Loading...

Summer Research Fellowship Programme of India's Science Academies

Design of a calibrated sensor for remotely monitoring and managing electrical load consumption

Rounak Goswami

Indian Institute of Engineering Science and Technology, Shibpur

Guided by:

K.C. James Raju

Professor, Department of Physics, University of Hyderabad

Abstract

Energy conservation is the need of the hour and it is at the top of sustainable energy hierarchy. Saving energy directly leads to large environmental and financial benefits. Energy Management Systems may help to reduce energy consumption, improve system utilization, increase reliability, predict electrical system performance and load patterns, and optimize energy usage to reduce cost. Although every end load is connected to the grid, irrespective of the geographical location and the electricity distribution network, consumption still remains monitored by a lump at a specific metering point. The problems with this model are numerous, poor demand visibility, large power factor variations, unpredictable power quality to name a few. A slight aberration in consumption could impact the overall grid. To bridge the existing gap between this connected consumption and disconnected information, an intelligent interconnected network of sensors which feed data into the cloud will lead to better Demand Side Management, the ability to act in real time, so that every electricity user may act proactively. This electrical sensor to sense the state of the load, measure the electrical parameters like voltage, current, power factor, phase, frequency, lead, lag, active power, reactive power, surge/spike and harmonics forms the fundamental block of the energy efficiency systems. This sensor can be hooked to every connected electrical load for recording the sense-measure-manage status into the wireless system for further analytics and actionable feedback. Sensing and measuring can be achieved by multiple techniques such as shunts, current transformers (CT), hall-effect sensors, and Rogowski coils, while managing requires additional intrusive control circuitry. The CT method has been chosen for its benefits of precision, electrical isolation between the load and the conditioning/ transmitting unit. Proper calibration of sensors is required to ensure reliable and reproducible measurements for predictable end use. This work mainly aims at studying the effects of the shape of the core, core material, SWG of the windings, and turns ratio of the Current Transformer to address load currents from possibly 1mA up to 1A and interface it with the signal conditioning and processing unit. Once done, the sensors can be integrated into the wireless network to provide real-time data, that can be analyzed for any actionable feedback.

Keywords: energy management systems, isolation transformers, energy metering, energyDSM, energy efficiency, true RMS

Abbreviations

List of Abbreviations
CTCurrent Transformer
PTPotential Transformer
BLEBluetooth Low Energy
FHMMFactorial Hidden Markov Model algorithm
DMMDigital MultiMeter
MSOMixed Signal Oscilloscope
PCBPrinted Circuit Board
MMMMonitor-Measure-Manage
SWGStandard Wire Gauge

INTRODUCTION

Background

Energy conservation and efficient use of energy and resources are required for the sustainable growth and development of the human society. The economic growth of India or any country for that matter has largely been associated with energy consumption. A major chunk of the industrial, commercial and domestic requirements is in the form of electrical energy i.e, irrigation, transportation & freight, utility, HVAC, industrial applications. Hence, focus on the management of energy, and its demand and supply, is the need of the hour. Understanding various components of energy demand is very much pertinent to understand the load patterns and predict the electrical system performance for the end loads and optimize energy usage.

Statement of the problems

Conventional electrical consumption monitoring systems usually rely on power monitoring & metering equipment such as a smart meter or a conventional metering device. But this type of monitoring of end load consumption is done by a lump from a specific metering point. There are some smart power consumption monitoring systems, but they are either cost-intensive or the sensors placed at each load consume a sizable power defeating the purpose of energy savings or they could not be remotely/centrally monitored & managed. Using wired/BLE/Ethernet as the transmission medium, the results populated on the server, which in turn are disaggregated/analyzed for the electricity consumption on each load using FHMM or similar algorithms for the user to observe the consumption of all or each of the loads. The parameters of each individual end load in this method heavily depends on the execution of the algorithm.

However, obtaining enough information for energy monitoring is limited by the number of measuring equipment that can be installed. Cost and privacy issues are involved. The status of devices is difficult to infer from limited information using the technology called non-intrusive appliance load monitoring, while it can be easily managed by intrusive techniques. However, because of the limited information in estimating the operational status of the devices, a lot of efforts to improve the estimation accuracy are in progress.

Objectives of the Research

The problems with all the previously stated models of energy monitoring are many, poor demand visibility, addressing large power factor variations, unpredictable power quality to name a few. A slight aberration in consumption could impact the overall grid. The objective is to bridge the existing gap between this connected consumption and disconnected information, through an intelligent interconnected network of sensors which feed data into the system and allows real-time monitoring of appliance states. At the very fundamental level of this monitoring system stands a calibrated sensor which will be hooked to every connected electrical load for M-M-M purpose.

Sensing and measuring can be achieved by multiple techniques such as shunts, current transformers (CT), hall-effect sensors, Rogowski coils and magnetic field based transducers. Out of these, shunt and CT methods are found to be more reliant for obtaining accurate and reliable measurements for end use. The CT sensor method has been chosen for isolating the electrical load from the conditioning and digital microcontroller circuit.

I worked on the calibration of CT sensors with toroidal core. Proper calibration of the sensor is required to ensure reliable and reproducible measurements for predictable end use. An attempt has been made to identify the best-suited material, shape for the core which serves the purpose, the proper wire gauge and optimum turns ratio of the primary to secondary windings to address a load current ranging from possibly 1mA up to 1A. Comparisons have been made between the performance of different toroidal cores made with different core materials. Performance for a given core material with different number of turns on primary for a fixed load current, and with fixed number of turns on primary while addressing varying load currents have also been studied.

The basic purpose of this sensor is to sense the state of the load, measure the current consumption and remotely control the AC loads if necessary.

Scope

The system when realized will have a lot of scope. It will not only be able to monitor the end loads, but it will also measure and manage the loads at the same time. This CT sensor which forms the fundamental building block of this management system can be further refined and improvised with additional circuitry to measure electrical parameters like voltage, power factor (lead/lag), phase, frequency, active power, reactive power, surge/spike and harmonics associated with the load.

LITERATURE REVIEW

Studies have been done on the various types of loads namely, continuous loads, discrete loads, highly inductive loads, non-linear loads,loads which demand a very high starting current etc. A paper by Kurmi, Sharma, Singh[1] introduces the various possible loads in power systems.

Thorough studies have also been done on the various existing methods of current detection and sensing and energy monitoring systems. An application note by Yang Zhen [2] discusses the concepts and fundamentals of various current sensing circuits. Forghani-Zadeh and Rincón-Mora [3], in their paper, have given a comparative study of the various methods of current sensing. A paper by Sung-Yong Son[4] gives an alternative algorithm based approach to management systems in smart homes and also gives an idea of the challenges faced in realizing such a system. Soft Ferrite, A User's Guide by Magnetic Materials Producers Association gives us an insight into the soft ferrites, the material which has been used as the transformer core in our case.

Information

Shunt resistor technique works on the simple principle that voltage drop across a resistor is proportional to the current going through it. The parasitic inductance present in the shunt affects high precision current measurement. They also introduce additional resistance in the source circuit and also limits high current sensing applications due to large power dissipation.

Rogowski coil is based on the principle of Faraday’s law of induction and the output voltage Vout of the Rogowski coil is given by integrating the current Ic to be measured. The Rogowski coil has a low sensitivity which can be compensated by either adding more turns or using an integrator with a higher gain.

Hall effect sensors are devices based on the Hall-effect, i.e, they are activated by an external magnetic field. Signal conditioning circuitry like an amplifier stage and temperature compensation are necessary to make the output usable for most applications.

Summary

The act of measurement of current ranges from picoamps to tens of thousands of amperes. The selection of a current sensing method depends on requirements such as magnitude, accuracy, bandwidth, robustness, cost, isolation or size. The current value may be directly displayed by an instrument, or converted to digital form for use by a monitoring or control system. The added benefit of isolation and non -intrusive monitoring along with high precision makes the CT technique best suited for our purpose.

PROJECT

Concepts

Transformer

Theory of operation of a real transformer

A transformer basically operates on the principle of mutual induction. Flux is present in the primary coil of a transformer due to excitation from some source. Not all the flux produced in the primary coil passes through the secondary coil - some of the flux lines leave the iron core and pass through the air instead. The portion of the flux that goes through one of the transformer coils, but not the other one is called leakage flux and the flux that links both the windings is called the mutual flux. The ratio of the primary voltage caused by the mutual flux to the secondary voltage caused by the mutual flux is equal to the turns ratio of the transformer. A current is drawn in its primary circuit, even when the secondary circuit is open circuited. This current is the current required to produce flux in a real ferromagnetic core. This no-load current can be divided into two parts, magnetization current and core-loss current.

Transformer Core

The core of the transformer transfers energy from the primary winding to the secondary winding through magnetic coupling. As a matter of fact, the core selection is the first designing step and after that, all other building details are defined: the number of turns and the conductor element (that can be copper foil, solid wire, litz wire), the isolation, the output connections. The common transformer cores used in various applications are:

• Solid Iron/Steel

• Laminated Silicon alloy/silicon steel

• Amorphous and Nanocrystalline Laminations

• Soft Ferrites

• Powdered cores (distributed air gap materials)

Out of the above, soft ferrites are the cheapest and they can operate over a wide range of frequencies. They are ceramic, homogenous material composed of various oxides with iron oxide as the main constituent. They are ferromagnetic in nature. They have saturation flux density in the range of 0.3 to 0.5 Tesla and relative permeability of the order of about 1000 . Soft ferrites have low eddy current and hysteresis loss owing to high electrical resistivity and small hysteresis area over a wide range of frequencies. Soft ferrites are available in many shapes such as ring core, E Core, I core etc. Nanocrystalline cores have desirable magnetic properties like very high saturation flux density, high magnetic permeability and low coercivity.

Losses in a transformer

1.  Copper (I2R) Losses: Copper losses are the resistive heating losses in the primary and secondary windings of the transformer. They are proportional to the square of the current in the windings.

2.  Eddy Current Losses: Eddy current losses are resistive heating losses in the core of the transformer. They are proportional to the square of the voltage applied to the transformer.

3.   Hysteresis Losses: Hysteresis losses are associated with the rearrangement of the magnetic domains in the core during each half-cycle. They are a complex, nonlinear function of the voltage applied to the transformer.

4.   Leakage flux: The fluxes which escape the core and pass through only one of the transformer windings are leakage fluxes. These escaped fluxes produce a self-inductance in the primary and secondary coils, and the effects of this inductance must be accounted for.

Instrumentation transformers

Instrumentation transformers are of two types:

Potential Transformers -Their primary is connected in parallel with the main circuit. Moreover, potential transformers are used as "sourcing elements".

Current Transformers -Their primary is connected in series with the main circuit and they are mostly used for metering and control purposes.

For the purpose of current detection, we have used a current transformer. The current transformer has two components, namely core and windings. The ideal current transformer may be defined as one in which any primary current is reproduced in the secondary circuit in the exact ratio and phase relationship.

Ip=Primary Current

Is=Secondary Current

Np=Number of turns in the primary

Ns=Number of turns in the secondary

H=Magnetic field strength (Amps/metre)

B=Magnetic flux density (Tesla)

le=Effective mean length of flux path

Ae=Effective cross-sectional area

For an ideal current transformer: INp = Is Ns IpIs=NsNp

Therefore the ratio of primary and secondary winding currents equal to the turns ratio.

In an actual transformer, the windings have resistance and reactance and also the transformer has magnetizing and loss component of current to maintain the flux. Therefore, in an actual transformer the ratio of current is not equal to the turns ratio and also there is a phase difference between the primary current and the secondary currents reflected back on the primary side and consequently, we have ratio error and phase angle error.

The following are the basic relations from magnetism:  

μ = BH\frac{B}{H}   [Definition of Relative Permeability]

H = NpIple\frac{{N_{p}}I_{p}}{l_{e}}       [Ampere's Circuital Law]

Erms = 4.44φfNsErms= 4.44BfNsAe [Faraday's Law of Electromagnetic Induction]

For a given toroidal transformer core, the maximum value of H is Hs which corresponds to the saturation flux density Bsat. So the maximum value of the NpIp product is given by

(NpIp)max = Hs×le

Knowing the value of maximum primary current Ip max we can get the value of maximum number of primary turns Np max.

The value of Bsat restricts the maximum value of Erms that can be obtained at the CT secondary.

It is to be noted that distributed capacitance and leakage inductance in a CT restricts the turns ratio from being too high. But if the turns ratio is very low then the output may be unstable because of distortion.

Burden Resistance

The secondary of a CT should not be kept open. So a burden resistor is kept across the secondary terminals of the CT. It determines the value of secondary current Is​​. It helps to avoid the loading effect at the CT secondary and also addresses the no load and full load connected condition. Burden Resistor Rb is also necessary if we want to connect multiple CTs in a cascaded fashion. Let VCT = Erms be the emf obatained at the transformer secondary. Then the secondary current Is is given by

Is = VCTRb\frac{V_{CT}}{R_{b}}

Now, Is = IsNpNs\frac{I_{s}N_{p}}{N_{s}}

Therefore, Rb​ = VCTNsIpNp\frac{V_{CT}N_{s}}{I_{p}N_{p}}

Corresponding to the value of Np max​​ we can obtain the value of Rb min.

B-H Curve

The B-H curve is a plot of magnetic flux density (B) v/s magnetic field strength (H) for a particular material. This is an important curve for selecting the material of the transformer. It helps us to compare the magnetic properties of one material over another while selecting a material for a specific application.

The following two graphs show the B-H characteristics for two different materials with the highlighted zone showing the desired region of operation for linearity.

F39 loop_1.png
F39
    P11 loop.png
    P11
      B-H curve (from material datasheet)

      The zone of operation on the B-H curve should be as linear as possible for our purpose since we want to replicate the primary current as a proportionally calibrated emf on the secondary side.

      Calculation of Imin and Imax:

      From the fig 1.a) it can be seen that as per the highlighted zone, Hmin =30A/m and Hmax =120A/m.

      Now for F39 core , le = 30.14 mm. Hence for a fixed number of turns in the primary Np =19, we have Imin = HminleNp\frac{H_{min}l_{e}}{N_{p}}Imin = 30×30.14×10319\frac{30 \times 30.14 \times 10^{-3}}{19} = 0.047 A = 47 mA

      Imax =  HmaxleNp    \frac{H_{max}l_{e}}{N_{p}}  Imax = 120×30.14×10319\frac{120\times30.14\times10^{-3}}{19}  =0.190 A = 190 mA

      The above calculation gives the range of current can be addressed in a linear relation using the F39 core.

      Zero Crossing Detector and Snubber

      A zero crossing detector detects the point of zero crossing of the voltage. This mechanism inside the optocoupler ensures that the gate of the power triac is triggered only when the supply voltage crosses zero so that the load is not exposed to sudden high voltage when turned on.

      The following waveforms clearly depict how a triac driver only triggers at the point of zero crossing and also stops at the point of zero crossing.

      zero crossing driver.JPG
        Waveforms for triggering of Triac gate using Optoisolator Triac Driver

        Since the current is not in phase with the voltage in case of capacitive and inductive loads, hence the semiconductor device which is acting as the electronic switch is exposed to high commutating dv/dt, which may be even higher than the static dv/dt rating of the device.Hence, the snubber circuit, which is basically an arc suppression circuit is put across the device to damp out the voltage spikes or high dv/dt which may result in false triggering of the device or may even cause electrical damage. ​​

        Zero crossing.JPG
          Waveform for depicting voltage across power triac while driving highly inductive loads

          Filter Circuits

          Noise can be a very disturbing factor in signal detection and measurement, especially if the signal is itself very weak. The filter circuits are those which eliminate high-frequency noise and only allow the signal to pass. These are called low pass filters and can also have the ability to filter harmonics if specially designed. The main problem with passive filter circuits is that the signal may get attenuated when passing through it.

          Circuit

          Basic Block Diagram

          The following figure shows the sensor circuit integrated into the total energy management system.

          block_1.jpg
            Block Diagram of the sensor circuit with the system 

            Main Components

            The main components in the circuit and their purpose are as described below:

            Intelligent Power Supply

            It is a power supply module for providing power to the microcontroller and signal conditioning circuit with·the following properties:

            ‣Highly accurate and low quiescent power consumption.

            ‣Delivers as much power as is demanded by the load.

            ‣Low footprint.

            ‣Low derating of power.

            ‣Wirelessly manageable sleep/wake up modes to avoid unnecessary power consumption.

            Failsafe Mechanism and Snubber

            It is for the purpose of protection. It protects the control circuit from any short circuit or overloading of the AC load and also protects the sensor circuit and the ac load from any switching or lightning-induced surges and spikes in the line causing electrical damage. It consists of components like fuse, MOVs etc. Snubber is for the dv/dt and di/dt protection of the semiconductor device.It can be any arc suppression circuit such as an R-C series circuit.

            Signal Conditioning Circuit

            It is the instrumentation and feedback circuitry which is directly fed with the CT sensor output. It has the following properties:

            ‣It has the capability to control multiple sensors (max 8 in this case).

            ‣It has high noise rejection capabilities.

            ‣It has a provision for configurable gain.

            ‣It consumes low power and has a minimal footprint.

            ‣It is wirelessly managed.

            Wireless Microcontroller

            It is the like the brain of the management system and it can have duplex communication with the remote server. It has the following properties:

            ‣It follows low power communication protocols.

            ‣It is a processor with offline computation and managing capabilities.

            ‣It has low quiescent power consumption and minimal footprint.

            Triac

            For the purpose of switching the load, we require a semiconductor switch. An SCR provides only unidirectional conduction controlled by a third terminal called the gate. A Diac is a bidirectional switch which can conduct in both directions above a certain threshold voltage but with no controlling third terminal.

            The features of bidirectional conduction in diac and third controlling terminal in thyristor has been combined into a single semiconductor device called triac. It means TRIode for Alternating Current. It is a three terminal semiconductor junction device. It works in two active quadrants Q1 and Q3. The two main terminals are MT1 and MT2 and gate(G) is the controlling third terminal. There are four possible combinations:

            Possible combinations for triggering of Triac
            Applied VoltageGate triggering Current

            Quadrant of Operation

            State of Conduction

            MT2+

            G+

            Q1

            ON

            MT2+

            G-

            Q1

            OFF

            MT2-

            G-

            Q3

            ON

            MT2-

            G+

            Q3

            Off

            The triac can be thought of as two back to back thyristors. It can drive ac loads. It is to be noted that in the above table the datasheet refers to MT2 and not MT1. This is because the triac is like a solid state switch and load will be connected to T2 only and not T1.

            The graph below clearly explains the schemes for the above possible operations:

            Triac characteristics.png
              Triac V-I characteristics

              In the circuit, we have used BT 136.

              Optoisolator

              Optocoupler is a device which contains infrared LED and some photodetector such as photodiode, darlington pair, phototransistor, triac etc. It acts like a solid state relay. It acts as an interface between digital control circuit and AC loads usually operating at different voltage levels (LV /HV). The gate of optoisolator internally works on an optical principle. Low voltage digital logic (DC) of microcontroller can be used as an input to the optoisolator to provide gate triggering pulse as an output to the power triac to drive and control the state of high voltage AC loads. The extremely high isolation voltage also protects the low voltage logic circuits (LHS) from surges and spikes arising out of load/power line (RHS). It also eliminates traditional methods of providing electrical isolation like the use of relay controlled contact or isolation transformers and can be used for fast electronic control of the load along with minimum size. Its output terminals also have very high dv/dt withstanding capability. It also has an internal zero crossing detector circuitry implemented within the chip which ensures that the power triac gate is triggered only when the supply voltage crosses zero so that the AC load does not experience any abruptly high voltage during on/off operation.

              In the circuit, we have used MOC 3043.

              Toroidal Current Transformer

              Structure of the core has been chosen as toroidal because such cores provide a continuous path for the flux lines to pass without any air gap. So the reluctance of the core is very low with minimum leakage flux. Four different core materials for the toroidal core of CT have been used. Those are F39, P11, T38 and Nanocrystalline. For the purpose of winding SWG 26 enamelled copper wire was used.

              Some important parameters related to toroidal core transformers are listed below:

              ✶C1, C2 → Geometric Core constants

              ✶A→ Outer Diameter of the toroid [A=2×r2]

              ✶B→ Inner Diameter of the toroid [B=2×r1]

              ✶C→ Height of the toroid [C=h]

              ✶le → Effective length

              ✶Ae → Effective area

              ✶AL →Inductance factor

              ✶L → Inductance

              Torus2.gif
                Toroidal Core

                The following relations mathematically relate the above parameters:

                C1 = 2πhloge(r2r1)\frac{2 \pi }{h {\log_e (\frac{r_{2}}{r_{1}}) }}  

                C2 = 2π(1r11r2)h2loge3(r2r1)\frac{2 \pi (\frac{1}{r_{1}}-\frac{1}{r_{2}})}{h^2 \log_e^3 (\frac{r_{2}}{r_{1}})}

                le = C12C2\frac{C_{1} ^ 2}{C_{2}}  

                Ae = C1C2\frac{C_{1}}{{C_2}}

                AL = μAele\frac{\mu A_{e}}{l_{e}}

                L = AL×N2

                Methodology

                • The circuit design was designed keeping the EMI/EMC considerations by using the DIP TraCe software which gives a file in xyz format. Then it was first converted into a gerber file from which the g code was generated. The circuit was printed on PCB using a CNC PCB milling machine rather than chemical processing (etching) for prototyping purpose. It was printed on single layer FR-4 glass epoxy laminated board.
                SensorPCB.jpg
                PCB Design
                  Circuit on board.jpg
                  Sensor Circuit
                    Sensor Circuit on PCB
                    • Respective components were then placed on the board and soldered at an appropriate soldering temperature, 300ºC. Thermal paste was used wherever necessary.
                    • Considering the limitations of Np and Ns, winding of transformer was done manually with bare hands carefully so that the enamel is not removed. At the ends, the enamel was removed to get the copper contact at the terminals. Tight winding is done ensuring that there is minimum air gap left between toroid and the wire. Elecreical isolation was maintained between primary and secondary windings using sleeve and the transformer was then placed in the circuit.
                    • The observations were taken under the following conditions for each of the four cores F39, P11, T38 and Nanocrystalline:

                    ★Keeping Np constant and varying Ip.

                    ★Keeping Ip constant and varying Np.

                    • Ip has been varied by varying the wattage of the bulbs connected as the load while Np has been varied from 1 to Np max. For the case of constant Np, it has been kept close to Np max (say for Np max =22, Np has been kept as 20). For the case of constant Ip, it has been kept close to Ip max (say for addressing current up to 1A, Ip has been kept as 0.868 A which is the current drawn by two 100W bulbs).
                    • Input is the line current applied to transformer primary and output is the emf taken across the CT secondary using high precision Scientific 4½ DMM incorporating True RMS using integral function and Tektronix 2002B 16 channel 70MHz MSO with noise filter.​​

                    RESULTS AND DISCUSSION

                    Comparative Graphs from collected data

                    Constant Np new graph_1.png
                    • 1
                    • 2
                    • 3
                    • 4
                    Np v/s Erms for a constant load of 200W
                    Constant Np new graph_1.png
                    • 1
                    • 2
                    • 3
                    • 4
                    Ip v/s Erms for constant turns ratio

                    From the above two graphs it is clear that higher values of emf can be achieved at the CT secondary using T38 core but the results, although less, are more linear for F39 and P11 cores. F39 and P11 cores give almost similar outputs though the P11 output is lower than the F39 output. This is possibly because of the lower value of initial permeability in the case Np or Ip . Nanocrystalline gives a lower value of emf due to a lesser number of turns in the primary and secondary.

                    Waveforms obtained from MSO

                    P11-120W-10Np100Ns new Ng.png
                    Load=120W, Np:Ns=10:100
                      P11-120W-10Np100Ns-50Hz-Magnified Sampled new Ng.png
                      Load=120W, Np:Ns=10:100
                        P11 Core

                        Fig 8.b) is the sampled version of Fig 8.a) The waveform obtained is not sinusoidal and there is a lagging tendency at the leading edges in both positive and negative cycles which is probably due to lower initial permeability value of P11.

                        T38-160W-10Np100Ns new Ng.png
                        Load=160W, Np:Ns=10:100
                          T38-30W-10Np100Ns Sampled new Ng.png
                          Load=30W, Np:Ns=10:100
                            T38 Core

                            From Fig 9.a) and Fig 9.b) it is clear that the waveform of the latter one is more close to a sinusoid. This is probably due to the highly linear B-H characteristics of T38 at lower values of magnetizing strength H i.e., at lower values of load current.

                            F39-200W-10Np100Ns new Ng.png
                              F39 Core,  Load=200W, Np:Ns=10:100

                              From Fig 10 it is quite clear that the waveform obtained for the F39 core is nowhere close to sinusoidal and has lots of induced harmonics due to noise.

                              NC-15W-10Np50Ns new Ng.png
                              Load=15W, Np:Ns10=:50
                                NC-200W-5Np50Ns-magnified new Ng.png
                                Load=200W, Np:Ns=5:50
                                  Nanocrystalline Core

                                  Fig 11.a)is the closest to sinusoidal. This is because of the near perfect magnetic properties of Nanocrystalline which helps to replicate the variation in secondary emf in the same way as the variation in primary current.

                                  Some noise introduced from the line in the primary winding was getting amplified in the secondary. The cores which have lower coercive field strength HC and lower remanent flux density Br are more susceptible to noise. This can be easily understood from the differences in the readings obtained from the DMM and the MSO. Most of the readings were higher in MSO as compared to DMM. The DMM has a fixed internal noise filter which rejects the noise to give the readings while the MSO has an adjustable noise filter which can be varied from 6 kHz to 70 MHz. If the filter frequency is set at higher levels, then the displayed values are higher because the noise also contributes to the output obtained.

                                  The best material out of the four tested as toroidal core of CT is definitely Nanocrystalline, as expected, due to its high permeability which offers low reluctance to flux path, high Hs which allows a more linear range of operation, higher Bsat due to which higher values of emf can be observed in the secondary, low Hc and Br which leads to lower power loss in the hysteresis cycle. It is to be noted that choosing a higher SWG allows a higher number of turns to be accommodated due to its thinner diameter but it also decreases the current handling capability of the wire. Keeping both the factors,i.e, current carrying capacity and number of turns required, SWG 26 was found to be the best suited for our purpose.

                                  CONCLUSION

                                  The selection of the core material depends on the application and the desired range of current detection. Although the very basic primary sensor circuit with four different toroidal core materials was realized and a comparative study was completed, further work is needed to find the best-suited material and optimum turns ratio, which will give a calibrated output emf against any primary load current.

                                  REFERENCES

                                  Web Resouces & Documents referred to for studying the components, datasheets and specifications:

                                  F39, P11 datasheets: https://www.mmgca.com/catalogue/MMG-Neosid.pdf

                                  T38 datasheets: https://en.tdk.eu/download/519704/069c210d0363d7b4682d9ff22c2ba503/ferrites-and-accessories-db-130501.pdf

                                  Nanocrystalline Core datasheet: https://www.mouser.com/pdfdocs/VACChokesandCoresDatasheet.pdf

                                  BT 136 datasheet: https://www.promelec.ru/pdf/BT136-600-NXP.pdf

                                  Optoisolator Triac Driver datasheet: http://www.mouser.com/ds/2/149/fairchild%20semiconductor_moc3031m-544476.pdf

                                  Information about ferrite and toroid parameters: https://product.tdk.com/info/en/catalog/datasheets/ferrite_summary_en.pdf

                                  Application note about Zero Voltage Crossing Optically Isolated Triac Driver: ​https://www.onsemi.com/pub/Collateral/AN-3004.pdf.pdf​

                                  Papers referred for literature review:

                                  [1] Kurmi, V.P., Sharma, D, Singh, D.K. (2016). Analysis of the behaviour of various electrical loads with variable voltage and variable frequency, International Journal of Science, Engineering and Technology Research (IJSETR),Volume 5, Issue 5, May 2016

                                  [2] Yang Zhen, Microchip Technology Inc., Current Sensing Circuit Concepts and Fundamentals, Microchip, Application Note AN-1332

                                  [3] Forghani-zadeh, Hassan P., Rincón-Mora, Gabriel A. (2005). Current-Sensing Techniques for DC-DC Converters, Georgia Tech Analog Consortium, School of Electrical and Computer Engineering, Georgia Institute of Technology

                                  [4] Sung-Yong Son (2015). Home Electricity Consumption Monitoring Enhancement Using Smart Device Status Information, International Journal of Smart Home, Vol. 9, No. 10, (2015), pp. 189-196

                                  ACKNOWLEDGEMENTS

                                  With great pleasure, I express my gratitude to my project guide Dr K.C. James Raju, Professor, Centre for Advanced Studies in Electronics Science and Technology (CASEST), School of Physics, University of Hyderabad for his guidance.

                                  I offer my sincere thanks to Mr G. Krishna Prasad and his team Conservision Technologies at TIDE Centre, University of Hyderabad for their constant support, motivation and guidance in developing the calibrated monitor-measure-manage sensor.

                                  I also want to thank my teammates for this project Mr Sadanand Powar and Mr Ninad Nikalje without the help of whom this work could not have been completed.

                                  I express my heartful thanks to my professor, Dr Abhijit Chakrabarty, Department of Electrical Engineering, Indian Institute of Engineering Science and Technolgy, Shibpur for his recommendation letter.

                                  I am extremely grateful to the Indian Academy of Sciences for providing me with the fellowship and such a great platform to do a fruitful summer project.

                                  The AuthorCafe platform was extremely user-friendly and helped me in compiling my report.

                                  Last but not the least I would like to thank my parents and friends who have always been very supportive, without the help of whom this project would not have been possible.

                                  References

                                  • https://www.onsemi.com/pub/Collateral/AN-3004.pdf.pdf

                                  Source

                                  • Fig 1 a: http://www.mmgca.com/catalogue/MMG-Neosid.pdf
                                  • Fig 1 b: http://www.mmgca.com/catalogue/MMG-Neosid.pdf
                                  • Fig 2: https://www.fairchildsemi.com/application-notes/AN/AN-3004.pdf
                                  • Fig 3:   https://www.fairchildsemi.com/application-notes/AN/AN-3004.pdf  
                                  • Fig 5: https://electronicspost.com/sketch-the-v-i-characteristics-of-a-triac-describe-some-of-its-important-applications/
                                  • Fig 6: http://g3ynh.info/zdocs/magnetics/mag-core.html
                                  More
                                  Written, reviewed, revised, proofed and published with