0093-9994 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information. This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIA.2018.2869118, IEEE Transactions on Industry Applications Standard Efficiency Determination of Induction Motors With a PWM Inverter Source Emmanuel Agamloh Advanced Energy Corporation Raleigh, NC, USA eagamloh@advancedenergy.org Andrea Cavagnino, Silvio Vaschetto Politecnico di Torino, Dipartimento Energia Corso Duca degli Abruzzi 24, 10129 Torino, Italy andrea.cavagnino@polito.it, silvio.vaschetto@polito.it Abstract—Induction motor loss segregation and efficiency measurement requires loading dynamometers and other equipment such as a variable voltage sinusoidal power supply. These are expensive and not often available even when a loading device is accessible. Variable frequency drives are now widely used for operating induction machines and are more widely available and less expensive. However, their use in place of a fixed frequency sinusoidal power supply to directly measure machine efficiency is intriguing, but inherently challenging due to the PWM output voltage. This paper presents some investigations for using variable frequency drives to perform IEEE 112B and IEC 60034-2-1 tests. The use of the drive, the measurement criteria and procedures are reported and discussed. The results presented demonstrate the feasibility of the proposed concept for measuring machine efficiency with a PWM power source. Keywords— Induction machines, efficiency, sinusoidal supply, PWM supply, Variable Frequency Drive (VFD), IEEE 112B test, filtering, fundamental component, Fourier transformation. I. INTRODUCTION The determination of induction motor efficiency by loss segregation has been well discussed in literature. The relevant standards such as IEEE 112 [1] and IEC 60034-2-1 [2] have been used for many years and have been revised and improved in harmony. It is well known that these standards yield accurate results that are considered equivalent. However, the test equipment required to carry out these tests are elaborate and expensive, leading to various researchers looking at different ways of estimating efficiency rather than perform actual testing. In particular, a fixed frequency variable voltage transformer (or any other three-phase variable-voltage sinusoidal power supply) is required to smoothly vary the voltage in order to perform the no-load tests which help to determine two of the five loss contributions in the machine. The absence of such equipment means that the loss segregation tests cannot be performed. In fact, many manufacturer test facilities, motor test laboratories and motor repair facilities either lack this critical equipment or do not have one with adequate capacity. With the widespread use and availability of Variable Frequency Drives (VFD), it will be desirable if a VFD can be used in place of the traditional fixed frequency power supply to perform the induction motor efficiency test. However, due to the harmonic content of the pulse width modulation (PWM) output waveform produced by VFDs, it is not straightforward to use such a device in place of a traditional sinusoidal power supply source. This is particularly important because the aforementioned international standards assume that a fixed frequency power supply of negligible harmonic content would be used for testing. Furthermore, the introduction of time harmonics into the machine potentially introduces additional losses compared to a sinusoidal power supply. Since the motor efficiency under sinusoidal supply is the target measurement, these additional losses could potentially reduce efficiency of the motor compared to the sinusoidal case. Nevertheless, the use of PWM inverters could be beneficial if techniques are developed to ensure accurate efficiency measurement. Reference [3] discussed the use of VFD for motor performance measurement. In a two-part paper, the authors tested two large machines, 0.5MW and 1MW, using an inverter power supply. Several machine performance parameters including efficiency and locked rotor torque were measured. However, no sinusoidal test results were reported and therefore results of the tests with an inverter supply was compared to a MATLAB simulation and “algorithmic induction motor analysis program”. The referenced paper therefore lacked the comparison necessary to establish validity of the use of VFDs in place of a sinusoidal power supply. In other words, although some result was obtained, it is not clear how the harmonic losses introduced by the VFD have affected the final value of efficiency. Extending the work presented in [4], this paper overcomes these shortcomings by establishing criteria necessary for comparability and goes a step further to perform tests with a sinusoidal power supply for comparison. More specifically, the constant current criteria is discussed as an alternative of the more used constant torque criteria. In addition, filtering techniques and Fourier analyses of the PWM waveforms have been identified as equivalent approaches to extract the fundamental components of electrical quantities, allowing the applicability of international standards [5]. II. BACKGROUND OF THE APPROACH From the expression in (1), the task of determining efficiency () comprises determining the total losses (P loss ) of the motor. loss out out in loss in P P P P P P      (1) stray rotor stator fw core loss P P P P P P      (2)