KEYWORDS: Cameras, Photodetectors, Sensors, Signal processing, Signal detection, Photodiodes, High speed cameras, Video, Data acquisition, Laser welding
Laser powder bed fusion (LPBF) is an additive manufacturing (AM) process in which a high power laser melts metal powder layers into complex, three-dimensional shapes. LPBF parts are known to exhibit relatively high residual stresses, anisotropic microstructure, and a variety of defects. To mitigate these issues, in-situ measurements of the melt-pool phenomena may illustrate relationships between part quality and process signatures. However, phenomena such as spatter, plume formation, laser modulation, and melt-pool oscillations may require data acquisition rates exceeding 10 kHz. This hinders use of relatively data-intensive, streaming imaging sensors in a real-time monitoring and feedback control system. Single-point sensors such as photodiodes provide the temporal bandwidth to capture process signatures, while providing little spatial information.
This paper presents results from experiments conducted on a commercial LPBF machine which incorporated synchronized, in-situ acquisition of a thermal camera, high-speed visible camera, photodiode, and laser modulation signal during fabrication of a nickel alloy 625 AM part with an overhang geometry. Data from the thermal camera provides temperature information, the visible camera provides observation of spatter, and the photodiode signal provides high temporal bandwidth relative brightness stemming from the melt pool region. In addition, joint-time frequency analysis (JTFA) was performed on the photodiode signal. JTFA results indicate what digital filtering and signal processing are required to highlight particular signatures. Image fusion of the synchronized data obtained over multiple build layers allows visual comparison between the photodiode signal and relating phenomena observed in the imaging detectors.
Accurate non-contact temperature measurement is important to optimize manufacturing processes. This applies to both additive (3D printing) and subtractive (material removal by machining) manufacturing. Performing accurate single wavelength thermography suffers numerous challenges. A potential alternative is hyperpixel array hyperspectral imaging. Focusing on metals, this paper discusses issues involved such as unknown or changing emissivity, inaccurate greybody assumptions, motion blur, and size of source effects. The algorithm which converts measured thermal spectra to emissivity and temperature uses a customized multistep non-linear equation solver to determine the best-fit emission curve. Emissivity dependence on wavelength may be assumed uniform or have a relationship typical for metals. The custom software displays residuals for intensity, temperature, and emissivity to gauge the correctness of the greybody assumption. Initial results are shown from a laser powder-bed fusion additive process, as well as a machining process.
In addition, the effects of motion blur are analyzed, which occurs in both additive and subtractive manufacturing processes. In a laser powder-bed fusion additive process, the scanning laser causes the melt pool to move rapidly, causing a motion blur-like effect. In machining, measuring temperature of the rapidly moving chip is a desirable goal to develop and validate simulations of the cutting process. A moving slit target is imaged to characterize how the measured temperature values are affected by motion of a measured target.
Under certain conditions, the polarization state of infrared light emitted by metal changes when the metal is strained.
During cutting, metal is severely strained. Assessing both strain and strain rate is of interest to the metal cutting research
community. Over large areas, Digital Image Correlation (DIC) performed on high-speed video can provide approximate
values for the average strain and strain rate. However, small areas such as the shear zone are difficult to image with
enough resolution to perform DIC. If the thermal radiation emitted by these small areas is polarized, there is the potential
to provide valuable information to the metal cutting community. This paper is an initial investigation into that
possibility, as well as the use of the polarization information for uncertainty analysis, reflection detection, and region of
interest classification. A rotating polarizer is used that triggers a thermal spectrum camera to acquire images at specific
polarization angles. When cutting, the metal is constantly moving and the material imaged is different from one moment
to the next. At each angle of the polarizer, a sufficiently long integration time is used so the material is severely motion
blurred, resulting in an image which estimates the typical intensity for that angle. By comparing the typical intensities,
and assuming the light is linearly polarized, the polarization state may be estimated.
Process models, including finite element modeling simulations, are important for optimizing the metal cutting process,
allowing industry to make parts faster, better, and at less cost. Measurements of the process can be used to improve and
verify the accuracy of these models. There are many error sources when using infrared radiation thermography to
measure the temperature distribution of the tool, workpiece, and chip during metal cutting. Furthermore, metal cutting
presents unique measurement challenges due to factors such as the high magnification required, high surface speeds,
micro-blackbody effects, and changing emissivity as chips form.
As part of an ongoing effort to improve our understanding of uncertainties associated with these thermographic
measurements, two sets of experiments were performed. One set explored how well the surface temperature of the
cutting tool accurately reflects the internal temperature. This was accomplished by simultaneously measuring the
temperature using both a thermal camera and a thermocouple embedded within the cutting tool.
The other set investigated correcting for motion blur, point spread function, and a less than ideal range of sensitivity of
the thermal camera when measuring the shear zone temperature of the chip. In theory, this correction could be performed
using deconvolution. Unfortunately, deconvolutions are sensitive to noise and it is difficult to gauge the uncertainty of
the computed values. Thus, convolutions of various assumed inputs were computed and compared to the measured
temperatures. Assumed inputs which yielded a good fit to the measured temperatures were considered candidate values.
The range of those candidate values yields a measure of the uncertainty of the calculation.
There are many error sources when using infrared radiation thermography to measure the temperature distribution of the
tool, workpiece, and chip during metal cutting. It is important to understand how these error sources affect the
measurement uncertainty. Some are familiar to anyone performing thermography measurements, such as uncertainties
in the basic camera calibration. However, metal cutting presents unique measurement challenges due to factors such as
the high magnification required, high surface speeds, polarization effects, micro-blackbody effects, and changing
emissivity as chips form. This paper presents highlights of the current state of efforts at NIST to catalog and characterize
error sources and the resulting uncertainties.
In order to investigate temperatures reached during orthogonal metal cutting, a novel approach for measuring temperatures at the tool-chip interface has been developed based on high-speed thermography. A thermal infrared camera and a visible camera combined through a dichroic beam splitter form the basis for a synchronized visible and
infrared imaging system. Pairing the infrared camera with a higher speed visible camera allows for assessment of thermal images with aberrant chip flow or an obstructed view of the tool/chip interface. This feature facilitates the use of the apparatus in machining environments where machining chips or other debris fly about. The measurement setup also includes a force dynamometer, custom timing circuitry, and a high-speed digital oscilloscope to enable timing of frames together with force measurements so that analysis of the infrared images can be compared against the energy levels measured through the cutting forces. The resulting infrared images were converted to radiance temperatures through comparison to a NIST calibrated blackbody. Emissivity was measured by thermally imaging the machining chips heated to known temperatures. Machining experiments were performed at various cutting speeds and at two different infrared wavelengths. Analysis of these experiments gives insight into the relationships between emissivity, temperature, surface condition, infrared wavelength and motion blur. The analysis shows that using the visible, thermal and force data together is a significant improvement over any of these alone. These insights lead to practical guidance for use of infrared imaging systems to image rapidly moving objects.
Since the modeling of machining processes relies on high-strain-rate, high-temperature material properties, NIST has built a split-Hopkinson (or Kolsky) bar to determine the stress-strain behavior of rapidly heated materials at high temperatures. Our Kolsky bar has been constructed in the NIST high current pulse-heating facility, which enables electrically heating the samples within ~ 100 milliseconds time duration, immediately before the mechanical impact in the bar. Due to the rapid heating, we avoid possible structural changes in the sample, and a stress-strain relationship can be determined at different temperatures for various test materials. We describe the design and the development of the resistively-heated Kolsky-bar apparatus. The incident and the transmitted bars are constructed of 1.5 m long, 15 mm diameter maraging steel, and a typical sample is a 4 mm-diameter, 2 mm-long cylinder of 1045 steel. The sample is placed between the bars and held by friction. The current is transmitted through the graphite-sleeve bushings of the two bars. The non-contact temperatures are measured using an InGaAs near-infrared micro-pyrometer (NIMPY) and an InSb focal-plane (320 by 256) array (thermal camera). The NIMPY and the thermal camera are both calibrated using a variable-temperature blackbody, and the thermodynamic temperature of the metal is determined using the emissivity determined from the measured infrared spectral reflectance of the metal. Thermal videos of the electrically-heated and the room-temperature impacts will be shown with 1 kHz frame rates, and the changes in the stress-strain curves with the temperature of the samples will be discussed.
The NIST (National Institute of Standards and Technology) virtual/physical surface roughness calibration standard consists of physical specimens whose surfaces are manufactured by a numerically controlled diamond-turning process using digitized profiles. These standards are designed for checking the characteristics and algorithms of surface measuring systems, and for sensing the amount of distortion of the surface information flow though different measuring systems. The digitized profiles can also be used for remote instrument calibration and surface measurement unification. The design, manufacture, test results, and potential uses of the NIST prototype specimens are discussed.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.