I am attempting to provide a solution to a problem relating to hazardous gas detection training, something firefighters often do. In a fire, a first responder will often carry a hazardous gas scanner, which indicates if there are toxic gases, flamable gases, and other hazards present in the air. Typically the scanner looks like a handheld gps unit.
The current problem is that training the people to use the scanners is difficult without actually having real threats present. Right now we "pretend" and someone walks behind us and says "ok now the O2 reading is XYZ. Now the reading is ABC." Basically it is very crude and not a good way to train people.
I am trying to develop a system the could utilize a transmitter and a signal strength meter to simulate scenarios for training purposes. I would like to be able to drop a portable "propane leak" transmitter, and the student would have a signal strength type meter that indicates how close he is to the source.
Max range of detection is 50ft. Should be able to be used indoors and outdoors. Ideally, the system would be simple and low power.
I have no idea where to start looking at hardware to do this. The problem seems simple enough, but oddly uncommon. I have looked into some XBee equipment, but frankly I am overwhelmed. Can anyone point me in a more senseable direction. All guidance and thoughts are very appreciated.
Well I don't know if there is a commercial field strength meter available that would meet your needs or not. However if I needed such a device I would build one using the Analog Devices AD8307 logarithmic RF amplifier chip. It's a simple 8 pin device that uses +5vdc power and outputs a simple 0-5vdc analog voltage that is a measurement value of the RF input voltage. It's a log scale output, meaning it can handle an incredible 90dbm power range. It's a little pricey at around $15, but requires little support components and is easy to interface to a analog display or micro.
Signal strength will be very patchy, because of interference patterns - basically the detector will see rapidly fluctuating readings as it moves or people move... These fluctuations will dwarf the 1/r^2 gradient that you want to pick-up on. Worse still the received strength will fluctuate (go through null-points) as you rotate the receiver.
You'll get the same problems with ultrasound for the same reason - you are using coherent radiation source.
You might be able to defeat this effect by clever means (rotating/vibrating transmitter or baffle? helical polarization? sub-millimeter wavelengths?), but it is going to be a major issue if you were expecting monotonically increasing readings with proximity to a radiation source.
Thanks so much for your help Lefty. This looks like the perfect "guts" of the system. I know this may be trivial, but there are some parts of how this works that I don't really understand.
If I am using the AD8307 in the receiver, how can I separate one transmitter frequency from another? (If I have to different "channels" indicative of different hazards, how can the AD8307 tell the difference between them?) Or would I need one AD8307 with some sort of band-pass filter involved for each transmitter frequency?
Also, on the transmitter end, is there a device I could build/buy that would output a signal that is receivable using the AD8307?
I appreciate your help with this a lot. I am very excited to get started on this project.
This variance that you are referring to is due to signal interference effects and antenna issues and things like that? I know that there is a similar effect in the actual gas detection scenarios due to drafts/wind/humidity etc. Some noise in the signal is ok, but I am hoping that I can find a way to get enough signal/noise ratio to still be able to see an increasing signal as I get closer to the source. I am not making any specific measurements from the reading just a (colder warmer HOT) type of readout. I see you mentioned some suggestions of how to mitigate these interference effects, but I am wondering if the problem wouldn't be able to be solved with some type of time-averaging. Could I just limit the readout to only change to reflect the 1 second average (or median) of the past signal response? Basically, could I filter the effects out and still see the proximity = strong signal response?
Thanks so much for you help. I am glad to see this kind of response from an online community. It is very encouraging.
How about some sort of scheme that uses RFID tags. I don't think you'd be able to get a range very well for reasons that Mark mentioned. But perhaps the tags could incorporate some sort of scenario. So instead of it just being - when detected then show CO2, it would be - when detected show CO2 rising from 20% to 70% over 2 minutes. Then you would have nice passive tags with no batteries to scatter around the training area
Just had an idea from a previous life - this technology already exists. It's called an avalanche transciever - used by snowboarders and skiers to find each other when one is buried under snow. It gives range and direction on modern devices and they also handle multiple buried devices perhaps these would offer a good place to research
The instructor walks around the place beforehand with a GPS enabled Arduino "hazmat-sensor", stops somewhere and selects CO leak. Then goes somewhere else and selects "Chlorine gas" (or whatever).
Then during training, the Arduino just reads GPS signals to get its position and calculates how far away from the "hazards" it is.
The only problem I can think of is GPS accuracy. I mean this would work great for a corporate campus, but not so good for a single lab. And also perhaps signal reception in basements and other places where these leaks would realistically occur?
Is there a better solution than GPS for smaller spaces, perhaps requiring that you plant transmitters in known locations?
If I am using the AD8307 in the receiver, how can I separate one transmitter frequency from another? (If I have to different "channels" indicative of different hazards, how can the AD8307 tell the difference between them?) Or would I need one AD8307 with some sort of band-pass filter involved for each transmitter frequency?
No field strength meter would be able to do that, they just measure the whole RF spectrum energy within it's bandwidth. What you are asking for could only be handled by some kind scanner type receiver that would be programmed for all the transmitter frequencies you plan to use and then measure the RSS signal strength on each channel it receives. Not a easy project by any means.
Hmmm. Ok, from the overall feeling I get from this conversation (Thanks so much to everyone for their feedback), this project is going to not be a simple "buy-snap together-turn on" type of operation. I am going to re-evaluate the need for this and decide if it is worth pursuing.
I have considered an alternative from rf that I would be interested in hearing your input on: Using non-audible frequency sound emmitted from speakers and picked up by microphones. Using some sort of filtering to distinguish between the different signals. Thoughts are appreciated!
It works pretty well, my nieces love it. The transmitter cannot be at all covered and reflections are not too bad -- in fact they are probably perfect for your app. For example you get a weak signal from the other side of a mostly-closed door.
The toy lights up LEDs based on the proximity, so it would probably be possible to literally use this toy in your project -- you could probably remove the leds and connect instead to Arduino IOs. Or you could take it apart to see how it works!
Thanks! I'm going to order a few and take them apart/ test them out. If nothing else, its a great starting point to look at what components are involved.
No problem! I'm looking for a cheap localization solution (localization is the opposite, so essentially the same -- its where are YOU relative to some fixed locations). So keep me updated on how this goes. Unfortunately I can't take apart my niece's toys without getting in trouble! I already have a bad enough reputation for that kind of thing :-)!
Unfortunately I can't take apart my niece's toys without getting in trouble! I already have a bad enough reputation for that kind of thing
lol, you too huh..
By the time I was 5 I'd pulled apart every radio and tape deck in the house, and who could resist pulling Barbie apart to see how her speech processor worked!