W4KAZ

The Project and Situation: After quite a bit of trying over the past 15 years to find the best way to pull dipoles up into the closely packed trees in the yard it is clear the options are limited. Having the dipoles favor the NE/SW directions are the goal, but the arrangement of the best supports make this difficult. To beat this problem a combination of single band and multi band fan dipoles were used. [No, the “chainsaw solution” is not an option – yet.]

The primary supports are now occupied with supporting a 160m inverted L and another with a vee dipole for 80m. These are not high enough for direction to make much difference, but are in convenient locations. So everything else needs to fit around those two primary constraints.

The current problem is that there is really only one support that easily allows stretching out the legs of a 40m dipole in the desired directions while also achieving a good height for 40m(almost 50′). The other high supports will only allow the antenna to be deployed favoring a N/S direction(i.e., legs are stretched out E/W).

Using fan dipoles has come with its own practical problems. The dense tree branch coverage tends to tangle in the multiple wires of the legs. Then the fan legs have become entangled in heavy winds. So it is both a problem deploying the antennas, but also the SWR issues when legs are entangled after bad WX. An ongoing maintenance issue.

Alternate solution: trap dipoles. With dual band trap dipoles, it seems like it may be easier to arrange the antennas in favorable directions AND at good heights. The traps are relatively small compared to the mess of multiple wires on a fan, so also maybe it will be a bit easier to navigate dense branch cover of the biological deciduous antenna support structures. The downside is in the extra effort required in constructing the traps, tuning them to desired frequencies, and tuning the antenna legs for each desired band.

What’s the frequency, Kenneth?!? Using EzNEC 6 I ran models with trap data. Based on those results I initially decided to use traps tuned for just above the top frequencies of any given band(e.g., on 20m tuned for 14.400). I’m willing to live with the trap losses for the advantage of maintenance simplicity. Models showed tuning traps for the top end resulted in wire lengths that are the same as a single band dipole, or slightly longer. I then chose to build antennas with traps above the high end of the band based on the following.

1. A trap resonant frequency above the band results in the dipole wires being the same length or slightly longer than the single band dipole at the trap frequency
2. A trap resonant frequency below the band requires the dipole wires to be shorter than a dipole for that band. This might be worth pursuing if trying to reduce the antenna length.
3. there were already a few spare dipoles laying about, and if the traps project flopped they would still be usable mostly intact if traps designed above band,
4. I’ll probably be mostly using them on CW so maybe a tad less loss with trap rez at opposite end
5. the gut feeling that a 20m trap resonated at 13.900 would maybe have more loss than the trap at 14.4 when used at 14.025.
6. NOTE: SEE Part 2 for notes on how these initial assumptions changed!

Research: The traps will inevitably add unwanted weight to the antennas, and I wished to keep them as lightweight as possible. The reasoning for light weight was to extend the project to portable dipoles deployable on telescoping fiberglass masts. So I ruled out using one of the many coaxial trap designs simply to save weight where possible. For coil forms I chose to use small pieces of 1.5″ plumbing waste pipe cut from small sections of what is sold in the US as a “drain tail piece”. This is thin wall pipe, and much lighter than ordinary schedule 40 PVC. The second form material tried AND ABANDONED is 3.4 inch PVC sched 40.

Excluding the coax trap articles, there are relatively few trap dipole projects written up or documented in places accessible via internet searches. The best[most relevant] source is an ARRL antenna book article on a 2 band trap dipole. W8JI also has some interesting trap info published. Although it does not cover the specifics on the options I chose, it led me to the final result. My choices were made based on materials already on hand(wire, capacitors, and coil form material). Engineer the possible.

Initial Trap Construction: The available values of capacitors also drove the selection of trap resonant frequencies. On this point I made an effort to follow W8JI’s information and make the traps resonant off of the desired operating frequencies to minimize trap losses. Beyond this guideline I could locate nowhere any info to indicate if certain values of inductance vs capacitance were better or worse. A larger inductor will allow the antenna to be shorter overall, but the length of the dipole legs was not a restricting parameter for my project. This was merely about having the dipole resonant on 2 bands. Also, the capacitors are 2KV and 3KV 5% tolerance ceramics from Panasonic that I have used previously in band pass filter projects with great success. (NOTE: MORE ON THIS LATER!!!)

Coil Guidelines????: Guidelines for winding the coils are also a bit of guesswork, beyond W8JI’s testing results that show the highest losses occur on the resonant frequency of the trap. I simply started with the inductors, targeting a value of 6uh, initial turns counts generated by a random calculator found via internet search. Then trial and error on actual coil winding. Calculated inductances are based on trap resonant frequency measurements recorded and on the assumption the 5% caps were the most accurate component. Inductances are then calculated from cap face values and resonant frequency.

Test coils for the traps were close wound with #14 THHN stranded housing wire. They were close wound by hand as tightly as possible onto the forms. Coil Q is probably lower than it could be, but the close winding was a compromise accepted for ease of construction and ease of replication. Four inch lengths of 1.5″ waste pipe and three inch lengths of 3/4 inch PVC were tried. The latter were discarded as unsuitable.

Experimenting With the Coils and Caps: The 5% Panasonic caps on hand typically measure very close to the marked nominal values, much better than any 5% or 10% silver mica caps I have used in similar projects. I found that coils wound with similar technique and the same number of turns would reliably resonate within a range of +/-100 to 200hz. Generally the accuracy and reproducibility is better at 7 and 14 Mhz than at 28Mhz. The coils at the higher frequencies have fewer turns, and smaller differences in inductance and capacitance have a larger effect on resonance.

A group of several capacitor values were used along with an MFJ-259C as a grid dip meter to find the resonant frequencies. Pickup coupling coil was a coax jumper terminated on the business end with gator clips and a short length of #14 wire formed into an adjustable sized loop.

Some initial experimenting with the number of turns on the inductors was based on these available values of fixed capacitors. The first inductor was 12 turns on the 1.5″ forms, then resonance was tested with the values of capacitance that were on hand, or able to be easily derived using series and parallel pairs. It was then relatively simple to find the number of turns needed to be able to produce a trap resonant at a given frequency. I also wound coils on the same form material using 7 and 9 turns, and measured resonance for these.

Experimenting With Trap Dipoles- Part 2