Bests & Worsts Reviews from Amazon

With emergency products like CO detectors, it's hard to know if they really work unless you're actually being exposed to dangerous amounts of CO. And if a CO detector doesn't work, you wouldn't know; instead, you'd just be poisoned while thinking that nothing was wrong. The only way to really find out if it works is to put it in an airtight container and generate CO. Because I like to play with dangerous chemicals at home, I decided to try this out. Adding formic acid to concentrated sulfuric acid causes the formic acid (HCOOH) to break down into CO and H2O. I got both of these chemicals for less than $10/liter apiece, plus shipping. I put some sulfuric acid into a graduated cylinder and placed it in a 1-gallon airtight plastic container. I then added small amounts of diluted formic acid and rapidly closed the container. Sure enough, it worked! It registered CO concentrations that were well within the ballpark of what I'd expect to get with the amount of formic acid I added (using the ideal gas law to figure out what the maximum CO level would be). In high concentration tests, I found that the alarm goes off within 3 minutes of hitting 500 ppm (which is dangerous but would take several hours to kill, giving plenty of time to get to fresh air). In another test, it went off after about 20 minutes of CO levels around 190 ppm, which is dangerous but not lethal. Its peak reading is in the 910-920 ppm range; it doesn't display levels higher than that. Needless to say, it goes off within a minute or two of such high CO levels and anybody who sees 900+ ppm on the display should leave their house and seek medical help immediately. It worked well in low concentration tests too. Due to some strange US regulation, it displays "0" for any reading below 30 ppm. If you press the "peak level" button, though, it will tell you what its peak reading was even if it was below 30 ppm. This works down to the 10-15 ppm range, below which you don't really have to worry about CO at all. Prolonged exposure around 50 ppm leads to an alarm after a couple of hours. Another, simpler test (which anyone with a CO detector should do to make sure it's working) is to light a candle in an enclosed container with the CO detector, then close the container. The candle will burn through the available oxygen and a fair amount of incomplete combustion (releasing CO instead of CO2) will occur as it runs out of O2. I tried this and got 96 ppm with a small candle in my gallon container; a larger candle would probably release more. I then tried it by lighting a crumpled piece of newspaper in the container and got >910 ppm, which is reasonable because paper and wood experience pyrolysis, which releases lots of CO. Combustion of small amounts of acetone (a fairly clean fuel relative to paper) resulted in 300-550 ppm depending on how hot the fire got before I put the lid on; the lower reading relative to paper helps confirm that it's reading accurately. In conclusion - this $20 detector works very well when tested with real CO. It's accurate, the alarm goes off at dangerous CO levels, it only goes off if CO levels are dangerous, and it might just save your life.