Probing RFID Tag with Network Analyzer

[ptarantino]

My company has purchased a network analyzer for use in comparing various RFID inlay and antenna types. I know how to use the network analyzer, and I know what information I am looking for. My problem is that I am unsure what the proper probing method would be for accurate measurements.

For your reference, most of the devices I test consist of a metallic-ink antenna printed on either paper or plastic backing, with no adhesive or other covering. But since there are no connectors built into these devices, I do not trust my measurements.

At the advice of a colleague, I have already implemented a macro to convert the normal S-parameters to differential mode. I am not convinced this is the way to go and I don't know if some external grounding is needed for the antenna. I generally probe two points, on either side of the mounted chip. So far I have tried one probe (with the other point touched by the ground of the coax probe) and two probes (with both cable's grounds soldered together).

One more issue is that I am not certain that the chip is functioning in active mode when driven by the network analyzer. I have theoretical impedance values for these chips in active mode at the operating frequency (915 MHz), but I don't trust these measurements either.

Any advice?

Thanks in advance!

 

[Higgler]

If you are just comparing different hardware, why not do a simple range test in an anechoic chamber. Largest range wins.

Small antenna devices totally change when probed, so be cautious about that.

kch (antenna eng.)

 

[ptarantino]

I have already performed rough range testing (no anechoic chamber yet) and I have a decent idea of comparative performance. The problem is that I am evaluating antenna/chip combinations, and many different print variations, so being able to measure impedance mismatch and reaction to the presence of conductive materials would be a huge help in finding the best solution.

I certainly got the impression from my testing so far that I was primarily seeing the effects of probing and not the devices themselves. Thanks for the caution, although it doesn't really make me optimistic