Area 51 for Antenna Modelers & Builders (See Post #1)

[300ohm]

I don't know what's up with the dip at channel 48.

Nothing really. Thats typical of what occurs on the GH at the breakup at the high end frequencies of its bandwidth. Also since the SWR dips at the same time, the net gain curve is smoother.

A prerequisite to building a 2nd or more iteration fractal antenna should be to have training in the Japanese fine art of origami, heh.

For some odd reason, I seem have a copy of "Fractal Programming in C" by Roger T. Stevens 1989, in front of me. The example programs are on 5 1/4 inch floppy, MS/PCDos format (unopened). If you think some may be useful, Ill see if the disk still reads once I dig thru my collection of old hardware crap, heh.

[firimari]

Thanks, but I think I'll pass on the ancient code offer. I seem to have Koch Curves figured out, and I don't think I'm going to go any further with them. If I get bored, I may do a more in-depth analysis of a full size GH fractal.

[300ohm]

What ? No Peano, Hilbert, Sierpinski, Dragon, Phoenix curves, not to mention the Mandelbrot set ? Heh, heh.

Yeah, theyre basically only practical for cell phone/PDA type antennas printed on some kind of pcb.

That Italian Job antenna, with its nice wide front pattern, may make a nice wifi antenna for my use once I scale it for 2.4 GHz. Of course, I got to find a cheap source for the cable/connectors. They seem to want a fortune for a measly 25 foot piece, heh.

[Xauto]

300ohm: If you make it for 2.4 Ghz., you may need 1/4 wave lenght separators between the diamonds. Google Quados.

[300ohm]

That doesnt look bad either. Im wondering if I use the same technique on the Italian Job, that it will be even better still, ie wide beam, high gain.

My biggest problem is finding cheap coax and connectors for it. It goes totally against my grain and principles to spend $50 to $100 for a measley 25 ft piece, or $10+ for a connector : http://www.newegg.com/Product/Produc...82E16833998039
If I did that, I would be kicking myself in the butt for a long time, heh.

Its pretty obvious that a lot of those Youtube wifi antennas havent been modeled.

[RamKat]

Here is the NEC file for the dual half folded dipole seen from BT

I discussed the modeling results here

http://www.digitalhome.ca/forum/showthread.php?t=83772 (Post #265)

CM Omnidirectional FM Antenna
CM Based on a photo in the Blonde Tongue catalogue
CM Optimized for 98 MHz
CE
SY h=3.009341
SY f=25.3431 'Inches
SY g=33.5266 'Inches
SY v=2.452685 'Inches
SY Wire=0.0404 'Inches
SY Tilt_angle=3.997235 'Tilt angle of the dipole
SY i=h*0.41421 'Circle approximated with an octagon
SY h1=0.70711*i
SY h2=h1+i
SY tilt1 = h1*tan(Tilt_angle)
SY tilt2 = h2*tan(Tilt_angle)
SY tilt = h*tan(Tilt_angle)
SY s0_z4=-v*0.5
SY s0_z3=-v*0.5-h1
SY s0_z2=-v*0.5-h2
SY s0_z1=-v*0.5-h
SY s0_z5=v*0.5-h
SY s0_z6=v*0.5+h1-h
SY s0_z7=v*0.5+h2-h
SY s0_z8=v*0.5
SY s0_y1=-f
SY s0_y2=-f+h1
SY s0_y3=0
SY s0_y4=0.75
SY s0_y5=g
SY s0_x1=-f
SY s0_x2=-f+h1
SY s0_x3=0
SY s0_x4=0.75
SY s0_x5=g
GW 1 23 Tilt s0_y3 s0_z1 Tilt s0_y2 s0_z1 wire
GW 2 3 Tilt s0_y2 s0_z1 Tilt2 s0_y1 s0_z2 wire
GW 3 3 Tilt2 s0_y1 s0_z2 Tilt1 s0_y1 s0_z3 wire
GW 4 3 Tilt1 s0_y1 s0_z3 0 s0_y2 s0_z4 wire
GW 5 55 0 s0_y2 s0_z4 0 s0_y5 s0_z4 wire
GW 6 1 Tilt s0_y3 s0_z1 s0_x4 s0_y4 s0_z1 wire
GW 7 23 s0_x3 Tilt s0_z5 s0_x2 Tilt s0_z5 wire
GW 8 3 s0_x2 Tilt s0_z5 s0_x1 Tilt2 s0_z6 wire
GW 9 3 s0_x1 Tilt2 s0_z6 s0_x1 Tilt1 s0_z7 wire
GW 10 3 s0_x1 Tilt1 s0_z7 s0_x2 0 s0_z8 wire
GW 11 55 s0_x2 0 s0_z8 s0_x5 0 s0_z8 wire
GW 12 1 s0_x3 Tilt s0_z5 s0_x4 s0_y4 s0_z5 wire
GW 13 5 s0_x4 s0_y4 s0_z1 s0_x4 s0_y4 s0_z5 wire
GS 0 0 0.0254
GE 0
GN -1
EK
EX 0 13 2 0 1 0
FR 0 21 0 0 88 1
EN

[Xauto]

RamKat: Is the feed to 75 ohms, and is it only off the 1/2 folded leg of each. I have a yagi that use's 1/2 folded. The feed is from the center of the dipole to the 1/2 folded leg. Have you tried flipping the bottom element to change the feed point space?

[RamKat]

XAuto, it feeds to 300 Ohms with a SWR of very close to 1.0. I set it up to feed between the short legs of the two elements (was not clear from the picture if this was where it should have been) Do you mean treating the two half folds as seperate antennas and then combining the feed?

Here is the layout

http://img17.imageshack.us/img17/9839/framee.th.jpg

Here is the donut

http://img14.imageshack.us/img14/7918/dhdp3d.th.jpg

Yes I tried the flip but was having some other finger trouble so maybe I should try it again.

[firimari]

My current optimization project: Take all of j3d's GH{4,6,8,10} and re-optimize them for my build style (6 gauge copper / quarter inch aluminium), and my channel range (17-47)

If only I had four cores instead of two on my windows box.

[Xauto]

RamKat: Thank you for the info. If you do treat the two half folds as seperate antennas and then combine them, it will just make the phase lines move complex.

[stampeder]

RamKat, can you try posting your images this way:

http://www.digitalhome.ca/forum/showthread.php?t=92167

[300ohm]

If only I had four cores instead of two on my windows box.

There is a benchmark thread on this site for that. It seems 4nec2 only will use 3 of the 4 cores. My gut feeling is, more mhz on a dual core is better overall.

[300ohm]

Yes I tried the flip but was having some other finger trouble so maybe I should try it again.

Yeah, that is somewhat weird. I was also expecting the gain to be a bit better on that design. Oh well.

[300ohm]

I found Ross Andersons, W1HBQ, site while perusing the web. He has posted the NEC files for his models. The Curtain Quad is the one with the most interest. 221 element, 26.55 dbi raw gain, and about a 1 degree beamwidth, BI-DIRECTIONAL, heh. Stick a reflector on that baby, and youll get over 30dbi !
http://home.comcast.net/~ross_anderson/

[300ohm]

Dont let this small picture fool you, this is one large antenna ! 241 inches wide by 307.20 inches high, heh. (20 by 26 feet !) (but still probably easier to build and get right than a 14 - 20 ft parabola)

At 635mhz:


Rescaled, I got 26.25 dbi Raw Gain bidirectional, with an SWR of 3.29 at 300 ohms at 635mhz. Net Gain is about 24.85 dbi. Not too shabby, heh. Will run frequency sweep later.

The NEC file : http://www.wuala.com/300ohm/Document...urtainQuad.nec