Shop OBEX P1 Docs P2 Docs Learn Events
IR Sensor for my robot? — Parallax Forums

IR Sensor for my robot?

lenny1337lenny1337 Posts: 26
edited 2007-12-11 04:47 in Propeller 1
Im currently in the process of designing building and programming a propeller-powered firefighting robot for the expert division of the trinity college firefighting robot competition this april.

Can anyone reccomend a good IR sensor for distinguishing between essentially 3 different temperatures - ambient room, baby (bulb painted black, around 100F) and candle (400F ish?)

Thanks...
·

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Lenny Bogdanov
Systems Concept Center

Post Edited (lenny1337) : 12/9/2007 7:39:41 PM GMT
1280 x 960 - 823K

Comments

  • Paul BakerPaul Baker Posts: 6,351
    edited 2007-12-10 19:09
    The main problem you face is that there isn't any silver bullet (one sensor that will work for all situations), the reason is that the measurement is not only based on source temperature but distance. So it is possible for a distant flame to generate the same signature as a nearby human body if only a small portion of the electromagnetic band is measured. The way to overcome this is by using multiple sensors where each sensor is sensitive to a different portion of the electromagnetic spectrum. By using a photoresistor, an ordinary IR transistor and a Pyroelectric IR sensor you should be able to discriminate between flame, human and ambient. A photoresistor is used to measure ambient light, the reading of this measurement is used as a "baseline measurement" and is used to discriminate light sources which output large amounts of IR (halogens are an example of this type of light source). The standard IR transistor is your main flame detector, it will readily respond to an open flame but not to the blackbody radiation of a human. The Pyroelectric sensor (such as this one: http://www.glolab.com/pirparts/pirparts.html) will respond to both an open flame and human blackbody radiation. So by constructing a decision tree you should be able to determine what (if any) IR source is in front of the robot:

    Photoresistor | IR transistor | Pyroelectric |  Object detected
    ----------------------------------------------------------------
       LOW        |     LOW       |    LOW       |     None
       HIGH       |     LOW       |    LOW       |     Ordinary light
       HIGH       |     HIGH      |     ?        |     Light with high IR content
       LOW/MED    |     HIGH      |    HIGH      |     Open Flame
       LOW        |     LOW       |    HIGH      |     Human body
    

    What constitutes a LOW, MEDium or HIGH reading will have to be determined experimentally.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Paul Baker
    Propeller Applications Engineer

    Parallax, Inc.
  • GavinGavin Posts: 134
    edited 2007-12-11 00:54
    Lenny,

    UV sensors are normally used for flame detection but the price of Thermopile sensors are falling.
    http://www.melexis.com/Sensor_ICs_Infrared_and_Optical/Infrared/Digital_plug__play_infrared_thermometer_in_a_TO-can_615.aspx

    I have even seen cheap non contact temperature probes now sold in stores.
    Similar sensors are used in the ear temp thermometers.
    Pyroelectric sensors are AC, piezo based, so need a mechnical chopper and amps·to get different temp levels.

    Gavin
    ·
  • Paul BakerPaul Baker Posts: 6,351
    edited 2007-12-11 02:24
    I agree that the UV sensor is a superior flame detector, but since he still needs·to detect humans, he's still left looking for a means of detecting the ~10 um wavelength our bodies emit. The MLX90614 you linked to looks like a suitible alternative.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Paul Baker
    Propeller Applications Engineer

    Parallax, Inc.

    Post Edited (Paul Baker (Parallax)) : 12/11/2007 2:29:56 AM GMT
  • lenny1337lenny1337 Posts: 26
    edited 2007-12-11 04:47
    Thank you all very very much for the input. Excellent ideas.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Lenny Bogdanov
    Systems Concept Center
Sign In or Register to comment.