[AMRadio] LINK COUPLING)

[Donald Chester]

Brian wrote:
 
<<<<<In the past, I have run full power for several minutes (1500 watts of
carrier only) from my AM transmitter through this link coupled tuner I
built, unkeyed and then felt the main output coil.  Barely even warm to the
touch.  Try that with a ferrite balun! >>>>>
 
 
Another problem with the ferrite balun is that they may be driven to
saturation (magnetically) so that they become non-linear. This can generate
a broad signal, just like non-linearity in the final tube.
 
Many of the "21st Century" Hammy Hambone tuners (actually they have been
doing this for decades, but undoubtedly some would deride link coupled
tuners as not "adapting") are unbalanced L- and T- networks feeding the OWL
through a balun. This would work fine  for an untuned (matched) balanced
transmission line if the balun were specifically designed for the impedance
of the line, but very poor design for working into a tuned transmission line
that may have a substantial reactive component and random resistive
component depending on frequency.  I would go so far as to classify those
things as BOGUS.

A better method would be to place the balun between the transmitter and the
input to the tuner, and then float the whole tuner above ground by mounting
it on stand-off insulators or, if nothing else, supporting it on a pile of
books or upside-down beer glasses. Of course  this makes the case of the
tuner, panel and adjustment knobs hot with RF, and besides an rf burn, you
may get a hand capacity effect causing the resonant frequency to vary a bit
when you take your hand away from the knob. Much better to build a balanced
tuner to begin with. There was an article on a balanced L-network several
years ago in QST, I believe.

Some of my "link coupled" tuners do not actually use a link, but the low-Z
line is tapped directly to a few turns of the coil near the midpoint. This
will sometimes, but not always, work; plus you have to be careful about
giving common mode currents a convenient path to ground or to the feed line
when using this method.

My tuner coils are made from EF Johnson edge-wound coil stock or equivalent.
This is much more convenient for moving taps around on the coil, than using
round wire, especially if it is enamelled or corroded. I do have a few clips
that attach to round wire,  but they don't work as well as the edge wound
coil & clips (if you  can find the clips). Rob and I each found a bunch of
edge wound coil clips at Dayton this spring for few bucks total, but newly
manufactured ones are listed by one company at $20 each, and when Ripoff
Sales of Nebraska had them in stock they wanted $9 each. When EFJ was still
making them, they were a few cents each.

I made up a special coil for matching my balanced 440-ohm untuned OWL
transmission  line to the 160m ground mounted vertical. The first prototype
was designed pretty much by trial and error, using some old junky coil
stock, which was corroded and the plastic support strips had virtually
self-destructed. I managed to  cobble the  coil assembly together using
epoxy and hot glue, and experimented with the number of turns in the primary
and secondary, resonating capacitance, etc. When I finally got it to do what
I wanted, I was getting noticeably better efficiency using the OWL than I
was able to achieve with coax. After thoroughly testing it and deciding this
was indeed to be my final design, I dug up some EFJ edge-wound coil stock
that happened to be in my junk collection and was able to make up an exact
replica of the prototype, using clean,  silver  plated, edgewound coil stock
instead of corroded round wire with disintegrated insulation supports. It
tuned up exactly like the prototype, so I put an rf ammeter in the line to
the base of the tower, eager to see how much more antenna current I would
get with the better coil. Turns out, I got EXACTLY the same results with the
silver plated edge wound stock as I was getting with the crappy old corroded
round wire coils with deteriorated insulation! No better, no worse, but it
certainly  looks prettier, and is practically bullet-proof, unlikely to fall
apart at any moment, as the prototype would have been.	

Don