0
$\begingroup$

I am trying to design a heat source for use with testing the functionality of an infrared thermometer. I am happy with my circuit, I have a surface heated to a nice, constant 50°C (as constantly monitored by a thermocouple).

The surface is just an aluminium alloy, which I have sprayed with a matt black paint. Unfortunately, finding the exact emissivity of the surface is proving difficult. I am unsure whether to keep this as my surface, or find something else. The matt black paint is supposed to have an emissivity of 0.93-0.97, yet aluminium alloy has a very low emissivity.

A table can be found HERE which shows the emissivity of different surfaces.

Thermal conductivity of different materials can be found HERE.

The reason I chose aluminium was because it is easy to get hold of, it is not expensive and has good thermal conductivity. Because of its low emissivity though, I had to paint the surface in order to test the infrared thermometers.

I did a lot of research on the subject and there are so many places that specify using a black surface to get the best emissivity (as in closer to 1.0), yet there are some black surfaces which have a very poor emissivity as seen HERE. That document also shows some white surfaces as having good emissivity, and white is the furthest from black you can get!

There are other articles such as this one HERE that look at the effect of the smoothness of a surface, which I found interesting as some materials with smooth surfaces have a good emissivity, even if not black.

So my question is: What is more important when it comes to emissivity (getting closer to 1.0), colour or surface finish? Do different colours or surface finishes affect the wavelength of the thermal radiation? If so, why and how do these factors have an effect?

Improve this question
$\endgroup$
8
  • 1
    $\begingroup$ White is not necessarily the furthest from black you can get. "White" is defined as spectrally flat across the narrow optical part of the spectrum (equivalent to about 6000K), while "black" in the context of emissivity covers the whole spectral region across the temperature range of interest (in your case, about 323K) $\endgroup$ – Brian Drummond Aug 17 '17 at 11:26
  • $\begingroup$ Ahh, I was looking at it in colour terms rather than engineering terms! $\endgroup$ – MCG Aug 17 '17 at 11:28
  • $\begingroup$ What's more important accurate or high values? The emissivity of a surface depends not only on the material but also on the nature of the surface. For example, a clean and polished metal surface will have a low emissivity, whereas a roughened and oxidised metal surface will have a high emissivity. The emissivity also depends on the temperature of the surface as well as wavelength and angle. I would use Kapton tape for accuracy $\endgroup$ – Tony Stewart Sunnyskyguy EE75 Aug 17 '17 at 12:21
  • $\begingroup$ Accuracy is more important. I'm quite happy with my overall design. The questions were more to satisfy curiosity! $\endgroup$ – MCG Aug 17 '17 at 12:23
  • $\begingroup$ If you use paint, or Kapton, then the substrate is less important. but thickness is. $\endgroup$ – Tony Stewart Sunnyskyguy EE75 Aug 17 '17 at 12:24
2
$\begingroup$

An article on Omega's website may help.

Five Ways to Determine Emissivity

There are five ways to determine the emissivity of the material, to ensure accurate temperature measurements:

  1. Heat a sample of the material to a known temperature, using a precise sensor, and measure the temperature using the IR instrument. Then adjust the emissivity value to force the indicator to display the correct temperature.
  2. For relatively low temperatures (up to 500°F), a piece of masking tape, with an emissivity of 0.95, can be measured. Then adjust the emissivity value to force the indicator to display the correct temperature of the material.
  3. For high temperature measurements, a hole (depth of which is at least 6 times the diameter) can be drilled into the object.
  4. This hole acts as a blackbody with emissivity of 1.0. Measure the temperature in the hole, then adjust the emissivity to force the indicator to display the correct temperature of the material.
  5. If the material, or a portion of it, can be coated, a dull black paint will have an emissivity of approx. 1.0. Measure the temperature of the paint, then adjust the emissivity to force the indicator to display the correct temperature. Standardized emissivity values for most materials are available (see pages 114-115). These can be entered into the instrument to estimate the material's emissivity value.

Masking tape is commonly available, self adhesive, removable and low cost.

Improve this answer
$\endgroup$
1
  • $\begingroup$ Interesting information... So what about the actual questions asked? $\endgroup$ – MCG Sep 19 '17 at 14:49
1
$\begingroup$

Since the paint covers the aluminum, the emissivity of Al is irrelevant. The heat is conducted from metal to paint, and about the best you can do is to assume that both are at the same (surface) temperature.

To calibrate a thermometer, what's important is not the emissivity value (1 vs 0.99 vs 0.6) but the uniformity across the spectral band of the thermometer. That's because the temperature is calculated based on the ratio of power at different wavelengths.

Next, you have to consider not only the paint's spectral emissivity but also whether you need a Lambertian output distribution. If you are just testing IR thermometers, you probably don't- just point the thermometer perpendicularto the radiating surface and be consistent in your alignment procedure.

Improve this answer
$\endgroup$
3
  • $\begingroup$ I probably shouldn't have included so much background info to the question.... The metal and the paint are the same temperature, I have verified the surface temperature with a thermocouple. I am not using this to calibrate the thermometers, that is done at a later stage, this is a pre calibration check. The actual questions I wanted answering were the ones at the end. The rest of it I am happy with. The actual product works as intended, I was asking some questions I was curious about. Unfortunately, this answer doesn't answer any of the questions I posed. $\endgroup$ – MCG Aug 17 '17 at 13:41
  • $\begingroup$ Just as a reminder, the actual questions were: So my question is: What is more important when it comes to emissivity (getting closer to 1.0), colour or surface finish? Do different colours or surface finishes affect the wavelength of the thermal radiation? If so, why and how do these factors have an effect? $\endgroup$ – MCG Aug 17 '17 at 13:44
  • $\begingroup$ @MCG well, the paint's visible color is not particularly correlated to its spectral emissivity in the IR. Surface finish drastically affects emissivity. In fact "gold black," a surface of pure gold but deposited in microdroplets, is oneof the better "black" surfaces available. To avoid specular behavior at low incidence angles, roughened surfaces are pretty much a must. This is not a trivial problem, to say the least. $\endgroup$ – Carl Witthoft Aug 17 '17 at 15:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.