On 9/28/2015 12:05 PM, John S wrote:
On 9/28/2015 10:47 AM, Ian Jackson wrote:
In message , rickman
writes



Definition of Return Loss

In technical terms, RL is the ratio of the light reflected back from a
device under test, Pout, to the light launched into that device, Pin,
usually expressed as a negative number in dB.

RL = 10 log10(Pout/Pin)

Here is a link for a table of return loss and VSWR....

http://www.jampro.com/uploads/tech_d.../VSWRChart.pdf

It shows a higher return loss (assuming you mean magnitude since the
values are all negative) for lower VSWR.

I'm surprised to see negative quantities. For 50 years, I've always
understood the Return Loss Ratio (RLR) to be exactly what it says on the
tin, ie the ratio (in dB) of the LOSS (the attenuation) of the reflected
signal wrt the incident signal. This is a +ve quantity. Things are
already sufficiently confusing without having to start thinking in
unnecessary -ve figures!



I think you are correct. I think the confusion is the word *loss*. If
you have a positive *loss* number, the return signal is reduced. To have
a negative return loss number, you need to refer to *gain*.

For example, a return *loss* of 20dB is the same as a return *gain* of
-20dB.

In my other replies on this I mistakenly said the return loss in dB is
negative for all but zero reflection. That should have been for all but
100% reflection.

However, I did a bit more research on the topic and I noticed that
Wikipedia says for the general case a loss ratio in decibels should be a
positive number. But they have a paragraph explaining that by
convention the return loss ratio is a negative number which makes it
identical to the reflection coefficient.

I could not check the reference wikipedia gives for this as it is a
subscription document. I did find in the Electronic Engineers' Handbook
a formula for "system" losses starting at the antenna terminal which are
expressed as

Ls = 10 log (pt/pa) = Pt - Pa, dB

where Ls is the loss in decibels, pt is the power delivered to the
antenna terminals, pa is the power at the receiving antenna. The
capital form of these power levels are in dB.

I think the reason we see a different formula for return loss is because
in the above equation the ratio is the power sent to the power received.
The formula for the return loss is the ratio of the power provided to
the power reflected which is not at all the same ratio. To make the
loss equations measure the same thing return loss would be

Lr = 10 log (pt/(pt-pr))

where pr is the power reflected yielding the same ratio as pt/pa. This
would still be a positive value in dB. So I'm not at all sure why the
convention of negative dB for return loss got started. Maybe it is one
of those issues where there is a lot of misinformation that never gets
corrected properly.

So I withdraw my statements since the issue is very confused in the
literature. Clearly there is no universal convention.

--

Rick