[TowerTalk] K8UR wire 4-SQR characteristics vs 1/4 vertical array design from N0AH

Dinsterdog@aol.com Dinsterdog@aol.com
Tue, 8 Aug 2000 14:12:48 EDT


After having a lot of good results with a 80 meter 1/4 wave vertical 4-SQR 
array, I decided to try a 4-SQR for 40 meters and contacted Jim at Comtek 
System for some ideas.  He sent a diagram showing a K8UR style 1/2 wave 
dipole array so I thought I'd give it a go- 

You can find out more on this design in ON4UN's Low Band DXing Book, third 
edition, page 11-79 (c) for a diagram.  I know Jim is doing a lot right now 
testing out various ideas with the design, so he is a great resource to 
contact as well-

After talking to Jim, I finished a Rohn 25 tower I started 2 1/2 years ago 
(one of those projects)  and topped it off at 65 feet with a 10 foot mast.  
Each end of the dipoles were connected at the top and bottom of the tower 
with the feedpoints pulled out 25 feet from the tower.  

I used 50 foot ropes tied to the W2DU current baluns to pull out the feed 
points and tied them to T-posts.  You really need to use the current baluns 
for this design.  The Hybrid coupler was placed in the center of the tower-

When all the dipoles were in, they formed a diamond shape pattern.  Each 
feedpoint was spaced at 1/4 wave distances from one another, (about 33 feet)  
forming a square pattern.  Diagonal spacing between the feedpoints was around 
48 feet.  I used 26 1/2 foot long feedlines using 75 ohm coax with a 78% 
velocity factor.  They sag a bit between the feedpoints and the tower-  

Per Comtek instructions,  I hung the top of the antennas (positive side (+) 
of the dipole towards the top of the tower) about 5 feet of the tower mast 
while to bottom leges were pulled out only around 18 inches.  This was to 
minimize the antenna(s)  coupling with the tower.  I also used insulated guy 
wires to minimize coupling intereaction.  

It was suggested that I get the top of the antennas as far away as possible 
from the top of the tower, but I compromised a bit on the lower legs, which 
represented the ground side (-) of the dipoles.  While the top wires were 
about 5 feet away form the tower, the lower legs were only around 18 inches 
from the tower in order to keep the dipoles from blowing in the wind.  I used 
parachute cord to tie the ropes from the dipole insulators to the tower legs.

The array's performance, front to back, was a lot like the 80 meter vertical 
array, around 20-25db's.  I was really surprised, almost shocked,  at this 
high level of performance considering I spent a fraction of time and $$$ on 
this array vs the 80 meter vertical design.

Tests on the dipole array,  thus far,  have been limited but promising-  Most 
testing done has been by listening to stateside stuff last night in heavy QRN 
and JA's and VK's in the last couple of mornings.  I've heard enough to see 
how this array might play and it is very responsive.   It is going to take a 
full season to really give this antenna a good test run but initial results 
have been great! 

The most significant test measurement between my 40 meter K8UR dipole array 
and my 80 meter 1/4 vertical array, is the minimum power dissipation into the 
dummy load.  

For those of you not up on 4-SQR's, if you use a Comtek hybrid coupler to 
properly phase the array, using a Collins type feed system, you test the 
efficiency of the antenna system by measuring how much of your power is not 
being radiated.  The less, the better.  This is done by reading the amount of 
power going into a dummy load port off the comtek coupler.  It represents 
power that is not being radiated out of the array due to resonant points. 

Normally, power not radiated out due to resonant points is measured by SWR 
readings.  But the hybrid coupler dumps this unused power into a dummy load 
port for reasons Jim at Comtek can better answer then I can...it's a neat 
part of the science of array design. 

Since the hybrid coupler hides true SWR readings,  you need to measure the 
power dumped into the dummy load to see whats up.

The 1/2 wave dipole array's minimum power dissipation into the dummy load was 
at or close to 1%.  I about fell out of my chair as It hardly moved the 
meter.  The 4-Sqr 1/4 vertical array design's minimum power dissipation was 
at around 3%.  It may not sound like a big difference, but trust me, when you 
get into building arrays, it seems huge on bandwidth performance.  For the 
lower your minimum power dissipation, the more potential you have for greater 
bandwidth.

For example, when my 80 meter vertical array was dumping a minimum of 7% at 
it's most resonant point, I only had 150KHz of operating room on bandwidth 
before the power dissipation readings exceeded 20%.  (When you go past 20%, 
you start to lose F/B performance and your dummy load can really get hot-  so 
20% is a good rule of thumb I've found to follow for measuring effective 
bandwidth of an array). 

However, by improving the minimum power dissipation by using more full sized 
radiators, I went from 7% down to 3% which gave me around 400KHz of band 
width between the 20% power dump edges.  Not bad considering I only 
lengthened the original top loaded verticals by 8 feet.

The moral here is that using full size 1/4 wave and/or full size 1/2 wave 
radiators in an array is really a lot better than using compromised shortened 
antenna designs. But this is only from what I've done here-  I'm sure others 
will have their opinions on the subject as far as what can effect minimum 
power dissipation levels.  Types of feedlines, styles of 4-SQR's, etc...... 
all have an impact-

While the 80 meter 1/4 wave vertical array's resonant point jump 180KHz from 
where the individual radiators were tuned due to the effects of mutual 
coupling, the K8UR array's resonant point stayed exactly where the dipoles 
were tuned.  

Jim at Comtek Systems told he not expect a jump with the dipole array and he 
was right on the money.  W0ETC told me his wire array jumped around 5%.  So I 
thought I'd end up somewhere in between- hihi

On the other hand, my 80 meter vertical array jump was expected to be around 
100KHz, but it jumped 180 KHz. So go figure-  The effects of mutual coupling 
seems to be a case by case issue.  But definitely,  there is less potential 
for a big jump in a K8UR array vs. a 1/4 wave vertical design.

If you have a tall enough tower, I really think you will get your money's 
worth by utilizing a K8UR 1/2 wave dipole design vs. the time, $$$ space 
needed for a 1/4 wave vertical array of some sort. 

Because I was using these arrays for different bands, a sided by side 
comparision was not possible.  But from a performance standpoint, they sure 
seem a lot alike. 

How a 1/4 wave full size vertical 40 meter array with a decent ground system 
would play against a K8UR design, I'm not as persuaded as I used to be that 
there would be a huge difference in performance.  But I'm sure the 1/4 
vertical design would have the edge- but for the $$ and time spent, the K8UR 
approach seems a decent alternative!

And I can't say enough about how nice it is to have these high quality Comtek 
System Hybrid coupler units to use for my arrays.

If you are building and/or considering an array, and you need any assistance, 
please email me at dinsterdog@aol.com and I will try and help where I can.  
While there are experts out there on array's, I still have a tough time 
finding all the information I'd like to on their performance and design. A 
lot of home brewing will go into your array projects-  But they sure are a 
neat way to go!

73  Paul  N0AH
Carpenter, Wy 


--
FAQ on WWW:               http://www.contesting.com/FAQ/towertalk
Submissions:              towertalk@contesting.com
Administrative requests:  towertalk-REQUEST@contesting.com
Problems:                 owner-towertalk@contesting.com