Infrared to detect heating or cooling losses from houses & other buildings

Discussion in 'Digital Photography' started by Ray, Aug 16, 2003.

  1. Ray

    Ray Guest

    does anyone know if there are digital cameras that are especially useful for
    getting images of either heating or cooling leaking from houses and other
    buildings? from what I hear the rates for heating fuel and electric are
    heading to the stratosphere later this year and now might be a good time to
    start plugging leaks...

    -- Ray
    Ray, Aug 16, 2003
    1. Advertisements

  2. Ray

    Bill Smith Guest

    (UK) fire service vehicles carry them as standard, to check for any
    remaining hot spots after a fire's extinguished. I'd *guess* they're
    specialist devices and probably out of the financial reach of ordinary
    punters for such a 'one-off' use, and I very much doubt if any consumer
    digital camera has such capabilities as the fire folks' cameras sense
    for heat, not light.
    Bill Smith, Aug 16, 2003
    1. Advertisements

  3. Ray

    Paul Cordes Guest

    It would require a really deep IR filter and a CCD that is cooled
    You are correct these are purpose built and expensive.
    For instance glass is opaque to long wave IR so that the lenses are often
    made of materials like sapphire or zinc selenide.
    Paul Cordes, Aug 17, 2003
  4. Ray

    Ray Guest

    well, if the outdoor temperature is 10 above (or -40 below here in Minnesota
    from time-to-time) the CCD would get down to that pretty quickly... keeping
    the batteries warm enough to keep the camera going would be a possible
    problem... but such an examination of a house or building probably wouldn't
    have to be up to the standards of deep-space astronomy... just needs to be
    *good enough* to tell you where your house or whatever is spewing out

    -- Ray
    Ray, Aug 17, 2003
  5. Ray

    Paul Cordes Guest

    Ok, let's say that the camera is cold. We'll use an external battery pack
    which we can keep warm inside your coat.
    Now all you need is a lens that passes longwave IR but not visible light and
    short wave IR. Those filters can be bought. Now if you use a pinhole as
    your lens you should be good to go. They make pinholes for the front of
    SLRs so a digital SLR should do the trick. Not much resolution there but
    it's a heck of a lot cheaper than a sapphire lens.
    One last problem suggests itself. I'm not sure how far down a typical CCD
    goes into the IR with that Bayer grid in front of it. Might want to look
    that up before you get to far into this science experiment.

    Paul Cordes, Aug 17, 2003
  6. Ray

    Don Stauffer Guest

    Yes, but they start at about ten grand. There are three types now,
    mechanically scanned, pyroelectric vidicons, and bolometer arrays. They
    are getting cheaper all the time, though since they cannot use silicon
    chips they will never get down to prices of consumer cameras. Big
    buyers are police and fire departments.

    BTW, only some of these are visible, many are analog. They are
    generically called 'thermal imagers'. Prices can run upwards of several
    hundred grand for really fancy ones used by military.
    Don Stauffer, Aug 17, 2003
  7. Ray

    Don Stauffer Guest

    No, even cooled a silicon chip will not go out far enough. You need at
    least four microns. Platinum silicide is only silicon-based chip to
    work, and it is MUCH more expensive than straight silicon CCD. PtSi
    still needs cooling.
    Don Stauffer, Aug 17, 2003
  8. Ray

    Tom Monego Guest

    Infrared covers a long spectrum, film infrared is short wavelength infrared
    while what you are looking for is a long wavelength thermal imager. Some house
    inspectors offer this service, our gs company was doing it for free one year,
    then the next year no one at the company knew anything about it or who to call
    to get the service.

    Tom Monego, Aug 17, 2003
  9. Yes, these are special products that have (liquid-nitrogen?) cooled
    sensors in them and are VERY expensive. Thousands of dollars.

    Basically you have somebody come out to your house and do the work.
    You don't do this yourself. These are industrial machines. They are
    used in many kinds of industries to measure temperatures of objects
    Michael Scarpitti, Aug 18, 2003
  10. Ray

    Ray Guest

    what's needed is a sensor that'll work in the 3 ~ 12 micron emission
    range... ferroelectric bolometers is what I've seen them called, and
    supposedly these things are getting cheap enough to be put in ATM
    stanchions... if they're getting that cheap, manufactured in appreciable
    volume, we ought to begin seeing cameras using them get cheaper... I wonder,
    however, if we really need that... My Sony TRV730 camcorder, used without
    its little red IR source, does a pretty fair job of picking up a 98 degree F
    face and distinguishing it from an 80 degree shirt and 70 degree walls - I'm
    wondering about this... I haven't gone outside in the present 90 degree
    weather to check my own house for heat loss for obvious reasons...

    -- Ray
    Ray, Aug 18, 2003
  11. Ray

    Scott Coutts Guest

    You're sure it's measuring temp differences rather than NIR reflectivity?

    Ray wrote:
    Scott Coutts, Aug 18, 2003
  12. You present it as a question, but it is a fact; the temperatures mentioned
    are Far Infrared, so the Near Infrared the camera is picking up is reflected
    light, not emitted.
    It would probably (depending on the camera's sensor) take a temperature of
    something like 600 Celsius (1112 F) or more to emit enough energy in the NIR
    band to register.

    Bart van der Wolf, Aug 18, 2003
  13. Ray

    Don Stauffer Guest

    Ferroelectric detectors are only one kind. Others are the bolometric
    arrays, Pt. Si. CCD chips, and various long wave arrays, usually
    HgCdTe. It is a holy grail to make a 2-D array of later, so most of
    later are mechanically scanned linear arrays. These are what military
    uses, although a few military systems for shorter range do use PtSi.
    PtSi and HgCdTe MUST be cryo cooled, ferroelectric, pyroelectric, and
    bolometer arrays do not HAVE to be, though they work better if they are

    FE and bolometer array sensors are in $10K range, PtSi cameras in $50K
    range, HgCdTe sensors $100K on up.

    And, to answer another question, yes these sensors do respond to changes
    in IR emissivity (1-reflectance) changes, but most subjects of interest
    have pretty high, relatively constant emissivity in far IR.
    Don Stauffer, Aug 18, 2003
  14. Ray

    Ray Guest

    boy, is there ever an opportunity here for the first folks to get affordable
    IR gear... think about home safety/fire prevention... or troubleshooting
    equipment... or home security... and probably scads of stuff I ain't even
    aware of.... those fellers at MIT, soon's they get through fiddling with
    reformable newsprint, could/should take this on (probably started on it
    already) ....

    -- Ray
    Ray, Aug 18, 2003
  15. Ray

    Ray Guest

    oh, that gave me a vision... hereabouts the houses are quite old (the core
    room in my house is the remnant of an 1856 log cabin - square cut logs
    pinned with wooden dowels - I found newspaper bits talking of Pres.
    Buchanan) and 2, sometimes 3 story, w steep pitched roofs to slough the snow
    off) ... I'd have to get myself a scissor-lift cherry-picker !!
    Ray, Aug 19, 2003
  16. Ray

    Don Stauffer Guest

    There is a SLOPE to the curve. IF you put enough energy into the eye,
    it can see quite a ways out. I was playing with an IR laser projector
    one day and accidentily looked into system and saw .693 very clearly.
    That is supposed to be beyond range, but if the curve is 20 or 30 db
    down, and you shine a bright enough laser into your eye, you will see
    it. Some people hit with 1.06 claim to have seen red light, though in
    this case it was so powerful that they did suffer damage. Fortunately
    in my case, I had no aftereffects.
    Don Stauffer, Aug 20, 2003
  17. Ray

    Jack Ferman Guest

    If your 0.693 is 6930 angstrom, then you should have seen it. The
    handbook arbitrary limit for red is 7000 angstrom.
    Jack Ferman, Aug 20, 2003
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.