RF Probe's magic EM suction

RFbeginner

I often use Spectrum Analyzers to probe RF signal powers at different points on multilayered PCBs. These probes have a single conductor (or at least only require a single conductor). Placing the probe on a signal trace on the top exposed surface of the PCBs instantly shows us the RF power at that point. BUTTTTT How can this be when the RF energy is flowing within the PCB (between two conductors and amongst the DIELECTRIC)? It would make sense if we connected one probe on the top conductor and the other probe on the other conductor (usually a ground plane), that way we can multiply a I and V value, but this works on 1 probe. They just touch these magic probes on the top layer and BINGO.. they get teh RF POWER reading. How can these devices somehow guide the underlying RF Energy to dive into them

Averagesupernova

I have done a LOT of troubleshooting with a spectrum analyzer. First and foremost, no transmission line whether it be conventional coax or strip-line on a PC board is going to be 100% effective at shielding. Second, how far down on the spectrum is the signal? In other words, what is the signal level sampled? If it is 40 dB down, then the sampled signal is 1/10,000th of the strength of the actual signal. Pretty insignificant.

RFbeginner

How does the probe measure the power of the wave that is mostly sandwiched between the layers of dielectric in the PCB? I do realize that a certain % of the energy leaks into the probe, but if that % was not fixed, how do we get proper measurements from the Spectrum Analyzer?

chroot

The analyzer is using earth ground as a reference, and so is your PCB. Of course, you'd be better off using a differential probe, but it sounds like your single-ended probe gets the job done for you.

- Warren