Voltage sags in industrial systems

G. Alonso Orcajo, J. M. Cano R., C. H. Rojas G., M. G. Melero, M. F. Cabanas, F. Pedrayes

Dpto. de Ingeniería Eléctrica, Electrónica, de C. y S.,

Edificio departamental nº 4, Campus de Viesques s/n.
University of Oviedo
33204. Gijón ASTURIAS
Phone: +34985182623 Fax: +34985182453
gonzalo@correo.uniovi.es
Abstract 2. Characterization of voltage sags
This paper presents results relating to the effect of The recorded results are shown for a sag due to a
voltage sags in industrial systems on adjustable distribution line fault at 22 kV and propagates
speed drives. The influence of the source down to 400 V system. The monitoring
impedance, dc link capacitor, characteristics of equipment is installed at the low voltage line side
voltage sags and the effect of the utility is of the circuit. The voltage sag is characterized by
computed. The results from the analysis can be [1], [2], [3], [4]:
used during the design task of the power system
- Voltage waveform record
protections and in order to improve the
susceptibility of the adjustable speed drives - Magnitude and duration
(ASD´s) under sags. Simulation and experimental
- Sequence components of the voltage
results are provided.
- Phase angle shift associated
Keywords: sags, PWM drives, effects
Both unbalance and phase angle shift are likely to
1. Industrial distribution system under have an important influence on nuisance tripping
study of equipment.
The power system components for voltage sag
studies have been modelled and simulated by
using a transients program [14]. The simulated
power system under study includes the following
components:
- Utility transmission impedance
- Distribution transformer
- AC line inductors
- PWM adjustable-speed drive
- Induction motor driven by the ASD
- Vector control system
- Controlled short-circuit connection to
establish voltage sag phenomena Fig. 2. Voltage waveform.

AC Motor Drive

Distribution Transformer
system w

G M Lo

Three-phase TL
electric utility
Control
system

System fault

Load

Fig 1. Industrial distribution system

Fig. 3: Magnitude and duration
 60

50

40

30

20

10

10

20

30
400 300 200 100 0 100 200 300 400

Fig 4. Sequence components of the voltage harmonic

spectrum

Fig. 8. Waveforms of the phase voltages during a type CS

- Normal conditions
sag
- Under the sag Case C. System under type D sag voltage. Sags
type D are dips with the main drop in one phase.
To identify the sagged phase a subscript is added,
hence DR would be a dip with the main drop in
phase R.
Fig. 5. Three-phase phasor diagram during and before
the sag

Fig 6. Phase angle between the voltage during the sag

and during the normal evolution
3. Voltage sags
Three cases have been studied.
Case A: System under type A sag voltage. The
first one is a sag under a three-phase fault. The
voltage during the event is equal in the three
phases.
Fig. 7. Phase voltages during a type A sag
Case B: System under type C sag voltage. Sags
type C are dips with the main drop between two
phases. A subscript identifies the less sagged
phase. Hence CR would be a sag with the main
drop between phase S and T.
Fig. 9.Phase voltages during a type DR sag
4. Effect of power system utility
The effect of the transformer and the location of the
fault point have been studied [5], [6], [7], [8], [9].
A- Influence of the transformer winding
connection
Depending on the winding connections of the
transformer the retained voltages seen at
secondary side due to an unsymmetrical fault at
the primary side may be modified. This means
that the dip type may change when going through
the transformer.
The transformer between the faulted point and the
observation bus is a Dyn transformer. The
recorded results are shown for a sag due to a
transmission line fault at 22 kV and propagates
down to 400 V. In this case the load and the fault
point are at different sides of a transformer with
delta winding. The positive-sequence impedance-
matrix is equal to the negative-sequence.
Case I: Three-phase fault
A three-phase fault causes a balanced sag, type A.
There is neither negative nor zero-sequence
voltage. The sag is characterised by its phase- Case III- Single Phase to ground fault
retained voltage, which is of positive-sequence.
The phase voltages will not contain a zero
4

1.6 10
4
2 10
sequence component because the transfer
4
1.2 10

8000
impedance is null. The non-faulted phases show
A

B
i

i
4000

0
the same magnitude and direction in their voltage
C
i
4000

8000
change. These voltage changes are half of the
1.2 10

1.6 10
4

4
value of the magnitude of the voltage change in
2 10
4
2.9 2.96 3.02 3.08 3.14 3.2

i. 10
5.
8
3.26 3.32 3.38 3.44 3.5 the faulted phase.
Fig.10.Phase voltages during a type A sag at the 22 kV bus In this case, the voltage sag is a DR sag. A single
500 phase to ground fault involving phase R causes a
400
DR sag. However the Dyn transformer change the
300

200
character of the unbalanced sag.
I
i 100
4
2 10
J 0
i 4
1.6 10

Q 100 1.2 10
4
i
200 8000
A
i 4000
300
B 0
i
400
C 4000
i
500 8000
3 3.1 3.2 3.3 3.4 3.5
4
6. 1.2 10
i . 10 80
4
1.6 10

Fig.11.Phase voltages at the 400 V bus. Sag A 2 10

4
2.98 3.004 3.028 3.052 3.076 3.1 3.124 3.148 3.172 3.196 3.22
5
i. 10 . 8
propagated down from 22 kV bus is an A sag at 400 V Fig.15.Phase voltages during a type DR sag at the 22 kV
Case II- Phase to phase fault bus
500

During this fault, the non-faulted phase does not 400

300

see any change in the voltage. The change in the F

200

i 100

voltages of the faulted phases is equal in G

i 0

magnitude but opposite in direction. In the H

i
100

200

analysed example, the voltage sag is a C sag. A 300

400

single phase to phase fault involving phase R and 500

2.98 3 3.02 3.04 3.06 3.08 3.1
6
i. 10 . 80
3.12 3.14 3.16 3.18 3.2 3.22

S causes a CT sag. Fig. 16.Phase voltages at the 400 V bus. Sag DR

2.4 10
4 propagated down from 22 kV bus is a CS sag at 400 V
4
1.6 10 250
A
i
8000 225
B
i
200
0
C
i 175
4
1.2 10 F
i 150
4
2 10 G 125
2.96 3.012 3.064 3.116 3.168 3.22 i
5
i . 10 . 8 H 100
i
Fig.12.Phase voltages during a type CT sag at the 22 kV bus 75

50

However the Dyn transformer change the 25

character of the unbalanced sag. Sag CT 0

2.9 2.95 3 3.05 3.1 3.15 3.2 3.25 3.3 3.35
6
i. 10 . 80
propagated down from 22 kV bus is a DR sag at
Fig.17.Shape of the sag in the secondary side
400 V.
400

320
5. Effect of voltage sags on adjustable
240

160
speed drives
I
i 80

J
i 0
The responses of ASDs to voltage sags conditions
Q
i
80
can significantly affect the industrial processes
160

240
that use them. The general layout for PWM ASDs
320 has a rectifier, a DC bus and an inverter. During a
400
2.95 3 3.05

i . 10
3.1
6.
80
3.15 3.2 3.25
voltage sag, the response of the electronic ASDs
Fig.13.Phase voltages at the 400 V bus. depends on the hardware design, the controller
150
software, and the control system response time.
Voltage sags cause a decrease of the DC bus
F

2.36
i
100
voltage in the ASD. During very brief sags it may
G
i
2.36
be possible to supply the energy from the DC bus
H
i 50
capacitor. During longer sags periods, the DC bus
2.36

voltage will drop to a lower level. If this falls

0
2.9 2.95 3 3.05 3.1 3.15 3.2 3.25 3.3 3.35
bellow the DC bus trip voltage then the inverter
6
i. 10 . 80 will trip [10], [11], [12], [13].
Fig.14 .Shape of the sag in the secondary side
 600
Case I- Three-phase sag 600
No load
300 550 0,5 p.u
VR, VS, VT [Vrms] Xi

X2i 500
200
X3i

450 1,0 p.u

100
0.73 p.u. 400 400
1.95 2 2.05 2.1
t[s] 1.95 −5 2.15
i ⋅4 ⋅10
0
1.95 2 2.05 2.1 2.15 Fig 22. Decrease of the DC bus voltage under different
load conditions (1.0 p.u., 0.5 p.u. and no load)
Fig.18.Shape of the sag in the secondary side
40
At the occurrence of a three-phase or balanced 1,0 p.u
sag, no energy is supplied from the AC-supply
into the DC bus because of the reverse biased 20 0,5 p.u
diode rectifier. All the energy needed to drive the
load is taken from the electric energy stored in the No load

DC bus capacitor. As a result, the dc bus voltage

0
starts to decay according the dc bus capacitor, the 1.95 2 2.05 2.1 2.15

load torque, the mechanical speed and the inverter Fig 23. Increase of the DC bus current under different
and motor efficiency [15], [16], [17], [18], [19]. load conditions (1.0 p.u., 0.5 p.u. and no load)
600
Case II- Sag due to a phase-to-phase fault
Vdc [V]
450
Sag CT propagated down from 22 kV bus is a DR
300 Discharge of
sag at 400 V. Under these conditions the three-
the capacitor phase rectifier is single phased during the voltage
150

t[s
sag event. Since the output power has to be
0
1.95 2 2.05 2.1 2.15
supplied via two phases, this mode increases the
rms input current of the ASD during the sag
Fig 19. Decrease of the DC bus voltage under sag event.
120.5

w[rad/s] IR, IS, IT [A] Imax =2.8 p.u. Imax =7.6 p.u.
100

120

t[s]
119.5
1.95 2 2.05 2.1 2.15
100 t [s]
Fig 20. Speed of the motor during the sag event under
the effect of the vector control system 1.98 2 2.02 2.04 2.06 2.08 2.1 2.12 2.14

If the supply voltage recovers before the dc bus Fig 24. Line currents during the sag event
voltage reaches the under-voltage protection 600

level, a high charging current is drawn from the Vdc [V]

supply. 450

0,93 p.u.
300
IR, IS, IT [A]
100

150

t [s]
0
0
1.95 2 2.05 2.1 2.15
Under the
discharge of the
capacitor the
100
currents are null t [s] Fig 25. Decrease of the DC bus voltage under sag
1.99 2 2.02 2.04 2.06 2.08 2.1 6. Commercial ASD during voltage sags
Fig 21. Line currents during the sag event Some experimental analysis were carried out to verify
the results obtained by using the transients program. In
this section, an experimental analysis is discussed for a
Case II- Sag due to a phase-to-ground fault
commercial available 400 V, 50 Hz, 12 kVA PWM
Sag DR propagated down from 22 kV bus is a CS adjustable speed drive. The analysis platform
sag at 400 V.The dc-link voltage under a voltage includes 11 kW induction motor, three-phase and
sag event is directly dependent on both the output programmable supply voltage to produce the
power and the effective inductance (including two voltage sags and electrical brake to permit motor
source inductance and dc-link inductance). operation under different load levels.
Test A
For this evaluation test the ASD was subjected to
DR voltage sag for a duration of 100 ms
( U R = 90∠0º , U S = 203∠ − 101º , UT = 203∠ − 100,46º ).
It is a no load test. During the sag the line
currents are null. The rectifier diodes cease to
conduct because are reverse biased due to
reduction in the phase voltages.

Fig. 29. Motor line voltages and supply line voltages

Origin of Test C
the sag

In this evaluation test the ASD was subjected to

DR voltage sag for a duration of 100 ms
( U R = 90∠0º , U S = 203∠ − 101º , UT = 203∠ − 100,46º ).
End of the
sag Three intervals have been seen. 1)At the
occurrence of the sag all the energy needed to
drive the load is taken from the electric energy
stored in the DC bus capacitor. The line currents
Fig. 26.Waveforms of the supply line currents during the sag are null. 2)When the dc bus voltage reaches the
new steady state, the rectifier is single phased and
the line currents are increased . 3)If the supply
voltage recovers before the dc bus voltage
reaches the under-voltage protection level, a high
charging current is drawn from the supply.

Origin of 3
the sag 2

Fig. 27.Motor line voltages and supply line voltages

1 End of the
sag

Test B

The ASD was subjected to DR voltage sag for a Fig. 30.Waveforms of the supply line currents during the sag
duration of 100 ms ( U R = 70∠0º , U S = 230∠ − 101º ,
U T = 230∠ − 100,46º ). During the sag, the rectifier
is single phased and the line currents are
increased. The dc-link voltage reduction under no
load and single phase behaviour is slight.

Origin of
the sag

Fig.31 . Motor line voltage

Test D
End of the
The ASD is set to trip when the dc voltage drops
sag 0,76 times the nominal value. This evaluation test
was conducted to show how the ASD goes to stop
as the dc voltage falls bellow the DC bus trip
Fig. 28.Waveforms of the supply line currents during the sag voltage. The response of the ASD can be seen in
figure 32.

Fig.32.Motor line voltages and supply line voltages
Acknowledgement
The authors gratefully acknowledge The Regional
plan of Research and Development FICYT for their
technical and financial support during this study.
Conclusions
In this paper, the effect of voltage sag on ASD´s has
been evaluated both experimentally and by
simulation. The power system components for
voltage sag studies have been modelled and
simulated by using a transients program. This paper
also gives a review of results from the effect of the
winding connections of the transformer, the kind
of fault, the resulting kind of sag and the load torque
on the dc voltage and input current fluctuations. The
model and the results can be used during the design
task of the mitigation devices and ride-through
alternatives.
References
[1] M. H. J. Bollen et al. “ Characterization of
three-phase unbalanced dips (as easy as one-two-
three ?)”. Proceedings of ninth international
conference on harmonics and quality of power.
Orlando- Oct.2000.
[2] N. Kagan et al. “ Influence of rms variation
measurement protocols on electrical system
performance indices for voltage sags and swells”.
Proceedings of ninth international conference on
harmonics and quality of power. Orlando-Florida-
Oct.2000.
[3] N. S. Tunaboylu et al. “Voltage disturbance
evaluation using the missing voltage technique”
Proceedings of 8th international conference on
harmonics and quality of power. Vol 1. Atenas.
Grecia 1998.
[4] L. Zhang et al. “Characteristic of voltage dips
(sags) in power systems”. IEEE Transactions on
Power Delivery, vol. 15, April 2000.
[5] M. H. J. Bollen et al. “Assessment of the
number of voltage sags experienced by a large
industrial customer”. IEEE Transactions on
Industry Applications, vol. 33, Nov./Dec. 1997.
[6] J. M. de Carvalho et al. “Electric power
system under voltage sag – a tool for achieving
compatibility”. Proceedings of ninth international
conference on harmonics and quality of power.
Orlando- Florida-Oct.2000.
[7] J. Lamoree et al. “Voltage sag analysis case
studies”. IEEE Transactions on Industry
Applications, vol. 30, July/August 1994.
[8] M. F. McGranaghan et al. “Voltage sags in
industrial systems”. IEEE Transactions on
Industry Applications, vol. 29, March/April 1993.
[9] G. Yalcinkaya et al.” Characterization of
voltage sags in industrial distribution systems”.
IEEE Transactions on Industry Applications, vol.
34, July/August 1998.
[10] M. H. J. Bollen. “Characterisation of voltage
sags experienced by three-phase adjustable speed
drives”. IEEE Transactions on Power Delivery,
vol. 12, October 1997.
[11] J. Durán G. et al. “Effect of voltage sags on
adjustable-speed drives: a critical evaluation and
an approach to improve performance “ IEEE
Transactions on Industry Applications, vol. 35,
Nov/Dec 1999.
[12] R. Langley et al. “Voltage Sag ride-through
testing of adjustable speed drives using a
controllable dynamic dynamometer” Proceedings
of 8th international conference on harmonics and
quality of power. Vol 1. Atenas. Grecia 1998
[13] A. Van Zyl et al. “Voltage sag ride-through
for adjustable-speed drives with active rectifiers”.
IEEE Transactions on Industry Applications, vol.
34, Nov/Dec 1998.
[14] M. H. J. Bollen. “The use of Electromagnetic
Transient Programs for voltage sag analysis”
Proceedings of 8th international conference on
harmonics and quality of power. Vol 1. Atenas.
Grecia1998
[15] M. H. J. Bollen. “The influence of motor
reacceleration on voltage sags”. IEEE
Transactions on Industry Applications, vol. 31,
July/August 1995.
[16] L. Guasch et al. “ Effects of unsymmetrical
voltage sag types E, F and G on induction
motors”. Proceedings of ninth international
conference on harmonics and quality of power.
Orlando- Florida-Oct.2000.
[17] A. von Jouanne et al. “Ride-through
alternatives for ASDs”. Part I, II, III. Web page
Darnell.com. Powerpulse.
[18] K. Stockman et al. “Embebed solutions to
protect textile processes against voltage sags”
[19]G. Newman et al. “Voltage sag ride through
improvement of modern ac drives: review of
methods and a case study”