Troubleshoot RF Immunity Problems
Simple techniques can uncover sensitive circuits.
Daryl Gerke and William Kimmel, Kimmel Gerke Associates, St. Paul, MN -- Test & Measurement World, 2/1/1999
| Your equipment has just failed an RF immunity test, one of several EMI immunity tests. Now what? You don’t have a shielded anechoic chamber with expensive amplifiers and antennas at your disposal, so you’ll have a difficult time measuring any design modifications in your design lab. You do have options, though. You can perform a basic immunity test with equipment you already have. Find some VHF and UHF handheld radio transceivers, and key them 1 to 3 ft away from your EUT. Check with any radio hams you know, as most have at least one handheld radio. (See Letter to the Editor, below.) Even a 1-W radio will generate electric-field levels of 1 V/m to 5 V/m at a distance of 3 ft, and of 3 V/m to 15 V/m at 1 ft. A useful formula to predict the electric field levels is: If your failures aren’t frequency-sensitive, you can use the radio transmitter to uncover problem areas. Key the radio transmitter and observe the EUT for disruptions in performance. As a courtesy, don’t key the radio for more than a second or two at a time, and don’t transmit on a frequency that is already in use in your area. During the compliance test, you may have found that the EUT is sensitive to RFI at specific frequencies. If so, you can use a signal generator and small broadband amplifier to find the sensitive circuit. We’ve used magnetic field “sniffer” probes for this application with some success. Using a sniffer probe, you can pick a particular frequency and poke around different circuits and cables. You can even add modulation, such as the 1-kHz modulation signal required for the European Union EMI tests. Most probes can handle 1 W or 2 W of power, but feel the probe to see if it is getting hot. If the probe gets too hot to touch, don’t use it. You may need to add a 50-V series resistor, as the magnetic field probes are low impedance devices. (A 1-W or 2-W resistor should suffice.) Move the probe close to various cables and circuits to determine their vulnerability to RF energy. Hold the probe within an inch or two of a susceptible circuit or cable. To further localize suspect circuits, you can decrease the output of the signal generator, thus decreasing the signal to the probe. Another method for cable testing involves taping a 3-ft length of wire to a cable and injecting current into the wire (Fig. 1). The wire couples RF energy into the cable through crosstalk. We’ve injected both continuous RF signals and ESD pulses on power and signal cables, with varying degrees of success.
You can accomplish the same objective with an RF current probe, but be sure the probe can handle the current so you don’t burn out the probe. If you don’t have an amplifier, these methods probably won’t work, as you generally need at least 1 W or 2 W of power. Instead, you can directly inject an RF signal from a signal generator into suspect circuits. You’ll likely need a 3-dB pad to terminate the generator, plus a series capacitor to block the DC. We recommend blocking capacitors of 0.1 mF for frequencies from 1 to 10 MHz, of 0.01 mF for frequencies from 10 to 100 MHz, and of 0.001 mF for frequencies from These methods don’t always work, and they are qualitative rather than quantitative. Nevertheless, you can often use them to troubleshoot RF immunity problems in the design lab to assess design changes. T&MW
The article "Troubleshoot RF Immunity Problems" (February 1999, p. 29) was excellent, however, it is necessary to point out that to utilize an amateur radio transceiver in the suggested manner is illegal for a couple of reasons. First, transmitting (even as little as keying) in the amateur bands by a non-licensed operator is against FCC regulations. Also, use of said transceiver by a licensed ham radio operator for any business reasons is against FCC regulations. It would be better to suggest that the person see about obtaining an inexpensive transceiver that does not break FCC regulations when used for transmission.' Judith Martin |
