It seems like every pyrometer manufacturer claims to be “accurate” and with each new model that is released, the big selling point is “improved accuracy” or “ensures accuracy.” Well, I’m sure if you did a comparison test where you aimed different brands of pyrometers at the same material, you would get different temperature readings. So… how do you know which one is “accurate” and which one is reading in error? Unless you are working in a high-tech lab with sophisticated and calibrated reference equipment it would be difficult to tell.
Most pyrometers are calibrated against blackbody furnaces and when you aim any type of pyrometer (long-wavelength, short-wavelength, ratio, etc.) against a blackbody furnace, it will be accurate. Just about any pyrometer can make an accurate reading while aimed at a blackbody furnace under ideal laboratory conditions. However, most industrial applications that use infrared pyrometers for process temperature control are far from ideal laboratory conditions. The chart below depicts a number of different sources of temperature error that could lead to less “accurate” measurements.
Out of the 5 sources of errors listed – only one is really in the control of the pyrometer manufacturer: Pyrometer Calibration. Again, this is making sure that when aimed at a blackbody furnace, the pyrometer reads correctly. The other 4 factors are all dependent on the end user’s process once the pyrometer is installed. All of these factors can influence the reading and “accuracy” of the pyrometer. However, with the right wavelength selection, and using the appropriate infrared technology, these sources or error can be reduced or even eliminated. I’ll go through these one by one.
1. Optical Obstruction
This is probably the biggest application challenge for most non-contact temperature measurements: How do we deal with viewing through any number of these interferences (steam, flames, combustion gasses, water, etc.)? The truth is that that depending on the wavelength of the pyrometer, you can actually view through these obstructions without interference.
Infrared energy is part of the electromagnetic spectrum, just like visible light, x-rays, microwaves, etc. If I shine a flashlight at my chest, the light does not pass through, but if I aim an x-ray at my chest, it goes right through. The only difference is the wavelength. So what does this mean? It means that at some wavelengths, objects can either be transparent or opaque. So if we apply the same logic to all of these interferences (flames, steam, water, combustion gasses, plasmas), at certain wavelengths these interferences can either be transparent or opaque. In order to avoid any interference, you would want to choose a pyrometer filtered at a wavelength where these interferences are transparent and the pyrometer can view through them without any error in the reading (see chart below). If the pyrometer is filtered at a wavelength where these optical obstructions are opaque, then the pyrometer will be influenced by these interferences and will not read “accurately.”
2. Emissivity Variation
Every type of material is different and emits infrared energy in a different way. Emissivity is a measure of how well an object or material emits infrared energy. In the most practical of terms, emissivity is the opposite of reflectivity – meaning shinier objects like aluminum have a very low emissivity and objects like paper have a high emissivity. Emissivity depends on the surface characteristics of a material and in an industrial process, emissivity can vary from one piece to the next. For single-wavelength pyrometers, you need to give an emissivity input in order for the pyrometer to make a temperature calculation. If that emissivity input equals the true emissivity of the material, then you will get an “accurate” reading. If the true emissivity of the material is different than the emissivity input to the pyrometer, then you will get an error. The size of the error is dependent on the degree of error between the true emissivity and the emissivity input: the larger the difference in emissivity, the greater the error.
Ratio pyrometers, on the other hand, automatically compensate for emissivity variations of most materials. As the name implies, ratio pyrometers take the ratio of energy at two different wavelengths and convert that ratio into a temperature value. If the change in emissivity affects both wavelengths equally then you are left with the same ratio measurement and a stable temperature – effectively eliminating any error from emissivity variation (see black and orange chart). This is a reason why ratio pyrometers are so popular for many steel applications.
Certain materials known as “non-greybodies” have complex emissivity characteristics where the change in emissivity is not the same across all wavelengths and a ratio pyrometer will produce an error. The most common non-greybody materials are aluminum, copper, brass, zinc, and stainless steels. For these materials, a multi-wavelength pyrometer using an algorithm to adjust to these complex emissivity materials is the only type of pyrometer that will produce an “accurate” and repeatable measurement.
3. Background Reflections
Infrared pyrometers measure energy and just like light, energy can be reflected off of different materials and picked up by the sensor. Common types of reflections come from hot furnace walls, open furnace doors, reflected sunlight, and even high energy overhead lights. Any of the reflected energy seen by the sensor will cause it to read artificially high. The most common way to avoid this interference is to aim the pyrometer in a location where there is no possibility of seeing any high energy reflections (i.e. away from the furnace door, not looking into it. Additionally, you can use a baffle or physical obstruction to shield the pyrometer’s viewing area from any reflected energy.
Single-wavelength pyrometers take an average temperature of what they see in their field-of-view. If they become misaligned or are only viewing partial view of the target, then they will read artificially low and will no longer be “accurate.” Ratio pyrometers do a better job at handling partially filled fields-of-view (the same principle applies here as it did with the emissivity variation mentioned above). However, dual-wavelength pyrometers can tolerate a much smaller filled field of view compared with two-color ratio pyrometers. Many pyrometers have different methods of aiming (line-of-sight, through-the-lens visual aiming, laser aiming) to ensure proper alignment to the target. But some processes measuring very small parts or wandering targets can make it difficult to constantly be aligned to the target, and so selecting the appropriate wavelength technology becomes crucial in order to maintain an “accurate” temperature measurement.
So for any given application, there are a number of potential interferences that can affect the “accuracy” of a pyrometer. One approach to avoiding these interferences and maintaining accuracy is to aim the pyrometer at a location in the process where none of these errors can occur – although this is not always ideal and usually not possible. Another approach is to have very tight optics and look at a very small spot – this way you know exactly where you are aiming and can always align to the process and because you are looking at a smaller spot, all of the other interferences will be minimized. However, if for some reason the small field of view becomes completely filled with an interference, then you have a large error and will get an inconsistent temperature reading and won’t always be “accurate.”
The far more effective way to remain “accurate” is to select the right wavelength to view through any optical obstructions in your process. This way, the pyrometer views through them like they were not even there. You would also want to choose the right wavelength technology that minimizes or eliminates any potential errors due to emissivity variation (ratio or multi-wavelength) or misalignment (ratio). Lastly, as a best practice, you should try to avoid any hot background reflections as these will cause an interference for just about all pyrometer technologies.