Electrical Maintenance and Safety (part 2)

<< cont. from part 1

The Elements of CBM

Real-Time Monitoring (RTM). Real-time monitoring provides a continuous stream of data from operating equipment. For example, rotating-equipment, RTM systems provide continuous monitoring of speed, vibration, voltage, current, and frequency data. Careful analysis of the collected data allows the user to diagnose incipient faults. An incipient fault is an impending failure that may not be predictable by the normal senses including sight, sound, smell, and touch. The only way to diagnose an incipient fault is through advanced testing techniques.

To be a valid candidate for the use of RTM, electrical equipment must satisfy two criteria:

1. The equipment must be critical or expensive enough to warrant the expenditures for the purchase and installation of the monitoring hardware and software.

2. Analysis of the monitored parameters (voltage, frequency, speed, etc.) must provide meaningful equipment diagnostics and prognostics.

RTM is being used widely on motors, generators, and transformers. However, not all equipment lends itself to RTM; moreover, there are a large variety of off-line tests that yield valuable, trendable results and may actually be less expensive than RTM. This is where PDM fits in.

Predictive Maintenance (PDM). In its simplest form, PDM uses test results taken from PM procedures. These results are statistically evaluated and a prognosis is developed allowing the system operators to increase, decrease, or even eliminate maintenance intervals. In addition, when properly analyzed, the test results can be used to provide a usable estimate of remaining equipment life. Data analysis Methods for CBM

The basic methodology for CBM depends upon the collection of either operational data (RTM) and/or maintenance test results (PDM). With either RTM or PDM, the collected data is then mathematically manipulated in one of several ways to determine the condition of the equipment and to develop a prediction of its future behavior. It should be noted that a properly written PDM analysis program is equally effective on data provided by RTM monitors.

Real-Time Monitoring vs. Predictive Maintenance. RTM data is constantly delivered as long as the equipment is on-line and the test equipment connected; conversely, because PDM tests may be performed only very infrequently, the PDM data set is exceptionally sparse. This means that the statistical evaluation of PDM data must use methods that are effective for sparse data sets.

Despite this challenge, PDM methods do offer one very significant advantage over RTM. Consider an insulation resistance test, normally performed on all types of insulation systems during both PM and PDM test intervals. Incipient insulation failures may manifest as only extremely small changes in current and resistance. In fact, changes of only a few milliamperes or milliohms may be the only precursors of an imminent failure. These changes are readily captured by quality PDM test equipment.

On the other hand, since RTM data is collected during the actual operation of the equipment, these very small changes have to be captured from within the normal values of cur rent and voltage. Normal operating values may exceed hundreds of volts or amperes.

Although RTM technologies and mathematical models exist that allow this level of discrimination, results are much more readily obtained using the very precise measurements made using specialized, off-line test equipment.

The ideal way to provide accurate diagnosis of impending problems is to combine RTM and PDM. These methodologies are often used together; however, if one of the two methodologies is clearly superior, it may be used alone. The development of these predictions is frequently referred to as trending.

What Is Trending and What Do We Expect from It? Trending is the act of using past data values to predict future data values. Consider FIG. 7, which shows an example of a simple, linear trending chart. This figure is a graph of insulation resistance readings (in megohms) taken every six months in June and December. As you can see, the insulation resistance tests taken through June were all in the neighborhood of 10,000 megohms. The last December reading showed a drop to approximately 7500 megohms, and the last June reading dropped again to slightly over 5000 megohms.

FIG. 7 linear trending graph.

FIG. 8 Parts per million (ppm) moisture graph.

It is important to note that every one of the readings falls within the "normal" range for this type of equipment; consequently, a technician might assume that the equipment is in good condition.

Comparison of the last two data points with the previous values clearly shows that the insulation resistance has started a downward slide that points toward complete failure, possibly within one year. Prior to the mid-1980s, records and graphs of this type were rarely kept. Rather, each reading was compared to an industry norm and given a pass or fail accordingly. Both of the last two readings would have passed using industry norms.

Although linear trending represents a major improvement over norm-based systems, even simple linear trending is not the final answer. FIG. 8 is a trending graph for parts per-million moisture in transformer insulating oil. The important data on this figure are the Upper Norm limit, the Actual, and the Predicted lines.

• The Actual curve plots the collected data over a period of years. Prior to the May 13, 2005 results, no trending or prognostication was performed. Note that there are several years when the results peaked above the Upper Norm limit line.

• The Upper Norm limit curve is a generally accepted industry value that represents the maximum moisture content that should be in the insulating liquid. Note that a correction was made within a few months each time the test values peaked above the upper limit.

• Perhaps the most important information on the graph is the Predicted curve. Note that when the last result was captured on April 4, 2005, the actual curve trend had dropped a great deal-probably due to a dehydration procedure that was performed after the norm was exceeded on May 13, 2004.

The Predicted line predicts test result values on an annual basis and is extended out many years. Like any statistical analysis, the further out it is extended, the less reliable it is.

The method used to develop the trend is a highly sophisticated mathematical algorithm that takes many variables into account, including the following:

• Previous results for this transformer

• Group results from a rich database of other transformers of the same type

• Environmental conditions where the transformer is located

Even though the specimen transformer results are trending lower in the actual, note that the predicted values show a gradual rise over the next several years. Still, the method used for this analysis is accurate enough that further testing could be delayed for a period of up to two or even three years. Two points will help to clarify the issue:

1. The trend line is counterintuitive. A visual analysis using simple trending would seem to predict a slow reduction in the moisture content. Although the specific algorithm is not available for this proprietary technique, it is clear that the more sophisticated analysis yields a result that is much more revealing than simple trending. The algorithms used have been vigorously tested and have shown that predictions are well within narrow statistical margins.

2. The last available data point for this measurement is from 2005. Although additional data points might modify the projection somewhat, rigorous testing of the algorithm has shown that over 90 percent of the predictions fall within one standard deviation out to three years beyond the last test.

This example shows that advanced trending methods can reveal useful results even from data that is not intuitively obvious like that shown in FIG. 7.

Using such advanced methods, the system owner can reasonably expect ample warning before major problems occur. A trend pattern that shows continued healthy predictions will allow the decision maker to extend the maintenance intervals-cutting the cost of the maintenance effort.

A trend pattern that indicates excessive deterioration or even failure can be used as the basis to schedule physical maintenance that will resolve the problem before it becomes an expensive, unplanned outage and/or failure.

What, then, are the types of mathematical mechanisms that can be and are being used for CBM prognostications? Fuzzy Logic. Increasing numbers of mathematical trending applications are using fuzzy logic. The principles of fuzzy logic were developed in the mid-1960s. Fuzzy logic is sometimes flippantly referred to as the "life is in shades of gray" theory. That is, any given data point or trend may be something other than true or false, good or bad, 1 or 0. Rather, the data may be some value between 1 and 0. This allows the system to more accurately match the broad spectrum of real-world values.

Fuzzy logic most often uses a set of queries instead of strict parameters for decision making. This makes the process attractive for control systems since it better resembles the way people make decisions. For example, an air-conditioning unit has set parameters for when to turn on and off based on the feedback from a temperature sensor. When the temperature reaches, say 8°F, the A/C unit will turn on and then turn off when the sensor once again reads the 7°F limit. A system implementing fuzzy logic could ask first if the temperature is rising quickly, then turn on the A/C to cool the temperature at a rate to counter act the quick rise. This allows for more control on the environment since it assumes, correctly, that the real world is not static but dynamic.

Artificial Neural Networks. In combination with fuzzy logic, artificial neural networks (ANNs) create an adaptive system for modeling system behaviors. ANNs use a nonlinear approach to statistical data modeling. They are exceptionally useful in identifying patterns in non-sparse, high-volume data sets.

As the name implies, the behavior of an ANN is designed to mimic biological neural networks by creating complex relationships among the components, or nodes, in a system.

Present-day ANNs do not self-learn in the quite same way that artificial intelligence (AI) programs can. Rather, ANNs use methodologies that strengthen (or weaken) the multitude of connections among nodes as more data become available. Thus, these networks are always improving to better fit the control system they emulate.

Classic Rule-Based (Algorithmic) Systems. A classic, rule-based approach is most useful when modeling systems that follow a clear pattern that can be quantified. An equation with set parameters, or rules, that can perfectly emulate a real-world process is rare, to say the least; consequently, most classic algorithmic systems are created by compromise between the accuracy of trending results versus the number of variables defined.

The simplest algorithmic system contains only one variable. For trending purposes, this is usually a linear equation over time where the rate of change is a set value. As variables are added, the equation becomes less rigid and can allow for other factors such as age, environment, and usage. There are various approaches to defining these additional variables based on empirical data samples, including-but not limited to-maximum likelihood estimation, inferential statistics, and the method of moments.

Combinations of Fuzzy Logic, ANN Techniques, and Rule-Based Systems. A more recent approach combines the older systems in a method that is especially applicable to sparse data sets. For example, the system used to generate the predictions in FIG. 8 applies a combination of a multivariant algorithmic approach and the artificial neural network's dynamic systems with self-training capabilities. Although there are a set number of variables and relationships among them as in a rule-based system, the values of the variables are for ever changing as in a self-training system like an ANN. Such an approach works well for sparse data sets in that each individual gains the benefit of the population for predicting future results. The impact of each individual on the population is based on the strength of the relationships so no one "rogue" unit can adversely affect the predictions of the others, a problem faced when trending most small data sets. As these data sets grow, the relationships play a greater part in the trending. When properly developed, the combined approach will work at least as well as ANNs for large data sets.

How About the Safety Aspects. Clearly CBM will improve the financial and technical efficiency of an electrical system, but how about the safety aspects of the system? There are at least two ways that CBM will enhance personnel safety:

1. Every time workers perform maintenance or testing on a piece of equipment, they are in a dangerous situation. An error can cause equipment failure either on the spot or later when some mistake makes itself known. Since using a CBM approach optimizes the testing intervals, workers will have minimal exposure to the electrical hazard.

2. By including the analysis and trending concepts, CBM will pinpoint incipient failures.

Using other methods such as a calendar-based approach, workers will not have advance warning of a failure and may be subject to injury or even death.

Maintenance Requirements for Specific Equipment and Locations

Although comprehensive maintenance procedures are beyond the scope of this section; the following should help to provide a starting point for those new to electrical maintenance or a midcourse correction for those with more experience. For detailed information, refer to the documents in the references of this section and other standard texts.

General Maintenance requirements Table 1 shows the general maintenance requirements for electrical systems that are laid out in NFPA 70E.

Some of the requirements have been slightly modified from that document to be consistent with the authors' beliefs about proper maintenance techniques.

Substations, switchgear, panelboards, Motor control centers, and disconnect switches.

Enclosures (Electrical Rooms, Buildings, and Other Such Locations). Enclosures must be kept clear of extraneous materials that could cause a tripping, fire, or any other type of hazard.

Area Enclosures (Fences and Other Such Enclosures). Fences and other such guards shall be maintained to prevent unauthorized access or accidental exposure to electrical hazards. This requirement is especially important for fences that protect the public from access to electrical facilities.

Conductors. All current-carrying conductors must be maintained so that they can carry rated current without overheating. They must also be maintained so they can carry available short-circuit current without failing and introducing arc and blast hazards.

====

Table 1 General Maintenance requirements for Electrical Systems

Topic required maintenance Qualified persons Only qualified persons are allowed to perform maintenance on electrical equipment and installations. This means that they must be trained in and familiar with the operation of the equipment, the safety hazards of the equipment, the specific maintenance procedures, and the tests that they will perform.

Single-line diagram A legible, accurate, and up-to-date single-line diagram should be maintained to assist in day-to-day operations, including maintenance, testing, and switching activities. (Note that NFPA 70E does not require a single-line diagram but does require that it be properly maintained if it exists.) Spaces about The working spaces that are required by NFPA 70 and electrical equipment 29CFr 1910.302-1910.308 must be maintained.

Grounding and bonding All grounding and bonding conductors must be maintained to ensure electrical continuity.

Guarding of live parts Enclosures around live parts must be maintained to guard against accidental contact.

Safety equipment locks, interlocks, and all other types of safety equipment (including personal protective equipment) must be maintained in proper working condition so that they provide the desired protection.

Clear spaces Access to working space and escape passages must be maintained.

(This refers to exits and aisles as opposed to actual working spaces about electrical equipment.) Identification of All electrical components and circuits should be identified with components and circuits legible, secure signs, tags, or other such devices. Operating procedures, such as switching and/or lockout-tagout instructions fall under this requirement.

Conductors and cables All types of conductors and cables, including flexible cord sets (extension cords), should be maintained so that they are not allowed to become damaged, frayed, or strained.

====

Electrical Insulation. Electrical insulation must be maintained so that it will continue to support the impressed voltage. The maintenance of insulation is very important. Insulation maintenance typically accounts for over 70 percent of all maintenance procedures.

Protective Devices. Protective devices must be maintained so that they can adequately withstand and/or interrupt the available fault current. They must also be capable of operating as they are designed to do. The next sections provide detail for many of the maintenance procedures for protective devices.

Fuse Maintenance requirements

Fuses may well be the most widely used electrical protective device. They are certainly the oldest, dating back to the late 1800s when it was noticed that the smallest piece of wire in a series of wires was always the first link to open. Although fuses are usually not electrically tested, the following specific steps should be employed during electrical maintenance:

1. Inspect fuses and fuse blocks closely for any signs of overheating that may be caused by loose connections, overload currents, or environmental conditions.

2. Make certain that the correct size fuse is being used in the application.

3. In a three-phase or two-phase fuse block, the fuses should be of the same size and from the same manufacturer.

4. Make certain that fuses have sufficient interrupting rating for the available fault current at their point of application.

5. Check the tightness of fuse mounting clips and hardware.

Although simple in application, these five easy steps can help to ensure that fuses will operate properly when called upon.

Molded-case circuit Breakers

The need for inspection of molded-case breakers will vary depending on operating conditions. Suggested inspection and testing is defined in ANSI/NEMA AB 4, Guidelines for Inspection and Preventive Maintenance of Molded Case Circuit Breakers Used in Commercial and Industrial Applications. As a part of these guidelines, AB 4 also provides some basic procedures for the inspection and maintenance of molded-case circuit breakers, by qualified persons.

Generally, maintenance on molded-case circuit breakers is limited to proper mechanical mounting, electrical connections, and periodic manual operation. Most lighting, appliance, and power panel circuit breakers have riveted frames and are not designed to be opened for internal inspection or maintenance. All other molded-case circuit breakers that are Underwriters laboratories (UL) approved are factory-sealed to prevent access to the calibrated elements. An unbroken seal indicates that the mechanism has not been tampered with and that it should function as specified by UL. A broken seal voids the UL listing and the manufacturer's warranty of the device. In this case, the integrity of the device would be questionable. The only exception to this would be a seal being broken by a manufacturer's authorized facility.

Molded-case circuit breakers receive initial testing and calibration at the manufacturers' plants. These tests are performed in accordance with UL 489, Standard for Safety, Molded Case Circuit Breakers, Molded-Case Switches and Circuit Breaker Enclosures. Molded-case circuit breakers, other than the riveted frame types, are permitted to be reconditioned and returned to the manufacturer's original condition. To conform to the manufacturer's original design, circuit breakers must be reconditioned according to recognized standards. An example of a recognized standard is the Professional Electrical Apparatus recyclers league ( PEARL) reconditioning Standards. To ensure equipment reliability, it is highly recommended that only authorized professionals recondition molded-case circuit breakers.

Circuit breakers installed in a system are often forgotten. Even though the breakers have been sitting in place supplying power to a circuit for years, there are several things that can go wrong. The circuit breaker can fail to open due to a burned-out trip coil or because the mechanism is frozen due to dirt, dried lubricant, or corrosion. The overcurrent device can fail due to inactivity or a burned-out electronic component. Many problems can occur when proper maintenance is not performed and the breaker fails to open under fault conditions. This combination of events can result in fires, damage to equipment, or injuries to personnel.

Very often, a circuit breaker fails because the minimum maintenance (as specified by the manufacturer) was not performed or was performed improperly. Small things, like failing to properly clean and/or lubricate a circuit breaker, can lead to operational failure or complete destruction due to overheating of the internal components. Common sense, as well as manufacturers' literature, must be used when maintaining circuit breakers. Most manufacturers, as well as NFPA 70B, recommend that if a molded-case circuit breaker has not been operated, opened, or closed, either manually or by automatic means, within as little as six months time, it should be removed from service and manually exercised several times. This manual exercise helps to keep the contacts clean, due to their wiping action, and ensures that the operating mechanism moves freely. This exercise, however, does not operate the mechanical linkages in the tripping mechanism, FIG. 9. The only way to properly exercise the entire breaker operating and tripping mechanisms is to remove the breaker from service and test the overcurrent and short-circuit tripping capabilities. A stiff or sticky mechanism can cause an unintentional time delay in its operation under fault conditions.

This could dramatically increase the arc/flash incident energy level to a value in excess of the rating of personal protective equipment.

FIG. 9 Principal components of a molded-case circuit breaker.

Another consideration is addressed by OSHA in 29 CFr 1910.334(b)(2), which states: reclosing circuits after protective device operation. After a circuit is deenergized by a circuit protective device, the circuit may NOT be manually reenergized until it has been determined that the equipment and circuit can be safely reenergized. The repetitive manual reclosing of circuit breakers or reenergizing circuits through replaced fuses is prohibited.

NOTE: When it can be determined from the design of the circuit and the overcurrent devices involved that the automatic operation of a device was caused by an overload rather than a fault condition, no examination of the circuit or connected equipment is needed before the circuit is reenergized.

The safety of the worker manually operating the circuit breaker is at risk if the short circuit condition still exists when reclosing the breaker as stated above. OSHA no longer allows the past practice of resetting a circuit breaker one, two, or three times before investigating the cause of the trip. This previous practice has caused numerous burn injuries that resulted from the explosion of electrical equipment. Before resetting a circuit breaker, it, along with the circuit and equipment, must be tested and inspected by a qualified person, to ensure a short-circuit condition does not exist and that it is safe to reset the breaker.

Any time a circuit breaker has operated and the reason is unknown, the breaker, circuit, and equipment must be inspected for a short-circuit condition. Melted arc chutes will not interrupt fault currents. If the breaker cannot interrupt a second fault, it will fail and may destroy its enclosure and create a hazard for anyone working near the equipment.

To further emphasize this point, the following quote is provided: After a high level fault has occurred in equipment that is properly rated and installed, it is not always clear to investigating electricians what damage has occurred inside encased equipment.

The circuit breaker may well appear virtually clean while its internal condition is unknown. For such situations, the NEMA AB4 'Guidelines for Inspection and Preventive Maintenance of MCCBs Used in Commercial and Industrial Applications' may be of help. Circuit breakers unsuitable for continued service may be identified by simple inspection under these guidelines.

Testing outlined in the document is another and more definite step that will help to identify circuit breakers that are not suitable for continued service.

After the occurrence of a short circuit, it is important that the cause be investigated and repaired and that the condition of the installed equipment be investigated. A circuit breaker may require replacement just as any other switching device, wiring or electrical equipment in the circuit that has been exposed to a short circuit. Questionable circuit breakers must be replaced for continued, dependable circuit protection.

The condition of the circuit breaker must be known to ensure that it functions properly and safely before it is put back into service.

Low-Voltage power circuit Breakers

Low-voltage power circuit breakers are manufactured under a high degree of quality control, of the best materials available, and with a high degree of tooling for operational accuracy. Manufacturers' tests show these circuit breakers to have durability beyond the minimum standards requirements. All of these factors give these circuit breakers a very high reliability rating when proper maintenance is performed per the manufacturer's instructions. However, because of the varying application conditions and the dependence placed upon them for protection of electrical systems and equipment as well as the assurance of service continuity, inspections and maintenance checks must be made on a regular basis. Several studies have shown that low-voltage power circuit breakers that were not maintained within a five-year period have a 50 percent failure rate.

Maintenance of these breakers will generally consist of keeping them clean and properly lubricated. The frequency of maintenance will depend to some extent on the cleanliness and environmental conditions of the surrounding area. If there were very much dust, lint, moisture, or other foreign matter present, then more frequent maintenance would be required.

Industry standards for and manufacturers of low-voltage power circuit breakers recommend a general inspection and lubrication after a specified number of operations or at least once per year, whichever comes first. Some manufacturers also recommend this same inspection and maintenance be performed after the first six months of service for a new circuit breaker, regardless of the number of operations. If the breaker remains open or closed for a long period of time, it is recommended that arrangements be made to open and close the breaker several times in succession. Environmental conditions would also play a major role in the scheduling of inspections and maintenance. If the initial inspection indicates that maintenance is not required at that time, the period may be extended to a more economical point. However, more frequent inspections and maintenance may be required if severe load conditions exist or if an inspection reveals heavy accumulations of dirt, moisture, or other foreign matter that might cause mechanical, insulation, or electrical failure.

Mechanical failure would include an unintentional time delay in the circuit breaker's trip ping operation due to dry, dirty, or corroded pivot points, or by hardened or sticky lubricant in the moving parts of the operating mechanism. The manufacturer's instructions must be followed to minimize the risk of any unintentional time delay.

FIG. 10 provides an illustration of the operating mechanism for such a circuit breaker and the numerous points where lubrication would be required and where dirt, moisture, corrosion, or other foreign matter could accumulate causing a time delay in, or complete failure of, the circuit breaker operation.

FIG. 10 Power-operated mechanism for a low-voltage power circuit breaker.

Medium-Voltage circuit Breakers

Most of the inspection and maintenance requirements for low-voltage power circuit breakers also apply to medium-voltage power circuit breakers. Manufacturers recommend that these breakers be removed from service and inspected at least once a year. They also state that the number and severity of interruptions may indicate the need for more frequent maintenance checks. Always follow the manufacturer's instructions because every breaker is different.

Figures 6.11 and 6.12 illustrate two types of operating mechanisms for medium-voltage power circuit breakers. These mechanisms are typical of the types used for air, vacuum, oil, and SF-6 (sulfur-hexafluoride) circuit breakers. As can be seen in these figures, there are many points that would require cleaning and lubrication to function properly.

FIG. 11 Operating mechanism for a medium-voltage air circuit breaker.

Protective relays

Relays must continuously monitor complex power circuit conditions, such as current and voltage magnitudes, phase angle relationships, direction of power flow, and frequency.

When an intolerable circuit condition, such as a short circuit (or fault) is detected, the relay responds and closes its contacts, and the abnormal portion of the circuit is de-energized via the circuit breaker. The ultimate goal of protective relaying is to disconnect a faulty system element as quickly as possible. Sensitivity and selectivity are essential to ensure that the proper circuit breakers are tripped at the proper speed to clear the fault, minimize damage to equipment, and reduce the hazards to personnel.

FIG. 12 Solenoid-operated mechanism for a medium-voltage circuit breaker.

A clear understanding of the possible causes of primary relaying failure is necessary for a better appreciation of the practices involved in backup relaying. One of several things may happen to prevent primary relaying from disconnecting a power system fault:

• Current or voltage supplies to the relays are incorrect.

• DC tripping voltage supply is low or absent.

• Protective relay malfunctions.

• Tripping circuit or breaker mechanism hangs up.

There are two groups of protective relays: primary and backup. Primary relaying is the so-called first line of defense, and backup relaying is sometimes considered to be a subordinate type of protection. Many companies, however, prefer to supply two lines of relaying and do not think of them as primary and backup. FIG. 13 shows the zones of protection for a primary relay system. Circuit breakers are found in the connections to each power system element. This provision makes it possible to disconnect only the faulty part of the system. Each element of the system has zones of protection surrounding the element. A fault within the given zone should cause the tripping of all circuit breakers within that zone and no tripping of breakers outside that zone. Adjacent zones of protection can overlap, and in fact, this practice is preferred because for failures anywhere in the zone, except in the overlap region, the minimum number of circuit breakers are tripped.

FIG. 13 Zones of protection for a primary protective relaying system.

In addition, if faults occur in the overlap region, several breakers respond and isolate the sections from the power system. Backup relaying is generally used only for protection against short circuits. Since most power system failures are caused by short circuits, short-circuit primary relaying is called on more often than most other types. Therefore, short-circuit primary relaying is more likely to fail.

Voltage and current transformers play a vital role in the power protection scheme. These transformers are used to isolate and protect both people and devices from high voltage and to allow current-carrying devices such as relays, meters, and other instruments to have a reasonable amount of insulation. It should be clearly understood that the performance of a relay is only as good as the voltage and current transformers connected to it. A basic under standing of the operating characteristics, application, and function of instrument transformers is essential to the certified relay technician.

Some overcurrent relays are equipped with an instantaneous overcurrent unit, which operates when the current reaches its minimum pickup point ( FIG. 14). An instantaneous unit is a relay having no intentional time delay. Should an overcurrent of sufficient magnitude be applied to the relay, both the induction disc and the instantaneous unit will operate. However, the instantaneous unit will trip the circuit breaker, since it has no intentional time delay.

FIG. 14 Electromechanical instantaneous unit.

The instantaneous trip unit is a small, ac-operated clapper device. A magnetic armature, to which leaf-spring-mounted contacts are attached, is attracted to the magnetic core upon energization. When the instantaneous unit closes, the moving contacts bridge two stationary contacts and complete the trip circuit. The core screw, accessible from the top of the unit, provides the adjustable pickup range. Newer designs also feature tapped coils to allow even greater ranges of adjustment.

The instantaneous unit, like the ICS unit, is equipped with an indicator target. This indication shows that the relay has operated. It is important to know which relay has operated, and no relay target should be reset without the supervisor's knowledge and permission.

As can be seen, several things can go wrong that would prevent the instantaneous unit from operating properly. These things include an open or shunted current transformer, open coil, or dirty contacts. Protective relays, like circuit breakers, require periodic inspection, maintenance, and testing to function properly. Most manufacturers recommend that periodic inspections and maintenance be performed at intervals of one to two years. The intervals between periodic inspection and maintenance will vary depending upon environment, type of relay, and the user's experience with periodic testing.

The periodic inspections, maintenance, and testing are intended to ensure that the protective relays are functioning properly and have not deviated from the design settings. If deviations are found, the relay must be retested and serviced as described in the manufacturer's instructions.

Rotating Equipment

The following maintenance requirements are taken verbatim from NFPA 70E.

Terminal Boxes. Terminal chambers, enclosures, and terminal boxes shall be maintained to guard against accidental contact with live parts and other electrical hazards.

Guards, Barriers, and Access Plates. Guards, barriers, and access plates shall be maintained to prevent employees from contacting moving or energized parts.

Portable Electric tools and Equipment

The maintenance requirements for these types of devices focus primarily on the attachment plugs, receptacles, cover plates, flexible cords, and cord connectors. These should be maintained to ensure that:

1. There are no breaks, damage, or cracks that may expose energized live parts.

2. All cover plates are in place, undamaged, and secure.

3. All terminations are complete and have no exposed strands or loose terminals.

4. That all pins and connectors in the plugs are intact, in place, and firm.

5. That the polarity is correct; that is, the hot, neutral, and ground leads are all connected to their correct terminations.

Personal Safety and Protective Equipment

The maintenance, inspection, and testing requirements for this type of equipment are all covered in detail in Section 4 of this handbook.

Electrical Safety by design

Introduction

Any electrical system design-whether for new construction or refurbishing-is always modified based on a number of competing criteria. Historically, these criteria included such factors as the following:

• Cost

• Available space

• required types of equipment

• required size and capacity of equipment

• Proper protective schemes for the power system

• Environmental considerations

If any safety considerations were included in the design, they were almost always minimum requirements as found in the National Electrical Code. Clearance distances, size of wire, and grounding requirements are examples.

Consider FIG. 15 for example. The classical approach to setting the protective devices shown is as follows:

1. The protective devices must be set in such a way that they will allow full load current to flow continuously. This means that Protective Device 1 (PD1) must carry all the Bus 1 load current without opening, PD2 must carry all the load at Bus 2, and PD3 must carry all the load at Bus 3.

2. All protective devices must operate fast enough so that no thermal or mechanical dam age occurs on any equipment that is not already faulted. If, for example, a short circuit were to occur on Bus 1, PD1 must trip fast enough so that T1 will not be damaged by the high levels of current flow.

3. When a short circuit occurs, only the closest upstream protective device to the fault must be the one to trip. This requirement is called selective tripping. If a fault occurs on Bus 1, PD1 should open, if the fault occurs on Bus 2, PD2 should open, and so on.

Setting the devices in this way allows the load current to continue on other, noninvolved parts of the system. For example, for a fault on Bus 2, PD2 would open and the load connected to PD5 would continue.

4. In the event that the first upstream device fails to trip, the next upstream device should operate after some time delay. If PD3 fails to open for a fault on Bus 3, PD2 should open up after some time delay. This helps to guarantee that the system is adequately and redundantly protected.

Incident energy from an arc is proportional to the length of time that the arc is present.

(See Section 4 of this handbook.) This means that we would like for every protective device to trip instantaneously. This requirement conflicts with criteria 3 and 4 in the previous list.

===

FIG. 15 Simple power distribution system.

Protective Device 1 Protective Device 2 Protective Device 3 Bus 3 Bus 2 Bus 1 Protective Device 4 Protective Device 5

===

Including Safety in Engineering Design Criteria

So how is a system engineered for good safety? Using the example of FIG. 15, the engineer may sacrifice selective tripping at the very high values of current flow. Where previously instantaneous tripping would be disabled at PD1 and PD2, now the engineer might enable instantaneous tripping for very high current faults. This will reduce the arc energy that would occur, but it would also eliminate selective tripping for those high current levels.

Another approach might be the inclusion of a zone-interlocking system. With this type of system, each of the protective devices includes a communications module. When the device senses an overcurrent condition, the communications module sends a signal to all the other protective devices. For a fault on Bus 3 ( FIG. 15), PD3, PD2, and PD1 would see the overcurrent and would send signals out. The signal from PD3 will override the others and force them into their normal time-delay operation. PD3 will trip instantaneously. A similar thing happens for a fault on Bus 2, except in this case PD2 will over ride PD1.

This example is but one specific situation in which an engineer can design safety into a power system. There are other broad examples:

• When considering working spaces, such as those required by NFPA 70 (NEC) in Sections 110.26 and 110.34, consider when and why such rules were developed. The working space distances found in such standards are based on two criteria: electric shock hazard and emergency egress. But how about the incident energy from an electrical arc? Even if a worker is wearing appropriate arc-flash protection, being confined in a 1-by-1-meter space is dangerous.

• What about the physical construction of switchgear and its ability to withstand a worst case electrical arc and blast? Even in the early part of the 21st century, there are many engineers who believe that standard switchgear will protect workers from electrical arc and blast. This is simply not true and never was. Unless the switchgear is designed, tested, and built according to the modern arc-resistance switchgear standards, a closed door or panel will simply not protect the worker from these hazards.

• Design systems that have adequate backup and bypass so that equipment can be taken out of service for repair and maintenance without having to stop production. This type of approach will greatly limit the number of times that a worker has to approach energized conductors.

• Avoid sizing motors, transformers, cables, and other types of equipment using the mini mum requirements of the National Electrical Code. Consider the fact that the minimum requirements by necessity can take into account only the median types of hazards that may be encountered. Short-term expenditures certainly must be taken into account; how ever, the effect of electrical accidents on long-term costs also must be considered.

Remember that one electrical accident can easily cost 10 times, or more, what it would cost to engineer the hazard out.

• There is a great deal of emphasis on utilizing optic sensors and high-speed microprocessor protective relaying to sense an arcing event and clear the circuit before the arc-flash and blast occur. This concept is not entirely true. Granted, the use of these devices will provide high-speed detection of an electrical arc event; however, the circuit can clear the fault only as fast as the electromechanical circuit breaker can operate. Maintenance of the circuit breaker must be considered a vital component for safely clearing the faulted condition, as noted previously in this section.

• Another consideration that many are using is current-limiting devices. There are many industries that have installed current-limiting reactors at the secondary of their power transformers to reduce the amount of available short-circuit current that the electrical system and equipment will see. Still another method that many are employing is the use of current-limiting fuses instead of circuit breakers. The advantage of using fast-acting fuses is that they have no moving mechanical parts that are prone to failure and require constant maintenance to help improve their reliability.

Improved Engineering standards

Starting in the 1990s, there has been a gradual improvement of engineering standards to allow for safety to be "built in." One of the new standards that is in the process of being developed is IEEE Standard Std 1814, titled recommended Practice for Electrical System Design Techniques to Improve Electrical Safety.

The working group comprises a team of highly skilled engineers, technicians, and safety professionals. The underlying concept of Std 1814 is to develop a reference for the types of procedures that engineers can use to improve the safety of their designs. They will collect, refine, and include information from a number of industry sources including:

• IEEE papers, conference presentations, workshops, and magazines

• Industry end-user best practices

• Manufacturer developments and white papers

• Trade publications

Industry efforts like Std 1814 along with continuing education for design engineers will greatly enhance safety in the workplace with improved, safer electrical equipment and installations.

Conclusion:

Electrical maintenance is critical to the safety of all personnel who must work in the vicinity of electrical equipment. Whether justified on a safety or an economic basis, all facilities should have a comprehensive electrical maintenance program.

Although the topic is too detailed for complete coverage in this handbook, a number of excellent publications and consultants are available to help with the establishment and continuation of a good electrical maintenance program.

==

Also see: Electrical safety systems