Test tool advice from expert Chuck Newcombe
I recently presented a seminar and demonstration on the long term cost advantages of incorporating high efficiency distribution transformers in modern commercial and industrial buildings.
Among the topics discussed were the presence of harmonic currents generated by electronic system loads, alternative transformer types, wiring alternatives, and, in the case of system rehabs and upgrades, the measurements needed to establish before and after performance benchmarks.
As the session broke to the demonstration phase I was approached by one of the attendees, an electrical contractor, who voiced this question, "You know, this all sounds good, but how do you convince a potential customer that the problems that you propose to solve really exist in their system? Harmonics and all that are pretty mysterious, you know, and I don't have a 'big gun', an expensive analyzer, just sitting on my tool shelf." Then he asked if there wasn't some way he could easily show the customer that the problem existed using only his DMM.
His question took me back to 1987, when I first encountered harmonics in power systems and was asking myself the same questions. At the time, I was tasked with explaining the importance of True RMS current measurements at a conference on the problems associated with power harmonics.
As part of the explanation, I devised an experiment in which a common average sensing meter and a True RMS meter (a rarity at the time), were used to measure the same signal. The signal in question was the typical current waveform of the single phase current drawn by a computer.
First, the meters were compared measuring a current without harmonics, to verify that they read the same. Then, a harmonic-rich source was used and the measurements were repeated.
The result was that with roughly 90% harmonics present, the average sensing meter read some 30% less than the True RMS meter. A power quality consultant who was present said, "That could be a quick way to verify the presence of harmonic distortion before I bring in my scope or power analyzer." And with that observation, the two-meter method for detecting and estimating harmonic content was born.
If you already have a True RMS meter, and you'd like to consider using this method, the solution is relatively inexpensive - just get yourself an average sensing clamp meter (they're the least costly ones) and implement the two-meter method. Many of you may already have one of these - that you tossed in a closet when you upgraded to True RMS.
The nice thing about using this method is that you can show the two measurements side-by-side while varying loads.
A year later, in 1988, the Fluke 87 True RMS DMM was introduced. And this workhorse, recently reintroduced as the Series V, also had the ability to read the peak value of the waveform in addition to its True RMS value. Now, looking at the ratio between the peak and rms values (the crest factor) the technician had yet another way to detect and estimate harmonic content. A 60 Hz sine wave will have a crest factor of 1.4, while the computer current waveform mentioned about may show a crest factor of 3 or more.
If you're dealing with a 3 phase 4-wire 208/120 system (most common in commercial and industrial offices), there's an even simpler telltale measurement you can make. All you need is a dmm that measures frequency, and access to a 120V receptacle at the end of a branch circuit with a shared neutral. With normal loads active, measure the neutral to ground voltage at the chosen receptacle.
In a troubled system, you may read 5 V or more from neutral to ground. Something less than 2 V is required by some office equipment suppliers, or they void their warranty. The thing that tags the likely problem as harmonics is to simply measure the frequency of that voltage. If harmonics are present, it will be 180 Hz (the third harmonic) rather than the expected 60Hz.
These DMM tests can be the first line of defense and will quickly sort out the most common problems without the expensive and time consuming process of a complete analysis using a full-blown power quality meter. But, once you've determined that significant harmonics are present, it is of course advisable to perform the more detailed tests and record them as a reference for the correction work that may follow.
Back to the seminar presentation I mentioned at the beginning of this column. I was able to demonstrate two of the methods described using my trusty 87-V (I didn't have the average sensing meter to use for comparison purposes.) The questioner seemed happy with the answer, so we moved on to the originally planned demonstration, which pointed out some of the transformer efficiency issues covered during the presentation.
The demonstration clearly showed that harmonic currents are a significant contributor to reduced system efficiency, and that in these days of soaring energy costs, reducing or removing harmonics represents a major opportunity for cutting plant operating costs.
In a future column, I'll discuss some innovative examples where the costs of upgrading or replacing outmoded distribution equipment, to deal with transformer efficiency in the presence of harmonics, actually resulted in power savings so significant that the payback for the more expensive transformers was in months, not years.
With an incentive like that, use of the big guns may be easily justified.