Factors Affecting Conducted EMI Tests

Conduction Emission in EMI testing is used to measure the RF signal transmitted through the power system via equipment such as LISN and isolation transformers. For conducted EMI tests, other signals may be picked up from the environment and wiring, causing measurement errors.

The purpose of this experiment is to help users understand that even with the same equipment, different results may occur due to differences in cable types, wiring, and cable placement.

Generation of Antenna Effect:

The possible generation of antenna effect in wiring is shown in the following diagram. Factors that can affect measurement results include cable types, cable placement, and whether it is floating or not, etc.

First, we used BNC cables of different types, including RG58U, RG400, and BNC cables purchased from an electronics store, to connect between the spectrum analyzer and LISN. Then we placed the test system in an RF isolation chamber and found that the performance of the three BNC cables was similar.

(Purple: RG58U, Blue: BNC cable purchased from an electronics store, Yellow: RG400)

Then we moved the entire test system out of the RF isolation chamber and performed tests indoors. We found that some noise appeared near the higher frequency bands, indicating that some signals may have been coupled in this frequency band. Except for RG400, the other two cables picked up some noise.

(Purple: RG58U, Blue: BNC cable purchased from an electronics store, Yellow: RG400)

Next, we tested a more extreme case by directly connecting the power cable from the LISN input port to LISN output port. Surprisingly, a lot of noise appeared. In fact, at this point, it indicates that a loop antenna effect has been generated on the LISN, which can easily receive noise from low frequencies.

(Purple/Blue: Switching to LISN 9 k/150kHz filter under normal connection. Yellow: Circuit condition.)

In EMI labs, there are regulations for the placement of RF cables or power cables. However, in the debug or pretest stage, it may be difficult to replicate the same environment as the lab, which may result in some measurement errors. Through the tests described in this article, we can see that even in conducted EMI tests, there may still be measurement differences due to the environment and cable types. Those are factors that require to be considered when establishing an EMI testing environment.

For more detailed application content, please go to www.gwinstek.com/en-global/products/detail/GSP-9330

Please contact us for further information of GSP-9330.

Contact us:
Overseas Sales Department
Good Will Instrument Co., Ltd
No. 7-1, Jhongsing Road, Tucheng Dist.,
New Taipei City 23678, Taiwan R.O.C
Email: marketing@goodwill.com.tw