rickman wrote:

On 9/28/2015 3:22 PM, Wayne wrote:


"Jerry Stuckle" wrote in message ...

On 9/28/2015 2:27 PM, Wayne wrote:


"Jerry Stuckle" wrote in message ...

On 9/28/2015 12:47 PM, rickman wrote:
On 9/28/2015 10:38 AM, Jerry Stuckle wrote:
On 9/28/2015 12:03 AM, rickman wrote:
On 9/27/2015 10:39 PM, Jerry Stuckle wrote:
On 9/27/2015 9:46 PM, Wayne wrote:


From LUNA web site regarding optical measurements which should be no
different from RF...


It "shouldn't be" - but optical measurements are handled differently
than electrical measurements. Fiber Optics have their own way of
measuring loss, reflection and refraction (which doesn't exist in
feedlines).

That's like applying electrician's color codes to electronics. They
both have color codes - but don't hook the electrician's black wire to
ground - or the transformer's green wires to safety ground.

I thought you would claim optical was different. That's why I included
the VSWR vs return loss table link. You didn't comment on that.


# I didn't because I thought it was obvious. But I guess not to you.

# Return loss is calculated with logs. Logs of values 1 are negative.
# And -10db is smaller than -5 db.

# As the SWR approaches 1:1, the reflected power approaches 0, and the
# returned loss approaches NEGATIVE infinity. Note that I said NEGATIVE
# infinity. At the same point, the returned power measured in watts
is 0.

Return loss is a positive number for passive networks. The equation has
(P out/P reflected). P out will never be less that P reflected, and
thus return loss will never be negative. (for passive networks)

As the SWR approaches 1:1, the return loss increases in a positive
direction, finally reaching infinity.


# No, return loss is calculated as P reflected / P out. P out is the
# constant with varying load; P reflected is the variable. The ratio is
# always less than one, hence the calculation is always negative DB.

# Please point to a reliable source which agrees with you.

I have never heard return loss expressed as a negative for passive RF
networks.
In fields other than RF I suppose anything is possible.

Here are some references, searching only for RF definitions:

http://www.ab4oj.com/atu/vswr.html

This page is clear, but it is just one guy citing no references, not
exactly authoritative.


http://www.mogami.com/e/cad/vswr.html

This page is not at all clear and only expresses return loss in terms of
the impedances. Following through on the calculations give negative
values for RL.


http://www.microwaves101.com/encyclo...swr-calculator

Offers a definition of RL in terms of gamma. Far from authoritative.
Definition of gamma is in terms of VSWR and VSWR in terms of gamma.


http://www.amphenolrf.com/vswr-conversion-chart/

Convention is to give positive values in tables.


From wikipedia: https://en.wikipedia.org/wiki/Return_loss
Return loss is the negative of the magnitude of the reflection
coefficient in dB. Since power is proportional to the square of the
voltage, return loss is given by,
(couldn't cut/paste the equation)
Thus, a large positive return loss indicates the reflected power is
small relative to the incident power, which indicates good impedance
match from source to load.

You didn't cite the relevant section...

Sign

Properly, loss quantities, when expressed in decibels, should be
positive numbers.[note 1] However, return loss has historically been
expressed as a negative number, and this convention is still widely
found in the literature.[1]


http://www.spectrum-soft.com/news/fall2009/vswr.shtm
The return loss measurement describes the ratio of the power in the
reflected wave to the power in the incident wave in units of decibels.
The standard output for the return loss is a positive value, so a large
return loss value actually means that the power in the reflected wave is
small compared to the power in the incident wave and indicates a better
impedance match. The return loss can be calculated from the reflection
coefficient with the equation:

This one is a bit confusing. The words in the first sentence say one
thing and the rest say something different. I don't see an equation in
terms of the power that I can understand without interpreting their
programming language.

At this point I guess the issue is very muddy with usage in both camps.
I think the wiki page indicates that with the section on Sign.

What really bugs me is why return loss is in terms of the loss while
other loss figures are in terms of the remaining power. Or did I find
an anomalous reference for that one?

It would probably make more sense to call return loss "return gain",
but, since it is always less than one, that would merely cause a
different set of ambiguities.


--
Roger Hayter