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

donate to eham
   Home   Help Search  
Pages: [1] 2 Next   Go Down
  Print  
Author Topic: First attempt at a phased wire array antenna  (Read 1450 times)
WD4ELG
Member

Posts: 183




Ignore
« on: November 17, 2017, 01:55:03 PM »

After reading VA2GU's QRZ.com webpage, I did some modelling of a 2-element 80 meter phased dipole array at 20 feet in height.  Separated by 64 feet and with a phase delay of 130 degrees (as suggested by VA2GU), it shows max gain broadside (reversible, of course) at 55 degrees of 11.86 dBi.  At 30 degrees, the gain is 6 dBi, and at 15 degrees the gain is 3 dBi.  Compared to my simulation for my inverted L, this phased array is far superior except below 15 degrees, where the two are equal.  On 80 meters, the angles of arrival are mostly above 15 degrees.  (From VA2GU's web page, the phased dipoles at 20 feet will absolutely wreck a single dipole at that height, which is to be expected).

So begins construction:  the dipoles, phasing coax, coax tee's.  I'm a little concerned about ground loss with the antennas being so low to ground.  VA2GU says the antenna operates fine this low, but we shall see.  Earlier eHam article by N4JTE shows quite a different approach here: http://www.eham.net/articles/26865  (higher wire, 84 degree phasing, 71 degree phasing line).  Mine is two 1/2 wave matching coax, one from each dipole, then an extra 130 degree coax line for the leading element.

Interested to hear what others have to say on this subject.
Logged
KD6RF
Member

Posts: 575


WWW

Ignore
« Reply #1 on: November 17, 2017, 04:45:11 PM »

Sounds like a fun project!

A quick sim here with EZNec shows much less gain than your modeling.  I'm seeing about 4.5 dB max gain at 59.6 degs.  Convergence and avg gain all look god, so there is something quite a bit different between our modeling results that you may want to check out.

Another issue that you may need need to address is the phasing amplitude split and phasing system.  The modeling and optimization often is done with equal current into the 2 elements.

However feeding with coax and electrical delay equal to the modeled value will probably not give equal current or correct phasing since the feedpoint impedance is quite a bit different on the 2 elements.....  Of course, if you already took this into account, then just ignore what I said  Wink

GL!
Logged

VTenn Antennas
Bay Area Technical Equip Rental and Test Range
http://vtenn.com/Blog/
K4SAV
Member

Posts: 2405




Ignore
« Reply #2 on: November 17, 2017, 05:02:42 PM »

I assume you haven't modeled this and those numbers you are quoting are from VA2GU's web page.  I suggest you model it before going to all the trouble of building it.  Gain of 4 to 5 dBi at a take off angle of about 60 degrees is more realistic for an antenna at 20 ft.

I don't see where VA2GU stated the antenna height for those simulations, but based on the gain plots it appears they were done at about 70 ft, for the dipole to which the antenna is being compared.

KD6RF seems to be right on target.

Jerry, K4SAV
« Last Edit: November 17, 2017, 05:04:43 PM by K4SAV » Logged
WB6BYU
Member

Posts: 17182




Ignore
« Reply #3 on: November 17, 2017, 05:19:10 PM »

Quote from: WD4ELG

...it shows max gain broadside (reversible, of course) at 55 degrees of 11.86 dBi...




That makes me think you are using MININEC ground for your model, which is notoriously
bad for low dipoles as it doesn't calculate ground losses properly.

You need to use "high accuracy"/Norton-Sommerfield ground to get any sort of
reasonable results.
Logged
WD4ELG
Member

Posts: 183




Ignore
« Reply #4 on: November 17, 2017, 05:34:11 PM »

Thanks for the replies, OM's.  This is how I learn, with your guidance.

WB6BYU - yep, that's what I did.  Once I switched to High Accuracy, I get the same numbers mentioned by KD6RF.  And I see my Inverted L outperforms the phased dipoles by 2-3 dB from 20 degrees down to 5 degrees.

David, I did NOT take into account different feedpoint impedances.  I assumed equal for each element.  Forgive my ignorance, but why would the two dipoles for 80 meters, separated by 1/4 wave, have different feedpoint impedances?  I was reading the N4JTE article, and he uses identical wavelengths of coax for 84 degrees feed...isn't he also assuming equal feedpoint impedance?

Jerry, I was using VA2GU's example for low-height (20 FT) phased dipoles on 80, about 3/4 down his QRZ.com page where the header says "80M LOW WIRE ANTENNAS FOR DXING"


Logged
WD4ELG
Member

Posts: 183




Ignore
« Reply #5 on: November 17, 2017, 05:43:06 PM »

Alright, what if I adjust the plan.  Let's design it for 40 meters.  Let's make the two dipoles separated by 10 meters, at a height of 20 meters and 130 degree phase delay.  THAT shows what I am looking for...about 5 dB more gain at max angle of 30 degrees, with attenuation on the back end compared to the dipole.
Logged
WD4ELG
Member

Posts: 183




Ignore
« Reply #6 on: November 17, 2017, 06:08:52 PM »

KD6RF, was it you or WB6BYU that said "rule of thumb...if you can get the wire a half wave high then go horizontal...anything less than a half wave should be vertical"

What I am trying to do is get more energy at a lower take-off angle with horizontal orientation. 

If I simply follow that rule of thumb, then I am wasting my time on 80 with horizontal unless the wire will be at least a half wave high...even if I use phasing of dipoles.

But here again VA2GU asserts that phased dipoles do work at low height. 

http://forums.qrz.com/index.php?threads/co-phasing-dipoles.169419/   

http://forums.qrz.com/index.php?threads/suggestions-for-low-band-antennas.470596/page-5#post-3443568

http://forums.qrz.com/index.php?threads/80m-dipole-beam.416728/#post-3038593

But Jerry/K4SAV, I think you chimed in on a few of those, right?
Logged
KB4QAA
Member

Posts: 3258




Ignore
« Reply #7 on: November 17, 2017, 07:29:20 PM »

But here again VA2GU asserts that phased dipoles do work at low height.  
WD4ELG:
You are falling into the trap of olden days when the performance of antennas was speculated, debated and argued.

We are decades past that point.  Numerical modeling gives us comparable numbers for analysis and graphs which are readily digested.

There is no need to keep referring to VA2GU or others who "assert" this or that.   Model or end the arguing.  b.
Logged
W9IQ
Member

Posts: 1711




Ignore
« Reply #8 on: November 17, 2017, 07:38:52 PM »

Quote
But here again VA2GU asserts that phased dipoles do work at low height.

Since it is very unlikely that he has an antenna range by which he is able to characterize his antenna, he has either modeled it or has made up the numbers. If the former, get a copy if it or at least an understanding of the model. If latter, then you need to remodel the design to test his assertions.

And then remember that an 80 meter dipole that low to the ground will have >5.8 dBi gain by itself. Of course that is straight up.

- Glenn W9IQ
Logged

- Glenn W9IQ

I never make a mistake. I thought I did once but I was wrong.
K4SAV
Member

Posts: 2405




Ignore
« Reply #9 on: November 17, 2017, 07:52:19 PM »

Vertical versus dipole choice on 80 meters is not clear cut.  It depends on the height you have, what you want to work, and which direction.  Vertical is omnidirectional, dipole is not.  If you go for a vertical with low angle signals you give up a lot of gain at higher angles.  Best choice, of course, is one of everything.

The plot for the vertical shown below was for a ground loss of 8 ohms.  Vertical plot is in black.  Forgot to label it.

Break even points for equal gain, in the dipoles best direction for various dipole heights is

70 ft  14 deg
55 ft  17 deg
30 ft  26 deg
20 ft  32 deg

An inverted L may be a consideration also if you need to go with only one antenna.  It will have some gain at higher elevations and only be down a small amount from a full size vertical.  But it seems you already have one of those.

Jerry, K4SAV

edit:  I see you are in NC.  That could mean not very level ground.  Running your terrain thru HFTA could change your choice of antennas, or al least give a more accurate estimate of what to expect.

« Last Edit: November 17, 2017, 07:59:36 PM by K4SAV » Logged
WB6BYU
Member

Posts: 17182




Ignore
« Reply #10 on: November 17, 2017, 08:03:15 PM »

Quote from: WD4ELG

... "rule of thumb...if you can get the wire a half wave high then go horizontal...anything less than a half wave should be vertical"




I would have qualified that statement to say that the trade-off height depends on your ground conditions.

For any particular take-off angle, there will be some height where vertical is better below it and horizontal
is better above it (though that will, of course, depend on the angle relative to the direction of the dipole, too.)

Sometimes it may be as low as 1/4 wavelength with poor soil.

But that is just to say that, in some cases, the vertical won't be as good as the rule of thumb might assume.
Generally, if you can get a horizontal antenna up 1/2 wavelength, it will probably out-perform a vertical
(unless you are over salt water).  And, generally, if you are under 1/4 wavelength then vertical is likely
to be better for DX.  But that range between 1/4 and 1/2 wavelength isn't as straight-forward as we
might like.


Meanwhile,  I tried a few designs on your 40m problem, and it looks like a pair of 20m wires spaced 10m,
fed with 4 and 8 meters of 50 ohm coax (Vf = 0.67) to the common junction, with the connections reversed
at one feedpoint, gives a pretty good pattern and low SWR at 7.2 MHz.  At least it could be a good starting point.

One problem with such designs is that the coax is operating under a high SWR (the mutual coupling between
the two antennas affects the load impedance of each), and the fact that they aren't perfectly matched means
that the impedances at the common junction get transformed by different amounts.)  I experimented with
different lengths of the short feedline, then varied the longer one for best reverse pattern.  Changing the
element lengths, even when they are kept the same, also changes the required phasing line lengths.
(In some cases I couldn't get 10 dB F/B, but changing one of the lengths by 1 or 2 m allowed me to get
a much better pattern.)  Getting the power to share equally between the two antennas by just using a
coax feed isn't trivial, and the phase shift of a line operated at high SWR isn't the same as when it is
properly matched.  The short 4m line may be too short for 10m spacing, depending on how you plan to
switch it, but it gave a better pattern.  Changing that piece to 5m required around 15m on the long piece
to get a good pattern, but with a higher SWR.  You can't just say that a certain number of feet of coax
will change the phasing a fixed number of degrees when the line isn't matched.
Logged
K4SAV
Member

Posts: 2405




Ignore
« Reply #11 on: November 17, 2017, 08:15:10 PM »

Something you may want to play with.  HFTA is fun to play with and very easy to use.  The only hard part is initially getting the terrain data, but you only have to do that once. 

I live on a small rolling hill.  Look at what that does to my 4 element Yagi on 20 meters, versus the same antenna on flat ground.

This is not an exact science, but it's a lot better than a wild guess.

Jerry, K4SAV

Logged
KM1H
Member

Posts: 2622




Ignore
« Reply #12 on: November 18, 2017, 07:18:36 AM »

An 8JK after W8JK is a better performer at low heights according to many.

I had one up for 75 at about 35-40' aimed at Europe and the South Pacific that did far better than a pair of 1/4 wave phased verticals hanging from tree branches when conditions were favorable.
It was far lower noise than the verticals or an inverted V at 90' yet signals were stronger in the favored directions and elevation angles.

A plus is multiband use with gain.

Then I moved here and had no simple way to support it.

Here two of many articles on the 8JK worth reading.

http://www.bvarc.org/Tech/CompactW8JKBeam-K5LJ.pdf

http://www.robkalmeijer.nl/techniek/electronica/radiotechniek/hambladen/radcom/1990/09/page31c/index.html

Carl
Logged
JS6TMW
Member

Posts: 1192




Ignore
« Reply #13 on: November 18, 2017, 07:20:24 AM »

Leaving aside the elevation and ground characteristics, you have a pretty simple array to model. But you do need to calculate the feed impedances as if fed from a current source, which can be done in EZNEC, but not MMANA. Once you know the feed impedances, I suggest you then use the small standalone calculator Feed2EL to look at your phasing network options.  You will be in the dark and out in the weeds if you just guess at coax lengths.

Have fun,


Steve in Okinawa
Logged
WD4ELG
Member

Posts: 183




Ignore
« Reply #14 on: November 18, 2017, 08:26:05 AM »

Thanks, all.  I will put up the wires and do some measurements on actual impedances, then report back.
Logged
Pages: [1] 2 Next   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!