As commercial, industrial and educational facilities become more dependent on information technology (IT) equipment for performing the day-to-day tasks, the quality of electrical supply that powers these growing loads becomes more of a factor. More and more people are becoming aware that power quality needs to be part of the productivity equation. However, there is a significant shortfall in the educational process as to what is a suitable level of power quality to allow the equipment within the facility to operate without malfunctions or interruptions.
There are two major factors that enter into the equation. The first factor is what is the susceptibility of the equipment to various types of power quality phenomena. And the second is what is the frequency and severity of the different power quality phenomena that one would expect at the point that the electricity is used. The last part of that sentence is a critical point, as the quality of supply will vary within a facility. Generally, the majority of disturbances are created by things happening within the facility, not from events occurring on the electric utility system.
To determine these factors, one must have an understanding of the various types of power quality phenomena (or disturbances) and the cause/effect relationships. There are several good references for this, including IEEE Std 1159-2001 Recommended Practice on Monitoring Electric Power Quality and NFPA 70B-2002 Recommended Practice for Electrical Equipment Maintenance. The types of disturbances are broken into several categories.
The most common disturbances are those where the energy-providing capacity of the voltage varies enough from the nominal value to make equipment malfunction. The method that we use to measure these variations is looking for changes in the RMS (root mean square) value of the voltage waveform. When the variation goes below typically 90 percent of the nominal value, we call this a sag (a “blink” to some people), which usually account for more than 60 percent of the disturbances. When it goes down further to below 10 percent of nominal, we call this an interruption (outage or blackout). When it goes above the nominal value by more than 110 percent, this is a swell (surge).
Not only is the remaining voltage magnitude important, but so is the duration of the variation. There are both short- and long-term RMS variations. Short-term variations are classified as instantaneous, momentary and temporary, ranging from 8 milliseconds in duration to 1 minute. Longer-term variations are classified as sustained undervoltage (brown-out) or overvoltage conditions.
There are also disturbances that are much shorter in duration but can also be much larger in magnitude. These are referred to as transients (spikes, impulses), which can be impulsive in nature (such as lightning strikes on or adjacent to a wire) or have an oscillatory or ringing nature (such as when a power factor correction capacitor is switched on or off).
Other types of disturbances can distort the normal sine waveshape voltage waveforms. This distortion includes harmonic distortion, which is the presence of signals that are integer multiples of the fundamental frequency (50 or 60Hz normally). Sometimes the distortion is generated from frequencies less than the fundamental frequency, which can give rise to voltage fluctuations (flicker). During large system-wide fault conditions or when operating from a generator, the fundamental frequency can vary, though this isn’t too common in the United States and other countries with large, interconnected electrical grids.
Each of these phenomena can have a different effect on the operation of a piece of equipment. The most common type of equipment that people associate with such vulnerabilities have been referred to in the past as “sensitive electronic equipment”, such as computers, printers, copy machines, PBXs, etc. In the industrial facility, there are also programmable logic controllers (PLC), adjustable speed drives (ASDs), and even plain old induction motors that the PQ phenomena can be the source of an interruption or malfunction of the process. This in turn results in either damaged product that must be scrapped, damaged equipment that must be taken off line and repaired, lost productivity due to restart time, or even lost revenue due to unavailability of the service, such as with credit card and other financial institutions. Elevator controls can malfunction, resulting in people stuck between floors in high rise office complexes.
Even personal and equipment safety equipment can be affected by such. If the photoelectric sensor devices (used to shut down equipment if part of someone interrupts the beam) are subjected to relatively minor sags, some of these devices will give a false trip indication, shutting a production line down. In other facilities with smoke/fire alarm systems, sags and transients that often occur when thunderstorms rumble through the area can cause false trips as well. When these systems are tied into automatic shutdown controls, once again, the production is interrupted for no apparent reason.
When high energy transients are coupled into unprotected power and communication lines, components can fail catastrophically, resulting in damage to the equipment and potential fire hazard. High harmonic levels can cause overheating of motors and transformers. In wye-type circuits with high harmonics, it is possible to cause overheating of the neutral conductor, due to excessive current levels flowing in a conductor that normally should have very little current. In extreme cases, this has resulted in electrical fires. If some of this current is diverted into the equipment ground conductor, then metal enclosures may have an elevated touch-potential, creating an electric shock hazard.
Even educational facilities are subject to problems associated with power quality phenomena. The classrooms from kindergarten to graduate school are becoming increasingly “wired” and dependent on computers for learning, as well as administrative functions. Whereas it might be more difficult to attach a productivity dollar cost when a lesson is interrupted in the classroom, these facilities still experience many of the similar problems found in the commercial and industrial world. The infrastructures of many educational facilities were not designed to handle the significant increase in electrical load from the proliferation of IT equipment in the classroom, labs, and office space. Heat levels go up in the room, so window air conditioners are brought in and plugged into the same circuit as the computers. When the AC unit’s compressor kicks on, the inrush current can cause a sag that trips the PCs off momentarily, which results in several minutes of waiting time for the computers to reboot and get running the applications again.
Universities add another dimension of vulnerability. Most universities have laboratories that conduct funded research that provides a significant revenue stream. Whether it is complex calculations on “supercomputers” or nano-technology experiments in the biochem labs, power quality phenomena can create havoc with these systems. It can take a much smaller level of harmonics, transients, and other disturbances to disrupt or distort the results of these very sensitive and costly experiments. In today’s times, loss of connection to the Internet has even lead to significant student “unrest.”
No matter where you work, your dependence on equipment that is vulnerable to power quality phenomena is most likely increasing every day. Just because you can’t see, smell, taste, or feel it, doesn’t mean that it isn’t out there affecting you on a daily basis.
BINGHAM, a contributing editor for power quality, can be reached at 732.287.3680.