Module 4: Three-phase systems

Belén Garcı́a
Electrical Engineering Department

This module introduces the main concepts of balanced three-phase systems, defining the
properties of this type of configuration and justifying its application to power systems. We
will learn how to analyse the currents and voltages of three-phase systems using the one-phase
equivalent approach for wye and delta connected configurations. Finally, some power related
concepts applicable to these systems are introduced.

1 Configuration of three-phase systems

Power systems transfer big amounts of power from generators to loads. As explained in
previous sections, the design of power systems is conceived to minimize the power losses at
the lines using to this end power and distribution transformers. Transformers enable the
transmission of energy at high voltage levels increasing the efficiency of power systems.
Power systems are commonly constructed with a three-phase configuration, as the one
shown in the following figure:

Generator Load

Transmission line

1.1 Three-phase generators

Generators in power plants are generally three-phase as these type of rotating machines are
more robust from the mechanical point of view, have a lower cost and require a smaller space
to generate the same amount of power.
Three phase generators incorporate three electric circuits that provide a three phase system
of voltages, that is: three sinusoidal voltages of the same frequency and the same amplitude

1
and with relative phase-shift 120 o .

u
ua ub uc
ua(t) ub(t)

� t

uc(t)

For circuit analysis purposes three-phase generators are commonly represented as three
voltage sources connected between them. As will be explained throughout this module the
connection between phases can be done in wye or delta. The phases of a generator are wye
connected if three terminals, one of each phase, share a common point, which is called the
neutral point, while the other three terminals are connected to the system. Alternatively,
a generator is delta connected, when each source is connected to the other two sources
forming a triangle, as shown in the following diagram:

Wye connection Delta connection

A A

+
ua(t)
+

uc(t) ua(t)
N
+
uc(t) ub(t)
+ +

C B C B
+
ub(t)

This module is focused towards the analysis of three-phase balanced systems, which
are those in which the voltages and the currents of the three phases have the same amplitude.
A balanced three-phase system of voltages is composed of three sinusoidal voltages of the
same amplitude and a relative phase-shift 120o :

√
ua (t) = 2 · U · cos(ωt) (1)

√
ub (t) = 2 · U · cos(ωt − 120o ) (2)

2
 √
uc (t) = 2 · U · cos(ωt + 120o ) (3)

The phasors that represent this type of systems are:

Ua = U ∠0o (4)

Ub = U ∠−120o (5)

Uc = U ∠120o (6)

If the relative phase shift between the voltages are as in the previous equations (ua leading
ub and lagging uc ) we say that the system has positive sequence. In some cases, the
sequence of the phases changes and the voltage of phase b leads the voltage of phase a; these
type of systems have a negative sequence.

Uc Ub

120o 120o

Ua Ua

Ub Uc

Positive sequence Negative sequence

It is important to note that the sum of three sinusouidal functions of the same amplitude
and relative phase shift 120o is zero:

ua + ub + uc = 0 (7)

That statement is also valid for the phasors that represent the sinusoidal functions:

Ua + Ub + Uc = 0 (8)

1.2 Three-phase loads

Three-phase loads can be represented, in the frequency domain, as a set of three impedances
connected between them. We will limit our analysis to systems in which the impedance
connected to each phase has the same value; this is the most common situation in power
systems. As with generators, the three-phase loads may be connected in wye or delta:

3
 Wye connection Delta connection

A A

ZY

Z� Z�
N
ZY ZY

C B C B
Z�

1.3 Balanced three phase systems

The analysis of this course is restricted to balanced systems; in those systems the amplitude
of the voltages and the impedance of the load in the three phases is the same.
The figure below represent a three phase wye-wye system, as the three impedances and
the three generators share one terminal while the other terminal is connected to the system.
In wye wye systems the neutral points of the generators (N) and the loads (N’) might be
connected by means of a neutral wire forming a four-wire configuration. In this case, the
current of each phase flows from the generator towards the load (i.e in phase A the load flows
from A to A’) and returns through the neutral wire and the the neutral wire carries the sum
of currents Ia , Ib and Ic

A A'
Ia
+
Ua
ZY
IN=Ia+Ib+Ic
N N'

+ +
ZY ZY
C Uc Ub C'
B Ib B'

Ic

In the circuit of the figure, the currents that circulate through the three phases are:

Ua U ∠0 Ub U ∠−120o Uc U ∠120o
Ia = = Ib = = Ic = =
ZY ZY ZY ZY ZY ZY

4
 If the impedances are ZY = |ZY |∠θ, the phasor diagram of the system for inductive loads
and capacitive loads is:

Inductive load Capacitive load

Uc Uc
Ic
� Ia
�
o
120 Ic
�
Ib � Ua Ua
� Ia �
Ub Ib
Ub

The current that flows through the neutral is the sum of Ia , Ib , and Ic ; however, as the
three currents form a balanced three phase system (i.e. they have the same magnitude and
relative phase-shift 120 o ), the current that flows through the neutral is zero:

U
IN = Ia + Ib + Ic = · (1∠0 + 1∠ − 120o + 1∠120o ) = 0 (9)
ZY

From a practical point of view, the fact that no current flows through the neutral wire
implies that this conductor is suppressed in many occasions, moving towards a three wire
system as the one shown in the figure below. The suppression of the neutral wire implies a
significant reduction in the system construction and management costs (i.e. investment in
materials, infrastructure building, maintenance costs):

A A'
Ia
+
Ua
ZY

N N'

+ +
ZY ZY
C Uc Ub C'
B Ib B'

Ic

5
2 Voltages and currents in three-phase systems
2.1 Nomenclature: phase and line magnitudes
Before introducing the methodology that will be applied to the analysis of three phase systems,
we need to define some nomenclature:
• Phase-voltage: Voltage drop across a single phase of the generator or the load.
The phase-voltage of the generator (UPHASE GEN in the diagram) is the voltage drop
across the terminals of one of the ideal voltage sources; the phase voltage of the load
(UPHASE LOAD in the diagram) is the voltage drop across one of the impedances that
constitutes the load.
• Line-votage: Voltage drop between any couple of lines. We could obtain the line
voltage at the generator side of the system (ULINE GEN in the diagram), or the line
voltage at the load side (ULINE LOAD ).
• Phase current: Current in a single phase, i.e current flowing through one of the ideal
sources or through one of the impedances (IPHASE GEN , IPHASE LOAD ).
• Line current: Current in a single line (ILINE in the diagram).

A A'
+ + + +
IPHASE GEN ZTL ILINE I PHASE LOAD
+
Ua UPHASE GEN
ZY UPHASE LOAD
ULINE GEN ULINE LOAD
N - N' -

+ +
ZTL ZY ZY
Ub - -
C Uc C'
B B'
ZTL

2.2 Relation between the line and phase magnitudes in a wye-wye system
The system in the previous figure represents a wye connected three-phase generator that
supplies energy to a wye connected three phase load.
Applying KCL to the different nodes of the circuit it can be seen that for the three phases:

IL = IP h (10)

Regarding the phase voltage and line voltage across at the generator side of the system:

UP ha = Ua UP hb = Ub UP hc = Uc (11)

6
 And the line voltages are:

√ √
ULa = UAB = Ua − Ub = U ∠0 − U ∠−120o = 3 · U ∠30o = 3 · UP ha ∠30o (12)

√
ULb = UBC = Ub − Uc = 3 · UP hb ∠30o (13)

√
ULc = UCA = Uc − Ua = 3 · UP hc ∠30o (14)

The same relations between the line and phase magnitudes are valid for the load end of the
system if the load is wye connected.
The relation between the phase and line voltages can be also derived using the phasor-
diagram:

-Ua
Uc Uab

Uca 30o 120o -Ub

30o
Ua

Ub
30o
Ubc

-Uc

2.3 Relation between the line and phase magnitudes in a delta-delta system
If we now consider a delta-delta system 1 and determine its phase and line voltages and
currents:
1
For the sake of simplicity we consider that the impedance in the transmission line is zero

7
 A
+ + A'
ILINE
IPHASE GEN I PHASE LOAD
+
+
+
Uc ULINE LOAD Z� Z�
Ua ULINE GEN
UPHASE GEN UPHASE LOAD
+
-

- -
C C'
+ B B'
Ub Z�

We see that in this case the phase and line voltages of each phase are the same.

ULIN E = UP HASE (15)

The phase-currents in the load are:

Ua U ∠0
IP ha = IB 0 C 0 = = (16)
Z∆ Z∆

Ub U ∠−120o
IP hb = IA0 B 0 = = (17)
Z∆ Z∆

Uc U ∠120o
IP hc = IC 0 A0 = = (18)
Z∆ Z∆

And the line-current can be obtained using KCL:

U ∠0 U ∠−120o √ U √
ILa = IP ha − IP hb = − = 3· ∠−30o = 3 · IP ha ∠−30o (19)
Z∆ Z∆ Z∆

√
ILb = 3 · IP hb ∠−30o (20)

√
ILc = 3 · IP hc ∠−30o (21)

8
3 Analysis of three-phase systems: one-phase equivalent cir-
cuit
3.1 One-phase equivalent
Lets imagine a wye-wye balanced three-phase system where the load is connected with the
generator by means of a transmission line of impedance ZT L
IA
A A'
ZTL
+
ZTL UAN
ZY
ZN IA+IB+IC
N N'

+ +
IB ZY ZY
C C'
B ZTL B'

IC

ZTL

We could apply KVL to the three mesh of the circuit:

−Ua + Ia · (ZT L + ZY ) + (Ia + Ib + Ic ) · ZN = 0 (22)

−Ub + Ib · (ZT L + ZY ) + (Ia + Ib + Ic ) · ZN = 0 (23)
−Uc + Ic · (ZT L + ZY ) + (Ia + Ib + Ic ) · ZN = 0 (24)

Given that Ia + Ib + Ic = 0 the equations result:

−Ua + Ia · (ZT L + ZY ) = 0 (25)

−Ub + Ib · (ZT L + ZY ) = 0 (26)
−Uc + Ic · (ZT L + ZY ) = 0 (27)

As the total current flowing through the neutral wire is zero, the obtained equations are
identical to the ones that would be found if the three independent circuits shown below were
analysed:

9
 IA
A A'
ZTL +
+
UAN
UA'N' ZY
-
N -
N'

IB
B B'
ZTL +
+
UBN
UB'N' ZY
-
N -
N'

IC
C C'
ZTL +
+
UCN
UC'N' ZY
-
N -
N'

Moreover, as the system is balanced, the electric magnitudes of the three phases have the
same amplitude and a known phase shift (120o ). Then, the behaviour of the whole system
could be derived from the analysis of the following one-phase system shown below. This is
the so called one-phase equivalent or phase-neutral equivalent of the system.
IA
A A'
ZTL +
+
UAN
UA'N' ZY
-
N -
N'

Example

In the circuit of the figure UL = 380V , ZT L = 1 + jΩ and ZY = 50 + 50jΩ per phase.

Calculate the line current, the line voltage at the load side of the system and the voltage drop
across the transmission line

10
 ILA
A A'
ZTL
+

ZY

N N'

+ +
ILB ZY ZY
C C'
B ZTL B'

ILC

ZTL

Solution

The problem can be solved with the one-phase equivalent approach. We study what happens
in phase A and extrapolate the variables of the other phases.
The problem only indicate the magnitude of the line voltage (not the phase), so we can
choose any phase origin. We take the phase voltage of phase A as phase origin: UAN = 380
√ ∠0o
3

IA
A A'
ZTL=1+j� +
+
UAN=380/\/ 3 UA'N' ZY=50+50j�
-
N -
N'

UAN
IA = = 1.83 − 1.83j = 2.59∠−45o A
ZT L + ZY

For the other phases the currents are:

IB = 2.59∠−45 − 120 = 2.59∠−165o A

IC = 2.59∠−45 + 120 = 2.59∠75o A

Voltages at the load end of the system:

UA0 N 0 = IA Y = 182.83∠0o V

11
 Line voltage:

√
ULload = Uphload · 3 = 316.66V

Phasors line voltage at the load end of the system:

UA0 B 0 = 316.66∠30o V
UB 0 C 0 = 316.66∠−90o V
UC 0 A0 = 316.66∠150o V

Voltage drop across the transmission line:

UAA0 = IA · ZT L = 36.56∠0o V

∆UT L = 36.56V

3.2 Analysis of systems with delta-connected loads or generators

The application of the one-phase equivalent is not possible in systems in which the generator or
the load are in delta. However we can always replace the delta configuration by an equivalent
wye.

3.2.1 Delta-connected generators

Three delta-connected generators can be redrawn as an equivalent wye configuration:

A
+ + A

+
UPhY
+
UL� ULY
UPh�
N
- +

+ +
- -
B + C B C

The two configurations are equivalent if the line voltages are the same for both of them:

ULY = UL∆ (28)

As can be seen in the diagram, the phase voltages of the delta and wye generators will be
different.

12
 UP h∆
UP hY = √ ∠−30o (29)
3

3.2.2 Delta connected loads: wye delta equivalent

We want to find what should be the value of ZY that makes the two loads in the figure
equivalent.

A A
+ + IAY
IA�

ZY

UAB� UABY
Z� Z� N
ZY ZY

-
- C
C IBY B
IB� B Z� ICY
IC�

The loads are equivalent if for the same applied line voltages:

UAB∆ = UABY UBC∆ = UBCY UCA∆ = UCAY (30)

the line currents are the same:

IA∆ = IAY IB∆ = IBY IC∆ = ICY (31)

We can limit our analysis to one phase, since the behaviour in the three phases is the same
except that there is a phase shift of 120o

√
√ o UAB · 3∠−30o
IA∆ = 3 · IP hA ∠−30 = (32)
Z∆
√
UAB / 3∠30o
IAY = IP hA = (33)
ZY

We determine for what value of the impedance ZY the identity IA∆ = IAY is true

√ √
UAB · 3∠−30o UAB / 3∠30o
= (34)
Z∆ ZY

13
 finding that:

Z∆
ZY = (35)
3

3.2.3 One-phase equivalent for systems with delta connected-elements

In systems with a delta-connected load or a delta-connected generator, it is not possible to
apply the one-phase equivalent approach directly, since there is no neutral point.However it
is possible to apply a wye delta transformation to obtain a YY connected system equivalent
to the original.
In Y∆ or ∆∆ or ∆Y systems:
1. The system is transformed into a YY equivalent system.
2. The one-phase equivalent approach is applied over the YY equivalent circuit.

Example

In the following circuit, the load is supplied with 400 V. Calculate the line currents and
the phase currents and the line voltage at the generator end of the system. ZT L = 1 + 4jΩ,
Z∆ = 30 + 60jΩ.
ILA
A A'
ZTL I PhA
+ I PhC

+
Z� Z�
N
-
+ +
ILB -
C C'
B ZTL B' I PhB

ILC

ZTL

Solution
We transform the original system into a YY equivalent circuit. As the systems are equiva-
lent, the line voltages and line currents remain unchanged.

14
 ILA
A A'
ZTL
+

ZY

N N'

+ +
ILB ZY ZY
C C'
B ZTL B'

ILC

ZTL

Where the impedances in Y are:

Z∆
ZY = = 10 + 20jΩ
3

and the line voltage at the load end is 400 V.

The one-phase equivalent of the system is:
IA
A A'
ZTL=1+4j� +
+
UAN UA'N'=400/\/ 3 ZY=10+20j�
-
N -
N'

In this case we take the phase voltage at the load end of the system as phase origin for the
sake of simplicity.

U A0 N
IA = = 4.62 − 9.24j = 10.33∠−63.43o A
ZY

For the other phases the line currents are:

IB = 10.33∠−63.43 − 120 = 10.33∠−183.43o A

IC = 10.33∠−63.43 − 120 = 10.33∠56.56o AA

15
 The line currents are the same for the YY and the Y∆ circuits. The phase currents for the
original system are:

I
IphA = √B ∠30o = 5.96∠−33.43o A
3

and for the other phases:

IphB = 5.96∠−153.43o A
IphC = 5.96∠86.57o A

Phase voltage at the generator end of the system:

UAN = IA · ZT L + UA0 N 0 = 272.66∠1.94o V

Line voltage at the generator:

√
ULg = Uphg · 3 = 472.27V

Phasors line voltage at the load end of the system:

UA0 B 0 = 472.27∠1.94 + 30o V = 472.27∠31.94o V

UB 0 C 0 = 472.27∠−88.06o V
UC 0 A0 = 472.27∠151.94o V

Voltage drop across the transmission line:

UAA0 = IA · ZT L = 42.58∠12.53o V

∆UT L = 42.58V

4 Power in three-phase circuits

4.1 Instantaneous power
To calculate the instantaneous power of a three phase load or generator, we must add the
power of each phase.

p(t) = ua (t) · ia (t) + ub (t) · ib (t) + uc (t) · ic (t) (36)

In a three phase system were:

√
ua (t) = 2 · U · cos(ωt)

16
 √
ub (t) = 2 · U · cos(ωt − 120o )
√
uc (t) = 2 · U · cos(ωt + 120o )

and

√
ia (t) = 2 · I · cos(ωt − ϕ)
√
ib (t) = 2 · I · cos(ωt − 120o − ϕ)
√
ic (t) = 2 · I · cos(ωt + 120o − ϕ)

p(t) = 2·U ·I·(cos(ωt)·cos(ωt−ϕ)+cos(ωt−120o )·cos(ωt−120o −ϕ)+cos(ωt+120o )·cos(ωt+120o −ϕ)

applying:

1
cos α · cos β = · (cos(α + β) + cos(α − β) (37)
2

p(t) = U · I · (3 · cos ϕ + cos(2ωt − ϕ) + cos(2ωt − 120o − ϕ) + cos(2ωt + 120o − ϕ))

| {z }
0

Since the sum of three sinusoidal functions with relative phase shift 120o is zero we find that
the instantaneous power of a three phase system is constant and depends on the amplitude
of the current, the amplitude of the voltage and the power factor:

p(t) = 3 · U · I · cos ϕ (38)

The fact that the three-phase instantaneous power is constant has practical implications,
as the fact that vibrations in the axles of three phase motors and generators are smaller that
those in one-phase devices. This makes them more stable from the mechanical point of view,
improve their reliability and extend their service life.

4.2 Active and reactive power of a three-phase load

4.2.1 Wye-connected systems
Lets consider a three-phase load with impedance per phase ZY = |ZY |∠θ

17
 ILA
+ A
I PhA +

ZY UPhA
ULA
-
N

- ZY ZY
C
B

The active power absorbed by the load is the sum of the active power absorbed by each
individual impedance:

P = PA + PB + PC (39)

where

PA = UP hA · IP hA · cos ϕ PB = UP hB · IP hB · cos ϕ PC = UP hC · IP hC · cos ϕ

as the three phase voltages and the three phase currents have the same magnitude (U):

PA = PB = PC = UP h · IP h · cos ϕ (40)

And the resulting active power is:

P = 3 · UP h · IP h · cos ϕ (41)

The power can√ also be expressed as a function of the line magnitudes. As in Y IL = IP h

and UP h = UL / 3:

√
P = 3 · UL · IL · cos ϕ (42)

Equations (41) and (42) are also valid to calculate the power generators
Following the same reasoning we can find equations to calculate the reactive power of a
wye-connected load or generator:

√
Q = 3 · UP h · IP h · sin ϕ = 3 · UL · IL · sin ϕ (43)

18
4.2.2 Delta connected systems
ILA
+ A

I PhA
+
ULA Z� Z�
UPhA
-

-
C'
B
Z�

For delta-connected loads and generators we can still apply:

P = 3 · UP h · IP h · cos ϕ (44)
Q = 3 · UP h · IP h · sin ϕ (45)

To express the active ans reactive power as a function of the line magnitudes we consider√the
relations between phase and line magnitudes in delta systems: UL = UP h and IP h = IL / 3:

√
P = 3 · UL · IL · cos ϕ (46)
√
Q = 3 · UL · IL · sin ϕ (47)

4.2.3 General expressions for the active and reactive power of a three-phase
system
From the previous reasoning it can be concluded that the expressions to calculate the active
and reactive power of a three-phase are the same for Y and delta systems and are:

√
P = 3 · UP h · IP h · cos ϕ = 3 · UL · IL · cos ϕ (48)

√
Q = 3 · UP h · IP h · sin ϕ = 3 · UL · IL · sin ϕ (49)

4.3 Complex power of three phase generators

To calculate the complex power of a three-phase generator, connected in wye or delta is the
sum of the complex power of the three phases

19
 A
A
I PhA
+ I PhA I PhC
UPhA
+ -
UPhA
UPhC
N
UPhB UPhC - +

+ +

I PhB I PhC
B C B - C
I PhB +
UPhB

Sg = UP hA · I∗P hA + UP hB · I∗P hB + UP hC · I∗P hC = 3 · UP hA · I∗P hA (50)

5 Power factor correction

5.1 Power factor
To characterize an AC load it is necessary to specify its rated power but also its power factor.
The power factor provides information about the proportion between the active power and
reactive power absorbed by it.

p.f. = cos ϕ (51)

The angle ϕ may be expressed as the phase-shift between the current and the voltage across
the impedance, the arch-tangent of the ratio between Q and P and the angle of the complex
impedance:

Im
U
� I
I U Z
-
� X
+

Z Re R

Q X
ϕ = ϕu − ϕi = arctan = arctan (52)
P R
If ϕ > 0 we say that the power factor is lagging or that it is inductive, if ϕ < 0 the power
factor is lagging or capacitive.
The following table summarizes the values of the reactive power and power factors of the
different types of loads. Resistive systems have power factor 1, what means that there is no
absorption of reactive power.

20
 Q ϕ cos ϕ character
Resistive loads 0 0 1 -
Inductive loads >0 >0 0 < p.f < 1 lagging or inductive
Capacitive loads <0 <0 0 < p.f < 1 leading or capacitive

5.2 Reactive power compensation

In AC systems there is a continuous exchange of energy between generators and capacitive
and inductive loads. The amplitude of the fluctuating power is defined as the reactive power
and is of different sign in both types of elements. As inductors absorb reactive power (QL > 0)
capacitors deliver reactive power (QC < 0).
Many real-life loads, as electric motors, are highly inductive and it is common that the
operation of power systems electric systems involve high amounts of reactive power transferred
from the generators towards the loads. The fluctuating power increases the current flowing
through the lines what impacts in the system losses and causes voltage drops. For that reason,
electric companies penalize the costumers that consume power with poor power factor.

P, Q
ZTL

Generator Inductive load

P, Q

The solution given to this problem by the industry consists of adding banks of capacitors
in parallel with the loads, to compensate part of the reactive power absorbed by them. In
this case the generator only supplies a portion of the reactive power demanded by the load
while the most of the required reactive power is delivered by the capacitors:

21
 P, Q'
ZTL

Generator Inductive load

P, Q

Capacitors
QC

It is important to note that capacitors do not absorb or deliver any active power, so the
active power of the system remains unchanged despite of connecting them.
If the powers are plotted in a power triangle (the active and reactive power absorption in
the line are neglected in the reasoning for simplicity’s shake), we see how the relation between
the active an reactive power changes and the angle ϕ0 becomes smaller if the capacitors are
added to the system:

Initial system System with capacitors

Q
S
Q S'
� �' Q'=Q+QC
P P'=P
QC

The reactive power delivered by the generator if the capacitors are added is:

Q0 = Q + QC (53)

5.3 Determination of the capacitance of the capacitor bank

5.3.1 Reactive power of a capacitor
If we have a capacitor of capacitance C and with a voltage drop U = U ∠ϕu and current flow
I = I∠ϕi :

I U
+ -

ZC=-j/�C

22
 Its reactive power is:

U2
QC = XC · I 2 = = −ω · C · U 2 (54)
XC

5.3.2 Reactive power of a bank of capacitors in wye

The total reactive power of a bank of capacitors connected in wye is the sum of the power
of the reactive power of the three capacitors and can be expressed as a function of the phase
voltage or the line voltage

ZTL
+
UL
Generator Inductive load
P, Q

CY

+ UPh

2
QCY = −3 · ω · CY · UP h = −ω · CY · UL2 (55)
√
|{z}
UL / 3

5.3.3 Reactive power of a bank of capacitors in delta

If the capacitors are delta connected the expression of the power in terms of the phase voltage
is the same but if it is expressed as a function of the line voltage, we see that for the same
capacitors we extract three times more power if they are connected in delta. For this reason
this is the most commonly used configuration for the capacitor banks that are used for power
compensation purposes.

23
 ZTL
+
UL
Generator Inductive load
P, Q

+ UPh

C�

2
QC∆ = −3 · ω · C∆ · UP h = −3 · ω · C∆ · UL2 (56)
|{z}
UL

5.3.4 Capacitance required to get a target value for the power factor
If we have a system that it is working with power factor cos ϕ and we want to compensate
the reactive power so that the power factor is corrected to cos ϕ0 , as in the next diagram:

�
�' Q'=Q+QC

P'=P
QC

Q = P · tan ϕ Q0 = P · tan ϕ0 (57)

QC = Q − Q0 = P · (tan ϕ − tan ϕ0 ) (58)

Then, if the capacitors are ∆-connected:

P · (tan ϕ − tan ϕ0 )
C∆ = (59)
3 · ω · UL2

And, if the capacitors are Y-connected:

P · (tan ϕ − tan ϕ0 )
CY = (60)
ω · UL2

24
6 Measure of power in three-phase systems
6.1 Working principle of watmeters
A wattmeter is a measuring device that provides information on the power absorbed by
electric dipoles. Wattmeters incorporate two measuring circuits: one to measure the current,
the current coil, and another to measure the voltage, the voltage coil. The current coil must
be connected in series with the dipole, while the voltage coil is connected in parallel with it.
Two signs * mark the terminals of the current and voltage coil of the same relative polarity:

A W
Electric
dipole
B

The reading of the wattmeter of the figure is:

W = IA · UAB · cos(U\
AB IA ) (61)

In three phase systems wattmeters may be connected in different ways to measure the power;
the connection form depends on the configuration of the system that is being characterized
and the variable that is searched. The following subsections provide some examples for
connections that are used in real systems to measure the active and reactive power of three
phase systems.

6.2 Measure of the active power in systems with accessible neutral point
In the system in the figure, the wattmeter is measuring the active power of the impedance in
phase A. The active power of the three phase load is calculated as 3 · W .

A W

B N

P
W = IA · UAN · cos(U\
AN IA ) = UP h · IP h · cos ϕ = (62)
3

6.3 Measure of the reactive power of a three-phase system with one wattmeter
The reactive power of a balanced three phase load can be measure with one wattmeter, even
if there is not neutral point or there is no access to it. The wattmetter must be connected to

25
measure the current flowing through one phase and the voltage drop between the other two
phases:

A W
Balanced
B three-phase
load

The wattmeter of the figure measures:

W = UBC · IA · cos(U\
BC IA ) (63)

The value of the angle U\ BC IA can be determined using a phasor diagram. As can be seen
in the phasor diagram provided below, the angle equals 90 − ϕ for an inductive load. If the
BC IA is 90 + ϕ:
load is capacitive, the current leads the voltage by ϕ, and the angle U\

UC
IC
�

UA
�
IB � ���� IA
o
30
UB
UBC

-UC

Q
BC IA ) = UL · IA · cos(90 − ϕ) = UL · IA · sin(ϕ) = √
W = UL · IL · cos(U\ (64)
3

6.4 The two wattmeters method

The two wattmeters method allows us to measure the active an reactive power of balanced
three-phase systems. Two wattmeters W1 and W2 are connected as shown in the following
diagram:

26
 A W1
Balanced
B W2 three-phase
load
C

The wattmeters measure:

W1 = UAC · IA · cos(U\
AC IA ) (65)

W2 = UBC · IB · cos(U\
BC IB ) (66)

The values of the angles can be found using the phasor diagram:

UC
IC
�

UA
���� �
IA -UC
IB � ���� 30o
30o
UB UAC
UBC

-UC

√
3 1
W1 = UL · IL · cos(U\
AC IA ) = UL · IL · cos(30 − ϕ) = UL · IL · ( cos ϕ + · sin ϕ) (67)
2 2

√
3 1
W2 = UL · IL · cos(U\
BC IB ) = UL · IL · cos(30 + ϕ) = UL · IL · ( cos ϕ − · sin ϕ) (68)
2 2

The active and reactive power of the three phase system can be obtained as the sum and
the difference of the readings of the two wattmeters.

√
W1 + W2 = 3 · UL · IL · cos ϕ = P (69)

27
 Q
W1 − W2 = ·UL · IL · sin ϕ = √ (70)
3

28