Sunday, May 19, 2013

The 80M elevated wire vertical array

The 80M elevated wire vertical array

Antenna Overview

The new "keeper" 80M antenna is a pair of raised 1/4-wave (68') verticals in the pine woods on our sloping lower property.
Two verticals, each is 68' of 16-gauge aluminum fence wire
(Princess Auto farm section, $30 for 1,300' roll)
Three elevated 66'-long radials for each vertical
Relay control box switches for NE, SW or broadside signal
Uses one SPDT relay to switch for end-fire patterns, and if you want a second SPDT to short the phasing line for broadside bidirectional at 1 dB gain over a single vertical.
Exhibits up to 5 S-units of front-to-back between endfire directions
Audio samples of antenna doing its thing

Parts List
Relay MOD 50-063-0 DPDT 12V 5A
Note: an SPDT relay is all you need... I had a DPDT on hand but only use one side of it.
Relay Omron SPDT 12V 30A
Switch MOD 48-514-1 4 pole 3 pos
Coax Provo 5813a 0.66 vF 50-ohm

Larger view of vertical array control box

Note: The 71-degree phasing line (on the left, partly hidden under the pine needles) is now connected to the E and W feedpoints directly. For broadside firing, a second relay shorts across the two antennas, so both 84-degree feedlines are fed at the same time, without the 71-degree phasing line.

Basic schematic for end-fire in two directions

Note: This diagram does not show a second relay used to create a broadside pattern (simply shorts across the A and B terminals).


Over the past four years, I have sought a good DX and contesting antenna for 80M. The best I found was the venerable delta loop -- about 90 feet on each side, it is a giant. Once, I had two of them in the air, spaced about 50 feet apart and working as a driver and passive reflector. Worked well to the U.S. and Canada, but was large, hard to keep in the air, and the support rope was not in a useful line on the property (it ran above my tower and yagi, which always made me worry about a rope failure).

After studying Google Earth views of our property, I saw how I could raise a pair of 1/4-wave 80M verticals from pine trees. Took a day to get ropes into the right trees at the necessary height, but when I was done the two 68' wire verticals were hanging from 75' high or higher branches. Vertical bases were about 5' off the ground, and about 70' apart -- close enough to 1/4-wave spacing for a good endfire pattern in each direction.

Phasing with 84-degree and 71-degree Coax Lines

In an antenna book (Low Band DXing, I think), I ran across a suggestion for "from the book" Christman phasing that could work for typical two-vertical arrays spaced 1/4-wavelength apart.

While every antenna installation varies, the suggestion I followed said that if you have 1/4-wave verticals on 80M, each with a self-impedance at the feedpoint of about 50 ohms (a reasonable assumption if you have less than ideal radials), the mutual impedance of such an array could lend itself nicely to feeding each antenna with an 84-degree line of 50-ohm coax, and connecting the two feedpoints with a 71-degree piece of coax -- the "phasing" line. (This is a replacement for the typical "90-degree" lines to each antenna, and a 90-degree phasing line).

By shorting the 71-degree phasing line with a relay, the pair of antennas becomes a broadside array.

For a 40M version of this antenna, be sure to read N4JTE's very good (and illustrated) article on eHam "Verticals... Got Two?" about how to wire the relay. Also, see notes on my 40M two-element array.

Measuring and Cutting the Coax: 84- and 71-degree lines

UPDATE Aug. 2010: I have created an easy calculator to figure out the proper lengths, and the frequency for your RF analyzer so you end up with 71 or 84 degrees at your operating freqency.

For those who want to know the inside details, here's how I manually determined the proper lengths:

Being a cheap-o ham, rather than using pricey RG213 for an experiment that might not work, I bought a bunch of the cheapest RG58 I could find (30-cents a foot) with a velocity factor of .66.

For a design frequency of 3.550 Mhz, I cut two 42'8" lengths to feed the antennas (that's 84-degrees at .66 vf) and one 36' length for the 71-degree phasing line.

Note: I got lucky because the lengths I calculated worked out very well for the particular brand of RG58 I used. To check that I was indeed getting 84 degrees and 71 degrees for 3.550 Mhz, I calculated the frequency for which I expected each piece of coax to be 1/4-wavelength (90 degrees), and checked for a null there using a Palomar RX Noise Bridge.

Manufacturer specs for velocity factor aren't always what you get. Coax ain't cheap, so before cutting, borrow an antenna analyzer or RX noise bridge and test a piece of coax to confirm its velocity factor. Then, before cutting your feedlines, check calculations for the intended frequency, think it over twice more, and cut carefully.

To figure out 84 degrees: VA7ST's custom Christman calculator
Figure out 360 degrees (1 wavelength) in feet
(3.550 mhz = 277.06 ft).
I used an online wavelength calculator to get the 1-wave length
Divide this by 360 degrees (to get the length of 1 degree)
277/360 = 0.77 feet per degree
Multiply this by 84 degrees
84 x 0.77 = 64.63 feet for 84 degrees
Multiply this by your coax's velocity factor
64.63 feet x 0.66 vf (RG58) = cut the coax at 42.66 feet (42'8")

To figure out 71 degrees: VA7ST's custom Christman calculator
Multiply 0.77 feet per degree x 71 degrees = 54.67 feet
Multiply this by 0.66 vf = 36.08 feet (the length to cut the coax)

Does It Work?

Why yes, it works very fine business, and thank you for asking. So far, I can happily report that the cheap-o RG58 lines and the 5-amp Mode relay, which I worried wouldn't handle much power, seem to handle 500-700 watts just fine!

The SWR, which the book predicts will be 2.4:1 or so, is actually very low -- under 1.5:1 in either end-fire direction. That low SWR indicates either: less than ideal efficiency (ground losses, due to having just four raised radials for each vertical), or the mutual impedance due to sloping raised radials just works well with this feeding method.

Larger view of vertical feedpoint

The elevated feedpoint for one of my two pine-tree suspended 80M wire verticals.

I have the system working and exhibiting all the right characteristics I'd hoped for, so I'll tweak later with more raised radials and just enjoy playing with the thing for a while first.

The first time I tried out the antenna (April 10, 2008), I heard Peter, ZS1JX, calling CQ on 3.501. I have never heard ZS on 80M before. I turned on the amp and worked him (received a paltry 349 report, but I got a report!).

South Africa is on a heading of 60 degrees from here, quite a long way off the vertical array's prime heading of 120 degrees -- so I was fortunate to hear him or be heard there. Over the next week, every night between 0330Z and 0430Z I heard Peter calling CQ, sometimes peaking at S5!

And guess what? Peter was trying out his own new 80M antenna. A two-element vertical array!

The bands have been dismal since then, with flux in the 68 to 70 range, but I've also worked V63JQ and FG5FR.

Domestic (North American) Performance

U.S. Eastern and Southeastern stations pound into British Columbia even better than they did with the big old delta loop, which was a pretty strong performer for that range. I worried that the array would not do so well for close-in stations (WA, OR, CA, ID, CO, WY, ND), but they have been just as plentiful during the regular Thursday night Northern California Contest Club (NCCC) ladder CW sprints.

One night, a few minutes after the first NCCC ladder sprint using the new verticals, I heard the NCCC Net fellas on 3.610 SSB commenting that they'd never heard VA7ST that loud before. Nothing different here except the new verticals!

I have not found the broadside pattern very useful. The two higher-gain end-fire lobes are so broad that they cover big sections of the horizon to the SE and NW from here. Broadside should point right at Europe and W6, but I have not heard any Europe on the antenna.

Update Nov. 14, 2008.... Yes I have! Heard F5IN (France) and CT3FT (Madeira Is.) tonight just before midnight local (0800z), both about 559 down low in the CW band.

Audio Recordings of 80M Antenna in Action

Conditions are not yet back to normal (may take a few more weeks monthsyears) but I've added a few 80M audio recordings to illustrate the difference in signals when switching between endfire patterns. The difference is at least as prominent as the difference between my 40M single half-squares.

VA2WDQ was calling CQ on 3.510 mhz around 0330Z on April 29, 2008. I recorded four brief samples of his signal, switching the vertical array from East to West and back again. You'll hear the difference when it's aiming his direction.
Audio recording 80M 1 (427kb, MP3)
Audio recording 80M 2 (449kb, MP3)
Audio recording 80M 3 (544kb, MP3)
Audio recording 80M 4 (493kb, MP3)
FT-2000 filter = 200hz

Audio Recordings of 40M vertical array in Action

Hear how the 40M version of this vertical array sounds

Here's ES3AX (Estonia) on 7.010 mhz at around 0140z on Feb. 17, 2010. You'll hear him clearly with the 40M array switched to the Northeast (about 30 degrees) then fade away when I switch the array to aim Southwest.
Audio recording 40M_1 (258kb, MP3)
FT-2000 filter = 300hz

Here's HK1KRY on 7.010 mhz at around 0150z on Feb. 17, 2010. You'll hear him clearly with the 40M array switched to the Southwest then disappear into the noise when I switch the array to aim Northeast toward Europe. There is some fading on the signal, so the rear null comes in and goes out enough to notice.
Audio recording 40M_2 (436kb, MP3)
FT-2000 filter = 300hz

Note: I have constant broadband B.C. Hydro power line noise from a faulty (diagnosed but unrepaired) power distribution substation about 1 mile from home, due East. While the noise isn't nearly as bad on 80M as it is on 20M and 15M, it's still there. That's why this antenna sounds a bit noisier pointing East and quieter pointing West. I heard JAs rather well in the mornings when the band was actually alive.

The 30M Bonus Antenna!

Oh, and I have discovered that this 80M vertical array also tunes to below 1.5:1 on 30M (10.1 Mhz). Since April 10, 2008, I've picked up 20 countries on 30M just casually tuning around. These include: ZF, LU, PJ2, CO, HP, KP4, TI, TG, FG, SM, LY, UA3, 4A3, ES, V31, T32, KH6, 9M6 and, of course, W. Not bad for a bonus band, hi.

No comments: