Thursday, February 12, 2009

Remote 3-position Antenna Switch

Part of the next phase of my station upgrade involves putting up some permanent antennas for 160, 80, and 40 meters. After a fair amount of research and staring at my QTH from across the street, I came to the conclusion that, at least for 80 and 160, my best bet is to put an inverted-L into my Ginkgo tree, and that I might as well keep the 40 meter vertical I currently have there.

When I went to model this using EZNEC, I tried to build a model with two inverted-Ls, one cut for 80, and one for 160, fed from a single feedline. Try as I might, I couldn't find a combination of lengths that resulted in resonance on both bands at once. So I decided against trying to feed all the antennas at once, and started planning a remote antenna switch.

Now, about a year ago, I did a bang-up job of making a trench across the yard, so I could run the feedline underground to the tree (otherwise the coax would need to lay on the ground and pose a tripping hazard). Unfortunately, I only ran a single feedline, and I am not digging that thing up!

So I started looking for a remote antenna switch where the power was fed to the switch via the coax. The hard part is finding one that can handle three antennas. With a DC circuit, you've got one "bit" of information - power on or off - to select an antenna, and that's only two. Every circuit I could imagine was fairly complicated - using voltage levels to select antennas, or some sort of pulse-counting system.

Finally, I happened upon a design in the ARRL Antenna Book that uses 12 VAC at the shack side. Imagine an AC-DC rectifier, with half in the shack, and half out at the antenna. The original design has two independent relays. One actuates when the positive-going AC is put on the wire, and the other actuates when the negative-going half is put on the wire. When a complete sine wave appears, both relays actuate. The design in the ARRL book is a little skimpy on details (e.g. the value of the RFC chokes aren't specified) but I found an article in the QST Archive that had a similar circuit for delivering DC voltage an a coaxial cable, and it gave recommended values (as well as instructions on how to modify a widely available Radio Shack choke to have the correct value).

I reworked the design a little so that with an SPDT switch with a center-off position, I can actuate no relays, one relay, or both relays, and then wired the relay contacts together so that I have:
  • Power off (center) = both relays off, antenna 1 selected
  • Switch in up position = relay 1 engaged, antenna 2 selected
  • Switch in down position = relay 1 and 2 engaged, antenna 3 selected.

I built the relay box into a waterproof plastic electrical conduit box, and sealed all the SO-239 connector holes with coax-seal. This thing should last a long time, I hope.

Now I'll be able to switch between 160, 80, and 40 for this weekend's WPX RTTY contest without going out in the rain!

Sunday, February 1, 2009

A Different Kind of Antenna Modeling

As I've been planning the improvements to my station, one thing I've wanted to do is to make sure that my planned antennas are something my family can live with, visually. To that end, I wanted to be able to visualize how the antennas might look. Here are some of the results:

Front view of house TGM MQ-34SR and G5RV

This is my planned antenna system, barring any major snafus. It's a TGM Communications MQ-34SR mini-beam for 20/15/10 meters, with a G5RV mounted just below the rotator.

I selected the mini-beam because it's about all I can fit at the rear of the house, where I've only got 11 feet of space to the fenceline. A tower on the side of the house would allow for something bigger, but we may be remodeling or moving, so there's not much point in building something permanent. At best, I would need to abandon it if we move, and at worst I might need to take it down, if we remodel.

Front View: Cushcraft R8 + G5RV

Here's plan B - a Cushcraft R8 vertical with a G5RV mounted at the base:

The R8 seemed to be the best choice for a 40-10m, no-radials installation, based on my reading N0AX and K7LXC's HF Vertical Performance - Test Methods and Results publication. If you're thinking about a vertical, I highly recommend ordering this - it was $17 well-spent.

How'd I do it?

First, I should say that I have very little artistic talent, and anyone who's at all competent with Photoshop probably will roll their eyes at my work. However, it did the trick - it let me visualize how the antennas will look.

Step 1: Get a good photo of the house. Make sure that there's enough sky in the shot to allow your planned antenna to fit. Ideally, take the photo on a clear day.

Step 2; Get some size references. Rather than trying to do all the math to calculate how big, say, a 28-foot high vertical should appear in the photo, I just temporarily set up a 30-foot high mast using the fiberglass stacking poles I mentioned in a previous post. I photographed the house with the mast in place, and used that as a reference.

Step 3: Find a photo of the antenna you're thinking of erecting. A Google search yielded photos of both antennas I was thinking of.

Step 4: Using photo editing software, import the antenna photo and erase all the background around the antenna. I use The Gimp, and am able to zoom and edit out the background.

Step 5: Create a new file, and set up two layers. One layer will have the house photo, the other will have the antenna.

Step 6: Resize the layer containing the antenna so that the apparent size of the antenna is correct. In the case of the R8 photo above, I made the antenna, which is 28 feet high, just slightly less tall than the 30 foot mast, then positioned the base of the vertical at the top of the mast.

Step 7: For a more geometrically complicated antenna like a beam, it's pretty likely that the perspective of the photo you found won't be right for the photo of your house. I used the Gimp's "Skew Perspective" feature to "eyeball" things so the perspective looked right. Again, I'm sure that if I dusted off my old computer graphics book from college, I could have gotten the skew exactly right, but as it is, it does the job.

Also, for sizing photos of beams, I found that it was easier to draw a line of known length, then skew it to the same angle as an element of the beam for which you know the length. For example, the TGM mini-beam has an 10-foot boom. I drew a line that was 1/3 the length of my reference 30-foot mast, then skewed that until it was parallel to the boom of the image of the antenna. I then resized the antenna image until its boom was the same length as the line, and finally deleted the line.
Another idea is to search around the web for real-life photos of the antenna you're thinking of, especially when mounted on a house of similar size to yours. That will help you determine if you've got the perspective and sizing about right.

Finally, here's a little bit of dreaming - A Force-12 C-3SS triband beam on a crank-up tubular tower.