72
Made to measure. Practical guide to electrical measurements in low voltage switchboards (a) (b) V t V t (f) 50 Hz A 0 50 100 150 200 250 500 A 0 50 100 150 200 250 500 V 0 80 60 40 20 Made to measure. Practical guide to electrical measurements in low voltage switchboards

Made to measure. Practical guide to electrical

  • Upload
    others

  • View
    3

  • Download
    0

Embed Size (px)

Citation preview

Mad

e to measure. P

ractical guide to electrical m

easurements in low

voltage switchb

oards

(a)

(b)V

t

V

t

(f)50 Hz

A

050

100

150

200250 500

A

050

100

150

200250 500

V

0

8060

40

20

Made to measure.Practical guide to electrical measurements in low voltage switchboards

2CS

C44

5012

D02

01 -

12/

2010

- 1

.500

- C

AL.

Contact us

ABB SACEUna divisione di ABB S.p.A.Apparecchi ModulariViale dell’Industria, 1820010 Vittuone (MI)Tel.: 02 9034 1Fax: 02 9034 7609

bol.it.abb.comwww.abb.com

The data and illustrations are not binding. We reserve the right to modify the contents of this document on the basis of technical development of the products,without prior notice.

Copyright 2010 ABB. All rights reserved.

Made to measure. Practical guide toelectrical measurements in low voltage switchboards

table of

contents

1 Electrical measurements

1.1 Whyisitimportanttomeasure?.......................................... 31.2 Applicationalcontexts.......................................................... 41.3 Problemsconnectedwithenergynetworks......................... 41.4 Reducingconsumption........................................................ 71.5 Tableofcharges.................................................................. 81.6 Absorptionpeaks................................................................ 81.7 Dividingconsumption.......................................................... 91.8 PowerFactorCorrectionandMaintenance.......................... 91.9 Remoteandhistoricalreadingofinformation....................... 9

2 Technical reference standards

2.1 IECstandards.................................................................... 102.2 MIDDirective..................................................................... 11

3 Measuring instruments

3.1 Analogueinstruments........................................................ 123.2 Digitalinstruments............................................................. 143.3 Measurementerrorsandaccuracyratings......................... 153.4 Comparisonbetweenthetwocategoriesofinstrument: advantagesandlimits........................................................ 18

4 Direct and indirect measurements: CTs, VTs, converters and accessories

4.1 Directmeasurements......................................................... 204.2 Indirectmeasurements...................................................... 204.3 Shuntsfordirectcurrent.................................................... 234.4 Convertersandaccessories.............................................. 23

5 Overview of ABB range

5.1 Analogueinstruments........................................................ 24 5.1.1 DINrailanalogueinstruments............................................ 24 5.1.2 Analogueinstrumentsonfrontofpanel.............................. 25 5.1.3 Advantages....................................................................... 275.2 Digitalinstruments............................................................. 28 5.2.1 DINraildigitalinstruments................................................. 29 5.2.2 Analogueinstrumentsonfrontofpanel.............................. 29 5.2.3DMTMEmultimeters.......................................................... 30 5.2.4MTMEandANRnetworkanalysers................................... 31 5.2.5Temperaturemeasuringunits............................................ 34 5.2.6Electronicenergymeters................................................... 355.3 Accessoriesformeasuringinstruments.............................. 36 5.3.1Serialcommunicationadapters.......................................... 36

5.3.2 Currenttransformers......................................................... 37 5.3.3Voltagetransformers......................................................... 38 5.3.4Shuntsfordirectcurrent.................................................... 38

6 The measurements

6.1 TRMSMeasurements........................................................ 40 6.1.1Linearloads....................................................................... 40 6.1.2Nonlinearloads................................................................. 40 6.1.3ProblemsconnectedwithTRMSmeasurements............... 416.2 HarmonicdistorsionandTHD............................................ 426.3 Cosfì(cosϕ)andpowerfactor(PF)...................................... 446.4 Practicaltipsforinstallingagood measuringsystem............................................................. 44

7 Digital communication

7.1 Communicationprotocols.................................................. 49 7.1.1 Thephysicallayer.............................................................. 49 7.1.2Theconnectionlayer......................................................... 52 7.1.3Theapplicationallayer....................................................... 52 7.1.4Compatibilitybetweenlevels.............................................. 537.2 Supervisionofelectricaldistributionplants......................... 537.3 TheModbusRS-485network............................................ 55 7.3.1Rulesforcorrectcabling.................................................... 55 7.3.2TheoperationoftheModbussystem................................. 57

8 Applicational examples of network analysers

8.1 Nominalvoltage(phase/neutral)andlinkedvoltage (phase/phase)astrueeffectiveTRMSvalue....................... 628.2 CurrentastrueeffectiveTRMSvalue onthethreephasesandonneutral................................... 628.3 Powerfactor(cosϕ)............................................................ 628.4 Activepower..................................................................... 638.5 Harmonicdistortionrate(HDR) upto31thharmonicdisplayedgraphically andasapercentagevalue................................................. 638.6 Harmonicdistortionupto31thharmonicdisplayed graphicallyandasapercentagevalue............................... 638.7 Activeenergyconsumedandgeneratedwithsubdivision ofthecountintototalcountersandaccordingtosettable timesofday....................................................................... 63

9 Annex

9.1 MeasurementsGlossary.................................................... 64

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 1

2 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

1 1Electric measurements

Measurement:ratiobetweenonequantityandanotherthatishomogenoustoit,whichisconventionallyaunit.Nevertheless,intheelectricalfielditisnotalwayseasytofindsamplestocompare,especiallyformeasurementsconductedoutsideequippedlaboratories.Inpractice,calibratedinstrumentsarethereforeusedthatdonotcomparetheelectri-cal quantity under examination with an electrical sample but with another type ofquantity(forexample,inanalogueinstruments,theforceexertedbyaspring).Thegeneraldefinitionofthemeasuringconceptmakesitimportanttodefinetheunitsofmeasurement,whichmustbeinvariableandingeneralreproducible.ThecorrectunitsofmeasurementsthathavetobeusedarethoseindicatedbytheSI(SystèmeInternational);Table1.1showsthefundamental(orbasic)unitsofmeas-urementoftheSI.

QuantityUnit

Standards Symbol

Length metre m

Mass kilogram kg

Time second s

Current ampere A

Thermodynamic kelvin K

Quantityofsubstance mole mol

Lightintensity candle cdTable 1.1 Basic SI unit

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 3

1

ElE

CT

riC

ME

As

ur

EM

En

Ts

Table1.2showstheelectricandmagneticquantitiesthataremostoftenencounteredandwhichneedtobemeasured.

QuantitySI Unit Dimensional

expressionname symbol

-Electriccurrent ampere I A

-Quantityofelectricity(load) coulomb C s·A

–Electricpotential•potentialdiff.•electromotiveforce•voltage

volt V m2·kg·s-3·A-1

-Electriccapacity farad F m-2·kg-1·s4·A2

-Permittivity faradpermetre F/m m-3·kg-1·s4·A2

-Resistance•impedance

ohm Ω m2·kg·s-3·A-2

-Resistivity ohmpermetre Ω·m m3·kg·s-3·A-2

-Conductance siemens S m-2·kg-1·s3·A2

-Conductivity siemenspermetre S/m m-3·kg-1·s3·A2

-Inductance henry H m2·kg·s-2·A-2

-Electricfield voltpermetre V/m m·kg·s-3·A-1

-Loaddensity coulombpermetre2 C/m2 m-2·s·A

-Currentdensity amperepermetre2 A/m2 m-2·A

-Frequency hertz Hz s-1

-Inductionflow weber Wb m2·kg·s-2·A-1

-Magneticinduction tesla T kg·s-2·A-1

-Magneticfield amperepermetre A/m m-1·A

-Magneticpotential weberpermetre Wb/m m·kg·s-2·A-1

-Dielectricconstant faradpermetre ε m-1·kg-1·s4·A

-Magneticpermeability henrypermetre μ m·kg·s-2·A-2

-Power watt W m2·kg·s-3

-Energy wattpersecond J m2·kg·s-2

1.1

Why is it important to measure?Asarticle2ofEuropeanDirective374of25July1985statesthat'alsoelectricity'isa'product',equatingitwithanyothertypeof'movablegoods',thefirst,immediateansweris:tobeabletomarkettheproductelctricity.Usingamoresophisticatedargument,evenifitislimitedtothemanagementaspectsofapowerplant (thus ignoringall thetechnicalandscientificquestions), there isaclearneed in today'smarket tocontainandreducecostsandensurecontinuityofservice.Ithasthusbecomevitaltothoroughlyfamiliariseoneselfwiththeoperationofanelectricplantinordertobeabletooptimise:consumption,loadcurves,harmonicinterference,voltagedisturbances,etc,i.e.alltheelementsthatcontributetoincreaseefficiency,improvingcompetetitivity,and,animportantfactortoday,reducingharmfulemissionsintheenvironment.Lastly,stillfromthemanagementpointofview,measuringandmonitoringelectricalquantitiesenablefaultpreventiontobeoptimisedandmaintenancetobescheduledowingtoearlyidentificationofproblemsthatresultsingreaterprotectionnotonlyoftheplantsbutalsooftheobjectsconnectedtothem.

Table 1.2 Main electric and magnetic quantities

continues1

4 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

1

ElE

CT

riC

ME

As

ur

EM

En

Ts

1.2

Applicational contextsAnefficientsystemofmeasuringandmonitoringelectricalquantitiesisimportantforensuringthesuccessofallinitiativesthatrequire:-energycoststobecontained;-qualityenergysupplies;-continuityofserviceoftheplants.Specifically,achievingtheaboveobjectivesrequirestheactivitiestobeimplementedthataresetoutintheflowchartinFig.1.1.

Functions:-submeteringof

consumptionanddividingcosts

-monitoringloadpatterns

-managingpeaks-improvingpower-

factorcorrection

Functions:-analysingharmonics-detectingovervoltages,

voltagevariations,andvoltagegaps

-detectingdischargesfromsteeptransients

-conformityofsupplytoEN50160

Functions:-controlofplantin

realtime-remotecontrol-managingalarms

anddividingcosts-preventive

maintenanceandmaintenanceintheeventofafault

Functions/objectives of electric measurements

Reducing energy costs

Energy qualityContinuity of service

ABBmeasuringinstrumentsareanalogueanddigitalnetworkanalysersandenergymetersthatoptimisetheabovefunctionsinthemostvariedapplicationalcontexts:-residentialandcommercialbuildings-industries-shoppingcentres-garages-collegesandcampuses-tradefairsites,exhibitionvenues-touristports-hotelsandcampsitesAlltheABBinstruments,bothoftheDINrailandpanelfronttype,aredistinguishedbythesuperiorityandexcellenceoftheirproperties,andlastbutnotleast,theyen-ablelowvoltagepowerpanelsandcabinetscabledintothepowercentrestobeen-hancedandtheiraestheticappealtobeapproved.

1.3

Problems connected with energy networksInordertodefinethefeaturesofthepowersupplyatthedeliverypoints,adistinctionmustbemadebetweennormalandemergencyoperationofanelectricalsystem.Anelectricalsystemisrunningnormallywhenitisabletomeettheloadsupply,elimi-natefaultsandresumeoperationwithordinarymeansandproceduresiftherearenoexceptionalconditionsduetoexternalinfluencesorsignificantcriticalsituations.Emergencyoperationoccurswhen,owingtoinsufficientgenerationcapacityorbe-causeofsituationshavingagraveimpactonthesystem,orowingtoeventsthatarebeyondthecontrolofthepowerprovider(deliberatedestruction,disasters,strikes,acts of the public authorities, etc), it becomes necessary to interrupt or limit the

Fig. 1.1: Functions and objectives of electric measurements

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 5

1

ElE

CT

riC

ME

As

ur

EM

En

Ts

service.Thethree-phasevoltagesuppliedtothedeliverypointsbythepublicdistributionsys-temduringnormaloperatingconditionshasthefollowingfeatures(seealsoTable1.3):-frequency-amplitude-waveform-symmetryofthethree-phasevoltagesystem.Thedistributingbodycanalso imposelowvoltagesignalsonthevoltagethathavetheaimoftransmittinginformationonoperation.Thesefeaturesaresubjecttovariationduringnormaloperationoftheelectricalsys-temduetoloadfluctuations,disturbancesgeneratedbycertaintypesofuserequip-mentorplantsandtheoccurrenceoffaults,whicharemostlyduetoexternaleventsthatmaycausetemporaryinterruptionstothesupply.Asaresult, thesecharacteristicsarechangeableovertime, if referredtoaspecificdeliverypoint;theyarechangeableinspaceifatagivenmomentallthedeliverypointsinadistributionnetworkareconsidered.Consequently,inbothcases,theymustbedescribedinstatisticalterms;Fig.1.2showsthedifferenttypesofvariationofvoltageamplitudeduetotransistorandpulsephenomena.Table1.4liststhemaindisturbingdevices,i.e.machinesandequipmentoftheuserthatmayintroduceelectromagneticdisturbances.Theyarelistedbytypeofapplica-tiontoshowhowthesametypeofdevicemaysimultaneouslygenerateseveraldis-turbances.Forexample,a resistanceweldingunitmaygenerate:dissymmetryandunbalance,voltagefluctuation,voltagevariations,respectivelyindicatedinthecolumnsontherightofTable1.4bythelettersSQ,VF,VV.

CharacteristicPhenomenon

Type Description

Frequency Variation %deviationfromnominalvalue

Amplitude Slowvariations %deviationfromnominalvaluewithdurationofvariation>10s

Rapidvariations %deviationfromnominalvaluewithdurationofvariation<10s

Overvoltages Risesinvoltagemeasuredasinstantaneousabsolutevalueoraspercentageofnominalvalue

Holes Partialdecreasesbelow90%ofratedvoltageanddurationcomprisedbetween10msand60s

Shortinterruptions Novoltagefor≤180s

Longinterruptions Novoltagefor>180s

Waveform Harmonics Theyaresinusoidalvoltagesorcurrentswithafrequencyequaltoacompletemultipleofthebasicfrequency,thepresenceofwhichcausesadistortiontothewaveformofthesupplyvoltage

Interharmonics Theyarevoltagesorcurrentsthatmaybesinglesinusoidalcomponentswithafrequencythatisnotanentiremultipleofthebasicfrequencyormaybeanextendedspectrumofsinusoidalcomponents

Symmetryofthethree-phasesystem

Dissymmetry Inconsistentamplitudeand/oranglebetweenthemeasuredphasesasadegreeofdissymmetry Table 1.3 Voltage

characteristics

continues 1.3

6 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

1

ElE

CT

riC

ME

As

ur

EM

En

Ts

Fig. 1.2: Schematisation oftypes of variation of

voltage amplitude

Key:SQ = dissymmetries and unbalanceVF= voltage fluctuationsVV = voltage variationsAR = harmonicsSF = spurious frequenciesRE = radioemission(1) if single-phase(2) at insertion, when power is not low

compared with the power of the short-circuit power of the network

(3) if remote control

Key:a) Voltage gaps:

durationfrom10msto60s,ifthevoltageiscompletelycancelledtheinterruptionsareknownasshortinterruptions

b) Non pulsed overvoltages:theoppositeofvoltagegaps

c) Slow variations:amplitudevariationsinrelationtothenominalvaluewithduration>10s

d) Pulsed voltages of long duration: durationcomprisedbetween0.1msandsomemsoriginatingfromfaultsormanoeuvres

e) Pulsed voltages of medium duration: durationcomprisedbetween1and100μsofatmosphericoriginorfromactionofswitchesorcircuitbreakersorfromtrippedfuses

f) Pulsed voltages of short duration: duration<1μsoriginatingfromactionsofswitchesorcircuitbreakersinspecialcases

g) Communication transients: originatingfromconverterdevicesandrectifiers

Table 1.4 Disturbing devices

Devices PowerDisturbances generated

SQ VF VV AR SF RE

Resistanceheating 1-40kW (1) (2) (3)

Domesticovens-microwaveovens-infraredovens 1-2kW

(1)(1)

••

• •

Industrialkilns-induction-HF-UHF-plasma-arc

10-2,000kW10-600kW10-100kWsomeMVA1-100MVA • •

••••

•••

•••

•••

Weldingmachines-resistance-arc

0.1-2MW1-300kW

• ••

•• (3)

Motors-asynchronous(e.g.compressors)-variablespeed

<10MVA

-20MVA

• •

Transformers <100MVA • •

Converters-ac/dc-ac/acandcycleconverters

<10MW<30MW

••

•• •

Electroerosion 10-30kW •

Dischargelamps •

Televisions • •

Radiology • •

continues 1.3

(a)

(b)V

t

V

t

VM

V

t

(c)

(d)

V

t

(e)

50 Hz

V

t

(g)

V

t

(f)50 Hz

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 7

1

ElE

CT

riC

ME

As

ur

EM

En

Ts

Fig. 1.3: How to reduce consumption

Thesamedevicemayalsogenerateseveraldifferenttypesofdisturbanceatthesametime.Emissionlevelsforthevariousdisturbancesarecalculatedinthefollowingmanner:-theemissionlevelofthesingledevicesiscalculated;-thetotalemission leveloftheuser iscalculatedasacompositionoftheemission

levelsofthesingledevices;-thetotalemissionleveloftheuseriscomparedwiththepermittedemissionlevel;

thisemissionlevelisgenerallydefinedbythedistributoronthebasisofcriteriathatensurecontrolofthecompatibilitylevels.

Theemissionlevelsaregenerallymeasuredatthe'commoncouplingpoints'thataredeemedtobeofparticularinterest:pointofcommoncouplingwiththenationalgrid(PCC)andinternalpointsofcommoncouplingwiththeuserdistributionnetwork(PIC).Thedisturbancesthatoccurmostfrequentlyandthathavetobeassessedandcon-tainedare:-harmonics;-rapidvoltagevariations;-flickers.Thelatterarevoltagefluctuationsthathaveamodulationfrequencybetween0.5and35Hzandgiverisetothephenomenonofflicker,i.e.thesensationoffluctuationsintheintensityofthelightofthelamps.

1.4

Reducing consumptionTherisingcostofelectricenergyisasignificantproblemandisoneoftheparametersthatisincreasinglycomingunderscrutinyinordertocontainthegeneralcostsofabusiness.Statisticsdrawnupbothnationallyandinternationallyshowthateverysinglecompanycanreduceitsenergybillfrom10%to30%.Thissavingspercentagevariesaccordingtotheevaluationoftheconsumptionmadeduringthepowerplantdesignphaseandevenmoreforolderplantsintermsoftheconsumptionanalysisandtherelativesolutionsadoptedtomanageconsumption.ThestepsrequiredtoachieveagoodresultaresummarisedinFig.1.3.

Reducing consumption

Contract analysisConsumption

analysisTechnical

interventions

continues 1.3

8 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

1

ElE

CT

riC

ME

As

ur

EM

En

Ts

1.5

Table of chargesTheanalysisofthesupplycontractforelectricpowerrevealsaseriesofusefulpiecesofinformation:-thepowerused,i.e.themaximumvalueofavailablepowerthat is limitedormust

notbeexceededinordernottoincurpenalties;-thetableofchargesthatcanbefixedormayvaryaccordingtothetimeofday;-thepeakorexcessivepowerthatexceedsthecontractuallyagreedsupply.

Thecommittedpower is themaximumusablevaluethat, forcontracts for levelsofpowerthatarenotparticularlyhigh(generallyupto35kW)ismanagedbyacurrentlimiterthat interruptstheenergysupplywhenconsumptionexceedsthecommittedvalue.Thepowercommitment issetduringthedesignstageonthebasisoftheeffectivepowerrequiredforsimultaneouslyrunningloadsduringpeakconsumptonperiods.EachkWcommittedhasa fixedcostand it is thereforeadvisable toassessactualneedstoavoidpayingforunnecessarypower.Thecontractmustbesignedafteranassessmentofthemostappropriateusernet-work architecture, taking into account the following of the most importantparameters:-numberofconnectionpoints;-lowormedium-voltagedeliveryorseverallowvoltagedeliverypoints;-creatinganemergencyplant;-makingconsumptionforecastsonthebasisoftheactualconsumptionandnoton

thesumofthenominalpoweroftheloads(todefineavailablepower).During the course of the supply contract the user should periodically examine theconsumptionstated in theelectricitybillsandmakeanalyses/recordswithsuitableinstruments;hencetheimportanceofmeasuringandmonitoringenergyconsumptionovertime.

1.6

Absorption peaksForpowerabove37.5kWthepowerproviderusesenergymetersthatmeasureab-sorptionovertime,recordingaverageconsumptionevery15minutes(Fig.1.4).

0 15 mi

Area proportionalto average value

Instantaneous power Integrated value

tip-measuring device

n

100 kW

0 15 min

200 kW

If,forexample,thecontractisfor100kW,thepeakvalueisdeemedtobewithinthecontractualvalueiftheaveragevalueofmaximumconsumptionis100kWin15min-utes,whichmaybetheequivalentofaverageconsumptionof200kWin7.5minuteswith0kWconsumptioninthefollowing7.5minutes.Inordertoavoidpenalties,itisimportanttocontrolandmanageabsorptionpeakssoastoneverexceedtheavailablepoweraverage.Acorrectanalysisofconsumptionenablesthesuitabilityofthetypeofcontracttobecheckedagainsttheuser'suseparameters,thusslashingcompanycostsandavoid-ingahighpriceadjustmentattheendoftheyear;forexamplerecordingenergycon-sumptionatdifferenttimesofdayenablesallthepowerconsumptionofthedayormonthtobemonitored,thusprovidingacompletepictureoftheenergysituation.

Fig. 1.4: Graphic representation of average consumption

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 9

1

ElE

CT

riC

ME

As

ur

EM

En

Ts

1.7

Dividing consumptionIf it isfundamentaltoknowconsumptiontooptimiseenergysaving,it isjustasim-portanttousetheenergyavailableunderthecontractrationallyinordertoavoidin-terruptionstothesupplyandpenaltycharges.Inresidentialortertiary-sectorenvironments,wheretheavailablepowerislimitedandloadneedschangecontinuouslyovertheday,itisimportanttoknowmomentarycon-sumptionand tobeable todisconnect less important installations if themaximumpowerlevelisreached.Forexample,ifinadomesticenvironmentseveralmachinesareworkingsuchas:awashingmachine,dishwasher,vacuumcleaner,etcthatexceedthecontractualpower,thelimiterintheelectricitymeteroftheproviderinterruptsthesupplyanddisconnectspower to theentire system. In simplecases suchas thisonea loadmanagementswitchmaysuffice(forexampletheLSS1/2switch),whilstinmorecomplexenviron-mentssuchasindustryandthetertiarysector,itispossibletouseABBenergymetersoftheODINsingleandDELTAsinglesingle-phasemeters,andODINandDELTAplusthree-phasemeters(seechapter5below)tomonitorcontinuouslyconsumptionandmakecontingencyplans forcases inwhich themaximumsetvalue is reached (forexamplebydisconnectingonlytheinstallationsthatareconsideredtobelessimpor-tant,maintainingthesupplytothemaininstallations).

1.8

Power Factor Correction and MaintenanceThepowerfactororcosϕ (whichisthephaseanglebetweenthephasorsofthevoltageofthecurrent),mustbemaintainedatavaluethatisasnearaspossibleto1,toavoidunnecessary inductive currents that overload the line of the supplier company. As isknown,theuserdevices,whichmostlyhaveinductiveloads(forexample):motorsandtransformers),needmagnetisingcurrenttooperatethatdoesnotproduceworkbutloadsthelines,reducingtheircapacity.Forthisreason,thesuppliersofelectricpowerapplyapenaltywhenthepowerfactorcosϕislessthan0.9.Itisthusimportanttomeasurethepowerfactorandifitdoesnotfallwithinthecontrac-tuallimits,power-factorcorrectingcapacitorsmustbefittedtotheout-of-phaselines.Measuringandrecordingconsumptionthusbecomesanimportantindicatorforschedul-ingmaintenance, inparticular in industrialenvironments,because identifying themostused linesanddevicesenablesthe interventionstobecontrolledandestablished inascheduledpreventivemaintenanceprogramme.

1.9

Remote and historical reading of informationInordertoconductathoroughanalysisofelectricparametersandofevents,itisim-portant for themeasuring instruments tohaveasystem forstoringmeasureddataandtobeabletotransferremotelysuchdatainordertocompareandanalysethem.Remotereadingandstoringinformationareusedespeciallyinplantswithagreatex-tentandinthepresenceofheavyloads,suchas,forexample,inthelargechainstoresandinindustry.

10 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

2Technical reference standards

Inanytechnicalenvironmentandinparticularintheelectricalsector,inordertomake'workmanlike'devices,allrelevantlegalandtechnicalstandardsmustbecompliedwith.Knowledge of standards and distinguishing between legal standards and technicalstandardsisthusthebasisforacorrectapproachtomeasuring-instrumentquestionsthatinvolvesnotonlytechnicalaspectsrelatingtoaccuracyandsafetybutalsotaxandaccountingmatters.Legalstandardsareallthosegoverningthebehaviourofpartiessubjecttotheauthorityofthestate,includingtheEuropeandirectivesthatarenormallyenactedinnationalleg-islationthroughlegislativedecrees.Technicalstandardsarealltheprescriptionsonthebasisofwhichthemachines,de-vices,materialsandplantshavetobedesigned,builtandtestedtoensuretheiroper-atingefficiencyandsafety.Thetechnicalstandardssetbynationalandinternationalbodies(CEI,CENELEC,IEC)havebeendrawnupinaverydetailedmannerandcanhavelegalsignificancewhenthisisassignedtothembyalegislativemeasure.

2.1

IEC standardsThreecommitteesarespecificallyresponsibleformeasuringinstrumentation.-TC85"Measuringequipmentforelectricalandelectromagneticquantities"-TC66"Safetyofmeasuring,controlandlaboratoryinstruments"-TC13"Electricalenergymeasurement,tariffandloadcontrol".Thefirstcommitteedrawsupandpublishesthereferencestandardsforalltheinstru-ments(voltmeters,ammeters,wattmeters,etc)ofbothanalogueanddigitaltypeandprovidestheprescriptionsfortheinstrumentsandthesampleequipment(batteries,resistors,recordinginstruments,etc).Committee85isalsoresponsibleforaseriesofstandards,allofwhichareEuropeaninorigin(fromIECEN61557-1toIECEN61557-10),dedicatedtoelectricalsafetyinlowvoltagedistributionsystems.Thesestandardsalsocontainsomesafetyprescrip-tionsandthefunctionalfeaturesrequiredfortheinstrumentsforthetests,measure-mentsandcontrolsoflowvoltageelectricplantssuchas,forexample:earthresist-ancemeasuringinstruments,instrumentsformeasuringimpedanceoffaultloop,in-struments for testing continuity of protection conductors, insulation measuringinstruments,etc.Thesearethereforeparticularlyimportantstandardsfordefiningtherequiredcharac-teristicsofthemeasuringinstrumentstobeusedforthetestsprescribedbystandardIEC60364thatgovernslowvoltageelectricplants.Committee66dealswiththesafetyprescriptionsforelectricalmeasuringdevicesthatmustbemetbythemanufacturertoensurethesafetyoftheoperator.

2

TE

CH

niC

Al r

EFE

rE

nC

E s

TAn

DA

rD

s

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 11

Lastly,Committee13isentirelydedicatedtopublishingthestandardsgoverningac-tiveenergyandreactiveenergymeasurementsandtherelativedevices:meters,inte-gratedunits,devicesofdifferenttypes.Inthisconnectionthefollowingstandardshaveparticular importance for the purposes of tests on energy meters: EN 50470-1,EN50470-2,EN50470-3,whichsetthetestprescriptionsbothforelectromechanicalactiveenergymetersandforstaticmeters.Fig.2.1summarisesthestandardstowhichmeasuringinstrumentationissubject.

Reference Standards for measuring instruments

Electricalenergymeasurement,tariffandloadcontrol

TC13

Safetyofmeasuring,controlandlaboratoryinstruments

TC66

Measuringequipmentforelectricalandelectromagneticquantities

TC85

2.2

MID DirectiveEuropeanDirective2004/22/ECof31March2004introducedaCommunityframe-worklawgoverningdevicesandsystemswithmeasuringfunctionsrelatingtocom-monconsumergoods:water,gas,fluidsingeneraland,inparticular,“active electric energy meters and measuring transformers”,whichareidentifiedinthedirectivebyMI-003.Thedirectivestates that themeasuring instrumentmustconform to 'the particular requirements which are applicable to the instruments in question';fortheactiveelec-tric energymeters, the annexdefines specific requirements in termsof: accuracy,operating conditions, maximum tolerated errors, procedures for ascertainingconformity.Thedirectiveappliestoallelectricenergymeters,whethertheybelongtotheutilitycompanyortoprivateindividuals,thatareinstalledforanyreasoninplantsformeas-uringand/ormeteringelectricenergy;itisalsospecifiedthatthemeterscanbeusedincombinationwithexternaltransformers.Thesignificanceofthedirectiveisconsiderable,notonlybecauseitproposeselimi-natingallunreliablemeasuring instruments thathavenotbeenconstructed incon-formity to theproductstandardandsometimesdonotevenhaveCEmarking,butalsobecauseitenablesinstrumentationtobeused(providedthatitconformstothedirective)alsotometerenergyfortaxpurposes.

Fig. 2.1: Reference Standards for measuring instruments

continues 2.1

12 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

3Measuring instruments

Forthelastfewdecadesbothanalogueanddigitalmeasuringinstrumentshaveex-istedalongsideoneanother.Theformeraredevicesinwhichtheinformationisassociatedwithphysicalquantitiesthatarecontinuouslyvariablewhereasindigitalinstruments(thatemergedlaterinthe1970sand1980swiththearrivalofelectronicsandIT)thequantitieshavediscretevalues(asintheword'digit').These instrumentsconsistofanA/D transducer-convertersystem for transforminganynonelectricinputquantityintoanoutletanalogueelectricquantity(generallyvolt-age)andsubsequentlyconvertingintodigitalform,andofacountersystemforpro-vidinginformationonthenumberofpulses.

3.1

Analogue instrumentsIn Fig. 3.1 a block diagram shows the initial configuration of an analogueinstrument.

Quantities to be measured

Drive torque Deflection angle Reading

Electromechanical converter

Force or torque meter

Angle meter

Theseinstrumentsexploitphenomenaforwhichtheinteractionofelectricormagneticquan-titiesgivesrisetomechanicalforceortorque.Theseinstrumentsconsistofamovablepartthathasaninitialrestposition,onwhichadrivingtorqueactsinfunctionoftheelectricormagneticquantitiesonwhichtheassociatedphenomenondepends.Thedrivingtorqueisopposedbyrestoring,normallyelastic,torque,which,dependingonthemovement,tendstoreturnthemovableparttotheinitialpositionwhentheactionpro-ducedbythedrivingtorqueceases.Fromthebalanceofthetwotorquesanangulardeviationisobtainedthatisproportionaltothequantitytobemeasured.Anindexfixedtothemovablepartrotatesalongascale.Ingeneral,themanufacturerplacesonthedialoftheinstrumentcertainconventionalsymbolsthatcharacterisenotonlytheunitofmeasurementbutalsotheoperatingprinciple,theconnectionnetwork(directoralternat-ing), theaccuracyrating,operatingposition(horizontal,vertical)andsafetyprescriptions(testvoltage).TheconventionalsymbolsthataregenerallyusedaresummarisedinTables3.1.and3.2.

Fig. 3.1: Block diagram of an electromechanical analogue instrument

3

ME

As

ur

ing

ins

Tr

uM

En

Ts

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 13

Table 3.1 Identification of the instruments and symbols shown on the dial

Table 3.2 Identification of the instruments and symbols of the operating principle

Circuits in which it can be inserted

Circuit Symbol Circuit Symbol

Directcurrent Three-phasealternatingcurrentwithacurrentcircuitandavoltagecircuit

Alternatingcurrent Three-phasealternatingcurrentwithtwocurrentcircuitsandtwovoltagecircuits

Directandalternatingcurrent Three-phasealternatingcurrentwiththreecurrentcircuitsandthreevoltagecircuits

Arrangement of the instrument

Arrangement Symbol Arrangement Symbol

Instrumenttobeusedwithverticaldial

Instrumenttobeusedwithtilteddial

Instrumenttobeusedwithhorizontaldial

Angleoftilt(optional)

Test Voltage

Voltage Symbol Voltage Symbol

Testvoltage500V Testvoltage5,000V

Testvoltage2,000V Instrumentexemptfromvoltagetest

Instrument Symbol Instrument Symbol

Fixedmagnetandmovingcoil Fixedmagnetandmovingcoilasratiomeasuringdevice

Movableiron Movableironasratiomeasuringdeviceorasdifferentialinstrument

Electrodynamic Electrodynamicasratiomeasuringdevice

Electrodynamicwithiron Electrodynamicwithironasratiomeasuringdevice

Induction Inductionasratiomeasuringdeviceorasdifferentialinstrument

Hot-wirethermaloverloadfuse Bi-metalbladethermaloverloadfuse

Electrostatic Vibratingblades

Movablethermocouplereel Movablereelwithrectifier

continues 3.1

3

ME

As

ur

ing

ins

Tr

uM

En

Ts

14 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

3.2

Digital instrumentsTheoperatingprincipleofdigitalinstrumentsisbasedonanalogue-digitalconversiontechniques;decodinganddisplaydevicesarealwaysassociatedwiththem,asareveryoftensamplefrequencyoscillatorsanddecimalcountcircuits.TheblockdiagramisshowninFig.3.2.

Decoding and display

AttenuatorConverter

ConverterA/D

Controller

Thedigitalinstrumentsareessentiallyvoltmetersfordirectcurrents;nevertheless,us-ingusualconversionsystemsfromVACtoVDC(especiallythermocoupleconversionsystems)andintroducingdirect-currentsources,theycanbecomeuniversal instru-ments for also measuring high-frequency voltage up to a few hundred kHz andresistences.If thesemeasuring instrumentsare inplace theycanbeused tostoreandsubse-quentlycallupmeasuringvalues,andprocessthemandcontrolthemremotely,astheycanbeinterfacedwithmicroprocessorsystemsuntilautomaticmeasuringstruc-turesofgreatfunctionalversatilityareobtained.Twoparticularaspectsmustbeborne inmindwhenconstructingandusingdigitalinstrumentsinordernottocompromiseoperationandsafety:-electromagneticinference;-earthsockets.Themanufactureroftheinstrumentprotectsitdirectlyagainstelectromagneticinter-ferencebyprovidingtheinstrumentwithanelectrostaticscreen(anonferrousmag-neticmetal)thatisalsoeffectiveagainsthigh-frequencyelectromagneticfields.Thisscreencanbeconnectedtooneofthemeasuringterminalsoritcanconstituteathirdterminalinitsownright.Inthefirstcaseso-called'unbalanced'measurementsareobtainedasoneofthetwoterminalshastobeconnectedtothemeasurementearth,sothatonly twovoltagemeasurementsarepossiblethatrefertotheearthpotential.Ontheotherhand,ininstrumentswiththreeterminals,twoarededicatedtomeasur-ingandonethatisdedicatedtoshieldingisearthed.Inthiscasepotentialdifferencescanbemeasuredalsobetweenthetwopoints thatarebothoutsideearthandthetypeofmeasurementisknownas'balanced'.Earthconnectionsaredefinedasapointthepotentialofwhichremainsconstantandwhichisassumedtobeareferencepotential;thisisobtainedthroughanearthcon-nectionwithverylowimpedance.Inelectronic/digitalinstrumentsitmaybenecessarytohaveseveralreferencepointstowhichdistinctpartsofthecircuitsoftheinstrumentbelong;thesepointsareknownasmassconnectionsandareohmicallyisolatedfromoneanother(thecapacitivecou-plingmustalsobeminimised).ThesymbolsthataremostcommonlyusedforearthandmassconnectionsaresetoutinFig.3.3.

Fig. 3.2: General configuration of a digital instrument

Fig. 3.3: Symbols most commonly used for earth (a) and mass (b, c)

connections

3

ME

As

ur

ing

ins

Tr

uM

En

Ts

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 15

3.3

Measurement errors and accuracy ratingsNomeasurementcanbeconsideredexact.Thelimitsmustthereforebeestablishedeachtimewithinwhichthevalueofthemeasuredquantityfalls,thusdefiningthemar-ginoferrorofthemeasurement.Errorsmayariseinameasuringoperationformanyreasonsandhavevariousorigins.Apartfromtheerrorsduetomajormistakes(e.g.fittinganinstrumentincorrectly),itispossibletodividethetypesoferrorintotwocategories:systematicandaccidental,asexplainedbetterintheblockdiagraminFig.3.4.

Fig. 3.4: The main causes of error in electric measurements(1)Parallaxandappreciationerrorsaretypicalonlyofanalogueinstruments

Causes of error

Theydonotdependontheoperator;theydependontheequipmentandonthe

measuringprocess

Systematic

Theydependontheinstrument

class

Instrumental

Theyareduetocurrentabsorptionbytheinstruments

andtovoltagedrops

Loss

Theyoccurwhenthescaleindexisnotreadperpendicularlytothe

scale

Parallax (1)

Theyarisefromvisualestimatesoffractionsofthescalewhentheindexdoesnotstop

aboveadivision.

Estimate (1)

Theydependontheoperator

Subjective

Reading Incorrect method

Theydependoninstrumentfaults,assemblyerrors,

blows,vibrations,unstablecontacts,etc.

On equipment

Theyareduetofortuitouscircumstances;theyvaryin

valueandtype

Accidental

3

ME

As

ur

ing

ins

Tr

uM

En

Ts

16 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

Regardlessofthecausesoftheerror,anabsoluteerrorisdefinedasεaofthemeas-urementofanyquantity,thedifferencebetweenthevalueVmsuppliedbythemeas-urementandthetruevalueVVofthequantityunderexamination.Thisisthusformu-latedas:

εa = Vm – VV

Inpracticepeopleprefertotalkofarelativepercentageerrorthatisobtainedbydi-vidingtheabsoluteerrorεaabythetruevalue(VV)ofthequantity,allmultipliedby100:

Vm – VV εa εr % = · 100 = · 100 VV VV

TheformulashowsthatthepercentageerrordecreasesasVmincreases,i.e.bythemeasuredvalue.AstheabsoluteerrordoesnotingeneraldependonVm,itisdeducedthattherelativeerrorislesswhenthevalueontheinstrumentistowardsthefullscale.Infact,if,forexample,thereisanabsolutevalueof0.5Vwithavoltmeterthatinonecasereads50Vandintheothercase100V,theerrorsare:

εr % = 0.5 · 100 = 1 % εr % = 0.5

· 100 = 0.5 % 50 100

Inotherwords,inthesecondcasethereisarelativeerrorthatisthehalfofthefirsterror.Thisfactmustbeborneinmindwhenchoosingthemeasuringinstrument,asinanalogueinstrumentsthereadingshouldalwaysbetakentowardstheendofthescale.Itisequallyimportanttoknowtheaccuracyratingofaninstrument,toknowaprioritheabsoluteerrorsthatwillbemadeandthusevaluatewhethertheaccuracyofthemeasurementcanbeconsideredtobesatisfactory.ElectricalinstrumentsareinfactdividedonthebasisoftheiraccuracyratingintothefollowingcategoriesinconformitytoIECstandards:

0.05 – 0.1 – 0.2 – 0.3 – 0.5 – 1.0 – 1.5 – 2.5 - 5

Thesenumbersrepresenttheabsoluteerrorsinrelationtonominalcapacityandarestatedasapercentageofnominalcapacity.Thismeansthata0.5ratingvoltmeterwithnominalcapacityof200Vmustnothaveatanypointofthescaleanabsolutepercentageerrorthatisgreaterthan±0.5%.Inotherwords,itsmarginofabsoluteerroris:

± 0.5 · 200 εa = = ± 1 V 100

Thus,whateverthevoltagevaluethatisreadontheinstrument,thisreadvaluemustnotbemorethan1Vhigherorlowerthantherealvalue.Theclassofaninstrumentthuscoincides,asanumericvalue,withtherelativeerrorevaluatedatfullscale,whichinthecaseoftheexampleis:

1 εr = · 100 = 0.5 % 200

Inthecaseofdigitalinstrumentsthepercentageerrorofthereaderror(withrespecttothetruevalueofthemeasuredquantity)isnormallyindicatedwithadoubleindex,asintheexampleshowninthenextpage.Inparticular,theindicationwithwhichtheerrorisstabilisedisshownbyaseriesofabbreviations and numbers and is generally shown in the instrument's technicaldata.

continues 3.3

3

Me

as

ur

ing

ins

tr

uM

en

ts

Made to measure. Practical guide to electrical measurements in low voltage switchboards 17

Example

Declared error: ±1% rdg. ±4 dgt.;where: rdg. is the abbreviation of reading and

dgt. is the abbreviation of digit.Chosen capacity of the instrument 300 VResolution 0.1 VRead value 30 V

To assess the measurement error, proceed as follows:- maximum error of read value ±1% di 30 = ±0.3 V- error due to movement of the last digit ±4 digits = ±0.4 V- maximum possible error 0.3 + 4 = ±0.7 V

All other things being equal, if instrument resolution were 1V instead of 0.1 V, the measurement error would be:- maximum error of read value ±1% di 30 = ±0.3 V- error due to movement of the last digit ±4 digits = ±4 V- maximum possible error 0.3 + 4 = ±4.3 V

In digital instruments particular attention must also be paid when the instrument is used to measure alternating current; in this case it is important that the instrument is able to detect the effective true value (TRMS) of the quantity. Many instruments (mul-timeters, amperometric sensors, etc) have been built and calibrated to measure only quantities with a sinusoidal shape and network frequency (50 Hz).If these instruments are used in plants with non-linear loads or in the presence of har-monics (user devices such as computers, dimmers, photocopiers, microwave ovens, inverters, televisions, etc) very great reading errors can be made (up to 50% less than the true effective value). In order to include the influence of harmonic currents in the measurement, instruments with a wide frequency response (at least up to 1000 Hz) must be used.When voltmeters are used to measure voltage in environments with strong magnetic fields (in transformer rooms, in the presence of large engines, near high-voltage lines, etc), great attention must be paid to the influence that these electromagnetic fields may have on the instrument.The voltmeters that are normally used to conduct voltage measurements in the elec-trotechnical-plant sector are generally voltmeters with high internal impedance. The high internal impedance of a voltmeter, which is typical of digital instruments or of instruments with an electronic input, is the feature that enables voltage measurements with high resolution to be conducted, i.e. enables small voltage values or small vari-ations to be appreciated even with little energy available. For this instrument also the connecting cables can cause measuring errors due to the presence of strong elec-tromagnetic fields.In fact, the cables inserted into an electromagnetic field are the seat of induced elec-tromotive forces.The longer and more extended measuring cables and the higher the internal imped-ance of the voltmeter, the higher the induced (disturbance) voltage value comprised in the measurement. These voltmeters can indicate voltage values above 100 V with one test probe connected to earth that is not carrying voltage and with the other test probe in the air.

continues 3.3

3

ME

As

ur

ing

ins

Tr

uM

En

Ts

18 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

3.4

Comparison between the two categories of instruments: advantages and limitsAnalogueinstrumentsweretheonlyonesthatexisteduntilafewdecadesagoandtheyperformed (andstillperform) their functionsoutstandinglywell; inparticular inpanelinstrumentationtheirtoughnessandreliabilityarestillvalidandappreciated.Digitalinstrumentsobjectivelyoffermanyadvantagesoverthecorrespondinganaloguedevices,inparticular:theyareeasytoread,asthereisnointerpolationbetweentwocontiguousdivisionsandthecalculationoftheconstantofthescale,thereisgreateraccuracyandhighresolution,lownoiselevel,highmeasuringspeed,thepossibilityof even direct insertion into an automatic measuring complex controlled by acomputer.Thechoiceof instrumenttypemustbemadebyassessingreal instrumentrequire-mentswithregardtotheelectricplant,thepanelorthemeasuringcircuitwhereithastobeinserted:ifontheonehanditispointlesstorequirefunctionsthatwillneverberequired,forexample,fromavoltmeterthathastobeinsertedintotheshopfloorpowerpanelofanengineeringcompanyforthesolepurposeofindicatingthepresenceofvoltage,itshouldstillberememberedthatelectronicinstrumentsthatcanstoreandprocess the values of the measured quantities are almost indispensable in plantswhere monitoring the quality of the energy and/or cost reduction (for example formonitoringthepatternofloads)areprimeobjectives.

3

ME

As

ur

ing

ins

Tr

uM

En

Ts

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 19

20 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

4Direct and indirect measurements: CTs, VTs, converters and accessories

In order tomeasure electrical quantities, themeasuring instrumentsmustbe con-nectedtothelinessafely,withmaximumsimplicityandconvenience.Generally,thefundamentalparameterstobedetectedarevoltageandcurrent,whichrespectivelyrequireaparallelconnectionandaserialconnectiontothelineonwhichthemeasurementistaken.

4.1

Direct measurementsThedirectconnectiontothelinedefinesadirectmeasurementofthequantityastheinstrument is connected in the measuring point without the interposition ofadapters.Thedirectmeasurementispossibleonlywhenthequantitytobemeasuredhasalevelthatiswithintheinstrument'scapacity.Forexample,if230Vvoltagehastobemeasured,theinstrumentmusthaveacapac-itythatisgreaterthanthisvalue(forexample300V).Thisalsoappliestothecurrentmeasurements:ifcurrentsupto5Ahavetobemeas-ured,aninstrumentwithatleast5Acapacityand0-5Ainputisrequired.Panelandcabinetinstrumentsfordirectmeasurementsgenerallyconsistofintrumentswithverylimitedcapacity(measurementofsmallcurrentandvoltagevalues)withoneorseveraladditionalresistancesinsertedinsideforthevoltmetersand/oroneormoreshuntsfortheammeters.Whenthecapacityresistancesare inserted intothe instrument, the instrumentcanbeconnecteddirectlytothelineswherethemeasurementisconducted.

4.2

Indirect measurementsWhenthequantitytobemeasuredislargerthanthecapacityofthemeasuringinstru-ment,atransformermustbeinterposedthatreducesthequantityandsuppliesthequantitytotheinstrumentwithvaluesthatarecompatiblewithitscapacity.Thismeth-odologyisdefinedasindirectmeasuring.Themeasurementconductedviaameasuring transformer isdefinedasan indirectmeasurementbecauseitdoesnottakeplacedirectlyonthelineunderexamination.Forexample,ifacurrentupto100Ahastobemeasuredwithancurrentthathasacapacityof5A,acurrenttransformer(CT)hastobeinterposedwithatransformationratioof100/5.Ifthecurrenttransformerisofthewoundprimarytype,itisconnecteddirectlyseriallytotheconductoronwhichthecurrenthastobemeasured.Ontheotherhand,ifitisofthetypewithathroughprimary,theinsulatedorbareconductorisinsertedinsidetheholeofthedevice.

4

Dir

EC

T A

nD

inD

irE

CT

ME

As

ur

EM

En

Ts

: AT

s, V

Ts

, CO

nV

Er

TE

rs

An

D A

CC

Es

sO

riE

s

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 21

Thecurrenttransformerhasanoutletthatwillsupplyacurrentthatisreducedby20timesthecurrentthatcirculatesintheconductorbeingmeasured,towhichthecur-rentwith5Acapacityisconnected.Incurrenttransformerstheprimarywindingisintendedtobeconnectedserialytothecircuittraversedbythecurrenttobemeasured,whilstthesecondarywindingsuppliesoneormoremeasuringinstruments(allseriallyconnectedtoeachother).ThewiringdiagraminFig.4.1.showsthesetransformers.Comparedwiththeoperatingprincipleofanormaltransformer,theCTisdesignedtomakethemagnetisationcurrentI0negligeablethatisrequiredtoproducetheflowΦinthecore.Intheseconditions,theprimaryandsecondarycurrentsareinexactphaseopposi-tionandtheirrespectiveeffectivevaluesareinaratiotooneanotherthatisinversetothenumberofcoilsN1andN2.Inotherwords:

Ip N2 = = n Is N1

fromwhich: Ip = n Is

Thecoilrationbetweenthesecondaryandprimarywindingisthustheidealtransfor-mationratiobetweentheprimaryandsecondarycurrent.Infact,themagneticcoreofthetransformercannothavenilreluctanceandIEC38-1standardsdefine,foreverysingletransformer,theprimaryandsecondaryreferencecurrents,whichconstitutethenominalcurrentsIPnandISnofthetransformer.Theratiobetweenthesetwocurrentsisthenominalratio:

IPn Kn = Isn

whichisindicatedbyalwaysspecifyingthenumeratoranddenominator:thecurrenttransformeris,forexample,saidtohaveanominalratioof75to5AandiswrittenforthesakeofbrevityasCT75A/5A.Table4.1showstheratioandangleerrors(phasedifferencebetweentheprimaryandthesecondarycurrent)permittedbyIECstandardsforcurrenttransformers.

Accuracy ratingCurrent as %

of nominal valueRatio errors

%

Angle errors

in arc minutesin hundredths

or percentages

0.1

1010100120

±0.25 ±0.2 ±0.1 ±0.1

±10 ±8 ±5 ±5

±0.3 ±0.24 ±0.15 ±0.15

0.2

1020100120

±0.5 ±0.35 ±0.2 ±0.2

±20 ±15 ±10 ±10

±0.6 ±0.45 ±0.3 ±0.3

0.5

1020100120

±1 ±0.75 ±0.5 ±0.5

±60 ±45 ±30 ±30

±1.8 ±1.35 ±0.9 ±0.9

1

1010100120

±2 ±1.5 ±1 ±1

±120 ±90 ±60 ±60

±3.6 ±2.7 ±1.8 ±1.8

350120

±3 ±3

noprescription

550120

±5 ±5

noprescription

continues 4.2

Fig. 4.1: Wiring diagram of current transformer (CT)

Table 4.1 CT ratio and angle errors permitted by IEC standard.

4

Dir

EC

T A

nD

inD

irE

CT

ME

As

ur

EM

En

Ts

: AT

s, V

Ts

, CO

nV

Er

TE

rs

An

D A

CC

Es

sO

riE

s

22 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

When there is theproblemofmeasuringhighvoltagesorvoltages thataregreaterthanthecapacityoftheinstrument,voltagetransformersareused(indicatedbythelettersVT),theprimaryofwhichissuppliedwiththeUPvoltagetobemeasuredwhilstthetransformersusethesecondarytosupplythemeasuringinstruments(allparalleltooneanother)attheUSvoltage.ThewiringdiagraminFig.4.2.showsthesetransformers.

Similarly to thecurrent transformers, the theoretical rationbetweenthenumberofcoilsofthetwowindings(idealtransformationratio)isgivenbytheformulas

UP EP NP = = = n Us Es Ns

However,inpracticethefallsinohmicandinductivevoltageofthetwowindingsmeanthattheratioUP/USdiffersfromthecoilsnratio,givingrisetoaratioerrorηV %.Ac-cordingly,foreverysingletransformer,themanufacturersetsthenominalprimaryandsecondaryvoltages,whichcorrespondtoasetloadcondition:thetwodefinedvolt-agesconstitutethenominalvoltagesofthetransformer,whichmustbeindicatedre-spectivelybythesymbolsUPnandUSn.Theratiobetweenthesetwovoltagesisthenominalratioofthetransformer:

UPn Kn = Usn

whichmustbeindicatedbyalwaysspecifyingthetwoterms:thevoltagetransformeris,forexample,saidtohaveanominalratioof10,000to100VandiswrittenforthesakeofbrevityasVT10,000V/100V.Also for the VTs Table 4.2 shows the ratio and angle errors specified by the IECstandard.

ClassesRatio errors

%

Angle errors

in arc minutes in hundredths

0.10.20.51.03.0

±0.1 ±0.2 ±0.5 ±1 ±3

±5 ±10 ±20 ±40

±0.15 ±0.3 ±0.6 ±1.2

noprescription noprescription

Toconcludethisdiscussionofvoltageandcurrentmeasuringinstruments,weremindthereaderthatwhenthemarginoferrorofthemeasurementisevaluated,theerroroftheinstrumentmustalwaysbeaddedtotheerrorofthetransformer;e.g.iftheac-curacyratingoftheinstrumentis1.5andtheaccuracyratingofthetransformeris0.5themarginoferrorcanbe±2%ofthereadvalue(accuracyrating2).

continues 4.2

Fig. 4.2: Wiring diagram of voltage transformer (VT)

Table 4.2 VT ratio and angle errors permitted by IEC standard.

4

Dir

EC

T A

nD

inD

irE

CT

ME

As

ur

EM

En

Ts

: AT

s, V

Ts

, CO

nV

Er

TE

rs

An

D A

CC

Es

sO

riE

s

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 23

4.3

Shunts for direct currentWhenthecapacityofaninstrumentislessthanthecurrenttobemeasured,shuntsareused:theseareadditionalresistorsthatareconnectedparalleltotheinstrumenttoshuntpartofthecurrenttobemeasuredandtolimitthecurrentthatpassesthroughtheinstrumenttoanacceptablevalue.Fig. 4.3 shows the wiring diagram of a shunt for measuring a direct current via amillivoltmeter.Inordertoreachthedesiredcapacity,theshuntmustbeproportioned(orselected)accordingtothecurrentdivider;intheFig.4.3wehave:

Rs 1 I I' = I' R+Rs m

fromwhich: I' = m I = K'A n

being R+Rs I' m = = I' Rs I

themultiplyingpoweroftheshunt,nthenumberofdivisionsreadonthescale,andK'Athenewreadingconstantoftheinstrument,expressedbytheproductK'A = m KA

I’

Is

U

I’I

I

(Ri)

Rs

A B

4.4

Converters and accessoriesTheconvertersaredevicesthat, if theyareconnectedtoelectricnetworkswithanalternatingcurrentsignal,areabletoprovideadirectcurrentorvoltagesignalthatisproportionaltotheinputsignal,regardlessoftheload.Theyareparticularlysuitableforacquiringhighlyreliableandaccuratedataandarenotaffectedbytemperaturevariationsandvibrations.Theconvertersgenerallyhaveseveraloutputsthatcanbeselectedtoadapttovari-ousneeds.InadditiontotheCTs,VTsandtotheconverters,themeasuringaccessoriesare:-the interchangeable scales to adapt analogue instruments to the desired

capacities;-thecurrentandvoltageswitchesforswitchingreadersonseveralcurrentandvolt-

agephases;-thetransducers,whicharenecessaryforthedirectinsertionoftheanaloguepower-

factormeters.Currentandvoltageconvertersconvertersproduceadirectcurrentorvoltageoutputsignalthatisindependentoftheloadthatisdirectlyproportionaltotheinputvoltageorcurrentsignal.Theirelectroniccircuitensurestheirreliabilityandoperatingaccuracy,theextensionofthemeasurementfield, insensitivitytotemperaturevariationsandvibrations,andlimitedabsorptionofpowerfromthecircuitbeingmeasured.Theirrapidcentralisedacquistionofdata,evenovergreatdistancesandtheavailabilityofdifferenttypesofselectableoutletsbyactingsimplyontheadjustingminidips,makesthemsuitableforbeinginstalledinplantsthatrequireparticularattentioninproduction,distributionanduseofelectricpower.

Fig. 4.3: Measurement of direct current with millivoltmeter and external shunt

Fig. 4.1 – Current and voltage convertersFig. 4.1 – Current and voltage converters

24 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

5Overview of ABB range

Themeasuringinstrumentsforinstallationinsideindustrialswitchboardsforprimaryandsecondarydistributionofmediumand lowvoltagearean idealcomplement toABBdeviceswithwhichtoconfigurethepanelasanintegratedfunctionssystem.The range comprises about 1000 products in the basic versions but engineering/standardisationofthecomponentsalsomakesmanyspecialversionsavailableinor-dertosatisfyanytypeofplantneed.Bothanalogueanddigitalinstrumentsareavailable:intheformerthereadingfunctionisprovidedbymovementofamovableindexalongagraduatedscale,whichenablesthedetectedvaluestobereadimmediately;thedigitalversionsare,ontheotherhand,equippedwith3or4digitdisplayLEDs,dependingonproducttype.Inbothversions,theoperatingtemperatureisbetween-10°Cand+55°C,withthepossibilityofoperatinginevenmoredifficultconditionswithoutsubstantialalterationstotheaccuracyrating.TheresistancetovibrationsandtheIPprotectiongradeareparticularlyhigh.

5.1

Analogue instrumentsInadditiontonormaldevices formeasuringelectricalparameters (voltmeters,am-meters,power-factormeters)therangeofanalogueABBinstrumentscomprisesspe-cialinstruments(meters)andaseriesofaccessories,includingthecurrenttransform-ers,whichextendtheiroperatingrange.Thereare twodistinctproduct ranges: theDINrailproducts,whicharesnap-fittedontoanordinaryDINrailandhavedimensions,compactnessanddesignthatperfectlysuit thecommandandprotectiondevicesof theSystemproMcompact®,Systemseriesandthepanelfrontinstrumentsthatcaneasilybemountedinthemediumandlowvoltageprimaryandsecondary industrialdistributionpanels.Theyare fittedbymeansofscrewbracketsthatenablethedevicetobeplacedbothinahorizontalandverticalposition, thusoptimisingspaceoccupiedandrationalisingaccess fromthefrontofthepanel.

5.1.1

DIN rail analogue instrumentsTable 5.1 summarises the characteristics ofDIN rail ABBanalogue instruments; forcompleteinformationonthetechnicalcharacteristicsofthedevices,seethetechnicalcatalogueSystemproM compact®.

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 25

5

PAn

Or

AM

iCA

DE

llA P

rO

Du

ziO

nE

AB

B

ABB analogue measuring instruments

AC DC

-Directvoltmeters-Directammeters-AmmeterswithoutscaleforCT-Frequencymeter45-65Hz-Power-factormeterwithscalefortransducers(1mAinput)

DirectammetersAmmeterswithoutscaleforshunt

Technical characteristics

RatedvoltageUn [V] AC300,500;DC100,300

Ratedalternating directreadingcurrents indirectreading

[A] fullscalevalues5...30fullscalevalues5...2500

Rateddirectcurrents directreading indirectreading

[A] fullscalevalues0.1...30fullscalevalues5...500

Frequency [Hz] 50/60

Overloadability [%] 20comparedtoratedvoltageorcurrent

Accuracyrating [%] 1.5(0.5forfrequencymeters)

Dissipatedpower [W] seeSystemproM compact®catalogue

Modules [n°] 3

Standards EN60051

Both thedirect insertion instrumentsandthose thatare insertibleviaCTorshunts(seefigure5.1forinsertionmethods)donotrequireanauxiliarysupply.Fortheformeritissufficienttoconnectafterchoosingtheratedvoltageorcurrent;fortheothers:-choosethenominalmeasurement(current,voltage,...);-selectthecurrentorvoltagetransformer,shuntortransducer;-selecttheappropriatescale;-connecttheinstrument.

V

1 2 3 4

L1

N

A

1 2 3 4

L1

N

A

1 2 3 4

L1

N

S1 S2P1 P2

A

1 2 3 4

L1

N

5.1.2

Analogue instruments on front of panelThe range includesvoltmeters,ammeters,power-factormetersand frequencymeterswithafixedormovablereel,dependingonversions.Withthepassageofcurrentinthedevicesprovidedwithafixedreel,thetorqueproducedbytheelectromagneticfieldmovesapieceofironalongthequadraticscale,theironbe-ingconnectedtothedisplayindex.Owingtotheirparticularresistancetocurrentpeaks,fixedreeldevicesaremoresuitableforalternatingcurrent.Withthepassageofcurrentinthedevicesprovidedwithafixedreel,thetorqueproducedbytheelectromagneticfieldmovesapieceofironalongthequadraticscale,theironbeingconnectedtothedisplayindex.Theclockwisemovementoftheindexdependsonpolarity,whichmeansthatthesede-vicescanbeusedfordirectcurrentonly.Thevoltmetersandtheammeters,whichareavailablebothintheversionforalternatingcurrentandintheversionfordirectcurrent,aresuppliedinthreestandarddimensionsof48mmx48mm,72mmx72mmand96mmx96mm(specialversionsavailableonrequest).Forammeterswithoutascale,theinterchangeablescalecodewithwhichaccessorisethemisindicated.Therangeoffrontpanelmeasuringinstrumentsiscompletedbypower-factormetersand frequencymeters forapplicationsonsingle-phaseand three-phase

continues 5.1.1

Table 5.1 ABB DIN rail analogue measuring instruments

Fig. 5.1: Insertion method (direct, via CT and shunt) of the analogue instrumentsDirect insertion Insertion via CT Insertion via shunt

5

OV

Er

ViE

w O

f AB

B r

An

gE

26 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

alternatingcurrentlinesinthethreestandarddimensionsof48mmx48mm,72mmx72mmand96mmx96mm.Fig.5.1showssomeoftheseinstrumentsandthetechnicalspecificationsaresetoutinTable5.2.Foracompletedescriptionoftheinstruments,thetypeandthecodeforordering,seethetechnicalcatalogue2CSC400002D00209SystemproMcompact®.

Technical characteristics

Max. rated insulation voltage V 650

Test Voltage V 2000effective(50Hz/1min)

Accuracy rating 1.5(0.5forfrequencymeters)

Overloadability(1) :

-ammeterwindings uptoInx10/<1sec.

uptoInx2/permanent

-voltmeterwindings uptoUnx2/<5sec.

uptoUnx1.2/permanent

Operating temperature °C -20…+40

Storage temperature °C -40…+70

Average and max. relative humidity (DIN 40040)(2)

65%(annualaverage)85%(+35°C/60daysperyear)

Vibration resistance (IEC 50-1) g(9.81m/s) 0.08-1.8(0.35mm/10-55Hz;3axes/6h)

Protection class IP52forinterior

IP00onterminals(IEC144,DIN40050)

IP40withtheterminalcovers

Manufacturing material:

-casesandfrontedge self-extinguishingthermoplasticmaterialconformingtoUL94V-0,resistanttofungiandtermites

-displayindices(DIN43802)(3) mouldedaluminium

-terminals brass

Assembly vertical/horizontalbymeansofscrewbrackets(4)

Dimensions L x H x D (DIN 43700/43718) mm 48x48x5372x72x5396x96x53

Reference Standards IECEN61010-1(1)IninstrumentswithinsertionbyCT,theoverloadmaybegreaterbecausethetransformer

containsthepeaksofsecondarycurrentwithin10In.(2)Tropicalisationenablesvaluesupto95%max.relativehumiditytobetolerated(+35°C/60

days).DIN40040requiresthemtobeprotectedfromhumditypenetratinginsidethem.Terminals,screws,washers,boltsandmagnetsaregalvanicallyprotectedagainstrustwhereastheelectricalcircuitsarecoatedwithspecialMulticolorPC52paint.

(3)Thedampingtimeforthedisplayindicesis1second.Thedetectedvaluesareresetbymeansoftheappropriateadjustment.

(4)Withpanelsthatare0.5mm–19mmthickthescrewsmustbefittedinthefixingpositionthatisnearestthefrontedgeofthemeasuringdevice.Thepanelsthatare20mm-39mmthickarefixedbyscrewsinthepositionthatisfurthestawayfromthefrontedge.

continues 5.1.2

Fig. 5.2: Analogue measuring instruments on front of panel

Table 5.2 Technical characteristics of front panel analogue measuring instruments

Max. rated insulation voltage

Test Voltage

Accuracy rating

Technical characteristics

Max. rated insulation voltageMax. rated insulation voltage

Technical characteristics

V 650

V 2000effective(50Hz/1min)

1.5(0.5forfrequencymeters)

Technical characteristics

2000effective(50Hz/1min)

1.5(0.5forfrequencymeters)

2000effective(50Hz/1min)

1.5(0.5forfrequencymeters)

5

OV

Er

ViE

w O

f AB

B r

An

gE

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 27

5.1.3

Fig. 5.3: DIN rail switches and front panel switches

Fig. 5.4: a – 90° full scaleb – 78° full scale

with extra scale

AdvantagesAnalogueABBmeasuringinstrumentsaredistinguishedbytheirreliabilityandstabilityinindicatingthemeasuredvalue,thusmakingevenremotereadingsimple;theyalsohavethefollowingfeatures,whicharemuchappreciatedduringtheinstallationphase:-reductionofoveralldimensions;-completerangeforpanelfrontinstruments(48x48,72x72,96x96mm);-donotrequireauxiliarysupply;-areabletoprovidemultiplereadingsthankstotheselectors.Itisveryconvenientforthefitterandwholesalertohaveasingleinstrumentwitham-plecapacity (from5A to2500A),completewithawide rangeofaccessoriesandaccompanyingdevicesforinsertion,whichincludetheDINrailswitches(fig.5.3).

A final noteon the typeof scales available that are interchangeable to extend thereading scope of electrical measurements detectable with analogue measuringinstruments.Forexample,infigures5.4aand5.4btwodifferenttypesofdialsforscalesareshown:thefirsthasatraditional90°fullscale,thesecondhasa78°fullscaleplusanextrascale,whichcanbeusefullyusedwhen,duringthemeasurement,peakcurrentsthatcouldexceedthefullscalearedetected(forexampleduringthestartupphaseofanasynchronousmotor).

100

0

90°

100

0

78°

100

0

90°

100

0

78°

SCL1/A1/100 SCL1/A5/100

A final noteon the typeof scales available that are interchangeable to extend thereading scope of electrical measurements detectable with analogue measuringinstruments.Forexample,infigures5.4aand5.4btwodifferenttypesofdialsforscalesareshown:

A final noteon the typeof scales available that are interchangeable to extend thereading scope of electrical measurements detectable with analogue measuring

Forexample,infigures5.4aand5.4btwodifferenttypesofdialsforscalesareshown:

A final noteon the typeof scales available that are interchangeable to extend thereading scope of electrical measurements detectable with analogue measuring

Forexample,infigures5.4aand5.4btwodifferenttypesofdialsforscalesareshown:

A final noteon the typeof scales available that are interchangeable to extend thereading scope of electrical measurements detectable with analogue measuring

Forexample,infigures5.4aand5.4btwodifferenttypesofdialsforscalesareshown:

5

OV

Er

ViE

w O

f AB

B r

An

gE

28 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

5.2

Digital instrumentsTherangeofABBdigitalinstrumentsisparticularlywide:alongsidetraditionalmeas-uringinstruments(voltmeter,ammeter,frequencymeter),bothintheDINrailandinthepanelfrontversion,thereare:-themultimetersoftheDMTMEseriesthatnotonlyenablethemainelectricalquanti-

tiestobemeasuredbutalsostorethemaximum,minimumandaveragevaluesofthemainelectricalquantitiesandcountactiveandreactiveenergy;

-MTMEandANRseriesnetworkanalysers thatnotonlymonitor thequalityof theenergyinrealtimebutarealsoabletodisconnectloadsandsendalarmsignals;

-energymeters;-temperaturemeasuringunits.Inaddition,awiderangeofaccessoriesmaketheseinstrumentsuniversalforelectricplantsandnetworksinthefollowingrange:-voltageupto600V-currentupto999A-frequency:from40to80HzItshould lastlybenoted that theabsenceofpartssubject towear through frictionensuresalongeroperatinglifeandparticularlyhighregulatingaccuracy.

5.2.1

DIN rail digital instrumentsTable5.3summarisesthecharacteristicsofABBDINraildigitalinstruments;forcom-plete information on the technical characteristics of the devices, see the technicalcatalogueSystemproM compact®.

Alltheinstrumentsofferhighmeasuringaccuracy(0.5rating)andeasyandaccurateread-ingofthemeasuredvalues:therangeiscompletedbyinstrumentswithaninternalrelaywhichdisplayandmonitorameasurementandwhenaprogrammablethresholdisexceededtheytriparelaycontactanddisplaythealarmcondition.Thealarmthresholdcanbepro-grammedasaminimumormaximumthreshold.Theminimumandmaximumrecordedpeakvaluesaresavedintheinstrument'spermanentmemory.Therelayactionisprogram-mable.Thefactorysettingisnormallyopencontactthatclosesonlyintheeventofanalarm.Inprogrammingmodetheinstrumentcanbeconfiguredinsuchawaythattherelayworkswithpositivesafety:inthiscasetherelaywillcloseduringcorrectoperationandwillopenintheeventofanalarmorpowerfailure.Theinstrumentwiththerealycanbeusedalter-nativelyeitherasaminimumrealyorasamaximumrealybutnotsimultaneouslyforbothalarms.Theintrumentsalsoenabletheminimumandmaximummeasurementvaluetobestoredanddisplayed.

Table 5.3 ABB digital measuring instruments

5

OV

Er

ViE

w O

f AB

B r

An

gE

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 29

5.2.2

Front panel digital instrumentsTheseinstrumentsareprovidedwiththree-digitredLEDsdisplaystoindicateimme-diatelytheelectricalvaluesdetected.Afewsimpleoperationsenablethemultiscalefunctiontobeaccessedthatenablestherangeofdisplayablequantitiestobevariedorextended.Theproduct rangecomprisesvoltmeters,ammeters fordirector indirectmeasure-mentsusingdevicetransformersandshuntsandtemperaturemeasuringunits.Theapplicationcanberuninbothalternatinganddirectcurrent.Theabsenceofmechanicalpartsthataresubjecttowearmakestheseinstrumentsveryreliableanddurable.Fig.5.1showssomeoftheseinstrumentsandthetechnicalspecificationsaresetoutinTable5.4.Foracompletedescriptionoftheinstruments,thetypeandthecodeforordering,seethetechnicalcatalogue2CSC400002D0208SystemproMcompact®.

continues 5.2.1

Fig. 5.5: Insertion method of different ABB digital instruments

Fig. 5.6: Panel front digital measuring instruments

Wiring diagrams for digital instruments, both DIN rail and front panel

VLMD-1-2 and VLMD-1-2-RVLMD P and VLMD-R P

� � � �� � � � � �� �

� � � �� � � � � �� �

FRZ-DIG

AMTD-1 and AMTD-1-RAMTD-1 P and AMTD-1-R P

AMTD-2 and AMTD-2-RAMTD-2 P and AMTD-2-R P

*Onlyforinstrumentswithoutputrelay

5

OV

Er

ViE

w O

f AB

B r

An

gE

30 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

Technical characteristics

Power supply [V] 230VCA

Rated frequency [Hz] 50÷60

Ammeter full scale value [A] 5,20,25,40,60,100,150,200,250,400,600

Voltmeter full scale value [V] 300,500

Frequency meter range [Hz] 35...400

Tripping delay [s] 1,5,10,20,30

Hysteresis [%] 5,10,20,30setthreshold

Output pins [V] 3-4

Output relay [A] NO

Rated voltage relay [In/Vn 230VCA

Rated current relay [%] AC116,AC153

Relay configuration NOrelayclosesinalarmstatus

NCrelayopensinalarmstatus,positivesafety

Overload [In/Vn] 1,2

Accuracy class [%] ±0,5fullscale±1digitat25°C

Max. signal input value for ammeters [°C] 5ACA/60mVDC

Display [°C] 3digitLEDdisplay

Operating temperature [VA] -10…+55

Storage temperature [mm] -40...+70

Protection degree IP20

Power consumption 4

Modules 3

Overall dimensions front panel devices

36x72x61.5(51.5depthinsidetheswitchboard)

Standard IECEN61010

5.2.3

DMTME multimetersTheinstrumentsoftheDMTMEseriesaredigitalmultimetersthatenablesthemainelec-tricalquantitiestobemeasured(TRMS)in230/400VACthree-phasenetworks,themaxi-mum,minimumandaveragevaluesofthemainelectricalquantitiestobestoredandtheactiveandreactiveenergytobecounted.ThemultimetersoftheDMTMEseriesenablethesameinstrumenttoactasavoltmeter,ammeter,power-factormeter,wattmeter,varmeter,frequencymeter,activeandreactiveenergycounter,hourcounter.Thispermitssignificantsavingsbecauseboththesizeofthepanelandwiringtimearereduced.Fig.5.7ashowsaDMTMEDINrail(6-modules)multimeterthatcanbeinsertedviaCT.../5Aformeasurementson230/400VAClines(displayablemeasurements:V-I-W-VA-Hz-kWh-kVARh);theDMTME-I-485versionisprovidedwithtwodigitaloutletsthatcanbeprogrammedasalarmthresholdsandpulseoutletsforremotecontrolofenergycon-sumptionandaserialdoorRS485.Fig.5.7bshowsthemultimeterstobeinstalledinthefrontofthepanelinthetwover-sions:traditional96x96mmand72x72mminthemorecompactversionthatisidealforinstallationinpowercentresinwhichcompactdimensionsarerequired.FromtheserialportRS485severalmultimetersandotherdigital instrumentscanbeconnectedtothenetworkviatheModbusRTUprotocol.AlltheversionsaresuppliedwithaCDcontainingthe instruction manuals, technical documentation, communicaion protocol andDMTME-SWtool.

continues 5.2.2

Table 5.4 Technical characteristics of digital measuring instruments

on front of panel

Fig. 5.7a DIN rail multimeter DMTME

5

OV

Er

ViE

w O

f AB

B r

An

gE

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 31

Technical characteristics

Rated voltage [Vrms] 230+15%-10% DMTME-72andDMTME-96

[Vrms] 240+15%-10% DMTME-72andDMTME-96

[Vrms] 400+10%-10% DMTME-72

[Vrms] 400+10%-10% DMTME-72

[Vrms] 115+15%-10% DMTME-96

[Vrms] 120+15%-10% DMTME-96

Frequency [Hz] 45…65

Power input [VA] <6

Safety fuse 0.1A

Voltmeter inputs

Range [Vrms] 10…500V(L-N)

Nondestructivemax. [Vrms] 550

Impedance(L-N) [MW] >8

Current inputs (only CT .../5 A external)

Range [Arms] 0.05…5

Overload 1.1permanent

Measurements accuracy

Voltage ±0.5%F.S.±1digitinrange

Current ±0.5%F.S.±1digitinrange

Activepower ±1%±0.1%F.S.fromcosj=0.3tocosj=-0.3

Frequency ±0.2%±0.1Hzfrom40.0to99.9Hz

±0.2%±1Hzfrom100to500Hz

Energy count

Maximumcountedvaluepersinglephase 4294.9MWh(MVarh)withKA=KV=1

Maximumcountedthree-phasevalue 4294.9MWh(MVarh)withKA=KV=1

Accuracy Class1

Max.powerconsumption 1.4foreachinput(withImax=5Arms)

Digital outputs

Pulseduration 50msOFF(min)/50msON

Vmaxoncontact 48V(peakDCorAC)

Powerconsumption 450mW

Maximumfrequency 10pulses/sec

Imaxcontact 100mA(maxDCorAC)

Insulation 750Vmax

Confi gurable parameters

VTtransformationratio 1…500

CTtransformationratio 1…1250

Freehourcounter [h] 0…10,000,000,resettable

Countdown [h] 1…32,000

Operating temperature [°C] 0…+50

Storage temperature [°C] -10…+60

Relative humidity 90%max.(withoutcondensation)a40°C

Overall dimensions [mm] 96x96x103 DMTME-96

[mm] 72x72x90 DMTME-72

5.2.4

MTME and ANR network analysersTheMTMEseriesnetworkanalysers(Fig.5.8a)enablethemainelectricalquantitiestobemeasuredastrueeffectivevaluesin230/400VACthree-phasenetworks,themaximum,minimumandaveragevaluesofthemainelectricalquantitiestobestoredandtheactiveandreactiveenergytobecountedontotalorpartialcounters.OwingtotheTHD(totalharmonicdistortion)measurementasanabsoluteandper-centagevalue, it ispossible tomonitor in real time thequalityof theenergyof theplantandtopreventdamagetothedevices.MTMEnetworkanalysersarealsoable,dependingon theversion, tomanageanddisconnectloadstosaveenergyandoptimiseconsumptionandtosendalarmsignalsrelatingto34quantitiesviatworelayoutlets.

continues 5.2.3

Figura. 5.7b: Multimetri fronte quadro DMTMEFigura. 5.7b: Multimetri fronte quadro DMTME

Fig. 5.8a Network analyser MTME-485-LCD-96

5

OV

Er

ViE

w O

f AB

B r

An

gE

32 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

VersionswithanRS484portenableallthequantitiesofaninstrumentorofanetworkofinstrumentstobereadandmontoredlocallyorremotely.ThelocaldisplayofthequantitiesisshownonanLCDdisplaywithhigh-visibilitybacklighting.Thefollowingfeaturesarealsoavailable:-automaticrecognitionofthedirectionoftheCT(selectable)-programmablemainscreen-accesspassword-frimwarethatisupdatableviaPCAlltheversionsaresuppliedwithaCDcontainingtheinstructionmanuals,technicaldocumentation,communicaionprotocolandDMTME-SWsoftware.

Main characteristics of the MTME-485-LCD-96 network analyser

Rated voltage [V rms] 230 +15% - 10%

[V rms] 240 +15% - 10%

[V rms] 115 +15% - 10%

[V rms] 120 +15% - 10%

Frequency [Hz] 45…65

Power input [VA] < 6

Safety fuse T0, 1A

Voltmeter inputs

Range [V rms] 10…500 V (L-N)

Non destructive max. [V rms] 550

Impedance (L-N) [MΩ] > 2

Current inputs (always use CT .../5 A)

Range [A rms] 0.05…5

Overload 1.1 permanent

Measurements accuracy

Voltage ±0.25% ±0.3% F.S.

Current ±0.25% ±0.3% F.S.

Active power ±0.5% ±0.1% F.S. from cosj = 0.3 to cosj = -0.3

Frequency±0.2% ±0.1Hz from 40.0 to 99.9 Hz

±0.2% ±1Hz from 100 to 500 Hz

Energy count

Maximum counted value per single phase 4294.9 MWh (MVarh) with KA = KV = 1

Maximum counted three-phase value 4294.9 MWh (MVarh) with KA = KV = 1

Digital outputs

Pulse duration 50 ms OFF (min)/ 50 ms ON

Vmax on contact 48 V (peak DC or AC)

Power consumption 450 mW

Maximum frequency 10 impulsi/sec

Imax contact 100 mA (max DC or AC)

Insulation 750 Vmax

Configurable parameters

VT transformation ratio 1…500

CT transformation ratio 1…1000

Operating temperature [°C] 0…+50

Storage temperature [°C] -10…+60

Relative humidity 90% max. (without condensation) at 40°C

Overall dimensions [mm] 96x96x103

Ifevenmoreadvancedanalysisfunctionsarerequested,therangeofABBpanelinstru-ments,andANRnetworkanalysersenablenetwork,informationandalarmparametersto be measured and recorded by routing the data to supervision and monitoringsystems.TheSW01softwarewithwhichtheyareprovidedmanagestheprogramming,displayandrecordingofthemeasurementdataandofthealarms.Performanceistoplevel:-itispossibletomeasure,recordandanalyseover60electricparameters;

continues 5.2.4

5

OV

Er

ViE

w O

f AB

B r

An

gE

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 33

-thevoltageandcurrentsaremeasuredastrueandeffectivevalues(“trueRMS”)with0.5ratingaccuracy;

-communcationsareprovidedon:programmableanalogueoutlets,digitaloutletsforcontrols,pulsesandalarms,statusand/ornon-electricparametersacquisition,Mod-bus,Profibus,ASCII,Ethernetprotocols;

TheANRnetworkanalysersareavailableinthe96x96mmrecessedformatorinthe144x144mmversion(thelatterhaveexpansioncards)andhavegraphicback-lighted128x128pixelgraphicLCDdisplays.Theirusepermitsextremelyefficientmonitoringofthequalityoftheenergyinbothsin-gle-phaseandthree-phasedistributionnetworksviainstantaneousandhistoricalvari-ationsofvoltage,ofsupplyinterruptions,ofmicrodisturbancesandofharmoniccom-ponentsuptothethirty-firstorderandwaveshapes,andoptimisationofenergycoststhroughpromptandhistoricalanalysisofconsumptionoverfourperiodsinadaythatarefreelyselected,withmonitoringanddisconnectionofloads.

Main characteristics: ANR 144-230 network analyser

Packaging

Overall dimensions [mm] 96 x 96 x 130 - 144 x 144 x 66 IEC 61554

Max setion of wires [mm2] 2.5

Protection class IP52 frontal-IP20 terminal boards EN 60529

Weight [g] 430

Display

Graphic LCD 128x128 points with adjustable contrast with LED back lighting

Display dimensions [mm] ANR96: 50 x 50-ANR144: 70 x 70 IEC 60529

Voltage (TRMS)

Direct measurement [V] 10 - 600

Transformation ratio range kTV [V] 0.01 - 5000,00

Permanent overload 750, above this value a voltage transformer must be used

Consumption [VA] 0.2

Input resistance [MW] > 2

Current (TRMS)

3 insulated inputs with internal CTs .../5 A [A] 0.01 - 5

Minimum current measurement [mA] 10

Consumption [VA] 0.2

Display

Overload [A] 10 (100 A for 1 second)

Transformation ratio range kCT 0.01 - 5000,00

THD

Voltage and current Up to 31st harmonic

Frequency

[Hz] 30 - 500

Accuracy

Current [%] < 0.5 EN 61036

Voltage [%] < 0.5

Power [%] < 1

Power factor [%] < 1

Active energy [%] < 1 IEC 62052-11

Reactive energy [%] 2 IEC 62053-23

Separate power supply

ANR96-230, ANR96P-230, ANR144-230 [V] 85 ÷ 265 AC/DC

ANR96-24, ANR96P-24, ANR144-24 [V] 20 ÷ 60 AC/DC

Internal fuse 5 x 20 mm 315 mA 250 V Fast

Conditions of use

Operating temperature [°C] -10 ÷ +50

Storage temperature [°C] -15 ÷ +70

Relative humidity [°C] 90% non condensated

continues 5.2.4

-thevoltageandcurrentsaremeasuredastrueandeffectivevalues(“trueRMS”)with

-communcationsareprovidedon:programmableanalogueoutlets,digitaloutletsforcontrols,pulsesandalarms,statusand/ornon-electricparametersacquisition,Mod-

TheANRnetworkanalysersareavailableinthe96x96mmrecessedformatorinthe144x144mmversion(thelatterhaveexpansioncards)andhavegraphicback-lighted

Theirusepermitsextremelyefficientmonitoringofthequalityoftheenergyinbothsin-gle-phaseandthree-phasedistributionnetworksviainstantaneousandhistoricalvari-ationsofvoltage,ofsupplyinterruptions,ofmicrodisturbancesandofharmoniccom-ponentsuptothethirty-firstorderandwaveshapes,andoptimisationofenergycoststhroughpromptandhistoricalanalysisofconsumptionoverfourperiodsinadaythat

Fig. 5.8b: Network analyser ANR 144-230

5

OV

Er

ViE

w O

f AB

B r

An

gE

34 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

Insulation

Insulation voltage 3700 V AC rms for 1 minute

Serial output

RS485

Programmable Baud rate [bps] 1.200 - 19.200

Communication protocols Modbus RTU, ASCII

Internal memory

For ANR96 and ANR144 [kbyte] 128 (usable 80)

For ANR96P [Mbyte] 1

Type of memory Permanent data memory using internal buffer battery

Data storage period 5 years at 25°C

Internal clock

RTC clock IEC EN 61038

Accuracy [ppm] 5

Digital outputs

Max setion of wires [mm² ] 0 ÷ 2.5

External pulse voltage [V] 12 ÷ 230 V AC/DC

Max. current [mA] 150

Digital inputs

Voltage [V] 12 - 24 DC

5.2.5

Temperature measuring unitsTheyareusedtomonitortemperaturelevelsandtheventilationfunctionsofelectricmachines,transformers,motors,etc.Preventivecontroloftemperatureenablesmal-functionsandoverloadingtobeavoided.PT100andRTDprobesareusedtomeasuretemperatures.Foreachmeasurementchannel, twoalarm levels (trippedalarms)canbeset that trip twooutput relays toremotelysignalifacriticaltemperaturelevelhasbeenreached.Therecordedvaluesandpossiblealarmstatusesaredisplayedonthe3-digitdoubledisplayonthefrontofthedevice,fromwhichtheadjustingfunctionsofthedevicescanbeaccessedviathe5programmingkeys.Inaddition,thecontrolunitsenablethemaximumvaluestobe stored, each intervention to be stored and ventilation inside the panel to becontrolled.Fig.5.9showsthepanelfrontcontrolunitTMD-T4/96

Main characteristics of the TMD-T4/96

Auxiliary supply voltage [V] 100 … 125, 220 … 240, 380 … 415/50-60 Hz

Max. consumption [VA] 4

Measuring inputs 2 RTD Pt100

Measuring range [°C] 0…+220 ±2 °C

Trip delay - hysterisis 5 s/2 °C

Measurement display LED display, 7 segments, digits

Outputs 1 12 V DC, 3 NA-C-NC relay, 8 A resistive load

Outlet functions alarm, trip, ventilation, autodiagnosis

Programmable functions ALARM, TRIP, HOLD, FAN, T. MAX

Connections extractable screw terminal boards, max. section 2.5 mm2

Insulation [Vrms] 2500/50 Hz - 1 min

Protection class

IP52 on front panelcan be increased to IP65 with optional

protection cap code 2CSG524000R2021

IP20 on back panel

Operating temperature [°C] -10...+55, max. humidity 90%

Storage temperature [°C] -25 ... +80

StandardsIEC EN 50081-2, IEC EN 50082-2,

IEC 14.1, IEC EN 60255

continues 5.2.4

Fig. 5.9 – TMD-T4/96 unit

5

OV

Er

ViE

w O

f AB

B r

An

gE

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 35

5.2.6

Electronic energy metersThewiderangeofDINrailABBenergymetersformeasuringenergy issummarisedinTable5.5.Foradescriptionofthespecifictechnicalcharacteristicsofeverysingledevice,seecatalogueSystemproM compact®.EveryelectricitymetermadebyABBistypeapproved,verifiedandissuedwithacertificateofcomplianceaccordingtoMIDrequire-ments,Measuring InstrumentsDirective.ABBhaschosen typeapprovalaccording toAnnexBandinitialverificationaccordingtoAnnexD.theproductmustfulfillrelevantpartsofEN50470andbeassessedbyaNotifiedBodythatthenissuesacertificateforthatproduct.ABBuseNMi,anindependentexpertformetrologytesting,certifyingandcali-brating,foranytask.Theenergycounterscanbeprofitablyusedinbothresidential/tertiaryenvironmentsandinindustrialenvironments.Atypicalexampleofthefirstcaseistheinteriorofashoppingcentrewherelocalenergyconsumptioncanbemeasured,arecordofconsumptioncanbeestablished,thebuildingcanbemanagedremotelyandbeintegratedwithamanage-mentsystemowingtodifferentprotocolschosenbythecustomerModbusRTU,M-bus,LonWorkandEthernet,GSM/GPRS,EIB/KNXthankstotheserialadapters.Asthecurrentdirectionisrecognisedautomatically,thecountersalsoensuresafeanderror-freeinstallation.Equallysignificantaretheinstallationadvantagesoftheenergymetersinindustrialplants,wherecertainspecificfeaturesofthedeviceareimmediatelyreflectedineconomicad-vantagesandreliability,asshowninTable5.6.

Single-phase energy meters Three-phase energy meters

ODINsingle DELTAsingle ODIN DELTAplus

Directmeasurementupto65A

Directmeasurementupto80A

Directmeasurementupto65AindirectbyCT(5/5-900/5A/A)

Directmeasurementupto80AindirectbyCT(1-999A)

Tab. 5.5 – DIN rail electronic energy meters

5

OV

Er

ViE

w O

f AB

B r

An

gE

36 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

5.3

Accessories for measuring instruments

5.3.1

Serial communication adaptersTheypermitserialcommunicationofdatabetweenenergycountersandtheremotesupervisionsystem;theyhavereduceddimensions (2DINmodules)andcaneasilybeinstalledontheDINsectionandcanbecoupledwiththeenergycounterasshowninFig.5.10.

Theirmainfunctionistoconvertopticalsignalscomingfromthecountersinthepow-erlineserialcommunicationmeans,pair,etcandintheselectedprotocols(ModbusRTU,Meter-bus,TCP/IP,KNX/EIB,GSM/GPRS).

Serial converter RS485 / RS232ThemultifunctionserialconverterCUS(fig.5.11)isusedinallcasesinwhichitisnecessarytoconvertormanagetheseriallinesEIA-232(RS-232),EIA-485(RS-485)andEIA-422(RS-422).Theconnectionbetweendevicesthatusethesetypesofcommunicationbus(suchasforexamplePLCs,measuringandcontrolinstruments,connectionbetweendevicesandcomputerswithspecificapplicationsoftwareinstalled,etc)oftenrequiresthetypeofseriallinetobeconverted,thesignalonthelinetobeamplied,differentpartsofthecommunicationnetworktobeinsulatedetc.TheCUSconverteristhususedwidely,asithasbroadapplicationalscopewithdifferentadjustmentsandsettingsthatenableittobeusedinthemostwidelyvaryingapplications.

CUSensuresthegalvanicallyinsulatedconversionbetweentheRS-232,theRS422-485sideandthesupplysource.Theversatilityofthedeviceallowsdifferentoperatingmodes:-RS-232conversiontoRS-422fullduplex-RS-232conversiontosingleRS-485halfduplex-RS-232conversiontodoubleRS-485halfduplex-RS-485repeater(andmonitorfunctiononRS-232)

Themainapplicationsare:-networksformultipointdatatransmission-long-distanceserialconnections-galvanicseparationofperipherals-extensionofRS-485lines

Fig. 5.10: Coupling energy meter-adapter

Fig. 5.11 – CUS serial converter

5

OV

Er

ViE

w O

f AB

B r

An

gE

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 37

Main characteristics of serial converter RS485 / RS232

Supply voltage [V] 230 V AC ±20%

Frequency [Hz] 50-60

Power input [VA] 7 max

Dissipated power [W] 3.5

Line fuse 500 mA internal

Dimension, supply terminals [mm2] 2.5 max

Dimension, RS485-422 terminals [mm2] 2.5 max

Connection RS232 Sub-D 9-pole female (DB9)

Length, max. RS232 line [m] 15

Length, max. RS485-422 line [m] 1200

Units that can be connected in multidrop mode Max 32

Operating temperature [°C] -20…+60

Storage temperature [°C] -20...+80

Modules [n°] 6

5.3.2

Current transformersTheyareusedtotransformprimarycurrents(max.6000A)intolow-currentsecond-arycurrents.../5Asupplyingindirectlyanalogueanddigitalmeasuringinstruments.Theyareavailablewithbothawoundwindingandathroughwinding.Inthefirstcasetheyaresuppliedtogetherwiththebarorprimaryterminal;inthesecondthehaveaholeinwhichtoinsertthebarorcablethatconstitutestheprimary.Therangeisverycomprehensive:foradescriptionofthespecifictechnicalcharac-teristicsofeverysingledevice,seethetechnicalcatalogueSystemproMcompact®.SystemForexample,infigure5.12threecurrenttransformerswithdifferentfeaturesareshown:1)DINrailtransformer.2)transformer,withwoundprimary,primarycurrentonbar,25mm,secondaryon

terminals;3)transformerwiththroughprimary: forprimarycurrentfromcable, fromhorizontal

barorfromverticalbar;

Current transformer range

DINrailtransformerTRFM Transformerwithwoundprimary

Transformerwiththroughprimary

Choice of primary

CT3 CT4 CT6 CT8 CT8-V CT12 CT12-V

Section conductor

[mm]

21 25 50 2x30 2x35 2x50 2x35

30x10 40x10 60x20 80x30 - 125x50 -

20x10 40x10 - - 3x80x5 - 4x125x5

continues 5.3.1

Fig. 5.12: Example of current transformers

5

OV

Er

ViE

w O

f AB

B r

An

gE

38 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

5.3.3

Voltage transformersTheyareusedtotransformprimaryvoltagesupto600Vintosecondaryvoltagesof.../100Vmaxwithwhichsupplyindirectlyallmeasuringinstruments,bothanalogueanddigital.Theyareavailableinthecasemadeofself-extinguishingrating1plastic(Fig.5.13a)orina0.5ratingmetalcase(Fig.5.13b).Theycanbeinstalledinthree-phasenet-works,withorwithoutneutral.Fortheselectionofthesingledevices,seethecata-loguefortheSystemproMcompact®.

Examples of voltage transformers

a)inplasticcase b)inmetalcase

5.3.4

Shunts for direct currentTheshuntshave60mVvoltageandmustbeusedwithamaximumloadof0.25ΩincombinationwiththeDCmeasuringinstrumentsformeasuringcurrent.Thetwo-polecablewithwhichtheyarefittedis1mlongandhasa1.4mm2sectionthatisequaltoaresistanceof0.025Ω.Toensurethattheshuntsoperatecorrectly,checkthat:-assemblycanbebothhorizontalorvertical(thehorizontalpositionenablestheheat

tobedispersedbetter);-thecontactsurfacemustbeusedcompletelyandbeclean;afterconnecting,cover

withaspecificgrease;-thescrewsandboltsmustbecorrectlytightened;-theshuntsmustbesufficientlyventilated;astheyarenot insulated,theymustbe

protectedagainstaccidentalcontacts.

Fig. 5.15: Method of insertion of shunt into measuring circuit

V

1 2 3 4

L1

N

A

1 2 3 4

L1

N

A

1 2 3 4

L1

N

S1 S2P1 P2

A

1 2 3 4

L1

N

INSTRUMENT

+ –

+

–G

+

–U

Fig. 5.13: Examples of voltage transformers

Fig. 5.14: Direct current shunt

5

OV

Er

ViE

w O

f AB

B r

An

gE

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 39

40 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

6The measurements

6.1

TRMS Measurements

6.1.1

Linear loadsWhentheelectricity isgeneratedbytheelectricpowerutility, thewaveformof thevoltageissinusoidal.Thetraditionalloadsare,forexample:-incandescentlampsandheaters(resistiveloads);-motorsandtransformers(inductiveloads),absorbsinusoidalcurrentiftheyarecon-

nectedtoasinusoidalvoltagesource.Thecurrentabsorbedbyapurelyresistiveorinductiveloadhasthesamepatternandthusthesamewaveformasthevoltagethatsuppliesit.Thus,inlinearloadsthewaveformof thecurrent is thesameasthevoltagewave form(botharesinusoidal)andtherearenoharmonics.

V

1 2 3 4

L1

N

A

1 2 3 4

L1

N

A

1 2 3 4

L1

N

S1 S2P1 P2

A

1 2 3 4

L1

N

1,5

1

0,5

0

-0,5

-1

-1,5

0 10 20 30 40 50 60 70 80

i(t)v(t) i(t) 90° delaycompared with

v(t)

inductive load

1,5

1

0,5

0

-0,5

-1

-1,5

0 10 20 30 40 50 60 70 80

i(t)v(t) i(t) 90°ahead of v(t)

capacitive load

V

1 2 3 4

L1

N

A

1 2 3 4

L1

N

A

1 2 3 4

L1

N

S1 S2P1 P2

A

1 2 3 4

L1

N

1,5

1

0,5

0

-0,5

-1

-1,5

0 10 20 30 40 50 60 70 80

i(t)v(t) i(t) 90° delaycompared with

v(t)

inductive load

1,5

1

0,5

0

-0,5

-1

-1,5

0 10 20 30 40 50 60 70 80

i(t)v(t) i(t) 90°ahead of v(t)

capacitive load

6.1.2

Non linear loadsTechnologyandtheneedtoreduceconsumption,whichisincreasinglyrequestedbythemarket,haveledtonewhigh-performanceloadsbeingdevelopedthatareabletooperatewithlessenergyabsorption.

Fig. 1: Sinusoidal linear voltage v(t) and current i(t) pattern

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 41

6

lE M

isu

rE

TheintroductionofsophisticatedcontrollogicsbymeansofstaticAC/DCconvertershasenabledalternating-currentmotorstobeobtainedthatprovidedynamicresponsesandperformancethataresimilartothoseofdirect-currentmotors.Thewaveformofthecurrentabsorbedbyadevicesuppliedbyaconverterisnotsi-nusoidalbutisanonsinusoidalalternatingperiodwaveshapewithanamplitudeandfrequencywithintheperiodthatistheequivalenttothesinusoid.Itswaveformisverydistortedcomparedwithasinusoidalwaveandforthisreason,whenaloadissuppliedbysuchatypeofsource,thesupplyissaidtobeanonlinearoradistorting load. Innon-linear loads theabsorbedcurrenthasadistortedwaveformthatdiffersfromthewaveformofthevoltageappliedtotheload.

i(t)v(t)

distorted i(t)

y(t)

0 T

t

Thefollowingareexamplesofnon-linearloads:-computers,printers,monitors;-UPS;-AC/DC,AC/ACstaticconverters;-inductionkilns;-electronicregulators;-switchingsupplyunits(alsoindomesticappliances):-illuminationsystemscontrolledbySCR/Triac;-variablespeeddrives;-X-raymachines;-magnetic-resonancemachines

6.1.3

Problems connected with TRMS measurementsMeasuringinstrumentscanbeoftwotypes:-instrumentsthatmeasuretheeffectivevalue(RMS)ofthequantity;-instrumentsthatmeasuretheeffectivevalue(TRMS)ofthequantity;Theinstrumentsthatmeasuretheeffectivevalueofthequantitiesassesstheaveragevalueoftherectifiedwavemultipliedbytheformfactor1.11(typicalofthesinusoidalwave),thustakinganapproximatemeasurementoftheeffectivevalueofthewave.Thevaluereadontheinstrumentisthus:readvalue=averagevaluexSinFFwhereSinFF=SinusoidalFormFactor,i.e.1.11Example:22.4Ax1.11=24.8ATheaveragevalueinthehalfcyclecanalsobeseenastheheightoftherectanglewithabasethatisthesameasthehalfcycleandhasthesameareaasthehalfwave.

i(t)v(t)

distorted i(t)

y(t)

0 T

t

continues 6.1.2

Fig. 3: Effective value of a sinusoidal signal

Fig. 2: Non sinusoidal pattern of a non-linear load

6

Th

E M

EA

su

rE

ME

nT

s

42 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

Theinstrumentsthatmeasurethetrueeffectivevalue(TRMS)ofthequantityperformthefollowingoperations:-samplingofthewaveovertheentireperiod;-squaringthesamples;-addingupthesquaresandmakingtheaverage;-thencalculatingthesquareroot.

YRMS=n

Yi

2

i=1

n

basic

200

100

0

-100

-200

7th HARM

5th HARM

resulting distorted wave

1 n/2 n

Theinstrumentsthatmeasureonlytheeffectivevalue(RMS)ofthequantitiesprovidethevaluecorrespondingtothetrueeffectivevalue(TRMS)onlywhentheymeasurequantitieswithaperfectlysinusoidalwaveform.Inordertohaveaccuratemeasurementsinthepresenceofdistortedwaves,anden-ablethepowertobedeterminedcorrectly,instrumentsthatareabletomeasurethetrueeffectivevalue(TRMS)ofthequantitiesmustalwaysbeused.

6.2

Harmonic distorsion and THDHarmonicsaresinusoidalwaveswithfrequenciesthatarethesameasentiremultiples(harmonicorder)ofthebasicwave.Atnetworkfrequency(50Hz),thedominantharmonicsgeneratedbythenonlinearloadsaretheunevenharmonics:-thethirdharmonic(150Hz);-thefifthharmonic(250Hz);-theseventhharmonic(350Hz)etc.

YRMS=n

Yi

2

i=1

n

basic

200

100

0

-100

-200

7th HARM

5th HARM

resulting distorted wave

1 n/2 n

Nonlinearloads,includingthoselistedbefore,aresourcesofcurrentharmonics.Whentheconcentrationof thesedevices increases inanelectricplant,theirinfluenceontheinternalelectricdistributionsystemalsoincreases.Whenthecurrentharmonicsreachasufficientamplitude,thereisthephenomenonofinteractionwiththeinternaldistributionsystemandwithotherdevicesinstalledinthesameplant.Thecurrentharmonicsinteractwiththeimpedanceofthedistributionsystem,creat-ingvoltagedistorsionsandenergyloss.Whentheharmonicdistorsionreachesexcessivelevels,thedevicesmayfacediffer-entproblems,inparticular:-untimelytrippingofthedifferentialrelays;-currentincreaseinthephaseconductors;-significantcurrentincreaseintheneutralconductorwithconsequentoverheating;-overheatingofthetransformersandincreaseinnoise;

continues 6.1.3

Fig. 4: True effective value of a sinusoidal signal

Fig. 5: Wave form with harmonic components

6

Th

E M

EA

su

rE

ME

nT

s

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 43

-increaseinthespeedofthediscintheinductionenergycounters;-prematureagingoftheelectricalcomponents;-faultsinthepower-factorcorrectioncapacitors;-faultsinthefiltercapacitorsandlittlestand-bypowerintheUPS;-reductionofpowerfactorandimpositionofpenaltiesbyelectricpowerproviderWhentheloadsarebalancedalsotheharmoniccurrents,likethephasecurrentsatthebasicfrequency(50Hz),tendtocanceleachotherout.Thisprincipleappliestoalltheharmonicswiththeexceptionofunevenharmonicsthataremultiplesofthree,which,unliketheothers,areaddeduptogetherandreturnex-clusivelythroughtheneutralconductor.Inelectricplantssuppliedbythree-phasesystems,star-connectednon-linearloadsthatgenerateharmonicsthataremultiplesofthreemaycausepossibleoverloadsandthusoverheatingoftheneutralconductors.Thevectordiagrambelowshowsthequantitiestrendforthebasicfrequency,forthefifthharmonicandforthethirdharmonic.Thefollowingtable,whichistakenfromanactualmeasurement,showshowthetotalneutralcurrent issubstantiallyequivalent tothesumof thethreephasecurrents inrelationtothethirdharmonic.

L2

L3

L1

L3

L2

L1

L1L2L3

5° HA 3° HAbasic

TRMS current measurements

Measurements with analyser

Line TRMS Line I fundamental I Third harmonic I Firfth harmonic

L1 143.5A L1 138.2A 35.5A 12.1A

L2 145.5A L2 140.7A 34.7A 11.6A

L3 147.8A L3 141.7A 39.6A 13.2A

Neutral 109.9A Neutral 10.6A 109.4A 3.1A

THDisthetotalharmonicdistortionofthebasicwave,whichconsidersthecontributionofallharmoniccomponentspresent.TheTHDisexpressedaspercentageofthebasicwaveandisagoodindicatorofthepresenceorabsenceofharmonics.THD(TotalHar-monicsDistortion)correspondstothetotalharmonicdistortionofthebasicwave,whichconsidersthecontributionofallharmoniccomponentspresent.Inotherwords,THDistheharmonicdistortionthatispresentinthemeasuredquantitycomparedwiththebasicwave.TheTHDvalueisexpressedasapercentageandisagoodindicatorofthepresenceofharmonics.TheIECEN50160standarddescribesthe'voltagecharacteristicsofelectricitysuppliedbypublicelectricitynetworks'.Article4.11'Harmonicvoltage'prescribesthatthetotalharmonicdistortionofthesupplyvoltage(includingallharmonicsuptothefortiethorder)mustbelessthanorthesameas8%.TheTHDindicationofcurrentharmonics,eveninpercentagesofafewunits,becomesanimportantindicatoroftheneedforathoroughharmonicanalysisinordertoidentifythepresenceofharmonics,suchasthethird,whichmaybepossiblecausesofmalfunc-tionsintheelectricplant.

continues 6.2

Table 1 Influence of the third harmonic on the neutral current

6

Th

E M

EA

su

rE

ME

nT

s

44 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

6.3

Cosfì (cosϕ) and power factor (PF)Thecosfi,ormoreexactlycosϕ,isthecosineofthephaseangleφbetweencurrentandvoltageinanalternatingcurrentelectricalsystem.In a purely resistive (also known as ohmic) system, the phase angle is nil, socosϕ =1.Inareal inductivesystem,i.e.withanon-nilresistivecomponent(e.g.anelectricmotor,asupplyunitforafluorescentlamp),thephaseangleisbetween0andπ/2(delayedphasedisplacement).Inasystemwithacapacitivecomponentthephasedisplacementisbetween0and-π/2(anticipatedphasedisplacement).Inbothcasesthevalueofcosϕisloweredbyoneuntilittheoreticallyreachesthevaluezero.Cosϕ isalsodefinedasthepowerfactorbecauseitistheequivalentoftheratiobe-tweenactivepowerandapparentpower.Acosϕwithaunitaryvaluemeansthattheapparentpowercorrespondstotheactivepowerandthatreactivepowerisnil.Inthepresenceofelectriclineswithaharmoniccontentwehavetotalkofapowerfactor(PF)astheeffectoftheharmonicsiscalculatedintheactivepower/apparentpowerratio.Reactivepowerisalwaysundesirable;acosϕvaluebecomesmoreun-desirablethefurtheritdeviatesfromone.Astheinductiveandcapacitivephasesoccurinoppositedirections,combiningthetwocomponentsinacircuitappropriately,by,forexample,addingcapacitorstoin-ductiveloadscanbedoneinsuchamannerthattheycanceloneanother,returningthecosϕneartoone.Cosϕisaparameterthatisrequiredtocalculatepower-factorcorrectionpower.

6.4

Practical tips for installing a good measuring system

Start from the need: what do I want to measure? A single electric parameter or all the electric parametersTherearedifferentproduct familieson themarket: instruments thatmeasurea singleelectricparameter(voltage,current,frequency,phaseanglecosϕ),generallyusedinsin-glephasesystems,asinstrumentationonthemachine,andinstrumentsthatenablealltheelectricparameterstobemeasuredanddisplayed,bothforthesinglephaseandinthethree-phasesystem.Thistypeofmultifunctioninstrumentisidealinpanelsinwhichspaceislimited,inpanelsofsubstationsandinmainindustrialpanels.Ifnotonlyelectricparametersneedtobemonitoredbutalsoenergyconsumptionneedstobechecked,measuring instrumentsthatalso includeanactiveandreactiveenergycounthavetobeselected.

Selecting the measuring system: single parameter, multifunction, analogue, digital instrumentTheinstrumentshouldbeselectedaccordingtothetypeofdistributionsystem.Inasingle-phasesystem,analogueanddigital instrumentsareselectedformeasuringvoltage,current,frequencyandthepowerfactor.Inathree-phasesysteminstrumentscanbeinstalledthatmeasurethesingleelectricpa-rameter,oneperphase,oravoltmeterandancurrentcanbeinstalledtogetherwiththevoltageandcurrentswitches,whichenable themeasurements tobedisplayed inse-quence,phasebyphase.Choosingananalogueinstrumentensuregoodreadingstability,duetothemechanicalinertiaoftheneedleandthefactthatthereaderimmediatelyknowswhethertheinstru-ment isworkingnormallyorwhetherthereadingisoff-scale.Theanalogueinstrumentindicatesthepointonthemeasuringscaleinwhichitfindsitself,showingtheupperandlowerlimits.Indigitalinstrumentsthisindicationisnotpossibleastheonlyreferenceisthereadingofthevalueonthedisplay,forexample,ofthecurrent.Somemeasuringinstrumentshavebarindicatorsthatshowthecurrentlevelasapercentageofthesetfullscale.Choosingadigitalinstrumentguaranteesbetterreadability,alsoinpoorlighting,speciallyforinstru-mentswithLEDdisplays,andanimmediatereactiontothemeasurementvariation.

6

Th

E M

EA

su

rE

ME

nT

s

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 45

Sizing the system, choosing the CTSizing themeasuringsystemstartswithknowing themainparametersof theplant; inparticular,startingfromthecharacteristicsoftheprotectionswitch,thetypeofdistribu-tionsystem,ratedcurrent,ratedvoltageandbartypecanbeknown.Afterthetypeofinstrumenthasbeendefinedthatismostsuitableforrequirements,ifthemeasurementisconductedthroughindirectinsertion,theaccessoriesofthemeasuringsystemsuchascurrentandvoltagetransformersmustbechosencarefully.Ifan800Acurrenthastobemeasured,inmostcasestheinstrumentcannotbecon-necteddirectlytotheline.Acurrenttransformerthatissuitablefortheapplicationmustthereforebeselected.Thechosenparametersofacurrenttransformerarenotonlyratedcurrent,secondarycurrentandpowerbutalsothetypeofassembly.FlexibleandstiffcablesorbarsforcarryingpowercanbeinstalledinapowerpanelThetransformerscanbeofdifferenttypes,dependingontheassemblysystem:athroughcableoracablewithawoundprimary,transformersforassemblyonhorizontalorverticalbars.

Cabling and wiring diagramsConnectinganalogueinstrumentsisverysimple;itinfactsufficestoconnectthephaseandneutralcablestotheinstrument's.terminal.Twocablesfotheauxiliarysupplymustalwaysbeconnectedfordigitalinstruments.Multifunction instrumentscanbeused indifferentdistributionsystems. In three-phasesystemswithdistributedneutralthreecurrenttransformersarerequired.Inthree-phasesystemswithoutdistibutedneutralinwhichtheloadsarebalancedandsymmetrical,anAroninsertioncanbecarriedout,i.e.twocurrenttransformersratherthanthreecanbeused;theinstrumentwillcalculatebydifferencethethirdphasethatisnotmeasureddi-rectly,consideringittobethesameastheothertwo.Inmultifunctioninstrumentsnotonlythecablesconnectedtothemeasurement,butalsotheRS485serialportandtheanalogueanddigitaloutputsandinputshavetobecabled.

Protecting the instrument and earthingInordertoensurethattheinstrumentisproperlyprotected,fusesmustalwaysbefittedtothesupplycablesofdigitalinstrumentsandtothevoltmetermeasuringinputs.EarthingthesecondariesoftheCTsensuresanearthconnectionifthetransformerde-velopsafaultanddoesnotaffectthemeasurement.Ifthereisagreatpotentialdifferencebetweenneutralandearth,thiscouldaffectthemeasurementnegatively,inthecaseofinstrumentswithmeasuringinputsthatarenotgalvanicallyinsulated.

Setting digital instrumentsBeforedigital instrumentsstartoperating theymustbesetwith theparametersof themeasuringsystemandthecommunicationparameters.ThemainmeasuringparametersarethetransformationratiosoftheCTsandoftheVTs,whicharedefinedasthemathematicalratiobetweenthenominalvalueandthevalueofthesecondary;forexample,settingthetransformationratioofanCTCT3/100withasec-ondaryat5AmeanssettingkCT=100:5=20.

Troubleshooting during final testingThemainproblemsthatariseduringthetestphasemaybeduetoincorrectinstallationoftheinstrumentsandtheaccessories.Alwayscheckthatthewiringcomplieswiththeinstructionmanual.Thefollowingerrorsarethosethataremostfrequentlycommittedwheninstallingameas-uringinstrument:-invertingthesecondariesoftheCTs-invertingthephasesofthecurrentandvoltagemeasuringinputs-failuretoeliminatetheshortcircuitofthesecondariesoftheCTs-settinganincorrecttransformerratio.

continues 6.4

46 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

Digitalcommunicationisadataexchange(inbinaryform,i.e.representedbybits)(1)

between“smart”electronicdevicesequippedwiththerelativecircuitsandinterfaces.Communication isnormallyserial, i.e. thebits thatconstituteamessageoradatapackagearetransmittedoneaftertheotheronthesametransmissionchannel(physi-calmeans).Thedevicesthathavetoexchangedataandinformationareconnectedtoeachotherinacommunicationnetwork.Anetworkgenerallyconsistsofnodesin-terconnectedtocommunicationlines:-thenode(a“smart”devicethat isabletocommunicatewithotherdevices) is the

datatransmissionand/orreceptionpoint;-thecommunicationlineistheelementthatconnectstwonodesandrepresentsthe

directpaththeinformationtakesinordertobetransferredbetweentwonodes;inpractice, it is thephysicalmeans (coaxialcable, twin-core telephonecable,opticfibres,infraredrays)alongwhichtheinformationanddatatravel.

0 receivingdevices

1 1 1 1 1 1 10000000

transmissiondevices signal element

(bit)

Themaincommunicationnetworkscanbeclassifiedasfollows:-Loopnetwork.Loopnetworksconsistofaseriesofnodes(representedbyPCsin

Fig.2),interconnectedsoastoformaclosedloop.

7Digital communication

Fig. 1: Bit sequence

Fig. 2: Loop network

(1)Abitisthebasicunitofinformationmanagedbyacomputerandcorrespondstothestatusofaphysicaldevice,whichisinterpretedas0or1.Acombinationofbitsmayindicateanalphabeticcharacteroranumericfigure,ormaygiveasignal,makeaswitchorperformanotherfunction.

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 47

7

lA C

OM

un

iCA

ziO

nE

Dig

iTAlE

-Starnetwork.Starnetworksarebasedonacentralnodetowhichalltheotherpe-ripheralnodesareconnected.

-Busnetwork.Thebusstructureisbasedonatransmittingmeans(usuallyatwistedcableorcoaxialcable)forallthenodesthatarethusparallel-connected.

Someexamplesofprocessmanagement inwhichcommunication isrequestedbe-tweenthedevicesinsertedinacommunicationnetworkare:

1)dataexchangebetweenthepersonalcomputersofafirmorcompanythatarecon-nectedtogetherinaLAN(2)network.

continues 7

Fig. 3: Star network

Fig. 4: Bus network

Fig. 5: Example of LAN network

(2)LAN(LocalAreaNetwork):localnetworks(e.g.Ethernet)thatlinkcomputersandterminalsthatarephysicallynearthatareforexamplelocatedinthesameofficeorinthesamebuilding.

7

Dig

iTAl C

OM

Mu

niC

AT

iOn

48 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

2)transceivingdataandcommandsbetweenasupervisionandcontrolsystemandthefielddevices(sensorsandactuators)ofanautomationsystemtomanageanin-dustrialprocess.

Acommunicationprotocolisrequiredtohandledatatrafficinthenetworkandallowthetwocommunicatingdevicestounderstandeachother.Thecommunicationpro-tocolisthesetofrulesandbehavioursthattwoentitieshavetofollowtoexchangeinformation;itisapreciseconventionassociatedwiththedataexchangedbetweenthecommunicationpartners.Thereareverymanydifferentprotocolsusedformakingdifferentdevicescommunicateinindustrialapplications.Theyvaryaccordingtothecommunicationneedsofeachapplication.Theseneedsmaybe:-amountofdatatobetransmitted;-numberofofdevicesinvolved;-thecharacteristicsoftheenvironmentinwhichthecommunicationoccurs;-timeconstraints;-whetherthedatatobetransmittedarecriticalornot;-possibilityornotofcorrectingtransmissionerrors;andstillothers.

ThereisthenafurtherwidevarietyofprotocolsusedtomakeITequipmentsuchascomputersand the relativeperipheralscommunicate.Weshallnotdealwith theseprotocolsbutshallmerelydescribetheprotocolsthatarededicatedtoindustrialcom-municationbetweenfielddevices,i.e.thedevicesthatinteractdirectlywiththephysi-calprocessthatistobekeptundercontrol.Inparticular,theconceptsofcommuni-cation, supervision and control will be applied to managing the electric plants fordistributinglowvoltageenergy.

continues 7

Fig. 6: Example of a supervision system for managing an industrial process Actuator Sensor Actuator Sensor

7

Dig

iTAl C

OM

Mu

niC

AT

iOn

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 49

7.1

Communication protocolsVerycomplexprotocolsarecurrentlyusedinindustrialcommunications.Inordertosimplifythedescription,theiroperatinglayersarenormallyseparated;eachprotocolhasaphysicallayer,adatalinkandanapplicationlayer.Eachlayerdescribesanop-eratingaspectofthecommunicationandinparticular:-thephysicallayerspecifiesthelinkbetweenthedifferentdevicesintermsofhard-

wareanddescribes theelectricsignalsused to transmit thebits fromone to theother;itdescribes,forexample,theelectricalconnectionsandthecablingmethods,thevoltageandcurrentsusedtorepresentthebits1and0andtheirduration.Inindustrialprotocols, thephysical layer isgenerallyoneof thestandard interfacessuchasRS-232,RS-485,RS-422etc;

-thelinklayerdescribeshowthebitsaregroupedbycharacterandhowthesearegroupedbypacketandhowerrorsaredetectedandcorrected.Ifnecessary,italsodefinestheshiftsorprioritiesthatthedeviceshavetofollowtoaccessthetransmis-sionmeans;

-theapplicationlayerdescribeswhatdatahavebeentransmittedandtheirmeaningfortheprocessbeingmonitored.Thisisthelayerinwhichthedataarespecifiedthathavetobecontainedinthepacketstransmittedandreceivedandhowtheyareused.

Generally,thelayersareindependentofoneanother;byapplyingtheconceptoflay-erstocommunicationbetweenpeoplewecanagreewhethertotalkoverthetelephoneorviaatransreceiver(physicallayer),whethertospeakEnglishorFrench(linklayer)andwhat the topic of conversationwill be (application layer). In order to establishcommunicationbetweentwoentitiessucessfully,allthelayersconsideredmustcor-respondtooneanother,i.e.forexample,ifweusethetelephonewecannottalktoapersonwhoisusingaradio;wecouldnotunderstandeachotherifweuseddifferentlanguages,etc.Withoutwantingtodescribeexistingprotocolsinacompletemanner,wewillneverthelessmentionsomefeaturesofthecommunicationsystemsthroughashortdescriptionofthethreelayersthathavejustbeenintroduced.

7.1.1

The physical layerAtthephysicallayerwehave:-wirelesssystemsthatuseasaphysicalmeansradiowaves,infraredraysorlumi-

noussignalsthatarefreelypropagatedinspace;-wired,orcabledsystemsinwhichthesignalsaretransmittedbycables(orpossibly

opticfibres).Thelatterare:-systemswithpointtopointcablinginwhicheachportionofcableconnectstwode-

vicesandisusedexclusivelyforcommunicationbetweenthem(aclassicexampleisthatofcommunicationbetweenaPCandaprinter).Thiscommunicationcanbeofthefullduplextypeifthetwodevicescantransmitsimultaneously,orbehalfdu-plexiftheycantransmitonlyalternately;

-systemswithmultipointcabling (alsoknownasmultidropcabling) inwhichmanydevicesshareinparallelthesamecommunicationcable(seeFig.8).Averyimpor-tanttypeofmultipointsystemarethosewithabusconnection,inwhichamaincablewithoutbranchesorwithveryshortbranchesconnectsallthedevicestogetherinparallel.

Device1

Device2

Device3

Device4

Branch(Stub)

Main cable(Backbone)

Fig. 8: Multidrop system with bus connection.

7

Dig

iTAl C

OM

Mu

niC

AT

iOn

50 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

Inindustrialnetworks,themostcommonlyusedphysical-layerinterfacesareRS-232forpoint-to-pointlinksandRS-485formultipointlinks.

The interfaces RS-232 and RS-485Atthephysicallayerwehave:The RS-232 interface,whichissocommoninpersonalcomputersthatitisknownasa'serialport'.Itisapoint-to-pointasynchronousserialcommunicationsystemthatcanoperateinfullduplex.

Wedescribeitscharacteristicssimply:-serialmeansthatthebitsaretransmittedoneaftertheother;-asynchronousmeans thateachdevice is free to transmitonecharacterata time

thatareseparatedbylongorshortintervals,asrequired;-point-by-pointmeansthatonlytwodevicescanbeconnectedtogetherinthisman-

ner.IfitisdesiredtousetheRS-232toconnectmorethantwodevices,eachpairmusthaveanindependentchannelavailable,withtwoportsdedicatedtoit.

-Full duplexmeans that thedevices can transmit and receive simultaneously. Fullduplexoperationispossiblebecausetwoelectricconnectionsexistthataresepa-ratedforthetwodirectionsinwhichthedatacantravel.

Thebitsaretransmittedintheformofvoltagelevelsfromthetransmissionterminal(Tx)ofadevicetothereceptionterminal(Rx)oftheotherdevice.Thevoltagesreferto an earth signal conductor (GND) connected to the GND terminal of the twodevices.

Rx1

Tx1

GND1Rx2

Tx2

GND2

Atleastthreewires(Tx,RxandGND)arerequiredfortheconnection:itispossibletouseextraconnections to regulate thedata flow (e.g. indicatingwhenadevice is ready totransmitandtoreceive);theseoperations,whichconstitutethehandshakingandflowcontrol(3)processes,willnotbediscussedhere.

continues 7.1.1

Fig. 11: Point-to-point connection between two PCs

Fig. 12: Basic connections for communication between two

devices with the RS-232 interface

(3)Flowcontrol:methodforcontrollingtheinformationflow.Handshaking:Exchangeofpresetsignalsbetweentwodevicesinordertoobtaincorrectcommunication.Withthisexchangeofsignalsthedevicescommunicatetohavedatatobetransmittedortobereadytoreceive.

Fig. 9: RS-232 9 pin serial connector

Fig. 10: RS-232 9 pin serial cable

Device port 1 Device port 2

RS-232

7

Dig

iTAl C

OM

Mu

niC

AT

iOn

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 51

continues 7.1.1

Eachcharacterthattransitsontheserialcableconsistsof:-oneormorestartbitsthatwarnthereceivingdeviceofthearrivalofanewcharacter

(astheinterfaceisasynchronousitisnotpossibleforthereceivingdevicetoknowwhenacharacterarrives,soitmustbewarnedbeforehand);

-acertainnumberofdatabits(forexample8);-apossibleparitybitthatrecogniseswhetheroneofthetransmittedbitsiswrong(in

thiscasetheentirecharacterisconsideredtobeinvalidandisrejected):ifthepar-itybitisuseditcanbeconfiguredinevenoroddmode;

-oneormorestopbitsthatconcludethetransmission.

Allthelistedbitshavethesameduration:theserialinterfaceisconfiguredfortrans-mittingacertainnumberofbitspersecond(bpsorbaud).Transmissionspeedsarestandardised,andtraditionallymultiplesof300bitsasecondareused.Forexample,adevicecould transmitat9600,19200or38400bauds,orbitspersecond.Inordertocommunicatecorrectly, thetwodeviceshavetousethesamesettings:baudrate(transmissionspeed),numberofdata,startandstopbits,useornonuseoftheparitybitand,ifitisused,themode(evenoruneven).Ifthisdoesnotoccur,nocharacterisrecognisedcorrectlyandsoitisimpossibletotransmitdata.

b0 b1 b2 b3 b4 b5 b6 b7

stopstart

1

0

Dispositivo1

DispositivoN

Dispositivo2

DispositivoN-1

R R

Data +

Data –Resistenza diterminazione

Forexample,inthestringofbitsshowninFig.13thefollowingcanbedefined:-astartbit;-8bits(b0….b7)thatmakeupthedatum;-astopbit.TheRS-485 interfacediffersfromtheRS-232throughitselectricalandconnectioncharacteristics.Itsmainadvantagesare:thepossibilitytomakemultidroplinks(4)orlinksbetweenmorethantwodevices(seeFig.14)andbetterimmunityfromelectricdisturbances.

b0 b1 b2 b3 b4 b5 b6 b7

stopstart

1

0

Device1

DeviceN

Device2

DeviceN-1

R R

Data +

Data –TerminationResistance

Thesecharacteristicshavemadetheinterfacemoreusedinindustrialenvironments,fromthefirstversionsofModbus(1960s)tothemoremodernModbusRTU,Profibus-DP,DeviceNet,CANopenandAs-Interface.IntheRS485,allthedevicesareconnectedinparallelonasinglebusformedbytwoconductorsknownas:Date+andDate-,orAandBoralsoData1andData2,depend-ingonthemanufacturersofthedevices.Thesignalsusedaredifferential; i.e. thebitsare thepotentialdifferencebetweeenData+andData-.Theconductorsaretwistedandkeptneartooneanothersothattheelectricaldisturbanceshitthemwithequalintensity,sothatthevoltagedifferenceisalteredaslittleaspossible.Whenadeviceisnottransmittingitgoesinto'recep-tion'mode,withgreatimpedanceonthecommunicationport.Thestandardspecifi-cationRS-485(EIA/TIA-485)(5)imposeslimitsontheinputimpedanceandprescribesthe current/power that each device has to be able to transfer to the line when ittransmits.

Fig. 14: Multidrop system with bus connection on RS-485

Fig. 13: Datum transmitted on 8 bits

(4)Inprinciple,inamultidropconnectionthedevicesareconnectedinparalleltoamaincable.

(5)EIA/TIA-485“DifferentialDataTransmissionSystemBasics”isthedocumentthatdescribesstandardRS485,towhichallthemanufacturersrefer.

7

Dig

iTAl C

OM

Mu

niC

AT

iOn

52 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

Inparticular,incompliancewiththeprescriptionsofthereferencestandard,correctdatatransmissionispossibleifupto32devices'inreceptionmode'areconnectedontheline.Thusinaccordancewiththestandard,RS-485ensuresthatthecommu-nicationcanoccurcorrectlywithamaximumnumberof32devicesconnectedtothebus;andineachcommunicationcycleadeviceisplacedin'transmissionmode'andtheother31areplacedin'receptionmode'.Infact,asallthedevicesareconnectedinparallelonasinglebus,onlyoneatatimecantransmit,becauseotherwisethesignalsoverlapandbecomeunrecognisable.The RS- 485 does not have any mechanism for defining the device authorised totransmit;thistaskisdelegatedtohigherlayersoftheprotocolused.Thestructureofeachtransmittedcharacter,itsdurationandthepossibilitiesofconfiguringtransmis-sionarethoseseenpreviouslyfortheRS-232serial;forexample,transmissionspeedsetat19200baud,with1startbit,1stopbitandausedparitybitcanbeobtained.Allthedevicesconnectedtothesamebuscanhavethesamesettingssoastobeabletocommunicatetogether.

7.1.2

The connection layerTheconnectinglevelusesmaster-slaveprotocolswhenoneofthedevices(themas-ter) has the task of controlling and managing the communication of all the others(slaves).Peer-to-peersystemsareusedwhenthishierarchydoesnotexistandthedevicesaccessthecommunicationmeansequally(inthiscasetheprotocolcomprisestheproceduresformanagingtheshiftsandtheaccessprioritytothecommunicationmeans;aprimeexampleisEthernet).

Themostwidelyusedcommunicationprotocolsare:-ModbusRTU,theconnectionprotocolmostwidelyusedwithelectronic-industrial

devices;-ProfiBus-DP,usedforfieldcommunicationwithsensorsandsmartactuators,gen-

erallyforfastandcyclicdataexchangesbetweenfieldequipmentandcontrollers;-DeviceNet,alsousedforinterfacingbetweenfielddevicesandcontrollers(PC,PLC);-AS-i,forcommunicationwithverysimplesensors,suchaslimitswitchesorcontrol

devices(e.g.pushbuttons).

7.1.3

The applicational layerTheapplicational layergivesameaning to the transmitteddata; i.e. itassociatesacommand(e.g.:open/closetheswitch)oranumber(e.g.voltagevalues)withthedatainbinaryformatthatthedevicesexchangethroughthecommunicationnetwork.Forexample,oftheModbusprotocolisusedtoreadthecurrentvaluesremotelythatarestoredinaDMTME-I-485multimeter.Themultimeterstoresthevaluesofthequantitiesandoftheparametersinsuitableregisters;theseregistersmayberead-onlyregisters(e.g.registersmeasuringthecur-rents)orreadingandwritingregisters(e.g.registersforsettingcurrenttransformersratio).Whenthemaster (e.g.aPC)wantstoreadthevaluesof thecurrents, itsendsthemultimeteramessagecontaining:-thenumberofregistersinwhichtoreadthedata(themeasuredquantitiesareas-

sociatedwiththeregisternumber)-thetypeofoperationtoberun(e.g.:readingthevaluescontainedintheregister).Theslave(inthiscasethemultimeter)respondsbysendingtherequestedvaluestothemaster.Thesevaluesarethenshowntotheoperatorinacomprehensibleformatthroughtheuser interfacesof thesoftwareandof thesupervisionapplicationprogrammesthat

continues 7.1.1

7

Dig

iTAl C

OM

Mu

niC

AT

iOn

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 53

facilitate presentation of the information of the data coming from the controlledprocess.Fig.15showsauserinterfaceoftheDMTME-SWsoftwarethroughwhichanoperatorcandisplaythevaluesofthecurrentsandalltheotherelectricparametersthatthemultimetermeasures.

7.1.4

Compatibility between levelsInindustrialcommunication,thedifferentdevicesthatexchangeinformationhavetousethesameprotocolsatallthelevelsinvolved.Forexample,themultimetersandtheABBnetworkanalysersusetheModbusRTUprotocolonRS-485.However, therealsoexist industrialdevices thatuseModbusRTUonRS-232orProfibus-DPonRS-485.

7.2

Supervision of electrical distribution plantsAlowvoltageelectricaldistributionplantcanbeconsideredtobeanindustrialproc-essfordistributingelectricenergyandassuchitalsorequiresasupervisionandcon-trolsysteminordertoincreaseitsreliabilityandoptimiseitsmanagement.Inordertointegratetraditionalplanttechnologyandcontrolsystemsinordertomanage,controlandmonitor inacentralisedandautomaticmannerdomesticand industrialplants,theelectricplantcanbeconsideredaffectedbytwoflows:-amainflow(energyflow)consistingofthepowerandtheenergythat,throughthe

lineconductorsandthecommandandprotectiondevices,issuppliedtotheserv-icesandtotheloadsofaplant;

-aninformationfloworaninformativeflow(digitalflow)consistingofalltheinforma-tion,thedataandthecommandsusedformonitoringandmanagingtheplant.

Thisisthesupervisionsystemtomanagethisinformativeflowthattravelsonthecom-municationnetwork.

Dependingontheextentandthecomplexityoftheplantstobemanaged,supervisionsystems with different architectures can be devised, from very simple architecture(two-levelarchitecture) tothemostcomplexarchitectures (multilevelarchitectures).Inthesimplesttwo-levelsystemthereare:

1)TheconnectionlayerconsistingoftheSCADA(SupervisoryControlandDataAc-quisition)system.system.Insimplerapplicationsthislevelcomprisesacomputeronwhichtheplant'sdataacquisition,controlorsupervisionsoftwareisinstalled.It isatthislevelthatthedatatransmittedbythesensorsareacquired,displayed

continues 7.1.3

Fig. 15: screenshot of reading software for DMTME-SW data of a series of multimeters.

7

Dig

iTAl C

OM

Mu

niC

AT

iOn

54 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

andprocessedandthecommandsaresent to theactuators. In thismanner,anoperatorcan, fromthesameposition,monitor thestatusof theentireplantandundertaketheoperationsrequiredtoensureefficiencyandcorrectoperation.Moreingeneral,inapplicationsinwhichtheelectricdistributionplantandthemanage-mentoftheprocessareintegratedthecontrollevelconsistsofthecomputersu-pervisingtheautomationsystemoftheentireindustrialprocess.

2)Thefieldlayerconsistingoffielddevicesprovidedwithacommunicationinterface(measuringinstruments,sensors,actuatorsandprotectionswitchesequippedwithelectroniccut-outmeans)installedintheelectricplantthatinteractdirectlywiththeplantandconnectitwiththecontrollevel.Themainfunctionsofthefieldlayerare:1)sendingtheplantdata(e.g.currents,voltages,energy,power,etc)tothecontrollayer2)actuating thecommands (e.g.opening/closureofswitches) received fromthecontrollayer

Thetwolayerscommunicatethroughabuscommunicationnetwork.Theinformation(e.g.measuredvalues)transmittedfromthefieldlayertothecontrolandcommandlevelthattravelsinanoppositedirection,constitutestheinformativeflowthattransitsonthebus.

supervision system for remote monitoring of electric parameters and energy consumption

mainpower panelANR144-230

motor 1 motor 2

Modbus RTU network

HVAC ANR96-24control

HVACsystem

monitoringof line 1

DMTME-I-485-96

productiveline 1

continues 7.2

Fig. 16: diagram of a supervision system with networked multimeters

an analysers.

Power flow

Info

rmat

ion

flow

Info

rmat

ion

flow

Info

rmat

ion

flow

Info

rmat

ion

flow

7

Dig

iTAl C

OM

Mu

niC

AT

iOn

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 55

7.3

The Modbus RS-485 network

7.3.1

Rules for correct cablingThecablingoftheindustrialcommunicationsystemsisdifferentinsomewaysfromthecablingusedforpowercablingandtheelectricianmayexperiencesomedifficultiesifheisnotanexpertinModbuscommunicationnetworks.AModbusRS-485connectsaMasterdevicetooneormoreSlavedevices.Henceforth,weshallconsiderSlavede-vicestobeABBmeasuringinstrumentswithserialcommunication,evenifthecablingissimilarforallModbusdevices.Themainrulesaresetoutbelowthatmustbefollowedforthecablingofthistypeofnetwork.1.Connection port Eachdevicehasacommunicationportwithtwoterminals,whichareindicatedforthesakeofconvenienceasAandB.Inthesetwoterminalsthecommunicationcableisconnectedsothatallthedevicesthattakepartinthecommunicationareconnectedinparallel.Allthe'A'terminalsmustbeconnectedtogetherandallthe'B'terminalsmustbeconnectedtogetherrespectively;invertingthe'A'and'B'connectionsofadevicedoesnotonlypreventitfromcommunicatingbutmayalsostoptheentirecom-municationsystemfromworkingowingtoincorrectdirect(polarisation)voltagefoundontheterminalsof the incorrectlyconnecteddevice. Inorder toavoiderrorswhenmanydevicesareconnected,cablesof thesamecolourshouldbeused forall theconnectionstotheterminalsAandcablesofthesamecolourshouldbeusedforalltheconnectionstotheterminalsBofthevariousdevices(e.g.whiteforAandblueforB);thismakesiteasiertoidentifycablingerrors.ThecommunicationportontheMasterdevice,whateveritis,hastwoterminalsthatcorrespondtoAandB.2.Connection between the devicesUnlikewhathappensinmanyenergydistributionsystems,themannerinwhichthedevicesareconnectedinparallelisimportant.TheRS-485systemusedforModbuscommuncationprovidesamaincable(Busorbackbone),towhichallthedeviceshavetobeconnectedwithbranches(alsoknownasstubs)thatareasshortaspossible.Thebranchesmustbenolongerthan1200m.Longerbranchescouldcausesignalreflections and generate disturbances and consequent errors in the reception ofdata.Fig.17showsanexampleofacorrectBusconnection.

Main cable/Backbone (Bus)

Imax= 1 mStub

3.Maximum distance and maximum number of devices.Themaincablemustbeno longer than700m.Thisdistancedoesnot include thebranches(whichmustneverthelessbeshort).Themaximumnumberofdevicesthatcanbeconnectedtoamaincableis32,includingtheMaster.

Fig. 17: Network with Bus structure

Fig. 18: Examples of incorrect Bus connections

7

Dig

iTAl C

OM

Mu

niC

AT

iOn

56 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

4.Use of repeatersInordertoincreasetheextentoftheModbusnetwork,repeaterscanbeused;andsignalamplifyingandregeneratingdevicesprovidedwith twocommunicationportsthattransfertoeachwhattheyreceivefromtheother.Usingarepeater,themainca-bleisdividedintodifferentsegments,eachofwhichcanbeupto700minlengthandconnect32devices (thisnumber includes the repeaters).Themaximumnumberofrepeatersthatshouldbeseriallyconnectedis3.Ahighernumberintroducesexces-sivedelaysinthecommunicationsystem.5.Type of cable to useThe cable to be used is a shielded twisted pair (telephone type). ABB specifies aBelden3105Acable,butdifferenttypesofcablewithequivalentcharacteristicscanbeused.Thetwinconsistsoftwoconductorsthataretwistedtogether.Thisarrange-mentimprovesimmunitytoelectromagneticdisturbancesbecausethecableformsaseriesofsuccessivecoils,eachofwhichfacesintheoppositedirectiontothenextone:inthismanneranymagneticfieldintheenvironmenttraverseseachpairofcoilsinoppositedirectionsanditseffectisthusveryreduced(theoretically,theeffectoneachcoil isexactlytheoppositeoftheeffectonthenextoneandthustheeffectiscancelled).Theshieldingmaybebraided (be formedbyameshof thinconductingwires)orbeafoil(consistingofasheetofmetalwoundaroundtheconductors):thetwotypesareequivalent.

sheath shield(foil-type)

twistedpair

earthingof shield

6.Connecting to the terminals Insomecountries,insertingtwocablesintothesamescrewterminalispermitted.Inthiscaseitispossibletoconnectthemaininletandoutletterminaldirectlytotheter-minalsofaninstrumentwithoutcreatingabranch.Ifontheotherhandeachterminalcanacceptonlyasinglecable,aproperbranchmustbecreatedusingthreeauxiliaryterminalsforeachinstrumenttobeconnected.7.Earth connection of the shieldThecableshieldmustbeearthedonlyinonepoint.Normally,thisconnectionismadeatoneendofthemaincable.8.Termination resistanceInordertoavoidsignalreflections,a120Ohmterminationresistancemustbefittedoneachendofthemaincable.Theendresistancemustbeusedonlyattheendsofthemaincable.Ifthetotallengthofthemaincableislessthan50mterminationre-sistancescanbeavoidedattheendsofthemaincable.

Fig. 19 : Detail of a shielded twisted pair.

Fig. 20 : 120 Ohm resistance connection

continues 7.3.1

7

Dig

iTAl C

OM

Mu

niC

AT

iOn

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 57

9.Connection to personal cmputerIfthemasterusedisapersonalcomputer,ingeneralanRS-232/RS-485serialconverterprovidestheconnectiontothebus.

7.3.2

The operation of the Modbus systemThetrafficoftheinformationonthebusismanagedbyaMaster/SlaveprocedurewiththePCorthePLCactingasaMasterandtheswitchesactingasSlaves.TheMasterdirectsallthetrafficofthebusandonlytheMastercaninitiatethecommunication.Ittransmitsdataand/orcommandstotheSlavesandasksthemtotransmitthedata.TheSlavestransmittothenetworkwhenrequestedtodosobytheMaster.TheSlavescan-notcommunicatedirectlywithoneanother:forexample,totransferadataitemfromoneSlavetoanothertheMastermustreadthedataitemfromthefirstSlaveandtrans-ferittothesecond.Thecommunicationsequencebetweeneachmultimeter(Slave)andthePC(Master)occursinthefollowingmanner:1)ThePCsendsacommandorquerytothebus.2)the interrogatedmultimeterprovidesa responseby taking theappropriateaction,

whichmaybe:-runningthereceivedcommand;-providingtherequesteddataor-informingitthattherequestcannotbemet.Thecommandortherequestcontainstheidentificationoftheinstrumenttowhichthecommunicationhasbeensentandtherefore,althoughthetransmissionhasbeenre-ceivedbyallthedevicesconnectedtothenetwork,normallyonlythedeviceconcernedwillrespond.TheswitchesareinterrogatedbythePCthroughcyclicalpolling,i.e.oneatatimecylicallysoastocompletelyscantheplantwithinasettime(pollingtime).Thecalculationofpollingtimeconsiderscomputerprocessingtime,tPC,tobenegligible,i.e.thetimethatelapsesbetweentheendoftheRESPONSEofaninstrumentandthestartofaQUERYthatthecomputersendstothenextinstrument.

WhatisneededtoimplementaModbusRTUsystemwithABBmeasuringinstrumentsandhowdoestheModbusprotocolactuallywork?

Whatisneeded:-master,whichmaybeaPCoraPLCoraSCADA-ifthemasterisaPCwithaserialinputportRS232,thenetworkofinstrumentsmust

beinterfacedwiththemasterbyaserialconverter232/485-connectioncablebetweentheconverterandPCthatmayhaveserialsocketsorUSB

inputs-shieldedtwistedpair(telephonetype),asdescribedinparagraph7.3.1-instrumentswithaserialportRS485,consistingofaterminalboardwith3terminals

ontheinstrumentmarkedABC.

InordertobeabletoimplementaModbusRTUcommunicationnetworkbetweensev-eralslavescomunicatinginModbusRTU,whethertheybemeasuringinstruments,pro-tectionswitches,ortemperaturecontrolunit, it isfundamentaltosetthesamecom-municationparametersonalltheobjectsinthenetwork.Thecommunicationparametersare:-datatransmissionspeed,knownasthebaudrate:from2400bpsto19200bps-databit:8-paritynumber:Even,Odd,None-stopbit1,2(ifparitynumber=none),1(ifparitynumber=even,oddornone)-addressforeachslave

Oncethesamebaudrate,paritynumberandstopbithasbeenset,andaftereachslavehasbeengivenitsownuniqueaddress,it ispossibletoacquireinformationfromthemaster.

continues 7.3.1

7

Dig

iTAl C

OM

Mu

niC

AT

iOn

58 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

Themastercommunicateswiththeslavesbysendingrequestsforinformationorqueries,andtheslavescommunicatewiththemasterbysendinganswers,knownasresponses,tothemaster.Theslavesareinterrogatedonebyonebythemaster,soifthenetworkisverycomplexintermsofthenumberofinstrumentsconnectedandthephysicaldistancebetweenthem,responsetimesincrease.TheModbusnetworkcanmanageupto247instruments.Themaximumdistancethatcanbecoveredis1200m;beyondthatdistanceasignalrepeatermustbeusedthatamplifiesthesignalandenablesdistancesthataregreaterthan1200mtobecovered.Themessagethatthemastersendstotheslaveisan8bitmessagewhereeachpartofthemessagehasameaning.Thefirstpartofthemessageisthephysicaladdressoftheslavetobeinterrogated.Subsequently,thefunctionisindicatedthatitisdesiredtoperform;thefunctionsaretypi-cally readingparameters,writingset-upsettings in the instrumentasa transformationratiooftheCTandoftheVT,andfunctionsforacquiringtherecordofthenetworkedproduct.Thecentralpartofthemessageindicateswhatandhowmuchinformationisrequired.Lastly,theclosurebitscheckthatthemessagehasarrivedandhasbeende-cypheredbythecorrectinstrument.

Theinformationthatthemasterrequestsfromameasuringinstrumentarethevaluesofthemeasuredandcalculatedelectricalparameters.Thelistofthesevaluesislocatedintheinstrumentonalist,eachparameterhasitspositioninsidethislist;thelistisknownasamemorymapandeachposition is indicatedasaregister,which iswhy it isalsoknownastheregistersmap.Thememorymapisthusthelistofalltheregisterscontain-ingtheparametersreadbytheinstrument.Thefollowingtableshowsthecorrespondencebetweentheaddressofeachposition,thelengthoftheresponsestring(2meansthattheslavewillrespondwithtwovalues,thefirstofwhichindicatestheparametermark),thedescriptionoftheelectricparameter,theunitofmeasurementandthebinaryformat.

Address Word Measurementdescription Unit Format1000h 2 3-PHASESYSTEMVOLTAGE Volt UnsignedLong1002h 2 PHASEVOLTAGEL1-N Volt UnsignedLong1004h 2 PHASEVOLTAGEL2-N Volt UnsignedLong1006h 2 PHASEVOLTAGEL3-N Volt UnsignedLong1008h 2 LINEVOLTAGEL1-2 Volt UnsignedLong100Ah 2 LINEVOLTAGEL2-3 Volt UnsignedLong100Ch 2 LINEVOLTAGEL3-1 Volt UnsignedLong100Eh 2 3-PHASESYSTEMCURRENT mA UnsignedLong1010h 2 LINECURRENTL1 mA UnsignedLong1012h 2 LINECURRENTL2 mA UnsignedLong1014h 2 LINECURRENTL3 mA UnsignedLong1016h 2 3-PHASESYS.POWERFACTOR *1000 SignedLong1018h 2 POWERFACTORL1i *1000 SignedLong101Ah 2 POWERFACTORL2i *1000 SignedLong101Ch 2 POWERFACTORL3i *1000 SignedLong101Eh 2 3-PHASESYSTEMCOSi *1000 SignedLong1020h 2 PHASECOS1i *1000 SignedLong1022h 2 PHASECOS2i *1000 SignedLong1024h 2 PHASECOS3i *1000 SignedLong1026h 2 3-PHASES.APPARENTPOWER VA UnsignedLong1028h 2 APPARENTPOWERL1 VA UnsignedLong102Ah 2 APPARENTPOWERL2 VA UnsignedLong102Ch 2 APPARENTPOWERL3 VA UnsignedLong102Eh 2 3-PHASESYS.ACTIVEPOWER Watt UnsignedLong1030h 2 ACTIVEPOWERL1 Watt UnsignedLong1032h 2 ACTIVEPOWERL2 Watt UnsignedLong1034h 2 ACTIVEPOWERL3 Watt UnsignedLong1036h 2 3-PHASES.REACTIVEPOWER VAr UnsignedLong1038h 2 REACTIVEPOWERL1 VAr UnsignedLong103Ah 2 REACTIVEPOWERL2 VAr UnsignedLong103Ch 2 REACTIVEPOWERL3 VAr UnsignedLong103Eh 2 3-PHASESYS.ACTIVEENERGY Wh*100 UnsignedLong1040h 2 3-PHASES.REACTIVEENERGY VArh*100 UnsignedLong1046h 2 FREQUENCY mHz UnsignedLong1060h 2 MAXLINECURRENTL1 mA UnsignedLong1062h 2 MAXLINECURRENTL2 mA UnsignedLong1064h 2 MAXLINECURRENTL3 mA UnsignedLong1066h 2 MAX3-PHASESYS.ACTIVEPOWER Watt UnsignedLong1068h 2 MAX3-PHASES.APPARENTPOWER VA UnsignedLong1070h 2 3-PHASESYS.ACTIVEPOWER15'AVER Watt UnsignedLong11A0h 2 CURRENTTRANSFORMRATIO(CT) 1-1250 UnsignedLong11A2h 2 VOLTAGETRANSFORMRATIO(VT) 1-500 UnsignedLong11A4h 2 PULSEENERGYWEIGHT 1-4ii UnsignedLong

Fig. 21 : Memory map registers map of DMTME multimeters

continues 7.3.2

7

Dig

iTAl C

OM

Mu

niC

AT

iOn

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 59

Forexample,ifIwanttoknowthevalueofthethree-phasevoltage,themasterwillhavetosendacommandconsistingof:-addressoftheinstrumentthatIwishtointerrogate(forexampleamultimeterlocated

onthemaincontrolpaneloftheplant)-readingfunction-addressoftheregisterofthe'three-phasevoltage'value-howmanyotherparametersiwanttoread,upto5-testingandcheckingthatthemessagehasreacheditsdestination

Thestringsentbythemasterhasthefollowingformat:

AddressField = 1FhFunctionCode = 03hStartAddressH = 10hStartAddressL = 00hNo.ofregisterH = 00hNo.OfregisterL = 14hCRCH = 42hCRCL = BBh

Intheaboveexample,themastersendsa03hreadingfunctiontotheslavewiththeaddress1Fh,startingfromtheparameterofthe1000hregisterfor14registers.

Theresponsesentbytheslavehasthefollowingformat:

AddressField = 1FhFunctionCode = 03hBytecount = 28hDataReg1000H = 10hDataReg1000L = EFh-------------------------------CRCH = XxhCRCL = Yyh

Analysingthemaponthe1000hregisterthereisthevoltageofthethree-phasesystem.Thusstartingfromthefirstregisterfor14registers,thePowerFactorvalueuptophase2isread.

Thevaluesoftheregistersinthememorymapareexpressedashexadecimalvalues.WhenusingfreewarereadingsoftwaredownloadedfromtheWebsuchasModbusPollorModbusContructorthatenablethedatareadbyamultimetertobeacquired,caremustbe takenoverwhether hexadecimal ordecimal valuesare requested from thesoftware.

Forexample,thethree-phasevoltagevalueisinthehexadecimalregister1000,whichbecome4096ifconvertedintodecimals.

Thememorymapissetbythemanufacturer,whodecideswhichregistershouldbeas-sociatedwiththeparameterreadbythemultimeter,andalsodecideswhetheralltheread and setting parameters of the instruments can be transmitted via serialcommunication.WhoneedstheregistersmapofaModbusRTUinstrument?ItisnormallytheSystemIntegratorwhohastoimplementthecommunicationnetworkviaPCorPLC.Thisisthepersonwhoimplementscommunicationbetweenthevariousdevicesconnectedtothebus.Theregistersmapisrequiredtoindicatetothemastertheaddressesinwhichelectricalquantitiesarefound.

continues 7.3.2

60 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

load

8Applicational examples of network analysers

Anapplicationalexampleisgivenbelow,withtherelativeinstructionsforsettinganduseoftheANRrangenetworkanalyser.Theapplicationintheexamplereferstoanindustrialplantortoatertiarysector(majorretailing)plantwithmixedlinearandnon-linearloads.TheANR144instrumentisinstalledinthelowvoltageMainPowerPanel.

Connectitinaccordancewiththeinstructionsthatfollow

Thesupplyfortheoperationoftheinstrumentcanbetakendirectlyfromthesupplyline(ANR144-230).

Fig. 1: Network analyser ANR144

Fig. 2: Wiring diagram for inserting ANR in three-phase network

with neutral. Insertion on 4-wire line with 3 CTs and 3 VTs

8

AP

PliC

AT

iOn

Al E

xA

MP

lEs

Of n

ET

wO

rk

An

Aly

sE

rs

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 61

continues 8

If theeventsdueto the interruptionof themainsupplyneedtobestoredanddis-played,theinstrumentmustbesuppliedwithacontinuitysubunit(UPS)ortheANR144-24modelmustbeusedthatenables20to60VDCandVACtobesuppliedalsofromgeneratorsthatareindependentofthemainsupply(e.gbatteries).

Oncetheinstrumenthasbeenconnected,thefollowingparameterscanbedisplayedandstoredandtheutilityofthedetectedvaluescanbeexamined:1.Ratedvoltage(phase/neutral)andlinkedvoltage(phase/phase)asatrueeffective

TRMSvalue;2.CurrentastrueeffectiveTRMSvalueonthethreephasesandonneutral;3.PowerfactorPF(cosϕ);4.Activepower5.Harmonicdistortionrate(HDR)upto31stharmonicdisplayedgraphicallyandasa

percentagevalue;6.Harmonicdistortionupto31stharmonicdisplayedgraphicallyandasapercentage

value;7.Activeenergyconsumedandgeneratedwithasubdivisionofthecountintototal

countsandaccordingtosettabletimesofday.

Fig. 2: Wiring diagram in three-phase network with neutral.

BatteriaAlimentatorec.a./c.c.

VOLTAGEINPUTS

CURRENTSINPUTS

Supply unitAC/DC

230 V AC input

outlet 12/24V DC

Battery

12/24V DC

230V AC

230V AC

8

AP

PliC

AT

iOn

Al E

xA

MP

lEs

Of n

ET

wO

rk

An

Aly

sE

rs

62 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

8.1

Rated voltage (phase/neutral) and linked voltage (phase/phase) as a true effective TRMS valueMeasuringvoltageisoneofthemainparametersforbeingabletoconductthenet-workanalysis.Itisalsousedtoassessthestateofbalanceofthevoltageonthethreephasesduringnormaloperationoftheplant.

8.2

Current as true effective TRMS value on the three phases and on neutralMeasuringcurrentsisoneofthemainparametersforbeingabletoconductthenet-workanalysis.Itisalsoimportantforcheckingthatloadsarecorrectlydistributedoverthethreephases.MeasuringtheneutralcurrentasatrueeffectiveTRMSvalueisimportantforestab-lishingwhethernon-linearloadsintroduceadistortionofthethirdharmonicasindi-catedinpoint6.2.-Iftheloadsarebalancedandtherearenoharmonicdistortions,thecurrentonthe

neutralconductorisvirtuallynil;-innormalconditions,withloadsthatarenotbalancedbutintheabsenceofharmon-

ics,theneutralcurrentismuchlessthanthephasecurrent;-inthepresenceofharmonicdistortionthethirdharmoniclinecurrentsareaddedto

neutralbecausetheyarephasedtogetherandaneutralcurrentoccursthatisgreaterthehigherthevalueofthethirdharmoniccurrents.

8.3

Power factor PF (cosϕ)Thepowerfactor,whichisbetterknownwiththetermcosϕ,mustbemaintainedatavaluethatisascloseaspossibleto1.MeasuringthepowerfactorPF is importanttoavoidhavingtopaypenaltiestothesupplieroftheelectricenergyforPFvaluesbelow0.9.AnalarmthresholdshouldbeinsertedforthismeasurementtowarntheuserifthePFgetsnearthe0.9value(e.g.alarmtrippedat0.92)

Fig. 5: Display of cosphi values

and power factor.

Fig. 4: Display of page dedicated to phase and neutral currents.

Fig. 3: Display of the parameters of the three-phase system.

8

AP

PliC

AT

iOn

Al E

xA

MP

lEs

Of n

ET

wO

rk

An

Aly

sE

rs

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 63

8.4

Active powerAsmentionedinchapters1.5and1.6,inordertoavoidpenalties,it isimportanttocontrolandmanageabsorptionpeakssoastoneverexceedtheaverageofavailablepower.Forcorrectmanagementandoptimisationofconsumption,theinstrumentcanbeprogrammedinsuchawaythat:-Consumptionisrecordedforanalysis,alsofortimesofday,inrelationtothesupply

contract;-Theloadsthatcanbedisconnectedwiththeleastgraveconsequencesbythein-

strumentshouldbesetfordisconnectionintheeventoftheavailablepowerthresh-oldbeingexceeded.

8.5

Total harmonic distortion rate (THD) up to 31st harmonic displayed graphically and as a percentage value DisplayingandstoringtheTHDenablesthetotalharmonicpercentageoftheloadsintheplanttobeassessedovertime.

8.6

Harmonic distortion up to 31st harmonic displayed graphically and as a percentage valueIftheprecedingmeasurementdetectsaharmoniccontentintheelectricplant,ananalysisoftheharmonicspresent,uptothe31stharmonic,canbeconducted,displayingthephe-nomenabothgraphicallyandasapercentagevalue.Whentheharmonicdistortionreacheshighlevels,differentproblemsmayarise,assetoutinpoint6.2.

8.7

Active energy consumed and generated with a subdivision of the count into total counts and according to settable times of day.Thisfunction isparticularlyuseful fortestingandthebalancebetweenenergycon-sumedfromthenetworkandenergyproducedinthecaseofself-generation.

Fig. 8: Display of harmonic analysis up to 31st order, numeric and graphic representation.

Fig. 6: Display of active power.

Fig. 7: Display of percentage THD values for voltages and currents.

64 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

9Annex

9.1

Measurements Glossary

Accessory Element.Groupofelementsordeviceassoci-atedwiththemeasuringcircuitofameasuringinstrument to supply characteristics specifiedforthemeasuringinstrument.

Amplitude of the measuring field Algebricdifferencebetween thevaluesof theupperlimitandofthelowerlimitofthemeasur-ingfield.Itisexpressedinunitsofthemeasuredquantity.

Measuring range (actual range) Field defined by two values of the measuredquantity, in which the margins of error of ameasuring instrument (and/or accessory) arespecified.

Auxiliary circuit Circuit,other thanameasuringcircuit, that isnecessaryfortheoperationoftheinstrument.

Measuring circuit (of an instrument) Partoftheelectriccircuitlocatedinsidethein-strumentanditsaccessories,togetherwithanyconnectioncords,suppliedbyvoltageoracur-rent,oneorboththesequantitiesbeinganes-sential factor for determining the measuredquantity(oneofthesequantitiescanbetheac-tualmeasuredquantity).

Accuracy rating Groupofmeasuringinstrumentsand/oracces-soriesthatmeetcertainmeasuringprescriptionsthataredesignedtokeeperrorsandvariationswithinthespecifiedlimits.

Reference conditions Appropriatesetofspecifiedvaluesandofspeci-fiedfieldsofvaluesof the influencequantitiesforwhich thepermissibleerrors foran instru-ment and/or for an accessory are specified.Each influencequantity canhavea referencevalueorareferencefield.

9

An

nE

x

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 65

continues 9.1

Measuring cord Cordcomprisingoneormoreconductors,whichhas been specially designed for connectingmeasuringinstrumentstoexternalcircuitsortoaccessories.

Shunt Resistorconnected inparallel toameasuringcircuitofameasuringinstrument.

Division Distance between any two consecutive seg-mentsofascale.

Error (absolute) Foraninstrument,avaluethatisobtainedbysubtracting the true value from the indicatedvalue. Foranaccessory,avaluethat isobtainedbysubtractingthetruevaluefromthe(expected)markedvalue. N.B.:

1 As the true value cannot be obtained through a measurement, a value is used in its place that is obtained in specified test conditions in a specified instant. This value derives from national measuring samples or from measur-ing samples agreed between the manufac-turer and the user.

2 Attention is drawn to the fact that an acces-sory error can be transformed into the oppo-site type of error when this accessory is as-sociated with an instrument.

Scale error Difference between the value indicated by ameasuringinstrumentandtheproportionalval-ueofthequantitymeasuredatdifferentpointsofthescaleaftertheinstrumenthasbeencali-bratedso that itdoesnothaveerrorsat twopoints.

Error (intrinsic) Errorofaninstrumentand/orofanaccessoryplacedinthereferenceconditions.

Phasemeter Instrumentthat indicatesthephaseanglebe-tweentwoelectricinputquantitiesofthesamefrequencyandwithasimilarwaveshape.

Thisinstrumentmeasures: -thephaseanglebetweenonevoltageandan-

othervoltageorbetweenonecurrentandan-othercurrent.

or -the phase angle between a voltage and a

current.

Distortion factor Ratio:(factor of total harmonic distortion effectivevalueofthenon-sinusoidalquantityof a quantity) effectivevalueoftheharmoniccontent

Peak factor Ratiobetweenthepeakvalueandtheeffectivevalueofaperiodicquantity.

9

An

nE

x

66 Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards

continues 9.1

Graduation Segmentsplacedonthedialtodividethescaleintoappropriate intervals soas toenable thepositionoftheindextobedetermined.

Index Component(means)thatisassociatedwiththescaleandindicatesthepositionofthemovableelementofaninstrument.

Class index Numberthatindicatestheaccuracyrating. Note: some instruments and/or accessories may have

more than one class index.

Length of the scale Lengthoftheline(cuvedorstraight)thatpassesthrough the middle points of all the shortestsegmentsofthegraduation,includingbetweenthefirstandthelastsegmentofthescale.Itisexpressedinunitsoflength.

Power factor meter Instrumentdesignedtomeasuretheratiobe-tweentheactivepowerandtheapparentpowerofanelectriccircuit.

Accuracy Forameasuringinstrument,itisthequalitythatcharacterisesthedegreeofproximitybetweentheindicatedvalueandthetruevalue.Foranaccessory,itisthequalitythatcharacterisesthedegreeofproximitybetweentheindicated(ex-pected)valueandthetruevalue.

Note: the accuracy of a measuring instrument and/or

of an accessory is defined by the intrinsic error and by the limits of the variations.

Dial Surfaceonwhichthescaleandotherinscrip-tionsandsymbolsarelocated.

Resistor (impedance) Resistor(impedance)seriallyconnectedtoan additional measuringcircuitofameasuringinstrument.

Scale Theentiregraduationandallthenumbersfromwhich,incombinationwiththeindex,thevalueofthemeasuredquantityisobtained.

Overshoot Difference(expressedasafractionofthelengthofthescale)betweenthemaximumtransitoryindicationandthepermanentindicationwhenthemeasuredquantitychangessuddenlyfromoneconstantvaluetoanotherconstantvalue.

Analogue display instrument Measuring instrument intended forpresentingordisplayingtheoutletinformationasacontinu-ousfunctionofthemeasuredquantity.

Note: an instrument in which a variation of the indica-

tion occurs by small discrete steps but which does not have a numeric display is considered to be an analogue instrument.

9

An

nE

x

Madetomeasure.Practicalguidetoelectricalmeasurementsinlowvoltageswitchboards 67

continues 9.1

Instrument with a response Instrument that in a specified frequency fieldas an effective value providesanindicationthathastobeproportional

totheeffectivevalueofthemeasuredquantity,evenwhenthemeasuredquantityisnotsinu-soidalorhasacontinuouscomponent.

Electric measuring instrument Measuringinstrumentdesignedtomeasureanelectricquantityoranon-electricquantityusingelectricmeans.

Electronic measuring instrument Measuringinstrumentdesignedtomeasureanelectricquantityoranon-electricquantityusingelectronicmeans.

Direct-action indicating instrument Instrumentinwhichtheindicatingdeviceiscon-nected mechanically to the movable elementandisdrivenbytheelement.

Ripple rate (content) Ratio:(of a quantity) effectivevalueofthecontinuouscomponent effectivevalueoftheripplecomponent

Response time Timerequiredfortheindicationtofirstenterandthenremainwithinarangecentredonthefinalpermanentindicationwhenthemeasuredquan-tityvariessuddenlyfromzero(correspondingtonon-suppliedstatus)toavaluethatissuchthatthe final permanent indication is a specifiedpointonthescale.

Assigned value Valueofaquantityfixed,generallybythemanu-facturer,foraspecifiedoperatingcondition.

Conventional value Clearlyspecifiedvalueofaquantitytowhichtheerrorsofaninstrumentand/orofanaccessoryrefer,inordertodefinetheirrespectivedegreeofaccuracy.

Rated value Valueofaquantitythatindicatestheenvisageduseofaninstrumentoranaccessory.

Theprescribedcharacteristicsoftheinstrumentsandtheaccessoriesarealsonominalvalues.

Graduation zero Segmentofthedialassociatedwiththenumberzero.

Contact us

System pro M compact®

DIN Rail components for low voltage installation

Technical catalogue - Edition 2011

Contact usContact us

2CS

C40

0002

D02

09 -

Jan

uary

201

1 -

14.0

00ABB SACE a division of ABB S.p.A.ABB STOTZ-KONTAKT GmbH

In consideration of modifications to Standards and materials, the characteristics and overall dimensions indicated in this catalogue may be considered binding only following confirmation by ABB

Copyright 2011 ABB. All rights reserved.

System

pro

M co

mp

act® D

IN R

ail components for low

voltage installation

2CSC400002D0209

Ifyourequirefurtherdetails,youcanorderthetechnicalliteratureillustratedheredirectlyon-linefromthehomepageofourwebsitehttp://www.abb.com/abblibrary/DownloadCenter/inthe“Orderdocumentationandtechnicalsoftware”section.

System pro M compact®

Mad

e to measure. P

ractical guide to electrical m

easurements in low

voltage switchb

oards

(a)

(b)V

t

V

t

(f)50 Hz

A

050

100

150

200250 500

A

050

100

150

200250 500

V

0

8060

40

20

Made to measure.Practical guide to electrical measurements in low voltage switchboards

2CS

C44

5012

D02

01 -

03/

2011

- 1

.500

- C

AL.

Contact us

ABB SACEUna divisione di ABB S.p.A.Apparecchi ModulariViale dell’Industria, 1820010 Vittuone (MI)Tel.: 02 9034 1Fax: 02 9034 7609

bol.it.abb.comwww.abb.com

The data and illustrations are not binding. We reserve the right to modify the contents of this document on the basis of technical development of the products,without prior notice.

Copyright 2011 ABB. All rights reserved.