Proceedings of the

23rd NORDIC INSULATION SYMPOSIUM

June JUNE9–12, 9-12, 2013 2013 Trondheim, Norway Trondheim, Norway

Department of Electric Power Engineering NORWEGIAN UNIVERSITY OF SCIENCE AND TECHNOLOGY

© NORD-IS &Akademika Publishing, 2013 ISBN 978-82-321-0274-7 This publication may not be reproduced, stored in a retrieval system or transmitted in any form or by any means; electronic, electrostatic, magnetic tape, mechanical, photo-copying, recording or otherwise, without permission. Layout: The authors Cover Layout: Akademika Publishing Paper: Gprint 90 gr Printed and binded by AiT Oslo AS Photo cover:

Pål Keim Olsen [email protected] We only use environmentally certified printing houses. Akademika Publishing NO–7005 Trondheim, Norway Tel.: + 47 73 59 32 10 www.akademikaforlag.no Publishing Editor: Lasse Postmyr ([email protected])

Nordic Insulation Symposium - Nord-IS 13 - Trondheim, Norway, June 9 - 12, 2011

Preface

This publication contains all the papers presented at the 23rd Nordic Insulation Symposium (Nord-IS 13) held in Trondheim, Norway, June 9 - 12, 2013. Before acceptance, the abstracts and then the 44 received papers were reviewed by members of the Organizing Committee and the Advisory Council with respect to relevance and quality. Challenges arising from use of HVDC is selected as the preferential subject for Nord-IS 13. All subjects dealt with at previous Nord-IS are, however, included. This means for example ageing and breakdown phenomena, condition assessment and measurement techniques. The Symposium is an interdisciplinary forum for open discussion of ideas, research results and practical experiences related to application of insulating materials and systems in electrical power apparatus. It is addressed to PhD students, researchers and engineers working within academia, research institutes, power industry and power utility companies. Nord-IS is held every second year in one of the Nordic countries; Norway, Denmark, Sweden and Finland. Young researchers are particularly encouraged to contribute. English is the working language of Nord-IS and participants from outside the Nordic area are welcome. I would like to express my gratitude to all those who have worked hard and contributed in many different ways to make Nord-IS 13 possible. Thanks are due to the members of the Organizing Committee and the Advisory Council for their cooperation in planning of the program and acting as session chairmen during the Symposium. I am particularly indebted to PhD fellow Pål Keim Olsen for his invaluable efforts as secretary, executing all the work associated with Nord-IS 13. – Last but not least I would like to thank all authors and participants for making Nord-IS 13 a success.

Trondheim, May 2013

Erling Ildstad Chairman, Nord-IS 13

III

Nordic Insulation Symposium - Nord-IS 13 - Trondheim, Norway, June 9 - 12, 2011

Organizing Committee Stanislaw Gubanski Chalmers

Sweden

Joachim Holbøll

Technical University of Denmark

Denmark

Erling Ildstad

Norwegian University of Science and Technology

Norway

Kari Lahti

Tampere University of Technology

Finland

Advisory Council Georg Balog

Subsea Cable Consultants

Norway

Jörgen Blennow

Chalmers

Sweden

Hans Edin

Royal Inst. Of Technology

Sweden

Rolf Hegerberg

Sintef Energy Research

Norway

Henrik Hilborg

ABB Corporate Research

Sweden

Claus Leth Bak

Aalborg University

Denmark

Petri Hyvönen

Aalto University

Finland

Anders Jensen

NKT Cables

Denmark

Harri Suonpää

Alstom Grid

Finland

Bjørn Sanden

StatNett

Norway

Juha Laakko

Terichem Tervakoski

Finland

V

Nordic Insulation Symposium - Nord-IS 13 - Trondheim, Norway, June 9 - 12, 2011

Secretary Pål Keim Olsen

Department of Electric Power Engineering, NTNU NO-7491 Trondheim

Mail: [email protected]

Phone: +47 73594722

History 1: 2: 3: 4: 5: 6: 7: 8: 9: 10: 11: 12: 13: 14: 15: 16: 17: 18: 19: 20: 21: 22: 23:

1968 - Nord-PD in Västerås, Sweden 1970 - Otnäs, Finland 1972 - Trondheim, Norway 1974 - Kollekolle, Denmark 1976 - Saltsjöbaden, Sweden 1978 - Vaasa, Finland 1980 - Røros, Norway 1982 - Odense, Denmark 1984 - Kungälv, Sweden 1986 - Hanaholmen, Finland 1988 - Trondheim, Norway 1990 - Lyngby, Denmark 1992 - Västerås, Sweden 1994 - Vaasa, Finland 1996 - Bergen, Norway 1999 - Lyngby, Denmark 2001 - Stockholm, Sweden 2003 - Tampere, Finland 2005 - Trondheim, Norway 2007 - Lyngby, Denmark 2009 - Gothenburg, Sweden 2011 - Tampere, Finland 2013 - Trondheim, Norway

VI

Fax: +47 73594279

Nordic Insulation Symposium - Nord-IS 13 - Trondheim, Norway, June 9 - 12, 2011

PROGRAM NORD-IS 2013 Sunday, June 9, 2013

17:00-19:00 Registration at NTNU. Mounting of Posters

Monday, June 10, 2013 08:00 - 09:00 Registration and mounting of posters 09:00 - 09:10 Opening of symposium:

Welcome to NTNU by head of department prof. Olav Fosso

09:10 - 09:40 Opening lecture: “Challenges arising from use of HVDC”…………p. XVII Erling Ildstad, NTNU, Norway 09:40 - 10:00 Coffee break and mounting of posters 10:00 - 12:00 Session 1 - HVDC Challenges Chair: Bjørn Sanden, StatNett (Norway) Conduction behavior of polyaniline/elastomer composites and the influence of carbon black addition………………………………………..p. 3 Björn Sonerud1, Knut Magne Furuheim1, Staffan Josefsson1, Jani Pelto2, Marjo Ketonen2, Outi Härkki2 1 Nexans Norway AS 2 VTT Technical Research Institute of Finland Short and long term behavior of functionally filled polymeric insulating materials for HVDC insulators in compact gas‐insulated systems……...p. 7 Michael Tenzer, Maximilian Secklehner, Volker Hinrichsen TU Darmstadt, High Voltage Laboratories Comparison of simulated and measured field dependent charge injection in mineral oil under dc bias………………………………………………p. 11 Olof Hjortstam, Christian Sonehag, Joachim Schiessling ABB Corporate Research Space Charge Accumulation in XLPE versus Temperature and Water Content…………………………………………………………………….p. 15 Torbjørn Andersen Ve, Frank Mauseth, Erling Ildstad NTNU

VII

Nordic Insulation Symposium - Nord-IS 13 - Trondheim, Norway, June 9 - 12, 2011

Surface Potential Decay on Silicon Rubber Samples at Reduced Gas Pressure……………………………………………………………………p. 18 Shahid Alam, Yuriy Serdyuk, Stanislaw Gubanski Chalmers University of Technology Challenges when measuring the DC electric field very close to an insulator surface……………………………………………………………………...p. 23 Birgitta Källstrand1, Daniel Borg1, Lars Walfridsson1, Charles Doiron2, Kenneth Johansson1 1 ABB AB, Corporate Research 2 ABB Schweiz AG, Corporate Research 12:00 - 13:00 Lunch 13:00 - 14:15 Poster Session 1 and coffee break 14:15 - 15:35 Session 2 - Breakdown and Ageing of Solid Insulation Systems Chair: Hans Edin, KTH (Sweden)

The Effect of DC Electro‐thermal Ageing on Electrical Treeing in Polyethylene……………………………………………………………….p. 29 Adrian Mantsch, Xiangrong Chen, Jörgen Blennow, Stanislaw Gubanski Department of Materials and Manufacturing Technology, Chalmers University of Technology Effect of Film Thickness and Electrode Area on the Dielectric Breakdown Characteristics of Metallized Capacitor Films………………………….p. 33 Ilkka Rytöluoto, Kari Lahti Tampere University of Technology Development of insulation system for variable speed driven motors; performance of a corona resistant magnet wire.......................................p. 39 Tomi Nuorala1, Janne Lehtonen2, Markus Takala1 1 ABB Oy, BU Motors and Generators 2 ABB Oy, BU Transformers Enhancement of Water Tree Initiation due to Residual and Applied Mechanical Strain on XLPE Cables……………………………………p. 43 Erling Ildstad1, Simon Årdal Aarseth1, Hallvard Faremo2 1 NTNU 2 Sintef Energy Research 15:30 - 16:00 Coffee break

VIII

Nordic Insulation Symposium - Nord-IS 13 - Trondheim, Norway, June 9 - 12, 2011

16:00 - 17:00 Session 3 - Breakdown and Ageing of Solid Insulation Systems Chair: Jørgen Blennow, Chalmers (Sweden) Thermal Ageing of XLPE Cable Insulation under Operational Temperatures – Does It Exist?...................................................................p. 49 Rasmus Olsen1, Joachim Holboell2, Mogens Henriksen2, Jens Hansen3 1 Energinet.dk 2 Technical University of Denmark 3 Danish Energy Association Influence of DC Stress Superimposed with High Frequency AC on Water Tree Growth in XLPE Insulation………………………………………..p. 53 Frank Mauseth1, Sverre Hvidsten2, Hans‐Helmer Sæternes2, Jørund Aakervik2 1 NTNU 2 SINTEF Energy Research Influence of antioxidants in epoxy‐anhydride resin used for HV applications………………………………………………………………..p. 57 Chau Hon Ho, Emmanuel Logakis, Andrej Krivda ABB Switzerland Ltd. ‐ Corporate Research

19:00 - 21:30 Symposium opening banquett at Banksalen, Trondheim city centre

IX

Nordic Insulation Symposium - Nord-IS 13 - Trondheim, Norway, June 9 - 12, 2011

Tuesday, June 11, 2013 08:00 – 09:00 Mounting of Poster Session 2 09:00 - 10:50 Session 4 - Condition Assessment and Test Procedures Chair: Petri Hyvönen, Aalto University (Finland) On‐line condition monitoring importance and evolution……………....p. 63 Nicolaie Fantana ABB DECRC Study of the dielectric response of ester impregnated cellulose for moisture content evaluation………………………………………………………...p. 67 Andrzej Graczkowski, Jarosław Gielniak, Piotr Przybyłek, Krzysztof Walczak, Hubert Morańda Poznan University of Technology Correction of Geometric Influence in Permittivity Determination……p. 71 Xiangdong Xu1, Tord Bengtsson2, Jörgen Blennow1, Stanislaw Gubanski1 1 Chalmers University of Technology 2 Chalmers University of Technology and ABB Corporate Research System for detection and analysis of partial discharges under transient voltage application………………………………………………………...p. 75 Søren Valdemar Kjær1, Joachim Holbøll2 1 DONG Energy 2 Technical University of Denmark VLF testing for High Voltage Cables, state of the art…………………..p. 79 Peter Mohaupt, Kurt Misteli, Harald Geyer Mohaupt High Voltage 10:50 - 11:00 Coffee break 11:00 - 12:00 Poster Session 2 12:00 - 13:00 Lunch 13:00 – 16:00 Technical visits – NTNU/SINTEF laboratories and Leirfossen Hydro Power Station 18:30 - 19:30 Greetings from the Mayor’s Office and Concert in Nidarosdomen 19:30 - 20:30 Tour Nidarosdomen

X

Nordic Insulation Symposium - Nord-IS 13 - Trondheim, Norway, June 9 - 12, 2011

Wednesday, June 12, 2013 09:00 - 10:30 Session 5 Breakdown and Ageing of Liquid Insulation Systems Chair: Henrik Hillborg, ABB Corporate Research (Sweden) Oil Aging due to Partial Discharge Activity……………………………..p. 85 Mohamad Ghaffarian Niasar, Respicius Cemence Kizza, Hans Edin KTH Royal Institute of Technology, Stockholm Streamer Propagation in a Long Gap in Model Liquids……………….p. 89 Van Dung Nguyen1, Hans Kristian Høidalen1, Dag Linhjell2, Lars E Lundgaard2, Mikael Unge3 1 Norwegian University of Science and Technology 2 SINTEF Energy Research 3 ABB Corporate Research Investigation of the Static Breakdown Voltage of the Lubricating Film in a Mechanical Ball Bearing………………………………………………….p. 94 Abhishek Joshi, Jörgen Blennow Chalmers University of Technology, Gothenburg Measurement techniques for identifying polarity dependence of ion injection in transformer oil……………………………………………….p. 98 Joachim Schiessling1, Deepthi Kubevoor‐Ramesh1, Yuriy Serdyuk2, Olof Hjortstam1 1 ABB Corporate Research 2 Chalmers University Gothenborg 10:30 - 10:45 Coffee break 10:45 - 12:05 Session 6 Gaseous and Impregnated Insulation Systems Chair: Rolf Hegerberg, Sintef Energy (Norway) Mechanical Simulations Regarding the Influence of Paper Insulation Degradation on the Radial Mechanical Strength of Continuously Transposed Conductors for Power Transformers…………………….p. 103 Daniel Geißler, Thomas Leibfried Institute of Electric Energy Systems and High Voltage Technology at Karlsruhe Institute ofTechnology (KIT) Effect of High Voltage Impulses on Surface Discharge at the Oil‐Paper Interface…………………………………………………………………..p. 108 Respicius Clemence Kiiza, Mohamad Ghaffarian Niasar, Roya Nikjoo, Xiaolei Wang, Hans Edin KTH

XI

Nordic Insulation Symposium - Nord-IS 13 - Trondheim, Norway, June 9 - 12, 2011

Radial Flow Paths for Oil in Mass Impregnated HVDC Subsea Cables ..……………………………p.112 Bendik Støa1, Erling Ildstad1, Magne Runde2 1 Norwegian University of Science and Technology 2 SINTEF Energy Research/Norwegian University of Science and Technology Corona at Large Coated Electrodes………..…………………………..p. 116 Mats Larsson1, Olof Hjortstam1, Håkan Faleke1, Ming Li1, Liliana Arevalo2, Dong Wu2 1 ABB Corporate Research 2 ABB HVDC 12:05 - 13:05 Lunch 13:05 - 14:45 Session 7 – Design and Modeling of Electric Components Chair: Anders Jensen, NKT Cables (Denmark) Strategies for Inclusion of Structural Mass Estimates in the Direct‐Drive Generator Optimization Process………………………………………..p. 123 Matthew Henriksen, Bogi Jensen Technical University of Denmark Estimating Transmission Line Parameters of Three‐core Power Cables with Common Earth Screen…………………………………………….p. 127 Yan LI1, Peter A. A. F. Wouters1, Paul Wagenaars2, Peter C. J. M. van der Wielen2, E. Fred Steennis2 1 Eindhoven University of Technology 2 DNV KEMA Energy & Sustainability Effects of Ambient Conditions on the Dielectric Properties of Thermally Sprayed Ceramic Coating……………………………………………….p. 131 Minna Niittymäki1, Tomi Suhonen2, Jarkko Metsäjoki2, Kari Lahti3 1 Department of Electrical Engineering, Tampere University of Technology 2 Advanced Materials, VTT Technical Research Centre of Finland 3 Department of Electrical Engineering, Tampere University of Technology Water Diffusion Barrier – A Novel Design for High Voltage Subsea Cables………………………………………………..…………………....p. 136 Knut Magne Furuheim1, Susanne Nilsson1, Svein Magne Hellesø2, Sverre Hvidsten2 1 Nexans Norway AS 2 Sintef Energy Research Robustness Analysis of Classical High Voltage Joint Design Under High Voltage DC Stress……………………………………………………......p. 140 Fredrik Fälth1, Santhosh Kumar BVMP2, Hossein Ghorbani1 1 ABB High Voltage Cables 2 ABB GISL

XII

Nordic Insulation Symposium - Nord-IS 13 - Trondheim, Norway, June 9 - 12, 2011

14:45 - 15:00 Closing of the symposium

Poster Session 1 Charge Decay Measurements on Polymeric Insulation Material under Controlled Humidity Conditions………………………………………..p. 149 Yvonne Späck, Sarath Kumara, Stanislaw M. Gubanski Chalmers University of Technology Dielectric Breakdown Strength of Polymer Nanocomposites‐The Effect of Nanofiller Content……………………………………………………….p. 153 Markus Takala ABB Oy, BU Motors and Generators Sensitivity Improvement of Acoustic Partial Discharge Detection Measurements through Wavelet Analysis……………………………...p. 157 Demetres Evagorou, Patrick Janus, Mohamad Ghaffarian Niasar, Hans Edin KTH Royal Institute of Technology Comparison of Test Setups for High Field Conductivity of HVDC Insulation Materials……………………………………………………..p. 161 Johan Andersson1, Villgot Englund1, Per‐Ola Hagstrand1, Carl‐Olof Olsson2, Andreas Friberg2 1 Borealis AB 2 ABB AB, Corporate Research Influence of Applied Voltage and Temperature on the Current through the Alumina‐filled poly(ethylene‐co‐butyl acrylate) Nanocomposites Under Constant Stress…………………………………………………………..p. 165 Nadja Jaeverberg, Bandapalle Venkatesulu, Lars Jonsson, Hans Edin KTH Mechanical Stress Distribution inside Dry Capacitor Elements……...p. 169 Linnea Petersson, Kun Wei, Göran Paulsson, David Stromsten, Johan Ekh ABB AB, Corporate Research

XIII

Nordic Insulation Symposium - Nord-IS 13 - Trondheim, Norway, June 9 - 12, 2011

Poster Session 2 Behavior of Rubber Materials under Exposure to High Electric Fields………………………………..p. 175 Anna Candela Garolera, Joachim Holböll, Mogens Henriksen Technical University of Denmark Thickness Dependency in Dielectric Breakdown Strength of Biaxially Oriented Polypropylene‐Silica Nanocomposite Films…………………p. 179 Hannes Ranta, Ilkka Rytöluoto, Kari Lahti Tampere University of Technology, Department of Electrical Engineering Lumped‐circuit Modeling of Surface Charge Decay in a Needle‐plane geometry………………………………………………………………….p. 183 Xiaolei Wang, Nathaniel Taylor, Mohamad Ghaffarian Niasar, Respicius Clemence Kiiza, Hans Edin KTH Capacitor performance limitations in high power converter applications…………………………………..p. 187 Walid Ziad El‐Khatib, Joachim Holböll, Tonny W. Rasmussen Denmark Technical Univertsity Positive Breakdown Streamers and Acceleration in a Small Point‐Plane Liquid Gap and Their Variation with Liquid Properties……………..p. 191 Dag Linhjell1, Stian Ingebrigtsen1, Lars Lundgaard1, Mikael Unge2 1 SINTEF Energy Research 2 ABB Corporate Research Axial Water Ingress MV XLPE Cable Designs with Watertight Barrier………………………………………………....p. 197 Knut Brede Liland1, Svein Magne Hellesø1, Sverre Hvidsten1, Karl Magnus Bengtsson2, Arve Ryen2 1 SINTEF Energy 2 NEXANS Norway Modelling of Partial Discharges in Polymeric Insulation Exposed to Combined DC and AC Voltage…………………………………………p. 202 Pål Keim Olsen, Frank Mauseth, Erling Ildstad Norwegian university of science and technology

XIV

Nordic Insulation Symposium - Nord-IS 13 - Trondheim, Norway, June 9 - 12, 2011

SESSION 5

Breakdown and Ageing of Liquid Insulation Systems

83

Nordic Insulation Symposium - Nord-IS 13 - Trondheim, Norway, June 9 - 12, 2011

Oil Aging due to Partial Discharge Activity M. Ghaffarian Niasar, H. Edin, R. Clemence Kizza Royal Institute of Technology (KTH)

Abstract

and this means that if the insulation condition getting worse and worse over time, the operator can repair transformer before it is too late. Most of transformers in use are oil filled transformer. The main insulations in these transformers are paper which covers conductor, pressboard which is used as support between disks and winding, and oil which is used both as a cooling fluid and insulation between windings. Many authors have investigated the effect of thermal ageing on paper and pressboard and correlate it with the mechanical strength of paper and pressboard [2]. Thermal ageing has also been applied to oil and the change of the oil parameters such as conductivity, breakdown strength, and acidity was reported [3-4]. In order to understand ageing process because of partial discharge it is necessary to investigate the effect of partial discharge on oil parameters such as breakdown strength and oil conductivity. The effect of carbon particle produced because of breakdown in oil was investigated in [5]. The main conclusion from that paper is that carbonization of oil lead to reduction of breakdown strength. In this paper transformer oil was aged by partial discharge activity for different duration of time and also by means of complete discharge produced by lightning impulses. Change in partial discharge parameters (number of discharge and average magnitude of discharge) over time is reported. Change of polarization and depolarization current, oil conductivity and oil breakdown strength due to application of impulse and partial discharge was compared with new oil.

Oil is the main insulation in power transformers and over long time of ageing its insulation properties can change. In this paper ageing of oil due to the exposure to electric discharges was investigated. The effect of high energy discharges (complete arc) and low energy discharges (partial discharges) on oil properties such as breakdown strength and oil conductivity was investigated. An experimental setup consisting of two spherical electrodes was designed. The adjustable distance between the two electrodes made it possible to have PD with different magnitude. The oil conductivity and breakdown strength was measured for three sets of experiments. The first group of experiments was performed on new oil in order to have a reference for comparison. In the second group of experiments the new oil samples was exposed to 1000 and 3000 lightning impulses. In the third group of experiments new oil samples was exposed to partial discharge for different duration of time. Oil conductivity and breakdown strength of these aged samples were compared with new oil. The results show that after exposure to lightning impulse oil conductivity increases and breakdown strength decreases, However PD activity for short time does not change the oil conductivity but it reduces the breakdown strength.

1. Introduction The power transformer is a critical component of the power system. Any failure in transformer could be very expensive because the transformer itself is expensive and the cost of power shut down is high. Previous studies show that the leading cause for transformer failure is insulation failure [1]. Insulation failures may occur at very short time such as surface flashover on power line insulator due to lightning overvoltage. However, since a transformer is expensive, during design process the safety margin which has been considered is quite high so the probability to get complete insulation breakdown at very short time is quite low. On the other hand, the insulation failure may occur over long time. This means ageing of insulation over time weakens the insulation which finally may lead to a complete breakdown. The aging process of insulating material can be due to thermal, mechanical or electrical stresses. In most cases, the insulation deterioration can initiate partial discharge activity inside insulation. If partial discharge continues inside insulation it can deteriorate the insulation further, and finally lead to complete breakdown. It is possible to correlate the insulation condition with the partial discharge activity. PD monitoring over time can show any change in insulation

2. Experimental setup In order to investigate the effect of partial discharge on oil parameter a setup was designed that is shown in figure 1. The setup simulates a metal conductor at floating potential which was used as a source of partial discharge. Partial discharge occurs between two metallic spheres with 20 mm diameter. One sphere is connected to high voltage electrode while another one is at floating potential. The oil gap between two electrodes was fixed to about 200 micro meters. The top lead of the container made of Teflon and is sealed with two O-rings in order to keep the generated gases inside the container and let them dissolved in oil in order to analyze the dissolve gases in oil as a function of ageing time (However this result is not presented in this paper). A magnetic stirrer is used to mix oil during ageing of oil with partial discharges. Oil Breakdown strength was measured by using two metallic spheres with the oil gap of 1 mm. for each oil sample with different level of ageing the breakdown strength was measured 20 times in order to get statistic in data.

85

Nordic Insulation Symposium - Nord-IS 13 - Trondheim, Norway, June 9 - 12, 2011

In order to measure oil conductivity a setup consists of two parallel electrodes with a distance of exactly 1 mm was used. The setup placed inside a metallic box in order to eliminate the effect of external charges on the measurement. Polarization and depolarization current and oil conductivity was measured by using a DC voltage source with maximum 3 kV output and a Keithley 6514 electrometer. The schematic of this measurement setup is shown in figure 2.

polarization and depolarization current of each sample of the processed oil was measured first in order to make sure the starting point is similar for all cases. While depolarization current showed very consistent result in all cases the polarization, or in other words, the time dependent conduction, was not consistent even for measurements on the same sample. For the measurement of the breakdown strength the voltage on the test sample was increased with rate of 1 kV/s until breakdown occurred. For each sample the experiment was repeated 20 times in order to get statistic in data. Mean value and standard deviation of the breakdown voltage was determined for each experiment.

4. PD ageing

A voltage equal to 150% of the PD inception voltage was applied to the test cell and PD activity was monitored up to desired time. Figure 3 shows the trend of number of PD over time and figure 4 shows the trend of average magnitude of PD over time. While the number of PD decreases over time the average magnitude of PD is almost constant during ageing time. Figure 1. Setup for oil ageing through partial discharge activity

Figure 3. Number of PD as a function of ageing time

Figure 2. Schematic used for polarization and depolarization current measurement Figure 4. Average magnitude of PD

3. Experimental procedure

5. Polarization and depolarization current measurement

The Oil that is used for this experiments was NYTRO 10XN, which is a common oil in Swedish transformers. First the new oil was dried and degasses under vacuum at 60 °C for 24 hours. 6 samples of oil with the volume of 1.5 litres were used for the experiments. Two samples, one for 1000 and one for 3000 impulse experiment and 3 other samples for partial discharge experiment and one sample for new oil breakdown strength experiment. Prior to each experiment

Measurement of polarization and depolarization current was performed on new oil samples several times. While all samples taken from the same container, still there was some variation in measurement especially on polarization current. All measurement performed for the electric field of 1 kV/mm. Measurement of polarization and depolarization current performed only up to 1 hour

86

Nordic Insulation Symposium - Nord-IS 13 - Trondheim, Norway, June 9 - 12, 2011

after connection and disconnection of the voltage. A general waveform of polarization and depolarization is shown in figure 5.

Depolarization current (A)

U(t)

-5

10

U0 tp

td

After 1 hour After 4 hour After 7 hour

-10

10

-15

10

-20

10

Time

ip

id

4

10

-9

Polarization current (A)

10

2

3

10

2

3

10 Time (s)

10

4

10

-8

Depolarization current (A)

-10

10

-12

10

-14

new oil oil exposed to 1000 impulse oil exposed to 3000 impulse

10

-16

10

0

10

1

10

2

10 Time (s)

3

10

4

10

Figure 9. Comparison between depolarization current of new oil and oil exposed to impulse

5.2. Effect of PD on depolarization current

polarization

and

Figure 10 and 11 show that polarization current (time dependent conductivity) is not varying a lot, however in the case of exposure to intense PD activity the polarization current decreases. The change on depolarization current is not significant similar to the case of exposure to impulse. Changing of oil conductivity for different experiment is shown in table 1.

-12

10 Time (s)

1

10

10

-11

1

0

Figure 8. Comparison between polarization current of new oil and oil exposed to impulse

10

10

new oil oil exposed to 1000 impulse oil exposed to 3000 impulse

10

After 1 hour After 4 hours After 7 hours

-10

-11

10

-12

-9

10

-10

10

10

10

Polarization current (A)

3

10

Figure 8 and 9 show that polarization current (time dependent conductivity) increases by applying impulse to oil. Even though that there is a little change in depolarization current, however it is not a big change.

According to figure 5, if we assume that the material has only constant DC conductivity, in this case the polarization current should be equal to a constant conduction current minus depolarization current. However for oil, the measurement results show that depolarization current decay to zero very fast while polarization current still is decaying. This means that oil conductivity is time dependent and it decrease until it reach to the dc conductivity. For each oil sample polarization and depolarization current was measured 3 times. The first experiment performed 1 hour after oil was poured into the metallic box, second and third experiment was performed 4 and 7 hours after oil was poured into the metallic box. Figure 6 shows the polarization current and figure 7 shows the depolarization current for new oil (dried and degassed). As it is clear in figure 6 the polarization current (time dependent conduction) is varying (decreasing) for each measurement however according to figure 7, depolarization current shows a good consistency. For most of experiment the same trend was observed. This behavior can be explained by particles in oil. The more time that the oil is resting in the metallic box the more particles can settle so because of that the conduction current decreases. The average of polarization (and depolarization) current obtained from the two last measurements on each oil sample is used in part 5.1 and 5.2.

0

2

10 Time (s)

5.1. Effect of impulse on polarization and depolarization current

Time

Figure 5. waveform of polarization and depolarization

10

1

10

Figure 7. Depolarization current for three consecutive measurements

i(t)

10

0

10

4

10

Figure 6. Polarization current for three consecutive measurements

87

Nordic Insulation Symposium - Nord-IS 13 - Trondheim, Norway, June 9 - 12, 2011

-9

50

10

Breakdown voltage (kV)

40

Polarization current (A)

-10

10

-11

10

1- new oil 2- oil exposed to 1000 impulse 3- oil exposed to 3000 impulse 4- oil exposed to 3.2 C PD 5- oil exposed to 9.0 C PD 6- oil exposed to 42.2 C PD

20

10 -12

new oil oil exposed to total 3.2 C PD oil exposed to total 9.0 C PD oil exposed to total 42.6 C PD

10

-13

10

0

10

1

10

2

10 Time (s)

0

3

10

-10

10

-12

10

new oil oil exposed to total 3.2 C PD oil exposed to total 9.0 C PD oil exposed to total 42.2 C PD 0

10

1

10

2

10 Time (s)

3

10

4

10

Table 1. Variation of oil conductivity

1000 impulse 3000 impulse total 3.2 C PD total 9.0 C PD total 42.2 C PD

4

5

6

7

This project was funded by SweGRIDS and also run with a connection to the innovation project KICInnoEnergy/CIPOWER which is gratefully acknowledged.

Figure 11. Comparison between depolarization current of new oil and oil exposed to PD

New oil exposed to

3

7. Acknowledgment

-16

10

2

In this paper oil ageing because of PD activity and lightning impulse was investigated. The results from PD activity show that while the number of PD decreases over time, the average magnitude of PD stays constant. Oil conductivity increases when it is exposed to complete breakdown, however when the electrical discharge is small it cause an increase in oil conductivity. Oil Breakdown strength decreases after it exposed to PD activity or complete breakdown due to lightning impulse.

-8

-14

1

6. Conclusion

10

10

0

Figure 12. Breakdown strength of different oil sample (new oil, aged with impulse and aged with PD)

4

10

Figure 10. Comparison between polarization current of new oil and oil exposed to PD

Depolarization current (A)

30

Conductivity (10−14 S/m) 60 s 3000 s 4.6 1.8 19.5 13.8 13.5 19 2.3 1.4 2.3 1.7 0.7 0.2

8. References [1] H. William, P. E. Bartley, “Analysis of Transformer Failures”, International Association of Engineering Insurers 36th Annual Conference – Stockholm, 2003. [2] Lars E. Lundgaard, Walter Hansen, Dag Linhjell and Terence J. Painter, “Aging of Oil-Impregnated Paper in Power Transformers”, IEEE Transactions on Power Delivery, Vol. 19, No. 1, January 2004 [3] Thomas JUDENDORFER, Alexander PIRKER, Michael MUHR, “Conductivity measurements of electrical insulating oils”, IEEE International Conference on Dielectric Liquids, 2011 [4] B.S.H.M.S.Y. Matharage1, M.A.A.P Bandara, M.A.R.M. Fernando, G.A. Jayantha, C.S. Kalpage, “Aging Effect of Coconut Oil as Transformer Liquid Insulation - Comparison with Mineral Oil”, IEEE International Conference on Industrial and Information Systems (ICIIS), 6-9 Aug. 2012 [5] M. Krins, H. Borsi, E. Gockenbach. “Influence of carbon particles on the breakdown and partial discharge inception voltage of aged mineral based transformer oil”, Seventh International Conference on Dielectric Materials, Measurements and Applications, 23-26 September 1996

6. Oil breakdown strength The breakdown voltage for six oil sample is shown in figure 12. Mean value and standard division for 20 experiments on each sample was used to calculate the breakdown strength of the specific sample. From the figure 12 it is clear that both PD activity and impulse cause reduction of breakdown voltage. The results show that while by increasing the number of impulse injected to oil the breakdown strength decreases, if the oil sample exposed to much more PD activity the breakdown strength may increase again.

88

Author index Aakervik, Jørund Aarseth, Simon Årdal Alam, Shahid Andersson, Johan Arevalo, Liliana Bengtsson, Karl Magnus Bengtsson, Tord Blennow, Jörgen Borg, Daniel Chen, Xiangrong Doiron, Charles Edin, Hans Ekh, Johan El-Khatib, Walid Ziad Englund, Villgot Evagorou, Demetres Faleke, Håkan Fantana, Nicolaie Faremo, Hallvard Friberg, Andreas Furuheim, Knut Magne Fälth, Fredrik Garolera, Anna Candela Geißler, Daniel Geyer, Harald Ghorbani, Hossein Gielniak, Jarosław Graczkowski, Andrzej Gubanski, Stanislaw Hagstrand, Per-Ola Hansen, Jens Hellesø, Svein Magne Henriksen, Matthew Henriksen, Mogens Hinrichsen, Volker Hjortstam, Olof Ho, Chau Hon Holbøll, Joachim Hvidsten, Sverre Härkki, Outi Høidalen, Hans Kristian

53 43 18 161 116 197 71 29, 71, 94 23 29 23 85, 108, 157, 165, 183 169 187 161 157 116 63 43 161 3, 136 140 175 103 79 140 67 67 18, 29, 71, 149 161 49 136, 197 123 49, 175 7 11, 98, 116 57 49, 75, 175, 187 53, 136, 197 3 89

Ildstad, Erling Ingebrigtsen, Stian Jaeverberg, Nadja Janus, Patrick Jensen, Bogi Johansson, Kenneth Jonsson, Lars Josefsson, Staffan Joshi, Abhishek Ketonen, Marjo Kiiza, Respicius Clemence Kjær, Søren Valdemar Krivda, Andrej Kubevoor-Ramesh, Deepthi Kumar, BVMP Santhosh Kumara, Sarath Källstrand, Birgitta Lahti, Kari Larsson, Mats Lehtonen, Janne Leibfried, Thomas Li, Ming Li, Yan Liland, Knut Brede Linhjell, Dag Logakis, Emmanuel Lundgaard, Lars Mantsch, Adrian Mauseth, Frank Metsäjoki, Jarkko Misteli, Kurt Mohaupt, Peter Morańda, Hubert Nguyen, Dung Van Niasar, Mohamad Ghaffarian Niittymäki, Minna Nikjoo, Roya Nilsson, Susanne Nuorala, Tomi Olsen, Pål Keim Olsen, Rasmus Olsson, Carl-Olof

15, 43,112, 202 191 165 157 123 23 165 3 94 3 85, 108, 183 75 57 98 140 149 23 33, 131, 179 116 39 103 116 127 197 89, 191 57 89, 191 29 15, 53, 202 131 79 79 67 89 85, 108, 157, 183 131 108 136 39 202 49 161

Paulsson, Göran Pelto, Jani Petersson, Linnea Przybyłek, Piotr Ranta, Hannes Rasmussen, Tonny W. Runde, Magne Ryen, Arve Rytöluoto, Ilkka Schiessling, Joachim Secklehner, Maximilian Serdyuk, Yuriy Sonehag, Christian Sonerud, Björn Späck, Yvonne Steennis, E. Fred Stromsten, David Støa, Bendik Suhonen, Tomi Sæternes, Hans-Helmer Takala, Markus Taylor, Nathaniel Tenzer, Michael Unge, Mikael Ve, Torbjørn Andersen Venkatesulu, Bandapalle Wagenaars, Paul Walczak, Krzysztof Walfridsson, Lars Wang, Xiaolei Wei, Kun Wielen, Peter C. J. M. van der Wouters, Peter A. A. F. Wu, Dong Xu, Xiangdong

169 3 169 67 179 187 112 197 33, 179 11, 98 7 18, 98 11 3 149 127 169 112 131 53 39, 153 183 7 89, 191 15 165 127 67 23 108, 183 169 127 127 116 71

JUNE 9-12, 2013 Trondheim, Norway

June, 9–12, 2013 Trondheim, Norway

9788232 102747

Department of Electric Power Engineering NORWEGIAN UNIVERSITY OF SCIENCE AND TECHNOLOGY

www.akademikaforlag.no