Does anyone know how these work?

Discussion in 'General Electronics Chat' started by marshallf3, May 12, 2011.

  1. marshallf3

    Thread Starter Well-Known Member

    Jul 26, 2010
    2,358
    201
    Those non-contact IR spot thermometers. I see a lot for sale but I need one with a very narrow detection beam angle spot, often I would like to know the temperature of a 4" pipe that's 20' off the ground and the common 8:1 models just pick up too much of the surfaces around the pipe.

    What's inside these things anyway? Wouldn't there be a way to combine whatever sensor they use with some good optics? I'm thinking maybe the sensor in a common PIR (turns lights on with movement) might be a potential sensor but I don't know how they work either, seems to me they just sense a change in the black body radiation in the area they're covering. I have a couple of spare sensors from those outdoor floodlight fixture to play with.

    I also understand that IR and common glass lenses don't get along very well?
    Or is that UV & common glass lenses?
     
  2. Markd77

    Senior Member

    Sep 7, 2009
    2,803
    594
    Some types of IR get through glass, otherwise it would be impossible to burn ants with a magnifying glass. I think it gets worse the shorter the wavelength is so maybe not too good for lower temperature measuring.
     
  3. mcgyvr

    AAC Fanatic!

    Oct 15, 2009
    4,769
    969
    The biggest issue/problem I have with these devices is the are highly dependent on the "emissivity" value to have entered.. If that is off so are your measurements.. And sometimes by quite a lot.
    Good luck finding one with a 60:1 field of view for cheap..
     
  4. someonesdad

    Senior Member

    Jul 7, 2009
    1,585
    141
    I suspect one of the reasons they have such large "spot sizes" is to keep a reasonably decent signal to noise. While it shouldn't be too hard to find a suitable lens that works in the IR, I'll bet you might have sensitivity problems.

    Why don't you mount the thing on a long pole and get the readings that way? You may have to use binoculars. :p Another question is why not mount a thermocouple on the pipe?

    mcgyvr's right about the emissivity problem, but there are some things you can do. Inorganic minerals and hydrocarbons (e.g., plastics and wood) often have emissivities in the 0.85 to 0.95 range -- with more probably in the 0.9 to 0.95 range. The problematic surfaces are uncoated metals, as they can range from 0.01 to 0.2 (IIRC, polished silver can get down to around 0.01). One tip (if you can use it) is to stick some vinyl or masking tape on a surface and meter off that to get your 0.95 emissivity. Paint and oxide layers (e.g., anodizing aluminum) are another way to get the emissivity up.

    If you'll be doing a bunch of repeated measurements, then the best thing is to use another temperature measurement (e.g., thermocouple or RTD) and calibrate the emissivities yourself. Note that emissivity can vary with wavelength, temperature, and angle off the normal.
     
  5. nigelwright7557

    Senior Member

    May 10, 2008
    487
    71
    I have one with a laser pointer on it and it isnt very accurate where it is pointing.
    I have to move the pointer around a little to get the hottest spot.
     
  6. strantor

    AAC Fanatic!

    Oct 3, 2010
    4,302
    1,988
    Yes, emissivity. If you need accurate and repeatble temperature measurement on a wide range of materials, don't use a noncontact thermometer. There's ony one place in our facility that uses noncontact and it is a calibratable unit mounted and focused on one thing that never moves, and calibrated to look only at that one thing. The handheld ones are not even allowed here because our calibration department cannot certify them due to emissivity. if you only need a rough estimate then fine, or if you are only looking for variations (i.e. you check the same pipe once per day in the same spot to see if the temp went up or down) then it may be acceptable for you.
     
  7. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    Seems you could use a high temperature tape with a decent emission signature to use for this. It should make it a bit more accurate. I tried one during my machine maintenance days, but was never happy with the results.
     
  8. marshallf3

    Thread Starter Well-Known Member

    Jul 26, 2010
    2,358
    201
    These sensors I have from some old PIR light sensors only have three leads coming from them, they're still on the circuit boards so a little reverse engineering may get one running to see what it outputs when looking closely at different temperature surfaces such as one of those flat stoves. I also have plenty of collimated laser diodes that could be used as the pointer.

    I just don't like trying to hunt down our 26' scissor lift and, if nobody's using it, going up 20' in the air to strap on my Fluke temp sensor adapter for one of my meters to it. I don't need an accurate measurement, +/- 5% is fine, but repeatability is the important part. These pipes run just a few feet below ceiling level and the temperature of the ceiling varies greatly.

    [EDIT:] http://en.wikipedia.org/wiki/PIR_sensor

    I still think this is what the cheap ones use but an alternative could be to buy a cheap one and change the optics but that would probably require changing the sensor to lens distance.
     
    Last edited: May 13, 2011
  9. strantor

    AAC Fanatic!

    Oct 3, 2010
    4,302
    1,988
    you could use the 26' scissor lift one time, go up and tape a thermocouple to the pipes, run the wire down a beam or wall, then just go plug your fluke into that every day.
     
  10. strantor

    AAC Fanatic!

    Oct 3, 2010
    4,302
    1,988
    here's another option:
    http://www.omega.com/ppt/pptsc.asp?ref=OM-CP-TC101A&Nav=dase01
    together with this software:
    http://www.omega.com/pptst/OM-CP-IFC110.html

    total 250$

    I just ordered this to monitor the temp of an oven stack so I haven't actually tried it yet. If it performs as stated, you could just run a USB cable down, or route an ethernet cable all the way to your desktop and you whouldn't even have to get off your rear to check the temperature. it exports to excel so you could keep a log of it. Set it to sample anywhere from 1/second to 1/day.
     
  11. tom66

    Senior Member

    May 9, 2009
    2,613
    214
    I had one, then I tried to stupidly measure the temperature of air coming out of my hot-air gun. I placed it too close to it and melted the plastic lens inside which threw all temp readings off. But it did give me a chance to have a look inside of a dirt cheap one. It looks to me like they are using a photodiode and have analog conditioning circuits as well as a small microcontroller (under an epoxy blob) driving an LCD. I bought a new one, luckily they are only £15 each but I learnt my lesson. Oh, and although mine is rated from -32 to +380°C, I was able to measure the temperature of a fire at +560°C, but I'm not sure how accurate that is.
     
  12. marshallf3

    Thread Starter Well-Known Member

    Jul 26, 2010
    2,358
    201
    Permanent sensor + wires are impractical, there would be some 100' runs.

    I noticed a lot of these PIR sensors have logic already in them and only output on or off, but the ones in the common floodlight fixtures must have a liner output or you could have the sensitivity control.
     
  13. strantor

    AAC Fanatic!

    Oct 3, 2010
    4,302
    1,988
    This is a follow-up on post #10. I just got the parts in and tried them out. I have attached the results. Not sure if this could work for you or not, or even if you ever found an alternative. You could set the time period longer (had it set @ 2 seconds, the minimum) and download the data once/week or however often you need it; or you could get the bluetooth model and wirelessly retrieve the data anytime.

    please inform if you can't see the document; I had to copy>paste from excel, then save in the old version of .doc to be able to upload it.
     
Loading...