monitoring power quality

7/28/2019 Monitoring Power Quality

1/6

Monitoring Power Quality

Franois D. Martzloff

National Institute of Standards and TechnologyGaithersburg, Maryland

Thomas M. Gruzs

Liebert Corporation

Columbus, Ohio

Reprinted from Powertechnics Magazine, February 1990

Significance:

Part 5 Monitoring instruments, laboratory measurements, and test methodsFirst part of a two-part trade magazine version of the IEEE paper Power Quality Site Surveys: Facts,Fiction, and Fallacies. (See file PQ surveys FFF for the original IEEE paper and file Site surveys for

the second part of the trade magazine publication.)

Provides background on the motivation for site surveys and a brief description of the types of monitorsthat have been used, concluding with remarks on how future site surveys should be conducted.

Introduces to the trade literature the term of swell as opposed to surge to descrive a momentaryovervoltage at the power frequency.


2/6

Monitoringpower qualm*Fran~ois . Martzloff .National Institute of Standards andTechnol yGaithersburg, 8arylandThomas M. GruzsLiebert CorporationColumbus, Ohio

Power disturbances that affectsensitive loads have a variety ofsources. Results of site surveys canassist in determining which lineconditioning methods are mostappropriate.

ite surveys are generally initiated to evaluate thequality of the power available at a specific loca-tion with the aim of avoiding equipment distur-bances in a planned installation or of explaining(and correcting) disturbances in an existinginstallation. In either case, survey results constitute oneof the inputs in the decision-making process of providingsupplementary line conditioning equipment, either beforeor after power disturbances have become a problem.Depending on the reliability requirements of the loadequipment, its susceptibility, and the severity of the dis-turbances, various line conditioning methods have beenproposed: surge suppressor (with or without filter), isola-tion transformer, voltage regulator, magnetic synthesizer,motor-generator set, or uninterruptible power supply(UPS).- - - ,Because this additional line conditioning equipmentmay require significant capital investment, the choice ofcorrective measures is generally made by economic trade-off which is the prerogative and responsibility of the enduser. However, if technical inputs to this trade-off areincorrect because erroneous conclusions were drawn as aresult of a faulty site survey, the whole process is worthless- or worse yet, misleading.For this reason, a good understanding of the merits andlimitations of site surveys is essential for reconcilingexpectation with reality before expensive line condition-ing equipment is called for; one should deal, not withfiction or fallacies, but with facts.Power disturbances that affect sensitive electronic loadshave a variety of sources. Lightning, utility switching,and utility outages are often-cited sources of power distur-bances. However, power disturbances are frequentlycaused by users themselves, through switching of loads,ground faults, or normal operation of equipment. As oneexample, computer systems are not only sensitive loads,but can also generate disturbances themselves. Their non-linear load characteristics can cause interactions with thepower system such as unusual voltage drops, overloadedneutral conductors, or distortion of the line voltage.Utility systems are designed to provide reliable bulkpower. However, it is not feasible for them to providecontinuous power of the quality required for a completelyundisturbed computer operation. Because normal use ofelectricity generates disturbances, and because unexpectedpower system failures will occur, every site will experigncesome power disturbances, but their nature, severity, andincidence rates will vary from site to site.Confusion InTerm DefinitionsAs will become painfully apparent in next month'sreview of site surveys, the terms used by the workersreporting their measurements do not have common defini-tions. An effort is being made within the IEEE to resolvethis problem, as described later in this paper, but con-sensus has yet to be reached. In this paper, terms describ-ing disturbances are consistent with the IEEE StandardDictionary of Electrical and Electronics Terms 111 and withestablished usage within the community of surge protec-tive devices engineers.The generally accepted meaning of surge voltage, in thecontext of power systems, is a short-duration overvoltage,typically less than 1 ms or less than one half-cycle of thet'wer frequency. This meaning is not that which haseen established by manufacturers and users of monitoringinstruments and line conditioners. This unfortunate sec-ond meaning is a momentary overvoltage at the funda-mental frequency with a duration of typically a few cycles.

22 FEBRUARY 1990 + POWERTECHNICS MAGAZINE


3/6

P o w e r ally accepted as meaning a momentary voltage reductionat the ac power frequency. However, details (threshold,duration, etc.) of what characterizes a sag are not well

WAVEFORM DISTORTION

Figure 1.Graphcc dep~ctionsof disturbances with multiple meanings

-defined.Motivation For Site SurveysWith the increasing dependency on computer-based systems for industry, commerce, and con-sumers, disruptions from power disturbances areless and less acceptable. Operational problemssuch as hardware damage, system crashes, and pro-cessing errors are the most visible indications ofpower disturbances. Some computer system usersmay accept (albeit reluctantly) such problemsbecause they see them as unavoidable. Othersmay be unaware that otherwise invisible power dis-turbances could be the cause of these problems. Asingle power disturbance can cost more in down-time and hardware damage than the investment inpower protection that would have prevented theeffect of the disturbance. Nearly all sites can bene-fit from a reduction of operational problems byimproving the quality of power supplied to thecomputer systems [ 2 ] .Power line monitoring with sophisticated powerdisturbance recorders has often been advocated asJ a way of determining if any line conditioning isrequired. While monitoring appears to be a logicalfirst step, it has limitations. For example, severedisturbances occur infrequently or on a seasonalbasis. Therefore, monitoring periods of less than aIn this article, this second meaning of the word "surge" (a year might not produce an accurate profile of power distur-momentary overvoltage) will be signaled by the use of bance, and most users are unwilling to wait at least a yearquotat ion marks. Wha t power engineers call surge is to characterize the problem. Also, power line monitoringreferred to as "impulse" or "spike" by the monitoring only produces data on past performance. Changes withininstrument community. Figure . .shows by graphic the site (or at neighboring sites) or by the utility candescriprions the contusion created by thedual meaning of the word surge. ~ c k n o w l -edging the desire of users for terse labels, wepropose for consideration the word "swell"instead of "surge" for a momentary over-voltage.The term "outage" is another example ofconfusion created by unsettled definitions.Most users agree that it means a completeloss of line voltage, but the duration of anoutage can be quite different when definedby computer users (as short as one half-cycle)or power engineers (seconds, perhaps min-utes). Part of the problem may be that thedefinition of "outage" has regulatory implica-tions for evaluating the performance of pub-l i c u t i l i t y co mp an ies . U se r s an dmanufacturers of line conditioners do notmake a clear distinction between completeloss of line voltage (zero voltage condition),severe undervoltages ("deep sags"), or thesingle-phasing of polyphase power systems.For example, a momentary flicker of fluores-cent lighting caused by a brief loss of voltagem~ght e considered an outage; however, abrief sag to less than 80 percent of nominalvoltage will produce the same visible effect.Some UPS manufacturers consider inputvoltage sags that cause transfer to the batterybackup operation as outages.The term "sag" has not yet been definedin the lEEE Dictionary, but it is now gener-

Peripheral Remote elementSeparate signal cablespower introduce an~ U P P ~ Y additional ground2 1- loopElectromagnetic field couples into loop =and induces common mode disturbance Composite lineon input port

Impinging common modedisturbance i s blocked byan isolation transformer Gneglecting interwindingcapacitance

Ground current from other systemscouple common mode disturbanceinto grounding conductor of powersupply

Figure 2.Origins of commo n mode disturbances, including ground cuwent omfther systems ( I ) , electromagnetic field coupling into loop to i w einput port disturbance (2), and introduction of an additional groundloop from remote peripherals (3)POWERTECHNICS MAGAZINE + FEBRUARY1990 23

- - -- -- - -


4/6

P o w e r M o n i t o r i n g three-phase grounded-wye power supplies typical of largecomputer systems, common mode disturbances can also bedefined as the potential difference between neutral and

Nonlinear or switching loadsinteract with power systemsource impedance to causedisturbance on bus bars (B)

. , \supplying load (A)(Bltiagn local systemther loads1 lmplnglng disturbance t A-----?: t Normal mode(lightning, remote Iswltch~ng) :----J ~~~d A) disturbanceI /I

Normal modedisturbances are coupled Ithrough an isolation Itransformer by Imagnetlc coupltng '- ! , - I

~o;clear l~-- definedFigure 3.Ongins of normal mode disturbances include impingingdisturbances such as lightning ( I ) , switching of other loads onlocal systems (2 ) , and interaction between nonlinearlswitchingloads with pwer system source impedance (3)

drastically alter the power disturbance profile in ways thatline monitoring cannot predict.While exact prediction of the disturbances to beexpected at a specific location is almost impossible - andattempting it would be a fallacy - general guidelines canbe formulated. A n attempt has been made by standards-writing groups to provide guidance [3] or specificationsreflecting expected disturbances [4-61.However, users willstill seek specific data for their particular case, and sitesurveys will still be necessary. Another fallacy would beto attempt correction of power line disturbances revealedby monitoring and then to expect operational problems ofequipment to disappear without having first determinedthe exact susceptibility of the equipment. Operationalroblems are the results of more than just power distur-Eances.

Types Of DisturbancesPower line disturbances can be classified into two cate-gories: common mode and normal mode disturbances.The two terms were first defined in the context of com-munication circuits; a recent IEEE Guide [7 ] has estab-lished an expanded definition which is used in this article,as outlined in the following paragraphs. The IEEE Dic-tionary [I] and the IEC Dictionary [8] define symmetricaland asymmetrical voltages akin to, but not interchangea-ble with, the definitions of normal mode and commonmode, respectively.Common mode disturbances are defined as unwantedpotential differences between any or all current-carryingconductors and the grounding conductor or earth. In

ground.Two different types of common mode distur-bances can affect a sensitive load. The first type is adisturbance on the input power conductors relativeto the input power grounding conductor. Points 1and 2 of Figure 2 show examples of the origins forthese disturbances. This type of disturbance can belimited somewhat by a line conditioner, but it isalso influenced by the location of the line condi-tioner and the wiring practices.The second type is a ground potential differencebetween elements of the computer or remoteperipherals connected to the computer. Point 3 ofFigure 2 is an example of this type of disturbance,which is more difficult to limit because it is influ-enced by factors such as system configuration andthe impedance of the grounding system. These twofactors are generally beyond the direct control of theuser except in the construction of a new facility.Because of the broad frequency band involved,wiring resonances can make equalizing groundpotentials difficult. Proper computer system ground-ing, including a signal reference grid, has beenfound to be effective against most common modedisturbances [9]. However, when remote elementsare connected to the computer systems by datacables, large ground potential differences are possi-ble. Proper surge protection of the power supplyand proper grounding of data cables will help elimi-nate hardware damage but might not prevent datacorruption. When dealing with the situation ofexample 3 in Figure 2, fiber optic links are veryeffective because they provide complete metallicse~aratio n f the various elements in the svstem - a

separation that might not be sufficiently achieved by thediscrete opto-isolation devices sometimes proposed for thatfunction [lo].Normal mode disturbances are defined as unwantedpotential differences between any two current-carrying cir-cuit conductors. Figure 3 shows three examples of theorigins of such disturbances. Usually a sine wave of nomi-nal voltage is desired for a computer power supply; anydeviation from this sine wave is a normal mode distur-bance. Computer users and monitoring instrumentsdesigners characterize these disturbances by a variety ofterms not always clearly defined such as sags, surges("swells"), outages, impulses, ringing transients, waveformdistortion, and high-frequency noise. Unfortunately,there is no consensus at the present time on the exactmeaning of these terms and their underlying quantitativedefinitions such as amplitude, duration, and thresholds.Types Of MonitorsThe instruments used in the various surveys reflecttechnological progress as well as logistics constraintsresulting in a diversity of approaches. Nevertheless, allmonitoring instruments used in past surveys were voltme-ters (with one exception, combining voltage and currentmeasurements) from which disturbance parameters werederived. Some of the monitors recorded a single param-eter such as the actual voltage peak or the fact that thevoltage exceeded a preset threshold. Other monitors com-bined time with voltage measurements describing voltagewaveforms. The recording functions of instruments usedin the surveys may be classified in broad categories:

FEBRUARY 1990 + POWERTECHNICSMAGAZINE


5/6

Threshold counters - The surge i sa p p l i e d t o a c a l i b r a t e d v o l t a g ed i v i d e r , t r i g g e r i n g a co u n t e r e achtime a preset threshold is exceeded.The ear ly types were ana log ; more ments are useful (and inexpens ive)recent types are digital. indicators of freq uen t disturbanc es;Digital peak recorders - T h e surge is othe r, more sophisticated (and moreconver ted to a digi tal value which is expens ive) ins t ruments can prov ide

Figure 4.Effec t upo n voltage peak and waveformswith varistor placed upstream of monitor

recorded in a buffer memory for laterplayback or printed out immediatelyafter i t occurs. In the early types of. r e c o r d e r s , o n l y t h e p e a k w a srecorded; in later types , the durat ionof the surge was also recorded, open-ing the way to the m ore complex dig-i t a l w a v e f o r m r e c o r d e r s n o wavailable.Oscilloscope with camera - T he surgetriggers a single sweep on the oscillo-scope's CRT, which is recorded as i toccurs by a shutterless camera withautomat ic f i lm advance. The osci l lo-s c o p e s a v a i l a b l e a t t h a t t i m e ( t h eearly 1960s) did not allow differentialmeasurements .Screen storage oscilloscope - T h esurge is displayed and stored on thecath ode ray tube. T h e writing-speedcapability of these oscilloscopes was al imitat ion in the late 1960s .Digital storage oscilloscope - T h esurge is digitized and stored in a shiftregister for subsequent playback anddisplay whenev er a preset threshold isexceeded. A n imp ortant feature ist h e cap ab i l i t y o f d i s p l ay i n g ev en t spr ior to the beginning of the surge.Digital waveform recorder - T h esurge is digitized and stored in a man-n e r s i m i l a r t o t h e d i g i t a l s t o r ag eoscilloscope, but additional data pro-cessing functions are incorporated inthe ins t rument , a l lowing r epor t s o fmany different parameters of the dis-turbance relat ing vol tage to t ime.Although some surveys might aima t g r ea t a ccu r acy , t h e r ea l w o r l dexperiences such a variety of distur-bances that any at tempt to descr ibethem in fine detail only restricts gen-eral usefulness of the data. Seekings u c h f i n e a n d d e f i n i t i v e d e t a i l i sanother fal lacy. Some s imple ins t ru-

q u i t e co m p r eh en s i v e d a t ao n d i s tu r b an ces ( b u t o n lyon past events from whichfuture dis turbances can beextrapolated only by assum-i n g t h a t t h e c a u s e s w i l lremain unchanged) . Thusthere is a practical limit tothe amount of detai l that asurvey can yield, and unre-alistic expectations of veryprecise information shouldbe avoided.Previous And FutureSurge RecordingsB e f o r e a t t e m p t i n g y e t a n o t h e rbroad survey of power quality, would-b e s u r vey o rs n eed t o co n s i d e r n o ton ly improvements in ins t rumenta-t i o n b u t a l s o c h a n g e s t h a t h a v eoccurred in modem power systems -in particular the proliferation of surgeprotect ive devices . These two dif -ferences between earlier surveys andthe more recent surveys should bekept in mind when comparing resultsand when planning future surveys.Prior to the proliferation of surgeprotective devices in low-voltage sys-tems (those defined by IEEE and IECas 1,000V or less), a limitation hadalready been recognized [4] for peakvoltages: the flashover of clearances,tvvicallv between 2 and 8 kV for low-vgltage ' wiring devices. Fort h a t r e as o n t h e e x p e c te dmaximum value cited in theIEEE Guide on Surge Voltages[3] reflects this possible trun-c a t i o n o f t h e d i s t r i b u t io naround 6 kV. Unfortunately,some readers of this Guideinterpreted the upper practi-cal limit of 6 kV as the basisfor a withstand requirement,a n d t h e y h a v e i n c l u d e d a6 - k V t e s t r e q u i r e m e n t i ntheir performance specifica-tions. A new version of thisGuide, currently under repa-ration as a Recommendec Prac-rice, will attempt to avoid thismisinterpretation.The number of surge-pro-tective devices such as varis-tors used in the United Statess i n ce t h e i r i n t r o d u c t i o n i n1972 on low-voltage ac powercircuits may he estimated at

500 million. Th erefore, a new lim-i t a t ion ex i s t s in the vo l t age surgesthat will be recorded. A surge-rec-ording instrument installed at a ran-dom loca t ion might be c lose to av a ri st o r c o n n e c t e d n e a r t h e p o i n tbeing moni tored. Such a proximityof surge protective devices and rec-ording instruments may impact pres-e n t a n d f u t u r e m e a s u r e m e n tcampaigns. Four are outlined below.1) L o ca t i o n s w h e r e v o l t ag e s u r g e sw e r e p r e v i o u s l y i d e n t i f i e d -assuming n o change in th e sourceof the surges - are now likely toexper ience lower vol tage surges ,while current surges will occur int h e n e w ly i n s t a l le d p r o t e c t i v edevices.2) Not on ly w i l l t he peaks o f theobserved voltages be changed, buta l s o t h e i r w a v e f o r m s w i l l b eaffected by the presence of nearbyvaristors as illustrated i n Figures 4,5, an d 6.

A ) If a varistor is located betweenthe source of the surge and ther eco r d i n g i n s t r u m en t ( F i g u r e4), the instrument will recordt h e c l a m p in g v o l t a g e of t h ev a r i s t o r . T h i s v o l t a g e w i llhave lower peaks but longert i m e t o h a lf -p e ak t h a n t h eoriginal surge.B) If t h e i n s t r u m e n t i s lo ca t edbetween t he source of th e surgeand a varistor, or if a parallelbranch circuit contains a varis-tor (Figure 5) , the ins t rument

Figure 5.Effect upon volta e peak and waveformswith varistor pk d downstream ofrecorder

POWERTECHNICS MAGAZINE + FEBRUARY 1990 25

-- - - -


6/6

P o w e r M o n i t o r i n gwill now record the clamping voltage of the varistor,preceded by a spike corresponding to the inductivedrop in the line feeding the surge current to thevaristor.

C ) If a varistor is connected between the line and neu-tral conductors, and the surge is impinging betweenline and neutral at the service entrance (normalmode), a new situation is created, as shown in Fig-ure 6. The line-to-neutral voltage is clamped asintended, but the inductive drop in the neutral con-ductor returning the surge current to the serviceentrance produces a surge voltage between the neu-tral and the grounding conductors at the point ofconnection of the varistor and any downstreampoint supplied by the same neutral. Because thissurge has a short duration, it will be enhanced bythe open-end transmission line effect between theneutral and grounding conductors [ll].

3) Th e surge voltage limitation function previously per-formed by flashover of clearances is now more likely tobe assumed by the new surge protective devices that areconstantly being added to the systems.4) These three situations will produce a significant reduc-tion in the mean of voltage-surge recordings from thetotal population of different locations as more and more

varistors are installed. However, the upper limit willremain the same for locations where no varistors havebeen installed. Focusing on the mean of voltage surgesrecorded in power systems can create a false sense ofsecurity and an incorrect description of the environ-ment. Furthermore, the need for adequate surge cur-rent handling capability of a proposed suppresser withlower clamping voltage might be underestimatedbecause some diversion is alrehdy being performed.Part 11 of this article, a detailed review of several site sur-veys, will appear in the March issue of Powertechnics. Thisarticle was based on "Power Quality Site Surveys: Facts,Fiction, and Fallacies," by F.D. Martzloff and T. M. Gruzs,which first ameared in lEEE Transactions on IndustrvApplications, \ id. IA-24, No. 6, pp. 1005-1018, ~ o v . / ~ e i1988. 1988 IEEE.

References[I] lEEE Standard Dictionary of Electrical and ElectronicsTerms, ANSIIIEEE Std. 100, 1988.[2] T.M. Gruzs, "Power Disturbances and Computer Sys-tems: A Comparison of the Allen-Segall and theGoldstein-Speranza Power Line Monitoring Stud-ies," in Proc. 1986 Electrical Overstress Exposition,Nelson Publishing Co., Nokomis, FL, Jan. 1986.[3] Guide on Surge Voltages in Low-voltage A C Power C ir-cuits, ANSIIIEEE C62.41, 1980.[4] lnsulation Coordination within Low-Voltage Systems,lncluding Clearances and Creepage Distances for Equip-ment. EIC 664. 1980.

NN-G

Figure 6.Protective device installed between line and neutral conductors, forsurge applied in normal mode at service entrance and varistorconnected at end of 75-m branch circuit ( a ) ; surge applied at"send" end, 3 kV initial peak, 100 kHz oscillation (b);(c) howssurge clamped at L-N at end of line, maximum voltage 400 V(same time scale as b, but different amplitude sc ak ); 2500 spike atN-G at end of line ( d ) , created by surge current in neutralconductor (sa me scales as b ) ( d ) will experience short durationsurge

[5] ~ l e c t r i m a ~ n e t i cmission and SusceptibilityRequirements for the Con tro l of Electromag-netic Interference, DOD-STD-461B.[6] interface Standard for Shipboard Systems,DOD-STD-1399, Sec. 300, 1978.[7 ] lE EE Guide on Surge Te sting for EquipmentConnected to Low-Voltage A C Power Cir-cuits, ANSIIIEEE C62.45, 1987.[8] IEC Multilingual Dictionary of Electricity, pub-lished by IEEE and distributed in coopera-tion with Wiley-Imerscience.[9] Guideline on Electrical Power for A D P Installa-tions, FlPS Pub 94, National Bureau ofStandards, 1983.[lo] F.D. Martzloff, "The Protection of Com-puter and Electronic Systems AgainstPower Supply and Data Line Distur-bances," General Electric Co., Schenec-tady, NY, Report 85CRD084, 1985.[l l] F.D. Martzloff and H.A. Gauper, "Surgeand High-Frequency Propagation inIndustrial Power Lines," IEEE Trans. Ind.Appl., vol. 1A-22, no. 4. , JulyIAug.1986.

Please rate this article ..UsefulSomewhat UsefulNot Useful

26 FEBRUARY1990 + POWERTECHNICSMAGAZINE

-- -- - A-

monitoring power quality

Documents