60099-8 IEC:2011 – 11 – 3.13 follow current Ifollow the current immediately following an impulse through an EGLA with the power-frequency voltage as the source 3.14 specified long-t
EGLA identification
An EGLA shall be identified by the following minimum information, which shall appear on a nameplate permanently attached to the arrester:
• rated frequency in Hz, only if it is less than 48 Hz or larger than 62 Hz;
• rated short-circuit current I s in kA;
• manufacturer’s name or trade mark;
• serial number (at least for arresters for U m > 52 kV);
• lightning discharge capability (only charge value) in C; example: "8 C"
Information on required gap spacing including tolerances shall be given in an appropriate way, for example in the manual.
EGLA classification
EGLAs are categorized based on their nominal discharge currents and their ability to withstand high current impulses, as outlined in Table 1 They must also fulfill the test requirements and performance characteristics detailed in Table 3 Notably, these arresters do not have operational responsibilities for slow-front surges and power-frequency overvoltages.
Table 1 – EGLA classification – “Series X” and “Series Y“
Nominal discharge current (kA), 8/20 5 5 10 20 Nominal discharge current (kA), 2/20 5 10 15 20
NOTE 1 "Series X" corresponds to the classification used in IEC 60099-4 A nominal discharge current of
8/20 wave shape and a high current impulse of 4/10 wave shape are used in IEC and in IEEE standards
"Series Y" refers to the classification used in Japan for shielded line applications, with a specific focus on the wave shape 2/20 This specification applies to both the nominal discharge current and the high current impulse, tailored for this particular application.
NOTE 2 According to service conditions, other high current impulse values than those specified in this table may be applied
5 Standard ratings and service conditions
Standard rated voltages
Standard values of rated voltages (r.m.s values) are specified in Table 2 in equal voltage steps within specified voltage ranges
Table 2 – Steps of rated voltages (r.m.s values)
Range of rated voltages (kV) Steps of rated voltage (kV)
NOTE Other values of rated voltage may be acceptable, provided they are multiples of 6
Standard rated frequencies
The standard rated frequencies are 48 Hz to 62 Hz.
Standard nominal discharge currents
The standard nominal discharge currents for 8/20 or 2/20 shapes are: 5 kA, 10 kA, 15 kA and
Service conditions
Normal service conditions
EGLAs that meet the specified standard are designed for normal operation under specific service conditions, including an ambient air temperature range of –40 ºC to +40 ºC, an altitude not exceeding 1000 m, and an a.c power supply frequency between 48 Hz and 62 Hz The power-frequency voltage applied between the EGLA terminals must not exceed its rated voltage While mechanical conditions and wind speed are not specified, the equipment is expected to operate in environments with pollution from dust, smoke, corrosive gases, vapors, or salt, provided that the pollution level does not exceed "heavy" as defined in IEC/TS 60815-1.
Mechanical and environmental factors play a crucial role in service; however, the wide range of potential installation configurations prevents the establishment of standard values for items e) and f).
Abnormal service conditions
Surge arresters operating under non-standard conditions may need unique design, manufacturing, or application considerations The implementation of this standard in such abnormal service scenarios must be mutually agreed upon by both the manufacturer and the purchaser.
Insulation withstand of the SVU and the complete EGLA
Insulation withstand of the housing of the SVU
The housing of the SVU must endure a lightning impulse voltage of 1.4 times the residual voltage at the nominal discharge current for "Series X," and 1.13 times the residual voltage at high current impulse for "Series Y," with a minimum threshold specified.
1,3 times the residual voltage at nominal discharge current
NOTE The factor of 1,4 in case a) covers variations in atmospheric conditions and discharge currents up to three times the nominal discharge current.
Insulation withstand of EGLA with shorted (failed) SVU
The EGLA must demonstrate insulation withstand performance by enduring the designated switching impulse withstand voltage level of the system, even in the event of a short circuit caused by overloading or failure of the SVU.
– 14 – 60099-8 IEC:2011 b) the EGLA shall be able to withstand the maximum temporary overvoltages phase to ground for their maximum durations even if the SVU has been shorted due to overloading
Residual voltages
The measurement of residual voltages aims to determine the maximum residual voltages for a specific design across all defined currents and wave shapes This data is obtained from type test results and the maximum residual voltage associated with a lightning impulse current, as outlined in the routine tests published by the manufacturer.
The maximum residual voltage for any EGLA design is determined by multiplying the residual voltage from SVU sections, tested during type tests, by a specific scale factor, and adding the calculated inductive voltage drop across the SVU, gap, and connection leads This scale factor is the ratio of the declared maximum residual voltage, verified during routine tests, to the measured residual voltage of the sections under the same current and wave shape.
The residual voltage of the EGLA at nominal discharge current and high current impulse, when multiplied by the specified factor in section 6.1.1, must remain below the minimum flashover voltage of the insulator assembly being protected.
High current duty
The capability of the SVU for discharging operations shall be demonstrated by injecting two high current impulses.
Lightning discharge capability
Metal-oxide resistors must demonstrate their ability to endure lightning discharges with current waveforms lasting several tens of microseconds for shielded line arresters and several hundreds of microseconds for unshielded line arresters Additionally, the testing will assess the impact of multiple lightning strikes on these resistors.
Short-circuit performance of the SVU
The manufacturer must specify the short-circuit rating of the SVU, ensuring that the short-circuit currents do not lead to violent shattering of the unit, and that any open flames will self-extinguish within a specified timeframe.
The gap is not included in the short-circuit tests for the SVU, and its short-circuit performance must be verified independently It is essential that the gap maintains its mechanical integrity when exposed to the rated short-circuit current of the EGLA, and it should not experience a reduction in sparkover voltage.
Mechanical performance
The EGLA must be installed on transmission towers or poles, demonstrating mechanical performance capable of withstanding tensile, bending, and vibration loads caused by wind pressure, conductor vibration, abnormal installation loads, and moisture ingress.
The applicable values of tensile and bending loads shall be agreed between the manufacturer and the purchaser
The SVU shall be able to withstand the vibration load to be expected in service
NOTE The complete EGLA including gap assembly and mounting structure should be able to withstand at least the same mechanical stress
Weather aging of SVU
The SVU must be able to withstand the environmental stress expected in service
Environmental tests using accelerated procedures confirm that the sealing mechanism and exposed metal combinations of the SVU remain unaffected by environmental conditions Additionally, SVUs with polymer (composite and cast resin) housings must also demonstrate resistance to UV radiation.
NOTE A revision of the UV test is currently under consideration by Cigré WG D1.14.
Internal partial discharges
The level of internal partial discharges in the SVU in the tests according to 9.1 and 10.3 shall not exceed 10 pC.
Coordination between insulator withstand and EGLA protective level
Effective coordination between the flashover characteristics of the insulator assembly and the sparkover voltage of the EGLA, under both front-of-wave and standard lightning impulses, is essential Additionally, it is crucial to demonstrate the residual voltage of the EGLA at nominal discharge current, and for “Series Y” arresters, at high current impulse.
Any sparkover operation for lightning impulse voltage shall occur in the external series gap of the EGLA, without causing any flashover of the insulator assembly to be protected
• for "Series X": 1,4 times the residual voltage at the nominal discharge current according to
• for "Series Y": 1,13 times the residual voltage at high current impulse, but not less than
The residual voltage at nominal discharge current must be less than the value calculated as \( U_{50, \text{Insulator}} - X \cdot \sigma \), where \( \sigma = 0.03 \) and \( X \) is a mutually agreed value between the manufacturer and user, with a recommended value of \( X = 2.5 \).
Follow current interrupting
Follow current interrupting operation of the EGLA under wet and polluted conditions shall be demonstrated by a test procedure which takes these operating conditions into account
Performing a follow current interrupting test is mandatory, either as a type test according to
8.8 or as an acceptance test according to 10.6.
Electromagnetic compatibility
Arresters are not sensitive to electromagnetic disturbances, and therefore no immunity test is necessary
In normal working operating conditions, the EGLA shall not emit significant disturbances
A radio interference voltage test (RIV) shall be applied as an acceptance test to the complete
EGLA (see 10.4) The maximum radio interference level of the EGLA energized at the maximum continuous phase to ground system voltage (U s /√3) shall not exceed 2 500 àV.
End of life
On request from users, each manufacturer shall give enough information so that all the arrester components may be scrapped and/or recycled in accordance with international and national regulations
Measuring equipment and accuracy
The measuring equipment shall meet the requirements of IEC 60060-2 and IEC 60099-4 The values obtained shall be accepted as accurate for the purpose of compliance with the relevant test clauses
All power-frequency voltage tests should utilize an alternating voltage with a frequency ranging from 48 Hz to 62 Hz, ensuring the waveform is approximately sinusoidal unless specified otherwise.
Test samples
For each test item, the complete test sequence must be conducted on the same test sample, as specified The number of test samples is detailed in Table 3 These test samples should be new, clean, fully assembled, and arranged to replicate the conditions in which they will be used.
When conducting tests on sections or units, it is essential to adhere to specific criteria: (a) define the ratio of the rated voltage of the complete EGLA to that of the section or unit as \( n \); (b) ensure that the volume of the resistor elements used for testing does not exceed the minimum volume of all resistor elements in the complete EGLA divided by \( n \); and (c) set the reference voltage \( U_{\text{ref}} \) of the SVU for the test section to be equal to the minimum reference voltage of the SVU of the EGLA divided by \( n \).
The SVU of the test section must exceed the minimum reference voltage of the complete EGLA divided by the factor n; if it does, n should be correspondingly reduced Conversely, if the SVU of the test section falls below this minimum reference voltage, the test section cannot be utilized.
The factor n of the test samples shall be recorded in the test report
General
Table 3 identifies the type tests that shall be performed on the complete EGLA or on components of the EGLA
Table 3 – Type tests (all tests to be performed without insulator assembly)
Test item Number of test samples EGLA Section
Section of SVU Clause number
1.1 Housing withstand test of SVU
1.2 EGLA withstand test with failed SVU
3 Standard lightning impulse sparkover test a) 1 Test 8.4
4 High current impulse withstand test 3 Test 8.5
5 Lightning discharge capability test 3 Test 8.6
6 Short-circuit tests 4 or 5 Test 8.7
7 Follow current interrupting test b) 1 Test c) Test c) 8.8
The weather aging test is a mandatory procedure that must be conducted if it has not been performed as an acceptance test according to sections 10.5, 10.6, and 10.7 This test can be carried out on either a complete EGLA or a section of it, as referenced in section 8.8.2 Additionally, a vibration test is required to be performed on one complete SVU, as detailed in section 8.9.2.1.
Insulation withstand tests on the SVU housing and on the EGLA with failed
General
The tests evaluate the lightning impulse withstand voltage of the SVU housing in dry conditions, as well as the EGLA's ability to withstand the maximum anticipated switching surges and power-frequency overvoltages in wet conditions.
SVU had failed and is shorted.
Insulation withstand test on the SVU housing
This test demonstrates the dielectric withstand capability of the external housing of the SVU against lightning impulse voltages
The SVU housing shall be subjected to a standard lightning impulse voltage dry test according to procedure A or B in 20.1 of IEC 60060-1
The test will be conducted on the SVU housing that experiences the highest specific voltage stress per unit length Additionally, the non-linear metal-oxide resistors will be either removed or substituted with insulating material components.
Fifteen consecutive impulses at the test voltage value shall be applied for each polarity
Test voltage: a) for "Series X": 1,4 times the residual voltage at the nominal discharge current according to
Table 1 and 8.3.3 b) for "Series Y": 1,13 times the residual voltage at high current impulse, but not less than
1,3 times the residual voltage at nominal discharge current according to Table 1 and 8.3.3 and 8.3.4
If the dry arcing distance or the sum of the partial dry arcing distances is larger than the test voltage divided by 500 kV/m, this test is not required
Evaluation: The SVU shall be considered to have passed the test if the number of external disruptive discharges does not exceed two in each series of 15 impulses.
Insulation withstand tests on EGLA with failed SVU
A switching impulse wet withstand voltage test and a power-frequency wet withstand voltage test are conducted to simulate a failed SVU These tests aim to ensure that no sparkover occurs under switching surge and power-frequency overvoltages, even in the worst-case scenario of a shorted SVU due to failure.
8.2.3.2 Switching impulse wet withstand voltage test
The test procedure shall be as follows:
Test sample: EGLA with shorted SVU The failed SVU shall be simulated by shorting the
The electrode condition for the SVU with a metal wire will be determined through mutual agreement between the manufacturer and the purchaser Additionally, the manufacturer will specify the minimum external series gap length required for the test.
The manufacturer must specify the withstand voltage value or agree with the purchaser based on the actual switching impulse withstand voltage level of the line The 50% flashover voltage (U 50, EGLA) is determined using the up-and-down method as per IEC 60060-1 for each polarity on the EGLA with the SVU shorted, utilizing a test voltage wave shape of 250/2500 Additionally, the rain characteristics must comply with IEC 60060 standards.
Evaluation: The withstand voltage of the EGLA is determined as
The EGLA value at U 10 is equal to U 50 multiplied by (1 - 1.3σ), where the 50% flashover voltage is measured and the standard deviation σ is assumed to be 6% (σ = 0.06) for switching impulse voltage The EGLA is considered to have passed the test if the withstand value meets or exceeds the claimed or agreed value.
NOTE For a normal distribution, as assumed here, the 10% probability value results from the 50% probability value minus 1,3 times the standard deviation
8.2.3.3 Power-frequency wet withstand voltage test
The test procedure shall be as follows:
Test sample: EGLA with shorted SVU The failed SVU shall be simulated by shorting the
The manufacturer or the user must specify the minimum external series gap length and the conditions for the gap electrodes in an SVU with a metal wire.
Test voltage and test condition: a) The power-frequency wet withstand voltage test shall be performed in accordance with
The IEC 60060-1 standard specifies that when testing the EGLA with the SVU shorted, the test voltage must be set to 1.2 times the rated voltage of the EGLA Additionally, the characteristics of the rain during testing must comply with established requirements.
Evaluation: The EGLA has passed the test if the sample withstands the test voltage for one minute.
Residual voltage tests
General
The test confirms that the residual voltages of the SVU and complete EGLA align with the specified values under lightning impulses All residual voltage assessments are conducted on the same three sections of an SVU, with adequate time between discharges to allow the samples to cool to near ambient temperature The EGLA's residual voltage is derived from the SVU sections' measured residual voltage, adjusted by a scale factor and accounting for the inductive voltage drop across the SVU, gap, and connection leads Similarly, the SVU's residual voltage is calculated using the measured values from its sections, multiplied by a scale factor and including the inductive voltage drop across the SVU.
Procedure for correction and calculation of inductive voltages
To determine the need for inductive correction in a current wave shape of 2/20, apply a current impulse to a metal block matching the dimensions of the tested resistor samples Record the peak voltage across the metal block If this peak voltage is less than 2% of the resistor samples' peak voltage, no correction is necessary If it falls between 2% and 20%, subtract the metal block's impulse shape from each resistor voltage to obtain corrected values Should the peak voltage exceed 20%, improvements to the test and voltage measuring circuits are required.
To ensure consistent current wave shapes during measurements, it is recommended to connect both the test sample and the metal block in series within the test circuit By simply swapping their positions, one can measure the voltage drop across either the metal block or the test sample effectively.
The sample impulse voltage waveform, adjusted as needed, with the maximum peak value, will be utilized to assess the current impulse residual voltage of the SVU and the complete system.
EGLA, respectively, according to one of the following procedures a) or b):
1) Multiply the sample impulse voltage wave shape by the scale factor (see 6.2)
2) From the wave shape of the current impulse, determine the rate of change of current
(di/dt) over the entire wave shape and multiply it by the inductance in order to determine the inductive voltage drop:
( ) d d i i u t L L h t t where u(t) is the inductive voltage drop in kV as a function of time;
The inductance per unit length, denoted as \$L'\$, is measured in àH/m, with a value of \$L' = 1\$ àH/m The variable \$h\$ represents the terminal-to-terminal length in meters of either the SVU or the entire EGLA, which includes the series gap and connection leads Additionally, \$di/dt\$ indicates the rate of change of current over time, expressed in kA/às.
3) Add the results of 1) and 2) on a wave shape basis; the peak value of the resulting wave shape shall be taken as the actual current impulse residual voltage of the arrester
1) Multiply the peak value of the sample impulse voltage by the scale factor (see 6.2)
2) Determine the inductive voltage drop between the arrester terminals using the following formula:
U L is the peak value of the inductive voltage drop in kV;
The inductance per unit length, denoted as \$L'\$, is measured in henries per meter (àH/m) and is equal to 1 The variable \$h\$ represents the terminal-to-terminal length in meters of either the SVU or the entire EGLA, which includes the series gap and connection leads.
T f is the front time of the current impulse in às; T f = 2;
I d is the actual discharge current amplitude in kA
3) Add the results of 1) and 2); the resulting value shall be taken as the actual current impulse residual voltage of the arrester.
Lightning current impulse residual voltage test
Each of the three samples will undergo a single lightning current impulse, tested at peak values of approximately 0.5, 1, and 2 times the nominal discharge current of the EGLA The current wave shape will be 8/20 for "Series X" arresters and 2/20 for others.
"Series Y" arresters according to Table 1
The tolerances for adjusting the equipment must ensure that the measured values of the current impulses fall within specified limits, particularly for 2/20 current impulses.
– from 1,7 às to 2,3 às for virtual front time;
– from 18 às to 22 às for virtual time to half value on the tail; b) for 8/20 current impulses:
– from 7 às to 9 às for virtual front time;
– from 18 às to 22 às for virtual time to half value on the tail
The lightning impulse residual voltage for "Series Y” arresters is determined as per procedure a) or b) in 8.3.2 For "Series X” arresters, no inductive effects are necessary to consider
The maximum values of the determined residual voltages shall be drawn in a residual voltage versus discharge current curve
• 1,4 times the residual voltage at the nominal discharge current according to Table 1 for
• 1,3 times the residual voltage at nominal discharge current according to Table 1 for
"Series Y" designs, shall be lower than the minimum flashover voltage of the insulator assembly to be protected
If a routine test cannot be performed on a complete SVU at its nominal discharge current, it is essential to conduct tests at a current between 0.01 and 1 times the nominal discharge current for effective comparison with the complete SVU.
High current impulse residual voltage test
The test is specifically designed for "Series Y" models, requiring the application of a high current impulse with a 2/20 wave shape and a peak value as specified in Table 1 to each of the three samples.
For the current impulses the tolerances on the adjustment of the equipment shall be such that the measured values of the current impulses are within the following limits:
– from 1,7 às to 2,3 às for virtual front time;
– from 18 às to 22 às for virtual time to half value on the tail
The high current impulse residual voltage is determined as per procedure a) or b) in 8.3.2
The value of 1,13 times the high current impulse residual voltage shall be lower than the minimum flashover voltage of the insulator assembly to be protected See also 10.5.3.
Standard lightning impulse sparkover test
A mandatory type test is required for each specific insulator assembly if an acceptance test, as outlined in section 10.5, has not been conducted This type test is carried out independently of the insulator assembly.
The purpose of this test is to determine the 50 % sparkover voltage of the EGLA under lightning impulse voltage stress
The test sample is one EGLA with the maximum gap distance for a given designated system, without the insulator assembly
Wave shape shall be 1,2/50 The 50 % sparkover voltage (U 50, EGLA ) shall be verified by the up-and-down method according to IEC 60060-1
The protective margin between the sparkover voltage of the EGLA and the flashover voltage of the insulator assembly can be assessed using the formula \( U_{50, \text{EGLA}} + X \times \sigma \), where \( U_{50, \text{EGLA}} \) represents the EGLA's sparkover voltage and \( \sigma \) denotes the standard deviation.
The insulator assembly must be protected such that the value of \( U_{50, \text{Insulator}} - X \cdot \sigma \) does not exceed \( U_{50, \text{Insulator}} \), where \( X \) is a value mutually agreed upon by the manufacturer and the user The standard deviation (\( \sigma \)) is established at 3% for 1.2/50 impulses.
NOTE 2 A recommended value for X is 2,5
NOTE 3 Experience during testing has shown that the sparkover voltage of the EGLA may be influenced by the close vicinity of the insulator assembly
High current impulse withstand test
Selection of test samples
The test will be conducted on three sections of an SVU, ensuring that the residual voltage at nominal discharge current is at the upper limit of the manufacturer's specified variation range To meet these requirements, two conditions must be satisfied: first, the ratio of the residual voltage at nominal discharge current of the complete SVU to that of the section, denoted as n, must be maintained, with the volume of resistor elements used for testing not exceeding the minimum volume of all resistor elements in the complete SVU divided by n Second, the reference voltage \( U_{ref} \) of the SVU in the test section must equal the minimum reference voltage of the SVU of the EGLA divided by n.
If the SVU of the test section exceeds the minimum reference voltage of the complete EGLA divided by the factor n, then n should be correspondingly reduced Conversely, if the SVU of the test section falls below this minimum reference voltage, the test section cannot be utilized.
Test procedure
Two high current impulses of same polarity, having peak values (tolerance
0 + %) and wave shapes according to Table 1, shall be applied to the three sections Time interval between the impulse applications shall allow the sample to cool to ambient temperature
The tolerances on the adjustment of the equipment shall be such that the measured values of the current impulses are within the following limits: a) for 2/20 current impulses:
– from 1,7 às to 2,3 às for virtual front time;
– from 18 às to 22 às for virtual time to half value on the tail; b) for 4/10 current impulses:
– from 3,5 às to 4,5 às for virtual front time;
– from 9 às to 11 às for virtual time to half value on the tail.
Test evaluation
The reference voltage must not vary by more than 10% before and after testing Additionally, any change in residual voltage at nominal discharge current should remain within the range of -2% to +5% A visual inspection of the test samples post-examination should show no signs of puncture, flashover, cracking, or other significant damage If the metal-oxide resistors are not removable for visual inspection, further tests are required to confirm that no damage occurred during the testing process.
Following the residual voltage test, two impulses at the nominal discharge current must be applied to the test sample The first impulse is administered after allowing the sample to cool to ambient temperature, while the second impulse is applied 50 to 60 seconds later Throughout both impulses, the oscillograms of voltage and current should not indicate any breakdown, and the difference in residual voltage between the initial measurement and the final measurement after the two impulses should remain within the range of -2% to +5%.
Lightning discharge capability test
Selection of test samples
This test shall be performed on three samples These samples shall include complete SVUs,
SVU sections or metal-oxide resistor elements which have not been subjected to any previous tests except as necessary for evaluation purposes of this test
For the lightning impulse discharge capability test, samples must exhibit a residual voltage at nominal discharge current at the upper limit of the manufacturer's specified variation range In multi-column arresters, the maximum uneven current distribution must be taken into account To meet these requirements, the following conditions must be satisfied: a) The ratio of the residual voltage at nominal discharge current of the complete Surge Voltage Unit (SVU) to that of the section is defined as \( n \) The volume of resistor elements used as test samples must not exceed the minimum volume of all resistor elements in the complete SVU divided by \( n \) b) The reference voltage \( U_{\text{ref}} \) of the SVU in the test section should equal the minimum reference voltage of the SVU of the External Ground Lightning Arrester (EGLA) divided by \( n \).
If the SVU of the test section exceeds the minimum reference voltage of the complete EGLA divided by the factor n, then n should be correspondingly reduced Conversely, if the SVU of the test section falls below this minimum reference voltage, the test section cannot be utilized.
Test procedure
Before commencing the tests, the lightning impulse residual voltage at nominal discharge current and the reference voltage of each test sample shall be measured for evaluation purposes
Each lightning impulse discharge capability test comprises 18 discharge operations, organized into six groups of three There should be a 50 to 60-second interval between individual operations, and sufficient time between groups to allow the sample to cool to near ambient temperature.
After completing 18 discharge operations and allowing the sample to cool to near ambient temperature, it is essential to repeat the residual and reference voltage tests conducted prior to the test The results must be compared to the initial values, ensuring that any changes fall within the acceptable range of (–2% to +5%).
Visual examination of the test samples after the test shall reveal no evidence of puncture, flashover, cracking or other significant damage of the metal-oxide resistors
For designs where resistors are not accessible for inspection, an additional impulse must be applied once the sample has cooled to ambient temperature If the sample endures this 19th impulse without sustaining damage, as verified by oscillographic records, it is deemed to have passed the test.
Test parameters for the lightning impulse discharge capability test
The manufacturer selects the current peak value to achieve a specific charge, ensuring that the current impulse shape is approximately sinusoidal The duration for which the instantaneous impulse current exceeds 5% of its peak value must be between 200 µs and 230 µs Additionally, the peak of any opposite polarity current wave should remain below 5% of the peak current value Each impulse's current peak value on test samples should fall within 100% to 110% of the chosen peak value.
Measurements during the lightning impulse discharge capability test
The charge and peak current must be documented for each impulse, along with the duration that the instantaneous impulse current exceeds 5% of its peak value Additionally, oscillograms illustrating the typical voltage and current waveforms should be displayed on the same time scale.
Rated lightning impulse discharge capability
Average peak current and charge shall be calculated from the 18 discharge operations
The rated lightning impulse discharge capability of the arrester is determined by two key factors: the lowest average peak current recorded among three test samples and a charge value from the specified list of 8.6.6 that is less than or equal to the lowest average charge of the three samples.
List of rated charge values
The following values, expressed in C, are standardized as rated charge values: 0,4; 0,6; 0,8;
Short-circuit tests
General
The manufacturer must assert the short-circuit rating of the SVU, which will undergo testing as specified in this subclause The purpose of the test is to ensure that any failure of the SVU does not lead to violent shattering of its housing and that any open flames self-extinguish within a specified timeframe Each type of SVU is evaluated using four different short-circuit current values Additionally, if the SVU features an alternative to a conventional pressure relief device, this arrangement must also be included in the testing process.
The frequency of the short-circuit test current supply shall be between 48 Hz and 62 Hz
With respect to short-circuit current performance, it is important to distinguish between two designs of SVUs:
– “Design A” SVUs have a design in which a gas channel runs along the entire length of the
SVU unit and fills ≥ 50 % of the internal volume not occupied by the internal active parts
"Design B" SVUs feature a robust construction that does not include an enclosed gas volume, or if there is an internal gas volume, it occupies less than 50% of the internal space not filled by active components.
NOTE 1 Typically, “Design A” SVUs are porcelain-housed SVUs, or polymer-housed SVUs with a composite hollow insulator which are equipped either with pressure-relief devices, or with prefabricated weak spots in the composite housing which burst or flip open at a specified pressure, thereby decreasing the internal pressure
"Design B" SVUs are solid units without pressure relief devices or enclosed gas volumes In the event of electrical failure in the resistors, an internal arc forms, leading to significant evaporation and potential damage to the housing and internal materials The effectiveness of these SVUs in short-circuit situations relies on their capacity to manage the cracking or tearing of the housing caused by the arc, preventing violent shattering.
NOTE 2 "Active parts" in this context are the non-linear, metal-oxide resistors and any metal spacers directly in series with them
The requirements for the number of test samples, initiation of short-circuit current, and the amplitude of the first short-circuit current vary based on the type of SVU and the test voltage used.
60099-8 IEC:2011 – 25 – circuit current peak Table 4 shows a summary of these requirements which are further explained in the following subclauses.
Preparation of the test samples
For high-current testing, the samples must consist of the longest SVU unit designed, utilizing the highest rated voltage specific to each SVU design.
For the low-current test, the test sample must be an SVU unit of any length, utilizing the highest rated voltage specific to each SVU design.
NOTE 1 Figure 2 shows different examples of SVU units
When selecting a fuse wire, it is essential to choose the appropriate material and size to ensure that the wire melts within the first 30 electrical degrees after the test current begins.
NOTE 2 In order to have melting of the fuse wire within the specified time limit and create a suitable condition for arc ignition, it is generally recommended that a fuse wire of a low resistance material (for example copper, aluminium or silver) with a diameter of about 0,2 mm to 0,5 mm be used Higher fuse-wire cross-sections are applicable to surge SVU units prepared for higher short-circuit test currents When there are problems in initiating the arc, a fuse wire of larger size but with a diameter not exceeding 1,5 mm, may be used since it will help arc establishment In such cases, a specially prepared fuse wire, having a larger cross-section along most of the SVU height with a short thinner section in the middle, may also help.“
Samples must be prepared to conduct the necessary short-circuit current using a fuse wire This fuse wire should be in direct contact with the MO resistors and positioned as close as possible to the gas channel, effectively short-circuiting the entire internal active component The specific location of the fuse wire during testing must be documented in the test report.
No differences with regard to polymer housings or porcelain housings are made in the preparation of the test samples However, differences partly apply in the test procedure
"Design A" SVUs featuring polymeric sheds, which are not constructed from porcelain or other hollow insulators and possess brittleness similar to ceramics, will be classified and evaluated as porcelain-housed SVUs.
"Design B" SVUs featuring polymeric sheds, which lack porcelain or other mechanical support structures and possess brittleness similar to ceramics, will be classified and evaluated as porcelain-housed SVUs.
No special preparation is required for the testing process Standard SVU units will be utilized, which must undergo electrical pre-failure through a power-frequency overvoltage This overvoltage test will be conducted on fully assembled units, and no physical alterations will be made to the units between the pre-failure and the subsequent short-circuit current test.
The manufacturer's specified overvoltage must exceed the reference voltage, leading to the failure of the SVU within (5 ± 3) minutes Resistors are deemed to have failed when the voltage across them drops below 10% of the initially applied voltage.
The short-circuit current of the pre-failing test circuit shall not exceed 30 A
The time between pre-failure and the rated short-circuit current test shall not exceed 15 min
NOTE The pre-failure can be achieved by either applying a voltage source or a current source to the samples
The voltage source method requires an initial current of 5-10 mA/cm² and a short-circuit current ranging from 1 A to 30 A Once set, the voltage source generally does not need further adjustments, although minor tweaks may be necessary to ensure the resistors fail within the specified time frame.
The current source method typically operates with a current density of approximately 15 mA/cm², allowing for a variation of ± 50% This setup can lead to resistor failure within the specified time range The short-circuit current should generally fall between 10 A and 30 A Once the current source is initially set, it usually does not require further adjustments, although minor tweaks may be needed to ensure resistor failure within the designated timeframe.
Samples must be prepared to conduct the necessary short-circuit current using a fuse wire This fuse wire should be in direct contact with the MO resistors and positioned as far from the gas channel as possible, ensuring it short-circuits the entire internal active component The specific location of the fuse wire during testing will be documented in the test report.
Mounting of the test sample
The SVU units can be installed either directly onto a base, as illustrated in Figures 3a and 3b, or suspended according to the manufacturer's installation guidelines The manufacturer has the discretion to choose the installation method For suspended installations, it is essential that the bottom end of the SVU aligns with the upper edge of the circular enclosure.
For a base-mounted SVU, the mounting arrangement is shown in Figures 3a and 3b The distance to the ground from the insulating platform and the conductors shall be as indicated in
For non-base-mounted SVUs, such as pole-mounted models, the test sample must be installed on a non-metallic pole using standard mounting brackets and hardware The mounting bracket is considered part of the SVU base for testing purposes If this setup conflicts with the manufacturer's instructions, the SVU should be installed according to the manufacturer's guidelines Additionally, the lead between the base and the current sensor must be insulated for a minimum of 1,000 V, and the top end of the test sample should be equipped with either a base assembly of the same design as the SVU or a top cap.
For base-mounted SVUs, the bottom end fitting must be installed on a test base that matches the height of a surrounding circular or square enclosure, which can be made of insulating material or conducting material if its dimensions are smaller than those of the SVU fitting Both the test base and enclosure should rest on an insulating platform Non-base-mounted SVUs follow the same bottom requirements The arcing distance between the top end cap and any metallic object, excluding the SVU base, must be at least 1.6 times the height of the SVU sample, with a minimum of 0.9 m The enclosure, constructed from non-metallic material, must be symmetrically aligned with the test sample and remain stationary throughout the test Its height should be 40 cm ± 10 cm, and its diameter or side length (for square enclosures) must be at least 1.8 m or determined by the specified equation.
H is the height of tested SVU unit;
D SVU is the diameter of tested SVU unit
Porcelain-housed SVUs shall be mounted according to Figure 3a Polymer housed SVUs shall be mounted according to Figure 3b
Test samples shall be mounted vertically unless agreed upon otherwise between the manufacturer and the purchaser
NOTE 1 Mounting of the SVU during the short-circuit test and, more specifically, the routing of the conductors should represent the most unfavourable condition in service
The routing depicted in Figure 3a is the least favorable option during the initial testing phase prior to venting, particularly when a pressure relief device is installed on the SVU Proper positioning of the sample is crucial in this context.
The orientation of the venting ports in Figure 3a, directed towards the test source, may lead to the external arc being positioned closer to the SVU housing, resulting in increased thermal shock and potential chipping or shattering of porcelain weather sheds However, this configuration also encourages the arc to move away from the SVU during the later stages of arcing, thereby lowering the risk of fire, particularly in SVUs with polymeric housing Both the initial arcing phase and the fire risk are critical considerations in this context.
For polymer-housed SVUs, it is essential to direct the ground conductor opposite to the incoming conductor, as illustrated in Figure 3b This configuration ensures that the arc remains near the SVU throughout the short-circuit current, thereby maximizing the potential fire hazard.
NOTE 2 In the event that physical space limitations of the laboratory do not permit an enclosure of the specified size, the manufacturer may choose to use an enclosure of lesser diameter.
High-current short-circuit tests
Three samples will be tested at currents determined by a chosen rated short-circuit current from Table 5 All samples will be prepared in accordance with section 8.7.2 and mounted as specified in section 8.7.3.
Tests shall be made in a single-phase test circuit, with an open-circuit test voltage of 77 % to
Testing high-voltage SVUs typically requires a voltage of 107% of the rated voltage, as specified in section 8.7.4.1 However, many laboratories may lack the necessary short-circuit power capability to conduct these tests at 77% or higher of the rated voltage Therefore, an alternative method for performing high-current, short-circuit tests at a lower voltage is provided in section 8.7.4.2 It is essential that the total duration of the test current flowing through the circuit is at least 0.2 seconds.
NOTE Experience from porcelain-housed arresters has shown that tests at the rated current do not necessarily demonstrate acceptable behaviour at lower currents
8.7.4.1 High-current tests at full voltage (77 % to 107 % of rating)
The prospective current shall first be measured by making a test with the SVU short-circuited or replaced by a solid link of negligible impedance
The duration of such a test may be limited to the minimum time required to measure the peak and symmetrical component of the current waveform
For "Design A" SVUs, the peak value of the first half-cycle of the prospective current must be at least 2.5 times the r.m.s value of the symmetrical component of the prospective current, which should equal or exceed the rated short-circuit current The test current is determined by dividing the peak value of the prospective current by 2.5, regardless of whether the r.m.s value of the symmetrical component is higher Due to the increased prospective current, the sample SVU may experience more severe conditions, so tests with an X/R ratio lower than 15 require the manufacturer's consent.
For “Design B” SVUs tested at rated short-circuit current, the peak value of the first half- cycle of the prospective current shall be at least √2 times the r.m.s value
The r.m.s value of all reduced short-circuit currents must comply with Table 5, and the peak value of the first half-cycle of the prospective current should be a minimum of \$\sqrt{2}\$ times the r.m.s value of this current.
The solid shorting link shall be removed after checking the prospective current and the SVU sample(s) shall be tested with the same circuit parameters
NOTE 1 The resistance of the restricted arc inside the SVU may reduce the r.m.s symmetrical component and the peak value of the measured current This does not invalidate the test, since the test is being made with at least normal service voltage and the effect on the test current is the same as would be experienced during a fault in service
NOTE 2 The X/R ratio of the test circuit impedance, without the SVU connected, should preferably be at least 15
In cases where the test circuit impedance X/R ratio is less than 15, the test voltage may be increased or the impedance may be reduced, in such a way that,
The peak value of the first half-cycle of the prospective current for the rated short-circuit current must be equal to or exceed 2.5 times the specified test current level.
– for the reduced current level tests, the tolerances in Table 5 are met
8.7.4.2 High-current test at less than 77 % of rated voltage
When conducting tests with a test circuit voltage below 77% of the rated voltage for the samples, it is essential to adjust the test circuit parameters to ensure that the root mean square (r.m.s.) value of the symmetrical component of the actual SVU test current meets or exceeds the required test current level of 8.7.4.
For "Design A" SVUs, the peak value of the first half-cycle of the test current must be at least 2.5 times the r.m.s value of the symmetrical component of the actual test current Additionally, this r.m.s value should be equal to or exceed the rated short-circuit current The test current is defined as the peak value of the actual SVU test current divided by 2.5, regardless of whether the r.m.s value of the symmetrical component is higher.
The following exception for the test at rated short-circuit current is valid for “Design A” polymer-housed SVUs only (see 8.7.2.1 for the definition of polymer- and porcelain-housed
SVUs): if the rated voltage of the test sample is more than 150 kV and a first peak value of
≥ 2,5 times the rated short-circuit current cannot be achieved, an additional test sample shall be tested This additional test sample shall be tested according to either 8.7.4.1 or 8.7.4.2
The equipment must have a rated voltage of at least 150 kV and should not be shorter than the shortest SVU unit utilized in the actual design The rated short-circuit current value must be the lower of the r.m.s current obtained from testing the longest unit or the r.m.s current determined from tests conducted according to sections 8.7.4.1 or 8.7.4.2 on the minimum 150 kV rated unit Both test results are required to be reported.
For "Design B" SVUs evaluated at the specified short-circuit current, the peak value of the initial half-cycle of the actual SVU test current must be a minimum of \$\sqrt{2}\$ times the root mean square (r.m.s.) value.
The r.m.s value of all reduced short-circuit currents must comply with Table 5, and the peak value of the first half-cycle of the actual SVU test current should be a minimum of \$\sqrt{2}\$ times the r.m.s value of that current.
NOTE 1 Especially for tall SVUs that are tested at a low percentage of their rated voltage, the first asymmetric peak current of 2,5 is not easily achieved unless special test possibilities are considered It is thus possible to increase the test r.m.s voltage or reduce the impedance so that, for the rated short-circuit current, the peak value of the first half-cycle of the test current is equal to, or greater than, 2,5 times the required test current level In case of testing with a generator, the first peak of 2,5 times the required test current can also be achieved by varying the generator’s excitation The current should then be reduced, not less than 2,5 cycles after initiation, to the required symmetrical value The actual peak value of the test current, divided by 2,5, should be quoted as the test current, even though the r.m.s value of the symmetrical component of the actual SVU test current may be
Testing at an X/R ratio lower than 15 for the sample SVU may lead to more severe duty due to the higher test current, and should only be conducted with the manufacturer's consent.
NOTE 2 For “Design B” polymer-housed SVUs, even the first current peak of √2 may not be easily achieved unless special test facilities are considered Pre-failed SVUs can build up considerable arc resistance, which limits the symmetrical current through the SVU It is therefore recommended to perform the short-circuit tests as soon as possible after the pre-failure, preferably before the test samples have cooled down
To ensure accurate testing of pre-failed SVUs, it is crucial to maintain a sufficiently low impedance before applying the short-circuit current This can be achieved by reapplying the pre-failing circuit for a maximum of 2 seconds prior to the short-circuit test current The short-circuit current of the pre-applied circuit may be increased up to 300 A (r.m.s), but the duration must not exceed the limit defined by the equation: \( t_{rpf} \leq \frac{Q_{rpf}}{I_{rpf}} \), where \( t_{rpf} \) is the re-pre-failing time in seconds.
Q rpf is the re-pre-failing charge in C; Q rpf = 60 C;
I rpf is the re-pre-failing current (r.m.s.) in A.
Low-current short-circuit test
The test must utilize a circuit capable of generating a current of 600 A ± 200 A (r.m.s value) through the test sample, measured approximately 0.1 seconds after the current begins The current should be maintained for 1 second, or until venting occurs for “Design A” porcelain-housed surge voltage suppressors (SVUs).
Refer to Note 2 of 8.7.6 with regard to handling an SVU that fails to vent.
Evaluation of test results
The test is considered successful if the following three criteria are met a) No violent shattering
NOTE 1 Structural failure of the sample is permitted as long as criteria b) and c) are met b) No parts of the test sample shall be allowed to be found outside the enclosure, except for
• fragments, less than 60 g each, of ceramic material such as from metal-oxide resistors or porcelain;
• pressure relief vent covers and diaphragms;
• soft parts of polymeric materials c) The SVU shall be able to self-extinguish open flames within 2 min after the end of the test
Any ejected part (in or out of the enclosure) shall also self-extinguish open flames within 2 min
NOTE 2 If the SVU has not visibly vented at the end of the test, caution should be exercised, as the housing may remain pressurized after the test This note is applicable to all levels of test current, but is of particular relevance to the low-current, short-circuit tests
NOTE 3 A shorter duration of self-extinguishing open flames for ejected parts may be agreed upon between the purchaser and the manufacturer
NOTE 4 It may be of particular importance for EGLA applications that safety considerations on ejected fragments, mechanical integrity and even a certain strength after failure are required In that case, different test procedures and evaluations may be established between the manufacturer and the user (as an example, it may be required that after the tests the SVU should still be able to be lifted and removed by its top end)
Table 4 – Test requirements R equi re d num be r of te st sam pl es
The initiation of short-circuit current is determined by the ratio of the first current peak value to the root mean square (RMS) value of the required short-circuit current, as outlined in Table 5 The test voltage ranges from 77% to 107% of the rated short-circuit current, while values below 77% indicate a reduced short-circuit current The design incorporates a porcelain-housed fuse wire along the surface of the resistors, positioned as close as possible to the gas channel to optimize performance.
The requirements for the design specifications indicate a minimum of 2.5 for the proposed values, while the actual requirements show no specific minimum Additionally, the actual values must be at least 2, with a focus on "Design A," which involves polymer-housed fuse wire positioned along the surface of the MO resistors, ideally located within or as close as possible to the gas channel.
The requirements for the project include a minimum of 2 for the proposed specifications, with no actual requirements specified Additionally, the actual specifications must meet or exceed 2, with a specific condition of 2.5 for the longest unit and 2.5 for units with a UR of 150 kV or higher The design incorporates porcelain-housed fuse wires along the surface of the MO resistors, positioned as far away as possible from the gas channel.
The requirements for the design include a minimum of two prospects, with no specific actual requirements The actual specifications also indicate a minimum of two, emphasizing the use of "Design B," which features a polymer-housed system This system is designed to pre-fail under constant voltage or constant current sources.
P ros p : ≥2 A ct ual : no req ui rem ent P ros p : ≥2 A ct ual : no req ui rem ent
A ct ual : ≥2 A ct ual : ≥2 A ct ual : ≥2A ct ual : ≥2
Table 5 – Required currents for short-circuit tests
Low short-circuit current with a duration of 1 s a)
NOTE 1 If an existing type of SVU, already qualified for one of the rated currents in Table 5, is being qualified for a higher rated-current value available in this table, it should be tested only at the new rated value Any extrapolation can only be extended by two steps of rated short-circuit current
NOTE 2 If a new SVU type is to be qualified for a higher rated current value than available in this table, it should be tested at the proposed rated current, at 50 % and at 25 % of this rated current
NOTE 3 If an existing SVU is qualified for one of the rated short-circuit currents in this table, it is deemed to have passed the test for any value of rated current lower than this one a) For SVUs to be installed in resonant earthed or unearthed neutral systems, the increase of the test duration to longer than 1 s, up to 30 min, may be permitted after agreement between the manufacturer and the purchaser
In this case the low short-circuit current shall be reduced to 50 A ± 20 A, and the test sample and acceptance criteria shall be agreed between the manufacturer and the purchaser.
IE C 289 7/1 0 Figure 2 – Examples of SVU units
Fl ex ib le ov er a l engt h of at le as t 0, 2 m Su rg e a rre st er Ba se Ins ul at in g pl at fo rm 0 m -2 m
V en ting s ys tem (if a ny ) 0, 4 m ± 0, 1 m
Fl exi bl e ov er a l engt h of at lea st 0, 2 m S ur ge ar res ter B as e Ins ul at ing pl at for m
V ent ing s ys tem (if a ny) E nc los ur e
SVU SVU IE C 289 8/1 0 IE C 289 9/1 0 Figure 3 – Short-circuit test setup
The circuit layout for porcelain-housed surge voltage suppressors (SVUs) features all leads and venting systems aligned in the same plane, as illustrated in Figure 3a Similarly, Figure 3b depicts the circuit layout for polymer-housed SVUs, which also maintains the same configuration of leads and venting systems in a single plane.
SW 1 is closed while SW 2 is opened to apply a pre-failing current level, capped at 30 A due to impedance Z After a maximum duration of 2 seconds, SW 2 is closed to initiate the specified short-circuit current through the test sample.
Figure 4 – Example of a test circuit for re-applying pre-failing circuit immediately before applying the short-circuit test current
Follow current interrupting test
General
This test aims to assess the current interrupting capabilities of an EGLA following a series gap that has been subjected to a lightning impulse The sample tested may consist of either a complete EGLA or a segment of it.
This test assesses the EGLA's performance in polluted environments by considering the current that flows over the SVU housing surface due to a wetted pollution layer.
The test can be conducted as a type test using a manufacturer-selected SDD level and EGLA configuration, or as an acceptance test with an SDD level mutually agreed upon by the manufacturer and purchaser, as outlined in section 10.6.
The follow current interrupting test shall be performed by either “Test method A” (see 8.8.2) or
If the pollution severity at the site is classified as "Very heavy" according to IEC/TS 60815-1, "Test method B" must be utilized Otherwise, the manufacturer has the discretion to select the appropriate test method.
In "Test method A," the impact of pollution on the external surface current of the SVU is simulated using an additional linear resistor in parallel with the SVU, conducted under clean and dry conditions Conversely, "Test method B" evaluates the SVU under artificial pollution conditions.
8.8.2.1 Requirements on the test circuit
The impedance of the power-frequency voltage source must be configured to ensure that the peak value of the power-frequency voltage, measured at the EGLA terminals, is maintained during the flow of follow current.
The test specimen's peak voltage must not drop below its rated voltage, and after the interruption of follow current, it should not exceed the rated voltage by more than 10% An example of a test circuit is provided in Annex A.
The preparation of the EGLA test sample requires that the non-linear metal-oxide resistor component be either a complete SVU, an SVU section, or a collection of metal-oxide resistor elements, with a scale factor \( n \) not exceeding five For complete EGLAs rated above 12 kV, the test sample's rated voltage must also be at least 12 kV Additionally, the volume of the resistor elements used as test samples must not exceed the minimum volume of all resistor elements in the complete SVU divided by \( n \) Furthermore, the reference voltage \( U_{\text{ref}} \) of the SVU in the test section should match the minimum reference voltage of the EGLA's SVU divided by \( n \).
The SVU of the test section must exceed the minimum reference voltage of the complete EGLA divided by the factor \( n \); if it does not, the factor \( n \) should be reduced accordingly If the reference voltage of the SVU in the test section falls below this minimum threshold, the test section cannot be utilized Additionally, a linear resistor must be connected in parallel with the SVU to ensure a sufficiently high follow current Furthermore, the external series gap should consist of the same electrodes used in the EGLA.
Its length shall be not greater than the minimum gap length specified by the manufacturer
It is not necessary to scale the gap
The test shall be conducted as follows:
A power-frequency voltage equal to the rated voltage of the EGLA or EGLA section shall be applied to the test sample
The follow current flowing through the external series gap during the test will result as the addition of the following two components:
• the leakage current on the SVU polluted surface simulated by means of the linear resistor connected in parallel to the SVU;
• the internal resistive current through the non linear metal-oxide resistor blocks when energised at the rated voltage
The resistance of the linear resistor necessary to simulate the leakage current on the SVU polluted surface shall be calculated as R = F/K being F the form factor (according to
IEC 60507) of the SVU housing and K the layer conductivity
The layer conductivity K shall be taken from Table 3 of IEC 60507 at the line corresponding to the selected SDD The accepted tolerance for the resistance shall be
NOTE 1 In the case of an EGLA, the pollution layer on the SVU is not under voltage until sparkover occurs In a worst-case scenario, the pollution layer will be totally wetted under rain conditions and will remain so since drying due to surface leakage currents does not occur As there is no dry band arcing activity, the pollution layer may be assumed as a linear resistance
NOTE 2 With this method, the current level is higher than in operating service conditions, because the calculation does not take into account the voltage drop across the external series gap of the EGLA
Lightning impulses will be applied to the EGLA to initiate sparkover and create a conductive channel across the external series gap The impulse generator will be calibrated to ensure consistent sparkover of the gap.
The lightning impulses, having the same or opposite polarity as the actual half cycle of the alternating voltage, shall be applied (30° to 0°) before the instant of peak voltage
A first test shall be performed with a gap length small enough to show that the power source is able to supply and maintain the specified follow current
The parallel linear resistor shall be adjusted such that the total follow current during the tests is at least equal to the estimated value
The gap length must be adjusted to the minimum specified value Subsequently, five sparkover operations should be conducted for each polarity during the actual half cycle of the alternating voltage If follow current is not established, additional sparkover operations will be necessary until follow current is achieved five times for each polarity.
Permanent oscillograms of power-frequency voltage and follow current related to each discharge must be recorded These oscillograms should display the voltage across and the current through the test sample from one complete cycle before the impulse application to ten complete cycles after the follow current's final interruption This final interruption must take place within the half-cycle during which the impulse is applied, and there should be no additional sparkover of the sample in any following half-cycle.
The sample is considered successful if, during ten sparkover operations, the follow current is interrupted within the first half cycle of the power-frequency voltage, and no additional sparkover occurs in any subsequent half cycle.
8.8.3.1 Requirements on the test circuit