Recently, someone whose blog I follow mentioned that instead of waiting until a project is finished, he makes several mini-blogs about it, along the lines of “New Blog Post – Soldered All Resistors in New Radio!”, New Blog Post “Now I’ve Soldered All The Capacitors!” etc etc. My approach has been to wait until a project is finished and do one big write-up on it, but now I’m having second thoughts. I spent this afternoon putting up the first section of my new trap dipole and I’d like to blog about it so that I can get this part of the project “out of the way”, so to speak. Besides, I’m going to England next week and if I wait until the trap dipole is completely installed and working before I blog about it, it could be several weeks.
The 30M dipole that I put up a week or so ago works great, but I’d like to be QRV on more than one HF band. The house I live in is shared with 9 other tenants and while I have a certain amount of freedom to put up antennas, I know that if I filled the sky above the house with wire, someone would object. A dipole fed with balanced line was considered, but the feeder would probably be too intrusive. I could tune a longwire against ground, but I preferred the idea of a balanced antenna for (hopefully) lower noise. It would also save me having to bang in ground rods and lay radials – another move that my fellow tenants might object to. The coax fed 30m dipole that I just installed has not drawn any complaints, which got me to thinking that a trap dipole would also be fairly low-key. It only has the one coax feeder, and the only thing that would make it stand out more than the current dipole would be the traps, which would be lightweight and fairly small anyway.
Having made lots of dipoles, but never a trapped antenna, I started by doing some reading on the subject (isn’t the internet wonderful?) Coax traps wound on thin plastic formers would be cheap and lightweight. I did read that coax traps tend to be lossy near the resonant frequency so decided that for this reason, instead of putting the resonant frequency of the traps close to the desired operating frequencies, I would make them resonant below the bottom edge of each band in question. There’s no point in operating on a frequency at which your trap is lossy.
Another way to minimise the loss of an antenna using coax traps is to use the minimum length of coax needed to achieve resonance. I let someone else do the math for me. R.C. Summer N4UU in his article “Optimizing Coaxial Cable Traps” in the December 1984 issue of QST, says that the minimum length of coax will be used when the length of the coil divided by the diameter of the former is about 0.45. Using VE6YP’s Coax Trap Program and some Nalgene bottles I picked up from the local outdoors/recreational store, I came up with the following for traps made with Belden 8259:
2″ diameter former – 5 turns of co-ax – resonant frequency 13.5MHz (length/diameter = 0.485)
2.4″ diameter former – 5.8 turns of co-ax – resonant frequency 9MHz (length/diameter = 0.475)
A lot of pill botttles are made of polyethylene, which has excellent RF properties (polyethylene is used as the dielectric in a lot of coax cables). If you look at the bottom of the bottle, it will usually tell you what it’s made of. HDPE means high-density polyethylene, which is good.
Here’s the bottle for the 20m-30m trap, which I want to be resonant on about 13.5MHz. I’ve already marked a line around the top with a Sharpie:
After cutting the top of the bottle off with a craft knife, I steady one finger against the end of the bottle to help in drawing a fairly straight line around the other edge:
A careful slice with the craft knife and – presto! – the former for our first antenna trap:
The coil has been wound and the center conductor of one end of the coax has been soldered to the braid of the other end, leaving one braid and one center conductor for connections to the antenna elements. Notice the 3 small holes drilled in the end of the former to act as a strain relief for the antenna wire (there are 3 similar holes in the other end too). I am not going to use nuts and bolts for the connections to the antenna as it adds extra weight – the antenna wires will be soldered directly to the traps:
Antenna wires soldered on and the whole thing covered with a good dose of Brush-On Electrical Tape. This is the finished trap, which resonated at about 13.2MHz:
For the center insulator, I decided to get rid of the SO-239 and PL-259. The traps will already add extra weight for the thin fiberglass mast to support – no point in carrying unnecessary weight. The braid and center conductor of the Mini RG-8X were tinned and my new favorite goop, Brush-On Electrical Tape applied to seal the end of the coax:
The center insulator was cut from the lid of an old storage container. It’s made of LDPE – low density polyethylene, and the whole affair looks strangely like some ancient fertility symbol:
This is stating the obvious to anyone who has built antennas before, but notice the loop of coax at the top, oriented so that any moisture will tend to run away from the end of the coax.
If the end of the coax were pointing upwards, moisture would have a tendency to sit on the end of the coax and would be more likely to creep inside. If this is your first antenna and you haven’t thought of these things before, notice how TV cables and electrical cables are looped or otherwise situated at the point of entry into a building, and then think of where moisture would collect.
I made the first two traps, which both resonated at about 13.2MHz, and fashioned the 20M part of the dipole thus:`
The idea with a trap dipole is to work from the center outwards, pruning dipole legs as you go, so my first objective was to get the 20M part working. I initially cut the 20M dipole legs to 17 feet each and found (with the MFJ-259B SWR analyzer) that the dipole resonant frequency was approximately 13.2MHz. At this resonant frequency, the SWR was 1.4:1 – not quite as low as I would like it. After pruning equal amounts off each leg, I got the resonant frequency to 13.975MHz with an SWR of 1.2:1 and a 1.5:1 bandwidth of 400KHz.
Next time I’ll trim about a little bit more off each dipole leg to get the center frequency close to 14.060. With a 1.5:1 bandwidth of 400KHz, I could place the center frequency at 14.175 and cover the whole band at less than 1.5:1, but I don’t care much about the phone portion and would rather cover the bottom 100KHz with as low an SWR as possible. Then I’ll add the 30 meter portion of the dipole and we’ll see how well these traps work!
UPDATE: – Next morning, I called CQ on 14060 and had a great QSO with WG0AT, who had been practicing the ukelele and said that his fingers were getting sore, so he jumped for the key to answer my CQ and give his fingers a break. What a thrill to have my first QSO with this antenna be with Steve, such a great QRP’er! Then I added the 30m section and noticed something that makes perfect sense, but having never built a trap dipole, wasn’t expecting. The addition of the 30 meter sections brought the resonant frequency of the 20m section down from 13.975 to around 13.55. Either the traps weren’t completely cutting off the outer sections, or there was some electrical interaction between the 2 sets of dipole sections, seeing as there was only an inch or so between the ends of the 20m sections and the beginning of the 30m sections. This is where trap dipole construction probably starts to involve a lot of pruning. Also, the SWR at the new resonant frequency was 1.4:1, so I was starting to experience the reality that a trap dipole, while able to operate on several different frequencies when correctly tuned, doesn’t give you as low an SWR on any of them as a simple one band wire dipole.
The light ran out, I didn’t get to finish pruning the 30m section, and then I made a mistake. I decided to leave the antenna outside for the night with the traps not yet completely weatherproofed. That would have been OK had it been a dry night, but the heavens opened and it rained quite heavily all night. So I’m not sure that my traps are compromised, but if I were to go ahead with this idea, it would be best to wind new ones with fresh coax just to make sure.
All I need is a low maintenance HF antenna that will work on most HF bands so I can switch the radio on any time of day or night and operate. It is very apparent that antennas follow the same basic rule for all devices, mechanical and/or electrical, and that is that they can either do one thing very well, or lots of things less well, but not lots of things very well. It’s the reason computers are much higher maintenance than your standard AM/FM radio and it’s the reason that until I find a QTH more amenable to antennas I will keep fussing and futzing with attempts to be QRV on all bands 80 thru 10.
Here's one of the traps, with a liberal coating of Brush-On electrical tape on the turns to hold them in place. You can also the see the same product over the connections inside. Unfortunately, the center conductor at the other end of the trap hadn't been coated. Same thing with the other trap. Aah shucks.
With a trip to England coming up in few days, it’s time to shelve this project until my return, but I’m already not too happy with the trap dipole idea – too much fussing and pruning required to end up with an antenna that will be a compromise. I think my next step will be to put up a simple 40 meter dipole, which will at least give me 40 and 15, then perhaps another dipole connected to the same center insulator for one other HF band. I’ve also been wanting to try out an end fed dipole that was so well described in the October 2010 issue of World Radio Magazine Online (download it now before it’s taken offline at the end of the year.)
Bloody antennas. Bane of my life. If they weren’t so much fun I’d forget them and take up stamp collecting 🙂