ATP SIMULATION OF FARADAY CAGE FOR THE ANALYSIS OF LIGHTNING SURGES

Transcription

1 ATP SIMULATION OF FARADAY CAGE FOR THE ANALYSIS OF LIGHTNING SURGES Mehmet Salih Mamis Cemal Keles 1 Muslum Arkan 1 Ramazan Kaya 2 Inonu University, Turkey 1 Inonu University, Engineering Faculty, Electrical & Electronics Eng. Dept., Malatya, Turkey 2 EMTA Energy, Turkey mehmet.mamis@inonu.edu.tr, cemal.keles@inonu.edu.tr, muslum.arkan@inonu.edu.tr Abstract - Determination of current and voltages in lightning protective structures is important for the design of system. However, simulation models are limited. In this paper, a Faraday cage with 4 air terminals and 2x6 earthling rods constructed to protect a 380 kv switchyard control building in a substation is simulated in Alternative Transient Program- ATP for the analysis of lightning surges. Down conductors are represented as non-uniform lines. Current waveforms through the system and voltages at some critical points after a lightning stroke are computed. The traditional lightning parameters needed in structural protection such as lightning peak current, maximum current derivative, current rise time and current duration are determined. The effects of grounding system parameters are also investigated. Keywords: Faraday cage, ATP simulation, modelling, lightning surge, protective structures. 1 Introduction A Faraday cage is an enclosure covered with electrically conductive metal or formed by mesh of conductors that protects the internal volume from outside electric fields. With the cage the electric field is prevented to get into and get out from the enclosure. The outermost valence electrons of the atoms forming conductive materials can be easily separated from their atoms and have the ability to move. Therefore, when an electrically conductive body having a closed surface is placed in the electric field, these electrons will move and create a rearrangement until the electric field inside the conductor is cancelled. By cancelling the electric field, the reasons for rearrangement are eliminated. The Faraday cage works based on this principle and protects the objects at the inside against external electric field. Therefore, an ideal Faraday cage is formed with grounded hollow metal surface composed of a closed conductor like a sphere. However, instead of being continuous conductive surface, it can be constructed in the form of mesh of conductors. In this case, a little electric field may penetrate inside the cage, but this does not result any problem if the intervals are sufficiently small. However, the geometry of the conductive material need not be spherical; any closed surface can fulfil the task of the cage. The purpose of Faraday cage is to prevent the inside magnetic field to radiate outside and outside magnetic field to penetrate inside. Even though application areas change, this aim does not change to a great extent.

2 Some application areas of Faraday cage are described below: i) Buildings that contain explosive and flammable materials: The outer side of such buildings is covered with mesh of conductors. Lightning rods are mounted outside the building at high points to capture lightning surges. All conductors and lightning rods are connected and grounded. By this way the system is got equipotential. ii) Radio frequency emitting devices: The outer metal sheaths of cabins of this type of devices are grounded to prevent spreading parasitic radio signals to environment. iii) Radio communication buildings: To avoid radio communication signals to leak out and to be listening by outsider, the Faraday cage is constructed at out of the building. iv) Modules in the radio frequency circuit boards To prevent an electronic card and its working environment from emitted interference signals, radio-television tuner and radio frequency modules such as GSM transmitter and receiver circuits or electronic circuit sections are covered with closed metal sheet and are earthed. According to EMC (Electromagnetic Compatibility) directives, it is mandatory to take out such measures. Electrical devices are not allowed to spread parasites as radio signals or through transmission line conductors. As a result, Faraday cages, though expensive, are used in the units where devices are sensitive to electric field. When the cost of devices to be protected is considered, even this cost, the construction is advantageous. If sensitive devices do not present, instead of Faraday cage, simple lightning protection methods are often successfully implemented and are usually sufficient. In this paper, a Faraday cage constructed in a substation to protect a 380 kv switchyard control building is simulated in Alternative Transient Program-ATP for the analysis of lightning surges. Down conductors are represented as non-uniform transmission lines with the parameters varying with the conductor height. Current waveforms through the system and voltages at critical points are computed. The traditional lightning parameters needed in structural protection such as lightning peak current, maximum current derivative (di/dt), current rise time and current duration are determined. The effect of cable spacing and grounding system parameters are also investigated. 2 Description of Protective Structures Calculation of peak current in protective structures such as Faraday cage is important for destructive arcing between exposed or buried elements of the lightning protective system. The efficacy of lightning protection has been well demonstrated in practice [1-3]. However, simulation models for design purpose are limited. One of the aspects of lightning protection design is the diversion and shielding, mainly intended for structural protection but also functioning to reduce the lightning electric and magnetic fields within the structure [4,5]. Lightning protection systems offer a preferred

3 attachment point for lightning and then a safe path for the lightning current to ground to reduce the lightning electric and magnetic fields within the structure without damage to the protected structure. Such systems basically consist of three components illustrated in Fig. 1. In the figure, air terminals located at suitable locations on the structure are used to intercept the lightning surge, down conductors are used to carry the lightning current from the air terminals toward the ground, and grounding electrodes are used to pass the lightning current into the earth. A modern structural lighting protection system is shown in Fig. 2. Fig. 1. Lightning protection system for houses proposed (most likely by G. Ch. Lichtenberg) in Adapted from Wiesinger and Zischank [6]. Fig. 2. Modern structural lightning protection [7].

4 3 ATP model of Faraday cage In this study the Faraday cage which is used for protection of a 380 kv switchyard control building shown in Fig. 3 is considered. The cage consists of 4 air terminals and 2x6 earthling rods. 50 mm 2 stranded copper wire is used for the cage. Mesh size is 5x5 m. Grounding rods are 3 mm copper covered 2.5 m long with 20 mm diameter. Two grounding rods are used at each grounding point. The simulation file of ATPDraw of the system is shown in Fig. 4. Vertical lines are considered as non-uniform lines. Propagation speed for the lightning surges is selected as 3x10 8 m/s. Time step for the simulations is taken to be 1/100 of travel time of the shortest line stub. The current source used to model the lightning current is a standard 1/50 wave with 1.0 unit peak value which is represented as, where =1.5x10 4 s -1 and =6x10 6 s -1. The following formula for the single conductor over ground is used for the horizontal line segments: where h represents the conductor height and r is the radius. As the characteristic impedance of a vertical line segments varies with height, conductor parameters are non-uniform and therefore the exponential line model is used. Considering measurements and theoretical studies on non-uniform lines carried out in the literature [8-12], the non-uniform variation of the vertical conductors is expressed as: where x represents height from the ground level. The characteristic impedance varies from 150 at ground level to 250 at conductor top. Three uniform line sections are used for the simulation; which has characteristic impedance of 177.6, and 250. Simplified circuit model of the system constructed by line stubs is shown in Fig. 4 and ATPDraw [13] model is shown in Fig. 5. (1) (2) Fig kv switchyard control building.

5 5m 5m 5m 5m V 2 V 3 A V 1 V m Fig. 4. Circuit model of Faraday cage constructed for 380 kv switchyard control building. Fig. 5. ATP model of the Faraday cage.

6 4 Simulation results In this study the peak current is normalized and it is taken to be 1.0 A. For design purposes, the results should be transformed to real values by considering the lightning peak current. Several measurements have been done to identify the peak values of the lightning currents [1-3] and median peak currents for first and subsequent strokes were found to be and ka. The injected current is shown in Fig 6. The current waveform at down conductor at stroke point is shown in Fig 7.a and its derivative is illustrated in Fig 7.b. The results are obtained by selecting the grounding resistance 2. The lightning peak current at down conductor at stroke point is about 0.58 percent of peak value of lightning current wave. Maximum current derivative is about 4.5x10 6 A/s, and current rise time is 0.5 µs. By a simple calculation it can be obtained that a lightning wave with 30 ka peak value produces 135 kv across each meter of the wire in another system (for example, in a communications tower and in an adjacent electronics building) having a mutual inductance per unit length of M=10-6 H/m. 1,0 [A] 0,8 0,6 0,4 0,2 0, [us] 50 (f ile atp13.pl4; x-v ar t) c:xx0005-xx0001 Fig. 6. 1/50 lightning current waveform. Current at down conductor at stroke point for two values of grounding resistance (0.5 and 5 ) is shown in Fig 8. It is observed from the figure that the current in the down conductor decreases more rapidly than the lightning current and after 50 µs its value is pu and pu for 0.5 and for 5 grounding resistance respectively, while source current is 0.5 pu at this instant. Also, it is observed that the peak value of the current increases as grounding resistance decreases. As it can be seen from the figure, when ground resistance increases current wave decreases sharply from its maximum value. The voltage waveforms for two values of ground resistance are illustrated in Fig. 9. The peak value of the voltage at stroke point is pu for all values of grounding resistance. The value of grounding resistance does not affect the first peak of the voltage which is due to

7 arrival time of the reflected waves at grounding points. Maximum values of the voltages at other critical points indicated at Fig.1 are about 0.37 percent of the voltage at stroke point. 0,6 [A] 0,5 0,4 0,3 0,2 0,1 0, [us] 5 (f ile atp13.pl4; x-v ar t) c:xx0001-xx0003 (a) Time (s) (b) Fig. 7. (a) Current and (b) derivative of current at down conductor at stroke point ( =2 ).

8 0,7 [A] 0,6 0,5 0,4 0,3 0,2 0,1 0, [us] 50 (f ile atp13.pl4; x-v ar t) c:xx0001-xx0003 c:xx0023-xx0025 Fig. 8. Current at down conductor at stroke point for different values of grounding resistance. 20 [V] [us] 5 (f ile atp13.pl4; x-v ar t) v :XX0001 v :XX0023 Fig. 9. Voltage at stroke point for different values of the grounding resistance =0.5 =5 for lightning current shown in Fig. 6. and The peak values of the voltages at critical points for various grounding resistance values for 30 ka peak lightning current are given in Table 1. Maximum value is kv, which is measured at stroke point. It is observed that the effect of grounding resistance on the peak value of the surge voltage is small.

9 Table 1. Maximum of voltages at critical points for various grounding resistance values in (0-0.5 µs) time interval. V (kv) V 1 V 2 V 3 V Conclusion Using ATP, Faraday cage is simulated in this paper and voltage and currents at some critical points are computed. Transmission line models are used for the cage conductors and vertical conductors are simulated as non-uniform lines with characteristic impedance varying by height. Simulation results have shown that the effect of ground resistance on the peak value of the generated surge voltage at stroke point is small. 6 References [1] Berger, K.: Mesungen und Resultate der Blitzforschung auf dem Monte San Salvatore bei Lugano, der Jahre , Bull. SEV (1972), no: 63, pp [2] Visacro, S.; Soares, J. A.; Schroeder, M. A. O.; Cherchiglia, L. C. L.; and de Sousa V. J.: Statistical analysis of lightning current parameters: Measurements at Morro do Cachimbo Station, J. Geophys. Res. (2004), vol. 109, pp. D D [3] Takami, J.; Okabe, S.: Observational results of lightning current on transmission towers, IEEE Trans. Power Del. (2007), vol. 22, no. 1, pp [4] Rakov, V.A.; Uman, M.A.: Lightning: Physics and Effects, Cambridge University Press, 687 p., ISBN , PB ISBN , [5] Rakov, V.A.: Lightning discharge and fundamentals of lightning protection, Curitiba, Brazil, 9th-13th November, [6] Wiesinger, J.; Zischank, W.: Lightning Protection, in Handbook of Atmospheric Electrodynamics, Boca Raton: CRC Press (1995), vol. II, ed. H. Volland, pp [7] UL 96A, Standard for Installation Requirements for Lightning Protection Systems, Underwriters Laboratories, [8] Wagner, C.F.; Hileman, A.R.: A new approach to the calculation of the lightning performance of transmission lines I11 - a simplified method: stroke to a tower,.41ee

10 Trans. Power Syst PAS-79 (1960), pp [9] Chisholm, W.A.; Chow, Y.L.: Lightning surge response of transmission towers, IEEE Trans. Power Syst PAS-102 (1983), no:9, pp [10] Menemenlis, C.; Chun, Z.T: Wave propagation on nonuniform lines, IEEE Trans. Power Syst., PAS-101 (1982), no:4, pp [11] Nguyen, H.V; Dommel, H.W; Marti, J.R.: Modelling of single-phase nonuniform transmission lines in electromagnetic transient simulations, IEEE Trans. Power Deliv. (1997), vol:12, no2, pp [12] Mamiș, M. S.; Köksal, M.: Lightning surge analysis using nonuniform, single-phase line model, IEE Proceedings-Generation, Transmission and Distribution (2001), vol:148, no:1, pp [13] Alternative Transient Program (ATP),