A Johnson Matchbox is itself a BalUn.
I think I made that point.
>> The link to coil is essentially a tunable balun. <<
If your station mandate shouted from mountaintops is there shall be no BalUns, you can't have a Johnson Flashbox.
I never wrote that. see below:
>>Anytime efficient power transfer and balance can be achieved in a transmitting system without the use of a toroidal or bead balun or unun, an improvement has been made. <<
Some of us cannot improve doublet layouts due to space constraints.
Tell me about it. 50 x 100 foot lot here.
fixed ratio line transformers aren't all bad. If you have a situation where R is transformed with no X and you never change frequency they're okay.
But for QRO AM operating all over HF, I think they suck.
Moving on....
The harmonic output of a single ended final with link coupling shouldn't be much different from that of a single ended final with a pi-network, if in each case the loaded Q of the tank circuit is at least 10 to 12, and the link is coupled to the coil at the cold end. Of course, the push-pull final has the advantage of theoretically balancing out even harmonics, and since with push-pull, the tuned circuit is excited with two kicks per RF cycle instead of one, the output is less dependent on the flywheel effect at the tank circuit to generate a pure sine wave, so push-pull should be cleaner in terms of harmonics. That has nothing to do with link coupling or pi network. A push-pull final may be built with a balanced pi network output. Circuits are described in the pre-WWII Radio Handbook, and that's what is in the military BC-339 using a pair of 833As.
I don't disagree with the facts above which appear to be sound, but in practice, rigs with link coupled outputs and non-p.p. finals don't seem to do well with second harmonics. For example, I have never run a BC610, however I've heard and read that the 250TH to the link coupled output on 160, puts out an 80 meter harmonic that's only about 20 dB down from the fundamental. That's enough to get a qrp signal out on 80 m. There could be other causes obviously but it seems to be widely known that the 610 unmodified to OWL and dipole is a N.O.V. risk.
"Tuned" or "resonant" feeders means that the OWL line is not matched, i.e., it is working into a load that is different from its characteristic impedance, Zo. With tuned feeders there IS a mis-match between the line and the load, so the tuned feeders operate with standing waves, and depending on the load at the opposite end (which may or may not have a reactive component) and the length of the line, the antenna tuning unit may end up seeing a wide variety of load impedances as the frequency is varied, and only rarely is that load likely to be purely resistive.
"Untuned" or "non-resonant" feeders means no mis-match between the line and the load, so that the OWL is "flat" or "matched", working into a purely resistive load that is the same as its characteristic impedance Zo. No standing waves exist along the line and varying the length of the transmission line has no effect on the load seen by the ATU. This is the balanced equivalent of coax working into a perfectly matched load with 1:1 SWR.
My mistake. I apologize for the incorrect information I posted above. I went to my 1934 ARRL Handbook and re-read the section on feeders. I had everything backwards. To me the term "Tuned Feeder" is counter-intuitive. I think of tuning the feeder, i.e. adjusting its length and width to achieve a characteristic Z that is the same as the load so it is flat and therefore tuned.
I'll have to study more of the section on feeders obviously.
The 1934 handbook is great. it is subtitled as "A manual of amateur high-frequency radio communication." It is written in a way that is plain spoken lacking in the egg-head engineer mumbo jumbo that renders the later editions slightly impenetrable. Somewhere in the mid to late 1930s, the ARRL decided to turn the handbook into a quasi-professional resource and began to toss out the clear and easy to understand language for impressive sounding buzz words that a regular ham had to de-code. The earlier editions were free of that as they were genuinely trying to help hams set up stations, instead of vying for space on a library reference shelf.
