gin&pez@arg
Greetings, and thank you for the kind words and your clear and thorough response to my long list of 18 questions. Please accept my apologies for the very long deferral in my response, but be assured that I have not been idle on this topic.
I replicated your BBC BASIC program in a spreadsheet and confirmed that my results are very close but not identical. I have not yet concluded the difference to be one of math precision, but this does appear to be the case. Beyond that; I have been using the spreadsheet as a study aid as I become more familiar with your work. This has been both fruitful and enlightening, but I am ignorant of much of the mathematical details. It has been a long journey of discovery for me, and I am grateful for your patience.
There are more than 100 posts in this thread. I return to review those that are focused and relevant; only to discover what you previously informed us about. My growth in understanding eases the task of interpreting your work that was not clear or sufficiently detailed to understand until now. Perhaps my ignorance may be serving as a benefit to you by sharing our communications with the English speaking world.
I am flattered by your offer to participate as a supporting contributor to your work, but I hardly have the credentials for doing so. If I had an advanced degree, it would be a PhD in tenacity and persistence from Hard Knocks University. :-) I don't quite know how that remark might translate; so I will explain that it is intended as a metaphorical way of saying that I work hard and overcome obstacles. I got paid but not graded for my efforts. Now that I'm retired; I do this for fun. You are certainly welcome to publish freely anything I have written here, and anything in the way of test results or data that you might ask me to contribute.
Now back on topic...
On your answer A3:
"A3 : Not at all - Our VNA data were taken exactly on the same reference plane.
Otherwise a comparison would not be possible at all - Please, be careful at this
very point - You have been warned "
I understand, and I concur. My question was imprecise, but is now answered,
Your answer A10:
A10 : No - The so called "calibration" procedure is hidden in our compact
computation of Reflection Coefficient of the Unknown load.
Perfect... This validates my observations precisely, and is what I have been reluctant to confess. I have not attempted to study the mathematics of how calibration is performed traditionally, but your algorithm is simple, concise, and clearly results in a calibrated reflection coefficient that overlays the corrected results of the NanoVNA almost exactly.
Your approach is both elegant and provocative. I most recently observed that the measurements (all reflection coefficients) are piecewise linear across significantly large spans of frequency. I am planning a simulation and experiment to validate the utility of this, and I will post any results that appear useful. I have also been looking at using your algorithm to compare my NanoVNA with a simulated "perfect" VNA as a candidate reference standard for comparing hardware much like an isotropic antenna serves as a reference to which all antennas are compared. The early results of this look exciting, and suggests that it may be possible to characterize and calibrate the response of the overall VNA test station itself. For now however; much of this remains observational speculation.
I respect and concur with your response (A14-16:)
A14-16 : We don't want to comment to the slightest on these - These are exclusively
your conclusions.
I agree... the evidence in the data is circumstantial and inconclusive.
Thank you again for sharing this openly in the amateur radio community. VNA's are new to many hams; but as they become common, your work will gain the recognition it deserves.
--
73
Gary, N3GO
