walterb,
I can understand why you have drawn some of the conclusions you have, but I'm afraid that those conclusions are not completely accurate. For instance.
The higher the 'Q' of an antenna the less usable bandwidth it will have. And, the lower the 'Q' the more usable bandwidth it will have. There's no disputing that, it's been proven too many times.
The only 'job' any feed line has is to carry energy from one source to it's destination as efficiently as possible. That depends on several things, but that's all it (coax) was ever designed to do. One aspect of that is the shielding. It should be shielded enough so that things don't affect that energy it's carrying to any great extent. There will always be some affect from less than 100% shielding, but if it's kept to a minimum, it doesn't amount to anything significant.
The relationship between the center conductor and the shiled, the diameters of both and the distance between them is what determines the impedance of the coaxial cable. If those distances/diameters are proportioned correctly, the cable can be of almost any size (within reason). The insulation between those conductors also plays a part in that, some insulators are more 'reactive' than others. It also plays a huge part in how much power the cable is able to handle without arcing.
All feed lines are subject to 'leakage'. That includes parallel feed lines (ladder line), and is very easy to demonstrate, just get it too close to something. It also applies to 'hard line'.
When the feed line starts to radiate, it isn't just a feed line anymore, it's 'part' of the antenna. That can be used to beneficial affect, but it also destroys the purpose of the feed line. If a feed line is properly matched to the source and load, then the only losses to that feed line is resistive. It just don't get no better than that.
The differences in your example about swapping feed lines for your friend is a classic matter of an improperly matched antenna system. No two sections of coax cable will ever have exactly the same characteristic impedance, that includes sections from the same batch of coax cable. It's called "characteristic" impedance for a reason. It means that the impedance will be at least 'close' to whatever is claimed, but there's no guarantee that it will ever be exactly what it's said to be.
If you take that 100 feet thingy and just extend it a bit, a coax cable long enough doesn't even need anything on the end of it for the transmitter to think it's looking into a 'perfect' load.
The thing to do is dig into the -why- of it to find out how come it seems to work out that way...
- 'Doc
