Efficiency Testing of Electric Induction Motors Anibal T. De Almeida and Fernando Ferreira ISR, Dep. Eng. Electrotecnica University of Coimbra, Polo II 3030 Coimbra, Portugal Tel: +351-39-796-218/201 Fax: +351-39-406-672 Email:
[email protected]
1 - INTRODUCTION Electric motors use over half of all electricity consumed in most countries. Typically 6080% of the electricity which is used in the industrial sector and about 20-35% of the electricity used in the commercial sector is consumed by motors. In industry, a motor consumes an annual quantity of electricity which corresponds to approximately 5 times its purchase price, throughout its whole life of around 12 to 20 years. Motors are by far the most important type of electric load. They are used in all sectors and in a wide range of applications, namely the following: fans, compressors, pumps, mills, winders, elevators, transports, home appliances, and office equipment, etc. It is their wide use that makes motor drive systems one of the main targets to achieve significant energy savings. As motors are the largest users of electrical energy, even small efficiency improvements will produce very large energy savings. Motor systems improvements produce substantial electricity savings and bring up opportunities to improve plant efficiency, to reduce the use of fossil fuels, and to reduce greenhouse gas emissions. Additionally, the profitability and competitiveness of many economic activities can be significantly improved through the use of energy-efficient motor technologies. As a pre-condition to the development of motor labelling, motor minimum efficiency standards and large-scale motor DSM programs, there is a need to assess the motor efficiency of the motors sold in the market, using as much as possible an uniform yardstick. Motor efficiency testing protocols differ around the world, and their application on any given motor can lead to significantly different efficiency values. Many countries with no domestic motor manufacturing capacity import motors from a variety of countries using different testing procedures. For instance, the Philippines, in which US AID is supporting technical assistance for motor efficiency DSM and standards programs, currently imports motors from Asia, Europe, and the U.S. Accordingly, 64% of the imported motors follow the International Electrotechnical Commission protocols (IEC 34-2), 16% follow the National Electrical Manufacturers Association (NEMA) protocols which conform to Institute of Electrical and Electronic Engineers protocols (IEEE 112), 8% follow the Japan Industrial Standards (JEC-37) and 12% follow other protocols (the percentage values are given on a dollar value basis).
2 - ENERGY EFFICIENT MOTOR SYSTEMS The vast majority of the motors used in industry are squirrel-cage induction motors due to their low cost, high reliability and fairly high efficiency. There are no electrical connections to the rotor, which means that there are no brushes, commutator or slip rings to maintain and replace. The speed of an induction motor is essentially determined by the frequency of the power supply and by the number of poles of the motor. However the speed decreases a few percent when the motor goes from no-load to full load operation. The efficiency of a motor driven process depends upon several factors which may include: Motor efficiency Motor speed controls Power supply quality System oversizing Distribution network Mechanical transmission Maintenance practices Load management and cycling Efficiency of the end-use device (e.g. fan, pump, etc.) It must be emphasised that the design of the process itself can also influence to a large extent the global efficiency (units produced/kWh). 2.1 Motor Losses There are four different kinds of losses occurring in a motor: electrical losses, magnetic losses, mechanical losses and stray load losses. These losses can be reduced by using quality materials, as well as by optimising the design. The electrical losses are of the type I2R, and consequently they increase rapidly with the motor load. These Joule losses can be decreased by increasing the cross-section of the stator and rotor conductors. Magnetic losses occur in the steel laminations of the stator and rotor. They are due to histeresis and eddy currents, and vary with the flux-density and the frequency. They can be reduced by increasing the cross-section of the iron in the stator and rotor, by using thinner laminations, and improved magnetic materials. Mechanical losses are due to friction in the bearings, ventilation and windage losses. They can be decreased using low friction bearings and improved fan design. Stray load losses are due to leakage flux, non-uniform current distribution, mechanical imperfections in the air gap, and irregularities in the air gap flux density. They can be reduced by optimal design and careful manufacturing.
Figure 1 – Variation of the different motor losses, motor efficiency and power factor as a function of the load. 2.2 - Energy-Efficient Motors (EEMs) After World War II and until the early seventies, there was a trend to design inefficient motors which minimised the use of raw materials (namely copper, aluminium and silicon steel). These motors had lower initial costs and were more compact than previous generations of motors, but due to their low efficiency, their running costs were higher. By the mid seventies electricity prices started escalating rapidly. Most of the large motor manufacturers started offering a line of energy-efficient motors (EEMs) besides standard efficiency motors. Efficiency improvements have also been achieved in the standard motor range. Recently the efficiency of the motors has already started to be a factor of competition among motor manufacturers. EEMs feature optimised design, more generous electrical and magnetic circuits and higher quality materials. Incremental efficiency improvements are still possible with the use of superior materials and optimised computer aided design techniques. EEMs typically have 30-50 % lower losses than the equivalent standard motors. Due to their lower rotor resistance, EEMs normally have lower starting torque than standard motors. EEMs normally carry a price premium of around 20-25% in relation to standard motors. This typically translates into a price premium of 5-20 US$/kW. In a new application, and for a motor with a large number of operating hours, the paybacks are normally under two years.
2.3 - System Considerations A number of important but often overlooked factors which may affect the overall motor efficiency include: power supply quality (high-quality power supply), careful attention to harmonics, system oversizing (proper equipment sizing), the distribution network that feeds the motor (attention to power factor and distribution losses), the transmission and mechanical components (optimised transmission systems), maintenance practices (careful maintenance of the entire drivepower system) and the match between the load and the motor. Power Supply Quality Electric motors, and in particular induction motors, are designed to operate with optimal performance, when fed by symmetrical 3-phase sinusoidal waveforms with the nominal voltage value. Deviations from these ideal conditions may cause significant deterioration of the motor efficiency and lifetime. Such deviations include: Voltage Unbalance Voltage unbalance wastes energy: it leads to high current unbalance which in turn leads to high losses. A phase unbalance of just 2% can increase losses by 25%. Additionally, long operation under unbalanced voltage can damage or destroy a motor (that is why many designers include phase unbalance and phase failure protection in motor starters). Another negative consequence of unbalance is the reduction of the motor torque. Undervoltage or Overvoltage When the motor is running at or nearly full load, voltage fluctuations exceeding 10% can decrease motor efficiency, power factor and lifetime. Harmonics Under ideal operating conditions, utilities supply pure sinusoidal waveforms (50 Hz frequency in Europe). However there are some loads, namely VSDs and other power electronic devices, arc furnaces, saturated magnetic cores (transformers, reactors), TVs and computers that cause voltage distortion. The resulting distorted waveform contains a series of sine waves with frequencies that are multiples of the fundamental 50 Hz frequency, the so-called harmonics. Harmonics increase the motor losses and noise, reduce torque, and cause torque pulsation and overheating. Vibration and heat can shorten the motor life, by damaging bearings and insulation. System Oversizing The load factor probably is one of the most discussed factors on motor system efficiency. Designers usually tend to oversize motor systems to warrant the capability of meeting the system peak requirements. They assume that the extra capacity cost is a small premium to pay to insure the system will be able to cope with the maximum mechanical demand.
Oversized motors will allow high starting torque loads, they will be able to accommodate load fluctuations, and they will be able to operate under adverse conditions, such as voltage unbalance. If the motor is significantly oversized and operates at partial load, its efficiency may be reduced, meaning an increment in energy consumption, poor power factor and additional costs. As can be seen in Figure 1, the peak efficiency occurs normally around 75% of the full load for most motors except for the smallest motors, and above it does not vary significantly. In most motors the efficiency starts dropping significantly below 50% of the full load. Some manufacturers recommend the replacement of motors operating below 50% of the full load by a smaller more efficient motor.
3 - MOTOR EFFICIENCY TESTING The efficiency of electric motors can be either measured directly or indirectly. Direct methods are generally more accurate, but its use has not been widespread around the world due to the need to minimise the costs associated with setting up the testing laboratories. 3.1 Direct Methods (IEEE 112-B, CSA-390) The measurement of the efficiency of electric motors can be made directly using the equation:
Efficiency % = Mechanical Output Power x 100 % Electrical Input Power
Therefore it is required to measure both the mechanical output power and the electrical input power. The electric input power can be measured with good accuracy with a simple set up and with moderately priced equipment. The mechanical power is given by the equation:
Mechanical Power = Torque x Angular Speed
Whereas speed measurement is a relatively simple procedure requiring inexpensive equipment to achieve accurate results (±1 RPM), torque measurement requires a more elaborate set up and much more expensive equipment to provide accurate results. The measurement of the torque normally requires coupling the motor to a dynamometer, which has the possibility of creating a controllable variable load, fitted with an accurate torque transducer. In North America the prevailing testing methods are based on the direct measurement of the efficiency. The Institute of Electrical and Electronic Engineers (IEEE) Standard 112B is the standard accepted both by the National Electrical Manufacturers Association (NEMA Standard MG-1) and by the US Department of Energy. In Canada the Canadian
Standards Association (CSA) Standard 390, follows a very similar procedure to the IEEE 112-B Standard. 3.2 Indirect Methods (IEC 34-2, JEC 37) In order to avoid the complexity and associated expenses of torque measurement, the motor efficiency can also be indirectly measured through the following equation:
Efficiency % =
Electrical Input Power- Losses x 100% Electrical Input Power
This calculation requires the measurement of the motor losses. Most of the motor losses (copper, iron, mechanical) can be measured with fairly good accuracy. However, the remaining losses (stray load losses) cannot be accurately measured. In indirect efficiency testing methods such as the International Electrotechnical Commission (IEC) 34-2 Standard, the stray losses at full load are arbitrarily assumed to be equal to 0.5% of the full load input power. This assumption grossly underestimates the stray load losses, particularly for small motors. The error is also larger in 60 Hz than in 50 Hz systems. In the Japanese Standard JEC 37, the error is even larger, since the stray load losses are totally ignored in the indirect measurement of the efficiency. Because of the way in which the stray losses are assessed, the indirect efficiency testing measurements generate efficiency values which can be several points above the values measured with direct efficiency methods. 3.3 Revision of IEC 34-2 Standard The availability of different testing standards in different parts of the world, is an obstacle to global competition and makes more difficult the implementation of motor efficiency programmes. The rationale for indirect measurements – to save costs in the testing equipment – is becoming less and less of an argument, because of the much larger costs associated with the design and manufacture of electrical motors, which is dominated by large companies. IEC 34-2 standard is currently being revised and taking into account the outcome of the IEC Rotating Machine Technical Committee meeting held in Montreal in July 1998, it is expected that in the near future, a new testing standard will become available giving two options for evaluating efficiency: -Method 1: the preferred method will use direct testing in a way very similar to IEEE 112-B -Method 2: if direct testing cannot be carried out, the stray load losses are assumed to be much higher than at present (going from 3% of full load input power for a 1kW machine, to 1.5 % of input power for a 200 kW motor). This assumption will penalise the manufacturers who keep on carrying indirect measurements This ongoing revision of the IEC Standard makes appropriate that new testing laboratories should be prepared to make efficiency measurements in accordance both with direct and indirect methods.
3.4 Comparison of Testing Standards The above mentioned motor efficiency testing protocols lead to different efficiency values due to the way in which stray load losses are evaluated. Whereas in North America the NEMA standard provides for direct measurement of the motor efficiency, in other parts of the world indirect measurement methods are used. This requires the measurement of the motor losses, which is a fairly accurate procedure with the exception of the measurement of the stray load losses. In IEC 34-2 standard, the stray load losses at full load are arbitrarily set at 0.5% of the nominal power consumption. In JEC-37 the stray load losses are totally ignored when evaluating the motor efficiency. With the NEMA standard the total losses are measured and it is possible to infer stray losses. Not surprisingly these different methods lead to differences in the measured efficiency values, which can be above 3%. This difference is unacceptable since it may be comparable to the efficiency gap between an energy-efficient motor and a standard motor. There is a need to promote the development of uniform world-wide testing standards to measure the efficiency of motors, since testing standards are significantly different depending on the test method. Table 1 shows an example of the efficiency results achieved, with the same set of 4 motors, by the different test methods used in Japan, Europe and USA.
5
JEC 37, Circle Diagram 88,8
IEC 34-2 Loss Summation 88,3
IEEE 112, Method B 86,2
10
89,7
89,2
86,9
20
91,9
91,4
90,4
75
93,1
92,7
90,0
hp
Table 1- Efficiency Determined by Different Testing Standards in Japan, Europe and USA Source: R. E. Osterlei, Proceedings of the National Conference on Power Transmission, Gould Inc., St. Louis, MO, 1980.
4 – MOTOR TESTING LABORATORY 4.1 Electrical Requirements The supply voltage should have the voltages with an unbalance less than 0.5% and the voltage magnitude should also be within 1 % of the nominal value. These requirements normally dictate the need for a voltage stabiliser.
Harmonic distortion can increase steeply the losses of electrical motors and therefore decrease the efficiency values of the motors being tested. The total harmonic distortion of the 3 phase supply should be under 1%. If there are nearby factories with arc furnaces, large variable speed drives or other significant non-linear loads, power quality may be low. In this case, there may be a need to install harmonic filters to reduce harmonic distortion to under 1%. 4.2 Instrumentation Accuracy Requirements Table 2 shows the minimum accuracy requirements of the instruments used to test the motor efficiency. It can be seen that the requirements for IEEE 112-B are significantly stricter than for IEC 34-2. For this reason and taking into account the revision of IEC standard, the minimum accuracy requirements for new laboratories should be based on the requirements of IEEE 112-B. Table 2: Instrumentation accuracy of efficiency testing standards Unit
IEC 34-2
IEEE 112 (Method B)
Power,W Current, I Voltage, V Frequency, Hz Speed, RPM Torque, N m Ohms Temperature, ºC
±1,0% ±0,5% ±0,5% ±0,5% ±2 --±0,5% ±2
±0,2% ±0,2% ±0,2% ±0,1% fN ±1 ±0,2% ±0,2% ±1
4.3 Data Acquisition and Control System Figure 2 shows the diagram of a possible configuration for a data acquisition and control system for motor testing, which is able to perform in an automated mode the efficiency and power factor test from no-load (or near condition due to minimum residual torque given by the dynamometer) to full-load. Inputs to the data acquisition system will include speed and torque (both usually provided by the dynamometer), operating temperatures (provided at least by 4 temperature sensors), currents, voltages, power factor, active power and reactive power (provided by a three-phase power analyser). The control system can be programmed to change the load applied by the dynamometer to the motor. The software can generate automatically the efficiency and power factor curves as a function of the load
Figure 2 - Diagram of a motor testing facility controlled by a personnal computer
5 - STANDARD TEST PROCEDURES AND OTHER RELEVANT PUBLICATIONS National Electrical Manufacturers Association (NEMA) MG-1: Motors and Generators This document provides guidelines and practical information on safety, testing, construction and operation of electric motors and generators. MG-2: Safety Standard for Construction and Guide for Selection, Installation and Use of Electric Motors This document sets design standards for specific motor and generator safety features, their proper selection, installation and use. MG-10: Energy Management Guide for Selection and Use of Polyphase Motors This guide provides information on the proper selection, installation, operation, and maintenance of polyphase induction and synchronous motors. MG-13: Frame Assignments for Alternating Current Integral-Horsepower Induction Motors This standard assists users in selecting the proper frame size, horsepower and speed when selecting motors for specific applications. Available from: National Electrical Manufacturers Association (NEMA) Publications Distribution Center P.O. Box 338 Annapolis Junction, MD 20701-0338, USA Tel: (301)604-8002 Fax: (301)206-9789 Institute of Electrical and Electronics Engineers (IEEE ) IEEE 112: Standard Test Procedures for Polyphase Induction Motors and Generators This standard defines the proper testing and reporting of induction electric motor and generator performance. It is the basis for most induction motor efficiency testing in North America.
IEEE 739: Recommended Practice for Energy Conservation and Cost-Effective Planning in Industrial Facilities This book provides guidelines for the correct design, operation and maintenance of industrial and commercial mechanical systems for cost effective energy conservation and efficiency. Available from: IEEE Customer Service 445 Hoes Lane, P.O. Box 1331 Piscataway, NJ 08855-1331, USA Tel: (800)678-IEEE; (908) 981-1393 Fax: (908) 981-9667
International Electrotechnical Commission (IEC) IEC 34-1: Rotating Electrical Machines Part 1: Ratings and Performance IEC 34-2: Rotating Electrical Machines Part 2: Methods for Determining Losses and Efficiency of Rotating Electrical Machinery from Tests Available from: International Electrotechnical Commission Bureau Central de la Commission Electrotechnique Internationale 1, Rue Varembe Geneva, Switzerland
BIBLIOGRAPHY: “Energy Efficiency Improvements in Electric Motors and Drives”, A. De Almeida, P. Bertoldi and W. Leonhard, Springer Verlag, Berlim, Germany, 1997
