Make:

Most gas sensors have a small heater inside of them. Once the unit is warm enough, it can start sensing particles. It looks like it is just a variable resistor that varies according to the amount of gas (kinda like a light sensor that requires a heater to operate)

Lets look at the cheap alcohol sensor:
you need 5v, +-.1 volt, connected through H and connect the other h to ground (polarity doesn't matter). This is 5v/33 ohms, so .152 amps of current will flow through this.
Connect A to +5, and B to a 200Kohm resistor that is connected to ground. (could use a potentiometer to change sensitivities or bias.)

Connect the AtoD on the arduino to B.

Now the voltage will vary according to the amount of alcohol present at the gas sensor.

Reply to this comment

I have a couple at home. The problem I ran into when using them is how to calibrate them. I was hoping to measure the levels of carbon monoxide in my apartment since I live next to a big street, but the challenge is to find a calibration reference that I can compare to. I think that's the toughest part of using these sensors. If anyone has a good idea on how to do the calibration, lemme know.

Akiba
FreakLabs, Open Source Zigbee Project
http://www.freaklabs.org

Reply to this comment

Note that the CO2 sensor is actually a bit different, outputting a voltage between A and B. It's in the range of mV, so you'd probably want to run it through an op-amp to ramp it up to the 0-5V range.

Also note that the CO2 sensor requires a 6V +- 0.1V heating voltage and pulls a bit more power than the Alcohol one, so a separate power supply for the heater in the sensor would probably be a good idea.

Reply to this comment

For what it's worth, I reverse-engineered an old commercial sensor a while back:

http://www.neufeld.newton.ks.us/electronics/?p=184

It had an adjustable reference, a power-on delay to allow the heater time to warm up, and a missing sensor detector (if the sensor gets unplugged or knocked loose). It also has another part of the circuit I was able to analyze but whose function I wasn't able to determine. :-)

It might give some ideas about the feature set one would want to build in, especially if using these for life- or property-safety.

Reply to this comment

It actually is quite simple to get it working with an arduino. Here's a translated project where someone did exactly that with one of the gas alcohol sensors. http://translate.google.com/translate?prev=_t&hl=en&ie=UTF-8&u=http%3A%2F%2Flusorobotica.com%2Findex.php%2Ftopic%2C111.0.html&sl=pt&tl=en&history_state0=

Reply to this comment

Posted by: zeni on April 13, 2009 at 6:14 PM

I'm using five of these for a noise performance. I used pots to have the same bias. The first comment explains it pretty much.
The nicest thing about the MQ-3 (available on sparkfun) is that it's quite responsive for using live.

http://www.cho-yaba.com/works

Reply to this comment

i remember seeing these a while ago and thought it would be a cool idea to somehow interface these with a synth/computer/some sort of sound making device. Have the sensor taped to a mic and that way as one drank during a performance the sound would change.

Pure Data generative alcohol music?
Atari Puke Console?

Does the CO2 content of one's breath vary much? I suppose that could be a teetotaler option for a sound controller.

Reply to this comment

Here are some links with info on the MQ-3 alcohol sensor from the sparkfun forums. There are a few gotchas, especially as far as current requirements for the heater are concerned:

http://forum.sparkfun.com/viewtopic.php?t=13784
http://forum.sparkfun.com/viewtopic.php?t=13064

Reply to this comment

Posted by: Mr_Alb on April 15, 2009 at 3:44 AM

Its in "Making Things Talk" - "Reporting Toxic Chemicals in the Shop" - and from the look of things its pretty easy to interface.

Reply to this comment