Cross wires

Cross wiresThere are a few angry equipment makers out there as the EU ponders a change to the new noise on the mains standard. Richard Wilson finds cause for complaint
An annoying humming noise has been rumbling away on the electricity mains for a number of years. Like many similar noises we become used to it and only notice it when something changes.
That is exactly what has happened with moves to draw up a European standard for mains harmonics.
As Kevin Packard of switched-mode power supply manufacturer Kingshill Electronics describes mains harmonics: “It is a form of electronic pollution which should be eliminated or drastically reduced.”
Manufacturers of electronic equipment from PCand TVs to industrial rack systems, or any equipment with a mains AC supply, are faced with the task of making their equipment conform to the rules. governing harmonic noise on the mains. European standards like the proposed EN61000-3-2 mains harmonic standard is the tool the authorities will use to ensure that all manufacturers tackle the problem.
It is the manufacturers of the AC/DC power supplies which find themselves at the sharp end. But they have a double problem. Not only must they tackle noise emissions in their products, but they must also deal with a standards situation which is far from helpful.
A product manager at another AC/DC supply firm described the 61000-3-2 standards issue as “very scary”. What is frightening power supply manufacturers is that after seven years of working towards the 61000-3-2 standard a last minute change to the rules could add unwanted and some say unnecessary design costs to their products.
The new standard does not become mandatory until 2001 in the meantime manufacturers must comply with the more generic, and arguably less exacting EN60555-2 harmonic standard within the EMCdirective. But with the 2001 deadline approaching the last thing designers need is a change to what is already a challenging standard.
What is even more annoying to the industrial and profession equipment manufacturers is that uncertainty over the standard is being caused by the lobbying of the powerful ITand consumer electronics sector.
They have used their European muscle to persuade the standards authorities that there should be a four year delay before the more onerous 61000-3-2 standard applies to IT equipment. In the meantime the professional equipment sector, which is a smaller portion of the electronics equipment market and hence a much smaller producer of harmonic noise on the mains than the IT sector, must struggle with the extra cost of the meeting 61000-3-2 in 2001.
“The cost to the industry due to the indecision has been huge,” adds Packard, who is chairman of PSMA’s technical committee.
Behind the scenes there is anger amongst power supply manufacturers who believe they are being unfairly messed around due to the indecision of the standards authorities. This will all come to a head in March when the industry will see if the IT sector has succeeded in getting its way. Noise measurement on switched-mode power supplies The basic operation of a switched mode power supply involves switching a current through the primary winding of a transformer at a very high frequency, typical 250kHz. This will induce a current to flow in the secondary winding of this transformer which is then rectified to produce a DC output. The voltage waveforms around the transformer and the output rectifiers are predominantly square wave in nature and the fast switching edges can generate RF noise. Careful filtering and layout of circuits within the PSU minimise the effect of this RF noise on the output voltage but it is inevitable there will be some evidence of these switching edges in the output ripple waveform. In general, output noise and ripple is usually measured using an oscilloscope with the measurement bandwidth limited to 20MHz. Noise and ripple specifications from some manufacturers are only valid for frequencies up to 20MHz. To measure this noise it is essential to adopt the correct techniques, if rational and consistent results are to be obtained. A standard scope probe has a relatively long ground connection which is inductive and can act as an antenna to any RF signals nearby. Any RF pickup in the ground lead will be manifest as an increase in the apparent measured output noise amplitude which results in potentially substantial measurement errors. This error can be minimised by careful use of the probe and its ground connection. The optimum measurement technique is relatively straightforward and involves removing both the clip and ground wire from the probe. The ground connection to the probe is made using a small coil of wire (e.g. tinned copper wire) fashioned so that it fits snugly around the ground collar near the tip of the probe. The new ground wire should be as short as possible to reduce the high frequency noise pickup. Careful installation and earthing of the power supply unit within the system should help to minimise the effect of noise. This noise can be further reduced by the addition of decoupling capacitors on the power supply output terminals. However, it is important to keep the component legs as short as possible if any benefit is to be achieved and the capacitor must be a low ESR, high frequency type, eg, ceramic. This helps reduce any radiated noise in the PSU connection leads.
Based on an application note from Astec.

Leave a Reply

Your email address will not be published. Required fields are marked *