Electromagnetic Interference (EMI) — alternately known as Radio Frequency Interference (RFI) when it arises in the radio frequency spectrum — is an obviously bad effect on a variety of levels. There are two transmission vectors, equally problematic.
Conducted EMI arises at lower frequencies. It is carried through conductors that touch, and it may be conveyed from the source to other equipment that is on the same power line. The neutral conductor can convey conducted EMI throughout a building and beyond. Back feeding through a step-down transformer, the voltage can rise. Then the transmission line becomes an antenna if the frequency is borderline.
Radiated EMI is transmitted without direct electrical contact. It arises when a conductive body acts as a transmitting antenna to another conductive body that acts as a receiving antenna. It may be a broadband phenomenon or more narrowly tuned.
In analog communication, even a small amount of EMI can cause an unacceptable amount of noise in the signal. Low-level EMI has little effect on digital transmission, where the distinction between 0 and 1 is not obliterated. But as noise increases, the bit error rate rises. At a certain point, there is an abrupt cliff effect and the consequences can range from trivial to catastrophic.
Harmful EMI emissions can be mitigated by modifying the offending electrical equipment. An example is the ubiquitous resistor spark plug. A ceramic resistor is incorporated in series with the arc gap. It suppresses EMI emission while having no effect on the high-voltage electric circuit.
Beginning in the 1930s, the International Electrotechnical Commission (IEC) formulated regulations designed to reduce EMI emissions in electrical equipment manufactured in Europe. The scope soon expanded. Provisions were introduced requiring electronic devices to incorporate defensive measures designed to make them immune to EMI from outside and inside sources. These measures were largely successful even as the potential for EMI increased rapidly in an ever more crowded electrical environment. The U.S. and other non-IEC countries joined in.
Now, compliance with applicable EMI regulations must be assured for all new products. Compliance testing generally takes place on a prototype prior to production.
The problem confronting manufacturers is that compliance testing is enormously expensive. An IEC-certified organization will perform EMI emission compliance testing in a carefully engineered environment to ensure that incidental EMI unrelated to the device under test will not enter the picture.
A test of this sort will have a price tag of thousands if not tens of thousands of dollars, depending upon the complexity of the product. What accounts in part for the high cost to the manufacturer is the fact that the compliance testing organization has tremendous liability exposure. They can’t have a product that is certified emit harmful amounts of EMI after thousands have been sold and are in the pipeline. Products that don’t pass EMI compliance testing must be redesigned, and the revised model must go through another round of testing.
This is the rationale for pre-compliance testing, usually conducted in house by manufacturer personnel. Naturally it is typically less rigorous and there aren’t the elaborate protocols and specialized equipment including the anechoic chamber that shields the test setup from outside radiation.
Typically, in-house pre-compliance testing makes use of field-fabricated ground planes placed strategically around the test setup. Moreover, in-house pre-compliance testing makes use of ordinary oscilloscopes and spectrum analyzers rather than the far more expensive EMI receiver that professional compliance testing organizations use.
For effective pre-compliance testing, in conjunction with an oscilloscope, near-field probes are used when EMI is the focus of the investigation. The boundary between near field and far field is usually taken to be one wavelength. At reasonable frequencies, there is plenty of room in the near-field zone to work the near-field probe without touching the radiating body, which would contaminate the investigation by introducing conducted EMI.
Within a given chassis, every wire, terminal, trace, semiconductor and conductive metal part becomes an antenna that may aggressively radiate EMI. Microprocessors in certain circumstances become prodigious electromagnetic factories. To comply with EMI standards, design engineers are obliged to contain or otherwise minimize these electromagnetic storms so that, conforming to various inverse propagation laws, they do not overwhelm the analog and digital information streams that have become so much part of our lives. Then there is conducted EMI, which requires a different set of solutions.
To deal with radiated EMI, its source first must be located. Then the field must be mapped and quantified so if it rises higher than an acceptable level, it can be reduced if not eliminated.
The test procedure for EMI pre-compliance testing is to locate the open chassis of the equipment to be tested in a tent that is set up in an open field or on a lab bench with ground planes in place. The equipment should be powered up and checked for EMI emission immediately and continuously during a half-hour or more warm-up period, the exact amount of time depending on the ambient temperature and nature of the equipment.
Historically, grounded metal enclosures were universal. Open cooling vents controlled temperature rise caused by heat from internal components, but they also provided a way out for radiated EMI. Contemporary electronic equipment such as high-end oscilloscopes are housed in durable air-tight plastic enclosures that often include thin grounded shielding on the inside. Highly efficient cooling methods have facilitated this shift in part to meet EMI mandates.
The near-field probe has a small downside: It perturbs the field that it intends to measure. Inaccuracies are inevitable, and that is one of the reasons for a final round of compliance testing. In the interim, competent pre-compliance testing requires an understanding of near-field and far-field probing and the nature of time- and frequency-domain interaction. This may seem difficult, but it is worthwhile in view of the expense involved in consigning the entire project to the compliance-testing community.
An issue arises in regard to whether to use an oscilloscope or spectrum analyzer. The spectrum analyzer is generally considered a more advanced instrument, but in pre-compliance testing the oscilloscope has advantages. If it is a choice between the two instruments, for a variety of reasons the oscilloscope is the way to go. An MDO oscilloscope will display the time- and frequency-domain signals simultaneously, and this is immensely useful in zeroing in on the exact source of EMI by scanning the circuit board(s) with a near-field probe.
The MSO oscilloscope is necessary for finding the cause of intermittent EMI emission, which otherwise may be highly problematic. The MSO is capable of simultaneously displaying signals from two different sources within the same piece of equipment. The classic miscreant is a power supply that intermittently, sometimes only for very brief intervals, becomes dysfunctional to the extent that distorted power is delivered to the component under investigation.
Using the MSO, a time-stamped view of the anomaly can be obtained and correlated to high EMI emission. Then, in a second-generation prototype, the power-supply problem can be eliminated and with it the non-compliant EMI emission.
Rich Markley says
One great application for using a scope in EMI testing is capturing sporadic events with the zone trigger. Intermittent or sporadic emissions are the most difficult to debug. They are difficult to capture and not easy to analyze. The zone trigger in the time and frequency domain makes it possible to trigger an oscilloscope with FFT capability only on such events. For further analysis, the user can adjust FFT settings such as resolution bandwidth or gate, and for a more detailed analysis use the cursors to get signal levels.