[TowerTalk] measuring old coax, return loss ?

[Wes]

Dave,

Without wading through the operation manual of your instrument I'm going to 
speak in generalities.  Generally when measuring something like a cable or 
attenuator, it's nice to be able to measure at both ends of the device under 
test (DUT).  A two-port measuring system like the DG9SAQ VNWA that I have does 
just that. Port 1 of the device outputs a signal that goes into the DUT and Port 
2 measures the signal that comes out the other end.  Before dong the 
measurement, the VNWA is calibrated by placing a short circuit between Port 1 
and Port 2. This condition is defined as zero loss.  It's a little more 
complicated than that but it demonstrates the concept.

Now what if your measurement device can only measure one port? Using a direction 
coupler or bridge, one port can serve two purposes, it can supply a signal in 
one direction and measure a signal coming from the other direction. All of the 
SWR measuring devices built into most modern radios do just that.  They measure 
the power (or voltage) going into the load and measure the power (or voltage) 
returning from the load.  Using some math they present SWR, forward and/or 
reflected power or return loss.  All of these labels are just different ways to 
express the same thing.

Now, if you want to know the attenuation of a cable connected to your radio or 
the test port on an analyzer you can't just terminate the cable with a dummy 
load and measure the ratio of forward to reflected power, since with a good 
load, there isn't any reflected power to measure.  Even with a so-so load, most 
of the incident (forward) power is absorbed in the load and what is reflected 
will give an inaccurate ratio.  What we need is a load that reflects 100% of the 
power back to the source.  There are only a few loads that do this, an open or a 
short are the usual suspects.

Now what happens (in simple terms) is that the source sends a signal down the 
line. On the way to the load the signal suffers loss, so not all of it gets to 
the load.  Let's say for demonstration purposes that only one-half of the source 
power makes it to the load.  At the load, that one-half is totally reflected and 
heads back to the source.  On the way it suffers the same loss in the other 
direction so only one-quarter of the original power made it back to the source.  
That's a total return loss of 6 dB, but the loss is due to two trips through the 
cable, so the one-way loss is 3 dB. That is the cable loss.

In the real world when better accuracy is required a lot of error sources come 
into play.  For instance that directional coupler isn't perfect.  There is 
always unwanted leakage signals between ports.  It's also assumed that the 
source doesn't re-reflect power returning from the highly reflective load.  
That's not true either.  There are more weeds but I'll leave it at that.

Wes  N7WS



On 5/8/2024 6:23 PM, Dave . wrote:
> I have a aligent 8920a..it has been calibrated...I have the Aligent return loss adapter...two 6db attenuators
> And a Huber Shuner power divider...and a 50ohm dummy load........I also have the RF Tools software for the Aligent..
>
> So..is it better to measure with the 50ohm dummy termination.....OR a Short?  And what is the difference in the readings shown?
>
> I'm sure this is easy peasy for you..however we are eager to learn and can't wait to learn more
>
> The videos on how to do this procedure are lacking ...very lacking......I was hoping that there would be some from Aligent..but none..
>
> Thank you in advance.
>
> Dave
>
>