During my more than 40 years of employment as an electronics engineer with pioneering PQ instrument companies, I often heard from nonengineers, especially in finance, that our products would soon be obsolete.

The reasons they gave were:

1. The susceptible loads would be made immune to PQ phenomena.
2. The susceptible loads would have PQ monitoring instrumentation built in.
3. The electric grid would be improved to eliminate PQ disturbances.
4. The grid would be converted from AC to DC voltage.

Reviewing more recent data

While the loads and electric supply have changed over the years, data from more recent studies show these were false prophecies. Let’s look at a few of those studies.

The Leonardo Power Quality Initiative published in 2007 that losses attributed to power quality, for those surveyed, were 150 billion euros ($165 billion) annually in Europe. Of respondents, 75% reported typical PQ disturbances, including dips (sags), harmonics, transients and interruptions that resulted in data loss, loss of sync in processing equipment, motor/process equipment damage, lights flickering and breakers tripping.

The 2001 Primen Study of 1,000 businesses, which are typically the most vulnerable, found that outages or interruptions made up 85% of attributed losses. The researchers assumed that nonsurveyed businesses had 25%–50% of the surveyed costs, which were summed up in the aforementioned 150 billion euros. The report concluded that spending on preventive measures for losses due to poor power quality was less than 5% of the total cost of the losses.

According to the Galvin Electricity Initiative, “the U.S. electric power system is designed and operated to meet a ‘3 nines’ reliability standard. This means that electric grid power is 99.97% reliable. While this sounds good in theory, in practice it translates to interruptions in the electricity supply that cost American consumers an estimated $150 billion a year.”

The Consortium for an Electric Infrastructure to Support a Digital Society’s 2001 report details responses to what was happening in their electrical systems and processes at that time. This report, “Understanding the Cost to U.S. Business from Unmitigated Reliability and Power Quality Events,” was updated for inflation in 2020 by the Electric Power Research Institute. It identified that those industries most vulnerable had average annual losses of $40,000 per business, or $60 billion total. Those less vulnerable had less than half the dollar loss per business. Because there were so many more businesses in the less vulnerable category, it resulted in estimated losses due to PQ disturbances for all businesses in the United States to be between $145 billion and $230 billion.

Since this seems higher than statistics from data in the 1980s and ‘90s, even after accounting for inflation, has the quality of the electrical supply decreased or is equipment more susceptible now? How does one separate the quality of the supply from the equipment’s susceptibility as the main cause of costs associated with process interruptions and equipment failures?

Electrical distribution need

As for the quality of the U.S. electrical distribution, the EPRI DPQ Project, which ran from 1989 to 1995, found the 95th percentile for voltage harmonic distortion (voltage thd) was 1.4% across 100 sites on the distribution system. The 3rd, 5th and 7th harmonics contributed the majority of the distortion, with little contribution above the 13th harmonic. 

The follow-up report in TPQ-DPQ III, which ran from 2009 to 2012, found that the voltage  had doubled, but was still well below the IEEE 519 accepted limit of 5% (which was raised to 8% in a subsequent revision of 519). 

However, it did note that high-frequency resonance conditions began to be observed several years after the report was published. Voltage harmonic distortion is primarily caused by the current of certain loads, not by voltage generation or distribution.

A power quality index SARFIx is the system average RMS variation frequency index. The X reflects the sag threshold used, such as the CBEMA, ITIC, SEMI F47 or numeric limit such as 70% of nominal, which is a typical point for equipment having issues. In the three DPQ studies over two decades, that number has changed from 23.4 to 13.7 to 15.7. This makes it seem that the quality of the supply has gotten better with regard to sags.

There have been very few studies of PQ levels published in the last two decades, though distributed resources that may or may not affect voltage regulation and stability have significantly increased. Loads, such as electric vehicle charging and the aforementioned power supplies that generate higher-frequency harmonics, called supraharmonics, are on the rise. One cannot ignore the tremendous increase in very large kilowatt-consuming data centers.

The answer to the question of better or worse will need focused research. Or will we just go along saying, “It’s good enough as long as my beer is cold and my pizza is hot”?

Maxim Stepanov / stock.adobe.com