Unfortunately, now I live in a jointly-owned building and using a Yagi antenna 2.5 metres long, that come out of the terrace, it's not so convenient and I think my neighbours could be scared. Moreover, after have seen the 14 elements Yagi antenna simulation with 4nec2 software, I had a wish to see if I find out a better features antenna compared to Yagi antenna.
As a matter of fact, Yagi antenna has several secondary lobes and frequently, with strong signals, it's hard to understand from where the signal come from, ever since secondary lobes capture the signal even if the antenna isn't aimed to the right direction. Besides, Yagi antenna has not a very good F/B. Yes, ok: -27 dB aren't so bad but secondary lobes range from -15 to -20 dB.
Then, if I can't go too much out of my terrace, why can't I develop the antenna in width dimensions, for example, like a broadside antenna? Moreover, if I do a wire mesh for reflector, maybe the antenna could be stealth.
Instinctively, I've always been attracted by double H antenna that I found at page 109 of "Le antenne riceventi e trasmittenti", a very nice Nuova Elettronica's book. Here it is:
Simulating the antenna with 4nec2, I played with all parametres: first with dipoles lenght, after with their relative distance, then with the distance between dipoles and the reflector, to end with reflector panel area, all of it in the purpose to obtain the perfect resonance and the best front to back ratio. Certainly this is NOT an easy to handle antenna: in fact reflector's dimensions are 1.8 metres by 1.6 metres, but this is like a parabolic dish and it fires the signal exactly where I want to and it hasn't secondary significant lobes, especially on the back and sideways, so this antenna could be excellent to unwanted signals.
Below, the double H antenna simulation with reflector for PMR446 band, with gain's graphs, with front to back ratio, VSWR curve, impedance and resistance curves, vertical and horizontal radiation patterns and tridimensional patterns:
Note that front to back ratio is -47.5 dB (8 signal points!), the only secondary lobe in the back side is -35.4 dB, the only secondary lobe in the front side is -16 dB.
The antenna's impedance is pure 73.6 ohm without reactance, so the antenna is perfectly in resonance in our narrow slice of frequencies, 446-446.1 MHz and even if transmitter doesn't see the expected 50 ohm, we'll have not to care about it, VSWR is under 1.5. What's the problem? If you look around, you'll find out worse things.
The distance between wires that connects the dipoles in phase/counterphase was fixed to 25 millimetres to stop the impedance line's to 386 ohm, accordingly the formula 276*log(D/d) where D = the distance between the wires centre (25 mm) and d = the wire's diametre (2 mm) and the logarithm is 10 based.
Below, the comparison between radiation pattern of the two antennas, the 14 elements optimized Yagi antenna (red line) and the double H antenna (blue line), we can see that gain's difference is only a little more than 1 dB, but front to back ratio is definitely the best for double H antenna:
Being this a balanced antenna too, I'll provide a balanced/unbalanced coupling with 1/4 wave cable.
Here you can find the .nec file of this simulation with 4nec2 software of double H antenna just analized.
And now, we only have to look for the material and do the job!