Saturday, July 26, 2008

EXPERT SYSTEMS

Expert System
     “an expert system is a computer program that represents and reason s with knowledge of a specialized subject with a view to solving problems directly or giving advice”.
Advantages of Expert System
    Expert system works by storing expertise of a domain expert. Then a user can interact with the system and let it solve a problem using the same logic generally used by the expert. Usually the expert systems is faster and more accurate than a human , and does not get tired or bored. A very strong benefit of Expert Systems is being able to widely distribute the knowledge of a single expert , or being able to accumulate in one place the knowledge of several widely represented experts. Some situations when ES should be considered Only one expert is available , but is needed in multiple locations. Expertise is divided among several experts. Expertise is required is a harsh environment.
An expert is retiring. Expert systems are especially helpful when a task is performed only occasionally and the expert has to relearn the procedure each time it is performed . one can use expert systems are especially helpful when a task is performed only occasionally and the expert has to relearn the procedure each time it is performed. One can use Expert systems to standardize operations . another use for expert system is as job aid for experts. This will allow them to perform more accurately, more consistently, and faster , freeing up time for the expert to be more creative in performing the task. This is especially helpful when dealing with tedious, repetitious tsks. Thus an expert system is needed to
completely fulfill a function that normally requires a human expertise or it may play the role of an assistant to a human decision maker. An expert system can be distinguished from a more conventional application program in that:
     It simulates human reasoning about a problem domain, rather than simulate the domain it self.
      It performs reasoning over representations of human knowledge, in addition to doing numerical calculations or data retrieval.
    It solves problems by heuristic or approximate methods, which unlike algorithmic solutions are not guaranteed to succeed.
   Traditional program is termed as procedure oriented programming where as for an expert system term it as object – oriented programming.

SOLAR POWER SYSTEM

     Can you ever imagine life without lights, fans, cars, computers and television or of fetching water from the well and river? This is what life would have been like had man not discovered the uses of energy both renewable and nonrenewable resources.are of different types for e.g. solar energy, wind energy, tidal energy, geothermal energy etc. Now let us focus on solar energy as it is one of the abundant forms of energy available. Power "For the successful technology, reality must take precedence over public relations, for nature cannot be fooled." Richard Feyman Energy. It has been already said that the rationale for going into space, apart from the fact that the human race must extend its limits and explore and then conquer space, has to do with retrieving energy, mainly the Sun's energy. About 80 % of the total energy demanded by our society is supplied from fossil fuels. 90% of the CO2 which is the major cause of the greenhouse effect comes from combustion. It is now widely accepted that the only way to reduce the environmental risks while sustaining the economic growth is to develop a large-scale alternative energy system which is ecologically benign.
     A scientific venture must be pursued when it follows certain logic and the solution is correct, even if technology for proper utilization is not current or not available. Such is this case. Although the human race would perhaps not be able in the very immediate future to exploit the untapped potential of solar energy, it is certainly a direction that must be
followed. Exclusive dependence on fossil fuels will inevitably lead to energy shortages. 
    It must be remembered that this scheme was one of the main determinants in choosing the location of the space colony. The liberationpoints along the Earth's path were chosen primarily for their constant exposure to sunshine Solar Energy here on Earth Why should we go into space to get solar energy and not profit directly from it here on Earth? The answer is twofold.
Solar Power Satellites:
   A possible scheme for producing power on a large scale contemplates placing giant solar modules alongside the colony where energy generated from sunlight would be converted to microwaves and beamed to antennas on earth for recon version to electric power. On ground, the microwave power is rectified and converted to the commercial electric power.
      To produce as much power as five large nuclear power plants (1 billion watts each) several square km of solar collectors, weighing more than 5 million kg would have to be assembled in the settlement. An earth-based antenna 5 miles in diameter would be required for reception. These vast assemblies are often referred to as Solar Power Satellites (SPS) .The concept of the SPS is revolutionary with a high potentiality to solve the global environmental problems, as it uses the limitless solar energy, it utilizes the space outside of the earth ecology system, and it has no by-product waste. The use of Even though one of its panels could never be deployed, Skylab effectively demonstrated solar energy.A large-scale receiving antenna, retina, is necessary to collect the microwave power from space.
The Atmosphere :
     The benign atmosphere protects us from the intensity of the sun's rays, that are filtered by our gaseous cover. That same protective effect which shields us and allows life on Earth also prevent us from fully receiving the Sun's energy. It is estimated that, in average, between 0.1 and 0.2 kW/m2 of solar energy can be received from the Sun on the Earth's surface. In near Earth space the quantity o energy that can be collected is approximately ten times as much, that is, around 1 to 2 kW/m2 in average. This first reason is obviously decisive. 
The Earth's rotation :
       But even if extra sensitive solar panels could be engineered, there is another problematic factor that complicates full utilization of the sun's energy. The rotation of the Earth, as we very well know, gives rise to days and nights, which means that during 12 hours in average no sunlight hits the surface of our planet. Because of this, solar energy devices have to trap the heat during the night period and great pains are taken to ensure that minimum heat gets lost. None of these problems will be met in space, where sunshine is constant and with far greater intensity

CONTROL OF SINGLE PHASEVOLTAGE CONVERTERS

In most power electronic applications, the power input is of 50 or 60 hertz AC voltage provided by the electric utility. It further converted to a DC voltage for various applications. The inexpensive rectifiers with diodes convert AC to DC and the output voltage is uncontrolled. The controlled rectifiers are used for providing variable/ constant output voltage. The dc output voltage of a controlled/uncontrolled rectifier should be ripple free. Therefore a large capacitor is connected as a filter on dc side. Due to this, controlled/uncontrolled rectifier has the following drawbacks; these rectifiers draw highly distorted current from the Utility and Power factor is very low. To take the edge off these drawbacks, High performance rectifiers are proposed in this paper. One of the authors has proposed a digital controller-based predictive instantaneous current control scheme for the single-phase voltage-fed rectifier and its modified analogue version. Although the control schemes obtain a sufficient performance, further simplification is necessary to realize an economical PFC converter system in practice. In this paper propose a simplified analogue controller-based predictive instantaneous current control scheme for single-phase voltage fed PFC converters, which is obtained from further modification of the original digital scheme. Since three phase converters have phase interference, a particular arrangement of the controller modulator for the three phase system is employed to avoid such phase interference. As a result, a simple predictive instantaneous current control scheme for three-phase voltage-fed PFC converters is obtained. In this paper, the predictive control methods for PFC are reviewed and their system configuration is highlighted in section II. The principles of the predictive control algorithm are discussed in same section. . In section III, the simulation results are presented. The conclusion is drawn in the last section.
CONFIGURATIONS:
A. Classification of different Topology
    Improved power quality converters are classified on the basis of topology and type of converter used. The topology-based classification is categorized on the basis of boost, buck, buck–boost, multilevel, unidirectional and bidirectional voltage, current, and power flow. The converter type can be step-up and step-down choppers, voltage source and current-source inverters, bridge structure, etc. These converters are developed in such vastly varying configurations to fulfill the very close and exact requirement in variety of applications. Some of these improved Power quality converters are improved to provide better performance from primitive configurations.
B. Current Control Techniques for PFC converters
      The current control techniques have gained importance in ac to dc converters used for high performance applications. Where the fast response and high accuracy are important. Various current control methods have been proposed and classified as hysteresis control, predictive control and linear control.

AUTOMATIC SUBSTATION CONTROL

   The electrical energy is transferred from large g enerating stations to distant load centres via various sub-stations. In every sub-station certain supervision, control and protection functions are necessary. Every substation has a control room. The relay and protection panels and control panels are installed in the control room. The various circuit breakers, tap changers and other devices are controlled by corresponding control-relay panels. In a small independent sub-station, the
supervision and operation for normal service can be carried out by the oper ator with the aid of analogue and digital control systems in the plant. The breakers can be operated by remote control from the control room. During faults and abnormal conditions, the breakers are operated by Protective relays automatically. Thus, the primary control in sub-station is of two categories.
1. Normal routine operation by operators command.
2. Automatic operation by action of protective relays and control systems.
SUB STATION CONTROL FUNCTION ARRANGED THROUGH SCADA SYSTEM:
1. Alarm Functions
    To sound alarm/annunciation regarding dangerous, uncommon events such as abnormal values of process parameters, fire, illegal entry in premises, over temperatures, low voltage of auxiliary supply, unusual happening etc. Alarms are obtained from data logger and are for alerting this operator in the control room.
2. Control and Indication
Control of two position devices such as circuit -breakers, isolators, earthing-switches,starters. Indication of ON/OFF state of the devices on control board/mimic diagrams.
Control of position of devices having positions (closed, middle open) e.g. values, input settings, indication of position on control panels.
Control positions of multi-position device e.g. tap changer, indication of position on control panels.
Indication without control.
Control without indication : e.g. raise or lower control of generator load by automatic load frequency control.
Set-point control to provide set point to a controller located at remote sub -station.
3. Data collection, recording, display.
4. Sequential operation of devices with predetermined time and conditions for operation of various devices e.g.
Auto-reclosing of circuit-breakers operation O-CO-Time-CO Operation of circuit-breaker, isolator and earthing switch in a particular sequence during opening of circuit and another sequence during closing of circuit.
5. By means of SCADA system, the operator in control centre can cause operations in a remote sub-station. The possible remote operations include :
Opening and closing of switching devices I
Tap-changing of transformers (voltage control)
 Switching of capacitor banks (voltage control)
 Load shedding (load frequency control)
6. Some of the remote operations are made automatic by one -line computer based system without human intervention e.g. Net work islanding, Backup protection. The automatic control function are segregated into :
 Interconnection functions
 Transmission line automatic function
 Distribution system automatic functions

WHAT IS SCADA AND ITS NEED:

SUPERVISORY CONTROL AND DATA ACQUISITION –we more frequently call it as SCADA. As the name implies SCADA system supervises, acquires and control data received from a distant data source from the control center. SCADA system is located in the control center and is operated in the scanning mode, communicating between the CONTROL CENTER and the REMOTE STATION by means of two-way communication channels. Such a supervisory control and data acquisition system is intended to facilitate the work of operator by acquiring and compiling information as well as locating, identifying and reporting faults. On the basis of information received, the operator makes necessary decisions via the control system he can the perform different control operations in power stations or influence the processing of 
information acquired. The main task of a modern day power system is to ensure quality and reliable power at an economic rate. Hence the system is to be updated at a very fast rate (real time mode/management), which helps to control the complex system effectively without any loss of time.
FUNCTIONS OF SCADA:
 DATA ACQUSITION- Furnishes status information & measurands data to operator
 CONTROL - Allows the operator to control the devices e.g. ckt breakers ,Xmer, tap changer etc from a remote centralised location .
DATA PROCESSING - Includes data quality & integrity check , limit check , analog value processing etc.
TAGGING - Operator identifies any specific device & subjects to specific operating restrictions to prevent from unauthorized operation.
ALARMS - Alerts the operator of unplanned events & undesirable operating conditions in the order their severity & criticality.
LOGGING- Logs all operator entries, alarms &selected entries.
TRENDING- Plots measurements on selected scale to give information on the trends e.g. one minute, one hour etc.
HISTORICAL REPORTING - To save & analyze the historical data for reporting, typically for a period of 2 or more years & to archive.
FUNCTIONAL UNITS OF SCADA:
 1.Data collection equipment.
2.Data transmission / telemetric equipment.
3. Remote terminal unit.
4. Data loggers.
5. Data presentation equipments.

HIGH FREQUENCY DC-DC CONVERTERS:

    In the past few years, the unprecedented demand for handheld electronic appliances and laptop computers with an ever-more impressive list of features and capabilities has meant more demand for processors. This demand directly translates to greater needs for higher current and power at voltages that will soon approach the fractional voltage range from the dc-dc converters in these appliances.
      These demands mean that converters must now be more efficient to allow for longer battery life, while fitting into ever-shrinking volumes and pc-board footprints. And, as if this were not a tough enough challenge for power-supply designers, better transient response and tighter load regulation are also required.
    To the experts it is clear that high switching-frequency operation — in the 5-MHz to 10-MHz range — is becoming mandatory. This single parameter holds the key to achieving designs that can meet all the demands of the next generation of portable appliances. 
     Higher switching frequencies will allow engineers to design supplies with considerably wider control-loop bandwidth, which is typically between a fifth and a tenth of the switching frequency. This is important because the wider the loop bandwidth, the fewer output filter capacitors needed, leading to cheaper design and smaller pc-board footprints. At present, a large number of output capacitors is needed in low switching-frequency applications to properly handle the large and fast transient load currents, while staying within transient voltage regulation limits. 
    As designs move to high switching frequencies, the filter inductor may also shrink, as less filter inductance is needed. This change further reduces the total volume occupied by the output filter and adds the benefit of lower cost. 
   The transition to 5-MHz to 10-MHz switching frequencies has been slow in coming due to the demands it places on almost all the components used in the design. Also, there is an immediate need for engineers to learn new processes for designing high current and high frequency converters. 
       For starters, if we really want to reap the full benefits of the high frequency approach, ceramic capacitors used in the output filter must be chosen to handle larger ripple currents for a given value and package at a lower effective series resistance for reduced output-ripple voltage. A variety of small-outline, low-profile, surface-mounted inductors are currently available for designs in the megahertz range and in the right inductance range of 30 nH to 100 nH. But some further reduction in size is still needed for currents in the neighborhood of 5 A to 10 A and even beyond to address the developing needs for the not-so-distant future. 
    This brings us to the switching power MOSFETs needed in these applications with their two ingredients, namely package and silicon technology. High frequency-power packages have matured in the last few years, and several packages are already available and can be used once this trend comes to fruition, becoming common practice. 
     Power-switching devices like MOSFETs that are used in these applications must offer low enough switching and conduction losses to achieve total power efficiencies of 85% to 95% at full current. At the same time, the control scheme must be designed to offer similar efficiencies across almost the entire range of load current to maximize the battery life. The demands on the power train create opportunities for the engineering community to develop the present topologies and/or invent new topologies and smarter gate drivers to deliver the performance levels needed. 
        In my opinion, the best approach would be “complete power modules” that are comprised of the controller, gate driver and switching devices where all the interfaces have been optimized to deliver the highest possible performance for a given application. Few such modules are available today from a small number of suppliers. Nevertheless, these modules open the door for further development to improve dc-dc converter performance in the portable and handheld market.

SOLVING MOTOR FAILURES DUE TO HIGH PEAK VOLTAGE AND FAST RIST TIME:

    The evolution of power semiconductors has been so dramatic that today an insulated gate bi-polar transistor (IGBT) can be turned on in just 0. I micro-second. This results in the voltage rising from zero to peak in only one-tenth of a microsecond. Unfortunately, there are many motors in existence that do not have sufficient insulation to operate under these conditions.
HIGH PEAK VOLTAGES
        High peak voltages can be experienced at the motor terminals especially when the distance between the inverter (drive) and the motor exceeds about 15 meters. This is typically caused by the voltage doubling phenomenon of a transmission line having unequal line and load impedance's. Motor terminal voltage can reach twice the DC bus voltage in long lead applications. When the characteristic load impedance is greater than the line impedance, then voltage (and current) is reflected from the load back toward the source (inverter). The absolute peak voltage is equal to the sum of the incident peak voltage traveling toward the motor plus the reflected peak voltage. If the load characteristic impedance is greater than the characteristic line impedance, then the highest peak voltage will be experienced at the load (motor) terminal. If the DC bus voltage is 850 volts, then motor terminal voltage could reach 1700 volts peak.
FAST VOLTAGE RISE TIMES:
Fast voltage rise times of 1600 volts per microsecond can be typical as the motor lead length exceeds just a few hundred feet. Voltage rise time is referred to as dv/dt(change in voltage versus change in time). When the rise time is very fast the motor insulation system becomes stressed. Excessively high dv/dt can cause premature breakdown of standard motor insulation. Inverter duty motors typically have more phase-to-phase and slot insulation than standard duty motors (NEMA design B).
         When motors fail due to insulation stress caused by high peak voltage and fast voltage rise times (high dv/dt) they have common symptoms. Most failures of these types occur in the first turn as either a phase-to-phase short or phase to stator short. The highest voltage is seen by the first turn of the winding and due to motor inductance and winding capacitance of the motor, the peak voltage and dv/dt decay rapidly as the voltage travels through the winding. Normally, the turn to turn voltage in a motor is quite low because there are many turns in the winding. However, when the dv/dt is very high the voltage gradient between turns and between phase windings can be excessively high, resulting in premature breakdown of the motor insulation system and ultimately motor failure. This problem is most prevalent on higher system voltages (480 & 600 volts) because the peak terminal voltage experienced often exceeds the insulation breakdown voltage rating of the motor.
STANDARD MOTOR CAPABILITIES
      Standard Motor Capabilities established by the National Electrical Manufacturers Association (NEMA)and expressed in the MG- I standard (part 30), indicate that standard NEMA type B motors can withstand 1000 volts peak at a minimum rise time of 2 u-sec (microseconds). Therefore to protect standard NEMA Design B motors, one should limit peak voltage to 1KV and reduce the voltage rise to less than 500 volts per micro-second.
SOLUTIONS
     There are several solutions available to solve this problem, each offering a different degree of protection at a different price.
1. Inverter Duty motors should be considered for all new IGBT drive installations. They offer increased winding slot insulation, increased first turn insulation, and increased phase- to-phase turn insulation. They are more expensive than standard design B motors but are the best motor for the job when it will be controlled by an IGBT variable frequency inverter. The NEMA Standard MG- I (part 3 1) indicates that inverter duty motors shall be designed to withstand 1600 volts peak and rise times of >0.1µsec. Nevertheless, it is wise to confirm the actual motor capability with the manufacturer.
2. Minimize Cable Length between the inverter and motor. Quite often this is somewhat uncontrollable, especially when the application is downhole pumping where the motor is required to be a great distance from the inverter. The longer the cable, the greater the capacitance of the cable, the lower the impedance of the cable and thus a greater mis-match will result between the characteristic line and load impedance's, resulting in higher peak voltage at the motor (load) terminals. Minimize this length whenever possible to avoid problems.
   3. Tuned Inductor & Capacitor (LC) Filters are an effective means of taming the output voltage waveform and protecting the motor. An "LC" circuit can result in the best voltage waveform but at a relatively high cost and with some future considerations. Of course these filters are "low pass shunt type filters" tuned for some specific frequency, often in the range of 1 kHz to 2Khz. Because these filters have essentially zero impedance at there resonant frequency, it is very important that the inverter switching frequency not be set too low. The threat exists that someone may vary the carrier frequency (at a later date) without consideration for the existence of a low pass filter resulting in damage to the inverter or filter. One should be very careful when applying this type of filter on the output of an inverter with variable carrier frequency. LC filters for this purpose cost approximately 3-4 times the cost of a load reactor.
   4. RC Snubber Networks can reduce the slope of the voltage waveform leading edge and reduce the peak voltage of the waveform but they have a minimal effect on the actual waveshape. They perform marginally when compared to the other solutions discussed herein. At an intermediate cost, they provide a marginal benefit. The cost of these network can be 2-3 times the cost of a load reactor.
   5. Load Reactors are the most cost effective means of solving high dv/dt and peak voltage problems associated with IGBT inverters. Typical experience is that peak voltage is limited to I 000 volts or less (actual value varies based upon system voltage). Voltage rise time (dv/dt) is typically extended to several micro-seconds resulting in only about 75 - 200 volts per micro- second rise times. Usually the load reactor is all that is needed to adequately protect the motor from dv/dt and to allow full warranty of the motor in IGBT inverter applications. (Some motor manufacturers do not offer a warranty in IGBT applications if a load reactor is not installed).
    Whether you install the load reactor at the inverter or at the motor, it will provide you with protection for your motor. It offers the best dv/dt reduction when it is placed at the inverter and this is usually the easiest place to add the reactor. Placement at the inverter also provides voltage stress protection for the motor cables. Of course there are some applications that may require the addition of the load reactor at the motor terminals. This will also provide very good protection of the motor because the IGBT protected reactor acts like the first turns of the motor. The motor is protected well in this case, however the motor cables are not protected from voltage stress.
GUARD-AC LINE/LOAD REACTORS
"Guard-Ac" Line/Load Reactors manufactured by MTE Corporation, are specially constructed with IGBT protection.
They have a 4000 volt rms(5600Vpeak) insulation dielectric strength and are approved by both CSA and UL (UL506 & UL508). Only reactors approved to UL506 have the high dielectric strength (4000 volts) required for IGBT applications.
MTE Corporation Line/Load reactors also feature "Triple Insulation" on the first two and last two turns of each coil providing over 10,000 volts strength. Our standard Line/Load Reactors are suitable for use on IGBT inverter outputs with switching frequencies up to 20Khz. 

HELPING MOTORS TO KEEP THEIR COOL:

An electric motor's insulation system separates electrical components from each other, preventing short circuits and thus, winding burnout and failure. Insulation's major enemy is heat, so it's important to be sure to keep the motor within temperature limits. There is a rule of thumb that says a 10 degree Celsius (1299 degrees Fahrenheit) rise reduces the insulation's useful life by half, while a 10 degree Celsius (50 degrees Fahrenheit) decrease doubles the insulation's life. This implies that if you can keep a motor cool enough, the winding will last forever, which ignores factors like moisture, vibration, chemicals and abrasives in the air that also attack insulation systems.
     The real issue is at what temperature the motor windings are designed to operate for a long and predictable insulation life?20,000 hours or more. NEMA, the National Electrical Manufacturers Association, sets temperature standards based on thermal classes, the most common being A, B, F and H. The accompanying table provides a summary.
      Class B or Class F insulation systems are usually used in today's industrial-duty NEMA "T frame" motors. Many manufacturers also design their motors to operate cooler than their thermal class might allow. For example, a motor might have Class F insulation but a Class B temperature rise. This gives an extra thermal margin. Class H insulation systems are seldom found in general-purpose motors, but rather in special designs for very heavy-duty use, high ambient temperature or high altitudes.
     Class A insulation, while not used on today's industrial-duty motors, was standard on industrial "U frame" motors built in the 1960s and earlier. Because Class A insulation has such a low temperature rating, older motors were required to have far lower maximum temperatures. This accounts for the perception among many long-time motor users that modern motors "run hot." In fact, they do compared with older motors, but modern insulation systems are so much better that the reliability and durability of new motors are equal to or better than older-design motors. Plus, better insulation systems have allowed motor manufacturers to put more horsepower in a smaller package.
Determining Correct Operation:
     Though many people believe they can judge a motor's operating characteristics by feeling its surface, that really isn't a very effective method. Design ratings for temperature apply to the hottest spot within the motor's windings, not how much of that heat is transferred to the motor's surface. Unless you have intimate knowledge of a specific motor model's design?including benchmark lab readings of heat runs that show "normal" surface temperatures for that specific model in exact locations on the frame?a motor's "skin temperature" provides little if any evidence of what's going on inside. This is true even if temperature measurement methods far more sophisticated than the human touch are used. In addition, for safety reasons, it is unwise to touch operating motors anyway.
     Specifying motors with inherent overload protectors, thermostats or resistive temperature devices, or installing similar protection in motor controls, can help ensure that a motor is taken off-line before winding damage occurs. Motor protection of one sort or another is advisable in almost any application. A common and reliable field test for motor heating involves checking the motor?s amp draw with a clamp-style ammeter. Use this to confirm that actual amps are less than or equal to the nameplate rating. A precise test for winding temperature is the resistance method. This involves careful measurements with sensitive equipment, calculations and several hours of time. Procedures to conduct such tests can be found in technical manuals. Or contact your motor manufacturer.
Common Sense Precautions:
     Sometimes a motor overheats because of a manufacturing or design defect. But far more often, overheating can be traced to misapplication. Overloading is the leading cause. This could take the form of using an undersized motor, a situation that may become more common as concern for energy efficiency puts the emphasis on eliminating oversized motors. Use an 80% loading as your guide. Most electric motors reach their peak efficiency at that load, and a comfortable overload margin remains.
     Other common causes of overloading include a load seizing up or misalignment of power transmission linkages. Plus, unanticipated changes in environment, aging of equipment, misuse and other factors can subject the motor to stresses for which it was not intended.
     Environmental conditions that can result in motor overheating include high ambients (especially look at the near vicinity of the motor for any heat-generating device) and high altitudes, (above 3,300 feet (1005.84m), where the "thin" air has less cooling potential). You might have to derate a motor under these conditions, probably choosing the next size up. Another environmental concern is dirt and fibers, which can clog ventilation openings, coat heat dissipating surfaces and cause a variety of mechanical problems.
     Power supply problems are another overheating cause. Low voltage will result in the motor drawing higher current to deliver the same horsepower, and the higher current means higher winding temperatures. A 10% drop in voltage could cause nearly that much greater temperature rise. Excessive or sustained high voltage will saturate a motor's core and lead to overheating as well. In three-phase motors, phase imbalances can result in high currents and excessive heat, the extreme being the complete loss of voltage in one phase (called single phasing), which, if correct protection is not in place, will result in motor burnout.
     Often overlooked as a cause of overheating is the number of start/stop cycles. It's not uncommon for a motor at starting to draw five times or more the current it does while running. This accelerates heating dramatically. Though various provisions are made relative to loading and off-time, NEMA essentially limits a three-phase continuous-duty motor to two starts in succession before allowing sufficient time for the motor to stabilize to its maximum continuous operating temperature rating. This is highly application-dependent, so it's best to check with your motor manufacturer if you're facing a high-cycle application.
     Finally, pay special attention when applying adjustable-speed inverter drives, especially if you are introducing an inverter in a system of older motors. Some additional heating to the motor windings will inevitably occur because of the inverter's "synthesized" AC wave form. A greater cooling concern involves operating for an extended time at low motor RPM, which reduces the flow of cooling air. Modern inverter-duty motors have higher insulation ratings to help alleviate this concern, and the robust insulation systems used in most of today's general-purpose industrial motors are adequate for many applications as well. In extreme cases, however, secondary cooling such as a special lower may be required.

EXTENDED MOTOR LIFE WITH BEARING CARE:

     Bearing failures are the root cause for the great majority of electric motor downtime, repair and replacement costs. Bearing and motor manufacturers are aware of the situation. Motor repair shops can attribute much of their business to bearing failures. And motor users see bearing failure as the fundamental cause of virtually every electric motor repair expense. Studies conducted by the Electrical Apparatus Service Association also demonstrate that bearing failures are by far the most common cause of motor failures.
       Knowing that shaft bearings are the Achilles’ heel of industrial electric motors is not a new idea in maintenance departments, but what is new is recognizing that something can be done to prevent most motor bearing failures.
Factors affecting bearing failure:
        Electric motors actually present a relatively easy duty for shaft bearings. The motor rotor is lightweight, yet because of its large shaft diameter, the bearings are large. For example, the bearings supporting the 140 lb. Rotor for a typical 40 hp. 1800 rpm industrial motor are so large that they have an L-10 minimum design fatigue life of 3000 years, or 10 percent of the bearings are statistically expected to fail from fatigue after 3000 years of operation. Plant operating experience, however, strongly contradicts such optimistic estimates of motor bearing life. In actual industrial environments, bearing failure is rarely caused by fatigue; it is caused by less-than-ideal lubrication. Because of contaminated lubrication, bearings fail well before they serve their theoretical fatigue life. There are many reasons for less than-ideal bearing lubrication. Lubricants can leak out; chemical attacks or thermal conditions can decompose or break down lubricants; lubricants can become contaminated with non-lubricants such as water, dust, or rust from the bearings themselves. These lubrication problems can be eliminated. Motor bearings can last virtually forever by simply providing an ideal contamination-free, well-lubricated bearing environment. Conventional wisdom teaches that such an ideal motor bearing environment can be provided by using a dry-running lip seal or using sealed (lubricated-for-life) bearings.
      Indeed, for many light-duty applications, such bearing protection techniques are often sufficient to allow bearings to last as long as the equipment itself. However, these bearing protection methods have not significantly reduced the rate of bearing failure in severe-duty industrial motors.
    Bearings in industrial applications continue to fail because of inadequate lubrication caused by lubricant loss, contamination, and decomposition and break-down. Lip seals invariably wear out well before the bearing fails, and sealed bearings inherently foreshorten the life of a bearing to the service life of the contained grease (usually only about 3,000 to 5,000 hours for most industrial services).
       Maintenance professionals may find the following suggestions on how to forestall motor hearing failure obvious, but some new techniques and technologies are available.
Lubricate Bearing at Correct Intervals: 
     Despite years of warnings from bearing manufacturers, over lubrication continues to plague many motor bearings. Too much grease can cause overheating of the bearings. The lubrication instructions supplied by the motor manufacturer will specify the quantity and frequency of lubrication. Generally, two-pole motors should be greased twice a year, four-pole and slower motors only once a year.
use the best available grease:
     The most commonly used bearing grease is polyurea-based, a low-cost, low-performance, highly compatible lubricant. However, it does not handle water well, a serious drawback for many industrial applications. It reacts readily with water and loses its ability to lubricate bearings.
      Industrial motor bearings should be lubricated with a synthetic-based aluminum complex grease. A high-quality grease pays for its additional cost in reduced motor downtime and repair costs.
Keep Out Moisture:
      Unless the motor is being hosed down or it operates in a humid environment, reasonably shielded motor bearings may not become seriously contaminated with moisture while the motor is running. However, when the rotor is shut down, moisture and condensation can collect on the surface of the bearing components. Eventually, this water breaks through the oil and grease barrier, contacts the metal parts of the bearing, and produces tiny particles of iron oxide. These rust particles make an excellent grinding compound when mixed with the grease. resulting in premature failure of the bearing because of surface degradation.
      Preventing water contamination is a major challenge to bearing housing design. Close shaft-to-endbell clearances cannot stop the movement of humid air. Contact seals will quit contacting, resulting in large gaps that allow movement of air and water vapor across the bearing. 
        Vapor-blocking bearing isolators, such as the one illustrated, are among the more successful devices presently available to prevent water vapor from entering a stationary bearing. When the motor shaft is rotating, the isolator opens, eliminating the possibility of friction and wear. However, when the shaft is stationary, the isolator closes, preventing movement of air or water across its face. With no wear from rotating friction, the seal may last indefinitely, and surely as long as the fatigue-failure life of the bearing.
Keep Out Dirt:
      Lip seals, contact seals, and frequent grease replacement help minimize the amount of dirt and other air-borne abrasives that can contaminate bearing lubricant. These solutions, however, have some drawbacks. Lip seals have a short service life, and frequent grease displacement is expensive and messy.
    One successful approach to keeping air-borne dirt and liquids out of an operating bearing is to install a labyrinth-type non-contact seal over the bearing housing. These bearing isolators, readily available from suppliers, combine a tortuous labyrinth path with impingement and centrifugal forces to trap and remove air-borne dirt and liquid; virtually no contamination can reach the bearing. Because the bearing isolator is a non-contact device, it will generally be the longest-lasting component of the motor.
      Although not intended as such, a bearing isolator could serve as an emergency sleeve bearing if the primary bearing fails, possibly preventing damage to the motor’s stator and rotor. In emergency situations, the bearing isolator can allow continued operation for a short time and still prevent the need to rewind the motor when the bearing is replaced. Bearing isolators constructed of bronze or other non-sparking materials also can prevent hazardous sparks that could otherwise occur when the bearing’s rolling elements fail.