RF probes for measuring spurious emissions from 30 MHz to 18 GHz.

[rawkey89]

Hii,

I am trying to find the source of certain radiated spurious emissions from a device. I am looking for some sharp edge probes if available which can easily help me find the source point accurately. The frequency range I am interested is 30 MHz to 18 GHz. Any suggestions on this?

BR
Rakesh

 

[ravindragudi]

I am afraid that you won't be able to isolate the problem on a device using probe ...
Moreover, the frequency range from MHz to GHz - thats too much.
Better way is to carry out analysis and isolate the issue on the device.

 

[FvM]

Miniature electrical and magnetical probes are available from various manufacturers. The idea of finding a source of radiated emission with a "sharp edge probe" involves however a simplistic understanding of EM phenomena, I fear. It probably works at low frequencies where relevant device dimensions are far below wavelength. At least above a few 100 MHz, emissions are radiated by the distributed structures, e.g. a ground plane or case parts. The field maxima may be observed at other locations than the primary source.

 

[D.A.(Tony)Stewart]

• Spurious emissions are calibrated with a special antenna such as dodecahedral dipole at 10m in an open field.
• Although you can locate spurious sources using any Spectrum Analyzer and a stub antenna ( short piece of wire ) and realize that the gain is proportional to length up to 1/10λ and use a 1/2λ length at the observed frequency. Usually LO leakage is dominant or any fast rise time pulses on SMPS or clocks on long cables.
• Near field levels will be much stronger and easier to locate by waving loop or stub over the interfering unit under test.
• Ambient noise can be stored in memory for nulling (A-B) in digital RF analyzers.
• Faraday Cages (e.g. Lingren) work better for ambient suppression but then interfere from standing waves.
• Determine if it is radiated out of slots or gaps or transmission lines.
• To get rough absolute levels, use an RF generator and similar stub on coax end parallel to Rx stub on SA for comparison at a known distance and path loss.
• A small loop or stub works well as hemispherical omni-directional antenna with a null spot looking at the end.
• Get a unit that passes and compare.

 

[chiques]

We used to do this by just waving a bare 50 Ohm semi rigid cable and our spectrum analyzer. We were only doing up to 3GHz though.

 

[biff44]

what they said.
you are looking to buy a set of "near field probes".