EMC tests at equipment level

Carrying out electromagnetic compatibility tests requires thorough preparation. The difficulties encountered can be many and unexpected. Lack of preparation, installation of the equipment under test which may be haphazard or not fixed, the sensitivity of the means used to control the equipment under test, the electromagnetic pollution generated by this control equipment, the operating mode of the equipment are all elements which can lead to tests which do not enable a system to be tested under the expected conditions.

All these factors lead to test configurations that complicate the reproducibility of a test (problem of an identical installation for tests carried out in different laboratories, or different Faraday cages).

These reproducibility problems are all the more obvious when testing (CE marking) electric field measurements in a Faraday cage at distances of one or three meters, with a fixed antenna height (usually one meter) and the front face of the equipment (by convention) presented to the antenna. These tests need to be confirmed by free-space tests at a distance of ten meters, with a variable antenna height of between one and four meters and a variation in the face of the equipment presented to the antenna that can vary by 360 degrees. The aim of these variations is to maximize each measured frequency. For measurements in excess of a few hundred megahertz, variations of several tens of decibels can be observed, due to the directivity of the measured disturbances (e.g. radiation from a housing slot).

One of the other problems encountered is that equipment tested in the prototyping phase is sometimes far removed from its future state of industrialization, which makes it impossible to determine the conformity of the final equipment. Tests carried out on electronic boards with wire modifications show that they behave differently from final boards. Indeed, the V-cycle shows that there is no point in testing a prototype whose representativeness cannot be demonstrated. However, many companies have not yet adopted the V-cycle in their EMC development quality process. What's more, many designers don't realize the impact that a simple change in capacitor technology can have on some of the results obtained in EMC testing.

All these tests can take a long time, which has an impact on costs. The aim of this article is therefore to raise awareness of good installation practices, as well as to identify the key points to avoid mistakes that lead to the need to repeat all or part of a test campaign.