[TowerTalk] 4 square question Part III

Sat Oct 31 16:51:10 EDT 2020

Bob,

My 2 cents here.  There is a practical answer and an ideal answer to the 
"which direction do my radials point."

My answer is that you space them so that they are not overlapping and 
share the maximum separation with respect to the feed & with each 
other.  I also choked on both ends of the feed my verts to try to 
contain the elements and the radials as the only intentional radiators.  
Opinions vary as to this need but I'm personally a big believer in 
conductive elements participating in the antenna thing unless you do 
something to prevent that in advance.

On the other hand, unless you specifically measure the actual radials 
and use a length somewhat off of 1/4  wl exactly, then the result is 
almost certainly that one of the radials will hog the current and that 
will bend your pattern on that one vertical.

So one answer is that it does not matter, and the other answer is that 
it matters a lot if you want to peel the onion back another layer and go 
for a more optimized installation.  The problem with that is the benefit 
of doing it in a more complicated way is probably lost on most guys 
anyway because the improvements are not determinable in advance.  A 4SQ 
is a magical beast that will work "good enough" even with terrible 
compromises and construction - where an optimized one is not really 
appreciated because most hams don't have a way of doing a before/after 
sort of evaluation.

Good luck!

73/jeff/ac0c
alpha-charlie-zero-charlie
www.ac0c.com

On 10/31/20 3:31 PM, jimlux wrote:
> On 10/31/20 12:11 PM, kq2m at kq2m.com wrote:
>> I want to thank KK9A, NA6O, W4EF, SM2EKM, DCharing (callsign?) and 
>> W6RMK, for all of their guidance, comments and links to helpful 
>> sites.  Greatly appreciated as always!
>>
>> Now for the final question...  let’s say that the “positions” of the 
>> Gauge 12 or 14 THHN wire 4-square elements are N, E, S, and W, 
>> arranged in a box with the sides broadside to NE, SE, SW, and NW and 
>> all the elements will all have either 3 or 4 elevated radials.  I 
>> will be using the Comtek ACB4 (80) phasing box fed with each element 
>> fed with 1/4 wavelengths of RG11 foam coax. (It’s the same phasing 
>> box I have been using for the past 25 years).
>>
>> Does it matter what directions the radials are pointing for each 
>> element? 
>
>
> Not really - what the radial(s) do is provide a return path for the 
> current that "intercepts" the ground - and reduce "ground loss" 
> (independent of the "counterpoise" function).
>
> The elements themselves are assumed omni directional. And it's only 
> some some sort of extreme cases (1 radial) that there's a more than 
> trivial effect on the directionality of the radiation pattern of a 
> single element.  What changes is the efficiency (i.e. the loss 
> resistance gets lower as the number of radials increases).
>
> In the usual 4SQ "current forcing" array with 3/4 wavelength feed 
> lines, the current in the element (which is what radiates) is the 
> right ampliude and phase to form the beam.
>
>
> So let’s say I am using the NE position on the ACB-4 controller, since 
> I am radiating on all 4 elements, does it matter what directions the 
> radials are pointing?  Do I want to have the same radial directions 
> with all 4 elements or do I want to spread out the radial directions 
> differently on each element?  Since NE is by far the most important 
> direction for me (Europe), do I want to have more radials pointing NE 
> or broadside NE or equally spread out so that the radiation and 
> current return is more uniform?   I am not clear on any of this or if 
> it is even important so I would appreciate any guidance that you can 
> provide!
>>
>> My understanding is that I want to get the radials at least .05 
>> wavelength above ground so at least 13’ above ground.  Is that correct?
>>
>> Tnx & 73
>>
>>
>> Bob, KQ2M
>>
>>
>>
>> -----Original Message-----
>> From: jimlux
>> Sent: Friday, October 23, 2020 8:44 AM
>> To: towertalk at contesting.com
>> Subject: Re: [TowerTalk] 4 square question
>>
>> On 10/22/20 9:45 PM, Michael Tope wrote:
>>> Bob,
>>>
>>> I found this by googling "4  square 8 directions":
>>>
>>> https://ncjweb.com/bonus-content/k3lc4squarea.pdf
>>>
>>> I think W8JI describes something similar on his website.
>>>
>>> 73, Mike W4EF....................
>>>
>>>
>>
>> and that 4 square handbook from the DF6QV.. He's got all possible boxes
>> and phasing
>> http://tm1o.free.fr/4SQ/80m/en_ver_final4-sq_03_04_15.pdf
>>
>>
>>
>>>
>>> On 10/22/2020 7:39 PM, kq2m at kq2m.com wrote:
>>>> I really want to thank Jim, Jeff, Les, Mike, John and Gary for all of
>>>> their help, answers and ideas to which I have been giving a lot of
>>>> thought this week. The information gleaned has made me confident of
>>>> the new design.  Unfortunately there is no way to avoid putting the
>>>> 4-square elements around the tower so I know that interaction effects
>>>> will likely be noticed.  Vicious storms have taken down all the other
>>>> tall trees that were previously in use to support my 80 meter 
>>>> 4-square.
>>>>
>>>> Note to Jim Lux – I am particularly intrigued by your idea of changing
>>>> the feeding of the elements so that I can radiate broadside from the
>>>> side of the array rather than on the diagonal since the only trees
>>>> that I have to support elements happen to be roughly N, E, S and West,
>>>> which, with the Comtek phasing box, would give me radiation N, E, S
>>>> and West rather than the desired NE, SE, SW and NW.
>>>>
>>>> The Comtek ACB-4 outdoor box uses the NW element in Pos 1 on the
>>>> outdoor , NE (2), SE (3) and SW in position 4 and arranged as shown in
>>>> the .pdf:
>>>>
>>>> www.static.dxengineering.com/pdf/comtek_4_square_info.pdf
>>>>
>>>> Which elements (N, E, S, W) would you put in which positions on the
>>>> box (so that the 4-square radiates toward the NE instead of East?
>>>>
>>>> How is the gain and F/B and F/S compared to the standard pattern that
>>>> radiates diagonally?
>>>>
>>>>
>>>> Tnx to all & 73!
>>>>
>>>> Bob KQ2MTowerTalk mailing list
>> TowerTalk at contesting.com
>> http://lists.contesting.com/mailman/listinfo/towertalk
>> _______________________________________________
>>
>>
>>
>> _______________________________________________
>> TowerTalk mailing list
>> TowerTalk at contesting.com
>> http://lists.contesting.com/mailman/listinfo/towertalk
>>
>
> _______________________________________________
>
>
>
> _______________________________________________
> TowerTalk mailing list
> TowerTalk at contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk