There was a recurring topic at the summer meetings of the IEEE Power Quality Subcommittee’s various working groups and task forces, and it came to a head at a panel session on the latest version of IEEE 519 2014, IEEE Recommended Practice and Requirements for Harmonic Control in Electric Power Systems. One attendee was a member of the IEEE 1547 Working Group (WG), which has written a series of standards that were driven by the Federal Energy and Regulatory Commission (FERC) to get electric utilities to allow the interconnection of distributed or dispersed generation that wasn’t their own. Solar, wind and other alternative-power sources are increasingly interconnected into the grid. Part of the 1547 WG’s latest efforts involved power quality phenomena limits, and interharmonics became the hot button.


Nearly everyone who works with electrical power is familiar with harmonics. In the early days of electric-power distribution, the generator produced voltage in sinusoidal waveshapes, and the loads of consumers used power with sinusoidal current waveshapes. As electric motors proliferated, the current and voltage remained sinusoidal, but the waveshapes were no longer in phase as the current lags the voltage in inductive loads, which is mainly what electric motors are.


Fast-forward a century and a half to the expansion of so-called “electronic loads.” To power systems, this meant the current being drawn by the loads was no longer a pure sine wave at the fundamental power frequency. The “smarts” of the power supplies didn’t conduct current through the entire cycle, rather just enough to provide the required power to the equipment that they were part of. In some cases, such as adjustable speed drives (ASDs), the fundamental frequency changes within the equipment to provide more efficient and productive equipment operation. Hence, harmonics became an important aspect of power providers and equipment designers.


A full wave rectifier on a three-phase circuit has six paths, pulses or poles of conduction. These are commonly used in the nonlinear power supplies of electronic loads, such as ASDs. The result is that the harmonic currents drawn by the load follow the pattern of h = n × p ±1, where h is the harmonic numbers, p is the poles of conduction, and n represents integers from 1 to infinity. For the three-phase full-wave rectifier, the harmonics are 5, 7, 11, 13, 17, 19 … When 12-pole converters came along, the harmonics shifted to 11, 13, 23, 27, 35, 37 … .


Generally, the first half-dozen or so harmonics are the most dominant, with the amplitude falling off in an exponential decaying pattern. Harmonics cause losses and resulting heat in electromagnetic equipment, such as motors. The higher pole converters generally draw less harmonic current, which is good, as some losses go up as the square of the harmonic number—the higher harmonic number, the higher the losses. Harmonics that are multiples of three are called triplens and are added in the neutral or grounded conductor. Even-order harmonics have lower limits than the adjacent odd-number harmonics. Evens are likely to cause more problems because they make the sine wave nonsymmetrical related to the zero axis.


The standards-making groups in the IEEE and International Electrotechnical Commission (IEC) have set limits on harmonics for more than 20 years, both from what can be in the voltage that the utilities provide and in what the current harmonics from equipment can produce. Since current multiplied by impedance equals voltage, harmonic currents from one facility’s loads can produce harmonic voltage that affects other facilities, hence the need for limits.


But not all equipment causes only harmonic currents. Asynchronous operations can result in frequency of harmonic currents and voltage that aren’t integer multiples of the fundamental frequency. Rather than 120, 180, … 300 hertz (Hz) on 60-Hz systems, there can be currents with a frequency of 145 Hz or 2,917 Hz. These are referred to as interharmonics. 


What should these limits be? Neither standards group had any definitive statement in their published standards. The Working Group for IEEE 1547 wanted to put numbers in their next revision, but what should they be? What about frequencies that are being found from power converters that have switching frequencies well above the 50th harmonic, which is the highest order that most standards consider?


One solution would be to make the limits the same as the adjacent harmonics, but limits for odd and even harmonic numbers are different. If the interharmonic has a frequency of 265 Hz, is it more like the 4th harmonic (240 Hz) or the 5th (300 Hz)? Are there any special considerations because they aren’t integer multiplies, and could the losses or some other interference require additional restrictions? In some parts of the world, these higher order interharmonic frequencies are used as communication signals for rate or tariff information in revenue meters. You can bet some utilities would get upset if the data gets corrupted because of this.


This is another topic without an answer. Hopefully, the groups are developing meaningful limits to the safe and productive electric power systems and equipment operation. Stay tuned.