[TowerTalk] decoupling feedline from a dipole

[Jim Lux]

At 09:07 AM 11/12/2006, Jim Miller WB5OXQ wrote:
>I hear that decoupling the feedline can reduce noise so what is the 
>best way to do to.  I have tried commercially made baluns in the 
>past only to have them burn out at legal limit power levels.  Is a 
>choke wound from the feedline like I made for my beam going to do 
>the job or something else.  I am plagued by 10 to 30 db over s9 
>noise on 75 especially using an inverted v at 55'.  Noise seems 
>worse on AM when no carrier is present than on ssb also.


The W1HIS paper on decoupling ia a good start
http://www.yccc.org/Articles/W1HIS/CommonModeChokesW1HIS2006Apr06.pdf

Part of the challenge is that a "single" choke may not do the 
job.  You might need several, at different places along the 
feedline.  The chokes/decoupling act in the same way as breaking up a 
long guy wire with insulators. Except that the choke isn't an 
insulator, but a combination of inductance and resistance.

So, heating would be because the impedance is too low (if it were 
"really, really" high, then not much current would flow) and the 
current that flows through the "choke" causes heating in the R component.

The usual guideline is that the choke impedance should be >10x the 
feedline impedance, but that's not necessarily an appropriate number 
(i.e. it's really the impedance of the "outside" of the feedline 
that's relevant). But even if you took 50 ohms as the nominal 
impedance.. and had a 500 ohm choke impedance, 1/100th of the power 
is going to flow through the choke, and with legal limit, that's 15 
W.  You're not going to successfully dissipate 15W in a little clamp 
on ferrite (at least not for very long).