Multi-spectral method for detection of anomalies during powder bed fusion additive manufacturing
Embodiments of the systems can be configured to receive electromagnetic emissions of a substrate (e.g., a build material of a part being made via additive manufacturing) by a detector (e.g., a multi-spectral sensor) and generate a ratio of the electromagnetic emissions to perform spectral analysis with a reduced dependence on location and orientation of a surface of the substrate relative to the multi-spectral sensor. The additive manufacturing process can involve use of a laser to generate a laser beam for fusion of the build material into the part. The system can be configured to set the multi-spectral sensor off-axis with respect to the laser (e.g., an optical path of the multi-spectral sensor is at an angle that is different than the angle of incidence of the laser beam). This can allow the multi-spectral sensor to collect spectral data simultaneously as the laser is used to build the part.
1. A method of performing spectral analysis, the method comprising:
receiving electromagnetic emissions from a surface of a substrate;
splitting the electromagnetic emissions into a first light beam and a second light beam, the first light beam corresponding to line emissions of at least one component of the substrate, the second light beam corresponding to continuum emissions of the at least one component of the substrate;
generating line data S Line representative of an intensity of the first light beam;
generating continuum data S Continuum representative of an intensity of the second light beam;
subtracting S Continuum from S Line to generate line-continuum difference data (S Line −S Continuum ); and
dividing (S Line −S Continuum ) by the S Continuum to generate continuum to line-to-continuum ratio data S Line-to-Continuum ; and
using S Line , S Continuum , (S Line −S Continuum ), and/or S Line-to-Continuum data to identify anomalies related to defects in the substrate.
2. The method recited in clim 1 , wherein the substrate comprises build material of a part made by an additive manufacturing apparatus.
3. The method recited in claim 2 , further comprising generating S Line , S Continuum , (S Line −S Continuum ), and/or S Line-to-Continuum data in situ as the part is being generated.
4. The method recited in claim 2 , further comprising:
generate a laser beam at an angle α relative to the surface of the build material;
receiving the electromagnetic emissions though an optical receiver having an optical receiver axis that forms angle β with respect to the surface of the build material; and
α does not equal β.
5. The method recited in claim 2 , further comprising generating a feedback loop so that anomaly information related to defects in the part is used to control operational parameters of the additive manufacturing apparatus.
6. The method recited in claim 2 , further comprising using anomaly information to guide post-build inspection techniques for the part.
7. The method recited in claim 2 , further comprising generating a compilation of S Line , S Continuum , (S Line −S Continuum ), and/or S Line-to-Continuum data for at least one layer of the part.
8. The method recited in claim 7 , further comprising identifying anomalies, based on the compilation of S Line , S Continuum , (S Line −S Continuum ), and/or S Line-to-Continuum data, related to defects in the part in real-time as the part is being built.
9. The method recited in claim 1 , wherein receiving electromagnetic emissions comprises:
receiving, by a first multi-spectral sensor, a first electromagnetic emission from a first position of the substrate, the first electromagnetic emission being generated from a first energy source; and
receiving, by a second multi-spectral sensor, a second electromagnetic emission from a second position of the substrate, the second electromagnetic emission being generated from a second energy source.
10. The method recited in claim 9 , further comprising deconvolution of the electromagnetic emission.
11. The method recited in claim 1 , further comprising:
filtering the first light beam so as to pass light having wavelengths within a 10 nm range about the line emissions of the at least one component of the substrate and block or attenuate light having wavelengths outside the 10 nm range about the line emissions of the at least one component of the substrate.
12. The method recited in claim 1 , further comprising:
filtering the second light beam so as to pass light having wavelengths within a 10 nm range about the continuum emissions of the at least one component of the substrate and block or attenuate light having wavelengths outside the 10 nm range about the continuum emissions of the at least one component of the substrate.
13. The method recited in claim 1 , further comprising:
filtering the first light beam so as to pass light having wavelengths within a 10 nm range about the line emissions of the at least one component of the substrate and block or attenuate light having wavelengths outside the 10 nm range about the line emissions of the at least one component of the substrate; and
filtering the second light beam so as to pass light having wavelengths within a 10 nm range about the continuum emissions of the at least one component of the substrate and block or attenuate light having wavelengths outside the 10 nm range about the continuum emissions of the at least one component of the substrate.
14. The method recited in claim 1 , further comprising:
focusing the received electromagnetic emissions at an image plane;
diffusing an angle of incidence of the electromagnetic emissions at the image plane; and
colliminating, using a lens, the electromagnetic emissions at a focal length of the lens from the image plane.
15. The method recited in claim 1 , wherein the electromagnetic emissions comprise emissions related to a melt pool and/or a plume generated on the surface of the substrate.
16. The method recited in claim 1 , wherein the at least one component of the substrate comprises build material of the substrate and a contaminate.