Current waveforms of household appliances for advanced meter testing Ronald van Leeuwen Electricity and Time Department VSL Delft, The Netherlands rvanleeuwen@vsl.nl Gertjan Kok Flow Department VSL Delft, The Netherlands gkok@vsl.nl Helko van den Brom Electricity and Time Department VSL Delft, The Netherlands hvdbrom@vsl.nl Gert Rietveld Electricity and Time Department VSL Delft, The Netherlands grietveld@vsl.nl Dennis Hoogenboom Electricity and Time Department VSL Delft, The Netherlands dhoogenboom@vsl.nl Abstract—Recent studies showed that static electricity meters can generate wrong metering results when exposed to conducted electromagnetic inference caused by electronic appliances, consisting of waveforms with very steep rising edges in combination with large peak amplitudes. To identify more waveforms that can cause errors, we captured a large series of waveforms of common household appliances, and after analyzing these waveforms we selected a number of waveforms for testing a static electricity meter. Keywords—Electromagnetic compatibility, static meters, interference, accuracy, testing, household appliances I. INTRODUCTION Throughout Europe traditional electromechanical electricity meters are being replaced by a new generation of static electricity meters. In total over 300 million electricity meters are replaced in Europe. It is therefore of importance that the static electricity meters are reliable and insensitive to interference. Due to increased use of power electronics the amount of unwanted interference signals on the electricity grid increased. Previous research from the university of Kassel and university of Eindhoven already showed that single frequency high current tones in the 2 – 150 kHz range can cause metering errors [1,2]. Updated standards for testing static electricity meters now include tests that deal with these single frequency high current peaks [3]. More recent studies executed by the University of Twente and verified by VSL showed that electromagnetic interference caused by certain household appliances also cause metering errors, resulting in too high or too low registration of consumed energy [4,5]. In both studies a laboratory setup has been developed, based on commercially available dimmers, LED and CFL lamps and heaters, to generate these signals. The current waveform of these new interference signals is different than the signals found in the earlier work. These new interference signals are strongly- distorted waveforms in the time domain, characterised by very steep rising edges and high peak amplitudes. In the frequency domain, this results in a broad spectrum of higher harmonics starting at the 50 Hz fundamental mode and up to 2 – 50 kHz. These signals strongly differ from the earlier- found interference signals in the 2 – 150 kHz range. The questions that arise is how realistic and relevant these new signals generated by the laboratorium setups are, and whether there are other household appliances that can generate comparable signals. Therefore, in this paper we would like to answer the following questions. How do the waveforms of modern household appliances look like, and are they comparable with the earlier found inference signals? Can the most disturbed waveform found in an actual household appliance generate metering errors? To answer these questions the following strategy is developed. With the help of a number of literature sources a selection of household appliances is made. From each of these household appliances the voltage and current waveform is measured. The gathered waveforms are then analysed by determining three parameters which give a good description of the waveforms causing metering errors at certain static electricity meters. After measuring the waveforms, a sensitive electricity meter will be tested with the household appliances which according to the earlier calculated parameters could generate metering errors. During these tests the metering error of the static meter will be determined. In the final part of the paper we further analyse the waveforms that caused metering errors in order to find a mechanism that causes the metering errors. II. MEASUREMENT SETUP AND TEST METHOD A. Setup for capturing waveforms To avoid interference of other electronic equipment and to ensure reproducibility at a voltage of 230 V rms at 50 Hz, a separate power source is used. The output impedance of the power supply is verified to be comparable to the impedance of the mains socket in the laboratory. The technique to simultaneously measure voltage and current is used in recent work in testing of static electricity meters and is proven to be a reliable technique to capture waveforms [6-8]. The applied voltage over the household appliance under test is recorded by measuring the voltage between the phase and the neutral. A 150/1 voltage divider is used to scale the voltage down. A 50 mΩ current shunt is used to convert the current in a measurable voltage [8]. The voltage from both the voltage divider and the current shunt is digitized by a broadband synchronized 24-bit 2-channel ADC at a sampling rate of 1 MSa/s. For each waveform 10 cycles at 50 Hz are captured [9].