- Email: [email protected]
An improved Current Voltage Transferring Device for high current high frequency measurement
G Model
ARTICLE IN PRESS
EPSR-5200; No. of Pages 7
Electric Power Systems Research xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Electric Power Systems Research journal homepage: www.elsevier.com/locate/epsr
An improved Current Voltage Transferring Device for high current high frequency measurement Norasage Pattanadech ∗ , Masaaki Kando Electrical Engineering Department, Faculty of Engineering, King Mongkut’s Institute of Technology Ladkrabang, Chalongkrung Road, Ladkrabang, Bangkok 10520, Thailand
a r t i c l e
i n f o
Article history: Received 31 July 2016 Received in revised form 26 May 2017 Accepted 20 June 2017 Available online xxx Keywords: Current Voltage Transferring Device Rogowski coil Current transformer Lightning current Lightning current generator Printed circuit board
a b s t r a c t This paper represents the characteristics of the improved Current Voltage Transferring Device (CVTD) for high current measurement including lightning impulse currents. The improved CVTD has been developed from the prototype CVTD to provide exceptional characteristics for high current with high frequency measurement. The improved CVTD in h-shaped configuration is made of copper and aluminium. The relationship between input current and output voltages of the new CVTD was investigated with short and long duration lightning impulse currents to establish the empirical formulas. The typical impulse current generator was used to generate 8/20 s lightning impulse current with the amplitude of 940 A while the impulse generator with crowbar elements was utilized to generate 10/350 s lightning impulse current with magnitude of 30.2 kA for the experiment. The lightning impulse current measurement characteristics of the CVTD and Rogowski coil were examined. From the experiments, the new improved CVTD shows the superior characteristics to measure the impulse currents with very fast rise time. Besides, nanosecond pulses superimposed on the lightning impulse currents were detected by only the improved CVTD. According to the experiments, there is no doubt that the new improved CVTD is a superior one for high current with high frequency measurement especially lightning impulse currents. © 2017 Elsevier B.V. All rights reserved.
1. Introduction There are many measuring techniques applied for high current measurements for examples an analog ammeter connecting in series in the test circuit, hall elements measuring electromagnetic field generated from the circuit under test, and a current transformer (CT) measuring the induced voltage generated from a conductor wire in the test circuit. These current measurement techniques are suitable for AC current with power frequencies. Besides, Rogowski coil measuring the induced voltage generated from the test circuit and a special designed low-ohmic resistor connecting in series in the test circuit are usually applied for impulse current measurement [1–4]. The measurement error may occur to some extent when the low-ohmic resistor is applied for impulse current measurement due to the inductive component of such resistor. To overcome this problem, the current measuring by a shunt resistor has been introduced. Shunts can be divided into two types: tubular and wirewound. Tubular coaxial shunts are applied for
∗ Corresponding author. E-mail addresses: [email protected], [email protected] (N. Pattanadech), [email protected] (M. Kando).
measurement of high impulse currents. The tubular coaxial shunt resistance is between 0.1 and some 100 m which is required to limit heat dissipation and loading effects on the test circuit. Due to free of a magnetic field within the current-carrying tube, the measurable current bandwidth of the tubular coaxial shunt is quite high. However, its upper frequency limit is determined by skin effect in the inner cylinder involving shunt dimension and material properties. Therefore, higher bandwidth or short risetime tubular coaxial shunts require very thin tubes made of material with high specific resistivity [5–9]. In case of wirewound shunt types, their resistance is in the range of some 50 m–10 . Details of wirewound shunts for measuring fast impulse currents have been well documented in Ref. [7]. In case of current measurement by CT, the induced voltage from external electromagnetic field may enhance the measurement error of this measuring system. Furthermore, the practical skill is needed to operate the CT to avoid the potential problems for the users or the CT itself. Some situations are avoided for example the opening of the secondary circuit of CT. Besides, CT’s dimension may quite large and also heavy which can make a big trouble with the installation in the test circuit for measuring a large current. As above mentioned, applying each current measurement technique needs to be considered its limitation especially frequency bandwidth which is very important for high current with
http://dx.doi.org/10.1016/j.epsr.2017.06.015 0378-7796/© 2017 Elsevier B.V. All rights reserved.
Please cite this article in press as: N. Pattanadech, M. Kando, An improved Current Voltage Transferring Device for high current high frequency measurement, Electr. Power Syst. Res. (2017), http://dx.doi.org/10.1016/j.epsr.2017.06.015
G Model EPSR-5200; No. of Pages 7 2
ARTICLE IN PRESS N. Pattanadech, M. Kando / Electric Power Systems Research xxx (2017) xxx–xxx
Fig. 1. Schematic diagram of the conventional current measuring method with the Rogowski coil.
Fig. 2. The improved CVTD for high current measurement.
high frequency measurements in high voltage field. Many efforts have been devoted for developing high current with high frequency measurement techniques for high voltage engineering applications and researches in order to deeply understand the breakdown mechanism of the lightning discharges and switching discharges, plasma phenomena and so on. However, there is mostly no progress on developing the high current and high frequency measuring devices. Therefore, the prototype CVTD has been developed continuously for these proposes by the authors [10,11]. Focusing on the conventional Rogowski coil circuit as shown in Fig. 1, it composes three parts as the Rogowski coil, a low pass filter, and an amplifier respectively. Each part has a frequency band affecting the high frequency current measurement. Currently, maximum frequency is limited as a few MHz for application in high voltage field. However, some electrical components or circuits such as PCB (Printed Circuit Board) utilized in such field can properly operate above the mentioned frequency.
trical and thermal conductivity while aluminium provides average electrical and thermal conductivity, low density and a stable oxide layer [12,13]. Besides, the interface layer of the aluminium to copper joint should provide sufficient resistance; consequently, the voltage across the junction layer should sufficiently large to be measured when the current flows through the CVTD. It is reported that the resistivity at the aluminium to copper interface might up to 7 times compared to copper’s resistivity [12]. The thickness of the CVTD is 8 mm. This CVTD can measure AC current up to 600 A for 1 minute which provides the maximum output voltage of 20 mA. Besides, it is expected to measure the impulse current with the maximum amplitude of 60 kA. The designed thickness of the CVTD is related to the maximum measurable current amplitude. Considering the CVTD configuration, the h shaped CVTD is the optimized dimension adapted from the comb shaped CVTD as reported in Ref. [10]. The improved CVTD dimension requires small size as possible for installation in the narrow space. The improved CVTD has three terminals by which two terminals, terminal (A) and terminal (b), are used as the input terminal and the output terminal connecting in series in a DC testing circuit, so the measuring current flows through the different materials of the CVTD. Besides, the terminal (b) is used as the common terminal for the voltage measurement. The output voltage is the voltage across the input terminal (A) with the common terminal (b), VAb , and the remaining terminal (a) with the common terminal (b), Vab , including the voltage across the input terminal (A) and the remaining terminal (a), VAa , which are measured and transferred to be the input current at the end. The output voltages obtained from VAb , VAa and Vab should be different due to the interface characteristics between copper and aluminium. For impulse current measurement, terminal (a) and terminal (A) are connected in series in the impulse current measurement test circuit. The voltages across each terminal are measured. 3. Test experiments and test results 3.1. Direct current measurement by the new improved CVTD
2. The improved Current Voltage Transferring Device An improved Current Voltage Transferring Device (CVTD) for high current with high frequency measurement was developed from the prototype as reported in Refs. [10,11]. The improved CVTD consists of two different metals (aluminium and copper) forming to be a small letter “h” with the specified dimension as depicted in Fig. 2. Aluminium to copper joints have been widely used for over 40 years in a variety of applications such as in heat exchangers, power plants, electrical transmission lines (i.e. busbars) and so on because of their unique properties. Copper has superior elec-
To characterize the new designed CVTD, the DC current, I, generated from a controllable DC power source (max. 8 V, 12 A) was controlled and measured by the DC ammeter (type: moving coil, class 0.5) while the voltages (VAb , Vab and VAa ) were measured as shown in Fig. 3. The relationship between the input current, I, and the output voltages were calculated with a commercial software. The experiment was performed in short time to avoid the raising temperature of the connecting wire. Furthermore, the connecting wires and the CVTD connections were tightened due to reducing contact resistance problem. The output voltages were measured
Please cite this article in press as: N. Pattanadech, M. Kando, An improved Current Voltage Transferring Device for high current high frequency measurement, Electr. Power Syst. Res. (2017), http://dx.doi.org/10.1016/j.epsr.2017.06.015
G Model EPSR-5200; No. of Pages 7
ARTICLE IN PRESS N. Pattanadech, M. Kando / Electric Power Systems Research xxx (2017) xxx–xxx
3
Fig. 3. Schematic diagram of DC test experimental circuits.
by a portable multi-meter (range mV, 4 digits). Moreover, alligator clips were used on the top of lead from the multi-meter. The clips were employed at each terminal (A, a, b) due to the elimination of increasing the contact resistance from lead wires extending from the DC generator. The empirical equations to demonstrate the relationships between the input direct currents and the output voltages measured by the new designed CVTD are represented in the following Eqs. (1) and (2) VAb = 0.264 I [mV]
(1)
VAa = 0.115 I [mV]
(2)
Fig. 5. View of the coaxial tubular shunt utilized in lightning current test (8/20 s).
where: I is in [A]. The obtained equations showed that DC resistances between terminals (RAb and RAa ) are lower than the order of milliohms as 0.264, 0.115 [m] respectively. 3.2. Short duration lightning current measurement by Rogowski coil and coaxial shunt
Fig. 6. The measured 8/20 s lightning current waveforms with the maximum amplitude of 700 A by the Rogowski coil (CH1) and the tubular shunt (CH2).
In this section, the comparative tests for lightning current measurement using a tubular shunt and the Rogowski coil were performed. The lightning current (8/20 s) was measured by the 1 m coaxial tubular shunt with thermal capacity of 2.5 kJ and the conventional Rogowski coil. The schematic diagram of the comparative testing circuit is shown in Fig. 4. In order to confirm the similarity of the two current waveforms measured by both measuring devices, the tubular shunt was connected in series in the testing circuit whereas the Rogowski coil measured indirectly the same currents. Fig. 5 depicts the coaxial tubular shunt used in the
experiment. The measured currents both waveforms and amplitudes from each measurement technique were compared as shown in Fig. 6. As can be seen, the current waveforms obtained from the conventional Rogowski coil and the coaxial tubular shunt were the same. Therefore, only the conventional Rogowski coil with the rise time of 200 ns was selected to measure the lightning currents compared with the new designed CVTD.
Fig. 4. Schematic diagram of the 8/20 s lightning impulse current measurement by the tubular shunt and the Rogowski coil.
Please cite this article in press as: N. Pattanadech, M. Kando, An improved Current Voltage Transferring Device for high current high frequency measurement, Electr. Power Syst. Res. (2017), http://dx.doi.org/10.1016/j.epsr.2017.06.015
G Model EPSR-5200; No. of Pages 7 4
ARTICLE IN PRESS N. Pattanadech, M. Kando / Electric Power Systems Research xxx (2017) xxx–xxx
Fig. 7. Schematic diagrams of the short and long duration lightning impulse current measurements by the Rogowski coil and the new designed CVTD. Where: L is the total inductance of the impulse current generator, R is the total resistance of the impulse current generator, Lc is the inductance of the crowbar connection, Rc is the resistance of the crowbar connection.
3.3. Short and long lightning current measurements by the new improved CVTD and the Rogowski coil To verify the performance of the new designed CVTD with the rise time of 5 ns for short and long duration impulse current measurements, the typical impulse current generator and the impulse current generator with the crowbar switch were utilized to generate the 8/20 s and 10/350 s lightning current waveforms respectively. The schematic diagrams for investigation of lightning impulse current measurement characteristics of the new designed CVTD and the Rogowski coil are shown in Fig. 7. The components of the impulse current generator were adjusted to generate the required impulse current waveforms for the experiments. Theoretically, the crowbar switch is trigged when the load voltage is zero because there is no current passing through the load. However, to synchronize the timing between the crest point of main current and crowbar operation point is a tough work. Therefore, the lightning current waveform may be distorted in some cases [14–16]. In the case of long duration impulse current generation, after the trigger gap (the spark gap shown in Fig. 7(b)) was activated for at least a few s or more, the crowbar switch (the crowbar gap shown in Fig. 7(b)) will start to operate. As shown in Fig. 7(b), the testing circuit has two triggering gaps to generate the long duration lightning current waveform. The electrical breakdown must occur at the spark gap and at the crowbar gap respectively in the tests to complete the lightning current waveform generation. These breakdown phenomena including nanosecond pulses must be detected
Fig. 8. The measured lightning current waveform (8/20 s) with the maximum amplitude of 940 A from the designed CVTD(CH2) and the Rogowski coil (CH1).
by the measuring equipment. However, when the lightning impulse current tests were carried out, the gap breakdown phenomena were not detected with the conventional CT or a Rogowski coil utilized normally in the industrial sectors. In order to examine the impulse current measurement capability of the improved CVTD and the conventional Rogowski coil, a lightning currents with short duration (8/20 s) generated by a conventional lightning current generator were measured. Fig. 8 represents the waveform of the lightning impulse current measured by the Rogowski coil compared with the improved CVTD.
Please cite this article in press as: N. Pattanadech, M. Kando, An improved Current Voltage Transferring Device for high current high frequency measurement, Electr. Power Syst. Res. (2017), http://dx.doi.org/10.1016/j.epsr.2017.06.015
G Model EPSR-5200; No. of Pages 7
ARTICLE IN PRESS N. Pattanadech, M. Kando / Electric Power Systems Research xxx (2017) xxx–xxx
5
Fig. 9. View of the two current measuring devices (the new designed CVTD and the Rogowski coil) for measuring the 8/20 s lightning impulse current.
In Fig. 9, two current measuring devices for the experiment are demonstrated. The lightning current waveforms were recorded by a digital oscilloscope with bandwidth of 100 MHz and sampling rate of 2 Gs/S. It can be seen that the lightning current waveform (blue line) measured by the Rogowski coil has the rise time about 20 s (CH1) which is conformity with the recommended waveform (8/20 s). While the rise time of the lightning current (light blue line) obtained from the new improved CVTD is of the order of nanosecond (CH2), which is different from the standard waveform. However, the wave tails of both lightning current waveforms are relatively similar. The differences of the rise time of the lightning impulse current waveforms may be because of the difference of the limitation of frequency band of each measuring technique. On the other words, the CVTD is only composed of two different metals (copper and aluminium) so that the CVTD may acts as pure resistance, therefore its frequency band is theoretically infinity. Generally, the lightning current generator is composed of main capacitor banks and resistors with small inductances, the ratio (di/dt) of current (i) and time (t) becomes small due to inductance in the circuit. As mentioned before, the lightning current waveform (8/20 s) obtained from the Rogowski coil was the same as that obtained from the tubular shunt resistor. The Rogowski coil was again used in the long duration lightning current tests. To investigate the performance of the improved CVTD for measuring the long duration lightning impulse current, the lightning impulse current with 10/350 s was generated by the impulse current generator with the crowbar elements and measured by the developed CVTD and the Rogowski coil. The long duration lightning currents with the maximum amplitude up to 30 kA were measured by the improved CVTD and the Rogowski coil. The installation of the improved CVTD and the Rogowski coil for this experiment was nearly the same as setting up for short duration lightning current measurement experiment as shown in Fig. 9. Besides, the CVTD and the Rogowski coil were put on a wooden made stand in order to avoid the potential problems such as flash over between the testing object and the grounding system. The digital oscilloscope as a measuring instrument was operated by a chargeable battery, floating from the ground, to avoid the damage of the digital oscilloscope in the case of over voltage occurring at the oscilloscope terminals during performing the test. The experiments were performed under circumstance of room temperature 18.9 ◦ C, atmospheric pressure 1018 hPa, and relative humidity 32%. The detected current waveforms by the new CVTD and the Rogowski coil are shown in Fig. 10. As can be seen in Fig. 10(a), the nanosecond
Fig. 10. Detected lightning impulse current waveforms by the new designed CVTD (CH1) and the Rogowski coil (CH2).
pulses are superimposed on the lightning impulse current measured by the improved CVTD after approximately 4 s from the triggering point (CH1). Moreover, the tail of waveform decreases with times slowly due to the discharge mechanism of a resistancecapacitance circuit, or a differential circuit. Considering the spike voltages on the lightning impulse currents which are difference of wave shape and original position as shown on (CH1) of Fig. 10(a) and (b), it may depend on the operation timing of the crowbar gap and current amplitude. In Fig. 10(b), the spikes occur before 4 s which may due to the breakdown of the crowbar gap driven by the higher current compared with Fig. 10(a), the spike discharges with nanosecond pulses were superimposed on the current waveforms measured by the improved CVTD. Whereas, the spike discharges were not detected using the conventional Rogowski coil. The similar phenomena of generating the spikes on the lightning current waveforms have been proved by the authors [17]. According to the former research work [17], the spike pulses, or the nanosecond pulses were observed when the standard impulse voltages were applied to the special electrode system. In that study, the special electrode system contained four identical sphere electrodes with diameter of 125 mm. The first and second electrodes were separated with the gap distance of 10 mm whereas the third and the fourth electrodes were separated with the gap distance of 30 mm. The first electrode was connected to the high voltage source; the second electrode was electrically connected with the third electrode while the fourth electrode was connected to the earth. In the experiment, the standard impulse voltage with 123 kV was applied to the electrode system. The oscillating pulse superimposed on the impulse current waveform was detected as shown in Fig. 11(a). As can be seen, the damping oscillation wave superimposes on the lightning current at about 10 s after the virtual origin of the applied voltage are detected. In order to examine the characteristics of the damping oscillation wave in detail, the waveform is zoomed in. The damping oscillation wave shows that the period is of 0.16 s and a half value of width of the damping oscillation wave is 80 ns. Moreover, the first peak is dropped which demonstrates that the breakdown may occur either at the first gap or the second gap of the electrode system. Considering retrospectively Fig. 10, the damping
Please cite this article in press as: N. Pattanadech, M. Kando, An improved Current Voltage Transferring Device for high current high frequency measurement, Electr. Power Syst. Res. (2017), http://dx.doi.org/10.1016/j.epsr.2017.06.015
G Model EPSR-5200; No. of Pages 7 6
ARTICLE IN PRESS N. Pattanadech, M. Kando / Electric Power Systems Research xxx (2017) xxx–xxx
For lightning impulse current waveform (8/20 s) experiment VAb = 5.22 I [V]
(3)
For lightning impulse current waveform (10/350 s) experiment VAa = 2.25 I [V]
(4)
where: I is in kA. According to the Eqs. (3) and (4), the impulse resistances between terminals (RAbi , RAai ) are as 5.22 and 2.25 [m] respectively. It can be seen that the impulse resistances between terminals are not corresponded to the DC resistances. The impulse resistance ratio (RAbi /RAai ) is approximately 2.26. On the other hands, the DC resistance ratio (RAb /RAa ) is approximately 2.32. That means that two ratios are quite similar values. However, the differences of the DC resistance and the impulse resistance maybe caused by the boundary property of the copper and aluminium and/or the types of testing waveforms. However, the difference of resistances is under consideration. In this study the improved CVTD is composed of different metals connected by the specialized welding technique. 4. Conclusions Fig. 11. Lightning impulse current with nanosecond pulse superimposed testing with the special electrode system [17].
oscillation waves superimposed on the lightning current waveform can be detected by using the new designed CVTD. However, such waveforms are not detected by the Rogowski coil. The reason could be the characteristics of frequency limitation band of the Rogowski coil. In general, a conventional Rogowski coil for electrical power application has not wide frequency band, only a few MHz. In order to measure the higher lightningimpulse currents with the current amplitude up to 30.2 kA, the lightning impulse characteristics of the designed CVTD were investigated. The output voltages were measured. Two lightning current waveforms, 8/20 s and 10/350 s, were generated and then measured by the new designed CVTD. The specified output voltage (VAb ) was measured in case of the lightning impulse current waveform (8/20 s) experiment and the voltage (VAa ) was measured in case of the lightning impulse current waveform (10/350 s) experiment. The relationships between lightning currents and output voltages across the terminals of the designed CVTD are shown in Fig. 12. The empirical equations to demonstrate the relationship between the input lightning impulse currents and the output voltages measured by the new designed CVTD are represented in Eqs. (3) and (4) and in Fig. 12 which are in concordance with the previous paper [11].
The new designed CVTD is developed from the prototype CVTD in order to measure two typical conventional lightning impulse current waveforms. The electrical characteristics of the new designed has been investigated both DC and impulse currents with the short duration lightning impulse current (8/20 s)and the long duration lightning impulse current (10/350 s) compared with that of the conventional Rogowski coil. The empirical equations described the relationships between the DC currents and DC output voltages including the relationships between the lightning impulse currents and the output voltages were established. The relationship of the input current and the output voltage of the improved CVTD was a linear function. The rise time of the new designed CVTD was in the order of nanosecond. The lightning current magnitude up to 30.2 kA with superimposed nanosecond pulses can be detected by using the new designed CVTD. It is clear that the new developed CVTD provides the higher frequency bandwidth to measure very short risetime pulse currents than the conventional Rogowski coil. Therefore, it will be very useful for applications not only in high voltage field especially for high current impulse measurements with very high frequency or with nanosecond pulses but also in transmission lines, distribution lines, and so on for any current measurements. However, the interface phenomena between copper and aluminium under difference of the power sources of DC and lightning impulse current are not clearly understand, which is under considerations.
Fig. 12. Relationships between lightning impulse currents, 8/20 s and 10/350 s, and output voltages, VAa and VAb , of the new designed CVTD.
Please cite this article in press as: N. Pattanadech, M. Kando, An improved Current Voltage Transferring Device for high current high frequency measurement, Electr. Power Syst. Res. (2017), http://dx.doi.org/10.1016/j.epsr.2017.06.015
G Model EPSR-5200; No. of Pages 7
ARTICLE IN PRESS N. Pattanadech, M. Kando / Electric Power Systems Research xxx (2017) xxx–xxx
Acknowledgments Authors express thanks to Dr. Sompob Polmai and Dr. Peerawut Yutthagowit for their encouragement and support. Authors give thanks also to OTWA ELECTRIC Co., Ltd., Japan for the permission of utilizing the lightning current generators in order to investigate the performances of the improved CVTD. References [1] High-voltage engineering and testing, in: Hugh M. Ryan (Ed.), Power and Energy Series, 66, third edition, Institution of Engineering and Technology, London, United Kingdom, 2013, pp. 571–572. [2] Wolfgang Hauschild, Eberhard Lemke, High-Voltage Test and Measuring Techniques, Springer-Verlag, Berlin, Heidelberg, 2014, pp. 356–359. [3] Klaus Schon, High Impulse Voltage and Current Measurement Techniques: Fundamentals, Measuring Instruments, Measuring Methods, Springer International Publishing, Switzerland, 2013, pp. 170–187. [4] Tatsuo Kawamura, Eiichi Haginomori, Yutaka Goda, Tetsuya Nakamoto, Recent developments on high current measurement using current shunt, in: IEEJ-EIT Joint Symposium on High Voltage Power Technology, Bangkok, Thailand, November 14–15, 2006, pp. 31–36. [5] B. Lago, R. Eatock, Coaxial shunt, Proc. IEE 114 (September (9)) (1967) 1317–1324. [6] Alf J. Schwab, Low-resistance shunts for impulse currents, IEEE Trans. Power Appar. Syst. PAS-90 (5) (1971) 2251–2257. [7] Ryszard Malewski, Wirewound shunts for measurement of fast current impulses, IEEE Trans. Power Appar. Syst. PAS-103 (October (10)) (1984) 2927–2933.
7
[8] W. Pfeiffer, Fast measurement techniques for research in dielectrics, IEEE Trans. Electr. Insulation EI-21 (October (5)) (1986) 763–780. [9] Farouk A.M. Rizk, Giao N. Trinh, High Voltage Engineering, CRC Press, 2014, pp. 740–743. [10] M. Kando, N. Pattanadech, New current-voltage transferring device with different metals, IEEJ Trans. Electr. Electron. Eng. 10 (September (5)) (2015) 603–604 (Willy & Sons, New York). [11] M. Kando, N. Pattanadech, New electrical shunt with two rectangular holes in a copper plate, IEEJ Trans. Electr. Electron. Eng. 10 (November (6)) (2015) 723–725 (Willy & Sons, New York). [12] M. Braunovic, N. Alexandrov, Intermetallic compounds at aluminum-to-copper electrical interfaces: effect of temperature and electric current, IEEE Trans. Compon. Packag. Manuf. Technol. A 17 (March (1)) (1994) 78–85. [13] M.P. Mubiayi, E.T. Akinlabi, M.E. Makhatha, Effect of process parameters on tensile strength and morphology of friction stir spot welds of aluminum and copper, in: 2017 8th International Conference on Mechanical and Intelligent Manufacturing Technologies (ICMIMT), Cape Town, South Africa, February 3–6, 2017, pp. 48–53. [14] F. Heidler, W. Zischank, Z. Flisowski, Ch. Bouquegneau, C. Mazzetti, Lightning current given in IEC 62305—background, experience and outlook, in: 29th International Conference on Lightning Protection, Uppsala, Sweden, 23rd–26th June, 2008, pp. 1–22. [15] Christof Drilling, Markus Droldner, Ernst Günther Jordan, Jan Meppelink, A New Generator for Testing of SPD’s Using Multiple Lightning Current Impulses for Combined Tests with Follow Currents, pp. 1–8. http://obo.de/ downloads/GENERATOR.PDF. [16] Ju-Hong Eom, Sung-Chul Cho, Tae-Hyung Lee, Parameters optimization of impulse generator circuit for generating first short stroke lightning current waveform, J. Electr. Eng. Technol. 8 (2013) 742–748. [17] M. Kando, T. Inaba, AC phase controlled testing apparatus superimposed with double pulses, IEEJ Trans. PE 129 (3) (2009) 447–452 (in Japanese).
Please cite this article in press as: N. Pattanadech, M. Kando, An improved Current Voltage Transferring Device for high current high frequency measurement, Electr. Power Syst. Res. (2017), http://dx.doi.org/10.1016/j.epsr.2017.06.015
ARTICLE IN PRESS
EPSR-5200; No. of Pages 7
Electric Power Systems Research xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Electric Power Systems Research journal homepage: www.elsevier.com/locate/epsr
An improved Current Voltage Transferring Device for high current high frequency measurement Norasage Pattanadech ∗ , Masaaki Kando Electrical Engineering Department, Faculty of Engineering, King Mongkut’s Institute of Technology Ladkrabang, Chalongkrung Road, Ladkrabang, Bangkok 10520, Thailand
a r t i c l e
i n f o
Article history: Received 31 July 2016 Received in revised form 26 May 2017 Accepted 20 June 2017 Available online xxx Keywords: Current Voltage Transferring Device Rogowski coil Current transformer Lightning current Lightning current generator Printed circuit board
a b s t r a c t This paper represents the characteristics of the improved Current Voltage Transferring Device (CVTD) for high current measurement including lightning impulse currents. The improved CVTD has been developed from the prototype CVTD to provide exceptional characteristics for high current with high frequency measurement. The improved CVTD in h-shaped configuration is made of copper and aluminium. The relationship between input current and output voltages of the new CVTD was investigated with short and long duration lightning impulse currents to establish the empirical formulas. The typical impulse current generator was used to generate 8/20 s lightning impulse current with the amplitude of 940 A while the impulse generator with crowbar elements was utilized to generate 10/350 s lightning impulse current with magnitude of 30.2 kA for the experiment. The lightning impulse current measurement characteristics of the CVTD and Rogowski coil were examined. From the experiments, the new improved CVTD shows the superior characteristics to measure the impulse currents with very fast rise time. Besides, nanosecond pulses superimposed on the lightning impulse currents were detected by only the improved CVTD. According to the experiments, there is no doubt that the new improved CVTD is a superior one for high current with high frequency measurement especially lightning impulse currents. © 2017 Elsevier B.V. All rights reserved.
1. Introduction There are many measuring techniques applied for high current measurements for examples an analog ammeter connecting in series in the test circuit, hall elements measuring electromagnetic field generated from the circuit under test, and a current transformer (CT) measuring the induced voltage generated from a conductor wire in the test circuit. These current measurement techniques are suitable for AC current with power frequencies. Besides, Rogowski coil measuring the induced voltage generated from the test circuit and a special designed low-ohmic resistor connecting in series in the test circuit are usually applied for impulse current measurement [1–4]. The measurement error may occur to some extent when the low-ohmic resistor is applied for impulse current measurement due to the inductive component of such resistor. To overcome this problem, the current measuring by a shunt resistor has been introduced. Shunts can be divided into two types: tubular and wirewound. Tubular coaxial shunts are applied for
∗ Corresponding author. E-mail addresses: [email protected], [email protected] (N. Pattanadech), [email protected] (M. Kando).
measurement of high impulse currents. The tubular coaxial shunt resistance is between 0.1 and some 100 m which is required to limit heat dissipation and loading effects on the test circuit. Due to free of a magnetic field within the current-carrying tube, the measurable current bandwidth of the tubular coaxial shunt is quite high. However, its upper frequency limit is determined by skin effect in the inner cylinder involving shunt dimension and material properties. Therefore, higher bandwidth or short risetime tubular coaxial shunts require very thin tubes made of material with high specific resistivity [5–9]. In case of wirewound shunt types, their resistance is in the range of some 50 m–10 . Details of wirewound shunts for measuring fast impulse currents have been well documented in Ref. [7]. In case of current measurement by CT, the induced voltage from external electromagnetic field may enhance the measurement error of this measuring system. Furthermore, the practical skill is needed to operate the CT to avoid the potential problems for the users or the CT itself. Some situations are avoided for example the opening of the secondary circuit of CT. Besides, CT’s dimension may quite large and also heavy which can make a big trouble with the installation in the test circuit for measuring a large current. As above mentioned, applying each current measurement technique needs to be considered its limitation especially frequency bandwidth which is very important for high current with
http://dx.doi.org/10.1016/j.epsr.2017.06.015 0378-7796/© 2017 Elsevier B.V. All rights reserved.
Please cite this article in press as: N. Pattanadech, M. Kando, An improved Current Voltage Transferring Device for high current high frequency measurement, Electr. Power Syst. Res. (2017), http://dx.doi.org/10.1016/j.epsr.2017.06.015
G Model EPSR-5200; No. of Pages 7 2
ARTICLE IN PRESS N. Pattanadech, M. Kando / Electric Power Systems Research xxx (2017) xxx–xxx
Fig. 1. Schematic diagram of the conventional current measuring method with the Rogowski coil.
Fig. 2. The improved CVTD for high current measurement.
high frequency measurements in high voltage field. Many efforts have been devoted for developing high current with high frequency measurement techniques for high voltage engineering applications and researches in order to deeply understand the breakdown mechanism of the lightning discharges and switching discharges, plasma phenomena and so on. However, there is mostly no progress on developing the high current and high frequency measuring devices. Therefore, the prototype CVTD has been developed continuously for these proposes by the authors [10,11]. Focusing on the conventional Rogowski coil circuit as shown in Fig. 1, it composes three parts as the Rogowski coil, a low pass filter, and an amplifier respectively. Each part has a frequency band affecting the high frequency current measurement. Currently, maximum frequency is limited as a few MHz for application in high voltage field. However, some electrical components or circuits such as PCB (Printed Circuit Board) utilized in such field can properly operate above the mentioned frequency.
trical and thermal conductivity while aluminium provides average electrical and thermal conductivity, low density and a stable oxide layer [12,13]. Besides, the interface layer of the aluminium to copper joint should provide sufficient resistance; consequently, the voltage across the junction layer should sufficiently large to be measured when the current flows through the CVTD. It is reported that the resistivity at the aluminium to copper interface might up to 7 times compared to copper’s resistivity [12]. The thickness of the CVTD is 8 mm. This CVTD can measure AC current up to 600 A for 1 minute which provides the maximum output voltage of 20 mA. Besides, it is expected to measure the impulse current with the maximum amplitude of 60 kA. The designed thickness of the CVTD is related to the maximum measurable current amplitude. Considering the CVTD configuration, the h shaped CVTD is the optimized dimension adapted from the comb shaped CVTD as reported in Ref. [10]. The improved CVTD dimension requires small size as possible for installation in the narrow space. The improved CVTD has three terminals by which two terminals, terminal (A) and terminal (b), are used as the input terminal and the output terminal connecting in series in a DC testing circuit, so the measuring current flows through the different materials of the CVTD. Besides, the terminal (b) is used as the common terminal for the voltage measurement. The output voltage is the voltage across the input terminal (A) with the common terminal (b), VAb , and the remaining terminal (a) with the common terminal (b), Vab , including the voltage across the input terminal (A) and the remaining terminal (a), VAa , which are measured and transferred to be the input current at the end. The output voltages obtained from VAb , VAa and Vab should be different due to the interface characteristics between copper and aluminium. For impulse current measurement, terminal (a) and terminal (A) are connected in series in the impulse current measurement test circuit. The voltages across each terminal are measured. 3. Test experiments and test results 3.1. Direct current measurement by the new improved CVTD
2. The improved Current Voltage Transferring Device An improved Current Voltage Transferring Device (CVTD) for high current with high frequency measurement was developed from the prototype as reported in Refs. [10,11]. The improved CVTD consists of two different metals (aluminium and copper) forming to be a small letter “h” with the specified dimension as depicted in Fig. 2. Aluminium to copper joints have been widely used for over 40 years in a variety of applications such as in heat exchangers, power plants, electrical transmission lines (i.e. busbars) and so on because of their unique properties. Copper has superior elec-
To characterize the new designed CVTD, the DC current, I, generated from a controllable DC power source (max. 8 V, 12 A) was controlled and measured by the DC ammeter (type: moving coil, class 0.5) while the voltages (VAb , Vab and VAa ) were measured as shown in Fig. 3. The relationship between the input current, I, and the output voltages were calculated with a commercial software. The experiment was performed in short time to avoid the raising temperature of the connecting wire. Furthermore, the connecting wires and the CVTD connections were tightened due to reducing contact resistance problem. The output voltages were measured
Please cite this article in press as: N. Pattanadech, M. Kando, An improved Current Voltage Transferring Device for high current high frequency measurement, Electr. Power Syst. Res. (2017), http://dx.doi.org/10.1016/j.epsr.2017.06.015
G Model EPSR-5200; No. of Pages 7
ARTICLE IN PRESS N. Pattanadech, M. Kando / Electric Power Systems Research xxx (2017) xxx–xxx
3
Fig. 3. Schematic diagram of DC test experimental circuits.
by a portable multi-meter (range mV, 4 digits). Moreover, alligator clips were used on the top of lead from the multi-meter. The clips were employed at each terminal (A, a, b) due to the elimination of increasing the contact resistance from lead wires extending from the DC generator. The empirical equations to demonstrate the relationships between the input direct currents and the output voltages measured by the new designed CVTD are represented in the following Eqs. (1) and (2) VAb = 0.264 I [mV]
(1)
VAa = 0.115 I [mV]
(2)
Fig. 5. View of the coaxial tubular shunt utilized in lightning current test (8/20 s).
where: I is in [A]. The obtained equations showed that DC resistances between terminals (RAb and RAa ) are lower than the order of milliohms as 0.264, 0.115 [m] respectively. 3.2. Short duration lightning current measurement by Rogowski coil and coaxial shunt
Fig. 6. The measured 8/20 s lightning current waveforms with the maximum amplitude of 700 A by the Rogowski coil (CH1) and the tubular shunt (CH2).
In this section, the comparative tests for lightning current measurement using a tubular shunt and the Rogowski coil were performed. The lightning current (8/20 s) was measured by the 1 m coaxial tubular shunt with thermal capacity of 2.5 kJ and the conventional Rogowski coil. The schematic diagram of the comparative testing circuit is shown in Fig. 4. In order to confirm the similarity of the two current waveforms measured by both measuring devices, the tubular shunt was connected in series in the testing circuit whereas the Rogowski coil measured indirectly the same currents. Fig. 5 depicts the coaxial tubular shunt used in the
experiment. The measured currents both waveforms and amplitudes from each measurement technique were compared as shown in Fig. 6. As can be seen, the current waveforms obtained from the conventional Rogowski coil and the coaxial tubular shunt were the same. Therefore, only the conventional Rogowski coil with the rise time of 200 ns was selected to measure the lightning currents compared with the new designed CVTD.
Fig. 4. Schematic diagram of the 8/20 s lightning impulse current measurement by the tubular shunt and the Rogowski coil.
Please cite this article in press as: N. Pattanadech, M. Kando, An improved Current Voltage Transferring Device for high current high frequency measurement, Electr. Power Syst. Res. (2017), http://dx.doi.org/10.1016/j.epsr.2017.06.015
G Model EPSR-5200; No. of Pages 7 4
ARTICLE IN PRESS N. Pattanadech, M. Kando / Electric Power Systems Research xxx (2017) xxx–xxx
Fig. 7. Schematic diagrams of the short and long duration lightning impulse current measurements by the Rogowski coil and the new designed CVTD. Where: L is the total inductance of the impulse current generator, R is the total resistance of the impulse current generator, Lc is the inductance of the crowbar connection, Rc is the resistance of the crowbar connection.
3.3. Short and long lightning current measurements by the new improved CVTD and the Rogowski coil To verify the performance of the new designed CVTD with the rise time of 5 ns for short and long duration impulse current measurements, the typical impulse current generator and the impulse current generator with the crowbar switch were utilized to generate the 8/20 s and 10/350 s lightning current waveforms respectively. The schematic diagrams for investigation of lightning impulse current measurement characteristics of the new designed CVTD and the Rogowski coil are shown in Fig. 7. The components of the impulse current generator were adjusted to generate the required impulse current waveforms for the experiments. Theoretically, the crowbar switch is trigged when the load voltage is zero because there is no current passing through the load. However, to synchronize the timing between the crest point of main current and crowbar operation point is a tough work. Therefore, the lightning current waveform may be distorted in some cases [14–16]. In the case of long duration impulse current generation, after the trigger gap (the spark gap shown in Fig. 7(b)) was activated for at least a few s or more, the crowbar switch (the crowbar gap shown in Fig. 7(b)) will start to operate. As shown in Fig. 7(b), the testing circuit has two triggering gaps to generate the long duration lightning current waveform. The electrical breakdown must occur at the spark gap and at the crowbar gap respectively in the tests to complete the lightning current waveform generation. These breakdown phenomena including nanosecond pulses must be detected
Fig. 8. The measured lightning current waveform (8/20 s) with the maximum amplitude of 940 A from the designed CVTD(CH2) and the Rogowski coil (CH1).
by the measuring equipment. However, when the lightning impulse current tests were carried out, the gap breakdown phenomena were not detected with the conventional CT or a Rogowski coil utilized normally in the industrial sectors. In order to examine the impulse current measurement capability of the improved CVTD and the conventional Rogowski coil, a lightning currents with short duration (8/20 s) generated by a conventional lightning current generator were measured. Fig. 8 represents the waveform of the lightning impulse current measured by the Rogowski coil compared with the improved CVTD.
Please cite this article in press as: N. Pattanadech, M. Kando, An improved Current Voltage Transferring Device for high current high frequency measurement, Electr. Power Syst. Res. (2017), http://dx.doi.org/10.1016/j.epsr.2017.06.015
G Model EPSR-5200; No. of Pages 7
ARTICLE IN PRESS N. Pattanadech, M. Kando / Electric Power Systems Research xxx (2017) xxx–xxx
5
Fig. 9. View of the two current measuring devices (the new designed CVTD and the Rogowski coil) for measuring the 8/20 s lightning impulse current.
In Fig. 9, two current measuring devices for the experiment are demonstrated. The lightning current waveforms were recorded by a digital oscilloscope with bandwidth of 100 MHz and sampling rate of 2 Gs/S. It can be seen that the lightning current waveform (blue line) measured by the Rogowski coil has the rise time about 20 s (CH1) which is conformity with the recommended waveform (8/20 s). While the rise time of the lightning current (light blue line) obtained from the new improved CVTD is of the order of nanosecond (CH2), which is different from the standard waveform. However, the wave tails of both lightning current waveforms are relatively similar. The differences of the rise time of the lightning impulse current waveforms may be because of the difference of the limitation of frequency band of each measuring technique. On the other words, the CVTD is only composed of two different metals (copper and aluminium) so that the CVTD may acts as pure resistance, therefore its frequency band is theoretically infinity. Generally, the lightning current generator is composed of main capacitor banks and resistors with small inductances, the ratio (di/dt) of current (i) and time (t) becomes small due to inductance in the circuit. As mentioned before, the lightning current waveform (8/20 s) obtained from the Rogowski coil was the same as that obtained from the tubular shunt resistor. The Rogowski coil was again used in the long duration lightning current tests. To investigate the performance of the improved CVTD for measuring the long duration lightning impulse current, the lightning impulse current with 10/350 s was generated by the impulse current generator with the crowbar elements and measured by the developed CVTD and the Rogowski coil. The long duration lightning currents with the maximum amplitude up to 30 kA were measured by the improved CVTD and the Rogowski coil. The installation of the improved CVTD and the Rogowski coil for this experiment was nearly the same as setting up for short duration lightning current measurement experiment as shown in Fig. 9. Besides, the CVTD and the Rogowski coil were put on a wooden made stand in order to avoid the potential problems such as flash over between the testing object and the grounding system. The digital oscilloscope as a measuring instrument was operated by a chargeable battery, floating from the ground, to avoid the damage of the digital oscilloscope in the case of over voltage occurring at the oscilloscope terminals during performing the test. The experiments were performed under circumstance of room temperature 18.9 ◦ C, atmospheric pressure 1018 hPa, and relative humidity 32%. The detected current waveforms by the new CVTD and the Rogowski coil are shown in Fig. 10. As can be seen in Fig. 10(a), the nanosecond
Fig. 10. Detected lightning impulse current waveforms by the new designed CVTD (CH1) and the Rogowski coil (CH2).
pulses are superimposed on the lightning impulse current measured by the improved CVTD after approximately 4 s from the triggering point (CH1). Moreover, the tail of waveform decreases with times slowly due to the discharge mechanism of a resistancecapacitance circuit, or a differential circuit. Considering the spike voltages on the lightning impulse currents which are difference of wave shape and original position as shown on (CH1) of Fig. 10(a) and (b), it may depend on the operation timing of the crowbar gap and current amplitude. In Fig. 10(b), the spikes occur before 4 s which may due to the breakdown of the crowbar gap driven by the higher current compared with Fig. 10(a), the spike discharges with nanosecond pulses were superimposed on the current waveforms measured by the improved CVTD. Whereas, the spike discharges were not detected using the conventional Rogowski coil. The similar phenomena of generating the spikes on the lightning current waveforms have been proved by the authors [17]. According to the former research work [17], the spike pulses, or the nanosecond pulses were observed when the standard impulse voltages were applied to the special electrode system. In that study, the special electrode system contained four identical sphere electrodes with diameter of 125 mm. The first and second electrodes were separated with the gap distance of 10 mm whereas the third and the fourth electrodes were separated with the gap distance of 30 mm. The first electrode was connected to the high voltage source; the second electrode was electrically connected with the third electrode while the fourth electrode was connected to the earth. In the experiment, the standard impulse voltage with 123 kV was applied to the electrode system. The oscillating pulse superimposed on the impulse current waveform was detected as shown in Fig. 11(a). As can be seen, the damping oscillation wave superimposes on the lightning current at about 10 s after the virtual origin of the applied voltage are detected. In order to examine the characteristics of the damping oscillation wave in detail, the waveform is zoomed in. The damping oscillation wave shows that the period is of 0.16 s and a half value of width of the damping oscillation wave is 80 ns. Moreover, the first peak is dropped which demonstrates that the breakdown may occur either at the first gap or the second gap of the electrode system. Considering retrospectively Fig. 10, the damping
Please cite this article in press as: N. Pattanadech, M. Kando, An improved Current Voltage Transferring Device for high current high frequency measurement, Electr. Power Syst. Res. (2017), http://dx.doi.org/10.1016/j.epsr.2017.06.015
G Model EPSR-5200; No. of Pages 7 6
ARTICLE IN PRESS N. Pattanadech, M. Kando / Electric Power Systems Research xxx (2017) xxx–xxx
For lightning impulse current waveform (8/20 s) experiment VAb = 5.22 I [V]
(3)
For lightning impulse current waveform (10/350 s) experiment VAa = 2.25 I [V]
(4)
where: I is in kA. According to the Eqs. (3) and (4), the impulse resistances between terminals (RAbi , RAai ) are as 5.22 and 2.25 [m] respectively. It can be seen that the impulse resistances between terminals are not corresponded to the DC resistances. The impulse resistance ratio (RAbi /RAai ) is approximately 2.26. On the other hands, the DC resistance ratio (RAb /RAa ) is approximately 2.32. That means that two ratios are quite similar values. However, the differences of the DC resistance and the impulse resistance maybe caused by the boundary property of the copper and aluminium and/or the types of testing waveforms. However, the difference of resistances is under consideration. In this study the improved CVTD is composed of different metals connected by the specialized welding technique. 4. Conclusions Fig. 11. Lightning impulse current with nanosecond pulse superimposed testing with the special electrode system [17].
oscillation waves superimposed on the lightning current waveform can be detected by using the new designed CVTD. However, such waveforms are not detected by the Rogowski coil. The reason could be the characteristics of frequency limitation band of the Rogowski coil. In general, a conventional Rogowski coil for electrical power application has not wide frequency band, only a few MHz. In order to measure the higher lightningimpulse currents with the current amplitude up to 30.2 kA, the lightning impulse characteristics of the designed CVTD were investigated. The output voltages were measured. Two lightning current waveforms, 8/20 s and 10/350 s, were generated and then measured by the new designed CVTD. The specified output voltage (VAb ) was measured in case of the lightning impulse current waveform (8/20 s) experiment and the voltage (VAa ) was measured in case of the lightning impulse current waveform (10/350 s) experiment. The relationships between lightning currents and output voltages across the terminals of the designed CVTD are shown in Fig. 12. The empirical equations to demonstrate the relationship between the input lightning impulse currents and the output voltages measured by the new designed CVTD are represented in Eqs. (3) and (4) and in Fig. 12 which are in concordance with the previous paper [11].
The new designed CVTD is developed from the prototype CVTD in order to measure two typical conventional lightning impulse current waveforms. The electrical characteristics of the new designed has been investigated both DC and impulse currents with the short duration lightning impulse current (8/20 s)and the long duration lightning impulse current (10/350 s) compared with that of the conventional Rogowski coil. The empirical equations described the relationships between the DC currents and DC output voltages including the relationships between the lightning impulse currents and the output voltages were established. The relationship of the input current and the output voltage of the improved CVTD was a linear function. The rise time of the new designed CVTD was in the order of nanosecond. The lightning current magnitude up to 30.2 kA with superimposed nanosecond pulses can be detected by using the new designed CVTD. It is clear that the new developed CVTD provides the higher frequency bandwidth to measure very short risetime pulse currents than the conventional Rogowski coil. Therefore, it will be very useful for applications not only in high voltage field especially for high current impulse measurements with very high frequency or with nanosecond pulses but also in transmission lines, distribution lines, and so on for any current measurements. However, the interface phenomena between copper and aluminium under difference of the power sources of DC and lightning impulse current are not clearly understand, which is under considerations.
Fig. 12. Relationships between lightning impulse currents, 8/20 s and 10/350 s, and output voltages, VAa and VAb , of the new designed CVTD.
Please cite this article in press as: N. Pattanadech, M. Kando, An improved Current Voltage Transferring Device for high current high frequency measurement, Electr. Power Syst. Res. (2017), http://dx.doi.org/10.1016/j.epsr.2017.06.015
G Model EPSR-5200; No. of Pages 7
ARTICLE IN PRESS N. Pattanadech, M. Kando / Electric Power Systems Research xxx (2017) xxx–xxx
Acknowledgments Authors express thanks to Dr. Sompob Polmai and Dr. Peerawut Yutthagowit for their encouragement and support. Authors give thanks also to OTWA ELECTRIC Co., Ltd., Japan for the permission of utilizing the lightning current generators in order to investigate the performances of the improved CVTD. References [1] High-voltage engineering and testing, in: Hugh M. Ryan (Ed.), Power and Energy Series, 66, third edition, Institution of Engineering and Technology, London, United Kingdom, 2013, pp. 571–572. [2] Wolfgang Hauschild, Eberhard Lemke, High-Voltage Test and Measuring Techniques, Springer-Verlag, Berlin, Heidelberg, 2014, pp. 356–359. [3] Klaus Schon, High Impulse Voltage and Current Measurement Techniques: Fundamentals, Measuring Instruments, Measuring Methods, Springer International Publishing, Switzerland, 2013, pp. 170–187. [4] Tatsuo Kawamura, Eiichi Haginomori, Yutaka Goda, Tetsuya Nakamoto, Recent developments on high current measurement using current shunt, in: IEEJ-EIT Joint Symposium on High Voltage Power Technology, Bangkok, Thailand, November 14–15, 2006, pp. 31–36. [5] B. Lago, R. Eatock, Coaxial shunt, Proc. IEE 114 (September (9)) (1967) 1317–1324. [6] Alf J. Schwab, Low-resistance shunts for impulse currents, IEEE Trans. Power Appar. Syst. PAS-90 (5) (1971) 2251–2257. [7] Ryszard Malewski, Wirewound shunts for measurement of fast current impulses, IEEE Trans. Power Appar. Syst. PAS-103 (October (10)) (1984) 2927–2933.
7
[8] W. Pfeiffer, Fast measurement techniques for research in dielectrics, IEEE Trans. Electr. Insulation EI-21 (October (5)) (1986) 763–780. [9] Farouk A.M. Rizk, Giao N. Trinh, High Voltage Engineering, CRC Press, 2014, pp. 740–743. [10] M. Kando, N. Pattanadech, New current-voltage transferring device with different metals, IEEJ Trans. Electr. Electron. Eng. 10 (September (5)) (2015) 603–604 (Willy & Sons, New York). [11] M. Kando, N. Pattanadech, New electrical shunt with two rectangular holes in a copper plate, IEEJ Trans. Electr. Electron. Eng. 10 (November (6)) (2015) 723–725 (Willy & Sons, New York). [12] M. Braunovic, N. Alexandrov, Intermetallic compounds at aluminum-to-copper electrical interfaces: effect of temperature and electric current, IEEE Trans. Compon. Packag. Manuf. Technol. A 17 (March (1)) (1994) 78–85. [13] M.P. Mubiayi, E.T. Akinlabi, M.E. Makhatha, Effect of process parameters on tensile strength and morphology of friction stir spot welds of aluminum and copper, in: 2017 8th International Conference on Mechanical and Intelligent Manufacturing Technologies (ICMIMT), Cape Town, South Africa, February 3–6, 2017, pp. 48–53. [14] F. Heidler, W. Zischank, Z. Flisowski, Ch. Bouquegneau, C. Mazzetti, Lightning current given in IEC 62305—background, experience and outlook, in: 29th International Conference on Lightning Protection, Uppsala, Sweden, 23rd–26th June, 2008, pp. 1–22. [15] Christof Drilling, Markus Droldner, Ernst Günther Jordan, Jan Meppelink, A New Generator for Testing of SPD’s Using Multiple Lightning Current Impulses for Combined Tests with Follow Currents, pp. 1–8. http://obo.de/ downloads/GENERATOR.PDF. [16] Ju-Hong Eom, Sung-Chul Cho, Tae-Hyung Lee, Parameters optimization of impulse generator circuit for generating first short stroke lightning current waveform, J. Electr. Eng. Technol. 8 (2013) 742–748. [17] M. Kando, T. Inaba, AC phase controlled testing apparatus superimposed with double pulses, IEEJ Trans. PE 129 (3) (2009) 447–452 (in Japanese).
Please cite this article in press as: N. Pattanadech, M. Kando, An improved Current Voltage Transferring Device for high current high frequency measurement, Electr. Power Syst. Res. (2017), http://dx.doi.org/10.1016/j.epsr.2017.06.015