1
\$\begingroup\$

We have a product that we successfully ran through radiated EMI testing (EN55022) with our highest peak approx. 13dB under the permissible limit.

Looking at our test setup and what is required to comply with the new EN55032 standard, I see our original test setup used loopback cables which were only 0.8 meters in length rather than 2 meters.

I'm looking for an equation to help calculate what would be the worst-case increase to our radiated EMI from loopback cables which are this much longer. I have a hard time believing a 2.5x longer cable would add 13db of noise, but I have no way to mathematically support that.

\$\endgroup\$
5
  • \$\begingroup\$ What prevents you from using longer cables and test again? \$\endgroup\$
    – PlasmaHH
    Jan 25, 2017 at 21:37
  • \$\begingroup\$ Easy- They're called Maxwell's equations. (Which won't be easy to actually solve, so you are better off re-testing.) \$\endgroup\$
    – John D
    Jan 25, 2017 at 21:41
  • \$\begingroup\$ It's approximately $3000 of time at our local lab to run through all of EN550x2, not including my own time. The EN55022 and EN55032 setups are very similar with a few exceptions. I'm trying to determine if those differences, in the worst case, could add up to us violating the limits. \$\endgroup\$
    – 1N4007
    Jan 25, 2017 at 21:43
  • \$\begingroup\$ My hope is that there are a number of assumptions which can be made here (same distance to antenna, same termination impedance, same frequencies, etc.) which can be made here so that it boils down an equation to something pretty straightforward. \$\endgroup\$
    – 1N4007
    Jan 25, 2017 at 21:50
  • 1
    \$\begingroup\$ For clarity - I don't mind spending the money and time to re-test, as long as it makes sense to do so. If it turns out a common equation dictates a cable twice as long would only emit 3dB more of noise in this situation, then it doesn't make as much sense to put the time and money into re-testing just on account of the cable length. \$\endgroup\$
    – 1N4007
    Jan 25, 2017 at 22:13

1 Answer 1

1
\$\begingroup\$

In short: to meet compliance you will probably have to retest, I haven't read the standard myself (It's paywalled). But there are resources that deal with the change in compliance.

I'm not sure that many standards let you prove the compliance without testing, and yes it will make a difference on the cable length.

A word of advice: it would be better to find an EMC consultant or someone who knows the standards and has done the testing before because there might be a way to get the device classified under a different standard OR knows a way to get it grandfathered it to the new standard without testing. This has worked for me in the past.

This is a good resource if you really want to calculate and learn about antennas. The cable could be approximated as a coaxial antenna, but what would be hard to model is what voltage source for the antenna is. On paper it might be possible to run through the equations backwards for the length of cable you have using the radiated test data you have, but the length of the cable will change the resonant frequency. This will make it hard to determine what the radiated emissions of the longer cable would be from the original test data.

\$\endgroup\$
1
  • \$\begingroup\$ Thank you for the links to the LearnEMC page. This may provide a good starting point to calculate approximately where we're at. We had tested with NWEMC originally and they provided that first link to us when a miscommunication on our part resulted in us testing to 55022 rather than 55032. \$\endgroup\$
    – 1N4007
    Jan 26, 2017 at 19:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.