Greetings fellow radio geeks!

Lately I've been trying to build a decent antenna for the reception of HDTV broadcasts. Not so much because I don't want to pay for a commercial unit but because I want that geeky DIY experience.

My first attempt involved a 300/75 ohm matching transformer like this:
http://www.radioshack.com/product/index.jsp?productId=2062054&cp=2032057.2032187 .2032189.2032207&pg=2&parentPage=search&sr=1&origkw=matching+transformer&kw=matc hing+transformer&parentPage=family
and some insulated solid core copper wire. I bent 2 pieces of wire into steep Vs and attached one to each screw on the transformer. The end result was like one of those old KECH bow-tie antennas, but open at the ends.

That worked okay, but I read about the wide bandwidth of the folded-dipole design (and that it was better suited to the 300 ohm transformer) and decided to try that. I used the same transformer and wire but this time bent it into a flattened "bar" shape, with the transformer connecting in the middle - lame ascii art version below.
.___________________
(________.._________)

This worked better, but still was weak on certain channels (6-DT and 32-DT, whatever those are).

One thing I was never able to nail down precisely is how long the elements should be - I found lots of dipole calculators online, but nothing that specifically addresses the folded design.

Lastly, I considered adding multiple elements, tuned for different frequencies. That's where I started getting way over my head in the theory - can I just tack on additional elements or do they need to be separately isolated? Should they be stacked or colinear? Any other thoughts or suggestions?

Thanks!

I have had good luck with this type of antenna:

http://www.dl5neg.de/23cm_ant/23cm_ant.htm

Since the design is for 1296 Mhz, you need to multiply the dimensions by about 2.1 to center on the Portland UHF HDTV range (about 550 to 680 mhz).

You might want to experiment with a reflector (like a cookie sheet) behind (start at about 6 inches and experiment) your dipole to increase gain.

Stacking your bow-ties (or dipoles), check the picture here:

http://www.amazon.com/Antennas-Direct-DB4-Directional-Antenna/dp/B000EHYG9K/ref= pd_bbs_sr_5?ie=UTF8&s=electronics&qid=1199500181&sr=8-5

For 2, just run parallel wires between them and connect your matching transformer 1/2 way between the two so tnat the signal for each antenna arrives at the connection point in phase. For 4, study the picture carefully, note that the connecting wires are crossed so that signal from all 4 antennas arrives in phase at the connection point.

"Lastly, I considered adding multiple elements, tuned for different frequencies"

This idea sounds like a log periodic design (see link):

http://www.vk2zay.net/article.php/43

You might find the following website interesting: http://www.avsforum.com Click on "Forums".

Is KGW dtv 8.1 and dtv 8.2 on?
I can't get them on my indoor antenna. I used to.
I get the other Portland dtv channels ok except for KGW and KPXG.

Just checked, 8.1 has a football game in widescreen and 8.2 has the weather.

I believe KGW-DT is channel 46. If you connect your cable TV line directly to the set without going through the box, you get them on 8-1 and 8-2. After February 17, 2009, KGW-DT will broadcast on channel 8.

I like the statement in the Amazon.com product jr_tech mentioned:
f you are looking for an antenna that is practically guaranteed to perform, the DB4 is one of your safest choices.

"practically guaranteed"?

Other than in the interest of science, why bother building one when you can just go out and buy one?

Because it's fun!

Seriously. Making antennas offers a lot of creative options. It's sometimes surprising what ordinary "around the house" materials can do. The Pringles can wi-fi directional being a great example.

You know you are a geek if...

You shop for canned foods with a caliper

Jimbo takes the practical view: time is money to a working engineer, so why spend 3-4 hours making one's own $30 part? But for a nearsighted fumble-fingers like me, building something simple and larger like an antenna is an opportunity to do satisfying about getting a signal with my own hands.

(I took one look inside my book-sized microprocessor-controlled shortwave receiver and promptly put it back together, despite several very interesting mods for it posted on-line by braver and better-coordinated tinkerers than I.)

I could never solder well, so when I discovered wire-wrap, it was like the arrival of sliced bread! (Or those Switchcraft mini-plugs with screw connectors.) I'd bought a full-travel keyboard for $30 by mail-order back in 1980, and was confronted by 200 tiny prongs that I realized I could never solder (after burning four fingers to make two bad connections). I wrap-wired the whole thing in an hour or so. (Tandy published full schematics even for their early computers, including the keyboard matrix.) I proceeded to make other mods to my home computers until multilayer boards put an end to that. (Remember when computer magazines featured circuit mods?)

Crimp connectors were the next godsend, after I'd soldered a few DB-25s (ouch! ouch! ouch!) back in the day before cable assemblies were hobbyist-affordable.

(A retired engineer made a gift to me of a "soldering armature" so I could finally quit trying to use clothespins or binder clips as my "third hand.")

I can't breadboard (much less make my own PC board) but can handle "dead-bug" construction using wire-wrap. Yep, it's sure ugly, but once you screw the lid onto the project box, who's to know? For the really impatient (or extraordinarily clumsy), double-sided foam tape can take the place of gun glue for component mounting. A child could do it (which makes it about my level).

RS hasn't carried Forrest Mim's "mini-notebooks" for years, so I greatly regretting losing mine. But I found they're still available (bound into four volumes) http://www.forrestmims.com/engineers_mini_notebook.html

They were the only electronics books I could comprehend myself, and they're cheerful and encouraging.

Ah, now I remember another(cheapskate) reason why I didn't want to buy one - many/most/all of the channels will be switching back to VHF in relatively short order, so why waste a bunch of money on UHF gear?

Re: Jr_Tech's hybrid quad- interesting that he's not using a balun. I thought they were necessary? Supposedly the impedance of my folded dipole is about 300 ohms, which plays nicely with the 4:1 matching transformer I have(matching transformer = balun?). I also note that his design specifies 50 Ohms.. isn't a TV receiver (and coax) setup for 75 ohms? Is it possible that the losses of the balun are greater than the gain from having a balanced ciruit? (am I asking too many questions?)

At this point I'm not sure that my design is working too well, because I get almost-as-good results by just sticking the stripped end of my wire spool into the center of the UHF jack.

Thanks for the input, I'll continue to tinker and report back.

He may have a little skewing of the current distribution in the antenna because he does not use a balun:

"is even ok to solder it together without insulation at the top and bottom of the driven element since this are the points of maximum current and zero voltage. But due to the fact that we are not using a balun, the current maximum is not exactly at these points so we would short circuit a little bit of the antenna signal."

But I suspect that as a practical matter, as a receiving antenna this (as well as the 50/75 ohm difference) is not very important. I connect this antenna to either a 50 ohm UHF communication receiver (Icom 8500) or 75 ohm TV set... no problem! A double quad has a lower impedance than a dipole.

A balun converts a balanced situation (such as twin-lead) to unbalanced (such as coax, where one side is grounded (the term "balun" is coined from balanced/unbalanced). It may or may not be designed as an impedance transformer.

A double-double quad here:

http://www.hamuniverse.com/doubletwinquad.html

This author does use a choke balun to connect to the driven elements.

But consider what you are doing by building your own antenna. You are learning something ! Gee Wiz huh, do ya think? You can calculate ( albeit crudely with out getting into a lot of math ) wavelengths and feedpoint impedances of the dipoles and what configurations might work best and perhaps eventually know * why * they work best for the frequency spectrum you are receiving.

After you play around with all this and learn from it, you could build either from a kit or a pre made circuit board or from scratch if you have the talent, an appropriate amplifier to raise the RF level several db above the field density at your recieving point for consistent reception of HD signals.

I tried to get my brother to play around with all this, he lives several hundred miles from me and so that didnt work real well because I wasnt around all the time to bug him or build him one better than the little circle single element antenna he had up there....so I just bought him a 23 dollar jobbie from Wal-Mart for Christmas, he was thrilled to get it now so he gets HD consistently out in Washougal.

I found that over here,after fiddling about with yaggis and stacked arrays , having an amp behind the array works best to deliver a consistent HD signal. I do have to readjust the antenna directivity on rare occasions, due to variant conditions. Reflection problems I suspect. A rather obvious problem exists on one UHF channel when my son opens the refrigerator door to make a sandwich and our ABC affiliate goes dark for a minute or so while he finds the Mayonaise.

I think that most hobbies don't make sense if they are analyzed closely. If they turned a profit, then they would be considered commercial activities. Assuming that you want to build the antenna because you want to learn about building antennas or you just want the intellectual challenge, go for it!

As far as the balanced vs. unbalanced and impedance questions go, receiving antenna setups generally aren't too picky about precise impedance matching. If you build your antenna with a folded dipole, its impedance will be 300 ohms; you can connect this directly to 300 ohm twin-lead or go through a TV balun to connect to coax. Either of these setups is completely kosher.

The impedance of a rectangular loop antenna, however, is 100 Ohms, rather than 300. If your hybrid quad's driven element looks like the one shown on this page ( http://www.dl5neg.de/23cm_ant/23cm_ant.htm ), what you are seeing is two 100 ohm loop antennas connected in parallel to get a 50 Ohm impedance. I can't look at the link that JR_tech posted right now, as my work Internet access has blocked this page.

Any antenna that has a symmetrical construction would prefer a balanced feed. However, it is often possible to get away with directly feeding from coax, as you have seen. Unless you have access to the test equipment required to optimize your antenna, I would recommend sticking to a simple design, such as a dipole Yagi or a quad.

I do a lot of the things I do for self sufficency reasons. --that, and it's just fun.

Ever want a DVD player that does exactly what you want it to always? That's Linux, a nice older machine, some remote control interfacing stuff, and open code.

(skip to the movie, over previews, etc... no restrictions)

Maybe you want your car to perform in a specific way, or maybe save a lot of gas without having to first invest in an expensive new vehicle.

Get stuck with something important broken and little cash. (been there too many times)

Taking the time to know how things work seriously cuts your dependence on dollars and services that come from others. It also builds real choice, as in "I don't like all the established choices, so I'll just do what I think is cool."

For most things in life, one can spend their way outta most tight jams, so that's what makes this kind of thing a hobby. If there ever is a time though... it sure is nice to just apply the grey matter and whatever is at hand toward a fix!

While trying to figure signal paths, I drew lines in GoogleEarth between sites with the display of elevation cranked up. I expected I'd have to mouse along the line to spot-check the elevations of intervening hills, but was surprised to find the drawn lines apparently track elevation between endpoints as well as direction, as they seemed to disappear into intervening hills.

Here is a nice Yagi calculator that lets you specify the desired gain, frequency, and element diameters

http://www.k7mem.150m.com/Electronic_Notebook/antennas/yagi_vhf.html

I ran the calculator for 15dB gain @ 550 MHz. A total of 19 elements would be needed. The boom would be slightly over 10' in length. The driven element could be either a dipole or a folded dipole; according to this page, either could be used without altering the spacing and lengths of the parasitic elements. I'd recommend getting some PVC pipe for the boom, some metal rods or heavy-gauge solid wire for the elements.