SO GMC es uo) {as current transducers for protective relays in industrial, rene mc ees eu CEC ne) application quides deal with the steady-state sine wave Peete eC ees recommended rules applied to avoid transient saturation, it always occurs in certain critical relay applications. Analyzing and PN yi meats Besa) ‘Analyzing and Applying Current Transformers is a reference book written for professionals concerned with the application and operation of CTs used as current transducers for input Pepe ee Ae cu hae ces Pesca ene cee en ey Potro cu acu ue cent ecicca eee eLearn nce COS ee Se RC cise Lee eect Sue cna a ease Brea aR aed eee eo a ee een ree Cue mike ce ee eae cee oe eee See ee en Ca Pena en hea) Cee Cu tac enicay 19471991, where he held various engineering positions, Penis casace unc a Bere mec ee eS TCC Ca Tee TTer ect g Bene Nee Sane Oe Cee eR) protection using solid state and microprocessor technology ‘and is the author of numerous IEEE and Protective Relay Conference papers. He received the Best Paper Award of the 1988 Petroleum and Chemical Industry Conference and the Power System Relaying Committee's Distinguished Service eens ISBN 0-9725026~2-9 Crete aie rs TET ey ALT) iiiCopyright © 2004 Schweitzer Engineering Laboratories, Inc. All rights reserved. No part of this publication may be reproduced, stored in 8 retrieval system, or transmitted, in any form or by any means, electronic, ‘mechanical, photocopying, recording, or otherwise, without the prior written permission of the author and/or publisher. For information, contact Schweitzer Engineering Laboratories, Inc.,2350 NE Hopkins Court, Pullman, WA 99163. Library of Congress Cataloging-in-Publication Data Zocholl, Stanley F. ‘Analyzing and Applying Current Transformers / Stanley E. Zocholl P om Includes bibliographical references (p. 95) ISBN 0-9725026-2-9 1. Eleotic transformers. 2. Electric ransformers-Protection. I. Title, "TK2551 263 2008 First Edition: August 2004 Printed and Bound in the United States of America SEL is a registered trademark of Schweitzer Engineering Laboratories, Inc. Transformer Differential Relay Applications .. Contents Introduction ..........6....05 ola BYERS o wees tue 1 Current Transformer Equivalent Circuit ..... aasee cme Volt-Time Area and Saturated Waveforms ............5 The Excitation Curve ............ Computer Simulation .... Flux Density Limit of the Volt-Time Area . The Criterion to Avoid Saturation .. Rating Current Transformers for Line Protection .......... 23 Current Transformer Selection Procedure ..... sa ows Calculating Current Transformer Burden. 6 Limits of the Criterion . a Statistics of Asymmetry 2200 a ameanns ogy ACase History ...... 33 kA Fault With Ideal Current Transformers ‘The Saturation Effect veeeeeees 33 kA Fault With Maximum Asymmetry in A-Phase . 33 KA Fault With Offset in B-Phase and C-Phase CT Application Rules . Generator Differential Relay Application . .. Bus Differential Relay Applications . Percentage Slope Characteristics . .Alpha Plane Differential Character Determining the Slope for a Given Degree of Saturation . ayaa ve Saturation Voltage 1 ana - 65 Bus and Machine Differential Applications - -66 The Effect of Remanent Flux on the Slope Setting .. . .69 The Impact of High Fault Current on Overcurrent Protection .......... as sien vines 72 Limitations of an 80 A Instantaneous Setting ........ 74 Defining a CT Selection Criterion ..... seine soning wel Maximum Fault Current With an 80 A Instantaneous Setting . 81 Minimum CT Rating With an 80 A Instantaneous Setting... 82 Implementing Instantaneous Overcurrent Elements . . . 83 Digital Implementation of Instantaneous Overcurrent Relays . 83 Digital Filter Performance With CT Saturation 84 ‘The Cosine-Peak Adaptive Filter . 86 A Simple Distortion Index ...... 87 Annex A: Volt-Time Area Computer Simulation 93 95 References Figure | Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 Figure 15 Figures Toroid Curent Transformer ........ Current Transformer Equivalent Circuit Secondary Current Waveforms for Various Levels of Excitation (Resistive Burden) Core Flux and Secondary Current in a CT With a 20-Per-Unit Symmetrical Rating Carrying 10-Per-Unit Current With the Offset Due to 0.010 Second Primary ‘Time Constant (75° Impedance Angle) (€400, 2000:5 Current Transformer Excitation Curve ‘Showing the 300:5 Tap With Knee-Point Tangent and Normal Lines fo tieeeees Magnetizing Impedance : Equivalent Circuit ofa Current Transformer Equivalent Circuit for the Differential Connection of Carrent Transformers B-H Loop andthe Slope ofthe Mean Curve (C400, 2000:5 Current ‘Transformer Computer Simulation (C400, 2000:5 Current Transformer With 40 kA Asymmetrical Current Burden Voltage for Asymmetrical Fault Curent Event 1—33 kA ABC Fault at 1.55 Miles on a 230 kV Line (Theoretical—No CT Saturation) Secondary Curzent in a C800, 3000:5 With 33 KA Offset ‘Three-Phase Fault at 1.55 Miles With 0.066 Primary ‘Time Constant Magnitude and Phase Difference of Serle Sesont Current a z . 10 13 4 16 18 18 20 31 32 34Figure 16 Figure 17 Figure 18 Figure 19 Figure 20 Figure 21 Figure 22 Figure 23 Figure 24 Figure 25 Figure 26 Figure 27 Figure 28 Figure 29 Figure 30 Figure 31 Figure 32 Figure 33 Figure 34 Figure 35, Event 2—33kA ABC Fault at 1.55 Miles on a 230 KV Line (CT Saturation Caused by Maximum Offset in A-Phase) sta Event 3—33KA ABC Fault at 1.55 Miles on a 230 KV Line (CT Saturation Caused by Offset in B- and C-Phase) Power Plant Auxiliary Differential Relay Restraint Currents h andy Ig [s the Differential. . Harmonic Content of the Differential Current in Figure 19 Plotted in Per Unit of the Fundamental Frequency Secondary Currents I, and I Due to a 58800 A ‘Tarough Fault Current With Offset. The Differential Carrent Ig Is Also Plotted. 6000:5 CTs Have Knee-Point Volts of 700 and 500. X/R=52. ... Secondary Currents I; and Ip Due to an Asymmetrical ‘Through Fault of $8800 A. Adding Burden Resistance Decreases the Differential Current ly. 6000:5 CTs. Knee-Point Voltages of 700 and 500. X/R = 52, Differential Protection .. Percentage Characteristic a xa Saturation Due to DC Offset. C200 2000:5 CT, 2.0 © Burden, Fault Current 10667 Amp. X/R Fundamental Waveforms Extracted by Digital Circle Characteristic Characteristic Resembling @ Cardioid MATLAB Program Graphic Interface VS = 20 Fault Current 10667 VS = 40 Fault Current 10667 VS = 80 Fault Current 10667 VS = 160 Fault Current 10667 VS = 80 Fault Curent 10667A With Cardoid-Shaped Characteristic VS = 160 Fault Current 10667A With Cardioid-Shaped Characteristic .. we 37 39 44 245 49 50 52. 52 54 35 58 . 58 - 59 2 2 8 6 Figure 36 Figure 37 Figure 38 Figure 39 Figure 40 Figure 41 Figure 42 Figure 43 Figure 44 Figure 45 Figure 46 Figure 47 Figure 48 Figure 49 Figure 50 Figure 51 Figure 52 Figure 53 Figure 54 Figure 55 Figure 56 Slope k Versus CT Saturation Voltage ‘Muhtrestraint Bus Differential Vs=40,k=30 Operate Signal Versus Restraint VS = 67, k= 44, 40 Percent Remanence Operate Signal Versus Restraint With 40 Percent Remanence Generator Auxiliary Bus With High Fault Curent. Relay Schematic Showing Measured Signals... ‘Signal Extracted With a 200 A Primary CT Current . CT and Relay Signals for a 40 KA Fault Using C50, 100:5 CTs Response to a 40 KA Fault Using C200, 20055 CTs Response to a 40 kA Fault Using C400, 400:5 CTs Diagram of Bipolar Peak Detector Bipolar Peak Detector Transient Overreach RMS Filter Transient Overreach Compared Bipaar Peak Detector Filte Response, Fault 40 kA, X/R = 20, C100, 2005 CT, 0.5 W Burden Instantaneous Element Using the Cosine-Peak | ‘Adaptive Filter Cosine Filter Providing Trip for a Waveform With Low Saturation C400, 200:5, 4500 A Fault With X/R =11.31 . Bipolar Peak Filter Providing Trip for a Waveform With High Saturation C50, 200:5, 20 kA Fault With 7R = 11.31 : Cosine iter Proving Trip fora Waveform With No Saturation C100, 200:5, 4kA Symmettical Fault Current . ‘Mathcad Plot of Current and Flux 65 0 68 = 8 n n 75 16 76 -78 ESS 85 85 87 89 90 1 93Introduction ‘The purpose of this book is to develop an understanding of how to analyze and apply current transformers (CTs) used as current sources for protective relays. A CT appears to be the simplest of electrical devices. For example, a bushing CT is simply a winding ‘on an insulated core, which becomes a transformer only when placed over a primary conductor as shown in Figure 1. Primary Conductor Figure 1 Toroid Current Transformer The opening paragraph of S, D. Moreton's classic paper [1] states that the art of calculating current transformer characteristics from excitation curve data has been known for some time. Moreton wrote this in 1943 and referred to papers written two years earlier. Today, nearly 60 years later, there is still an abiding interest in this seemingly simplest of electrical devices.Introduction For example, the Power System Relaying Committee Working Group I16 is presently revising IEEE Std C37.110-1996 Guide for the Application of Current Transformers Used for Protective Relaying Purposes [2] to include current transformer selection criteria for transmission line protection. A former working group completed a paper addressing the problems of applying low ratio current transformers and the adverse effect of saturation on relay performance with extremely high fault currents. At the same time, the Electric Power Research Institute (EPRI) funded a project to study models used to determine CT transient performance for relay testing using the Electromagnetic Transient Program (EMTP). Tdeally, the CT functions to produce ratio current in the secondary winding that is an undistorted replica of the primary waveform with the magnitude of the primary current divided by the turns ratio. The ‘current transformer performs its function accurately over a range of primary current determined by its current rating, burden, and the X/R ratio of the protected circuit. ‘Standard ratings and application guides deal with the steady-state sine wave behavior of a current transformer. However, despite the use of recommended ru'es applied to avoid it, transient saturation always occurs in certain critical relay applications. Consequently, the relay engineer needs to understand the nonlinear characteristics of current transformers, accuracy ratings, and the transient bebavior of current transformers. Our purpose is to present current transformer concepts that provide these fundamentals. Current Transformer Equivalent cuit We study the behavior of a current transformer using an equivalent Circuit. The model is a constant current source where ratio current injected into a magnetizing impedance that is in parallel with the burden as shown in Figure 2. Using a reactance to represent the ‘magnetizing leg of a current transformer provides a useful visual concept. However, magnetization is a nonlinear phenomenon that Tequires the value of reactance to chang? for each level of excitation. For example, the three diagrams, shown in Figure 2, show flux @ versus magnetizing current /,,, representing low, moderate, and high levels of excitation, At low excitation, the slope d¢/d! representing the inductance is Tow. The low slope indicates a disproportionate amount of ‘magnetizing current compared to the burden current at low excitation. At moderate excitation, the d6/d/ is relatively high and the magnetizing current is small compared to the current in the burden, At high excitation, the versus J,, curve exhibits the ‘maximum slope in transition between scturated states. The fact that ‘magnetizing current is so small compared to the ratio current during the transition suggests that it can be igncred. Consequently, the core can be viewed simply as a volt-time switch as shown in Figure 2, which opens during a rate of flux change and closes during saturation.Current Transformer Equivalent b 3 t Circuit % A lof Low Ty Excitation Moderate Ty Excitation High Ty Excitation Figure 2 Current Transformer Equivalent Circuit Volt-Time Area and Saturated Waveforms Secondary terminal voltage ratings designate relaying accuracy ratings. The concept of the voltage rating can be derived from the fundamental relation: e=v@ a Where ¢ isa voltage induced by N’ turns and a rate of change of flux 4 with respect to time ¢, and where the flux is limited to the transition from the saturation flux $5 to 9. and e = Vsin, then integrating Equation 1 gives: 2GgN = fede = Vi" sin(Oyar @ Equation 2 shows that the saturation flux 26, and the number of turns 1 limit the area under the secondary voltage sine wave, The Voltage rating of the CT is the rms value of the sine wave where the ‘lux-limited volt-time area just fits under the half cycle of the sine wave. Furthermore, if the current is increased beyond this point, saturation occurs and the sine wave is cut off at an angle less than 180 degrees. Expressing the instantaneous value e as a function of the volt-time area can show the process of saturation. To do this, letVolt-Time Area and Saturated Waveforms Vs = 2dsN. Making this substitution and integrating Equation 2 gives: cos(®,) ® Expressing cos(0,) in terms of sin(@,) and noting that sin(®,) = (e/V) gives: v, ? af @) Solving Equation 4 for the instantaneous voltage e during saturation gives: — é prdy-2 (Ss) Equation 5 can be expressed in multiples of a CT-rated current by letting V = MV, and V, = M,V, where V, is the peak secondary voltage when rated current is applied to the transformer. With these substitutions, Equation 5 becomes: My e a it (w © Equation 6 is plotted in Figure 3 with the resulting voltage ‘waveforms at various multiples of rated current. As shown Figure 3, an input of (5/2) produces a sine wave output with a volt-time area equal to that of the transformer core. When the input «> the output exists for 90 degrees of the sine wave and the last \Volt-Time Area and Saturated Waveforms linear peak occurs. When the input exceeds Mg, the volt-time area ‘occurs in less than 90 degrees and the peak output follows the parabola described by Equation 6. ' Ney ty 4 te = [as Mall? ‘ ‘ Tem 5 oO 2 4 (Muti of CT Rating) Tine Figure 3. Secondary Current Waveforms for Various Levels of Excitation (Resistive Burden) The volt-time area concept gives a clear illustration of the voltage rating and the limit of sine wave operation of a CT. It also provides an uncluttered view of the saturation process and the waveforms produced. The simple Mathcad® worksheet listed in Annex A. implements the volt-time concept for demonstration purposes. Figure 4 shows a program output plot fora CT that has a 20-per-unit rating but is subjected to a 10-per-unit current with a de component from a 0.010-second time constant (75-degree impedance) primary circuit.\Volt-Time Area and Saturated Waveforms Gore Fiux Secondary Current -al a am ne 08 08 Seconds Figure 4 Core Flux and Secondary Current in a CT With a 20-Per-Unit Symmetrical Rating Carrying 10-Per-Unit Current With the Offset Due to 0.010 Second Primary Time Constant (75° impedance Angle) ‘The plot shows the saturation during the de offset. Please note that the volt-time area derivation is based on a resistive burden. The general impedance burden case is addressed in “Computer Simulation” on page 13. The Excitation Curve ‘The volt-time concept assumes the magnetic core is a volt-time switch. This concept assumes no magnetizing current when there is rate of change of flux and the entire ratio current flows to the burden, When saturation flux is reached, as indicated by volt-time ‘area, and there is no longer a change of flux, the switch closes, shunting the entire ratio current away from the burden until a reversal of current and integration becomes negative to reduce the flux. Here, saturation occurs at a well-defined point indicated by a specific value flux and turns, AA finite amount of ampere-turns is required to establish flux in a CT ‘core and can be expressed as magnetizing current measured at the secondary terminals. The excitation current, which is subtracted from the ratio current, has definite values for each value of voltage as shown in Figure 5. This curve depicts steady-state voltage versus excitation current where the voltage is measured with an average reading voltmeter calibrated in rms. It is actually a plot of flux versus magnetizing current since the average voltage is the volt-time integral averaged over the period of the sine wave. ‘The excitation curve for a C400, 2000:5 multiratio bushing CT, shown in Figure 5, is a measure of CT performance and can be used to determine ratio correction factors at various levels of steady-state‘The Excitation Curve excitation. While the excitation curve has a well-defined knee point, it has no discernible point of saturation. For this reason, relaying accuracy ratings are based on a ratio correction not exceeding 10 percent, and ratings are designated by classification and secondary voltage. The secondary voltage rating is the voltage the (CT will support across a standard burden with 20 times rated current without exceeding 10 percent ratio correction. 1000 7 e008 SAI Secor Exctaton Vos FU) a ast oy + “0 Figure 5 C400, 2000:5 Current Transformer Excitation Curve Showing the 300:5 Tap With Knee-Point Tangent and Normal Lines A classification and a secondary terminal voltage rating designate ANST relaying accuracy ratings. Toroid current transformers with a distributed winding have negligible leakage flux in the core and are given the C classification, indicating that the ratio can be calculated. Current transformers with a wound secondary winding and significant leakage flux in the core are given the T classification, indicating that the ratio must be determined by test. 10 ‘The Excitation Curve The standard burdens for relaying are 1,2, 4. and 8 ohms, all with an impedance angle of 60 degrees. Consequently, at 20 times the ‘S-ampere rated secondary current, a multirario CT is voltage rated ig the maximum tums ratio. The standard ratings are 100, 200, 400, and 800 vols. Since the CT rating occurs with 100 amperes of secondary current at a 10 percent ratio correction factor, the voltage rating can be read from the excitation curve at 10 amperes of excitation current, We must first subtract the internal voltage drop due to the de resistance of the winding. For the 2000:5 ratio winding in Figure 5, the voltage read from the curve at 10 amperes is 496 volts. In this case, the voltage is less than the 800 rating and greater than 400. Therefore, the rating is C400, provided the 400 turn winding has less than a 0.0024 ohms-per-turn de resistance to ‘guarantee an internal voltage drop less than 96 volts. ‘The magnetizing impedance of the current t-ansformer can be obtained by dividing each value of voltage read from the curve by the corresponding excitation current. The impedance is plotted in Figure 6. The magnetizing impedance is nonlinear, increasing from 3000 ohms at 0.001 amperes excitation current to a maximum of 8000 ohms at 0.0248 amperes of excitation. This is the point of maximum permeability and is located by the 45-degree tangent to the curve. The impedance values decrease from this point as the excitation is increased, reaching 49 ohms at {0 amperes of excitation current. 1"The Excitation Curve 0000 = Zmerms velle a Excitation Curent Figure 6 Magnetizing Impedance Any point on the maximum ratio curve can be referred to a lower ratio tap by using a constant volts-per-turn relation for the voltage and a constant ampere-turn relation for the current. Consequently, y, 1 ane a N, “aa H so that the knee-point of each curve lies along the normal line to the 45-degree tangent drawn in Figure 5, which also shows the curve for the 30:5 ampere minimum tap. 2 Computer Simulation ‘Because current transformers are subject to saturation during the de transient of fault current, there is a growing interest in computer simulations like the EMTP program, which produces detailed plots of current corresponding to an oscillograph obtained from a full- scale system test. The primary aim of the simulation is to obtain digitized records, which can be reconstituted as secondary analog signals using D/A conversion and amplification, for the purpose of relay testing, ‘The simple equivalent circuit for the simulation is shown in Figure 7, where the primary current referred to the secondary is a constant current feeding a nonlinear magnetizing inductance and the resistance and inductance of the burden. Figure 7 Equivalent Circuit of a Current Transformer 3Computer Simulation ‘The equivalent circuit in Figure 8 is needed to simulate the interaction between the two CTs and the restraint and operating, ‘windings of a differential relay. In Figure 8, the side A and side B CTs are represented by the ratio currents /, and I, and nonlinear ‘magnetizing inductances Ly and Lg. Linear elements Rj, Ly,Ry, and L, represent resistance and inductance of the leads and include the resistance of the CT secondary windings. R, and Ls represent the linear elements of the relay operate winding. Lg. is the nonlinear inductance of a saturable reactor, a necessary element in the representation of machine differential relays. Figure 8 Equivalent Circuit for the Differential Connection of Current Transformers ‘The simulation is the time-discrete solution of the differential equation of the voltage drops caused by the loop currents. A recursive solution using a finite time increment is used because the magnetizing inductance Ly, changes continually as a function of the voltage. Thus, the output of the program is an increment-by- increment time plot of the current, voltage, and flux density. Programs like the EMTP program use a piece-wise linear representation of the saturation curve of the current transformer that the user must enter as a series of points. Other programs use an 14 ‘Computer Simulation equation to emulate the B-H curve of the CT. The characteristic ‘S-shaped curve of the Frohlich equation is made to fit the B-H curve for oriented grain steel by equating the slope of the curve at the origin to the maximum incremental permeability (on order of 5000), and the slope during saturation to the permeability of fice space (4x 10°’). The inductance Ly is then calculated using an incremental permeability equal to the slope of the B-H curve as emulated by the Frohlich equation: Mot A 1 (8) where: Ly isthe magnetizing inductance in Henries N- isthe number of turns A isthe core cross-sectional area (sq. meters) hip is the permeability of free space h, isthe core material incremental relative permeability 1 is the mean length of the core (meters) ‘The plot of the B-H curve emulated by the Fréhlich equation is, shown in Figure 9. The permeability is taken to be the slope of the curve through the center of the hysteresis loop. The curve of the slope of the B-H curve in Figure 9 emphasizes that the magnetizing inductance has no single value but changes for every value of instantaneous excitation during a cycle. In addition, the excitation ‘curve indicates that the slope of the B-H loop varies with the level of excitation, Since simblations are used to analyze high excitation involving saturation, the slope at the origin is set equal to the maximum permeability of the core, The maximum relative permeability for the C400, 2000:5 current transformer was found to 5Computer Simulation be 6300, calculated using the 8000 ohm maximum reactance of the impedance plot in Figure 6 on page 12 and using typical core dimensions. aT = 7 5 i Figure 9 8-H Loop and the Slope of the Mean Curve ‘The variable core permeability is the slope of the B-H curve of the core material at any specific point. The current transformer model needs a defined relationship between B and H so that the permeability can be calculated. The empirical Frohlich equation is used to emulate the S-shaped anhysteretic B-H curve: H c+b- lH) ° where: B isthe Magnetic Flux Density, Tesla His the Magnetic Field Intensity, (amp-turns)/m 16 Computer Simulation ‘The core material using Equations 10 and 11 determines the constants 6 and ¢: (10) ay Bsar where: hi, is the initial permeability Hy _ is the permeability of free space Bsyy is the saturation flux density ‘The graphical user interface of a current transformer simulation is shown in Figure 10. In this case, the C400, 2000:5 current transformer with standard burden (2 + j3.464) ohms is subjected toa current of 20 times its primary rating. The simulation provides what amounts to an oscillograph of the ratio, burden, and magnetizing current. Note that in accordance with the C400 rating, the ‘magnetizing current causes not more than a 10 percent ratio correction factor. 7Computer Simulation ca = = a = = cat —_ en) x Figure 10 C400, 2000:5 Current Transformer Computer Simulation qj a 4d qa 144 rz r= i cI roe ro om re) | nae pe | Figure 11 C400, 2000:5 Current Transformer With 40 kA Asymmetrical Current ‘The simulation in Figure 11 shows the saturation caused by the added volt-time area during an asymmetrical fault of 40 kA. The ‘computer simulation will be used to supply the samples needed to test the subsequent current transformer selection criteria, 18 Flux Density Limit of the Volt-Time Area ‘The burden voltage v is related to core turn N and the rate of change of the core flux @ by the induction Equation 12: di ven (a2) ‘We can integrate Equation 12 to show that the area under the voltage Waveform represents the flux density in the core. Therefore, the flux linkages in the core are given by integral Equation 13 where the flux is expressed as flux density B times the core cross-sectional area A ON = BAW = fivedt (a3) ‘We can now recognize the significance of the ANSI voltage rating because the area under the sine wave of that magnitude represents the saturated flux density B,. That volt-time area signifies the threshold of saturation and marks the boundary of saturation-free operation. Figure 12 shows the shaded volttime area produced by asymmetrical fault current. Here, J, is the magnitude of the fault current in the secondary, Zy is the burden impedance, and L/R is the time constant of the primary fault circuit. The sine wave and 19Flux Density Limit of the Volt-Time Area exponential components of the wave are shown dashed for ‘comparison. The plot emphasizes the fact that although we think of the C-rating as a sine wave, we must consider the increased volt- time area of the asymmetrical fault when selecting a CT. Figure 12 Burden Voltage for Asymmeti Using the asymmetrical voltage in Equation 2 we can write: z boNeAcw <1 jo 8 (Bat) — Keotoneal (4) In Equation 14, the limit of the integral of the exponential term is the X/R ratio of the primary circuit. Since the limit integral of the cosine term is unity, we can write Equation 15: By N-A-@ (R+1) -te25 (as) 20 Flux Density Limit of the Volt-Time Area Equation 15 expresses the C-rating voltage in terms of the physical parameters of the current transformer, namely the saturated flux density B,, the turns ratio N,, the core cross-sectional area A , and the system frequency . Moreover, it determines the saturation-free operation range of the current transformer in terms of the system X/R ratio, the maximum fault current Jp, and burden Z,. 2The Criterion to Avoid Saturation We can derive a more versatile form of Equation 15 by recognizing that the rating voltage is 20 times the voltage across the standard burden at rated current. If we then express the fault current /,. in per unit of the rated current, and the burden Z, in per unit of the standard burden, Equation 15 becomes the simple criterion to avoid saturation: 202 (+1) Ip 2y (16) where: J, is the maximum fault current in per unit of CT rating Z, is the CT burden in per unit of standard burden X/R isthe X/R ratio of the primary fault circuit Here is an example of how the criterion is used: A transmission line has an 85,24° impedance angle (ie., the X/ ratio is 12). The ‘maximum fault current is 4 times the rated current of the C800 current transformer. Find the maximum burden in ohms that can be used and still avoid saturation. 22 — ‘The Criterion to Avoid Saturation In this case, we can solve Equation 16 for Z,, which is the burden in per unit of the standard burden 20 20 24s = +i Gey @eP = 0.385 an Zeurven = Zspp°% = 8-0.385 = 3.08 ohms Equation 16 is satisfied when Z,, is equal to or less than 0.385 per tunit of the standard 8-ohm burden. Therefore, saturation is avoided by keeping the burden at 3.08 ohms or less. Rating Current Transformers for Line Protection In practice, modern line relays clear faults in cycles to preserve stability, accurately identify fault type for single-pole reclosing applications, and determine an accurate fault location. To do this, line relays require undistorted CT secondary current to perform phasor measurement in the presence of the de offset. How well are CTs rated for line protection? The criterion stated in Equation 16 can be used to check any given application. For example, the line relaying for a 4.5 mile 138 KV transmission line uses a VT ratio of 1200:1 and a CT ratio of 300:1. The CT is C800, 2000:5 multiratio on the 1500:5 tap. The maximum fault is 4625 MVA or 19349 amperes and the line and source impedance angle is 74 degrees. The parameters for Equation 16. = 193494 _ x = pe 71299 = tan(7d) = 3.48 (ig) 23‘The Criterion to Avoid Saturation ‘Substitute these values in Equation 17 to determine the maximum burden in per unit of the standard burden: 20 —20___ 9343 19 Bae + iy a5 = 45 as, Equation 19 indicates that the burden should be less than or equal to 0.345 per unit of the standard burden. Although the standard burden is 8 ohms for the C800 rating, based on constant volts per turn we ‘must use three-quarters of the standard burden value because the 1500:5 tap is used. Consequently, saturation is avoided if the total burden is equal to or less than: Zy = 0345-6 = 2.0730. (20) ‘The burden budget for the installation is: CT leads 200 feet full circuit run of 0.2520 #11 AWG (1.2610/1000") CT winding 300 turns at 0.00250/tam 0.7500 Allowable relay burden 1.071 ‘Total burden 2.0739 The relays, consisting of a microprocessor-based distance relay with negligible burden and an overcurrent relay used as a current detector, have a combined burden easily less than the allotted 1.071 ohms. Consequently, CT saturation is avoided for all asymmetrical as well as symmetrical faults on the line, 24 The Criterion to Avoid Saturation Current Transformer Selection Procedure ‘The above example suggests the following current transformer selection procedure using Equation 16 in any given line relay application: 1, Determine the maximum fault current I;in primary amperes. 2. Determine the corresponding primary circuit X/R ratio. 3. Select the current transformer voltage rating. Then, determine the total burden in per unit of the standard burden. 4. Using Equation 16, calculate /,, which is the fault current in per unit of current transformer nominal rating. 5. Divide the maximum primary fault current /, by the per unit current to determine the CT nominal current rating. Select the nearest standard rating greater than the calculated value. Over what range of applications can the procedure be successful? Whether or not Equation 16 can be satisfied depends on the X/R ratio and the magnitude of the maximum fault current. You can specify the burden and the X/R ratio and then use Equation 17 to calculate the maximum fault current for a given CT ratio, Table 1 lists the maximum fault currents versus X/R ratio for which is avoided using 3000:5, 2000:5, or 1500:5 current 25The Criterion to Avoid Saturation TABLE1 Maximum Current to Avoid Saturation (€800, 3000:5 €800, 2000:5 C400, 150035, WR 22512207220 Line Angle Ratio (amperes) (amperes) (amperes) 1 37 40,547 33,812 12,680 7 43 3601230010128, so 578803983, 8994 71 3689.71 7393 sy 81 97) T79 6361 se 95 sz 15217 53107 sya saG 272, 4g27 sh 4328 10.457 3921 sr sa 9,561 7968 2988 se 286 6478 5:99 2025 Calculating Current Transformer Burden In the process of CT selection we are interested in minimizing the total burden that consists of the internal resistance of the CT ‘winding itself, the resistance of the leads from the CT to the relay, including the return path, and the burden of the connected relays. Higher ratio current transformers (3000:5) contribute a resistance of 0.0025 ohms-per-turn to the burden; and lower ratios (300:5), 0,005 ohms-per-turn. Consequently, applying a 600-turn (3000:5) current transformer contributes an internal resistance of 1.5 ohms. In new installations, we can choose the wire size to control the 26 The Criterion to Avoid Saturation resistance for a circuit run, For example, the log of resistance per 1000 feet of wire is proportional to the AWG gauge of the wire. A benchmark for copper wire is 0.9989 ohms/1000 fect for #10-gauge AWG wire. Decreasing the gauge by three numbers halves the resistance and increasing it by three doubles the resistance. Consequently, the handy formula for resistance versus AWG wire eauge is: 2/(1900") = €67926-232 ap where: 9/1000" is the resistance in ohms per 1000 feet G is the AWG gauge number A good practice is to size the leads to limit the lead resistance to 0.5 ohms or less. Under this rule, #10-gauge AWG wire would be the choice if the current transformers were located 250 feet from the relays, for a total run of 500 feet to and from the current transformers. Finally, we can contrast the almost negligible burden of a multifunction microprocessor-based relay with the burden in the order of ohms contributed by electromechanical relays for the same function. The advantage of the low burden is seen in view of the burdens and fault current restrictions versus X/R ratio listed in Table 1 Limits of the Criterion Can saturation be avoided in all applications? A limit to the criterion is indicate Table 1, where the permissible maximum fault arThe Criterion to Avoid Saturation current for a given CT rating severely decreases with increasing X/R ratio. The limit occurs where high X/R ratio and high fault current are experienced near a generator. It then becomes. impractical to size the CT to avoid saturation during an asymmetrical fault. We must then abandon the criterion and rate the CT for the reasonable sensitivity for Tine-end faults. It then remains that we must assess the effect of saturation during the offset Statistics of Asymmetry When the current is less than 20 times the current transformer nominal rating and the burden is less than the rated standard burden, no saturation will occur for symmetrical fault current. Furthermore, ‘an insulation breakdown or a flashover is more likely to occur at a voltage peak where the reactive current is at a natural zero. Consequently, line-to-ground faults are more likely to be symmetrical faults. However, in any three-phase fault, all currents cannot be at zero simultaneously in each phase and de offset is inevitable in one or more phases, In addition, the 120-degree phase displacement causes unequal de offset to occur in each phase. 28 A Case History How does CT saturation affect the response of a distance relay? The effect is less dramatic than one might imagine and is best illustrated by an example, Consider a 31-mile 230 kV line protected by a microprocessor distance relay where the VT and CT ratios are 2000:1 and 600:1 respectively. The CT has a C800, 3000:5 rating with a total burden of 2.5 ohms. The line is one of several lines connecting a generating station to the system where the maximum three-phase fault duty is 17184 MVA and the X/R ratio of the line and the source is 25. The fault current of a three-phase fault on the line at 1.55 miles from the station is 33195 amperes per phase. The relay response to a severe fault of this type at various inception angles is presented in the following three event reports. 33 kA Fault With Ideal Current Transformers Event 1, shown in Figure 13, is the ideal case of the three-phase fault with no CT saturation. This report contains quarter-cycle samples in a sequence-of-events format. Samples 9 through 36 of the report, show pre-fault, fault, and post-fault current and voltage samples, and the states of all the fault-measuring elements, outputs, and inputs. Currents and voltages are labeled left to right at the top with the clement labels (read vertically) toward the right. Time progresses from top to bottom in quarter-cycle intervals. With a quarter-cycle 29ACase History ACase History representing 90 degrees between samples, any sample and the previous sample form the real and the imaginary parts of the phasor- PLE: US 8, OREAKER 3 ate: 07/1/96 Tine: 08:48:42,702 yielding magnitude and phase angle of the measured quantity. ‘CURRENTS (pr) ‘VOLTAGES (KY pri) RELAY ELEMENTS = OUT IN xia oes I ‘The event, triggered at the 16th sample, shows pickup of the wwe es supervisory instantaneous overcurrent element 50H. The negative- Ra WED ob coche Mok sequence element 51Q picks up momentarily as a complete cycle of © 0 8 0 oeo.e 78.9 1382 ce vou data is accumulated. However, the significant point is that the first of ME Ds anne te Sean HE the phase-to-phase mho elements (ZC) declares a Zone 1 fault and 23 5 Ve SPe te a actuates the trip contacts 1 and 2 (indicated by the B in that column) a oo 152 Le before the full cycle of data is acquired. The relay then continues to oe Tis ses oe ee che i ‘measure until the current is interrupted and in this interval ay iss wast aso ane Fee determines an accurate fault location. The relay then appends the usa 25761 24000 1518 Gs ees fault type, the front-panel targets, the fault location, and the tracking Sones ere S87 ae hoe frequency at the end of the report. Hl tae eae ao. ts oe Te -s66s 716 SR 13.8 17.3 312 16 “1565-19710 Ne 27.8 26.0 =20 So 2 ies 2716s ane 12.7 7.3 -312 he a8 “2413659 19710 36K -2718 26.020 Dil 6 “60 -25059 17940 7058 16.3 18.4273 Dae 22-7659 1002 20673. 6 U.S “10 ime 2 97) 08 AMS 9.8 4b 8 3 9 me se “test pe as oa oo 2 0 6 (3-80 08 90 a 20 3 2 00 0 00 Sa 9 2 2 3 00 30 00 2 ° « bo 08 00. 4 0 0 ¢ 0 08 8 90% Sas $3 3-3 08 0.0 00 x vets A Lacatons 1.55 Frawengs «0.0 Targets: UST. 2ONEL HAGE Mi tens 195.0 / 39 Figure 13 Event 1-33 kA ABC Fault at 1.55 Miles on a 230 KV Line (Theoretical-No CT Saturation)ACase History ‘This event is, of course, purely theoretical because 1) de offset cannot be avoided for a three-phase fault, and 2) the specified CT rating, burden, and X/R ratio indicate that de offset will cause saturation with as little as 7413 amperes of fault current. The severe current transformer saturation caused by the 33 kA fault with ‘maximum offset in A-Phase is shown in Figure 14, i he { INVA a Tio (secon) Figure 14 Secondary Current in a C800, 3000:5 With 33 kA Offset Three-Phase Fault at 1.55 Miles With 0.066 Primary Time Constant 32 A Case History The Saturation Effect We can see the effects of saturation in the waveforms of Figure 14. A large portion of the A-Phase ratio current has gone to magnetizing current. What is left of the ratio current in the burden is severely reduced and distributed toward the leading edge of the waveform. The digital filtering used in a microprocessor relay extracts the fundamental component of the waveform.' Consequently, the relay extracts a fundamental current reduced in magnitude and advanced in phase when compared to the ratio current. At the same time, because the maximum offset is in the A-Phase, saturation is delayed in B- and C-Phases. The result of sampling the A-Phase current with a 16-sample-per- cycle cosine filter is shown in Figure 15. The saturated secondary and the magnitude of the fundamental are shown solid, with the ratio current and the magnitude of its fundamental shown dashed. Immediately below is a plot of the phase angle by which the fundamental of the saturated signal leads that of the ratio current, ‘The ratio of the fundamental of the saturated secondary current to ideal secondary ratio current results in a complex factor. We may think of this factor as being applied to the samples of the ratio current to produce the saturated waveform, For example, the sample-by-sample factors for the first two cycles of the waveforms of Figure 14 are listed in Table 2, The factors show little change. However, at 1.5 cycles, A-Phase secondary current is 58.3 percent of the ratio and leads the ratio current by 23.7 degrees. ‘BO. Schweitzer II and Daquing Hou, “Filtering for Protective Relaying,” Proceedings of the Western Relay Conierence, Spokane, WA 1992 33A Case History Cunent (Amps) TABLE 2 Phase Current Samples ACase History ‘Phase Current Phase Current 3 Cycles Figure 15 Magnitude and Phase Difference of Sampled Secondary Current ees TT anal 088 Telos Tangle ° 0.1250 ors 0.2500 03125 03730 oars 0.5000 0.3625 0.6250 oss 0.7500 osi2s 08750 0937s 1.0000 1.0625 1.1250 Laws 1.2500 131s 13750 ass 1.5000 1.5698 1.6250 1.6895 17500) iaIas 1.8730 1.9315 2.0000 (0.9993 0.9588 0.9983 039976 0.9963, 908s 0.9938 0.9993 Lou 1025s Loz 1.0233 1.0206 1.0202 Loz4s Loss 006s 0.951 07017 osx 0.5240 0.5866 o.si80 0.5830 oes 03674 0.3089 03226 oa6 03499) 03510 osers ‘0.0008 0.0020 0061 ooisi 0361 (0.0893 0.2330) 0.6050) 1.2230] 147s 1.0879] 0.9637 o4sr7 0.3630 x08 07519 0.8500| “1381 133 +9300} 29531 24.3849] 30.8397 28.9589 23.7362 19.4597] 23.0543 37.5746 49.7819 sisi 53.6025 54.1780 53.8137 0 0.1250 oars 0.2500 312s 03780 oars 0.5000 0562s 0.8250 6875 07500 ogi2s 08750 oss7s 1.0000 1.0625 1.1250 118s 1.2300 isis 13150 14375 1.5000 1.5625 1.6250 1.6875 1.7500 L815 1.8750 19375 2.0000 09978 ows 0.9994 10004 0002 09998 0994 0990 998s 0.9983 0.9986 0.9996 1.0008 1.0010 1.0010 1.0008 1.0008 1.0008 1.0008 1.0008 1.0006 1.0003 09997 0.9988 978 oer 0.9980 1.0009 1.0048 1.0068 1.0060 1.0083 1.0037 ‘O01 ines 0.1686 0.1018 0.0562 00877 oon, 0.0833 0.0590) oss oma 0.2079 02012 0.1699 oe 0.1208 ousis 0.1383 0319 0.1268 0.1197 oan oss ous 02016 0.3659 05993 0.7760 gaa 07386 0.6681 07319 sao 35ACase History 33 kA Fault With Maximum Asymmetry in A-Phase ‘The effect of the saturation factors is apparent in the samples recorded in the event report shown in Figure 16, The Zone | trip is similar to that of the ideal case. The SOM instantaneous element asserts first, followed by the momentary pickup of the negative- sequence directional element 32Q and the negative-sequence overcurrent element 51Q. This happens during the acquisition of a full cycle of three-phase information. As in the previous case, the event is triggered at the 16th sample and the phase elements ZAB, ZBC, and ZCA all indicate aZone | trip in the 18th sample before a full cycle of samples has accumulated. Once the trip is issued, the relay continues measurement to identify the fault type and location. The effect of A-Phase saturation is seen with the appearance of false residual current at sample 21 and 22. ‘Consequently, the fault identification logic detects a ground fault at sample 23 and declares a BC-to-ground fault to block two of the phase elements. In this case, despite the reduced A-Phase current and the false residual, the relay calculated the fault location at 1.52 miles using BC voltage and current. 36 ACase History IMMLE: BUS, SREAER 5 - SU S6OMGOS Oates 07/15/96 Times 10:02:13.990, CURRENTS (prt) YOLIARGS (et prs) RELAY CUMENTS. UT 18 gcoca 21110770 RMR Hah overt 1h Te ae Bees ZAQPNGE 2UE8 Dee Saple © 9 6 nse wt TA .. Let © 9 6 8 eo -mns ue7 Lie “206 2 a8 use ol 7s ula 6-8 a 80 ais 123 koe 29 ae n-s3 wl m2 om 8 mw stage ms | % 10S 4512-096 swe 12.9 110.0 “30.3 coe 5105. 6909-1178 -103.4 Was Le 2-178 HRD 7021 as Loe 8 pss 7 “9 8 was san, osm 81 W232 Lo ss vase 2.7 3 we 87 PM 08 Ns 3 ion 2016 suns as pa 288 ve sio2 1505 s0200 -272u 21.0 wa Lee 38 wie Oe Lek une 12 28 209 nize S28) ain tease aisle me W2 1 haa se 6150 2682 15216 0.2 se noe no 0 Wim a7 a 26 on) 138-420, ot 75.9 9.0, Le saw 1026 5 3 a 33 @a8.0 05 7007 1 7 a 4 0 13.6 1086 3 s 304 ima er * 04 7 iam 3 a) as 787 | o vents 8S Lacetion: +1.52 Freuengy: 69.0 Targetss INST AONEL LA BG MLNS 129.9 / 96 Figure 16 Event 2-33 kA ABC Fault at 1.55 Miles on a 230 kV Line (CT Saturation Caused by Maximum Offset in A-Phase) 37ACase History 33 KA Fault With Offset in B-Phase and C-Phase An incorrect fault identification is the only consequence of the current transformer saturation condition in Event 2. The last case ‘considered is Event 3, shown in Figure 17, where the fault inception angle causes no offset in A-Phase with equal offset in B- and C-Phase As in Event 2, the event is triggered at sample 16 and the Zone 1 trip is declared in sample 17 with the assertion of both trip contacts, indicated by the B in that column. The false residual is caused by saturation in both B- and C-Phases. This effect causes the assertion of the A-Phase-to-ground distance element ZAG. After the trip, the relay uses the subsequent samples to correctly identify a three-phase fault. However, the reduced B- and C-Phase current caused the calculation of an erroneous fault location of 3.61 miles. 38 A Case History ROUGE 3 5/8 KOSDNS. Oates 07/18/86 Tins t1e06:68.197 CURRENTS (ort) ——_VOLAGES pri) aKLAY GLENS. UT. ON mune sss60 3 1397 SCHED 9110770. AMM SHER Cvene Ae vR vee eubes aigouge Sach 2888 Sample © 6 0-10. 27.9 120.7 . hee 2 8 0 0 sie a. dese +1 4 213m 32 i low 2-80 teks 2eat at x te as oo es 7365 8 corse Ca as st tas tient 751s 16 ath... Goo hee si iene? “lait 18623 7A hs thea ine w ua tern ares et sen goat nop 2190 aasoh 35 25205 S68 Altes pul or) 1s 22006 2720-1980 Sasi 2a16s “4s 19208 66 1087-2065 4268 6303-905 9196 2150 16.3 1.0 14. es “sss 98 2am 10D Saas 2 “107s sta 02 “Le 8 0-00-90 0.0 2 © 2 9 0 08 00 09 2 ° 0 0 0a 00 oD » 3b 3 00 00 00 = 6 0 3 3 09 00 20 6 vents 196 Lonetions 49.61 Fromoneys 60.0 Targets: NST. DOWEL ABE QM Mens 182.3/ 12 Figure 17_ Event 3-33 kA ABC Fault at 1.55 Miles on a 230 kV Line (CT Saturation Caused by Offset in B- and C-Phase) 39CT Application Rules Generally, protective relays are designed for sine wave operation and their performance is not specified for other waveforms. Therefore, in a protective relay application, the voltage rating and the burden of the CT should be specified to ensure undistorted secondary current for the maximum fault condition. IEEE/ANSI Standard C57.13-1993 [3] suggests that CTs for relaying be applied on the besis that the maximum symmetrical fault current not exceed 20 times the CT current rating and that the burden voltage not exceed the accuracy class voltage of the CT. Applying CTs for relaying is an art rather than a science because the ‘engineer is left to choose the specific operating point on the excitation curve, However, there is a rationale for choosing a CT to produce the knee-point excitation at the maximum symmetrical fault ‘current since the magnetizing reactance is at a maximum, Observe that the knee-point of a typical excitation curve, as shown in Figure 5 on page 10, is about 46 percent of excitation voltage, corresponding to a 10-ampere excitation current. A popular rule of thumb suggests that the C-rating be twice the excitation voltage developed by the maximum fault current. By good planning or dumb luck, the rule of thumb guarantees operation near the knee-point of the excitation curve for the maximum symmetrical fault. CT Application Rules Preventing saturation due to the exponential component of fault current requires a C-rating exceeding the symmetrical rating by a factor equal to the X/R ratio of the faulted primary system plus one. ‘The dilemma is that itis routinely impossible to achieve such ratings despite the fact that saturation affects the peifurmauce uf high-speed relays. Itis in these cases that simulations are used to assess the effects of transient saturation on relay performance. a4Transformer Differential Relay Applications Figure 18 shows a 4160-volt power plant auxiliary bus supplied by a 5000 kVA delta-wye resistance grounded transformer. The transformer has a 4.95 percent impedance on a 5 MVA base and is fed by a standby and an emergency bus. The bus is protected by phase and ground directional overcurrent relays. The transformer is protected by percentage differential relays with harmonic restraint and also a ground overcurrent relay fed by a CT in the transformer neutral A.C-200, 1200:5 rating was selected for the low-side CTs and a C-200, 600:5 rating was selected for the high-side CTs. The taps for the 87T relay were set at 2.9 amperes for the high-side winding and at 8.7 amperes for the low-side winding, leaving a ratio mismatch of 4.4 percent. The percentage differential for tripping was set at 25 percent. ‘The complication in this application is the fact that the 600:5 CTs are located 1400 feet from the switchgear, requiring at least 2800 feet of leads. Only 25 feet of leads are required for the 1200:5 CTs ‘mounted in the switchgear with the relay. The long run was installed with paralleled #10 gauge AWG wire, which has a calculated resistance of 3.36 ohms, including the totally resistive relay burden. 42 ‘Transformer Differential Relay Applications ‘The 1200:5 CT was installed with a calculated total burden of 0.365 ohms. The maximum through-fault current was calculated as 12312 amperes on the 4160 bus (3711 amperes at 13.8 kV) with an X/R ratio of 11 In the low-side 1200:5 CT, the maximum symmetrical secondary fault current of 51.3 amperes flowing in the total burden of 0.365 ohms produces 18,72 volts. This voltage multiplied by(X/R+1) = 12 gives 225 volts. Consequently, the low-side CT can support undistorted asymmetrical ratio current in the secondary. In the 600:5 CTs, a maximum symmetrical fault current of 30.9 amperes flowing in the total circuit burden of 3.4 ohms produces 105 volts, or about half the 200-volt accuracy rating. These CTs, therefore, are adequately rated by the accepted rules. However, their core will saturate with asymmetrical fault current. Since only one set of CTs will saturate, relay operating current will persist for asymmetrical faults, 43Transformer Differential Relay Applications 13.8 kV Standby Bus | 13.8 KV Emergency Bus |, 36005 —— > 258" a (con) é¢ XY ~ (sors) (sr (e132) — (exs0n) & =X I 12005 l 4160 V Bus Figure 18 Power Plant Auxi Transformer Differential Relay Applications swcordary amps yates Figure 19 Differential Relay Restraint Currents |, and Ip. ly IS the Differential. ‘The computer simulation of Figure 19 shows the implications of this archetypical application. The graph shows the time plot of saturated 600:5 CT secondary current with the unsaturated 1200:5 CT secondary current. The plot shows the large difference current, which decays with the de transient. The plot is literally the ‘magnetizing current during saturation caused by the asymmetry. It has the same wave shape and is caused by the same phenomenon as the inrush current of the power transformer itself, Figure 20 is a plot of the harmonic content of the difference current. The magnitudes of the harmonics are plotted in per unit of the fundamental. The analysis shows a typical 42 percent second harmonic, 15 percent fourth harmonic, and 10 percent fifth harmonic. As it turns out, the second harmonic is high enough during the offset that a differential relay with harmonic restraint units will restrain the relay from‘Transformer Differential Relay Applications operating. Therefore, the rule of thumb proved adequate for this transformer differential application. 100 20 2 7 Percent of Fundamental 10 ° SRESRRSSIEBRERES Frequency Figure 20 Harmonic Content of the Differential Current in Figure 19 Plotted in Per Unit of the Fundamental Frequency 46 Generator Differential Relay Application It is impractical to size CTs to avoid transient saturation in @ generator differential because of the high X/R ratios encountered. The rule is to select the largest practical rating and match the terminal and neutral-side CTs. The pitfall is that the highest CT accuracy class is C800 and tha: any CT with an excitation voltage exceeding 800 volts is classified C800 no matter how high the voltage. For example, one 6000:5 CT may have an excitation voltage of 1500 volts at 10 amperes of excitation current and be classified C800. A second 6000:5 CT from a different manufacturer may have 978 volts at 10 amperes of excitation and also be classified C800. ‘The generator CT's must have the same excitation curve with ‘matching knee-point voltage and the same excitation voltage at 10 amperes excitation current. Consider the application of a generator differential relay for a 111 MVA, 13.8 KV generator. The machine has an X/R ratio of 32 and can contribute 58800 amperes to an external bus fault. All the CTs are rated 6000:5, C800. However, the set of generator terminal CTs has a 700-volt knee-point voltage while the set of neutral-side CTs has a 500-volt knee-point voltage. We have observed that the knee-point voltage is generally 46 percent of the excitation voltage occurring at 10 amperes of excitation. The actual rating can be a7Generator Differential Relay Aoplication estimated as 700/0.46 = 1522 and 500/046 = 1086 volts, respectively, The table of input data for the maximum fault case is as follows: Side ACT Data ‘Crating (actual) 1522 Ri 1200 Core mean length (m) 102 Side A Lead and Side B Lead and fing Data Winding Data Lead Resistance 26 23 Restraint Winding, oot ool Resistance Restraint Winding oot oor Reactance Operator Winding and External Reactor Data ‘Operating Winding Resistance 0.52 Operating Winding Reactance oot Reactor Knee-Point Volts 2 Ext. Ohms Reaetance 170 Ext. Reactor Core Length 0.229 Primary Current Primary Current 38800 Primary Time Constant os Incident Angle (Max. Offset) 00 48 NN Generator Differential Relay Application Figure 21 is a plot of the restraint currents and the differential current for the maximum asymmetrical fault case. In this application, the CT voltage rating exceeds the burden voltage produced by the maximum symmetrical fault by a factor of 11.9 in the side-A CT and by a factor of 9.6 in the lesser-rated CT. These conservative ratings delayed saturation for nearly two cycles, allowing time for a subsequent instantaneous trip. Figure 22 shows the effect of adding 0.6 ohms to the side-A burden to even the time to saturate. In this case, the pulse of differential current is virtually climinated. This study shows the margin for CT mismatch. 200 ‘secondary ange L ay Figure 21 Secondary Currents |; and lz Due to a 58800 A Through Fault Current With Offset. The Differential Current ly Also Plotted. 6000:5 CTs Have Knee-Point Volts of 700 and 500. X/R = 52.Generator Differential Relay Application secondary pe cycles Figure 22 Secondary Currents land Iz Due to an Asymmetrical Through Fault of 58800 A. Adding Burden Resistance Decreases the Differential Current ly. 6000:5 CTs. Knee-Point Voltages of 700 and 500. X/R 2. 60 Bus Differential Relay Applications Differential relays provide speed and sensitivity for clearing internal faults in critical bus and generator applications. These relays are fast enough to operate during the offset portion of fault currents where high current magnitudes, together with high X/R ratios, conspire to produce CT saturation and false differential current. Such high fault current applications require full-rated CTs to prevent saturation and, for security, rely on the percentage differential slope characteristic to override incidental false differential. The total scheme must restrain from operating for the most severe external fault and yet have the sensitivity to operate on the minimum fault, In existing schemes, the application engineer must live with the ratings of existing CTs. Even in new installations, the engineer must contend with switchgear space limitations or with CT ratings pro- cured by company standards rather than by specific application. The application engineer must determine the percentage slope setting in relation to the degree of saturation that provides the required security, Here, we will explore the relation between the percentage slope characteristic and the degree of saturation produced by a range of CT ratings encountered in utility and industrial applications. We will show the plots of the trajectories of operating current versus the restraint characteristic for field applications, and establish settings in terms of fault currents and CT ratings. 51Percentage Slope Characteristics Figure 23 shows the single-line diagram of a machine differential relay where the currents 1, and J are shown with the polarity of an internal fault. As shown in Figure 24, in a percentage differential relay, the magnitude of the sum of the currents entering the zone must exceed a predetermined percentage of a restraint quantity. Figure 23 Differential Protection Operate (trip) t Restrain (No Trip) res Figure 24 Percentage Characteristic 52 Percentage Slope Characteristics ‘The restraint quantity may have one of several definitions. Two ‘common characteristics are given in Equations 22 and 23: Mattel = 22) Mattel = (Ul + ald (23) ‘The left-hand side in each equation is the operate signal equal to the magnitude of the sum of the currents. The right-hand side in each ‘equation is the restraint signal that is multiplied by the slope k. The restraint signal in Equation 22 is the magnitude of the difference of the currents, In Equation 23, the restraint signal is the sum of the magnitudes of the currents. Both characteristics produce the relay characteristic shown in Figure 17, Some relays use the average magnitude of the currents, in which case the slope & in the equations is half the slope setting of the rela. The only purpose for the percentage differential characteristic is to override false differential current caused by CT ratio error. Since the CT ratio errors are much less than ten percent in the linear range, the only purpose for the percentage slope characteristic is to override the error due to saturation caused by the de offset. What is the nature of the errors due to saturation, and how significant are they? Figures 25 and 26 show the severity of the CT saturation error. Figure 25 compares the saturated secondary current with the true ratio sine wave current for a C200 2000:5 CT for @ 10667 A fault. In a microprocessor relay, the digital filter removes the de component and all harmonics, leaving only the fundamental of the waveform as shown in Figure 26. In the saturated waveform, the latter part of the half cycle of ratio current is missing. The missing current flows in 53Percentage Slope Characteristics the magnetizing impedance and not in the burden. As a result, the fundamental extracted from the saturated waveform is reduced in magnitude and is shifted, leading in phase with respect to ie current as shown in Figure 25. These are the signals processed by the relay; the worst case is when the left-hand set of CT's saturates and the right-hand set does not, Ratio Current ! cycies Figure 25 Saturation Due to DC Offset. C200 2000:5 CT, 2.0 © Burden, Fault Current 10667 Amp. X/R = 14 Percentage Slope Characteristics ‘Secondary Amps Figure 26 Fundamental Waveforms Extracted by Digital Filter 55Alpha Plane Differential Characteristic ‘The percentage differential characteristic can be visualized by expressing Equations 22 and 23 in terms of the ratio [4] (p/l,) = a+b (24) Equation 22 written in terms of the ratio becomes: Qs) Substituting the complex number (a +b) for the ratio and expanding gives the equation: (26) By completing the square, Equation 26 can be recognized as a circle with radius and center at: Radius = 2k 7) 1k 2 Center tks; 0 (28) 1k Alpha Plane Differential Characteristic ‘Consequently, the percentage slope characteristic is a circle enclosing the stability (no trip) region in the complex plane called the alpha plane. Similarly, Equation 23 written in terms of the ratio of currents becomes: (29) ‘Substituting the complex number (a + jb) for the ratio and expanding gives the equation of a characteristic resembling a cardioid that circumscribes the circle characteristic: P+ 5(c0s(0)- 0) r+ 1 = 1-8 (30) Po isa ee? a is r+ cos(@) 87Alpha Plane Differential Characteristic Figures 27 and 28 show the plots in the alpha plane. m bad 2 if i & Po o asa as a Real Figure 27 Circle Characteristic 1; ‘0 el CCarcoid Shaped ® (Characteie : Pr 2 Pe ° 4 2 Real Figure 28 Characteristic Resembling a Cardioid Determining the Slope for a Given Degree of Saturation ‘The alpha plane characteristic provides a convenient way of determining the slope setting in relation to a given degree of CT saturation. The method is to plot the ratio of currents (J,/J,) in the alpha plane for an external fault and then choose slope k such that all the points are enclosed by the circle- or cardioid-shaped characteristic. The current samples are obtained from a MATLAB® simulation program, GUI_2CTS.m, that calculates current samples at 57.87 microsecond intervals for 6 cycles. The program interface shown in Figure 29 lists the CT data. Figure 29 MATLAB Program Graphic Interface 59Determining the Slope for a Given Degree of Saturation ‘The 1815 samples of each of the two currents are resampled by the 16-samples-a-cycle cosine filter to obtain the fundamental phasor values. The ratios of the samples are then calculated. The samples are plotted and appear as a cluster of points in the left complex plane for an external fault. A slope k is then selected so that the plot of the characteristic encloses all of the samples. ‘The degree of saturation depends on the saturation voltage Vs across the burden impedance given in Equation 31 Vs = (1+4) 4%, GBD where is the maximum fault current in a unit of the CT rating Z, is the burden in a unit of the standard burden X/R isthe X/R ratio of primary fault current Our procedure will be to start with the left and right set of CTs rated. ‘at C800 and apply the maximum fault current that just produces the threshold of saturation for a predetermined X/R ratio. The value of Vg equals 20 in this case. This value of fault current will be retained in all subsequent cases. The saturation voltage V, will then be doubled from the previous case by lowering the ANSI voltage class of one set of CTs by one standard rating. In this way, the saturation voltage will take on the values 20, 40, 80, and 160. Each simulation will produce 101 phasor samples that plot as a cluster of points in the alpha plane. In each case, the value of k will be selected such ‘that the circle characteristics will enclose all the points. The values of k will then be plotted as a function of the saturation voltage V.. Each simulation was run for a 2000:5 CT with a lead and relay Determining the Slope for a Given Degree of Saturation burden of 1.0 ohm total and a winding resistance of 1.0 ohm. The fault current in each case is an offset 10667 amperes, where the X/R ratio is 14. The cases are plotted in the following Figures 30 through 33, The darker points in each plot are the ratios (IR/IL). The lighter points are the reciprocal ratios (IL/IR). Note that the k in each plot is in percent. Plots using the cardioid-shaped characteristic are shown in Figures 34 and 35. 61Determining the Slope for a Given Degree of Saturation ati RAL and WR 2a 0 eathiny Figure 30 0 Fault Current 10667 Fate Lond aR est) Figure 31. Vs = 40 Fault Current 10667 Determining the Slope for a Given Degree of Saturation Reto Land UR ey Figure 32 Vs = 80 Fault Current 10667A Fein} a 23 3. oat) Figure 33 Vs = 160 Fault Current 10667A 63Determining the Slope for a Given Degree of Saturation ate a dcr 2 lan. Figure 34 Vg = 80 Fault Current 10667A With Cardiol Shaped Characters ae and UR 70 71 oO ea Figure 35 Vs = 160 Fault Current 10667A With Cardioid- Shaped Characteristic Determining the Slope for a Given Degree of Saturation Saturation Voltage ‘The relation between the slope seting and the degree of saturation is given by plotting the slope & as a function of the saturation voltage. ‘The curve with its empirical equation, shown in Figure 36, determines the slope setting for a given degree of saturation. This ‘equation is valid for V. < 160. Slope k Figure 36 Slope x Versus CT Saturat 80 k = 0,824(V4)- 0.00242( V5) 60 a a x ve= (14412, n Voltage 65Bus and Machine Differential Applications In machine differential applications, there is no reason for the terminal-side CTs and the neutral-side CTs to have a different ANSI voltage rating. In fact, itis impractical to size CTs to avoid transient saturation in a generator differential because of the high current magnitude and high X/2 ratios encountered. Consequently, the rule is to select the largest practical rating and match the terminal- and neutral-side CTs, so that in the presence of saturation both sets of CTs produce the identical waveform and the ratio (IR/IL) remains at =I during an external fault Unlike the machine differential application, identical CTs in a multirestraint bus differential scheme do not guarantee the same waveform in all CT’. For example, the bus differential scheme in Figure 37 shows the maximum extemal fault on a feeder. The CT on the faulted feeder carries the total fault current of 10667 A. For an X/R ratio of 14, Vs is 40 and the fault current produces saturation. However, the CTs of the transformer and tie carry only half the total current, Consequently, these CTs act as one equivalent C800 CT with a saturation voltage, V,, of 20 that produces undistorted secondary current. 66 Bus and Machine Differential Applications 5333 5233 b | 4 Sem fe 1 10867 A ‘ALCTs C400, 200055 Figure 37. Multirestraint Bus Differential ‘These are the exact conditions of the case shown in Figure 9 on page 16, where the & of 30 is obtained by using V, equal to 40 in the equation developed from the plot in Figure 36. The alpha plane plot and a plot of the operate versus the restraint current for the case are shown in Figures 38 and 39. erBus and Machine Differential Applications Figure 38 V,= 40, 4= 30 — sma fnecsnebe IST = 2x( H+ RZ operate Curent Restaint Curent Figure 39 Operate Signal Versus Restraint 68 The Effect of Remanent Flux on the Slope Setting A high offset fault current is usually interrupted in a few cycles. Consequently, the duration of the fault current can be much shorter than the time constant of the primary circuit. The interruption leaves remanent flux in the core that is not affected by normal load current and that can be removed only by demagnetization. A survey of 141 CTs on a230 kV system is reported in Annex C of Reference {2}. In this survey, the remanent flux ranged from 20 to 80 percent of the flux at the threshold of saturation, ‘The ANSI voltage rating of a CT is the voltage caused by 20 times the rated 5-ampere secondary current flowing in the standard burden. Furthermore, the volt-time area of the rated voltage represents the flux at the threshold of saturation. Consequently, the Temanent flux in percent of the flux at the threshold of saturation reduces the voltage rating of the CT by that percentage. For example, C400 CT with 40 percent remanent flux has been effectively reduced to a C240 rating, The effect of remanent flux can be included in the calenlation of the saturation voltage Vs by modifying the standard burden as follows: Y, emanence Zit = “aes ~ RB Remanence (32)‘The Effect of Remanent Flux on the Slope Setting The calculation of k for the case shown in Figures 31 and 32 for a C400 CT with 40 percent remanent flux is as follows: Maximum fault current 10667 X/R ratio of primary circuit X/R = 14 CT primary current rating CT rating = 2000 CT burden in ohms Zburden = 2.0 ohms Percent remanent flux % Rem = 40 Standard burden y, = YANSI() 400 ( _ 40, 2nd = “ig (1- 55) = feo) = 4 Saturation voltage X) lt thurden R/ CTrating = Zstd - 10667 20. = (1-414) 10662. 20 = 6 (0.824Vs — 0.00242Vs* = 44 Slope setting ‘The calculation shows that the 40 percent remanent flux increases the saturation voltage from 40 to 67 and the secure slope from 30 to 4, The alpha plane plot and the plot of the operate versus the restraint current for this case are shown in Figures 40 and 41. 70 The Effect of Remanent Flux on the Slope Setting External Faut Interna Fault Figure 40 Vs = IST = 2k + RO ‘operate Curent Restrnt Curent Figure 41 Operate Signal Versus Restraint With 40 Percent Remanence nThe Impact of High Fault Current on Overcurrent Protection Current transformers for transmission lines are rated to avoid saturation during the asymmetrical portion of the maximum fault current. Saturation is avoided by selecting an ANSI voltage rating larger than the maximum fault burden voltage with the (1 +X/R) factor applied. This is possible in line protection applications since large load currents result in the use of high ratio CTs, and fault ‘currents are typically limited from three to five, and rarely exceed ten, times the CT primary current rating. It becomes more difficult to avoid saturation with transformer differential protection. In these applications, the high-side CTs tend to be mounted on the transformer bushings and require long lead runs to the relay. The lower current ratio required on the high side of the transformer, together with longer leads, conspires to cause saturation during offset, while the well-rated low-side CTs have a higher ratio and remain linear. However, the relay detects the second harmonic present in the false differential current and blocks tripping for external faults. Consequently, you can apply an ANSI rating as low as twice the maximum fault burden, rather than applying the (1+X/R) factor. 2 The Impact of High Fault Current on Overcurrent Protection ‘Whereas IEEE guides address CT selection for line and differential relays, there is an unfamiliar and neglected area where no guidelines ‘exist. We now discuss CT ratings used in power plant auxiliaries and related applications, where fault currents can exceed 200 times the CT primary current rating. A CT selection criterion is developed for these applications based on a two-cycle trip time. We review the limitations of conventional digital filtering used in modern instantaneous overcurrent elements, We then introduce proper digital measurement techniques in order to maintain speed and reliability when instantaneous overcurrent elements are applied with highly saturated current waveforms. 73Limitations of an 80 A Instantaneous Setting Relays protecting power plant auxiliaries may experience fault currents as high as 40 KA, where the X/R ratio exceeds 20. In addition, low ratio CTs may be used with relays that allow the instantaneous overcurrent element to be set as high as 80 A. What is, the response of a relay with an 80 A instantaneous setting, and what are the current waveforms? Are there operating limits? Are there published guidelines? Such an auxiliary bus is shown in Figure 42. ‘The following example addresses these issues. In Figure 42, the 600-hp water pump motor on the plant auxiliary bus uses C50, 100:5 CTs. The motor relay has a motor full-load current setting of 6.7 A corresponding to 135 A primary current. The locked rotor setting is 40 A, which is six times the full load current. The instantaneous element is set to 80 A, which is twice the locked rotor value. 74 Limitations of an 80 A Instantaneous Setting GEN) ‘Aux Bus To | 7 AOKALL M) 600 hp Ke Figure 42 Generator Auxiliary Bus With High Fault Current Figure 43 is the schematic diagram of the microprocessor-based relay showing the location of the signals measured. As illustrated by Figure 43, the relay auxiliary transformer converts the CT secondary ‘current to a scaled voltage signal. The anti-aliasing filter removes any high frequency present in the waveform, and the A/D converter converts the signal to the digital value of the current at a typical sampling rate of 16 samples-2-cycle. The function of the digital filter is to reject all harmonics and to extract the magnitude of the fundamental content of the signal. Figure 44 shows the extracted fundamental magnitude resulting from a 200 A primary current. The sine wave current and the extracted fundamental magnitude are the 75Limitations of an 80 A Instantaneous Setting expected waveforms. However, what waveform can be expected for the 40 KA fault? Figure 45 shows the severely saturated CT secondary current and the limited magnitude A/D output. What is more, the fundamental extracted from the A/D output falls short of the 80 A trip threshold. Sampling Frequency cea aD Baa ane | ccomersion />|_Fiter OF Teay A Primary ST cr Figure 43 Relay Schematic Showing Measured Signals Figure 44 Signal Extracted With a 200 A Primary CT Current 76 Defining a CT Selection Criterion The problem associated with high fault current and the 80 A instantaneous setting is shown by the failure to clear the 40 kA fault (Figure 45). The C50, 100:5 CT is clearly inadequate, but what criteria should be used? As a criterion, we will select the CT rating to guarantee a relay trip time in no more than two cycles. Using the CT simulation, we can increase standard CT ratings until we obtain the desired result, Figure 46 shows that a C200, 200:5 CT produces a trip time longer than two cycles. Figure 47 shows that a C400, 400:5 CT produces a trip time of less than two cycles and is the minimum rating that meets the criterion. The degree of saturation is defined by saturation voltage, Vs. For line protection, V5 was limited to 20 and used as the criterion to avoid saturation. In bus protection, values of V; exceeding 20 were Used to define and correlate various degrees of CT saturation to percentage slope settings for low-impedance bus differential applications. Here we will base the CT selection criterion on the degree of saturation voltage produced by the 40 kA fault in the €400, 400:5 CT. 7Defining a CT Selection Criterion ‘The equation for the saturation voltage, Ve, is: vse (E41) 1-2 G3) where is the maximum fault current in per unit of CT rating 2, isthe CT burden in per unit of standard burden XR isthe X/R ratio ofthe primary fault circuit 7 2 a = . 3 cyCLES Figure 45 CT and Relay Signals for a 40 kA Fault Using C50, 100:5 CTs 8 Defining a CT Selection Criterion g ores ay ry a 8 __ SECONDARY CURRENT 88 a ‘SECONDARY CURRENT crcues Figure 47 Response to a 40 kA Fault Using C400, 400:5 CTs 79Defining a CT Selection Criterion Using the 400 A CT primary rating and 40 kA fault current: 1, = [max = 40000 - 199 ery 1° CTraring 400 Using the C400 standard burden of 4 © and the actual CT burden of 050: Zourden . 95 ~ 0,125 G5) Zsrp % With an X/R ratio of 20: Vg2(20+1)- 100- 0.125 = 262.5 36) Consequently, for power plant auxiliary applications, CTs used with the microprocessor relay should meet the following criterion: 262.52 (%+ NI Zy Bn where: is the maximum fault current in per unit of CT rating Z, isthe CT burden in per unit of standard burden X/R isthe X/R ratio of the primary fault circuit The following examples show how the criterion is used 80 Defining a CT Selection Criterion Maximum Fault Current With an 80 A. Instantaneous Setting Maximum fault current in terms of primary CT and ANSI voltage rating, burden in ohms, and X/R ratio: ANSI 100-Z_ CTRarive 8) Example: A microprocessor overcurrent relay has an 80 A instantaneous setting. The relay will be used with C400, 400:5 CT with a 0.5 Q total burden. The X/R ratio is 20, How high can the ‘maximum fault be for secure operation? ‘The burden is primarily due to the CT windings and external leads to the relay (the microprocessor relay burden itself is negligible): 300 feet full-circuit run of #10-gauge AWG wire (1.0.9 /1000-f) 03 CT winding of 80 tums at 0.0025 O/tumn 02 Total burden 05a 262.5. ANSI Iaax= T0rzs CTrarine 09) (+9) = 2625, 400 499 = ‘wax 20)" Too. 03 40° = 40000 0) 81Defining a CT Selection Criterion Minimum CT Rating With an 80 A Instantaneous Setting CT rating in terms of maximum fault current, X/R ratio, ANSI rating, and burden: (ay CTrarina Example: With an 80 A instantaneous setting, what is the minimum, CT rating that can be used when the maximum fault current is 40 KA, X/R = 20, and the burden is 0.5 Q: (+) 14 100 © CTparine* “apps aNST MAX 28 (2) 1420) 100 7 43) CTexrive = TEZO) pg 40000 05 = 400 (43) ‘The criterion allows protection engineers to determine the limits of high-current applications in terms of the fault current magnitude, the X/R ratio, and the relay burden. The criterion addresses the limitation of CT ratings. We must now address the tripping speed. 82 Implementing Instantaneous Overcurrent Elements Digital Implementation of Instantaneous Overcurrent Relays Microprocessor relays use a pair of orthogonal finite impulse response filters for phasor acquisition of current and voltage Quantities. Phasors are usually obtained using Fourier or cosine filters, which eliminate de and harmonic components while contributing minimal transient overreach. In addition, root-mean-square (rms), peak, or averaging filters can also be realized. The rms filters respond to the total energy content and the de component, and every frequency present in the waveform. Digital peak detectors also respond to the harmonics and the de component. Although they are not generally applied, we will discuss the distinct advantage of peak detectors. Figure 48 shows a variation of the peak detector that has a reduced transient overreach. The filter computes the waveform peak value as the average of the absolute value of two consecutive positive and negative peaks. The maximum transient overreach of 112 percent occurs with a I-cycle time constant and decreases as the time constant is increased, as shown in Figure 49. 83Implementing Instantaneous Overcurrent Elements Primary SompingFrouoncy BF oay aux a er ‘antag iar eg contarsion Sami botctor Figure 48 Diagram of Bipolar Peak Detector Figure 49 Bipolar Peak Detector Transient Overreach Digital Filter Performance With CT Saturation Digital filters cannot make an accurate measurement of fault current once saturation occurs. We have seen, in Figure 45 on page 78, that the magnitude of the fundamental in a severely saturated current waveform is a poor representation of the actual fault current, However, the fast rising responses of the rms and the peak filter are Implementing Instantaneous Overcurrent Elements ‘more representative of the actual magnitude. The rms filter has a fast rising signal but exhibits a prohibitive 150 percent transient Overreach because it must respond to de offset as shown in Figure 50. The responses of the peak, rms, and cosine filters are compared in Figure 51. Of the three filters, the comparison shows that the bipolar peak detector makes the best magnitude acquisition. Figure 50 RMS Filter Transient Overreach Compared to Bipolar Peak Detector Bho L Figure 51 Filter Response, Fault 40 kA, X/R = 20, C100, 200:5 CT, 0.5.9 Burden 85The Cosine-Peak Adaptive Filter ‘The cosine filter has excellent performance with respect to de offset and removal of harmonics. The bipolar peak detector has the best ‘magnitude acquisition in situations of extreme CT saturation, ‘Combining the two filters provides an efficient solution for the ideal instantaneous element. This instantancous element, shown in Figure 52, is called a cosine-peak adaptive filter since it incorporates oth filters. The cosine filter supplies the magnitude for normal sine ‘wave operation, The bipolar peak detector provides magnitude for saturated waveforms. A detector measures the degree of saturation by evaluating the level of distortion and switches the input to the bipolar peak detector when the distortion reaches a predetermined value. 86 ‘The Cosine-Peak Adaptive Filter meses Figure 52__ Instantaneous Element Usi Figure 52. tn Ising the Cosine-Peak A Simple Distortion Index A simple form for the distortion index measures the ratio of the sum of P harmonic magnitudes over the fundamental magnitude: a4) As an example, if we use the fundamental, second, and third harmonics, the formula becomes: = Ault al + Di= aa lay (45) 87‘The Cosine-Peak Adaptive Filter In Equation 45, Aj, Ao, and Ay are the fundamental, second, and third harmonic phasors. For waveforms with no distortion (or no harmonics), the distortion inéex will be equal to one. For highly saturated current waveforms, a distortion index of this nature will reach levels greater than two. The basic concept is to then switeh the magnitude measurement from the conventional digital filtering system to the peak detector when the distortion index becomes greater than a fixed threshold. Typically, this threshold will be set at 1.75, ‘The distortion index is computed sample-by-sample and compared to the threshold. The comparator, in turn, controls a switch. When the distortion exceeds the threshold value, the waveform magnitude ‘measurement is taken from the bipolar peak detector output. When the distortion is less than the threshold value, the waveform ‘magnitude is taken from the cosine filter output, The cases shown in Figures 53 through 55 demonstrate the action of the switch. Figure 53 shows a case of low saturation where the distortion index falls below the threshold, and the cosine filter produces the instantaneous trip. Figure 54 shows a case of high saturation where the instantaneous trip is produced by the bipolar peak filter. Figure 55 shows a case with no saturation where the distortion index settles below the distortion threshold, and the cosine filter provides the measurement. The distortion index registers a high initial value in response to an abrupt change but settles quickly to the correct value. The 3/4-cycle security timer overrides the settling time to ensure an accurate measurement. 88 ‘The Cosine-Peak Adaptive Filter ‘SECONDARY CURRENT 88 SECONDARY CURRENT 200 aah £ CYCLES Figure 53 Cosine Filter Providing Trip for a Waveform With Low Saturation C400, 200:5, 4500 A Fault With X/R = 89The Cosine-Peak Adaptive Filter The Cosine-Peak Adaptive Filter = » vs an 5 i boo En 3 | = Ewa 3 S 8 | 5 8 x | Bro SECONDARY CURRENT 1 © 5 3 cycles Figure 54 Bipolar Peak Filter Providing Trip for a Waveform Figure 55 Cosine Filter Providing Trip for a Waveform With No With High Saturation C50, 20:5, 20 kA Fault With X/R = 11.31 Saturation C100, 200:5, 4kA Symmetrical Fault Current 90 aAnnex A: Volt-Time Area Computer Simulation ‘The steps of the Mathead® worksheet on the following page use core volt-time area to determine the waveform of CT secondary current, herein the user specifies the primary fault current and X/R ratio, the CT primary current rating, and the ANSI voltage rating and burden (plotted in Figure 56). Tratiog 200 100 Amos +100 200 ° @ ™ 00 —O4 oj Seconds he Ee Cone Seema Cane Figure 56 Mathcad Plot of Current and Flux 93‘Annex A: Volt-Time Area Computer Simulation ‘Fault Current (Primary Amps) (CT Burden (Ohms) IR Ratio of Faut Creut (CT Primary Rating (with Samp Secondary) CT ANSI Voliage Rtg Por Unit Curent M=20 (CT Burden in Per Unit of Standard Burden er Unit to Saturato Number of cycles Ciosing Angle Por Unt Residual Fux T= 000001 o= 22 TO . ew toa Ege eT te Ups Maze 6 as(cr)—eo(}z0 + C4) A= [Agee ape SE ot ra IS ApH HAL, He AO AL <-@ vy, ate US. WJ] 2 Q0.4) eq Zia tao = Ze Inal Constants Constants CCaleutaton indax. Core Voltage VoleTime Area. @ ‘Amps 94 References (1) _S.D. Moreton, “A Simple Method for the Current-Transformer Characteristics." PP. 581-585, Sept. 1943. Determination of Bushing- AIEE Transactions, vol. 62, {2} TEBE Std C37.110-1996, “IEEE Guide for the Application of Current Transformers Used for Protective Relaying Purpases.” (3) IBEE Std C57.13-1993, “IBEE Standard Requirements for Instrument Transformers.” [4] AR. van CWarrington, “Protective Relays: Their Theory and Practise-Volume 1," 2d ed. London: Wiley/Chaprcan & Hall, 1968, p. 99-140, 95