𝑖𝑐

• = 𝐼𝑚 𝑐𝑜𝑠 𝑤𝑡 − 120𝑜 − ∅ − 𝑐𝑜𝑠 𝑤𝑡 − 240𝑜 − ∅

• = 𝐼𝑚[𝑐𝑜𝑠 𝑤𝑡 − ∅ 𝑐𝑜𝑠 120𝑜 + 𝑠𝑖𝑛 𝑤𝑡 − ∅ 𝑠𝑖𝑛 120𝑜 − 𝑐𝑜𝑠 𝑤𝑡 − ∅ 𝑐𝑜𝑠 240𝑜

− 𝑠𝑖𝑛 𝑤𝑡 − ∅ 𝑠𝑖𝑛 240𝑜 ] = 𝐼𝑚𝑠𝑖𝑛 𝑤𝑡 − ∅
16

8
 12-Oct-21

• Farther, Let
• 𝑣𝑎 = 𝑉𝑚𝑐𝑜𝑠 𝑤𝑡 − ∅
• 𝑣𝑏 = 𝑉𝑚𝑐𝑜𝑠 𝑤𝑡 − 120𝑜 − ∅
• 𝑣𝑐 = 𝑉𝑚𝑐𝑜𝑠 𝑤𝑡 − 240𝑜 − ∅
• Since the per-phase turns in both the machines are equal and their rotating
m.m.fs. have the same magnitude, factor will not appear in the voltage
transformations, i.e.
• 𝑣𝛼 = 𝑣𝑎 = 𝑉𝑚𝑐𝑜𝑠 𝑤𝑡
• 𝑣𝛽 = 𝑣𝑏 − 𝑣𝑐 = 𝑉𝑚 𝑐𝑜𝑠 𝑤𝑡 − 120𝑜 − 𝑐𝑜𝑠 𝑤𝑡 − 240𝑜
• = 𝑉𝑚[𝑐𝑜𝑠 𝑤𝑡 𝑐𝑜𝑠 120𝑜 + 𝑠𝑖𝑛 𝑤𝑡 𝑠𝑖𝑛 120𝑜 − 𝑐𝑜𝑠 𝑤𝑡 𝑐𝑜𝑠 240𝑜 −
𝑠𝑖𝑛 𝑤𝑡 𝑠𝑖𝑛 240𝑜 ] = 𝑉𝑚𝑠𝑖𝑛 𝑤𝑡
17

• From above, it is seen that

𝑣𝑜𝑙𝑡𝑎𝑔𝑒 𝑝𝑒𝑟 𝑝ℎ𝑎𝑠𝑒 = 𝑉
• For a 3-phase motor,
𝑐𝑢𝑟𝑟𝑒𝑛𝑡 𝑝𝑒𝑟 𝑝ℎ𝑎𝑠𝑒 = 𝐼
𝑣𝑜𝑙𝑡𝑎𝑔𝑒 𝑝𝑒𝑟 𝑝ℎ𝑎𝑠𝑒 = 𝑉
• For a 2-phase motor,
𝑐𝑢𝑟𝑟𝑒𝑛𝑡 𝑝𝑒𝑟 𝑝ℎ𝑎𝑠𝑒 = 𝐼

• Now
⁄
= ⁄
= ⁄
=

18

9
 12-Oct-21

• This shows that the parameters of the 2-phase induction motor are
times the corresponding parameters of the 3-phase induction motor. In
view of this, the parameters of the equivalent two-phase induction
motor are :
• Stator resistance = ∗ 0.30 = 0.20 
• Rotor resistance = ∗ 0.45 = 0.30 
• Stator and rotor leakage reactances = ∗ 2.1 = 1.4  each
• Magnetising reactance = 30 = 20 

19

• (b) When per phase turns of the 2-phase motor are times that of the
3-phase motor, then it can be shown that
per phase voltage = V
• for a 3-phase motor
per phase current = I

per phase voltage = V

• for a 2-phase motor,
per phase current = I
• Thus the parameters of the equivalent 2-phase induction motor are
times the corresponding parameters of the 3-phase induction motor
and their magnitudes are :

20

10
 12-Oct-21

• Stator resistance = ∗ 0.30 = 0.45 

• Rotor resistance = ∗ 0.45 = 0.675 
• Stator and rotor leakage reactances = ∗ 2.1 = 3.15  each
• Magnetising reactance = ∗ 30 = 45 
• (c) In this case it can shown that per phase values of 3-phase induction
motor are V, I and the corresponding per phase values of 2-phase
induction motor are 𝑉, 𝐼. Since the ratio of per-phase voltage
to per-phase current is the same in both the induction motors, the
parameters of the equivalent 2-phase induction motor are the same as
that of the 3-phase induction motor.
21

• The transformation equations giving the new currents iα, iβ in terms of ia, ib,
ic, i.e. can be expressed in matrix form as follows :
1 − − 𝑖𝑎
𝑖𝛼
• = 𝑖𝑏
𝑖𝛽 0 − 𝑖𝑐
• The transformation matrix, containing the constant coefficients, is a singular
one and thus ia, ib, ic can't be obtained from iα, iβ; since inverse of a singular
matrix does not exist.
• The matrix can be made square matrix if there is a third equation of
constraint between ia, ib, ic. Since the magnitude and direction of the m.m.f,
produced by two and three-phase systems are identical, the third current in
terms of ia, ib, ic should not produce any resultant air-gap m.m.f.
• The only obvious choice is the zero sequence current io which, for
convenience, is defined as :
• 𝑖𝑜 = ia, ib, ic
22

11
 12-Oct-21

• Note that zero sequence current does not produce any rotating m.m.f.
and here the factor is chosen arbitrarily to suit the
transformations.
• The three phase currents ia, ib, ic are thus replaced by two phase
currents iα, iβ plus zero sequence current io according to the equations,

• 𝑖𝛼 = 𝑖 − 𝑖 − 𝑖

• 𝑖𝛽 = 0+ 𝑖 − 𝑖

• 𝑖𝑜 = 𝑖 + 𝑖 + 𝑖

23

• In matrix form
1 − −
𝑖𝛼 𝑖𝑎
• 𝑖𝛽 = 0 − 𝑖𝑏
𝑖𝑜 𝑖𝑐

• The transformation matrix now is non singular and its inverse can be
easily obtained.

24

12