Strongly directional antenna / homing beacon

[badreplicant]

I'm looking to construct a homing beacon of sorts: a cheap 433MHz transmitter connected to some sort of very directional antenna such that you'll observe a large increase in signal strength when you're right on top of it.

The problem is that I don't know much about antennas. I've set up a few long-distance wifi links, so I'm well versed in antenna gain vs. credit card debt, but that's about it.

My understanding is that an antenna's job is to take all the energy that would normally radiate in all directions and focus it in one particular direction or plane. The better you can do this, and the more efficient your antenna is at radiating what you feed it, the more gain you have.

The problem is that high-gain antennas tend to be rather large, especially as frequency drops. So perhaps what I'm looking for isn't necessarily high *gain*, just high directivity: is it possible to have one without the other? An small antenna that may be very inefficient at radiating the power sent to it, but is nonetheless very directional?

I'm looking for something that, if facing straight up, could be homed in on to within a meter. Is that even possible? My intuition says to just put an antenna into a big grounded metal can, but layman's intuition and antenna design probably don't go well together.

Any ideas? Much appreciated...

[calexanian]

Lets start here.

http://www.i1wqrlinkradio.com/antype/ch13/chiave896.htm

We plunk in a 4 element beam design for 433mhz and we get a element spacing length of 4.79 inches per element making the total boom about 15 inches long. That seems manageable to me. As for your particular application i am a little unclear. You want to have a large signal drop off when you get a certain distance away? I suppose you could just point a directional antenna straight up as 433 would just go up and never bounce back, or very little of it will anyways. You will basically have a stubby figure 8 pattern with some weird lobes coming off of it at ground level and the signal will be quite low if thats what you are going for. There is a chance that 433 being a ISM band in certain regions, with some stipulations, you may find some ready made antennas floating around out there for very reasonable prices. I have used antenna factor in the past for just some basic 433 duck antennas and they seem fine. Also they have just little pcb helix antennas too. https://www.linxtechnologies.com/en/products/antennas/small-external

Again i am not sure if this is in the direction you are wanting to go in. Just taking a stab at it.

[Dago]

Antenna gain and directivity are directly related. Gain just takes the antennas efficiency into account. Basically you cannot have a high gain antenna with low directivity but you can have high directivity and low gain.

[pickle9000]

Google

fox hunt antenna

That is a good starting point.

[badreplicant]

I guess some more detail would be helpful: I'm looking to hand a quadcopter on a target by seeking out a radio beacon, which would be transmitting straight up into the sky. I'd like to be able to position the quad on top of the beacon to within a meter or so, so I'm looking for an antenna that radiates in as narrow a beam as possible. I'd aim it straight for the sky and the quad would float around until it found the signal, then home in on it.

So what I'm looking for is an antenna that radiates in as tight a beam as possible, at 433MHz, without being huge. The signal strength should decrease dramatically if you're not right on top of it.

Is it possible to take something like the Yagi antenna you've linked me to and enclose it such that all the energy leaking out the sides is just absorbed / grounded?

[Nuno_pt]

The more dirrectivity the antenna have, the more elements it will have and the bigger it will be, putting it inside something will detonate the pattern of the antenna, putting an antenna inside a can it's good on 2.4Ghz and above because of the wavelength, it's just a waveguide needed, no long boom or elements, can you put the beacon on 1.2Ghz or above?

[pickle9000]

Try a

cantenna

I guess you are planning on sending telemetry to the copter.   

[Nuno_pt]

An cantenna will not work, he is trying to make an antenna for 432MHz.

300.000/432MHz = 69.444444 cm, that's the wavelength for 432MHz, so for a yagi it's the minimum for the reflector, the only antennas that work inside a can it's from 2.4GHz and above, because it's measures it's just an waveguide, like on 10GHz.

[Rory]

You may want to investigate other methods than detecting signal strength.  Real world aviation has been doing this kind of thing for years. Research glide slope, DME and ILS for some ideas. You should be able to exploit one of these technologies.

On the other hand, are you completely dependent on RF or can it be a machine vision targeting system instead? You can get tons more resolution out of a camera than an antenna array.

[C]

For radio direction finding it is a lot easer to use the difference between two signals.
You could for example put two receivers on your copter and detect the difference between the two but you would be adding weight to the copter.

A better option would be to use two signals on the ground and let the copter detect the difference of the two signals. But even two signals will not work. To land the copter you need both X and Y with a hint of Z.

On the ground, take two omni antennas and put a vertical medal plate between the two antennas. When you are edge on to the plate the two signals will be equal strength. As you move from this, one will get stronger while the second gets weaker.
To be able to land you need at least three antennas. To home, the copter just flies to increase the signal it can hear. At some point the copter will start receiving a second signal. The flight path changes so that this signal is also increasing.  Again at some point the copter will start receiving the third signal. The copter just keeps going until all three signals are the same strength at which point the copter starts descending while keeping the signals the same strength.     
So your landing platform is setting on three medal plates with 120 degree difference between the plates.
A small gap between the three plates center would allow some of the signal of the shielded transmitter to leak through the gap. This could be used as a getting close to over the center. The size of the shields especially the hight aspect will make this work better.

You do not have to transmit from all three antenna's at the same time just rotate through the three antennas. This actually makes it easer to process on the copter.
Remember for this you need to allow time for the copter end to adjust to the different signals if you are just using it's receive strength level for this.

http://en.wikipedia.org/wiki/Corner_reflector

For 3 antennas you want a 120 degree corner. If you decide that 4 antennas is easer then 90 degree corner.
Note that even if you have very good corner reflector antennas, at a close distance there will be some signal off the side. If this is still to strong a signal to center the copter on the platform then some space between the reflectors and more hight could help.

C

[Frant]

I guess some more detail would be helpful: I'm looking to hand a quadcopter on a target by seeking out a radio beacon, which would be transmitting straight up into the sky. I'd like to be able to position the quad on top of the beacon to within a meter or so, so I'm looking for an antenna that radiates in as narrow a beam as possible. I'd aim it straight for the sky and the quad would float around until it found the signal, then home in on it.

So what I'm looking for is an antenna that radiates in as tight a beam as possible, at 433MHz, without being huge. The signal strength should decrease dramatically if you're not right on top of it.

The frequency you have chosen is too low for such a system to be practical.

How about a laser? The light emitted from the laser can be modulated in order to send the location ID or other information to the vehicle.

[pickle9000]

Maybe a beer cantenna

http://fpvlab.com/forums/showthread.php?10584-Beer-CanTenna-A-ground-antenna-for-433MHz-TX

[pickle9000]

Sorry I should have read that first, here is a much more informative link.

http://www.electro-tech-online.com/threads/433mhz-directional-antenna.106477/

[C]

You may want to investigate other methods than detecting signal strength.  Real world aviation has been doing this kind of thing for years. Research glide slope, DME and ILS for some ideas. You should be able to exploit one of these technologies.

433 Mhz is a whole lot higher then some of the frequencies air planes use.
So not so low.
All good sound ideas if you can do what is needed and stay in their limits.
just adapting above and what I know to OP's problem.

As a guess a 433Mhz receiver will have a poor RSI (receive strength indicator). I would not expect it to be so bad that a copter flying in a circular pattern could not guess what direction to go by using just RSI. But this circle will most likely get larger the greater the distance from the antennas. A reflector for this Freq will not be huge. Also you are not talking a large cost for the reflector behind an omni antenna. It also may turn out to be easer and cheaper to just use three transceivers on the ground instead of switching three antennas to one transceiver. It is also very easy to test before adding two more antennas. Keep in mind that the reflectors are shooting the signal out in the horizontal direction not UP. And the up side of the reflectors can also have a shield added to make UP signal even weaker. So as the copter gets closer at the same hight the signal will get weaker. Behind the reflector should be the lowest RSI indication for an antenna, granted that a directional antenna has many strong side lobes. The net result of trying to get the RSI from three antennas to match should only happen in one area and that is above the edge of all three reflectors. The radio signal would look somewhat like three different color balloons with the fill valves all tied to the ground at the same spot when the back of the reflectors touch for the three antennas. The result is the copter landing in an area where a small horizontal move makes a large difference in signal strength.
 
The copter just needs to track how the RSI changes with flight direction and does not really need to know what antenna it is.

If you have problems thinking in three antennas then the pattern for four will act somewhat like X & Y for direction.

I would be very surprised if this did not work well by it's self to do the job. But if more precision is needed at the final stages, the radio antenna shields would also be a good IR shield so you could have a rotation of three IR leds and a simple IR receive chip on the copter for a precision final landing with the radio system getting the copter in range of this.         

C

[badreplicant]

You may want to investigate other methods than detecting signal strength.  Real world aviation has been doing this kind of thing for years. Research glide slope, DME and ILS for some ideas. You should be able to exploit one of these technologies.

On the other hand, are you completely dependent on RF or can it be a machine vision targeting system instead? You can get tons more resolution out of a camera than an antenna array.

Thanks, I'll look into that. It has to be RF, because the beacon is actually going to be buried

[badreplicant]

I guess some more detail would be helpful: I'm looking to hand a quadcopter on a target by seeking out a radio beacon, which would be transmitting straight up into the sky. I'd like to be able to position the quad on top of the beacon to within a meter or so, so I'm looking for an antenna that radiates in as narrow a beam as possible. I'd aim it straight for the sky and the quad would float around until it found the signal, then home in on it.

So what I'm looking for is an antenna that radiates in as tight a beam as possible, at 433MHz, without being huge. The signal strength should decrease dramatically if you're not right on top of it.

The frequency you have chosen is too low for such a system to be practical.

How about a laser? The light emitted from the laser can be modulated in order to send the location ID or other information to the vehicle.

I'm looking for a compromise: something at low enough frequency that it will penetrate through the ground or light tree cover, but high enough frequency that the antenna doesn't have to be enormous. This means that light, or even 2.4GHz, isn't going to work very well.

[C]

Thanks, I'll look into that. It has to be RF, because the beacon is actually going to be buried

A antenna under ground is a very different thing from an antenna mounted on the ground.
Even an antenna just laying on the ground can kill a transmitter.
http://en.wikipedia.org/wiki/Voltage_standing_wave_ratio

The electronic dog fences that I know of which use an underground wire work at a below am radio band in the usa which is below 540Khz and the receiver works in a range of a few feet.
I think there could be higher freq used but I would guess that such a system has to be built special with a very good protection system to keep working. I think I would be very safe in saying this is not a 443mhz transmitter can buy off the shelf. If the omni was sticking up from the ground then it is a different story as you have a antenna on a ground plane.
And there are rules for what frequencies you can use for radio and big fines if the radio is transmitting out of a allowed band or with to much power.

C