Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: [1]   Go Down
  Print  
Author Topic: Franklin collinear array - how to match  (Read 2707 times)
WD4ELG
Member

Posts: 863




Ignore
« on: May 18, 2013, 04:28:41 PM »

ARRL antenna book says about 1K for feedpoint impedance, and unlike collinear arrays of just two half waves that can be matched by a quarter wave of 450 ohm line....Franklin arrays apparently don't have that capacity.

I am considering running 50 ohm coax from the shack, and then using a 9:1 transformer to connect to the ladder line.  The problem is...what kind of SWR will I see if I don't use a tuner?  Should I still try to use a 1/4 wave length of 450 ohm line to transform the impedance...even if it is not down to 50 ohms?  Not worried about the loss in the 450 ohm ladder line, but I am concerned about the loss in the 300 foot of coax...and I am wondering if I need a tuner at the shack?
Logged
WB6BYU
Member

Posts: 13029




Ignore
« Reply #1 on: May 18, 2013, 07:30:50 PM »

Any resonant antenna can be matched with a quarter wave transmission
line, as long as you can come up with one of the proper impedance.

To find the required matching section impedance just multiply together
the two impedances you want to match and take the square root.  In
this case, matching 1000 to 50 ohms would require an impedance of
223 ohms.  You can build such a line (might be easier using 4 wires rather
than 2, however.)

I checked my model of a Franklin in EZNEC and it was closer to 750 ohms,
which is very close to a 200 ohm matching impedance.  Or you can use
300 ohm twinlead and get a pretty good match to 75 ohms.

There are other options as well.  If you start with 1000 ohms and use
a quarter wave of 450 ohm line you come out very close to 200 ohms,
so you could use a 4 : 1 balun at that point with coax to the shack.  Or
from the 200 ohm point you can use a quarter wave of 100 ohm line
(made from two parallel lengths of 50 ohm coax) and transform it down
to 50 ohms.  Using two quarter wave sections of different impedances
gives you a lot more flexibility to choose convenient lines.
Logged
WD4ELG
Member

Posts: 863




Ignore
« Reply #2 on: May 18, 2013, 07:50:54 PM »

Thanks, Dale.  Ah yes, love that magical 1/4 wave transformation magic!

I have the 4:1 balun, so if I assume 1000 ohms feedpoint impedance then I will be all set with using the 1/4 wl of 450 ohm line for the transformer.

If the feedpoint impedance is 750 ohms, then I still come out OK with the 1/4 wl of ladder line and the 4:1 balun.

Do I need to worry about an excessive value for the reactive element when considering this transformation effect, or just look at the total impedance as 1000 ohms and not worry about the reactive versus pure resistive component? 
Logged
WD4ELG
Member

Posts: 863




Ignore
« Reply #3 on: May 18, 2013, 07:55:44 PM »

Dale

This pic shows the tap happening at 1/4 wave point, but the length of the transmission line is longer, a total of 1/2 wave.  And it is shorted.

http://www.qsl.net/va3iul/Antenna/Wire%20Antennas%20for%20Ham%20Radio/31-Colinear%20Franklin%20Antenna.jpg

So I am still a bit confused.  Should I just use 1/4 wavelength and hook direct, or do I need 1/2 shorted and tap at the 1/4 wave point?
Logged
WB6BYU
Member

Posts: 13029




Ignore
« Reply #4 on: May 18, 2013, 09:05:24 PM »

Hmmm...  I wouldn't think of feeding it that way...

If you make the short circuit at the end of the matching stub adjustable so it
can slide along the stub, then the matching method as shown should allow
a match over a wide range of impedances by varying the tap points of both
the shorting bar at the bottom and the attachment point of the feedline.  But
a fixed short circuit at the end of a 1/2 wave length of feedline doesn't seem
to make a lot of sense.  (More likely you would get a match with the total stub
length closer to 1/4 wavelength.)


Actually, my model was the original "Franklin Uniform Aerial", where the stubs
were folded back parallel to the main elements.  The example you gave is
"four half waves in phase", and a model of that suggests an impedance around
1600 ohms instead.  Still, using 1000 ohms as a starting point isn't bad.

This assumes that the antenna is resonant, hence the reactive component is
negligible compared to the resistance.  As you get off frequency there will be
added reactance:  you can compensate to some extent by varying the length
of the matching section slightly.  But as long as you can cut the wires reasonably
close to start with, you'll probably be close enough to reduce the losses in the
coax even if you still need to use a tuner in the shack for final matching.

Another option with similar gain and about the same length is to use 3 half
wavelengths with wider spacing between the ends (achieved by using 1/8 wave
stubs rather than 1/4 wave ones.)  When fed in the middle of the center section
that design is around 160 ohms, and feeding it directly with a 4 : 1 balun should
work pretty well.
Logged
WD4ELG
Member

Posts: 863




Ignore
« Reply #5 on: May 19, 2013, 08:49:20 AM »

Thanks Dale.  I will disregard the picture and go with the 1/4 wave matching section of 450 ohm ladder line as discussed.

I am expecting about 4.5 db gain over a dipole, per the ARRL antenna handbook, in broadside lobes.  I will have the antenna about a half wavelength high which should give me a pretty good pattern and some low angle DX over the poles...as I will have the wires running E/W.
Logged
WD4ELG
Member

Posts: 863




Ignore
« Reply #6 on: May 21, 2013, 07:40:37 PM »

Antenna is up. 108 feet long.  about 3/8 wave high on 17 meters.  Did not have a chance to listen much other than for a few minutes Sunday evening.  I took a 1/4 wave of 450 ohm and hooked it to a 4:1 balun to 50 ohm coax.

In the shack, no need for a tuner as the SWR shows 1.6:1.  So I will give it a listen and compare it to the hex at 45 feet.  The Franklin array does does sound about 10 dBm quieter than the hex.  But that was just one evening.  So I shall see.

Thanks for the help, Dale.
Logged
WB6BYU
Member

Posts: 13029




Ignore
« Reply #7 on: May 21, 2013, 08:03:39 PM »

You could tweak the feedline length to see if it makes any difference, but if
you're happy with the SWR then just use it.

Isn't it great when you put something together and it works first time?  And
that SWR is pretty much what EZNEC predicted when it came up with a
feedpoint impedance of 1600 ohms.
Logged
WD4ELG
Member

Posts: 863




Ignore
« Reply #8 on: May 22, 2013, 08:50:51 AM »

Yes Sir!  VERY COOL!  Would you mind emailing me the NEC file, Dale?

wd4elg@arrl.net

Thanks
Logged
WB6BYU
Member

Posts: 13029




Ignore
« Reply #9 on: May 22, 2013, 10:28:54 AM »

Unfortunately I have EZNEC running on a computer that is not connected
to the internet.  I may be able to transfer the file off using a thumb drive,
but it probably is faster to create your own.  It isn't difficult:  Start with
a full wave dipole (two half waves) fed in the center, then add a quarter
wave wire going down, a single segment horizontal wire a couple inches
long, another quarter wave wire going up, and the final 1/2 radiator on each
end.  Just make sure you use the same number of segments in each of the
vertical wires that make up the stubs.
Logged
SA3BYC
Member

Posts: 7




Ignore
« Reply #10 on: June 29, 2013, 12:25:14 PM »

What a great help finding this thread! I have just started to read about Franklin antennas and was trying to find information regarding feeding of it. I say thank you very much! think you have answered most of my questions previously in this thread.

Any chance that you could give some info on your findings and experiences with this antenna? anything would be very much appreciated Smiley

73 from SA3BYC in Sweden
Logged
WD4ELG
Member

Posts: 863




Ignore
« Reply #11 on: June 29, 2013, 09:34:02 PM »

Very limited experience so far as I have been traveling and working almost full time.  I will be off this week on vacation so I hope to make some observations and share them with everyone.
Logged
WB6BYU
Member

Posts: 13029




Ignore
« Reply #12 on: June 30, 2013, 08:41:15 AM »

While a 4-element array shows about 4dB gain over a dipole, the radiation pattern is
quite narrow, with a half power beamwidth of +/- 12 degrees or so.  (Bidirectional
when constructed for horizontal polarization.)  So it is more useful for specific paths
(when properly aligned) than for general ham contacts.

Though I did at one point develop a method of rotating such a wire array over a limited
range in azimuth by using 3 or 4 supports with pulleys and ropes.   That might work when
the desired coverage area was wider than the beamwidth, but less than about 90 degrees.

A quarter wave of 300 ohm twinlead looks like it would give a better match to 50 ohms
at reasonable heights above ground.  Make sure you correct for the velocity factor, of course.
Logged
WD4ELG
Member

Posts: 863




Ignore
« Reply #13 on: June 30, 2013, 09:37:46 PM »

Thanks again.

So the reason for limited testing is I need to specifically focus on stations N/S compared to the hex.  So far I have not turned on the rig in June, until this weekend.
Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!