Date: Mon, 16 Apr 2012 09:16:13 -0700
From: Jim Brown <email@example.com>
Subject: Re: [TowerTalk] Balun Recommendation
On 4/16/2012 4:11 AM, Jim Thomson wrote:
> 22 x type 43 beads, of the 1.125 inch long x 1 inch diam x .515 inch ID
> variety, slid over RG-213 or RG-393 coax will also work.
Jim, you are looking at the elephant through a pin hole. You build and
run big power amps, and your measure of the effectiveness of a choke is
that there it doesn't blow up. A string of beads on coax that are not
lossy at the operating frequency CANNOT overheat -- there's not enough
resistive component to their impedance to dissipate power -- but they
are also essentially useless for all of the reasons outlined in my RFI
#### Look at Steve Hunts bar charts. His bead baluns depict a MAINLY
across the spectrum. That’s the thin black line below.
Steve only did 2 x bead baluns, one with 6 beads, and the other with 10
beads..and both bead
baluns used type 31 material. Id like to see a bar graph, but this time using
say 22 x type 43 beads,
slid over 213 U. Although its over 2k,and mainly resistive, its not a sky
The primary function of a common mode choke in an antenna that is used
for receiving is to MINIMIZE RECEIVE NOISE, and to do that effectively,
the choke must have a high resistive common mode impedance at the
frequency(ies) of interest. Ferrite chokes operate as parallel resonant
circuits. A single turn through cores made of #31 and #43 ferrite
materials is resonant around 150 MHz, so it is USELESS on the HF bands.
To be useful on the HF bands, we must wind multiple turns through the
material to increase both the inductance and the stray capacitance so
that the resonance moves down to the frequencies where we want to use
them. Until you understand that fundamental concept, you are going to
continue to make incorrect assessments of the value of various
solutions. I refer you to my RFI tutorial, which addresses the
technical issues in detail, including an analysis of dissipation at high
power levels that doesn't "hide behind the math."
73, Jim Brown K9YC
### I did read your tutorial, several time in fact..and also steves material.
I’m not prepared to use
RG-174, or RG-58 or Rg-303, RG-8X on any of my baluns. Nor am I prepared to
use bifilar wound
schemes with magnet wire etc. On some of your designs, using coax on cores,
they will work, but are impractical to use on a yagi. On some of them, the
loops are hanging and flopping
about in mid air . how are you supposed to keep the loops of coax in one place.
And how are you supposed
to keep the entire mess away from the Aluminum boom, and DE on a yagi.
## I use a remote switch box at top of tower, so only one feed line up the
tower. Each yagi gets its own type
43 string of beads over 393 coax. At the INPUT of the remote switch box goes
a line isolator. Just a balun with
coax connectors on each side. That line isolator consists of 4 turns of 393
coax wound onto 4 x FT-240 size
cores. The 4 x cores are type 31 material. The idea here is, the type 43
bead baluns are way out onto the booms.
They have to be 100% reliable, as I can’t get to them without taking the yagi
down. The type 43 bead baluns will
handle the power, and the single 4 x core type 31 line isolator balun at the
input of the remote coax switch box, will kill
any residual rf spill over. Between the 2 of them, the Z should be plenty
high enough. The individual runs of coax from
remote switch box to each yagi are not too long. I put a 2nd line isolator
in the basement.
## each line isolator has its 4 x stacked type 31 cores + coax windings
mounted dead center in a plastic nema box.
That minimizes any stray C from coax windings to the mast its mounted to. If
the same nema box is mounted to a boom,
same deal. That’s the best compromise I could come up with.
## aside from RX noise, the ultimate test is to use a clamp on RF ammeter on
the coax, and take measurements
in several locations on the coax. That RX noise you speak of can be broken
down into 2 x parts. One is band noise,
the other is local noise,coming from a myriad of sources.
TowerTalk mailing list