[TowerTalk] Coax Losses on 160 and 75?

[Richard (Rick) Karlquist]

1.  For the special case of a uniform transmission
line, you can completely characterize it with two
1-port measurements, having open and short terminations.
These measurements are denoted Z11 and 1/Y11.
The short and open will not be perfect, so there will
be some error.  It is possible to account for this
error, but it would be beyond the ability of most
hams to do this.  [For the general case of an
arbitrary 2 port network, you can completely
characterize it with 3 measurements, but the math
is quite messy.]

2.  There is a great dilution of precision with a
1 port measurement if the loss is small.

3.  Short cuts involving 1/8 wave or 1/4 wave simplify
the math but you still need a perfect short or open and
you need to determine the exact frequency for the 1/8
wave.  And you still have dilution of precision.
And it presupposes you have at least as much as 1/8 wave of coax
to work with.  On 160 meters, that is at least 50-60 feet.

4.  If you have the luxury of a 1/4 wave (or multiple
thereof), the preferred way to measure loss is to measure
the *unloaded* bandwidth of the resonance.  Very accurate, doesn't
even require a network analyzer.  You do have to understand
what you are doing however.  But there is no complex math.
I have tried to teach this technique to clients with varying success.

4.  Which brings us back to the 2 port network analyzer.
Plug and play.  Anyone can do it.  Always works.
With the weak pound, the VNWA is down to a half a kilobuck or so.

Rick N6RK