Harmonics Problems in Distribution Systems Case Study AEP

Harmonics Problems in Distribution Systems Case Study AEP Texas Central Company Scott Lee, Reliability Engineer

WHAT’S WRONG WITH MY LIGHTS? !

SYMPTOMS OF A PROBLEM Residential customer calls complaining of problems with electronics: TV, thermostat – AC, motors making strange noises Serviceman and Operations Tech troubleshoot and have transformer replaced because of unusual humming and arcing sounds from coil New transformer makes the same sound, customer still has problems – call the engineer Tech goes to new pipe plant down the road and finds 2000 k. VA pad transformer making same noise

THE SOURCE Houston based company built a pipe coating and storage plant in George West to better serve the Eagle Ford Shale area Two 1000 k. W induction furnaces to heat and coat 2 inch to 24 inch diameter casing with fusion bonded epoxy Two solid state 6 -pulse rectifiers and step up transformers provide variable frequency 4 k. V power to induction coils For more info. . . see texisle. com and ajaxtocco. com

INSTALLATION Previously during the plant planning stages, AEP provided system, line, and transformer impedances to drive manufacturer for design. Distance and PQ were discussed. 3 miles of distribution feeder was upgraded with 477 MCM aluminum, with a total length of 6. 4 miles Drive manufacturer wanted two 1500 k. VA or one 3000 k. VA transformer, we supplied one 2000 k. VA. They were also very insistent on our transformer meeting their specs. Actual peak is 1380 k. W. A 1000 k. VA pad was installed for other plant power that peaks at 600 k. W AEP standard for three phase padmounts is wye-wye IEEE-519 requirements were not discussed, customer did not tell us the inverter was a 6 -pulse drive

Plant * Residence George West Sub 138/12 k. V 9. 4 MVA Feeder upgraded to 477, 6. 4 circuits miles from sub to customer

AJAX TOCCO, PACER-2 DRIVE 1000 k. W 500 Hz Variable Frequency 6 -pulse rectifier No filtering Water cooled

INDUCTION HEATING LINE

FUSION BONDED EXPOXY COATING

MEASUREMENTS PMI recorders installed at plant transformer and at residential customer. Set to record harmonics and capture waveforms. Voltage THD at pipe plant on secodary, 14% Voltage THD at residential secondary, 5. 1% PMI stands for Power Monitors Incorporated, see www. powermonitors. com, i. Vs-3 and Eagle 440

Unfiltered, running at 1550 k. VA 5 th 7 th 3 rd 2000 k. VA Padmount 6 -Pulse Drive 11 th 13 th 12. 4% THD-Voltage 5 th 7 th 11 th 23. 4% THD-Amps

Unfiltered, 2000 k. VA Padmount, 6 -Pulse Drive 12. 4% THD-V 23. 4% THD-A

THE FIX Met with pipe company and drive manufacturer to request they install filtering and comply with IEEE-519 (5% THD limit on voltage) Move other distribution customers in area to another feeder Drive manufacturers said they would install passive filters (reactors and capacitors) but parts would have a 14 week lead time plus engineering and installation time I countered to pipe company that this was unacceptable, and offered three filter manufacturers with on the shelf parts that offered active filtering and engineering support. Recomended MTE Matrix Pure Sine active filters, Schaffner EMC ECOsine active filters, and TCI H 5 active filters.

THE FIX CONT. There was a debate over passive versus active, but after some research, I pressed the customer to go with active filters because of the varying load and the possibility of hitting resonance with a passive unit, also recommend a 5% line reactor which is standard for 6 pulse drives The customer chose a vendor, and we had several phone conferences about what to install, and what would be acceptable with AEP The vendor chosen was Schaffner EMC and the ECOsine FN 3420 filter with additional 5% reactors For more info. . . List location or contact for specification (or other related documents)

ACTIVE FILTER Schaffner EMC ECOsine filter injects up to 300 amps of current Compensates for harmonic currents Compensates for reactive current Can choose percentage of overall current dedicated to each harmonic Two installed with 5% reactor ahead of each drive

MEASUREMENT AFTER FILTER INSTALL Before at 1850 amps of load: Voltage THD 12. 6% Current THD 23% After at 2150 amps of load: Voltage THD 5. 5% Current THD 8. 8% After at 1320 amps of load: Voltage THD 7% Current THD 5% Adjacent Transformer Voltage THD 3. 5%

HARMONIC CURRENTS BEFORE/AFTER Total Load 1800 amps Harmonic Amps 5 th 427 7 th 225 11 th 113 13 th 60 Amps with Filter 49 60 52 59 THD-V from 12. 2% to 8. 1%

Testing, filter on and off ----- Filter off 2300 A rms 5. 9% THD-V 9. 6% THD-A 8. 4% THD-V 24. 2% THD-A Neutral volts and amps

Waveforms showing commutation notching 6. 9% THD-voltage 5% THD-current 1300 amps

UG Residence, one mile from pipe plant, plant running at 7% THD-V 3. 5% THD-V 3 rd, 7 th, and 5 th V I

STILL PROGRESS TO BE MADE Not below 5% THD on voltage at meter Filtering has done all it can do Work with drive manufacturer to control commutation notching using saturating reactors on SCRs to speed gate closing

IN RETROSPECT Should have required drive manufacturer to meet IEEE 519 standard Get a written proposal from drive manufacturer on how they will meet the standard Use delta primary transformer to trap Triplen harmonics, this is 2. 7% of total Recommend active filtering Recommend 12 pulse drives or technology to reduce commutation notching

Questions