Power Influence On Telecom Systems - Jacobson Engineering

EE530 Power QualityPower Influence on TelecomSystemsEE530 Power QualityKevin Jacobsonkevin@jacobsonengineering.caFebruary 7, 2005www.jacobsonengineering.ca1EE530 Power QualityIntroduction Power system harmonics have a detrimental effecton power system operation: power quality, efficiencyof machine operation, premature failure ofcomponents, etc. Not many people are aware of the detrimental effecton telecom systems: noise, failures, serviceinterruptions, etc. Often power quality issues affect telecom systemsbefore power system people are aware that there is aproblem This talk will focus on power influence - the effect ofutility system harmonics on telecom systemsFebruary 7, 2005www.jacobsonengineering.ca21

EE530 Power QualityOutline Overview of power influence due to harmonics– How it works– How it’s measured– Standards Noise mitigation on the telecom system– Shielding– Twisted pairs– Noise filtering devices Real world noise caseFebruary 7, 2005www.jacobsonengineering.ca3EE530 Power QualityWhy Does Interference Occur? Power systems transmit very high energy– MW, kV, kA Telecom systems transmit much smaller signals– mW, µW, (lowest audible noise is 1 pW 10-12 W) MW (106 W) to µW (10-6 W) – ratio of 1012 Although telecom systems are designed with interference inmind and can reject a lot of it, small amounts of spillage of MWcan still have a significant effect Power and telecom cables are often close together, and runparallel for long distances (utility side: many kilometers), whichmeans electromagnetic couplingFebruary 7, 2005www.jacobsonengineering.ca42

EE530 Power QualityAerial Joint UseCable TVTelephone cablePrimary power (13kV)Secondary power (120V/240V)February 7, 2005www.jacobsonengineering.ca5EE530 Power QualityBuried Telephone CableFebruary 7, 2005www.jacobsonengineering.ca63

EE530 Power QualityTypes of Induction ProblemsInduction60 Hz Voltage/CurrentDamage(fault)HarmonicsSystem disruption(fault, transient, steadystate)Noise(steady state, transient)Safety(fault, transient, steadystate)February 7, 2005www.jacobsonengineering.ca7EE530 Power Qualityihineutral,hiphase,hiphase,hPower SubLoadineutral,hMagnetic field(60 Hz and harmonics)Telecom cableInduced noise voltage: vh Zm ihFebruary 7, 2005www.jacobsonengineering.ca84

EE530 Power QualityMutual Impedance Carson’s Equation (Bell System Technical Journal,1926) jωµ0 d12'Zm ln 2 j 2π d120[](u2 j ) u e u α (h1 h2 ) cos(uαs )du Exact expression, derived from electromagnetic fieldtheory, but too complicated in practicewww.jacobsonengineering.caFebruary 7, 20059EE530 Power QualityMutual Impedance Simplified approximation of Carson’s Equation(Parker, Bell Labs, 1980 and IEEE Std 367-1996)4 jωµ0 D1 1 2 Zm Rm jXm ln (Ω/km)2π D0 12 γD1 D0 s (h1 h2 )222 2 D1 s h1 h2 γ 2γ j ωµ0 σFebruary 7, 2005www.jacobsonengineering.ca105

EE530 Power QualityMutual Impedance100 ohm*m soil resistivity, 7.62 m height difference, 13.2 m separationFebruary 7, 2005www.jacobsonengineering.ca11EE530 Power QualityHow Much Interference?The severity of a power induction problem depends on: Influence – “How bad is the power line?”– Conditions on the power line: balance, harmonic content,earth return current, phases, grounding, impedance ofneutral– Limits defined in ECUC, IEEE 519, IEEE 776, IEC 61000,CSA C22.3 No. 3-98, CEA Electrical Coordination Susceptibility – “How bad is the phone system?”– Conditions on the phone system: cable balance, sheathintegrity, equipment, grounding– Limits defined in IEEE, CSA C22.3 No. 3-98, CEA ElectricalCoordination, Telcordia specs Coupling – “Geometry”– Separation, length of parallel, earth resistivityFebruary 7, 2005www.jacobsonengineering.ca126

EE530 Power QualityTIF Quantifies the Influence Telephone Interference Factor (TIF) is a key quantityused in the measure of a power influence problem Each harmonic is assigned a relative interferingweight Wf (or Wh) which is determined by threefactors– The bandpass filter used in telephone circuits. Theintelligible portion of speech is approximately 300 Hz to 3000Hz, so the remaining is filtered out.– “Annoyance” factor: how the human ear responds to noise atdifferent frequencies. This varies from person to person, soan average value is used.– Frequency: vf Zm(f) ifwww.jacobsonengineering.caFebruary 7, 200513EE530 Power QualityTelephone Influence Factor Curve10000TIF(W f)5000010002000300040005000F r e q u e n c y in H e r tzFebruary 7, 2005www.jacobsonengineering.ca147

EE530 Power QualityTIF Weighting 62069705533008090* 0 Sequencewww.jacobsonengineering.caFebruary 7, 200515EE530 Power QualityI*T and kV*T Products Analysis of individual harmonics is important, but it isalso useful to define an aggregate quantity to sum itall up:I T (i W )2hhkV T h (v W )2hhh Units: weighted amperes or weighted kilovolts CSA limit on probe wire I*T is 500 AFebruary 7, 2005www.jacobsonengineering.ca168

EE530 Power QualityDifferent I*T Measurements I*T is most commonly used, and has a few differentinterpretations depending on which document youread (some are not very clear)– Balanced I*T: includes only positive and negative sequenceharmonics– Unbalanced I*T: includes only zero sequence harmonics– Residual I*T: includes all harmonics, vector summed overthe three phases (some of these currents may flow on theneutral)– Ground return I*T: includes all harmonics, vector summedover the three phases and the neutral– Probe wire I*T: the total harmonic content as seen by a 100foot test probe wire placed parallel to the power lineFebruary 7, 2005www.jacobsonengineering.ca17EE530 Power QualityDifferent I*T Measurements Balanced, unbalanced and residual I*T are measuredon the power system, and can be done on either theprimary or secondary side (need to consider turnsratio of transformer) Probe wire I*T is the best representation of what willbe induced on a parallel telephone cable Ground return I*T on the primary can be related toprobe wire I*T If there is no neutral on the primary (rural powersystems) residual and ground return I*T are the sameFebruary 7, 2005www.jacobsonengineering.ca189

EE530 Power QualityProbe Wire MeasurementPower SourceIphaseLoadIneutral100 foot wireTest setTest set measures individual harmonics and gives an I*T readingwww.jacobsonengineering.caFebruary 7, 200519EE530 Power QualityLoop Coil MeasurementPower SourceIphaseLoadIneutralLoop coil - mount of roof of vehicleTest setTest set measures individual harmonics - relative strengthCan track a single harmonic with an audible Geiger modeFebruary 7, 2005www.jacobsonengineering.ca2010

EE530 Power QualityToolsFebruary 7, 2005www.jacobsonengineering.ca21EE530 Power QualityToolsGround pegsLoop coil100 ft. wireFebruary 7, 2005www.jacobsonengineering.ca2211

EE530 Power QualityStandards and Legislation Inductive Coordination–––– Power System–––– CSA C22.3 No. 3 Electrical CoordinationCEA Electrical Coordination GuideIEEE 776-1992 Recommended Practice for Inductive Coordination of Electric Supplyand Communication LinesIEEE 1137-1991 Guide for the Implementation for Inductive Coordination MitigationTechniquesIEEE 519-1992 Recommended Practices and Requirements for Harmonic Control inElectric Power SystemsIEEE 367-1996 Recommended Practice for Determining the Electric Power StationGround Potential Rise and Induced Voltage From a Power FaultIEC 61000 Electromagnetic CompatibilityAlberta Electrical and Communication Utility CodeTelecom System––––IEEE 820-1984 Standard Telephone Loop Performance CharacteristicsIEEE 455-1985 Standard Test Procedure for Measuring Longitudinal Balance of .Telcordia PracticesAlberta Electrical and Communication Utility CodeFebruary 7, 2005www.jacobsonengineering.ca23EE530 Power QualityIndividual Harmonic LimitsMeasured directly on power conductorsat load(Isc/IL 20, 120V through 69kV)Voltage measured on 100 ft. probe wire interfaceplaced 50 ft. from power line primary(limit for 4.752 km to 15.240 km parallel)Current on primary:calculated using mutual impedanceFebruary 7, 2005www.jacobsonengineering.ca2412

EE530 Power QualityOutline Overview of power influence due to harmonics– How it works– How it’s measured– Standards Noise mitigation on the telecom system– Shielding– Twisted pairs– Noise filtering devices Real world noise caseFebruary 7, 2005www.jacobsonengineering.ca25EE530 Power QualityShielding Many think that because the telecom wires areencased inside a grounded metallic shield that thewires are shielded completely from noise - not true! Encasement inside a metallic structure forms a“Faraday cage” which blocks electric fields Power influence due to induction is a magnetic fieldphenomenon A Faraday cage does nothing to block magnetic fields Magnetic shielding occurs due to a differentphenomenonFebruary 7, 2005www.jacobsonengineering.ca2613

EE530 Power QualityMagnetic ShieldingPhase CurrentPower sourceLoadMagnetic fieldCurrent induced in pair by phase currentCentral OfficeCurrent induced in pair by sheath currentCable pairCable sheathCurrent induced in sheath by power lineFebruary 7, 2005www.jacobsonengineering.ca27EE530 Power QualityMagnetic Shielding Current induced in grounded shield induces alongitudinal current in the cable pair which reducesthe current induced by the power line Magnetic shielding occurs due to any parallelgrounded conductor Effectiveness of shielding depends on frequency,resistance of conductor, resistance of grounds, sizeof cable, length of cable, and phases of individualharmonic components There is almost no shielding at 60 Hz, shieldingincreases with frequencyFebruary 7, 2005www.jacobsonengineering.ca2814

EE530 Power QualityTwisted Pair A significant reduction in noise is accomplished byusing twisted pair cable Impedances of the two conductors of a pair (called tipand ring) must be very close (limited by practicalmanufacturing processes) - balanced pair Any equipment connected to the pair must also bewell balanced otherwise the noise immunity is ruinedwww.jacobsonengineering.caFebruary 7, 200529EE530 Power QualityIdeal CableMagnetic fieldIring- 0Central OfficeItip If the impedances of the two conductors of a pair areexactly the same throughout the cable length, thenthe currents (called longitudinal currents) inducedwill be exactly the same The longitudinal currents will then cancel out in theterminating circuit, and there will be no noiseregardless of the amount of induction presentFebruary 7, 2005www.jacobsonengineering.ca3015

EE530 Power QualityReal World CableMagnetic fieldIring- Central OfficeItip The impedances of the two conductors are neverexactly the same throughout the cable length, so thelongitudinal currents are not the same The longitudinal currents do not cancel, resulting in adifferential noise current which will affect thequality of serviceFebruary 7, 2005www.jacobsonengineering.ca31EE530 Power QualityCable Balance Standards say balance should be greater than 60 dB reduce induced voltage or current by a factor of1000 or more Impedance difference: R, j2πf L, 1/(j2πf)(1/CT-1/CR) Balance is frequency dependent: ability to rejectnoise gets worse with frequencyFebruary 7, 2005www.jacobsonengineering.ca3216

EE530 Power QualityNoise Filtering Devices Noise choke Induction neutralizing transformer (INT) or inductioncontrol transformer (ICT) Harmonic drainage reactor (HDR) Transformer excited network (TEN) Single Noise Interfering Exterminator (SNIX)February 7, 2005www.jacobsonengineering.ca33EE530 Power QualityNoise ChokeFebruary 7, 2005www.jacobsonengineering.ca3417

EE530 Power QualityInduction Neutralizing TransformerFebruary 7, 2005www.jacobsonengineering.ca35EE530 Power QualityHDR and TENFebruary 7, 2005www.jacobsonengineering.ca3618