who is willing to experiment with the cube antenna

[PA0NVD]

About 40 years ago during a 4 month trip to El Paso, TX, I did experiment with a wire antennea that could swith 4 directions. It alway was in the back of my mind, and today I made a model at 330 MHz. It seems to me a very nice antenna for 7 MHz, but I don't have the space to experiment at full size 7 MHz, perhaps in future. If someone has the space, please be so kind to report the results to me. Off course it can be made at other frequencies as well to prevent a rotor.
At 330 MHz indoor, the results were very promising, Gain did approximate a 3 el yagi and FB ratio was approx 15 dB. Because it can be made as a wire antenna supported by a single pole of 0,4 wavelength or by wires in a tree, it is quite simple to construct. It actually is a cubical quad where the sides are at 45 degrees forming a cube. The bottom and top wires are bend towards its other  with the top wires all connected together forming a cube at his point. At the bottom, the four wires can be connected to a relay circuit connecting the reflector to a stub or coil and the director to the feedline. Doing so one can swith 4 directions. The gain will be a little less than a full size cubical quad. Bending the top and bottom towards each other will compromise gain and FB ratio.
The sides of the cube are 0,27 wavelength, the height is a little compressed to be 0,35 wavelength. I used a stub of 300 Ohm open line of 0,033 wavelenth for the reflector, but one can use a coil instead.
So, take a vertical pole of any material of 0,4 wave length. It doesn't have to be strong, it has the antenna wires as guide lines. Connect 4 wires at the top all together. Spread these wire at 90 degress and approx 40 degress downwards. After 0,27 wavelength, connect a plastic guide wire to strech the wire and support the vertical pole and bend the antenna wires back to the bottom of the mast . They should come together at 0,35 wavelength from the top. That will give you a cube at his point and the sides are all spread by the plastic lines to a support point giving tha pole a nice guide for wind. 
The top of the antenna can also be supported by guidelines to trees etc to prevent a vertical pole.

[PA0NVD]

here a sketch of the antenna

[WBear2GCR]

Presumably, the base needs to be not close to the ground/earth?? Or??

Also, why would the 90deg element(s) effect or cause there to be F/B ratio - or what direction does the thing radiate in?
I'd have expected more like a cardioid effect??

I'd have guessed broadside to a "cube"??

This looks interesting... I've acquired a 10M military portable telescoping mast that I'm threatening to use fixed mobile,
and have been pondering various schemes to utilize it for antennas on bands starting at 40m and going higher...

On 15m or 20m the thing would be fairly small in dimension!
Have you done any simulation on it??

                            _-_-bear

[PA0NVD]

I don't see any reason that the base should be close to ground other than that you need a larger mast or higher support. The antenna is in principle a cubical quad though bend together. The radiation is at the direction where the feedline is connected while the other element is a reflector. When you short the other element in stead of the coil, that element becomes a director and the radiation is reverse. But I measured a few dB less gain than in the refelectro situation though the FB ratio is similar.

[PA0NVD]

Bear, WB6BYU did some modelling and got quite nice results. He got approx 7.4 dBi gain and two sidelobes at 135 degrees just as the X beam. When the elements are shaped as a triangle with a flat bottom, the feedpoint is raised from the ground and the gain increases to 8.4 dBi
He got 1 dB more gain as director-DE than a reflector-DE . My tests showed the opposite, 1 dB more gain from a reflector configuration, but that are indoor tests at 330 Mhz, so no ground effects included and reflections from the walls. The discussion is running at Eham

[WBear2GCR]

I asked about the ground placement, since it is easier to set up in a portable/fixed mobile situation with the bottom
closer to the ground with 40M size elements...

And, I misspoke, I ought to have said bi-directional, not cardioid, so sort of a bi-directional cardioid? (figure 8?)

                                  _-_-

[PA0NVD]

The antenna has a back lobe that is approx 15 dB down and two side lobes at 135 degrees similar to a hex beam

[WBear2GCR]

So, which direction is the radiation??

Broadside to the driven element, or in line with the plane of the driven element??

Hard to visualize how the 90deg element can be a director/reflector... pix of the simulation would be very useful.

[PA0NVD]

Here is the bottom view

[WBear2GCR]

DOH!

Roger that.
Makes sense now...

Ok, what does it do when the elements IN the same plane are driven/connected??

[KD6VXI]

Wish I was better at modeling.....   I can do a model,  etc.  But when it comes to lag or lead current I get lost.

That said:  wonder what would happen if you could vary the phasing between the two cubes......?   I seem to recall a similar antenna used for RDF back when......

--Shane
KD6VXI

[PA0NVD]

Hello Shane
I can't model, but did some tests at the 330 Mhz model I made. When changing the phase of the reflector by the length of the stub, there is a clear optimum. I tried a phasing line between the driven lement and the reflector, but that never gave the result as good as the stub or coil, whatever lenght I tried. May be the impedance of the phasing line is the problem, I tried 75 Ohms and 100 Ohms. I tried as well a director configuration, but that gave 1 - 2 dB less gain though the FB ratio was similar. In this case thare is a difference between the simulation and the 330 MHz test. In the simulation the director configutation ia approx 1 dB better

[Steve - K4HX]

Could you show the antenna pattern plots as compared to a dipole at the same height above ground?

[PA0NVD]

WB6BYU did some modelling Steve, I can't. I made a 330MHz model to measure, so I can't sent plots