Мodeling approaches for externally gapped line

arrester through ATP-EMTP model study

Danail Stanchev
Power Engineering Department
Technical University of Varna
Varna, Bulgaria
[email protected]
2020 12th Electrical Engineering Faculty Conference (BulEF) | 978-1-7281-9439-4/20/$31.00 ©2020 IEEE | DOI: 10.1109/BulEF51036.2020.9326048

Abstract— Line surge arrester (LSA) application is widely Despite it’s proven advantages, EGLA are less frequently
known and used for solution of the power system transient. They mounted. One of the advantages of EGLA is less weight and
are used for improving lightning performance of overhead lines sizes than the non-gapped arresters. Other very important
(lightning protection, bad tower grounding conditions), advantages of EGLA are lower residual voltage of SVU and
increasing insulation level, reducing clearances, compacting less aging due to less often energizing. Due to its importance
lines or saving overhead line ground wire. One of the types of and difficulties for testing procedures, modeling and case
line arresters is the externally gapped line arrester also known studies through programs for transient analysis are preferable.
as EGLA. Model studies for proving their effectiveness is In this paper model approaches for externally gapped line
preferred due to expensive live tests. Approaches for modeling
arrester through ATP-EMTP studies are shown. Case studies
externally gapped line arresters through ATP-EMTP software
is simulated in the paper. The models are tested with several
with direct lightning strike to a phase conductor and tower are
lightning waves and different approaches are compared each considered. The approached models and the obtained results
other. The presented models could be implemented into case from the study can be useful in case studies, student training
studies with EGLA applications. [5] and other.

Keywords— externally gapped line arresters (EGLA), ATP- II. MODEL DESCRIPTION
EMTP software, lightning overvoltages, modeling Air gap model
I. INTRODUCTION Modeling the air gap flashover with assumption that flash
happen when voltage reach some value is not recommended.
Nowadays the requirements for reliable and unbreakable That is why more accurate models are needed. Air gap
power supply from the consumers are growing. With this flashover can be modeled by volt-time characteristics
growth more reliable and efficiency performance of the power models[6], [7], integration method models[7] and leader
system is needed [1]. Lightning is one of the main problems development models [8]. These three models are used in the
for the power systems worldwide. It‘s known that lightning study. Flashover models based on volt-time curve and they use
causes large part of power system faults and outages. In order dielectric strength of the insulation and flashover happen
to reduce flashover rate, failure rate, faults and outages of the when the applied voltage exceed dielectric strength of the
overhead lines line surge arresters are installed. Their electrode system. In the model based on V-T characteristic
efficiency is widely known. Line surge arresters are two types, flashover voltage is given in function of length as:
Non-gapped line arresters (NGLA) [2] and Externally gapped
line arresters (EGLA)[3]. The both types of line arresters are 710
well known with their characteristics. Externally gapped line VFO = (400 + )D (1)
arresters have active part called series varistor unit (SVU) and tc0,75
an external gap which is connected series to the active part.
This is shown on fig. 1. Where: D [m] is the insulation length,
tc [μs] is the time to flashover.
This model is presented in ATP-EMTP software with the
model of ISF [9]. There are a lot of knowledge for different
insulations, electrode systems based on standard conditions,
but this information its not enough for prediction the
performance under non-standard conditions. Experimental
volt-time curves give information for every specific electrode
system and insulation, but not every time is possible to apply
experiment. In order to obtain accurate presentation of the
time to flashover under non-standard impulses integration
method models and leader development method models are
used. Integration method is based on the following equation:
t
DE =  (U (t ) − U 0 ) k dt (2)
t0

Where:
U(t) [kV] is the applied voltage,
Fig. 1. Standard configuration of EGLA[4]

978-1-7281-9439-4/20/$31.00 ©2020 IEEE

Authorized licensed use limited to: Escuela Superior de Ingeneria Mecanica. Downloaded on February 03,2022 at 00:50:08 UTC from IEEE Xplore. Restrictions apply.
 U0 [kV] is the minimum voltage before any breakdown Series varistor unit model
process start, The active part of the externally gapped line arrester is
t0[µs] is the time when the voltage exceeds the minimum called series varistor unit (SVU). According to [3] this unit is
voltage U0, non-linear metal-oxide resistor. This resistor is connected
series to the air gap. In some configurations may be included
t [µs] is the elapsed time after the lightning stroke,
several units. Typically, the residual voltage of the varistor is
k is constant, to 20-30% lower than the residual voltage of the non-gapped
line arresters. Parameters of the EGLA are shown on table Ⅰ
DE [kVkµs] is disruptive effect of the surge.
[4]. Recommended distance of the air gap is taken from [11]
Flashover happen when the integral become equal to DE. and it is equal to 114.5cm.
Disruptive effect of surge, k and U0 are constants for each air-
gap configuration and polarity. These constants are TABLE I. EGLA PARAMETERS
determined by experimental results for every configuration. Rated voltage 204 kV
Presentation of the integration method in ATP-EMTP is Max. residual voltage at
though model ISF[9]. According to [7] this method is 483 kV
10kA
recommended for air gaps smaller than 1.2 meters. Nominal discharge
10 kA
current
Leader development models are used for discharge Maximal discharge
40 kA
process for long air gaps. They are based on the physics of the current
discharge process. This process involves three phases: corona Rated short-circuit
65kA
inception, streamer propagation and leader propagation. Then current
time to flashover is given as: Lightning discharge
0.6C
capability
tc=ti+ts+tl (3) In this study the varistor is presented with nonlinear
resistor type 92 in the ATP-EMTP software. V-I
Where:
characteristic is shown on fig. 2.
ti [µs] is corona inception time,
ts [µs] is streamer propagation time,
tl [µs] is leader propagation time.
There are formulas for calculating the streamer and leader
propagation time in [7]. Leader velocity is function of the gap
length and applied voltage. Presentation of this dependence
derived from experimental work is presented in [9]. CIGRE
leader development model [10] is used in this study. Leader
velocity is express as function of gap length and applied
voltage:
dL U
= k3U ( − E0 ) (4)
dt ( D − L)
Where:
L [m] is leader length,
Fig. 2 V-I characteristic of SVU
k3 [m2kV-2s-1] is constant,
The complete representation of EGLA in ATP-EMTP is
U [kV] is the instantaneous voltage across the gap,
shown on fig. 3.
E0 [kV/m] is the critical field strength,
D [m] is the air gap length.
Also, the flashover conditions that the representation is
valid is for L≥D. Parameters k3 and E0 depend on the electrode
configuration (air gap configuration) and the polarity of the
applied voltage. Values for these parameters can be found in
[10]. Investigations show that the model is working for long
air gap configurations. The model is represented in ATP-
EMTP environment through ISF model [9].
The physics of the breakdown process require to add small
inductance to represent the arc inductance during the
discharge. Typical value for this inductance is 1µH/m[7]. This
value is taken in the study and is connected series to the ISF
model block.
Fig. 3. EGLA configuration in ATP-EMTP

Authorized licensed use limited to: Escuela Superior de Ingeneria Mecanica. Downloaded on February 03,2022 at 00:50:08 UTC from IEEE Xplore. Restrictions apply.
 III. CASE STUDIES Obtained results for the voltages at the line side of the
The models described in second part of the paper are EGLA are shown on figure 5. The colors for the approaches
implemented in ATP-EMTP case study. On this base two are the same as the previous figure.
cases are considered:
• Shielding failure at the externally gapped line
arrester installation place.
V-t
• Direct lightning strike on tower with mounted IM
EGLA.
LDM
Case studies are performed with waves 8/20µs and
8/77.5µs for different peaks and polarity. For the first
approach with volt-time characteristic only distance is needed
but for the integration method and leader development method
other constants are needed. The values of the constants and
voltage for the integration method for this distance of the air
gap and positive polarity are shown in table Ⅱ [7].

TABLE II. VALUES FOR U0 , K, DE FOR INTEGRATION METHOD

Fig. 5. Voltage on air gap (on the line side)
U0, kV k DE
Voltages on the nodes of the active part (SVU) is shown
577.5 1 641.2
on figure 6.
The values for the leader development method of CIGRE
[10] for positive and negative polarity are shown on table Ⅲ.

TABLE III. VALUES OF CONSTANT K3 AND E0

Configuration Polarity K3(m2kV-2s-1) E0(kV/m)

V-t
Positive 0.8 600
Air gap
Negative 1 670
IM
LDM
IV. RESULTS
Shielding failure
This case considered shielding failure at the place of
mounting of the line arrester. Lightning current is presented
with 8/20µs wave with 10 kA amplitude with positive polarity.
Obtained results for the voltages are observed on the air gap Fig. 6. Voltages on the active part of EGLA
and series varistor unit (SVU). Results for the voltages across
the air gap are shown on figure 4 and every approach is with Direct lighting strike on tower
different color (V-t (volt-time characteristic), IM (integration In this case direct lightning strike on tower is investigated.
method) and LDM (leader development method). As it can be Approaches for the air gap are the same as the previous case.
seen results for the IM and LDM modeling are similar. The Negative polarity of the wave is considered, the current
results for the case with volt-time characteristic are lower than amplitude is 200kA. Voltages for the approaches are the same
the integration method. as the previous case. Voltages across the air gap are shown
on figure 7.

V-t V-t
IM IM
LDM LDM

Fig. 4. Voltages across the air gap of EGLA

Fig. 7. Voltages across the air gap

Authorized licensed use limited to: Escuela Superior de Ingeneria Mecanica. Downloaded on February 03,2022 at 00:50:08 UTC from IEEE Xplore. Restrictions apply.
 Voltages on the series varistor unit are shown on figure 8. recommended for the externally gapped line arrester
configurations with air gaps up to 1,2m and leader
development method with nonlinear resistance is
recommended for externally gapped line arrester
configuration with air gap higher than 1,2m. Modeling with
volt-time characteristic is not recommended because this type
of testing is not widely known for different configurations
and polarities. Proposed modeling approaches can be useful
V-t in case studies for externally gapped line arrester
IM implementation and mounting. Also, models can be used in
LDM student education.
ACKNOWLEDGMENT
The model studies and the results presented in this paper
are supported by the National program "Young scientists and
Postdoctoral candidates", 2020 of Ministry of Education and
Science, Bulgaria. Author also thanks to Mr. Plamen Bunov
and Mr. Jonathan Woodworth for the help during the research
Fig. 8. Voltages on series varistor unit process.
Voltages across the EGLA nodes for this case is shown on REFERENCES
figure 9. [1] V. Kolev, I. Draganova - Zlateva, “The development of additional
services as a tool for balance of the electric power system and the
functions of slow tertiary regulation over the years,” Energetika, vol.7,
pp.19-23, September 2020.
V-t [2] IEC 60099-4, “SURGE ARRESTERS – Part 4: Metal-oxide surge
IM arresters without gaps for a.c. systems,” 2014.
LDM [3] IEC, “IEC 60099-8: Surge arresters – Part 8: Metal-oxide surge
arresters with external series gap (EGLA) for overhead transmission
and distribution lines of a.c. systems above 1 kV,” 2017.
[4] P. Bunov, L. Klingbeil, B. Goßler, and L. Montaz, “Design and Testing
of a 225-kV Externally Gapped LineArrester for installation under live
conditions,” in CIGREC4 International Colloquium on Lightning &
PowerSystems, 2014.
[5] M. Mehmed-Hamza, M. Vasileva, P. Stanchev, ‘’Increasing the
education quality by means of computer-aided visualization of the
processes in electric power systems’’, Advances in Intelligent Systems
and Computing, 2018, 680, pp. 386-395, 2018.
[6] I. S. Grant et al., “A simplified method for estimating lightning
performance of transmission lines: A Report Prepared by the Working
Group on Lightning Performance of Transmission Lines,” IEEE Trans.
Fig. 9. Voltages acros the EGLA nodes Power Appar. Syst., vol. PAS-104, no. 4, pp. 918–932, 1985.
[7] International Electrotechnical Commission, “IEC TR 60071-4,
V. CONCLUSION Insulation Co-ordination—Part 4: Computational Guide to Insulation
Co-ordination and Modeling of Electrical Networks,” 2004.
In this paper, investigation on different approaches for [8] H. Motoyama, “Experimental study and analysis of breakdown
externally gapped line arresters modeling through ATP- characteristics of long air gaps with short tail lightning impulse,” IEEE
EMTP model study have been made. Volt-time characteristic, Trans. Power Deliv., vol. 11, no. 2, pp. 972–979, 1996.
integration method and leader development method are used [9] Z. G. Datsios, P. N. Mikropoulos, and T. E. Tsovilis, “Insulator string
flashover modeling with the aid of an ATPDraw object,” Proc. Univ.
for modeling the air gap of EGLA. Obtained results are Power Eng. Conf., vol. 2011-Janua, no. January, 2011.
compared with values from the literature and values [10] CIGRE WG01 SC33, “Guide to procedures for estimating the lightning
corresponded to the real ones. From the observed model performance of transmission lines,” 1991.
studies and international standards can be concluded that [11] J. Woodworth, “Externally gapped line arresters a critical design
modeling with integration method for the air gap combined review,” in Proceedings of the IEEE Power Engineering Society
with nonlinear resistance (for the series varistors unit) is Transmission and Distribution Conference, 2014.

Authorized licensed use limited to: Escuela Superior de Ingeneria Mecanica. Downloaded on February 03,2022 at 00:50:08 UTC from IEEE Xplore. Restrictions apply.