yorkemc.co.uk » Consultancy & Research » EMC Research »

Development of improved test methods for assessing the EMC emissions of luminaires and ancillary devices

A report for the Radiocommunications Agency by York EMC Services Ltd, University of York

Authors: I D Flintoft, A J Rowell


The European Energy Directive which is due to come into force in 2004 will require that reduced power lighting technologies become more widely used. The major impact of this on the design of luminaires is that more extensive use will be made of electronic switching power supplies and control ballasts in lighting.

In this study, the radiofrequency emissions from lighting technologies have been considered and measurements made on a number of different luminaires and lamps.

The results show that there may be significant emissions at frequencies above 30MHz, a frequency band presently not considered in the lighting EMC emissions standard - CISPR15. Investigations showed no significant emissions to be present at frequencies in excess of 300MHz for commonly available lighting equipment.

There is therefore the need to introduce a test into CISPR15 to cover frequencies above 30MHz. Several possible test methods were considered, but for most luminaires the only effective measurement of interference potential is to perform a radiated emissions test similar to that specified in the generic EMC standards and CISPR22. The fluorescent tubes themselves can be a significant radiating source so this limits the usefulness of any of the alternative test methods considered.

A propagation model was applied to the existing (below 30MHz) limit and the likelihood of interference to radio services investigated. This was then compared to likely interferences which would occur above 30MHz with the new proposed limit. The proposed new limit was found to give superior protection to radio services above 30MHz.

In summary the main recommendations of this report are:

  1. The introduction of a radiated emissions test in the 30 to 300MHz range, the test would use the same methods as CISPR22.
  2. The current Van Veen loop test below 30MHz is retained but an alternative test method using a standard loop antenna is specified in a future version of CISPR15.
  3. The requirement for lamp 'burn in' be removed from CISPR15.
  4. A CDN or ISN test should be specified in the 30 to 300MHz range in CISPR30 so that ballasts and power supplies can be tested prior to incorporation in lighting fittings.
  5. An additional recommendation is that a guide to EMC good practice in lighting installations be produced. This would increase the understanding of the interference effects that might be expected and give guidelines for installation to minimise these effects wherever possible.


You can download this report as a PDF file (2.11 MB).