1
\$\begingroup\$

I am in the midst of setting up an EMC test bench for Conducted Emissions testing. This question goes out to the EMC experts regarding the essential equipment needed for the same.

Currently available: Spectrum Analyzer; Ground plane; LISN;

Question:

Several sources suggest connecting the LISN through an attenuator (10dB is what I came across the most) to the spectrum analyzer. How do we choose the attenuation level as there are several available in the market?

Note: The CE test is for a power electronics application board so the possible standards would be considered as for commercial/industrial applications.

Apart from the above-mentioned equipment what other equipment would be essential in order to set up the test bench?

\$\endgroup\$

1 Answer 1

1
\$\begingroup\$

It's always a good idea to have external attenuation in emission testing.

10 dB external attenuation should be a good starting point. You might also consider getting an attenuator with integrated transient protection. Some LISNs also have an internal attenuator that you can turn on and off.

But it's hard too tell how much attenuation is reasonable in your case. With significantly more attenuation than 10 dB the noise floor of the setup could prevent you from getting useful results.

What I always do is to start with the highest attenuation possible and then reduce the attenuation until I get useful results. But I always try to have at least 10 dB of external attenuation with a dedicated attenuator that you can't accidentally turn off.

Apart from that, I guess you are good to go. Just don't forget the little stuff like cables, gender-changers and adapters (e. g. BNC to N-Type).

\$\endgroup\$
5
  • \$\begingroup\$ Thank you @feynman for your reply. Currently, the spectrum analyzer has EMI measurement settings. If I choose the CISPR setting, it considers 6dB attenuation inherently. In this case, do I attach the attenuator? If yes, how would I know that the emission hasn't indeed been reduced in amplitude? \$\endgroup\$
    – ORion
    May 16, 2023 at 14:13
  • 1
    \$\begingroup\$ @ORion I think you are confusing the CISPR filter's roll-off characteristics (6 dB) with input attenuation. You need input attenuation to protect the mixer of your spectrum analyzer (SA) from damage. The mixer of a SA is very fragile and can easily be damaged by transients and/or too much RF power during conducted emissions testing. You can inform your SA about any external attenuation and the SA will adjust the y-axis accordingly. If you use the SA's internal attenuation, the y-axis will be adjusted automatically. \$\endgroup\$
    – feynman
    May 16, 2023 at 21:16
  • \$\begingroup\$ you are right. In that case, I will go forward with procuring a 10dB attenuator. I had one more doubt about capacitive coupling clamp: Do I need one for a conducted emission testing? \$\endgroup\$
    – ORion
    May 17, 2023 at 6:03
  • 1
    \$\begingroup\$ I'm not familiar with all standards. But for the standards I'm familiar with, you don't need a capacitive coupling clamp for conducted emission testing. To my knowledge a capacitive coupling clamp is used for conducted immunity tests (e. g. DIN EN 61000-4-4 burst tests). \$\endgroup\$
    – feynman
    May 18, 2023 at 5:58
  • \$\begingroup\$ thank you for the response. \$\endgroup\$
    – ORion
    May 22, 2023 at 4:41

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.