Cheap Direction Finding

Hi,

I'm working on a project which involves emulating an avalanche rescue beacon. These are essentially radio transceivers with basic direction finding (more info in the link).

Unfortunately they're much more expensive than I'm able to justify, so I've been looking into cheaper, alternative methods. This thread has some interesting ideas in, but everyone seems to focus either on using IR emitters and sensors, which wouldn't work if the transmitter is buried under snow, or some sort of directional radio antenna, which seems pretty tough to build.

I'm wondering what my best options are for a low cost approximation of a direction finding beacon that I can implement easily with Arduino? Thanks in advance.

This general subject has come up in several Forum Threads without any obvious solution emerging. I guess that's why the commercial system can get away with the price charged for it.

...R

Some of the LoRa radios have ranging capabilities. In a relatively benign environment the blogger in the link below got resolution to a few meters with these radios at short range and useful operation at many kilometers. How they might perform buried in the snow is a possible downfall, but the technology might be worth exploring for this sort of application.

This of course isn't direction finding per se, but with ranging information from multiple receiver locations, geolocation should be possible. Alternatively a searcher would simply proceed along a path that reduces range to the transmitter.

https://www.loratracker.uk/semtech-sx1280-2-4ghz-lora-ranging-tranceivers-2/

A long long time ago, and long before LoRa came along, I produced a very simple locator beacon using an RFM22B (Si4432) @ 100mW that emmited a series of decending power tones at 434Mhz. The tones were easy to hear at long distance with a simple very cheap Baofeng UHF tranciever.

Because the tones decended in power and frequency you could tell by listening when you were getting closer to the beacon. A UHF yagi (easy to build) helped you to get a direction fix, but you could use body fade RDF (look it up) and get a reasonable direction fix with no additional antenna.

For really long distance RDF one method that works extremly well, again with the RFM22B as the transmitter, is to have the RFM22B send out a plain carrier. The cheap Baofeng UHF tranciever, which is an FM receiver, gets confused by the carrier and instead of the background static, the sound from the Baofeng goes quiet. By moving a directional antenna around for minimum noise level, which is easy, you can get an accurate direction fix. I used this technique to get a very accurate direction on a 'beacon' from my back garden, with a 7 element portable yagi, the 'beacon' was 1,200km away at the time.

All of the important work goes into making the transceiver. The Arduino really has nothing to do with it.

With a single transistor, a ferrite loop antenna and 4-5 other minor components, you can make a 457 kHz transmitter, but the receiver is quite a bit more difficult.

Unfortunately they're much more expensive than I'm able to justify

The manufacturer can certainly justify it, as a lot of work goes into designing and making reliable equipment, with relatively few sales to be made. If you ski avalanche-prone slopes, you can decide for yourself how much your life is worth.

What is it about the beacons you wish to simulate?

If you are training people how to use the beacons then there are 2 major elements of the simulation: what it feels like in your hand and how the radio performs.

Both of these are relatively difficult. Here "relative" means "compared to just buying a beacon."

sdunnim:
Hi,

I'm working on a project which involves emulating an avalanche rescue beacon.

Unfortunately they're much more expensive than I'm able to justify, so I've been looking into cheaper, alternative methods.

By all means research and investigate.

However do appreciate that its extremly unlikely that you will come up with a satisfactory and reliable alternative thats a lot cheaper.

If there really was a reliable way to do a much cheaper beacon, you would be able to buy it already.

Additionaly think about the insurance implications, you develop a cunning wizard super cheap beacon, give them to your mates, the worst happens and they dont work.

Hi all, thanks for the helpful replies - over the next few days I’ll look into LoRa, body fade RDF etc.

Just to clarify, I don’t intend to use this to actually locate people in the snow, just to emulate the function for a project I’m working on - so no lives at risk!

MorganS:
What is it about the beacons you wish to simulate?

The main function I’m after is getting a rough direction to the transmitter - i.e. something that can tell an Arduino ‘it’s in that direction’, ideally with only a single receiver and little to no movement. Although, after digging around and reading these replies, it seems that’s no easyfeat.

I found this thread online going into detail on designing a transceiver from scratch, but if I’m honest a lot of their discussion is over my head, and I had been hoping for a somewhat simpler solution

Thanks again

it seems that's no easyfeat.

Actually, it is pretty easy if you have a decent receiver. All you need is a directional antenna, like a ferrite rod loop antenna.

The antenna tells you the direction to move, and you move in that direction as the signal becomes stronger.

You can model the entire operation with an old fashioned AM band transistor radio (which has a built in, directional ferrite loop antenna) and a one-transistor AM band transmitter.

Or, make the transmitter using an Arduino, as demonstrated by forum member Nick Gammon.

sdunnim:
over the next few days I'll look into LoRa, body fade RDF etc.

As for the LoRa part I am the guilty party for working on the ranging capability of the SX1280 2.4Ghz LoRa device, and publishing the Arduino library.

In LOS terms I have had the ranging measuring distances accuratly at 89km, 1/4 vertical at the remote end, cheap WiFi yagi at the receiver end.

I have tried the SX1280 ranging for an Easter Egg hunt over the local park which is a mixture of open playing fields hedges and trees. With the remote device on the ground hidden in the trees I was getting 'only' circa 250m on simple 1/4wave vertical rubber duck type antennas.

Using the SX1280 ranging would be a interesting way of doing this sort of location and I dont think there is another device out there that has the distance measuring capabilty at this cost, £5 per module. Whether the range would be good enough, not sure.

Foxhunting, eh?

With a strong signal, a good method is Doppler-scanning using a commutated antenna array and an FM receiver.

I have not had the time to develop this, but a hybrid method using a simple two-antenna (whip) array and manual (mechanical) scanning sounds very promising.

Paul__B:
I have not had the time to develop this, but a hybrid method using a simple two-antenna (whip) array and manual (mechanical) scanning sounds very promising.

I have tried this;

http://theleggios.net/wb2hol/projects/rdf/tdoa2.htm

But whilst it works the ever so simple decending power tones method worked better, lots simpler too.

sdunnim:
or some sort of directional radio antenna, which seems pretty tough to build.

Really? Have you actually tried? A simple dipole in front of a groundplane can't be very difficult, or a 3-element Yagi or even a short helical. You don't want too much directionality or you'll have trouble finding the signal if its weak.

Wavelength is everything with antennas though, so that must be thought through.

srnet:
http://theleggios.net/wb2hol/projects/rdf/tdoa2.htm
But whilst it works the ever so simple decending power tones method worked better, lots simpler too.

Yes, that's the commutation system, but my intent was to use a Doppler discriminator to indicate direction. :grinning: