ROGOWSKI COIL SENSOR FOR PARTIAL DISCHARGE DETECTION
PUOVIN MURUGAIAH
UNIVERSITI SAINS MALAYSIA
2018
ROGOWSKI COIL SENSOR FOR PARTIAL DISCHARGE DETECTION
by
PUOVIN MURUGAIAH
Thesis submitted in partial fulfilment of the requirements for the degree of Bachelor of Engineering
(Electrical Engineering)
JUNE 2018
i
ACKNOWLEDGEMENTS
I would like to express my sincere gratitude to my project supervisor Assoc.Prof.Ir.Dr. Mohamad Kamarol bin Mohd Jamil for being the pillar of support for me in completing my project successfully. His support, patience, motivation, enthusiasm and immense knowledge helped me to focus throughout the whole project.
I also want to thank my dad, Murugaiah A/L Muniandy, my mother, Suganthi A/P Maruthamuthu and my siblings who has given me continuous moral support and motivation throughout the period the of my studies.Without all of them, I wouldn’t able to complete this project smoothly and successfully.
ii
TABLE OF CONTENTS
ACKNOWLEDGMENT i
LIST OF TABLES iv
LIST OF FIGURES v
LIST OF ABBREVIATION vii
ABSTRAK viii
ABSTRACT ix
CHAPTER 1: INTRODUCTION
1.1 Research Background 1 1.2 Problem Statement 3 1.3 Objective 41.4 Scope of Work 4 1.5 Thesis Overview 4
CHAPTER 2: LITERATURE REVIEW
2.1 Introduction 6 2.2 Partial Discharge 62.3 Partial Discharge Detection 7 2.4 Working Principle of Rogowski Coil (RC) 10 2.5 Geometrical Parameters 12
iii
CHAPTER 3: METHODOLOGY
3.1 Introduction 15
3.2 Flow of Research 15
3.3 Development of Rogowski Coil 17
3.4 Geometrical Parameters of Rogowski Coil 18
3.5 Characteristic Determined by Geometrical Parameters 20 3.6 Circuit Design 23
CHAPTER 4:
RESULT AND DISCUSSION
4.1 Introduction 30
4.2 Results of PD Measurement 30
4.3 Analysis 40
4.4 Discussion 42
4.5 Summary 43
CHAPTER 5:
CONCLUSION AND RECOMMENDATION
5.1 Conclusion 44
5.2 Recommendation for Future Works 44
REFERENCES 46
iv
LIST OF TABLES
Table 1.1: Requirements of High Frequency Pulse Current Measuring Sensors 2
Table 2.1: Advantages and Disadvantages of RC 10
Table2.3: Recommendations for RC geometrical parameters 14 Table 3.1: Geometrical parameters used in this research 20 Table 4.1: PD Input Voltage and Output Peak Value for each Set 40 Table 4.2: PD Input Charge and Output Peak Value for each Set 41
v
LIST OF FIGURE
Figure 2.1: RC with various diameters 9
Figure 2.2: RC mounted around conductor 12
Figure 2.3: Geometrical parameters of RC 13
Figure 3.1: Flowchart of research 16
Figure 3.2: Electrical equivalent model for Rogowski coil 17
Figure 3.3: Constructed RC 18
Figure 3.4: Return Conductor 19
Figure 3.5: BNC connector with RC 19
Figure 3.6: PCB placed in the metal box with its casing 24 Figure 3.7: Covered metal box with its terminals labelled 24
Figure 3.8: Amplifier circuit 25
Figure 3.9: Working principle of high pass filter 26
Figure 3.10: Capacitor and resistor of high pass filter 26 Figure 3.11: Amplifier circuit with high pass filter and series 50Ω 27 Figure 3.12: Amplifier circuit with 50Ω in parallel 28 Figure 3.13: Amplifier circuit with 1kΩ in parallel 29
vi
Figure 4.1: PD measurement of RC with circuit A 31
Figure 4.2: PD measurement of RC with circuit A 32
Figure 4.3: PD measurement of RC with circuit A 32
Figure 4.4: PD measurement of RC with circuit A 33
Figure 4.5: PD measurement of RC with circuit B 34
Figure 4.6: PD measurement of RC with circuit B 35
Figure 4.7: PD measurement of RC with circuit B 35
Figure 4.8: PD measurement of RC with circuit B 36
Figure 4.9: PD measurement of RC with circuit C 37
Figure 4.10: PD measurement of RC with circuit C 38
Figure 4.11: PD measurement of RC with circuit C 38
Figure 4.12: PD measurement of RC with circuit C 39
Figure 4.13: Relationship between actual PD and measured voltage 40 for Circuit A, Circuit B and Circuit C
Figure 4.14: Relationship between actual PD charge and measured voltage 41 for Circuit A, Circuit B and Circuit C
vii
LIST OF ABBREVIATIONS
RC Rogowski Coil
CCC Current-carrying conductor
PD Partial Discharge
CT Current Transformer
viii
ABSTRAK
Istilah nyahcaj merujuk kepada ancaman yang boleh menyebabkan kelesuan pada penebatnya di zon medan elektrik yang sangat tinggi. Oleh itu, untuk mengelakkan kelesuan berlaku, pemerhatian yang kerap dilakukan dengan menggunakan kaedah pengesanan yang tepat. Pengukuran nyahcaj jenis ini adalah penting kerana kewujudannya berkaitan dengan penuaan penebat dalam peralatan elektrik. Pengukuran PD biasanya digunakan untuk menilai kualiti penebat peralatan voltan tinggi. Terdapat banyak alat pengesan PD yang ada di pasaran, dan pengesan menggunakan banyak teknik yang berbeza dalam mengukur PD. Makalah ini membentangkan sistem pengesanan PD peranti mudah alih dan kos rendah dengan penggunaan gegelung Rogowski (RC).
Kekerapan frekuensi tinggi kerana PD dikesan oleh RC. RC dibangunkan untuk julat frekuensi tinggi dari satu MHz sehingga pesanan seratus MHz. Analisis eksperimen dilakukan untuk menilai prestasi gegelung Rogowski dan kesan litar yang dirancang pada bentuk gelombang RC untuk pengesanan PD. Eksperimen diatur dalam makmal voltan tinggi. Ujian dijalankan mengenai keadaan pengukuran seperti sistem pengukur PD dalam talian. Isyarat PD yang dijana oleh penjana pulse dilalukan melalui pusat RC di mana keluarannya telah disambungkan ke litar yang dirancang. Pengukuran dibuat dengan menggunakan osiloskop untuk mendapatkan bentuk gelombang yang dihasil.
Reka bentuk litar dilakukan untuk meningkatkan output RC untuk sistem pengesanan yang lebih baik serta mendapatkan isyarat yang mereplikasi isyarat PD sebenar. Daripada eksperimen ini disiasat bahawa kaedah pengesanan RC ini berfungsi secara linear dan dengan kepekaan tinggi ke input terendah (5pC).
ix
ABSTRACT
The term partial discharge refers to the partial breakdown of insulation that develops in zones of highly concentrated electric fields. So, to prevent the breakdown occurs frequent observations must be done by utilising precise detection methods. The measurement of this kind of discharges is of importance because its existence is related to the aging of insulation in electric equipment. Partial discharge (PD) test is normally used to evaluate the insulation quality of high-voltage equipment. There are many PD detectors available in the market, and the detectors employ many techniques in measuring the PD. This paper presents a simplified and low cost device PD detection system with the use of the Rogowski coil (RC). The high frequency pulses due to the PD are detected by the RC.
RC is developed for high frequency range from one MHz up to order of hundred MHz’s.
An experimental analysis is performed to evaluate Rogowski coil performance and the effect of designed circuit on output waveform of RC sensor for PD detection. The experimental set-up is arranged in high voltage laboratory. The tests are carried out concerning measurement conditions such as on-line PD measuring systems. Input PD signal generated by pulse generator is passed through the centre of RC where the output of it is already connected to the designed circuit. The measurement is made by using oscilloscope to capture the output waveform. The circuit design is done to improve the RC output for better detection system as well as to obtain the signal that is replicates the input PD signal. From the experiment it is investigated that this RC detection method works linearly and with high sensitivity to the lowest input (5pC).
1
CHAPTER 1
INTRODUCTION
1.1 Backgroud
The usage of high voltage equipment determines the stability of any power system network [1]. This has become a major concern to a handful of distribution companies since the requirements of power supply networks and optimization of network assets has increased significantly. It is to be noted that, failures of equipment are commonly related to the activity of partial discharge (PD). [2]
PD are localized electric discharges in a partial region of a liquid or solid electrical insulation system under high- voltage field pressure which deteriorates the system performance and can lead to breakdowns, fires or irreparable damage to the system [3].
PD is a small start of flash that happens inside the insulation of electrical equipment.
There are currently three types of PD that are commonly known today which are corona discharge, surface discharge and internal discharge [3]. Each PD event generates low amplitude pulses of extremely short duration. These pulses can be observed as voltage and current pulses, superimposed on the mains voltage and current. Due to the very short rise time, the spectrum of such transients can reach well into the radio frequency region.
This is the reason partial discharge detection is used as a part of energy system to monitor the high voltage equipment’s health condition so that it is not exposed to any unnecessary hazards [1]. A sensor for detecting and accurately quantifying the pulses is required to have a wide operating bandwidth and high sensitivity. Table 1.1 shows some of the requirements of an ideal high frequency pulse current measuring sensors [2].
2
Table 1.1: Requirements of ideal high frequency pulse current measuring sensors
FEATURES PARAMETERS
Physical Cost effective
Light weight
Flexibility of installation.
Measurement Suitable for online non-intrusive measurements,
Safety of the sensor and measuring system.
Operational Higher sensitivity
wide bandwidth
linearity of operation
In the event where PD takes place, the risk of insulation system dielectric instability is high. Hence, the measurement and monitoring of PD is very much significant [4]. The IEC standard 60270 is commonly used as a benchmark during the detection of PD current and charge by commercially available instruments [5]. There are various types of sensors to detect PD pulse such as antenna, acoustic sensor, chemical sensor and Current Transformer (CT). However, The Rogowski coil (RC) has been considered to be one of the most favourable sensors with respect to the above requirements [2].The RC is defined as a transducer which is able to detect PD current signal. The RC is used in order to measure the PD currents which has high frequency signal.In addition, and the RC is a favorable sensor due to its lightweight, fast response and low cost equipment for PD detection. For applications such as PD detection and measurement, the design of the coil
3
is focused on high frequencies up to tens to hundreds of MHz. The geometrical parameters of the RC will very much affect the features mentioned above [2].
In this research, the developed RC detects the PD pulses generated by pulse generator. To achieve the features above, geometrical and electrical parameters are specified. Experimental test is done to investigate the effect of high pass filter and amplifier circuit with three types of modification on output waveform of RC. The final output waveform is analyzed and the result obtained are briefly discussed in this paper. It is essential that a power system designer and a power system installation maintenance engineer to have a good understanding with regards to PD mechanisms, characteristics and its development processes.
1.2 Problem Statement
Presence of PD need to be prevented to avoid electrical equipment from getting damaged. To detect the presence of PD a suitable RC sensor is needed. Besides, parameters of the RC sensor is important to detect high frequency PD signals. A proper RC sensor able to provide more reliable and accurate measurement results. The RC produces output which has very low amplitude voltage pulses.
At this moment, the experiment conducted in the high voltage laboratory USM to diagnose the PD in insulation oil is using the impedance matching circuit (IMC). The PD measurement using IMC should have the reference device to ensure that the measurement is accurate. Thus the RC is determined as one of the detecting device to be the reference of the PD detection.
4 1.3 Objective
The objective of this research are as follow:
1. To detect partial discharge pulse.
2. To develop Rogowski Coil for partial discharge measurement.
3. To amplify the PD signal detected by RC.
1.4 Scope of Work
The scope of this research is to develop a Rogowski Coil that can detect Partial Discharge signals with high frequency. Besides, the recommendations to specify the parameters of the coil that can affect the sensitivity and performance. Design amplifier circuit to improve the detected signal by RC. Three different types of amplifier circuit is designed to see the effect of amplification on RC measurement. Do analysis on obtained signal and data to find out the sensitivity and the linearity of this RC sensor.
1.5 Thesis Outline
The thesis is classified into 5 chapters, the outline of the thesis can composed as follow;
Chapter 2 summarizes literature review of PD measurement device. The PD detection methods and disadvantages are described. The definition, design and working principle are discussed.
5
Chapter 3 describes the methodology used to conduct the project. This chapter explains the project flow and the test method to determine PD. This chapter describes how the RC is constructed and how the amplifier circuit is designed.
Chapter 4 presents the actual PD signal and RC measurement signals for different PD magnitude. Analysis is made in this chapter to find out the linearity and sensitivity of this detection sensor. A brief discussion is made on the RC measurements.
Chapter 5 concludes the thesis work and the recommendation for future work addressed.
6
CHAPTER 2
LITERATURE REVIEW
2.1 Introduction
This chapter review on the PD detection methods. The review includes the definition, the geometrical parameters and electrical equivalent circuit of RC. This chapter ends with the summary literature review.
2.2 Partial Discharge
The failures caused due to electrical insulation are one of the reasons of unexpected disruption in power systems and associated electrical machines and power cables. In general, the power generation, transmission and distribution are stated to be the most expensive electrical assets in high voltage power systems, which are subjected to numerous electrical, thermal, environmental and mechanical stresses [8].
The PD issues are a common cause leading to power system failure. PD can be stated as an electrical discharge or pulse generated in a dielectric solid surface or liquid insulation system. PD occurs when insulation containing defects or voids is subject to high voltages. If left untreated PD can degrade insulation until insulation failure occurs.
7 2.3 Partial Discharge Detection
PD detection system is a system that can detect and display PD signals for better reliability. This system can work in many ways with various types of detection methods.
The PD emitted energy as electromagnetic emission, acoustic emission and ozone and nitrous oxide gases. PD signals can be detected by using these emitted energies [9].
The first stage is the pre-breakdown operation, starting with slow damage of the insulation followed by further insulation degradation with time. Second one is a post- breakdown condition when the complete failure of the insulation occurs, and the component is damaged beyond recovery. If an upcoming fault can be detected during the pre-breakdown operation, the faulty component could be repaired in advance, which would lead to significantly shorter power outage duration due to the faulty component.
Currently, there are two types of measurement techniques carried out to test if there are any faults along the CC line. They are on-line and off-line measurements. With the on-line measurement, the power line continues working as normal and no disruption of service is required. The methods currently available are using very high frequency (VHF) antennas, infrared sensors, acoustic, radiometric, capacitive coupling and Rogowski coil [10].
2.3.1 Acoustic Detection
The non-electric type measurement like acoustic emission method is often concerning localization of the PD origin. Acoustic detection, concentrates on the capturing and recording of an acoustic signal produced by a PD. The acoustic emission Methods can be very effectively used for the online PD Detection, since this method is immune to Electro-
8
magnetic noise, which can greatly reduce the sensitivity of electrical methods, especially when, applied under field conditions [11].
2.3.2 Radiometric Detection
Radio frequency detection of PD is a long-established principle, being described in the international standard for PD measurement. In its simplest form, a directional radio receiver can be used to determine the general location of an electrical discharge within a high voltage equipment’s. As the ability to perform RF measurements became more readily available, studies of these measurements for PD applications were carried out.
Radiometric detection method basically can be related to electric detection method. The electrical pulses take place in the nanosecond order. Besides that, it has a frequency component that can be measured over 1 MHz [12].
2.3.3 Electrical Detection
The electrical detection methods are very sensitive to the Electro-magnetic interference, also such interference signals may be controllable in a Laboratory. There are many types of current detecting hardware has been proposed for transient current measurements in Electrical field [13]. The Rogowski coil has been proposed to be a standout among the most suitable sensor which fulfils most of the requirements.
Rogowski coil was named after Walter Rogowski.
Rogowski coils be made up of helical coil of wire with the lead from one end returning through the centre of the coil to the other end, so that both terminals are at the same end of the coil and the two ends are usually connected to a cable. It is usually designed in shape of a donut or toroid. Circular, oval and rectangular are the most familiar
9
core cross sections. These electrical devices are unique in that their coil turns are wound around a non-magnetic air core, rather than an iron core [2].
In this research, the design of a Rogowski coil to measure very low currents, but with a very high frequency, such as partial discharges, is exposed. The Rogowski coil used as a PD transducer is very advantageous because it is inexpensive and easy to use.
Besides, it provides the needed bandwidth for this application [15]. D. Ward and J. Exton [16] showed that Rogowski coils have better performance in a vast majority applications;
lightning test, partial discharge monitoring and sudden short circuit test, as compared to the other current measuring devices. Kojovic [17] proposed a new configuration of Rogowski coil based on printed circuit board (PCB) being useful in transient current measurement. In this paper, the number of PCB coils was increased without decreasing its bandwidth to improve the amplitude of the coil’s output signal. Figure 2.1 shows typical RC sensor with various diameter size.
Figure 2.1: RC with various diameters
10
Table 2.1: Advantages and Disadvantages of RC
ADVANTAGE DISADVANTAEGE
Flexible core coils
Non-intrusive to the live conductor
Easy temperature compensation
No risk of secondary winding opening
Excellent linearity (have no magnetic materials to saturate)
The output of the coil must be passed through an integrator circuit to obtain the current waveform.
Rogowski coil does not have a response down to DC
Table 2.1 shows the advantages and disadvantages of the RC sensor. Because of these disadvantages as in Table 2.1, Rogowski Coil are essentially utilized when ease-of- establishment is a high priority. Traditional or split- centre current transformers are utilized when precision, noise immunity and currents during power quality events are a high priority.
2.4 WORKING PRINCIPLE OF ROGOWSKI COIL (RC) [8]
The output voltage produced by rogowski coil as in equation 2.1,where A is the area of one of the small loops, N is the number of turns, l is the length of the winding(the circumtance of the ring), 𝜇0= 4π ×10−7 V.s/(A.m) is the magnetic constant, R is the major radius of the toroid and r is its minor radius.
11
2.1
Flux induced
Φ = ∫ dΦ = ∫ μ0ANH dl cosα 2.2
And
M=ANμ0
Where N is the number of turns of the coil. Equation (2.4) shows the importance of the mutual inductance M in current measurement sensitivity. That is as M increases the induced voltage across the coil, Vrc increases.
iin (t) = ∫ H dl cosα 2.3
Where iin is the total electric current bounded by the enclosed path (2.3). H is the magnetic field intensity in Amperes/m, and is related in free space, to the magnetic flux density B (in Tesla).
Vrc(t) = -M diin
dt 2.4
This formula assumes the turns are evenly spaced and that these turns are small relative to the radius of the coil itself. The output of the Rogowski coil is proportional to the derivative of the wire current. The output is often integrated so the output is proportional to the wire's current:
12
Figure 2.2: RC mounted around conductor [2]
Rogowski coil around Conductor
The Figure 2.2 shows RC sensor with the current conductor placed on the centre for detection. The RC winding is a helical coil of wire which wind on the core surface.
The return wire is lead from one end returning through the centre of the coil to the other end to complete the loop. The PD detection is applied around conductor without disturbing the flow of the current in conductor. After the detection the RC gives Induced voltage as the output [2].
2.5 Geometrical Parameters of RC
The Figure 2.3 shows all geometrical parameters included in RC development and how the changes in parameters can affect the performance of the RC sensor. Geometrical parameters need to be considered before design a RC. RC can be designed in various
Conductor
Return wire RC winding
𝑉
𝑟𝑐(t)
13
types with different sets of parameters and materials. As shown in Table 2.3 Parameters can influence the performance of these Rogowski coil. Selection of parameters according to recommendations will improve the sensitivity and bandwidth of this coil in PD detection [2]. Current range and frequency range up to one to hundreds order of MHz of the signal that need to be measured can be adjusted by changing the parameters of the coil.
Figure 2.3: Geometrical parameters of RC [2]
14
Table2.3: Recommendations for RC geometrical parameters
PARAMETERS
RECOMMENDATIONS
Changes Advantages
Diameter of coil , dm
Decreasing
Improved performance
Sensitivity increase
Bandwidth increase
Lower weight
Smaller size
Easy placement of test line in the center of coil
Diameters of core , drc Decreasing Easy installation
Easy for bend and to get shaped
Less space required
Number of turns , N Less turns Improve the bandwidth significantly
Reduce weight
Diameter of wire , dw Larger Reduce resistance of wire
Good mechanical strength
15
CHAPTER 3
METHODOLOGY
3.1 Introduction
The chapter discuss about how this research was conducted. The apparatus and devices that used in experimental setup is explained. This chapter proceeds with Section 3.2 which elaborate about the flow of research. In Section 3.3, development of RC is discussed and diagram of the assembled RC is included. It continues with Section 3.4 which contains the geometrical parameters of RC designed in this research. In addition, calculations and formulas are briefed to determine the characteristic of RC corresponding to geometrical parameters used. Lastly, the experimental test setup diagram is presented.
3.2 Flow of Research
The knowledge about RC is important in this research. The main cause of designing the Rogowski Coil is determined. The significant facts that affect the sensitivity, bandwidth and performance of this RC were emphasized and jotted down to improve the understanding of this device. The parameters which include electrical and geometrical of RC were identified and shortlisted for the purchasing purposes. Then, RC is constructed and tested in lab. Experimental test with PD pulse generator carried out.
The output waveform and result is than analysed and discussed briefly.
16
Figure 3.1: Flowchart of research
START
literature review
Design a RC and brief the process
Determine parameters,list and purchase of components needed for
Rogowski coil
Constuct Rogowski coil
Test Rogowski coil with three types of amplifier
circuit
Output waveform amplified ?
Record the data
Discuss the characteristic of Rogowski coil
END
NO
Yes
17 3.3 Development of Rogowski Coil
Development process of Rogowski Coil needs very clear understanding from other references and journals. Before the development process starts, should have a clear view on definition of Rogowski Coil and the main purpose of Rogowski coil is designed.
Figure 3.2 shows the electrical equivalent circuit for RC which consist of Rc, Lc and Cc.
Figure 3.2: Electrical equivalent model for Rogowski coil.
Rogowski coil is mainly an elastic or flexible type sensing equipment in electrical field. Therefore, a flexible type material have to be chosen as core for the Rogowski Coil development. So the designed RC will be easy for handling and easy for installation process. In installation process the Rogowski Coil have to bended into circular shape to be mounted around the current-carrying conductor.
A PVC flexible pipe was chose as the material of non-magnetic core of Rogowski Coil. Advantages of using this mentioned pipe are its flexible, lightweight, elastic and low coast. The parameters of the PVC pipe are suitable for the objective of this research.
Furthermore, copper wire is used as the winding wire for the development of RC. The copper wire was chosen because it is a good conductor of current (good conductivity).
The diameter of winding wire will effect on the resistivity. According to past research in chapter 2, a proper diameter of copper wire is chose for a low resistivity. The Copper wire is than wind all over the surface of PVC pipe With N number of turns and the return loop
18
of the RC is passed through the centre of the hollow shaped PVC pipe to make both terminals accessible at the same end of RC.
3.4 Geometrical Parameters of Rogowski Coil
The Figure 3.3 shows the constructed RC sensor according to specified parameters as in Table 3.1.The Figure 3.4 shows the end of the return loop which passed through the center of the RC core from the other end. Figure 3.5 shows the RC sensor after the ground wire and the return loop soldered with BNC connector for easier connection method with oscilloscope.
Figure 3.3: Constructed RC
19
Figure 3.4: Return Conductor
Figure 3.5: BNC connector with RC
20
Geometrical parameters used for the development of Rogowski coil are listed in Table 3.1. All parameters chose based on recommendations given in [8].This parameter are reported to design a RC with high frequency range of detection from one MHz to order of hundred MHz.
Table 3.1: Geometrical parameters used in this research
PARAMETERS MEASUREMENTS
Outer diameter of coil , do 155 mm
Inner diameter of coil , di 125 mm
Mean diameter of coil , dm 140 mm
Core diameter , drc 15 mm
Length of coil , l 475 mm
Radius of wire , r 0.5 mm
Number of turns , N 30
3.5 Characteristics Determined by Geometrical Parameters
Calculation was made to determine all the electrical parameters of RC based on the geometrical parameters specified.
21 3.5.1 Mutual Inductance
Mutual inductance is a decisive parameter for the assessment of a coil’s sensitivity. Its value depends on the shape of the cross section of the core used for the coil winding. The most common core cross-sections are circular, oval or rectangular. For different shapes of the coil core cross-section, different expressions are available to determine the mutual inductance. For a circular shaped RC the mathematical formulae is as below [2]:
Mc =μ° N(√Rout−√Rin)2
2 3.1
Where N = number of turns, permeability constant μ° = 4π × 10−7NA−2, Rout=outer diameter and Rin=inner diameter
3.5.2 Self – resistance
The coil’s self-resistance is due to the resistance of the wire that is used for construct the winding. The copper coil resistance is given by [2]:
Rc = ρ l
πr2 3.2
Where copper resistivity, ρ=1.68 × 10−8 , l = length of wire, r = radius of winding wire.
3.5.3 Self - inductance
The coil inductance of single circular loop of wire is given by equation (3.3) [2]:
Lc =μ° N2h
2π logRout
Rin 3.3
h = height of RC
22 3.5.4 Self – capacitance
The self-capacitance of a Rogowski coil with the construction specified above could be observed as the sum of two capacitance values. The first one is the turn-to-turn capacitance of the coil turns, present for each multi-turn inductor. The second is a specific capacitance that is due to the return wire in the centre of the winding. In the latter case, the capacitance is formed between the turns of the coil and the return winding.
The capacitance value is given by [2]:
Cc =4π2ε°(Rout+Rin )
log(Rout+Rin
Rout−Rin) 3.4
Where air permittivity, ε°= 8.85 × 10−12
3.5.5 Turn-to-Turn Gap
The distance between each turn, g is calculated as [2]:
g = 2π
N (Rout+Rin
2 ) 3.5
3.5.6 Turn-to-Turn Capacitance
Capacitance between gaps of turns is given by [2]:
Cgap = πε°l
N log( g Drc+ √( g
Drc)2 −1)
3.6
23 3.6 Circuit Design
This section presents the circuit design of amplifier circuit. All the designed circuit is printed and soldered in PCB. Three types of modification made using this amplifier circuit to see the effect on the output signal of RC. The circuit is designed to improve the produced output waveform from the RC sensor. The first type of circuit is amplifier added with high pass filter and 50 ohm resistor in series. The second type is the amplifier is added with 50 ohm resistor in parallel in the beginning. The third type of circuit is amplifier circuit is modified with 1k ohm resistor at the beginning of the circuit.
These all three types of circuit are tested using same range of PD pulse generator input range to see the effect on the output signal of RC.
3.6.1 Metal Casing for the PCB
The designed circuit is printed in the PCB and placed in a metal box with its casing as shown in Figure 3.6.This metal box helps to prevent any external low frequency noises and signal from disturbing the performance of the circuit. So this helps to improve the stability of this RC output. This also provides more reliable PD detection system. Figure 3.7 shows the metal box which already covered with its casing and all the terminals like Vcc, ground, input and output labelled properly.
24
Figure 3.6: PCB placed in the metal box with its casing
Figure 3.7: Covered metal box with its terminals labelled.
25 3.6.2 Amplifier circuit
Amplifier circuit that is used in this research is shown in Fig. 8 consisting of biasing resistors bias), blocking capacitors (C1, C2 and C3) and radio frequency choke (RFC). R1 is used to set the bias currents that determine the performance of the circuit.
Referring to equation (3.8), Cblock is used to protect the input and output terminal of amplifier that blocks the DC signal and neglects the low frequency disturbance in this amplification. RFC is used to increase the separation of the output signal from the amplifier DCsupply as well as to serve as a "peaking coil" to increase the gain at high frequencies in order to compensate the gain decrease at high frequency.
𝑋
𝐶=
1𝑗2𝜋𝑓𝐶 3.8
Figure 3.8: Amplifier circuit
26
3.6.3 Amplifier Circuit with High Pass Filter and Series 50ohm Resistor (Circuit A)
Figure 3.9: Working principle of high pass filter [18]
Figure 3.9 shows theory of high pass filter with low and high frequency signal at the input but the output just contains the high frequency signals. A high pass filter is a filter which passes high-frequency signals and blocks, or impedes, low-frequency signals.
In other words, high-frequency signals go through much easier and low-frequency signals have a much harder getting through. High pass filters can be constructed using resistors with either capacitors or inductors. A high pass filter composed of a resistor and a capacitor is called a high pass RC filter. To create a high pass RC filter, the capacitor is placed in series with the power signal entering the circuit, such as shown in the Figure 3.10:
Figure 3.10: Capacitor and resistor of high pass filter
10 nF
1kΩ
27
In this research, resistor and capacitor values were determined based on calculation for better cut-off frequency as in 3.7. Than this combination of resistor and capacitor is tested in the lab to ensure that this high pass filter can reduce the low frequency noises from interrupting the RC sensor output.
3.7
Figure 3.11: Amplifier circuit with high pass filter and series 50Ω
Figure 3.11 shows a combination of a series resistor, high pass filter and the amplifier circuit. The 50 ohm resistor is purposely added to prevent the following components in the circuit from getting damaged by unexpected excess input.
𝑓𝑐 = 1 2𝜋𝑅𝐶
= 1
2𝜋(1000)(10 × 10−9) = 16 𝑘𝐻𝑧
28
3.6.4 Amplifier Circuit with 50ohm in Parallel as Matching Resistor (Circuit B) In electronics, impedance matching is the practice of designing the input impedance of an electrical load or the output impedance of its corresponding signal source to maximize the power transfer or minimize signal reflection from the load.Impedance matching is also to minimize reflections is achieved by making the load impedance equal to the source impedance. This section the amplifier circuit is modified with matching resistor of 50ohm. The experiments is conducted to see the effect of this 50Ω resistor to the output of RC Resistor with value of 50ohm represents the Bnc cable resistance. Figure 3.12 shows the amplifier circuit connected with 50 ohm resistor in parallel at beginning of the circuit.
Figure 3.12: Amplifier circuit with 50Ω in parallel
29
3.6.5 Amplifier Circuit with 1k ohm in Parallel as Matching Resistor (Circuit C) This section the amplifier circuit is modified with matching resistor of 1k ohm.
The experiments is conducted to see the effect of this 1kΩ resistor to the output of RC.
Figure 3.13 shows the schematic diagram of the amplifier circuit with 1k ohm resistor connected at the beginning of the circuit.
Figure 3.13: Amplifier circuit with 1kΩ in parallel
30
CHAPTER 4
RESULT AND DISCUSSION
4.1 Introduction
This chapter presents the data that was obtained by conducting experiments in high voltage laboratory. There are three types of amplifier circuit in this experiment;
Circuit A, Circuit B and Circuit C. PD pulse generator was used as the input device to test the designed Rogowski coil and the three types of amplifier circuit. The input PD pulse range is set to be fix at 5pC, 10pC, 50pC and 100pC. The effect of the circuit designed on output of RC has been investigated by performing experiments and measurements. Outputs of all three circuit is compared according to the input range. The peak value that recorded from oscilloscope is gathered and compared.
Besides, this chapter presents the waveforms for all the test measurements with variable input range. There also analysis was made for three types of circuit to find out which circuit has good linearity and the sensitivity of this detection method of PD. In addition, the results obtained from the measurements were discussed. The last part of this chapter is the summarization section of the overall Chapter 4.
4.2 Results of PD Measurement
4.2.1 PD measurement using RC with Circuit A.
Figure 4.1 shows the output waveform of RC measurement with circuit A and the actual PD. The signal in blue colour is the actual PD waveform of 5pC magnitude. The
31
orange colour waveform represents the output waveform of RC measurement when the actual PD injected. The actual PD signal reaches peak voltage of 11.3 mV and the RC measurement waveform had its peak at 7 mV.
Figure 4.1: PD measurement of RC with circuit A for 5pC PD charge
Figure 4.2 shows the output waveform of RC measurement with circuit A and the actual PD. The signal in blue colour is the actual PD waveform of 10pC magnitude. The orange colour waveform represents the output waveform of RC measurement when the actual PD injected. The actual PD signal reaches peak voltage of 22.5 mV and the RC measurement waveform had its peak at 12.7 mV.
32
Figure 4.2: PD measurement of RC with circuit A for 10pC PD charge
Figure 4.3: PD measurement of RC with circuit A for 50pC PD charge
33
Figure 4.3 shows the output waveform of RC measurement with circuit A and the actual PD. The signal in blue colour is the actual PD waveform of 50pC magnitude. The orange colour waveform represents the output waveform of RC measurement when the actual PD injected. The actual PD signal reaches peak voltage of 87.6 mV and the RC measurement waveform had its peak at 45.8 mV.
Figure 4.4: PD measurement of RC with circuit A for 100pC PD charge
Figure 4.4 shows the output waveform of RC measurement with circuit A and the actual PD. The signal in blue colour is the actual PD waveform of 100pC magnitude. The orange colour waveform represents the output waveform of RC measurement when the actual PD injected. The actual PD signal reaches peak voltage of 133.0 mV and the RC measurement waveform had its peak at 68.0 mV.
34 4.2.2 PD measurement using RC with Circuit B
Figure 4.5 shows the output waveform of RC measurement with circuit B and the actual PD. The signal in blue colour is the actual PD waveform of 5pC magnitude. The orange colour waveform represents the output waveform of RC measurement when the actual PD injected. The actual PD signal reaches peak voltage of 11.3 mV and the RC measurement waveform had its peak at 4.1 mV.
Figure 4.5: PD measurement of RC with circuit B for 5pC PD charge
Figure 4.6 shows the output waveform of RC measurement with circuit B and the actual PD. The signal in blue colour is the actual PD waveform of 10pC magnitude. The orange colour waveform represents the output waveform of RC measurement when the actual PD injected. The actual PD signal reaches peak voltage of 22.5 mV and the RC measurement waveform had its peak at 7.4 mV.
35
Figure 4.6: PD measurement of RC with circuit B for 10pC PD charge
Figure 4.7: PD measurement of RC with circuit B for 50pC PD charge
36
Figure 4.7 shows the output waveform of RC measurement with circuit B and the actual PD. The signal in blue colour is the actual PD waveform of 50pC magnitude. The orange colour waveform represents the output waveform of RC measurement when the actual PD injected. The actual PD signal reaches peak voltage of 87.6 mV and the RC measurement waveform had its peak at 25 mV.
Figure 4.8: PD measurement of RC with circuit B for 100pC PD charge
Figure 4.8 shows the output waveform of RC measurement with circuit B and the actual PD. The signal in blue colour is the actual PD waveform of 100pC magnitude. The orange colour waveform represents the output waveform of RC measurement when the actual PD injected. The actual PD signal reaches peak voltage of 133.0 mV and the RC measurement waveform had its peak at 36 mV.
37 4.2.3 PD measurement using RC with Circuit C
Figure 4.8 shows the output waveform of RC measurement with circuit C and the actual PD. The signal in blue colour is the actual PD waveform of 5pC magnitude. The orange colour waveform represents the output waveform of RC measurement when the actual PD injected. The actual PD signal reaches peak voltage of 11.3 mV and the RC measurement waveform had its peak at 9.5 mV.
Figure 4.9: PD measurement of RC with circuit C for 5pC PD charge
Figure 4.10 shows the output waveform of RC measurement with circuit C and the actual PD. The signal in blue colour is the actual PD waveform of 10pC magnitude.
The orange colour waveform represents the output waveform of RC measurement when the actual PD injected. The actual PD signal reaches peak voltage of 22.5 mV and the RC measurement waveform had its peak at 16.5 mV.
38
Figure 4.10: PD measurement of RC with circuit C for 10pC PD charge .
Figure 4.10: PD measurement of RC with circuit C for 50pC PD charge
39
Figure 4.11 shows the output waveform of RC measurement with circuit C and the actual PD. The signal in blue colour is the actual PD waveform of 50pC magnitude.
The orange colour waveform represents the output waveform of RC measurement when the actual PD injected. The actual PD signal reaches peak voltage of 87.6 mV and the RC measurement waveform had its peak at 50.0mV.
Figure 4.12: PD measurement of RC with circuit C for 100pC PD charge
Figure 4.12 shows the output waveform of RC measurement with circuit C and the actual PD. The signal in blue colour is the actual PD waveform of 100pC magnitude.
The orange colour waveform represents the output waveform of RC measurement when the actual PD injected. The actual PD signal reaches peak voltage of 133.0 mV and the RC measurement waveform had its peak at 70.0 mV.
40 4.3 Analysis
4.3.1 Linearity
Table 4.1 contains all actual PD voltage values and RC measurements with Circuit A, Circuit B and Circuit C. Figure 4.13 is a plot of a graph to see the relationship between actual PD and measured voltage by RC with Circuit A, Circuit B to find out the linearity of this detection method.
Table 4.1: PD Input Voltage and Output Peak Value for each Set Actual PD Voltage
(mV)
Output Circuit A (mV)
Output Circuit B (mV)
Output Circuit C (mV)
11.3 7 4.1 9.5
22.5 12.7 7.4 16.5
87.6 45.8 25 50
133 68 36 70
Figure 4.13: Relationship between actual PD and measured voltage for Circuit A, Circuit B and Circuit C
41 4.3.2 Sensitivity
Table 4.2 contains all actual PD charge that injected in RC and the measured voltage by RC with Circuit A, Circuit B and Circuit C. Than the graph in Figure 4.14 is plotted to find out the sensitivity of each type of designed circuit. The RC measured voltage values are labelled on the graph at 5pC of PD charge with respective colours.
Table 4.2: PD Input Charge and Output Peak Value for each Set Actual PD Charge
(pC)
Output Circuit A (mV)
Output Circuit B (mV)
Output Circuit C (mV)
5 7 4.1 9.5
10 12.7 7.4 16.5
50 45.8 25 50
100 68 36 70
Figure 4.14: Relationship between actual PD charge and measured voltage for Circuit A, Circuit B and Circuit C
42 4.3 Discussion
Rogowski Coil sensor is applicable detection method of PD. The geometrical parameters of the RC plays the most important role in the design process. The geometrical parameters can determine the sensitivity of the coil. It can be determined whether by using formula or by measurement in laboratory. The measured RC voltage waveform with Circuit A, Circuit B and Circuit C shows that the waveform after amplification resembles the PD input signal the most. The measured waveform of RC before amplification for every PD signals could not be seen clearly and the peak value are too small compared to the actual PD signal. So the RC sensor needs a better amplifier circuit which can improve its output signal so that the waveform can be seen clearly.
The measured voltage waveform of RC when connected with Circuit A shows that, having high pass filter in the circuit does not make any changes to the output.
Because the amplifier circuit is already designed with capacitor on both side. Even capacitors can prevents low frequency signal from disturbing the amplification process.
So the high pass filter did not contributes to improvise the RC measured signal. After RC measurement is tested with Circuit B connected to it, the measured output voltage peak values had a huge different. The peak values almost decreases around 50% compared to RC measurement with Circuit C. In Circuit A the series resistor dumbs and reduces the peak value of RC sensor.
RC works more effectively when it connected with Circuit C. The RC measurement of voltage gives the highest peak values compared to Circuit A and Circuit B. Circuit C amplifies 50% more than Circuit B and 9% greater than Circuit A. This proves that Circuit C has better amplification characteristic. This is because of that 1k ohm resistor. According to ohm’s law, voltage is directly proportional to resistor,
43
(V=IR).However all three amplifier circuit capable to improve the measured voltage waveform. After amplification the measured value becomes clear, visible for better detection system and the peak values are greater. The output waveform is nearly same with the input PD signal but it does have some pre-shoot and oscillations.
After that the results were tabulated and used to plot two graphs to perform analysis on linearity and sensitivity of this PD detection device for Circuit A, Circuit B and Circuit C. By referring to the Figure 4.13, it is found that all three types of circuit works linearly with the input. When the input PD signal is increased, the produced peak output voltage also increase. These proves that this RC sensor is reliable and changes according to the PD input signal. As shown in Figure 4.14, Circuit A, Circuit B and Circuit C have the ability to detect the lowest PD input signal which is 5pC. Even all three types of circuit can detect at 5pC, Circuit C produced higher peak output voltage value compared to Circuit A and Circuit B. This proves that Circuit C have better sensitivity compared to other two circuit designs.
4.4 Summary
RC measuring device is one of the approved sensor that can detect PD signal. The geometrical and electrical parameters are the crucial key for RC’s performance in high frequency environment. In addition, for the application like PD detection the sensitivity and wide bandwidth of the sensor coil is significantly affected by these parameters. The designed amplifier circuit made improvement on the output of this RC detection sensor.
This RC sensor works linearly and has good sensitivity.
44
CHAPTER 5
CONCLUSION AND RECOMMENDATION
5.1 Conclusion
To achieve the high frequency range like one MHz followed by order of hundred MHz, the geometrical parameters of RC that plays the most important part in this sensor.
The parameters which act as a beneficial part in this sensor were selected and specified based on the frequency range. Calculation were made by considering the geometrical parameters and also to determine the electrical parameters.
The input PD signal waveforms are compared with the output waveform of RC connected to amplifier Circuit A, Circuit B and Circuit C. Investigation shows that output waveform without amplification is not clearly visible due to low peak values but after the amplification is done the waveform becomes clearly visible with greater peak values.
Analysis shows that this RC sensor works linearly for all three types of amplifier circuit.
Circuit C has better sensitivity at 5pC PD input signal.
5.2 Recommendtion for Future Works
The experimental set up for testing the RC sensor can be improved by involving the real PD input in high voltage lab rather than using the PD pulse generator as the input.This can produce the output waveform which are more precise and accurate.
45
Next recommendation is to improve the work done is by measuring PD signal in any one of available high voltage equipments such as transformer.The RC can be installed either on live phaseor on ground terminals in order to measure the PD.
46
REFERENCES
1. Yaacob, M.M., Alsaedi, M.A., Rashed, J.R., Dakhil, A.M. and Atyah, S.F., 2014.
Review on partial discharge detection techniques related to high voltage power equipment using different sensors. Photonic sensors, 4(4), pp.325-337.
2. Shafiq, M., Hussain, G.A., Kütt, L. and Lehtonen, M., 2014. Effect of geometrical parameters on high frequency performance of Rogowski coil for partial discharge measurements. Measurement, 49, pp.126-137.
3. Chonpathomphikunloed, W., Paophan, B., Kunakorn, A., Yutthagowith, P. and Leelachindakrairerk, M., 2017, September. Analyses of Rogowski coils for partial discharge measurement. In Electrical Insulating Materials (ISEIM), 2017 International Symposium on (pp. 378-381). IEEE.
4. Isa, M., Elkalashy, N.I., Hashmi, G.M. and Lehtonen, M., 2012. Experimental evaluation of Rogowski coil performance for locating PD in energized overhead covered-conductor feeder. Journal of Energy and Power Engineering, 6(6), p.949.
5. Paophan, B., Kunakorn, A. and Yutthagowith, P., 2016, November.
Implementation of a Rogowski's coil for partial discharge detection. In Electrical Machines and Systems (ICEMS), 2016 19th International Conference on (pp. 1- 4). IEEE.
6. Boggs, S.A., 1990. Partial discharge: overview and signal generation. IEEE Electrical Insulation Magazine, 6(4), pp.33-39.
7. Samimi, M.H., Mahari, A., Farahnakian, M.A. and Mohseni, H., 2013. A Review on the Rogowski Coil Principles and Applications. measurements, 4, p.5.
8. Mohamed, H., Lazaridis, P., Upton, D., Khan, U., Saeed, B., Jaber, A., Zhang, Y., Mather, P., Vieira, M.F., Barlee, K.W. and Atkinson, D.S.W., 2016, May. Partial discharge detection using low cost RTL-SDR model for wideband spectrum
47
sensing. In Telecommunications (ICT), 2016 23rd International Conference on (pp. 1-5). IEEE.
9. Yaacob, M.M., AlSaedi, M.A. and Abdullah Al Gizi, N.Z., 2013. Partial discharge signal detection using ultra high frequency method in high voltage power equipments: a review. International Journal of Scientific and Engineering Research, 4(1), pp.1-6.
10. Ward, D.A. and Exon, J.L.T., 1992, February. Experience with using Rogowski coils for transient measurements. In Pulsed Power Technology, IEE Colloquium on (pp. 6-1). IET.
11. Lundgaard, L.E., 1992. Partial discharge. XIV. Acoustic partial discharge detection-practical application. IEEE Electrical Insulation Magazine, 8(5), pp.34- 43.
12. IEC, I.S., 2000. 60270, High Voltage Test Techniques-Partial Discharge Measurements. International Electrotechnical Commission.
13. Xiao, C., Zhao, L., Asada, T., Odendaal, W.G. and Van Wyk, J.D., 2003, October.
An overview of integratable current sensor technologies. In Industry Applications Conference, 2003. 38th IAS Annual Meeting. Conference Record of the(Vol. 2, pp. 1251-1258). IEEE.
14. Shafiq, M., Hussain, G.A., Kütt, L. and Lehtonen, M., 2014. Effect of geometrical parameters on high frequency performance of Rogowski coil for partial discharge measurements. Measurement, 49, pp.126-137.
15. Argüeso, M., Robles, G. and Sanz, J., 2005. Implementation of a Rogowski coil for the measurement of partial discharges. Review of scientific instruments, 76(6), p.065107.
48
16. Ward, D.A. and Exon, J.L.T., 1993. Using Rogowski coils for transient current measurements. Engineering Science & Education Journal, 2(3), pp.105-113.
17. Kojovic, L., 2002. PCB Rogowski coils benefit relay protection. IEEE computer applications in power, 15(3), pp.50-53.
18. Imaizumi, H. and Shioda, T., Sanyo Electric Co Ltd and System Solutions Co Ltd, 2009. High-pass filter. U.S. Patent Application 12/196,077.
