[TowerTalk] Comments by K9YC

Mon Jun 23 00:03:27 EDT 2014

> There are commercial antennas which use a quarter wave line with the
> top near the DE and the bottom connected to the feedline outer shield
> a quarter wave down the feed line. This should develop a high Z at
> the spillover point if there is one, Hi.

Yes, the quarter wave stub (Pawsley stub) raises the impedance seen
by the "grounded" side of the antenna.  It does not change the SWR
on the antenna or effect any mismatch between the impedance of the
feedline and the antenna.  The only thing it does is reduce the shunt
impedance from the outside of the coax shield.

The shield of the coax is directly analogous to the grounding strap
Mosley used on one side of its antennas - it's just a little longer.
If one were to correctly model a dipole fed with coax, one would
include *a third wire* the length of the feedline running from one
side of the feedpoint to ground.  With that third wire, one no longer
has a *balanced* antenna.

In fact, get EZNEC or another modeling program that will show you the 
currents in each segment.  Then play with the length of that third
wire to see what happens to those currents at various lengths of
wire ... what happens to "feedline radiation" and how beam patterns
can become corrupted/skewed.  It's quite interesting and educational.

73,

    ... Joe, W4TV

On 2014-06-22 11:28 PM, Dan Hearn wrote:
> Thanks for your comments guys. I am very confused. There have been a number
> of things in QST saying that the current from the center conductor goes
> into one half of the dipole DE and the remaining current goes partiallly
> into the other half of the dipole and some of it goes down the outside
> shield surface. I have never questioned that. There are commercial antennas
> which use a quarter wave line with the top near the DE and the bottom
> connected to the feedline outer shield a quarter wave down the feed line.
> This should develop a high Z at the spillover point if there is one, Hi.
> Here is what G0ksc says about it and he is a highly respected antenna
> modeler
> http://www.g0ksc.co.uk/creatingabalun.html
>    Frankly, I do not know what to believe. I have a rf current clamp on
> meter which I may use to explore this further.
>
> 73, Dan, N5AR
>
>
> On Sun, Jun 22, 2014 at 7:36 PM, Joe Subich, W4TV <lists at subich.com> wrote:
>
>>
>> On 2014-06-22 10:12 PM, Dan Hearn wrote:
>>
>>> The coax saw many places on the 3 bands where the feed point
>>> impedance was not 52 ohms. That will result in spill over of current
>>> from the inside of the shield to the outside at the attachment point
>>> and ultimately radiation which screws up the beam pattern and sends
>>> rf into your shack.
>>>
>>
>> The fact that the feedline is not working into a matched load does *not*
>> create common mode current.  Common mode on a feedline is created *only*
>> when the impedance to ground is different on each leg of the antenna.
>> If the impedance is the same - even if it does not match the feedline
>> impedance - the current into each leg of the antenna is the same, the
>> antenna is balanced and there can be no common mode current.
>>
>> Mosley used to connect one side of the driven element to the boom with
>> a strap.  Grounding one half of the driven element and connecting the
>> shield of the coax to the junction of the half-element and strap is
>> guaranteed to seriously unbalance the antenna, causing beam skew, feed-
>> line radiation, and common mode current.  In addition, if the boom plus
>> end elements happened to be a multiple of a half wave, the 1/4 wave
>> each side of the feed point along the boom could easily cause all kinds
>> of strange behavior.
>>
>> It is unbalance *not SWR* that causes common mode current to flow in
>> a feedline.  In a properly balanced system, the currents on the center
>> conductor and inside the coax are equal and opposite regardless of
>> their level or their phase relationship to the voltage and no current
>> appears on the *outside* of the shield.
>>
>> 73,
>>
>>     ... Joe, W4TV
>>
>>
>>
>> On 2014-06-22 10:12 PM, Dan Hearn wrote:
>>
>>> I recently sent the following to towertalk reflector. It appears that K9YC
>>> does not understand it.
>>> _________________________________________________________________
>>> On 6/21/2014 3:23 PM, Dan Hearn wrote:
>>>
>>>   It is interesting to note that the Mosley claimed gain is about 2db
>>>> greater than other tri band beams. They use dbd reference instead of dbi
>>>> while I think their claims would be about right if they used dbi
>>>> reference.
>>>>
>>>>   ------------------------------------------------------------
>>> -------------------------------------------------
>>> Maybe I should have used a few more words to explain. In the fine Yagi
>>> tests described by K7LXC and  N0AX in their book on tri band Yagis. each
>>> companies reference for their gain ratings  is given. Mosley gave dbd as
>>> their reference. That implies that they have about 2db greater gain than
>>> other companies who use dbi. People who do antenna modeling get their gain
>>> answers referred to dbi, an isotropic model. If this is confusing, You can
>>> go to our clubs web page, www.sdxa.org and look under Articals to find a
>>> clear discussion of dbi,dbd, and db gains.
>>> ------------------------------------------------------------
>>> -------------------------------------------------
>>> K9YC comment.
>>>
>>> "Exactly the opposite. The peak gain of a dipole is 2.2 dBi. So a gain
>>> specification of 4 dBd is equivalent to 6.2 dBi."
>>> ------------------------------------------------------------
>>> ----------------------------------------------
>>> This is true but I fail to see what it has to do with my note.
>>> ------------------------------------------------------------
>>> -----------------------------------------------
>>>
>>>   My second comment
>>>>
>>>
>>>
>>>
>>>   I suspect if you used a good line choke at the coax feed point and
>>>> subtract
>>>> 2.1db from their claimed gain you would have a typical triband beam.
>>>>
>>>>   ------------------------------------------------------------
>>> ------------------------------------------------
>>>
>>>
>>> I don't agree with that logic or that statement. A simple common mode
>>> choke
>>> simply reduces common mode current, which mostly prevents RF received on
>>> the feedline from filling in the nulls in the pattern.
>>> ------------------------------------------------------------
>>> -----------------------------------------------
>>> Here I do not believe Jim understands the problem. The choke will do what
>>> he says ON RECEIVING but that is not the problem we are trying to solve
>>> with understanding the lousy test results for the Mosley tri bander. They
>>> connected the coax directly to the driven element as Mosley said. The coax
>>> saw many places on the 3 bands where the feed point impedance was not 52
>>> ohms. That will result in spill over of current from the inside of the
>>> shield to the outside at the attachment point  and ultimately radiation
>>> which screws up the beam pattern and sends rf into your shack. That is why
>>> you need a good line choke at the coax attachment point.
>>>
>>>     I have a lot of respect for Jim and his work on line chokes and other
>>> things. I have built and measured many of his designs with excellent
>>> results. I do not understand why he went on such a rampage over my post.
>>> ------------------------------------------------------------
>>> ----------------------------------------------
>>>
>>>
>>> In his book, "Array of Light," N6BT shows optimized designs of 2-element,
>>> 3-element, and 4-element MONOBAND Yagis. He put these IDEAL gain figures
>>> in
>>> a table with the CLAIMED gains of a dozen or so competing products. In
>>> nearly all cases, the advertised gains were 2-3 dB better than the best
>>> monoband beam. In other words, they claimed impossible gains. After the
>>> first edition of "Array of Light" was published, nearly all of those
>>> manufacturers revised their gain claimed downward by several dB.
>>>
>>> 73, Jim K9YC
>>>
>>>   _______________________________________________
>>
>>
>>
>> _______________________________________________
>> TowerTalk mailing list
>> TowerTalk at contesting.com
>> http://lists.contesting.com/mailman/listinfo/towertalk
>>
>
>
>