Uncertainty Representation of return loss values in dB

[hansjoer...@gmail.com]

Hello,

I noticed, that there seems to be an issue when viewing uncertainty of return loss magnitude in logarithmic mode (especially with "very good" return loss values - when the measured s11 magnitude is (much) smaller than the given uncertainty):
I may show you here two screenshots from the same return loss with uncertainty in lin and log representation:

 

The linear scaled plot shows a pretty good S11 uncertainty of 0.008, the plot in dB shows an uncertainty range of about -500 dB to + 400 dB!

In this case the nonlinearity of the logarithmic scale becomes important and - i believe it becomes invalid to work with just one (symmectric) uncertainty value.
It might be a solution to calculate the uncertainty still as linear values and just convert the "border-value" (depending on the selection in the UI - standard or U95) to dB for viewing it in the diagram) to dB for visualization...

As comparison to the above log plot I converted the linear magnitude uncertainty U95-borders in excel to dB and generated an diagram:

But of course the effect of this issue is not that dramatic for usual return loss values,
here for comparison an example of a cable return loss where the nonlinearity effect is visible, but the deviation is not as big as in the first example:

  

Of course it is possible for me to use the "Mag lin log" representation where everything works fine
- it´s just the thing that I´m used to the dB representation of S11 and therefore it would nice for me to work here also with dB values...

Best Regards,
Hansjög Spiegel