A linear-loaded 40M rotatable dipole

Written by admin On August - 27 - 2006 0 Comment

 (Update 2007, 2016)

In the SARTG RTTY contest in mid-August 2006, my 40M performance was terrible — just over 20 contacts. That was using a two-element wire yagi (inverted-V elements) up nearly 70′.

Disgruntled, for the following weekend’s SCC RTTY contest I vowed to improve my 40M antennas. I had a week of vacation with most of the week free to play with antenna stuff.

By Tuesday, I decided on building a rotary dipole to put at the top of the tower. When fully cranked up, that would be about 55′ above ground level. Clearly this is not optimum — 70′ is the “industry standard” half-wavelength height for good 40M operation. But a lot of reading on the web told me a rotatable dipole even 50′ up will be as good as, or better than, an inverted-V at the same height. And it can be pointed to null out noise or peak a signal (a bit, at least).

Fully aware of the constraints I was accepting, I was ready to try something different anyway.

Choosing a Design
The first step was to review all the designs out there. I looked at the Cushcraft D40 dipole  that Don Hill, AA5AU, installed following the terrible damage by Hurricane Katrina last September (2005). His was at 70′, and he reported great results with it. Mine would be at 55′ and quite a bit shorter, but I hoped I’d still get good performance.

Knowing the kind of antenna I wanted, and not being very inventive (or at all handy for that matter), I needed a specific model to follow. I looked at two kinds of shortened dipoles:

Materials: Two Poles and Some EMT Conduit
I examined my on-hand materials: I had two good 16.5′ fiberglass “crappie” fishing poles that I had purchased for $9.99 each from Cabelas.com in 2002 to make a Spider Beam yagi. In 2011, they’re still available for $9.99. Just look for part number CCt-165!

I had removed the whippy end sections, and added a couple of feet by fabricating a pair of three-foot long 1.5″ PVC plumbing pipes reinforced with dowel (or broom handles), and inserted into the 1.5″ end of each pole — they fit snuggly. This gave me 17.5 feet for each side, or 35′ tip to tip.

To connect the two fiberglass poles, I found a piece of 1.5″ EMT thin-wall steel conduit that would snuggly accommodate the butt end of each element’s PVC extension. I cut this to about 4.5′ in length.

I used a pair of bolts on each end to secure the connection of the EMT tubing and the inserted pole ends. It’s plenty strong enough for this purpose.

Mounting Plate on the Cheap

The “chopping board” mounting plate, with U-bolts for the mast and around the EMT tubing.

For a mounting plate on the tower’s mast, I used two white chopping boards, each about half an inch thick of tough acrylic. Got them for $1 each at the Dollar Store — quite a bargain.

These chopping boards were bolted together for a 1″ thick mounting plate. Half-inch holes were drilled for the 1.5″ U-bolts — two around the mast, and two around the EMT tube of the antenna.

Total length of the antenna was now about 38′ tip-to-tip. Pretty short for a 40M dipole. But I was going to load the antenna with mid-element coils, per GM4JMU’s design. I spent a morning making the coils according to the design (40 turns on a 1.5″ PVC pipe 6″ long), and by evening had the whole thing ready to test in its stand in the back yard.

False Start With Mid-Element Coils

Using my Palomar RX Noise Bridge ($5 used from a local ham — and once I figured out how to use it, it’s fabulous), I determined that the antenna was too short.

I lengthened the elements with alligator clips so wire hung over the ends. After an evening and a morning of experimenting in the test stand, it was resonant at 7.040 Mhz with 2′ more wire on each end — or total length of 43′ or so.

I didn’t want a dipole up on the tower with extra wire flapping off the ends. I know I could make this work with shorter elements by changing the coil diameter or adding more windings, but this meant a lot of work and it was a crap shoot (purely guesswork).

I have no grid dipper or analyzer to give me an idea of where the coil I’m building is going to resonate. I wanted something even simpler — it was already Wednesday and my week was running out.

(Update: Nov. 2008 — I plan to try the GM4JMU mid-element loaded design again. From what I’ve read, the inductance loading can provide more bandwidth than linear loading, which is useful for contesting when running CW and SSB.)

Changing Course: Linear Loading with Twinlead

Schematic of the dipole. Linear loading twinlead nearly as long as the radiating elements.
Maybe I am breaking laws of physics here.

Then I remembered the K4VX design I’d read on the web, using 450-ohm ladder line as a linear loading method. I had miles of very good (big wire) 300-ohm twinlead picked up cheap when Radio Shack dropped “Radio” from their inventory and, in my opinion, became just “Shack” — and then in Canada dropped the whole name to become The Source by Circuit City. But I digress (I am bitter — can’t even get a chunk of RG8X locally any more).

I decided to give this 300-ohm twinlead loading a try. The original design called for 22’6″ elements and 11’10” linear loading sections of ladderline. I figured seeing as my wire elements were going to be just 19′ long, I better start with about 16.5′ of 300-ohm twinlead.

After the coil-loaded experiment, which took more than a day to get together and adjust, the entire W4VX system took less than half an hour to build. It’s SO EASY!

Finding the Sweet Spot

SWR plots for the Crappie 40M Rotary Dipole.

I was very close in my guesswork. The RX Noise Bridge measured the antenna’s resonance at 7.460 Mhz. Too short.

So I added 16″ first, then up to 24″ of twinlead to each end of the loading sections. (Total twinlead length on each side was now 18.5′).

Bingo. Resonance smack-dab in the CW/DX RTTY band — 7.035 Mhz, which was my design goal. (Blue line) SWR 1.1:1 and below 2:1 all the way up to 7.120 Mhz. Quite wide-banded considering its size!

Final Details: Making It Weatherproof

Next step: making the thing robust. I taped the wire elements (#18 insulated copper) to the 300-ohm twinlead, then taped and cable-tied them to the fiberglass poles. Resonance went way off due to closer coupling of the wire with the linear loading twinlead, so I re-adjusted the 300-ohn twinlead lengths (shortened them by a couple inches) and resonance was back at 7.035 Mhz.

Element wire taped to TV twinlead.

It took some doing to make the thin wire connections with the SO239 connector nice and strong. Not sure it is going to hold up in the weather, but I soldered everything, tape it up, and tied it in as best I could.

The antenna bolted onto the tilt-over tower’s mast very easily. I left it about 1.5′ down from the very tip, and added fishing downrigger stainless steel guys from the pole to just past the midpoint on each element. (I worried this would throw the antenna out of resonance, but it had no impact). It keeps the fiberglass from supporting its own weight — eliminating the 14″ of tip sagging that became apparent in the test stand.

Detuning With Tower Height
When the tower was fully extended, (red line) the SWR flattened out a bit, and climbed to 1.4:1 at 7.035 Mhz. I can live with that. (The resonant midpoint moved up 20 Khz to 7.055). With the tower nested down, dipole at only 35′, the SWR is one or two tenths lower across the board.

Next time the tower’s tilted over (next summer I hope), I will retune the loading sections by taking out a couple of inches to bring the “in the air” resonance point down 20 Khz.

Tried the antenna in the SCC RTTY contest August 26, 2006, and had a hard time working anyone with any antenna, so the verdict isn’t in yet. It’s a lot quieter than the vertical. Made about 30 out of 40 contacts with it that night (the remainder using a ground-mounted vertical).

Will post an update once I’ve had a chance to try it out when conditions are better than terrible.

UPDATE: February 2007

Finally ordered some new 100′ coax cables, freeing up some good RG8U from other antennas in the yard. Replaced the run of RG58 from the shack to the base of the tower, where it barrel-connects to the RG58 going up to the Crappie dipole. The old cable was the problem and now this antenna really works well, considering its short, very linearly loaded elements and the low height of 55′.

Even with the very thin-guage copper wire, the antenna handled the amplifier’s 1000W in ARRL DX CW and CQWW WPX RTTY without a problem. I didn’t stay at 1KW for long, as I didn’t need that much power to run stations when I wanted. Typical power was closer to 500W.

Kept the dipole aimed at 100 degrees (to most of the U.S.) in NAQP RTTY to work about 85 of the 102 contacts on 40M. Rest were on the ground-mounted vertical.

Also worked lots of European stations on the second night of ARRL DX CW, after the cable was replaced. Not as many as the previous year using the delta loop array, but still darn good for a simple, low 40M dipole.

Here’s what I worked from 4:56 p.m. (Pacific) to 5:53 p.m. on the second night of ARRL DX CW on 40M with the dipole:

Station Country
ZF2AM Cayman Is.
YV5OHW Venezuela
CT9L Madiera Is.
IK4ZGO Italy
HA7RY Hungary
OM8A Slovakia
S53F Slovenia
OM7M Slovakia
ES5RR Estonia
PJ5NA Sint Maartin
ZS6AAA South Africa
UU7J Ukraine
P49Y Aruba
IR4X Italy
UA6LV European Russia
HB9FAP Switzerland
EA8/OH6CS Canary Is.
IR2W Italy
CU2JT Azores
AY7X Argentina
UA9UZZ Asiatic Russia
RX1AA European Russia

From British Columbia, this is a pretty good array of 40M DX using a simple shortened dipole. South Africa was an especially great surprise.

Verdict: Highly recommended, if you can’t put up elements closer to full size.

Antennas, DXing, Uncategorized

Enjoy this Post? Share or cast your vote!

E-mail RSS Del.icio.us Digg Reddit Twitter

You must be logged in to post a comment.