A three-phase power meter is more than a few DMMs and some duct tape.
Portable power analyzers are not a one-size-fits-all commodity product. They come in many different shapes and sizes with a wide range of capabilities. It’s obvious that for something to be portable, it can’t be permanently installed at a fixed location. But attempting to be more specific, such as using the term hand-held, is too restrictive a definition in the case of portable power analyzers.
Several very comprehensive instruments are ruggedized versions of analyzers that often are permanently installed. An example of this type of product is the General Electric portable power quality meter (PPQM). As described in the datasheet, “The portable version of the power quality meter (PQM) has all the same features as the panel-mount PQM available in a rugged carrying case.” The PPQM case measures 12.8² × 11.6² × 10.5².
Applications that require this kind of meter are found in industrial, commercial, or utility three-phase installations. Compared with a simple single-phase load characterized by one voltage and one current, three-phase loads have three or four voltages and currents depending on configuration.
A so-called delta-connected three-phase system and a wye-connected system are shown in Figures 1a and 1b, respectively. The wye connection has a neutral circuit associated with the three phases, making it a four-wire system. The delta connection does not involve neutral, so it requires only three wires.
According to Clayton Wilson of Yokogawa’s measurement and control product marketing group, “The typical portable power-analyzer user is a field-service or maintenance engineer. These customers will be looking for improvements in production lines or ways to reduce costs through better management of power-consuming equipment such as air conditioners, elevators, or lights. It’s inconvenient to use a benchtop power analyzer for field checking equipment or inspection during maintenance cycles.”
Chunky Yet Portable
The complement of voltage and current inputs varies among the meters in this class, although all provide at least three voltage and three current channels. Some examples will help to illustrate what is available and why you might choose one product instead of another. The PPQM has four voltage and three current inputs. It can measure the neutral voltage in a wye-connected system, but you can’t measure the neutral current in addition to the three phases.
Ametek’s Meridian Power Quality Analyzer provides four voltage and four current inputs. Reliable Power Meters has a portable power recorder that features four voltage channels and five current channels. The three phases, neutral, and ground currents add up to five, and you can measure them simultaneously. The Megger PA-9Plus Power Quality Analyzer also accommodates four voltages and five currents.
The PA-9Plus draws power from the source being measured, or you can use an auxiliary power input. The lack of dependence on a separate 120-V power input means that this instrument can be used in remote locations.
Capable and Lightweight
In a much smaller form factor, the ELITEpro Load Profiling Power Meter from Dent Instruments/Optimum Energy Products packs four voltage and four current channels into a 12-oz 3.2² × 5.9² × 2.4² package. It is battery powered and connected to a PC via a built-in modem.
Chauvin Arnoux and Hioki also make instruments in this class that perform a complete range of three-phase measurements but are much more portable than the so-called chunky group. Products made by these companies include color LCDs that display voltage, current, and power waveforms as well as measurement values. The Yokogawa CW140/CW120 Power Meters feature monochrome LCDs with comprehensive numeric displays and graphs of harmonic content.
Mostly Single Phase
For lower power applications, single-phase meters may be all that you need. AEMC has a Power Quality Logger that plugs into the 120-V line. You then plug your 120-V load into the meter. There are no probes to connect, although the logger is configured by and works with your PC over an RS-232 link.
Fluke’s Model 43B Power Quality Analyzer combines the capabilities of power and harmonics meters, a trend recorder, and a 20-MHz oscilloscope. David Pereles, electrical marketing manager at Fluke, described how the meter is used: “When monitoring the quality of power delivered by a utility, a three-phase monitor at the service entrance is a necessary tool. But surveys have consistently shown that utility power is, in fact, relatively predictable and that most power problems originate downstream, inside the facility.
“These problems typically are caused by a combination of load interactions and inadequate wiring,” he continued. “A hand-held test tool is most useful for troubleshooting these kinds of problems. The 43B is designed for bottom-up troubleshooting from load to source to complement the top-down view offered by a three-phase monitor.”
What Is Power Quality?
When referred to the load, power quality describes the degree to which the circuit under test differs from a pure resistance. Power meters measure the difference in a number of ways. The most basic quantities are voltage and current amplitudes. However, any load other than resistance has a non-zero phase angle associated with it.
For undistorted current and voltage sine waveforms, the power factor is equal to the cosine of the phase relationship between current and voltage. More generally, the power factor is the ratio of the actual power in units of watts divided by the apparent power in units of VA.
This definition holds for all wave shapes including switched loads. Power not contributing to useful work in the load gives rise to the concept of VARs, reactive power or the power that circulates among the reactive components of the load.
When referred to the supply, power quality relates to the continuous provision of sinusoidal voltage at a constant level regardless of current load. The supply should not include anomalies such as transients or dropouts.
On its website, the Los Angeles Department of Water and Power emphasizes the change in customer power-quality expectations that has resulted from the introduction of sensitive electronic equipment and computers. Voltage spikes that may not have affected lighting or motors now cannot be tolerated. Fortunately, there are good measurement instruments available today that pinpoint the sources of problems. Equally important, there also are many affordable remedies for common power problems.
Even before the proliferation of sensitive electronic equipment, industrial power users were striving for high power factors. A low power factor increases the load on a facility’s wiring because large harmonic currents actually do flow although they do no useful work. Also, the utility supplying the power will be affected by a large user with a low power factor, so the utility may charge a penalty.
What’s Being Measured?
Voltage and current amplitude and phase, phase-to-phase voltage unbalance, power factor, crest factor, frequency, VA, power in watts, VARs, short-term voltage flicker, transformer k factors, the amplitude of individual harmonics, and total harmonic distortion are among the important power parameters commonly measured. Typically, you are trying to determine if your power source starts out clean but is corrupted by the loads it drives, or is the supply initially faulty?
Assuming the incoming power is not the problem, troubleshooting becomes a matter of anomaly hunting within the facility. Nonlinear loads such as switching and capacitor-input power supplies generate current harmonics that can distort the voltage waveform. Or, you may have sensitive equipment powered from the same circuit as heavy air-conditioning or industrial motors that switch on and off. Of course, some wiring simply may be misconnected.
Modern electronic power-quality instruments benefit from the low-cost availability of high-speed analog-to-digital converters (ADC) and signal processing. Signals often are digitized with 128 samples per cycle, and 256 is not uncommon. In addition, flicker and transient applications may have a much higher sampling rate, such as 2 MS/s used in Hioki’s Model 3196.
The length of time that signals and measurements can be logged depends on the amount of internal memory. For example, the Megger PA-9Plus comes with 12 MB as standard, and this can be extended by adding an optional 64-MB or 128-MB memory card. This means that you can continuously monitor and capture the behavior of a three-phase system for several days. Alternatively, many instruments effectively extend their storage capabilities by capturing only abnormal events.
Converter resolution varies from 12 to 16 bits for normal voltage and current acquisitions but degrades for fast transient capture. One model specifies 16 bits degrading to 12; another starts with 12 bits and degrades to 10. Together with the 256 samples per cycle, this means that enough detail is known about the waveforms to generate information out to the 63rd harmonic.
The accuracy specifications vary widely among models and require more than a cursory glance. The comparison chart that accompanies this article lists some representative specifications.
At issue are two terms: percent of range (rng) and percent of reading (rdg). To illustrate, assume you had a voltage meter with a 5% of reading accuracy specification. The actual value of a signal that read 1.0 V could be anywhere between 0.95 V and 1.05 V, regardless of the meter range selected.
The situation is not so straightforward for a meter specified to have 5% of range accuracy. On the 1.0-V range, this amounts to a possible error of ±50 mV for any input. So, the previous example of a 1.0-V signal still would have a possible ±50-mV error, but a full-scale input is the only signal for which 5% of range and 5% of reading specifications are equivalent. An input that read 0.5 V also could be in error by ±50 mV; that’s a 10% of reading error. The effect of a percent of range specification gets more pronounced as the input signal gets smaller.
Figure 2 shows the relationship between percent of range and percent of reading for the special case where the two parts of the accuracy specification add up to 5%. It’s easy to see that you must choose the smallest range that will accommodate your signals to minimize the effect of percent of range specifications.
The matter is further complicated if you consider that the specifications given in the comparison table represent the accuracy of the basic meter. Probe errors have to be added to these figures.
For the Yokogawa CW120, the current accuracy limits listed correspond to the 50-A, 200-A, and 500-A current clamps. Moving up to the 700-A current clamp Model 96032, the aperture size increases considerably, and the probe accuracy specification degrades from ±0.5% of reading to ±1.0%. The overall accuracy for the 96032/probe combination is ±0.8% of range ±1.2% of reading.
A clamp-on style current probe is convenient and surprisingly accurate, but the output does vary slightly according to the position of the lead in the probe window. To reduce this effect, try to center the lead if it does not almost fill the window or use a physically smaller probe if it has the necessary current rating.
Accuracy can be greatly affected by sample timing. Most meters specifically state that current and voltage inputs are simultaneously sampled. This is especially important if significant harmonics are present because even a small phase misalignment between voltage and current affects the instantaneous power calculation.
If high-frequency harmonics are important to your work, verify that the phase shift introduced by current transformers is sufficiently low to be ignored or account for it. Current clamps generally have much lower bandwidth than voltage probes, but the effects are small up to a few kilohertz.
Conclusion
A wide range of products is available to address your power quality analysis needs. Those that have no built-in display present results on your computer screen. Generally lower cost than meters with integral displays, these instruments also are not as convenient and easy to use. On the other hand, the analysis software has access to the high-performance microprocessor in your PC, and there is the possibility to customize the measurements being made.
Although portable, the so-called chunky meters are too bulky and heavy to use as troubleshooting tools in many locations. They are portable monitoring products intended to stay connected at a particular location for at least a few hours. It would be more convenient to use one of the so-called capable and lightweight models if you intend to move it around a great deal.
Whatever instrument is used, three-phase monitoring involves from six to nine probes, and it is important to connect voltage leads to the voltage inputs and current leads to the current inputs. Polarity must be observed. So, any three-phase investigation will take time just to connect and check everything.
What do you lose by using a portable power-monitoring product compared to a benchtop model? Very little, according to the industry experts who responded to our question. In fact, if you can use one of the so-called chunky models, its functionality may be identical to that of a monitor usually installed in a control panel.
As Fred Hensley, director of power quality sales at Megger, explained, “The functional differences between a benchtop analyzer and a ruggedized portable instrument are growing increasingly slim. The sampling rates of benchtop instruments tend to be faster, the graphical displays larger, and the cost significantly higher.”
(See power analyzer chart.)
Return to EE Home Page
Published by EE-Evaluation Engineering
All contents © 2002 Nelson Publishing Inc.
No reprint, distribution, or reuse in any medium is permitted
without the express written consent of the publisher.
December 2002