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: adealmeida@isr.uc.pt

1 - INTRODUCTION

Electric motors use over half of all electricity consumed in most countries. Typically 60-
80% 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 112-
B 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.

JEC 37, IEC 34-2

hp Circle Loss IEEE 112,
Diagram Summation Method B
5 88,8 88,3 86,2

10 89,7 89,2 86,9

20 91,9 91,4 90,4

75 93,1 92,7 90,0

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 ±1,0% ±0,2%

Current, I ±0,5% ±0,2%
Voltage, V ±0,5% ±0,2%
Frequency, Hz ±0,5% ±0,1% fN
Speed, RPM ±2 ±1
Torque, N m --- ±0,2%
Ohms ±0,5% ±0,2%
Temperature, ºC ±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