No announcement yet.

Accuracy Of Infrared Thermometers?

  • Filter
  • Time
  • Show
Clear All
new posts

  • Accuracy Of Infrared Thermometers?

    I'm considering buying on of those non-contact infrared thermometers, but am curious about there accuracy on different surfaces. I am probably looking for one that does about 0 to 900°.

    My biggest concern is the effect of the surface being measured on the accuracy. Is color anodized aluminum a problem? Rusty cast iron block? How about a highly reflective surface like chrome? Or trying to check the temperature of something clear, like the area near the seals on a transmitting tube? How far away can they measure?

    Another question is ruggedness. Can they be damaged by pointing at something too hot? Are there some brands to avoid or certain features that point to a lousy unit?

    Last edited by The Doctor; 02-28-2007, 02:38 AM.

  • #2
    Can't say I know a huge amount about these. Regards accuracy I suppose usual case of as accurate as you pay for. I use one of these to measure exhaust manifold temps when setting up carburettors. In this case I'm not concerned about absolute values just comparative. The surface of the material is important as this controls the emmisivity of the body. Forexample on my headers I wil spray a patch of black heat resistant paint and take my measurements off that. If I move to a polished area of the header (Stainless steel) I will see different values. I'd like to investigate the use of these for heat treating steel, up till now my heat treating has been rough and ready blacksmith end of the trade.


    • #3
      My experience with these has not given accurate results on steel, but then my application may have been different to what you might need.

      We used one of these a few years in our injection moulding business when we were develoing a product moulded in Polyphthlamide where we needed tool surface temperatures of 160 deg C+ , and we needed to both accurately measure the tool temps and the core temp of the moulded parts that came off the tool. Our normal type K pyrometers just weren't fast enough so we thought these would be a good idea.

      No problem at all on the moulded black plastic parts - super fast and accurate readings. However every time we needed to measure off the tool surface we get massively varying readings or none at all, and it seems that it did not like reflective surfaces at all.



      • #4
        Measured temperature is proportional to wavelength, not emissivity. If the instrument is measuring wavelength then it should still read within the limits of it's accuracy regardless of surface finish as long as the signal is strong enough and not affected by spurious sources. A black surface will provide a stronger signal as long as it is black at the wavelength of interest. That certainly isn't always the case.

        I have anodized aluminum black and then photographed it under halogen light, which is rich in infrared. It appeared light brown although under natural light to the eye it seemed jet black. This effect can upset measurements by reflecting spurious ambient radiation without being obvious.

        For instance, if sunlight is falling on what appears to be a black surface but it isn't at long infrared, then it can confound the reading. There is a new roofing product that takes advantage of this effect to reduce heat gain. Black roofing shingles that reflect 90% of infrared are available.

        Best results will be obtained by measuring in the dark or under lights that have little infrared. LED is best followed by fluorescent. Sunlight, incandescent and halogen will cause errors.
        Free software for calculating bolt circles and similar: Click Here


        • #5
          I have both the pocket type purchased from tower hobbies a couple years ago for RC engine head temps and a Harbor fright $39.00 one which I thick is the same as most of the off wall named ones sold by any of the tool places. I haven't checked either one down real close. But they follow the color change of Tool steel O=1 in class when we made chasing and repousse tools. And it was always close with the pocket one on the heads.
          Probably as good as the ambinent infrared will allow like Evan said. I have a couple analog dial pockets and a digital pocket and to there end of the scale they are with in a couple degrees. I wouldn't drop the hammer on them or throw em in the bottom of a tool pile. Even Harbor Frights electrial meter equipped with the pryrometer leads is pretty accurate. FWIW when you do something they don't like or their chip can't handle you get an error on display
          Been there, probably broke it, doing that!
          I am not a lawyer, and never played one on TV!
          All the usual and standard disclaimers apply. Do not try this at home, use only as directed, No warranties express or implied, for the intended use or the suggested uses, Wear safety glasses, closed course, professionals only


          • #6
            I digress for a moment, but do not denigrate the old smiths.

            "up till now my heat treating has been rough and ready blacksmith end of the trade."

            My old FIL started as a smith, he was a past master at the art of hardening and tempering, blacksmith welding, etc. Even did some nice work with knives and one or two swords.

            After his demob from WWII, hairsprings and mainsprings for ladies watches were unavailable. Jimmy procured the wire, rolled, hardened and tempered these minute springs and fixed the watches. Made some tidy pocket-money
            Just got my head together
            now my body's falling apart


            • #7
              I have a $100 Raytech model for shooting tire temps during asphalt days that is at least consistent. Tops out at 500*F.

              Was shooting the headers one day and found the back cylinders off the chart! Couldn't believe it until I realized the fan was cooling the front ones ....

              Used to get a hot inside corner on the right rear from the exhaust ....

              Needless to say it mostly kept me confused!



              • #8
                I don't have any experience with the cheap ones but a friend of mine is a private investigator. He (while I was in business) would bring his toys to me to figure them out. He bought a $12,000 handheld imaging thermograph and brought it to me to so I could show him how to hook it up to his laptop.

                It covers 0F to 500F and is accurate +- 4F. Very cool toy. Sure made it easy to see how warm the various parts were running inside a computer. Perfect for diagnosing cooling problems. Barry wouldn't let me borrow it for a while.
                Free software for calculating bolt circles and similar: Click Here


                • #9
                  Mine has some provision to recalibrate based on the material's characteristics if you need added precision. It seems pretty close most of the time.



                  Try G-Wizard Machinist's Calculator for free:


                  • #10
                    My Ray tek Says most organic materials and painted surfaces have an emissilvity of .095( pre set in the unit)As others have said shiny and polished metal surfaces .will give inaccurate readings. and to use flat black paint or masking tape. They also mention aluminum and stainless as shiny. So I guess the tape or paint is the safe way,then you don't have to worry about how shiny is shiny. It says the unit cannot measure through trans parent surfaces.It will measure the surface temp of the glass instead.Dust steam or smoke can prevent accurate readings as it obstructs the units optics. Chris