Just my 2 cents ...
I agree that it is a convention ... but to me it still is a "historical error" (not sure, but I think the formula with the minus sign was introduced by HP) .
It is generating more confusion than is needed.
Everywhere else in electronics we have formulas to convert something into dB (voltage, power, sound) , and we write 20 log x/y or 10 log x/y, NEVER with a minus sign.
I have been teaching this stuff for over 35 years, and my aim was always to keep things simple for my students, it was already "complex" (pun intended) enough for them.
So, from my textbook :
Return Loss = 20 log |rho|, NO minus sign. (|rho| being 0 ...1)
Then this agrees nicely with |s11| in dB , which is also a negative number (and everyone agrees on that one).
Why have two different signs for two things that are essentially the same ?
Now why is there this confusion all the time ?
One is because IEEE says so ... but do we always have to agree ?
But it comes also from the difference in what you SAY , and what you WRITE.
I will also say "my antenna has a return loss of 26 dB" (positive), and people will get what I mean from the context.
But when I WRITE it down in a table for a report, I will note " - 26 dB".
Just as I would measure a filter, I will SAY that it has a loss of 3 dB ... but in a level diagram I will neatly WRITE next to the filter block : -3 dB
It would be strange to write +3 dB, how will you then explain the input and output levels of that block ?
Or say you measure a bandpass amplifier, will you say in band it has a GAIN of 20 dB, and out of band it has a LOSS of 10 dB, how will you draw a response curve with all positive numbers ... ?
Again, you can say it so, but you will WRITE down +20 dB and -10 dB.
A good thing that Nano VNA Saver at least has the option to select the sign for Return Loss, I think you can guess what I selected ? ;-)
And whatever belief you have, the line of measured RL will still be BELOW the 0 dB line ...
73,
Luc ON7DQ
