Critically review your measurements
A recent post on an online forum provides a relevant example to discussion of this subject.
I have personally seen ratios similar to 3:1 or higher at the feed point become 1:1 at the rig over 100 or so feet of coax cable.
First point is that in good transmission line, it takes an infinite length to deliver the observations made above. Less might deliver almost VSWR=1 at the input end of the line.
Let us consider a practical scenario, 100′ of RG58A/U with a load of 150+j0Ω at 14MHz, the load end VSWR(50) is 3, the input impedance is 32.50-j22.86Ω and input VSWR(50) is 2.01. In this scenario, the line loss is 2.5dB which might be unacceptable for some applications.
The original statement should sound a warning that something does not seem right and bears detailed investigation to determine if the measurements were sound, and whether the coax was good.
So let’s say antenna A has a 1:1 feed point SWR at the design frequency and a 1:1 SWR at the radio. Let’s also say antenna B has a 3:1 feed point SWR at the design frequency and a 1:1 SWR at the radio.
Which antenna or antenna system is better and why?
Case A is possible, indeed if VSWR=1 at the load end, VSWR=1 at the source end (assuming good transmission line).
For case B, a repeat of the implication that it is practical to see VSWR=3 at the load end of the line and VSWR=1 at the source end. The scenario implies extreme loss in the line section.
Case B should sound a warning that something does not seem right and bears detailed investigation to determine if the measurements were sound, and whether the coax was good.
Let’s use a real world example. I have a 12/17 trapped dipole that I am about to put on the tower. When I configure the antenna as per specifications and test it at the feed point, the SWR is 4:1 at the intended frequency. The edge to edge bandwidth should be (and is advertised as) easily less than 2:1 due to the design. However, the dip occurs well above the intended frequency and the intended frequency does not fall within the dip. But of course, my radio sees a much lower SWR (maybe 1.5:1) at the end of the coax cable.
So, load end VSWR=4, and ‘maybe’ VSWR=1.5 at the source end.
These figures imply an approximate transmission line loss. The load end ReturnLoss=4.44dB and source end ReturnLoss=13.98dB. The one way matched line loss is approximately half the difference in these, (13.98-4.48)/2=4.75dB. Loss under standing waves would be higher, closer to 6dB.
Again the observations should sound a warning about accuracy of the observations and the transmission line quality.
Supposition and fact
Supposition is a really important input to fault diagnosis, the ability to create relevant suppositions assists diagnosis.
Suppositions are not fact until proven so.
Confusing fact with supposition frustrates logical diagnosis, always distinguish between fact and supposition.
Investigate apparent inconsistency
When you make measurements, think about the implications and whether they are inconsistent with the thing you are measuring, or inferences that they make.
One of the causes of apparently inconsistent results is disturbing the thing that you are measuring, that a pair of measurements might not be of the same thing and therefore are not comparable. An example would be above if the source and load end measurements were not of the same thing, that some side effect of measurement was to change the antenna system, then the results are not comparable and conclusions may be very limited. Exploiting your antenna analyser #9 dealt with this in more detail.
Chasing these things down finds the correct answer and contributes to learning and sound knowledge.
The original poster makes a number of statements as fact, and they are worthy of review.
Many of the stated ‘facts’ may be just supposition, they may be explained by loose talk, poor measurement accuracy, and/or defective transmission line.
One of the possible causes of apparently inconsistent measurements is disturbing the thing you are measuring.
- Duffy, O. 2001. RF Transmission Line Loss Calculator (TLLC). VK1OD.net (offline).