How to make a fractal antenna for HDTV / DTV plus more on the cheap

Step 5 Create fractals

You will repeat this step for each wire. Each bend on the wire will be 60 degrees exactly as we will be making equilateral triangles with this fractal. I used two pairs of pliers and a protractor. Each bend will be made at the 1 marks. Make sure you visualize the direction of each bend first before making it! Use the diagram below to help.

Comments

[captsomer]

I understand what you are doing here but I think the diagram is a bit confusing. At the top of the picture, the first bend is labeled "60deg". Really its a 120 deg bend. I know its bent 60deg from straight, but I think the picture is confusing. The next bend is correct but the third bend is the same problem as the first. I just used 120 deg bend then 60 deg bend and then 120 bend and so on...

[da winksta]

YUP! cool instructable but fix that diagram please

[williamruckman]

You are correct. i imagine it as if there are straight lines there. I say 60 degrees because they are suppose to be equalateral triangles. And equalateral triangles have 3 60 degree angles in them.

[baddahbing]

My mom wanted to dump cable, but could not get a dish because of her condo association. Doing a little research on TV antennas, she opted to buy a Philips indoor/outdoor antenna, with full low/high vhf & uhf receiving capacity (channel 2 thru 69), with a 40 mile range. Her neighbor bought a Terk indoor antenna with similar specs, but with a 45 mile range. They live about 35 miles southwest of the Chicago TV broadcast towers.

The Philips received only 2 channels, very badly. The Terk received none. So I decided to build her an antenna. After weeks of research, I decided to combine a few designs. I found the basic fractal design on Ruckman's sight, but changed it slightly. See attached drawing/photos.

I came up with a picture frame antenna that works really well. I live 25 miles northwest from the broadcast towers. The day I tested this antenna it was raining and jets were flying directly over head of the house ( 200 to 300 feet above). It should be noted that the jets disrupt the roof antennas signal.

Every station came in clear, no problems, even channel 2 (cbs) which is the hardest station to get (low vhf band). So I took it into the basement. Totally below grade, steel, concrete, not signal ever was received before. This antenna received all stations clearly, even channel 2. I couldn't believe it.

Took it to moms. Placed it on her TV (54 inches off the floor. Got every channel but channel 2, very cleanly. She's using it as I write this. Channel 7 (abc) which is also hard to get (also vhf), comes in better than it did on her cable.

I'm working on a better vhf antenna to try and get her channel 2. Hope the pictures make it.

Baddahbing

[hankuruski]

It is incredible how this simple design grew legs and run like this. I watched NOVA: Fractal on PBS once and was searching about and found this instructable. INCREDIBLE! To the first INVENTOR "thank you, Sir!" And, this refined model I thought to claimed this first once realized to expend the design by the fractal pattern from wikkipedia but you posted first so you are the MAN! Regardless, "WHO" was on first since this first simple design was free and the subsequent designs are also F.R.E.E and I am here to say "why would anyone pay more...for an antenna?" I reside in San Diego area but reaching to get LA stations with expensive antenna...yup DB8 but no luck so now for cosmetic reason to the house I reverted to this design and settle for just the local stations it is absolutely fabulous!

Gentlemen, you are geniusesseess!

[DIYSlacker]

I've made your version of the antenna with 1/2" bends (for a total of 9" per branch or 18" per element with 12 gauge wire). the only variation is that I've also kept the 9" distance between the elements and connected the balun at the middle (4.5").

from my tests (15 or so miles away from the towers, all in the same direction) I'm indeed getting very good results and constant signal for UHF, as expected (all channels 80-100%, mostly between 90-100%)

on the other hand one channel using VHF 7 is only 40% and quite constant, but when cars pass by the street i get breakups. i assume i could improve signal by adding a reflector and moving the antenna to find the "sweet spot" where all channels are in best shape.

overall I'm not sure there is much difference between this one and the normal "V" shaped element DB4 antenna, other than smaller size. it may work better if drawn on a PCB.

[DIYSlacker]

OK i tried adding a reflector in various positions up to 6" from the elements and it made the signal degrade significantly up to being lost on most channels. not sure if it's just my particular case or if the reflector for this kind of antenna shape makes the signal worse generally.

moving the antenna (without a reflector) around a bit increased channel VHF 7 to about 50-60% and slightly degraded the others to around 80%. I'll do some more experimenting, maybe adding an amplifier into the mix over the next few days.

[DIYSlacker]

i've tried your original 1" distance between the stacked elements (about 6" on center), but it's getting worse signal, especially on VHF 7 so i'm back to my 9" distance on center between them. the reflector at any distance made it definitely worse.

overall my bi-quad antenna with reflector (looks like a squared number "8" with a 5" arm length) beats this one in both UHF and VHF (channel 7).

[curiousgeorge03]

Hello,

Could you tell me how you connected the 4 dipoles to the two center wires and the tabs for the matching transformer?  Did you solder them?  And did you bend the copper wires by hand?  That must have been tedious.  I would like to try your design and was trying to figure out how to go about doing it.  Thanks.

[runninutes]

I've been thinking about making one of these, but was thinking about iterating once more on the fractal. However, there's an important question that I hoped someone could help with. Which is the important length? The overall length of the wire, or the length of each segment?

That is, the pattern calls for 8 inch wires bent into 8 segments of 1 inch each. Is the 8 inch length the important part for the calculation of what frequencies I'm trying to catch, or is it the 1 inch measurement? Thanks!

[nathenial]

what is the trick to maximising signal strangth, is it more acute bends or less obtuse bends, and can i add more then just two of these fractal shapes on either side, would that increase reception

[williamruckman]

acute or obtuse doesn't matter as long as it is a fractal shape. the fractal shape itself doesn't give you more gain at the center frequency but does give you a wider receive bandwidth at the cost of slightly lowering your center frequency gain. It is a trade off but well worth it for stability across the spectrum. This antenna design is based of dipoles, and with dipoles more elements means more gain. You could increase your gain by adding more elements but it will increase the total size of the antenna. But...it will still be smaller than a normal dipole of the same frequency.