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: Prev 1 [2]   Go Down
  Print  
Author Topic: Help: Need the Formula for spacing verticals one qtr wave apart  (Read 7887 times)
W3HKK
Member

Posts: 660




Ignore
« Reply #15 on: March 15, 2011, 02:32:02 PM »

JTE:  I have used $10 telescoping 10-25 ft fishing pole verticals with 4 elevated qtr wave radials on 40-30-20-17-15-12 and 10m
at two different qth's fpr tje [ast 8 years.  Results  were always very good, working  most anybody on a particular band.  These were my stealth antennas at the last qth, and my   present antennas here on 3 acres in th ecountry. ( No tower yet and no tall trees.)

I never found the 4 radials wanting, altho I never did any performance testing.  SWRs were usually around 1.5 to 1 or less.

My 40m  vertical  originally had 4 radials 2-3 ft off the ground and worked the world easily, even thru 40 m ssb pile ups at times. to ZS, VK6 KH6 and KL7.

But once I put up the phased array the long haul performance dropped while the domestic performace was outstanding.  The source of  my higher than expected angle of radiation is still uncertain altho several  guys have blamed my radial pattern and lack of more radials.  Apparently  what you can get away with wih a single vert you cant with a phased array of two. 


Once spring temps get to Ohio, I will resume the quest!
Logged
RFRY
Member

Posts: 424


WWW

Ignore
« Reply #16 on: March 15, 2011, 03:23:25 PM »

RFRY, just for your info ON4UN did some precise calculations and found that 84 degree phase feed lines and an 71 degree lag line are much more accurate than the basic approach of 90 degrees .He calculated the exact point on the lines where the voltage are equal, makes it work as designed.

Interesting, but would the accuracy of his calculations exceed those of NEC?  And if so, what optimal directivity did he determine using his approach to analyze this array?
Logged
W3HKK
Member

Posts: 660




Ignore
« Reply #17 on: March 15, 2011, 04:52:36 PM »

71 and 84 degrees are the lengths I used.  However, cutting them accurately  was a bit difficult since  measuring the  qtr wave frequency was via  a very slow dip and rise.  Using  new RG213 as ON4UN reco'd.

But the F/B results  look very good so I must be pretty close.  Just need to lower the radiation angles.  In effect, for DX,  the single vert works better than the phased array the way its currently configured.
Logged
N4JTE
Member

Posts: 1165




Ignore
« Reply #18 on: March 15, 2011, 04:57:21 PM »

RFRY my answer is this, build one with 90 degree phase lines and then build one to ON4UN phase angles, the differance will be obvious. NEC assumes unknown ground and a voltage match at both antennas, usually not my experience in the backyard.
I offer this as someone who has built them both ways along with many variations including 135 degrees, force fed 1/4/ 3/4 and many other variables, for the 90 degree cardoid pattern/ reversable, the Christman method is the best.
Regards,
Bob
Logged
RFRY
Member

Posts: 424


WWW

Ignore
« Reply #19 on: March 15, 2011, 05:19:33 PM »

RFRY my answer is this, build one with 90 degree phase lines and then build one to ON4UN phase angles, the differance will be obvious. NEC assumes unknown ground and a voltage match at both antennas, usually not my experience in the backyard.

My NEC model did not assume an unknown ground, and it was driven with equal currents -- as shown in my link to that NEC calculation.

Would you please post the specific advantage in the directivity of this 2-element array that you and/or ON4UN claim, compared to my NEC analysis?
Logged
N4JTE
Member

Posts: 1165




Ignore
« Reply #20 on: March 15, 2011, 06:00:34 PM »

RFRY, no problem, build one in your backyard and we can converse further.
Bob
Logged
W6RMK
Member

Posts: 680




Ignore
« Reply #21 on: March 16, 2011, 12:05:56 PM »

I think you're getting wrapped around the axle, as it were, because you're confusing the phasing of the EM waves and the lengths of the coax to do the phasing.

ON4UN's calculations are correct for the Christman phasing technique (which takes into account the mutual Z of the two antennas and the phase shift in the coax with non-50ohm termination)... you basically set the two element currents (90 degrees apart, equal magnitude), then calculate what length coax you need on both of them to get the voltages equal at the end of the coaxes. At that point, you use a simple coax T.  I'll note that the Z at that T is probably not 50 ohms.

The real burning question is what is the actual feedpoint Z matrix... NEC will happily calculate it, but my experience has been that for ground mounted verticals, it's never right (and it's not just a matter of adjusting soil parameters).  Is it "significantly different"?  Not really, at least in terms of the pattern "degradation" for a two elements 1/4 wavelength apart sort of thing. 

I suppose the other way to attack the feed problem is to use 1/4 wavelength "current forcing" lines and a lumped 90 degree phase shift network.  ON4UN describes that approach too.


The Christman feed is great for things like a pair of UHF or VHF dipoles (because their self Z and mutual Z is going to closer to the ideal free space vallues).  It might also be useful for a pair of 10m horizontal dipoles spaced 1/4 wave apart.

Here's a page with a link to an Excel spreadsheet and more info
http://home.earthlink.net/~w6rmk/antenna/phased/christman.htm
Logged
WX7G
Member

Posts: 6993




Ignore
« Reply #22 on: March 16, 2011, 12:47:22 PM »

The two phased verticals should provide a gain of about 3 dB or half an S-unit. The elevation pattern should be the same as a single vertical.
Logged
W6RMK
Member

Posts: 680




Ignore
« Reply #23 on: March 16, 2011, 01:14:11 PM »

I find that the fact that two phased antennas has a null (or two) is actually more useful than the small forward gain.  You can null an interfering source (or region).  Sort of the idea behind the cophased whips on each side of a vehicle.. put a null on signals coming from the side of the road.

(Yes, I know that a more practical reason is that when your antennas are essentially at the end of a 40 foot long metal box, it helps provide ANY propagation the rear)
Logged
N4JTE
Member

Posts: 1165




Ignore
« Reply #24 on: March 16, 2011, 09:47:04 PM »

WX7G; I always take your comments to heart, but in the case of 2 correctly phased verticals, I have to disagree with your comment that a single vertical has same pattern. 2 elements correctly phased in 90 degree quadrature, Christman, assuming equal currents/ voltage and phase, the tricky part, will give a very nice low angle take off, cardoid pattern, with a 3db bandwidth of about 120 degrees. Mine is showing about a 20 BD front to back when changing directions, aint gonna get that with a single vertical. perhaps I misread your post as I am going from memory, hi.
Regards,
Bob
Logged
RFRY
Member

Posts: 424


WWW

Ignore
« Reply #25 on: March 17, 2011, 07:52:59 AM »

The two phased verticals should provide a gain of about 3 dB or half an S-unit. The elevation pattern should be the same as a single vertical.

The elevation pattern from an array of two 1/4-wave monopoles spaced 1/4-wave apart in free space wavelengths, and driven with equal currents in phase quadrature produces an elevation pattern that is approximately the same as that of a single 1/4-wave monopole in the direction of maximum gain.  It isn't exactly the same, because the radiating sources are not coincident in space.

But at azimuth angles toward the minima of the azimuth pattern, the elevation pattern varies greatly from that of a single monopole used in that array.

The link below leads to a graphic showing this for the azimuth angle of maximum gain, and at +/- 45 degrees azimuth relative to maximum gain.

http://i62.photobucket.com/albums/h85/rfry-100/40-m_Array_Elev_Pats.gif
« Last Edit: March 17, 2011, 04:28:58 PM by RFRY » Logged
W1ZY
Member

Posts: 7


WWW

Ignore
« Reply #26 on: March 26, 2016, 02:41:43 AM »

SPACING VERICALS 1/4 wl:
300/f = wavelength.
Divide by 4 to arrive at one-quarter wavelength (in meters).
Multiply by 3.28 to convert to feet.
The velocity factor of the coax used for feeding & phasing does not figure into the free-space separation between the two verticals.
___

When calculating 71 and 84 degree lines:
300/f = wavelength (meters)
Divide by 360 to get length (in meters) per degree.
Multiply by 3.28 to convert from meters to feet.
This number equals how many feet per degree.

Multiply this number by 71 and/or 84 to arrive at how many feet are required for the 71 and 84 degree lines.
Then multiply these lengths by the velocity-factor of the coax you are using to arrive at real-world coax lengths.

_____

How to cut phasing lines using antenna analyzer:

Go back to the beginning, to its first step.
300/frequency = wavelength
Rearrange the equation to solve for frequency:
300/wavelength = frequency

So, "300 divided by some length equals some frequency".
You can use this equation to determine what frequencies the 71 and 84 degree coax lengths (before they've been multiplied by the velocity-factor) will appear as 1/4 wavelengths on.
So let's do it.

First you gotta take the FREE SPACE lengths of the 71 and 84 degree lines and multiply them by 4, since the equation above calculates in wavelengths.
Or, you can divide the 300 constant (speed of light) by 4, so that the equation solves for the frequency of a 1/4 wavelength, rather then the frequency of a wavelength.
Take your pick.

In my case, I will multiply the lengths by 4.
Then divide 300 by these "4x"lengths.
This gives the frequencies at which the 71 and 84 degree coax lengths will respectively appear as 1/4 wavelength long.

_______

How to use an antenna analyzer to cut feed and phasing lines.

The calculations you have made thus far are for free space. The velocity factor does not figure in until the last step--e.g. when you are out in the driveway with the wire cutters in one hand, and the coax in the other. Do not incorporate the velocity-factor when calculating the physical FREE SPACE separation of the verticals. Or when calculating their respective lengths. The velocity-factor is only a last-step conversion undertaken when all the calculations have been done. So not incorporate it into any calculation of the verticals, feed or phasing lines, or any "advice" given about "tweaking" the array's spacing. This error occur repeatedly ion many posts about this topic.

A 1/4 wl of coaxial line, when open at one end, will appear to be shorted (low impedance/zero reactance) at the other.
A 1/4wl of coaxial line, when shorted at one end will appear to be open at the other (high impedance/high reactance)


Take the 71 and/or 84 length of coax and leave it open at one end. This means it will appear to have "X=0" at the other end.
Screw the PL-259 you've already soldered at the other end into the antenna analyzer.
Set the analyzer to the "Resonant/impedance" mode, and set it to the 1/4 wl frequency for the line you are cutting.
Snip the line until the the analyzer reads "X=0".
To take into consideration the inaccuracy of the analyzer, sweep the frequency above and below the 1/4wl frequency you've calculated to get an idea of where the meter is reading "x=0" as you are snipping the line. Try to make the final snip when the deviation in the frequency sweep showing "X=0" is equal, e.g. +/- the same amount.












Logged
Pages: Prev 1 [2]   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!