I am working on measuring propagation loss in different types of air pressure and temperature. The introduced absorption loss in my test environment is less than 0.0001 dB, which I want to detect.
In other words, I have a received CW signal at -60.0000 dB and I want to measure an extremely small change in the received power in the order of 0.0001 dB, i.e. I should be able to read -60.0001 dB on the power meter.
N.B. The commercially available RF power detectors are designed for accuracies no more than +/- 0.05 dB, very far from my target accuracy +/- 0.0001 dB.
What are you using as a source, and in what frequency range? Stabiliy of frequency and amplitude may also pose a problem. Accuracies you are mentioning seem too high even for standard simulations.
I am using a CW signal generator, with frequency 40 GHz.
You are quite correct the source may have some amplitude, and phase fluctuations. Only comparative measurement methods might tackle this issue, i.e. to compare the input power (source) with the output power and then amplify the difference.
I've never seen any commercial gear with this kind of precision. Are you characterizing a gas inside a closed chamber? If so, the resonant modes of the cavity will be important as well, since small motions in the probes, i.e. vibrations, thermal expansion, will cause fluctuations in the signal level.
You may be able to take a long sequence of data at a given temperature and pressure and average the sequence over thousands of individual data collections to average out the effects of the thermal, vibration, and source uncertainty / fluctuations.
When you measure the power, a coupler is necessary.So, precisely measuering the coupler parameters is important as well? If you want to obtain 0.0001 accuracy. High resolution ADC should be exploit in the project ？
Dear all
to measure in absolute terms this small amout (0.1 milli-dB on a 60 db attenuation level ) in a static way is extremy hard if not impossible. it also may be not meaningful since the DUT is drifting much more for already 1 dec C temperature change.A relative measurements is already closer to reality (I guess thats what you want) but still quite hard.
The easy and proper way to measure those very small changes is via modulation (i.e. slowly modulated S-paramaters) with this one can measure down to nano-dB and micro deg) examples are microwave absorption cells using STARK modulation. In principle you need a CW generator with very low phase noise and in principle you can measure the high of the modulation related sidebands down to -180 dB (-180dbc) below the carrier...example:1 % AM gives you AM sidebands which are 46 dB below the carrier..0ne permille AM iis 66dB down etc. (1 permille=10E-3 change in voltage is roughly 0.01 dB)
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Electromagnetics