Fractal TV Antenna Research & Development - Page 3 - Canadian TV, Computing and Home Theatre Forums

Reply
 
LinkBack Thread Tools Search this Thread Display Modes

post #31 of 341 (permalink) Old 2008-11-22, 01:46 AM Thread Starter
Veteran
 
Join Date: Mar 2008
Location: Somewhere in Delaware on the flat side
Posts: 7,360
Pictures of the GH UHF antennas fractalized so far. The pictures are to scale :



On the left is the Gen2 fractalized with the /4 fractal. In the middle is the Gen1 fractalized with the /4 fractal. On the right is PEH's Gen1 fractalized with the /3 fractal.

My builds/plans (not the latest models) are located here.
300ohm is offline  
Sponsored Links
Advertisement
 
post #32 of 341 (permalink) Old 2008-11-28, 01:59 PM
 
Join Date: Sep 2008
Posts: 551
With gen2 fractal element and 30"X40" with hour-glass 1" by 3 1/4" gap reflector makes my best set up.
Xauto is offline  
post #33 of 341 (permalink) Old 2008-11-29, 12:19 PM Thread Starter
Veteran
 
Join Date: Mar 2008
Location: Somewhere in Delaware on the flat side
Posts: 7,360
Best in what way ? Did it help your far lower uhf channels and your far upper uhf channels ?

My builds/plans (not the latest models) are located here.
300ohm is offline  
 
post #34 of 341 (permalink) Old 2008-11-29, 01:31 PM
Veteran
 
Join Date: Nov 2006
Posts: 1,543
Mmm.. I wonder if all that this "fractalization" is doing is shortening the overall antenna dimensions, while maintaining about the same element lengths (but bent)?

In other words, is it simply approximating the trued'n'true methodology of partially coiling up the elements to make them physically shorter, yet the same (more or less) electrical lengths?

I wonder if some short coils in the middles of each element leg, maintaining the original zig-zag shapes, would produce an even better effect?

Sure I could model it.. but that's not my fortÚ here!
mlord is offline  
post #35 of 341 (permalink) Old 2008-11-29, 02:48 PM
 
Join Date: Mar 2006
Location: Ottawa ON
Posts: 391
Quote:
Originally Posted by mlord View Post
Mmm.. I wonder if all that this "fractalization" is doing is shortening the overall antenna dimensions, while maintaining about the same element lengths (but bent)?

In other words, is it simply approximating the trued'n'true methodology of partially coiling up the elements to make them physically shorter, yet the same (more or less) electrical lengths?

I wonder if some short coils in the middles of each element leg, maintaining the original zig-zag shapes, would produce an even better effect?

Sure I could model it.. but that's not my fortÚ here!
mlord,

The GH4 fractal model by 300ohm is interesting from a modeling point of view. It was based on the original GH4 VHF-Hi "Monster" that had a total height of about 106 inches. The fractal nec file posted by 300ohm, had similar results in gain and swr, with a max height of about 89 inches.

I assume you are suggesting another fractal idea to further reduce the total height, but it's not clear to me, exactly what you mean.
Can you explain what you mean by "an even better effect...using some short coils in the middle of each leg" ?
If you don't intend to model it, could you post a sketch of your proposed model? ....thanks
Autofils is offline  
post #36 of 341 (permalink) Old 2008-11-29, 03:07 PM Thread Starter
Veteran
 
Join Date: Mar 2008
Location: Somewhere in Delaware on the flat side
Posts: 7,360
I know what mlord means. For example, I modeled a bowtie fractalized (/4) and compared it to a non-fractalized bowtie of the same width legs. In other words, I took a 9 1/2 inch bowtie, fractalized it /4, and its new width legs were about 7 3/4 inches. Then I compared it to a straight 7 3/4 inch leg bowtie. The results on the bowtie were almost identical.

But the GH doesnt behave in that way when fractalized.

The biggest benefit to fractalization is to the VHF antennas, due to their size. Fractalization does knock the gain down a little in the models, but if that can be tolerated, the size difference of about 20 % plus the added stiffness of the structure is well worth it IMO.

Quote:
Fractalization does knock the gain down a little in the models
But Ive also read that current modeling software doesnt handle fractals that well. Since Xauto says he gets better gain with the fractalized version, I just dont how well the model reflects real life.
But if real life gain is better than the model, Im happy.

My builds/plans (not the latest models) are located here.
300ohm is offline  
post #37 of 341 (permalink) Old 2008-11-29, 04:07 PM
Veteran
 
Join Date: Nov 2006
Posts: 1,543
Quote:
Originally Posted by Autofils
I assume you are suggesting another fractal idea to further reduce the total height, but it's not clear to me, exactly what you mean.
I'm suggesting that perhaps just partially coiling up the elements (coils, as in 3-dimensional round spirally things) might achieve the aim of a shorter overall dimension better than other approaches.

I suspect the "fractalization" stuff might really be just a bunch of short bends which approximate the effects of (very) small coils -- nothing to do with fractals themselves. Just a guess, but..

Coils are very common in antenna design, especially at very long wavelengths and/or when physical antenna size is important (mobile devices, FM radios, ..).

Cheers
mlord is offline  
post #38 of 341 (permalink) Old 2008-11-29, 04:15 PM
 
Join Date: Mar 2006
Location: Ottawa ON
Posts: 391
Quote:
Originally Posted by 300ohm View Post
The biggest benefit to fractalization is to the VHF antennas, due to their size. Fractalization does knock the gain down a little in the models, but if that can be tolerated, the size difference of about 20 % plus the added stiffness of the structure is well worth it IMO.
Another option for reducing the size of a GH VHF antenna, is to keep the original zig-zag geometry, but reduce the array's half-angle to less than 45 degrees.

I posted a Small Footprint GH4 VHF-Hi version with a half-angle of 22 degrees, that results in a max height of 41 inches.
The RawGain and SWR results, as well as the Nec file are available at this link...
Small Footprint (41"h x 86" w) GH4 VHF-hi for Ch 7-13

Possible Improvements:
=================
I did not spend much time on optimizing the parameters, so you can probably squeeze more gain from this design with further optimizations of the 4 pair collinear reflectors.

For a given number of reflector pairs, the gain increases with increasing half-angle values, so you can customize your own design with whatever gain/height trade-off that you choose.

Increasing the number of reflector pairs to a GH6, would provide some additional gain without any increase in total height.
I'd expect a GH6 version might have higher net gain than any of the commercial antennas posted on Ken Nist's web site.
Anyone have the time to model and optimize a GH6 version and find out if this is true ?
Autofils is offline  
post #39 of 341 (permalink) Old 2008-11-29, 07:00 PM
 
Join Date: Mar 2006
Location: Ottawa ON
Posts: 391
Quote:
Originally Posted by mlord View Post
I'm suggesting that perhaps just partially coiling up the elements (coils, as in 3-dimensional round spirally things) might achieve the aim of a shorter overall dimension better than other approaches.

I suspect the "fractalization" stuff might really be just a bunch of short bends which approximate the effects of (very) small coils -- nothing to do with fractals themselves. Just a guess, but..

Coils are very common in antenna design, especially at very long wavelengths and/or when physical antenna size is important (mobile devices, FM radios, ..).

Cheers
mlord,

Thanks for this description.
I think you are suggesting to keep the zig-zag geometry but collapse each diagonal sub-section of the array into a coil, so the diagonal length of each sub-section becomes shortened due to a coil of N turns with diameter D.
The max reduction would occur if the entire sub-section was a coil with very short stubs at the bend points, but in your first post you used the phrase "some short coils in the middle of each element leg", so I assume you consider less coil turns and longer straight stubs per diagonal sub-sections are likely better.

For anyone that has an interest in modeling and time available, this would present an interesting challenge and a good learning experience.
Autofils is offline  
post #40 of 341 (permalink) Old 2008-11-29, 08:52 PM Thread Starter
Veteran
 
Join Date: Mar 2008
Location: Somewhere in Delaware on the flat side
Posts: 7,360
Quote:
Coils are very common in antenna design, especially at very long wavelengths and/or when physical antenna size is important (mobile devices, FM radios, ..).
Yeah, coils are very common in mobile vhf antennas. And flat fractal antennas are now very common in cell phones. (I dont consider the coil in AM/FM radios as an "antenna", but rather part of a tuning circuit even though it does the job of an antenna too. But it does remind me of something I wanted to experiment with a long time ago, heh. )
There are examples in the 4nec2 sample files of helix antennas. And the Geometry Builder has an automatic helix model builder. However, coils still look to be a bear to model, and to build correctly without some fancy test equipment, like GerryB has. Its much easier just to make the additional bends like in the above fractal models IMO.

Quote:
Another option for reducing the size of a GH VHF antenna, is to keep the original zig-zag geometry, but reduce the array's half-angle to less than 45 degrees.
I looked over that design. But doesnt reducing the array angle go against Doyt Hoverman's and your own original findings ? So tuning the co-linear rods really changed all that ?

My builds/plans (not the latest models) are located here.
300ohm is offline  
post #41 of 341 (permalink) Old 2008-11-29, 10:31 PM
 
Join Date: Mar 2006
Location: Ottawa ON
Posts: 391
Quote:
Originally Posted by 300ohm View Post
I looked over that design. But doesnt reducing the array angle go against Doyt Hoverman's and your own original findings ? So tuning the co-linear rods really changed all that ?
300ohm,

Don't fall into the trap of assuming Hoverman's original statements about the array angle stated in his patent, as being totally true for all cases.

When I first modeled generation 1 SBGH for UHF, I did look at the effect of reducing the angle and indeed found that for the UHF design, the gain was reduced with half-angles less than 45 degrees. That validated Hoverman's patent claim and hence there was no apparent value in reducing the array half-angle for the UHF design.

Since fading from this forum, I have been exploring two private projects:
1. GH version for WiFi (~2400 Mhz)
2. GH version for FM (88-108 Mhz).

A normal GH4 FM antenna (half-angle of 45), had a total height of over 200 inches. The gain/swr was excellent with net gain around 12dBi, but.... it is obviously just not practical.

That's when I began investigations on the GH with reduced half-angles.
While this produced a "useable" FM antenna, the total height is still over 80 inches, which is too much to fit in my attic.

That lead to a Combo FM/VHF-Hi GH4 design with a total height of about 53 inches. As you can appreciate, most combos force trade-offs, where the gain is reduced due to maintaining a low swr for the 2 to 1 frequency coverage ratio. The FM gain of this combo is nothing near 12dBi, but it is higher than the single quad FM presently being used, which receives 89.5 (NCPR Potsdam, NY) well in mono, but noisy in stereo mode. I'm about ready to try this combo in my attic; I'm expecting much better FM stereo reception.

As a result of this work, the next step was a reduced half-angle GH4 designed just for the 174 to 216 Mhz bandwidth. That design was posted recently as the GH4 Small Footprint VHF-Hi design.
The model results of gain/swr looks quite good, considering the overall size of 41"h x 86"w, but I have not tested any proto builds.

This VHF-Hi design could use some further optimizations, and I suspect an optimized GH6 version might show better results than any of Ken Nist's models of commercial antennas for Ch 7-13.

Out of interest, I also posted a GH4 UHF design with an reduced array angle of 35 degrees, and was surprised to see quite a reasonable gain for a total height of 24.2 inches. This might have applications, for those who need both VHF-Hi and UHF on a single mast, and want to keep total overall height under 6 ft.


Finally, let me close out my discussion of array half-angles:
For very high frequencies, such as WiFi, it might be an advantage to stretch the array rather than squish it, and that is what I am currently investigating for my WiFi GH antenna.


Cheers
Autofils is offline  
post #42 of 341 (permalink) Old 2008-11-29, 10:42 PM
OTA Forum Moderator
 
Join Date: Jan 2005
Posts: 24,878
Quote:
Originally Posted by 300ohm
There are examples in the 4nec2 sample files of helix antennas
AFAIK helical antennas cannot be configured for broadband - they are one frequency/wavelength wonders.
stampeder is offline  
post #43 of 341 (permalink) Old 2008-11-29, 11:04 PM
 
Join Date: Mar 2006
Location: Ottawa ON
Posts: 391
Quote:
Originally Posted by stampeder View Post
AFAIK helical antennas cannot be configured for broadband - they are one frequency/wavelength wonders.
Stampeder,

FWIW,
Helix antennas have applications at very high frequencies and there are designs with a significant bandwidth.

Here's a tapered helix antenna for 5.5Ghz that claims > 40% bandwidth.
Incidentally has anyone tried the trial version of the SuperNEC modeling software package, available via the download tab at this web site ?
It appears quite impressive, but is also very expensive. I wonder if there is a low cost student version.

http://www.supernec.com/thelix3.htm
Autofils is offline  
post #44 of 341 (permalink) Old 2008-11-30, 12:07 AM Thread Starter
Veteran
 
Join Date: Mar 2008
Location: Somewhere in Delaware on the flat side
Posts: 7,360
Quote:
It appears quite impressive, but is also very expensive.
What impresses you about it over 4nec2 ? (other than the fancy airplane antenna modeling).

Quote:
I wonder if there is a low cost student version.
Free, but 300 segment limit.

Q3: What is the difference between the trial SuperNEC Licenses?


A: There are 2 trial licenses:
Free Academic License:
Duration - 1 year
Limitation - 300 segments
Cost - Free
Requirements - You must be a student or lecturer and fax or e-mail proof (normally a University Card) to +27 11 262 5156.
You must have MATLAB v 6.0 or higher.

My builds/plans (not the latest models) are located here.
300ohm is offline  
post #45 of 341 (permalink) Old 2008-11-30, 12:48 AM
 
Join Date: Mar 2006
Location: Ottawa ON
Posts: 391
Quote:
Originally Posted by 300ohm View Post
What impresses you about it over 4nec2 ? (other than the fancy airplane antenna modeling).
You might be interested that SuperNEC supports dielectric-coated wires , but I'm not sure if that is in the trial version.

I am always interested in seeing what new info can be gained from other modeling software packages. Just reading the software's help files contain can provide very useful info and sometimes offer other explanations for the complexities of modeling.
For these reasons, I will take a look at the trial version of SuperNEC.
Autofils is offline  
Reply

Quick Reply
Message:
Options

Register Now



In order to be able to post messages on the Canadian TV, Computing and Home Theatre Forums forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.

User Name:
Password
Please enter a password for your user account. Note that passwords are case-sensitive.

Password:


Confirm Password:
Email Address
Please enter a valid email address for yourself.

Email Address:
OR

Log-in










Thread Tools Search this Thread
Show Printable Version Show Printable Version
Email this Page Email this Page
Search this Thread:

Advanced Search
Display Modes
Linear Mode Linear Mode



Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On

 
For the best viewing experience please update your browser to Google Chrome