Notes 5: Symmetrical Components

5.0 Introduction

When performing steady-state analysis of

high voltage transmission systems, we make
use of the per-phase equivalent circuit.

For distribution systems, however, there are

two major impediments to using per-phase
equivalent circuits for steady-state analysis.
These impediments are:
1. The flows are not balanced.
2. The mechanical design of the conductors
is asymmetric.

The meaning of #2 needs some elaboration.

A symmetric mechanical design is one in
which each phase conductor sees exactly the
same inductive effects as every other
conductor comprising the circuit. The
equilateral configuration, shown in Fig. 1, is
a symmetric mechanical design.

1
 Fig. 1: Equilateral Configuration

Because of many different influences on

design of conductor configuration (e.g., cost
of right-of-way, clearance between phases,
costs and strength of tower/pole and
corresponding supports, etc.), the equilateral
configuration is not often utilized.

Nonetheless, it is possible to achieve the

effect on inductance of a symmetric design
by transposing the conductors so that each
conductor resides in the same physical
location for 1/3 the total length of the line.

Yet, transposition costs money, and because

distribution systems typically operate
unbalanced anyway, it is usually not done
for distribution feeders.

2
The effect of asymmetry is often ignored,
however, and analysis is done on the feeder
assuming the design is symmetric. Although
this approach does incur error, it is common
because the alternative is more laborious.

In that case, we have only the problem of

unbalanced loading on the phases.

There is a very elegant approach available

for analyzing symmetric three-phase circuits
that are unbalanced. The approach was
developed by a man named Charles
Fortescue and reported in a famous paper in
1918. It is now called the method of
symmetrical components. We will spend a
little time studying this method before we
move on to a more general method.

3
5.1 Symmetrical Components: Motivation

Def: A symmetrical set of phasors have

equal magnitude & are 120º out of phase.

Goal: Decompose a set of three

unsymmetrical phasors into
 One unsymmetric but equal set
 Two symmetrical sets
Then we can analyze each set individually.

In what follows, we will illustrate that:

 Step 1: A unsymmetrical set, not summing

to 0, can be decomposed into two

unsymmetrical sets:
o an “equal” set and an
o unsymmetrical set that does sum to 0;
 Step 2: An unsymmetrical set that sums to

0 can be decomposed into two

symmetrical sets

4
Step 1: Consider a set of phasors that do not
add to zero (because of different magnitudes
or because of angular separation different
than 120º or because of both). Assume that
they have phase sequence a-b-c.

Add them up, as in Fig. 2, i.e.,

V R0 = Va + Vb + Vc (1)

a-b-c a-b-c
Vc Vc
Va Va

VR0

Vb Vb

Fig. 2: Addition of Unsymmetrical Phasors

So we see from (1) that

Va +Vb +Vc −VR0 = 0 (2)
Define:
1
V0 = VR0
3 (3)
Then:
Va +Vb +Vc −3V0 = 0
(4)
(Va −V0 ) + (Vb −V0 ) + (Vc −V0 ) = 0 (5)
5
Define:
VA = (Va −V0 )
VB = (Vb −V0 ) (6)
VC = (Vc −V0 )

Then:
(7)
V A +VC +VC = 0

Conclusion: We obtain an unsymmetrical set

of voltages that sum to 0 by subtracting V0
from each original phasor, where V0 is 1/3
of the resultant phasor, illustrated in Fig. 3.
a-b-c
VC Vc
VA

VR0 Va

VB
Vb

-V0=-VR0/3

Fig. 3: Subtracting V0 from unsymmetrical

phasors
Step 2: How to decompose VA, VB, and VC
into two symmetrical sets?

Can we decompose VA, VB, VC into 2 a-b-c

symmetrical sets?

6
As a test, try to add any 2 a-b-c symmetrical
sets and see what you get. See Fig. 4.
Vc1
a-b-c a-b-c
Vc2

Va1 Vb2

Vb1 Va2

VC

a-b-c
Vc1
Vc2

Vb2
Va1

VB

Vb1 Va2
VA

Fig. 4: Adding 2 symmetrical a-b-c sets

Note that in adding the 2 phasor sets, we add
the two a-phase phasors, the two b-phase
phasors, and the two c-phase phasors.

7
One can observe from Fig. 4 that the
resultant phasor set, denoted by the solid
lines, are in fact symmetrical!

In fact, it is possible to prove

mathematically that the sum of any 2 a-b-c
symmetrical sets is always another
symmetrical set.

Let’s try a different thing. Let’s try to add

two symmetrical sets, but let’s have one be
a-b-c (called positive sequence) and another
be a-c-b (called negative sequence).

As before, in adding the 2 phasor sets, we

add the two a-phase phasors, the two b-
phase phasors, and the two c-phase phasors.
The result of our efforts in shown in Fig. 5.

8
 Vc1
a-b-c a-c-b
Vb2

Va1 Vc2

Vb1 Va2

a-c-b

VC Vc1 Vb2

Vc2 VB
Va1

Vb1 Va2
VA

Fig. 5: Adding a symmetrical a-b-c set to a

symmetrical a-c-b set

The resultant phasor set is unsymmetrical!

We can guarantee that these unsmmetrical

phasors sum to zero, since we just added
two phasor sets that sum to zero, i.e.,

9
Va1+Vb1+Vc1=0
Va2+Vb2+Vc2=0
------------------- (8)
VA+VB+VC=0

Now consider Fig. 5 again. Assume that

someone hands you the unsymmetrical set of
phasors VA, VB, and VC.

Can you decompose them into the two

symmetrical sets?

Can you be assured that two such

symmetrical sets exist?

The answer is yes, you can be assured that

two such symmetrical sets exist. Fortescue’s
paper contains the proof.

I will simply argue that the three phasors

given in Fig. 6, VA, VC, and VC, are quite
general, with the single exception that they
sum to zero.

10
Claim: We can represent ANY
unsymmetrical set of 3 phasors that sum to 0
as the sum of 2 constituent symmetrical sets:
 A positive (a-b-c) sequence set and
 A negative (a-c-b) sequence set.

Given this claim, then the following theorem

holds.

Theorem: We can represent ANY

unsymmetrical set of 3 phasors as the sum of
3 constituent sets, each having 3 phasors:
 A positive (a-b-c) sequence set and
 A negative (a-c-b) sequence set and
 An equal set

These three sets we will call, respectively,

 Positive (V , V , V )
1 1 1
a b c

 Negative (V , V , V )
2 2 2
a b c

 zero (V , V , V )
0 0 0
a b c

sequence components.

11
The implication of this theorem that any
unsymmetrical set of 3 phasors Va, Vb, Vc
can be written in terms of the above
sequence components in the following way:
Va = Va0 + Va1 + Va2

Vb = Vb0 +Vb1 +Vb2 (9)

Vc = V +V +V
c
0
c
1
c
2

We can write the equations of (9) in a more

compact fashion, but first, we must describe
a mathematical operator that is essential.

5.2 The a-operator

To begin on familiar ground, we are all

conversant with the operator “j” which is
used in complex numbers.

Remember that “j” is actually a vector with

a magnitude and an angle:
j =1∠90 °
(10)

12
In the same way, we are going to define the
“a” operator as:
a =1∠120 ° (11)

It is easy to show the following relations:

a = 1∠−120 °
2
(12)
a =1∠0°
3
(13)
a = 1∠120 ° = a
4
(14)
We also have that:
1 + a = −a = 1∠60 °
2
(15)
as illustrated in Fig. 6

Fig. 6: Illustration of 1+a

Note that
(16)
− a 2 = −1∠240 ° =1∠60 °

Similarly, we may show that:

1 + a = −a = 1∠ − 60 °
2
(17)
1 −a = 3∠−30 ° (18)
1 −a = 3∠30 °
2
(19)
13
a −1 = 3∠150 ° (20)
a −1 = 3∠−150 °
2
(21)
And there are many more relations like this
that are sometimes helpful when dealing
with symmetrical components. (See the text
called “Analysis of faulted power systems”
by Paul Anderson, pg. 17.)

5.3 Symmetrical components: the math

We repeat equations (9) below for

convenience:
Va = Va0 + Va1 + Va2

Vb = Vb0 +Vb1 +Vb2 (9)

Vc = V +V +V c
0
c
1
c
2

We can relate the three different quantities

having the same subscript.

 Zero sequence quantities: These quantities

are all equal, i.e.,
V =V =V
a
0
b
0
c
0
(22)

14
 Positive sequence quantities: The relation
between these quantities can be observed
immediately from the phasor diagram and
can be expressed using the a-operator.
Vc1
a-b-c

Va1

Vb1

Fig. 7: Positive sequence components

V 1
=a V
2 1

Vc1
b

= aV a
a
1 (23)

 Negative sequence quantities: The relation

between these quantities can be observed
immediately from the phasor diagram and
can be expressed using the a-operator.
a-c-b
Vb2

Vc2

Va2

Fig. 8: Negative sequence components

15
Vb2 = aV a2
Vc2 = a 2Va2 (24)

Now let’s use equations (22), (23), and (24)

to express the original phasor Va, Vb, Vc in
terms of only the a-phase components
V ,V ,V , a
0
a
1
a
2

i.e., we will eliminate the b-phase

components
Vb0 ,Vb1 ,Vb2

and the c-phase components

Vc0 ,Vc1 ,Vc2

This results in
Va = Va0 + Va1 + Va2

Vb = Va0 + a 2Va1 + aV a2 (9)

Vc = V + aV + a V
a
0
a
1 2
a
2

So we have written the abc quantities (phase

quantities) in terms of the 012 quantities

16
(sequence quantities) of the a-phase. We can
write this in matrix form as:

Va  1 1 1  Va0 
V  = 1 a 2  
 b  a  Va1  (25)
Vc  1 a a 2  Va2 

Defining
1 1 1
A = 1 a 2 a 
1 a a 2 
(26)
we see that eq. (25) can be written as:
Va  Va0 
V  =  
 b AVa1 
Va2 
(27)
Vc   

We may also obtain the 012 (sequence)

quantities from the abc (phase) quantities:
Va0  Va 
 1 −1  
Va  = A Vb 
Va2 
(28)
  Vc 

where
1 1 1
1
a2 
A
−1
= 1
3

a
a2 a
 (29)
1 

Equations 22-29 hold for

 Line-to-line voltages
 Line-to-neutral voltages

17
 Line currents
 Phase currents

18