Electromagnetic compatibility (EMC) is the desired condition for electronic devices. In an ideal world, radio-frequency emissions and transient events would not cause failures or disruptions to electronic equipment. But as much as we (and the product manufacturers) wish otherwise, we are not in an ideal world.
Radiated emissions, whether intentional from transmitters or unintentional from electronic circuits, are in the air all around us. Sharing that space is a host of transient events like lightning, power surges, and electrostatic discharge (ESD).
Regulatory and contractual obligations require that radiated and conducted immunity tests be performed to determine whether the device being tested will operate normally when exposed to typical levels of interference.
The standards defining and governing immunity testing have been on the books for many years. Some are regulatory, such as those from the European Union (EU), that can prohibit product marketing unless the product has been shown to pass the applicable immunity tests.
Some are contractual, issued by manufacturers with application-specific immunity requirements. These tests are well-defined and understood.
Testing laboratories around the world have deep experience running these tests and providing reports on their findings. The responsibility falls to the product manufacturer to see to it that the applicable tests have been performed and that the results demonstrating standards compliance can be shown.
Immunity testing takes a number of forms, but the common feature among them is to verify that the device under test (DUT) does not experience a failure.
Testing labs have the equipment and know-how to run the tests, but the test lab cannot say definitively that the DUT is not operating normally unless the symptoms of DUT failure are understood.
RADIATED, CONDUCTED, AND ESD IMMUNITY
A radiated immunity test places the DUT in an anechoic or semi-anechoic shielded chamber and is subjected to a modulated radiofrequency (RF) signal across a swept range of frequencies. The field strength of that signal is specified in the standard that applies to that DUT’s product category and application.
The standard most often referenced for radiated immunity testing is IEC/EN 61000-4-3. It requires a pre-calibrated field from an antenna a fixed distance from the EUT.
The field is swept from 80 MHz to 1000 MHz in steps not exceeding 1% of the fundamental frequency, with sufficient dwell time at each step to note the DUT’s response.
The dwell time is the amount of time the signal rests at a single frequency as it sweeps across the frequency range.
The challenge for the EMC engineer in the test lab is knowing how long the dwell time needs to be for a given DUT.
There is not a one-size-fits-all answer since every product serves different functions and employs different designs to perform those functions.
Conducted immunity tests are performed by coupling electrical and RF disturbances into power and signal cables connected to the DUT.
The basic reference standard is IEC/EN 61000-4-6, which defines the RF waveform and its voltages between 150 kHz and 80 MHz, along with the coupling schemes between the test-signal generator and the DUT.
Electrical surge (IEC 61000-4-5) and burst/electrical fast transient (EFT) conducted immunity (IEC/EN 61000-4-4) are also part of the suite of tests.
These tests rely on rapid repetition of surge and burst pulses to look for the DUT’s vulnerability to conducted disturbances that can be coupled into its attached cables.
Electrostatic discharge (ESD) immunity tests are performed on the exterior of the DUT. Standard IEC/EN 61000-4-2 defines the waveform, voltage levels, and discharge techniques to approximate a static discharge from a human finger, furniture, or indirectly from a nearby discharge that can induce a transient charge into the DUT.
Discharges are done at positive and negative polarities at voltage levels ranging from 2 kV to 15 kV, depending on the product category.
Selection of discharge test points on the DUT are chosen to represent the product’s intended environment. The DUT is observed to note any change in its behavior during the discharges.
Enjoying this article?
Subscribe to Interference Technology for expert coverage of EMI, EMC, and signal integrity challenges—plus immediate access to new digital magazine issues.
Subscribe here →
IDENTIFYING A FAILURE
But throughout these tests, how does the engineer in the lab know whether the DUT is failing? On the surface it seems straightforward: Either the DUT fails to operate, or it doesn’t.
But contemporary electronic devices don’t simply “fail,” as in they cease to operate. More often than not, they have soft failures like resets or data corruption that might not be immediately obvious.
As the RF field, conducted pulses, or ESD transients sweep through the test, the DUT is monitored to observe its behavior.
It’s possible, even likely, that there will be a response when a single frequency, a range of frequencies, or transient pulses wash over the product. A response, however, is not necessarily a failure.
An example is a flickering LED on the DUT: Is that a failure or a benign response?
Depending on the type of product and its intended environment, it could be either. If the DUT is a low-cost toy for children, a flickering LED in the presence of a significant RF field is probably not a failure as long as the DUT continues to operate as intended.
But if the DUT is a piece of medical equipment, such as a cardiac monitor or an infusion pump, a flickering LED might lead to misinterpretation of a patient’s condition or something more serious.
Clearly, interpretation matters in an immunity test. Responses are going to happen in the course of an immunity test, and some may be a failure of the product’s intended function.
The EMC engineer cannot make that judgment alone but has to consult with the designer or application engineer responsible for the product.
THE IMPORTANCE OF A TEST PLAN
But designers or application engineers aren’t always available to answer those questions, and even if they were, they probably have a tight project schedule and aren’t going to be happy with questions that might cause a delay. And delays are inevitable if there’s an unexpected response from the DUT during the test.
It’s not unusual to find that nettlesome responses don’t consistently repeat, and it’s often difficult to isolate what part of the device is reacting to the electromagnetic disruption. In those circumstances, the process becomes a game of whack-a-mole, where the designer and the EMC engineer are trying one solution only to find that a new issue has developed.
Trying to find the root cause of a failure in that situation is difficult, to say the least. The most effective way to inoculate the developers and EMC engineers from the pain is to begin the project with a well-constructed test plan.
Since compliance testing is typically scheduled near the end of a development project, there is time far ahead of the planned immunity test for the EMC engineer(s) and development engineer(s) to meet and define the test conditions:
- What standard is the DUT being tested against? Is it a common regulatory requirement like IEC 61000-4-3 (radiated immunity), IEC 61000-4-6 (conducted immunity), IEC 61000-4-2 (electrostatic discharge)? Or is it a manufacturer-specific contractual requirement specifying different criteria and severity levels? The actual test steps will derive from the chosen standard.
- The configuration of the DUT: its hardware, the ancillary equipment (power supply, external interface, etc.), power and signal cabling, power requirements and connectivity, and any other parameters that will replicate the DUT’s operation in the field.
- A clear definition of normal operation: What should it be doing as it runs? What type of external monitoring is required? How is that monitoring set up and recorded? What cycles are in its operation?
- If software operation is integral to the DUT’s function, how many different modes are required to check real-world conditions?
- Most importantly, how exactly is a failure defined? Failure is a continuum ranging from no response at all (as might be expected of hardened military equipment) all the way to catastrophic failure (like overheating or shutdown). Everything in between is a “response,” which may be benign, annoying, or critical. All responses should be noted so that any corrective action can be taken.
CONCLUSION: PLAN YOUR IMMUNITY TEST
Long experience has demonstrated the value of a carefully constructed immunity test plan. Long experience has also demonstrated how painful the absence of an immunity test plan can be. Not every project and test strategy neatly follow these steps, of course.
Circumstances may find the EMC engineer and the project developer facing the one-two punch of a product delivery deadline and immunity failures that must be resolved.
Cooperation between the developers and the compliance engineers is critical to identifying the failure condition, isolating the root cause, and doing what can be done to mitigate the vulnerability before the product ships.
After the crisis has passed, the lessons learned are precious jewels to be kept and shared with colleagues. We think we know a failure when we see it. The real failure is, as Benjamin Franklin once wrote, “By failing to prepare, you prepare to fail.”
Write that immunity test plan. You won’t regret it.
