Topband: Wideband Balun...?

[Jim Brown]

On 8/28/2012 12:28 PM, Tom W8JI wrote:
> Generally a 1:1 ferrite core balun has the least problem with grossly
> mismatched impedances, the widest useable frequency range, and the best
> balance for a given size and cost.

In this tutorial      http://audiosystemsgroup.com/RFI-Ham.pdf

There are graphs of measured data for five different ferrite materials, 
with the number of turns varied from 1 to 14, for several useful designs 
using a bfilar winding configured as a parallel wire transmission line, 
and for a dozen or so practical chokes formed by winding multiple turns 
of coax through multiple ferrite cores. There's also a "Cookbook" that 
makes specific recommendations for all the HF bands individually and in 
combination.

> Air core would be the most restricted in impedance and bandwidth, and have
> the poorest balance over a wide range of load conditions. It also would have
> very high Q, being mostly reactive in any "choking" impedance, which means
> it could actually make balance worse under many load conditions.
>
> Let me give an example. If I had an aircore balun with a common mode
> impedance of 5 +j500 ohms (Q=100) in series with a feedline having a common
> mode impedance of any reasonable R with a reactance of around -j250 or
> higher, the balun would make unbalance increase.

Yes. I published this in an AES Paper in 2003.

> Those big air core coax baluns you see will actually make many systems have
> worse balance, because they are largely a pure reactance. They only help
> balance under certain conditions, contrary to what is often assumed.

Right.

> As a general rule we do not want a high-Q balun, but something with very
> high resistance. We have to compromise that at times because of heat and
> impedance limitations, but almost always one of the very last things we want
> is an air core.

Right, and exactly for situations like Tom's example above.

> I use air core baluns on some of my antennas, but I control the common mode
> impedance to ensure the baluns actually do what I want. I never use them in
> random situations.

I think we need to stop using the word "balun" to describe a half dozen 
different things that are fundamentally different.  That generates 
massive confusion and lack of understanding.  Everything being discussed 
in this email is a common mode choke, and it works by placing a high 
resistive impedance in series with the common mode circuit (the shield 
of the coax, or in the case of parallel wire line, current that is in 
phase and in polarity on both conductors, and that radiates).

The most effective common mode choke is a very low Q parallel resonant 
circuit whose impedance is within about 55 degrees of resistive at the 
frequencies of interest and whose resistive component is at least 5K 
Ohms.  The higher the resistive component of the impedance, the more 
power the choke can handle. I consider 5,000 ohms a minimum design 
value, and higher is better.

BTW -- Tom's been designing and building chokes and various sorts of 
commercial baluns for a long time, and I published my first coax choke 
designs in 2006.

73, Jim K9YC