3.9 6.

Communication

Process Monitoring Using

Synchronized Path Infrared
Thermography in PBF-LB/M

Dennis Höfflin, Christian Sauer, Andreas Schiffler and Jürgen Hartmann

Special Issue
Opto-Thermal Sensor Technologies
Edited by
Prof. Dr. Jürgen Hartmann and Dr. Jochen Manara

https://doi.org/10.3390/s22165943
 sensors
Communication
Process Monitoring Using Synchronized Path Infrared
Thermography in PBF-LB/M
Dennis Höfflin 1, *, Christian Sauer 1 , Andreas Schiffler 1 and Jürgen Hartmann 1,2

1 Institute of Digital Engineering, University of Applied Sciences Würzburg Schweinfurt,

97070 Würzburg, Germany
2 Bavarian Center of Applied Energy Research e.V., 97074 Würzburg, Germany
* Correspondence: dennis.hoefflin@fhws.de

Abstract: Additive manufacturing processes, particularly Laser-Based Powder Bed Fusion of Metals
(PBF-LB/M), enable the development of new application possibilities due to their manufacturing-
specific freedom of design. These new fields of application require a high degree of component
quality, especially in safety-relevant areas. This is currently ensured primarily via a considerable
amount of downstream quality control. Suitable process monitoring systems promise to reduce
this effort drastically. This paper introduces a novel monitoring method in order to gain process-
specific thermal information during the manufacturing process. The Synchronized Path Infrared
Thermography (SPIT) method is based on two synchronized galvanometer scanners allowing high-
speed and high-resolution observations of the melt pool in the SWIR range. One scanner is used to
steer the laser over the building platform, while the second scanner guides the field of view of an
IR camera. With this setup, the melting process is observed at different laser powers, scan speeds
and at different locations with respect to the laser position, in order to demonstrate the positioning
accuracy of the system and to initially gain thermal process data of the melt pool and the heat-affected
zone. Therefore, the SPIT system shows a speed independent overall accuracy of ±2 Pixel within the
evaluated range. The system further allows detailed thermal observation of the melt pool and the
Citation: Höfflin, D.; Sauer, C.;
surrounding heat-affected zone.
Schiffler, A.; Hartmann, J. Process
Monitoring Using Synchronized Path
Infrared Thermography in
Keywords: SPIT; PBF-LB/M; additive manufacturing; process monitoring; SWIR; melt pool;
PBF-LB/M. Sensors 2022, 22, 5943. galvanometer scanner
https://doi.org/10.3390/s22165943

Academic Editor: Jose Antonino

Daviu
1. Introduction
Received: 25 July 2022 Since the earliest days of invention, the tremendous potential of metal processing
Accepted: 6 August 2022 additive manufacturing technologies has excited many researchers and industrial commu-
Published: 9 August 2022 nities. The PBF-LB/M process, which is currently the most widespread technology with
Publisher’s Note: MDPI stays neutral the highest degree of industrial maturity, plays a key role [1]. This technology enables the
with regard to jurisdictional claims in production of material- and weight-efficient innovative components without additional
published maps and institutional affil- tools or clamping devices. However, the layer-wise melting of metal powder via laser
iations. heating results in complex and time-dependent temperature profiles that decisively depend
on process and material parameters. The application of high scan velocities results in
exposure times in the range of milliseconds and, in combination with high laser powers,
induces extremely high heating and cooling rates, causing unique material properties and
Copyright: © 2022 by the authors. microstructures [2]. The control of these extreme conditions, with the aim of ensuring the
Licensee MDPI, Basel, Switzerland.
required component quality, is still the major challenge of this technique. The immense in-
This article is an open access article
crease in published research studies on this topic in recent years (cf. Figure 1) shows that the
distributed under the terms and
development is not yet complete, and there is still a need for dedicated research, especially
conditions of the Creative Commons
in the area of in situ process data acquisition and validated in situ process monitoring.
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).

Sensors 2022, 22, 5943. https://doi.org/10.3390/s22165943 https://www.mdpi.com/journal/sensors

Sensors 2022, 22, 5943 2 of 11

Figure 1. Number of publications on the topic of in situ sensing and/or in situ monitoring of powder
bed fusion processes [3].

Numerous in situ monitoring approaches for specific process parameters and char-
acteristics have been investigated, with the aim of deriving indicators or key amounts for
the subsequent evaluation of process and component quality. Among other things, laser
power [4], powder coating and powder bed surface [5], powder bed compaction [6], smoke
plumes and spatter behavior [7], particle emissions [8], part distortion [9] and vibration
during powder coating [10] were already considered. However, the main focus of research
efforts in the field of in situ process monitoring is in the investigation of the spatial and
temporal temperature distribution in and around the melt pool, i.e., the laser–material –
interaction zone [11,12]. The systems used for this purpose are mainly designed for measur-
ing thermal radiation, and are based on non-contact measurement devices such as diodes,
pyrometers and cameras sensitive at different wavelength ranges [13–15]. Based – on the
arrangement of the optical path, these systems can be divided into off-axis and on-axis
systems (cf. Figure 2) [16].

Laser
on-axis
off-axis
Scanner
Beam Sensor Unit
Sensor Unit Splitter

Figure 2. Schematic representation of the differently arranged sensor units of on-axis and
off-axis systems.

In off-axis systems, the field of view of the sensor unit is arranged at a fixed angle
to the working plane, independently of the processing optics. This allows measurements
in broad wavelength ranges within the visible [17,18] as well as the IR range (SWIR [19],
MWIR [20], LWIR [21]). Off-axis monitoring systems of commercial PBF-LB/M systems,
in which the whole processing area is to be considered, focus either on a high recording
speed or a high spatial resolution due to limited hardware properties [22,23]. Zhang
presented an investigation based on an off-axis setup with a limited viewing ρ area of
12 mm × τ12 mm, using high frame α rate and high spatial resolution [7]. Emitted radiant
intensityhwas measured
known to investigate
emissivity, various
the radiance process-specific
of a real body can befeatures as indicators
determined for
via Planck’s
evaluating manufacturing quality. The single-track scenarios were detected with an image
b,λ

λ
−1
2ℎ𝑐 2 ℎ𝑐
𝐿𝑏,𝜆 (𝜆, 𝑇) = (𝑒𝑥𝑝 ( ) − 1)
Sensors 2022, 22, 5943 3 of 11

acquisition rate of 2000 fps and spatial resolution of 12 µm per pixel, in a wavelength range
from 350 nm to 800 nm.
In on-axis systems, thermal process emissions are directed through the optical path
of the laser onto the sensor unit. The field of view follows the beam of the laser over
the entire processing surface, which is why these systems are particularly suitable for
observing and characterizing the region in and around the melt pool. The small measuring
field allows high image acquisition rates and high spatial resolution, with no limitations
in the overall viewing area. To avoid imaging errors caused by the optical path, the
observation wavelength is in the range of the wavelength of the processing laser [24]. Using
an on-axis setup consisting of a high-speed camera and a photodiode, Berumen et al. [25]
succeeded in capturing the melt pool during processing, with a resolution of 10 µm per
pixel and an acquisition rate of 16.666 fps. The camera observed the melt pool dimensions,
while the diode detected the average emitted thermal radiation. Based on the quantitative
measurement data, a closed-loop control system was implemented to stabilize the melt
pool. Clijsters et al. [26] adopted a similar arrangement, which is under development for
real-time monitoring and detection of when melt breakage or material discontinuity occurs
in the process.
Due to the complexity of the processes involved in PBF-LB/M, alongside the fact
that individual sensors are limited to detect specific characteristics only, multi-sensor
systems have been used more frequently to increase the detection quality. Harbig et al. [27]
integrated an additional on-axis high-speed camera (HSC1; plasmoEye) into the existing
melt pool monitoring system of an EOS M290, in order to detect defects based on process
anomalies. The used melt pool monitoring system consisted of an on-axis and an off-axis
photodiode, collecting the intensity data of the process in a range from 400 nm to 900 nm,
while a high-speed camera provided a spatially resolved 2D intensity distribution of the
melt pool at a wavelength of 900 nm ± 50 nm. Using a new methodology for the data
fusion enabled a significant increase in the sensitivity of defect detection by up to 20%.
In order to achieve a sufficiently high signal, especially for narrow-wavelength band
measurements, the selected sensitivity range of the process monitoring system for thermal
radiation measurement is important.
Therefore, the spectral characteristics of the material under investigation are of great
interest. Here, the emissivity is probably the most important factor to be determined.
Emissivity is the ratio of the thermal radiation an object emits compared to that of a
perfect black body at the same temperature (a black body is defined by reflectance ρ = 0,
transmittance τ = 0 and absorptance α = 1) [28].
With known emissivity, the radiance of a real body can be determined via Planck’s
law, given in Equation (1). It describes the emitted thermal power per wavelength and
area, represented by the radiance Lb,λ of a black body, as a function of temperature T and
wavelength λ.
 −1
2hc2 hc
  
Lb,λ (λ, T ) = 5 exp −1 (1)
λ λkT
Here, h is the Planck’s constant, c is the speed of light and k is the Boltzmann constant.
Wien´s displacement law further shows that the wavelength λmax , at which the radiation
power has its maximum value in the black-body spectrum, displaces to the side of the
shorter wavelengths with increasing temperature T [29].

λmax T = 2897 µm K (2)

Figure 3 shows the spectral radiance distribution of a black body according to Planck,
as well as the maximum wavelengths according to Wien for temperatures between 100 ◦ C
and 1500 ◦ C. The picture depicts that the most relevant temperature regions of solidified
surface temperatures have their maximum intensity in short- and mid-wavelength infrared
(SWIR, MWIR).
Sensors 2022, 22, 5943 4 of 11

Spectral range off-axis Spectral range on-axis Spectral range SPIT Laser Wavelength

Spectral radiance Lb, λ (λ,T) [W/m3]

Wavelength λ [µm]

Figure 3. Visual representation of the spectral radiance distribution of a black body (Planck spectrum)
trum)Wien’s
and and Wien’s displacement
displacement lawtemperatures
law for for between 100 ◦ C and 1500 ◦ C [20]. The common
spectral range of on-axis (light green area) and off-axis (light red area) systems, as well as the spectral
range of the SPIT system (yellow bar) and the laser wavelength (red bar) are highlighted. The melting
range of the material used in this investigation (AlMg3) is between 595 ◦ C and 645 ◦ C [30].

In this work, a novel system for in situ process monitoring is presented (Synchronized
nized Path Infrared Thermography―SPIT) based on two separated Galvanomete
Path Infrared Thermography—SPIT) based on two separated Galvanometer scanners.
With this setup, it is possible to reliably detect a roaming melt pool with a sensor unit
whose movement can be controlled independently of the laser. The additional optical
path is designed for wavelength ranges favorable for process monitoring in additive
manufacturing (SWIR).

2. Materials and Methods

2.1. SPIT Setup
The experimental setup was designed to represent the basic processes taking place
in the PBF-LB/M process (components are given only for clarity, which does not mean
the used component is the best or the only applicable one). A Yb-fiber laser from JPT (JPT
Opto-electronics Co., Shenzhen, China) operating at a wavelength of 1080 nm was used as
the energy source. The laser output can be adjusted in the range from net 61 W to 458 W,
as continuous wave or pulsed, with a frequency of up to 20 kHz. After being collimated
and widened, the laser was guided into a SCANLAB intelliSCAN III 20 galvanometer
scanner (SCANLAB GmbH, Puchheim, Germany)—the ― so-called laser scanner. The scanner
directed the emitted radiation through an F-θ-Lensθ optimized for a wavelength range from
1030 nm to 1080 nm.
The lens focused the laser beam onto a platform placed 421 mm below in a sealed
aluminum containment box. The approximated spot size at the focal point had a diameter
of 65 μm,
of 65 with aa Rayleigh
µm, with Rayleigh length
length of 1.5 mm.
of 1.5 mm. The
The laser
laser and
and scanner
scanner unit
unit could
could be
be controlled
controlled
using a laserDESK software interface. The platform itself was movable along the Z-axis
to adjust the different thicknesses of the processed parts. To control the status inside the
sealed box, an additional diode was used to monitor the radiation emission occurring while
the laser was activated.
To make the measuring point independent from the position of the processing laser, the
SPIT setup was fitted with a second scanner unit—the so-called sensor scanner. A schematic
th a second scanner unit―the so
Sensors 2022, 22, 5943 5 of 11

representation of the setup is shown in Figure 4. The additional sensor scanner is, in
principle, equivalent to the laser scanner, but with its optical components optimized for
wavelengths in the SWIR range (1940 nm to 2050 nm). To adjust the slightly longer focal
distance, an aluminum plate with a thickness of 16 mm was placed underneath the sensor
scanner. This sets up a second optical path which is controlled by the same software
interface as the main scanner. To superimpose the zero point of the coordinate systems of
the two scanners, a basic offset was added, respectively subtracted from the X-axes. Within
the maximum operating parameters of the scanner units, the arrangement resulted in an
intersection area of 125 × 218 mm2 . Additionally, ScanLab provided correction parameters
θ
for each F-θ-Lens regarding the different refraction angles between the center and the outer
zones. These datasets increased the precision of the positioning significantly, and were
implemented as correction factors.

B C
D

A: Infrared Camera
B: Sensor Scanner
C: Laser Scanner
D: Laser

laser direction

Figure 4. Schematic setup of Synchronized Path Infrared Thermography (SPIT). The measuring field
of the camera (A) is moved by the sensor galvanometer scanner (B) synchronously with the working
galvanometer scanner (C) of the laser (D).

To gain infrared thermal radiation information about the melting process, an InfraTec
IR 8300 camera (InfraTec GmbH, Dresden, Germany) was placed in front of the aperture of
the sensor scanner. The camera itself used a 50 mm lens, resulting in an optical resolution
of the complete path of 135 µm for each pixel. Since the setup allows the measuring field of
tion of theto
the camera complete
be movedpath of 135 μmtofor
synchronously theeach
laser,pixel.
even aSince
smallthe setup
image allows
section the measuring
is sufficient
to fully capture the region of interest during the process. To maximize the possible frame
rate, the field of view was therefore set to (40·40) pixels in the center of the image. This
corresponded to an area of about (5.5·5.5) mm2 . With this configuration, including an
integration time of 89 µs, the camera provided a framerate of 2000 fps. The extraction of
the raw data was realized via a customized interface based on the Software Development
Kit (SDK) provided by InfraTec. After defining a region of interest, the software surveilled
this area for temperatures above a certain threshold and recorded a predefined number of
pictures when triggered. The integration time of the camera system and the adjustment
of the spectral sensitivity was configured by the SDK as well. The information provided
within the processed images shows the currently uncalibrated grayscale values of the laser
interaction zone and the surrounding area.
For downstream data analysis, the collected data were also stored as comma-separated
value files that were analyzed and visualized with the help of MATLAB R2021a. Assuming
that the brightest pixel always represents the laser–material interaction zone, the detection
of this pixel can be used to draw conclusions about the positioning accuracy of the sensor
system. Therefore, statistics about the position of the brightest pixel were carried out.
–
Sensors 2022, 22, 5943 6 of 11

Pictures taken during the time when the laser was deactivated, e.g., during jumps between
separated characters, were not considered. The entire setup is shown in Figure 5.

Laser Sensor
scanner scanner
Beam
Oscilloscope expander Optical
Laser Collimator fibre
scanner

z-axis
Laser Inert gas
nozzle IR Camera
Sensor
Cooler
scanner
Process
platform Containment
box

(a) (b)

Figure 5. (a,b) Representation of the laboratory setup for the Synchronized Path Infrared Thermogra-
phy system.

2.2. Methods, Samples and Test Parameters

To initially evaluate the performance of the laser source and optical path, a series of
tests were performed. The laser power was measured by a Primes Cube M, the beam caustic
and the focus diameter with the help of a Primes FMW+ and, finally, the precision of the scan
field with a Primes SFM (PRIMES GmbH, Max-Planck-Str. 2, 64319 Pfungstadt, Germany).
–
To further validate the accuracy of the setup and the ability to detect the thermal effects
of laser–material interaction, a series of sample exposures were performed at different laser
speeds and laser – powers on plain aluminum plates. The parameter range is shown in Table 1.
For every combination of laser speed and laser power, the path consisted of a straight line
along the X-axis with a length of 32 mm. After completion of the line, the current test
parameters and date were engraved (cf. Figure 6). The data produced while engraving
the current parameter set and date were also part – of the evaluation in order to represent
complex laser paths in the x–y plane. Therefore, a font size of approximately 2.5 mm was
chosen. Sandblasted plates made of AlMg3 with – the dimensions of 40 × 40 mm2 and a
thickness of 2 mm were used as the base material.

Figure 6. Processed AlMg3 base plate.

Sensors 2022, 22, 5943 7 of 11

Table 1. Laser energy outputs and travel speeds.

Energy Output Travel Speed

105 W 0.05 m/s
240 W 0.5 m/s
460 W 1.0 m/s

The investigation of the melt pool dimensions was carried out metallographically.
For this purpose, the samples were cut perpendicular to the scan direction of the line,
embedded in epoxy, ground, polished and etched. The microscopic investigation was
be well within the expected parameters, with a spot diameter of 66 μm, a focal length of
performed using an incident light microscope (Olympus AX70).

3. Results
In the first step, the characteristics of the laser beam were determined. It proved to
be well within the expected parameters, with a spot diameter of 66 µm, a focal length of
427 mm and a Rayleigh length of 1.43 mm.
Based on static pointwise measurements at different locations in and near the zero
point of the superimposed coordinate systems, the laser spot could be assigned to a defined
single pixel at the sensor unit. Further, the evaluation of the distinctive thermal images
of these measurements showed the single brightest pixel at the place where the beam
interacts with the surface. Surrounding pixels also showed significantly increased gray
values, representing the zone of thermal influence (cf. Figure 7). During initial dynamic
measurements near the zero point, the brightest pixel of the measurements corresponded
with the previously determined laser spot location on the sensor. Therefore, the brightest
pixel in each image was used as the reference point for the precision evaluation of the setup.

(a) (b) (c)

Figure 7. IR-images for (a) 460 W, 0.05 m/s; (b) 240 W, 0,5 m/s; (c) 105 W, 1 m/s.

The dimensions and intensity of the measured brightness depended– on the chosen
laser power and scan speed. The signal-to-noise ratio variated from 20.3 dB at 460 W and
0.05 m/s to 8.1 dB at 105 W and 1 m/s.
The observation of the brightest pixel showed that tracking the laser–material interac-
tion point is possible with high precision within the parameters applied in this study. The
migration of this pixel was minimal in the tests performed and showed a direct dependency
on the position within the processing area. The high repeatability revealed that the causes
of this movement are to be found in an optical distortion of the thermographic path.
While moving along the x-axis to carry out the straight line, no deviation in the
brightest pixel along the y-axis was observed. Meanwhile, the roaming within the field of
view along the x-axis was consistent in all sets with a span of 5 pixels. After finishing the
straight line, there was a jump along the y-axis to the beginning of the first digit. This jump
can also be correlated to the omission of a whole senor line when analyzing the position
of the brightest pixels. This skipped line is recognizable across all settings and, again,
indicates the geometric relation between position and deviation (cf. Figure 8).
Sensors 2022, 22, 5943 8 of 11

Figure 8. Frequency distribution of the single brightest pixel on the sensor at a laser power of 460 W
and a scan speed of 1 m/s.

In addition to the brightest pixel, a heat-affected zone can be observed in the thermog-
raphy images (cf. Figure 9a), the dimension of which depended on the energy input into the
surface, varying from approximately 2.2 mm at 460 W laser power and 0.05 m/s scan speed
to 0.8 mm at 105 W laser power and 1 m/s scan speed. Subsequently carried-out metallurgi-
cal examinations (cf. Figure 9b), in which the dimension of the melt pools were determined,
have shown that the areas of increased temperature visible in the thermography exceeded
the width of the melt pool.

(a) (b)
Figure 9. Comparison between: (a) A thermal image of an exposure with a laser power of 460 W and
a scan speed of 1 m/s and (b) the metallurgical microscopy of the aluminum plate. The melt pool has
has ahas a measured width
ameasured
measuredwidth
widthof
of 165.1and
of165.1
165.1μm
μm and a measured
µm andaameasured
measured depth
depth
depth of
of 282.6 μm.
of 282.6
282.6 μm.
µm.

4. Discussion
The test series carried out in this research demonstrates the high precision that the
synchronized galvanometer scanners can follow, even for complex paths. The brightest
pixel selected as a benchmark provides a reliable reference point for the operating laser.
The discrepancy between the strongly focused spot diameter and the resolution capability
of the camera system provides a clear drop-off between the zone of energy input and
the surrounding heat-affected area. In all combinations of power and traverse speed
investigated, a single pixel always stood out clearly, fitting the characteristic properties
Sensors 2022, 22, 5943 9 of 11

of the laser melting process of local extremely high temperatures and a rapid drop close
beside it.
The accuracy of the setup reaches or exceeds the resolution capability of the camera
system. Across all tests, oscillating deviations during the movement were not observed.
The shift within the field of view, nevertheless occurring with high repeatability, can clearly
be attributed to a distortion of the optical path that was not completely equalized for the
observed wavelengths.
The initial results of this study illustrate the combination of the benefits of on- and off-
axis monitoring within the SPIT approach. The system allows the use of wavelength ranges
in the SWIR that were previously described by Mohr [20] as favorable for observations in
the PBF LB/M process. In contrast with off-axis systems, there is no need to compromise
between resolution, frame rate and observable working space [7]. Furthermore, due to the
characteristics of the data acquisition consisting of a small movable measuring field, the
SPIT setup enables high frame rates and resolutions with a simultaneously large observable
working space. These attributes are particularly characteristic of on-axis systems [26]. In
addition, it is possible to design the travel paths of the two independent galvanometer
scanners differently, in principle.
By utilizing metallurgical investigations, the dimensions of the generated melt pools
could be mapped onto corresponding infrared images. It was noticeable that the heated
zone detected by the camera was significantly larger than the actual melt pool. This
indicates that even heated areas that have not been melted can be captured in detail with
the SPIT setup, and subsequently evaluated.

5. Conclusions
In this paper, the SPIT setup was introduced as a novel monitoring system that enables
high-resolution observation of the melt pool and the surrounding heat-affected zone in
the PBF-LB/M process. The use of two independent optical paths permitted the usage of
wavelength ranges favorable for thermography.
Initial trials with a variety of simple and complex laser paths approved the repeatabil-
ity, with very high precision of two synchronized galvanometer scanners, even at high scan
speeds. Furthermore, the point of interaction between the laser and the aluminum was
clearly recognizable as a single pixel with prominent brightness. The expansion of the melt
pool was limited to a very small spot in the field of view and surrounded by a significant
area of thermal influence, which was dependent on the test setup for laser energy and
travel speed.
In the next steps, dedicated upgrades in the optical path will take place. The currently
used 50 mm lens will be upgraded to an InfraTec telephoto lens with a focal length of
100 mm. Preliminary experiments with this advanced setup have shown a significant
improvement in the achievable resolution without a significant loss of signal strength.
In order to enable more control of incident radiation to reduce the influence of inter-
ferences and allow more precise temperature mapping, a narrow band-pass filter will be
installed into the sensor path. Tests already carried out have shown that an additional loss
of signal power remains within manageable limits.
The observed shifts of the laser spot within the field of view caused by the optical
refraction showed that a customized correction factor to equalize the movement completely
is necessary. This factor will be implemented after the conduction of additional trails to
map the currently existing distortion.
For future measurements of absolute temperatures, there will be investigations into
moving the focal point apart from the zone of direct laser interaction and towards more
stable areas regarding the aggregate state of the matter. Therefore, a predefined delay of the
sensor path in regard to the laser will be applied. This disengagement of the momentarily
linked scanners will be achieved via a custom software solution with a customizable gap
between laser and sensor focus.
Sensors 2022, 22, 5943 10 of 11

The goal of an obtainment of absolute temperatures requires accurate calibration of

the camera, including the entire optical path with all its elements.
Following these steps, the presented setup consisting of a high-speed SWIR camera
and two synchronized galvanometer scanners will enable the measurement of absolute
temperatures and temperature gradients in a defined region around the laser–material
interaction point in great detail. The information gained will then be used as indicators to
register anomalies occurring during manufacturing, to evaluate the process stability and
predict the achieved part quality.

Author Contributions: Conceptualization, D.H. and C.S.; methodology, D.H. and C.S. software,
C.S.; validation, C.S. and A.S.; formal analysis, D.H. and C.S.; investigation, D.H.; resources, J.H.
and A.S.; data curation, C.S. and D.H.; writing—original draft preparation, D.H. and C.S.; writing—
review and editing, D.H., C.S., A.S. and J.H.; visualization, C.S. and D.H.; supervision, A.S.; project
administration, J.H.; funding acquisition, D.H., A.S. and J.H. All authors have read and agreed to the
published version of the manuscript.
Funding: This research was conducted as part of the ViPAF project funded by StMWI—Bayern Inno-
vativ Bayerische Gesellschaft für Innovation und Wissenstransfer mbH, grant number 41-6562b/25/2-
VAL-2103-0006.
Data Availability Statement: The data presented in this study are available on request from the
corresponding author.
Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design
of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or
in the decision to publish the results.

References
1. Baumers, M.; Carmignato, S.; Leach, R. Introduction to Precision Metal Additive Manufacturing. In Precision Metal Additive
Manufacturing; CRC Press: Boca Raton, FL, USA, 2020; pp. 1–10.
2. Kempen, K.; Yasa, E.; Thijs, L.; Kruth, J.-P.; Van Humbeeck, J. Microstructure and Mechanical Properties of Selective Laser Melted
18Ni-300 Steel. Phys. Procedia 2011, 12, 255–263. [CrossRef]
3. Grasso, M.L.G.; Remani, A.; Dickins, A.; Colosimo, B.M.; Leach, R.K. In-Situ Measurement and Monitoring Methods for Metal
Powder Bed Fusion–An Updated Review. Meas. Sci. Technol. 2021, 32, 112001. [CrossRef]
4. Trapp, J.; Rubenchik, A.M.; Guss, G.; Matthews, M.J. In Situ Absorptivity Measurements of Metallic Powders during Laser
Powder-Bed Fusion Additive Manufacturing. Appl. Mater. Today 2017, 9, 341–349. [CrossRef]
5. Wehnert, K.K.; Ochs, D.; Schmitt, J.; Hartmann, J.; Schiffler, A. Reducing Lifecycle Costs Due to Profile Scanning of the Powder
Bed in Metal Printing. Procedia CIRP 2021, 98, 684–689. [CrossRef]
6. Ali, U.; Mahmoodkhani, Y.; Shahabad, S.I.; Esmaeilizadeh, R.; Liravi, F.; Sheydaeian, E.; Huang, K.Y.; Marzbanrad, E.;
Vlasea, M.; Toyserkani, E. On the Measurement of Relative Powder-Bed Compaction Density in Powder-Bed Additive Manufac-
turing Processes. Mater. Des. 2018, 155, 495–501. [CrossRef]
7. Zhang, Y.; Fuh, J.Y.H.; Ye, D.; Hong, G.S. In-Situ Monitoring of Laser-Based PBF via off-Axis Vision and Image Processing
Approaches. Addit. Manuf. 2019, 25, 263–274. [CrossRef]
8. Mohr, G. Measurement of Particle Emissions in Laser Powder Bed Fusion (L-PBF) Processes and Its Potential for In-Situ Process
Monitoring. In Proceedings of the Euro PM 2019, Maastricht, The Netherlands, 13–16 October 2019.
9. Dunbar, A.J.; Denlinger, E.R.; Heigel, J.; Michaleris, P.; Guerrier, P.; Martukanitz, R.; Simpson, T.W. Development of Experimental
Method for in Situ Distortion and Temperature Measurements during the Laser Powder Bed Fusion Additive Manufacturing
Process. Addit. Manuf. 2016, 12, 25–30. [CrossRef]
10. Kleszczynski, S.; zur Jacobsmühlen, J.; Reinarz, B.; Sehrt, J.T.; Witt, G.; Merhof, D. Improving Process Stability of Laser Beam
Melting Systems. In Proceedings of the Fraunhofer Direct Digital Manufacturing Conference, Berlin, Germany, 12–14 March 2014.
11. Schmidt, M.; Merklein, M.; Bourell, D.; Dimitrov, D.; Hausotte, T.; Wegener, K.; Overmeyer, L.; Vollertsen, F.; Levy, G.N. Laser
Based Additive Manufacturing in Industry and Academia. Cirp Ann. 2017, 66, 561–583. [CrossRef]
12. Mani, M.; Feng, S.; Lane, B.; Donmez, A.; Moylan, S.; Fesperman, R. Measurement Science Needs for Real-Time Control of Additive
Manufacturing Powder Bed Fusion Processes; NISTIR: Gaithersburg, MD, USA, 2015.
13. Höfflin, D.; Rosilius, M.; Seitz, P.; Schiffler, A.; Hartmann, J. Opto-Thermal Investigation of Additively Manufactured Steel
Samples as a Function of the Hatch Distance. Sensors 2022, 22, 46. [CrossRef]
14. Grasso, M.; Colosimo, B.M. Process Defects and in Situ Monitoring Methods in Metal Powder Bed Fusion: A Review. Meas. Sci.
Technol. 2017, 28, 044005. [CrossRef]
15. Everton, S.K.; Hirsch, M.; Stravroulakis, P.; Leach, R.K.; Clare, A.T. Review of In-Situ Process Monitoring and in-Situ Metrology
for Metal Additive Manufacturing. Mater. Des. 2016, 95, 431–445. [CrossRef]
Sensors 2022, 22, 5943 11 of 11

16. Krauss, H. Qualitätssicherung Beim Laserstrahlschmelzen Durch Schichtweise Thermografische In-Process-Überwachung; Herbert Utz
Verlag: Munich, Germany, 2017; Volume 325.
17. Mohr, G.; Altenburg, S.J.; Ulbricht, A.; Heinrich, P.; Baum, D.; Maierhofer, C.; Hilgenberg, K. In-Situ Defect Detection in Laser
Powder Bed Fusion by Using Thermography and Optical Tomography—Comparison to Computed Tomography. Metals 2020,
10, 103. [CrossRef]
18. Zenzinger, G.; Bamberg, J.; Ladewig, A.; Hess, T.; Henkel, B.; Satzger, W. Process Monitoring of Additive Manufacturing by Using
Optical Tomography. AIP Conf. Proc. 2015, 1650, 164–170.
19. Oster, S.; Maierhofer, C.; Mohr, G.; Hilgenberg, K.; Ulbricht, A.; Altenburg, S.J. Investigation of the Thermal History of L-PBF
Metal Parts by Feature Extraction from in-Situ SWIR Thermography. In Proceedings of the Thermosense: Thermal Infrared
Applications XLIII, Online. 12–16 April 2021; International Society for Optics and Photonics: Bellingham, WA, USA, 2021; Volume
11743, p. 117430C.
20. Mohr, G.; Nowakowski, S.; Altenburg, S.J.; Maierhofer, C.; Hilgenberg, K. Experimental Determination of the Emissivity of
Powder Layers and Bulk Material in Laser Powder Bed Fusion Using Infrared Thermography and Thermocouples. Metals 2020,
10, 1546. [CrossRef]
21. Bartlett, J.L.; Heim, F.M.; Murty, Y.V.; Li, X. In Situ Defect Detection in Selective Laser Melting via Full-Field Infrared Thermography.
Addit. Manuf. 2018, 24, 595–605. [CrossRef]
22. Boone, N. Near Infrared Thermal Imaging for Process Monitoring in Additive Manufacturing. Ph.D. Thesis, University of
Sheffield, Sheffield, UK, 2020.
23. Bamberg, J.; Zenzinger, G.; Ladewig, A. In-Process Control of Selective Laser Melting by Quantitative Optical Tomography. In
Proceedings of the 19th World Conference on Non-Destructive Testing, Munich, Germany, 13–17 June 2016; pp. 1–8.
24. Craeghs, T.; Clijsters, S.; Kruth, J.-P.; Bechmann, F.; Ebert, M.-C. Detection of Process Failures in Layerwise Laser Melting with
Optical Process Monitoring. Phys. Procedia 2012, 39, 753–759. [CrossRef]
25. Berumen, S.; Bechmann, F.; Lindner, S.; Kruth, J.-P.; Craeghs, T. Quality Control of Laser-and Powder Bed-Based Additive
Manufacturing (AM) Technologies. Phys. Procedia 2010, 5, 617–622. [CrossRef]
26. Clijsters, S.; Craeghs, T.; Buls, S.; Kempen, K.; Kruth, J.-P. In Situ Quality Control of the Selective Laser Melting Process Using a
High-Speed, Real-Time Melt Pool Monitoring System. Int. J. Adv. Manuf. Technol. 2014, 75, 1089–1101. [CrossRef]
27. Harbig, J.; Wenzler, D.L.; Baehr, S.; Kick, M.K.; Merschroth, H.; Wimmer, A.; Weigold, M.; Zaeh, M.F. Methodology to Determine
Melt Pool Anomalies in Powder Bed Fusion of Metals Using a Laser Beam by Means of Process Monitoring and Sensor Data
Fusion. Materials 2022, 15, 1265. [CrossRef]
28. Hartmann, J. High-Temperature Measurement Techniques for the Application in Photometry, Radiometry and Thermometry.
Phys. Rep. 2009, 469, 205–269. [CrossRef]
29. Wien, W. Temperatur Und Entropie Der Strahlung. Ann. Phys. 1894, 288, 132–165. [CrossRef]
30. Ostermann, F. Anwendungstechnologie Aluminium; VDI trimet; 3., neu bearb. Aufl.; Springer: Berlin/Heidelberg, Germany, 2014;
ISBN 978-3-662-43806-0.