Following on from a recent blog by my esteemed colleague Manfred in December, we continue the discussion on thermal imaging versus video imaging with a focus on furnace monitoring. We often get asked something along the lines of “why would I install a thermal imager when there are perfectly good high temperature video cameras available for my furnace?” It’s quite simple; accurate temperature data is extremely valuable.
IMAGEPro displays the thermal image whilst Regions of Interest (ROIs) are used to analyse the hottest, coldest, and average temperatures of groups of pixels
Our radiometric thermal imagers capture thermal energy emitted by objects in the form of an image made up from thousands of pixels that contain temperature data. In order to both measure high temperatures accurately and penetrate furnace gases, carefully selected wavelengths are specified, and cameras are calibrated in infrared laboratories containing blackbody furnaces. A visual imager/video camera, on the other hand, captures visible light and operate at wavelengths within the visible spectrum from approximately 400 to 700 nanometres. Over the years, we’ve started to understand how thermal imagers can add additional value to manufacturers in ways that visual imagers cannot:
1. Cost versus value. Even though a radiometric thermal imaging system can be significantly higher in price, the additional benefit of thermal imaging outweighs the extra cost in many applications, especially high temperature furnaces located in hazardous area. Having visual only hazardous area certified cameras on a furnace itself is quite expensive, especially when including installation, so the question is, how much value does a video imaging only system provide? Accurate temperature data on high-temperature processes is extremely valuable, so installing cameras without that capability is potentially a missed opportunity.
2. AMETEK Land thermal imagers will have a higher image quality than most visual cameras when installed in furnaces, largely because they are designed at carefully chosen wavelengths and with special filters that penetrate carbon dioxide and water vapour absorbing molecules. That means crisp, high contract images versus blurry, saturated, or low contrast images than can often be seen with video cameras on high temperature applications.
3. Plant yield can be increased by improving tube temperature uniformity and identifying burner/refractory issues. Continuous, comprehensive tube temperature data can help to identify both local flame impingement or underheated tubes that manual methods, outlet TCs, or skin thermocouples may miss.
4. Tube lifetime can be significantly extended with early detection and correction of issues that may cause premature wear and creep. Hot spots and physical tube movements can be detected and alarmed, avoiding upsets or critical incidents, and improving the overall safety, reliability, and efficiency.
The consequence of undetected hot spots can be significant, including tube rupture
5. Identifying poor uniformity and general overheating or underheating can reduce fuel consumption and CO2 emissions.
6. Labour cost savings and safety improvements through reduction in the frequency of manual inspections.
FoV analyses are conducted to ensure that thermal cameras can collect temperature data from all critical surfaces including tubes, burners, and refractory
7. From a maintenance perspective, continuous remote monitoring of the reformer's condition can make it easier to plan intervals, inspections, repairs, and other actions.
8. Thermal imaging during start-up, including flame behaviour recording and analysis, makes this process easier, safer, and reduces the risk of catastrophic incidents.
Radiometric thermal imagers are calibrated in blackbody furnaces
9. The system can help improve general productivity by increasing the availability and visibility of important data. Intuitive dashboards and historical analysis tools mean the software is a valuable training tool
10. Visual camera manufacturers occasionally claim to be able measure temperatures, but such claims should be verified for relevance to the specific conditions of the furnace being considered. Radiometric furnace thermal imaging and pyrometry is historically only performed with either a 1um or 3.9um instrument – if not in these wavelengths, limited data is available demonstrating accuracy. Accuracy errors can be extremely high and unknown if emissivity and background correction is not performed, especially when there is a high temperature delta between tube and refractory temperatures, as we well know from testing our Cyclops C055L versus the pyrometers more typically used on furnaces, the Cyclops C100L and C390L. Put simply, with shorter wavelengths, background errors increase. A non-radiometric thermal camera/video camera operates at even shorter wavelengths than 1um (e.g., 400-700nm), so the error is higher and also unknown. We radiometrically calibrate our imagers, which is the reason we can state a 1% accuracy and 1-degree Celsius repeatability. If temperature accuracy and repeatability is not stated, the device cannot really be considered a radiometric thermal camera.