Optics-Based Analysis of Error in Infrared Radiation Cameras for Large Format Additive Manufacturing

In large format additive manufacturing (LFAM), the large repeating curvatures from layer-based deposition indicates planar approximation, often required for the use of IR Cameras for accurate temperatures, is not valid. This work assessed established error from using radiant flux IR cameras in LFAM. Simulated electromagnetic interference maps (SEIM) were generated for a planar surface and representative bead geometries with varying properties like surface roughness. Two single bead wall geometries were observed where one was machined, and the other maintained the bead structure. The walls were reheated and measured via surface thermocouples to determine any temperature differences in machined and unmachined segments during cooling, as well as camera reported temperature differences in machined and unmachined relative to the thermocouples. Results showed that the unmachined panel had a 5% increase in absolute IR camera percent error relative to the machined panel, supporting the prediction that the bead-induced structure impacts the accuracy of radiant flux sensors. This work provided an interference-based recommendation for placement of cameras and insight to the impact on sensor accuracy from surface properties, which is needed for accurate and representative measurements.