However, if you have coax with good dielectric (polyethylene or
Teflon), at HF and below the loss is strongly dominated by the R term.
You can verify through measurements, if you are careful, that G can
be assumed zero unless you've done something to degrade your line's
dielectric. BUT...it's much easier to measure the line's attenuation
directly than to measure (accurately) the impedance's real and
imaginary parts anyway, so why would one try to do it that way?

Cheers,
Tom

(Example: RG174 at f=30MHz will have a bit more than 3.4dB/100 feet
loss because of R, and probably well under .025dB/100 feet loss
because of G. See Roy's suggested reading for the source of those
numbers.)

"David Robbins" wrote in message ...
"Cecil Moore" wrote in message
...
David Robbins wrote:

"Cecil Moore" wrote:
Does (R+jXL)/(G+jXC) really equal 2500 for RG-174 on 12m? The specs
say the Z0 of RG-174 is a nominal 50 ohms.

of course its not exactly 2500, otherwise there would be no loss. but
its
close, maybe 2500+j10 or something like that. and even the resistive
part
may not be exact, the nominal 50 ohms could be 45 to 55 depending on the
tolerances of the manufacturer.

Comparing the 6dB loss of RG-174 to the 0.14 dB loss for hardline -
is all that extra loss accounted for in the +j10 term?

no, its more complicated than that.

the attenuation constant (usually alpha) = Re(gamma) where gamma is
sqrt((R+jwL)(G+jwC)) Zo is sqrt((R+jwL)/(G+jwC)) so there is not a simple
way to relate the characterisitic impedance to loss. for a low loss line
the approximation for alpha is (R/2Zo)+(GZo/2) which can probalby be applied
for most normal cases, but again, you have to get the R and G values of the
line which can not be directly calculated from Zo.