EXPERIMENTAL INVESTIGATION OF A TRANSIENT INDUCED VOLTAGE TO AN OVERHEAD CONTROL CABLE FROM A GROUNDING CIRCUIT Akihiro AMETANI, Tomomi OKUMURA, Naoto NAGAOKA, Nobutaka, MORI Doshisha University - Japan aametani@mail.doshisha.ac.jp 1. Introduction and sheath voltages and currents are summarized in Table 1. Electromagnetic disturbance of low voltage control circuits in generator stations and substations and of consumer electronic/digital circuits are becoming a significant problem as a number of electronic/digital circuits composed of sensitive semiconductors have been adopted. In fact, a few disturbances have been experienced in Japan [1-3]. During last 10 years more than 300 disturbances in the control circuits in the power stations and substations are informed, and about 70% is estimated due to lightning. Among the 220 cases, 104 cases are of the voltage class 66-77kV, 45 cases/110-154kv, 14 cases above 154kV, and 57 cases the voltage class unknown. The incoming path of a lightning surge causing the disturbances has been investigated in the hydraulic generator stations and substations, and the result indicates two dominant paths : (1) via VTs and CTs, (2) via grounding systems due to arrester operation in a main circuit and etc. It is statistically found that more than 20% of the lightning surges are coming through the grounding systems, and another 20% through VTs and CTs. The present paper carries out an experimental investigation of a transient voltage and current of an overhead control cable induced from a single grounding conductor, i.e. a counterpoise, when a lightning current flows into the grounding conductor probably by an arrester operation. The transient voltages and currents on the cable core and sheath are measured under various termination of the cable. 500Ω PG 10m 4m 3D-2V 0.3m 1 node 0 1 2 1m 2m Fig. 1 Experimental setup 0.1m 1m 2. Experimental setup (a) Transient voltage Fig. 1 illustrates an experimental setup. A naked Cu conductor (counterpoise) with the radius of 1cm is buried underneath the earth surface at the depth of 30cm with the total length of 4m. A coaxial single core cable (3D2V) with the length of 2m is suspended above the earth surface at the height of 10cm. A pulse generator (PG) is connected to one end of the counterpoise through a resistance 500Ω via a lead wire of 10m. When measuring a sheath voltage, one terminal of an oscilloscope is connected to a voltage reference line with the length of 60m of which the other end is grounded independently from the counterpoise. Terminating conditions of the both ends of the overhead cable core and sheath are changed as summarized in Table 1. For examples, the sheath ends are either connected (ON) / disconnected (OFF) to the counterpoise or independently grounded (IG) from the counterpoise. The core ends are either open-circuited or connected through a matching resistance to the sheath. Measured results of core (b) Transient current Fig. 2 Measured results of transient voltage and current waveforms along a counterpoise DOS_Ametani_A1 Session 2 Paper No 4-1 -
case No. sheath to counterpoise core - to - sheath resistance[ Ω ] 3. Measured Results and Discussions 3.1 Voltage and Current along Counterpoise Fig. 2 shows measured results of transient voltages and currents along a counterpoise. Wave propagation characteristics along a counterpoise have been discussed in detail in reference [4]. 3.2 Sheath Ends Open-Circuited Fig.3 shows measured results of transient voltage and current waveforms on the cable core ( core to sheath ) and sheath ( to voltage reference wire ) in the case of the sheath both ends being open-circuited (OFF, OFF). (a) is the core to sheath voltage at the receiving end, (b) the sheath receiving-end voltage and (c) the sheath current in case11. It is clear in Fig. 3(a) that the core voltage is nearly the same as the sheath voltage at the receiving ends. The maximum difference is less than 2V as observed in Table 1 independently from the core termination. The phenomenon is readily explained by a traveling wave theory. Define the surge impedance matrices between nodes 0 to 1, 1 to 2 and 2 to right in Fig.1 as [Z 1 ] and [Z 2 ] respectively. Then, the refraction coefficients λ 1f from the left to the right at the node 1 is given in the following equation[5]. [λ 1f ] = 2 [Z 2 ] ( [Z 1 ] + [Z 2 ] ) -1 (1) Z g 0 0 where [Z 1 ] = 0 R1c 0, 0 0 R1s Zg [Z 2 ] = Zc Zm (2) Zm Zs Table 1 Experimental conditions and measured results max. core to sheath voltage [V] max. sheath voltage to reference wire [V] 60 60 1.60 1.32 15.47 15.03 0.0065 0.0044 0.0209 0.0271 and Z g : self surge impedance of a counterpoise Z c : self surge impedance of a cable core Z s : self surge impedance of a cable sheath Z m : mutual impedance between a core and a sheath Z n : mutual impedance between a counterpoise and a cable R 1c, R 1s : terminating impedance of a core and a sheath at node 1 When the both ends of the cable core and sheath are open-circuited, the terminating impedances are difined by : R 1c = R 1s = infinite ( ) (3) Substituting the above condition into eq. (1), the following refraction coefficient matrix is obtained [5]. 1 0 0 [λ 1f ] = Z n / Z g 0 0 (4) Z n / Z g 0 0 The above equation indicates that the sending-end voltages on the core and the sheath are the same. When the core sending end is short-circuited to the sheath, the refraction coefficient matrix [λ 1s ] is defined in the following equation [6]. 1 [λ 1s ] = 2( Z T Z ) T 1 U + 1' 2 t (5) Z g 0 where [ Z 1 '] =, 0 1 0 0 [T ] = : rotation matrix (6) 0 1 1 and subscript t for a transposed matrix. Then, the refraction coefficient is given in the following 2 2 matrix. 1 0 [λ 1s ] = Z n / Z g 0 core Max. current [A] send receive R 1c R 2c send receive send receive send receive send receive 11 OFF OFF 0.82 0.89 35.99 34.12 0.0077 0.0129 0.0231 0.0177 14 OFF OFF 60 60 1.32 0.47 35.20 34.38 0.0088 0.0060 21 ON OFF 1.85 1.03 52.42 43.50 0.0088 0.0127 0.0481 0.0236 22 ON OFF 60 0 1.09 1.40 52.17 43.73 0.0091 0.0076 23 ON OFF 0 60 1.19 1.50 51.90 43.28 0.0094 0.0094 24 ON OFF 60 60 1.33 1.45 51.75 43.82 0.0108 0.0094 31 OFF ON 1.71 0.72 45.53 38.00 0.0092 0.0129 0.0132 0.0262 32 OFF ON 60 0 1.57 0.83 47.75 37.22 0.0106 0.0094 33 OFF ON 0 60 1.66 0.90 47.05 37.77 0.0117 0.0079 34 OFF ON 60 60 1.30 0.97 46.67 38.38 0.0100 0.0063 41 ON ON 1.71 0.79 44.97 36.56 0.0104 0.0094 0.2106 0.2095 42 ON ON 60 0 1.85 1.03 46.98 37.31 0.0104 0.0062 43 ON ON 0 60 1.61 1.20 44.78 37.53 0.0113 0.0050 44 ON ON 60 60 1.61 1.05 45.64 37.72 0.0099 0.0051 51 independent 1.45 0.80 14.89 15.49 0.0221 0.0267 52 grounding 60 0 1.64 1.53 15.00 14.99 0.0073 0.0083 0.0205 0.0272 53 0 60 1.53 1.38 14.80 14.84 0.0069 0.0052 0.0208 0.0278 54 sheath (7) DOS_Ametani_A1 Session 2 Paper No 4-2 -
(a) Core to sheath voltage at the receiving end 65 16 16 Z 0 = 16 381 328 at f = 10MHz (8) 16 328 328 Thus, the transient induced voltage to the cable is estimated to be about 30% of the counterpoise voltage. The maximum counterpoise voltage being about 120V in Fig. 2, the cable voltage is evaluated roughly to be 30V which is smaller than the measured result in Fig. 3(b). The discrepancy is estimated due to a direct induction from the current lead wire. The sheath current in Fig. 3(c) is quite oscillatory, similarly to the core-to-sheath voltage in Fig. 3(a). The high frequency component is estimated due to multiple reflection of the coaxial and the earth-return modes of propagation (coaxial v 1 = 195m / µ s, earth 236 m / µ s ) at the cable ends. On the contrary, the sheath voltage shows a rather smooth waveshape, similarly to the counterpoise voltage. The propagation velocity on the counterpoise is measured to be about 100 m / µ s, which gives a dominant transient frequency = 1/ 4τ 6.3MHz, or one cycle T = 160ns. f g g 3.3 Sheath Sending-End Connected to Counterpoise (b) Sheath voltage at the receiving end (c) Sheath current in Case 11 Fig.3 Transient voltage waveform on a cable in Case 1i It is observed from eq. (4) and (7) that the core voltage is the same as the sheath voltage at the sending end independently from the open and the short circuit of the core and the sheath, and is given as ( Z n / Z g ) E0, where E 0 is a traveling-wave voltage incoming to node 1 along the counterpoise. The sending end voltage V g of the counterpoise is given by E 0 as in eq.(7), if no circuit is connected to the cable sending end. The characteristic impedances calculated by the Cable Parameters of the ATP-EMTP[7] are given in eq.(8) at 10MHz. Fig.4 shows transient sheath voltages when the sending end of the cable sheath is short-circuited to the counterpoise at the node 1 in Fig. 1, i.e. Case 2i in Table 1. The receiving end of the sheath is open-circuited. It is observed that the sheath voltage at the sending end is similar to that of the counterpoise voltage at the node 1 in Fig. 2(a). The receiving-end voltage is attenuated at the wavefront and the oscillatory waveform becomes distinct. The sheath current in Fig. 4(c) is less oscillatory than that in Case 11, Fig. 3(c) and the peak current at the sending end is nearly twice of Case 11. The node 1 voltage is analytically determined in the same manner as that in Sec.3.2. The sheath voltage is the same as the counterpoise voltage because those are short circuited. The voltage is roughly given by : g s ( Z Z ) V = V = E 2 V c (9) 0 2Z s / g + where E 0 : traveling-wave voltage incoming to node 1 along the counterpoise Applying the characteristic impedances in eq.(8), V g = V s V c ( 10 /11) E 0 With E0 50V at the node 1 in Fig. 2(a), the above equation gives the following result. V g = V s V c 45.5V The value agrees well with the measured results in Table 1. It should be noted that the sheath voltage in Case 2i is greater than that in Case 1i. s DOS_Ametani_A1 Session 2 Paper No 4-3 -
same, except the (a) Sending end (a) Sending end (b) Receiving end (b) Receiving end (c) Sheath current in Case 21 Fig.4 Transient sheath voltages in Case 2i 3.4 Sheath Receiving-End Connected to Counterpoise Fig.5 shows transient voltages in the case of the sheath receiving end being short-circuited to the counterpoise at node 2 (Case 3i). The maximum voltage is smaller in Case 3i than in Case 2i as observed in Table 1. Fig. 5(c) shows the voltage difference between the counterpoise and the sheath at the sending end. The counterpoise voltage is composed of Fig. 5(a) and (c), and becomes similar to that in Fig. 2(a). 3.5 Both Ends of Sheath Connected to Counterpoise Fig.6 shows transient voltages and current waveforms on the cable sheath in Case 4i. It is clear that the sending-end voltage waveform in Fig. 6(a) is nearly the (c) Voltage difference between the counterpoise and the sheath Fig.5 Transient sheath voltages in Case3i maximum value, as that in Fig. 4(a) in the case of the sheath sending end short-circuited to the counterpoise. The receiving-end voltage in Fig. 6(b) is similar to Fig. 5(b) except the initial part. As a result, the maximum sheath voltage is the smaller in Case 4i than that in Cases 2i and 3i as observed in Table 1, and this agrees with the conventional practice of the both ends of a cable sheath to be short-circuited to a counterpoise. Currents waveforms in Fig. 6(c) are quite different from those in Cases 1i to 3i which are highly oscillatory. The peak value of the current is far greater in Fig. 6(c). The reason is simply due a closed circuit composed of the sheath and the counterpoise. DOS_Ametani_A1 Session 2 Paper No 4-4 -
is (a) Sending end (a) Core-to-sheath voltage at the receiving end (b) Receiving end (b) Sheath voltage at the sending end (c) Current in Case 41 Fig.6 Transient sheath voltage and current waveforms in Case 4i (c) Sheath voltage at the receiving end 3.6 Sheath Grounded Independently from Counterpoise A transient voltage and current induced to a control cable from a counterpoise is desired as small as possible from the viewpoint of the insulation and the electromagnetic immunity of the control circuit in a power station and a substation. To achieve this, the paper investigates an independent grounding (Case 5i in Table 1) of the cable sheath from the counterpoise. The cable sheath is grounded at the both ends to a buried vertical rod apart by 2m from the counterpoise. Fig.7 shows measured results of transient voltage and current waveforms. (a) is the core-to-sheath voltage at the receiving end. A maximum voltage difference between the core and the sheath is observed to be less than 1.5V, which (d) Sheath current in Case 51 Fig.7 Transient voltage and current waveforms in Case 5i (independent grounding) DOS_Ametani_A1 Session 2 Paper No 4-5 -
similar to those in the Cases 1i to 4i in Table 1. Fig.7(b) and (c) are the sheath voltages at the sending and the receiving ends respectively. A comparison with the sheath voltage waveforms in Case 1i to 4i (Figs. 3 to 6) indicates that there is no fast rise of the voltage at the wavefront, and the maximum sheath voltage in Case 5i is less than 15.5V, while it reaches about 36V in Case 1i, 52.5V in Case 2i, 48V in Case 3i and 47V in Case 4i. Thus, it is clear that the maximum sheath voltage is reduced to 1/3 of that in the cases of the cable sheath being connected to the counterpoise. Fig. 7(d) shows the sheath current in Case 51. The sheath current is observed to be similar to that in Fig. 3(c), Case11 in the beginning, and to be far smaller than that in Fig. 6(c), Case 41. [4] A. Ametani, et.al. : Basic investigation of wave propagation characteristics on an underground naked conductor, Proceedings of ICEE 02, Jeju (Korea), pp.2141-2146, 2002 [5] A. Ametani : Distributed-Parameter Circuit Theory, Corona Pub.Co. (Tokyo), 1990.2 [6] N. Nagaoka and A. Ametani, Transient calculations on crossbonded cables, IEEE Trans., vol.pas-102, pp.779-786, 1983 [7] A. Ametani : Cable Parameters Rule Book, B.P.A. 1996. 4 4. Conclusions The paper has investigated experimentally transient voltages and currents on an overhead cable induced from a counterpoise, representing an induced voltage and current on a control cable in a power station and a substation. Based on the investigation, the following remarks have been obtained. (1) No significant voltage difference appears between the cable core and the sheath. (2) The sheath voltage is smaller in the case of the sheath being open-circuited, i.e. not connected to a counterpoise, than that in the case of the sheath connected to the counterpoise. (3) When a cable sheath is connected to a counterpoise, the sheath voltage becomes the smallest and the sheath current is the largest in the case of the both ends connected to the counterpoise. The sheath voltage is the largest in the case of the sending end connected to the counterpoise. (4) The sheath voltage is far smaller in the case of the sheath being grounded independently from the counterpoise than that in the case of the sheath connected to the counterpoise. Thus, the independent grounding of a control cable could be an effective means to reduce the transient overvoltage on a control cable in a power station and a substation, unless a voltage difference between the independent grounding and the counterpoise (or mesh) dose not cause a problem. References [1] T. Hasegawa, et.al., A fact-finding analysis of lightning disturbances and substations, IEE of Japan, Annual Meeting Records, Paper 1218, 1992 [2] Technologies of Countermeasures against Surges on Protection Relay and Control Systems, Report of Japanese ETRA, vol.57, No.3, Jan.2002 [3] T. Sonoda, et.al., An experimental study on surges induced from grounding grid to low-voltage control circuits, IEE of Japan, Research Meeting, Paper HV-01-129, 2001 DOS_Ametani_A1 Session 2 Paper No 4-6 -