|
Harmonics & Flicker EN61000-3-2 & 3
By David Hambley MSc. C.Eng. MIEE Principal Engineer York EMC Services Ltd
This article reviews the need to test for Harmonics & Flicker. Gives the background to Standards development and identifies a number of problems with the tests.
Why test Harmonics When an electrical apparatus presents a non-linear load to the AC power network it draws reactive power from the supply. Most modern electrical and electronic apparatus use some form of AC to DC power supply within their architecture and it is these supplies that provide the majority of non-linear loads. Contrary to popular belief “Linear” power supplies are no better than switch mode power supplies due to the fact that both topographies draw pulses of current from the AC network during each half cycle of the supply waveform. The amount of reactive power drawn by a given apparatus for example a domestic television may be small. However within a typical street there may be 100 or more TVs drawing reactive power from the same supply phase resulting in a significant amount of reactive current flow. This reactive current is not detected by the domestic tariff meter but has to be generated by the power utilities. This mismatch between the power generated and that used results in a loss of revenue to the utilities.
Furthermore 3-phase unbalance can also be created within a housing scheme since different streets are supplied on different phases. When a 3-phase system is unbalanced the unbalance current flows in the Neutral line of a star configuration. Traditionally the neutral line within the feeder was a smaller cross sectional area than the phase conductor, as it only had to carry a residual current. However with proliferation in recent years of electrical and electronic devices containing none linear supplies a dramatic increase of Neutral current has occurred. This current overloads the Neutral line causing heating and in the extreme has been known to cause burn out of the conductor.
Also the reactive current manifests itself as distortion of the voltage waveform of the ac supply. If co located apparatus is sensitive to such voltage distortion an EMC problem exists.
The test for evaluating the harmonic emissions generated by an apparatus is defined within EN (IEC) 61000-3-2.
Flicker When an electrical apparatus presents a changing load to the AC power network it draws fluctuating power from the supply. A good example of such an apparatus is a washing machine since it contains electrical heaters and electric motors both of which draw significant current. As the machine runs through its washing cycle the load presented to the supply changes as the program progresses. This changing load draws fluctuating current from the supply via the impedance of the supply wiring. A fluctuating voltage drop is therefore seen within the supply wiring. If the wiring provides power to other electrical apparatus in the locality this fluctuating voltage can affect the function of the co located apparatus. If the co located apparatus is incandescent lighting the fluctuation can manifest itself as a modulation of the light output from the luminaire, i.e. flicker.
The human eye/brain combination is particularly sensitive to such light variations particularly via peripheral vision. Flicker Perception is a measure of this sensitivity and is based on the effect of supply variations at different rates when applied to a 60Watt incandescent light bulb. The response of human eye/brain combination is most sensitive at a flicker rate of 500 changes per minute, this equating to a square wave at 8.33Hz. In extreme cases this rate can trigger an epileptic fit in vulnerable people.
In order to make measurements of any phenomena one or more parameters have to be defined. In the case of flicker, two flicker indicators have been defined. These are Short Term (PSt) and Long Term (PLt). The short-term indicator is measured over a 10-minute period whilst the long term over a period of up to 2 hours.
Like harmonic distortion if co located apparatus is sensitive to voltage fluctuations an EMC problem exists.
The test for evaluating the voltage fluctuations generated by an apparatus is defined within EN (IEC) 61000-3-3.
Problems with the tests EN61000-3-2 This standard was ratified as EN60555-2 Issue 2 in November 1994 and superceded EN555-2. It was subsequently renumbered before publication as EN61000-3-2 in July 1995. There was then considerable debate as to when 61000-3-2 superseded EN555-2 for demonstrating compliance to 89/336/EEC. This was finally resolved in November 1997 by the EU commission EMC government experts accepting the view that products out of the scope of EN555-2 need only be tested to EN61000-3-2 from January 2001.
As the standard became used it became apparent that there were serious problems with it. These revolved around interpretation of the wording of the standard resulting in test equipment manufacturers implementing test software/firmware in different ways. The inevitable outcome of these interpretations was that different results were possible at different test houses.
An Independent consultant and United Kingdom Accreditation Service (UKAS) assessor did considerable work in defining the problems. The EMC Test Laboratories Association (EMCTLA) then organized a Round Robin in which a standard piece of equipment under test (EUT) was tested at a number of test laboratories. As expected this work showed considerable variation in results particularly when the fluctuating harmonic test was being performed.
It was recognized that all the interpretations of the standard were valid and so UKAS issued a Technical Policy Statement (TPS21). This document required laboratories to provide in the test report the interpretation used by the particular test equipment. This obviously did not solve the problem of different results being obtained at different laboratories and hence a manufacturer could obtain a “pass” at one lab and a “fail” at another.
In an attempt to ease the problem the European standards making body CENELEC drafted amendment A14. After 2 years of “will they, won’t they” A14 was ratified and published in double quick time for immediate implementation 1 January 2001. This amendment dramatically changes the standard resulting in the majority of apparatus requiring testing to the benign Class A limit. There is a 3-year transitional period during which manufacturers may opt for the standard as is or with A14 applied. This period also allows the test equipment manufacturers and test laboratories to update their systems to implement the amendment. Subsequently the standard was issued as EN61000-3-2 (2000), this edition contains all the amendments and becomes mandatory January 2004.
EN61000-3-3 This standard was ratified as EN60555-3 Issue 2 in July 1994 and superceded EN555-3. It was also renumbered before publication as EN61000-3-3 in July 1995 and became mandatory from January 2001. Amendment A1 was published in June 2001 with a date of cessation of presumption of conformity to 89/336/EEC of 1/5/2004. This amendment makes changes to the limits that effectively add an inrush current test at switch on. It is generally agreed within the EMC test community that most apparatus if not fitted with a soft start will be non compliant with this test. Again this anomaly has been investigated. The EMCTLA advice to manufacturers is to have the inrush current test performed during the transitional period thus the need for a soft start for the apparatus may be assessed.
From a practical point of view there are problems with the tests required by the standard.
Figure 1
With reference to figure 1 the test relies on a statistical analysis of the measured data within the Flicker meter. The person performing the test therefore relies upon the test equipment manufacturer’s correctly interpreting the statistics. There are however further influences that have a significant effect on the measurement. The most crucial of these is the standard impedance.
The flicker meter measures the voltage at the input to the EUT and then calculates various parameters from this value. Therefore the absolute accuracy of the impedance is important since the voltage drop across the impedance is directly proportional to the value of the impedance. There is no tolerance specified but the uncertainty of the measurement must be 8%. The nominal value of the impedance is 0.47-ohm hence minor changes within the system can have significant influence on this value.
The stability of the impedance both long term due to aging and short term due to ambient temperature have to be considered. More importantly perhaps is the change of impedance due to self-heating of the components making up the complex impedance. These components can potentially carry 16A and so have to be rated accordingly. Finally the wiring and the non-zero source impedance add to the nominal value of the impedance increasing the error in the measurement.
Solutions The need for verification With the problems associated with these tests it is beneficial to have a means of verifying the test system. These verifications should enable any gross errors between systems to be spotted and track any drift within a system. Some commercially available systems have an in built verification. If however these verifications do not use an independent “reference” then errors may be masked. Use of a “Standard” EUT overcomes the objection to the in built system. In response to this need York EMC Services designed a standard EUT for its own use. The HFG01 is the development of this in house verification and is now available to the EMC testing community.
For product details see page 25. To download Application Note in PDF format log on to home page of this journal's web site www.compliance-club.com
|
