Technical Tidbit - June 2003

High Frequency Measurements Web Page
Douglas C. Smith

Address: P. O. Box 1457, Los Gatos, CA 95031
TEL:      800-323-3956/408-356-4186
FAX:      408-358-3799
Mobile:   408-858-4528
URL:      www.dsmith.org
Email:    doug@dsmith.org

Technical Tidbit - June 2003
A Simple Horn Antenna for Emissions Troubleshooting (test equipment need not be expensive)

Figure 1. Can Horn Antenna - Side View

Abstract: Test equipment does not always have to be expensive. A simple horn antenna made from an asparagus can is described and its use outlined. This device has been successfully used for debugging emissions problems in new designs.

Discussion: As equipment designs increase in speed, emissions above 1 GHz have become more important. There is a multitude of pre-compliance tests and simple equipment that have worked well below 1 GHz for years, but more of this type of equipment is needed for the higher frequency ranges. Radiation above 1 GHz can occur in narrow beams and can be difficult to find without expensive antennae. Figure 1 shows a simple horn antenna constructed from an asparagus can which turns out to have good dimensions for such a troubleshooting antenna. Its length of about two to three times its diameter helps give the antenna good directivity.

Figure 2 shows an end view of the antenna. The monopole length is the radius of the can and it is placed that distance from the back of the can. Copper tape is used to form a slit in the end to make the antenna more sensitive to the polarization of the received wave. The useful frequency range is that for which the can radius is about one quarter wavelength. Below that frequency, sensitivity falls off (waveguide beyond cutoff) and well above that frequency multiple modes of transmission can occur in the can reducing its usefulness with respect to polarization and directivity.

Figure 2. Can Horn Antenna - End View

A set of cans of varying sizes (tomato paste, asparagus, etc.) can be used to cover a wide range of frequencies.

Use the can horn antenna by positioning it a few feet from an equipment under test (EUT). Keeping it generally pointed at the EUT, move the can around the surface of an imaginary sphere enclosing the EUT, rotating the can to pickup different polarizations of an emitted wave. For large EUTs, one may want to point the antenna at various parts of the EUT from each position on the sphere. When a signal is picked up on a receiver or spectrum analyzer, one can move the antenna closer, following the signal. Often the source of the emission can be resolved this way.

Note that radiated emissions at these elevated frequencies can "bounce and scatter" in complex random interactions to the point of making the specific source difficult to locate. In this situation, the "horn" antenna can be placed directly above the suspected product area, such as gaps, seams, or gap arrays (such as finger-stock). By sliding the aperture around in contact with the product surface, the emissions can be directed into the antenna itself, increasing the resolution of the source identification.

While not exactly a "calibrated" antenna, it can find problems in the development lab simply and inexpensively.

The idea for this antenna was contributed by W. Michael King. It is based on a previous urgent need for a probe to detect problems at these frequencies when he was working in a very remote location in the Andes Mountains - in 1966! Michael has been known to say "I looked at the concept of the "feed horn" in our X-band parabola, and couldn't resist the temptation to call this a "food horn." His website is http://www.systemsemc.com.

Top of page
Home

Questions or suggestions? Contact me at doug@dsmith.org