Testing above 1 GHz

Richard A. Quinnell, Contributing Technical Editor -- Test & Measurement World, 9/1/2004

Wireless communications systems, including networking and domestic telephones, have moved into gigahertz frequencies, but outside of the US, most EMC standards for other electronic devices stop at the threshold. This leaves regulatory bodies struggling to define requirements and test procedures to cover the higher frequencies. Simply extending existing methods, however, leaves EMC test engineers in a bind.

The FCC Title 47 Part 15 standard in the US requires measurement of radiated emissions up to 40 GHz. That can be challenging to implement, according to EMI compliance engineer Boris Schusterman of information-storage company EMC Corp. (Hopkinton, MA; www.emc.com). He notes that not many vendors offer equipment capable of making the necessary measurements, and the available equipment is expensive. Furthermore, he says, the test setup requires special facilities and cabling to keep signal-to-noise levels high enough to get valid results, so testing above 1 GHz is too costly for smaller labs.

For labs that do tackle the challenge, an even greater problem awaits: the time required to make the measurements. Schusterman explains that the typical practice for measuring emissions is to rotate the equipment under test (EUT) while capturing signals with a fixed-height antenna.

This practice recognizes that radiators in the EUT may interact to create directionality in the emissions, and it ensures that the maximum emissions in any direction will be identified. The problem, Schusterman says, is that the radiation patterns at the higher frequencies are small—less than 10° wide. The EUT must rotate slowly to give the test equipment enough time to scan for inappropriate signals across a wide frequency range. In practice, test engineers break the range up into narrower bands for scanning. Even so, he says, it can take up to 4 hrs to examine the EUT in the 1-GHz to 7-GHz range.

A fixed-height antenna, however, only catches signals radiating in a plane. Studies conducted by UK-based York EMC Services conclude that the EUT will need a full three-dimensional scan (Ref. 1). The studies suggest that better results could be obtained with a test configuration that uses a variable-height antenna with the EUT mounted so the antenna can view the device across a 90° range of angles as the antenna moves (see figure). The antenna's motion combines with standard rotation of the EUT while mounted in each of two orthogonal positions to provide the full three-dimensional scan.

The EMC community is seeking less time-consuming procedures. One suggestion comes from York EMC itself, which believes that statistical techniques may be appropriate. The company's proposed method is to take a limited number of measurements and then make a prediction of the maximum emission based on a statistical analysis. The result is a reduced test time but an increase in the uncertainty of the result. The company has conducted experiments that it claims validate the approach.

Meanwhile, standards for radiated emissions and testing above 1 GHz are being drafted by technical committees in Europe. Although the drafts are not yet at the comment stage, test engineers are already concerned that the final standards will cause them even more problems. "Europe wants standards that are much tighter than ours," says Schusterman, "and that's going to cause trouble with around-the-world consistency." Testing to multiple standards based on a geographic market, he adds, will simply compound the difficulty of testing above 1 GHz.