Development of New Earthing System ... - J-STAGE Home

40 QR of RTRI, Vol. 61, No. 1, Feb. 2020

Hiroki TANAKAWest Japan Railway Company

Gaku MORITAPower Supply Systems Laboratory, Power Supply Technology Division

Development of New Earthing System Inspection Method for the Evaluation of Lightning Protection

Koji MICHISHITAShizuoka University

This paper describes new inspection equipment and a new circuit structure used for evaluating the lightning protection performance of earthing systems in railway substations. The new inspection equipment consists of an impulse generator and a measurement unit, and allows the inspector to measure the earth resistance, the high frequency earth impedance and the voltage difference of an earthing system without specialist knowledge. The new measure-ment circuit for the inspection of the earthing system only makes use of existing infrastructure components, i.e. independent earthing for a remote terminal unit, the earthing system of an adjacent substation and an overhead contact line. Therefore, the proposed circuit requires no temporary test earth electrodes for measurement.

Keywords: earthing system, lightning protection, earth resistance, earth impedance, voltage difference, inspection equipment

1. Introduction

The aim of earthing systems installed in fixed power supply installations, such as substations, is to prevent hu-man damage, and to protect electric and electronic devices inside or near the installation from overvoltage caused by an earth fault or a lightning strike. The Japanese government (the Ministry of Economy, Trade and Industry and the Min-istry of Land, Infrastructure, Transport and Tourism) has implemented a maximum limit value for earth resistance and set earthing construction procedures based on technical criteria. Therefore inspection techniques are important to be able to validate a new or replaced earthing system and to maintain the performance of the earthing system. However, there are problems with conventional earthing system in-spection methods, which are described below:

(1) The main target frequency of the conventional earth-ing system inspection (earth resistance measurement) based on technical criteria is DC and power frequen-cies such as 50 Hz and 60 Hz. Therefore, lightning strikes containing higher frequency components of be-tween 10 kHz to 1 MHz are not covered in the inspec-tion scope.

(2) Earthing system inspections for lightning protection have until now been carried out by inspection compa-nies (not railway operators) [1]. However, the inspec-tions require two types of independent measurement (measurement of earth impedance and voltage differ-ence), specialist skills and special equipment.

(3) Two test electrodes connected to the conventional in-spection equipment are required to inspect the earth-ing system. However, this requirement is difficult to fulfil, particularly in inner-city substations, due to the many buildings and public roads surrounding substa-tions.

Accordingly, new inspection equipment and a new cir-cuit structure were developed for the evaluation of light-ning protection performance of earthing systems.

2. Conventional earthing system inspection

The inspection of earthing systems includes measure-ment of earth resistance (low frequency characteristics) based on the technical criteria, measurement of earth impedance for evaluation of lightning protection (high fre-quency characteristics), and measurement of voltage differ-ence (high frequency characteristics).

2.1 Measurement of earth resistance

Measurement of earth resistance is a test to confirm that the earthing system meets the earth resistance refer-ence value (such as 10 Ω) specified by the technical crite-ria. There are typically two types of inspection method to perform the test [2]; they are the fall-of-potential method and measurement with an earth resistance ohmmeter.

The fall-of-potential method shown in Fig. 1 is specified as an inspection method for a large-scale earthing system such as an earth grid, which is used as an earthing system for fixed power supply installations in the Japan Railway (JR) companies. The test circuit consists of two indepen-dent circuits: a current injection circuit which supplies a test current (approximately 20 A or more) into the earthing system, and a potential measurement circuit which mea-sures the ground potential rise (GPR) of the earthing sys-tem. Two test earth electrodes are required as the termi-nals of the two test circuits: one is a current probe (C) - as the terminal of the current injection circuit, and another is a potential probe (P) - as the reference point of the poten-tial measurement circuit. In addition to preparing these probes, three time measurements and post-processing are

PAPER

41QR of RTRI, Vol. 61, No. 1, Feb. 2020

required to remove the effect of electromagnetic induction from surrounding transmission and distribution lines.

An earth resistance ohmmeter is used for measuring earth resistance of a small-scale earthing system composed of several earth electrodes. The circuit configuration of the earth resistance ohmmeter is almost the same as that of the fall-of-potential method. The ohmmeter can evaluate the earth resistance in one operation by supplying a small test current of several milliamperes to several amperes without specialist data processing skills. By using the ohmmeter, measurement and data processing are made simpler than in the fall-of-potential method.

Fig. 1 Fall-of-potential earth test circuit

measurement circuit which measures the ground potential rise (GPR) of the earthing system. Two test earth electrodes are required as the terminals of the two test circuits: one is a current probe (C) - as the terminal of the current injection circuit, and another is a potential probe (P) - as the reference point of the potential measurement circuit. In addition to preparing these probes, three time measurements and post-processing are required to remove the effect of electromagnetic induction from surrounding transmission and distribution lines.



2.2 Measurement of earth impedance

The earth impedance corresponds to the GPR of the earthing system, particularly during a lightning strike, which is measured by simulated lightning (lightning impulse) injection. In general, when a lightning current flows into an earthing system, the earthing system acts as a nonlinear distributed constant circuit [1, 2]. Therefore, earthing impedance has a characteristic which varies with time and converges with the earth resistance in several tens of microseconds as shown in Fig. 2.

The impulse impedance, which is the peak earth impedance value, is an important value in lightning resistance evaluation. In a large-scale earthing system such as in a substation, the impulse impedance usually has a higher value than low-frequency earth resistance as shown in Fig. 2.

The test circuit for the earth impedance measurement is almost the same as the fall-of-potential method except for the inspection equipment. The earth impedance is obtained by dividing the instantaneous value of GPR by the instantaneous value of the current. In the measurement, simulated lightning is injected by an impulse voltage generator (PG) owned by an inspection company. The front time of the current of PG, which is the most effective parameter to the transient characteristics of the earth impedance, is set to a very short time of about 0.05 μs. The value is sufficiently shorter than the

front time of 0.22 μs, which covers 95% of the subsequent negative short-time lightning strike. The value also represents the most severe past lightning conditions based on current statistics, shown in Table 1. However, it is difficult to manufacture an impulse generator which has both a short front time and a large current. Therefore, the current peak of PG is about 1 A, which is not suitable for the measurement of voltage differences described later.

Fig. 2 Time dependency of earth impedance

Table 1 Lightning current statistics [3]

Parameter Cumulative frequency Type of stroke 98% 95% 80% 50% 5%

Peak 4 kA ― 20 kA ― 90 kA First negative short ― 4.9 kA ― 11.8 kA 28.6 kA Subsequent negative short ― 4.6 kA ― 35 kA 250 kA First positive short (single)

Front time ― 1.8 s ― 5.5 s 18 s First negative short ― 0.22 s ― 1.1 s 4.5 s Subsequent negative short ― 3.5 s ― 22 s 200 s First positive short (single)

Stroke duration

― 30 s ― 75 s 200 s First negative short ― 6.5 s ― 32 s 140 s Subsequent negative short ― 25 s ― 230 s 2000 s First positive short (single)

2.3 Measurement of voltage difference

If a large voltage difference occurs in the earthing system caused by a lightning strike, surge voltage or current propagates through power or low-voltage lines. The surge increases the risk of equipment damage. In particular, equipotential-bonding has been the basis of recent lightning protection technology [3], and it can be evaluated by measuring the voltage difference between two points in the earthing system. Similar to the relationship between the earth impedance and the earth resistance, the voltage difference at a high frequency such when there is a lightning strike has a higher value than the voltage difference at low frequency.

The voltage difference is evaluated by the difference between the potential measurement data at two or more points on the earthing system. However, in order to accurately perform this measurement, the testing current should be propagated through the entire earthing system. Therefore, an impulse current generator (IG) owned by an inspection company, which is capable of outputting current with a peak value of several tens of amperes or more, is used as a power supply. Since the IG is installed and operated from a medium-sized car, the IG has a less portability than that of the earth resistance ohmmeter and the PG. Since the peak value and the front time of the output current from an impulse generator are traded off, the front time of the output current of an IG is several microseconds, which is about one digit longer than that of a PG. Compared to the past lightning current statistics, an evaluation covering approximately 50% of the first negative short-time lightning strike is possible using an IG (see Table 1).

A

V

VE0

Earthing systemunder test (E)

Current probe(C)

Potential probe(P)

ISVS

Current flow in the ground

Voltage sensor

Current sensor

ACsupply Surface

Current injectioncircuit

Potential mesurement circuit

GPR

Potential risearound earthing system (E)

Potential droparound pole (C)

Earth resistance

: at , : under normal phase current, : under opposite phase current

300m - 600m 4 to 5 times of a side of earth mesh

Transformer

: recommended to be >20AGPR: Ground Potential Rise

Time

Earthimpedance

Transient periodabout 1s to 10s

Steady stateabout >100s

Impulse impedance

Earth resistance


The earth impedance corresponds to the GPR of the earthing system, particularly during a lightning strike, which is measured by simulated lightning (lightning im-pulse) injection. In general, when a lightning current flows into an earthing system, the earthing system acts as a non-linear distributed constant circuit [1, 2]. Therefore, earth-ing impedance has a characteristic which varies with time and converges with the earth resistance in several tens of microseconds as shown in Fig. 2.

The impulse impedance, which is the peak earth imped-ance value, is an important value in lightning resistance evaluation. In a large-scale earthing system such as in a substation, the impulse impedance usually has a higher value than low-frequency earth resistance as shown in Fig. 2.

The test circuit for the earth impedance measurement is almost the same as the fall-of-potential method except for the inspection equipment. The earth impedance is obtained by dividing the instantaneous value of GPR by the instanta-neous value of the current. In the measurement, simulated lightning is injected by an impulse voltage generator (PG) owned by an inspection company. The front time of the current of PG, which is the most effective parameter to the transient characteristics of the earth impedance, is set to a very short time of about 0.05 μs. The value is sufficiently shorter than the front time of 0.22 μs, which covers 95% of the subsequent negative short-time lightning strike. The value also represents the most severe past lightning condi-tions based on current statistics, shown in Table 1. Howev-er, it is difficult to manufacture an impulse generator which has both a short front time and a large current. Therefore, the current peak of PG is about 1 A, which is not suitable for the measurement of voltage differences described later.


If a large voltage difference occurs in the earthing sys-tem caused by a lightning strike, surge voltage or current propagates through power or low-voltage lines. The surge increases the risk of equipment damage. In particular, equipotential-bonding has been the basis of recent light-ning protection technology [3], and it can be evaluated by measuring the voltage difference between two points in the earthing system. Similar to the relationship between the earth impedance and the earth resistance, the voltage dif-ference at a high frequency such when there is a lightning strike has a higher value than the voltage difference at low frequency.

The voltage difference is evaluated by the difference between the potential measurement data at two or more points on the earthing system. However, in order to ac-curately perform this measurement, the testing current should be propagated through the entire earthing system. Therefore, an impulse current generator (IG) owned by an inspection company, which is capable of outputting current with a peak value of several tens of amperes or more, is used as a power supply. Since the IG is installed and oper-ated from a medium-sized car, the IG has a less portability than that of the earth resistance ohmmeter and the PG. Since the peak value and the front time of the output cur-rent from an impulse generator are traded off, the front time of the output current of an IG is several microseconds, which is about one digit longer than that of a PG. Com-pared to the past lightning current statistics, an evaluation covering approximately 50% of the first negative short-time lightning strike is possible using an IG (see Table 1).

3. Development of inspection equipment

3.1 Basic concept

In order to overcome the issues faced with the conven-tional inspection equipment for earthing systems, new in-spection equipment was developed based on the following premises:

Table 1 Lightning current statistics [3]Parameter Cumulative frequency Type of stroke98% 95% 80% 50% 5%

Peak4 kA ― 20 kA ― 90 kA First negative short― 4.9 kA ― 11.8 kA 28.6 kA Subsequent negative short― 4.6 kA ― 35 kA 250 kA First positive short (single)

Front time― 1.8 μs ― 5.5 μs 18 μs First negative short― 0.22 μs ― 1.1 μs 4.5 μs Subsequent negative short― 3.5 μs ― 22 μs 200 μs First positive short (single)

Strokeduration

― 30 μs ― 75 μs 200 μs First negative short― 6.5 μs ― 32 μs 140 μs Subsequent negative short― 25 μs ― 230 μs 2000 μs First positive short (single)


measurement circuit which measures the ground potential rise (GPR) of the earthing system. Two test earth electrodes are required as the terminals of the two test circuits: one is a current probe (C) - as the terminal of the current injection circuit, and another is a potential probe (P) - as the reference point of the potential measurement circuit. In addition to preparing these probes, three time measurements and post-processing are required to remove the effect of electromagnetic induction from surrounding transmission and distribution lines.




The earth impedance corresponds to the GPR of the earthing system, particularly during a lightning strike, which is measured by simulated lightning (lightning impulse) injection. In general, when a lightning current flows into an earthing system, the earthing system acts as a nonlinear distributed constant circuit [1, 2]. Therefore, earthing impedance has a characteristic which varies with time and converges with the earth resistance in several tens of microseconds as shown in Fig. 2.

The impulse impedance, which is the peak earth impedance value, is an important value in lightning resistance evaluation. In a large-scale earthing system such as in a substation, the impulse impedance usually has a higher value than low-frequency earth resistance as shown in Fig. 2.

The test circuit for the earth impedance measurement is almost the same as the fall-of-potential method except for the inspection equipment. The earth impedance is obtained by dividing the instantaneous value of GPR by the instantaneous value of the current. In the measurement, simulated lightning is injected by an impulse voltage generator (PG) owned by an inspection company. The front time of the current of PG, which is the most effective parameter to the transient characteristics of the earth impedance, is set to a very short time of about 0.05 μs. The value is sufficiently shorter than the

front time of 0.22 μs, which covers 95% of the subsequent negative short-time lightning strike. The value also represents the most severe past lightning conditions based on current statistics, shown in Table 1. However, it is difficult to manufacture an impulse generator which has both a short front time and a large current. Therefore, the current peak of PG is about 1 A, which is not suitable for the measurement of voltage differences described later.


Table 1 Lightning current statistics [3]

Parameter Cumulative frequency Type of stroke 98% 95% 80% 50% 5%

Peak 4 kA ― 20 kA ― 90 kA First negative short ― 4.9 kA ― 11.8 kA 28.6 kA Subsequent negative short ― 4.6 kA ― 35 kA 250 kA First positive short (single)

Front time ― 1.8 s ― 5.5 s 18 s First negative short ― 0.22 s ― 1.1 s 4.5 s Subsequent negative short ― 3.5 s ― 22 s 200 s First positive short (single)

Stroke duration

― 30 s ― 75 s 200 s First negative short ― 6.5 s ― 32 s 140 s Subsequent negative short ― 25 s ― 230 s 2000 s First positive short (single)


If a large voltage difference occurs in the earthing system caused by a lightning strike, surge voltage or current propagates through power or low-voltage lines. The surge increases the risk of equipment damage. In particular, equipotential-bonding has been the basis of recent lightning protection technology [3], and it can be evaluated by measuring the voltage difference between two points in the earthing system. Similar to the relationship between the earth impedance and the earth resistance, the voltage difference at a high frequency such when there is a lightning strike has a higher value than the voltage difference at low frequency.

The voltage difference is evaluated by the difference between the potential measurement data at two or more points on the earthing system. However, in order to accurately perform this measurement, the testing current should be propagated through the entire earthing system. Therefore, an impulse current generator (IG) owned by an inspection company, which is capable of outputting current with a peak value of several tens of amperes or more, is used as a power supply. Since the IG is installed and operated from a medium-sized car, the IG has a less portability than that of the earth resistance ohmmeter and the PG. Since the peak value and the front time of the output current from an impulse generator are traded off, the front time of the output current of an IG is several microseconds, which is about one digit longer than that of a PG. Compared to the past lightning current statistics, an evaluation covering approximately 50% of the first negative short-time lightning strike is possible using an IG (see Table 1).

A

V

VE0


Current probe(C)

Potential probe(P)

ISVS

Current flow in the ground

Voltage sensor

Current sensor

ACsupply Surface

Current injectioncircuit

Potential mesurement circuit

GPR

Potential risearound earthing system (E)

Potential droparound pole (C)

Earth resistance

: at , : under normal phase current, : under opposite phase current

300m - 600m 4 to 5 times of a side of earth mesh

Transformer

: recommended to be >20AGPR: Ground Potential Rise

Time

Earthimpedance

Transient periodabout 1s to 10s

Steady stateabout >100s

Impulse impedance

Earth resistance


(1) Earth resistance, earth impedance, and voltage differ-ence can all be measured with one piece of inspection equipment.

(2) Evaluation using a current with a front time of 0.2 μs is possible, to cover approximately 95% of all lightning strikes.

(3) A current with a stable waveform (low oscillation, and reproducible rise form and front time) can be supplied regardless of the surge impedance of the current injec-tion circuit.

(4) The calculation process from the measurement data to the evaluation values is automated, to allow inspec-tions to be carried out even without specialist skills.

(5) The inspection equipment can be carried on a mainte-nance car commonly used by the railway operators.

3.2 Specifications

The developed inspection equipment consists of a pow-er supply unit and a measurement unit. By applying the test circuit configuration shown in Fig. 3, earth resistance, earth impedance, and voltage difference can be encom-passed in one measurement. Figure 4 is an image of the developed inspection equipment.

Figure 5 shows the schematic diagram and specifica-tion of the inspection equipment. By insulating the frame ground (Ef) of the power supply unit from the negative output terminal (C) and positive output terminal (E-c) with a withstand voltage of 40 kV, no additional protective mea-sures are required to protect against electric shocks.

Fig. 3 Circuit configuration for earthing system inspec-tion using developed equipment

Fig. 5 Schematic diagram and specification of developed inspection equipment

Fig. 4 Developed inspection equipment


3.1 Basic concept

In order to overcome the issues faced with the conventional inspection equipment for earthing systems, new inspection equipment was developed based on the following premises: (1) Earth resistance, earth impedance, and voltage difference

can all be measured with one piece of inspection equipment. (2) Evaluation using a current with a front time of 0.2 μs is

possible, to cover approximately 95% of all lightning strikes.

(3) A current with a stable waveform (low oscillation, and reproducible rise form and front time) can be supplied regardless of the surge impedance of the current injection circuit.

(4) The calculation process from the measurement data to the evaluation values is automated, to allow inspections to be carried out even without specialist skills.

(5) The inspection equipment can be carried on a maintenance car commonly used by the railway operators.

3.2 Specifications

The developed inspection equipment consists of a power supply unit and a measurement unit. By applying the test circuit configuration shown in Fig. 3, earth resistance, earth impedance, and voltage difference can be encompassed in one measurement. Figure 4 is an image of the developed inspection equipment.

Figure 5 shows the schematic diagram and specification of the inspection equipment. By insulating the frame ground (Ef) of the power supply unit from the negative output terminal (C) and positive output terminal (E-c) with a withstand voltage of 40 kV, no additional protective measures are required to protect against electric shocks.

The following two measures were introduced into the measurement unit to reduce noise and protect the unit: first, equipotential-bonding was performed between the frame ground (Ef) and internal components. Second, the input terminal (P) connected to the potential probe (P) was insulated from the frame ground with a withstand voltage of 40 kV.

Fig. 3 Circuit configuration for earthing system

inspection using developed equipment


3.3 Features

3.3.1 Power supply unit

A lightning impulse generator was used as the simulated lightning current generator in the power supply unit. A new circuit configuration was designed to exploit the advantages of both the PG and the IG, despite the trade-off between the front time and peak value of current.

The output resistance is adjustable between 1 kΩ and 4 kΩ. Since the output resistance is sufficiently larger than the surge impedance of the earthing system (several tens to several hundreds of ohms), the power supply unit acts as a current source to avoid the output current problems encountered with conventional power supplies, such as waveform oscillation and instability. Since natural lightning can be considered to be a current source [3], this power supply behavior is reasonable from a physical phenomenon point of view. The peak value of the output current is approximately 10 A to 40 A, which is

V

A

～～

Injection point

Terminal point

V

Power supply

～～

distancePotential

probe(P)

Current probe(C)

Measurementunit


Evaluation value RemarkEarth resistance Technical criteriaImpulse impedance Estimating GPRVoltage difference Testing equipotential bonding

Measurement result for evaluationAutomaticcalculation

distance

Power supply

about 0.7m

about 1m

Impulse generator

Control panel

PC basedcontroller

Oscilloscope

Current probe

Measuring unit


Calculation of GPRs

Calculation of voltage difference

Components

OscilloscopeHigh frequency current sensorHigh voltage sensorControl notebook PC

Sampling 1GHz (band width: 20MHz)

Vertical resolution 8 bit

Input voltage Single phase 100V

Input terminal insulation 40kV for voltage input terminals

Input terminals

CH1: currentCH2: E-p1

(potential of injection point)CH3: E-p2

(potential of terminal point)CH4: P

(potential of potential probe)

Max About 100kg

Specification of measurement unit

Type Impulse current generator

Charge circuit Inverter and rectifier

Charge voltage 40kV at maximum

Crest value 40A as maximum

Output resistance 1kW, 2kW, 3kW, 4kW

Front time 0.2s, 0.5s, 1.2s,2.0s, 5.0s


Output terminal insulation

> 40kV for positive and negative terminals

Discharge switch Mechanical switch

Mass Main body: about 40kgControl panel: about 10kg

Specification of power supply


3.1 Basic concept







3.2 Specifications







3.3 Features




V

A

～～

Injection point

Terminal point

V

Power supply

～～

distancePotential

probe(P)

Current probe(C)

Measurementunit




distance


Calculation of GPRs


Components






Input terminals





Max About 100kg






Output resistance 1kΩ, 2kΩ, 3kΩ, 4kΩ

Front time 0.2µs, 0.5µs, 1.2µs,2.0µs, 5.0µs








3.1 Basic concept







3.2 Specifications







3.3 Features




V

A

～～

Injection point

Terminal point

V

Power supply

～～

distancePotential

probe(P)

Current probe(C)

Measurementunit




distance


Calculation of GPRs


Components






Input terminals





Max About 100kg






Output resistance 1kΩ, 2kΩ, 3kΩ, 4kΩ

Front time 0.2µs, 0.5µs, 1.2µs,2.0µs, 5.0µs








3.3 Features


A lightning impulse generator was used as the simu-lated lightning current generator in the power supply unit. A new circuit configuration was designed to exploit the advantages of both the PG and the IG, despite the trade-off between the front time and peak value of current.

The output resistance is adjustable between 1 kΩ and 4 kΩ. Since the output resistance is sufficiently larger than the surge impedance of the earthing system (several tens to several hundreds of ohms), the power supply unit acts as a current source to avoid the output current problems en-countered with conventional power supplies, such as wave-form oscillation and instability. Since natural lightning can be considered to be a current source [3], this power supply behavior is reasonable from a physical phenomenon point of view. The peak value of the output current is ap-proximately 10 A to 40 A, which is approximately the same as that of the conventional IG, and is both necessary and sufficient for measuring both the earth impedance and the voltage difference.

The front time of the output current is adjustable from 0.2 μs to 5.0 μs. The lower limit of the front time of 0.2 μs is larger than the PG (0.05 μs). The reason for this is that it was deemed that a front time of 0.05 μs for the simulated


waveform of the natural lightning current was too severe (See Table 1). The tail time of the output current was de-signed to be over 100 μs, to measure the earth resistance at a steady state.

3.3.2 Measurement unit

The measurement unit consists of an oscilloscope, a control laptop computer (control PC), and other attached devices. In order to reduce noise propagating from the low-voltage power supply line, the low-voltage power supply is disconnected from the power supply unit before the current injection work. The uninterruptible power supply (UPS) supplies the operation power to all the components of the measurement unit during the current injection. The mea-surement unit has two potential measurement terminals for measuring the earth impedance and the voltage differ-ence and one potential terminal for measuring the voltage difference between the potential probe (P) and the frame ground of the measurement unit (Ef). The measurement unit also has a high-frequency current sensor (through-type) to measure the injected current.

In addition, a newly developed algorithm to calculate the evaluation value was implemented in the measurement unit as an automatic evaluation function. The algorithm is as follows:(1) Since the measurement data, which is a waveform in

the time domain, generally contains noise, the mea-sured data is processed through a digital second-order low-pass filter.

(2) The time region to be analyzed, where the current value is large enough to negate the noise, is calculated by the waveform particularly around the front of the impulse current.

(3) The offsets of the measurement unit to be subtracted from the evaluation results are calculated by the mea-sured data.

(4) The earth impedance zE (t) is calculated dividing the GPR at the injection point v1 (t) by the injection cur-rent i(t) (t: time). The impulse impedance is calculated as the peak value of zE (t). The earth resistance is cal-culated as the convergence value of zE (t).

(5) The voltage difference v12 (t) is calculated from the difference between v1 (t) and the GPR at the terminal point v2 (t).

3.4 Verification of measurement accuracy

The measurement accuracy of the developed inspec-tion equipment was verified in field tests. The verification was performed at three substations (A, B and C as shown in Table 2) on a DC electrified line (single track), which had suffered severe winter lightning damage. As shown in the table, the surrounding soil properties of each substa-tion were dry, general and wetland respectively. In order to improve lightning protection by reducing earth imped-ance, conductive concrete was used to fill the ground under each substation around the down conductor of the surge arrester for receiving line. The verification was carried out by comparing measurements taken using the developed inspection equipment with the results of an inspection car-ried out on completion of work, by a railway constructor or

an inspection company.The test circuit was configured as shown in Fig. 1. The

testing impulse current was injected through the down conductor of phase S of the surge arrester for the receiving line. The distance between the testing earthing system and the potential probe P was about 50 m to 100 m, and that between the earthing system and the current probe C was about 10 m to 50 m, respectively. The terminal point for the voltage difference measurement was the point near the rectifier approximately 10 m away from the injection point.

3.4.1 Current waveform

The stability (reproducibility and oscillation) verifica-tion of the waveform of the output impulse current of the inspection equipment was carried out by comparing the current waveforms injected at the testing substations (A, B and C). Figure 6 shows the waveforms of the output current under the strictest condition where the front time (expected) is set to 0.2 μs. When the output resistance was 1 kΩ, the waveforms from 0 μs (the beginning of discharge) to about 3 μs were different for each substation, and the ac-tual front times were from approximately 0.2 μs to 1.0 μs.

Table 2 Substations for testing

Substation Line Main power converter(s) Earthresistivity

ASingle

track with DC electric

traction

A rectifier (3000 kW) anda distribution transformer (500 kVA)

820 Ω・m(dry soil)

B A rectifier (4000 kW) 80Ω・m(general)

C A rectifier (4000 kW) 8.2 Ω・m(wetland soil)

Fig. 6 Output current waveform of inspection equipment (expected front time: 0.2μs)

(a) Output resistance: 1kΩ

(b) Output resistance: 2kΩ

(c) Output resistance: 4kΩ

Fig. 6 Output current waveform of inspection equipment (expected front time: 0.2μs)

3.4.2 Earth resistance

The accuracy of the earth resistance evaluated with the inspection equipment (the evaluation values) was verified by comparing the obtained values with the results of inspections carried out when constructions of the substations were originally completed using the fall-of-potential method (the reference values), as shown in Table 3.

The evaluation and reference values agreed well regardless of the output resistance value at substation A, where the earth resistance was the highest among the three substations. On the other hand, the error between the two values seemed to be larger in substations B and C, where the earth resistances were lower than substation A, because of the limit of the resolution of the oscilloscope (8 bits) in the inspection equipment. In substations B and C, because the order of impulse impedance and earth resistance differed in seconds in the order of magnitude, the vertical resolution insufficient for the earth resistance.

However, the accuracy described above for earth resistance is enough for a confirmation, such as in the case of a maintenance inspection, to the limit set by technical criteria (such as 10 Ω). Accordingly, it was confirmed that the inspection equipment was sufficiently applicable for measuring the earth resistance in substation maintenance inspections.

Table 3 Earth resistances evaluation results Developed inspection equipment Completion

inspection (reference)

Output resistance 1 kW 2 kW 3 kW 4 kW

Substation A 4.2 W 4.4 W 3.9 W 4.1 W 4.33 W Substation B 0.0 W 0.6 W 0.3 W 0.2 W 0.39 W Substation C 0.0 W 0.1 W 0.2 W 0.2 W 0.18 W

3.4.3 Impulse impedance

The accuracy of impulse impedance evaluations using the inspection equipment, which is the peak value of earth impedance, was verified through comparison with the measurement results using PG by an inspection company (the reference values) as shown in Table 4.

As a result, when the output resistance was about 1 kΩ to 2 kΩ, the evaluation and reference values agreed well with a maximum error of approximately 20%, when the output resistance was 1 kΩ or 2 kΩ.

Table 4 Earth impedances evaluation results

Developed inspection equipment Inspection company

(reference) Output

resistance 1 kW 2 kW 3 kW 4 kW

Substation A 43.7 W 50.1 W 60.5 W 53.3 W ― Substation B 40.4 W 40.9 W 39.6 W 39.7 W 37.6 W Substation C 37.2 W 43.4 W 49.6 W 55.5 W 42.0 W

3.4.4 Voltage difference

Table 5 compares voltage differences measured using the inspection equipment and by the inspection company using IG. The current front time of the inspection device was set to 2.0 µs, the same value as IG. The relative values obtained by dividing the peak value of the voltage difference by the peak value of the injection current were used for the comparison, since the peak values of current were not the same. The evaluation and reference values agreed with a maximum error of approximately 20% to 50% in substation B and C.

Examining the relationship between the output resistances and the evaluation values, the evaluation values tended to increase as the output resistance increased in substations A and B, while this tendency was not found in substation C. The reason for the difference is considered to be that the power supply unit with an output resistance of 3 kΩ or more (peak current is less than 10 A) cannot inject the impulse current into the whole earthing systems of substation A and B, which have larger earth resistivities and lower slopes of GPR than substation C.

In addition, the inspection equipment can supply an impulse current with a front time of 0.2 μs at minimum, which is one tenth of that of the IG. While the IG can cover approximately 50% of all lightning strikes, the developed inspection equipment is expected to cover approximately 95%. In other words, the inspection equipment is expected to perform tests in more sever conditions than with the IG. Table 6 shows the evaluated voltage differences obtained using a current with the minimum front time (0.2 μs). The evaluation values

※ Reference values from an inspection company are given in [1]. ※ Current front time of inspection equipment is about 0.2µs, and that of PG of the inspection company is about 0.05µs. ※ Methods used to define impulse impedance in this paper and that of the inspection company are different.


When the output resistance was 2 kΩ, the front time stabi-lized around the expected value of 0.2 μs, and the instabil-ity of waveforms was lower in comparison to the previous condition although there were some vibration components. By increasing the output resistance to 4 kΩ, the stability of the waveforms was improved compared to that of 2 kΩ.


The accuracy of the earth resistance evaluated with the inspection equipment (the evaluation values) was veri-fied by comparing the obtained values with the results of inspections carried out when constructions of the substa-tions were originally completed using the fall-of-potential method (the reference values), as shown in Table 3.

The evaluation and reference values agreed well re-gardless of the output resistance value at substation A, where the earth resistance was the highest among the three substations. On the other hand, the error between the two values seemed to be larger in substations B and C, where the earth resistances were lower than substation A, because of the limit of the resolution of the oscilloscope (8 bits) in the inspection equipment. In substations B and C, because the order of impulse impedance and earth re-sistance differed in seconds in the order of magnitude, the vertical resolution insufficient for the earth resistance.

However, the accuracy described above for earth re-sistance is enough for a confirmation, such as in the case of a maintenance inspection, to the limit set by technical criteria (such as 10 Ω). Accordingly, it was confirmed that the inspection equipment was sufficiently applicable for measuring the earth resistance in substation maintenance inspections.

Table 3 Earth resistances evaluation resultsDeveloped inspection equipment Completion

inspection (reference)

Output resistance 1 kΩ 2 kΩ 3 kΩ 4 k Ω

Substation A 4.2 Ω 4.4 Ω 3.9 Ω 4.1 Ω 4.33 ΩSubstation B 0.0 Ω 0.6 Ω 0.3 Ω 0.2 Ω 0.39 ΩSubstation C 0.0 Ω 0.1 Ω 0.2 Ω 0.2 Ω 0.18 Ω


The accuracy of impulse impedance evaluations using the inspection equipment, which is the peak value of earth impedance, was verified through comparison with the mea-surement results using PG by an inspection company (the reference values) as shown in Table 4.

As a result, when the output resistance was about 1 kΩ to 2 kΩ, the evaluation and reference values agreed well with a maximum error of approximately 20%.

Table 4 Earth impedances evaluation resultsDeveloped inspection equipment Inspection

company (reference)

Output resistance 1 kΩ 2 kΩ 3 kΩ 4 kΩ

Substation A 43.7 Ω 50.1 Ω 60.5 Ω 53.3 Ω ―Substation B 40.4 Ω 40.9 Ω 39.6 Ω 39.7 Ω 37.6 ΩSubstation C 37.2 Ω 43.4 Ω 49.6 Ω 55.5 Ω 42.0 Ω

※ Reference values from an inspection company are given in [1]. ※ Current front time of inspection equipment is about 0.2μs, and that of PG of the inspection company is about 0.05μs. ※ Methods used to define impulse impedance in this paper and that of the inspection company are different.


Table 5 compares voltage differences measured using the inspection equipment and by the inspection company using IG. The current front time of the inspection device was set to 2.0 μs, the same value as IG. The relative values obtained by dividing the peak value of the voltage differ-ence by the peak value of the injection current were used for the comparison, since the peak values of current were not the same. The evaluation and reference values agreed with a maximum error of approximately 20% to 50% in substation B and C.

Examining the relationship between the output resis-tances and the evaluation values, the evaluation values tended to increase as the output resistance increased in substations A and B, while this tendency was not found in substation C. The reason for the difference is considered to be that the power supply unit with an output resistance of 3 kΩ or more (peak current is less than 10 A) cannot in-ject the impulse current into the whole earthing systems of substation A and B, which have larger earth resistivities and lower slopes of GPR than substation C.

In addition, the inspection equipment can supply an impulse current with a front time of 0.2 μs at minimum, which is one tenth of that of the IG. While the IG can cover approximately 50% of all lightning strikes, the developed inspection equipment is expected to cover approximately 95%. In other words, the inspection equipment is expected to perform tests in more sever conditions than with the IG. Table 6 shows the evaluated voltage differences obtained using a current with the minimum front time (0.2 μs). The evaluation values shown in Table 6 were 2 to 3 times of those of Table 5 with a front time of 2.0 μs. Accordingly, it was confirmed that the inspection equipment can be used to make evaluations under more severe conditions than with the IG.

Table 5 Voltage differences evaluation results with cur-rent front time of 2.0 μs


(reference)Output

resistance 1 kΩ 2 kΩ 3 kΩ 4 kΩ

Substation A 11.5 V/A 13.9 V/A 15.5 V/A 16.0 V/A ―Substation B 12.8 V/A 13.8 V/A 15.4 V/A 16.9 V/A 9.5 V/ASubstation C 8.8 V/A 11.3 V/A 11.7 V/A 9.0 V/A 7.2 V/A

※ Reference values from an inspection company are given in [1].

Table 6 Voltage differences evaluation results with cur-rent front time of 0.2 μs


(reference)Output


Substation A 34.7 V/A 40.2 V/A 50.3 V/A 45.7 V/A ―Substation B 32.1 V/A 35.9 V/A 37.8 V/A 40.7 V/A ―Substation C 18.6 V/A 20.0 V/A 20.9 V/A 19.9 V/A ―

4. Proposal of new measurement circuit

4.1 Circuit structure

A new measurement circuit structure for earthing sys-tem inspections was proposed without temporary test elec-trodes and temporary measurement lines. Figure 7 shows the proposed circuit structure, whose equivalent circuit is


the same as the conventional circuit shown in Fig. 1. The differences with the conventional circuit are as follows: the earth pole for remote terminal unit (RTU) of the tested substation is used as the potential probe (P), the earthing system of the next substation is used as the current probe (C), and the overhead contact line system is used as an earth return path in the current injection circuit. There-fore, no temporary devices are required for the proposed circuit.

Yamashita H. et al. previously reported on experimen-tal examination results using a similar circuit structure for an earthing system in a Shinkansen traction substation [4]. In this previous study two adjacent fixed power sup-ply installations were used as the test electrodes (P and C). However, they reported that it was difficult to use the earthing system of the adjacent installation as the poten-tial probe (P), due to large interference (mutual resistance) in the potential measurement circuit [4].

The improvement in this proposal in relation to the previous study is that the earth pole for RTU is used as the potential probe (P), and the target frequency range is ex-tended from power frequencies to 1 MHz of lightning cur-rent using the same circuit structure.


The measurement accuracy of the proposed mea-surement circuit was verified in field tests in the above-mentioned substations A, B, and C. The verification was carried out by comparing the evaluation results of the pro-posed circuit with those of the conventional circuit shown in Fig. 1. The earth resistance, impulse impedance and voltage difference were measured using the developed test equipment in each circuit structure. The front time of the injection current was set to 0.2 μs.


Table 7 compares the evaluated earth resistance values obtained with the conventional and proposed (new) circuit structures. As shown in Table 7, the values obtained with the inspection equipment using the conventional and pro-posed circuits agreed approximately. This demonstrates that the interference between the potential measurement circuit and the earth pole for RTU, which was a concern in the previous study [4], was sufficiently small in the low fre-quency range. Accordingly, the proposed circuit structure could be applied for measuring earth resistance.

Fig. 7 New earthing system inspection circuit structure

shown in Table 6 were 2 to 3 times of those of Table 5 with a front time of 2.0 μs. Accordingly, it was confirmed that the inspection equipment can be used to make evaluations under more severe conditions than with the IG.

Table 5 Voltage differences evaluation results

with current front time of 2.0 µs Developed inspection equipment Inspection

company (reference)

Output resistance 1 kW 2 kW 3 kW 4 kW

Substation A 11.5 V/A 13.9 V/A 15.5 V/A 16.0 V/A ― Substation B 12.8 V/A 13.8 V/A 15.4 V/A 16.9 V/A 9.5 V/A Substation C 8.8 V/A 11.3 V/A 11.7 V/A 9.0 V/A 7.2 V/A

Table 6 Voltage differences evaluation results with current front time of 0.2 µs


(reference) Output


Substation A 34.7 V/A 40.2 V/A 50.3 V/A 45.7 V/A ― Substation B 32.1 V/A 35.9 V/A 37.8 V/A 40.7 V/A ― Substation C 18.6 V/A 20.0 V/A 20.9 V/A 19.9 V/A ―

4. Proposal of new measurement circuit

4.1 Circuit structure

A new measurement circuit structure for earthing system inspections was proposed without temporary test electrodes and temporary measurement lines. Figure 7 shows the proposed circuit structure, whose equivalent circuit is the same as the conventional circuit shown in Fig. 1. The differences with the conventional circuit are as follows: the earth pole for remote terminal unit (RTU) of the tested substation is used as the potential probe (P), the earthing system of the next substation is used as the current probe (C), and the overhead contact line system is used as an earth return path in the current injection circuit. Therefore, no temporary devices are required for the proposed circuit.

Yamashita H. et al. previously reported on experimental examination results using a similar circuit structure for an earthing system in a Shinkansen traction substation [4]. In this previous study two adjacent fixed power supply installations were used as the test electrodes (P and C). However, they reported that it was difficult to use the earthing system of the adjacent installation as the potential probe (P), due to large interference (mutual resistance) in the potential measurement circuit [4].

The improvement in this proposal in relation to the previous study is that the earth pole for RTU is used as the potential probe (P), and the target frequency range is extended from power frequencies to 1 MHz of lightning current using the same circuit structure.

Fig. 7 New earthing system inspection circuit

structure


The measurement accuracy of the proposed measurement circuit was verified in field tests in the above-mentioned substations A, B, and C. The verification was carried out by comparing the evaluation results of the proposed circuit with those of the conventional circuit shown in Fig. 1. The earth resistance, impulse impedance and voltage difference were measured using the developed test equipment in each circuit structure. The front time of the injection current was set to 0.2 µs.


Table 7 compares the evaluated earth resistance values obtained with the conventional and proposed (new) circuit structures. As shown in Table 7, the values obtained with the inspection equipment using the conventional and proposed circuits agreed approximately. This demonstrates that the interference between the potential measurement circuit and the earth pole for RTU, which was a concern in the previous study [4], was sufficiently small in the low frequency range. Accordingly, the proposed circuit structure could be applied for measuring earth resistance.

Table 7 Earth resistance evaluation results Output


Circuit Conv. New Conv. New Conv. New Conv. New Substation A 4.2 W 4.3 W 4.4 W 4.4 W 3.9 W 4.7 W 4.1 W 4.1 W Substation B 0.0 W 1.2 W 0.6 W 1.1 W 0.3 W 1.0 W 0.2 W 1.1 W Substation C 0.0 W 0.6 W 0.1 W 0.5 W 0.2 W 0.5 W 0.2 W 0.4 W


Table 8 compares the evaluated impulse impedance values obtained with the conventional and proposed circuits, where the current front time was set to 0.2 µs. The evaluation values using the conventional and proposed circuits were almost the same. This demonstrates that the interference reported in the previous study [4] is sufficiently small even in the high frequency range concomitant with a lightning strike. Therefore, the proposed circuit structure can also be applied to the measurement of earth impedance.

Since the proposed circuit structure uses an overhead contact line system, which is much longer than the conventional measurement circuit, as the earth return path, the attenuation of the traveling wave is expected to be higher than that of the conventional circuit. There were some cases using the

CB

Substation under test Next substation

Overhead contact line

Surgearrester

Disconnected

DS CBDS

Earthing systemunder testing

V

Used as a currentinjection circuit

Current flowin the ground

Earthing system

Earth polefor RTU

> 50m

Rectifier

A

Current flowin OCS

Used as acurrent probe

Used as a potentialprobe

※ Reference values from an inspection company are given in [1].

Table 7 Earth resistance evaluation resultsOutput


Circuit Conv. New Conv. New Conv. New Conv. NewSubstation A 4.2 Ω 4.3 Ω 4.4 Ω 4.4 Ω 3.9 Ω 4.7 Ω 4.1 Ω 4.1 ΩSubstation B 0.0 Ω 1.2 Ω 0.6 Ω 1.1 Ω 0.3 Ω 1.0 Ω 0.2 Ω 1.1 ΩSubstation C 0.0 Ω 0.6 Ω 0.1 Ω 0.5 Ω 0.2 Ω 0.5 Ω 0.2 Ω 0.4 Ω


Table 8 compares the evaluated impulse impedance values obtained with the conventional and proposed cir-cuits, where the current front time was set to 0.2 μs. The evaluation values using the conventional and proposed circuits were almost the same. This demonstrates that the interference reported in the previous study [4] is suffi-ciently small even in the high frequency range concomitant with a lightning strike. Therefore, the proposed circuit structure can also be applied to the measurement of earth impedance.

Since the proposed circuit structure uses an overhead contact line system, which is much longer than the conven-tional measurement circuit, as the earth return path, the attenuation of the traveling wave is expected to be higher than that of the conventional circuit. There were some cas-es using the conventional circuit structure where the peak value of reflected wave of the earth impedance was higher than that of the first traveling wave (the impulse imped-ance to be evaluated). The inspection equipment incorrect-ly evaluated the values of the impulse impedances in these cases as shown as the blue curve in Fig. 8(a). However, by using the proposed circuit structure, this problem was abated, as shown with the red curve in the figure, due to a reduction in the reflected wave.

Table 8 Earth impedance evaluation resultsOutput


Circuit Conv. New Conv. New Conv. New Conv. NewSubstation A 43.7 Ω 41.4 Ω 50.1 Ω 44.8 Ω 60.5 Ω 48.1 Ω 53.3 Ω 51.2 ΩSubstation B 40.4 Ω 36.5 Ω 40.9 Ω 36.9 Ω 39.6 Ω 38.3 Ω 39.7 Ω 40.1 ΩSubstation C 37.2 Ω 33.2 Ω 43.4 Ω 35.2 Ω 49.6 Ω 35.2 Ω 55.5 Ω 37.8 Ω

※ Current front time of inspection equipment was set to 0.2μs.


Table 9 shows the comparative results of the peak value of voltage difference using the conventional and proposed circuits, where the current front time was set to 0.2 μs. As with the earth impedance, the evaluation values of the conventional and proposed circuits were almost the same under all conditions. Therefore, it was confirmed that the proposed circuit structure is applicable to the measure-ment of voltage difference. Figure 8 (b) shows an example of the waveform of the potential difference measured in substation A.

Table 9 Voltage difference evaluation resultsOutput


Circuit Conv. New Conv. New Conv. New Conv. NewSubstation A 34.7 V/A 26.5 V/A 40.2 V/A 32.1 V/A 50.3 V/A 37.2 V/A 45.4 V/A 40.8 V/ASubstation B 32.1 V/A 26.0 V/A 35.9 V/A 30.0 V/A 37.8 V/A 34.0 V/A 40.7 V/A 37.2 V/ASubstation C 18.6 V/A 20.3 V/A 20.0 V/A 22.2 V/A 20.9 V/A 22.3 V/A 19.9 V/A 17.7 V/A

※ Current front time of inspection equipment was set to 0.2μs.


Fig. 8 Comparison of waveform results in substation A with output resistance of 2 kΩ

Authors

　　　　　

Gaku MORITASenior Researcher, Power Supply SystemsLaboratory, Power Supply TechnologyDivisionResearch Areas: Lightning Protection andEarthing Systems, Electric Equipment

　　　　　

Hiroki TANAKA, Dr.Eng.Deputy Manager, Electrical EngineeringDepartment, West Japan Railway CompanyResearch Areas: Lightning Protection andEarthing Systems, Remote Monitoring

　　　　　

Koji MICHISHITA, Dr.Eng.Professor, Dept. of Electrical and ElectronicEng., Shizuoka UniversityResearch Areas: Lightning, LightningProtection, Lightning Location System

5. Conclusions

New inspection equipment and a new circuit structure for earthing systems for fixed power supply installations were developed. The developed inspection equipment can evaluate conformity of earth resistance of an earthing system with given technical criteria and lightning resis-tance of the earthing system. Results of tests to verify the method in field tests confirmed that the inspection equip-ment can be used to measure and evaluate both the low-frequency parameter (earth resistance) and high-frequency parameters (earth impedance and voltage difference) with sufficient accuracy.

The proposed test circuit structure employs the earth-ing system of adjacent substations and earth poles for RTUs instead of test electrodes, and does not require any

temporary devices to be used. Field test results confirmed that the earth resistance, earth impedance and voltage dif-ference could be measured and evaluated with sufficient accuracy.

The results of this paper can be applied for mainte-nance inspections of earthing systems, verifying the ef-fectiveness of lightning protection countermeasures, and prioritizing protection countermeasures.

Acknowledgements

The authors would like to express their sincere thanks to Mr. Y. Koga of the Otowa Electric Co. for his cooperation.

References

[1] Maniwa K., Nobuta K., Kobara T., Tahara T., Wakat-suki Y., Ito K., Tanaka H., Yamanoi T., Kawahara K.,

“A study of rise in potential by the lightning surge re-sponse on traction substation with grounding mesh,” Workshop of the Institute of Electrical Engineers of Japan, TER-14-036, 2014 (in Japanese).

[2] IEEE Std. 81-2012, “IEEE Guide for Measuring Earth Resistivity, Ground Impedance, and Earth Surface Po-tentials of a Grounding System,” 2012.

[3] “Lightning Protection for Electrical and Electronic Equipment,” Institute of Electrical Installation Engi-neers of Japan, 2011.

[4] Yamashita H., Tanaka H., Akagi M., Morita G., Taka-hashi K., Shoda T., “Studies of subject in Ground Re-sistance Measurements using AT Feeding Circuit for Realization,” 2014 IEE-Japan Industry Applications Society Conference, the Institute of Electrical Engi-neers of Japan, 2014 (in Japanese).

conventional circuit structure where the peak value of reflected

wave of the earth impedance was higher than that of the first

traveling wave (the impulse impedance to be evaluated). The

inspection equipment incorrectly evaluated the values of the

impulse impedances in these cases as shown as the blue curve

in Fig. 8(a). However, by using the proposed circuit structure,

this problem was abated, as shown with the red curve in the

figure, due to a reduction in the reflected wave.

Table 8 Earth impedance evaluation results Output

resistance 1 kΩ 2 kΩ 3 kΩ 4 kΩ Circuit Conv. New Conv. New Conv. New Conv. New

Substation A 43.7 Ω 41.4 Ω 50.1 Ω 44.8 Ω 60.5 Ω 48.1 Ω 53.3 Ω 51.2 Ω Substation B 40.4 Ω 36.5 Ω 40.9 Ω 36.9 Ω 39.6 Ω 38.3 Ω 39.7 Ω 40.1 Ω Substation C 37.2 Ω 33.2 Ω 43.4 Ω 35.2 Ω 49.6 Ω 35.2 Ω 55.5 Ω 37.8 Ω


Table 9 shows the comparative results of the peak value of

voltage difference using the conventional and proposed circuits,

where the current front time was set to 0.2 µs. As with the

earth impedance, the evaluation values of the conventional and

proposed circuits were almost the same under all conditions.

Therefore, it was confirmed that the proposed circuit structure

is applicable to the measurement of voltage difference. Figure

8 (b) shows an example of the waveform of the potential

difference measured in substation A.

(a) Earth impedance

(b) Voltage difference

Fig. 8 Comparison of waveform results in

substation A with output resistance of 2 kΩ

Table 9 Voltage difference evaluation results Output

resistance 1 kΩ 2 kΩ 3 kΩ 4 kΩ Circuit Conv. New Conv. New Conv. New Conv. New

Substation A 34.7 V/A 26.5 V/A 40.2 V/A 32.1 V/A 50.3 V/A 37.2 V/A 45.4 V/A 40.8 V/A Substation B 32.1 V/A 26.0 V/A 35.9 V/A 30.0 V/A 37.8 V/A 34.0 V/A 40.7 V/A 37.2 V/A Substation C 18.6 V/A 20.3 V/A 20.0 V/A 22.2 V/A 20.9 V/A 22.3 V/A 19.9 V/A 17.7 V/A

5. Conclusions

New inspection equipment and a new circuit structure for

earthing systems for fixed power supply installations were

developed. The developed inspection equipment can evaluate

conformity of earth resistance of an earthing system with given

technical criteria and lightning resistance of the earthing system.

Results of tests to verify the method in field tests confirmed that

the inspection equipment can be used to measure and evaluate

both the low-frequency parameter (earth resistance) and high-

frequency parameters (earth impedance and voltage difference)

with sufficient accuracy.

The proposed test circuit structure employs the earthing

system of adjacent substations and earth poles for RTUs instead

of test electrodes, and does not require any temporary devices

to be used. Field test results confirmed that the earth

resistance, earth impedance and voltage difference could be

measured and evaluated with sufficient accuracy.

The results of this paper can be applied for maintenance

inspections of earthing systems, verifying the effectiveness of

lightning protection countermeasures, and prioritizing

protection countermeasures.

Acknowledgments

The authors would like to express their sincere thanks to

Mr. Y. Koga of the Otowa Electric Co. for his cooperation.

References

[1] Maniwa K., Nobuta K., Kobara T., Tahara T., Wakatsuki Y.,

Ito K., Tanaka H., Yamanoi T., Kawahara K., “A study of rise in

potential by the lightning surge response on traction substation

with grounding mesh,” workshop of the Institute of Electrical

Engineers of Japan, TER-14-036, 2014 (in Japanese).

[2] IEEE Std. 81-2012, “IEEE Guide for Measuring Earth

Resistivity, Ground Impedance, and Earth Surface Potentials of

a Grounding System,” 2012.

[3] “Lightning Protection for Electrical and Electronic

Equipment,” Institute of Electrical Installation Engineers of

Japan, 2011.

[4] Yamashita H., Tanaka H., Akagi M., Morita G., Takahashi

K., Shoda T., “Studies of subject in Ground Resistance

Measurements using AT Feeding Circuit for Realization,” 2014

IEE-Japan Industry Applications Society Conference, the

Institute of Electrical Engineers of Japan, 2014 (in Japanese).

Authors

Gaku MORITA

Senior Researcher, Power Supply Systems

Laboratory, Power Supply Technology

Division

Research Areas: Lightning Protection and

Earthing Systems, Electric Equipment

Hiroki TANAKA, Dr.Eng.

Deputy Manager, Electrical Engineering

Department, West Japan Railway Company

Research Areas: Lightning Protection and

Earthing Systems, Remote Monitoring

0102030405060

0 2 4 6 8 10Eart

h im

peda

nce

(Ω)

Time (µs)

Conventional circuitProposed circuit

Impulse impedance evaluated by proposed circuit

Impulse impedance evaluated by conventional circuit

-60-40-20

0204060

0 2 4 6 8 10Volta

ge d

iffer

ence

(V/A

)

Time (µs)

Conventional circuit

Proposed circuit

※ Current front time of inspection equipment was set to 0.2µs.

※ Current front time of inspection equipment was set to 0.2µs.

Documents

Development of New Earthing System ... - J-STAGE Home