Tuesday, October 20, 2015

Boxing the 80 Meter Vertical L Network

The L network at the feed point of my new 80 meter vertical is now complete. The rat's nest of components and wires lying at the base of the tower have are gone, replaced by a box of components with (mostly) fixed components. You can look back at that earlier article for details on the L network design and tuning process.


The photo at right shows how I put the network together and how it will be attached to the antenna. The box is an ordinary 4"x4"x2" plastic junction box for use outdoors. These are inexpensive items founds in every hardware and electrical supply store. Notice the seal on the cover.

The SO-239 is attached with 4-40 screws and nuts, with a ⅝" hole for the main body. For the latter I used a wood bit at a slow drill speed to avoid melting or burning the plastic. The centre pin supports one end of the series inductor.

Aluminum flashing was hand cut, holes punched for terminals and deburred with sandpaper. Not only do these serve to mechanically stabilize the parts, they are quicker to make than wires with lugs. The only wire lugs are for the coil tap and the external connection to the radials.

The shunt capacitor is overkill in this application, even for a kilowatt. I chose this 500 pf doorknob because it measured close to the 440 pf value measured for the variable capacitor after tuning the network. Also, it has been lying neglected in my junk box for many years.

The long stainless steel screw at the top secures the network to the tower leg, while also providing the electrical connection to the tower as the vertical monopole. The shorter stainless steel screw at the bottom is for the wire connection to the radials. Nuts secure both screws to the box walls. One more nut per screw is used for their external attachments. Washers protect the tower leg and box walls from being abraded when the nuts are tightened.

Tuning woes

To my disappointment the L network did not offer the same match at the test setup. I underestimated the difference between the rat's nest wiring and the finished product. All those dangling wire affected the match. The precision needed in component values is tight enough that modest differences can have a large effect.

Ultimately I needed to increase the shunt capacitance by 250 pf to about 700 pf total, with a final coil tap to give 2.6 μH series inductance. This is quite close to the tuner design I got from TLW and verified with EZNEC. My rationalization in the previous article was off base. The design was fine as it was. Classify this among the cases where theory trumps practice!

While I have smaller doorknobs to make up the additional capacitance they are a pain to work since they have screw terminals. Instead I used 3 tubular 82 pf capacitors in parallel (not pictured). These are inadequate for a kilowatt though good enough to handle 100 watts.

The result is a vertical with an SWR of 1.1 at 3.5 MHz and rising to 1.9 at 3.8 MHz. To better centre resonance I would have had to substitute capacitors until I had the perfect combination. Since I primarily operate CW and it is adequate for a few SSB contest contacts I decided I had spent enough time on this antenna. I boxed it up and moved on.

Perhaps the lesson here is to always allow a means for tuning in a matching network. Even if only to tweak it. Going to the other extreme of making every network fully adjustable is a waste of money (high-power variable capacitors and coils) and time.


The final photo show the box on the tower and ready for use. The only things not done at that point were soldering the wire to the radial nexus and sealing the boot on the coax connector.

The installation is quite clean looking and is almost ready to winter. However I did encounter one significant problem that had to be remedied before I could turn to other projects.

The box was on the tower for 2 days and nights while I dealt with the tuning problem and tested it on the air, including the Worked All Germany contest. The cover was in place but not screwed down. It rained. After the contest on a frosty Sunday afternoon I lifted the cover and found a small pool of water in the box, and a coil partially encased in ice! More waterproofing was necessary.

I put the box in the sink to melt and drain the ice then returned to my workshop. Coax-Seal was used to line the outside of the screws and coax connector to discourage water leaking in. The cover, when screwed down, is watertight by design. Yet I know that water is pernicious, not to mention that by late winter the snow line might reach the box. I drilled holes in the 4 bottom corner of the box to drain any moisture that does get in either by leakage or condensation.

Performance update

As mentioned above, I used the antenna in the Worked All Germany contest. I made 2 contacts on 80 CW (both Germans, per contest rules). While that seems pathetically low it is not bad for a QRP (5 watt) effort under conditions that were not good. I made a few other DX contacts (Europe) outside the contest though not enough yet to get a strong sense of its performance. Certainly it's no worse than I did with the loaded half sloper. Neighbourhood QRN is equally as loud on receive, unfortunately.

Perhaps I'll know better after CQ WW SSB this weekend. I don't expect to make any DX contacts on 80 SSB, but you never know. But right now I need to put up the supplementary 40 meter inverted vee before the weekend.

No comments:

Post a Comment

All comments are moderated, and should appear within one day of submission.