Understanding Insulation Coordination

Transcription

1 Alberta Power Industry Consortium & University of Alberta Professional Development Course Understanding Insulation Coordination Organized By Alberta Power Industry Consortium & University of Alberta AESO AltaLink ATCO Enmax Epcor FortisAlberta Instructed By Dr. David Peelo May 13 & 15, 2014 Calgary & Edmonton, Alberta, Canada

2

3 Abstract Insulation coordination is fundamental to the design and operation of power systems. In turn, overvoltage limitation is an important part of insulation coordination. Means for overvoltage limitation have evolved from rod gaps through gapped type surge arresters to the metal oxide surge arresters of today. This course will discuss the basic principles of insulation coordination and will then cover arrester application and selection. The course is intended for APIC company staffs who are not specialized in insulation coordination but want to gain an adequate understanding on the basics of insulation coordination and surge arrester application. It is hoped that the knowledge will help them to better appreciate the challenges and requirements of insulation coordination and equipment protection, which, in turn, will facilitate the execution of engineering projects involving multiple technical subjects including insulation coordination for T&D equipment. i

4 Confidentiality Requirement This course material was prepared by the University of Alberta for the ultimate benefit of the Alberta Power Industry Consortium members (hereinafter called SPONSORS ). It may contain confidential research findings, trade secrets, proprietary materials (collectively called Proprietary Information ). The term Proprietary Information includes, but is not limited to, plans, drawings, designs, specifications, new teaching materials, trade secrets, processes, systems, manufacturing techniques, model and mock-ups, and financial or cost data. The document is made available to the sponsors only. The Sponsors will use all reasonable efforts to treat and keep confidential, and cause its officers, members, directors, employees, agents, contractors and students, if any, ( Representatives ) to treat and keep confidential, and Proprietary Information in the document and the document itself. This course material shall not be disclosed to any third party without the consent of the Alberta Power Industry Consortium. Disclaimer This document may contain reports, guidelines, practices that are developed by the University of Alberta and the members of the Alberta Power Industry Consortium (APIC). Neither the APIC members, the University of Alberta, nor any of other person acting on his/her behalf makes any warranty or implied, or assumes any legal responsibility for the accuracy of any information or for the completeness or usefulness of any apparatus, product or process disclosed, or accept liability for the use, or damages resulting from the use, thereof. Neither do they represent that their use would not infringe upon privately owned rights. Furthermore, the APIC companies and the University of Alberta hereby disclaim any and all warranties, expressed or implied, including the warranties of merchantability and fitness for a particular purpose, whether arising by law, custom, or conduct, with respect to any of the information contained in this document. In no event shall the APIC companies and the University of Alberta be liable for incidental or consequential damages because of use or any information contained in this document. Any reference in this document to any specific commercial product, process or service by trade name, trademark, manufacture, or otherwise does not necessarily constitute or imply its endorsement or recommendation by the University of Alberta and/or the APIC companies. ii

5 About the Alberta Power Industry Consortium: The Alberta Power Industry Consortium consists of six Alberta utility companies (AESO, AltaLink, ATCO, Enmax, Epcor and FortisAlberta) and the University of Alberta. Established in the fall of 2007, its goal is to bring Alberta power companies together, with the University of Alberta as the coordinating organization, to solve technical problems of common interest, to produce more power engineering graduates, to support the professional development of current employees, and to promote technical cooperation and exchange in Alberta s power utility community. iii

6 About the instructor: Dr. David Peelo, P.Eng., is an independent consultant. He graduated from University College Dublin in 1965 and worked first for the ASEA Power Transmission Products Division in Sweden. He joined BC Hydro in 1973 where he rose to the position of specialist engineer for switchgear and switching and took early retirement in 2001 to pursue a second career as a consultant. In 2004 the Eindhoven University of Technology awarded him a PhD for original research on current interruption using air-break disconnect switches. He has published more than 60 papers on switching and surge arrester application and is actively involved with IEEE, CIGRE and IEC. He is a past convener of IEC Maintenance Team 32 Inductive Load Switching and IEC Maintenance Team 42 Capacitive Current Interrupting Capability of Disconnectors and a member of the Canadian IEC National Committees for switching equipment and for surge arresters. He is the author of a textbook on current interruption transients calculation and a coauthor of a textbook on switching in power systems both due for publication in iv

7 Course Outline 1. Basic principles of insulation coordination Overvoltages in power systems The concept of insulation levels The role of surge arresters Simplified approach to insulation coordination 2. Power system overvoltages Origins of overvoltages: lightning, switching and temporary overvoltages Characteristics of overvoltages in power systems Traveling waves 3. Metal oxide surge arresters Evolution of overvoltage protection devices The design of metal oxide surge arrester Characteristics and applications of metal oxide surge arrester 4. Surge arrester standards Evolution of surge arrester standards IEC versus IEEE standards Which is the recommended standard to follow? 5. Arrester application in substations Types and characteristics of incoming surges Distance effects Arrester selection v

8 vi

9 APIC 2014 Understanding Insulation Coordination David Peelo DF Peelo & Associates Ltd. DF Peelo & Associates 2014 APIC 2014 Basic principles of insulation coordination Overvoltages in power systems Concept of insulation levels Role of surge arresters Simplified approach to insulation coordination DF Peelo & Associates

10 APIC 2014 What is insulation coordination? IEEE definition: The selection of insulation strength consistent with expected overvoltages to obtain an acceptable risk of failure IEC definition: The selection of dielectric strength of equipment in relation to voltages which can appear on systems for which the equipment is intended and taking into account the service environment and the characteristics of available protective devices DF Peelo & Associates

11 Types of overvoltages and their causes: 5 Insulation types and strengths: External insulation Internal insulation Self-restoring insulation Non self-restoring insulation 6

12 APIC 2014 DF Peelo & Associates Internal insulation - transformers 8

13 Insulation strengths: Atmospheric air: statistical based on overvoltage type and insulation terminal configuration Liquids and gases: deterministic or statistical based on type of equipment 9 Overvoltage limitation: Air gaps: dependent on electrode configuration and type of overvoltage Silicon carbide surge arresters: voltage limited rating basis Metal oxide surge arresters: energy absorption limited rating basis 10

14 Basic definitions: Terms that we will encounter throughout the various part of the course 11 Nominal voltage of a system: A suitable approximate value of voltage used to designate or identify a system: example 230 kv system. Highest voltage of a system: The highest value of operating voltage which occurs under normal operating conditions at any time and at any point in the system: 245 kv on 230 kv system. Highest voltage for equipment (Um): The highest rms value of phase-to-phase voltage for which the equipment is designed in respect of its insulation as well as other characteristics which relate to this voltage in the relevant equipment Standards 12

15 Isolated neutral system: A system where the neutral point is not intentionally connected to earth (ground) Solidly earthed (grounded) neutral system: A system where the neutral point or points are directly connected to earth (ground) Impedance earthed (grounded) system: A system whose neutral points are earthed (grounded) through impedances to limit fault current 13 Earth (ground) fault factor: The ratio of the highest power frequency voltage on an unfaulted phase during a line-to-earth (ground) fault to the phase-to-earth (ground) power frequency voltage without the fault Overvoltage: Voltage, between one phase and earth (ground) or between two phases, having a crest value exceeding the corresponding crest of the highest voltage of the system. Overvoltages may be classified by shape or duration as either temporary or transient 14

16 Performance criterion: The basis on which the insulation is selected so as to reduce to an economically and operationally acceptable level the probability that the resulting voltage stresses imposed on the equipment will cause damage to the equipment insulation or affect continuity of service. This criterion is usually expressed in terms of an acceptable failure rate of the insulation configuration 15 Examples of performance criteria Transmission lines: 2 lightning flashovers per 100 km-years exposure 1 switching surge flashover per 100 switching operations Substations: Generally, station reliability criteria is 10 times line criteria Also, transformers and other non-self restoring insulation equipment arrester protected due to failure consequences Air insulated stations: MTBF of 50 to 200 years Gas insulated stations: MTBF up to 800 years due to failure consequences 16

17 Withstand voltage: The value of the test voltage to be applied under specified conditions during which a specified number of disruptive discharges may be tolerated: conventional assumed withstand voltage: number of disruptive discharges tolerated is zero and corresponds to a probability of withstand Pw = 100% statistical withstand voltage: number of disruptive discharges tolerated is related to a specified withstand probability and is specified at Pw = 90% 17 Representative overvoltages (Urp): Overvoltages assumed to produce the same dielectric effect on the insulation as overvoltages of a given class occurring in service due to various origins Coordination withstand voltage (Ucw): For each class of voltage, the value of the withstand voltage of the insulation configuration, in actual service conditions, that meets the performance criterion 18

18 Insulation coordination procedure 19 Required withstand voltage (Urw): The test voltage that the insulation must withstand in a standard withstand test to ensure that the insulation will meet the performance criterion when subjected to a given class of overvoltages in actual service conditions and for the whole service duration Standard withstand voltage (Uw): The standard value of the test voltage applied in a standard withstand test. It is the rated value of the insulation and proves that the insulation complies with one or more required withstand voltages 20

19 Coordination factor (Kc): The factor by which the value of the representative overvoltage must be multiplied in order to obtain the value of the coordination withstand voltage Atmospheric correction factor (Ka): The factor to be applied to the coordination withstand voltage to account for the difference between the average atmospheric conditions in service and the standard reference atmospheric conditions. The factor applies only to external insulation 21 Standard reference atmospheric conditions: The standard reference atmospheric conditions are: temperature to = 20 C pressure bo = kpa (1013 mbar) absolute humidity hao = 11 g/m³ 22

20 Safety factor (Ks): The factor to be applied to the coordination withstand voltage, after application of the atmospheric correction factor (if required), to obtain the required withstand voltage, accounting for all the differences between the conditions in service and those in the standard withstand test Test conversion factor (Kt): The factor to be applied to the required withstand voltage, in the case where the standard withstand voltage is selected of different shape, so as to obtain the lower limit of the standard withstand test voltage that can be assumed to prove it 23 Insulation coordination procedure 24

21 25 Insulation coordination simplified approach 26

22 Source IEC Source IEC

23 References: 1. IEC and IEEE insulation coordination standards noted earlier. 2. IEC Surge arresters Part 4: Metal-oxide surge arresters without gaps for a.c. systems. 3. IEC Surge arresters Part 5: Selection and application recommendations. 4. Insulation Coordination for Power Systems (Book), A.R. Hileman, Marcel Dekker, Inc Insulation Coordination for High-Voltage Electric Power Systems (Book), W. Diesendorf, Butterworth & Co Surge arrester manufacturer websites: ABB and Siemens in particular have comprehensive selection and application guides. 29 Power system overvoltages Lightning, switching and temporary overvoltages Characteristics of overvoltages in power systems Travelling wave basics How do surge arresters work? 30

24 Spectrum of power system overvoltages: 31 Overvoltage class: transient 32

25 Overvoltage class: low frequency continuous and temporary 33 Temporary overvoltages: causes and characteristics 1. Ferranti effect: Steady state voltage (V 2 ) at the open receiving end of an uncompensated line (no shunt reactors) is always higher than the voltage (V 1 ) at the sending end; occurs because the (leading) capacitive charging current flows through the series inductance of the line V V 2 = 1 1 cos(bl) where L is the line length in km and B the phase constant (7.2 degrees/100 km at 60 Hz) 34

26 Ferranti effect V 1 and V 2 sending and receiving end voltages. 1: no compensation 2: 50% series capacitor compensation 3: 50% series capacitor and 70% shunt reactor compensation Faults: A line to ground fault represents a typical example of a temporary undamped overvoltage that may be sustained on the unfaulted phases for up to hundreds of milliseconds The magnitude of the overvoltages on the unfaulted phases depends on the shift of the electrical neutral caused by the fault earth fault factors 36

27 SLG EFFs: Maximum line to ground voltage at any fault location and under any fault condition for effectively grounded system. Numbers on curves are maximum line to ground voltages on any phase in percent of line-to-line voltage. 37 Earth fault TOVs: Category EFF (pu) Duration Grounded Network, high SC* Line radial line, low SC Resonant grounded Network or mesh Long radial line Isolated/Ungrounded Distribution with O/H lines Industrial with cable s 1 s 8 hours 2 days 1 s s (with fault clearing) SC short circuit 38

28 3. Load rejection: Load rejection occurs when a remote circuit breaker on a transmission line carrying a substantial load is opened due to system condition or an error. A voltage rise follows because: the reduced current means a lower voltage drop across the internal system impedance generators tend to overspeed to produce higher voltages 39 Load rejection overvoltages: Category Magnitude Duration (s) Load rejection in a system generator-transformer steam turbine hydro > Note: Load rejection may be combined with the occurrence of a fault producing a combined effect 40

29 4. Resonance: Resonance in power systems can take two forms: linear resonance when inductive and capacitive elements in series form a series resonant circuit; for example, an unsaturated transformer and a shunt capacitor bank ferroresonance when saturated iron-cored inductive and capacitive elements in series form a series circuit; for example, the grading capacitors on an open circuit breaker in series with a magnetic PT 41 Overvoltages associated resonance situations can be very significant and rapid intervention is necessary Resonance Magnitude (pu) Duration (s) Unsaturated phenomena < Saturated phenomena Coupled circuits

30 5. Other TOV situations: TOVs due to Magnitude (pu) Duration Line energization and re-energization 1.5 < 1 s Stuck breaker pole 2 Steady state Backfeeding 2 Seconds 43 Interesting facts about TOVs: 1. TOVs DO NOT CONTRIBUTE TO INSULATION DESIGN HOWEVER 2. TOVs DETERMINE THE VOLTAGE RATINGS OF THE METAL OXIDE ARRESTERS TO BE APPLIED ON THE SYSTEM 44

31 Slow front switching surges: Line switching Making and breaking reactive currents TRVs across circuit breakers 45 Line switching: Due to travelling wave effects, line switching can result in overvoltages of significant magnitude. Worst case is re-energization of a line with a full DC trapped charge The resulting switching surges are slow-front with times to peak in the order of hundreds of microseconds 46

32 Switching surge overvoltage distribution with various control measures 47 Voltage profile on line 48

33 Typical switching overvoltage magnitudes: Condition Magnitude (pu) Energizing discharged line Re-energizing with no overvoltage control Re-energizing with pre-insertion resistors Re-energizing with controlled closing Making and breaking reactive currents: Inductive current switching transformer magnetizing current shunt reactors at EHV, HV and MV Capacitive current switching shunt capacitor banks lines and cables 50

34 51 Fast front overvoltages: Fast front overvoltages are those due to lightning and have the dimensioning role at system voltages below EHV levels (at EHV levels switching surges have the dimensioning role) The overvoltages are characterized by fast rise times of 0.1 to 20 µs and associated in the range 5 to 200 ka. 52

35 Very fast front overvoltages: Very fast front overvoltages are generally associated with gas insulated switchgear (GIS) applications and in particular with the live operation of disconnect switches The overvoltage are not dimensioning with respect to the insulation coordination of GIS; however, present standards for disconnect switches now incorporate very severe tests to demonstrate that live operation of the switches does not compromise the GIS insulation structure 53 What is lightning? Lightning is the breakdown of atmospheric air; in dry air at sea level breakdown voltage is 30 kv/cm but, at higher altitudes in a region filled with water droplets, it is about onethird this value Lightning strikes are not instantaneous starting as local events and progressing in steps to the other electrode which could be earth (ground) or another cloud Lightning strikes are very statistical in terms of current magnitude, number of strokes per strike and striking point 54

36 Global electric circuit 55 56

37 57 58

38 59 60

39 61 Travelling wave basics: Fundamental circuit for travelling waves V i = incident voltage Z i = surge impedance of V i Z t = surge impedance of V t V r = reflected voltage = αv i where α = (Z t Z i )/(Z t + Z i ) V t = transmitted voltage = βv i where β = 2Z t /(Z t + Z i ) 62

40 63 Voltage waves: Case Surge Impedance V r Reflected Voltage V t Transmitted Voltage No change Z i = Z t 0 V i Open circuit Z t >> Z i V i 2 V i 50% reduction Z t = 0.5 Z i V i 0.67 V i Short circuit Z t = 0 -V i 0 64

41 Surge current (directional): I = V/Z Case Surge Impedance I r Reflected Current I net (Z i side) No change Z i = Z t 0 I i Open circuit Z t >> Z i I i 0 50% reduction Z t = 0.5 Z i I i 0.67 I i Short circuit Z t = 0 -I i 2 I i 65 CO operation on 100 km unloaded transmission line Sending end Receiving end 66

42 We can now consider the principle of overvoltage surge protection using arresters: V1 V2 Z1 I3 Z2 V2 =V3 V3 Z3 67 We can write: V 2 = V 3 2ZT = Z1 + Z T V 1 where Z V 2 T = 2V Z 1 Z2Z3 = Z + Z 2 Z ( 1+ Z ) ( ) 2 Z3 Z2 1+ Z2 Z3 68

43 69 If no arrester, Z 3 = If Z 2 also as at open-ended line or transformer V 2 V V = = (voltage doubling) Z Z 2V Z V V + = = 70 We can also write: If ideal arrester (short-circuit), Z 3 = 0 and Z I = 2V Z V Z V I = = = Z Z Z Z Z Z Z Z 2V

44 We now have the two extreme points on the circuit load line: 2V1*Z2 Z1+Z2 V Z ( 2V Z ) 2 2 = 1 1 I3 Z1 + Z 2 V2 I3 2V1 Z1 71 Add the arrester VI-characteristic: 2V 1*Z 2 Z 1+Z 2 V ar Z2 = Z1 + Z 2 ( 2V Z I ) 1 1 ar V 2 V ar Arrester VI-characteristic I 3 Iar 2V 1 Z 1 72

45 Load line for transformer case: Z 2 = 2V 1 V ar = 2V Z I 1 1 ar V 2 V ar Arrester VI-characteristic I ar I 3 2V 1 Z 1 73 Question: how does travelling wave theory treat the effect of a surge arrester in clamping the voltage to the protective level at its location? 74

46 APIC 2014 Metal oxide surge arresters Evolution of overvoltage protective devices Design of metal oxide surge arresters Characteristics and application of metal oxide surge arresters DF Peelo & Associates History of surge arresters: 1898 rod gap design 1908 electrolyte arrester with non-linear resistive element to limit follow current and enable arc interruption 1930 silicon carbide arresters with silicon carbide resistive elements in series with gaps 1957 silicon carbide arresters with current limiting gaps 1976 metal oxide surge arresters with extreme non-linear characteristics 76

47 APIC 2014 Structure of metal oxide surge arresters: Consist simply of zinc oxide varistor disks of varying sizes connected in series Zinc oxide varistors are ceramic semiconductor devices with very non-linear voltage-current characteristics and very good energy absorption capability DF Peelo & Associates APIC 2014 Zinc oxide disks are about 90% zinc oxide (ZnO) by weight plus various other metal oxides to provide specific properties for example: bismuth + CaO, CoO, BaO, SrO, MnO: non-linearity of the VI characteristic K 2 O: inhibits grain growth Cr 2 O 3 : enhances thermal stability Ga 2 O 3 : increases exponent (α) of VI characteristic Sb 2 O 3 : grain grown enhancer DF Peelo & Associates

48 APIC 2014 Varistor microstructure key elements: Zinc oxide grains Bismuth-rich intergranular layer Spinel grains DF Peelo & Associates Varistor microstructure: Source: ABB 80

49 VI characteristic: three conductive regions 2.5 Per unit peak rated voltage Region 1 Region 2 Region 3 LIPL SIPL Rated voltage MCOV 25 C 150 C Resistive leakage current 0 1E Current (A) 81 APIC 2014 Region 1: Region 2: Region 3: pre-breakdown region is the low current region associated with steady state operation; material resistivity temperature dependent with negative temperature coefficient breakdown region is the highly non-linear region associated with TOVs and switching surges; very small temperature dependence and exponent α = 30 to 50 high current region is the region associated impulse currents > 1 ka due to lightning; non-linearity much less than in the breakdown region DF Peelo & Associates

50 Current conduction: equivalent circuit 83 APIC 2014 Pre-breakdown: Breakdown: High-current: in this region current conduction is determined by the high resistance associated with grain boundaries with a significant temperature dependence at a certain applied voltage across grain boundaries, the intergranular layer resistance drops allowing a large increase in current; energy is therefore being absorbed in the intergranular layer the zinc oxide grain resistance dominates in this region DF Peelo & Associates

51 APIC 2014 Comparison: metal oxide versus gapped silicon carbide arresters fundamentally different: Metal oxide arresters are rated on the basis of ability to absorb energy and maintain thermal stability at rated voltage followed by MCOV Gapped silicon carbide arresters are rated on the basis of ability to reseal interrupt follow current after discharging a lightning or switching surge DF Peelo & Associates Active gaps SiC resistors Grading resistors and capacitors Courtesy of ABB SiC gapped arrester ZnO arrester 86

52 87 Rated voltage and energy considerations: metal oxide arresters are energy (temperature) limited 88

53 Effects of aging: over lifetime 89 IEC specifies five (5) line classes: 7 Specific energy (kj/kv U r) Line class 1 6 Line class 2 5 Line class 3 Line class 4 4 Line class U res /U r Former IEC rating basis now changed to a thermal energy rating and a repetitive charge transfer rating 90

54 Former IEC rating basis now changed to a thermal energy rating and a repetitive charge transfer rating 91 APIC 2014 Thermal Energy Rating W th Maximum specified energy given in kj/kv of U r that may be injected into an arrester within 3 minutes time duration without causing thermal runaway. This rating is verified in the revised operating duty test above injection preceded by conditioning consisting of two high current impulses only. The rating is strictly thermal and no longer relates rated energy to protective levels; temperature coefficient for metal oxide material 0.33 ⁰C/J/cm 3. DF Peelo & Associates

55 Repetitive Charge Transfer Rating Q rs Maximum specified charge transfer capability of an arrester in the form of a single event or group of surges that may transferred through an arrester without causing mechanical failure or unacceptable electrical degradation to the MO resistors. The charge is calculated as the absolute value of current integrated over time. This is the charge that is accumulated in a single event or a group of surges lasting for no more than 2 seconds and which may be followed by a subsequent event at a time interval not shorter than 60 seconds. 93 TOV versus time curves: TOV factor ktov With prior energy Without prior energy MCOV Time (s) 94

56 Overvoltages: types and shapes Source: IEC APIC 2014 Energy absorption: TOVs: system is a voltage source lightning and switching surges: system is a current source For energy absorption studies, the minimum VI characteristics should be used DF Peelo & Associates

57 Overvoltages and associated energies: Type/Cause Magnitude Waveshape Energy kj/kv rated voltage Atmospheric Overvoltages: Lightning strikes or induced by lightning (multiple strikes may occur) > 5 pu Front: 1 6 µs 50%: 50 µs 0.5 Switching Overvoltages: Line autoreclosing, switching capacitor banks, shunt reactors, issue more at EHV 2 4 pu Front: µs 50%: µs 3 5 Temporary Overvoltages: SLG faults in ungrounded systems, Ferranti effect, loading shedding pu Power frequency (may be distorted) Very high (fast remedial action required) 97 Protective and related characteristics: 192 kv arrester Voltage (kv peak) Rated voltage 272 kv peak MCOV 218 kv peak LIPL 470 kv peak SIPL 384 kv peak E-05 1E E+05 Current (A) 98

58 APIC 2014 Because we are dealing with a VI characteristic, the respective protective levels are defined at standard normal discharge currents. For lightning protection: System voltage Standard nominal discharge current* (8/20 µs) Distribution 5 ka Sub-transmission up to 72.5 kv 5 or 10 ka 72.5 to 245 kv 10 ka 245 kv and up 10 ka or 20 ka 500 kv 20 ka * values also define so-called Arrester Classification DF Peelo & Associates APIC 2014 For switching surges: Arrester classification Peak currents* (A) 20 ka 500 and ka 250 and ka distribution class 125 and 500 * in switching impulse residual voltage test, higher numbers used to define the protective level DF Peelo & Associates

59 VI characteristic: switching and lightning currents Voltage (pu of 10 ka value) /10micros 1.3 8/20micros /90micros Current (A) 101 Extract from ABB surge arrester catalog: Guaranteed protective characteristics Recommended for system voltage Rated voltage Max. cont. operating voltage (COV) MCOV as per ANSI tests kv rms kv rms kv rms kv rms EXLIM P-A and P-B TOV capability for 1 s kv rms 10 s kv rms 1 ka kv crest Maximum residual voltage with current wave Switching surges 8/20 µs 2 ka kv crest 3 ka kv crest 5 ka kv crest 10 ka kv crest 20 ka kv crest 40 ka kv crest

60 Energy absorption: TOVs: system is a voltage source lightning and switching surges: system is a current source For energy absorption studies, the minimum VI characteristics should be used 103 Selection procedure: Electrical characteristics Rated voltage and MCOV Nominal discharge current Thermal energy rating W th Repetitive charge transfer rating Q rs Lightning and switching surge protection levels Mechanical characteristics Strike and creepage distances Short-circuit withstand Seismic and tensile loads 104

61 Selecting the housing: briefly the following applies Arresters are self-protecting and the housings do not require the same withstand capability as other station equipment Creepage distances should be the same as for all station equipment Pollution or exposure to salt contamination can require longer creepage distances but consideration should also be given to using higher rated voltage arresters or more appropriate arrester designs 105 Short-circuit testing is required on all arrester types over a range of current; for example, an arrester with a rated short-circuit current rating of 50 ka would also be tested at 25 ka, 12 ka and 600 ka Selection of the short-circuit current rating should be based on the maximum expected short-circuit current magnitude at the arrester location 106

62 Insulation coordination iterative process 107 Station layout considerations: Location of the arrester relative to the protected equipment is important and will be discussed later Location of the arrester relative to other energized equipment on the same phase also requires attention Question: how does the protected equipment know that the arrester is ahead of it? 108

63 Testing: Type testing is covered in great detail in IEC as is basic routine testing As noted earlier routine testing is at least as important as type testing; in making comparisons between arresters from different manufacturers, compare the tests performed on each block before considering award 109 Surge arrester standards Evolution of surge arrester standards IEC versus IEEE standards Which is the recommended standard to use 110

64 IEEE and IEC differ in approaches: IEEE: IEC: rated voltage is the duty cycle rating which is a holdover from gapped silicon carbide arresters and has no relationship to absorbed energy rated voltage is a TOV to be withstood following absorption of defined energy under specific circumstances 111 IEEE rating basis: from IEEE C duty cycle voltage rating: The designated maximum permissible voltage between its terminals at which an arrester is designed to perform its duty cycle Duty-cycle test The purpose of the duty-cycle test is verify that the arrester can withstand multiple lightning type impulses without causing thermal instability or dielectric failure. 112

65

66 Former IEC rating basis now changed to a thermal energy absorption rating 115 Importance of routine testing: Absolutely no redundancy in metal oxide surge arresters: one bad block and the arrester will fail guaranteed! Routine testing has therefore equal or arguably greater importance than type testing 116

67 APIC 2014 Arrester application in stations Type and characteristic of incoming surges Distance effects Arrester selection DF Peelo & Associates Stations: insulation coordination Transmission lines are normally shielded except where the soil resistivity is such that it is not possible to achieve low tower footing impedances at a justifiable cost Lines, however, are typically shielded about 1 km or more out from the station in order to limit the occurrence, magnitude and steepness of incoming surges Two types of lightning related failures are of interest: shielding failures where a lightning strike occurs directly to the line conductor and back flashover (also known as a backflash) where the tower or ground wire is struck and a flashover then occurs to a phase conductor. 118

68 119 Lightning strike to a phase conductor Only close-in strikes are of interest because of the damping effect of corona on the traveling wave. 120

69 Travelling wave damping due to corona Conductors Zo(ohms) Kc (km-kv/us) Single Conductor Bundle or 4 Conductor Bundle or 8 Conductor Bundle Voltage Surge Distortion Due to Corona 0.5 km 1.0 km Voltage (V) km 3.0 km 4.0 km Time (us) V@ 3.0 km V@ 4.0 km V@ 0.5 km V@ 1.0 km V@ 2.0 km Electrotek Concepts TOP, The Output Pro 122

70 Back flashover 123 Lightning currents Shielding failure: 4 ka or greater at 89% probability: 20 ka or greater at 80% probability. Back flashover: 20 ka or greater at probability of 80%; 90 ka or greater at probability of 5% Median current: 32 ka 124

71 Station insulation coordination: determine the representative incoming lightning overvoltage. Two approaches are possible: 1. Comprehensive approach: requires a computer study because of the complexity of the calculations. 2. Deterministic approach: a more general approach but does take into account the lightning stress severity of the station under study. 125 Comprehensive approach flow diagram: 126

72 Comprehensive approach - need to consider: Statistical (stochastic) variation of the lightning flash and stroke parameters Dependence of the flash strike point on these parameters Response of the line to the lightning flash including the tower footing impedance current dependence Propagation of the surge from the strike point to the station and the deformation effect of corona 127 Station insulation coordination: deterministic approach 1. Determine the lightning performance of incoming transmission line based on parameters discussed earlier ground flash density, line height etc. 2. Determine steepness of incoming surge based on the above and system fault tolerance. 3. Determine the protective zones based on selected arrester locations; may need a number of iterations and this type of study is usually done using EMTP simulations. 4. Transformer protection tends to get most attention but open breaker protection is also a consideration 128

73 129 Insulation Coordination

74 Basic case: arrester at transformer only CASE1>B-B1 (Type 1) Voltage (V) VB-B1 0 V-CB V-JTN VSA VTX Time (ms) 131 Entrance Bus Voltages : Normal, Wih Ccvt, With SA2 & Ccvt Voltage (V) Time (us) Electrotek Concepts Vb: normal Vb: With Ccvt Vb: With SA2 & Ccvt TOP, The Output Processor 132

75 Transformer Voltages: Normal, With Ccvt, With SA2 & CCVT Voltage (V) Time (us) Vtxc: normal Vtx: With Ccvt Vtx: With SA2 & Ccvt Electrotek Concepts TOP, The Output Processor Vsa : Normal, With Ccvt, With SA2 & Ccvt Voltage (V) Time (us) Vsa: normal Vsa: With Ccvt Vsa: With SA2 & Ccvt Electrotek Concepts TOP, The Output Processor 134

76 10000 SA Currents: Normal, With Ccvt, With SA2 & Ccvt 8000 Current (A) Time (us) Isa: normal Isa: Ccvt Isa: SA2 & Ccvt Electrotek Concepts TOP, The Output Processor 135 Surge arrester application: Arrester selection recap Arrester location relative to protected equipment Clearances associated with arresters 136

77 Arrester selection based on: system voltage and grounding temporary overvoltages desired safety factor relative to the equipment withstand voltage but its physical location relative to the protected equipment is also a consideration (why?) 137 Simple estimation of protective distance: where U pl S L c Urp = Upl + = lightning impulse protective level of the arrester (kv) = steepness of incoming surge (kv/µs) = d 1 + d A + d 2 + d (m) SL 2 c = velocity of light (300 m/µs) 138

78 Example: 245 kv transformer, LIWL 850 kv 192 kv rated surge arrester, U pl 440 kv, 1000 kv/µs L = d 1 + d A + d 2 + d = d = d U rp = d = (7.1 + d) ( ) 139 U rp (kv) Distance d (m) Voltage at transformer 850* *

79 Let s take a look at what is really happening when a surge approaches a transformer protected at a certain distance by an arrester: Incoming surge steepness 1000 kv/μs Arrester with protective level of 800 kv installed at distance of 75 m from the transformer Remember the questions: how does traveling wave deal with arresters and their voltage limiting effect and how does the transformer know there is an arrester ahead of it? 141 Arrester at distance 75 m and surge 1000 kv/µs 142

80 APIC 2014 EMTP study result same case DF Peelo & Associates What conclusions can we draw? U rp = U pl + 2ST for U pl > or = 2ST U rp = 2U pl for U pl < 2ST where T is the travel time of the lightning surge L/c. Actual voltage at the transformer oscillates due the arrival and departure of the traveling waves but the maximum value is U rp. 144

81 Disadvantages of simplified method: 1. Does not take the capacitance of transformers or reactors into account 2. Real VI characteristic of the arrester is not used 3. Does not take the effect of the power frequency voltage into account 145 Switching surges within stations: Are man-made events and generally more an issue for lines than for stations; however, station equipment and clearances must be capable of withstanding the applicable rated switching surge withstand voltages and surge arresters must provide the corresponding overvoltage protection. Equipment standards fix a certain ratio between rated lightning and switching impulse voltages as we have seen earlier. Surge arresters are applied to limit switching overvoltages within stations for the switching of shunt capacitor banks. This is a special case usually requiring a computer study to determine energy requirements in particular but some generalization is possible. Key parameters are discussed in the following. 146

82 Clearances: two types of clearances to consider Station air clearances Surge arrester clearances to ground to energized equipment on same phase 147 Statistical nature of electrical breakdown 148

83 Statistical nature of electrical breakdown (cont/d) Applied voltage must exceed critical value (field strength) where cumulative ionization is possible Statistical time lag t s from application of voltage to the time of creation of the first free electron time lag decreases with increasing voltage Once electron found must further generate a streamer, then streamer converts to a spark formative time lag t f 149 Statistical nature of electrical breakdown (cont/d) V s must be high enough for ionization to occur and sustained for duration longer than the total time lag Total time lag t s +t f not the same for each voltage application same voltage waveform may or may not cause breakdown and therefore it is a probabilistic event Question: what does this mean for lightning impulses and switching surges? 150

84 Influencing factors on breakdown: Waveshape: breakdown voltage dependent on impulse voltage and its profile over time Gap configuration: more of an influence on switching impulses than lightning impulses Polarity: breakdown voltage lowest for positive impulses 151 Influence of waveshape: lightning impulses: Lightning impulses represented by a standard 1.2/50 µs waveshape Duration of voltage around the peak does not give enough time for leaders to develop and breakdown is dependent on streamers only Breakdown (aka sparkover) voltage V s given by: V s = E s d where E s is the electric field gradient and d is the gap length 152

85 Rod-plane gaps have the lowest breakdown voltages for all gap configurations; for positive standard lightning impulses for such gaps 1 to 10 m (IEC ): U 50RP = 530d U 50RP is the 50% probability of flashover voltage in kv crest d is the gap spacing in m 153 For other gap configurations such as rod-rod, U 50 can be as high as 700d (rod-plane and rod-rod gaps tend to represent two extremes) For rod-plane gaps up to 6 m and negative standard lightning impulses: U 50RP = 950d 0.8 and can be as low as 700d 0.8 for other gaps 154

86 U50 flashover value (kv crest) Rod-plane positive Rod-rod positive Rod-plane negative Gap length (m) 155 Important notes for lightning impulses: 1. Gap factors (to be discussed for switching impulses) are generally not directly applicable 2. Because positive impulses give the lowest breakdown values does not mean that negative impulses can be ignored; most lightning surges have negative polarity and internal insulation (e.g. in a transformer) has lower withstand for negative impulses 3. Above-noted equations apply at sea level 156

87 Influence of waveshape: switching impulses Switching impulses represented by a standard rise time of 250 µs and a time to half-value of 2500 µs Breakdown is dependent on gap configuration and hence the use of gap factor Breakdown also dependent on voltage rise time giving socalled U-curves 157 Rod-plane gaps: K = 1 U50 flashover value (kv crest) EdF Harbec-M enemeniis 1000 IEC M eek-craggs Gap length (m) 158

88 So what is the gap factor? The gap factor represents the influence of the electrode configuration Configuration Gap factor Rod-plane 1 Conductor-plane 1.15 Vertical rod-rod 1.4* * dependent on length of grounded rod (5 m in this case) 159 Gap factors for phase-to-phase configurations*: Configuration α=0.5 α=0.33 Ring-ring Crossed conductors Conductor-conductor Supported busbars (including fittings) * IEC

89 Lightning versus switching impulse: rod-plane and rodrod gaps: U50 flashover values (kv crest) 2500 IEC Meek-Craggs 2000 IEC*1.4 Meek-Craggs* Lightning Gap length (m) 161 U 50 flashover values (kv crest) IEC Meek-Craggs IEC*1.4 Meek-Craggs*1.4 Lightning Gap length (m) 162

90 Arrester minimum clearances phase to ground: Because arresters are self-protecting lower minimum clearances than station minimum clearances can be used Consult the appropriate manufacturer catalog or instruction manual for their recommended minimum clearances 163 Surge Arrester Manufacturer Recommended Minimum Clearances Phase-to-Ground Arrester rated voltage (kv rms) Minimum clearance phase-to-ground (mm) for Mfr 1 Mfr 2 Mfr 3 Mfr 4 Mfr

91 Surge arrester clearances to energized equipment on the same phase: Energized equipment on the same phase includes transformer bushings (arrester on an outrigger), CVTs or PTs, post insulators and so on Arrester voltage distribution determined by grading rings and proximity to other equipment on same phase really only an issue where pollution is a consideration; however what about tight fit scenarios? 165 Distribution system arrester practices: On overhead systems arresters applied at the cable riser pole from substation and possibly on pole-top transformers in very lightning prone areas On underground systems, arresters applied at the cable riser pole and at open points in the underground circuit; other measures may include paralleling arresters at the riser pole or applying a surge arrester at next pole with or without a groundwire 166

92 TWENTY-ONE QUESTIONS 1. What are the voltages considered in insulation coordination? 2. Which overvoltage determines the rated voltage of the surge arrester to be used? 3. Metal oxide material consists of zinc oxide plus various additives: which additive gives the non-linear VI characteristic? Which four points are used to define the VI characteristic of the required surge arrester? 5. Which quantities define arrester energy handling capability? 6. Which test defines the relationship between arrester rated voltage and rated thermal energy? 168

93 7. What determines the withstand capability of the arrester housing? 8. How is the required creepage distance for the arrester housing selected? 9. How should the required pressure relief capability for the arrester be selected? With respect to lightning, what representative quantity is used for station insulation coordination? 11. If the required protective levels are not met, what additional measures can be applied to resolve it? 12. Why is there no distance effect for switching surges? 170

94 13. What determines the required clearances for surge arrester installation? 14. Why is the separation distance arrester to transformer so important? 15. What is the effect on added capacitance on incoming surges? What are the characteristic waveshapes for lightning and switching surges? 17. What is a gap factor? 18. Which electrode configuration has the lowest switching surge withstand capability? 172

95 19. Rated lightning withstand capability for a transformer is deterministic: what is the expected probability of withstand? 20. Rated lightning withstand capability for a circuit breaker is probabilistic: what is the expected probability of failure? 21. More questions? 173 Thanks for listening! 174