What's the most interesting antenna(s) you've encountered, or wanted to build?

[cdev]

I want to build a conical log spiral, which is a log spiral (broadband circular polarization) antenna with the addiition of gain. How would it look? Basically like a cross between a wizards hat and a barber pole. So i am trying to figure out a way to fabricate something like that. Obviously the material would have to be flexible.

The original papers are on Archive.org by a gentleman whose last name is Dyson.  At the point of the antenna (the feed) must go a balun, which is another challenge.

[hendorog]

So i am trying to figure out a way to fabricate something like that. Obviously the material would have to be flexible.

Perhaps 3d print a cone as a template and wrap wire around it. Something like this (see picture #2) but with a taper:
http://www.thingiverse.com/thing:586400

[cdev]

Great find, yes, that's a very good way to go with a helical antenna!

Here is a variation on the unidirectional log spiral that would lend itself to that kind of form.

http://cira.ivec.org/dokuwiki/lib/exe/fetch.php/events/unidirectional_equiangular_spiral_antenna.pdf

No reflector necessary, the two arms act as all of them, 'driven element' reflector and directors! Electrically, this is quite similar to a LPDA except circular. Can you see why?

 A similar antenna for a linear polarization can be accomplished extremely easily with the two arms in a zig zag, each made with a single piece of wire.

You can put two zig zag LPDAs on the same "stick" to get a crossed, yagi-like antenna, and then you could accomplish circular polarization - in both directions as well by means of a 90 degree hybrid.

So i am trying to figure out a way to fabricate something like that. Obviously the material would have to be flexible.

Perhaps 3d print a cone as a template and wrap wire around it. Something like this (see picture #2) but with a taper:
http://www.thingiverse.com/thing:586400

[T3sl4co1l]

There's more than a few standard designs for planar and conical wideband antennas, with smooth, periodic, or log-periodic (self-similar) design.

There's actually a theorem that any structure you build, if it's self complementary and repeats forever, is exactly half Zo (377 / 2 ohms), at all frequencies.  And all that resistance is radiation resistance, so, it's one heck of a wideband antenna.

BTW, these aren't to be made with wire -- you need wide, flat (or round) conductors to do this.  I mean, you can use a mesh equivalent too, but that's way more work than cutting up foil or sheet metal!

Personally, I rather like the idea of a monstrously large ridged horn antenna.  I doubt one is very practical for, say, even the upper half of the shortwave band (it'll be some meters wide!), and construction needs to be extremely accurate to work over a wide bandwidth (obviously, surfaces need to be within 1/4 wavelength at whatever the upper "still pretty okay" frequency should be).  But they do look cool.

Easy, lower frequency, wide-ish designs include the bowtie and biconical dipole.  Commonly seen in EMC test labs, for good reason.

Tim

[cdev]

There's actually a theorem that any structure you build, if it's self complementary and repeats forever, is exactly half Zo (377 / 2 ohms), at all frequencies.  And all that resistance is radiation resistance, so, it's one heck of a wideband antenna.

BTW, these aren't to be made with wire -- you need wide, flat (or round) conductors to do this.  I mean, you can use a mesh equivalent too, but that's way more work than cutting up foil or sheet metal!

Yes yes,, I have had much better results with the flat sheeted, 50% cover flat conductor anything.

Other kinds of structures sometimes behave similarly. Two conductive spheres, apparently, together can make an excellent broadband antenna.

Rhombic and half rhombic antennas - which are terminated at the front end, act like a infinitely long wire and take advantage of the lobes -

Personally, I rather like the idea of a monstrously large ridged horn antenna.  I doubt one is very practical for, say, even the upper half of the shortwave band (it'll be some meters wide!), and construction needs to be extremely accurate to work over a wide bandwidth (obviously, surfaces need to be within 1/4 wavelength at whatever the upper "still pretty okay" frequency should be).  But they do look cool.

they do!

I suspect they are not so precise but (online) I have seen several quite doable small projects that use cheap materials, for example, foil covered cardboard- discarded refrigerator boxes, plain (foil glued onto) foam core and other easily found or fabricated materials to make horns for use in backyard radio astronomy projects and the like.

Perhaps horns are among the easiest way to get a really high gain antenna for use in radio astronomy.

Easy, lower frequency, wide-ish designs include the bowtie and biconical dipole.  Commonly seen in EMC test labs, for good reason.

Tim

Both of them work well, also there is an antenna you can make with two planar disks that behaves similarly. Surprisingly, even if you only use two identical dishes (i.e. pizza plates or similar) its pattern is still omnidirectional in the H plane, with most gain concentrated at the horizon.