Shop OBEX P1 Docs P2 Docs Learn Events
Protective shield for IR temp sensor — Parallax Forums

Protective shield for IR temp sensor

Brian SmithBrian Smith Posts: 44
edited 2005-07-02 04:38 in General Discussion
I was wondering if placing a clear plastic lexan shield infront of an IR temp sensor (such as the Melexis MX90601 style sensor)·would mess up my temperature readings. I could use clear tape if this would work better. I need to protect the sensor from small flying rocks/road debris. The sensor is encased inside a steel tube except for the exposed lense.

Comments

  • steve_bsteve_b Posts: 1,563
    edited 2005-06-24 18:53
    hmmm....not sure. there would still be defraction....but the IR element itself is housed behind a lens/shield of its own.

    How about some small screening with the IR laser shooting thru a large enough hole. It will protect against 'MOST' debris....but like anything, it's not totally protected (some black boxes are unrecoverable too!)

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    ·

    Steve
    http://ca.geocities.com/steve.brady@rogers.com/index.html
    "Inside each and every one of us is our one, true authentic swing. Something we was born with. Something that's ours and ours alone. Something that can't be learned... something that's got to be remembered."
  • edited 2005-06-24 20:06
    Brian,

    I agree with Steve. Lexan is treated to reflect harmful UV rays, this is bound to have an effect on IR emissions. I have seen sumo bots with their frames constructed out of Lexan. On each of every example I've seen, the IR detectors are either peeking out of an exposed slot, or are mounted outside of the frame. I'd try Steve's idea or maybe instead of screening, you might want to look at using an optical filter that will allow IR emissions.


    Regards,
    Michael G. Jessat
  • Paul BakerPaul Baker Posts: 6,351
    edited 2005-06-24 20:34
    How about taking a peice of lexan or acrylic, drilling a hole large enough for the sensor element to be placed within and mounting so that it is recessed within the hole. This way debris striking the module will contact the acrylic instead, only debris small enough to fit within the hole and which actually strikes exactly inside would be an issue, this should help protect the sensor from the vast majority of debris.

    90601_300.jpg
    For this product I would first drill a hole the diameter of the sensor window, then drill the same hole with a bit that is the outer diameter of the sensor to a depth equal to the height of the sensor and mount it inside.
  • steve_bsteve_b Posts: 1,563
    edited 2005-06-24 22:17
    what's the image of Paul?

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    ·

    Steve
    http://ca.geocities.com/steve.brady@rogers.com/index.html
    "Inside each and every one of us is our one, true authentic swing. Something we was born with. Something that's ours and ours alone. Something that can't be learned... something that's got to be remembered."
  • Paul BakerPaul Baker Posts: 6,351
    edited 2005-06-25 00:26
    The Melexis MX90601, the IR temp sensor he's using.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
  • DaveGDaveG Posts: 84
    edited 2005-06-25 14:47
    Brian,
    The Melexis MX90601 is designed to detect infrared photons with a wavelength of 7.5 to 13.5 microns(long wave IR).
    Photons in this band are absorbed by glass so you cannot use a glass window to protect your sensor.
    However, a number of plastics such as polyethylene and the acrylates(Lucite, Plexiglas) are fairly transparent
    to long wave IR and you could try these.
    I would suggest setting up your sensor and placing various materials in front of it to see how much the signal
    from a warm emitter(like your hand) is attenuated. You might try the following materials:
    1. A clear plastic freezer bag.
    2. The bottom of a plastic cup.
    3. Thin Lucite, 1/8" max.
    4. Thin Lexan
    5. A thin piece of silicon (the kind that ICs are made of)
    6. A piece of window glass (this should block the signal)

    Dave G
  • Brian SmithBrian Smith Posts: 44
    edited 2005-06-27 16:20
    Thanks. I will try different types of plastic and see what works best once I get it set up and running.·

    I could also use some help getting the code right to read this sensor. I plan on using the digital output mlx90601-daa and im not exactly sure how to read the duty cycle or error mesages for that matter.



    ' {$STAMP BS2}
    ' {$PBASIC 2.5}

    timeCounter······· VAR· Word·············· 'Declare variable to store time.

    DO··············································· ·'begins main loop
    ··· timeCounter = 0························· · 'Set timeCounter to zero.

    ·PAUSE 13
    ··················································· ·'omits t1 from dutycycle calc
    ·DO
    ·· timeCounter = timeCounter + 1
    ·LOOP UNTIL IN3 = 0

    · DO
    ··· DEBUG "temp is ", DEC (timeCounter),
    ········· " deg.", CR,CR
    ········· DEBUG HOME
    · LOOP UNTIL IN3 = 1
    LOOP




    this is a very basic start and it wouldnt give the propper temperature output or tell me about errors..·Is there a better way? Melexis says that the value for T can varry and will not always be 102.4 ms. What do I need to do about that?
  • Jon WilliamsJon Williams Posts: 6,491
    edited 2005-06-27 17:44
    If you link the datasheet(s) to a post if it will be easier for others to help. I'm quite sure the reason you're having troubles with the loop code is that you have no timing control in it.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Jon Williams
    Applications Engineer, Parallax
  • Brian SmithBrian Smith Posts: 44
    edited 2005-06-27 18:11
    hows this? http://www.melexis.com/prodfiles/MLX90601_1.PDF

    Post Edited (Brian Smith) : 6/27/2005 8:27:00 PM GMT
  • edited 2005-06-27 22:45
    Brian,

    Which version of the MLX90601 do you have? It states that several versions exist. Looks like some have analog output, PWM output or an SPI interface. That might point us in the right direction.

    Regards,
    Michael G. Jessat
  • Brian SmithBrian Smith Posts: 44
    edited 2005-06-28 01:46
    Oh Yeah that would help smilewinkgrin.gif ....

    its the MLX90601EZA-DAA with PWM output
  • Brian SmithBrian Smith Posts: 44
    edited 2005-07-02 03:31
    well it does work... kinda. timecounter will change from 1 to about 40 depending on what I point it at in the room. It takes a glass of ice to get it to 1. Anyways I cant get the stamp to go through the loop fast enough to get decent resolution. And the code is still incorect to adjust in changes in T and it dosent recognize errors (i can do without that though). I also dont know how fast the loop is going so calibrating it will be difficult. Im sure theres a better way...

    So what is time controll and how can I use it?

    I havent tried any dikfferent shielding for the lens yet.
  • DaveGDaveG Posts: 84
    edited 2005-07-02 04:38
    Brian,

    The MX90601 uses a thermopile detector, which has a very slow response time, 1/2 sec. When you do any
    calibration, you should let it stare at the target for at least 2.5 seconds(5 time constants) before reading the output.

    Dave G
Sign In or Register to comment.