[Top] [All Lists]

Re: [TowerTalk] 1/4w matching, Common mode chokes and BAL-UN

Subject: Re: [TowerTalk] 1/4w matching, Common mode chokes and BAL-UN
From: Steve Hunt <>
Date: Fri, 31 Dec 2010 20:05:57 +0000
List-post: <">>
I have some modelled examples on my web site which illustrate the effect 
Jim is talking about. Take a look at the section "Why reactive chokes 
are undesirable" on this page:

The section finishes: "We conclude that a high value Resistive choke is 
the safe option for all scenarios. We also conclude that Rules-of-Thumb 
which equate the required choke impedance to some multiple of the 
differential-mode load impedance are unsound."

Steve G3TXQ

On 30/12/2010 04:52, Jim Brown wrote:
> On 12/29/2010 7:35 PM, wrote:
>> I think a more accurate description of several coiled turns of coaxial
>> cable would be a RF Choke, serving to limit common mode currents on the 
>> outside
>>    of the coax cable.
> That's an Alice In Wonderland view of things. Yes, the coil of coax is
> an inductance in the common mode circuit, but the common mode circuit
> also includes the series impedance of the transmission line. If that
> line is less than 1/4 wave, it is a CAPACITIVE REACTANCE, so the
> inductive reactance CANCELS all or part of it and the common mode
> current INCREASES rather than decreases.  The same thing happens if the
> line is between 1/2 and 3/4 wavelengths. Further, the electrical length
> is based on a Vf of 0.98, NOT the Vf of the coax (because the common
> mode current flows on the outside of the shield).  The 0.98 factor
> accounts for the slight shortening effect of the outer jacket.
> The only effective common mode choke is a very lossy parallel resonant
> circuit, which is what you get if you wind multiple turns of coax around
> a lossy ferrite core (like #31 or #43). All you have to do is use the
> number of turns that puts the very broad resonance where it covers the
> bands you work.
> Study
> 73, Jim K9YC
> _______________________________________________
> _______________________________________________
> TowerTalk mailing list

TowerTalk mailing list

<Prev in Thread] Current Thread [Next in Thread>