Gas Sensor w/MQ-7B

Hello All,
I just received 1 PCB & 2 Sensor (new) 2 days ago.
Thought it was going to be very easy. Almost. I looped the DEMO CODE into my application and tweaked Phase 1 & 2 so they function within the application. But the detection is NOT working when I spray the "Mag-Torch Butane" on the Sensor.

I switched back to the original DEMO CODE and the same thing happens.
NOTE: I can adjust the LED resistor and it will turn on & off and the program will detect and clear as required. I have not tried the 2nd sensor in the PCB yet.
My guess is either the (one) sensor is bad or the Butane is not the proper kind. There is no full break down of contents on the small container that I got from AutoZone.
Does anyone have any insight and/or the exact name brand of Butane to purchase with location so that I can properly test this sensor?

Thanks for any help and feed back.

Comments

  • Tracy AllenTracy Allen Posts: 6,405
    edited 2016-10-31 - 21:17:36
    "But the detection is NOT working when I spray the "Mag-Torch Butane" on the Sensor. ...snip... There is no full break down of contents on the small container that I got from AutoZone.

    Not sure what you're doing there. The CO would come from burning the butane in the torch or dryer, not from a direct unlit spray. For best results burn it a bit starved for air.

    You can get some CO from any burning that is oxygen limited. A cigar or an incense stick or a candle in a bell jar, say, but the amount will be ppm and it will be swamped with volumes of CO2, H2O and particulates. The sensor will have cross-sensitivity to those, the H2O especially. There are CO detector test sprays that are better, for example, COcheck. (~25ppm CO in argon carrier gas).
  • OH!!!
    Thanks Tracy. I just thought that there was some detectable quantity in a can without a fire or sticking it directly on a running cars tail pipe.

    Thanks again.
    Just ordered a can from Zoro ($8.00/ $15.00 total)

  • The sensor does have cross-sensitivity to other gases, as you can see in the data sheet found on the Parallax product page for the MQ-7B.

    The response to hydrogen gas H2 is as strong as the response to CO. So you could test the sensor for operation by electrolyzing water and passing the resulting H2 over the sensor. The data sheet also shows a weak response to methane (CH4) and to LPG (propane C3H8, plus some butane C4H10). The butane you used from might have given a small analog response, but maybe not enough to trigger an alarm.

    The Parallax breakout has a couple of test points. TP1 is the analog output of the sensor. Monitor that with a voltmeter as you apply the gas. If you want to measure CO concentration, route that TP1 to an analog to digital converter. The only way to check CO response directly is by use of a pure CO source like the COcheck can.

    Usually home protection CO alarms are set to trigger at around 70ppm CO.

  • UPDATE:
    Started thinking about different test setups and the 1st is a large Ziplock bag with "Smooth Sheen" (SS) hair spray that has (...Butane & Propane..lots of other stuff) in a pressurized can. I added a 10 min. count down (2 nested for/next) routines and started the test all over. I set the voltages again then placed the Unit Under Test (UUT) in the bag. No errors. I then sprayed the (SS) in the bag and the Sensor went off. I've repeated this about 20 times and it seems to be working. I also cleared the Sensor by blowing thru a regular drinking straw and that appears to be working too.
    Will be waiting for the CoTester to arrive in a couple of days. Will set up another test jig (small cardboard box) next time for further testing.
    Then the rear tail pipe of the car.
    The bag seems to hold the fumes long enough for repeated test.
    Thanks again Tracy.
  • UPDATE:
    Hello Tracy/All,
    COcheck can arrived today. Got home late and had other things to do first. Will re-calibrate and test with 2nd CO unit and the original set (negative) unit. Also I want a different Test Fixture (Box).
    Will also tweak code and move back to my application.

  • Tracy AllenTracy Allen Posts: 6,405
    edited 2016-11-03 - 06:18:59
    FYI, there is more information about COcheck and how to use it with a cap or ziplock at this site:
    testproducts.com
    Or the manufacturer hsifiresafety
    Look on the page for the MSDS, which is where I found its composition. I should have said 0.25% CO, not 25ppm--a lot different. Use a voltmeter on the test point T1.
  • UPDATE:
    COtester
    Did some testing and at first things seem to be going OK. But then I started to get a LOT of false positives. I switched to the DEMO code with a 10 min. pre-clean air purge and the false positives got worse. Finally had to shut it down and go to bed.
    NOTE: I'm using my own PCB with the Parallax 48 pin 32 I/O MC. I don't know if this could make a difference in code response.
    I did double the code "HSW" ON/OFF time to see if that made a difference and it did not.
    Next step:
    Will hook up with a BASIC Stamp and try again...Stumped for now.
    Thanks for the extra info Tracy. Also voltages are all set correctly.
Sign In or Register to comment.