[TowerTalk] Current Baluns
Jim Brown
jim at audiosystemsgroup.com
Sat Mar 20 00:36:37 EDT 2021
On 3/19/2021 8:01 PM, john at kk9a.com wrote:
> That is interesting, a balun wastes power. How by minimizing feedline
> currents?
If you include audio and video systems, the word "balun" has been used
to describe nearly a dozen VERY different things over the years. I can't
think of any commonly used in ham radio that waste any more transmitter
power than the very small loss in the short section of transmission line
used to wind them. This table shows the loss for my RG400 designs for
2.4-in o.d. cores. Depending on the band, worst case differential
dissipation in recomended choke for bands where they are intended to
operate is in the range of 0.06 dB, and is due to ordinary transmission
line loss. For legal limit operation, two chokes in series may be
required. My computations for loss assume about 6-in for leads.
http://k9yc.com/Chokes-2r4inRG400.png
There is also some dissipation in the common mode circuit, but the
greater the choking impedance, the lower the loss.
>
> FWIW, I use Array Solutions W1JR baluns which are really chokes.
Yes. But unless Joe has changed his design, they are not very effective,
because he uses Fair-Rite #61, so the choking impedance inductive at HF.
He and I talked about this 10-12 years ago at Visalia. An inductive
choke can have its impedance cancelled if it is inserted in a line whose
common mode impedance is capacitive at the frequency of interest.
My designs use the resistance at the choke's very broad parallel
resonance to limit common mode current. This, by the way, is exactly
what W2DU did many years earlier when he chose #73 material for his
"string of beads" designs. Those beads, by their nature, have a parallel
resonance in the middle of the HF spectrum, but HUNDREDS of them are
required to achieve high enough choking Z for even moderate power
levels, and all of those I've seen claiming to use his design use less
than a quarter of those used in his higher power designs.
73, Jim K9YC
More information about the TowerTalk
mailing list