Below are the questions asked during the live event, along with their respective answers.
Q: How can you tell if you have an EMC issue?
A: Intermittent issues are difficult to trace and figure out. Try and notice the time of day and what is on or off when the event is noticed. If you can see a pattern that relates to a machine starting up, someone starting their job for the day, maybe an event on the power line that occurs around the same time each day. If you can plug the device being affected into another outlet that is on a different circuit this is a quick test.
If you can re-route suspect cables that may be radiating disturbance or re-route the cables related to what is being interrupted. If something improves or gets worse you know you’re on the right track.
Review wires and how they are routed. Are they all in the same conduit? Are they in the same wire rack or run? Separate these. Remove any loops that may have been created during install. Extra wire length is sometimes added for ease of assembly with the extra just “stuffed” inside. Add grounds to the affected device and/or nearby devices generating noise.
Q: Are there any quick tests you can do?
A: EMC noise is challenging, as it is an unknown path. If you have a receiver or spectrum analyzer with a voltage probe, current clamp, sniffer probes or an antenna you can do more tests. However, this is not typical equipment people have lying around.
You can turn devices on and off and see the effects. Or if a particular operation of a unit is only being affected, track that systems wires and circuity. What is around it (including all wire runs and where they go.
You can add grounds to see if there is any improvement. Grounding is typically an issue one way or another. Remember, a 12AWG wire at frequencies in the kHz and above will have impedance and therefore not be an effective conductor as it turns into an inductor. A ground braid, sheet metal or very short wire length are more effective.
You can move and/or shield cables (power, signal, output drives). You can use a handheld radio tuned to the AM band and no nearby station and see if it picks up any noise.
Q: You mentioned your test equipment could induce failing results on compliant equipment. Is this generally caused by the ground reference of the test equipment and equipment being tested?
A: This can be caused by many factors. Each situation can have its own challenge. However, ground is a big part of this. Ideally, you want to power your support equipment from a separate circuit and power your test equipment from a separate circuit. Another example of a grounding issue is if you add a cable adaptor you may inadvertently lift a ground connection within the cable.
Q: Is there a simple way to determine if the noise is differential mode or common mode?
A: When measuring conducted emissions this is always a question. You can get an idea regarding the noise by placing caps between the phases. If you affect the noise its looking differential. If you place caps to ground and affect the noise its looking common mode. However, keep in mind it is most likely both. Another test you can do is if you have a current clamp. Run the phases through the clamp in one direction and then run one phase through one way and the other the opposite way. If the noise cancels out common mode may be the prevalent one.
Q: Given that there is no such thing as a low pass filter, i.e., parasitic always limit the high-frequency attenuation, do you have any tips on extending the filtering effectivity of chokes apart from increasing the spacing between the turns?
A: Focusing just on the winding of the choke. Try and keep the windings separate from each other and go in only one direction. If you need more room and have to double up the windings go forward 4 windings then back 2 and forward again. Or something similar.
Q: What components make up a sine filter?
A: A sine filter is made up of an inductor and capacitors.