Measurement of dielectric strength of 5 kv ethylene propylene rubber (EPR) cable under AC voltage

Transcription

1 SOP TRANSACTIONS ON POWER TRANSMISSION AND SMART GRID In Press SOP TRANSACTIONS ON POWER TRANSMISSION AND SMART GRID Measurement of dielectric strength of 5 kv ethylene propylene rubber (EPR) cable under AC voltage Bishal Pradhan*, Clayborne D. Taylor, Stanislaw Grzybowski Mississippi State University, MS, USA *Corresponding author: bp623@msstate.edu Abstract: This study deals with AC voltage stress imposed on ethylene propylene rubber (EPR) cable. The cable terminal was prepared and installed in different test sets constructed inside the lab. The AC voltage was increased till breakdown of cable. The proper breakdown of cable was achieved by the test set using de-ionized water as dielectric liquid medium surrounding cable terminal. The electrical breakdown strength of 5 kv EPR cable under AC voltage has been measured by constructing a suitable test set in Mississippi State University High Voltage Lab. Keywords: Dielectric Strength; Electric Stress; Permittivity; Semicon Terminus 1. INTRODUCTION Cable is a current carrying conductor surrounded by insulation. It is used for transmission and distribution of electricity. During the operation of cable, insulation of cable has to withstand different electrical, thermal, environmental stresses etc. including AC voltage stress. These stresses slowly age the insulation, thereby, decreasing its dielectric strength over time of voltage application. Different mechanisms of breakdown (such as intrinsic, avalanche, electromechanical, thermal, erosion, electrochemical breakdown) of solid insulation occur according to the time of voltage application [1]. Dielectric strength of solid insulation is higher during early period of voltage application which implies higher test voltage needed for breakdown of the insulation. Cable is tested multiple times during various phases of its lifespan such as design phase, manufacture, installation and operation phases to assure the product reliability against all operational stresses. The testing may be voltage withstand test, impulse voltage test, dielectric response, partial discharge, time-domain reflectometry, thermal infrared imaging test etc. The testing may also include destructive disruptive discharge voltage test under AC voltage, also known as AC breakdown voltage test, where the AC voltage is raised until breakdown occurs on the insulation, thereby, reducing the voltage between the electrodes practically to zero. The breakdown voltage test is used to measure the dielectric strength of the cable as well as to predict the cable-life. Institute of Electrical and Electronics Engineer (IEEE) and American Society for Testing and Materials (ASTM)-D149 recommend the procedures for disruptive discharge breakdown test [2, 3]. 1

2 SOP TRANSACTIONS ON POWER TRANSMISSION AND SMART GRID 2. CABLE TERMINATION When cable termination is not prepared at the ends of cable, there is external flashover between the surface of inner conductor and outer metallic shield during application of AC breakdown test voltage due to short separation distance between them and applied electric field stress being higher than the dielectric strength between them. The cable termination is made by stripping out the insulation shield at the end of cable making the distance between the surface of conductor and surface of metallic shield far enough so as to prevent external flashover between them. During this process, tangential electric field is introduced in the region of insulation semicon terminus. The superposition of tangential and radial electric field causes high electric stress in the semicon terminus. Hence, the preparation of cable termination solves the problem of external flashover but introduces new problem of higher stress in the semicon terminus [4] as shown in figure 1.a. When AC test voltage is applied, the cable breaks down early through this region. This problem has to be addressed to achieve proper breakdown of cable. The tangential stress has to be within tolerable value which can be achieved by field control methods such as geometric field control and resistive-capacitive field control. The geometric field control is the way to control and re-arrange electric field distribution to tolerable electric stress value by the arrangement of suitable geometry between primary and auxiliary electrode by using corona ring, stress cone etc. The resistive-capacitive field control method uses the stress grading material whose electrical characteristics is such that when they are placed in the electric field enhanced region, the material will re-arrange the electric field distribution in the desired tolerable limit. A material characterized by a rather high permittivity is positioned between the two electrodes providing the electric stress reduction as shown in figure 1.b. (a) High electric stress region in untreated semicon terminus. (b) Reduced electric stress due to field control method. Figure 1. is adapted from [6]. 3. LABORATORY TEST AND RESULTS Southwire s Envirotect CT1-09ET Type MV-105 Cable having 2.92 mm as nominal thickness of EPR insulation was chosen as 5 kv cable sample for the experiment [5]. A 250 kv AC test transformer was used as AC source for the test. A 300 kv potential transformer was used to measure the applied voltage. The water resistor was used between test transformer and cable as a protective device for test transformer to limit the current flow in case of breakdown of cable. The voltage was applied on copper conductor and copper tape was grounded. Different configuration of test set up was performed as the experiment progressed [6]. The first experimental set up was done by immersing cable terminal in the bucket filled with mineral oil as shown in figure 2. Cable termination was prepared from 3 m length sample. The geometrical field control method was implied on the region of semicon terminus, where maximum electric stress occur, with the help of corona ring and 2

3 Measurement of dielectric strength of 5 kv ethylene propylene rubber (EPR) cable under AC voltage (a) Cable termination. (b) Experimental set up with bucket. (c) Puncture. Figure 2. is adapted from [6]. the steel bowls with double cone structure was attached on the conductor termination. The voltage was increased till breakdown occurred. Slight variation was done each time in the experimental configuration for this first experimental set up with bucket and five measurements were taken for AC breakdown voltage as 93 kv, 55 kv, 56 kv, 51 kv and 50 kv. The puncture mostly occurred in the semicon terminus and the AC breakdown voltage had lower value so this set up didn t provide the result as expected. (a) Cable termination. (b) Experimental set up with bucket. (c) Puncture. Figure 3. is adapted from [6]. The second experimental set up was done by inserting cable sample in the pipe filled with mineral oil as shown in figure 3. This set up reduced moisture penetration in oil and the volume of oil used. The length of cable sample was increased to 480 cm. The semiconducting tape was wrapped around semicon-insulation interface. Corona rings were placed on conductor end and around semicon-insulation interface. The measurements were conducted with slight variation of cable termination length for each measurement and the AC breakdown values were recorded as 57 kv, 62 kv, 84 kv, 51 kv, 10 kv. The puncture was on semicon terminus giving lower breakdown voltage value so this set up didn t provide the expected result. The third experimental set up was done by inserting cable termination in cable test termination (CTT) filled with de-ionized water as shown in figure 4. The 123 cm long PVC tube with internal diameter of 10 cm was used as CTT. The 300 cm long cable sample was stripped with 3 cm bare conductor, 40 cm insulation and 40 cm semicon for cable termination and the steel electrode was fitted at the cable end. De-ionized water was poured inside the CTT surrounding cable termination. The conductivity of de-ionized water was measured to be 0.5 µs/cm. The voltage was applied through manual control of voltage regulator in test transformer from zero voltage to 80 kv uniformly with constant but with faster rate of rise and then again uniformly with slower constant rate of voltage rise till breakdown occurred in test object. The breakdown occurred as expected in the jacketed area of cable. The insulation resistance test and visual inspection were done to confirm the breakdown of cable. The resistance of cable was measured to be 1-80 kω after breakdown. It is to be noted that resistance of new cable before breakdown was 3 TΩ. Similarly, for 3

4 SOP TRANSACTIONS ON POWER TRANSMISSION AND SMART GRID (a) CTT. (b) Cable sample. (c) Puncture. (d) Experimental set up with CTT. Figure 4. is adapted from [6]. visual inspection, the jacketed layer was scored off and the copper tape was taken out to see the punctured region. The black dust was observed over the cable surface due to burning of the insulation and puncture on the break down area. However, in some samples, the puncture was seen on the region of semicon surface. The reason for puncture on semicon was due to gas bubbles sticking on the surface of semicon. The table 1 for AC-breakdown using de-ionized water in CTT is shown as follows. Table 1. The AC breakdown results for set up with de-ionized water in CTT as adapted from [6]. Sample No AC breakdown value Remarks (kv) Average AC breakdown value 121 kv 4

5 Measurement of dielectric strength of 5 kv ethylene propylene rubber (EPR) cable under AC voltage 4. ANALYSIS A. The test set up of de-ionized water in CTT was based on resistive field control method. It was successful to obtain desired AC breakdown voltage value by controlling electric stress in semicon terminus due to following properties of de-ionized water. Firstly, the conductivity of de-ionized water was measured to be 0.5 µs/cm in the experiment i.e. resistivity of water as 20 kω?m, this lower resistivity of de-ionized water caused uniform voltage distribution along the cable termination [7]. The resistive grading due to de-ionized water made the electric field distribution uniform and controlled the enhanced electric stress on semicon-insulation interface to tolerable limit. Secondly, the configuration of cable inside de-ionized water can be taken as composite dielectrics having different values of permittivities (ε) and dielectric strength. We know that flux density (D) is independent of permittivity (ε), hence, D c = ε c E c and D w = ε w E w are equal where subscript c represents cable and w represents water. This relation gives interesting information about ratio of electric field stress being inversely dependent on ratio of permittivity. E w E c = ε c ε w (1) The permittivity of de-ionized water (ε w ) is 80.1 [8] whereas the permittivity of EPR is (ε c ) 2.3 [9] so, the electric field stress becomes lower by multiplication of to the electric field stress in EPR insulation when it enters de-ionized water medium. Hence, enhanced electric stress along the semicon-insulation interface in water medium is reduced to a tolerable value due to permittivity. Thirdly, the dielectric strength of water is kv/cm [8] whereas the dielectric strength of EPR is typically kv/cm [9]. The increased dielectric strength of water retains the integrity of dielectric properties of water for higher electric stress in comparison to EPR insulation. Hence, breakdown of EPR is more likely to occur than that of water when AC test voltage is increased in conductor. B. The electrical dielectric strength of EPR insulation due to radial electric field stress is calculated from AC breakdown voltage data obtained by test set up with de-ionized water in CTT using the following formula. E maximum = V rln R r (2) The average AC voltage breakdown value (V) is measured to be 121 kv in the experiment. Also, conductor radius (r) = 4.60 mm and insulation radius (R) = 8.25 mm. Using equation (2), E maximum is calculated as 45 kv/mm = 450 kv/cm. Hence, the electrical dielectric strength of EPR cable under AC voltage is 450 kv/cm. C. The experimental set up with oil in bucket and pipe were not able to control electric stress on semicon terminus so the breakdown of cable was observed around that region. The conductivity of mineral oil is around 5.85e-16 S/cm to 3.94 e-17 S/cm [10]. The conductivity of water is calculated around 0.5e-6 S/cm during experiment. As liquid dielectric of lower value of conductivity (i.e. liquid dielectric of higher resistivity) is used in CTT, the voltage distribution becomes non-linear. With very lower value of conductivity, the voltage around semicon terminus even increases above 1 per unit value due to capacitive effect. As the conductivity of liquid dielectrics is lowered to smaller value, the voltage drop is concentrated only on the semicon terminus so voltage gradient, which is defined as electric field, is maximum on that region [7]. This caused maximum electric stress around semicon terminus when oil was used as liquid dielectric medium. The voltage distribution obtained with mineral oil as well as de-ionized water is shown in 5 calculated from mathematical model with conductivity values of oil and de-ionized water. It is noted that oil medium was used in cable termination instead of air medium because oil medium has better permittivity and dielectric breakdown properties than air; however, permittivity and dielectric breakdown strength of oil is in the similar range to that of EPR, hence, not much can be done to control electric stress through electrical characteristics of mineral oil. The corona rings were also used along with oil in semicon terminus for electric stress control up to 2.5 kv/cm or lesser to avoid the discharge [11] through geometrical field control method. However, the experiment set 5

6 SOP TRANSACTIONS ON POWER TRANSMISSION AND SMART GRID with oil was unable to control electric stress along semicon-insulation region. The mathematical model for voltage distribution is developed for cable termination as follows [7]: V (x) = V H V Hsinhγx sinhγl (3) (a) Cable termination dimension. (b) Voltage distribution for oil and water. Figure 5. Voltage distribution along cable termination length as adapted from [7]. D. The water termination was able to control electrical stress around semicon terminus, however, some breakdown of cable was around semicon region giving the breakdown value nearly equal to the breakdown found in jacketed region. There were formation of air bubble inside the water which increased as temperature of water increased. These air bubbles were attached to semicon layer due to adhesive force by morphology of three phase contact line [12]. The air bubble had lower permittivity of [8] due to which electric stress was very high than surrounding medium which had higher permittivity. Also, electric breakdown of air was 30 kv/cm [8] which was comparatively lower. This caused breakdown of air and partial breakdown in CTT caused production of ions and electrons which further increased to full disruptive discharge in whole system. Thus, there was breakdown through semicon surface where air bubble was attached in the water termination of CTT. The future work for improved water termination construction in HV lab needs to address this issue. 6

7 Measurement of dielectric strength of 5 kv ethylene propylene rubber (EPR) cable under AC voltage 5. CONCLUSION Different types of test set up were conducted to obtain the proper AC breakdown voltage value. The set up with mineral oil in bucket, the test setup with pipe containing mineral oil, and test set of CTT with de-ionized water were experimented. The desired result was obtained through test set up with de-ionized water in CTT. The average breakdown voltage for eight cable samples with desired experimental configuration was measured to be 121 kv. The electrical breakdown strength of 5 kv ethylene propylene rubber (EPR) cable under AC voltage was calculated as 450 kv/cm. References [1] J. Kuffel, E. Kuffel, and W. S. Zaengl, High Voltage Engineering:Fundamentals, vol. 2. Butterworth-Heinemann, [2] IEEE-Std.4, IEEE Standard Techniques for High-Voltage Testing, [3] ASTM-D149, Standard Test Method for Dielectric Breakdown Voltage and Dielectric Strength of Solid Electrical Insulating Materials at Commercial Power Frequencies, pp. 1 13, [4] J. P. Mackevich and J. Hoffman, Insulation Enhancement with Heat- Shrinkable Components Part III: Shielded Power Cable, Electrical Insulation Magazine, IEEE, vol. 7, no. 4, pp , [5] ENVIROTECT CT1-09ET with SIMpull Jacket, pp , [6] B. Pradhan, Electrical breakdown strength of 5 kv ethylene propylene rubber (EPR) cable under AC voltage, in Electrical and Computer Engineering, p. 71, Mississippi State University: Mississippi State, [7] S. Strauch, S. Dieterich, K. Aniol, J. Annand, O. Baker, W. Bertozzi, M. Boswell, E. Brash, Z. Chai, J.-P. Chen, et al., Calculation of Electric Field Distribution at High Voltage Cable Terminations, in High Voltage Engineering and Application (ICHVE), 2010 International Conference on, pp , IEEE, [8] CRC Handbook of Chemistry and Physics [9] W. T. Shugg, Handbook of Electrical and Electronic Insulating Materials, vol. 2. IEEE Press, [10] T. Judendorfer, A. Pirker, and M. Muhr, Conductivity measurements of electrical insulating oils, in Dielectric Liquids (ICDL), 2011 IEEE International Conference on, pp. 1 4, IEEE, [11] R. Gleyvod and P. Mohaupt, Operating of water terminations for testing power cables, [12] A. T. Paxson and K. K. Varanasi, Self-similarity of contact line depinning from textured surfaces, Nature Communications, vol. 4, pp. 1 8,