# Why does S11 seem to be independent of transmission media in this RF transmission experiment?

Hi friends I am starting to characterize a system in which we will plan to measure the response of a sample of liquid to radio waves. In order to do this I am exciting two antennas with the ports of an Agilent two-port network analyzer as shown in the image below.

So far I see no significant difference in the data when the container of liquid is present or absent between the two antennas. The data shows vey high values for S11 and S22, and low values for S12 and S21 mostly independent of the sample presence.

May I know if I should expect some difference due to the transmission media, or what I might test or change in the setup in order to become more sensitive to the sample?

• Does your network analyzer have a manual? If so, have you followed a recommended test set up? How are you coupling the VNA to your experiment? Can you add a diagram? Right now your question contains very little information and your measurement could be the result of a problem with your setup. Start with this then explain clearly how your are coupling between the four ports of the analyzer to your liquid sample. In general, unless you have a perfect network, you should see all four S parameters non-zero. – uhoh Jan 2 '19 at 7:16
• @uhoh Sir, the lab setup is very minimal. And since i require assistance I want to share you the pictures. Please find the lab setup [here] (drive.google.com/file/d/1ElXC_hUq_FBNb-N4igR5QWJgG0T_7pfH/…). The S-Parameters of antenna in freespace/air is here. I did read that for coupling both the antenna it is necessary to see that both the antenna have similar free space Scattering parameters. Please do help me. – shiva Jan 2 '19 at 9:15
• @uhoh I forgot to mention that red-s22 blue-s11 green-s21 cyan-s12 – shiva Jan 2 '19 at 9:36
• Thanks, looks good! Have a look at an answer I've posted, feel free to comment further! – uhoh Jan 2 '19 at 11:09

I've added your images back to your question, slightly modified, I hope you don't mind. Now you might consider those arrows I've added, then go back and check to see if you are using ~950 MHz antennas (which are $$\lambda/4$$ and will resonate at integer multiples) and then ask if that's really the best thing to use. Away from resonance, they will naturally reflect most of the power.

You are mostly measuring the bandwidth of the antennas, except near the resonance where they can effectively radiate.

The network analyzer will measure the whole system, and so unless you use really good antennas that are nearly perfectly matched, mostly you will see reflections between the coax and the antenna connected to it.

An excellent, isolated antenna would dip down much deeper than -20 dB, but for a cheap WiFi antenna that's not bad, it means that roughly 99% of the power in the coax is leaving the driven antenna. However, at that point, still only 1% of the power is being received by the adjacent antenna and coupled into the other cable, so most of the power is probably going into free space.

## As a first step:

What you can try is to focus on only the frequency where the antenna resonates naturally, which look like about 950 MHz. Scan only around there, with many small steps in frequency in order to make a smooth curve.

I think that if you use patience and study this narrow range, (say 800 to 1100 MHz or so) you might really see a difference between sample-in and sample-out.

However, the difference might be related to the effect of the dielectric sample on the antenna resonance, so you can also repeat the experiment with your sample near one antenna but away from the other; to the left of the left antenna, and to the right of the right antenna.

This may give you a better understanding of the limitations of your current set-up, and some ideas how to improve it. I'd recommend reading further as well.

Good luck, and have fun!

• @shiva I'm an old Ham radio operator, I read some books a million years ago but can't recommend any now. There are so many ways to look at this subject, the best thing is for you to find the one you are most comfortable with. There is the ARRL antenna book that is read worldwide, if you can find a HAM radio club you may find someone who can help as well. Best thing is to get to a library and see what they have, sit down, flip through several and see which book "speaks to you" the best. – uhoh Jan 2 '19 at 12:22
• @shiva after seeing the changes I've made to your question (adding data, and other technical details) you can also consider asking further questions in Electronics SE. However, those guys can be a little rough on new users sometimes, other times they are extremely helpful. The more details you've added, plus a link to any reading you might have done (even Wikipedia) the better. – uhoh Jan 2 '19 at 12:24
• Thank you so much.... I am also an HAM Radio operator in india with call sign VU3HWW. Sadly here I didn't get time to go to local club. But now i have to find some time to go. I just want to know, do the most of radio clubs have VNA with them? And thank you a lot for helping me out – shiva Jan 2 '19 at 12:26
• @shiva Probably not. But they do have experience with SWR and matching antennas and some clubs might have a small library of books they can loan. But now it sounds like you may have enough background already. In the end I think you need a very different kind of antenna, or at least a different test set up. Another thing you can do is start a discussion in the Ham Radio Stack Exchange chat room The Ham Shack and ask people there. Post a link to your question here to start he conversation. – uhoh Jan 2 '19 at 12:29
• @shiva okay, quickly speaking, dB is the log of a ratio only there are no physical units, so -10dB means 10%, -20dB is 1%, etc. On the other hand, dBm is the ratio to 1 milliwatt (that's the "m") so +10dBm is 10 mW of power, -30 dBm is 1 uW (1 microwatt), etc. Sometimes people use dBW (Watts instead of milliwatts) but that's rare. I mention dBW in this answer. – uhoh Jan 2 '19 at 12:39