We know that signals from radios, televisions, or cell phones can interfere with one another. All this equipment is both a transmitter and receiver, whether intentional or unintentional. So when a product’s emissions fall into the wrong band, interference is produced.
The military developed MIL-STD-461 and -462 to test products for these phenomena. MIL-STD-462 also concerns itself with test methods, as do the industry standards IEC1000-4-X series and ANSI C63.4.
Unlike connecting a DMM to a pair of wires, it is a real challenge to take accurate and repeatable EMC measurements. First, you must learn which factors adversely affect the outcome and then take steps to overcome them. If an error can’t be eliminated, it must be stabilized, measured, and applied as a correction factor.
Test Types and Equipment
There are four types of EMC tests: radiated emissions, radiated immunity, conducted emissions, and conducted immunity, each with a different test setup. Every piece of equipment in an EMC test setup has a role to play, but it also is a potential source of error.
The choice of test equipment depends on whether you are conducting immunity testing, emissions testing, or both. It also is important to know whether you are purchasing a precompliance or a full compliance system. For this article, the focus is on radiated tests only.
Most precompliance systems are basically sniffers that check the radiated emissions of your product. The associated test equipment generally is comprised of lower-cost antennas and EMI receivers or spectrum analyzers and doesn’t include a preselector filter. Precompliance systems are excellent troubleshooting tools, but your results may not correlate well with full compliance measurements.
Radiated Emissions Test Setup
In the radiated emissions test setup the equipment under test (EUT) is the transmitter (Figure 1). An antenna placed 3 or 10 meters away picks up this signal, which then goes to the receiver through two cables and a preamplifier. The receiver is a tuned voltmeter which scans the frequency spectrum and measures the signal’s amplitude at each frequency.
The E-field signal level is derived using Equation 1 which takes into account the receiver output, the cable losses, the preamplifier gain and the antenna factor.
|
E = V + CL1 – PAG + CL2 + AF (1) where: E(dBµV/m) = measure of E-field V(dBµV) = Receiver reading CL1 (dB) = Loss in Cable 1 PAG (dB) = Preamplifier gain CL2 (dB) = Loss in Cable 2 AF(dB-1) = Antenna Factor |
Radiated Immunity Test Setup
For the radiated immunity test setup, the signal generator and amplifier drive the antenna to transmit the signal (Figure 2). The EUT is the receiver. A power meter monitors the power of the signal driving the antenna, and a field probe measures the field at the EUT 3 meters away.
The E-field signal level is comprised of the signal generator output, the amplifier gain, and the transmit antenna factor as shown in Equation 2.
|
E= SG+ AG+ TAF (2) where: E(dBµV/m) = E-field output test level SG(dBµV) = Signal generator output AG(dB) = Amplifier gain TAF(dBm-1) = Transmit antenna factor |
Test-Site Selection
Since air is the transmission medium, testing can be performed indoors or outdoors. The choices are compared in Table 1.
There are a variety of indoor test environments, such as anechoic chambers, semi-anechoic chambers, transverse electromagnetic mode (TEM) cells, and gigahertz transverse electromagnetic (GTEM) cells. In a test chamber, field strength and uniformity must be maintained.
The walls are designed to have an impedance of 377 W , which duplicates results generated outdoors by simulating the characteristic impedance of free space. The walls also should be lossy to absorb signals, not reflect them. Some walls are covered with ferrite tiles for uniform field strength.
An outdoor area test site (OATS) is best located in a quiet place, away from communications interference or at least where the background noise is measurable and repeatable. With the proliferation of communications technology, how many quiet places will be left in a short time?
Migration Toward Compliance
Since a common path to compliance is controlling a product’s emissions and then tackling immunity, the choice and use of an EMI receiver or spectrum analyzer takes on added significance. A full-compliance EMI receiver must meet the requirements of CISPR 16-1 and is used with a preselector filter to match the specification’s bandwidth measurement requirements.
The filter also prevents preamplifier saturation due to a broadband input. The receiver’s calibration also is traceable to a competent body, such as the National Institute of Standards and Technology (NIST) in the United States or National Accreditation of Measurement and Sampling (NAMAS) in the United Kingdom.
Peak, Quasipeak, and Average Levels
Most EMI receivers cover a frequency range of 9 kHz to 1 GHz and above and incorporate a provision for measuring the amplitude of emission signals according to peak, quasipeak, and average levels. “The peak detector measures the peak value of a signal, regardless of the duty cycle of that signal,” said Cliff Morgan, EMI marketing manager of Wireless Communications Test at Tektronix. “A quasipeak detector applies an amplitude-weighting factor based on the repetition rate of the signal.
“For continuous-wave signals such as sine waves, a peak detector and a quasipeak detector will read the same value. For impulsive type signals, however, the quasipeak detector will yield a lower reading than the peak detector,” he continued.
Peak measurements can be easily performed and are a useful way to begin an investigation. “Initial measurements are made in the peak detector mode,” noted Dennis Handlon, an EMC engineer at Hewlett-Packard, “Then the operator performs quasipeak measurements on the suspect signals. Any signal below the regulatory limit in peak mode also will be below the limit in quasipeak mode.”
A peak measurement is a direct reading, but there is some math involved in making quasipeak and average measurements. “Quasipeak and average detection,” pointed out Pradeep Wahi, president of Antenna Research, “are used to determine if a detected field has the weighted energy to be disruptive.”
“Peak detection should be used on aircraft or life-support equipment,”
indicated Paul Sikora, EMI/EMC product manager at Electro-Metrics. “Measurements taken with a quasipeak detector are less than or equal to those of a peak detector and may provide a passing margin. But for critical applications, this decision could come back to haunt you.”
Some receivers include provisions to test for clicks, which are defined as a burst of less than 200 ms spaced at more than 2 s. The clicks may be subtracted from the normal measurement. An optional tracking generator allows site attenuation measurements.
Separating Background and EUT Emissions
How can you separate the emissions of the EUT from background emissions? “Tune in certain signals that may be in question and use the AM/FM demodulator on the receiver to see whether the signals are coming from the EUT or the ambient environment,” said Cliff Morgan of Tektronix. “Or you can create a list of known ambients and have the software exclude such signals. Subtracting the ambient reading from the ambient + EUT reading is erroneous unless the phase of the two signals is accounted for and EMI receivers don’t measure phase.”
“To separate emissions significantly larger than ambient, locate the maximum EUT emission by rotating the device while adjusting the antenna height,” said Paul Sikora of Electro-Metrics. “Then turn off the EUT to see if it was the device.
“It’s time-consuming, but test software can save time, and it also may generate a suspect list. To separate high spectral density signals like AM and FM broadcast bands,” he continued, “view the EMI receiver’s IF output on an oscilloscope or listen to its audio output.”
Continuous Interference Sources
How can emissions be measured when a continuous interference is present? “The substitution method can be used,” said HP’s Dennis Handlon. “Measure the amplitude with the EUT on, then off, and note the difference. Replace the EUT with a dipole antenna connected to a signal generator, tune to the frequency in question, and raise the power level until you get the same response as before on the receiver. Calculate the field strength, and this is equal to the field strength radiating from the EUT.”
Due to it’s high gain and broad bandwidth, the preamplifier goes into saturation without the operator’s knowledge. “To prevent this, the EMI receiver should have preselection filters with high out-of-band rejection and good sensitivity with high dynamic range,” explained Electro-Metrics’ Paul Sikora. “In cases of severely high ambient signal density, the operator may need to test that spectral region in a chamber and correlate the results with an open site.”
Calibrations and Measurement Uncertainty
Because EMC measurements may differ significantly under similar circumstances, the concept of measurement uncertainty has been introduced. NAMAS specifies measurement uncertainty in standards NIS80 and NIS81.
An error budget is given for every EMC test setup. CISPR 16 allows ±3-dB uncertainty for receiver, antenna, cable loss, and signal mismatches, but this does not include the test site. A typical normalized site attenuation uncertainty is ±4 dB.
The combination of measurement and test site attenuation uncertainties can give a possible overall uncertainty of 7 dB. This means the recorded measurement must be 7 dB or more under the specified limit to pass without question. The message is clear: Find out how to reduce measurement uncertainty to an acceptable level. There are EMC test software packages which perform calculation for measurement uncertainty based on calibrations and other factors in a given setup.
Conclusion
We have come full circle. If an error can’t be eliminated, we must stabilize it, measure it, and apply it as a correction factor. Reducing measurement uncertainty demands accurate, repeatable calibrations on test equipment and test setups.
Now that EMC testing is taking place at 1 GHz and beyond, it is truly a pioneering effort to manufacture equipment and set up tests with predictable calibration curves over the full frequency range. The challenge is to produce results that are certifiable for full compliance testing.
References
1. Editorial, IEEE EMC Society Newsletter, Summer 1996, p. 2.
2. “A Guide for Choosing Essential EMC Immunity Test Equipment,” EE-Evaluation Engineering, April 1998, p. 68.
3. “Set Up Your Emissions Test System With Confidence,” EE-Evaluation Engineering, January 1998, p. 62.
4. “Immunity and Emissions Test-Software Features That Make a Difference,” EE-Evaluation Engineering, April 1998, p. 78.
5. “Should You Invest in Precertification EMC Test Equipment?,” EE-Evaluation Engineering, February 1997, p. 126.
6. “Build Your Test Site Right,” International Product Compliance, January 1998, p. 19.
NOTE: The EE-Evaluation Engineering articles can be accessed on EE’s TestSite at www.nelsonpub.com/ee/. Select EE Archives and use the key word search.
Table 1
|
Parameter |
Indoor Test Site |
Outdoor Test Site |
|
EUT Size |
Limited to size of chamber |
Unlimited |
|
Field |
Must be controlled to be uniform |
Uniform (the earth’s field) |
|
Local Signals |
Chamber wall must absorb reflections |
No surfaces from which to reflect |
|
Ambient Noise |
Shielded out |
Due to surrounding communications media |
|
Test Type |
Emissions and immunity, to 5 GHz |
Emissions only, to 25 GHz |
Spectrum Analyzer Designed
For Precompliance Testing
The HP 8590EM Series of EMC Analyzers measures frequencies from 9 kHz to 1.8, 2.9, 6.5, 12.8, 22, or 26.5 GHz, specified at ±1.5 dB from 9 kHz to 1 GHz. Features include peak, quasipeak, and average detectors; CISPR 16- specified bandwidths; an AM/FM demodulator with a speaker; a ROM card with built-in agency limits and transducer correction factors; a RAM card for storing traces, lists, and setups; and IEEE 488 and a parallel port. Hewlett-Packard, (800) 452-4844.
Receiver Used for Compliance
Testing to CISPR, EN, FCC
The ESCS 30 EMI Test Receiver is a portable instrument used for compliance testing to CISPR, EN, and FCC EMC standards. The instrument has a 9-kHz to 2.75-GHz frequency range and an amplitude accuracy of £ 1.0 dB @ 1 GHz and £ 1.5 dB @ 2.75 GHz. The 6.5″ color display uses a bar graph to show the peak, quasipeak, and average signal levels simultaneously. The unit includes a time-domain oscilloscope mode for click testing and measures radiation from mechanical switching with a 100-µs resolution. Tektronix, (800) 426-2200, press 3, code 1037.
Harmonics and Flicker Tester Is PC-Based
The Compliance Test System combines an AC power source with a PC-based data acquisition system to provide an automated method to test for IEC1000-3-2 harmonics and IEC1000-3-3 flicker. Windows-based IEC test software performs required tests, logs the results to disk, and prints out a pass or fail report. Both single phase- and three-phase versions are available, with power levels from 1,250 VA to 15,000 VA. The system also tests to IEC1000-4-11, IEC1000-4-14, and IEC1000-4-28. From $9,975. California Instruments, (800) 422-7693.
Full Compliance Receiver Tests to 1.2 GHz
The PMM 9000 EMI Receiver, fully compliant to CISPR 16, tests for frequencies from 9 kHz to 1.2 GHz. Three simultaneous detectors provide peak, quasipeak, and average measurements. Features include an internal tracking generator, floppy and hard disk drives, preselector filters, an AM/FM demodulator, and a 7.2″ display. RS-232, PC keyboard, VGA monitor, and printer interfaces are standard; IEEE 488 is optional. Spectral analysis is performed in the sweep mode. Antenna Research, (301) 937-8888.
System Performs Six
Immunity Tests
The BEST EMC System tests to IEC1000-4-2, -4, -5, -8, -9, and -11 in one unit. The multifunction generator provides a 4.4-kV/100-kHz burst and a 4.4-kV/2,200-A surge. An ESD gun delivers 16.5 kV in air, or 9 kV contact. The unit simulates power line dropouts and voltage dips without added equipment and is small enough to perform on-site investigations. Magnetic testing under EN 50082 and a three-phase extension unit are optional. Schaffner EMC, (973) 379-7778.
Hand-Held Field Meter
Logs EMF Readings
The HI-4460 is a graphical, hand-held instrument for site measurements of electromagnetic fields. The display shows individual axis, resultant, and peak readings as well as bar-graph indications and running time plots. An RS-232 interface transfers data to a PC. Alarm, threshold, and percent exposure settings are provided. Holaday Industries, (612) 934-4920.
Semi-Anechoic Chamber System
Tests Emissions and Immunity
The ARcell System offers four cell sizes to accommodate test objects from 30 cm to 100 cm on a side. It includes an antenna, an rf absorber, an amplifier, a preamplifier, a signal source, a power meter, a field probe, and a computer. Pre-loaded software contains specification support and a pass/fail margin which, along with the equipment, make in-house precompliance testing possible for both emissions and immunity requirements. Amplifier Research, (215) 723-8181.
Copyright 1998 Nelson Publishing Inc.
October 1998
|
|