... it is 1/3 the loss, but the total amount of loss is insignificant in any case.
Dale, I guess I'm just a myth buster wannabe. We can argue over the definition of "insignificant" but not over the definition of "most efficient". The question is: Is the following statement true or false?
"If you're feeding resonant antennas that have a 50-ohm feed point, that would mean 50-ohm coax would be the most efficient path between your radio and the antenna."
I know, it's tough to be technically correct without becoming far too verbose to get
all the conditions properly stated!
I agree that the statement is not ALWAYS true, because it is possible to choose conditions
where it isn't (as you did). On the other hand, the examples I gave at 28.5 MHz the coax
(particularly choosing a low-loss type such as LMR-400) actually had slightly LESS loss than
the open wire line (again, depending on the exact parameters.)
Choosing the specific materials on hand (RG-8X and #14 copper to make open wire line)
with a 50 ohm load at 28.5 MHz (where losses will be higher) the calculated numbers are
1.6dB loss for coax and about 0.7dB for open wire line plus about 0.1 to 0.3dB loss in the
tuner depending on the electrical length of the line (assuming a lossless 1 : 1 balun.)
On 80m the corresponding losses are about 0.57dB vs. 0.25dB + 0.3dB tuner loss. At that
point we're splitting hairs.
Meanwhile, on 80m (and more-so on 160m) we have to consider the skin effect losses if the
open wire line is made using Copperweld(R), which VK1OD's calculator doesn't take into account,
as well as the losses in the balun, tuner, how the feedline is routed with respect to brick
walls, whether the insulation is removed from the copper wire, etc.
So the best we can say in general is that the losses are similar, while providing specific
examples where it could go one way or the other.