common-power-quality-factors-affecting-transformers_an.pdf

8/10/2019 common-power-quality-factors-affecting-transformers_an.pdf

1/6

Application Note

Commercial buildings commonlyhave a 208/120 V transformer ina delta-wye configuration tofeed receptacles. Single-phase,non-linear loads connected tothe receptacles produce triplenharmonics, which add up in the

neutral. When this neutral cur-rent reaches the transformer, it isreflected into the delta primarywinding where it causes over-heating and transformer failures.

Another transformer problemresults from core loss and copperloss. Transformers are normallyrated for a 60 Hz phase currentload only. Higher frequency har-monic currents cause increasedcore loss due to eddy currentsand hysteresis, resulting in moreheating than would occur at the

same 60 Hz current.Transformers supplying non-linear loads should be checkedperiodically to verify operation

within acceptable limits. Trans-formers are also critical to theintegrity of the grounding system.

Factors

1. Transformer loading (kVA)

Start by measuring kVA anddetermine wether the transformerload is balanced. Connect voltage probes on

Phase 1 and Neutral andclamp current probe on samephase. Repeat for Phase 2 and3.

Use a single phase powerquality analyzer to read kVA ofeach phase and sum all threefor total transformer kVA.

Or, connect all four currentclamps and all five test leadsfor the three phase powerquality analyzer to read kVAfor each phase and the total.

F r o m t h e F l u k e D i g i t a l L i b r a r y @ w w w . f l u k e . c o m / l i b r a r y

Common powerquality factors

affecting transformers

Unbalanced load:kVATOTAL= kVA1+ kVA2 + kVA3

kVA1Red

Red

Red

Black Black

Black

kVA2

kVA3

1

2

3

N

Compare actual load kVA tonameplate kVA rating to deter-mine % loading.

When using a single phaseanalyzer on a balanced load, asingle measurement is sufficient.Transformers loaded at less than

50 % are generally safe fromoverheating. However, as loadsincrease, measurements shouldbe made periodically. At somepoint the transformer may requirederating.

2. Harmonic spectrumThe harmonic spectrum of thesecondary (load) current will giveus an idea of the harmonic ordersand amplitudes present:

In a transformer feeding sin-gle-phase loads, the principalharmonic of concern is the3rd. The 3rd will add arith-metically in the neutral andcirculate in the delta primaryof a delta-wye transformer.The good news is that thedelta-wye tends to isolate therest of the system from the 3rd(though not the 5th, 7th orother non-triplen harmonics).The bad news is that thetransformer pays the pricewith additional heat.

Figure 1. Measuring transformer load (unbalanced) using a single phase power quality analyzer.

Figure 2. Harmonic spectrum.


2/6

2 Fluke Corporation Common power quality factors affecting transformers

In a transformer feeding three-phase loads which includedrives or UPS systems with 6-pulse converters, the 5th and7th harmonic will tend to pre-dominate. Excessive 5th is ofparticular concern because it isnegative sequence. It will tendto produce counter-torque andoverheating in polyphasemotors.

Harmonic amplitudes normallydecrease as the frequencygoes up. If one frequency issignificantly higher in ampli-tude than lower frequencies,we can suspect a resonant

condition at that frequency. If

such a condition is detected,be sure to take readings atcapacitor banks to see if thecaps are experiencing overcur-rent/overvoltage conditions.

Before-and-after harmonicspectrum measurement isextremely valuable to deter-mine if harmonic mitigationtechniques, like trap filters,which are tuned to specificfrequencies, are sized properlyand are working as expected.

Different harmonic frequenciesaffect equipment in differentways (see below).

3. Total Harmonic DistortionCheck for THD of both voltageand current: For voltage, THD should not

exceed 5 %

For current, THD should notexceed 5-20 % (see OddHarmonics table)

IEEE 519 sets limits for har-monics at the PCC (Point ofCommon Coupling) between theutility and customer (EN50160 isthe European standard). IEEE 519is based on THD measurementstaken at the PCC. Technically, thePCC is the primary of the utilitysupply transformer (althoughthere are cases where the PCC isat the secondary if the secondary

feeds a number of customers). Inpractice, these measurements areoften made at the secondary ofthe customers main transformer,since that is the point most easilyaccessible to all parties (and alsosince that is generally a LowVoltage measurement).

Some PQ practitioners havebroadened the concept of PCC toinclude points inside the facility,such as on the feeder system,where harmonic currents beinggenerated from one set of loads

could affect another set of loadsby causing significant voltagedistortion. The emphasis is onimproving in-plant PQ, ratherthan on simply not affecting util-ity PQ.

3a. Voltage THDTHD has a long history in theindustry. The underlying conceptis that harmonic currents gener-ated by loads will cause voltagedistortion (E=IZ) as they travelthrough the system impedance.This voltage distortion then

becomes the carrier of harmonicssystem-wide: if, for example, thedistorted voltage serves a linearload like a motor, it will then cre-ate harmonic currents in thatlinear load. By setting maximumlimits for voltage distortion, weset limits for the system-wideimpact of harmonics.

Table 1: Measurements at the distribution transformer

Measurement Look for

1. kVA Transformer loading. If loading exceeds 50 %, check for harmonicsand possible need for derating.

2. Harmonic Harmonic orders/amplitudes present:

spectrum 3rd harmonic (single-phase loads)5th, 7th (primarily three-phase loads) Resonance of higher order harmonics Effectiveness of harmonic trap filters

3. THD Harmonic loading within limits:Voltage %THD < 5 %Current %THD < 5-20 % (Table 2)

4. K-factor Heating effect on transformer from harmonic loads

5. Ground currents Objectionable ground currents are not quantified but areprohibited by the NEC

Neutral-ground bond in place ESG (Electrical Safety Ground) connector to ground electrode

(typically building steel) in place

Harmonic Sequences

Name F 2nd 3rd 4th 5th 6th 7th 8th 9th

Frequency 60 120 180 240 300 360 420 480 540

Sequence + 0 + 0 + 0

Rule: If waveforms are symmetrical, even harmonics disappear.

Effects of Harmonic Sequences

Sequence Rotation Effects (from skin effect, eddy currents, etc.)

Positive Forward Heating of conductors, circuit breakers, etc.

Negative Reverse Heating as above + motor problems

Zero None Heating, + add in neutral of 3-phase, 4-wire system

Harmonics are classified as follows:1. Order or number: Multiple of fundamental, hence, 3rd is three times the fundamental, or

180 Hz.2. Odd or even order: Odd harmonics are generated during normal operation of nonlinear

loads. Even harmonics only appear when there is dc in the system. In power circuits, thisonly tends to occur when a solid state component(s), such as a diode or SCR, fails in aconverter circuit.

3. Sequence: Positive sequence. Main effect is overheating. Negative sequence. Create counter-torque in motors, i.e., will tend to make motors gobackwards, thus causing motor overheating. Mainly 5th harmonic. Zero sequence. Add in neutral of 3-phase, 4-wire system. Mainly 3rd harmonic.


3/6


Voltage distortion, however,depends on source impedance,i.e., on system capacity. It wasquite possible for the first (or sec-ond or third) customer to injectsignificant harmonic currents intothe system and not cause voltageTHD to exceed 5 %. The entireresponsibility for harmonic miti-gation could fall on the lastcustomers unlucky enough topush V-THD over 5 %, even if

their particular harmonic loadwas relatively small-literally thestraw that broke the camelsback.

3b. Current THDTo restore some fairness to thissituation, standards for maximumcurrent harmonics were added,since current harmonics wereunder the control of the localfacility and equipment manufac-turer (remember, harmonicloads act as generators of har-monics). This emphasis on the

mitigation of current harmonics atthe load, including the not-too-distant requirement that the loadgenerate virtually no harmonics,has become the prevailing regu-latory philosophy. It puts theburden of responsibility on thelocal site and on the equipmentmanufacturers.

For equipment manufacturers,IEC 1000-3-2, published in 1995,is the applicable standard. Itspecifies maximum current levelsout to the 40th harmonic. Itsexpected effective date is pro-jected to be early 2001. To certifyfor CE, a requirement for theEuropean market, manufacturerswill have to meet this standard.This edict will have a major effecton power supply design.

For the facility, IEEE 519 is thestandard (EN 50160 in Europe).The limits set in IEEE 519 forharmonic currents depend on thesize of the customer relative to thesystem capacity. (See Table 2.)

The SCR (Short Circuit Ratio) isa measure of the electrical size ofthe customer in relation to theutility source. The smaller thecustomer (higher SCR), the less

the potential impact on the utilitysource and the more generousthe harmonic limits. The largerthe customer (smaller SCR), themore stringent the limits on har-monic currents.

3c. TDD and THDTDD (Total Demand Distortion) isthe ratio of the current harmonicsto the maximum load (IL). It dif-fers from THD in that THD is theratio of harmonics to the instan-taneousload. Why TDD insteadof THD? Suppose you were run-

ning a light load (using a smallfraction of system capacity), butthose loads were nonlinear. THDwould be relatively high, but theharmonic currents actually beinggenerated would be low, and theeffect on the supply systemwould in fact be negligible. Sowho cares? TDD acknowledgesthis, and allows harmonic load tobe referenced to the maximumload: if harmonic load is high atmaximum load, then we have towatch out for the effect on the

supply source. So where doesthat leave current THD as a usefulmeasurement. The closer the cur-rent THD reading(s) is taken toconditions of maximum load, thecloser it approximates TDD.

Table 2: IEEE 519 limits for harmonic currents at the point ofcommon coupling (All percentages are % of IL, maximum demand load current)

Odd Harmonics

SCR=Isc/IL 35 TDD

1000 15.0 % 7.0 % 6.0 % 2.5 % 1.4 % 20.0 %

SCR = Short circuit ratio (Isc/IL)Isc = Available short circuit current at PCCIL = Maximum demand load current (rms amps)

TDD = Total demand distortion

Note: IEEE allows these limits to be exceeded for up to one hour per day, while IEC allowsthem to be exceeded for up to 5 % of the time.

The concept of IL, maximum demand load current, is key to using Table 2. For existing facili-ties, IL is calculated by averaging the maximum demand current for 12 consecutive months(information available in billing records). For new installations, IL must be estimated.Transformer rating could be used and would be the most conservative estimate (i.e., it wouldresult in the lowest SCR), since it assumes that the transformer would be used at full capacity.

Table 3

Inspection of Transformer Ground Explanation

Check for N-G bond. A high impedance N-G bond will causevoltage fluctuation.

Check for grounding conductor and Fault currents will return to the source viaintegrity of connection to building steel these connections, so they should be as low(exothermic weld). impedance as possible.

Check for tightness of all If the conduit is not itself grounded, itconduitconnections. will tend to act as a choke for higher

frequencies and limit fault current(remember that fault currents are not justat 60 Hz but have high-f components).

Measure for ground currentson the Ideally there should be none, but there willgrounding conductor. always be some ground current due to

normal operation or leakage of protectivecomponents (MOVs, etc.) connected fromphase or neutral to ground. However,anything above an amp should be causefor suspicion (there is no hard and fast rule,but experienced PQ troubleshooters developa feel for possible problems).


4/6


A final word on measuringTHD: the one place not to applythe specs is at the individual har-monic-generating load. This willalways be a worst-case distortion

and a misleading reading. This isbecause as harmonics travelupstream, a certain amount ofcancellation takes place (due tophase relationships which, forpractical purposes, are unpre-dictable). Measure at a PCC, or atthe source transformer.

4. K-factorK-factor is a specific measure ofthe heating effect of harmonics ingeneral and on transformers inparticular. It differs from the THDcalculation in that it emphasizes

the frequency as well as theamplitude of the harmonic order.This is because heating effectsincrease as the square of the fre-quency.

A K-4 reading would meanthat the stray loss heating effectsare four times normal. A standardtransformer is, in effect, a K-1transformer. As with THD, it ismisleading to make a K-factorreading at the load or receptaclebecause there will be a certainamount of upstream cancellation;

transformer K-factor is whatcounts. Once the K-factor isdetermined, choose the nexthigher trade size. K-factor ratedtransformers are available in

standard trade sizes of K-4, K-13,K-20, K-30, etc. K-13 is a com-mon rating for a transformersupplying office loads. The higherratings tend to be packaged into

PDUs (Power Distribution Units)which are specially designed tosupply computer and other PQ-sensitive installations.

5. Ground currentsTwo prime suspects for excessiveground current are illegal N-Gbonds (in subpanels, receptaclesor even in equipment) and so-called isolated ground rods: Subpanel N-G bonds create a

parallel path for normal returncurrent to return via thegrounding conductor. If the

neutral ever becomes open, theequipment safety groundbecomes the only return path;if this return path is highimpedance, dangerous voltagescould develop.

Separate isolated ground rodsalmost always create twoground references at differentpotentials, which in turncauses a ground loop currentto circulate in an attempt toequalize those potentials. Asafety and equipment hazard is

also created: in the case oflightning strikes, surge currentstravelling to ground at differentearth potentials will createhazardous potential differences.

Transformer groundingThe proper grounding of thetransformer is critical. (Table 3.3.)NEC Article 250 in general and250-26 in particular address the

grounding requirements of theSDS. A ground reference is estab-

lished by a groundingconnection, typically to build-ing steel (which, in turn, isrequired to be bonded to allcold water pipe, as well asany and all earth groundingelectrodes). Bonding should beby exothermic weld, notclamps that can loosen overtime. The grounding electrodeconductor itself should have

as low a high-frequencyimpedance as possible (notleast because fault current hashigh frequency components).Wide, flat conductors are pre-ferred to round ones becausethey have less inductive reac-tance at higher frequencies.For the same reason, the dis-tance between the groundingelectrode conductor connectionto the system (i.e., N-G bondat the transformer) and thegrounding electrode (building

steel) should be as short aspossible: in the words of theCode, as near as practicableto and preferably in the samearea...

The neutral and ground shouldbe connected at a point on thetransformer neutral bus.Although permitted, it is notadvisable to make the N-Gbond at the main panel, inorder to maintain the segrega-tion of normal return currentsand any ground currents. Thispoint at the transformer is theonly point on the systemwhere N-G should be bonded.

480V

208 Y/120 V

Neutral

Grounding electrode nearby,preferably structural metal

Figure 3. Transformer grounding.


5/6


There are a number of solutions

for transformer-related PQproblems: Install additional distribution

transformers (SeparatelyDerived Systems)

Derate transformers Install K-rated transformers Used forced air cooling1. Separately Derived System(SDS)The distribution transformer isthe supply for a SeparatelyDerived System (SDS), a termwhich is defined in the NEC(Article 100). The key idea is thatthe secondary of this transformeris the new source of power for allits downstream loads: this is apowerful concept in developing aPQ distribution system. The SDSaccomplishes several importantobjectives, all beneficial for PQ: It establishes a new voltage

reference. Transformers havetaps which allow the second-ary voltage to be stepped upor down to compensate forany voltage drop on the feed-ers.

It lowers source impedancebydecreasing, sometimes drasti-cally, the distance betweenthe load and the source. Thepotential for voltage distur-bances, notably sags, isminimized.

It achieves isolation. Sincethere is no electrical connec-tion, only magnetic coupling,between the primary and sec-ondary, the SDS isolates itsloads from the rest of the elec-

trical system. To extend thisisolation to high frequency dis-turbances, speciallyconstructed isolation trans-formers provide a shieldbetween the primary and sec-ondary to shunt RF (radiofrequency) noise to ground.Otherwise, the capacitive cou-pling between primary andsecondary would tend to passthese high-frequency signalsright through.

A new ground reference isestablished. Part of the defini-tion of the SDS is that it hasno direct electrical connection,including a solidly connectedgrounded circuit conductor, tosupply conductors originatingin another system. (NEC 100)The opportunity exists to seg-regate the subsystem servedby the SDS from ground loopsand ground noise upstreamfrom the SDS, and vice versa.

2. K-rated transformers

Harmonics cause heating intransformers, at a greater ratethan the equivalent fundamentalcurrents would. This is becauseof their higher frequency. Thereare three heating effects in trans-formers that increase withfrequency: Hysteresis. When steel is

magnetized, magnetic dipolesall line up, so that the Northpoles all point one way, theSouth poles the other. Thesepoles switch with the polarityof the applied current. Thehigher the frequency, the moreoften the switching occurs,and, in a process analogous tothe effects of friction, heatlosses increase.

Eddy currents. Alternatingmagnetic fields create localizedwhirlpools of current that cre-ate heat loss. This effectincreases as a square of the

frequency. For example, a 3rdharmonic current will havenine times the heating effectas the same current at the fun-damental.

Skin effect. As frequencyincreases, electrons migrate tothe outer surface of the con-ductor. More electrons areusing less space, so the effec-tive impedance of theconductor has increased; atthe higher frequency, the con-ductor behaves as if it were alower gauge, lower ampacity,higher impedance wire.

The industry has respondedwith two general solutions to theeffects of harmonics on trans-formers: install a K-factor rated

transformer or derate a standardtransformer. Lets look at prosand cons of the K-factor approachfirst. K-factor is a calculationbased on the rms value, %HD(harmonic distortion) of the har-monic currents, and the square ofthe harmonic order (number). It isnot necessary to actually performthe calculation because a har-monic analyzer will do that foryou. The important thing tounderstand is that the harmonic

Figure 4. Typical K-factor in commercialbuilding.

Solutions


6/6


3. Derating standardtransformers

Some facilities managers use a50 % derating as a rule-of-thumb

for their transformers servingsingle-phase, predominantlynonlinear loads. This means thata 150 kVA transformer wouldonly supply 75 kVA of load. Thederating curve, taken from IEEE1100-1992 (Emerald Book),shows that a transformer with60 % of its loads consisting ofSMPS (switched-mode powersupplies), which is certainlypossible in a commercial officebuilding, should in fact bederated by 50 %.

The following is an acceptedmethod for calculating trans-former derating for single-phaseloads only. It is based on the veryreasonable assumption that insingle-phase circuits, the thirdharmonic will predominate andcause the distorted current wave-form to look predictably peaked.

Use a true-rms meterto makethese current measurements:1. Measure rms and peak current

of each secondary phase.(Peak refers to the instanta-neous peak, not to the inrush

or peak load rms current).2. Find the arithmetic average of

the three rms readings and thethree peak currents and usethis average in step 3 (if theload is essentially balanced,this step is not necessary).

3. Calculate Xformer HarmonicDerating Factor:xHDF = (1.414 * IRMS) / IPEAK

4. Or, since the ratio ofPeak/RMS is defined as CrestFactor, this equation can berewritten as:xHDF = 1.414 / CF

If your test instrument has thecapability, measure the CF ofeach phase directly. If the loadis unbalanced, find the aver-age of the three phases anduse the average in the aboveformula.

Since a sine wave currentwaveform has a CF=1.414, it willhave an xHDF=1; there will beno derating. The more the 3rdharmonic, the higher the peak,

the higher the CF. If the CF were2.0, then the xHDF=1.414 / 2=.71. A CF=3 gives us an xHDF=.47. A wave with CF=3 is about

as badly distorted a currentwaveform as you can expect tosee on a single-phase distributiontransformer.Caution: This method does not apply totransformers feeding three-phase loads,where harmonics other than the third tend topredominate and CF is not useful as a simplepredictor of the amount of distortion. A calcu-lation for three-phase loads is available inANSI/IEEE C57.110. However, there is somecontroversy about this calculation since itmay underestimate the mechanical resonantvibrations that harmonics can cause, and thataccelerate transformer wear above andbeyond the effects of heat alone.

4. Forced air coolingIf heat is the problem, cooling isthe solution. Break out the fan,turn it on the transformer and useforced air cooling. Some experi-enced hands figure thats worth20-30 % on the up side. In anycase, it can only help.

order is squared in the equationand that is precisely where thehigh- frequency heating effects,like eddy current losses, aretaken into account.

K-rated transformers aredesigned to minimize and accom-modate the heating effects ofharmonics. K-rated transformersdo not eliminate harmonics(unless additional elements likefilters are added). They accom-modate harmonics withtechniques such as the use of anumber of smaller, parallel wind-ings instead of a single largewinding: this gives more skin forthe electrons to travel on. Theprimary delta winding is up-sized

to tolerate the circulating thirdharmonic currents without over-heating. The neutral on thesecondary is also up-sized forthird harmonics (typically sized attwice the phase ampacity).

Application issues withK-factor transformers

K-rated transformers have beenwidely applied, but there are cer-tain issues with them. Manyconsultants do not see the needfor using transformers with a rat-

ing higher than K-13 althoughK-20 and higher might be sup-plied as part of an integratedPower Distribution Unit (PDU).Also, early applications some-times overlooked the fact thatK-rated transformers necessarilyhave a lower internal impedance.Whereas a standard transformerhas an impedance typically in the5-6 % range, K-rated transform-ers can go as low as 2-3 %(lower as the K-rating increases).In retrofit situations, where a

standard transformer is beingreplaced by a K-rated transformerof equivalent kVA, this mayrequire new short circuit calcula-tions and re-sizing of thesecondary overcurrent protectivedevices.

Fluke CorporationPO Box 9090, Everett, WA USA 98206

Fluke Europe B.V.PO Box 1186, 5602 BDEindhoven, The Netherlands

For more information call:In the U.S.A. (800) 443-5853 orFax (425) 446-5116In Europe/M-East/Africa (31 40) 2 675 200 orFax (31 40) 2 675 222In Canada (800) 36-FLUKE orFax (905) 890-6866From other countries +1 (425) 446-5500 orFax +1 (425) 446-5116Web access: http://www.fluke.com

2004 Fluke Corporation. All rights reserved.Printed in U.S.A. 10/2004 2403202 A-US-N Rev A

Fluke. Keeping your worldup and running.

00

20

40

60

80

100

20 40 60 80 100

Transformer Capacity (%)After Derating forElectronic Load

Switched-Mode Power Supply Load (% of Overall Load)

Figure 5. Transformer derating curve (IEEE 1100-1992)

common-power-quality-factors-affecting-transformers_an.pdf

Documents

Transcript of common-power-quality-factors-affecting-transformers_an.pdf