- What is Power Quality?
- What is Electrical Grounding
- Why Grounded Systems are Preferred
- Sensitive Electronic Equipment
- The Switched-Mode Power Supply
- The Effects of Non-Linearity on 3-Phase Systems
- Techniques That Help
- General Wiring
- Grounding Considerations
The proliferation of computers and other sensitive devices throughout our manufacturing and office environment has fostered the need to design the electrical systems of buildings with an eye toward power quality issues. There have been numerous books and articles concerning diagnosis and remediation of power quality problems in existing structures after these problems have manifested themselves.
The primary focus of this presentation will be wiring and grounding techniques and practices that are recommended to be part of the design of new or renovated structures. These practices will help prevent power quality problems from occurring in the first place, or diminish their effect to the point that they are not significant. Back to Top
What is Power Quality
The term "power quality" means different things to different people. One definition is the relative frequency and severity of deviations in the incoming power supplied to electrical equipment from the customary, steady, 60 Hz, sinusoidal waveform of voltage or current. These deviations may affect the safe or reliable operation of equipment such as computers.
Thus, while not having a strict basis of measurement, terms like "poor power quality" generally mean there is sufficient deviation from norms in the power supply to cause equipment mis-operation or premature failure. "Good power quality", conversely, means there is a low level of such deviations or mis-operations.
Because the sensitivity to such deviations varies from one piece of equipment to another, what may be considered poor power quality to one device may be perfectly acceptable power quality to another.
Poor power quality affects the reliable operation of computers and computer-based equipment, which are now so ubiquitous. Often more important than the physical effect on the equipment is the loss of productivity resulting from computer equipment failure, mis-calculations and downtime. In fact, it has been estimated that the total cost to US businesses of this lost productivity is a staggering $15-30 billion per year. A recent survey by E-Source indicated that, while most respondents did not calculate the cost of their annual losses due to power quality (or may even erroneously attribute power quality glitches to software or hardware causes), roughly a third of those that did report a loss figure said it exceeded $1 million per year 1 .
The vast majority of power quality problems in a building originate within the same building. The Institute of Electrical And Electronic Engineers (IEEE), various government agencies and other organizations have been studying these problems and effects for several years. As a result, they have issued design guidelines and recommended practices that are known to greatly reduce, if not eliminate, the incidence and severity of power quality related problems.
In many cases, simply installing enhanced electrical systems and better grounding systems will prevent (or cure) the problem. Many of the simple techniques explored in this document are relatively inexpensive to install during construction, or during major building renovation. Further, since the use of a particular building, or area within a building, may vary considerably over the years, the recommended infrastructure improvements will serve to make the building more useful over time, despite changes in tenants, end uses or equipment.
Ten or more years ago, few builders and electrical designers could imagine the level of computerization we find today in buildings of every sort. Who could have foreseen a PC on every desk? Business computers were large machines located in special "computer rooms". Lighting fixtures had low harmonics output. Telephones were hard-wired. Motors ran only at their design speed. An office at home was a rarity. Laser printers were uncommon. And, considering all this, who can predict what the future holds with respect to electronics?
Generally speaking, by following generally well-known formulae for electrical loads to be expected per given floor area, the designer of past decades was reasonably assured of designing an adequate electrical installation that could be expected to serve the needs of the building and its occupants well into the future. There was seldom a need to be concerned about harmonics, or transients. But time, progress, and micro-computerization marched on.
How often does a power quality problem arise? According to a study of 112 sites of differing location, size and type, performed by National Power of Neceda, WI, the average site had 106 disruption events per month, with the worst location having over 4,000 such events 2 . Most disruptions show up in random, difficult-to-reproduce ways, such as a PC that locks up, a PBX that loses calls or a motor that fails prematurely.
According to the Electric Power Research Institute, as much as 80% of power quality problems relate to inadequate wiring or grounding, 3 so as some power quality issues are examined below, particular emphasis will be placed on wiring and grounding.Back to Top
What is Electrical Grounding
The term "ground" refers to the earth, or a large body that serves in place of the earth. The term "grounded", then, refers to a system in which one of the elements is purposely connected to "ground". The British use the terms "earth" and "earthing" instead of "ground" and "grounding", which are probably more appropriate, but this publication will use the American convention since the terms referencing ground appear throughout the literature and US codes.
Electrical systems need not be grounded to function, and indeed not all electrical systems are grounded. But the voltages referred to when talking about electrical systems are usually voltages with respect to ground. Ground, therefore, represents the reference point, or zero potential point, to which all other voltages refer. Indeed, as computerized equipment communicates with other equipment, a zero reference voltage is critical for proper operation.
The ground (earth), then, is a good choice as the zero reference point in most cases since it surrounds us everywhere. When one is standing on the ground, one's body is approximately at the voltage potential of the earth. If the building is metal framed, the metal components of the building structure, or the water piping (if metallic), are approximately at ground potential.
In most cases, the electrical service to most buildings installed over the past several decades is "grounded". There are numerous exceptions. Whether or not a given electric service to a building is "grounded" - that is, purposely connected via a low impedance connection to the "ground" - is determined by the rules of the National Electrical Code © (NEC) 4 and the electric utility serving the facility.Back to Top
Why are Grounded Systems Preferred?
Primary purpose of grounding electrical systems is to protect personnel and property if a fault (short circuit) were to occur. In simple terms, if one of the three hot legs (phases) of an ungrounded electric service becomes grounded, intentionally or accidentally, nothing happens. No circuit breaker trips, no equipment stops running. Ungrounded electrical systems were popular in industrial buildings of the first half of the 20th century precisely for the reason that motor-driven loads, which were the most common at the time, would not stop simply because of a short.
But a consequence of this type of system is that it is possible for the frame of a piece of equipment to become energized at some voltage above ground, and present a shock hazard for personnel who may be touching the equipment and a grounded component of the structure simultaneously.
A second purpose of a grounding system is to provide a controlled, low impedance path for lightning-induced currents to flow to the earth harmlessly.
The assumption in this document is of a grounded service installed in accordance with the National Electrical Code © (NEC). There are some cases where this practice is not desirable, and the NEC provides for those exceptions.Back to Top
Sensitive Electronic Equipment
Earlier, the proliferation of personal computers in the office and home environment was discussed. That description is really a metaphor for the proliferation of all the microprocessor-controlled equipment found throughout the commercial and manufacturing environments. Today, most factory environments are computer-controlled.
Concurrent with the proliferation of these sensitive devices, the devices themselves have been changing in ways that make them more sensitive to power irregularities. Operating speeds have been increasing (in the radio frequency range), making the circuits more susceptible to (and emitting) electromagnetic interference. Circuits have been miniaturized, with less space between adjacent conductors on a circuit board, increasing susceptibility to overvoltages, and increasing adjacent-channel interference. The microprocessor chips themselves have become smaller and more densely packed. This decreases heat dissipation, and makes them less robust. Operating voltages have and continue to decrease to allow for this miniaturization. A digital "1" may be in the vicinity of 3.5 - 5.0 volts or less, and a "0" in the range of 0 - 1.5 volts. So smaller overvoltages from transient conditions may result in operating errors.
It is easy to see where it becomes important to keep transient overvoltages and high frequency harmonics away from the microcircuits.
As this continuing miniaturization was taking place, a new type of power supply was developed that offered dramatic weight and component savings, a necessary step to development of smaller, lighter and less costly computers. That was the "switched mode" power supply, to be discussed in more detail shortly.
Among the types of equipment that both can cause power quality problems, and are susceptible to them, are:
- Uninterruptible Power Supplies
- Variable Frequency Drives
- Battery Chargers
- Large Motors During Startup
- Electronic Dimming Systems
- Lighting Ballasts (esp. Electronic)
- Arc Welders, and Other Arc Devices
- Medical Equipment, e.g. MRIs and X-Ray Machines
This list includes equipment that breaks a smooth sine-wave into stepped increments, for control of the downstream device, by varying the voltage or frequency of the output.
Arc operated devices, including general purpose "universal" motors with brushes, arc welders, and even arc-discharge lighting (fluorescent or HID) can be a strong source of electromagnetic interference. (The arc itself is rich in energy of all frequencies.) This interference can be picked up by improperly shielded or improperly grounded wiring, and then conducted into sensitive devices.
Fourier analysis (if you remember your calculus) tells us that a wave of any shape can be created by a defined combination of sine waves of varying frequency and amplitude. Very simply put, the math tells us that square waves and quasi-square waves, which are the output of switched mode power supplies and variable frequency drives, contain elements of sine waves. But rather than just the fundamental 60 Hz sine waves, these square waves also contain many higher frequency components, which are harmonics (multiples) of the 60 Hz fundamental, as well as spiked components that are transient overvoltages. These harmonics can result in heating of circuits and neutrals and possible mis-operation of the digital logic. In addition, the leading edge of a square wave or spike behaves like a high frequency (radio frequency) sine wave, and can be mistaken for such.Back to Top
The Switched Mode Power Supply
Historically, devices requiring DC (direct current) to operate (as all electronic circuits do) had hefty and bulky power supplies that typically had a stepdown transformer supplying a low voltage to a half-wave (simple diode) or full-wave (bridge) rectifier. The power supply was heavy, bulky, and fairly inefficient.
Recently (in the last ten years or so), partly because of the need for lighter weight and higher efficiency, the "switched-mode" power supply was developed. The switched-mode power supply has a full-wave bridge rectifier (BR1 in diagram) directly connected to the incoming 120 V AC line.
The switching circuit draws stored energy from capacitor C1 in short pulses (thus quasi-square waves) before sending the now pulsed DC on to the transformer (TR in diagram). The transformer is now operating on high-frequency, pulsed DC, instead of the historically used 60 Hz AC. This change in operation enables the transformer to be made much smaller and lighter than was possible in the 60Hz, 120 volt version. Thus, overall power supply efficiency is greatly improved, from about 50% in standard power supplies to about 80% for the switched-mode type.
Equipment can now be made smaller and lighter. Power consumption decreases, and batteries for portable models can last much longer. But not without a downside. Because of the pulsed nature of the output, it contains a fairly high level of harmonics, which can flow back out onto the power distribution system, adversely affecting other equipment and even the wiring itself.Back to Top
The Effects Of Non-Linearity On 3-Phase Systems
The net result of harmonic and transient generation is possible mis-operation of sensitive electronic equipment, and overheating of phase and particularly neutral conductors. How does this happen?
In a balanced 3-phase circuit (equal linear load on each phase), operating with a smooth 60 Hz sine wave voltage on each phase, the neutral carries the vector sum of the three phase currents, which is zero. But if one or more of the phase conductors is also carrying significant currents at harmonic frequencies (multiples of the 60 Hz fundamental), they may not cancel by vector addition, but may add in the neutral. Standard test instruments cannot even measure them.
If the harmonic currents are sinusoidal, we find mathematically that the even multiples cancel. But the odd multiples, because they are in phase, are additive, and appear in the neutral, where they can cause overheating. The current in the neutral can actually be higher than that in any one of the phase conductors. (Fires in fact have been reported that resulted from harmonics.) If the fundamental or harmonics are non-sinusoidal, such as square waves that may be caused by a pulsed power supply, mathematical analysis becomes very difficult.
The phase wires themselves may now be carrying a sinusoidal or non-sinusoidal 60Hz fundamental, plus non-sinusoidal, high frequency, pulsed currents, which may result in overheating of the phase conductors. As predicted by Ohm's Law, these distorted currents will cause distorted voltage wave forms in the building wiring system, which can, in turn, cause equipment failure in other equipment. So we have a situation where some equipment is creating problems that can affect other equipment in the building.Back to Top
Techniques That Help
There are a variety of techniques that can help prevent or alleviate the effects of poor power quality. Most simply involve better electrical designs and installation of some additional wiring. These techniques are inexpensive to install, especially when a building is undergoing construction, and they may also be cost effective during retrofits.
The most serious consequence of poor power quality, frequently, is not the physical hardware that may be damaged, but the lost data, reduced productivity and costly downtime. Like most ailments, they are much easier and cheaper to prevent than to diagnose and cure.
Most of the following techniques are part of the current IEEE recommended practice, and are contained in IEEE Standard 1100-1992 and/or Standard 142-1991 6 . Unfortunately, they are not part of any required code, although some of them should be, since safety may be affected in some cases.Back to Top
- Double-Size Neutrals, or Separate Neutrals per Phase
The sources of harmonics on building wiring have already been discussed. Harmonics are much more than an inconvenience or source of equipment malfunction. They can be a serious safety concern. Fortunately, they can be easily handled by using double-size neutrals, as recommended by the former Computer and Business Equipment Manufacturers Association (CBEMA), now the Information Technology Industry Council. Alternatively, separate neutrals can be used for each phase conductor. At least one cable manufacturer makes a Type AC or MC cable with oversized or extra neutral conductors built-in. The additional cost of oversizing the neutral is minimal. And the safety provided will be functional even if there are changes in the equipment that affect the frequencies involved.
- Harmonic Filters
Filters are sometimes most cost effective in an existing structure where rewiring is difficult or costly. The filters are used to block or trap the offending currents, lessening the harmonic loads on the wiring. But the filter design is dependent on the equipment on which it is installed, and may be ineffective if the particular piece of equipment is changed. Filtering characteristics need to be carefully designed for a given installation, and seeking professional design advice is recommended. Filters are also fairly expensive on a per-kVA basis.
- Shielded Isolation Transformers
Shielded isolation transformers are filtering devices that lessen feed-through of harmonic frequencies from the source or the load. They are a plausible retrofit technique where power problems have already been encountered, but are also quite expensive per-kVA.
- K-Rated Transformers
K-rated transformers have beefed-up conductors and sometimes cooling to safely handle harmonic loads. Alternatively, standard transformers are sometimes de-rated to allow for the extra heating due to harmonics. Depending on the conditions encountered, a load limit of as little as 50% of the nameplate rating is observed. This may be adequate to handle harmonics, but lowers effective transformer efficiency. A careful comparison of the relative costs of K-rated vs. de-rated standard transformers should be made.
- Harmonic-Rated Circuit Breakers and Panels
Overheating due to harmonics is the danger here, and beefed-up components used in these elements offer protection. Neutral buses should be rated for double the phase current.
- Separation of Sensitive Electronic Loads From Other Equipment
A dedicated "computer" circuit in each office is a good idea, at least back to the branch circuit panel. A better idea, and required in some cases, is to power sensitive equipment from separate branch circuits emanating from separate panel boards, fed from separate feeders back to the main service entrance.
The neutrals and grounding conductors need to be kept separate also. A dedicated circuit means separate phase wires, a separate neutral, with a separate grounding conductor, run in its own separate metal conduit, back to the source. See the section on conduit (below) for further discussion.
Avoid having sensitive equipment on the same circuits, or even panelboards, as motor loads. Such equipment as laser printers, copying machines and fax machines should be kept separate from computers.
- Limited Number of Outlets per Circuit
Three to six outlets per circuit is recommended instead of the thirteen allowed by Code on a 20 amp circuit. This will minimize the number and variety of sensitive equipment sharing circuitry, tend to minimize voltage drop (discussed later), minimize the chance for interaction, and leave some room for later growth or equipment changes.
- Metal Conduit
Metal conduit, properly grounded, provides shielding of the conductors from RF energy. However, do not omit the grounding conductor (green insulated copper wire), irrespective of the conduit material. It is needed for safety, as well as assurance of a continuous, low impedance path to ground. The grounding conductor is run inside the metal conduit, not outside.
All connections should be made properly and maintained to avoid possible rectification of RF at poor joints. Corrosion and joint loosening need to be addressed on a regular maintenance schedule to ensure low impedance electrical continuity at all conduit joints.
According to the IEEE Standard 142 (Green Book), rigid steel conduit offers better performance as a grounding conductor than aluminum, if a separate copper grounding conductor is not used. But the best advice is to always use a separate, full-size copper grounding conductor, irrespective of the conduit material, due to the concern for corrosion and loosening.
- Voltage Drop
Although the NEC allows up to a 3% voltage drop in a branch circuit, recommended practice is to design for no more than a 1% voltage drop at full load on branch circuits feeding sensitive equipment. Feeder voltage drop should not exceed 2%.
That means conductor gages should often be larger than required as code minimums. But a side benefit of larger conductor gage is that larger conductors frequently save enough energy, due to their lower resistance, to compensate for higher initial cost, with a short payback. Copper Development Association Inc. has free information on upsizing conductors to save energy, available on request.
Another factor to be considered in computing voltage drop is the crest factor (ratio of peak to average value of the wave shape.) In a sine wave, the crest factor is 1.414 ( ), and most tables, formulae and codes are based on this common traditional waveform. But a non-sinusoidal waveform, containing harmonics and irregular shapes, may have a crest factor of 3, 4, or higher.
Thus, the voltage drop at the current peaks may be several times higher than usually expected from the sinusoidal case. The question arises as to the value of current to employ when computing the voltage drop, as well as the value of circuit impedance at the higher harmonic frequencies. One engineer has suggested using three or four times the nameplate loads of the connected equipment to account for this increased crest factor and to compensate for the skin-effect and higher inductive reactance of the higher frequency components of current that may be present. This degree of conservatism may not be required in most cases, but prudence would suggest that phase conductors not be loaded to their published ampacity limits.
The combination of upsizing conductors beyond the gage needed for the load, combined with a 1% design voltage drop limit, should preclude excess voltage drop in the branch circuit in most cases. Again, it is a case of the extra materials being an inexpensive part of the overall installation cost during construction.
- Conductor Material The chances of problematic connections which could cause voltage fluctuations in mild cases, and catastrophic failure in extreme cases, are decreased with the use of copper conductors. Copper is the standard conductor metal against which all other conductor materials are measured, And for good reason. It has lower electrical resistance for given gage size. That means smaller gages and conduit sizes for a given load requirement. Copper oxide is a relatively good conductor, whereas aluminum oxide is an insulator. Special installation precautions are not needed, and maintenance requirements are reduced when using copper. Special corrosion inhibitors are not needed. Because of its superior connectability, there is less risk of a power quality-related failure.
- Metallic Enclosures
All metal objects that enclose electrical conductors, or are likely to become energized in the event of a fault or electrostatic discharge, should be effectively grounded to provide personnel safety, as well as equipment performance. It is best to use solidly grounded AC supply systems.
All metal enclosures, raceways, equipment grounding conductors and earth grounding electrodes should be solidly joined together into one continuous electrically connected system. All structural building steel should be bonded into a single electrically conductive mass, and connected to the required electric service ground at the service entrance, as well as the equipment grounding conductor system and the metallic cold water system. Ground in accordance with Article 250 of the NEC.
- Isolated Grounds (IG)
Isolated grounding is a loosely defined technique that attempts to reduce the chances of "noise" entering the sensitive equipment through the equipment grounding conductor. The exact methods used in IG wiring vary somewhat from case to case, and there is no defined standard method.
In a typical branch circuit, the grounding conductor of the equipment is connected to the metallic outlet box through the connection of the grounding conductor screw to the mounting yoke (mounting strap), as well as to the green grounding conductor for that circuit. It is then further connected to the metallic panelboard enclosure where the branch circuit originated. There, it can pick up noise from adjacent circuits sharing the panelboard.
In the case of an IG receptacle, usually orange colored and identified with an orange triangle symbol on its face, the grounding pin is not electrically connected to the device yoke, and so is not connected to the metallic outlet box. It is, therefore, "isolated" from the green wire ground. A separate conductor, green with a yellow stripe, is run from the insulated grounding pin of the outlet to the panelboard with the rest of the circuit conductors, but usually is not connected to the metallic enclosure (Figure 4). In some cases, the isolation may terminate here. Instead it is insulated all the way through to the ground bus of the service equipment or to the ground connection of a separately derived system, i.e., an isolation transformer.
In the opinion of many designers, the IG wiring method sometimes helps reduce power quality problems, and sometimes it makes them worse! Thus, one may consider installing the IG conductor, to be available if needed, but experiment with reverting to a solidly grounded method if proven superior.
- Ground Rings
A buried exterior ground ring is a technique to help achieve a low impedance from the building's grounding system to the earth itself, and a convenient means to connect various grounds leading from the building. One recommended approach is to bury a bare copper conductor (minimum gage is permitted to be as small as #2 AWG, but sizes of 4/0 and 250 kcmil are more often specified, and 500 kcmil sometimes used), at a depth below the frost line (36"-42" in most of the US). Larger gages increase the contact surface area, helping lower resistance. The ring is set in a trench a few feet offset from the building's footprint, and completely surrounds the structure. Ground-enhancing backfill materials (bentonite, a natural clay material, or other proprietary materials) may be used to enhance earth conductivity.
To this buried ring is connected the building steel, the lightning protection down-conductors, the grounding electrode system, any metal piping systems crossing its path, and any other grounding electrodes present.
Sometimes the ground ring is further supplemented by vertical ground rods. In the design used at a large research university, for example, triple ground rods are placed at each corner of the building, sometimes supplemented by triple ground rods at mid-points, bonded to bare 500 kcmil copper conductor, at 36-40" depth, encircling each new building. (As long as the site is open, trenches dug, and personnel on premises, it makes little sense to skimp on the ground conductor.) This university is an example of a long-term owner/occupier, in whose case the use of particular buildings may change over the decades. For what has been estimated as an extra 1-2% of the cost of construction (that includes enhanced neutrals, extra circuits, superior grounding and many of the other suggestions here), they are assured that the electrical system will serve their needs into the future, without the need for excavation or other costly retrofits that can be prohibitively expensive or impossible in their urban campus setting.
- Grounding Resistance
The grounding resistance should be checked upon installation, using a ground resistance checker (such as a Megger® ), and checked again periodically, depending on experience encountered, annually or semiannually. Significant changes in readings require further investigation as to cause and needed corrective action.
Even though the National Electrical Code alludes to a "desired" ground resistance of 25 ohms or less, that standard is based on the level of ground resistance deemed adequate to cause the overcurrent device (circuit breaker) to trip under a fault condition. Proper operation of sensitive electronic equipment is not a consideration of the Code. Indeed, if the 25 ohm level is not achieved at first, the Code allows the installer to place a second ground rod, do no further checking, and stop there. The resultant ground resistance may be 100 ohms, 200 ohms, or whatever.
Many telephone and telecommunications companies specify a ground resistance of 5 ohms or less. There is no one figure that will guarantee trouble free operation of all equipment but, in general, the lower the figure the better, with 10 ohms or less being a reasonable target for most soil conditions. During the construction phase, while the site is excavated and personnel are on the scene, it is prudent and economical to install the best grounding electrode system possible for the site.
- Depth of Grounding
Where there is insufficient real estate to work with, or under conditions of unusually high ground resistivity, deep grounds may be required. Long copper pipe-type ground rods, sometimes tens or hundreds of feet long, in bored holes, are not unheard of in rare cases. In mountaintop locations, for example, in order to achieve the target ground resistance value, it may be more economical to bore a deep ground than to spread out a shallow ground system over rocky terrain or steep slopes.
Generally speaking, deeper ground rods are more effective than shallow rods, so a twenty foot rod is preferred to a ten foot rod, etc. As Figure 5 shows, resistance falls quickly as rod length increases, due to more stable temperatures and increased moisture at lower depths.
Electrode spacing is also important. The general rule of thumb is that multiple rods should be spaced apart at least twice the length of one rod. That is, two ten-foot rods should be placed no closer than twenty feet apart.
- Lightning Protection Systems In simple terms, if part of the "path of least resistance" to ground the lightning sees is through your wiring or equipment, that is where it will flow. Lightning produces very high currents, for a short time interval, but enough to cause fires or to destroy microcircuits even miles away. The idea of air terminals, or lightning rods as commonly known, goes back to Benjamin Franklin. The purpose is to provide a convenient, controlled point for lightning to strike, and then be safely conducted to ground. To provide the least resistive path, heavy-gage copper wire should be employed in the leaders and down conductors.
- Grounding of Lightning Systems The down conductors tie directly to the ring ground described above, or other grounding electrode system, along with all building steel and electric service grounds. Use heavy-gage copper conductors to minimize impedance.
Detailed design considerations covering lightning systems are found in the National Fire Protection Association's Code #780, Code For Protection Against Lightning.
By following the recommendations above, the chances of power-quality problems are minimized. During construction or major renovation, when structures are exposed and workmen are on-site, the cost of extra materials or larger conductors is minimal. The potential savings in lost production and downtime make these precautions a good investment.
In cases where power quality problems are encountered in an existing facility, a careful study will be necessary to determine the best course of action. Solutions may be as simple as moving some loads between branch circuits, some minor rewiring, or additional branch circuits. In some cases installation of shielded isolation transformers or harmonic filters may be the best course of action. In difficult cases, professional engineering assistance is recommended.
Ring grounds, combined with vertical rods, are recommended for new construction. They are usually not practical for retrofits, especially in urban areas or where there is limited space. In those retrofit cases the best solution may be a lengthy vertical ground rod or a chemically enhanced ground rod (or rods). Make sure any chemicals or backfill materials placed in the earth are environmentally acceptable and approved by such organizations as the National Sanitation Foundation and the relevant state environmental agency.
In diagnostic testing, be sure to use test instruments capable of accurately measuring harmonic frequencies (usually called "True RMS Meters").
Additional information on power quality, including a bibliography of information sources and a video on harmonics, is available from the Copper Development Association Inc. by calling 1-800-CDA-DATA, or by visiting Power Quality or Power Quality Pulications sections of our website).
Power quality problems frequently can be avoided entirely by careful design of building systems. In existing buildings, they are sometimes alleviated or eliminated through simple, often inexpensive, changes.Back to Top
- Bill Howe, "Corporate Energy Managers Express Their Views in Second Annual E-Source Survey", E-Source Strategic Memo, SM-97-5, July 1997
- Dr. Edwin Brush, private communication, BBF and Associates, September 1995
- "Wiring and Grounding for Power Quality", Electric Power Research Institute, Palo Alto, CA
- "Wiring and Grounding for Power Quality", Electric Power Research Institute, Palo Alto, CA
- From "EC&M Practical Guide to Quality Power for Sensitive Electronic Equipment," Intertec Electrical Group, Overland Park OK (1-800-543-77771)
- Power and Grounding of Sensitive Electronic Equipment, Standard 1100-1992 (Emerald Book)©, Institute of Electrical and Electronic Engineers, Piscataway, NJ 1992 (1-800-678-IEEE)
- Courtesy, AFC Cable Systems, New Bedford, Mass.
- Courtesy Sacramento Municipal Utility District
- Courtesy Sacramento Municipal Utility District
- Intertec Electric Group, op. cit.