[CQ-Contest] Coax Stubs for SO2R

[Jim Brown]

On Mon,7/18/2016 12:08 PM, Kenneth Wolff wrote:
> Jim right of course -- the exception being the old Alphas with Pi-L
> outputs.

There are many amps with Pi-L outputs.  There's a table of those I've 
been able to learn about (with help from friends) at this link 
http://k9yc.com/Coax-Stubs.pdf


> At the the old K1AR M2 we ran from my house we used a lot of stubs because
> we only had two towers.
>
> Stubs made with hard line have less loss and are deeper,  but narrower
> nulls.

Yes, and they can be TOO narrow, locking you into a narrow range of 
operating frequencies.

>   Fixing 40 into 15 is a little trickier because it as a third
> harmonic. I used an open 1/4 stub on 15m with a compensation stub to tune
> out the imaginary part of the leftover conductance on 40.
>
> Double and even triple stubs slightly offset across the band can be used to
> make a broader null.

Yes.

> The most important thing to do is to measure the coupling from each
> transmit antenna into each other antenna using 10w transmit and either a 50
> ohm scope or a calibrated receiver.

Yes, a very good move. But when we're talking stubs for harmonic 
suppression, we must measure that coupling at the frequency of the 
harmonic.

> I made a spreadsheet of all the coupling (transcontuce) between antennas to
> plan my defense between bands. Some bands needed only a receive filer.
> 1500W on 160m used to couple over 50W into the 80m vertical array. You can
> smoke receivers if you don't watch out!

Great advice.  A vector network analyzer or a calibrated spectrum 
analyzer and signal generator can also do this well.

73, Jim K9YC
>
> - Ken K1EA
>
> On Mon, Jul 18, 2016 at 12:04 PM, Jim Brown <k9yc at audiosystemsgroup.com>
> wrote:
>
>> Your logic is exactly backwards, Jukka, and it applies ONLY to optimizing
>> the relationship between the stub and the power amp. When the output
>> circuit element is a capacitor, the stub should be 1/4 wavelength from the
>> amplifier at the harmonic it is intended to suppress. When the output
>> circuit element is an inductor, the stub should be directly at the
>> amplifier output, OR a halfwave up the line at the harmonic.
>>
>> But the relationship between the antenna and the stub also matters a LOT.
>> A stub works by shorting the line at the frequency of the harmonic, and
>> will be most effective when applied at a HIGH Z point on the line AT THE
>> HARMONIC FREQUENCY. Most single-band antennas present a HIGH Z at the
>> second harmonic, so the stub should be some multiple of half waves from the
>> antenna.
>>
>> Note also the advantage of double stubs in portable setups where length of
>> the line between antenna and stub cannot easily be controlled. Even if the
>> first stub (nearest the antenna) is poorly placed, the second stub, a
>> quarter-wave nearer the transmitter, provides its full attenuation.
>>
>> All of this is covered in the applications note for which I posted a link
>> last night.
>>
>> http://k9yc.com/LocatingStubs.pdf
>>
>> The next link shows how we implement stubs in the trailer that W6GJB put
>> together for or CQP and 7QP expeditions.  Note how close the 80 and 40
>> antennas were to each other in the last setup, and the 40M dipole and C3SS
>> were to each other in the daytime setups. At 600W from an Elecraft KPA500,
>> we had zero problems with harmonics with CW stations on both bands.
>>
>> http://k9yc.com/7QP.pdf
>>
>> 73, Jim K9YC
>>
>> On Sun,7/17/2016 10:37 AM, Jukka Klemola wrote:
>>
>>> Placing a stub at output connector helps a tad in case your output circuit
>>> that shows a low impedance to ground at the harmonic. That is typical for
>>> nearly all our equipment that have a low pass circuit with a capacitor
>>> going to ground as the last component of the output circuitry; a Pi
>>> circuit
>>> output.
>>>
>>
>> _______________________________________________
>> CQ-Contest mailing list
>> CQ-Contest at contesting.com
>> http://lists.contesting.com/mailman/listinfo/cq-contest
>>
> _______________________________________________
> CQ-Contest mailing list
> CQ-Contest at contesting.com
> http://lists.contesting.com/mailman/listinfo/cq-contest
>