[Elecraft] Non-Resonant Antennas and Chokes

[Jim Brown]

On 1/13/2019 6:46 PM, Al Lorona wrote:
>>>> the feedline is a wire dangling from the antenna
>>>> that isn't connected to anything on the other end.
> That "dangling wire" is actually *two* wires, and the field of one cancels the field of the other for no net radiation or reception -- at least that is the condition we're trying to achieve. That isn't the same as a single wire which by definition would be a common-mode conductor as you correctly say.
You're confusing common mode with differential mode. Differential mode 
is the nicely equal currents at every point on the line, which do, 
indeed, cancel. But common mode current is the DIFFERENCE between 
currents that are not equal.
>
> Once common-mode current is reduced to a small enough value, then the open-wire line isn't radiating nor receiving. But... you gotta measure the common-mode current to know. I have done so.
WHERE are you measuring it? You do realize that, like any antenna, 
current varies along the wires that make up the antenna. A choke 
attempts to force the current to zero AT THE POINT WHERE THE CHOKE IS 
PLACED. You're putting it at the tuner, so it forces current to zero 
there. But a quarter wave up the feedline, the current reaches a maximum 
value.
> The instrument used to measure this is quite simple.
  I'd be very interested in how you're measuring the current at the 
antenna feedpoint. :) That is NOT so simple.

>
>>>> a choke that doesn't fry with TX
>>>> power probably isn't doing anything useful.
> A choke dissipates power only in it's resistance, not it's reactance, and only due to the common-mode current. If the parameters of the choke are chosen correctly for the frequency band, and if the choke reduces common-mode current to a low enough value, then the power dissipated in the choke can be very low even when you're operating at high power.
Yes.  Because power is I squared R, it is changing twice as fast as R is 
changing, and current is determined by R.  So making R very large makes 
current very small, which minimizes power dissipated by the choke.
> A choke that burns up at high power is certainly not inevitable nor normal and can be fixed by re-designing the choke. There are a number of ferrite mixes available and they seem to each be optimum for a slightly different part of the HF spectrum.

There are, in fact, nearly two dozen different ferrite mixes in the 
Fair-Rite catalog, but only a small fraction of them are useful for 
common mode chokes on frequencies that we care about. #31 material is 
the most useful between 160M and 2M; #75 is useful between 630M and 40M. 
I'm currently investigating a relatively new material that MAY be useful 
on the higher HF bands, but it's a fairly high Q material, and 
Fair-Rite's cores are a pretty wide-tolerance part. My recent Cookbook 
is based on having measured more than 200 cores, then winding chokes on 
cores that are at the limits of those I measured, and making 
recommendations on the basis of worst-case results from those cores at 
the limits. That works for #31, because it's a very low Q part in this 
range.

Well designed ferrite chokes are very low-Q parallel resonant circuits 
are resonant (or near resonant) in the frequency ranges where they will 
be used, and it is the very high value of resistance that makes it a 
good choke. Why? Because a choke that is mostly inductive, with very 
little resistance, can resonate with the rest of the transmission line 
in the common mode circuit, INCREASING the common mode current rather 
that decreasing it. But resistance ALWAYS reduces the common mode 
current. A choke without a lot of resistance is quite sensitive to the 
ELECTRICAL length of the feedline, which, of course, increases with 
increasing frequency.  That's why a coil of coax, or coax wound on a low 
loss toroid like Fair-Rite #61 is a lousy choke.

> There's also the twist that permeability is actually a complex quantity (real and imaginary parts) which has a direct bearing on the resistance and reactance of the choke, but I won't go into that at this time.

Yep. Folks can read about it in k9yc.com/RFI-Ham.pdf  which is a 
tutorial that I wrote more than ten years ago. There are two components, 
mu' and mu''; the first is what we've always called mu, and describes 
the inductance; the second tells us the value of a resistance in series 
with the inductor that, from a circuit analysis point of view, defines 
the loss coupled to the wire from the core. And both mu' and mu'' both 
vary (a LOT) with frequency.

I'm not going to go beyond this on the reflector, but will refer 
interested readers to that tutorial, and to my latest work, a major 
update of the Choke Cookbook published in the first edition of the 
tutorial. The update is k9yc.com/2018Cookbook.pdf

73, Jim K9YC