0
\$\begingroup\$

I can't go into great detail but in our EMC room we have vertical polarity antennas as well as horizontal polarity antennas. If the device being tested is failing at roughly 65MHz in the vertical antenna position but passed in the horizontal antenna position does that mean it could be the orientation of the device's wiring that is is being run? Most of the wiring is run horizontal at a length up to 5 feet at the most.

\$\endgroup\$
3
  • \$\begingroup\$ It is too hard to answer this question without more details.. The problem is something to do with how the EM signal is getting coupled onto the wiring.. Are the wires shielded? \$\endgroup\$ Jun 28, 2017 at 20:44
  • \$\begingroup\$ It is ALWAYS hard to answer. Maybe some very expensive 3-D electromagnetic solver will help, or just a common sense of properly applied shielding might fix the issue. The answer is yes, some conductor/case/wiring geometry is responsible for different immunity to different polarizations. \$\endgroup\$ Jun 28, 2017 at 21:20
  • \$\begingroup\$ I guess my main question is if the antenna is polarized in the vertical position what wires are more susceptible to noise, horizontal wires or vertical wires? We are testing at 20 V/m at a distance of 3 meters. Initially the wires (CAN communication) were unshielded and twisted but we have tried a twisted pair of wires with braided shielded around it bonded to the frame but still did not do the trick. Also tried the typical adding ferrites and chokes method. \$\endgroup\$
    – rsmith8236
    Jun 30, 2017 at 14:26

2 Answers 2

1
\$\begingroup\$

There are a myriad of things it could be. Some things to consider: Have you double checked grounding/bonding/shielding? Need to know the distance you're testing at, power of received signal at different distances, and characteristics of the signal. Is it found at other harmonics of 65MHz? Does it coincide with clock signals on the DUT? Seems like if it were the horizontal wires acting as an antenna, then the signal would probably be horizontally polarized. At 65MHZ, vertically-polarized, look for a vertical radiating element at whole-integer fractions (1/2, 1/4, 1/8, etc.) of the wavelength. Then, look for circuits coupled to it or within it's near-field (wavelength/2pi).

\$\endgroup\$
4
  • \$\begingroup\$ I guess my main question is if the antenna is polarized in the vertical position what wires are more susceptible to noise, horizontal wires or vertical wires? We are testing at 20 V/m at a distance of 3 meters. Initially the wires (CAN communication) were unshielded and twisted but we have tried a twisted pair of wires with braided shielded around it bonded to the frame but still did not do the trick. Also tried the typical adding ferrites and chokes method. \$\endgroup\$
    – rsmith8236
    Jun 30, 2017 at 14:29
  • \$\begingroup\$ Vertical, most likely. Are you bombarding the DUT with a CW to check how it responds to interference or are you receiving radiated emissions from the DUT? I probably should have asked earlier...my answer was based on the assumption that you were picking up a 65MHz spurious emission from the DUT \$\endgroup\$
    – GroundRat
    Jun 30, 2017 at 15:47
  • \$\begingroup\$ We are doing an immunity test which is checking how the DUT responds to interference. We are beaming a 20 V/m from 0 to 100MHz at the DUT. The DUT loses communication around 65MHz \$\endgroup\$
    – rsmith8236
    Jul 1, 2017 at 21:24
  • \$\begingroup\$ Gotcha. Most of my experience in EMC is the other way around, but the concepts are still pretty much the same. analog.com has some good tutorials, and the following is one that I've used on the job: analog.com/media/en/training-seminars/tutorials/MT-095.pdf I still think you're looking for a vertical conductor as the likely culprit. \$\endgroup\$
    – GroundRat
    Jul 3, 2017 at 13:06
0
\$\begingroup\$

I've seen LIN transceivers fail at 600MHz, because the PCB trace (under the LIN IC) resonated with the ceramic SMT VDD bypass capacitor. Cure? Install a lossy SMT bypass capacitor.

The development team had to learn the lesson ----- there are differences in brands of capacitors, differences in dielectrics of capacitors.

\$\endgroup\$
1
  • \$\begingroup\$ It could mean your “room” has standing waves adding in one direction more than the other. \$\endgroup\$ Mar 16, 2018 at 18:21

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.