electrical testing measurement handbook vol 7

Electrical Testing &Measurement

HandbookVolume 7


HandbookVolume 7

Electrical T

esting & M

easurement H

andbook Volu

me 7

The E

lectricity Forum

testing and measurement cover 5/15/07 2:44 PM

testing and measurement cover 5/15/07 2:46 PM

Fax: (905) 272-1425E-Mail: [email protected]: www.gtwood.com/flash/splash.html

Specializing in High-Voltage Electrical Testing, inspec-tions, maintenance and repairs. Refurbishing and repair of Newand Reconditioned Transformers, Structures, Switchgear andAssociated Equipment. Infrared Thermography, EngineeringStudies and PCB Management.

High Voltage, Inc.31 Rt. 7A, P.O. Box 408Copake, NY 12516USATel : (518) 329-3275Fax : (518) 329-3271Contact : Bob Tighe, E-Mail : [email protected]

Manufacturers of High Voltage Test Equipment. Productsinclude portable AC-VLF, .1Hz, .05 and 0.2Hz Very LowFrequency hipots with sine wave output, switchgear and bottletesters up to 100 kVac. Portable DC hipots up to 300 kV DC.Aerial lift and bucket truck AC test sets up to 300 kVac accord-ing to ANSI standards. Controlled energy cable fault locators,oil test sets and burners also offered.

LIZCO SALESR.R. #3Tillsonburg, ON N4G 4G8Toll Free: 1-877-842-9021Fax: (519) 842-3775Contact: Robin CarrollWebsite: www.lizcosales.com

We have the energy with Canada’s largest on-site directory:• New and Rebuilt Power/Padmount/Dry Transformers• New Oil-Filled “TLO” Unit Substation Transformers• New HV S&C fuses/loadbreaks/towers• High and low voltage:

- Air Circuit Breakers – Molded Case Breakers- QMQB/fusible switches – Combination Starters

• Emergency Service and Replacement Systems• Design/Build custom Application Systems

Megger4271 Bronze Way Dallas, TX 75237-1088 USATel: 1-800-723-2861 Ext. 7360 (Toll Free)

Tel: 214-331-7360 (Direct) Fax: 214-331-7379 Email: [email protected] www.megger.com

Megger is a leading provider of electrical test and measuringequipment for power, industrial, building wiring and communi-cation applications. Its wide range of products extends fromequipment to test protective relays and other substation electricalapparatus, to insulation resistance and ground testers. With threemanufacturing facilities and sales offices located around theworld, Megger is strategically positioned to provide customerswith innovative products, hands-on technical assistance andsuperior service. For additional information, visit our web sitewww.megger.com.

OPTIMUM ENERGY PRODUCTS LTD.#333, 11979 - 40 St SECalgary, AB T2Z 4M3Toll Free (877) 766-5412Main (403) 256-3636Fax (403) 256-3431E-mail: [email protected] Energy Products Ltd are specialists in Power Qualityand Power Metering products. We represent Fluke, AEMCInstruments, Electro Industries, and many other manufacturers.We sell portable PQ instruments for engineers and troubleshoot-ers in many industries. From Plug based voltage disturbancemeters to three phase Class A Power Quality instruments. Wealso supply permanent power and power quality meters for usein residential, commercial and industrial applications.For complete product range and information, please visit ourspecialty websites:www.PQMeterStore.comwww.PowerMeterStore.comwww.ElectricityMetering.comwww.MyMeterStore.com

Raytech USA90 C Randall AvenueWoodlyn, PA 19094Tel: 610-833-3017Fax: 610-833-3018email: [email protected]: www.raytechusa.com

RAYTECH is an employee owned company that special-izes in the design and manufacture of precision test equipmentfor the Electrical Industry. With extensive experience in thedesign and application of test equipment, RAYTECH offersproducts that truly meet the needs of the testing industry. Ourdurable products are used by Manufacturers, Rebuild Shops,

testing and measurement cover 5/15/07 2:47 PM Page 6

Field Test Crews, Utilities, Rural Electrical CO-OP's, Universitiesand Research Engineers.

RHCtest.com610 Ford Drive Suite 248OakvilleOntario L6J 7W4CanadaTel : (905) 828-6221Fax : (905) 828 -6408Contact : John RiddellE-Mail : [email protected]

RHCtest.com Inc. is a Canadian owned and operatedDistributor of Electrical Test and Measurement Equipment. Wecarry various products lines such as Kyoritsu, Thurlby Thandar,Dataq Instruments, Topward Instruments, Nidec Shimpo, HighVoltage and Midtronics. We distribute products such as;Multimeters, Voltage Testers, Clamp Meters, Clamp Adapters,Voltage and Current Loggers, Power Loggers, Power Analyzers,Insulation Testers, Earth Resistance Testers, Test leads, DC/ACHipots, VLF Hipots, TAN Delta Cable Diagnostics, Thumpers,Cable and Fault locating products, Power Supplies, SpectrumAnalyzers, RF Generators, DDS Generators, Arbitrary WaveformGenerators, Function Generators, LCR Meters, Micro Ohm Meters,Frequency Counters, DMM’s DC Loads, Strobescopes, Hand HeldTachometers, Panel Mount Tachometers, Data Acquisition StarterKits, Stand Alone Data Loggers, Thermocouple Data AcquisitionSystems, DC Connected Data Acquisition Systems and BatteryTesters.

SKM Systems Analysis Inc.1040 Manhattan Beach Blvd.Manhattan Beach, CA 90266USAToll Free : 1-800-232-6789Fax : 1-310-698-4708E-Mail : [email protected]

SKM Power*Tools software helps you design and analyze elec-trical power systems. Interactive graphics, rigorous calculations anda powerful database efficiently organize, process and display infor-mation. Associate projects with multiple one-line diagrams andTCC drawings with customized data fields. Generate better designwith 'what if' scenarios by comparing study results in a single table.Also includes thousands of validated equipment libraries and theability to export project data into AutoCAD DXF and XREF format.Multiple one-line diagrams can be associated with each project forbetter systems organization and presentation. Powerful drawingtools quickly create a structured, interactive one- line diagram sys-tem model.

SKM Systems Analysis, Inc. is a California-based corporationfounded in 1972 with a desire to automate electrical design calcula-tions. SKM has been a leader in the electrical engineering softwareindustry for more than 30 years, providing quality software, trainingand support to thousands of satisfied customers throughout theworld. SKM Systems Analysis, Inc. is also chosen by 39 of the top40 Electrical Engineering firms in the world.

techniCAL Systems 2002 Inc.436 Jacqueline Blvd.Hamilton, Ontario L9B 2R3Canada: 1-86-MEASURE-1 (1-866-327-8731)Tel: 905-575-1941 Fax: 905-575-0386E-mail: [email protected]: www.technical-sys.com

techniCAL provides electrical contractors and utilities withTest, Measurement, Calibration, Control & RecordingInstrumentation. Representing Best-of-Breed Manufacturers;techniCAL provides such products as; Power Quality Analyzers,Micro-Ohmmeters, Megohmmeters, Insulation Testers, LeakageCurrent Meters, Ground Resistance Testers, Data Loggers, HighVoltage Ammeters, Power Transducers, Panel Meters, CT’s, PT’s,Shunts, etc…


Electrical Testingand MeasurementHandbook Vol. 7

Published by The Electricity Forum

The Electricity Forum215 -1885 Clements Road

Pickering, Ontario L1W 3V4Tel: (905) 686-1040 Fax: (905) 686 1078

E-mail: [email protected]

The Electricity Forum Inc.One Franklin Square, Suite 402

Geneva, New York 14456Tel: (315) 789-8323 Fax: (315) 789 8940

E-mail: [email protected]

Visit our website at

www.e lec t r i c i ty forum.com

The Electricity ForumA Division of the Hurst Communications Group Inc.

All rights reserved. No part of this book may be reproduced withoutthe written permission of the publisher.

ISBN-978-0-9782763-2-4The Electricity Forum

215 - 1885 Clements Road, Pickering, ON L1W 3V4

© The Electricity Forum 2007

Prin

ted

in C

anad

a

ELECTRICAL TESTINGAND

MEASUREMENT HANDBOOKVOLUME 7

Randolph W. HurstPublisher & Executive Editor

Khaled NigimEditor

Cover DesignDon Horne

LayoutAnn Dunbar

Handbook SalesLisa Kassmann

Advertising SalesCarol Gardner

Tammy Williams

TABLE OF CONTENTS

Electrical Testing and Measurement Handbook – Vol. 7 3

ELECTRICAL MEASUREMENT AND TESTING CONTACT-LESS SENSING AND THE AUTO-DETECT INFRASTRUCTUREForward - Khaled Nigim . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5

DON’T RISK IT: USE CORRECT ELECTRICAL MEASUREMENT TOOLS AND PROCEDURES TO MINIMIZE RISK AND LIABILITYLarry Eccleston . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7

ISOLATION TECHNOLOGIES FOR RELIABLE INDUSTRIAL MEASUREMENTSNational Instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11

RESISTANCE MEASUREMENTS, THREE- AND FOUR-POINT METHOD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15

CLAMP-ON GROUND RESISTANCE TESTER, MODELS 3711 & 3731 STEP-BY-STEP USAGEChauvin Arnoux, Inc. and AEMC® Instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21

MEASURING MAGNETIC FIELDS, ELECTRIC AND |MAGNETIC FIELDSAustralian Radiation Protection and Nuclear Agency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23

ELECTRIC AND MAGNETIC FIELDS, MEASUREMENTS AND POSSIBLE EFFECT ON HUMAN HEALTH, WHAT WE KNOW AND WHAT WE DON’T KNOW IN 2000California Department of Health Services and the Public Health Institute California Electric and Magnetic Fields Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .25

A NEW APPROACH TO QUICK, ACCURATE, AFFORDABLE FLOATING MEASUREMENTSTektronix IsolatedChannel Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .31

HIGH-VOLTAGE MEASUREMENTS AND ISOLATION -GENERAL ANALOG CONCEPTSNI Analog Resource Center. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35

STANDARD MEASUREMENTS: ELECTRIC FIELDS DUE TO HIGH VOLTAGE EQUIPMENTRalf Müller and Hans-Joachim Förster . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .39

IDENTIFICATION OF CLOSED LOOP SYSTEMSNI Analog Resource Center . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .43

SELECTING AND USING TRANSDUCERS FOR TRANSFORMERS FOR ELECTRICAL MEASUREMENTSWilliam D. Walden . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .45

HOW TO TROUBLESHOOT LIKE AN EXPERT, A SYSTEMATIC APPROACHWarren Rhude, Simutech Multimedia Inc. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .53

ELECTRICAL INDUSTRIAL TROUBLESHOOTINGLarry Bush . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .55

THE ART OF MEASURING, LOW RESISTANCETee Sheffer and Paul Lantz, Signametrics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .59

STANDARDS FOR SUPERCONDUCTOR AND MAGNETIC MEASUREMENTSNational Institute of Standards and Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .63

MULTI CHANNEL CURRENT TRANSDUCER SYSTEMSDANFYSIK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .67

FALL-OF-POTENTIAL GROUND TESTING, CLAMP-ON GROUND TESTING COMPARISONChauvin Arnoux, Inc.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .69

AN INTRODUCTION TO ANTENNA TEST RANGES, MEASUREMENTS AND INSTRUMENTATIONJeffrey A. Fordham Microwave Instrumentation Technologies, LLC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .71

44 Electrical Testing and Measurement Handbook – Vol. 7

DERIVING MODEL PARAMETERS FROM FIELD TEST MEASUREMENTSJ.W. Feltes, S. Orero, B. Fardanesh,E. Uzunovic, S. Zelingher, N. Abi-Samra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .79

TESTING ELECTRIC STREETLIGHT COMPONENTS WITH LABVIEW-CONTROLLED VIRTUAL INSTRUMENTATIONAhmad Sultan, Computer Solutions, Inc. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .85

ASSET MANAGEMENT, THE PATH TO MAINTENANCE EXCELLENCEMike Sondalini, Feed Forward UP-TIME Publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .87

THINK SYNCHRONIZATION FIRST TO OPTIMIZE AUTOMATED TESTni.com . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .89

USING NATIONAL INSTRUMENTS SYSTEM IDENTIFICATION, CONTROL DESIGN AND SIMULATION PRODUCTSFOR DESIGNING AND TESTING A CONTROLLER FOR AN UNIDENTIFIED SYSTEMni.com . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .93

MAGNETO-MECHANICAL MEASUREMENTS FOR HIGH CURRENT APPLICATIONSJack Ekin, NIST- Electromagnetic Division . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .101

A BASIC GUIDE TO THERMOGRAPHYLand Instruments International Infrared Temperature Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .105


Maintaining a highly functional electric system is depend-ent on the operational and maintenance level of the integratedcomponents that are geared together to serve the customer. Aneffective preventive maintenance setup is dependent on the relia-bility of the sensing devices and relaying instrumentation as wellas on the operator’s understanding of the process functionality.

Early measuring devices were designed and based onelectromechanical indicating instrumentation. Their solo oper-ability necessitated around the clock operator attention. Suchdevices were accurate but provided limited adaptability for inter-facing with today’s centralized centers.

As the semi-conducting integrated circuits devices start toinvade the market, many instruments are now inter-actable witheach other and some can be used to sense and record data fromvarious sensing elements in a sequential manner and generatetheir own diagnostic reports within a very brief time. Today’ssensors are built around plug-and-play infrastructure which isbased on the IEEE 1451.4 standard that brings plug-and-playcapabilities to the world of transducers. With plug-and-play tech-nology, the operator stores a Transducer Electronic Datasheet(TEDS) directly on a sensor. The sensor identifies itself with allneeded information once and is hooked to a data bus. TEDS-compatible measurement systems can auto-detect and automati-cally configure these “smart sensors” for measurement, reducingsetup time and eliminating transcription errors that commonlyoccur during sensor configuration. This enables the operator tofocus on overall system operation rather than on individual com-ponent operation.

Furthermore, measuring relaying units and associatedsensing elements technologies has advanced rapidly over thepast 20 years. A particular advancement is noted in the contact-less measuring sensors and measured data handling capability.This progression in the testing and measurement field provides awider scope of applications and shorter time for interruptingearly failure signals. As an example, the cases where infra-redimaging techniques are used are now part of the routine mainte-nance of distribution transformers. The infrared image indicatesthe hottest spot and temperature distribution inside a large distri-bution transformer without the need of embedding sensors.Earlier techniques for measuring temperature were based on col-lecting data from various temperature sensors entrenched insidethe transformer windings. If one or more sensors were faulty, thegathered data would be incomplete and the transformer has to betaken out of service. Replacing the sensors is a timely and cost-ly procedure. Today’s data handling and processors that eithercontrol the data flow from one or more sensors or part of thehuman machine interface supervisory system, have the capabili-ty to run self-diagnostics routines to alert the operator to anyabnormal behavior from the various sensing elements, and gen-erate a check list to help figure out any culprits.

This edition of the Electrical Testing and Measurement

Handbook introduces the fundamental applications of electricaltesting and instrumentation and guidelines on the correct proce-dures, and how to interpret and diagnose measured reports thatenable the operator to maintain a high degree of functionality ofthe system with minimum interruption.

This handbook addresses various practical aspects oftoday’s electrical engineering infrastructure through selectedarticles available for scientific sharing.

The articles are grouped into 4 sections. Section 1 address-es the basics and fundamentals of electric testing techniquesusing various measuring sensors normally incorporated in manyof today measuring instruments. Section 2 addresses safe opera-tion, procedures and handling of instruments. Section 3 intro-duces various sensing and measuring devices that can be used ina wide area of application. And finally, section 4 showcases fieldapplications of instrumentation in various parts of the electricalengineering industry.

The Electricity Forum endeavors to provide correct andtimely information for their readers in their handbook series. Wewelcome readers’ suggestions and constructive feedback, andcontributions. Please submit your technical articles that showcase your experience in testing and measurement tools and sys-tems directly to the handbook editor’s desk ([email protected]).

ELECTRICAL MEASUREMENT AND TESTING CONTACT-LESSSENSING AND THE AUTO-DETECT INFRASTRUCTURE

Forward by Khaled Nigim


1. EXECUTIVE SUMMARY Between five and ten times on any given day, arc flash

explosions sufficient to send a burn victim to a special burn cen-ter take place in the U.S. These incidents and other less seriouselectrical accidents result in injury – sometimes death – lostwork time, medical costs and insurance claims, downtime, thelist goes on. The cost to both the victim, the victim’s family andthe company involved, are high. Yet many of these accidents canbe prevented. The combination of training, good measurementtechnique, and the use of proper tools can significantly reducethe chance of an accident occurring.

IS YOUR COMPANY AT RISK? HOW WOULD YOU ANSWERTHE FOLLOWING QUESTIONS?

1. Do you have a documented electrical measurementsafety program?

2. Do you regularly inspect your electrical measurementequipment for damage that could imperil safety?

3. Do your workers involved in taking electrical measure-ments receive annual, intensive training on how to worksafely?

4. Does your organization insure that only properly ratedtest instruments are used in your facility?

If you answered yes to three of the questions above, con-gratulations – you’re doing a better job than most employers toreduce the chance of accidents associated with taking electricalmeasurements. But there’s still room to do more. This resourcekit was designed to help you develop an electrical measurementsafety program that significantly reduces your risk.

The high-energy electrical systems common in today’sworkplace bring not only increased efficiency, but increased lev-els of hazard and risk for electrical workers and their employers.

Workers taking electrical measurements on high-energysystems frequently work close to potentially lethal electrical cur-rents. This danger can significantly increase due to the presenceof transient voltage spikes. Transient spikes riding on these powerfulindustrial currents can produce the conditions that cause the extremelyhazardous phenomenon of arc flash.

To help manage the risks inherent in high-energy electri-cal systems, national and international standards bodies havedeveloped rules that categorize electrical environments accordingto their potential danger. Personal protective equipment, includingtest instruments, is categorized according to the NFPA-70EStandard for Electrical Safety Requirements for Employee Work-places, related to the incident energy levels and arc flash bound-ary distances.

To help ensure safety in today’s high-energy, high-hazardenvironments, leading manufacturers have re-engineered theirtest instruments to enhance both reliability and safety. Such toolscan help companies avoid the many perils caused by high-energy

electrical accidents: disruption of operations, higher insurancecosts, litigation and, most importantly, human suffering.

In today’s society, where medical costs are escalating andlawsuits are common, wise managers will take every step toreduce the level of risk, help increase employee safety and mini-mize the organization’s operational and financial exposure. Thismeans that management must ensure that employees use appropri-ate personal protective equipment, including new-generation testtools independently tested to help ensure that they perform up tospecification. And employees must use that equipment correctly,and receive training in safe electrical measurement procedures.

2. INTRODUCTION: MANAGING HAZARDS IN THEELECTRICAL ENVIRONMENT

Today’s industrial and business electrical supply systemsdeliver high levels of electrical energy – up to 480 volts in theUnited States, and up to 600 volts in Canada. Such high-energycircuits can create significant hazard and risk.

Another characteristic of most high-energy electrical supplysystems is the presence of short-duration voltage kickback spikes,called transients.

When such spikes occur while measurements are beingmade, they can cause a plasma arc to form – inside the measurementtool, or outside. The high fault current available in 480-volt and 600-volt systems can make the resulting arc flash extremely hazardous.

Mitigating such risks requires the use of Personal ProtectiveEquipment (PPE) including test instruments engineered and testedto meet appropriate standards, adherence to safe measurement pro-cedures, and proper inspection and maintenance of test instruments.

In this paper we will cover: • Understanding the High-Energy Environment • Voltage Transients • The Danger of Arc Flash • Measurement Categories CAT I, CAT II, CAT III and

CAT IV• Measurement Tools as Part of Personal Protective Equipment• Safety Requirements for Measurement Tools • Test Tool Inspection and Maintenance • Safe Measurement Processes and Procedures • Conclusions and Recommendations

3. UNDERSTANDING THE HIGH-ENERGY ENVIRONMENT Businesses simply could not survive without large

amounts of electrical power. Manufacturing operations and officeheating, ventilation and air conditioning systems require largeamounts of power, and computer systems have now becomemajor power users.

The need to supply large amounts of power in the mostcost-effective way has led firms to choose higher-energy, higher-voltage supply systems, which cost less to install.

DON’T RISK IT: USE CORRECT ELECTRICAL MEASUREMENTTOOLS AND PROCEDURES TO MINIMIZE RISK AND LIABILITY

Larry Eccleston, Product Testing Manager, Fluke Corporation, Member, IEC Standards Committee


As a result of these trends, industrial and business opera-tions today incorporate higher levels of electrical energy, whichcan lead to increased hazard and risk for those who build andmaintain these systems. It is common for industrial and commer-cial maintenance workers and electricians to work with high levelsof energy. In the U.S., 480-volt, three-phase electrical supplysystems are commonplace. In Canada, systems use up to 600 volts.Although classified as “low voltage”, both 480-volt and 600-voltsystems can easily deliver potentially lethal amounts of currentsufficient to fuel an arc flash – an extremely hazardous occurrence.

4. VOLTAGE TRANSIENTS: DANGER IN A MICROSECOND The presence of voltage kickback spikes, called tran-

sients, is another characteristic of electrical supply systems thatadds to the potential danger encountered when taking electricalmeasurements.

Transients are present in almost every electrical supplysystem. In industrial settings, they may be caused by the switchingof inductive loads, and by lightning strikes. Though such transientsmay last only tens of microseconds, they may carry thousands ofamps of energy from the installation. For anyone taking measure-ments on electrical equipment, the consequences can be devastating.

When such spikes occur while measurements are beingmade, they can cause a plasma arc to form – inside the measure-ment tool, or outside. The high fault current available in 480-voltand 600-volt systems can generate an extremely hazardous con-dition called arc flash.

5. UNDERSTANDING ARC FLASH How can such a problem develop? A transient of suffi-

cient magnitude can cause an arc to form between conductorswithin an instrument, or across test leads. Once an arc occurs, thetotal available fault current similar to the bolted current can feedthe arc and cause an explosion.

The result may be an arc flash, which can cause a plasmafireball fueled by the energy in the electrical system. Temperaturescan reach about 6,000 degrees Celsius, or 10,000 degreesFahrenheit.

Transients are not the only source of arc-flash hazard. Avery common misuse of handheld multimeter can trigger a sim-ilar chain of events.

If the multimeter user leaves the test leads in the ampsinput terminals and connects the meter leads across a voltagesource, that user has just created a short through the meter. Whilethe voltage terminals have a high impedance, the amps terminalshave a very low impedance. This is why a meter’s amps circuitmust be protected with fuses.

Another common and dangerous misuse of test equipmentis measuring ohms or continuity on a live circuit. These measure-ments should be made only on circuits that are not energized.

6. ARC FLASH AS A SAFETY ISSUE Detailed information on the frequency and cost of arc flash

accidents is difficult to find. Accident reports may not distinguisharc flash from electric shock. In addition, employers may bereluctant to discuss or report incidents that can be so dangerousand costly.

Dr. Mary Capelli-Schellpfeffer of the University ofChicago provides the most authoritative estimates of arc flash fre-quency. Her firm, CapSchell, Inc., a Chicago-based research andconsulting firm, estimates that between five and ten times a day,arc flash explosions sufficient to send a burn victim to a special

burn center take place in the U.S.

7. MEASUREMENT CATEGORIES: CAT I, CAT II, CAT III AND CAT IV

To provide improved protection for users, industry stan-dards organizations have taken steps to clarify the hazards pres-ent in electrical supply environments. The American NationalStandards Institute (ANSI), the Canadian Standards Association(CSA), and the International Electro-Technical Commission(IEC) have created more stringent standards for voltage testequipment used in environments of up to 1000 volts.

The pertinent standards include ANSI S82.02, CSA 22.2-1010.1 and IEC 61010. These standards cover systems of 1000volts or less, including 480-volt and 600-volt, three-phase cir-cuits. For the first time, these standards differentiate the transienthazard by location and potential for harm, as well as the voltagelevel.

ANSI, CSA and IEC define four measurement categoriesof over-voltage transient impulses. The rule of thumb is that thecloser the technician is working to the power source, the greaterthe danger and the higher the measurement category number.Lower category installations usually have greater impedance,which dampens transients and helps limit the fault current thatcan feed an arc.

• CAT (Category) IV is associated with the origin ofinstallation. This refers to power lines at the utility con-nection, but also includes any overhead and under-ground outside cable runs, since both may be affected bylightning.

• CAT III covers distribution level wiring. This includes480-volt and 600-volt circuits such as 3-phase bus andfeeder circuits, motor control centers, load centers anddistribution panels. Permanently installed loads are alsoclassed as CAT III. CAT III includes large loads that cangenerate their own transients. At this level, the trend tousing higher voltage levels in modern buildings haschanged the picture and increased the potential hazards.

• CAT II covers the receptacle circuit level and plug-inloads.

• CAT I refers to protected electronic circuits. Some installed equipment may include multiple categories.

A motor drive panel, for example, may be CAT III on the 480-voltpower side, and CAT I on the control side.

8. MEASUREMENT TOOLS PART OF PERSONALPROTECTIVE EQUIPMENT

Another organization playing an important role in estab-lishing safety standards for electrical workers is the National FireProtection Association (NFPA). NFPA establishes guidelines forelectrical measurement tools in its standard 70E, “Standard forElectrical Safety Requirements for Employee Workplaces, 2004Edition”.

Standard 70E also includes important requirementsregarding the use of other Personal Protective Equipment (PPE)in various environments and installation/maintenance activities.

The NFPA standard makes it clear that test instrumentsand accessories must be matched to the environment where theywill be used. These are the pertinent sections:

• “Test instruments, equipment, and their accessories shallbe rated for circuits and equipment to which they will beconnected.” (Part II, Chapter 3, Paragraph 3-4.10.1)

• “Test instruments, equipment, and their accessories shall


be designed for the environment to which they will beexposed, and for the manner in which they will be used.”(Part II, Chapter 3, Paragraph 3-4.10.2)

A table included in NFPA Standard 70E, Table 3-3.9,“Hazard Risk Category Classifications,” provides additionalguidance regarding the personal protective equipment recom-mended for use in work on a variety of equipment types at vari-ous voltage levels.i

9. SAFETY REQUIREMENTS FOR MEASUREMENT TOOLS Management must ensure that, in compliance with NFPA

70E, test tools meet the standards for the environment wherethey are used. The entire testing ‘system’, including the meterand its internal fusing system, as well as the test leads and attach-ments, must comply with regulations for measurement environ-ment and hazard level.

In addition, tools must be included as an integral part ofthe Personal Protective Equipment that technicians are requiredto use when working on high-energy systems.

Beyond these requirements, however, management mustensure that the measurement tools in use are designed, certifiedand maintained so that they will meet the more advanced andstringent safety requirements of today. Management must accountfor three factors when assessing test tool safety: Category rating(older, unrated tools were not made for today’s electrical environ-ment), independent testing and certification, and regular inspectionand maintenance. It is important to note that the category rating forpersonnel protective equipment has no relationship to the CATratings identified as part of the markings of test and measure-ment equipment.

Category rating for PPE – Testers should be rated forthe electrical environment in which they will be used. For example,a 220-volt, three-phase system requires a tester rated CAT III orIV. Old, unrated test instruments do not meet IEC guidelines forrequired PPE. While they may be perfectly accurate and appearto perform well, even the best meters of yesterday were designedfor a world where working conditions and safety standards werefar different. Such test tools may not meet contemporary standards.

Independent Testing and Certification – Even in thevital area of safety, some tools may not perform as promised bythe manufacturer. Measuring devices rated for a high-energyenvironment may not actually deliver the safety protections,such as adequate fusing, claimed on their specification sheets.

THE CRUCIAL DIFFERENCE BETWEEN ‘DESIGNED’ AND ‘TESTED’ It is important to understand that standards bodies such as

ANSI, CSA and IEC are not responsible for enforcing their stan-dards. This means that a meter designed to a standard may notactually have been tested and proven to meet that standard. It isnot uncommon for meters under test to fail before achieving theperformance their manufacturers claim for them.

The best assurance for users and their employers is toselect test instruments that have been tested and certified to per-form up to specification by independent testing laboratories. Toprovide an extra measure of confidence, select test tools labeledto show that they have been certified to meet the appropriatecontemporary standards by two or more independent labs. Thisensures that test instruments have passed the most rigorous testsand meet every applicable standard. Such independent testinglabs include Underwriters Laboratories (UL) in the UnitedStates, Canadian Standards Association (CSA) in Canada andTUV Product Service in Europe.ii

10. TEST TOOL INSPECTION AND MAINTENANCE Regular Inspection and Maintenance – To perform

accurately and safely, test tools must be regularly inspected andmaintained. The need for inspection is clearly recognized by theNational Fire Protection Association. NFPA Standard 70E laysout the requirement that test tools must be visually inspected fre-quently to help detect damage and ensure proper operation. PartII, Chapter 4, Paragraph 4-1.1 provides the details:

Visual Inspection. Test instruments and equipment andall associated test leads, cables, power cords, probes, and con-nectors shall be visually inspected for external defects and dam-age before the equipment is used on any shift. If there is a defector evidence of damage that might expose an employee to injury,the defective or damaged item shall be removed from service,and no employee shall use it until repairs and tests necessary torender the equipment safe have been made.iii

Visual inspection alone, however, may not detect all pos-sible test instrument problems. To help ensure the highest levelof safety and performance, additional inspection and testingshould be conducted:

Additional Visual Inspection – Test tools should bechecked for the following points:

• Look for the 1000-volt, CAT III or 600-volt, CAT IV rat-ing on the front of meters and testers, and a “doubleinsulated” symbol on the back.

• Look for approval symbols from two or more independenttesting agencies, such as UL, CSA, CE, TUV or CTICK.

• Make sure that the amperage and voltage of meter fusesis correct. Fuse voltage must be as high or higher thanthe meter’s voltage rating. The second edition ofIEC/ANSI/CSA standards states that test equipmentmust perform properly in the presence of impulses onvolts and amps measurement functions. Ohms and con-tinuity functions are required to handle the full metervoltage rating without becoming a hazard.

• Check the instrument’s manual to determine whether theohms and continuity circuits are protected to the samelevel as the voltage test circuit. If the manual does notindicate, your supplier should be able to determinewhether the meter passed the second edition of IEC61010or ANSI S82.02.

• Check the overall condition of the meter or tester, look-ing for such problems as a broken case, worn test leadsor a faded display.

• Use the meter’s own test capability to determinewhether fuses are in place and functioning properly.

Step 1: Plug test lead in V/ Ω input. Select Ω.Step 2: Insert probe tip into mA input. Read value.Step 3: Insert probe tip into A input. Read value.

Typically a fuse in good condition should showmvalue of close to zero, but you should alwayscheck your meter owner’s manual for the speci-fied reading.

Inspecting Test Leads and Probes – As integral compo-nents of the test tool system, test leads, probes and attachmentsmust meet the requirements of the testing environment and bedesigned to minimize hazard. Test leads must be certified to acategory that equals or exceeds that of the meter or tester.

• Examine test leads for such features as shrouded con-nectors, finger guards, CAT ratings that equal or exceedthose of the meter, and double insulation.


• Visually inspect for frayed or broken wires. The lengthof exposed metal on test probe tips should be minimal.

• Test leads can fail internally, creating a hazard that can-not be detected through visual inspection. But it is pos-sible to use the meter’s own continuity testing functionto check for internal breaks.

Step 1: Insert leads in V/ Ω and COM inputsStep 2: Select Ω, touch probe tips. Good leads are 0.1 –

0.3 Ω.

11. SAFE MEASUREMENT PROCESSES AND PROCEDURES In addition to the consistent use of safe, correctly rated

and inspected test tools discussed in the preceding sections, safeelectrical measurement requires adherence to correct measure-ment procedures. Safety training programs should incorporateboth elements of safe measurement – equipment and procedures.

In addition to equipment inspection (detailed in Section10 above), safe measurement procedures include:

• Lockout/Tagout procedures – NFPA provides exten-sive information and guidance on lockout/tagout prac-tices and devices in Part II, Chapter 5 of NFPA 70E.iv

• Three-step test procedure – Before making the determi-nation that a measured circuit is dead, it is important toverify that test instruments are operating correctly. To doso, the technician should use a three-step test procedure.

First, check for correct test tool operation by using thetool to test a circuit known to be live. Then, test the target circuit.Finally, as a double check on test tool operation, test the originalknown circuit once again. This procedure provides the user astrong measure of confidence that the test tool is operating cor-rectly, and that the target circuit is performing as measured.

• Neutral first and last – The user should attach the testlead to a neutral contact first, then attach a lead to a hotcontact to conduct the test. In detaching test leads, firstremove the hot contact, then remove the neutral test lead.

• One hand only – When possible, it is good practice tofollow the old electrician’s advice and keep one hand ina pocket when testing. But common sense must rule.Conditions at the test location may make it impracticalto use this technique.

12. CONCLUSIONS AND RECOMMENDATIONS Unlike some other important safety initiatives, the measures

required to bolster the safety of electrical measurement tools andprocedures are not difficult or costly. Yet these steps can provideimportant benefits by improving worker safety, avoiding the dis-ruption of business operations, reducing risk and avoiding possibleincreases in insurance costs.

Employers should begin by ensuring that technicians arefully trained in correct use of all personal protective equipment,including test instruments.

As a companion measure, make sure the required PPE isreadily available, meets today’s standards, and is inspected toensure it is in optimum condition.

Test instruments are an essential component of PPE.Employers should inspect all test instruments to ensure they arerated, tested and certified by independent testing agencies tomeet safety requirements for the environments where they areused. Replace test instruments that do not meet current stan-dards, because they may create extra hazard, risk and liability.

Finally, personnel should be trained in the correct proce-dures for taking safe measurements, including methods for per-sonally inspecting and testing their instruments to ensure theyare in good condition and function correctly.

i NFPA 70E Standard for Electrical Safety Requirements forEmployee Workplaces, 2000 Edition, pages 55 through 58. ©2000 NFPA

ii For more information on these testing organizations, visit theirwebsites: http://www.ul.com/http://www.csa.ca/Default.asp?language=Englishhttp://www.tuvamerica.com/services/electrical/lowvolt.cfm

iii NFPA 70E Standard for Electrical Safety Requirements forEmployee Workplaces, 2000 Edition, page 63. © 2000 NFPA

iv Ibid, pp 64-66.


OVERVIEW Voltage, current, temperature, pressure, strain, and flow

measurements are an integral part of industrial and process con-trol applications. Often these applications involve environmentswith hazardous voltages, transient signals, common-mode volt-ages, and fluctuating ground potentials capable of damagingmeasurement systems and ruining measurement accuracy. Toovercome these challenges, measurement systems designed forindustrial applications make use of electrical isolation. Thiswhite paper focuses on isolation for analog measurements, provides answers to common isolation questions, and includesinformation on different isolation implementation technologies.

UNDERSTANDING ISOLATION Isolation electrically separates the sensor signals, which

can be exposed to hazardous voltages1, from the measurementsystem’s low-voltage backplane. Isolation offers many benefitsincluding:

• Protection for expensive equipment, the user, and datafrom transient voltages

• Improved noise immunity • Ground loop removal • Increased common-mode voltage rejection Isolated measurement systems provide separate ground

planes for the analog front end and the system backplane to sep-arate the sensor measurements from the rest of the system. Theground connection of the isolated front end is a floating pin thatcan operate at a different potential than the earth ground. Figure 1represents an analog voltage measurement device. Any common-mode voltage that exists between the sensor ground and the meas-urement system ground is rejected. This prevents ground loopsfrom forming and removes any noise on the sensor lines.

NEED FOR ISOLATION Consider isolation for measurement systems that involve

any of the following: • Vicinity to hazardous voltages • Industrial environments with possibility of transient

voltages • Environments with common mode voltage or fluctuat-

ing ground potentials • Electrically noisy environments such as those with

industrial motors • Transient sensitive applications where it is imperative

to prevent voltage spikes from being transmitted throughthe measurement system

Industrial measurement, process control, and automotivetest are examples of applications where common-mode voltages,high-voltage transients, and electrical noise are common.Measurement equipment with isolation can offer reliable measure-ments in these harsh environments. For medical equipment indirect contact with patients, isolation is useful in preventing powerline transients from being transmitted through the equipment.

Based on your voltage and data rate requirements, youhave several options for making isolated measurements. Youcan use plug-in boards for laptops, desktop PCs, industrial PCs,PXI, Panel PCs, and Compact PCI with the option of built-inisolation or external signal conditioning. Isolated measurementscan also be made using programmable automation controllers(PACs) and measurement systems for USB.

ISOLATION TECHNOLOGIES FOR RELIABLEINDUSTRIAL MEASUREMENTS

National Instruments

Figure 1. Bank Isolated Analog Input Circuitry

Hazardous Voltages are greater than 30 Vrms, 42.4 Vpk or 60 VDC Figure 2. Isolated Data Acquisition Systems


METHODS OF IMPLEMENTING ISOLATION Isolation requires signals to be transmitted across an isola-

tion barrier without any direct electrical contact. Light emittingdiodes (LEDs), capacitors, and inductors are three commonlyavailable components that allow electrical signal transmissionwithout any direct contact. The principles on which these devicesare based form the core of the three most common technologiesfor isolation – optical, capacitive, and inductive coupling.

OPTICAL COUPLING LEDs produce light when a voltage is applied across

them. Optical isolation uses an LED along with a photo-detectordevice to transmit signals across an isolation barrier using lightas the method of data translation. A photo-detector receives the lighttransmitted by the LED and converts it back to the original signal.

Optical isolation is one of the most commonly used methodsfor isolation. One benefit of using optical isolation is its immunityto electrical and magnetic noise. Some of the disadvantagesinclude transmission speed, which is restricted by the LEDswitching speed, high-power dissipation, and LED wear.

CAPACITIVE COUPLING Capacitive isolation is based on an electric field that

changes based on the level of charge on a capacitor plate. Thischarge is detected across an isolation barrier and is proportionalto the level of the measured signal.

One advantage of capacitive isolation is its immunity tomagnetic noise. Compared to optical isolation, capacitive isola-tion can support faster data transmission rates because there areno LEDs that need to be switched. Since capacitive couplinginvolves the use of electric fields for data transmission, it can besusceptible to interference from external electric fields.

INDUCTIVE COUPLING In the early 1800s, Hans Oersted, a Danish physicist, dis-

covered that current through a coil of wire produces a magneticfield. It was later discovered that current can be induced in a

second coil by placing it in close vicinity of the changing mag-netic field from the first coil. The voltage and current induced inthe second coil depend on the rate of current change through thefirst. This principle is called mutual induction and forms thebasis of inductive isolation.

Inductive isolation uses a pair of coils separated by alayer of insulation. Insulation prevents any physical signaltransmission. Signals can be transmitted by varying currentflowing through one of the coils, which causes a similar currentto be induced in the second coil across the insulation barrier.Inductive isolation can provide high-speed transmission similarto capacitive techniques. Because inductive coupling involvesthe use of magnetic fields for data transmission, it can be sus-ceptible to interference from external magnetic fields.

ANALOG ISOLATION AND DIGITAL ISOLATION Several commercial off-the-shelf (COTS) components

are available today, many of which incorporate one of the abovetechnologies to provide isolation. For analog input/output chan-nels, isolation can be implemented either in the analog sectionof the board, before the analog-to-digital converter (ADC) hasdigitized the signal (analog isolation) or after the ADC has digitized the signal (digital isolation). Different circuitry needsto be designed around one of these techniques based on the loca-tion in the circuit where isolation is being implementing. You canchoose analog or digital isolation based on your data acquisitionsystem performance, cost, and physical requirements. Figure 6shows the different stages of implementing isolation.

Figure 3. Optical Coupling

Figure 4. Capacitive Isolation

Figure 5. Inductive Coupling

Figure 6a. Analog Isolation

Figure 6b. Digital Isolation


The following sections cover analog and digital isolationin more detail and explore the different techniques for imple-menting each.

ANALOG ISOLATION The isolation amplifier is generally used to provide isolation

in the analog front end of data acquisition devices. “ISO Amp”in Figure 6a represents an isolation amplifier. The isolationamplifier in most circuits is one of the first components of theanalog circuitry. The analog signal from a sensor is passed to theisolation amplifier which provides isolation and passes the signalto the analog-to-digital conversion circuitry. Figure 7 representsthe general layout of an isolation amplifier.

In an ideal isolation amplifier, the analog output signal isthe same as the analog input signal. The section labeled “isolation”in Figure 7 uses one of the techniques discussed in the previoussection (optical, capacitive, or inductive coupling) to pass thesignal across the isolation barrier. The modulator circuit pre-pares the signal for the isolation circuitry. For optical methods,this signal needs to be digitized or translated into varying lightintensities. For capacitive and inductive methods, the signal istranslated into varying electric or magnetic fields. The demodu-lator circuit then reads the isolation circuit output and convertsit back into the original analog signal.

Because analog isolation is performed before the signal isdigitized, it is the best method to apply when designing externalsignal conditioning for use with existing non-isolated data acquisi-tion devices. In this case, the data acquisition device performs the analog-to-digital conversion and the external circuitry provides isolation. With the data acquisition device and external signal con-ditioning combination, measurement system vendors can developgeneral-purpose data acquisition devices and sensor-specific signalconditioning. Figure 8 shows analog isolation being implementedwith flexible signal conditioning that uses isolation amplifiers.Another benefit to isolation in the analog front end is protection forthe ADC and other analog circuitry from voltage spikes.

There are several options available on the market formeasurement products that use a general-purpose data acquisi-tion device and external signal conditioning. For example, theNational Instruments M Series includes several non-isolated, gen-eral-purpose multifunction data acquisition devices that providehigh-performance analog I/O and digital I/O. For applicationsthat need isolation, you can use the NI M Series devices withexternal signal conditioning, such as the National InstrumentsSCXI or SCC modules. These signal conditioning platformsdeliver the isolation and specialized signal conditioning neededfor direct connection to industrial sensors such as load cells, straingages, pH sensors, and others.

DIGITAL ISOLATION Analog-to-digital converters are one of the key compo-

nents of any analog input data acquisition device. For best performance, the input signal to the analog-to-digital convertershould be as close to the original analog signal as possible.Analog isolation can add errors such as gain, non-linearity andoffset before the signal reaches the ADC. Placing the ADC clos-er to the signal source can lead to better performance. Analogisolation components are also costly and can suffer from longsettling times. Despite better performance of digital isolation,one of the reasons for using analog isolation in the past was toprovide protection for the expensive analog-to-digital convert-ers. As the ADCs prices have significantly declined, measure-ment equipment vendors are choosing to trade ADC protectionfor better performance and lower cost offered by digital isola-tors (see Figure 9).

Compared to isolation amplifiers, digital isolation compo-nents are lower in cost and offer higher data transfer speeds. Digitalisolation techniques also give analog designers more flexibility tochoose components and develop optimal analog front ends formeasurement devices. Products with digital isolation use current-and voltage-limiting circuits to provide ADC protection. Digitalisolation components follow the same fundamental principles ofoptical, capacitive, and inductive coupling that form the basis ofanalog isolation.

Figure 7. Isolation Amplifier

Figure 8. Use of Isolation Amplifiers in Flexible Signal Conditioning Hardware

Figure 9. Declining Price of 16-Bit Analog-to-Digital Converters Graph Source: National Instruments and a Leading ADC Supplier


Leading digital isolation component vendors such asAvago Technologies (www.avagotech.com), Texas Instruments(www.ti.com), and Analog Devices (www.analog.com) havedeveloped their isolation technologies around one of these basicprinciples. Avago Technologies offers digital isolators based onoptical coupling, Texas instruments bases its isolators on capac-itive coupling, and Analog Devices isolators use inductive coupling.

OPTOCOUPLERS Optocouplers, digital isolators based on the optical cou-

pling principles, are one of the oldest and most commonly usedmethods for digital isolation. They can withstand high voltagesand offer high immunity to electrical and magnetic noise.Optocouplers are often used on industrial digital I/O products,such as the National Instruments PXI-6514 isolated digitalinput/output board (see Figure 10) and National InstrumentsPCI-7390 industrial motion controller.

For high-speed analog measurements, optocouplers,however, suffer from speed, power dissipation, and LED warelimitations associated with optical coupling. Digital isolatorsbased on capacitive and inductive coupling can alleviate manyoptocoupler limitations.

CAPACITIVE ISOLATION Texas Instruments offers digital isolation components

based on capacitive coupling. These isolators provide high datatransfer rates and high transient immunity. Compared to capac-itive and optical isolation methods inductive isolation offerslower power consumption.

INDUCTIVE ISOLATION iCoupler® technology, introduced by Analog Devices in

2001 (www.analog.com/iCoupler), uses inductive coupling tooffer digital isolation for high-speed and high-channel-countapplications. iCouplers can provide 100 Mb/s data transfer rateswith 2,500 V isolation withstand; for a 16-bit analog measure-ment system that implies sampling rates in the mega hertzrange. Compared to optocouplers, iCouplers offer other benefitssuch as reduced power consumption, high operating temperaturerange up to 125 °C, and high transient immunity up to 25 kV/ms.

iCoupler technology is based on small, chip-scale trans-formers. An iCoupler has three main parts – a transmitter, trans-formers, and a receiver. The transmitter circuit uses edge trigger

encoding and converts rising and falling edges on the digitallines to 1 ns pulses. These pulses are transmitted across the iso-lation barrier using the transformer and decoded on the otherside by the receiver circuitry (see Figure 11). The small size ofthe transformers, about three-tenths of a millimeter, makes thempractically impervious to external magnetic noise. iCouplerscan also lower measurement hardware cost by integrating up tofour isolated channels per integrated circuit (IC) and, comparedto optocouplers, they require fewer external components.

Measurement hardware vendors are using iCouplersto offer high-performance data acquisition systems at lowercosts. National Instruments industrial data acquisitiondevices intended for high-speed measurements, such as theisolated M Series multifunction data acquisition devices,use iCoupler digital isolators (see Figure 12). These devicesprovide 60 VDC continuous isolation and 1,400 Vrms/1,900VDC channel-to-bus isolation withstand for 5 s on multipleanalog and digital channels and support sampling rates up to250 kS/s. National Instruments C Series modules used in the NIPAC platform, NI CompactRIO, NI CompactDAQ, and otherhigh-speed NI USB devices also use the iCoupler technology.

SUMMARY Isolated data acquisition systems can provide reliable

measurements for harsh industrial environments with hazardousvoltages and transients. Your need for isolation is based on yourmeasurement application and surrounding environments.Applications that require connectivity to different specialty sen-sors using a single, general-purpose data acquisition device canbenefit from external signal conditioning with analog isolation.Where as applications needing lower-cost, high-performanceanalog inputs benefit from measurement systems with digitalisolation technologies.

Figure 10. Industrial Digital I/O Products Optpcouplers

Figure 11. Introduction Coupling-Based iCoupler Technology from Analog DevicesSource: Analog Devices (www.analog.com/iCoupler)

Figure 12. National Instruments Isolated M Series Multifuntion DAQ Uses


FOUR-POINT RESISTANCE MEASUREMENTSOhmmeter measurements are normally made with just a

two-point measurement method. However, when measuring verylow values of ohms, in the milli- or micro-ohm range, the two-pointmethod is not satisfactory because test lead resistance becomes asignificant factor.

A similar problem occurs when making ground mat resist-ance tests, because long lead lengths of up to 1000 feet are used.Here also, the lead resistance, due to long lead length, will affectthe measurement results.

The four-point resistance measurement method eliminateslead resistance. Instruments based on the four-point measure-ment work on the following principle:

• Two current leads, C1 and C2, comprise a two-wire cur-rent source that circulates current through the resistanceunder test.

• Two potential leads, P1 and P2, provide a two-wire volt-age measurement circuit that measures the voltage dropacross the resistance under test.

• The instrument computes the value of resistance fromthe measured values of current and voltage.

THREE-POINT RESISTANCE MEASUREMENTSThe three-point method, a variation of the four-point

method, is usually used when making ground (earth) resistancemeasurements. With the three-point method, the C1 and P1 terminalsare tied together at the instrument and connected with a shortlead to the ground system being tested. This simplifies the test inthat only three leads are required instead of four. Because thiscommon lead is kept short, when compared to the length of theC2 and P2 leads, its effect is negligible. Some ground testers areonly capable of the three-point method, so are equipped with

only three test terminals. The three-point method for ground sys-tem testing is considered adequate by most individuals in theelectrical industry and is employed on the TPI MFT5010 and theTPI ERT1500.

The four-point method is required to measure soil resistivity.This process requires a soil cup of specific dimensions into whicha representative sample of earth is placed. This process is not oftenemployed in testing electrical ground systems although it may bepart of an initial engineering study.

PURPOSE/TPI INSTRUMENT FEATURES

PURPOSEThe purpose of electrical ground testing is to determine

the effectiveness of the grounding medium with respect to trueearth. Most electrical systems do not rely on the earth to carryload current (this is done by the system conductors) but the earthmay provide the return path for fault currents, and for safety, allelectrical equipment frames are connected to ground.

The resistivity of the earth is usually negligible becausethere so much of it available to carry current. The limiting factorin electrical grounding systems is how well the grounding elec-

trodes contact the earth, which is known as thesoil/ground rod interface. This interface resistance com-ponent, along with the resistance of the grounding con-ductors and the connections, must be measured by theground test.

In general, the lower the ground resistance, thesafer the system is considered to be. There are differentregulations which set forth the maximum allowableground resistance, for example: the National ElectricalCode specifies 25 ohms or less; MSHA is more strin-gent, requiring the ground to be 4 ohms or better; electricutilities construct their ground systems so that theresistance at a large station will be no more than a fewtenths of one ohm.

TPI GROUND TEST INSTRUMENT CHARACTERISTICS• To avoid errors due to galvanic currents in the earth, TPI

ground test instruments use an AC current source.• A frequency other than 60 hertz is used to eliminate the

possibility of interference with stray 60 hertz currentsflowing through the earth.

• The three-point measurement technique is utilized toeliminate the effect of lead length.

• The test procedure, known as the Fall-of-PotentialMethod, is described on the following page.

RESISTANCE MEASUREMENTSTHREE- AND FOUR-POINT METHOD

Figure 1


THREE-POINT FALL-OF-POTENTIAL TEST PROCEDURE

GROUND TEST PROCEDURE In the Fall-of-Potential Method, two small ground rods –

often referred to as ground spikes or probes – about 12" long areutilized. These probes are pushed or driven into the earth farenough to make good contact with the earth (8" – 10" is usuallyadequate). One of these probes, referred to as the remote currentprobe, is used to inject the test current into the earth and is placedsome distance (often 100') away from the grounding mediumbeing tested . The second probe, known as the potential probe, isinserted at intervals within the current path and measures thevoltage drop produced by the test current flowing through theresistance of the earth.

In the example shown on the following page, the remotecurrent probe C2 is located at a distance of 100 feet from theground system being tested. The P2 potential probe is taken outtoward the remote current probe C2 and driven into the earth atten-foot increments.

Based on empirical data (data determined by experiment andobservation rather than being scientifically derived), the ohmic valuemeasured at 62% of the distance from the ground-under-test to theremote current probe, is taken as the system ground resistance.

The remote current probe must be placed out of the influ-ence of the field of the ground system under test. With all but thelargest ground systems, a spacing of 100 feet between the ground-under-test and the remotecurrent electrode is adequate.When adequate spacingbetween electrodes exists, aplateau will be developed onthe test graph. Note: A remotecurrent probe distance of lessthan 100 feet may be ade-quate on small ground sys-tems.

When making a test where sufficient spacing exists, theinstrument will read zero or very near zero when the P2 poten-tial probe is placed near the ground-under-test. As the electrodeis moved out toward the remote electrode, a plateau will bereached where a number of readings is approximately the samevalue (the actual ground resistance is that which is measured at62% of the distance between the ground mat being tested and theremote current electrode). Finally, as the potential probeapproaches the remote current electrode, the resistance reading willrise dramatically.

It is not absolutely necessary to make a number of measure-ments as described above and to construct a graph of the readings.However, we recommend this as it provides valuable data for futurereference and, once you are setup, it takes only a few minutes totake a series of readings.

The electrical fields associated with the ground grid andthe remote electrodes are illustrated on AN0009-5. An actualground test is detailed on AN0009-6, and a sample Ground TestForm is provided on AN0009-7. See AN0009-8 for a simpleshop-built wire reel assembly for testing large ground systems.

SHORT-CUT METHODThe short cut method described here determines the

ground resistance value and verifies sufficient electrode spacing –and it does save time. This procedure uses the 65' leads suppliedwith the TPI instruments.

• Connect the T1 instrument jack with the 15' green leadto the ground system being tested.

• Connect the T3 instrument jack with the red lead to theremote current electrode (spike) placed at distance of 65'(full length of conductor) from the ground grid beingtested.

• Connect the T2 instrument jack with the black lead tothe potential probe placed at 40 feet (62% of the 65' dis-tance) from the ground grid being tested and measurethe ground resistance.

• Move the P2 potential probe 6' (10% of the total dis-tance) to either side of the 40' point and take readings ateach of these points. If the readings at these two pointsare essentially the same as that taken at the 40' point, ameasurement plateau exists and the 40' reading is valid.A substantial variation between readings indicates insuf-ficient spacing.

THREE-POINT FALL-OF-POTENTIAL METHOD

INSTRUMENT SET-UP

Figure 2

Figure 3


A NOTE ON INSTRUMENT LABELING CONVENTIONSThe TPI MFT5010 and TPI ERT1500 use

the terminal designations T1 (C1/P1), T2 (P2), andT3 (C2).

The corresponding lead designations on theMFT5010 are E (Earth), S & H.

The corresponding lead designations on theERT1500 are E (Earth), P (Potential), C (Current).

TEST CURRENT PATH• Test Current (AC ) flows from instrument

T3 to remote current probe C2 on the redlead.

• Test Current flows from remote currentprobe C2 back through the earth to theground being tested as shown by dashedblue line.

• Test current flows out of ground grid backto instrument T1 on the short green lead.

• Black potential lead P1 is connected to instrumentT2 and is taken out at 10' increments. It measuresvoltage drop produced by the test current flowingthrough the earth. (P1 to P2 potential)

EQUAL-POTENTIAL PLANES

THE EXISTENCE OF EQUAL-POTENTIAL PLANES• When current flows through the earth from a remote test

electrode (in the case of a ground test) or remote fault, the volt-age drop which results from the flow of current through theresistance of the earth can be illustrated by equal-potentialplanes. The equal-potential planes are represented in the dashedlines in drawings below where the spacing between concentriclines represents some fixed value of voltage.

• The concentration of the voltage surrounding a ground-ing element is greatest immediately adjacent to that ground. Thisis shown by the close proximity of lines at the point where thecurrent enters the earth and again at the point where the currentleaves the earth and returns to the station ground mat.

• In order to achieve a proper test using the Fall-of-PotentialGround Test Method, sufficient spacing must exist between thestation ground mat being tested and the remote current electrodesuch that the equal-potential lines do not overlap. As shown by theblack line in the Sample Plot, adequate electrode spacing willresult in the occurrence of a plateau on the resistance plot. Thisplateau must exist at 62% of the distance between the ground matand the remote electrode for the test to be valid. Insufficient spac-ing results in an overlap of these equal-potential planes, as illus-trated at the bottom of this page and by the red line on the SamplePlot.

• See the Safety Note on AN0009-6 for information on thehazards of Step and Touch-Potentials.

Figure 4

Figure 5


ACTUAL FIELD TEST

This actual ground test was conducted on a pad-mounttransformer in a rural mountain area. The single-phase trans-former is supplied by a 12470/7200 volt grounded wye primaryand the transformer is grounded by its own ground rod as well asbeing tied to the system neutral which is grounded at multiplepoints along the line. The distribution line is overhead with justthe “dip” to the transformer being underground.

Ground Test Data Remote Current Probe C2 @ 100 Feet

P2 Distance from Transformer in Feet Instrument Reading in Ohms

10 1.83

20 3.59

30 3.85

40 3.95

50 4.0

60 4.25

62* 4.3

70 4.5

80 5.4

90 7.3

100 25.02

* Actual Ground resistance.

Figure 6

TEST PROCEDURETerminal T1 of the TPI MFT5010 tester was connected to

the transformer case ground with the short green lead. Theremote Current Probe C2 was driven in the ground at a location100 feet from the transformer and connected to Terminal T3 ofthe instrument with the red test lead.

Terminal T2 of the tester was connected, using the 100'black lead, to the P2 potential probe. This ground stake was insertedinto the ground at 10' intervals and a resistance measurement wasmade at each location and recorded in the table above.

The relatively constant readings in the 4 ohm range between40 and 70 feet are a definite plateau that indicates sufficient lead


spacing. The initial readings close to the transformer are lower, andthere is a pronounced “tip-up” as the P2 probe approaches theremote current electrode C2.

The measured ground resistance at 62 feet (62% of thedistance) was 4.3 ohms and is taken as the system ground resist-ance. This is an excellent value for this type of an installation.


SAFETY NOTE – POSSIBLE EXISTENCE OF HAZARDOUSSTEP AND TOUCH POTENTIALS

It is recommended that rubber gloves be worn when drivingthe ground rods and connecting the instrument leads.

The possibility of a system fault occurring at the time theground test is being conducted is extremely remote.

However, such a fault could result in enough current flowthrough the earth to cause a possible hazardous step potentialbetween a probe and where the electrician is standing, or hazardoustouch potential between the probes and the system ground. Thelarger the system, in terms of available fault current, the greater thepossible risk.

REEL ASSEMBLY

A SHOP-BUILT GROUND TEST WIRE REEL ASSEMBLYThis simple, low-cost, and easy-to-build wire reel assembly

is handy for making Ground (Earth) Resistance measurements onlarge ground systems. The unit shown below has 500 feet of wirefor testing medium-to-large ground fields typical of those found inindustrial plants and substations. For testing even larger systems,such as those installed for power generating plants, wire lengths of1000 feet can be used. Wrap-on wire markers are installed everyten feet on the current lead to simplify placement of the remotecurrent and potential probes. Your electrical distributor will prob-ably have empty surplus reels available for the asking – the onesshown below are about 12 inches in diameter. The conductor isstandard #12 THHN. Even though the TPI ERT1500 and theMFT5010 use an AC test signal, the test results are unaffected bythe inductance of any wire left on the reels.


1. Turn instrument on by pressing the green “ON/OFF”button (far right). Continue holding the green buttondown until the battery life indicator comes on.

2. Check battery life indicator – make sure at least 20percent remains.

3. Check calibration – locate the 25W calibration gaugesupplied with the tester and clamp the meter aroundany leg of the gauge.

4. Observe instrument reading – the reading should bewithin 1.0W of gauge specification (25W). If readingis correct, proceed to step 5. If not, clean instrumentand repeat steps 3 and 4. If you are not able to get theinstrument to read within 1.0W after cleaning instru-ment, do not proceed. Have the instrument repaired.

5. Remove instrument from gauge. Observe instrumentreading with nothing in the clamps. The readingshould be greater than 1000W OR read. If either ofthese conditions is observed, continue to step 6. If not,clean instrument (see instructions below) and repeatsteps 3 through 5. If, after cleaning instrument, you arestill unable to get the instrument to perform asdescribed in steps 4 and 5, open the jaws approximate-ly 1/2 inch and let them snap shut. Make sure that thejaws close properly. If the unit still does not performproperly, do not proceed. Have the instrumentrepaired.

6. Switch instrument to Current Mode. (Press buttonlabeled “A” for Amps)

7. Clamp instrument around the ground wire or rod.

8. Observe reading – if less than 1.0A, proceed to step 9.If between 1.0 and 5.0A, make note of reading andcontinue to step 9. If greater than 5A, terminate testand remove instrument from the ground wire or rodand correct the problem before re testing.

9. Switch instrument to Resistance (W) Mode. (Pressbutton labeled with Ohm (W) symbol)

10. Wait for reading to stabilize and record reading. Lockreading by pressing “HOLD”.

11. Remove instrument from ground wire or rod andreclamp to gauge.

12. Observe reading – the reading should be within 1.0Wof gauge value. If reading is OK – measurement isvalid. If reading is wrong, clean instrument (seeinstructions below) and repeat from step 4.

CLEANING THE HEADSTo ensure optimum performance, it is important to keep

the probe jaw mating surfaces clean at all times. Failure to do somay result in erroneous readings. To clean the probe jaws, use avery fine sandpaper (600 grit) to avoid scratching the surface,then gently clean with a soft cloth. Make sure that the instru-

CLAMP-ON GROUND RESISTANCE TESTERMODELS 3711 & 3731STEP-BY-STEP USAGE

Chauvin Arnoux, Inc. AEMC Instruments

step 2

step 3


ment is oriented such that no debris or filings will fall into theunit while cleaning. Check with your finger afterwards to besure that no foreign material remains on the jaw surfaces (bothtop and bottom).

CLAMP-ON GROUND RESISTANCE TESTINGThe clamp-on ground resistance testing technique offers

the ability to measure the resistance without disconnecting theground. This type of measurement also offers the advantage ofincluding the bonding to ground and the overall grounding con-nection resistances.

PRINCIPLES OF OPERATIONUsually, a common distribution line grounded system can

be simulated as a simple basic circuit as shown in Figure A oran equivalent circuit, shown in Figure B. If voltage E is appliedto any measured grounding system. Rx through a special trans-former (used in Models 3711 and 3731), current I flows through

the circuit, thereby establishing the following equation.Therefore, E/I = Rx is established. If it is detected with E

kept constant, measured grounding resistance can be obtained.Refer again to Figures A and B. Current is fed to a special trans-former via a power amplifier from a 2.3 kHz constant voltageoscillator. This current is detected by a detection CT. Only the 2.3kHz signal frequency is amplified by a filter amplifier. This occursbefore the A/D conversion and after synchronous rectification. Itis then displayed on the LCD of the Model 3711/3731 meter.

The filter amplifier is used to cut off both earth current atcommercial frequency and high-frequency noise. Voltage isdetected by coils wound around the injection CT, which is thenamplified, rectified, and compared by a level comparator. If theclamp is not closed properly, an “open jaw” annunciator appears

on the LCD.The important points to consider for proper use of the

clamp-on ground tester are:1. There is a series-parallel resistance path down stream

from the measurement point that is lower in resistancethan the point being measured.

2. That the earth is the return path to the point where theclamp-on meter is connected and not wire or othermetal structures (see Figure C).

3. If the measurement point is not connected to a series-parallel low resistance network (such as the case witha single rod), a temporary path may be created by con-necting a jumper cable from the measurement point toa low resistance like a pole ground (see Figure D).


Every thing electrical from a toaster to a high-voltagepower line produces electric and magnetic fields. Both the electricand magnetic fields are strong close to an operating source. Thestrength of the electric field depends on the voltage and is presentin any live wire whether an electrical appliance is being used or not.Magnetic fields, on the other hand, are produced by electric cur-rents and are only present when an appliance is operating i.e. thereis no magnetic field when an electrical appliance is turned off.

HEALTH EFFECTSCurrently there is no evidence that exposure to electric

fields is a health hazard (excluding electric shock). Whetherexposure to magnetic fields is equally harmless remains an openquestion. A large number of scientific studies performed on ani-mals and cells have not found a health risk. Some epidemiologicalstudies, however, have suggested a weak link between intense andprolonged exposure to magnetic fields and childhood leukaemia.

MAGNETIC FIELD UNITSThe strength of the magnetic field is expressed in units of

Tesla (T) or microtesla (µT). Another unit, which is commonlyused is the Gauss (G) or milligauss (mG), where 1 G is equiva-lent to 10-4 T (or 1 mG = 0.1 µT).

THE GAUSS METERThere is a range of different instruments that can measure

the magnetic field strength. The gauss meter is a hand-helddevice that provides a simple way of performing such measure-ments. ARPANSA has two different gauss meter models availablefor hire, which are a Teslatronics Model 70 and a Sypris Model4080. Both these instruments operate in a similar manner and theyare shown in the figure below.

Both gauss meters measure alternating fields from 25 Hzto 1000 Hz in units of mG. They do not measure and will givefalse readings from mobile phones. Readings taken very close (afew cm) to other electronic devices (as distinct from electricaldevices such as heaters, washing machines etc) may also givefalse readings. Shaking or vibrating either unit may also givefalse readings. Since the meters only measure varying magneticfields, they will not measure the earth’s magnetic field which isstatic and has a value of approximately 500 mG.

When either meter is turned on, it will perform an initialself-diagnostic test by showing all available readouts on its digitaldisplay. Following the initial test, the meter will display the mag-netic field intensity at the location where it is held or placed andthe intensity will change if moved accordingly. If the negative signis still showing after the initial test, that indicates that the meter isrunning low on power and the battery needs to be replaced.

PERFORMING MEASUREMENTSMeasurements of the magnetic field in the home are general-

ly taken in the middle of the room at about one metre from the groundor in locations where people spend a significant amount of time, forexample, the bed. Measurements should also be performed severaltimes over the course of a day. This is to allow for possible variationsto electricity demand which presumably would peak during theevening at about 7.00 pm. Measurements can also be made at anyother locations of interest.

It is important to remember that, as mentioned earlier,research suggests that if any health effects exist, they are associat-ed with prolonged magnetic field exposure. Measurements takenwith the gauss meter are instantaneous (i.e. measured at one pointin time) and do not accurately reflect prolonged exposure levels.

TYPICAL MAGNETIC FIELD STRENGTHSMagnetic fields within homes can vary at different locations

and also over time. The actual strength of the field at a given loca-tion depends upon the number and kinds of sources and their dis-tance from the location of measurement. Typical values measuredin areas away from electrical appliances are of the order of 2 mG.

Magnetic fields from individual appliances can vary con-siderably as well, depending on the way they were designed andmanufactured. One brand of hair dryer, for example, may gener-ate a stronger magnetic field than another. In general, appliances,which use a high current (such as those which have an electricmotor) will lead to relatively high readings. It should also benoted that different body parts will be exposed to different mag-netic field levels from the same appliance, depending on how farthat part of the body is from the appliance when in use. Typicalvalues of magnetic fields measured at normal user distance fromsome common domestic electrical appliances are listed in thefollowing table.

MEASURING MAGNETIC FIELDSELECTRIC AND MAGNETIC FIELDSAustralian Radiation Protection and Nuclear Agency


HOMES NEAR POWER LINESThe power lines that are present in typical neighbour-

hoods are called “distribution” lines and they usually carry lessvoltage than “transmission” lines, which carry very high voltages.As stated earlier, however, it is the current and not the voltage thatis associated with the strength of the magnetic field. Therefore,proximity to high voltage lines will not necessarily give a highreading unless those lines are also carrying a large current.Typical values of magnetic fields measured near power lines andsubstations are listed in the table below.

Electricity Testing and Measurement Handbook – Vol. 7 25

INTRODUCTION Our daily use of electricity is taken for granted, yet scien-

tific and public concern has arisen about possible health effectsfrom electric and magnetic fields (EMF) that are created by theuse of electricity. Because of this concern, the California PublicUtilities Commission authorized a statewide research, educationand technical assistance program on the health aspects of expo-sure to magnetic fields and asked the Department of HealthServices to manage it. Even though both electric and magneticfields are present with the use of electrical power, interest andresearch have focused on the effects of 50 and 60 Hertz (Hz)magnetic fields, called “power frequency” fields, from sourcessuch as power lines, appliances and wiring in buildings. This isbecause it is known that magnetic fields are difficult to shieldand because early scientific studies showed a possible relationshipbetween human exposure to certain magnetic field sources andincreased rates of cancer.

Even now, scientists are not sure if there are health risksfrom exposure to 50 and 60 Hz magnetic fields, or if so what isa “safe” or “unsafe” level of exposure. People frequently ask aboutEMF risk when they are choosing where to live. This choiceshould include consideration of proven risks of the location, suchas the possibility of earthquake, flooding, or fire, or the presenceof traffic, radon, or air pollution. To some people even limitedevidence for a possible EMF risk weighs heavily in their deci-sions. For others, different considerations take precedence. Therereally is no one right answer to these questions because each sit-uation is unique.

The California EMF Program developed this fact sheet togive an overview of the present state of knowledge and providea basis for understanding the current limitations on the ability ofscience to resolve questions about the possible health risks ofmagnetic field exposure. This paper describes electric and magneticfields, high field sources and how to interpret field measurementsonce they are made. It includes discussions of the controversy aboutpossible health effects, as well as current California state policy andwhat the government is doing to address public concern.

WHAT ARE ELECTRIC AND MAGNETIC FIELDS OR “EMF”Before man-made electricity, humans were exposed only

to the magnetic field of the earth, electric fields caused bycharges in the clouds or by the static electricity of two objectsrubbing together, or the sudden electric and magnetic fieldscaused by lightning. Since the advent of commercial electricityin the last century we have been increasingly surrounded byman-made EMF generated by our power grid (composed of pow-erlines, other electrical equipment, electrical wiring in buildings,power tools, and appliances) as well as by higher frequencysources such as radio and television waves and, more recently,cellular telephone antennas.

EMF: INVISIBLE LINES OF FORCE Wherever there is electricity, there are also electric and

magnetic fields, invisible lines of force created by the electriccharges. Electric fields result from the strength of the chargewhile magnetic fields result from the motion of the charge, or thecurrent. Electric fields are easily shielded: they may be weak-ened, distorted or blocked by conducting objects such as earth,trees, and buildings, but magnetic fields are not as readilyblocked. Electric charges with opposite signs (positive and neg-ative) attract each other, while charges with the same sign repeleach other. The forces of attraction and repulsion create electricfields whose strength is related to “voltage” (electrical pressure).These forces of attraction or repulsion are carried through spacefrom charge to charge by the electric field. The electric field ismeasured in volts per meter (V/m) or in kilovolts per meter(kV/m). A group of charges moving in the same direction iscalled an “electric current.” When charges move they createadditional forces known as a “magnetic field.” The strength of amagnetic field is measured in “gauss” (G) or “tesla” (T), whilethe electric current is measured in “amperes” (amps). Thestrength of both electric and magnetic fields decrease as onemoves away from the source of these fields.

FIELDS VARY IN TIME An important feature of electric and magnetic fields is the

way they vary in time. Fields that are steady with respect todirection, rate of flow, and strength are called “direct current”(DC) fields. Others, called “alternating current” (AC) fields,change their direction, rate of flow, and strength regularly overtime. The magnetic field of the earth is DC because it changes solittle in one year that it can be considered constant. However, themost commonly used type of electricity found in power lines andin our homes and work places is the AC field. AC current doesnot flow steadily in one direction, but moves back and forth. Inthe U.S. electrical distribution system it reverses direction 120times per second or “cycles” 60 times per second (the directionreverses twice in one complete cycle). The rate at which the ACcurrent flow changes direction is expressed in “cycles per sec-ond” or “Hertz” (Hz). The power systems in the Untied Statesoperate at 60 Hz, while 50 Hz is commonplace elsewhere. Thisfact sheet focuses on “power frequency” 60 Hz fields and not thehigher frequency fields generated by sources such as cellularphone antennas.

DESCRIBING MAGNETIC FIELDS The concentration of a chemical in water can be described

by citing a single number. Unlike chemicals, alternating electricand magnetic fields have wave-like properties and can bedescribed in several different ways, like sound. A sound can beloud or soft (strength), high or low-pitched (frequency), have

ELECTRIC AND MAGNETIC FIELDSMEASUREMENTS AND POSSIBLE EFFECT ON HUMAN HEALTH

— WHAT WE KNOW AND WHAT WE DON’T KNOWCalifornia Department of Health Services and the Public Health Institute

California Electric and Magnetic Fields Program

26 Electricity Testing and Measurement Handbook – Vol. 7

periods of sudden loudness or a constant tone, and can be pure orjarring. Similarly, magnetic fields can be strong or weak, be ofhigh frequency (radio waves) or low frequency (powerline waves),have sudden increases (“transients”) or a constant strength, consistof one pure frequency or a single dominant frequency with somedistortion of other higher frequencies (“harmonics”). It is alsoimportant to describe the direction of magnetic fields in relation tothe flow of current. For instance, if a magnetic field oscillates backand forth in a line it is “linearly polarized.” It may also be impor-tant to describe how a field’s direction relates to other physicalconditions such as the earth’s static magnetic fields.

MEASURING MAGNETIC FIELDS AND IDENTIFYING THESOURCES OF ELEVATED FIELDS

MEASURING MAGNETIC FIELD STRENGTH The strength or intensity of magnetic fields is commonly

measured in a unit called a Gauss or Tesla by magnetic fieldmeters called “gaussmeters.” A milligauss (mG) is a thousandthof a gauss, and a microtesla (uT) is a millionth of a tesla (onemilligauss is the same as 0.1 microtesla). The magnetic fieldstrength in the middle of a typical living room measures about0.7 milligauss or 0.07 microtesla. As noted above, the strength ofthe magnetic field is only one component of the mixture thatcharacterizes the field in a particular area. Measuring only mag-netic field strength may not capture all the relevant informationany more than the decibel volume of the music you are playingcaptures the music’s full impact. The main health studies to datehave only measured magnetic field strength directly or indirect-ly and assessed its association with disease. Some scientists won-der if the weak association between measured magnetic fieldsand cancer in these studies might appear stronger if we knewwhich aspect of the EMF mixture to measure. Other scientistswonder if any such aspect exists.

WHERE ARE WE EXPOSED TO 60 HZ EMF? There are “power frequency” electric and magnetic fields

almost everywhere we go because 60 Hz electric power is sowidely used. Exposure to magnetic fields comes from manysources, like high voltage “transmission” lines (usually on metaltowers) carrying electricity from generating plants to communitiesand “distribution” lines (usually on wooden poles) bringing elec-tricity to our homes, schools, and work places. Other sources ofexposure are internal wiring in buildings, currents in groundingpaths (where low voltage electricity returns to the system inplumbing pipes), and electric appliances such as TV monitors,radios, hair dryers and electric blankets. Sources with high voltageproduce strong electric fields, while sources with strong currentsproduce strong magnetic fields. The strength of both electric andmagnetic fields weakens with increasing distance from the source(table 1). Magnetic field strength falls off more rapidly with distancefrom “point” sources such as appliances than from “line” sources(power lines). The magnetic field is down to “background” level(supposed to be no greater than that found in nature) 3-4 feet froman appliance, while it reaches background level around 60-200 feetfrom a distribution line and 300-1000 feet from a transmission line.Fields and currents that occur at the same place can interact tostrengthen or weaken the total effect. Hence, the strength of thefields depends not only on the distance of the source but also the dis-tance and location of other nearby sources.

IDENTIFYING SOURCES OF ELEVATED MAGNETIC FIELDS Sometimes fairly simple measurements can identify the

external or internal sources creating elevated magnetic fields.For example, turning off the main power switch of the house canrule out sources from use of power indoors. Magnetic field meas-urements made at different distances from power lines can helppinpoint them as sources of elevated residential magnetic fields.Often, however, it takes some detective work to find the majorsources of elevated magnetic fields in or near a home. Currentsin grounding paths (where low voltage electricity returns to thesystem in plumbing pipes) and some common wiring errors canlead to situations in which source identification is difficult andrequires a trained technician. It is almost always possible to findand correct the sources of elevated magnetic fields when they aredue to faulty electrical wiring, grounding problems, or appli-ances such as lighting fixtures.

60 HZ MAGNETIC FIELD EXPOSURE DURING A TYPICAL DAY Exposure assessment studies of adults who wore meas-

urement meters for a 24- to 48-hour period suggest that the aver-age magnetic field level encountered during a typical 24 hours isabout 1 mG. About 40% of magnetic field exposures found inhomes come from nearby power lines, while 60% come fromother sources such as stray currents running back to the electri-cal system through the grounding on plumbing and cables, cur-rent “loops” due to incorrect internal wiring in the home, andbrief exposure to appliances and electrical tools.

Table 1. Examples of magnetic field strengths at particular distances from appliance surfaces.


MAGNETIC FIELD SURVEY OF HOMES IN THE SAN FRANCISCO BAY AREA The California Department of Health Services surveyed

homes in the San Francisco Bay Area in the mid-1990s. In thisstudy, magnetic field measurements were taken in the middle ofthe bedroom, family room and kitchen and at the front door ofthese homes under normal power conditions (any appliances orelectrical devices turned on at the onset of the measurement periodwere left on). As shown in table 2, about half the houses in theBay Area had an average level below 0.71 mG and 90 percenthad average levels below 1.58 mG.

MAGNETIC FIELDS GENERATED BY CURRENT FLOWING THROUGH WIRES CANBE REDUCED

Two wires with current flowing in opposite directions createmagnetic fields going in opposite directions. If the wires areplaced close together and have currents of similar magnitude themagnetic fields cancel each other. This principle is often used tolower magnetic fields. For example, an underground distributioncable has a “hot” line (carrying current to the user) and a “neutral”line (carrying it away) that generate low magnetic fields when theyare placed close together. The underground cables can be placedclose together because it is possible to insulate them heavily to pre-vent arcing. Overhead power lines cannot be placed this closetogether because of the weight of the needed insulation and theneed for worker safety. For most distribution and transmissionlines, however, California utilities use three-wire or four-wire sys-tems. The current in these lines alternates in strength and directionin slightly different phases (not alternating completely together). Itis sometimes possible to optimize these phase differences so thatthe magnetic fields from the wires cancel each other.

WHAT CAN WE SAY ABOUT A MEASUREMENT ONCE WE HAVE IT? A concerned person would like to know if the measure-

ments found in his or her home are “safe” or “unsafe.” Right now,most scientists do not feel that the data are solid enough to makepredictions about the health risks of magnetic field strength.When magnetic field exposure (or its estimate) increases there isno evident orderly increase of a health risk. The highest level ofmagnetic field strength measured in homes is below the intensityfound in almost all the cellular and animal experiments that haveproduced subtle biological effects. This makes scientists and pol-icy makers reluctant to set health-based standards for magneticfield exposures. However, it is possible to find out how measure-ments in your home compare to other homes and if these meas-urements are “typical” or not. The information in tables 1 and 2may be helpful in deciding if your home is typical.

DOSE-RESPONSE RELATIONSHIP A special problem in the study of health effects of envi-

ronmental factors is how to measure exposure in a way that ade-quately reflects the true amount of the person’s exposure to thesubstance being studied. This true amount is called the “dose.”With cigarette smoke and toxic chemicals, there is a positiverelationship between the size (or strength) of the dose and theadverse health effect it produces: the higher the dose, the greaterthe effect. With magnetic fields, however, some laboratory evi-dence suggests that this is not always the case, and very confus-ing relationships have been seen. Biological effects or changesappear at strengths of certain levels, disappear at higher levels,only to appear again at still higher levels. Varying the frequency(speed of alternation), for example from 60 Hz to 120 Hz, showssimilar “effect windows” of magnetic fields. To complicatethings further, some laboratory experiments have shown aneffect with intermittent (“pulsed”) exposures, others with“spikes” or transients, and still others with continuous exposure.There is some evidence that the orientation of alternating fieldsin relation to the direction of the earth’s static magnetic field isalso important in making a biological effect. Generally, theeffects observed are only biological changes that may or may nottranslate into true health effects.

LIMITATIONS OF DIRECT MAGNETIC FIELD MEASUREMENTS Those human health studies investigating the relationship

of magnetic field exposure and cancer measured magnetic fieldsusing one-time, short-term measures (i.e., for 24 hours) of onearea such as the bedroom, or one-time spot measurements (i.e.,for one minute) in several different rooms of the participants’homes. It was assumed that these home measurements adequate-ly estimate a person’s total exposure. However, these measurescan not be used to assess the biological importance of the lengthof exposure, the number of times there are high exposures, or thepresence of other components of the field such as harmonics.Also, field intensity (strength) varies at different times of dayand different seasons, depending on electricity use. Dinnertimereadings are often higher than readings in the middle of the night.In addition, an area measure may not reflect a personal exposurethat is dependent on the amount of time a person spends in thearea measured.

CONTROVERSY ABOUT POSSIBLE HEALTH EFFECTS The controversy about EMF health effects derives from:

1) the fact that many scientists believe power line magneticfields emit little energy and are therefore too weak to have anyeffect on cells; 2) the inconclusive nature of laboratory experi-ments; and 3) the fact that epidemiological studies of peopleexposed to high EMF are inconclusive.

1. WEAK FIELDS MAY HAVE TOO LITTLE ENERGY TO CAUSE BIOLOGICAL EFFECTSThe electromagnetic spectrum covers a large range of fre-

quencies (expressed in cycles per second or Hertz). The higherthe frequency, the greater the amount of energy in the field. X-rays have very high frequencies, and are able to ionize moleculesand break chemical bonds, which damages genetic material andcan eventually result in cancer and other health disorders. Highfrequency microwave fields have less energy than x-rays, butstill enough to be absorbed by water in body tissues, heatingthem and possibly resulting in burns. Radio frequency fieldsfrom radio and TV transmitters are another step weaker than micro-waves. Although they alternate millions of times per second, they

Table 2. Distribution of average magnetic field strength of San Francisco Bay Area homes.


can’t ionize molecules and can only heat tissues close to thetransmitter. Electric power fields (50 and 60 Hz) have muchlower frequencies than even radio waves and hence emit verylow energy levels that do not cause heating or breakage of bonds.They do create electrical currents in the body, but in most casesthese currents are much weaker than those normally existing inliving organisms. For these reasons, many scientists argue that itis unlikely that 60 Hz power frequency magnetic fields at thestrengths commonly found in the environment have any physicalor biological effects on the body.

2. INCONSISTENT LABORATORY RESULTSAs stated above, 60 Hz power frequency magnetic fields do

create weak electric currents in the bodies of people and animals.In the mid-1970s a variety of laboratory studies in cell cultures andwhole animals demonstrated that these fields produce biologicalchanges when applied in intensities of hundreds or thousands ofmilligauss. Some scientists observed effects at lower strengths, butaverage daily personal exposure is only about 1 mG. Biologicaleffects that seem to be attributable to magnetic fields are subtle anddifficult to reproduce. These studies are continuing in an effort tounderstand how magnetic fields affect living tissue. Some labora-tory scientists have found that magnetic fields can producechanges in the levels of specific chemicals the human body makes(such as the hormone melatonin), as well as changes in the func-tioning of nerve cells and nervous systems of other animals.However, the jury is still out as to whether this type of change canlead to any increased risk to human health.

In the mid-1990s, scientists conducted a series of EMfanimal studies. Most of these studies showed little or no associ-ation between EMF and cancer or adverse reproductive effects.This convinced some scientists that EMF’s were harmless.However, others pointed out that the animals’ EMF exposures inthese studies might not adequately capture some aspect of EMFexposure that could have biological effects on humans.

3. INCONCLUSIVE EPIDEMIOLOGICAL STUDIES Epidemiology examines the health of groups of people,

and epidemiological studies make statistical comparisons abouthow often diseases occur in “exposed” and “nonexposed”groups. Studies in which the disease rate is higher for theexposed group than nonexposed (said to have “positive” results)do not necessarily show a direct cause for disease, but ratherindicate that there is some sort of relationship between exposureand disease. Most epidemiological studies of magnetic fieldshave been of two types. One kind focused on children with cancerto see whether their home magnetic field measurements werehigher or if they were more likely to live in homes with overheadpowerlines carrying high current than a comparable group ofchildren without cancer. The other type of study looked at ratesof death and disease of adults assumed to be heavily exposed tomagnetic fields at work, with exposure often indirectly assessedby using job titles, to determine if their rates were higher thanadults assumed to be working in low magnetic field environments.

CHILDHOOD CANCER STUDIES Public concern has arisen because of media reports about

epidemiological studies that showed an association betweenchildhood cancer and proximity to high current-carrying over-head power lines. In 1996, a special committee of the NationalResearch Council (NRC) made a careful review of 11 epidemio-logical studies examining the relationship between childhood

leukemia and residential proximity to this type of power lines.1

For these studies, a child’s exposure to magnetic fields was esti-mated three ways. First, the type and proximity of power lines(“wire codes”) near the child’s home was assessed. Those houseswith lines nearby with the potential to carry high current wereclassified as “high current configuration” and were assumed tohave higher magnetic field levels (due to higher current) thanhouses near lower current configuration power lines (figure 1).Second, exposure was estimated by measurements of magneticfields taken in the child’s home at the time of the study – oftenmany years after diagnosis of their cancer. And third, exposurewas approximated by estimating what the home magnetic fieldlevels were right after the children were diagnosed, using linedistance from the house and past utility records of current flowin the lines during the appropriate time period.

The NRC made a statistical summary and comparison ofthese eleven studies. They concluded that children living in highcurrent configuration houses are 1.5 times as likely to developchildhood leukemia than children in other homes. Despite thisconclusion, the NRC was a unable to explain this elevated riskand recommended that more research be done to help clarify theissue. One reason for this uncertainty is that wire-code classifi-cation assumes that houses with high wire-codes have highermagnetic field levels than low wire-code houses, but high wire-codes may also be a proxy for some type of exposure besidesmagnetic fields that is not yet understood. For example, highwire-code houses tend to have higher traffic density nearby,resulting in higher air pollution levels. However, traffic densityseems to be an unlikely explanation for the wire-code associationfound in these studies.

In 1997, the NRC statement seemed to be contradicted bythe findings of Dr. M. S. Linet of the National Cancer Institute ina large epidemiological study1i. Her researchers estimated expo-sure to magnetic fields in two ways, wire-codes as defined above(based on distance of different types of power lines near thehome) and home area measurements. The study found no associ-ation between living in high wire-code houses and childhoodleukemia. On the other hand, the study found that children livingin houses with high average magnetic field levels did have higherrates of cancer in general.

Figure 1. Summary of results of power line distance(“wire code”) and childhood leukemia studies.


THE EMF RAPID PROGRAM WORKING GROUP STATEMENT ON CHILDHOOD LEUKEMIAIn 1998, a working group of experts gathered by the feder-

al EMF RAPID program (see “Governmental Regulation,” below)reviewed the research on the possible health risks associated withEMF. A majority felt that the epidemiology studies of childhoodleukemia provide enough evidence to classify EMF as a “possiblehuman carcinogen,” meaning they think it might cause cancer.This does not mean that it definitely causes cancer, however. Theworking group’s findings are published in a report posted on theprogram’s Web site (see address below).

IF REAL, HOW IMPORTANT WOULD THIS RISK OF CHILDHOOD LEUKEMIA BE? Each year an average of six cases of leukemia are diag-

nosed per 100,000 children. Six percent of American houses arenear high-current-carrying power lines.2 If the epidemiologicalassociation is correct that means that in such houses there wouldbe three additional cases of leukemia among 100,000 childrendue to the effects of EMF from the nearby power lines. (This isalmost the increased risk of lung cancer of an adult nonsmokerwho lives in a smoking household.) Among the 500,000 childrenin California who live nearest high-current-carrying power linesthere could be a theoretical 15 extra cases of leukemia each yearcompared to the number of cases if they lived further away. InCalifornia, we regulate chemicals whose typical exposures gen-erate a theoretical life-time risk of one per 100,000. An addedrisk of three sick children per 100,000 per year is larger than this.From an individual’s point of view, this risk, if real, would besmall: 99,991 out of 100,000 children would not get leukemiaeach year.

OCCUPATIONAL STUDIES The occupational studies looking at magnetic field expo-

sure and various health outcomes show mixed results.Occupations assumed to have higher than normal magnetic fieldlevels included electricians, telephone linemen, electric welders,electronic technicians, utility workers, electrical engineers andsewing machine operators. In general, but not always, workers ofthese occupations were more likely to have higher rates of braintumors, leukemia, testicular tumors and male breast cancer thanexpected. A particular brain tumor (astrocytoma) occurred moreoften among men who worked for many years in jobs with highestimated exposure levels such as electricians, linemen, and elec-trical engineers.3 A large study of Canadian and French utilityworkers found an association between estimated high magneticfield exposures based on area measures of certain occupationsand myeloid leukemia, a rare type of blood cancer.4 On the otherhand, another large study found no increase in mortality frombrain tumors, leukemia or other cancers among electrical work-ers with estimated high magnetic field exposure over manyyears.5 Differences among study results may exist simplybecause the studies used different study populations and methodsfor estimating high occupational magnetic field exposure. Also,these surrogate measures estimating high occupational magneticfield levels could be proxies for other types of exposure at workbesides magnetic fields.

COMPARING THE SCIENTIFIC EVIDENCE ON MAGNETIC FIELDS TO THAT OFENVIRONMENTAL TOBACCO SMOKE

There are regulations in place protecting us from environ-mental tobacco smoke. They are based on the strength of its asso-ciation with disease and the consistent epidemiological evidencefor it. What’s the difference between this evidence and that for

magnetic fields? First, no magnetic field epidemiological studyhas found an association with disease that is as strong as thatimplicating a two-pack-a-day smoking habit. The strength of theassociation found for leukemia in electric train engineers, whoare exposed to magnetic fields of hundreds of milligauss all daylong, is no stronger than the strength of the association relatingresidential magnetic field levels (generally less than 10 mG) tochildhood leukemia. Second, there is no laboratory evidenceabout magnetic field exposure that is as convincing as that forlung cancer and smoking— magnetic field animal studies havebeen inconsistent. These differences make scientists much morecautious about interpreting the magnetic field epidemiology asdangerous than the environmental tobacco smoke epidemiology.

GOVERNMENTAL REGULATION

STATE REGULATIONS Lack of understanding has kept scientists from recom-

mending any health-based regulations. Despite this, several stateshave adopted regulations governing transmission line-generatedmagnetic fields at the edge of the “right-of-way” (“ROW,” the areaimmediately surrounding power lines left clear for access formaintenance and repairs) because of concern about the risk ofelectric shock from strong electric fields present in these areas(table 3). All current regulations relate to transmission lines; nonegovern distribution lines, substations, appliances or other sourcesof electric and magnetic fields.

The California Department of Education requires minimumdistances between new schools and the edge of transmission linerights-of-way. The setback guidelines are: 100 feet for 50-133 kVlines, 150 feet for 220-230 kV lines, and 350 feet for 500-550 kVlines. Once again, these were not based on specific biologicalevidence, but on the rationale that the electric field drops tobackground levels at the specified distances.

The California Public Utilities Commission (CPUC),upon the recommendation of a Consensus Group composed ofcitizens, utility representatives, union representatives, and publicofficials, recommended that the state’s investor-owned utilitiescarry out “no and low cost EMF avoidance measures” in con-struction of new and upgraded utility projects. This means that4% of the total project cost is allocated to mitigation measures ifthese measures will reduce magnetic field strength by at least15%. The strategy is to address public concern and cope with

Table 3. Transmission line EMF standards and guide-lines adopted by certain states forutilities’ rights-of-way (ROW).


potential but uncertain risks until a policy based on scientific factcan be developed. The CPUC also followed the ConsensusGroup’s recommendation to establish the research, education andtechnical assistance programs of the California EMF Programunder the guidance of the California Department of HealthServices. It is expected to provide information that will be usefulto those responsible for making public policy in the future.

FEDERAL EFFORTS At the Federal level, the Federal Energy Policy Act of 1992

included a five-year program of electric and magnetic field (EMF)Research and Public Information Dissemination (EMF-RAPID).The EMF-RAPID Program asked these questions: Does exposureto EMF produced by power generation, transmission, and use ofelectric energy pose a risk to human health? If so, how significantis the risk, who is at risk, and how can the risk be reduced?

In 1998, a working group of experts gathered by theEMF-RAPID Program met to review the research that has beendone on the possible health risks associated with EMF. Thisgroup reviewed all of the studies that have been done on the sub-ject, and then voted on whether they believed that exposure toEMF might be a health risk. They then published a reportdescribing their findings. A majority of the scientists on thisworking group voted that the epidemiology studies of childhoodleukemia and residential EMF exposures provide enough evi-dence to classify EMF as a “possible human carcinogen.”6 Thismeans that, based on the evidence, these researchers believe thatit is possible that EMF causes childhood leukemia, but they arenot sure. About half of the group’s members thought that there isalso some evidence that workplace exposure to EMF is associat-ed with chronic lymphocytic leukemia in adults. The group alsoconcluded that there was not enough evidence to determinewhether EMF exposure might cause other diseases.6

The EMF-RAPID Program released its final report toCongress in 1999. This report explains the program’s findings,including the results of its working group and many researchprojects. The final report states that “the NIEHS believes thatthere is weak evidence for possible health effects from [powerfrequency] ELF-EMF exposures, and until stronger evidencechanges this opinion, inexpensive and safe reductions should beencouraged.”7 (page 38) The report specifically suggests educat-ing power companies and individuals about ways to reduce EMFexposure, and encouraging companies to reduce the fields creat-ed by appliances that they make, when they can do so inexpen-sively7 (page 38). For more information on the EMF-RAPIDprogram or to look at these reports, contact the EMF-RAPIDProgram, National Institute of Environmental Health Sciences,National Institutes of Health, P.O. Box 12233, Research TrianglePark, North Carolina27709, or visit their Web site athttp://www.niehs.nih.gov/ emfrapid. When ordering a copy ofthe final report, refer to NIH publication number 99-4493.

CONCLUSION Public concern about possible health hazards from the

delivery and use of electric power is based on data that givecause for concern, but which are still incomplete and inconclu-sive and in some cases contradictory. A good deal of research isunderway to resolve these questions and uncertainties. Until wehave more information, you can use “no and low cost avoidance”by limiting exposure when this can be done at reasonable costand with reasonable effort, like moving an electric clock a fewfeet away from a bedside table or sitting further away from the

computer monitor. Table 1 shows how quickly fields fall off asone moves away from appliances – they virtually disappear at 3-5 feet. You might stop using an electric appliance you do not real-ly need. You may also consider home testing, which can identifyfaulty electrical wiring that can produce shock hazards and cur-rent code violations as well as elevated magnetic fields. InCalifornia, the investor-owned utilities are required by the CPUCto provide magnetic field measurement at no charge to their cus-tomers. So far, in the absence of conclusive scientific evidence,there is no sufficient basis for enacting laws or regulations to limitpeople’s exposure to EMF, so it is up to individuals to decide whatavoidance measures to take, based on the information available.

REFERENCES 1. a) Wertheimer N et al. Electrical wiring configurations and

childhood cancer. American Journal of Epidemiology.1979; 109:273-84.

b) Fulton JP et al. Electrical wiring configurations and child-hood leukemia in Rhode Island. American Journal ofEpidemiology. 1979; 111:292-96.

c) Savitz DA et al. Case control study of childhood cancer andexposure to 60-Hz magnetic fields. American Journal ofEpidemiology. 1988; 128:21-38.

d) Coleman M et al. Leukaemia and residence near electricitytransmission equipment: A case-control study. BritishJournal of Cancer. 1989; 60:793-98.

e) London SJ et al. Exposure to residential electric and mag-netic fields and risk of childhood leukemia. AmericanJournal of Epidemiology. 1991; 134:923-37.

f) Feychting M. et al. Magnetic fields and cancer in childrenresiding near Swedish high-voltage power lines. AmericanJournal of Epidemiology. 1993; 138:467-81.

g) Fajardo-Gutierrez AJ et al. Residence close to high-tensionelectric power lines and its association with leukemia in chil-dren (Spanish). Biol Med Hosp Infant Mex. 1993; 50:32-38.

h) Petridou ED et al. Age of exposure to infections and risk ofchildhood leukaemia. British Medical Journal. 1993; 307:774.

i) Linet MS et al. Residential exposure to magnetic fields andacute lymphoblastic leukemia in children. New EnglandJournal of Medicine. 1997; 337:1-7.

2. Zaffanella L. Survey of residential magnetic sources. EPRIFinal Report. 1993; No. TR 102759-v1. No. TR 102759-v2.

3. Savitz DA et al. Magnetic field exposure in relation toleukemiaand brain cancer mortality and electric utility workers.American Journal of Epidemiology. 1995; 141: 1-12.

4. Theriault G et al. Cancer risk associated with occupationalex-posure to magnetic fields among utility workers in Ontario andQuebec, Canada and France. American Journal of Epidemiology.1994; 139: 550-572.

5. Sahl JD et al. Cohort and nested case-control studies ofhematopoietic cancers and brain cancer among electric utilityworkers. Epidemiology. 1993; 4: 104-114.

6. National Institute of Environmental Health Sciences.Assessment of health effects from exposure to power-line fre-quency electric and magnetic fields. NIEH Working GroupReport. 1998.

7. National Institute of Environmental Health Sciences. Healtheffects from exposure to power-line frequency electric and mag-netic fields. NIEH Final Report ot Congress. 1998.


Engineers and technicians often need to make “floating”measurements where neither point of the measurement is atground (earth) potential. This measurement is often referred to asa differential measurement. “Signal common” may be elevated tohundreds of volts from earth.

In addition, many of these differential measurementsrequire the rejection of high common-mode signals*1 in order toevaluate low-level differential signals. Unwanted ground currentscan also add bothersome hum and ground loops. Too often, usersresort to the use of potentially dangerous measurement techniquesto overcome these problems.

The TPS2000 Series oscilloscopes use innovative IsolatedChannel technology to deliver the world’s first 4-isolated-channel,battery-operated oscilloscope to allow engineers and techniciansto make multi-channel isolated measurements quickly, accuratelyand affordably – all designed with your safety in mind.

FLOATING AN OSCILLOSCOPE: A DEFINITION“Floating” a ground-referenced oscilloscope is the tech-

nique of defeating the oscilloscope’s protective grounding system –disconnecting “signal common” from earth, by either defeating thegrounding system or using an isolation transformer. This techniqueallows accessible parts of the instrument such as chassis, cabinet,and connectors to assume the potential of the probe ground leadconnection point. This technique is dangerous, not only from thestandpoint of elevated voltages present on the oscilloscope (ashock hazard to the operator), but also due to cumulative stress-es on the oscilloscope’s power transformer insulation. This stressmay not cause immediate failure, but may lead to future danger-ous failures (a shock and fire hazard), even after returning theoscilloscope to properly grounded operation.

Not only is floating a ground-referenced oscilloscope dan-gerous, but the measurements are often inaccurate. This potentialinaccuracy results from the total capacitance of the oscilloscopechassis being directly connected to the circuit-under-test at thepoint where the ground lead is connected.

A GUIDE TO MAKING QUICK, ACCURATE ANDAFFORDABLE FLOATING MEASUREMENTS

There are several products that enable you to make float-ing measurements, but they may lack the versatility, accuracy oraffordability that you need. In addition, there are four key meas-urement considerations that a user needs to take into accountwhen selecting the right product to make an accurate floating ordifferential measurement:

1 – What is the differential measurement range?2 – What is the common mode measurement range?3 – What are the loading characteristics of the probe?

Are they balanced or unbalanced?4 – What is the Common Mode Rejection Ratio (CMRR)

over the measurement frequency range?

TRADITIONAL OSCILLOSCOPESTraditional oscilloscopes are limited to making ground-

referenced measurements. Let’s examine why:Most oscilloscopes have their “signal common” terminal

connected to the protective grounding system, commonlyreferred to as “earth” ground or just “ground”. This is done sothat all signals applied to, or supplied from, the oscilloscopehave a common connection point. This common connectionpoint is usually the oscilloscope chassis and is held at (or verynear to) zero volts by virtue of the third-wire ground in the powercord for AC-powered equipment. It also means that, with fewexceptions, all measurements must be made with respect to earthground. This constrains the typical oscilloscope (at least in a singlemeasurement) from being used to measure potential differencesbetween two points where neither point is at earth ground.

A common, but risky, practice is to disconnect the oscillo-scope’s AC main power cord ground and attach the probe groundlead to one of the test points. Tektronix strongly recommendsagainst this unsafe measurement practice. Unfortunately, thispractice puts the instrument chassis, which is no longer groundedto earth, at the same voltage as the test point that the probeground lead is connected to. The user touching the instrument

A NEW APPROACH TO QUICK, ACCURATE,AFFORDABLE FLOATING MEASUREMENTS

Tektronix IsolatedChannel Technology

*1 A “common-mode signal” is defined as a signal that is present at both pointsin a circuit. Typically referenced to ground, it is identical in amplitude, frequency,and phase. Making a floating measurement between two points requires rejectingthe “common-mode signal” so the difference signal can be displayed.

Management and Safety in the Workplace

While the subject of this technical note is floating measurements, some defini-tions of terms and general precautions must be understood before proceeding.Historically, floating measurements have been made by knowingly defeating thebuilt-in safety ground features of oscilloscopes or measurement instruments invarious manners.

THIS IS AN UNSAFE AND DANGEROUS PRACTICE AND SHOULD NEVER BE DONE!

Instead, this technical note describes instruments, accessories, and practices thatcan make these measurements safely as long as standard safety practices andprecautions are observed.

When making measurements on instruments or circuits that are capable of deliv-ering dangerously high-voltage, high-current power, measurement techniciansshould always treat exposed circuits, bus-bars, etc., as being potentially “live,”even when circuits have been shut off or disconnected. This is particularly truewhen connecting or disconnecting probes or test leads.


becomes the shortest path to earth ground. Figure 1 illustratesthis dangerous situation. V1 is the “offset” voltage above trueground, and VMeas is the voltage to be measured.

Depending upon the unit-under-test (UUT), V1 may behundreds of volts, while VMeas might be a fraction of a volt.

Floating the chassis ground in this manner threatens theuser, the UUT, and the instrument. In addition, it violates indus-trial health and safety regulations, and yields poor measurementresults. Moreover, line-powered instruments exhibit a large par-asitic capacitance when floated above earth ground. As a result,floating measurements will be corrupted by ringing, as shown inFigure 2.

Battery-operated oscilloscopes, such as the TDS3000BSeries oscilloscopes, when operated from AC line power using astandard power cord, exhibit the same limitations as traditionaloscilloscopes. However, AC power is not always available whereyou want to make oscilloscope measurements. In the case of theTDS3000B Series oscilloscopes, the optional battery pack(TDS3BATB) allows you to operate the oscilloscope without theneed for AC power. However, it can only make safe floatingmeasurements up to 30 VRMS.

Traditional oscilloscopes emphasize performance (bandwidth,versatility), trading off the ability to make floating measurements.

DIFFERENTIAL OR ISOLATED PROBESDifferential or isolated probes offer a safe and reliable way

to adapt a grounded oscilloscope to make floating measurements.Neither of the two probe contacts need be at earth ground and theprobe system as a whole is isolated from the oscilloscope’s chassisground.

Differential probes offer a balanced impedance load to thedevice-under-test (DUT). However, they add a layer of cost and

complexity to the measurement apparatus. They may require anindependent power supply, and their gain and offset characteristicsmust be factored into every measurement. Differential probe-equipped oscilloscopes emphasize performance and safety (band-width, isolation), trading off form-factor benefits such as portabilityand cost.

SIGNAL FIDELITY BEGINS AT THE PROBE TIP An oscilloscope is actually a measurement system consist-

ing of preamplifiers, acquisition/measurement circuits, displays,and probes. The role of the probe is sometimes overlooked.Nevertheless, improper probes or probing techniques can affectthe measurement outcome. Obviously, it’s essential to use compat-ible probes that match the instrument’s bandwidth and impedance.

Less understood is the effect of ground-lead inductance. Aslead length increases, parasitic inductance increases (Lparasitic inFigure A). Lparasitic is in the signal path and forms a resonant LCcircuit with the inherent parasitic capacitance of the oscilloscope(Cparasitic). As Lparasitic increases, the resonant frequencydecreases, causing “ringing” (see Figure 2) that visibly interfereswith the measured signal. Simply stated, the common lead must beas short as physical constraints of the circuit-under-test will allow.

In regard to capacitance, even isolated, battery-poweredoscilloscopes exhibit capacitance with respect to earth ground. InFigure A, Cparasitic describes the oscilloscope’s parasitic capaci-tance from its ground reference (through the isolated housing) toearth ground. Like parasitic inductance, Cparasitic must be kept to aminimum in order to force the resonant frequency of the LC circuitas high as possible. If Cparasitic is large, ringing may occur withinthe test frequency range, hampering the measurement.

An instrument’s parasitic capacitance to ground is dictat-ed by its internal design. The physical environment can alsoprompt ringing. Holding the instrument or placing it on a largeconductive surface during measurements can actually increaseCparasitic and lead to ringing. For extremely sensitive measure-ments, it might even be necessary to suspend the oscilloscope inmid-air!

A NEW APPROACH TO QUICK, ACCURATE, AFFORDABLEFLOATING MEASUREMENTS

The most common method of isolation in a wide bandwidthoscilloscope system in use today is a two-path approach in whichthe input signal is broken up into two signals: low frequency andhigh frequency. This approach requires expensive optocouplersand wideband linear transformers for each input channel.

The TPS2000 Series uses an innovative approach,Isolated Channel technology, which eliminates the two-pathmethod and uses only one wideband signal path for each inputchannel – from DC to the bandwidth of the oscilloscope. This

Figure 1: A floating measurement in which dangerous voltages occur on the oscilloscopechassis. V1 may be hundreds of volts.

Figure 2: Ringing caused by parasitic inductance and capacitance distorts the signal andinvalidates measurements

Figure 2: Parasitic inductance and capacitance can affect measurement quality


patent-pending technology enables Tektronix to offer the world’sfirst four-input Isolated Channel, low-cost, battery-operatedoscilloscope, featuring eight hours of continuous battery opera-tion. The TPS2000 Series oscilloscopes are ideal for engineersand technicians who need to make four-channel isolated measure-ments and need the performance and ease-ofuse of a low-cost,battery operated oscilloscope.

The TPS2000 Series’ four Isolated Channel input architec-ture provides true and complete channel-to-channel isolation forboth the “positive” input and the “negative reference” leads,including the external trigger input. Figure 3 illustrates the IsolatedChannel concept.

The most demanding floating measurement requirementsare found in power control circuits, such as motor controllers anduninterruptible power supplies, and industrial equipment. In suchapplication areas, voltages and currents may be large enough topresent a threat to users and test equipment.

Isolated Channel technology is the preferred solution formeasurement quality and is designed with your safety in mind.*2

The TPS2000 oscilloscopes offer an ideal solution when a largecommon mode signal is present. True channel-to-channel isolationminimizes parasitic effects; the smaller mass of the measurementsystem is less prone to interaction with the environment.

A properly isolated battery-powered instrument doesn’tconcern itself with earth ground. Each of its probes has a “NegativeReference” lead that is isolated from the instrument’s chassis, ratherthan a fixed ground lead. Moreover, the “Negative Reference” leadof each input channel is isolated from that of all other channels.This is the best insurance against dangerous short circuits. It alsominimizes the signal degrading impedance that hampers measure-ment quality in single-point grounded instruments.

The TPS2000 Series oscilloscope inputs are always float-ing whether operated from battery power or connected to ACpower through an AC power adapter. Thus, these oscilloscopesdo not exhibit the same limitations as traditional oscilloscopes.

SPEED DEBUG AND CHARACTERIZATION WITH DRTSAMPLING TECHNOLOGY (TIP)

The TPS2000 Series oscilloscopes offer digital real-time(DRT) acquisition technology that allows you to characterize awide range of signal types on up to four channels simultaneous-ly. Up to 2 GS/s real-time sample rate is the key to the extraordi-nary bandwidth – 200 MHz in the TPS2024. This bandwidth/sample rate combination makes it easy to capture the high-fre-

quency information, such as glitches and edge anomalies thateludes other oscilloscopes in its class, so that you can be sure to geta complete view of your signal to speed debug and characterization.

MAKING QUICK, ACCURATE FLOATING MEASUREMENTSWITH TPS2000 SERIES OSCILLOSCOPES

POWER CONTROL CIRCUITS:Power control technologies use both high-power silicon

components and low-power logic circuits. The switching transis-tors at the heart of most power control circuits require measure-ments not referenced to ground. Moreover, the power circuit mayhave a different ground point (and therefore a different groundlevel) than the logic circuit, yet the two often must be measuredsimultaneously.

The channel-to-channel isolation of the TPS2000 Seriesprovides a real-world measurement advantage in addition to itsobvious safety benefits. Figure 4 is a screen image depictingwaveforms taken at two different points in a power control cir-cuit. Notice that the lower waveforms are about 200 A p-p, whilethe upper trace is about 5 V p-p. Because each of the TPS chan-nels is fully isolated from the other (including the negative refer-ence leads), and equipped with its own uncompromised DigitalReal Time digitizer, there’s no cross-talk between the two sig-nals. Were the oscilloscope channels not adequately isolated,there might be misleading artifacts coupled from the 200 A sig-nal to the smaller waveform; these might be misinterpreted as acircuit problem when in reality it’s an instrument problem. Theability of the TPS Series to discretely capture two waveforms ofvastly differing amplitudes reduces guesswork and improvesproductivity.

HARMONICS MEASUREMENTS REVEAL UNSEEN POWER PROBLEMS

An understanding of the harmonics within a power grid isessential to the safe and cost-effective use of electrical power.Line harmonics are a growing problem in a world movingincreasingly toward nonlinear power supplies for most types ofelectronic equipment. Nonlinear loads, such as switching powersupplies, tend to draw non-sinusoidal currents. Their impedancevaries over the course of each cycle, creating sharp positive andnegative current peaks rather than the steady curve of a sine

Figure 3: TPS2000 Series oscilloscope’s Isolated Channel architecture provides completeisolation from dangerous voltages

*2 Do not float the P2220 probe common lead to > 30 VRMS. Use the P5120probe (floatable to 600 VRMS CAT II or 300 VRMS CAT III) or a similarly ratedpassive high-voltage probe, or an appropriately rated high-voltage differentialprobe when floating the common lead above 30 VRMS, subject to the ratings ofsuch high-voltage probe.

Figure 4: The 4-channel TPS2024oscilloscope’s channel-to-channel isolationeliminates cross-talk effects when large and small signals are captured simultaneously


wave. The rapid changes in impedance and current in turn affectthe voltage waveform on the power grid. As a result, the linevoltage is corrupted by harmonics; the normally sinusoidal shapeof the voltage waveform may be flattened or distorted.

There’s a limit to the amount of harmonic distortion thatequipment can tolerate. Load-induced harmonics can cause motorand transformer overheating, mechanical resonances, and danger-ously high currents in the neutral wires of three phase equipment.In addition, line distortions may violate regulatory standards insome countries.

The TPS2024’s comprehensive, four-channel capability,along with its optional power analysis software, enables connec-tion to all three conductors of a three-phase system to measureand analyze line harmonics. Its “Harmonics” mode – invokedwith a single button–captures the fundamental frequency plusharmonics 2 through 50. Using only the oscilloscope’s standardvoltage probe, it’s possible to execute a harmonic voltage meas-urement. An optional current probe acquires current harmonicswith the same ease.

Figure 5 illustrates a current harmonic measurement. Theamplitudes are computed by the instrument’s internal DFT(Discrete Fourier Transform) algorithm. In this case the bar graphreveals a very strong fifth harmonic level. Excessive fifth harmoniclevels (along with certain other odd harmonics) are a classic causeof neutral-wire currents in three-phase systems.

POWER READINGS – MORE THAN JUST WATTSVoltage and current measurements are by nature straight-

forward and absolute. A test point has only one voltage and onecurrent value at a given instant in time. In contrast, power meas-urements are voltage-, current-, time-, and phase-dependent.Terms like “reactive power” and “power factor,” which weredevised to characterize this complex interaction, are not so muchmeasurements as computations.

The power factor is of particular interest in these compu-tations. This is because many electrical power providers chargea premium to users whose power factor is not sufficiently closeto 1.0, the ideal value. At a power factor of 1.0, voltage and cur-rent are in phase. Inductive loads – especially large electricmotors and transformers – cause voltage and current to shift phaserelative to each other, reducing the power factor. Some utility com-panies apply a surcharge in such cases because the inefficiencycauses energy loss in the form of heat in the power lines. There areprocedures to remedy power factor problems, but first the powercharacteristics must be quantified.

The TPS Series embraces a full suite of power measure-ments. Among these are true power, reactive power, crest factor,phase relationships, di/dt and dv/dt, and of course power factor.Figures 6, 7 and 8 show TPS Series screen images summarizingthese and other power measurements. All of the measurements,with the exception of waveform analysis and phase relationships,

require a current probe (or itsequivalent) and a voltage probeworking in tandem. All of thesemeasurements employ the instru-ment’s one-button applicationfunction.

MEASURING SWITCHING LOSS TO IMPROVE PRODUCT EFFICIENCY

Today’s power designers face increasing pressure toimprove the efficiency of their power designs. A major factoraffecting the efficiency is the power loss occurring in the switchingsection of the design. Optimizing this factor can prove complex.

The TPS Series allows the designer to look at switchinglosses in their design through the instrument’s one-button appli-cation function. The switching loss will be characterized as turn-on loss, turn-off loss, conduction loss and total device loss.Figure 9 is a TPS Series screen image showing the switching lossmeasurements.

CONCLUSIONEngineers and technicians confront high voltages and cur-

rents and must often make potentially hazardous floating measure-ments. Where other alternatives may lack the versatility, accuracy oraffordability to make floating measurements, the TPS2000 Seriesemploys unique IsolatedChannel technology to allow engineers andtechnicians to make these measurements quickly, accurately andaffordably.

Figure 5: Harmonic distortion measurements

Figure 6: TPS Series’ instantaneouspower analysis

Figure 7: TPS Series’ waveformanalysis

Figure 8: TPS Series’ dv/dt anddi/dt cursors (dv/dt cursors shown)

Figure 9: TPS Series’ switching loss display showing turn-on, turn-off and conduction losses


OVERVIEW This tutorial is part of the NI Analog Resource Center.

Each tutorial will teach you a specific topic by explaining thetheory and giving practical examples. There are many issues toconsider when measuring high voltage. When specifying a dataacquisition system, the first question you should ask is whetheror not the system will be safe. Making high-voltage measure-ments can be hazardous to the equipment, to the unit under test,and to you and your colleagues. To ensure that the system is safe,you should provide an insulation barrier, using isolated measure-ment devices, between the user and hazardous voltages.

WHAT IS ISOLATION? Isolation is a means of physically and electrically separating

two parts of a measurement device, and can be categorized intoelectrical and safety isolation. Electrical isolation pertains to elimi-nating ground paths between two electrical systems. By providingelectrical isolation, you can break ground loops, increase the com-mon-mode range of the data acquisition system, and level shift thesignal ground reference to a single system ground. Safety isolationreferences standards have specific requirements for isolatinghumans from contact with hazardous voltages. It also characterizesthe ability of an electrical system to prevent high voltages and tran-sient voltages from transmitting across its boundary to other elec-trical systems with which you can come in contact.

Incorporating isolation into a data acquisition system hasthree primary functions: preventing ground loops, rejecting com-mon-mode voltage, and providing safety.

GROUND LOOPS Ground loops are the most common source of noise in

data acquisition applications. They occur when two connectedterminals in a circuit are at different ground potentials, causingcurrent to flow between the two points. The local ground of thesystem can be several volts above or below the ground of thenearest building, and nearby lightning strikes can cause the dif-ference to rise to several hundreds or thousands of volts. Thisadditional voltage itself can cause significant error in the meas-urement, but the current that causes it can couple voltages innearby wires as well. These errors can appear as transients orperiodic signals. For example, if a ground loop is formed with 60Hz AC power lines, the unwanted AC signal appears as a periodicvoltage error in the measurement.

When a ground loop exists, the measured voltage, Vm, isthe sum of the signal voltage, Vs, and the potential difference, Vg,which exists between the signal source ground and the measure-ment system ground, as shown in Figure 1. This potential is gen-erally not a DC level; therefore, the result is a noisy measurementsystem, often showing power-line frequency (60 Hz) componentsin the readings.

GROUND-REFERENCED SYSTEM INTRODUCES GROUND LOOP To avoid ground loops, ensure that there is only one

ground reference in the measurement system, or use isolatedmeasurement hardware. Using isolated hardware eliminates thepath between the ground of the signal source and the measurementdevice, therefore preventing any current from flowing betweenmultiple ground points.

COMMON-MODE VOLTAGE An ideal differential measurement system responds only

to the potential difference between its two terminals, the (+) and(-) inputs. The differential voltage across the circuit pair is thedesired signal, yet an unwanted signal can exist that is commonto both sides of a differential circuit pair. This voltage is knownas common-mode voltage. An ideal differential measurementsystem completely rejects, rather than measures, the common-mode voltage. Practical devices however, have several limitations,described by parameters such as common-mode voltage range andcommon-mode rejection ratio (CMRR), which limit this ability toreject the common-mode voltage.

The common-mode voltage range is defined as the maxi-mum allowable voltage swing on each input with respect to themeasurement system ground. Violating this constraint results notonly in measurement error, but also in possible damage to com-ponents on the board.

Common-mode rejection ratio describes the ability of ameasurement system to reject common-mode voltages. Amplifierswith higher common-mode rejection ratios are more effective atrejecting common-mode voltages. The CMRR is defined as thelogarithmic ratio of differential gain to common-mode gain.

CMRR (dB) = 20 log (Differential Gain/Common-Mode Gain). (Equation 1)

Common-mode voltage is shown graphically in Figure 2.In this circuit, CMRR in dB is measured as 20 log Vcm/Voutwhere V-= Vcm.

HIGH-VOLTAGE MEASUREMENTS AND ISOLATION –GENERAL ANALOG CONCEPTS

NI Analog Resource Center

Figure 1. A Grounded Signal Source


In a non-isolated differential measurement system, anelectrical path still exists in the circuit between input and output.Therefore, electrical characteristics of the amplifier limit thecommon-mode signal level that can be applied to the input. Withthe use of isolation amplifiers, the conductive electrical path iseliminated and the common-mode rejection ratio is dramaticallyincreased.

ISOLATION CONSIDERATIONS There are several terms with which to be familiar when

configuring an isolated system:

Installation Category: A grouping of operating parameters thatdescribe the maximum transients that an electrical system cansafely withstand. Installation categories are discussed in moredetail later.

Working Voltage: The maximum operating voltage at which thesystem can be guaranteed to continuously safely operate withoutcompromising the insulation barrier.

Test Voltage: The level of voltage to which the product is sub-jected during testing to ensure conformance.

Transient Voltage (Over-voltage): A brief electrical pulse orspike that can be seen in addition to the expected voltage levelbeing measured.

Breakdown Voltage: The voltage at which the isolation barrierof a component breaks down. This voltage is much higher thanthe working voltage, and often times is higher than the transientvoltage. A device cannot operate safely near this voltage for anextended period of time.

ISOLATION TYPES Physical isolation is the most basic form of isolation,

meaning that there is a physical barrier between two electricalsystems. This can be in the form of insulation, an air gap, or anynon-conductive path between two electrical systems. With purephysical isolation however, we imply that no signal transfer existsbetween electrical systems. When dealing with isolated measure-ment systems, you must have a transfer, or coupling, of energyacross the isolation barrier.

There are three basic types of isolation that can be used ina data acquisition system:

OPTICAL ISOLATION Optical isolation is common in digital isolation systems.

The media for transmitting the signal is light and the physical

isolation barrier is typically an air gap. The light intensity is pro-portional to the measured signal. The light signal is transmittedacross the isolation barrier and detected by a photoconductiveelement on the opposite side of the isolation barrier.

ELECTROMAGNETIC ISOLATION Electromagnetic isolation uses a transformer to couple a

signal across an isolation barrier by generating an electromagnet-ic field proportional to the electrical signal. The field is createdand detected by a pair of conductive coils. The physical barriercan be air or some other form of non-conductive barrier.

CAPACITIVE ISOLATION Capacitive coupling is another form of isolation. An elec-

tromagnetic field changes the level of charge on the capacitor.This charge is detected across the barrier and is proportional tothe level of the measured signal.

ISOLATION TOPOLOGIES It is important to understand the isolation topology of a

device when configuring a measurement system. Differenttopologies have several associated cost and speed considerations.

Figure 2. CMRR Measurement Circuit

Figure 3. Optical Isolation

Figure 4. Transformer

Figure 5. Capacitor


CHANNEL-TO-CHANNEL The most robust isolation topology is channel-to-channel

isolation. In this topology, each channel is individually isolatedfrom one another and from other non-isolated system components.In addition, each channel has its own isolated power supply.

In terms of speed, there are several architectures fromwhich to choose. Using an isolation amplifier with an analog todigital converter (ADC) per channel is typically faster because youcan access all of the channels in parallel. A more cost-effective, butslower architecture, involves multiplexing each isolated inputchannel into a single ADC.

Another method of providing channel-to-channel isolationis to use a common isolated power supply for all of the channels. Inthis case, the common-mode range of the amplifiers is limited tothe supply rails of that power supply, unless front-end attenuatorsare used.

BANK Another isolation topology involves banking, or grouping,

several channels together to share a single isolation amplifier. Inthis topology, the common-mode voltage difference betweenchannels is limited, but the common-mode voltage between thebank of channels and the non-isolated part of the measurementsystem can be large. Individual channels are not isolated, butbanks of channels are isolated from other banks and from ground.This topology is a lower-cost isolation solution because thisdesign shares a single isolation amplifier and power supply.

SAFETY AND ENVIRONMENTAL STANDARDS When configuring a data acquisition system, you must take

the following steps to ensure that the product meets applicablesafety standards:

• consider the operational environment, which includesthe working isolation voltage and installation category.

• choose the method of isolation in the design based onthese operational and safety parameters.

• choose the type of isolation based on the accuracy needed,the desired frequency range, the working isolation voltage,and the ability of the isolating components to withstandtransient voltages.

Not all isolation barriers are suitable for safety isolation.Even though measurement products may have components ratedwith high-voltage isolation barriers, the overall product design,not just the components, dictates whether or not the device meetshigh-voltage safety standards. Safety standards have specificrequirements for isolating humans from contact with hazardousvoltages. These requirements vary among different applications andworking voltage levels, but often specify two layers of protectionbetween hazardous voltages and human-accessible circuits or parts.

In addition, the standards for test and measurement equip-ment are not only concerned with dangerous voltage levels andshock hazards, but also with environmental conditions, accessi-bility, fire hazards, and valid documentation for explaining theuse of equipment in preventing these hazards. They maintainspecific construction requirements of isolation equipment toensure that the integrity of the isolation barrier is maintainedwith changes in temperature, humidity, aging, and variations inmanufacturing processes.

When dealing with safety standards, the EuropeanCommission and Underwriters Laboratories, Inc. (UL) have out-lined the standards that cover the design of high-voltage instru-ments. There are approximately 200 individual safety standardsharmonized (approved for use to demonstrate compliance) to theLow Voltage Directive, which was the initial document that out-lined the specifications for the voltage levels that require safetyconsideration.

The relevant standard for instrument manufacturers is EN61010 – Safety Requirements for Electrical Equipment forMeasurement, Control, and Laboratory Use. EN 61010 statesthat 30 Vrms or 60 VDC are dangerous voltages. In addition tohigh-voltage design requirements, EN 61010 also includes othersafety design constraints (such as flammability and heat).Instrument manufacturers must meet all the specifications in EN61010 to receive the CE label.

There are two other standards very similar to EN 61010–IEC 1010 and UL 3111. IEC 1010, which was established bythe International Electrotechnical Commission, is the precursorto EN 61010. The European Commission adopted it andrenamed it EN 61010. UL 3111 is also a child of IEC 1010. ULtook IEC 1010, made some modifications and adopted it as UL3111. This new, strict UL standard replaces the older, morelenient UL 1244 standard for measurement, control, and labora-tory instruments. For new designs, instrument manufacturersmust meet all of the specifications in UL 3111 to receive a ULlisting.

INSTALLATION CATEGORIES The IEC defined the term Installation Category (some-

times referred to as Over-voltage Category) to address transientvoltages. When working with transient voltages, there is a level

Figure 6. Channel-to-Channel Multiplexed Topology

Figure 7. Bank Topology


of damping that applies to each category. This damping reducesthe transient voltages (over-voltages) that are present in the system.As you move closer to power outlets and away from high-voltagetransmission lines, the amount of damping in the system increases.

The IEC has created four categories to partition circuitswith different levels of over-voltage transient conditions.

• Installation Category IV – Distribution Level (transmis-sion lines)

• Installation Category III – Fixed Installation (fuse panels)• Installation Category II – Equipment consuming energy

from a Category III fixed installation system. (wall outlets) • Installation Category I – Equipment for connection to

circuits where transient over-voltages are limited to asufficiently low level by design.

TYPICAL APPLICATIONS REQUIRING ISOLATION

SINGLE-PHASE AC MONITORING To measure power consumption with 120/240 VAC power

measurements, you record instantaneous voltage and current val-ues. The final measurement, however, may not be instantaneouspower, but average power over a period of time or cost informa-tion for the energy consumed. By making voltage and currentmeasurements, software can make power measurements or doother analyses. To make high-voltage measurements you needsome type of voltage attenuator to adjust the range of the signal tothe input range of the measurement device. Current measurementsrequire a precision resistor. The voltage drop across the resistor ismeasured, and Ohm’s Law (I = V/R) produces a current value.

FUEL CELL MEASUREMENT Fuel cell test systems make a variety of measurements

that require signal conditioning before the raw signal is digitizedby the data acquisition system. An important feature for the test-ing of fuel cell stacks is isolation. Each individual cell can gen-erate about 1 V, and a stack of cells can produce several kV. Toaccurately measure the voltage of a single 1 V cell in a large fuelcell stack requires a large common-mode range and high com-mon-mode rejection ratio. Because adjacent cells have a similarcommon-mode voltage, bank isolation is sometimes acceptable.

HIGH COMMON-MODE THERMOCOUPLE MEASUREMENT Some thermocouple measurements involve high common-

mode voltages. Typical applications include measuring tempera-ture while a thermocouple is attached to a motor, or measuring thetemperature dissipation capabilities in a conductive coil. In thesecases, you are trying to measure small, millivolt changes withseveral volts of common-mode voltage. It is therefore importantto use an isolated measurement system with good common-moderejection specifications.

SERIAL COMMUNICATION Reliability is a number one concern when designing

equipment to be resistant to the interference inherent in a harshenvironment. Commercial and industrial applications such asPOS networks, ATMs, bank teller stations, and CNC-based pro-duction lines are susceptible to voltage spikes and noise.Isolation reduces the possibility of damaging control systemsand ensure that systems can remain operational. Other applica-tions that may require isolation are industrial process control,factory automation, serial networking devices, high speedmodems, monitoring equipment, long distance communicationdevices, printers and remote serial device control.

Figure 8. Installation Categories


The standards for personal safety in electric and magneticfields have been tightened. Three-dimensional measurement of thefields and the combination of these components into the equivalentfield strength is now required. Is this extra effort justified? As partof a study project at the Fachhochschule Reutlingen, high voltagelines, transformer stations and the working environment wereinvestigated. The results show that three-dimensional measure-ment is indeed necessary.

MEASUREMENT METHODS An E field sensor basically consists of a pair of condenser

plates placed side by side, across which the dielectric current ismeasured. The disadvantage of this simple arrangement is itsdirectional characteristic. To measure accurately, the direction ofthe field lines has to be known and the sensor positioned accord-ingly. This is seldom possible in practice. As a result, the tradeassociation [1] requires the measurement to be made in each ofthe three orthogonal spatial axes and the so-called equivalentfield strength calculated by summing the squares of the threefield components. This is theoretically possible with a simpleprobe by making three consecutive measurements in the threedirections, assuming that the field remains constant over time.The practical answer is to use a sensor that has a three dimen-sional structure. Modern measuring equipment uses sensorsmade up from three plate condensers arranged at right angles toeach other, and calculate the equivalent field strength automati-cally. The isotropy, i.e. the actual non-directionality of the sensor,is important in this context. This can be assessed by rotating thesensor in an homogeneous field; the indicated field strength mustremain constant [3]. This is the only way to ensure that danger-ous field strengths are not present.

MEASUREMENT CONDITIONS Several factors must be observed if measurements are to

conform to relevant standards [1]: • No person should be present in the immediate vicinity of

the measurement. • Objects in the vicinity that distort the field, such as trees,

bushes, machinery, etc., must be noted. • Environmental effects such as air humidity, temperature,

type of terrain, etc., must also be noted. • No condensation may be present on the sensor or its sup-

porting tripod as this will lead to measurement errors. • The persons operating the measuring instrument must

ensure that they do not stand between the field source andthe probe during the measurement.

These measures are required in order that comparable andreproducible results can be obtained under varying operatingconditions.

SIMPLEST CASE: THE HIGH VOLTAGE LINE Our first example is a high voltage line running across open

land. If the field is measured at the lowest point of the cable sag,i.e. as far as possible from the masts, it can even be assumed thatthe field lines are vertical. As expected, the measurement results ofa three dimensional (isotropic) and a one dimensional (so-called yonly measurement) differ only slightly from one another. Themaximum difference is below 5%. The slight unsymmetry in themeasurement curve is due to the terrain which showed a slightupward slope from left to right. The phase relationships betweenthe conductors are of no consequence in this case, as the measure-ment distance from the conducting cables is too large.

MORE INTERESTING: LINE CROSSING The second example shows the field profile in the area

where two lines cross. The measurement conditions were: • Voltages 110 kV and 220 kV• Three-phase conductors • Line 1 (top to bottom): 220 kV, christmas tree masts

approx. 40 m high • Line 2 (left to right): 110 kV single layer masts approx.

26 m high The ambient conditions at the time of the measurement

were: Temperature 16°C, average air humidity, very damp ground. Figure 2 shows the basic measurement path. Some trees

and a number of small bushes were located in the immediatevicinity of the measurement. The distorting effects of theseobjects on the field profile are discussed in the evaluation.

STANDARD MEASUREMENTS: ELECTRIC FIELDSDUE TO HIGH VOLTAGE EQUIPMENT

Ralf Müller and Hans-Joachim Förster

Figure 1: High-voltage line. Results of electric field measurements in one and three dimensions


Figure 3 shows the fieldstrength profile that was measured.The starting point of the measure-ment in the diagram is at the positionof the mast. The last measurementwas made at a distance of 60 m fromthis point. The effect of the mast canbe clearly seen up to the area wherethe lines cross. The crossing begins30 m from the starting point. A fieldstrength maximum occurs at the 28m point. This is due to the additionof the field strengths of the two lines.In the area of the crossing, the fieldcomponents of the upper line arecompensated, resulting in a mini-mum at this point. The field strengthincreases again rapidly after thecrossing area, at 52 m. This is due tothe fact that the screening effect ofthe mast is now reduced and the areaof the crossing has been left.

Figure 4 shows the relativedifference between the one-dimen-sional and three-dimensional meas-urements. The maxima are found atthe entry and exit of the crossing.The difference is up to 13%. Thelower conductors in the crossing areacompensate out the field compo-nents of the upper conductors. Thevariation in the field within the areaof the crossing in figure 3 is due tothe uneven terrain. This section istherefore shown in more detail infigure 5.

Figure 2: Measurement path beneath two high voltagelines that cross. Green areas indicate bushes and trees

Figure 3: Electric field profile where two high voltage lines cross.

Figure 4: Relative difference between one-dimensional and three-dimensional measurements


COMPLEX: TRANSFORMER STATION The Neckarwerke Esslingen AG kindly allowed us to

make measurements in a transformer station. A measurementpath was selected that included several conductor arrangements,insulators and carriers. It is depicted in figure 6.

The ambient conditions at the time of the measurementwere: Temperature 5°C, average air humidity, very damp ground.

The measurement results clearly show that significant dif-ferences in the results of one-dimensional and three-dimension-al measurements occur in the vicinity of crossing conductors,switching equipment, current busbars and the like. The relativeerror is very dependent on the measurement position. Directlybeneath the conductors, it is small, but it can be as much as 60%at points between the conductors. This difference cannot beaccepted when measurements are made for personal safety, espe-cially where legal settlements are involved. The difference clear-ly shows that the indicated field strength is lower than the actualfield strength and hence the assumed safety is not given. Thisexposes a weakness in IEC standard 833 [4] which exclusivelydefines measurement in the vicinity of high voltage lines and istherefore not applicable in cases where labor laws are involved.

Figure 5: Zoomed representation of the crossing area from figure 3

Figure 6: Measurement path in a transformer station

Figure 7: Direct comparison of one-dimensional and three-dimensionalmeasurement results.

Figure 5 clearly shows the effect of theterrain on the measurement result. A three-dimensional measurement is clearly to be pre-ferred where the terrain is very uneven. Accurateresults are not given by a one-dimensional meas-urement or by a computer simulation.

View from above: View from side:


UNCLEAR: MOST WORKING ENVIRONMENTS The conditions of most working environments in industry

are far removed from the simple case of a high-voltage line;switching equipment, transformer stations, induction heaters andmachinery may all play a part in the field profile. It is thus notpossible to predict the spatial field profile or its variation withtime. Further uncertainty results from the frequency spectrum.Several standards specify different limit values for different fre-quencies. Broadband measurement equipment cannot, therefore,be used if the frequency of the field is unknown or if severalfields are superimposed. As an example of this, an inductionheater emits radiation at the AC. line frequency of 50 or 60 Hzand its harmonics and also at the frequency of the heating cur-rent. The latest test equipment copes with this situation byemploying built-in filters to detect the main radiation compo-nents and evaluate their frequencies. The use of three-dimension-al measurement techniques coupled with filters is an absolutemust if personal safety measurements are to be made that arereproducible and which conform to the relevant standards.

REFERENCES[1] Precision Engineering and Electrical Engineering TradeAssociation: Rules for health and safety at work involving expo-sure to electric, magnetic or electromagnetic fields (in German)

[2] Electric and Magnetic Fields Everyday Electricity (in German)Electricity Industry Information Center (Informationszentrale derElektrizitätswirtschaft e.V.) 60596 Frankfurt

[3] Progress Report VDI Series 8: Measurement, Control andRegulation Dipl.-Ing. Georg Bahmeier, Untermeitingen Fieldprobes for calibration and for determining the magnitude anddirection of electric field strength (in German)

[4] International Standard IEC 833: Measurement of power fre-quency electric fields

[5] German Standard VDE 0848 Part 1: Endangerment due toelectromagnetic fields. Measurement and calculation methods(in German)

[6] German Standard VDE 0848 Part 4: Safety in electromagnet-ic fields. Field strength limit values for personal safety in the fre-quency range from 0 Hz to 30 kHz.

Figure 8: Relative error between one-dimensional and three-dimensionalmeasurements


Often it is necessary to identify a system that must operatein a closed-loop fashion under some type of feedback control.This may be due to safety reasons, an unstable plant that requirescontrol, or the expense required to take a plant offline for test. Inthese cases, it is necessary to perform closed-loop identification.

There are three basic approaches to closed-loop identifi-cation. These approaches are direct, indirect, and joint input-out-put. In this article we outline each approach and the system iden-tification techniques that may be used to implement them.

DIRECT The first method of interest is the Direct Approach. In this

method, we measure the output of the system y(t) and the inputto the plant u(t), ignoring any feedback and the reference signal,to obtain the model. This is illustrated in Figure 1. This has theadvantage of requiring no knowledge about the feedback in thesystem and becomes an open-loop identification problem.

The suggested system identification model structureswhen using this method are ARX, ARMAX and state-spacemodels. Optimal accuracy occurs if the chosen model structurecontains the true system (including the noise properties) and themain drawback to the method is that a poor noise model canintroduce bias into the model. This bias will be small when anyor all of the following hold

• The noise model is representative of the actual noise • The feedback contribution to the input spectrum is small• The signal to noise ratio is highSpectral analysis will not provide correct results in the

closed-loop case when using the direct approach so avoid non-parametric methods of identification such as impulse responseand bode response estimation.

INDIRECT The second method of interest in closed-loop identifica-

tion is the Indirect Approach as shown in Figure 2. In thismethod we identify the closed loop system (Gcl) using measure-ments of the reference input r(t) and the output y(t) and retrievethe plant model making use of a known regulator structure. Thetransfer function for the open loop plant G, with regulator H, canbe retrieved from

The advantages in using the indirect approach are that anymethod will work in determining the closed-loop transfer func-tion Gcl and the need for an accurate representation of the noisemodel is alleviated. The main disadvantage is that any error in H(including deviations due to saturations or anti-windup logic)will be imposed directly into G resulting in bias errors.

JOINT INPUT-OUTPUT The last method is the Joint Input-Output Approach. As

shown in Figure 3, we consider the plant input u(t) and the sys-tem output y(t) as outputs of the system. The inputs to the sys-tem are the reference signal r(t) and the noise signal v(t).

IDENTIFICATION OF CLOSED LOOP SYSTEMSNI Analog Resource Center

Figure 1 Direct Approach to Closed-Loop System Identification.

Figure 2 Indirect Approach to Closed-Loop System Identification.


This identification method results in a multidimensionalsystem of the form

Where the system matrix A is comprised of two models,the closed-loop model Gcl and the model relating u(t) to r(t),Gru. The plant model, G, is then estimated from the relation.

This approach is advantageous because the regulatorstructure is not needed nor is an accurate noise model necessary.It suffers from the disadvantages of requiring additional acquisi-tion hardware (sensors) and requires acquiring a greater quantityof data.

When using the indirect and joint input-output methods,the reference signal r(t) should be as informative as possible.This means it should provide good spectral coverage of thedomain of interest. This may be done by adjustments to the sys-tem set points (or adjustments to the regulator) as much asallowed by the system being identified.

CONCLUSION It is often necessary to perform identification under

closed-loop conditions to increase safety or reduce the costs of themodeling. The three approaches outlined in this article provideaccurate estimations of plant dynamics under feedback controlusing simple measurements. Using the LabVIEW SystemIdentification Toolkit provides the necessary identification algo-rithms to aid in these closed-loop identification problems.

Figure 3 Joint Input-Output Approach to Closed-Loop System Identification.


Transducers for electrical measurement are an essentialpart of any monitoring, measuring, or controlling system whereelectrical quantities are involved. In order to use these transduc-ers, it is important to know what they do, what kind of signalthey provide, and how to connect them.

Part I provides an introduction in using voltage, current,and power (watt) transducers along with using potential and cur-rent transformers.

POTENTIAL TRANSFORMERS Most manufactures’ transducers accept up to a maximum

of 600 volts AC direct. For AC voltages greater than 600 volts,potential transformers are required. Potential transformers areprecision transformers that step the voltage down to 120 voltsAC, a standard transducer input. These transformers, particularlywhen used with power or watt transducers, must be instrumentgrade transformers. They must not only be precise in steppingdown the voltage but in maintaining the phase or time relationshipof the voltage. This is very important. Do not attempt to save moneyby using control class transformers.

Transducer and meter loads are connected in parallel tothe potential transformer. Take care not to exceed the transformerburden rating. This burden is expressed in VA for volt-amperes(the product of volts and amps).

CURRENT TRANSFORMERSFor AC applications, most manufacturers’ transducers

will not accept direct current input over 20 amperes. For higheramperages, current transformers are utilized.

These transformers are most often the ‘donut’ type. Thecurrent carrying conductor is passed through the opening or win-dow of the ‘donut’. The secondary winding of the current trans-former is wound by the manufacturer on the toroidal iron corewhich makes the ‘donut’ shape. On most North American manu-factured current transformers, the secondary is wound to produce5 amperes when rated current is passed through the window. Theturns ratio is expressed as 100:5 or 3500:5 (read as 100 to 5 and3500 to 5). The first number represents the rated full-scale pri-mary current. The primary winding consists of the single pass ofthe current carrying conductor through the window. The secondnumber represents the full-scale secondary current in amperes. A100:5 ratio current transformer steps the current from 100amperes down to 5 amperes. The 3500:5 ratio current trans-former steps the current from 3500 amperes down to 5 amperes.

As with potential transformers, only use instrument gradecurrent transformers with power measuring transducers.

Connect the loads on current transformers in series beingcareful not to exceed the burden rating. The phase angle shiftintroduced by current transformers is sensitive to the loading.Therefore, keep the burden to a minimum by using adequate sizesecondary leads and keeping secondary leads as short as possible.

CAUTION: Current transformers can and will develop alethal voltage and possibly self destruct if the secondary is openwhen primary current is present! People have been hurt andequipment damaged when the secondary winding of a currenttransformer was opened. Never disconnect the secondary orleave it open when there is the possibility of primary current.

It is essential that experienced persons install currenttransformers. If you must make a connection to the current trans-former while it is in use, SHORT THE SECONDARY WIND-ING before doing anything. Some current transformers have ashorting block for this purpose. Auxiliary shorting blocks areavailable for this purpose too.

SELECTING AND USING TRANSDUCERS FORTRANSFORMERS FOR ELECTRICAL MEASUREMENTS

William D. Walden


Current transformers are rated for the voltage class forwhich they are to be used. These classes are: 600 volts, 5000volts, 8700 volts, 15 kilovolts, 25 kilovolts, and 34.5 kilovolts.Make certain that the current transformers are rated for the voltagewith which they are working or that the conductor is insulated forthe class voltage. Current transformers being used on conductorswith voltages greater than 600 volts must have the secondarygrounded to an earth ground.

VOLTAGE TRANSDUCERSVoltage transducers provide a DC current or voltage out-

put directly proportional to the AC input voltage. AC voltagetransducers typically have a transformer input to isolate thetransducer from the voltage input. Following the transformer arethe electronics.

There are two types of AC voltage transducers.• Absolute average measuring, rms than 600 volts, one

should use a calibrated (or mean value measuring,potential transformer. rms calibrated). These inexpensivetransducers simply convert the AC input to DC Currenttransducers provide a DC current or and have the outputcalibrated to voltage output directly proportional to therepresent the root mean square AC input current. ACcurrent transducers (RMS) value for sine wave input.Typically have a transformer input to isolate. This typeis very adequate for the transducer from the currentinput. This type is very adequate for situations in whichthe voltage wave shape is not distorted. Any odd har-monic or discontinuity will introduce large error. Usethe true RMS measuring type when distortion of a sinewave is present.

• True RMS (root mean square) measuring. These trans-ducers calculate the RMS value of the current input andprovide a DC output directly proportional to the effectivevalue of the current input. This type should be usedwhenever the current is distorted. Ohio Semitronics, Inc.has a wide range of models available for various situa-tions. Models are available with or without current trans-formers, with current transformers built in, and with splitcore current transformers.

• True RMS (Root Mean Square) measuring. These trans-ducers calculate the RMS value of the voltage input andprovide a DC output directly proportional to the effectivevalue of the voltage input. This type should be used when-ever the voltage is distorted.

Transducer models are available for nominal input voltagesof 69, 120, 240, and 480 volts. These typically have a measuringrange of 0 to 125% of the nominal input rating. Thus, a 120-voltmodel has a range of 0 to 150 volts. For voltage input higher than600 volts, one should use a potential transformer.

CURRENT TRANSDUCERSCurrent transducers provide a DC current or voltage out-

put directly proportional to the AC input current. AC currenttransducers typically have a transformer input to isolate thetransducer from the current input. Following the transformer arethe electronics.

There are two types of AC current transducers.• Absolute average measuring, rms calibrated (or mean

value measuring, rms calibrated). These inexpensivetransducers simply convert the AC input to DC and havethe output calibrated to represent the root mean square(RMS) value for sine wave input. This type is very ade-quate for situations in which the current wave shape isnot distorted. Any odd harmonic or discontinuity willintroduce large error. Use the true RMS measuring typewhen distortion of a sine wave is present.

• True RMS (root mean square) measuring. These trans-ducers calculate the RMS value of the current input andprovide a DC output directly proportional to the effectivevalue of the current input. This type should be usedwhenever the current is distorted.

Ohio Semitronics, Inc. has a wide range of models avail-able for various situations. Models are available with or withoutcurrent transformers, with current transformers built in, and withsplit core current transformers.


POWER OR WATT TRANSDUCERSA watt or power transducer measures true electrical power

delivered to a load and converts that measurement to a DC voltageor current signal proportional to the power measured. To measurepower, the watt transducer must monitor both the voltage and cur-rent in a circuit. Further, it must be able to accurately determine thephase relationship between the voltage and current. This is the angleby which the current leads or lags the voltage. This measurement isvery important to accurately determine true power.

The watt transducer must also measure the power in eachof the branches of the circuit. Your house, apartment, or smalloffice is wired in what is often referred to as the Edison system.This is a three-wire, single phase system with two power linesand a neutral. The watt transducer must measure the power ineach of the power lines or mains. This circuit requires a two-ele-ment watt transducer. A two-element watt transducer has two-watttransducers in the same case. The outputs of the two transducers ormultipliers are summed so that the output signal of the entire watttransducer represents total power. One, two, and three elementwatt transducers are discussed in Part II.

What type of watt transducer to use?• Analog watt transducers including Hall effect provide

good accuracy even with distorted wave shapes, discon-tinuity, or where there is poor frequency regulation.

• Electronic watt transducers with sampling or pulse-width, pulse-height type multipliers provide excellentaccuracy but may have problems with discontinuity orwhere there is poor frequency regulation. Before orderingwatt transducers, it is to your advantage to assess yourspecific needs and conditions.

SINGLE-PHASE WATT TRANSDUCERSThe most common application for a watt transducer is

monitoring a single-phase load such as a heater element or smallmotor. This requires a single element watt transducer connecteddirectly between the power line and the load as illustrated below.

The single-phase watt transducer shown above has a singlemultiplier or element inside the electronics package. Often thecombined loads of an entire house, apartment, or office are mon-itored with a watt transducer. This requires a two-element modelwith current transformers. The two-element, single-phase watttransducer is connected as shown below.

The two-element watt transducer shown above has twomultipliers inside the electronics package. The output of thesetwo multipliers is summed to obtain the total power. The outputsignal of this watt transducer thus represents the total powerbeing used.

THREE-PHASE WATT TRANSDUCERSMost motors in industry are three-phase, three-wire

motors. These require two-element watt transducers. Do notattempt to save money and use a single element transducer – itwill not provide correct or useful information. Smaller three-phase motors may be connected directly to the watt transducer.Larger three-phase motors will require the use of current and/orpotential transformers. All three cases are shown in the three dia-grams that follow.


Factories and large stores are typically supplied withthree-phase, four-wire power. Heavy loads such as motorsare connected line-to-line in a three-phase, three-wire con-figuration and lighter loads are connected line to neutral.Three element watt transducers are required to monitor theentire facility. This requires the use of current transformers.The connections are shown below.

In special cases where a three-phase, four-wire load isknown to be balanced in load and voltage, a single element watttransducer may be used to give an indication of total power bymultiplying the value represented by the transducer output bythree.

THE 2 1/2 ELEMENT WATT TRANSDUCERMonitoring three-phase, four-wire systems frequently

involves using potential transformers. These transformers cancost much more than the watt transducer. To reduce cost, twopotential transformers instead of three are used. The watt trans-ducer from the two voltages can derive the third voltage.

OUTPUT SIGNALS FROM TRANSDUCERSThe voltage, current, and watt transduc-

ers discussed above are available with DC cur-rent or voltage output. The least expensive andsimplest voltage and current transducers areavailable only with a current output.

How are these outputs used?The most common is for metering. The

transducer output is driving either an analog ordigital meter. The use of either is simple.

ANALOG METERIf you are using an analog meter, buy transducers that are

supplied with a 0 to 1 mADC output and a 0 to 1 mADC metermovement. The meter supplier can scale the meter face to matchthe transducer range. Some examples are shown on the nextpage.


DIGITAL METERSome digital meters allow the user to scale the meter to dis-

play to the transducer range. If you use one of these meters with awatt transducer that has a 4 to 20 mADC output representing 0 to960 kilowatts, simply adjust the meter to read 0 at 4 mADC and960 at 20 mADC. If you are using a 0 to 2 volt DC input meter thatdoes not allow scaling, use a scaling resistor. Some examples areshown below.

In both examples a 2-volt DC digital meter is being used.By applying Ohm’s Law (R=E/I, the value of the resistor equalsthe voltage divided by the current), one can determine the valueof the resistor required. Remember that the output of the trans-ducer is in milliamperes, 1/1000 of an ampere.

How did we figure the value of the resistor? Always base the resistor on the rated output of the transduc-

er, the rated output is the wattage level or current value that is rep-resented by 1 mADC.

In the first example the rated output of the transducer is1000 watts. We would like the digital meter to read 1000. If wesupply 1 volt to the meter, it will read 1.000.

Digital meter manufacturers build their meters so that thedecimal point can be moved. This is done using wire jumpers onthe connection strip of the meter, by DIP switches on the meter,or by wire jumpers or foil jumpers that the user cuts. Follow themeter manufacturer’s instructions. In our example, set the meterto display 1000 or 1000.0 when 1 volt is applied.

How did we get the 1-volt from the transducer? Use Ohm’s law. The value of the resistor equals the

desired voltage divided by the current. Or for our example,R=1/0.001 or 1000. Use of a 1000 Ω resistor will provide 1 voltat full scale of 1 mADC. Our meter will read 1000 for 1000watts.

In the second example we used a twenty ampere currenttransducer with a digital meter. We want the meter to read 19.99at full scale to take advantage of the four digits. (A 3-_ digitmeter will read to 1.999 volts. Above this it will flash at you tolet you know that the meter is over ranging.)

How do we get 2 volts?Again use Ohm’s Law. The value of the resistor will be

R=2/0.001 or 2000 Ω. Set the decimal point so that the meterwill read 19.99 at 1.999 volts. Your meter is now scaled to matchthe transducer.

OTHER OUTPUT SIGNALS AVAILABLEMost transducers manufacturers have transducers avail-

able with 0 to 1 mADC, 0 to 5 volts DC, 0 to 10 volts DC, or 4to 20 mADC outputs. The 0 to 10 volt and 0 to 5 volt outputs aretypically (but not exclusively) used with data acquisition equip-ment, strip chart recorders, analog input cards for computers, orcontrol interface devices.

The 4 to 20 mADC output is used with process controlequipment, for long (over 200 feet) transmission of the signal,and frequently as a ‘fail safe’ monitoring of the signal.

If the watt transducer output is 4 mADC, then one knowsthat the power being monitored is zero. However, if the outputsignal is zero, something is wrong – the transducer may havefailed or it may have lost instrument power. The user can take cor-rective action.

CABLES FOR ANALOG SIGNALSOhio Semitronics, Inc. recommends using a shielded

twisted pair of 22 gauge or larger wire to conduct an analog volt-age or current signal from the transducer to the meter or instru-ment. If you are using a 1-mADC-output transducer and a loadresistor, we recommend putting the load resistor on the meter orinstrumentation package. Ground the shield at the receiving endonly. Do not ground at both ends. Doing so can cause severeproblems. I have known the shields to melt when a lighteningstrike has occurred nearby.

EXAMPLES OF POWER, VOLTAGE, AND CURENT MONITORINGMonitoring voltage, current, and power delivered to a test

load. In this application a refrigerator is being examined.

Transducers used:• CT5-010A current transducer is wired in series with the

load.


• VT-120A voltage transducer is wired in parallel with theload.

• PC5-010A is wired in series with the load for monitoringthe current and in parallel with the load for monitoring thevoltage. These are used as examples. Other transducersthat may be used include the multifunction board leveltransducer PTB. This board provides analog outputs pro-portional to each phase of true RMS Current, each phaseof true RMS voltage, and total power.

This example uses digital meter which are scaled usingprecision load resistors.

Load resistors are selected as follows:• Remember Ohm’s Law: R=V/I where R is the resistance

in ohms (Ω ), V is the voltage that we want to apply tothe digital meter, and I is the current from the transducer.

• The CT5-010A provides an output of 1 mADC at 10amperes AC through terminals 3 and 4. 1 mADC repre-sents 10 amperes AC. Adjust the decimal point of thedigital meter so that it displays 10.00 with 1 mADCthrough a 1000Ω load resistor.

• The VT-120A provides an output of 1 mADC at 150volts AC applied to terminals 3 and 4. 1 mADC repre-sents 150 volts. Adjust the decimal point of the digitalmeter so that it displays 150.0 with 1 mADC through a1500Ω load resistor.

• The PC5-010A provides an output of 1 mADC at 1000watts. 1 mADC represents 1000 watts. Adjust the deci-mal point on the digital meter so that it displays 1000with 1 mADC through the 1000Ω resistor. Now all threemeters are scaled correctly and may be labeled amperes,volts, and watts. Note that the power does not equalvolts times amperes. This is because the refrigerator hasa power factor of 0.866 that is normal for older refriger-ators. For the single-phase situation, power factor maybe determined by dividing the power reading by theproduct of volts and amperes.

POWER, VOLTAGE, CURRENT, AND POWER FACTORIn one example, we are monitoring only one phase for

current, between two lines for voltage, and using a two-elementwatt transducer that measures two lines of current and two linesof voltage with respect to the third.


We are assuming a balanced condition to compute powerfactor given one current reading, one voltage reading, and totalpower.

Power Factor:

PF= watts ÷ (apparent power in VA). Apparent power for a three-phase,three-wire load may be calculated from the product of voltage, current,and the square root of 3 (1.732) or

PF = watts ÷ (V*I*1.732)

= 81,000 ÷ (479*231*1.732)

= 0.423

This is a low power factor and is very typical of somelightly loaded induction motors.

Where does the 1.732 come from? It is the square root of3 rounded to three decimal places. The square root of threecomes from the ratio of line to line voltage to line to neutral voltagein the three-phase system. Please refer to POWER MONITORINGIN PART TWO of this brochure.

PITFALLSMonitoring AC voltage and AC current is simple enough,

but in monitoring power, one must follow the connection dia-grams exactly.

• Watt transducers are polarity sensitive. They sense notonly the power but also the direction in which it is flow-ing. Should a current transformer be installed backwards,the watt transducer will sense this as reverse power flowand provide an output reversed in polarity, a negativeoutput

• Watt transducers are also phase sensitive. If a currenttransformer is installed on the wrong phase line, the watttransducer will interpret this as a 120-degree phase angleshift and give the wrong result.

The most frequent complaint I receive on three-phase watttransducers is “I am not getting the correct output.” Conservativelystated, 90% of the time, the watt transducer is not correctly con-nected – a current transformer may be installed backwards or onthe wrong line, voltage connections may be cross phased, or volt-age connections may reference the wrong line. The other 9.5% ofthe time, the following gives the user trouble.

The electrical quantity – WATT – is a measure of the rateat which work is being done. If an electric motor is not doing anywork or is doing very little work, it will not consume very muchpower in watts even though the electric current is relatively high.The power factor will be low and a watt transducer monitoringthis motor will have a low output. This is to be expected! Theoutput from a watt transducer reflects the rate at which the motoris doing work.

If you encounter incorrect readings from a watt transduc-er, double check your connections against the connection dia-gram on the transducer case or connection sheet.

COMMENTSA watt transducer monitoring a three-phase, three-wire

load must be a two-element watt transducer because the voltage,as measured and the current are out of phase by ± 30º at unitypower factor, +30º on one leg and -30º on the other leg. Totalpower measured by the watt transducer is as follows:

Ptotal = [Ia * Vac * Cos (_+ 30°) +Ib * Vbc * Cos (_ - 30°)]

Where:• Ia is the current in leg A• Ib is the current in leg B• Vac is the voltage between leg A and C• Vbc is the voltage between leg B and C• θ is the phase angle shift between the voltage and cur-

rent – the power factor angle.At a power factor of 0.867 one reading between two of the

legs will be double that between the other two legs. The sum ofthe two is the correct total power.

At a power factor of 0.500 one reading between two of thelegs will be greater than 0 and the other will be 0. The total of thetwo is the correct total power.

At a power factor of 0 the readings between the two sets oflegs will be the same but opposite in sign. Again, the total of thetwo is the correct total power – zero!


To expertly troubleshoot electrical equipment, problemsmust be solved by replacing only defective equipment or compo-nents in the least amount of time. One of the most important fac-tors in doing this, is the approach used. An expert troubleshooteruses a system or approach that allows them to logically and sys-tematically analyze a circuit and determine exactly what is wrong.

The approach described here is a logical, systematicapproach called the 5 Step Troubleshooting Approach. It is aproven process that is highly effective and reliable in helping tosolve electrical problems.

This approach differs from troubleshooting procedures inthat it does not tell you step by step how to troubleshoot a partic-ular kind of circuit. It is more of a thinking process that is usedto analyze a circuit’s behavior and determine what component orcomponents are responsible for the faulty operation. This approachis general in nature allowing it to be used on any type of electricalcircuit.

In fact, the principles covered in this approach can beapplied to many other types of problem solving scenarios, notjust electrical circuits.

The 5 Step Troubleshooting Approach consists of the following:PreparationStep 1 ObservationStep 2 Define Problem AreaStep 3 Identify Possible CausesStep 4 Determine Most Probable CauseStep 5 Test and RepairFollow-upLet’s take a look at these in more detail.

PREPARATIONBefore you begin to troubleshoot any piece of equipment,

you must be familiar with your organization’s safety rules andprocedures for working on electrical equipment. These rules andprocedures govern the methods you can use to troubleshoot electri-cal equipment (including your lockout/tagout procedures, testingprocedures etc.) and must be followed while troubleshooting.

Next, you need to gather information regarding the equip-ment and the problem. Be sure you understand how the equipmentis designed to operate. It is much easier to analyze faulty operationwhen you know how it should operate. Operation or equipmentmanuals and drawings are great sources of information and arehelpful to have available. If there are equipment history records,you should review them to see if there are any recurring problems.You should also have on-hand any documentation describing theproblem. (i.e., a work order, trouble report, or even your notestaken from a discussion with a customer.)

STEP 1 – OBSERVEMost faults provide obvious clues as to their cause.

Through careful observation and a little bit of reasoning, mostfaults can be identified as to the actual component with very lit-tle testing. When observing malfunctioning equipment, look forvisual signs of mechanical damage such as indications of impact,chafed wires, loose components or parts laying in the bottom ofthe cabinet. Look for signs of overheating, especially on wiring,relay coils, and printed circuit boards.

Don’t forget to use your other senses when inspectingequipment. The smell of burnt insulation is something you won’tmiss. Listening to the sound of the equipment operating maygive you a clue to where the problem is located. Checking thetemperature of components can also help find problems but becareful while doing this, some components may be alive or hotenough to burn you.

Pay particular attention to areas that were identified eitherby past history or by the person that reported the problem. A noteof caution here! Do not let these mislead you, past problems arejust that – past problems, they are not necessarily the problemyou are looking for now. Also, do not take reported problems asfact, always check for yourself if possible. The person reportingthe problem may not have described it properly or may havemade their own incorrect assumptions.

When faced with equipment which is not functioningproperly you should:

• Be sure you understand how the equipment is designedto operate. It makes it much easier to analyze faultyoperation when you know how it should operate;

• Note the condition of the equipment as found. Youshould look at the state of the relays (energized or not),which lamps are lit, which auxiliary equipment is ener-gized or running etc. This is the best time to give theequipment a thorough inspection (using all your senses).Look for signs of mechanical damage, overheating,unusual sounds, smells etc.;

• Test the operation of the equipment including all of itsfeatures. Make note of any feature that is not operatingproperly. Make sure you observe these operations verycarefully. This can give you a lot of valuable informationregarding all parts of the equipment.

STEP 2 – DEFINE PROBLEM AREAIt is at this stage that you apply logic and reasoning to your

observations to determine the problem area of the malfunctioningequipment. Often times when equipment malfunctions, certainparts of the equipment will work properly while others not.

The key is to use your observations (from step 1) to ruleout parts of the equipment or circuitry that are operating properlyand not contributing to the cause of the malfunction. You should

HOW TO TROUBLESHOOT LIKE AN EXPERTA SYSTEMATIC APPROACH

Warren Rhude, Simutech Multimedia Inc.


continue to do this until you are left with only the part(s) that, iffaulty, could cause the symptoms that the equipment is experi-encing.

To help you define the problem area you should have aschematic diagram of the circuit in addition to your noted obser-vations.

Starting with the whole circuit as the problem area, takeeach noted observation and ask yourself "what does this tell meabout the circuit operation?" If an observation indicates that asection of the circuit appears to be operating properly, you canthen eliminate it from the problem area. As you eliminate eachpart of the circuit from the problem area, make sure to identifythem on your schematic. This will help you keep track of all yourinformation.

STEP 3 – IDENTIFY POSSIBLE CAUSESOnce the problem area(s) have been defined, it is neces-

sary to identify all the possible causes of the malfunction. Thistypically involves every component in the problem area(s).

It is necessary to list (actually write down) every faultwhich could cause the problem no matter how remote the possi-bility of it occurring. Use your initial observations to help you dothis. During the next step you will eliminate those which are notlikely to happen.

STEP 4 – DETERMINE MOST PROBABLE CAUSEOnce the list of possible causes has been made, it is then

necessary to prioritize each item as to the probability of it beingthe cause of the malfunction. The following are some rules ofthumb when prioritizing possible causes.

Although it could be possible for two components to failat the same time, it is not very likely. Start by looking for onefaulty component as the culprit.

The following list shows the order in which you shouldcheck components based on the probability of them being defective:

• First look for components which burn out or have a ten-dency to wear out, i.e. mechanical switches, fuses , relaycontacts, or light bulbs. (Remember, that in the case offuses, they burn out for a reason. You should find outwhy before replacing them.)

• The next most likely cause of failure are coils, motors,transformers and other devices with windings. Theseusually generate heat and, with time, can malfunction.

• Connections should be your third choice, especiallyscrew type or bolted type. Over time these can loosenand cause a high resistance. In some cases this resistancewill cause overheating and eventually will burn open.Connections on equipment that is subject to vibrationare especially prone to coming loose.

• Finally, you should look for is defective wiring. Pay par-ticular attention to areas where the wire insulation couldbe damaged causing short circuits. Don’t rule out incor-rect wiring, especially on a new piece of equipment.

STEP 5 – TEST AND REPAIRTesting electrical equipment can be hazardous. The elec-

trical energy contained in many circuits can be enough to injureor kill. Make sure you follow all your companies safety precau-tions, rules and procedures while troubleshooting.

Once you have determined the most probable cause, youmust either prove it to be the problem or rule it out. This cansometimes be done by careful inspection. However, in many

cases the fault will be such that you cannot identify the problemcomponent by observation and analysis alone. In these circum-stances, test instruments can be used to help narrow the problemarea and identify the problem component.

There are many types of test instruments used for trou-bleshooting. Some are specialized instruments designed to measurevarious behaviors of specific equipment, while others, like themultimeters, are more general in nature and can be used on mostelectrical equipment. A typical multimeter can measure AC andDC Voltages, Resistance, and Current.

A very important rule when taking meter readings is topredict what the meter will read before taking the reading. Usethe circuit schematic to determine what the meter will read if thecircuit is operating normally. If the reading is anything other thanyour predicted value, you know that this part of the circuit isbeing affected by the fault.

Depending on the circuit and type of fault, the problemarea as defined by your observations, can include a large area ofthe circuit creating a very large list of possible and probablecauses. Under such circumstances, you could use a “divide andeliminate” testing approach to eliminate parts of the circuit fromthe problem area. The results of each test provides information tohelp you reduce the size of the problem area until the defectivecomponent is identified.

Once you have determined the cause of the faulty opera-tion of the circuit you can proceed to replace the defective com-ponent. Be sure the circuit is locked out and you follow all safetyprocedures before disconnecting the component or any wires.

After replacing the component, you must test operate allfeatures of the circuit to be sure you have replaced the propercomponent and that there are no other faults in the circuit. It canbe very embarrassing to tell the customer that you have repairedthe problem only to have him find another problem with theequipment just after you leave.

Please note, Testing is a large topic and this article hasonly touched on the highlights.

FOLLOW UPAlthough this is not an official step of the troubleshooting

process, it nevertheless should be done once the equipment hasbeen repaired and put back in service. You should try to deter-mine the reason for the malfunction.

• Did the component fail due to age? • Did the environment the equipment operates in cause

excessive corrosion? • Are there wear points that caused the wiring to short out? • Did it fail due to improper use? • Is there a design flaw that causes the same component to

fail repeatedly? Through this process, further failures can be minimized.

Many organizations have their own follow-up documentationand processes. Make sure you check your organization’s proce-dures.

Adopting a logical and systematic approach such as the 5-Step Troubleshooting Approach can help you to troubleshoot likean expert!


TROUBLESHOOTING IN THE FIELD – MOTOR TESTING –MOTOR CONTROLLER – PROGRAMMABLE LOGICCONTROLLERS (PLC)

A laptop computer with PLC programming, communica-tion, and operating programs is a necessary tool in today’s mod-ern plant. Engineers, production supervisors, maintenance super-visors, maintenance technicians, electricians, instrument techni-cians, and maintenance mechanics all need to have PLC andcomputer knowledge, training and skills in troubleshooting.

On-the-job training on PLCs is usually not very effectiveuntil the person being trained has reached a certain level ofexpertise in several areas. Knowledge and skills in electricity,troubleshooting, and computer operation are necessary prerequi-sites to effectively assimilate basic PLC training. The authorfound that long-term retention of material studied was higherfrom a vocational course taken at a local junior college than froma fast-paced, cram-course through a manufacturer.

The manufacturer’s course covered essentially the samematerial as a course at the junior college (JC). The major differ-ences were the amount of study time and shop time. The JCcourse was four hours of class time per week for 15 weeks. Therewere three hours of shop time doing actual hands-on work relat-ing to the problems and material covered in the first hour.Additional time was spent at home studying the manual and writ-ing programs. Also, the JC was open at night for extra shop timeon PLCs and computers.

In contrast, the manufacturer’s course was five, eight hourdays. Class work was extremely fast and condensed in order tocover the amount of material involved. The instructor was veryknowledgeable and covered the course material as we tried toinput the programs into desktop training equipment in order tosee how it worked. By the end of each day, our minds werejammed with information. By the end of the week, we all passedthe course, but I had a hard time remembering what we had stud-ied on the first day.

Basic troubleshooting techniques apply to every situationand occupation. Positive identification of the problem(s) is absolutelyessential to solving the problems. Many times, the inexperiencedtroubleshooter will mistake one or more of the symptoms for theproblems. Solving the symptom(s) will normally just postpone theproblems to a later date, by which time, the problems may havegrown to mountainous proportions.

An example is when a person experiences a headache andtakes a mild pain reliever, such as aspirin. The actual problemmight be any number of things: eyes need to be checked, med-ication or lack of medication, muscle strain, stress, tumor, bloodvessel blockage, or old war injury. The same thing occurs inindustry, a fuse in a circuit blows and the maintenance persongets the replacement fuse and inserts it into the fuse holder.There are many things that could have caused the fuse to blow,

depending on the complexity of the circuit.Excess current caused the fuse to open (blow). Excess

current could have been caused by: overload on the load; shortcircuit between the wires, grounded wires, short circuit in theload, ground in the load, voltage spike, voltage droop, etc. If themaintenance person does not troubleshoot the circuit prior toreplacing the fuse and restoring power, negative consequencescould arise.

It is not uncommon for a process to develop a number ofsmall problems and continue to function at a degraded level ofoperational capability. Then, one more small problem occurs andthe whole process breaks down. Finding and correcting the lastproblem will not necessarily restore the operational capability ofthe process. The process continued operations with the smallproblems, but the small problems may not allow the process torestart from a dead stop. All the other small problems must beidentified and corrected before the process is restored to fulloperational capability.

This situation arises in industry as well as a person. Theperson can continue to function with a number of small prob-lems, such as fatigue, blood pressure problems, hardening of thearteries, artery blockage, but one more small blood clot in thewrong place could easily cause the death of the person. Clearingthe blood clot does no good to the person. He/she will not berestored to full operational capability.

TROUBLESHOOTING IN THE FIELD:Unless prior experience dictates otherwise, always begin atthe beginning.

Ask questions of the Operator of the faulty equipment:• Was equipment running when the problem occurred?• Does the Operator know what caused the problem, and

if so, what, in their opinion, caused the problem?• Is the equipment out of sequence?• Check to ensure there is power• Turn on circuit breaker, ensure motor disconnect switch

is on, and operate start button/switchUse voltmeter to check the following at incoming and

load side of circuit breaker(s) and/or fuses, ensure that voltagesare normal on all legs and read voltage to ground from each leg:

• main power, usually 460 VAC between phases and 272to ground

• control & power, 208/240 between phases and 120 toground and 120 VAC to neutral on a grounded system

• low voltage control power, usually 24 to 30 VAC and/orVDC between phases and possibly to ground, usuallynegative is connected to ground

Check controlling sensors in area of problem, then makecomplete check of all sensors, limit switches to ensure they arein correct position, have power, are programmed, set, and arefunctioning correctly.

ELECTRICAL INDUSTRIAL TROUBLESHOOTINGLarry Bush


If and when a problem is found, whether electrical ormechanical, the problem should be corrected and the fault-find-ing begun anew, a seemingly unrelated fault or defect could bethe cause of the problem.

When there is more than one fault, the troubleshooting isexponentially more difficult. Do not assume that all problems aresolved after completing one. Always test the circuit and opera-tion prior to returning the equipment to service.

If available, check wiring diagrams and PLC programs toisolate problem.

Variable Frequency Drive (VFD) can be reset by turningpower off. Wait until screen is blank and restore power; on someVFDs, press Stop/Reset – then press Start.

Check that wiring is complete and that wires and connec-tions are tight with no copper strands crossing from one terminalto another or to ground.

Ensure that the neutral reading is good and that the neu-tral is complete and not open.

MOTOR TESTING IN SHOP:Prior to connecting a motor:• Move motor to electric shop motor test and repair station• Connect motor leads for 460 volt operation and wrap

connections with black electrical tape• Check motor windings with an ohmmeter, each reading

between phases should be within one or two ohms ofeach other; A to B, B to C, A to C

• Use megohmmeter to check insulation resistance toground of motor windings on 500 volt scale; minimumreading is 1000 ohms of resistance per volt of incomingpower that motor will be connected to

• Connect motor to power test leads and safety groundafter checking that test lead power is shut off; securemotor to table to prevent motor from jumping whenstarted; turn disconnect on; press start button; check “T”leads for motor amperage; check for abnormal soundsand heat in bearings or windings; clean motor shaft; shutdown and disconnect

MOTOR TESTING IN FIELD: When a motor overload or circuit breaker trips and/or

blows fuses, certain procedures and tests should be carried out:• Lockout and tagout main circuit breaker;• Test insulation resistance of motor wires and windings

by using megohmmeter between T1, T2, & T3 leads andground, then;

• Test “T” leads to motor with ohmmeter for continuityand ohmage of windings between A to B, B to C, A to C;each resistance should be within 1 or 2 ohms of eachother; if the ohms readings are significantly different, or,if there is no continuity; go to the motor disconnect box,turn it off, perform the continuity and resistance test onthe “T” leads, again; if the readings are good, the prob-lem is in the wires from the motor controller to the dis-connect switch.

• Check the three wires by disconnecting all three wiresfrom switch and twist together; go to controller andcheck for continuity between A to C, B to C, A to C; oneor more wires will be open or grounded.

• Correct solution is to pull all new wires in from con-troller to motor disconnect switch, whatever caused theproblem may have damaged the other wires, also,

replace all wires.• If problem is on motor side of disconnect switch, open

motor connection box and disconnect motor;• Check motor for resistance to ground with megohmme-

ter. If reading is below 500,000 ohms, motor is ground-ed and must be replaced.

• Test motor windings for ohms between phases withohmmeter A to B, B to C, A to C. Readings should bewithin 1 or 2 ohms of each other. If readings indicateopen or a significant ohmage difference, replace motor;

• If motor test readings are good, test the motor leadsbetween the disconnect switch and the motor connectionbox for continuity and ground resistance. If readings arenot good, replace wires.

• If all readings are OK, reconnect motor, remove lockout,and restore to service. The problem could have beenmechanical in nature; an overload on motor caused bythe chain, belt, bad bearings, faulty gearbox, or powerglitch.

MOTOR CONTROLLER:• Check motor Full Load Amps (FLA) at motor and check

setting on controller overload (OL) device; most newerOL devices are adjustable between certain ranges, someolder OL devices use heaters for a given amperage.

• If circuit disconnecting means in controller is a circuitbreaker, it should be sized correctly.

• If the disconnecting means is a Motor Circuit Protector(MCP), the MCP must be correctly sized for the motorit is protecting and the MCP has a trip setting unit whichhas to be correctly set based on the Full Load Amperageof the motor. Using a small screwdriver, push in on thescrew head of the device and move to a multiple of thir-teen of the FLA. Example: a motor FLA of 10 ampswould require that the MCP trip device be set to aninstantaneous trip point of 130 amps.

• Fuses protecting the motor should be the dual element orcurrent limiting type and based on the motor FLA.

PROGRAMMABLE LOGIC CONTROLLERS (PLC):• Check to ensure main power is on( 120 VAC.• Check 24V power available.• Identify problem area.• Check sensor operation in problem area.• Check sensor Inputs to PLC.• Check on PLC that a change in sensor state causes the

corresponding Input LED on the PLC to go on or off.• Identify Output controlled by Input on PLC ladder dia-

gram.• Ensure that Output LED is cycling on/off with Input.• Check that Output voltage is correct and cycling on/off

with Input.• Locate Output device and ensure that voltage is reaching

device and cycling with Input.• Ensure that Output device is working correctly (solenoid

coil, relay coil, contactor coil, etc.)• An Input or Output module can be defective in one area

or circuit and work correctly in all other circuits• If each field circuit is not fuse protected, the modular

internal circuit becomes a fuse and can be destroyed bya field short circuit or any other over-current condition

• Check modular circuit. If bad, module must be replaced


after correcting field fault.• Shut down PLC prior to changing any module – main

power and 24V power.• Locate fault in field circuit by disconnecting wires at

module and field device. Check between wires for short circuit and to ground for short circuit. Replace wire is short circuit found

• Check device for ground, short circuit, mechanical andelectrical operation, even when problem found in wires.Always check device for another fault. Problem in wirescan cause problem in device or vice versa. If devicedefective, replace device and then check total circuitbefore placing in operation and after restoring circuit,check again to ensure circuit and module are operatingcorrectly.

• Check power supply module. If no output, shut downpower and replace supply module.

• Back plane can go bad, some of the modules with powerand others with no power. Replace backplane.

• Sometimes, the PLC can be reset using the Reset keyswitch. Ensure that turning the PLC off won’t interruptother running sub-set programs, turn key switch to farright. After 15 seconds, turn to far left, wait, then returnto middle position. This operation should reset programand enable a restart.

• The PLC program can have a latch relay with no resetunder certain conditions. The key switch reset may haveno effect on the latch. Try turning the power to the PLCoff and back on. This operation may reset the latch andallow the program to be restarted.

• The PLC is usually part of a control circuit supplied with120VAC through a 460V/120V transformer as part of a system with motors, controllers, safety circuits, and other controls. Occasionally, cycling the main 480V power off/on will be necessary to try to reset all the safety and control circuits.

• Possession and use of an up-to-date ladder diagram, ele-mentary wiring diagram, manufacturer’s manuals & dia-grams, troubleshooting skills, operator’s knowledge,and time are all required to solve issues involved inmaintaining a modern manufacturing production line.


Don’t heap all the blame for a wrong measurement on theDMM (Delayed Neutron Monitor). There can be several lessobvious sources of the errors.

Testing assemblies and components usually includeschecking the continuity of connectors, wires, traces, and low-value resistive elements. Such applications typically require botha DMM and a switching system.

Many users select a DMM and switching cards based onlyon the specifications of the DMM and later are surprised to find thattheir measurements are an order of magnitude less accurate thanexpected. Many users don’t recognize the error as a system prob-lem and conclude that the DMM is not meeting its specification.

Making accurate, stable, and repeatable resistance meas-urements is an art. There is plenty of technology involved, butthe art is an important part, especially when you are measuringlow resistance values.

To achieve your accuracy goal, you need to understandthe error sources in your application. It is important to start witha good DMM. But, there are significant error sources outside theDMM, some of which may not be obvious. Things may be morecomplex than they seem, and some types of errors may be mis-interpreted.

LIMITATIONS AND ERROR SOURCESNot all materials are created equal. Most connectors and

test probes are made of beryllium-copper or phosphor-bronzematerials that closely match the electromotive force (EMF) ofcopper. For this reason, they do not cause significant thermallyinduced voltage errors.

However, relays and some other devices use nickel-ironalloys that do not match the properties of copper. These cancause significant thermal EMF errors. Thermal voltages are gen-erated when there is a mismatch of materials combined with atemperature difference.

This is the same principle that makes thermocouples workas temperature sensors. If you expect readings that are accuratewithin a few milliohms, this is a big issue. This error source alsoaffects higher value resistance measurements, but to a lesserdegree.

It is easy to overlook second-order specifications of aDMM, such as current drive levels used for resistance measure-ment. These specs may be in small print or missing, but they areimportant. For measuring low resistance, this spec tells what youcan expect from the DMM. The accuracy specs of the DMMdon’t tell the whole story. For example, the SignametricsSMX2064 PXI DMM uses a 10-mA current source, while mostother DMMs are limited to 1 mA or less. Remember Ohm’s Law:V = I x R means that 10 times the current produces 10 times thevoltage being sensed across the resistor. This larger signal is lesssusceptible to external errors and noise and provides more signalto measure.

The larger signal almost always produces a more accuratemeasurement. It is confusing to compare two DMMs havingsimilar specifications in ohms if one has 10 times the currentdrive. The two are not the same. The one with the higher currentwill perform better, especially in a system.

Good DMMs can measure signals down to a few micro-volts. If you need to measure a resistance down to a few mil-liohms, a 1-mA test current only produces 1 µV of signal per 1mΩ of resistance. In other words, you are operating right at theresolution limit of the DMM.

With a 10-mA test current, there are 10 µV of signal per 1mΩ of resistance. As a result, a DMM that uses 10x as much cur-rent for this test will give about 10x improvement in accuracy,stability, and repeatability for very low values.

If your test has serious throughput requirements and youneed to make hundreds of measurements per second, having astronger signal combined with good noise performance in theDMM makes a huge difference. Remember that the DMM’saccuracy at higher speeds may be much more important than itsbest accuracy.

TWO-WIREEveryone knows how easy it is to measure resistance

using a two-wire connection. However, for low resistance, a two-

wire connection has disadvantages (Figure 1).

Test leads frequently add >1 Ω of resistance, and your testprobe may add another 0.1 Ω of contact resistance to the meas-urement. These errors are significant if you are measuring 20 Ω.

You can eliminate most of the test-lead errors from a two-wire connection by shorting the leads and setting the Relative-Ohms control. This enables the DMM to subtract the test-leadresistance from the readings that follow. This is a very handy toolwhen you are doing manual testing, but it is less useful in anautomated test.

FOUR-WIRE ΩA four-wire connection is the standard method for meas-

uring low resistance. It eliminates the resistance of the test leads

THE ART OF MEASURING LOW RESISTANCETee Sheffer and Paul Lantz, Signametrics

Figure 1. Two-Wire


from the measurement. One pair of leads carries the test currentwhile the other pair of leads senses the voltage across the resis-tor under test (Figure 2).

The resistance of the current-carrying leads doesn’t mat-ter because they are not in the measurement path. The resistanceof the sensing leads doesn’t matter since they don’t carry anycurrent.

A four-wire connection is not immune to thermal EMFerrors caused by mismatched materials. This usually is notimportant in manual testing situations, but it is a major issue inautomated systems where a relay switch is used.

SIX-WIREWhat if the resistor you want to measure is in a circuit

with other components or resistors as in networks or on a loadedcircuit board? Then you need a six-wire guarded connection.This method makes it possible to measure resistance in situationswhere it would be impossible otherwise. The SMX2064 DMMoffers this capability (Figure 3).

A guard amplifier drives the junction of parallel compo-nents to a voltage level that prevents any of the test current fromleaking away from the resistor under test. This is a standardmethod used by large ATE in-circuit test systems. With the rightDMM, you can implement it too.

MEASURING THROUGH A SWITCH MATRIX Many applications are for production test. In these cases,

it is almost always necessary to perform multiple tests and meas-ure multiple points. You usually do this by putting a switchingcard or a matrix ahead of the DMM. It is important to note thatthe switching card can be a major source of error, particularlywhen measuring low resistance.

TWO-WIRE WITH A SWITCHING CARDHow does adding a switching card affect your two-wire

resistance measurements? Two-wire resistance measurements cer-tainly are attractive because you can fit twice as many two-wiremeasurements onto a card as you can four-wire measurements.

The economics are attractive. Perhaps you can put a shortcircuit on one of the inputs to the switching card and measurethat short to make a Relative-Ohms measurement? This line ofreasoning also might lead you to select the highest density switch-ing card possible.

However, there are reasons to be careful. A typical switch-ing card does not have the same resistance through all of itschannels. Channel 0 may add 0.2 Ω to the reading while Channel20 may add 0.8 Ω. Consequently, measuring a short on one doesnot give a good compensation for the other because they do nothave the same resistance.

Even if you could correct for the difference in channel-to-channel resistance, relays typically have about 50 mΩ of contactresistance that will shift around by 20 mΩ from one reading tothe next. You might think that high-current relays will havelower contact resistance, but it doesn’t work that way. High-cur-rent relays usually have silver-plated contacts that give lowresistance for currents above 100 mA. Unfortunately, silver-plat-ed contacts have a high and unpredictable contact resistance forcurrents less than 50 mA.

Relays are made of nickel-iron materials, and they allhave problems with thermal EMFs. Frequently, this error sourceis not specified for high-density switching cards. If not, the ther-mal voltages probably are around 100 µV. If your DMM uses 10mA to make this measurement, the switching card adds 10 mΩof error to the measurement. If your DMM uses only 1 mA, theswitch will add 100 mΩ of error to the measurement.

Keep in mind that this error voltage is made up of all ofthe closed relay contacts connected to the sense lines of theDMM. The more complex the switching system is, the higher theerror will be.

FOUR-WIRE WITH A SWITCHING CARDUsing a four-wire connection through your switching card

takes care of the resistance issues associated with the switchingcard. This accuracy improvement happens at the expense ofreduced channel density. However, it does not take care of thethermal EMF problems that come with some switching cards(Figure 4).

One way to reduce this error is to use a DMM with theOffset-Ohms function. However, this method is very slow, itadds noise, and it is limited in its capability to reduce the error.For best results, start with a high-quality switching card that isspecified for low thermal EMF error.

How big a problem are thermal EMF voltages in relayswitches? A high-quality switching card has about 10 µV while

Figure 2. Four-Wire

Figure 3. Six-Wire Guarded In-Circuit Measurement

THERMAL VOLTAGE ERRORS, NOT CORRECTED

Figure 4. Four-Wire Switching Card


a typical one has >100 µV of thermal voltages. There are a fewinstrumentation quality switches that exhibit 1µV or less.

Take a look at Figure 5 to see the effect. The yellow plotsdepict the specs of two similar DMMs. One of the DMMs uses1-mA excitation current while the other uses 10 mA. There aresome things to note:

• Both DMMs have very similar specifications as shownby the yellow lines.

• As soon as you combine them with a relay card that has10-µV offset, the system error is considerably greaterthan the DMM spec. For the DMM with 10-mA excita-tion, the system error is almost two times the DMMspec. For the DMM with 1-mA excitation, the systemerror is almost 10 times the DMM spec.

• If you combine the DMMs with a relay card that has100-µV offset, the error becomes huge. For the DMMwith10-mA excitation, the system error is almost 10times the DMM spec. For the DMM with 1-mA excita-tion, the system error is almost 100 times the DMMspec.

• The effect of the relay offset voltages overwhelms theDMM specifications in both cases, but the DMM thatuses 10-mA excitation current produces a system specbetween five and 10 times better than the DMM thatuses 1-mA excitation.

SIX-WIRE WITH A SWITCHING CARD A six-wire resistance connection works just fine with a

switching card as long as the card is organized to support it.Remember that a six-wire connection does not increase the accura-cy of your measurement unless other resistors in the circuit need tobe guarded. This is still the only way to guard-out parallel resistorsthat otherwise would make the measurement impossible.

EXAMPLESA manufacturer of semiconductor protection devices uses

an SMX2064 on its low-resistance four-wire range to accuratelymeasure resistances around 20 Ω before and after hitting thedevice with a high test voltage. Because the SMX2064 can takean accurate measurement in as little as 1 ms, test throughput ishigh.

A manufacturer of hybrid circuits uses an SMX2064 tomeasure resistance values of less than 100 mΩ. In this case,speed is not an issue, but getting a useable measurement is. OtherDMMs that use only 1-mA excitation current did not qualify todo the job.

CONCLUSIONIf you need to measure low resistance values, you benefit

by using a DMM that has a 10-mA excitation current. A 1-mAsource gives a much weaker signal to measure and presents sys-tem-level problems, particularly if there are switching cardsinvolved. If you expect a stable, accurate result, you almost cer-tainly need to use a four-wire connection.

The accuracy spec of the DMM is important but not thewhole story. Remember that everything in the measurement pathaffects the accuracy of the measurement, especially switchingcards. Your best bet is to combine a DMM with good ohms spec-ifications and high test current and a switching card with a lowthermal EMF spec, preferably an instrumentation type.

Figure 5a. 1-mA Ohms Excitation Figure 5b. 10-mA Ohms Excitation


GOALSThis project develops standard measurement techniques for

critical current, residual resistivity ratio, and magnetic hysteresis loss,and provides quality assurance and reference data for commercialhigh temperature and low-temperature superconductors. Applicationssupported include mag-netic-resonance imaging,research magnets, mag-nets for fusion confine-ment, motors, generators,transformers, high-quali-ty-factor resonant cavitiesfor particle accelerators,and superconductingbearings. Superconductorapplications specific tothe electrical powerindustry include transmission lines, synchronous condensers,magnetic energy storage, and fault-current limiters. Projectmembers assist in the creation and management of internationalstandards through the International Electrotechnical Commissionfor superconductor characterization covering all commercialapplications, including electronics. The project is currently focus-ing on measurements of variable-temperature critical current,residual resistivity ratio, magnetic hysteresis loss, critical currentof marginally stable superconductors, and the irreversible effectsof changes in magnetic field and temperature on critical current.

CUSTOMER NEEDSThis project serves the U.S. superconductor industry,

which consists of many small companies, in the development ofnew metrology and standards, and in providing difficult andunique measurements. We participate in projects sponsored byother government agencies that involve industry, universities,and national laboratories.

The potential impact of superconductivity on electricpower systems, alternative energy sources, and research magnetsmakes this technology especially important. We focus on: (1)developing new metrology needed for evolving, large-scalesuperconductors, (2) providing unique databases of supercon-ductor properties, (3) participating in interlaboratory compar-isons needed to verify techniques and systems used by U.S.industry, and (4) developing international standards for super-conductivity needed for fair and open competition and improvedcommunication.

Electric power grid stability, power quality, and urbanpower needs are pressing national problems. Superconductiveapplications can address many of them in ways and with effi-ciencies that conventional materials cannot. “Second-genera-tion” Y-Ba- Cu-O (YBCO) superconductors are approaching the

targets established by the U.S. Department of Energy. The demon-stration of a superconductor synchronous condenser for reactivepower support was very successful and has drawn attention to thepromise of this technology. Previous demonstration projects hadinvolved first-generation materials, Bi-Sr-Ca-Cu-O (BSCCO).Variable-temperature measurements of critical current and mag-netic hysteresis loss will become more important with these ACapplications, and methods for reducing losses are expected toevolve as second-generation materials become commercial.

Fusion energy is a potential, virtually inexhaustible energysource for the future. It does not produce CO2 and is environmen-tally cleaner than fission energy. Superconductors are used to gen-erate the ultrahigh magnetic fields that confine the plasma infusion energy research. We measure the magnetic hysteresis lossand critical current of marginally stable, high-current Nb3Snsuperconductors for fusion and other research magnets. Becauseof the way superconductors are used in magnets, variable-temper-ature critical-current measurements are needed for more completecharacterization.

The focus of high-energy research is to probe and under-stand nature at the most basic level, including dark matter anddark energy. The particle accelerator and detector magnets need-ed for this fundamental science continue to push the limits ofsuperconductor technology. The next generation of Nb3Sn andNb-Ti wires is pushing towards higher current density, less sta-bilizer, larger wire diameter, and higher magnetic fields. Theresulting higherectronics and Electrical Engineering Laboratorycurrent required for critical-current measurements turns manyminor measurement problems into significant engineering chal-lenges. For example, heating of the specimen, from many sources,during the measurement can cause a wire to appear to be thermallyunstable. Newer MgB2 wires may be used for specialty magnetsthat can safely operate at the higher temperatures caused by highheat loads. We need to make sure that our measurements and themeasurements of other laboratories keep up with these challengesand provide accurate results for conductor development, evaluation,and application.

Possible spin-off applications of particle accelerators areefficient, powerful light sources and free-electron lasers for bio-medicine and nanoscale materials production. The heart of theseapplications is a linear accelerator that uses high-efficiency, pureNb resonant cavities. We conduct research on a key materialsproperty measurement for this application, the residual resistivi-ty ratio (RRR) of the pure Nb. This measurement is difficultbecause it is performed on samples that have dimensions similarto those of the application. Precise variable-temperature meas-urements are needed for accurate extrapolations.

TECHNICAL STRATEGYInternational Standards – With each significant advance

in superconductor technology, new procedures, interlaboratory

STANDARDS FOR SUPERCONDUCTOR ANDMAGNETIC MEASUREMENTS

National Institute of Standards and Technology

Probe for the measurement of the critical current of asuperconductor wire as a function of temperature. Theprobe is inserted into the bore of a high field super-conducting magnet.


comparisons, and standards are needed. International standardsfor superconductivity are created through the InternationalElectrotechnical Commission (IEC), Technical Committee 90 (TC90).

Critical Current Measurements – One of the mostimportant performance parameters for large-scale superconductorapplications is the critical current. Critical current is difficult tomeasure correctly and accurately; thus these measurements areoften subject to scrutiny and debate. The critical current is deter-mined from a measurement of voltage versus current. Typical cri-teria are electric-field strength of 10 microvolts per meter andresistivity of 10–14 ohm-meters.

Critical-current measurements at variable temperaturesare needed to determine the temperature margin for magnetapplications. The temperature margin is defined as the differencebetween the operating temperature and the temperature at whichcritical current Ic is equal to the operating current. When a mag-net is operating, transient excursions in magnetic field H or cur-rent I are not expected; however, many events can cause tran-sient excursions to higher temperatures T, such as wire motion,AC losses, and radiation. Hence the temperature margin of awire is a key specification in the design of superconducting mag-nets. Variable-temperature critical-current measurements requiredata acquisition with the sample in a flowing gas environmentrather than immersed in a liquid cryogen. Accurate high-current(above 100 amperes) measurements in a flowing gas environ-ment are very difficult to perform.

Residual Resistivity Ratio Measurements – The RRRis defined as the ratio of electrical resistivity at two temperatures:

273 kelvins (0 degrees Celsius) and 4.2 kelvins (the boiling pointof liquid helium). The value of RRR indicates the purity and thelow-temperature thermal conductivity of a material, and is oftenused as a materials specification for superconductors. The lowtemperature resistivity of a sample that contains a superconductoris defined at a temperature just above the transition temperatureor is defined as the normal-state value extrapolated to 4.2 kelvins.For a composite superconducting wire, RRR is an indicator ofthe quality of the stabilizer, which is usually copper or aluminumthat provides electrical and thermal conduction during conditionswhere the local superconductor momentarily enters the normalstate. For pure Nb used in radio-frequency cavities of linearaccelerators, the low temperature resistivity is defined as the nor-mal-state value extrapolated to 4.2 kelvins. This extrapolationrequires precise measurements. We have studied some funda-mental questions concerning the best measurement of RRR andthe relative differences associated with different measurementmethods, model equations for the extrapolation, and magneticfield orientations (when a field is used to drive the superconductorinto the normal state).

Magnetic Hysteresis Loss Measurements – As part ofour program to characterize superconductors, we measure themagnetic hysteresis loss of marginally stable, high-current Nb3Snsuperconductors for fusion and particle-accelerator magnets. We usea magnetometer based on a superconducting quantum interfer-ence device (SQUID) to measure the magnetic hysteresis loss ofsuperconductors, which is the area of the magnetization-versus-field loop. In some cases, especially for marginally stable con-ductors, we use special techniques to obtain accurate results.Measurement techniques developed at NIST have been adoptedby other laboratories.

ACCOMPLISHMENTS

• Superconductor Data Enables U.S. Company toOffer Product to Korean Project – New bismuth-based high-temperature superconductor wires are under active considerationfor a 600 kilojoule superconducting magnetic energy storage(SMES) project lead by the Korea Electrotechnology ResearchInstitute. The purpose of the SMES system is to stabilize theelectric power grid. The magnet will be wound with 10-kilo-ampere superconducting cables composed of many round wires.It will be cooled to 20 kelvins by cryocoolers. A U.S. companyturned to us for critical current measurements at 20 kelvins to

Illustration of a superconductor’s voltage-current characteristic with two common criteriaapplied.

Electric field versus current at temperatures from 7.0 to 8.3 kelvins in steps of 0.1 kelvinsfor a Nb3Sn wire.These are typical curves for the determination of critical current.

Critical current versus temperature of a high-Tc Bi2Sr2CaCu2O8+x wire at various magneticfields. Such curves are used to determine the safe operating current at different tempera-tures and fields.


determine whether its conductor could meet the project’s speci-fications for critical current. Critical current, the largest current asuperconducting wire can carry, is a key performance and designparameter. Critical current depends on temperature, magneticfield and, in many cases, the angle of the magnetic field withrespect to the conductor.

We made variable-temperature critical-current measure-ments on three wire specimens in magnetic fields up to 8 teslas,at various magnetic-field angles, and at temperatures from 4 to30 kelvins. NIST has the only such multiparameter, high-current,variable-temperature measurement capability in the U.S. Thelargest current applied to the 0.81 millimeter diameter wire sam-ples was 775 amperes.

The results showed that the angle dependence of criticalcurrent for the wires was less than just 3 percent over the usefulrange of field and temperature, and that the round wires could beused at higher magnetic fields and temperatures than tape con-ductors. These data will be used to design the safe operating lim-its of the SMES magnet system.

• Key Measurements for the International ThermonuclearExperimental Reactor – Superconducting magnets are used infusion energy projects such as the International ThermonuclearExperimental Reactor (ITER), to confine and heat the plasma.The superconductors for ITER’s large magnet systems are all“cable-in-conduit conductors” (CICC), which provide bothmechanical support for the large magnetic forces and a flow pathfor the liquid helium required to cool the cable. The supercon-ducting magnet must be operated below the critical current of thecable, which is a function of magnetic field and temperature.Temperature is an important variable, and the local temperatureof the conductor depends on the mass-flow rate of the coolantand the distribution of the heat load along the CICC.

We designed and constructed a new variable temperatureprobe that allows us to make measurements in our 52-millimeterbore, 16-tesla magnet. This probe replaces one that was designedfor our 86-millimeter bore, 12-tesla magnet. Fitting everythinginto the smaller bore was difficult, but the new probe performedas expected and allows us to make measurements at the ITERdesign field of 13 teslas. We made measurements up to 765 ampereswith a Nb3Sn sample in flowing helium gas. Measurements weremade at temperatures from 4 to 17 kelvins and magnetic fields from0 to 14 teslas. Some measurements were made at 15 and 16 teslasfor temperatures from 4 to 5 kelvins; however, these magnetic fieldscan be generated only when a sample is measured in liquid helium.

The results of our unique variable-temperature measurements pro-vide a comprehensive characterization and form a basis for evaluat-ing CICC and magnet performance. We used these data to generatecurves of electric field versus temperature at constant current andmagnetic field. In turn, these give a direct indication of the temper-ature safety margin of the conductor.

• International Standards on Superconductivity –Many of the 14 published IEC/TC 90 standards on superconduc-tivity contain “precision” and “accuracy” statements rather thancurrently accepted statements of “uncertainty.” NIST has advo-cated that TC 90 adopt a more modern approach to uncertainty.In collaboration with the Information Technology Laboratory,we have developed a 50-page report on the possibility of chang-ing statements of “accuracy” to statements of “uncertainty” inIEC/TC 90 measurement standards, which was presented at TC90 meetings in June 2006. They included proposed change sheetsfor 13 of the 14 TC 90 document standards. Ultimately, all TC 90delegates voted in favor of changing to uncertainty statementsduring the maintenance cycle of existing standards and duringthe development of new standards.

• Current Ripple a Source of Measurement Errors –All high-current power supplies contain some current ripple andspikes. New high-performance conductors have high critical cur-rents that require current supplies over 1000 amperes. High-cur-rent power supplies with the lowest level of current ripple andspikes are often more than a factor of ten times more expensivethan conventional supplies. In addition, current ripple and spikesare a greater problem for short-sample critical current testingthan for magnet operation because of the smaller load induc-tance. Therefore, we need to understand the effects of ripple andspikes on the measured critical current (Ic) and “n-value”, theindex of the shape of the electric field-current curve. We focusedon how ripple changes the n-value and showed that, in terms ofpercentage change, the effect of ripple on n-value was about 7times that on Ic Interlaboratory comparisons often show varia-tions in n-value much larger than the variations in Ic. We exam-ined models and use the measurements on simulators to attemptto reproduce and understand the effects observed in measure-ments on superconductors. We believe that current ripple andspikes are sources of differences in n-values measured at differ-ent laboratories.

• New Method to Evaluate the Relative Stability ofConductors – We recently started measuring voltage versusmagnetic field (V-H) on Nb3Sn wires to assess their relative sta-

Critical current versus temperature at various magnetic fields for a Nb3Sn wire. Thesecurves show the current carrying limits for various combinations of temperature and mag-netic field.

Electric field versus temperature at currents from 66 to 84 amperes in steps of 1.5 amperesfor a Nb3Sn wire. These are typical curves for the determination of temperature margin.


bility. Voltage versus current (V-I) at constant field is usuallymeasured to determine Ic. Low-noise V-H measurements weremade at constant or ramping current with the same electronicinstruments, apparatus, and sample mount as used in Ic measure-ments. High-performance Nb3Sn wires exhibit flux-jump insta-bilities at low magnetic fields, and low-noise V-H curves onthese wires show indications of flux jumps. V-H measurementsalso reveal that less stable wires will quench (abruptly and irre-versibly transition to the normal state) at currents much smallerthan Ic at the lower magnetic fields. This new method needs tobe further understood and may be standardized to ensure that itprovides accurate and reliable data.

STANDARDS COMMITTEES• Loren Goodrich is the Chairman of IEC/TC 90, the U.S.

Technical Advisor to TC 90, the Convener of Working Group 2(WG2) in TC 90, the primary U.S. Expert to WG4, WG5, WG6and WG11, and the secondary U.S. Expert to WG1, WG3, andWG7.

• Ted Stauffer is Administrator of the U.S. Technical AdvisoryGroup to TC 90.

STANDARDSIn recent years, we have led in the creation and revision of

several IEC standards for superconductor characterization:

• IEC 61788-1 Superconductivity – Part 1: Critical CurrentMeasurement – DC Critical Current of Cu/Nb-Ti CompositeSuperconductors

• IEC 61788-2 Superconductivity – Part 2: Critical CurrentMeasurement – DC Critical Current of Nb3Sn CompositeSuperconductors

• IEC 61788-3 Superconductivity – Part 3: Critical CurrentMeasurement – DC Critical Current of Ag-sheathed Bi-2212and Bi-2223 Oxide Superconductor

• IEC 61788-4 Superconductivity – Part 4: Residual ResistanceRatio Measurement – Residual Resistance Ratio of Nb-TiComposite Superconductors Critical current vs. temperature ofa Bi-2212 tape at a magnetic field of 0.5 tesla and various mag-netic field angles. Such curves are used to determine the safeoperating current at various temperatures and field angles.

• IEC 61788-5 Superconductivity – Part 5: Matrix to SuperconductorVolume Ratio Measurement – Copper to Superconductor VolumeRatio of Cu/Nb-Ti Composite Superconductors

• IEC 61788-6 Superconductivity – Part 6: MechanicalProperties Measurement – Room Temperature Tensile Test ofCu/Nb-Ti Composite Superconductors

• IEC 61788-7 Superconductivity – Part 7: Electronic CharacteristicMeasurements – Surface Resistance of Superconductors atMicrowave Frequencies

• IEC 61788-8 Superconductivity – Part 8: AC Loss Measurements– Total AC loss Measurement of Cu/Nb-Ti CompositeSuperconducting Wires Exposed to a Transverse AlternatingMagnetic Field by a Pickup Coil Method

• IEC 61788-10 Superconductivity – Part 10: CriticalTemperature Measurement – Critical Temperature of Nb-Ti,Nb3Sn, and Bi-System Oxide Composite Superconductors bya Resistance Method

• IEC 61788-11 Superconductivity – Part 11: ResidualResistance Ratio Measurement – Residual Resistance Ratio ofNb3Sn Composite Superconductors

• IEC 61788-12 Superconductivity – Part 12: Matrix toSuperconductor Volume Ratio Measurement – Copper to Non-Copper Volume Ratio of Nb3Sn Composite SuperconductingWires

• IEC 61788-13 Superconductivity – Part 13: AC LossMeasurements – Magnetometer Methods for Hysteresis Loss inCu/Nb-Ti Multifilamentary Composites

• IEC 60050-815 International Electrotechnical Vocabulary –Part 815: Superconductivity


Modular up to six channels• High linearity• Low offset• High bandwidth up to 1MHz• Extremely low phase shift• High CMR due to galvanic insulation• Transducer heads from 200A to 5000A• Current and voltage output• Optimised for power electronics needs

CURRENT ANALYSIS IN POWER ELECTRONICSAPPLICATIONS

To optimise power electronics components like invertersor drives, the electrical signals current and voltage must be meas-ured with very high accuracy. The quality of the measurementresults depends on linearity, offset, and width and phase shift ofused instruments and connected current and voltage sensors.

Standard current transformers have a limited bandwidth,fast impulse transducers and very low accuracy. In addition thesetransducers are not able to measure DC components in the signalor do not work at all at DC since the iron core gets saturated.Wideband-MHz-shunts are accurate and fast but do not allow agalvanic isolated connection of the instrument from the powerelectronics circuit. Also, high common mode signals result indisturbances and inaccuracy.

The measurement of electrical power and the calculation oflosses are even more problematic. At low power factor linearity,offset and phase shift of the transducers have a much higher influ-ence on the resulting error than the instrument itself. Losses arenormally calculated by the subtraction of the output power fromthe input power. The efficiencies of power electronics componentsare quite high, and therefore the actual losses compared to themeasured input or output power are very small. Consequently, theerror of measured input or output power can easily be as high asthe loss itself if the sensor is not accurate enough.

MULTI CHANNEL SYSTEM MCS FROM DANFYSIKThe new multi channel systems MCS from DANFYSIK

combine highest accuracy and bandwidth with lowest phase shiftand common mode influence. The systems measure AC signalsas well as DC signals with a linearity in the ppm range and workup to 1MHz. Different current transducer heads from 200Apk to5000Apk are available as standard transducers.

Optimised systems for the needs of power electronics anddrives applications

The modularity of the MCS systems allows all types ofanalyses in the power electronics field to be connected. Many ofthe measurements are made on three-phase outputs of frequencyinverters or three phase sinus inputs of electric motors. Toanalyse a complete frequency inverter, a six-phase system is nec-essary. Automotive applications normally measure the DC fromthe battery in addition to the three-phase output to the load. TheMCS systems can be ordered from three to six channels. A three-phase system can be updated with additional channels easily.

ZERO-FLUX-PRINCIPLEThe transducer consists of a transducer head and an elec-

tronic module. In the transducer head there are three iron coreswith a common secondary winding but with separate auxiliarywindings. The primary current lp, via the winding Lp, produces amagnetic field in the three iron cores of the transducer. Thereby Lpmostly consists of the primary conductor, which is lead throughthe transducer. The compensation current lc compensates the mag-netic field of the primary current and provides a steady zero-fluxin the iron core. This compensation current is driven by an opera-tion amplifier to which both inputs are connected with a signalwhich is proportional with the AC- and DC-component of the pri-mary conductor current. The AC-component is thus induced intothe auxiliary winding Lh1. The DC-component and the very low-frequency com-ponent comesfrom the so-called Zero-Flux-Detector (sym-metry detector).Via an oscillatorand the auxiliarywindings Lh2and Lh3, theother two ironcores are driveninto saturation indifferent direc-tions.

MULTI CHANNEL CURRENT TRANSDUCER SYSTEMSDANFYSIK


Both iron cores and the auxiliary windings Lh2 and Lh3are built identically. The currents via Lh2 and Lh3 are thus iden-tical. In this case, the main core flux is zero.

A direct current via the primary conductor results in a fluxvia the core. Therefore, both Zero-Flux-Detector cores can nolonger be driven into saturation identically, and the two currentsvia Lh2 and Lh3 are no longer equal. The difference between thecurrents is proportional with the DC component of the current lp.The Zero-Flux-Detector processes this signal and leads it to theDC-input of the operation amplifier which drives the compensa-tion current. This way the DC-component of the primary currentcan also be compensated. The compensation current is an accuratereproduction of the primary current, and can be evaluated as a gal-vanic separated signal by all types of measuring instruments. Theburden resistor is only to be used if the measuring instrument onlyhas voltage inputs. The advantage of this technology is mostly thehigh accuracy of the transducer. The sensitivity of the Zero-Flux-Detector and of the iron cores allows the best possible ppm-accu-racy. A transducer bandwidth of a few hundred kHz can easily beobtained.


On April 14, 2002, a ground resistance test was conduct-ed to compare the results obtained from the Fall-of-Potential 3-Point testing method to the clamp-on testing method. Thegrounding system consisted of four copper clad rods installed inan approximate 20 ft square. Three of the rods are 5/8" in diameterand 10 ft in length. The fourth rod is 1/2" in diameter and 8 ft inlength. All rods were coupled together with 3-gauge aluminumwire. Figure 1 shows the schematic of the system.

The tests were conducted with the following equipmentmanufactured by AEMC instrument:

• Model 4500, 4-Point Ground Resistance Tester• Model 4630, 4-Point Ground Resistance Tester• Model 3731, Clamp-On Ground Resistance Tester.Additionally, we used the AEMC Model 5600, a

micro-ohmmeter to verify the bonding of the aluminumwire to the individual ground rods.

The soil conditions in the test area were predomi-nately loam with some gravel. Conditions on the day of thetest were dry and sunny, some light rain had occurred theprevious day to the test. Therefore, the soil was somewhatmoist at the surface.

The AEMC Model 5600 Micro-Ohmmeter was usedto measure bonding resistance at each rod and was the firsttest completed. Measurements from each conductor to therod were taken as well as measurements from conductor to

conductor through the rod and clamp. Readings on rod numberthree ranged from 615 to 733mΩ at each bonding point, indicat-ing that all connections were good. See Figure 2 for full results.

Measurement Point Resistance (µOhms)

A to B 713

C to B 615

A to C 733

In the first test, the AEMC Model 4500 was used as 3-Point ground tester. Rod number three was first disconnectedfrom the other rods in the system so that its individual resistancecould be measured. The X lead was attached to rod number three(see Figure 3). The Z lead was attached to an auxiliary electrode100 feet away and the Y lead was initially connected to the aux-iliary electrode 60 feet away. Readings were taken with the Yelectrode at 90, 80, 70, 60, 50, 40, 30, 20 and 10 feet. Figure 3shows the results of this test.

FALL-OF-POTENTIAL GROUND TESTING, CLAMP-ONGROUND TESTING COMPARISON

Chauvin Arnoux, Inc.

Figure 1. The Grounding System

Figure 2. Bonding resistance measurements

Figure 3. Three-Point test connection


The same test was repeated using the AEMC Model 4630fall-of-potential ground tester. The results are shown in Figure 5.

Finally, the AEMC Model 3731 was used to meas-ure the resistance at rod number three with all other rodsdetached from it. A temporary cable was installed betweenrod number three and the municipal grounding system,thus setting up the required parallel paths necessary foraccurate measurement using a clamp-on ground tester (seeFigure 6). Under these conditions, the reading was 84.5Ω.

The results of these tests showed that the clamp-onground tester is indeed an effective tool in measuringground resistance when used under the proper conditions.Readings between the clamp-on ground testing and thefall-of-potential ground testing method correlate. Theadvantages of using the clamp-on tester were the ability totest without disconnecting the rod from service and theability to test without the need for auxiliary ground elec-trodes. These two points saved considerable amount oftime in conducting the test

Figure 4. Model 4500 test results

Figure 5. Model 4630 test results

Figure 6. Single rod test using the Model 3731 clamp-on ground resistance tester


INTRODUCTION By definition, all of today’s wireless communication sys-

tems contain one key element, an antenna of some form. Thisantenna serves as the transducer between the controlled energyresiding within the system and the radiated energy existing infree space. In designing wireless systems, engineers must choosean antenna that meets the system’s requirements to firmly closethe link between the remote points of the communications system.While the forms that antennas can take on to meet these systemrequirements for communications systems are nearly limitless,most antennas can be specified by a common set of performancemetrics.

ANTENNA PERFORMANCE METRICS In order to satisfy the system requirements and choose a

suitable antenna, system engineers must evaluate an antenna’sperformance. Typical metrics used in evaluating an antennaincludes the input impedance, polarization, radiation efficiency,directivity, gain and radiation pattern.

INPUT IMPEDANCE Input impedance is the parameter which relates the antenna

to its transmission line. It is of primary importance in determiningthe transfer of power from the transmission line to the antenna andvice versa. The impedance match between the antenna and thetransmission line is usually expressed in terms of the standingwave ratio (SWR) or the reflection coefficient of the antenna whenconnected to a transmission line of a given impedance. The reflec-tion coefficient expressed in decibels is called return loss.

POLARIZATION The polarization of an antenna is defined as the polariza-

tion of the electromagnetic wave radiated by the antenna along avector originating at the antenna and pointed along the primarydirection of propagation. The polarization state of the wave isdescribed by the shape and orientation of an ellipse formed bytracing the extremity of the electromagnetic field vector versustime. Although all antennas are elliptically polarized, most anten-nas are specified by the ideal polarization conditions of circular orlinear polarization.

The ratio of the major axis to the minor axis of the polar-ization ellipse defines the magnitude of the axial ratio. The tiltangle describes the orientation of the ellipse in space. The senseof polarization is determined by observing the direction of rota-tion of the electric field vector from a point behind the source.Right-hand and left-hand polarizations correspond to clockwiseand counterclockwise rotation respectively.

DIRECTIVITY It is convenient to express the directive properties of an

antenna in terms of the distribution in space of the power radiatedby the antenna. The directivity is defined as 4p times the ratio ofthe maximum radiation intensity (power radiated per unit solidangle) to the total power radiated by the antenna. The directivityof an antenna is independent of its radiation efficiency and itsimpedance match to the connected transmission line.

GAIN The gain, or power gain, is a measure of the ability to con-

centrate in a particular direction the net power accepted by theantenna from the connected transmitter. When the direction is notspecified, the gain is usually taken to be its maximum value.Antenna gain is independent of reflection losses resulting fromimpedance mismatch.

RADIATION EFFICIENCY The radiation efficiency of an antenna is the ratio of the

power radiated by the antenna to the net power accepted at itsinput terminals. It may also be expressed as the ratio of the max-imum gain to the directivity.

RADIATION PATTERN Antenna radiation patterns are graphical representations

of the distribution of radiated energy as a function of directionabout an antenna. Radiation patterns can be plotted in terms offield strength, power density, or decibels. They can be absoluteor relative to some reference level, with the peak of the beamoften chosen as the reference. Radiation patterns can be dis-played in rectangular or polar format as functions of the sphericalcoordinates q and f. A typical antenna pattern in a rectangularformat is shown below1.

AN INTRODUCTION TO ANTENNA TEST RANGES,MEASUREMENTS AND INSTRUMENTATION

The basic principles of antenna test and measurement are discussed along with an introduction to various range geometries, and instrumentation

Jeffrey A. Fordham, Microwave Instrumentation Technologies, LLC.


ANTENNA RANGE SITING CONSIDERATIONS The choice of an antenna test range is dependent on many

factors, such as the directivity of the antenna under test, frequen-cy range and desired test parameters. Often the mechanical fea-tures of the antenna (size, weight and volume) can have as muchinfluence on the selection of an antenna range as do the electricalperformance factors. In selecting an antenna range to evaluateantenna performance, care must be taken to ensure the perform-ance metrics are measured with sufficient accuracy.

A few of the more com-monly used antenna test rangesare shown here. Regardless ofthe chosen test range, three keyfactors must be addressed andcontrolled to ensure a successfulmeasurement. These factors arethe phase variations of the inci-dent field, the amplitude varia-tions of the incident field and the stray signals created by reflec-tions within the test range.

VARIATIONS OF THE PHASE OF THE INCIDENT FIELD In order to accurately measure an antenna’s far zone per-

formance, the deviation of the phase of the field across its aperturemust be restricted. The criterion generally used is that the phaseshould be constant to within p/8 radian (22.5°). Under normaloperating conditions, this criteria is easily achieved since there isusually a large separation between transmitting and receivingantennas. During antenna testing, it is desirable because of variouspractical considerations to make antenna measurements at as shorta range as possible. Since the measurements must simulate theoperating situation, it is necessary to determine the minimum sep-

aration between the transmitting antenna and the receivingantenna for a reasonable approximation of the far field gain andradiation patterns. At distances from a transmitting antenna,which are large compared with the antenna dimensions, thephase front of the emergent wave is nearly spherical in shape.For extreme separations, the radius of curvature is so large that forall practical purposes the phase front can be considered planarover the aperture of a practical antenna. As the antennas arebrought closer together, a condition is reached in which, becauseof the short radius of curvature, there is an appreciable separationD between the wavefront and the edges of the antenna aperture.

A criterion that is commonly employed in determining theminimum permissible value of R is to hold D to a maximum of1/16 wavelength (equivalent to 22.5° of phase variation). If thiscondition is met, the receiving antenna is said to be in the farfield of the transmitting antenna. The mathematical expressionfor this minimum range:

The major effect of a small deviation D is to produceminor distortions of the sidelobe structure. Larger values of Dwill cause appreciable errors in the measured gain and lobe struc-ture. Conversely, this condition can mask asymmetrical sidelobestructures which are actually present.

VARIATIONS OF THE AMPLITUDE OF THE INCIDENT FIELD A second and important siting consideration is the variation

of the amplitude of the incident field over the aperture of the testantenna. Excessive variations in the field will cause significanterrors in the measured maximum gain and sidelobe structure. Thiseffect can be seen better from the viewpoint of reciprocity.Variations in the amplitude of the field over the aperture on receivingare analogous to the transmitting case of a modification of theaperture illumination by the primary feed. If the variation acrossthe antenna under test is limited to about 0.5 dB, error in themeasurements will be negligible for most applications. It is essen-tial that the transmitting antenna be accurately directed so that thepeak of its beam is centered on the aperture of the antenna undertest. Improper alignment, which may not cause a noticeable loss

Rectangular Anechoic Chamber Compact Antenna Test Range

Outdoor Elevated Range Ground Reflection Range

Planar Near-Field Cylindrical Near-Field

Spherical Near-Field

Spherical Phase Front Tangent to a Plane Antenna Aperture


of signal level, results in an asymmetrical aperture illuminationand error in the measurement of the sidelobe structure.

INTERFERENCE FROM REFLECTIONS The requirement of providing adequate separation between

antennas to prevent excessive phase error makes it difficult to sat-isfy a further requirement that the site be free of large reflectionsfrom the ground or other sources of reflection. Addition of reflectedfields at the test antenna can produce erroneous gain and patternmeasurements. For instance, an interfering field which is 30 dBbelow the direct path signal can cause a variation of ±0.25 dB inthe measured maximum gain and can seriously affect the meas-ured sidelobe structure of the pattern. The usual method of mini-mizing the effects of fields caused by reflections are to (1) mountthe transmitting antenna and test antenna sites on towers, (2)employ a directive transmitting antenna, (3) avoid smooth surfaceswhich are oriented so that they produce direct reflection into thetest antenna, and (4) erect screens or baffles to intercept the reflectedwave near the reflection point.

An alternate procedure is to locate the transmitting andreceiving antennas over a flat range and to take into account thespecular reflection from the ground in making measurements.The heights of the antenna under test and the transmitting antennaare adjusted for a maximum of the interference pattern betweenthe direct and ground reflected wave. Generally, it is more conven-ient to mount the test positioner and antenna on a fixed heighttower or building and vary the height of the transmitting antenna.This can be accomplished with the transmitting antenna mountedto a motor driven elevator/carriage assembly that can travel up anddown a tower.

In cases where the antenna range length is reasonablyshort, the entire range can be housed indoors in an anechoicchamber. An indoor far-field anechoic chamber has the samebasic design criteria as an outdoor range except that the surfacesof the room are covered with RF absorbing material. Thisabsorber is designed to reduce reflected signal over its designfrequency range. Testing indoors offers many advantages to con-ventional outdoor ranges including improved security, avoidingunwanted surveillance and improved productivity due to lesstime lost because of weather and other environmentally relatedfactors. The advantages of testing indoors are primarily respon-sible for the trend toward more advanced test ranges such as thecompact range and near-field ranges.

AMPLITUDE VARIATION – ELEVATED RANGES Variations in the amplitude of the field incident over a test

aperture must also be restricted for accurate far-zone measure-ments. For range geometries employing comparatively largetransmitting and test tower heights (i.e., elevated range geome-tries), it is advisable to restrict amplitude taper to the order of 1/4dB or less by using the following criterion:

From the viewpoint of suppressing range surface reflec-tions, it is also desirable to maintain the test height H, greaterthan or equal to 6D. If one must, for practical reasons, employtest heights less than approximately 4D, the ground-reflectiontechnique should be considered.

GROUND-REFLECTION ANTENNA TEST RANGES Ground-reflection antenna range geometries are often

advantageous when the test situation involves low directivitiesand high accuracy requirements or when practical test heightsare less than approximately four times the maximum verticaldimension of the test aperture. In this technique specular reflec-tion from the range surface is caused to create constructive inter-ference with the direct-path energy in the region of the test aper-ture, such that the peak of the first interference pattern lobe iscentered on the test aperture. Four basic criteria are applicable toground-reflection range geometries:

COMPACT ANTENNA TEST RANGES Compact ranges are an excellent alternative to traditional

far-field ranges. Any testing that can be accomplished on a far-field range can be accomplished on a compact antenna test range.This method of testing allows an operator to employ an indooranechoic test chamber at a reasonable cost and avoid the problemsassociated with weather and security often encountered whenusing an outdoor test range. In a research and development situ-ation, the small size of a compact range allows it to be locatedconvenient to the design engineers. In a manufacturing environ-ment, the compact range can be located near to the final testingand integration facilities. By placing a compact range in a shieldedchamber, one can eliminate interference from external sources.This last feature has become more important in the last severalyears as the proliferation of cell phone and wireless systems hascreated a background noise environment which has made anten-na testing in a quiet electromagnetic environment more difficult

The principle of operation of a compact range is based onthe basic concepts of geometrical optics. Diverging spherical wavesfrom a point source located at the focal point of a paraboloidal sur-face are collimated into a plane wave. This plane wave is incidenton the test antenna. The resultant plane wave has a very flat phasefront, but the reflector-feed combination introduces a small (butgenerally acceptable) amplitude taper across the test zone.

In principle, the operation of a compact range is straight-forward; however, its ultimate design, construction, and installationshould be carefully considered.


NEAR-FIELD ANTENNA TEST RANGES Near-field ranges are used where large antennas are to be

tested indoors in a relatively small space. This type of range uses asmall RF probe antenna that is scanned over a surface surroundingthe test antenna. Typically, separation between the probe and theantenna structure is on the order of 4 to 10 wavelengths. During themeasurement, near-field phase and amplitude information is col-lected over a discrete matrix of points. This data is then transformedto the far-field using Fourier techniques. The resulting far-field datacan then be displayed in the same formats as conventional far-fieldantenna measurements.

In addition to obtaining far-field data, Fourier analysistechniques are used to back-transform the measured electromag-netic field to the antenna’s aperture to produce aperture field distribution information. This offers the ability to perform elementdiagnostics on multi-element phased arrays.

In near-field testing, the test antenna is usually aligned tothe scanner’s coordinate system and then either the probe or thetest antenna is moved. In practice, it is easier and more costeffective to scan the RF probe over linear axes or the test anten-na over angular axes. But this does not have to be the case. Thereare many scanning coordinate systems possible for collecting thenear-field data. Three techniques are in common usage:

Planar Near-Field Method – With planar near-fieldscanning, the probe usually is scanned in X and Y linear coordi-nates over the aperture of the test antenna. A large planar scan-ner is used to move the probe over a very accurate plane locatedin front of the test antenna’s aperture. Once aligned to the scanplane, the test antenna is not moved during the collection of thenear-field data. Planar near-field provides limited angular cover-age of the test antenna’s field due to the truncation caused by thescanner’s dimensions.

Cylindrical Near-Field Method – For this method theprobe typically is scanned in one linear dimension using a singleaxis linear positioner. The test antenna is stepped in angle on arotary axis oriented parallel to the linear axis. The resulting scandescribes a cylindrical surface around the test antenna. Cylindricalnear-field scanning can provide complete angular coverage of thetest antenna’s field in one plane. The orthogonal plane has limitedangular coverage due to truncation caused by the finite length ofthe linear scanner.

Spherical Near-Field Method – Spherical near-fieldscanning normally involves installing the test antenna on a spher-ical scanning positioner. The probe antenna is normally fixed inspace. The test antenna is normally scanned in one angular axisand stepped in an orthogonal angular axis. The resulting data iscollected over a spherical envelope surrounding the test antenna.Full or nearly-full coverage of the test antenna’s radiating fieldcan be evaluated with this type of near-field system.

PULSED ANTENNA MEASUREMENTS Characterizing antennas under pulsed RF condition is

becoming increasingly commonplace. Advanced radar and wire-less systems and their enabling technologies such as monolithicmicrowave integrated circuits (MMICs) require testing methodsto verify performance over a wide range of operating parameters.In addition to the pulse parameters, the major factors influencingpulsed RF testing include high transmit power levels, thermalmanagement of the antenna under test (AUT) and its supportingequipment in the test environment, and interfacing to a highly

integrated antenna assembly with its associated transmitting andcontrol circuitry. Due to these issues, pulsed RF operation pres-ents an additional set of test problems not often encountered inCW operation. As a result, instrumentation complexity increasesand measurement system timing issues become critical.

The basic pulsed antenna test parameters are identical tothose encountered in CW measurements. Gain, sidelobe levels,pointing accuracy, beamwidth, null locations and depths, andpolarization parameters are essential to fully characterizing anantenna. In addition to the traditional time invariant antenna per-formance parameters, some new time dependent parametersemerge when testing under pulsed conditions. These include tran-sient effects such as beam formation and distortions as a functionof time within a pulse or over an ensemble of pulses, power output(i.e. gain) as a function of time within a pulse or pulse burst, etc.

Compounding these measurements is the additional burdenof multi-channel, multi-frequency, and multi-state measurementsas a function of pulse repetition frequency (PRF), duty factor (DF)and operating frequency. Due to the increasingly integrated natureof antennas with their transmitters, the measurement system mustbe responsive to external RF pulse generation and timing for bothsingle and multiple pulse measurements.

ANTENNA RANGE INSTRUMENTATION Regardless of the type of antenna range to be implemented,

the complement of instruments to operate the range is very sim-ilar. Differences occur due to the location of the various instru-ments with respect to the source and test antennas, types ofmeasurements to be performed and the degree of automationdesired. A description of the basic instrumentation subsystemsand typical applications of different types of antenna ranges, willbe presented here.

The instrumentation for measuring antenna patterns con-sists of four subsystems, which can be controlled from a centrallocation. These subsystems are:

1. Positioning and Control 2. Receiving3. Signal Source4. Recording and Processing The test antenna is installed on a positioner and is usually,

tested in the receive mode. The motion of the positioner (rotationof the test antenna) is controlled by a positioner control unit locat-ed in the control room. The positioner is equipped with synchrotransmitters or high accuracy encoders to provide angle data forthe position indicator and the recording/processing subsystem.

To process the received signal for recording, the RF signalmust be detected. In most cases, microwave receivers areemployed on the antenna range to accept the very low-level sig-nals from the test antenna and to downconvert these signals tolower frequencies for processing. Microwave receivers offer manyadvantages including improved dynamic range, better accuracy,and rejection of unwanted signals that may be present in the area.Also phase/amplitude receivers provide the ability to measurephase characteristics of the received signal. Phase information isrequired for many types of antenna measurements.

A signal source provides the RF signal for the remotesource antenna. The signal source can be permanently fixed onthe ground or floor, or located on a tower near the source antenna,depending on the frequency of operation and mechanical consid-erations. The signal source is designed for remote operation. Thesource control unit is usually located in the control room with themeasurement and control instrumentation.


Often, a computer subsystem is added to the instrumentationto automate the entire measurement sequence. This computer sub-system employs a standard bus interface, like the IEEE-488, tosetup and monitor the individual instruments. High-speed databusses are utilized for the measurement data to maximize datathroughput and productivity.

An automated antenna measurement system offers a highdegree of repeatability, speed, accuracy, and efficiency with min-imum operator interaction. Data storage is conveniently handledby a variety of media including a local hard drive, floppy disk,removable drives or bulk data storage on a local area network.After data acquisition is completed, an automated system sup-ports analysis of the measured data such as gain and polarizationplus a wide variety of data plotting formats such as rectangular,polar, three-dimensional, and contour plots.

TYPICAL APPLICATIONS OF ANTENNA RANGEINSTRUMENTATION

OUTDOOR FAR-FIELD RANGE In an outdoor far-field range configuration, the test anten-

na is installed on the test positioner located on a tower, roof orplatform outside the instrumentation control room. The receiverfront end (Local Oscillator) is usually located at the base of thetest positioner, with the mixer connected directly to the testantenna port. This configuration requires only a single RF paththrough the positioner, greatly simplifying system design. Use ofthe remote front end also minimizes local oscillator power lossto the mixer and maximum system sensitivity. An outdoor enclo-sure protects the local oscillator from the weather and temperatureextremes. For multi-ported antennas, simultaneous measurementscan be made on all ports through the use of multiplexers installedin front of the mixer. The receiver front end is remotely controlledfrom the control console through interfaces with the receiver.

The test positioner axes are controlled and read out by thepositioner control and readout units. A typical control system con-sists of a control unit located in the operator’s console. It is inter-faced to a power amplifier unit located near the test positioner.This configuration keeps the high power drive signals near thepositioner and away from sensitive measurement instrumentswhile providing remote control of positioner functions from theequipment console. The position readout unit is located in theequipment console to provide real time readout of position axes tothe operator or, in the case of an automated system, to the computer.

The source antenna is normally located at the oppositeend of the range on a tower or other supporting structure. Thesignal source is installed near the source antenna to minimizesignal loss. An outdoor enclosure protects the source from theelements. For some applications a multiplexer can be usedbetween the signal source and a dual polarized source antenna.This configuration allows simultaneous co- and cross-polarizationmeasurements to be performed. Motorized axes to position thesource antenna’s polarization, height and boresight are controlledby a positioner control and indicator system.

The signal source and positioner axes are remotely con-trolled from the operator’s console via serial digital link(s).Twisted pair cable, fiber optics or telephone lines can be used tointerface the digital link from the source site to the control console.

One or two positioner control systems may be used on anoutdoor range depending upon the length of the range and thetotal number of axes to be controlled. On very long ranges, or incases where the control room is not close to either positioner, itmay be advantageous to use a separate control unit for each endof the range. Also, since outdoor ranges frequently have manyaxes due to the source tower axes, multiple controllers may berequired to control all axes.

A block diagram of a typical outdoor range is shown below.

Outdoor Range with Manual Control1


INDOOR FAR-FIELD RANGE Anechoic Chambers are instru-

mented essentially the same way as out-door ranges with range lengths the pri-mary difference. The receiver front endis typically positioned near the test posi-tioner with the mixer connected directlyto the test antenna port. The source islocated near the source antenna. Thecontrol room is generally centrallylocated and connected to both ends ofthe range via cables or digital links.Since these systems are located indoors,special enclosures for the receiver frontend, positioner control, and signalsource subsystems are not required.

Usually, the source antennarequires only polarization control. This,as well as the short range length, usuallyallows a single positioner control unit tobe used to control all the range axes.

Anechoic chambers can be con-figured for either manual or automaticcontrol.

COMPACT RANGE In a point-source compact range,

the feed is usually located just in front ofand below the test antenna. In this con-figuration, the receiver local oscillatorand signal source can be located veryclose together. Special care must betaken to guard against direct leakage ofthe signal source into the test antenna.High quality RF cables and specialshielding are sometimes used to insureagainst this stray leakage. Otherwise,instrumentation for the compact range isvery similar to an anechoic chamber.

Indoor Range with Automatic Control1

Compact Range with Automatic Control Configured for Multiple Port Measurements1


NEAR-FIELD RANGE Near-field ranges usually are configured for automatic

control. The large numbers of measurements required, and theneed to transform the near-field data to the far-field, requires theuse of a computer system both for data acquisition and for datareduction and display.

The configuration of a near-field range is similar to a veryshort indoor range. The antenna may be tested in the transmitmode, receive mode, or both. Consequently, the design of the RFsystem and the location of the source and receiver front end mustbe considered for each application. The figure below is oneexample of a planar near-field application where the test antennais to be tested in both transmit and receive modes.

CONCLUSIONS As technology progresses, the requirements placed upon

wireless communication systems and their associated antennaswill continue to become more stringent. For example, the desireto increase network capacity will result in the requirement toreduce adjacent channel interference within the system, whichwill result in more stringent antenna sidelobe and cross-polariza-tion requirements.

The verification of the performance of antennas selectedto meet these and other requirements will, in turn, require testranges with higher accuracy measurement capability.Fortunately, the technologies used to advance the art of antennadesign is also being used to advance the design of antenna testand measurement ranges and instrumentation. Many of the sim-ulation tools available to antenna designers are also used todesign antenna ranges. The increased use of commercial off-the-shelf hardware and software, in conjunction with the increaseduse of automated test instrumentation networked into the localarea network, will ensure that current state-of-the-art antennameasurement systems meet the needs of the advanced antennasand systems coming to the wireless marketplace.

REFERENCES: [1] Product Catalog, Microwave Measurements Systems andProduct, Microwave Instrumentation Technologies, LLC.

[2] R. Hartman and Jack Berlekamp, “Fundamentals of Antenna Testand Evaluation,” Microwave Systems New and CommunicationsTracking, June 1988.

[3] J.S. Hollis, T.J. Lyon, and L. Clayton, eds., MicrowaveAntenna Measurements, Scientific-Atlanta, Inc., 1985.

[4] R.C. Johnson and Doren Hess, “Conceptual Analysis ofMeasurements on Compact Ranges,” Antenna ApplicationsSymposium, September 1979.

[5] R.C. Johnson editor, Antenna Engineering Handbook,McGraw-Hill Inc., 3rd edition, 1993.

A Typical Planar Near-Field Application1


The purpose of DeriveAssist is to speed up the parameterderivation process and to allow engineers less versed in param-eter matching and identification to get involved in the process.

A major component of any power system simulationmodel is the generating plant which comprises three major sub-components of interest: the generator, excitation system, and theturbine/governor. Accuracy of representation is dependent bothon the structure of the component models and the parameter val-ues used within those models.

Since the accuracy of power system stability analysisdepends on the accuracy of the models used to represent the gen-erators, excitation control systems, and speed governing sys-tems, the parameters used in those models could affect the calcu-lated margin of system stability. Use of more accurate modelscould result in increases in overall power transfer capability andassociated economic benefits. Alternately, inaccurate simulationmodels could result in the system being allowed to operatebeyond safe margins.

To assist in these efforts, Power Technologies, Inc. (PTI),the New York Power Authority (NYPA), and EPRI solutions,Inc. have developed an automated tool to assist engineering staffin the derivation of model parameters from the recorded resultsof staged tests.

NEED FOR BETTER PARAMETERS Modern power systems are highly dependent on the proper

use of dynamic control. Special-purpose computer programs havebeen developed to simulate the dynamics of large complex inter-connected power systems. At the same time, power systems havebecome more highly stressed through heavier system loadingcaused by transfer of low-cost energy and increased use of systemcontrols to increase transfer limits with existing transmission.

Inaccuracies in equipment modeling can be both due toinadequate model structures and, more often, due to lack of dataon equipment model parameters. Model parameters currentlyused for stability analysis are usually provided by equipmentmanufacturers and calculated from design data and, in somecases, factory tests. Generally, they are not verified by field tests.Some pieces of equipment are tuned by field personnel withresults of that initial tuning rarely incorporated into simulationmodels. Parameters may have changed from initial values due toretuning, aging, and equipment changes, such as generatorrewinding. The extensive use of computer simulation requires ahigh degree of confidence in the computer models. The only wayto assure that study assumptions are accurate is to field testequipment and validate simulation models by comparing modelresponses with those obtained from field tests. Thus, there is amajor industry need to enhance equipment model developmentas well as model parameter identification and validation.

This need has been recognized by organizations responsi-ble for system reliability. For example, the Western SystemCoordinating Council (WSCC) instituted a program requiringtesting and model validation for all generating units greater that10 MW. The North American Electric Reliability Council(NERC) is presently formulating its requirements in this area.

STAGED TESTS Staged field tests provide sufficient information to identify

the values of the key parameters of the computer simulation models.Such tests are selected to minimize the effect on plant operation,allow ease of simulation of the staged tests, and, to the extent possi-ble, reduce the complexity of the parameter derivation problem byhaving the response of an individual test significantly affected byonly a few parameters.

The test methodology described is just one testing method-ology; other methods are used successfully also. However, thetools and procedures developed and used in this parameter deri-vation software could be adapted to these other variants in thetesting process.

The testing process is divided into two phases. One phaseinvolves collecting steady-state measurements, which are used toestablish base values of quantities and to identify values forparameters that are associated with steady-state operation. Thesecond phase involves collecting the dynamic response of the gen-erator, excitation system, and governor/turbine system to stageddisturbances.

STEADY-STATE MEASUREMENTS The steady-state measurements are divided into two

groups: the open circuit saturation curve measurements andonline measurements. The open circuit saturation curve is meas-ured with the unit operating offline at rated speed. The generatorfield excitation is varied, and measurements of terminal voltage,field voltage, and field current are taken.

The online meas-urements (also some-times called V-curvemeasurements) are per-formed with the unitconnected to the electri-cal network and placedat a given load. At thatload level, the generatorfield excitation is variedto change the reactivepower output. Typicalmeasurement points aregiven in graphical formin Figure 1.

DERIVING MODEL PARAMETERS FROM FIELD TEST MEASUREMENTS

J.W. Feltes, S. Orero, Power Technologies, Inc., B. Fardanesh, E. Uzunovic, S. Zelingher, the New York Power Authority, N. Abi-Samra, EPRIsolutions, Inc.


DYNAMIC TESTS The gains and time constants of the models can be deter-

mined only from tests that excite the dynamic response of theequipment. The models of concern are those of the generator,excitation system, and governor. The purpose of the dynamictests is to provide a simple and safe disturbance to excite the unitin order to record its dynamic response. The usual approach is aseries of load rejection tests with the unit initially carrying par-tial load.

Each of the tests has identification of certain parametersas its primary goal. The loading of the machine is selected to iso-late those parameters as much as possible in order to reduce thecomplexity of the derivation process. The initial conditions for atypical set of load rejections are listed in Table 1.

TRADITIONAL GENERATOR PARAMETER DERIVATIONPROCESS AND THE DERIVATION SOFTWARE

The traditional method of determining the model parame-ters based on the recorded test results forms the basis for theapproach used by the parameter derivation automation software.The traditional methodology has been highly dependant on skilledengineers applying their knowledge to select initial parameters,perform calculations using those parameters, and, based on the dif-ference between measured and calculated values, adjust theparameters manually to improve the fit between model and real-world response. This iterative approach is quite time consumingand requires a skilled, experienced engineer to make the adjust-ments so as to accomplish a good match in a reasonable amount oftime. The purpose of the software described here is to speed upthat process and also allow engineers less versed in parametermatching and identification to get involved in the process.

PROGRAM OVERVIEWDeriveAssist works through a graphical user interface

(GUI) and is written to operate on a personal computer. The pro-gram uses the MATLAB/Simulink and the Optimization Toolboxas the calculation engine for the parameter derivation process.MATLAB is a high-performance language for technical comput-ing. Simulink is a graphical tool for modeling, simulation, andanalysis of dynamic systems. The Optimization Toolbox is a col-lection of routines that extend the capability of MATLAB forsuch problems as nonlinear minimization, equation solving, andcurve fitting. By combining these tools with the experiencegained through years of testing and parameter derivation, theprocess of parameter identification and derivation has been sig-nificantly advanced. The program is organized to facilitate thederivation of the parameters in a logical order, starting with thesteady-state tests and then proceeding to the dynamic tests. Themain entry point into the program is a window with several pulldown menus. Each menu item has several submenu choices.Each of these submenu choices performs a particular task. Thesubmenus were also designed to reflect a particular test.

The five menus are:• File I/O• Steady-state tests• Dynamic tests• Control tuning• Help.The parameter derivation program is organized to facilitate

the derivation of the parameters in a logical order, starting withthe data from the steady-state tests and then proceeding to thedynamic tests for the generator, excitation system, and governor.Where a certain sequence of actions must be observed (i.e., readsaturation curve first before trying to calculate the saturationparameters), the program performs a check and warns if the pre-requisite tasks have not been performed.

DERIVATION OF STEADY-STATE PARAMETERSThe steady-state parameters are derived first. They can be

derived based on a series of measurements of steady-state quan-tities as described previously. The first task is to analyze the opencircuit data (obtained from the offline open circuit test) to estab-lish the base values for field current and field voltage. This hasbeen traditionally accomplished by plotting terminal voltage ver-sus field current and drawing the air gap line. The value of fieldcurrent corresponding to rated terminal voltage on the air gapline is identified as the base value. Next, the saturation valuesS(1.0) and S(1.2), the parameters used to describe the shape ofthe saturation curve, are identified using the open circuit data.The program automates this task, calculating the base value andsaturation parameters using a least squares fit. Figure 2 shows anexample of the output of the program, showing the close matchachievable between test and calculated values. Tabular outputdemonstrating the fit of the measured data and calculated resultsare also given.

The online steady-state measurements are used to identifythe values for Xd and Xq. Recordings of voltage, power, reactivepower, field current, field voltage, and power angle are made atdifferent power levels and reactive power outputs. The user canselect which points to use in the calculation. The program calcu-lates the reactances Xd and Xq that best fit the measured data,again using a least squares optimization. Each reactance can becalculated separately or both at the same time. If the user selects


Xd only, then the error function to minimize uses only field cur-rent, Ifd. However, if the user selects Xq, the error function will bebased on power angle. The last option is only possible whenrotor angle measurements have been made during the field tests.Figure 3 shows the program output screen following the deriva-tion of the generator steadystate reactances.

DERIVATION OF DYNAMIC MODEL PARAMETERSThe traditional parameter derivation process for the

parameters of the dynamic models required numerous simula-tions to see exactly what happens as one changes each of themodel parameters. With each parameter change, a comparison ismade between response of the model and that obtained from theactual tests. The determination of the parameters to adjust andthe amount by which to adjust them to get a close match betweenthe model response and the actual performance requires the skillsof an experienced engineer. If there are nonlinear dynamic inter-actions among variables, it is usually very difficult to know howto set each of the parameter values to give the desired perform-ance. The process of choosing the appropriate model parametersthat provide the desired response can be automated by the use ofoptimization tools. MATLAB, Simulink, and the OptimizationToolbox provide a suitable advanced programming environmentthat can allow an optimization engine to interact with a dynamicsimulation package.

Thus the optimization phase of the model parameter deriva-tion involves the automatic adjustment of the model parametersuntil the difference between the model response and the desiredresponse (obtained from field tests) is minimized. The optimizationprocess tries to find the combination of model parameters that mostclosely matches the measured response.

A large task involved in the development of the DeriveAssistprogram was the building of generator, exciter, and governor mod-els in Simulink, testing them against a widely used commercialpower system simulation program (PTI’s PSS/E program) as abenchmark, and making the interface between all three componentsof the software: Simulink, MATLAB, and the optimization toolbox,as integrated as possible. Tests are performed to record the equip-ment response with different initial conditions and with disturbancesdesigned to target the derivation of specific generator, excitation,and governor model parameters.

The derivation process is similar for the generator, excita-tion system, and governor parameters, although the staged tests

and measured quantities are quite different. For the generator, theinformation from the load rejection tests is used to calculate thetime constants, the transient reactances and the subtransient reac-tances of the generator. For the excitation system, the AVR andexciter gains and time constants can be determined.

The governor models vary significantly depending on thetype of prime mover, that is, steam, gas, or hydroelectric turbine.However, the process determines the gains and time constantsrepresenting the governor and turbine dynamics. In all cases, acomparison was made between a measured signal and a simulat-ed signal to define an error function. The program attempts toimprove the model performance by adjusting the model parame-ters in the appropriate direction and repeating this simulationuntil the error signal is minimized. The process will be illustrat-ed using the derivation of excitation system parameters, but thereader should keep in mind that the general process would besimilar for the other equipment models.

EXCITATION SYSTEM PARAMETER DERIVATIONThe first step is, of course, to choose an appropriate model

structure for the excitation system. The selection is usually guidedby the manufacturer’s recommendation or from industry stan-dards. In some cases, the schematics of the excitation system mayneed to be examined to make the proper model selection or, if astandard model structure is not appropriate, to create a new model.

As noted above, the traditional strategy to identify the val-ues for the excitation parameters involves an iterative hill-climbingtechnique by the engineer, who changes the value of one parameterat a time until a match between simulation results and recordedmeasurements are made. This process requires good familiaritywith the specifics of how the equipment functions and of the effectthat a change in a parameter or a set of parameters has on theirdynamic response; unfortunately, such familiarity is quite rare. Theparameter derivation program greatly simplifies the process.

The user must read in the test data, which is easily per-formed using the program GUI. The GUI also facilities the selec-tion of signals for the derivation process and manipulation of thetest data as necessary, for example, conversion to per unit usingthe base values derived in the steady-state derivations.

Figure 4 shows the connections of round rotor generatormodel (GENROE in PSS/E) with an IEEE type 1 (IEEET1) exci-tation system. Inputs to the generator model are field voltage,generator currents Id and Iq, and mechanical power Pm. Outputsare terminal voltage magnitude and angle and generator speed.


Inputs to the excitation model are terminal voltage (fromthe generator) and reference voltage, while the output is fieldvoltage Efd, which is fed back as an input to the generator model.

The inputs and outputs allow data to be passed betweenMATLAB and Simulink and between the models. The parame-ters are defined such that they can be changed and passed toSimulink in the optimization process.

As an example, Figure 5 shows what lies under the exci-tation system block in Figure 4. The primary input is the voltageEcomp and the output is the field voltage Efd. The reference Vrefis calculated from the initial condition of the test. Auxiliary sig-nals such as those from the power system stabilizer and under- orover-excitation limiters are present in the model structure but arenot exercised purposely by the selection of the tests, concentrat-ing on the excitation parameters.

The optimization phase of the model parameter derivationinvolves the automatic adjustment of the model parameters untilthe difference between the Simulink model response and thedesired response (measured response) is minimized. The opti-mization process tries to find the combination of model systemparameters that best provide the desired response, that is, to findthe values of the excitation system model parameters that willmove the initial model response as close as possible to the meas-ured response.

A comparison of the simulation output and the measured(desired) output is displayed for each successive pass of the opti-mization process. The user sees the simulation output change

every few seconds as the model parameters are adjusted, a newsimulation is performed, and the new output is displayed. Thesimulation output gradually shifts from the original response tovery closely match the desired response. Figure 6 shows the

shows a comparison of a simulationusing an initial set of parameters andthe measured response. Note that theoriginal parameters are not a goodapproximation of the actual equip-ment. The simulation, in this case, isquicker and much better damped. Theresponse gets progressively closer tothe measured output following eachpass of the derivation process until; atlast, the two curves are essentially oneon top of the other. The final plot isshown in Figure 7. The whole opti-mization process to determine theparameters takes only a minute or twoon a typical PC.

Generator and governor model parameters are derived in amanner quite similar to that described for the excitation systems.The tests performed on the unit are considerably different, ofcourse, as the determination of the governor response character-istic requires a test resulting in a power imbalance and subsequentmovement of generator speed while the tests for the generatorparameters require isolation of the generator dynamics by placingthe AVR on manual.

ONGOING EFFORTSThe DeriveAssist parameter derivation software

described so far allows the derivation of all the generator steadystate parameters and includes most of the Simulink models forgenerators, exciters and governor systems. However, additionalwork is required to further develop the software. Some of thetasks for future work include:

• Derivation of excitation system parameters from AVRreference step tests

• Extension of the methodology to brushless excitationsystems.

• Develop algorithms designed to assist the user in thetuning of equipment such as exciters and stabilizers.


There are also a few technical issues that require furtherattention, including:

• Additional investigations to overcome some problems inthe automatic initialization of Simulink models

• Improvements to the model library and the ease ofselecting the model structures for the excitation systemand governor

• Improvements to the GUI and data passing betweenSimulink, MATLAB, and the Optimization Toolbox

• Additional reporting routines.The ongoing further development work will improve its

functionality and expand its capabilities.

ACKNOWLEDGMENTSThis article describes research sponsored by EPRI and

NYPA. The authors would also like to acknowledge F.P. deMello for his contributions to the original ideas behind theparameter derivation process used in this project and Ricardo J.Galarza for his contributions in the development of this parame-ter derivation software. The idea of developing this MATLAB-based tool was originally conceived by Bruce Fardanesh.

FOR FURTHER READING“Synchronous machine parameter derivation program,” EPRI,Palo Alto, CA, Rep. 10006653, 2001.

F.P. de Mello and J.R. Ribeiro, “Derivation of synchronousmachine parameters from tests,” IEEE Trans. Power App. Syst.,vol. PAS-96, no. 4, pp. 1211-1218, Jul./Aug. 1977.

F.P. de Mello and L.N. Hannett, “Determination of synchronousmachine electrical characteristics by tests,” IEEE Trans. PowerApp. Syst., vol., PAS-102, no. 12, pp. 3810-3815, Dec. 1983.

J.W. Feltes and L.N. Hannett, “Derivation of generator, excitationsystem and turbine governor parameters from tests,” presented atInt.

Conf Large High Voltage Electric Systems, Colloquium onPower System Dynamic Performance, Florianopolis, Brazil,Sept. 1993.

L.N. Hannett, J.W. Feltes, and B. Fardanesh, “Field tests to val-idate hydro turbine-governor model structure and parameters,”IEEE Trans. Power Syst., vol. 9, no. 4, pp. 1744-1751, Nov.1994.

L.N. Hannett, B. Fardanesh, G. Jee, “A governor/turbine modelfor a twin-shaft combustion turbine,” IEEE Trans. Power Syst.,vol. 10, no. 1, pp. 133-140, Feb. 1995.


The Challenge: Automated testing of magnetic ballasts used inelectric streetlights.

The Solution: Developing a PC-based virtual instrumentationsystem using a DAQ board controlled byLabVIEW.

INTRODUCTION Our task was to develop an automated test system for

magnetic ballasts used in high-pressure sodium (HPS) street-lights. Our client, who manufactures ballasts for the NorthAmerican and international markets, believes that product devel-opment and quality assurance require thorough and completetesting of prototypes and production samples to verify compli-ance with national and/or international standards.

The test system needed to accommodate the following: • Different types of core and coil ballasts, such as reactor,

autotransformer, constant wattage autotransformer(CWA), and constant wattage isolated transformer (CWI)

• Operating voltages from 120 to 600 V and rated lampwattage from 50 to 400 W

• Capacitors for wattage control and/or power factor correction

• Different lamp igniters • Open-circuit, short-circuit, lamp-starting, and lamp

running tests At the ballast input and output ports, we needed to meas-

ure true rms values of current and voltage, true power, and theratio of watts to volt-amperes (power factor, if the voltage andcurrent waveforms are clean sinusoids). Because HPS lamps arenonlinear loads, we monitor current and voltage peak values andcrest factors, along with total harmonic distortion.

SYSTEM INTEGRATION APPROACH With the tight budget of a growing company, establishing

a test bench with the functionality we required using conventionaltest equipment becomes difficult. We implemented a virtualinstrumentation approach to achieve project objectives withinbudget while maintaining flexibility for future needs.

Virtual instrumentation consists of using mainstreamcomputers, off-the-shelf plug-in instrumentation boards, andsoftware. Because the virtual instruments you create with theseproducts are user-defined, not vendor-defined, you can tailorapplications to meet your needs exactly. Some of the benefits ofvirtual instrumentation are ease of use, flexibility, and savings oftime and money.

We used LabVIEW software as the heart of the instru-mentation system. BallastVIEW is the name of the LabVIEWapplication we wrote to acquire signals, process data, and presentresults to the user on the computer screen.

The instrumentation system hardware consists of: • 486 DX2-66 PC (12 MB RAM, 340 MB hard drive) run-

ning Windows • Variac (manually adjustable transformer) supplying AC

power to the ballast under test through the system testfixture

• System test fixture containing switches and wiringrequired for the different test configurations

• Transducers for sensing current and voltage signals(such as resistive dividers and current shunts)

• Antialiasing RC filters, with components selected toavoid loading the board input amplifiers

TESTING ELECTRIC STREETLIGHT COMPONENTS WITHLABVIEW-CONTROLLED VIRTUAL INSTRUMENTATION

Ahmad Sultan, Computer Solutions, Inc.

Hardware Block Diagram of the Ballast


• 5B signal conditioning modules, to amplify and isolatethe filtered signals

• National Instruments Lab-PC+, installed in the PC, todigitize the conditioned signals

The cut-off frequency of the antialiasing filters was set tohalf the sampling frequency; the RC filters also serve to protectthe electronics items from the high-voltage spikes generatedwhen the igniter starts the lamp.

We configured the Lab-PC+ board for bipolar differentialinput (four channels). We set the sampling frequency to 7680Hz/channel. Acquisition was software-triggered on the risingslope of the input voltage.

BALLASTVIEW PRESENTATION The LabVIEW screen on the next page is the front panel

of BallastVIEW. It illustrates a stack of VIs representing an inputAC power analyzer, an output AC power analyzer, a waveformgraph, and a harmonic analyzer. The controls at the top of thescreen are switches for controlling acquisition, metering, har-monic analysis, and program execution. The user can capture asingle shot or continuously acquire signals.

For the power analyzers, the indicators (from left to rightin each row) display the rms, maximum, minimum, peak average,and crest factor of each signal. The active and apparent power,and their ratios, are displayed in the right column.

The waveform graph displays the signals acquired by thedata acquisition (DAQ) board.

Because both voltage current waveforms are displayed,the ordinate is labeled in relative units (PU). To find the trueamplitude of a particular signal, multiply its measured valuefrom the graph, in PU, by the respective base value from the PUBase table (to the right of the waveform).

The line spectrum, shown in the bottom right corner, dis-plays harmonic magnitude in either peak volts/amperes or perunit values normalized to the fundamental component of therespective signal. Magnitude of harmonics can be checked byflipping the cursors of the harmonic magnitude indicator (bottomcenter). The user can window signals before applying the FastFourier Transform.

EXAMPLE RESULTS The results presented in the BallastVIEW screen are test

results for a 200 W CWI ballast. The output power analyzer indi-cates that the lamp is operating at rated lamp power. Lamp volt-age and current are very close to the ANSI reference specifications(100 V and 2.4 A). Lamp current crest factor (CCF) is 1.6 (1.8 isthe maximum permissible). The input power analyzer indicatesthat the ballast draws 2.037 A at rated input voltage. Ballast loss isapproximately 39 W and the power factor is high (0.973 lagging).

The waveform graph shows almost clean input voltage andcurrent signals. Output (lamp) voltage is the square waveform ofa typical arc in a high-intensity-discharge (HID) lamp, containingthe full odd harmonics spectrum. The magnitude of the lamp volt-age third-harmonic component is 39 percent of the fundamental.Total harmonic distortion (THD) of lamp voltage and lamp cur-rent are 33.84 percent and 3.73 percent, respectively.

We verified the credibility of this system by obtainingagreement with test results from an independent test laboratory,electric utility companies, and customers of the ballast company.

The result is a flexible, high-performance, easy-to-use,and cost-effective PC-based measurement system, which savedtime in both product development and production testing.

CONCLUSION BallastVIEW measures and displays the electrical param-

eters required to test and develop ballasts and performs on-linewaveform analysis. The result is a flexible, high-performance,easy-to-use, and cost-effective PC-based measurement system,which saved time in both product development and productiontesting. An advantage of using LabVIEW is our ability toincrease BallastVIEW functionality in the future, for example,by monitoring the ballast-lamp characteristic curves and compil-ing results. The core of the BallastVIEW program constitutes thecornerstone for testing other electrical products, such as trans-formers, rectifiers, inverters, and UPSs, as well as for power linemonitoring.

LabVIEW Front Panel showing BallastVIEWTest Results


This article tell the Japanese way of doing asset manage-ment and maintenance. If you think you already have a good sys-tem then you will enjoy reading this month’s newsletter as youcompare yours and theirs. If you have a poor system then you willget a totally different view of how great maintenance can be done.

OVERVIEWI spent a week in Japan at the chemical plant of an inter-

nationally renowned chemical manufacturer. While there I askedthem about how they do their maintenance. They told me abouttheir maintenance philosophy. And I want to pass on to you whatI learnt about the Japanese way of doing maintenance on thattrip.

You will read about how this Japanese company determinesits equipment and component criticality. You will learn about a new,truly effective way, of making next year’s maintenance plan. Wewill cover condition monitoring the Japanese way. The Japanese aregreat maintenance investigators and you will be impressed whenyou learn how they do their failure analyses. We will also cover theirpsychology of maintenance – the way they think about maintenanceand how they look at it. You will be astounded at their mind-set.

A JAPANESE WAY TO DECIDE EQUIPMENT CRITICALITYHow do you decide what level and type of maintenance to

use on an individual item of plant and its sub-assemblies? Not allequipment is equally important to your business. Some are criticalto production and without them the process stops. Others areimportant and will eventually affect production if they cannot bereturned to service in time. While other items of plant are notimportant at all and can fail and not affect production for a verylong time.

As a maintainer you want to know which equipment inyour plant falls into each of those categories so you can deter-mine your response. Furthermore you want to know which sub-assemblies in each item of equipment are critical to the operationof the machine.

From this information you can decide which spares tohold on-site and which to leave as outside purchases. The equip-ment criticality also determines what level of preventive mainte-nance to use, what type and amount of condition monitoring touse and what type and amount of observation is required fromthe operators. You can also use it to justify on-line monitoringsystems to protect against catastrophic failure.

The western approach to determine criticality is often touse either Reliability Centered Maintenance or Risk BasedMaintenance to determine consequences of failure and thenaddress the appropriate response to prevent the failure. TheJapanese chemical manufacturing company I visited had a novelway of determining their equipment criticality. They based theequipment and component criticality on the knock-on effect of afailure and the severity of the consequences. It is the same inten-

tion as the previously mentioned methods but they arrive at therating and the response to it in a unique, quick four-step process.

They used a simple flow chart that production and main-tenance worked through together, equipment by equipment.Those failures that caused safety and environmental risks werenot allowed to happen and either the parts were carried as sparesand changed out before failure or the plant item was put on acondition monitoring program. Those failures that caused pro-duction loss or affected quality also were either not allowed tohappen or put into a condition-monitoring program. And thosefailures that didn’t matter were treated as a breakdown.

The flowchart let one arrive at a rating and a correctiveaction for each piece of equipment and component fast. No needto spend hours and days looking at failure modes and decidingwhat to do about them. If an equipment or component loss pro-duced dangerous situations, or if the failure stopped productionor affected quality, it was either changed out before the end of itsworking life or it was put on a monitoring program.

The maintenance philosophy for every bit of plant couldbe arrived at in a four-step decision process. It was very easy touse and to decide what action to take.

HOW TO TURN A MAINTENANCE PLAN INTO A STRATEGYThe maintenance plan my Japanese hosts showed me in

August 2002 was on a big spreadsheet. It listed all the equipmentin a plant by tag number covering the period 1994 through to2003. The maintenance histories of problems on a piece ofequipment for the past eight years were listed. A short notedetailing the month of occurrence and the failure was made inthe column of the year it happened. For this year, 2002, and thenext, 2003, the spreadsheet listed what maintenance and modifi-cations were going to be done on the equipment.

It was a ten-year plan the like that I had never seen before!But now, as I write, it has become clear why it’s worth-

while doing it like that. What I saw was not a plan! What I sawwas a strategy! It was a strategy to reduce the known productionstoppages and to focus the maintenance effort.

Can you see how something like that would work? Youknow what has gone wrong with the equipment over the lasteight years, it’s listed right there in front of you. You can see howeffective the past practices, methods and solutions have been.From that you can wisely decide what to do over the next twoyears to prevent the reoccurring problems. Instead of writing theusual ‘blue sky’ 5 or 10 year maintenance plan that no onebelieves anyway, you only plan for the believable two yearsahead. You write down exactly what can really be done in theforeseeable future to reduce or prevent the real problems.

The plan for the next two years would include proposedmodifications, equipment replacements, new condition monitor-ing plans, etc.

Now that is a great way to make next year’s maintenance

ASSET MANAGEMENTThe Path to Maintenance Excellence

Mike Sondalini, Managing Editor, Feed Forward UP-TIME Publications


plan! It would be one that is totally defendable and fully justifi-able to upper management because it is well thought out, rootedin getting the best return for your money and based on the impor-tant business requirements to continue in operation.

My suggestion to cover the period beyond the next two orthree years (and only if it is necessary in your company), is to usethe spreadsheet to make forecasts. Project ahead based on whatyou plan to do in the coming two to three years to fix the currentproblems. If you aren’t going to fix the problems then don’tassume less maintenance in the future. Remember that a forecastis not a plan! A forecast is a best-guess suggestion, often knownas ‘blue sky dreaming’. A plan is a set of action steps that overtime will produce a desired result. They are totally different toeach other.


OVERVIEWLatencies and timing uncertainties involved in orchestrat-

ing the operation of multiple measurement components present asignificant challenge in building automated test systems. Theseissues, often overlooked during the initial system design, limitthe speed and accuracy of the system. However, with a goodunderstanding of timing and synchronization technologies, youcan address these issues from the onset and deploy a system opti-mized for throughput and performance.

Before we proceed, first consider that most automatedmeasurements for test fall into one of two categories. The firstcategory, often called time-domain measurements, characterizesthe variation of a device under test (DUTs) output over time. Forthese measurements, the accuracy of the measured responsedepends not only on the accuracy of its magnitude, but also onthe time at which the signals are measured.

The second type or steady-state measurements occurswhen one or more inputs of known value are applied to the DUTand its outputs can settle to their steady-state value before youmeasure the signals. In this case, the measurement processdepends on the time of the measurement – if you measure thesignals too early, accuracy suffers because the source output maynot have fully settled. Although you can measure the signalsaccurately any time after the output has settled, you must mini-mize the delay to reduce test time. Many test developers insert anarbitrary delay in their test programs to ensure accurate results.While this is a simple fix, test time suffers.

Analog electronic component evaluation and manufacturingtest often involves measurements of both transient and steady-stateparameters.

WHAT IS INVOLVED IN SYNCHRONIZATION?The main objective of synchronization of multiple meas-

urement devices is correlated measurements and/or precise controlof process execution. In most cases, you are interested in correla-tion in terms of time, but correlation can occasionally be in dif-

ferent terms, such as position. For temporal correlation, you mustsynchronize measurements to correlate with the sample clock ofyour measurement device. In other words, it is pointless to exam-ine measurements synchronized to within nanoseconds if yoursampling clocks are 1 MHz. The objective is a system with syn-chronized devices that are synchronized to sub-microsecondaccuracy. Precise timing of measurements is a prerequisite of themeasurement device.

Let us take a digitizer as an example to elaborate on keytiming technologies. The heart of a digitizer is an ADC, whichsamples your signal and converts it to digital data. The sampleclock, which controls the timing of the ADC, is most often derivedfrom an onboard crystal oscillator. Thus, the synchronization ofmeasurements across multiple devices, such as a source or otherdigitizers, implies that you must synchronize all sample clocks towithin the uncertainty of the period of the sampling clocks.

Another important element to the measurement is collectingdata. This is usually accomplished with trigger signals. Externalevents or triggers are the main methodologies for initiating anacquisition. Triggers come in three forms – analog, digital, and soft-ware. Analog triggering refers to trigger generation when a moni-tored analog signal passes the imposed triggering condition. Youcan measure the analog signal itself or an auxiliary analog signal.Digital triggering refers to trigger generation when a digital signal,such as a TTL level signal, is received. Software trigger refers totrigger generation on software command. The software trigger canbe as simple as hitting a ‘start’ button on the soft front panel orgraphical user interface (GUI). Thus, synchronization of measure-ments requires not only synchronized sample clocks, but also thedistribution of a trigger to all measurement devices to initiate oper-ation at the same time. In synchronization applications, it is com-mon to designate a ‘master’ measurement device to monitor theoperation of the entire measurement system. When the systemmeets triggering conditions on this device, it distributes a commontrigger signal to all other devices that are ‘slaved’ to the master.

To achieve tight synchronization across multiple devices,you need to examine the distribution of clocks and triggers.There are three main schemes for synchronization:

1. START/STOP TRIGGERS CONTROL OPERATION

ON ALL DEVICESThis scheme for synchronization is the simplest. It

involves a single start or stop trigger signal to all measurementdevices involved. One device, designated as the ‘master’ device,monitors the operation. The master is set to look for an externaltrigger (analog or digital), or to generate a synchronizing triggeron a software command. When triggering conditions are met onthe master device, or the software command is issued, the masterdistributes a trigger signal to all ‘slave’ devices to start operationas shown in Figure A.

THINK SYNCHRONIZATION FIRST TO OPTIMIZE AUTOMATED TEST

www.ni.com


Some examples are:• Rotationally oriented measurements – A master digitiz-

er or oscilloscope, monitoring defects found on rotatingcircular or cylindrical devices such as computer harddrives, industrial cylindrical tubes, and automotive wheelshafts, passes a digital trigger to a slave counter/timerdevice making quadrature encoder measurements (posi-tion measurements). The system can correlate defects andanomalies to angular and radial position rather than time.

• High channel count measurements – Multiple digitizersacquire data on reception of an external digital triggerfrom an external triggering module or a master digitizerin the system.

With the examples above, two issues arise: The trigger signal should arrive at each slave device with

minimal delay and skew between each other. The delay and skeware separate issues and need equal consideration. With a signifi-cant delay from the master to the slaves, you lose synchronization.Minimal path length for signal propagation from master to slavesis crucial for tight synchronization. The other important, but sub-tle, consideration is the skew between slave devices. So that eachslave triggers at precisely the same time, you need to minimizethe device-to-device skew in time. At the least, the delay andskew should be identified to some uncertainty. Measurementsthat require relatively low sampling rates can tolerate a degree ofslack in the specifications of a system set up. At high samplingrates, these issues can affect the measurement integrity.

The second issue concerns the intrinsic accuracy of themeasurement device – you should identify or calibrate the timethat the device received the trigger signal to the first pre-triggeror post-trigger sampled point in each device. You can programmany measurement devices, such as digitizers, to continuouslyacquire samples into a circular onboard memory buffer that con-

tinually rewrites until it receives a trigger. After the devicereceives the trigger, the digitizer continues to acquire post-trig-ger samples if you specified a post-trigger sample count. Theability to correlate waveforms acquired on the various devicesdepends on the accuracy of the time-stamp of the trigger.

2. TRIGGERS AND A DIRECT SAMPLE CLOCK

INITIATE AND CONTROL THE TIMING ON ALL DEVICESThis scheme takes synchronization a step higher. It involves

trigger signals and a sample clock to all the devices involved. Onedevice, designated as the ‘master’ device, controls the operation ofthe entire measurement system. This device exports its sampleclock to all slave devices. For example, a system comprised of mul-tiple digitizers and analog output sources has a common sampleclock from an appointed ‘master.’ As illustrated in Figure B, themaster sample clock directly controls ADC and digital to analogconversion (DAC) timing on all devices.

The master is set to look for an external trigger (analog ordigital), or to generate the trigger on a software command. Whentriggering conditions are met on the master, the device distributesa trigger signal to all of the ‘slave’ devices to commence opera-tion. The same issues that arose in the previous scenario are alsopresent in this situation. The trigger and sample clock signalsshould arrive at each slave device with minimal delay and skewbetween each other. At the least, the delay and skew should beknown to an uncertainty. The significant advantage of this schemecompared with the previous scheme is that you use a commonsample clock to control all devices. With a common sample clock,all waveforms are precisely sampled at the same time. Thisresolves the central issue of synchronized measurements.

With this technique, you benefit in another important way.If you employ the clock on each measurement device, you haveto take the jitter and drift inherent in each clock into considera-tion. On each digitizer, different clock jitter and drift may giverise to sampling periods, which means you cannot correlate themwith relative accuracy.

The disadvantage of this scheme is that it is not optimalfor high-speed sampling because of the propagation delay of thesample clock. The sample clock simply takes time to get to theslaves from the master. This issue does not arise if the samplingrate is slower than the propagation delay. For example, in a givensystem the propagation delay is measured to be 10 ns. If the sam-pling rate is 5 MSamples/s, the period between each rising edgeof the clock is 20 ns. The sample clock reaches the slave devicesbefore the delay time encumbers the measurements. Additionally,the path lengths from the master to each slave device have to becarefully matched so the skew time is shorter than the samplingclock period.

3. TRIGGERS AND A REFERENCE CLOCK TO INITIATE ANDCONTROL THE TIMING ON ALL DEVICES

This scheme of synchronization is usually for high-speedsynchronization. It involves start/stop trigger signals and a referenceclock (typically 10 MHz) to all devices involved. The samplingclock of each measurement device is derived from the referenceclock by dividing the reference clock to obtain higher speed sam-pling clocks. The master is set to look for an external trigger (analogor digital), or to start acquisition on a software command. Whentriggering conditions are met on the master, this device distributesa trigger signal to all slave devices to start operation.


With the previous scheme, you could have a direct feed ofthe sample clock to each device. This is the ideal scenario, how-ever, it is not easy to pass a high-speed sampling clock (such as100 MHz clocks) across cables and/or trigger buses because ofline integrity and propagation delays. So, this scheme shares acommon reference clock for generation of all sample clocks.

The method usually employed to synchronize and gener-ate sampling clocks is phase lock looping (PLL). This methodbasically monitors the phase of the reference clock and producesa high-speed sampling clock that is phase locked to the referenceclock, as shown in Figure C above.

Third-party frequency sources, such as rubidium andoven-controlled crystal oscillator (OCXO)-based frequencysources, are ideal for synchronization applications because oftheir accuracy. These are frequency sources with accuracies ofbetter than 100 parts per billion (ppb). Thus, an OCXO sourcewith 100 ppb accuracy yields a 10 MHz clock with 1 Hz uncer-tainty. Another important property of your reference clock ismultiple output capability for multiple instrument synchroniza-tion. The reference clock from either the master instrument or aprecision frequency source should be capable of being driven tomultiple destinations without any loss of signal integrity. An exam-ple of this would be a minimal phase offset between the referenceclock outputs from the frequency source.

The same issues that arose in the previous scenarios arealso relevant in this scheme. The trigger and reference clock sig-nals should arrive at each slave device with minimal delay andskew between each other. At the least, the delay and skew shouldbe known to some uncertainty. The issue of minimal skewbetween each device is crucial for high-speed digitization. If theskew is large, the time stamp of the incoming trigger on eachdevice will not be coincident in time, and you cannot accuratelycorrelate events captured on separate devices.

SYNCHRONIZATION OPTIONSMeasurement devices come with three main options for

connecting synchronization signals – user-supplied cabling, pro-prietary vendor-defined cabling, and connections integrated withthe measurement platform.

User-Supplied Cabling – User-supplied cabling of sig-nals for synchronization is available for both computer-basedand stand-alone measurement devices. For example, you canoften externally synchronize your function generator or digitalstorage oscilloscope (DSO) to a reference frequency source.When you decide to synchronize your instrumentation, you haveto ensure that your cables from your frequency source to theother components of your measurement system are preciselymatched in length in order to avoid skew. The same criteria needto apply in distribution of your trigger signal from master to allslave devices. As noted above, your frequency source shouldhave the ability to distribute a common reference clock to multipledestinations. This is the only synchronization option for traditionalstand-alone instruments.

Proprietary Vendor-Defined Cabling – Some vendorsof computer-based measurement devices, such as data acquisi-tion boards, address synchronization by providing a proprietarybus, which may be external or internal to the computer. Samplingclocks, reference clocks, and triggers are distributed from masterto slaves through the bus. These dedicated high-speed digitalbuses are designed to facilitate systems integration. The physicalbus interface is a multipin connector on the board, and signals

are shared via a ribbon. You can serially chain two, three, four,or five boards together, thus achieving synchronization of severalI/O channels. Another attractive feature of these trigger buses isbuilt-in switching, so you can route signals to and from the buson-the-fly through software programming. This eases the burdenof having to manually configure your timing and triggering signaldistribution on your boards. You can find examples of these fea-tures in National Instruments measurement products in the formof the RTSI bus.

Connections Integrated with the MeasurementPlatform – Some of the computer-based measurement devices areimplemented in form factors such as VME/VXI andCompactPCI/PXI. VME/VXI, an older industrial form factor, andPXI/CompactPCI, a newer industrial form factor, both address testand measurement, telecommunications, defense, industrialresearch, and many other markets. VXI and PXI extended VMEand CompactPCI by adding timing and triggering buses to the formfactors. This greatly simplifies synchronization of multiple devices.

MORE ON VXI AND PXIVXI and PXI are open standards and many companies

make products for both variations. VXI is traditionally used inlarge test and measurement applications. Though relatively newto the market, PXI is gaining acceptance because of its relative-ly smaller footprint, portability, high throughput due to the PCIbus, and lower costs, made possible through use of standardcommercial technologies spawned by the large PC Industry.

Electrically, VXI and PXI add a trigger bus, a star triggerbus, a 10 MHz reference clock, and local buses. For synchro-nized measurements, the trigger bus, the 10 MHz referenceclock, and STAR trigger bus are key features. The PXI featuresdescribed below broadly apply to VXI as well.

System Reference Clock – The PXI back-plane providesa built-in common reference clock for synchronization of multi-ple modules in a measurement or control system. Each peripher-al slot features a 10 MHz TTL clock. Equal-length traces fromthe clock to each peripheral slot yield low skews of less than 1ns between slots. The accuracy of the 10 MHz clock is usually25 ppm (dependent on individual chassis vendors), making it arelatively reliable clock for synchronization applications thatrely on PLL methods. If you need a more accurate referenceclock, you can insert a PXI counter/timer device with an OCXO-based clock source into the second slot of the chassis. The slot’sOCXO 10 MHz clock can be driven onto the PXI backplaneclock lines in lieu of the PXI backplane clock. Then, the wholePXI chassis can inherit the OCXO clock stability.

Trigger Bus – The PXI eight-line trigger bus providesintermodule synchronization and communication. Trigger or clocktransmission can use the trigger bus lines. You can pass triggersfrom one module to any number of modules, so you can distributedigital trigger signals from master to slave measurement devices.With variable frequency sampling clock transmission, multiplemodules can share a timebase that is not a derivative of the 10MHz reference clock. For example, four data acquisition modulesusing a 44.1 kS/s CD audio sampling rate can share a clock that isa multiple of the 44.1 kHz or the direct 44.1 kS/s clock. For high-speed synchronization, the propagation delay and skew betweenslots can reach up to a maximum of 10 ns on a single PXI back-plane.

Star Trigger for Ultra High Speed Synchronization –The Star trigger bus has an independent trigger line for each slot


that is oriented in a star configuration from a special Star triggerslot (defined as slot 2 in any PXI chassis). The trigger can pro-vide an independent dedicated line for each of up to 13 peripher-al slots on a single PXI backplane. The PXI Star line lengths arematched in propagation delay to within one nanosecond from theStar trigger slot. This feature addresses ultra high-speed synchro-nization where you can distribute start/stop trigger signals fromthe master measurement module in the Star trigger slot with lowdelay and skew.

Platform Trigger Bus Reference Clock Star Bus

VXI 8 TTL, 2 ECL 10 MHz ECL Yes

PXI 8 TTL 10 MHz TTL Yes

CONCLUSIONSComputer-based measurement components are transforming

creation of synchronized measurement systems from integration ofloosely coupled, and often incompatible instruments, into anorderly engineering process that results in tightly integrated,high-performance systems. For synchronized measurements,timing and triggering details are critical keys to your automatedmeasurements. Precise synchronization requires proper distributionof clocks and triggers. The three main synchronization schemesand proper knowledge of the pros and cons of each and the capa-bilities of your measurement devices help you to make the rightdecision in choosing your solution.


1. INTRODUCTIONThis article describes the process of designing a closed

loop control system, or plant, using the NI System Identificationand Control Design Assistants. A DC Motor will be the plant(Figure 1).

The Quanser Engineering Trainer will be used in velocitymode. A voltage signal commands the motor to move and thetachometer output determines the velocity. The motor system isconnected to a National Instruments Data Acquisition (DAQ)device, where Analog Input 0 (AI0) is connected to the tachome-ter and Analog Output 0 (AO0) is connected to the motor com-mand input. For demonstration purposes, you can replace the DCmotor with an RC circuit.

This example uses the following LabVIEW add-ons:• NI LabVIEW System Identification Toolkit• NI LabVIEW Control Design Toolkit• NI LabVIEW Simulation ModuleYou can purchase these products together in the Control

Design and Simulation Bundle.To use these add-ons, you must install the following

software:• NI LabVIEW 7.1• NI Signal Express 1.0The closed-loop system acts on the difference between

two quantities: the process variable (the voltage output of the

tachometer as a function of motor velocity) and the set point (thecommand voltage you specify). The controller then determinesthe next voltage level to command to the motor to meet the spec-ifications defined while designing the controller. Figure 2 showsthe final closed-loop system.

This example describes the process of designing a simplePI controller for a system with unidentified dynamics. Note thatall functionality described in the Express Workbench environ-ment is also available in LabVIEW. All project scripts andLabVIEW VIs described in this document are available asattachments to this document.

2. IDENTIFYING THE SYSTEMTo identify an open loop system we need to excite it with

a signal that has voltage levels and frequency content that corre-sponds to its actual operating conditions. For more informationabout this process, refer to “Stimulus and AcquisitionConsiderations in the System Identification Process,” located atwww.ni.com > NI Developer Zone > Development Library >Analysis and Signal Processing > PID Control/SystemCharacterization/Stability.

You can use many different signal types to identify a sys-tem, including chirp signals, square waves, square waves over-laid with white noise, and so on. For this example, the stimulussignal is a 3V p-p triangular wave. You create this signal usingthe “Create Signal” step available in the NI Express Workbench.Figure 3 shows how you create this signal.

USING NATIONAL INSTRUMENTS SYSTEMIDENTIFICATION, CONTROL DESIGN AND

SIMULATION PRODUCTS FOR DESIGNING ANDTESTING A CONTROLLER FOR AN UNIDENTIFIED

SYSTEMwww.ni.com

Figure 1: The Quanser Engineering Trainer (QET) will be the plant for which we will design aclosed loop controller.

Figure 2. The final closed loop system. The Plant Model is the QET (Figure 1).


Table 1 shows where to find the settings for this particu-lar step and what values to use:

Table 1. Settings for creating the signal

Step Settings/Actions

Signal Input/Output-> Create Signal Signal Type = Triangle Wave

Frequency = 1 Hz

Amplitude = 3 V

Sample Rate = 1kS/s

Block Size = 5000 samples

To display the created signal on the data viewer inExpress Workbench, drag the “Calculated Signal” Output to theData View (Figure 4)

Next, you must have the DAQ device generate this signalas an analog output. Use the DAQmx Generate step to performthis function, shown in Figure 5.

Table 2 shows the settings for this step:

Table 2. Settings for generating the signal on the appropriate “Device” and “Channel” (AO0 in this case).


Signal Input/Output Config Tab: Device: Make sure to-> Generate Signals select appropriate DAQ Device and-> NI DAQmx Generate Channel

NOTE: This example does not synchronize the AO and AIchannels of the DAQ device. Typically you should synchronizethese channels, which you can accomplish using the AdvancedTiming page, because any delays caused by the difference in timingbetween AI and AO would be described by the transfer function ofthe open loop system, resulting in some error in the identification. Inthis example, the sample rate for AI and AO is 1 kHz, so the maxi-mum jitter between the two channels is 0.5 ms. This amount of jitteris negligible compared to the plant dynamics.

Use the DAQmx Acquire step to acquire the response ofthe plant to the stimulus signal. Figure 6 shows this acquisition.

Figure 3. Creating a 3 Volt p-p triangular wave.

Figure 4. The created signal in the data viewer in Express Workbench.

Figure 5. DAQmx Generate step to output the created signal as an analog signal on theDAQ card.


Table 3 shows the where to find and settings for this step:

Table 3. Settings for acquiring the signal on the appropriate “Device” and “Channel” (AI0 in this case).


Signal Input/Output Config Tab: Device: Make sure to-> Acquire Signals select Device and Channel-> NI DAQmx Acquire

Config Tab: Acq. Timing: 5000 samples to read

Config Tab: Acq. Timing: 1 kHz sample Rate

Next, drag the output of the DAQmx Acquire step to thedata viewing window. Express Workbench notifies you that thedata from the current step appears unrelated to the data alreadyon the display. A small disconnect symbol, circled in Figure 7, isalso displayed between the DAQmx Generate step and DAQmxAcquire step, which indicates that the steps below the disconnect

symbol are not dependent on the steps above. This symbol dis-appears after you use the Create and Acquire steps in the systemidentification process.

When this dialog box appears, select the No button to createa new display for the signal.

Next, run the project script once by clicking the greenrun arrow. This project generates and acquires 5000 data pointsat 1 kS/s for a total of five seconds of plant response data. Thisresponse data appears in the display you added in the previousstep. Figure 8 shows the stimulus signal and the plant responsedata.

You use the stimulus signal and the response data todefine a transfer function for the open loop DC motor system. Todefine this model, you will use a parametric estimation of themotor model. For more information about parametric estimation,refer to “System Identification Model Structure Selection,”located at www.ni.com >> NI Developer Zone >> DevelopmentLibrary >> Analysis and Signal Processing >> PID Control /System Characterization / Stability.

In this example, you use the default settings of theParametric Estimation step, shown in Figure 9, to create a first-order transfer function. The model order is based on the plantdynamics.

Figure 6. DAQmx Acquire step to acquire the response back from the DC motor plant as ananalog signal to the DAQ card.

Figure 7. Create a new display to view the output from the DAQmxacquire step. The yellow circle locates the disconnect symbol dis-played between the DAQmx Generate and DAQmx Acquire steps. Thissymbol indicates that the steps below the disconnect symbol are notdependent on the steps above.

Figure 8. The stimulus signal is in the upper display. The plant response to this signal is inthe lower display.


Table 4 shows where to find and the settings for this step:

Table 4. Settings for identifying a parametric estimation of the QET DC motor plant system.


System Identification Input Signals and Model Tab:-> Model Estimation Stimulus Signal: Calculated signal> Parametric Estimation Response Signal: Device and

Channel from NI DAQmx Acquirestep.

Add Display under “DAQmx Acq”Output Display

Drag “Estimated Response” to newDisplay

Notice that the disconnect symbol, shown in Figure 7, nolonger appears. NI Express Workbench removes this icon becauseyou used the Create and Acquire steps in the system identificationprocess. Also, notice that the largest prediction error typicallyoccurs in the beginning of the signal. This error occurs for tworeasons: the initiation effects of spinning up the system (which istypically not in perfect mechanical balance) and because thenumerical algorithm used to identify the model requires severaltime steps to initialize itself. For example, the disc that this par-ticular QET DC motor spins has 2 holes drilled through it, anddepending on the location during startup the motor, might startslower or faster. Therefore, the coefficients of the resulting transferfunction change slightly every time you run the final ExpressWorkbench Project Script.

After you identify a model, you must save the transferfunction for further analysis. Select System Identification»Import-Export Model»Save System Identification Model, shownin Figure 10, to save this model.

At this step in the example, the transfer function is dis-crete. Although you can design a discrete proportional-integral

(PI) controller in Express Workbench, this example converts thetransfer function model to a continuous one because the motor isa continuous plant. To facilitate this design in the continuousdomain, also known as the s-domain, this example transfers themodel into a Control Design type function and then converts themodel into continuous representation. Figures 11a and 11b showthis process.

To display the step shown in Figure 11a, select SystemIdentification»Import-Export Model»Convert to Control DesignModel.

To display the step shown in Figure 11b, select ControlDesign»Model Transformation»Discretize Model. On theConfiguration page of this step, select Make Continuous fromthe Operation pull-down list.

Figure 9. Identifying a parametric estimation of the DC motor plant system.

Figure 10: Saving the System Identification Model

Figure 11a: Converting the System ID Model to a Control Design Model Type (Transfer Function)


Next, create a new display for the continuous transferfunction. To create this display, right-click the DAQmx AcquireResult display in the data viewer window and select AddDisplay»Below from the shortcut menu. Figure 12 shows thiscreation.

Then, drag the output of the Discretize Model step to thenew display, shown in Figure 13.

The system has now been identified as a first order trans-fer function. Every time the Express Workbench project script isrun, the coefficients of the transfer function will change slightly.This is due to the spinning wheel and other mechanics of themotor itself (and the tachometer, and the fact that AI and AO isnot 100% synchronized etc.), as explained above during theEstimation of the Parametric Model.

3. DESIGNING THE CONTROLLERNow that the transfer function of the plant is available, the

next step is to design a controller for this plan. This exampledescribes how to design a controller that meets requirements forrise time, settling time, overshoot, and so on. This controller willcomplete the closed-loop system.

This example designs a simple proportional-integral (PI)controller using the PID Synthesis step, shown in Figure 14. Youalso can perform a root locus or interactive bode design.

Table 5 lists the settings for designing a PI controller.

Table 5. Settings for designing a PI controller for the DC Motor Plant System.


Control Design -> Controller Synthesis Tab: Check Controller Design -> “Gain” and "Integral (s)" boxesPID Synthesis

Adjust P and I gains to obtaindesired step response. Refer to Figure14 for the recommended settings.

Figure 11b. Making the discrete model continuous

Figure 12. Adding a window to display the transfer function for the open-loop system.

Figure 13. Displaying the transfer function of the open-loop system plant in the s-domain.

Figure 14. Designing a PI controller for the DC motor.


As you adjust the values of the P and I gains, the stepresponse graph changes to show the resulting rise time, over-shoot, ringing, settling time, and so on. Adjust the P and I gainsso the step response looks similar to the step response shown inFigure 14. This step response has a rise time of approximately 25ms and overshoot of less than 50% of the steady state value.Optionally, you can check these time domain specifications byadding a Time Domain Analysis step after the PID Synthesis step

WARNING: Too much overshoot can causes the outputof the controller to command a voltage much higher than theAnalog Output board and the motor can handle. However, lateron in this example, you will use the Simulation Module toenforce a limit on the valid range of the output.

After you have properly adjusted the P and I gains, savethe model by using the Save Control Design Model step, locat-ed at Control Design»Import-Export Model. Figure 15 showsthis step.

4. SIMULATING THE CLOSED LOOP SYSTEM

In this example, the previous sections provided informationabout identifying the plant model and designing a PI controllerbased on this plant model. Before you use this controller on theactual DC motor, you use the Simulation Module to verify thecontroller behaves as you expect. The Simulation Moduleincludes several ordinary differential equation (ODE) solversyou use to integrate the continuous transfer function model overa period of time. For more information about the SimulationModule, refer to www.ni.com >> Products & Services >> Real-Time Measurement and Control >> NI Real-Time Software >>Add-On Toolkits >> Simulation Module.

Figure 16 shows the LabVIEW block diagram, includingthe Simulation Loop that defines the simulation diagram. Noticethe pale yellow color of the simulation diagram to distinguish itfrom the LabVIEW block diagram. Also notice that the SimulationModule allows you to directly implement feedback, completingthe closed-loop system.

Notice LoadController.vi and LoadPlant.vi. These subVIsload the models that you created in Express Workbench andtransfer the models into the appropriate Simulation functions.The LoadController subVI also converts the discrete transferfunction into a continuous one. Recall that you implemented thisstep in the Express Workbench. However, all functionality availablein the Express Workbench environment is available in LabVIEW.

Figure 17 shows the response of the closed-loop systemto a square wave input. Refer to Figure 14 to verify that this isthe expected behavior.

Notice the knobs on the front panel of Figure 17. You usethese knobs to change the type, amplitude, and frequency of thestimulus signal while immediately viewing the response of theclosed-loop system.

5. DRIVING THE MOTOR WITH THE CLOSED LOOP SYSTEMNow that you have verified the closed-loop response of the

plant and controller models, the next step is to use this controllerto drive the actual DC motor. First, this example demonstrates anopen-loop system.

NOTE: This example does not synchronize the input andoutput values of the system, because, the short jitter (≤0.5mswith 1000 kS/s analog input and output) ensures that the outputdoes not display any significant difference.

Figure 15. Saving the PI controller.

Figure 16. Using the Simulation Module to simulate the behavior of the identified plant andthe PI controller in a closed-loop configuration.

[+] Enlarge Image

Figure 17. Stimulating the closed-loop system with a square wave input and showing the response.


Figure 18 shows the LabVIEW block diagram for drivingthe DC motor in an open-loop configuration.

Figure 19 shows the front panel of this block diagram.

Figure 19 shows how you specify the motor speed in rota-tions per minute (RPM). This example converts this value to thecorresponding analog voltage as directed by the manufacturer ofthe DC motor. In this situation, the multiplier is 0.0015 volts/RPM.The Analog Output Channel 0 (AO0) of the DAQ device thensends this value to the DC motor.

This example then uses Analog Input Channel 0 (AI0) ofthe DAQ device to acquire the data from the tachometer of theDC motor. This example then converts the tachometer value toRPM by using the manufacturer-supplied multiplier of 666.6RPM/volts.

After you press the Stop button, this example stops themotor by sending a value of 0 volts to AI0.

Notice in Figure 19 that the motor is slow to respond toany change in specified RPM. This example also demonstratessteady-state error, which is a permanent difference between thespecified and actual motor speeds. This error is due to the cali-bration uncertainty in the multiplication constants Figure 18shows. The steady-state error is particularly noticeable at highspeeds, because high speeds increase the relative error thatresults from not multiplying with the exact conversion factor.

By closing the loop and adding the PI controller to theopen-loop system, the response of the motor becomes faster andmore accurate with respect to the RPM you specify. The con-troller compares the actual speed of the motor with the speed youspecified and adjusts the motor speed accordingly. Figure 20shows this increase in response time and accuracy.

The integration term in the PI controller minimizes thesteady-state error by taking the history of the error into account.Figure 21 shows the LabVIEW block diagram that correspondsto the front panel shown in Figure 20.

The block diagram in Figure 21 converts the speed it isconverted to and from corresponding Analog Voltage using thesame multipliers described in Figure 18. The actual speed of themotor is compared with the speed you specify, or the Set Point.The controller is loaded from file as shown in Figure 15.

Figure 21 shows how Saturation function limits the outputvoltage of the motor. This figure also shows how you can use theSIM Set Diagram Params VI to programmatically change theODE solver and other parameters of the simulation.

NOTE: In the real world, the Saturation function is notnecessary, because the DAQ Analog Output Assistant ExpressVI has a control that sets a limit on the output voltage. However,this example demonstrates the capabilities of the SimulationModule and how you would place Saturation function in a closed

Figure 18. Driving the plant (DC Motor) in an open loop configuration.

[+] Enlarge Image

Figure 19. Driving the plant (DC Motor) in an open loop configuration. The response fromthe motor is slow.

Figure 20. Driving the plant (DC Motor) in a closed loop configuration. The response fromthe motor is fast, with overshoot, settling time characteristics as defined while designing thecontroller (Figure 14).

Figure 21. Driving the plant (DC Motor) in a closed loop configuration.


loop. Also, if you place a LabVIEW data probe before and afterthe Saturation function, changing the Set Point suddenly cancause the motor to overshoot.

6. CONCLUSIONSThis example described how you can use LabVIEW and

related software to identify, control, and simulate a real-worlddynamic system. Although this example did not use any real-time (RT) hardware, you can use the LabVIEW Real-TimeModule in conjunction with the Simulation Module to deploy acontroller to any National Instruments RT Series hardware.

Refer to Using CompactRIO, located at http://sine.ni.com/csol/cds/item/vw/p/id/538/nid/124200, for an example thatdemonstrates how to build a full-authority FPGA-based enginecontrol system for a high-performance motorcycle engine.

NOTE: You also can describe the simulation itself in theExpress Workbench Project Script by adding a User-DefinedStep. You also can translate an Express Workbench project scriptinto LabVIEW code by launching LabVIEW and selectingTools»Express Workbench»Convert Express Workbench Projectfrom the pull-down menu.

More complex systems, such as the high performancemotorcycle engine described above, may have multiple inputsand multiple outputs. In these situations, you can use state-spacemodel identification and control design methods to operate in themultiple-input multiple-output (MIMO) environment. TheControl Design Toolkit, System Identification Toolkit andSimulation Module support these design methods.


GOALSThis project specializes in measurements of the effect of

mechanical strain on superconductor properties such as critical-current density for applications in magnetics, power transmis-sion, and electronics. Recent research has produced the firstelectromechanical data for the new class of high-temperaturecoated conductors, one of the few new technologies expected tohave an impact on the electric-power industry. The StrainScaling Law, previously developed by the project for predictingthe axial-strain response of low-temperature superconductors inhigh magnetic fields, is now being generalized to three-dimen-sional stresses, for use in finite-element design of magnet struc-tures, and to high-temperature superconductors. Recent researchincludes extending the high-magnetic-field limits of electro-mechanical measurements for development of nuclear-magnetic-resonance (NMR) spectrometers operating at 23.5 teslas and 1gigahertz, and the next generation of accelerators for high-ener-gy physics. The project has diversified its research to includemagnetoresistance studies on a new class of carbon nanostruc-tures using our highfield superconducting magnet facility and anewly developed, variable-angle, variable-temperature measure-ment capability.

CUSTOMER NEEDSThe project serves industry primarily in two areas. First is

the need to develop a reliable measurement capability in thesevere environment of superconductor applications: low temper-ature, high magnetic field, and high stress. The data are beingused, for example, in the design of superconducting magnets forthe magnetic-resonance-imaging (MRI) industry, which providesinvaluable medical data for health care, and contributes 2 billiondollars per year to the U.S. economy.

The second area is to provide data and feedback to industryfor the development of high-performance superconductors. Thisis especially exciting because of the recent deregulation of theelectric power utilities and the attendant large effort being devot-ed to develop superconductors for power conditioning andenhanced power-transmission capability. We receive numerousrequests, from both industry and government agencies, for reliableelectromechanical data to help guide their efforts in research anddevelopment in this critical growth period.

The recent success of the second generation of high-tem-perature superconductors has brought with it new measurementproblems in handling these brittle conductors. We have theexpertise and equipment to address these problems. Stress andstrain management is one of the key parameters needed to movethe second-generation high-temperature coated conductors to themarket place. The project utilizes the expertise and uniqueelectromechanical measurement facilities at NIST to provideperformance feedback and engineering data to companies and

national laboratories fabricating these conductors in order toguide their decisions at this critical phase of coated-conductordevelopment.

TECHNICAL STRATEGYOur project has a long history of unique measurement

service in the specialized area of electromechanical metrology.Significant emphasis is placed on an integrated approach. Weprovide industry with first measurements of new materials, spe-cializing in cost-effective testing at currents less than 1000amperes. Consultation is also provided to industry on developingits own measurements for routine testing. We also provide con-sultation on metrology to the magnet industry to predict and testthe performance of very large cables with capacities on the orderof 10 000 amperes, based on our tests at smaller scale. In short,our strategy has consistently been to sustain a small, well con-nected team approach with industry.

Electromechanical Measurements of Superconductors –We have developed an array of specialized measurement systemsto test the effects of mechanical stresses on the electrical per-formance of superconducting materials. The objective is to sim-ulate the operating conditions to which a superconductor will besubjected in magnet applications. In particular, since most tech-nologically important superconductors are brittle, we need toknow the value of strain at which fractures occur in the supercon-ductor. This value is referred to as the irreversible strain limit,since the damage caused by the formation of cracks is perma-nent. The effect of cracks is extrinsic. In contrast, below the irre-versible strain, there exists an elastic strain regime where theeffect of strain is intrinsic to the superconductor. In this elasticregime, the variation in the critical-current density (Jc) withstrain, if any, is reversible and is primarily associated withchanges in the superconductor’s fundamental properties, such asthe critical temperature (Tc) and the upper critical field (Hc2), aswell as changes in the superconductor’s microstructure due tothe application of strain.

Measurement Facilities – Extensive, advanced measure-ment facilities are available, including high-field (18.5 teslas) andsplit-pair magnets, servohydraulic mechanical testing systems,and state-of-the-art measurement probes. These probes are usedfor research on the effects of axial tensile strain and transversecompressive strain on critical current; measurement of cryogenicstress-strain characteristics; composite magnetic coil testing; andvariable-temperature magnetoresistance measurements. Ourelectromechanical test capability for superconductors is one ofthe few of its kind in the world, and the only one providing spe-cialized measurements for U.S. superconductor manufacturers.

Collaboration with Other Government Agencies –These measurements are an important element of our ongoingwork with the U.S. Department of Energy (DOE). The DOEOffice of High Energy Physics sponsors our research on electro-

MAGNETO-MECHANICAL MEASUREMENTS FORHIGH CURRENT APPLICATIONS

Jack Ekin, NIST – Electromagnetics Division


mechanical properties of candidate superconductors for particle-accelerator magnets. These materials include low-temperaturesuperconductors (Nb3Sn, Nb3Al, and MgB2), and high-temper-ature superconductors – Bi-Sr-Ca-Cu-O (BSCCO) and Y-Ba-Cu-O (YBCO) – including conductors made on rolling-assisted,biaxially textured substrates (RABiTS) and conductors made byion-beam-assisted deposition (IBAD). The purpose of the data-base produced from these measurements is to allow the magnetindustry to design reliable superconducting magnet systems. Ourresearch is also sponsored by the DOE Office of ElectricTransmission and Distribution. Here, we focus on high-temperaturesuperconductors for power applications, including power-condi-tioning systems, motors and generators, transformers, magneticenergy storage, and transmission lines. In all these applications, theelectromechanical properties of these inherently brittle materialsplay an important role in determining their successful utilization.

Scaling Laws for Magnet Design – In the area of low-temperature superconductors, we have embarked on a funda-mental program to generalize the Strain Scaling Law (SSL), amagnet design relationship we discovered two decades ago.Since then, the SSL has been used in the structural design ofmost large magnets based on superconductors with the A-15crystal structure. However, this relationship is a one-dimension-al law, whereas magnet design is three-dimensional. Currentpractice is to generalize the SSL by assuming that distortionalstrain, rather than hydrostatic strain, dominates the effect. Recentmeasurements in our laboratory suggest however that thisassumption is invalid. We are now developing a measurementsystem to carefully determine the three-dimensional straineffects in A-15 superconductors. The importance of these meas-urements for very large accelerator magnets is considerable. TheStrain Scaling Law is now also being developed for high-temper-ature superconductors since we recently discovered that practicalhigh-temperature superconductors exhibit an intrinsic axial-strain effect.

ACCOMPLISHMENTS• New Measurement

Method for Marginally StableSuperconductor Wires – Thenext generation of particleaccelerators for high-energyphysics, and magnet systemsfor nuclear magnetic resonance(NMR) spectroscopy, will require the development of a new typeof superconducting niobium-tin wire able to carry extremelyhigh currents at high magnetic fields. One way to achieve highcurrents is to push the density of superconductor filaments incomposite wires to new limits. Oxford SuperconductorTechnology (Carteret, NJ) has successfully demonstrated thefeasibility of this concept. However, this could significantlyreduce the beneficial “pre-compressive strain” in these conduc-tors upon cooling, an important parameter for magnet design.Our superconductor electromechanical testing system is the onlyone in the U.S. that utilizes stress-free cooling, which is essentialfor a direct measurement of pre-compressive strain. Unfortunately,the new niobium-tin wires, owing to their relatively small amountof copper stabilizer, are only marginally stable, which makes elec-trical characterization extremely challenging. Hence, a new meas-urement technique was required that did not compromise thestress-free cooling advantage.

The technique consists of measuring critical-current den-sity (the maximum lossless current density that a superconductorcan carry) versus axial strain for a number of copper-plated spec-imens of the same wire with different amounts of copper. Wethen deduced the strain properties of the virgin (noncopper-plated)wire by an extrapolation technique. Copper plating made the nio-bium-tin wires electrically stable enough to characterize, but theextra copper also influenced the value of the pre-compressivestrain (εmax); hence the need for extrapolation. We confirmed thatεmax indeed decreased linearly with increasing niobium fraction.However, we found that other parameters such as the matrixmaterial and wire diameter also influence εmax.

The pre-compressive strain for high-niobium-fractionwires can be reduced to about 0.1 percent, a very small strainwindow for magnet design. Fortunately, we also found that theuse of copper alloys, instead of pure copper – along with smallwire diameters – substantially mitigates the problem and providesreasonable strain operating margins in these high performanceconductors. The data were used by Oxford SuperconductorTechnology to make immediate decisions regarding the conduc-tor design for a new NMR system.

• Copper Stabilizer Improves Coated Superconductors’Strain Tolerance – High-temperature superconductor (HTS)wires are now being fabricated in kilometer lengths, providingthe basis for a new generation of electric power devices, includinghigh power-density motors and generators, transmission lines,and power conditioners. The development of HTS technology isexpected to play a crucial role in maintaining the reliability of thepower grid and upgrading power delivery to core urban areas.The most promising superconductor candidate for replacing age-ing utility equipment is the highly textured Y-Ba-Cu-O (YBCO)compound deposited on buffered flexible metallic substrates.These “coated conductors” have a much higher current-carryingcapacity compared to the Bi-Sr-Ca-Cu-O (BSCCO) tapes nowcommercially available. Whereas BSCCO tapes experience per-manent damage when subjected to axial strains less than 0.2 per-cent, we demonstrated last year that the formation of cracks inthe new YBCO system does not commence until subjected tostrains higher than 0.38 percent, almost a two-fold increase instrain tolerance. This resilience of YBCO to strain is providing astrong motivation to produce commercial lengths of this “secondgeneration” conductor, especially for the design of electric gen-erators for which strain tolerance requirements have been raisedto 0.4 percent.

This year, we found that adding a Cu layer to the YBCOcoated-conductor architecture extends the irreversible strainlimit (εirr) of this composite even further, from 0.38 percent tomore than 0.5 percent. This markedly widens the strain windowfor coated-conductor applications and takes it beyond even themost demanding benchmark for large-scale superconductinggenerators. These measurements were undertaken in close collab-oration with conductor manufacturers American Superconductor(Westborough, MA) and SuperPower (Schenectady, NY), whoare incorporating the stabilizer layers either by Cu-lamination orCu-plating. The original motivation for adding the Cu layers wasto improve the electric and thermal stability of the conductor; thestrain-tolerance dividend was unexpected. We can relate thisremarkable result to the mismatch of thermal contraction betweenCu and the other components of the composite. During samplecooling from processing temperatures to the cryogenic operatingtemperatures, the Cu layer exerts an additional pre-compressivestrain on the YBCO film, and hence extends the irreversible

Pre-compressive strain εmax versusNb fraction for several niobium-tinwires with high niobium density.Data were obtained using a newmeasurement method developedby EEEL researchers for marginallystable superconductor wires.


strain εirr where permanent damage occurs. The Cu may also be act-ing as a crack arrester, which further improves the strain tolerance.

• New Magnetoresistance Apparatus to Probe CarbonNanostructures – Electronic properties of materials changemarkedly as their dimensions approach those of a few atomiclayers. Carbon nanostructures (including graphite sheets, single-walled carbon nanotubes, and multi-walled carbon nanotubes)are prime examples of such potentially useful materials,although some of their very fundamental properties remain con-troversial. Characterization of these structures at high magneticfields is one of the principal methods for determining the exis-tence of ballistic conduction, for example, which could be thefoundation for a new generation of nanoelectronic devices.

We have designed and recently commissioned an appara-tus to measure magnetoresistance of these highly directionalstructures in fields up to 18.5 teslas. (For comparison, the Earth’smagnetic field is only about 0.05 millitesla.) The apparatus auto-matically acquires data as a function of magneticfield magnitude,angle, and temperature. It was designed to also be compatiblewith the very-highfield magnet facilities at the National HighMagnetic Field Laboratory at Florida State University, permittingthe extension of EEEL’s measurements to fields up to 30 teslas.Magnetic field mapping has commenced for nanotubes fabricatedat NIST and Rice University as well as for graphitic sheet struc-tures manufactured by a nanotechnology research team at GeorgiaInstitute of Technology. Magnetic-field angle can be varied with aresolution of better than 0.1 degree over a range of 130 degrees,and sample temperature can be varied over an extended range of4.1 to 120 kelvins, with a stability of better than 3 millikelvins at4.2 kelvins.

• Textbook on Cryogenic Measurement Apparatusand Methods – A new textbook has been written on experimen-tal techniques for cryogenic measurements to be published byOxford University Press. It covers the design of cryogenic meas-urement probes and provides cryogenic materials data for theirconstruction. Topics include thermal techniques for designing acryogenic apparatus, selecting materials appropriate for suchapparatus, how to make high-quality electrical contacts to asuperconductor, and how to make reliable criticalcurrent meas-urements. The textbook is written for beginning graduate stu-dents, industry measurement engineers, and materials scientistsinterested in learning how to design successful low-temperaturemeasurement systems. The appendices are written for experts inthe field of cryogenic measurements and include electrical, ther-mal, magnetic, and mechanical properties of technical materialsfor cryostat construction; properties of cryogenic liquids; andtemperature measurement tables and thermometer properties.These appendices aim to collect in one place many of the dataessential for designing new cryogenic measurement apparatus.

Normalized critical current density as afunction of mechanical tensile strainfor unlaminated and CulaminatedYBCO coated conductor. The Cu stabi-lization layer extends the irreversiblestrain limit εirr of the composite from0.38 percent to more than 0.5 percent.

Photograph of a new magnetoresistanceprobe designed to investigate carbonnanostructures. At the right end of theprobe, the photo shows the steppermo-tor-controlled worm-gear system andsample stage, which allow precise angle-dependent, high-field measurements.


THERMOGRAPHYThermography is a method of inspecting electrical and

mechanical equipment by obtaining heat distribution pictures.This inspection method is based on the fact that most compo-nents in a system show an increase in temperature when mal-functioning. The increasein temperature in an electri-cal circuit could be due toloose connections or a wornbearing in the case ofmechanical equipment. Byobserving the heat patternsin operational system com-ponents, faults can be locatedand their seriousness evalu-ated.

The inspection toolused by Thermographers isthe Thermal Imager. These are sophisticated devices whichmeasure the natural emissions of infrared radiation from a heatedobject and produce a thermal picture. Modern Thermal Imagersare portable with easilyoperated controls. As phys-ical contact with the systemis not required, inspectionscan be made under fulloperational conditionsresulting in no loss of pro-duction or downtime.

The Land CyclopsThermal Imager is a devicedesigned for plant conditionmonitoring, preventativemaintenance and process monitoring applications.

Potential applications include: • Inspection of electrical equipment • Inspection of mechanical equipment • Inspection of refractory lined structures

MEASUREMENT OF TEMPERATURE USING INFRAREDMETHODS

When using a Thermal Imager it is helpful to have a basicknowledge of infrared theory.

BASICS PHYSICS An object when heated radiates electromagnetic energy.

The amount of energy is related to the object’s temperature. TheThermal Imager can determine the temperature of the objectwithout physical contact by measuring the emitted energy.

ELECTROMAGNETIC SPECTRUM The energy from a heated object is radiated at different

levels across the electromagnetic spectrum. In most industrialapplications, it is the energy radiated at infrared wavelengthswhich is used to determine the object’s temperature. Figure 3shows various forms of radiated energy in the electromagneticspectrum including X-rays, Ultra Violet, Infrared and Radio.They are all emitted in the form of a wave and travel at the speedof light. The only difference between them is their wavelengthwhich is related to frequency.

The human eye responds to visible light in the range 0.4to 0.75 microns. The vast majority of infrared temperature meas-urement is made in the range 0.2 to 20 microns. Although emis-sions are mostly unable to be detected by a standard camera theThermal Imager can focus this energy via an optical system on toa detector in a similar way to visible light. The detector convertsinfrared energy into an electrical voltage which after amplificationand complex signal processing is used to build the thermal picturein the operator’s viewfinder on board the Thermal Imager.

ENERGY DISTRIBUTION Figure 4 shows the

energy emitted by a targetat different temperatures.As can be seen, the high-er the target temperaturethe higher the peak ener-gy level. The wavelengthat which peak energyoccurs becomes progres-sively shorter as tempera-ture increases. At lowtemperatures the bulk ofthe energy is at longwavelengths.

A BASIC GUIDE TO THERMOGRAPHY Land Instruments International Infrared Temperature Measurement

Figure 1. The Thermal Image of electrical connector

Figure 2. The inspection of electrical equipmentusing a Thermal Imager

Figure 4. Infrared energy and distribution acrossthe Electromagnetic spectrum


EMISSIVITYThe amount of energy radiated from an object is dependant

on its temperature and its emissivity. An object which has theability to radiate the maximum possible energy for its temperatureis known as a Black Body. In practice, there are no perfect emit-ters and surfaces tend to radiate somewhat less energy than aBlack Body.

Figure 5 shows why objects are not perfect emitters ofinfrared energy. As energy moves towards the surface a certainamount is reflected back inside and never escapes by radiativemeans. From this example, it can be seen that only 60% of theavailable energy is actually emitted. The emissivity of an objectis the ratio of the energy radiated to that which the object wouldemit if it were a Black Body.

Hence emissivity is expressed as

Emissivity is therefore an expression of an object’s abili-ty to radiate Infrared energy.

EMISSIVITY VALUES The value of emissivity tends to vary from one material to

another. With metals, a rough or oxidised surface usually has ahigher emissivity than a polished surface.

Here are some examples:

It can be shown that there is a relationship between emis-sivity and reflectivity.

For an opaque object this is Emissivity + Reflectivity = 1.0

Hence, a highly reflective material is a poor emitter ofinfrared energy and will therefore have a low emissivity value.

EFFECTS OF EMISSIVITY If a material of high emissivity and one of low emissivity

were placed side by side inside a furnace and heated to exactlythe same temperature, the material with low emissivity wouldappear to the eye much duller. This is due to the different emis-sivities of the materials causing them to radiate at different levels,making the low emissivity material appear cooler than the highemissivity material, even though they are at exactly the same tem-perature.

The Thermal Imager would see this in the same way asthe eye and produce an error in making the temperature measure-ment. The temperature of an object cannot be determined by simplymeasuring its emitted infrared energy, a knowledge of theobject’s emissivity must also be known.

The emissivity of an object can be determined as follows: 1) Consult manufacturers literature (always ensure these

have been evaluated at the operating wavelength of yourThermal Imager as emissivity can vary with wave-length).

2) Have the object’s emissivity evaluated by a laboratorymethod.

There are two main ways to overcome the problem ofemissivity.

a) Mathematically correct the temperature measurementvalue. This is usually carried out within the signalprocessor of the Thermal Imager. Most modernThermal Imagers have a compensation setting whichcan quickly and easily be set by the operator.

b) It may be possible to paint the surface of a low emis-sivity target with a high and constant emissivity coating.This tends to elevate the target to a much higher emis-sivity level, but this may not be possible on all processplants.

When carrying out Thermographic inspections, faults areoften identified by comparing heat patterns in similar compo-nents operating under similar loads. This is an alternative to veryprecisely predicting the emissivity of each individual componentand obtaining absolute temperature values.

Figure 5. The Infrared energy reflected at a body surface

Thermal Imager being used to inspect electrical equipment. With equal load and emissivitiesthe temperature of the three measurement points should be the same.


THERMAL IMAGERS Thermal Imagers are sophisticated devices which measure

the natural emissions of infrared radiation from a heated objectand produce a thermal picture. Modern Thermal imagers such asthe Land TI814 are usually very flexible containing many stan-dard and optional features. Here are some of those of the TI814.

OPTICAL: A motorised focus is used to obtain a clear image at dif-

ferent distances from the thermal imager. The focus distance isfrom 380mm/15 inches to infinity. An electronic zoom functionenables 2X and 4X magnification of the image.

IMAGE DISPLAY: The real time thermal image is displayed in colour on a

102mm / 4 inch LCD screen. The image may be colourised by any one of the eight dif-

ferent palettes available. The real time thermal image is also displayed on the built-

in high resolution colour viewfinder.

DIGITAL MEMORY: A built in non volatile memory system enables the simple

capture of a large number of thermal images. Thermal images arestored on a removable compact flash memory card. This on boardfacility enables stored image recall to the viewfinder and selectiveimage deletion.

Several seconds of digital voice clip may be stored witheach image and replayed or re-recorded on board the imager. Thesound file can be replayed in by the imager or with image pro-cessing software.

A 256MB card is capable of storing up to 1000 thermalimages and up to an eight second digital voice clip with eachimage. Image file size including voice annotation is 256 KB.

Transfer to image processing software for further imageprocessing and report generation is via a USB Compact Flashmemory card reader.

TEMPERATURE MEASUREMENT: Temperature measurement at single point in the scene is

possible.

POST PROCESSING: This facility enables the generation of further temperature

analysis in the imager viewfinder on stored images. A singlemovable point enables spot measurement at any point in thescene and a movable cursor generates a temperature profile trace.

IMAGE PROCESSING SOFTWARE Frames of interest may be stored as an image file for

record purposes, or be subjected to a range of processing func-tions as follows:

a) File handling: save, delete and directory facility b) Image colouring: the image may be colourised using

any one of five colour palettes. c) Temperature measurement: a variety of different modes

are available to enable temperature measurement at anypoint in the scene, calculation of maximum, minimumor mean from within any defined area in the scene, pro-files, histograms, and isotherms.

d) Parameter changes: parameters saved with the storedimage may be changed within the software. Theseinclude emissivity, and background temperature.

e) Image enhancements: filtering, and zoom facility. Figures 7 to 12 show some of the available temperature

measurement modes.

Figure 7. Measuring the temperature at several points in the scene

Figure 8. Measuring the average temperature within several rectangles in the scene

Figure 9. Measuring the average temperature within several polygons in the scene

Figure 10. Measuring the temperature along several profiles in the scene


The software system is menu driven, making it extremelyeasy to use.

Report Writer: The image processing system provides areport writing facility. This may be used to provide a hard copyrecord of the thermal image accompanied by an imported photo-graph and any other information for reference purposes.

THERMAL IMAGERS IN PREDICTIVE MAINTENANCE APPLICATIONSIn today’s industrial plants it is essential that unplanned

breakdowns and the resultant costly loss of production is kept toan absolute minimum. Predictive maintenance schemes havebeen introduced to identify potential problems and reduce down-time.

Thermography in maintenance applications is based onthe fact that most components show an increase in temperaturewhen malfunctioning and faults steadily get worse before failure.

Routine inspection programmes using Thermal Imagerscan often offer the following benefits:

Inspections can be made under full operational conditionsand hence there is no loss of production.

• Equipment life can be extended • Plant downtime may be reduced • Plant reliability may be increased • Plant repairs scheduled for the most convenient time • Quality of repair work may be inspected • Thermal Imagers are mainly used for industrial predic-

tive maintenance in the following areas:• Electrical Installations • Mechanical Equipment • Refractory lined Structures

INSPECTING ELECTRICAL INSTALLATIONS Faults in an electrical installation often appear as hot-

spots which can be detected by the Thermal Imager. Hot spotsare often the result of increased resistance in a circuit, overloading,or insulation failure. Figure 14 shows a hot-spot created by a badconnection in a power distribution system.

Some of the components commonly inspected are as follows:Connectors: When looking at similar current carrying

connectors, a poor connection shows a higher temperature due toits increased resistance. Hot-spots can be generated as a result ofloose, oxidised, or corroded connectors.

Figure 11. Measuring the temperature distribution within a defined area in the scene

Figure 12. Using Isotherm to highlight areas of the scene within a selected temperatureband

Figure 13. Typical items page in a report generated by the report writer facility

Figure 14. Inspection of a power system

Figure 15. Inspection of connectors


Figure 15. Shows the fuses in the control panel of amachine. A faulty connection on the top of a fuse has created thehot-spot which can easily be seen by the imager.

Three phase motors: Require balanced phases and correctoperating temperatures. It has been shown that if correct operatingtemperatures are exceeded, the insulation life can be considerablyshortened.

Other commonly inspected components are: • Relays • Insulators • Capacitors • Switches

INSPECTION OF MECHANICAL EQUIPMENT The type of mechanical equipment inspected is often rotat-

ing machinery. Increased surface temperatures can be the result ofinternal faults.

Excessive heat can be generated by friction in faulty bearingsdue to wear, misalignment or inadequate lubrication.

As with electrical installations, it is desirable to performthe inspection with the system in operation wherever practicallypossible. Interpretation of results should be based on comparisonbetween components operating in similar conditions under similarloads or by trend analysis.

Equipment commonly inspected using thermal Imagers isas follows:

• Bearings• Gears• Drive Belts• Couplings• Shafts and Pumps.

INSPECTION OF REFRACTORY LINED STRUCTURESThe refractory structures of process plants can often have

an increased lifetime if the degree of wear and erosion can beassessed.

Thermal patterns produced by viewing the outer walls ofa structure can indicate hot-spots caused by worn refractorieswhich may be corrected by appropriate maintenance.

Figure 17 shows an abnormal heat pattern on the wall ofa cement kiln, which has been caused by erosion of the refracto-ry brick liner.

Equipment commonly inspected using Thermal Imagersis as follows:

• Electric Arc Furnaces• Ladles, Heat Treatment Furnaces• Glass Furnaces• Rotary Kilns and Dryers

Figure 16. Inspection of bearing housing

Figure 17. Inspection of a Kiln shell

3M CanadaPO Box 5757London, Ontario N6A 4T1Tel: (800) 3M HelpsFax: (519) 452-6286E-mail: [email protected]: www.mmm.com

Description of products/services:• Terminations and splices, using Cold Shrink®

Technology,moulded rubber, resin and heat shrink• Motor lead connection systems• Scotch® vinyl insulation tapes, splicing and terminatingtapes, corrosion protection sealing and general use tapes• Scotchloc® terminal, wire connectors and insulation displacementconnectors, lugs, copper and aluminum connectors• Scotchtrak® infrared heat tracers and circuit tracers• Fastening products, coatings and lubricants• Duct- , packaging, filament-, and masking tapes• Abrasive products• Personal safety products, sorbents.

B.G. High Voltage Systems Ltd.1 Select Avenue, Units 15 & 16Scarborough, ON M1V 5J3Tel: (416) 754-2666 ext. 202Fax: (416) 754-4607E-mail: [email protected]: B. J. (Bert) Berneche, C.E.T., President

Description of products/services: B.G. High VoltageSystems offers a comprehensive approach to electrical projectmanagement, providing design, construction and engineeringservices to meet all your requirements. We team up with ourclients to ensure that all their needs are defined and met at eachstage of the project. Our experts will coordinate with your engi-neering personnel to ensure minimal disruption to facility oper-ations. As well as complete electrical project management weoffer: material procurement, maintenance and training services,emergency repair, overhead and underground distribution con-

struction and engineering, street and parking lot lighting instal-lation and maintenance. Now available - Power Quality fieldsurvey, monitoring and solutionsn to power quality problems.

CD Nova Ltd.5330 Imperial St.Burnaby, BC V5J 1E6Tel: (604) 430-5612Fax: (604) 437-1036Contat: Don BealleE-mail: [email protected]: www.cdnova.com

CD NOVA companies distribute and service, in Canada,Energy and power Systems and devices, Transducer, Test andMeasurement Instruments, Batteries, Chargers, UPS, Wirelineand Wireless Comm. systems, SCADA systems, Power QualityAnalysers and systems. Teleprotection, Transformers, BreakersProtective Relays, Gas and chemical Analysers, Stack samplinesystems.

Duncan Instruments Canada Ltd. 121 Milvan DriveToronto, Ontario M9L 1Z8 Tel: 416 742-4448 Fax:416 749-5053 Email: [email protected] www.duncaninstr.com

Description of products/services: Duncan InstrumentsCanada is a leading manufacturers’ representative and master dis-tributor for a wide range of utility and electrical instrumentation.

We can offer you data loggers, power line analyzerspower/energy/harmonics analyzers, power disturbance monitorsand fused test leads/accessories.

In addition to sales, Duncan Instruments Canada can alsoprovide: calibration – traceable to NRC, technical product sup-port and application training, instrument repair/modifications,and rental of selected electrical instruments. Registered to ISO9001:2000

BUYER’S GUIDE


Flir Systems5230 South Service Road #125Burlington, ONTel: (905) 637-5696Fax: (905) 639-5488Web: www.flirthermography.com

FLIR Systems Ltd. (Agema Inframetrics) designs, manu-factures, calibrates, services, rents and sells many models ofinfrared imaging cameras and accessories. Complete predictivemaintenance solutions include the ThermaCam PM 695 radio-metric camera with thermaland visual images, autofocus, voiceand text messaging and of course Reporter analysis softwarewith "drag-n-drop" image transfer software. Level's 1, 2 and 3Thermography training conducted on site or at ITC facility.

Camera accessories, such as close-up and telescopic optics,batteries, etc. can be sourced directly from Canadianservice/sales depot in Burlington, ON. Ask about trade inallowances.

FLUKE ELECTRONICS Canada LP400 Britannia Rd. East Unit 1Mississauga, ON, Canada

L4Z 1X9Toll Free : 1-800-36-FLUKETel : (905) 890-7600Fax : (905) 890-6866Contact : Robin BrickerE-Mail : [email protected]

Fluke Electronics Canada (www.flukecanada.ca) offerscomplete families of professional test tools, including powerquality, thermography, digital multimeters, clamp meters, insu-lation resistance testers, portable oscilloscopes, thermometers,process testing equipment and accessories, as well as education-al and training resources. A subsidiary of Fluke Corporation,Everett, Washington, Fluke Electronics Canada is headquarteredin Ontario with offices across Canada. The Fluke brand has areputation for quality, portability, ruggedness, safety and ease ofuse and Fluke test tools are used by technical professionals in avariety of industries throughout the world.

G.T. WOOD CO. LTD.3354 Mavis RoadMississauga, ON L5C 1T8Tel: (905) 272-1696

Fax: (905) 272-1425E-Mail: [email protected]: www.gtwood.com/flash/splash.html

Specializing in High-Voltage Electrical Testing, inspec-tions, maintenance and repairs. Refurbishing and repair of Newand Reconditioned Transformers, Structures, Switchgear andAssociated Equipment. Infrared Thermography, EngineeringStudies and PCB Management.

High Voltage, Inc.31 Rt. 7A, P.O. Box 408Copake, NY 12516USATel : (518) 329-3275Fax : (518) 329-3271Contact : Bob Tighe, E-Mail : [email protected]

Manufacturers of High Voltage Test Equipment. Productsinclude portable AC-VLF, .1Hz, .05 and 0.2Hz Very LowFrequency hipots with sine wave output, switchgear and bottletesters up to 100 kVac. Portable DC hipots up to 300 kV DC.Aerial lift and bucket truck AC test sets up to 300 kVac accord-ing to ANSI standards. Controlled energy cable fault locators,oil test sets and burners also offered.

LIZCO SALESR.R. #3Tillsonburg, ON N4G 4G8Toll Free: 1-877-842-9021Fax: (519) 842-3775Contact: Robin CarrollWebsite: www.lizcosales.com

We have the energy with Canada’s largest on-site directory:• New and Rebuilt Power/Padmount/Dry Transformers• New Oil-Filled “TLO” Unit Substation Transformers• New HV S&C fuses/loadbreaks/towers• High and low voltage:

- Air Circuit Breakers – Molded Case Breakers- QMQB/fusible switches – Combination Starters

• Emergency Service and Replacement Systems• Design/Build custom Application Systems

Megger4271 Bronze Way Dallas, TX 75237-1088 USATel: 1-800-723-2861 Ext. 7360 (Toll Free)

Tel: 214-331-7360 (Direct) Fax: 214-331-7379 Email: [email protected] www.megger.com

Megger is a leading provider of electrical test and measuringequipment for power, industrial, building wiring and communi-cation applications. Its wide range of products extends fromequipment to test protective relays and other substation electricalapparatus, to insulation resistance and ground testers. With threemanufacturing facilities and sales offices located around theworld, Megger is strategically positioned to provide customerswith innovative products, hands-on technical assistance andsuperior service. For additional information, visit our web sitewww.megger.com.

OPTIMUM ENERGY PRODUCTS LTD.#333, 11979 - 40 St SECalgary, AB T2Z 4M3Toll Free (877) 766-5412Main (403) 256-3636Fax (403) 256-3431E-mail: [email protected] Energy Products Ltd are specialists in Power Qualityand Power Metering products. We represent Fluke, AEMCInstruments, Electro Industries, and many other manufacturers.We sell portable PQ instruments for engineers and troubleshoot-ers in many industries. From Plug based voltage disturbancemeters to three phase Class A Power Quality instruments. Wealso supply permanent power and power quality meters for usein residential, commercial and industrial applications.For complete product range and information, please visit ourspecialty websites:www.PQMeterStore.comwww.PowerMeterStore.comwww.ElectricityMetering.comwww.MyMeterStore.com

Raytech USA90 C Randall AvenueWoodlyn, PA 19094Tel: 610-833-3017Fax: 610-833-3018email: [email protected]: www.raytechusa.com

RAYTECH is an employee owned company that special-izes in the design and manufacture of precision test equipmentfor the Electrical Industry. With extensive experience in thedesign and application of test equipment, RAYTECH offersproducts that truly meet the needs of the testing industry. Ourdurable products are used by Manufacturers, Rebuild Shops,


Field Test Crews, Utilities, Rural Electrical CO-OP's, Universitiesand Research Engineers.

RHCtest.com610 Ford Drive Suite 248OakvilleOntario L6J 7W4CanadaTel : (905) 828-6221Fax : (905) 828 -6408Contact : John RiddellE-Mail : [email protected]

RHCtest.com Inc. is a Canadian owned and operatedDistributor of Electrical Test and Measurement Equipment. Wecarry various products lines such as Kyoritsu, Thurlby Thandar,Dataq Instruments, Topward Instruments, Nidec Shimpo, HighVoltage and Midtronics. We distribute products such as;Multimeters, Voltage Testers, Clamp Meters, Clamp Adapters,Voltage and Current Loggers, Power Loggers, Power Analyzers,Insulation Testers, Earth Resistance Testers, Test leads, DC/ACHipots, VLF Hipots, TAN Delta Cable Diagnostics, Thumpers,Cable and Fault locating products, Power Supplies, SpectrumAnalyzers, RF Generators, DDS Generators, Arbitrary WaveformGenerators, Function Generators, LCR Meters, Micro Ohm Meters,Frequency Counters, DMM’s DC Loads, Strobescopes, Hand HeldTachometers, Panel Mount Tachometers, Data Acquisition StarterKits, Stand Alone Data Loggers, Thermocouple Data AcquisitionSystems, DC Connected Data Acquisition Systems and BatteryTesters.

SKM Systems Analysis Inc.1040 Manhattan Beach Blvd.Manhattan Beach, CA 90266USAToll Free : 1-800-232-6789Fax : 1-310-698-4708E-Mail : [email protected]

SKM Power*Tools software helps you design and analyze elec-trical power systems. Interactive graphics, rigorous calculations anda powerful database efficiently organize, process and display infor-mation. Associate projects with multiple one-line diagrams andTCC drawings with customized data fields. Generate better designwith 'what if' scenarios by comparing study results in a single table.Also includes thousands of validated equipment libraries and theability to export project data into AutoCAD DXF and XREF format.Multiple one-line diagrams can be associated with each project forbetter systems organization and presentation. Powerful drawingtools quickly create a structured, interactive one- line diagram sys-tem model.

SKM Systems Analysis, Inc. is a California-based corporationfounded in 1972 with a desire to automate electrical design calcula-tions. SKM has been a leader in the electrical engineering softwareindustry for more than 30 years, providing quality software, trainingand support to thousands of satisfied customers throughout theworld. SKM Systems Analysis, Inc. is also chosen by 39 of the top40 Electrical Engineering firms in the world.

techniCAL Systems 2002 Inc.436 Jacqueline Blvd.Hamilton, Ontario L9B 2R3Canada: 1-86-MEASURE-1 (1-866-327-8731)Tel: 905-575-1941 Fax: 905-575-0386E-mail: [email protected]: www.technical-sys.com

techniCAL provides electrical contractors and utilities withTest, Measurement, Calibration, Control & RecordingInstrumentation. Representing Best-of-Breed Manufacturers;techniCAL provides such products as; Power Quality Analyzers,Micro-Ohmmeters, Megohmmeters, Insulation Testers, LeakageCurrent Meters, Ground Resistance Testers, Data Loggers, HighVoltage Ammeters, Power Transducers, Panel Meters, CT’s, PT’s,Shunts, etc…



HandbookVolume 7


HandbookVolume 7

Electrical T

esting & M

easurement H

andbook Volu

me 7

The E

lectricity Forum

electrical testing measurement handbook vol 7

Documents