Methods and systems for quality assessment and quality control in additive manufacturing processes

Abstract

Additive manufacturing process that includes: Monitoring the temperature of a section of a build plane during an additive manufacturing process using a temperature sensor while a heat source passes through the section of the build plane; Capturing a peak temperature associated with one or more passes of the heat source through the section of the build-up plane; Determining a limiting temperature by reducing the peak temperature by a predetermined amount; Determining a time interval in which the monitored temperature exceeds the limit temperature; Identifying a change in manufacturing conditions that is likely to lead to a manufacturing defect, based on the time interval; and Changing a process parameter of the heat source in response to a change in the manufacturing conditions.

Description

BACKGROUND OF THE INVENTION

[0001] Additive manufacturing, or the sequential build-up or creation of a part through a combination of material addition and energy input, can take various forms, with many concrete implementations and embodiments currently existing. Additive manufacturing can be performed using any of several different processes related to the formation of a three-dimensional part in virtually any shape. These processes include the layer-by-layer sintering of a liquid material using ultraviolet light, the layer-by-layer curing of a powdered material using a high-power laser, or the layer-by-layer melting of a granular raw material using an electron beam. Unfortunately, established methods for determining the quality of a part manufactured in this way have limitations.Conventional quality assurance inspections typically involve destroying the part after manufacturing. While destructive testing, as it allows for a precise examination of various internal areas, is a recognized method for verifying part quality, such testing is not feasible for a production part for obvious reasons. Therefore, there is a need for methods and systems for the non-destructive verification of the integrity of an additively manufactured part.

[0002] According to US 2015 / 0048064A1, a direct metal laser melting (DMLM) system for improving the build parameters of a DMLM component includes a confocal optical system configured to measure a melt pool size and / or a melt pool temperature. The DMLM system further includes a computing device configured to receive at least either the melt pool size or the melt pool temperature from the confocal optical system. Additionally, the DMLM system includes a controller configured to control the operation of a laser device based on at least one build parameter. BRIEF DESCRIPTION OF THE INVENTION

[0003] The described embodiments relate to a large subgroup of additive manufacturing processes that utilize an energy source in the form of a moving zone of intense heat energy. If the heat energy leads to the physical melting of the added material, these processes are generally referred to as smoldering processes. In smoldering processes, the material, which is added incrementally and continuously, is melted by the energy source in a manner similar to molten smoldering.

[0004] If the added material is in the form of powder layers, the heat source melts the progressively added powder after each layer has been added to the part being manufactured. This melting process creates a moving melt zone, referred to below as the heat bath. After solidification, these melted and solidified layers become part of the previously added, melted, and solidified layers beneath the new layer that forms part of the manufactured part. Because additive manufacturing processes can be lengthy and involve any number of heat bath cycles, it can be difficult to avoid even slight variations in the volume and temperature of the heat bath when using it to solidify the part.It should be noted that due to the high traverse speeds of the heating element and the complex patterns required to form a three-dimensional structure, additive manufacturing processes are typically controlled by one or more processors connected to a computer numerical control (CNC).

[0005] A primary objective of the described embodiments is to apply optical scanning techniques, such as quality inference, process control, or both, to additive manufacturing processes. Optical sensors can be used to monitor the evolution of physical phenomena during manufacturing by tracking the development of the associated physical process variables. "Optical" in this context can encompass the portion of the electromagnetic spectrum that includes the near-infrared (IR), visible, and near-ultraviolet (UV) ranges. Generally, the optical spectrum, also known as the visible spectrum, is assumed to extend over the wavelength range of 380 to 780 nm. However, near-UV and IR could extend over wavelength ranges down to 1 nm and up to 3000 nm, respectively.The sensor measurements collected by optical sensors can be used to determine in-process quality metrics (IPQMs). Half-power bandwidth (HPBW), based on thermal data acquired using a pyrometer or photodiode, represents one such IPQM.

[0006] HPBW is a metric that responds to changes in user-defined laser powder bed melting process parameters such as beam power, beam traverse speed, path spacing, and the like. HPBW allows the determination of the pyrometer's field of view position relative to the laser scan pattern and path size. HPBW can be used for analysis by comparing IPQM with a baseline dataset. The resulting IPQM can be calculated and displayed for each scan. IPQM comparisons with the baseline dataset can indicate manufacturing defects and be used to generate control signals for process parameters. HPBW enables a more accurate estimation of the correct thermal profile from the data acquired by the pyrometer.

[0007] An additive manufacturing process is disclosed, comprising the following: monitoring the temperature of a section of a build plane during an additive manufacturing process using a temperature sensor while a heat source passes through the section of the build plane. The process further comprises detecting a peak temperature associated with one or more passes of the heat source through the section of the build plane and determining a limit temperature by reducing the peak temperature by a predetermined amount. The process uses the limit temperature to determine a time interval during which the monitored temperature exceeds the limit temperature and, based on this time interval, identifies a change in the manufacturing conditions that could lead to a manufacturing defect. In response to the change in the manufacturing conditions, the process modifies one or more process parameters associated with the heat source.

[0008] An additive manufacturing process is disclosed, comprising: performing an additive manufacturing process using an energy source and receiving voltage data associated with a stationary photodiode during a scan of the energy source across a section of a powder bed to produce a part. The process further comprises detecting a peak voltage associated with one or more passes of the energy source across the powder bed section and determining a threshold voltage by reducing the peak voltage by a predetermined amount. The process further comprises determining a time interval during which the voltage data exceeds the threshold voltage and identifying, based on this time interval, a change in the manufacturing conditions that could lead to a manufacturing defect.The process changes a process parameter of the energy source in response to a change in the manufacturing conditions. List of characters

[0009] The following detailed description, in conjunction with the accompanying figures, in which the same reference symbols denote the same structural elements, contributes to an easy understanding of the revelation, whereby: Fig. Figure 1A shows a schematic representation of a system with a strong heat source, in this specific case a laser beam. Fig. Figure 1B shows a schematic representation of a system with a strong heat source, in this specific case an electron beam. Fig. 2A is an example of the field of view of a pyrometer set up as an Eulerian sensor, Fig. Figure 2B schematically shows the field of view of the pyrometer and the path of the thermally affected and molten material formed by the moving heat source. Fig. Figure 2C shows an example of a scan pattern that can be traversed by a heat source. Fig. 2D - Fig. 2E show different alternative scan patterns, Fig. 3A shows the raw data of a stationary (Euler) pyrometer when multiple scans of a heat source pass through the field of view; Fig. 3B shows the test results of a stationary (Euler) photodiode when multiple scans of a heat source pass through the field of view, Fig. 3C shows the test results of a stationary (Euler) pyrometer when multiple scans of a heat source pass through the field of view, Fig. 4 shows a subset of the temperature profile generated by a stationary pyrometer and linked to process quality metrics, Fig. 5 shows a peak temperature profile of the stationary pyrometer and the parameters for determining a full width at half maximum (FWHM), Fig. 6 shows a temperature profile that is used to calculate a mean half-width based on more than one peak temperature, Fig. 7A shows a heat source scan pattern and the section covered by a sensor's field of view, Fig. 7B the one with the field of vision of Fig. 7A linked temperature profile shows, Fig. 8A shows a heat source scan pattern and the section covered by a sensor's field of view, Fig. 8B, the one with the field of vision of Fig. The linked temperature profile shown in 8A is... Fig. Figure 9 shows a flowchart for process control at a high level of abstraction, Fig. Figure 10 shows a process in which data recorded by an optical sensor, such as a non-imaging photodetector, can be processed to characterize a build process of additive manufacturing. Fig. 11A - Fig. 11D illustrative representations show how multiple scans can contribute to the energy introduced into individual raster areas, Fig. 12A - Fig. Figure 12F illustrates how a grid can be dynamically created to characterize and control an additive manufacturing process, and Fig. Figure 13 shows an example of a closed control loop that includes a feedback control loop for setting up and maintaining feedback control of an additive manufacturing process. DETAILED DESCRIPTION OF SPECIFIC EXECUTION FORMS

[0010] Fig. Figure 1A shows an embodiment of an additive manufacturing system in which one or more optical sensor devices are used to determine the full width at half maximum (FWHM). The FWHM responds to changes in process parameters such as the power of the energy source, the speed of the energy source, and the path spacing. In the additive manufacturing system of Fig. In embodiment 1A, a laser 100 is used as the energy source. The laser 100 emits a laser beam 101, which passes through a partially reflective mirror 102 and enters a deflection and focusing system 103, which then projects the beam onto a small area, the beam interaction area 104, on a work platform 105. In some embodiments, the work platform comprises a powder bed. Due to the high material temperatures, the beam interaction area 104 emits optical energy 106.

[0011] In some embodiments, the deflection and focusing system 103 can be configured to detect a portion of the optical energy 106 emitted by the beam interaction area 104. The partially reflective mirror 102 can reflect the optical energy 106, represented as an optical signal 107. The optical signal 107 can be queried by several optical sensors 109, each of which receives a portion of the optical signal 107 via a series of further partially reflective mirrors 108. It should be noted that in some embodiments, the additive manufacturing system may have only one optical sensor 109 with a fully reflective mirror 108.

[0012] It should be noted that the detected optical signal 107 may not have the same spectral content as the optical energy 106 emitted from the beam interaction area 104, because the signal 107 has undergone some attenuation after passing through several optical elements, such as 103, 102, and 108. The optical elements may each have their own transmission and absorption properties, resulting in inconsistent attenuation and thus limiting certain parts of the energy spectrum emitted by the beam interaction area 104. The data generated by the optical sensors 109 may correspond to an amount of energy transferred to the work platform.

[0013] Examples of optical sensors 109 include, but are not limited to, transducers for converting optical signals into electrical signals, such as pyrometers and photodiodes. The optical sensors can also include low- or high-speed spectrometers and cameras operating in the visible, ultraviolet, or infrared frequency spectrum. The sensors 109 are located in a reference frame that moves with the beam; that is, they "see" all areas toward which the laser beam is directed and are capable of detecting optical signals 107 from all areas of the work platform 105 upon which the laser beam 101 strikes. Since the optical energy 106 detected by the deflection and focusing system 103 travels a path nearly parallel to the laser beam, the sensors 109 can be considered axial sensors.

[0014] In some embodiments, the additive manufacturing system can comprise a first stationary sensor 110 and a second stationary sensor 124, which are located in a stationary reference frame with respect to the laser beam 101. Stationary sensors can be considered Eulerian sensors. The first stationary sensor 110 can have a specific first field of view 111, which may be very narrow and cover only a portion of the work platform 105. The second stationary sensor 124 can have a specific second field of view 126, which can encompass a larger area than the first field of view 111. The second field of view 126 can span an area with a size extending to the entire work platform 105.

[0015] Examples of the first stationary sensor 110 and the second stationary sensor 124 may include, but are not limited to, pyrometers, photodiodes, spectrometers, high-speed or low-speed cameras operating in the visible, ultraviolet, or IR spectral range. In some embodiments, the sensors may be temperature sensors that process raw voltage or current data and output temperature data. The first stationary sensor 110 and the second stationary sensor 124 are not aligned with the power source and can be considered non-axial sensors. In some embodiments, the second stationary sensor 124, which has the second field of view 126, may be configured to detect temperature changes in substantially any area of ​​the top of the work platform 105.In some embodiments, the first stationary sensor 110 can be configured to provide the second stationary sensor 124 with calibration information, enabling the second stationary sensor 124 to accurately detect the temperature of any point on the top of the work platform 105. A further description of the sensor calibration is contained in U.S. Patent Application No. 14 / 945,247, filed on November 18, 2015, the disclosure of which is hereby incorporated herein by reference for all purposes.

[0016] The sensors could also be sensors that combine several physical measurement modalities, such as a laser-ultrasonic sensor that can actively excite or "ping" the deposition with a laser beam and then measure the resulting ultrasonic waves or "vibration" of the structure with a laser interferometer to measure or predict mechanical properties or the mechanical integrity of the deposition during build-up. The laser-ultrasonic sensor / interferometer system can measure the elastic properties of the material, which can provide information about the material's porosity and other material properties. Furthermore, the laser-ultrasonic sensor / interferometer system can measure the formation of defects that lead to material vibrations.In some embodiments, the HPBW can be determined from an observed temperature generated from pyrometer data acquired by the first stationary sensor 110 and the second stationary sensor 124.

[0017] Furthermore, the mechanical device distributing the powder, the recoating arm 112, can include contact sensors 113. These sensors can be accelerometers, vibration sensors, etc. Finally, other types of sensors 114 can also be present. These can include contact sensors such as thermocouples for measuring macrothermal fields or acoustic emission sensors, which can detect cracks and other metallurgical phenomena in the deposition during build-up. The contact sensors can be used during powder application to characterize the operation of the recoating arm 112. The data acquired by the optical sensors 109 and the stationary sensors 110 can be used to record process parameters related to the recoating arm 112. Accordingly, irregularities on the surface of the applied powder can be detected and addressed by the system.Rough surfaces resulting from changes in the powder coating process can be characterized by contact sensors 113 to anticipate possible problem areas or irregularities in the resulting part.

[0018] In some embodiments, the laser beam 101 can melt a raised area in the powder coating, so that the subsequent powder layer has a corresponding raised area. Eventually, this raised area may touch the recoating arm 112, potentially damaging it and resulting in further inconsistencies in the applied powder. Accordingly, embodiments of the present invention can detect inconsistencies in the applied powder before they lead to inconsistencies in the build-up area on the work platform 105. Many variations, modifications, and alternatives are familiar to the person skilled in the art.

[0019] In some embodiments, the optical sensors 109, the stationary sensors 110, the contact sensors 113, and other sensors 114 can be configured to generate raw data during the process. In further embodiments, the optical sensors 109, the stationary sensors 110, the contact sensors 113, and other sensors 114 can be configured to process the data and generate reduced-order sensor data.

[0020] In some embodiments, a computer 116 is provided with a processor 118, a computer-readable medium 120, and an input / output interface 122, and is coupled with suitable system components of the additive manufacturing system to collect data from the various sensors. The data received by the computer 116 can include raw process data from sensors and / or reduced-order sensor data. The processor 118 can use raw process data from sensors and / or reduced-order sensor data to determine the power of the laser 100 and control information, including coordinates related to the work platform 105. In further embodiments, the computer 116, which includes the processor 118, the computer-readable medium 120, and an input / output interface 122, can enable the control of the various system components.The computer 116 can send, receive and monitor control information linked to the laser 100, the work platform 105 and the recoating arm 112 to control and adjust the respective process parameters for each component.

[0021] The processor 118 can be used to perform calculations using data acquired by the various sensors to generate process quality metrics. In some embodiments, the data generated by the stationary sensors 110 can be used to determine the HPBW (High Heat Flow Rate) during the build process. Control information associated with the movement of the heat source across the build plane can be received by the processor. The processor can then use this control information to correlate data from the stationary sensors 110 with the corresponding position of the heat source. The correlated data can then be combined with the HPBW to determine changes in process parameters.In some embodiments, the HPBW and / or other metrics can be used by the processor 118 to generate control signals for process parameters, such as laser power, laser speed, path spacing, and other process parameters, in response to HPBW or other metrics that are outside the desired ranges. This allows for the correction of a problem that would otherwise lead to the production of a defective part. In embodiments where multiple parts are produced simultaneously, timely corrections of the process parameters in response to metrics that are outside the desired ranges can prevent adjacent parts from receiving too much or too little energy from the heat source.

[0022] In some embodiments, the input / output interface 122 can be configured to transfer acquired data to another location. The input / output interface can also be configured to receive data from another location. The received data can include basic datasets, historical data, post-processing data, and classification data. The other computer system can calculate process quality metrics from the data transferred by the additive manufacturing system. The other computer system can then transmit information to the input / output interface 122 in response to specific process quality metrics.

[0023] In the case of an electron beam system, it shows Fig. 1B Possible configurations and arrangements of sensors. The electron beam gun 150 generates an electron beam 151, which is focused by the electromagnetic focusing system 152 and then deflected by the electromagnetic deflection system 153, resulting in a finely focused and aligned electron beam 154. The electron beam 154 creates a hot beam-material interaction zone 155 on the workpiece 156. Optical energy 158 is emitted, which can be detected by a series of sensors, each with its own field of view, which in turn can be localized to the interaction zone 155 or encompass the entire workpiece 156. Furthermore, the sensors 159 can have their own tracking and scanning system for monitoring the electron beam 154 as it moves across the workpiece 156.

[0024] Regardless of whether the sensors 159 have optical tracking or not, they can consist of pyrometers, photodiodes, spectrometers, and high-speed or low-speed cameras operating in the visible, UV, or IR spectral range. The sensors 159 can also be sensors that combine a variety of physical measurement methods, such as a laser-ultrasonic sensor that can actively excite or “ping” the deposition with a laser beam and subsequently measure the resulting ultrasonic waves or “vibration” of the structure with a laser interferometer to measure or predict mechanical properties or the mechanical integrity of the deposition during build-up. Furthermore, contact sensors 113 can be located on the recoating arm. These sensors can be accelerometers, vibration sensors, etc.Finally, other types of sensors 114 may also be present. These may include contact sensors such as thermocouples for measuring macrothermal fields or acoustic emission sensors that can detect cracks and other metallurgical phenomena in the deposition during build-up. In some embodiments, one or more thermocouples may be used to calibrate the temperature data acquired by the sensors 159. It is noted that the [missing information] in connection with the [missing information] Fig. 1A and Fig. The sensors described in 1B can be used in the manner described to characterize the quality of any additive manufacturing process with sequential material build-up.

[0025] Fig. Figure 2A shows an example of a pyrometer's field of view configured as an Eulerian sensor. The first stationary sensor 110 and the second stationary sensor 126 can be pyrometers configured as Eulerian sensors. An Eulerian sensor is based on an Eulerian frame of reference. An Eulerian frame of reference specifies the quantities associated with a transport phenomenon by observing values ​​at specific time intervals that correspond to specific points in space. The grid of points is fixed in space, and the medium "flows" through this grid. This is comparable to sitting on the bank of a river and watching the flowing water. In some embodiments, the Eulerian frame of reference is a stationary frame of reference through which the energy beam passes.

[0026] In Fig. In Figure 2A, the Eulerian reference frame is represented by a circular field of view 200. The temperature field within a field of view 200 can change and can be both spatially and temporally dependent. Each individual surface element 202 contributes to the temperature averaged over all, which is determined in proportion to the fraction of its area to the total area of ​​the pyrometer's field of view. The small differential surface element 202 is in Fig. Figure 2A is shown schematically. The field of view 200 can comprise a variety of radiating surface elements 202. A time interval can be assigned to an area of ​​the assembly plane covered by the differential surface element 202 as the scan passes through the area. In general, these surfaces are distributed within the field of view 200 of the pyrometer at a specific radius R 204 from the center 210 and at a specific angular orientation θ 206. Furthermore, each individual surface 202 can have a different temperature-dependent emissivity.

[0027] Fig. Figure 2B shows a schematic view of the pyrometer's field of view 200 and the region 212 of the thermally affected and molten material produced by the moving heat source. The pyrometer's field of view 200 is represented as a circular field of view with a radius R 204. The area heated and melted by the moving heat source is shown as a rectangular region 212 that intersects the pyrometer's field of view 200. This region generally varies in size, area, and the position at which it intersects the field of view.

[0028] The heat source is not an instantaneous heat source; that is, it cannot be switched on immediately, nor does it instantly emit a finite amount of heat. Rather, the heat source is a moving, continuous heat source. As the heat source moves through the field of view while passing over the heated areas, the temperature in different regions within the field of view steadily increases and decreases. The heated areas correspond to region 212 of the thermally affected material. Therefore, the observed temperature should be interpreted as the time-integrated mean of the time-dependent thermal behavior of the hot and cold zones, each weighted by its respective area fraction. Fig. Figure 2C shows an example of a scan pattern along which a heat source can be moved. As shown, the field of view 200 can cover a relatively small part of the scan pattern 216, so only the heating and cooling within the field of view 200 can be accurately quantified. It should be noted that other scan patterns are conceivable and can include narrower or wider scan patterns, executed at different speeds and output powers. The scan rate, power, and scan pattern each influence how much energy is delivered during the additive manufacturing process.

[0029] The Fig. 2D - Fig. 2E show different alternative scan patterns. Fig. 2D shows how the scan pattern can be divided into a checkerboard pattern 220, whose fields are scanned successively from left to right 222 and from top to bottom 224.

[0030] The numbers 226 in each square 220 indicate the order in which the squares 220 of the chessboard are scanned. Fig. In 2E, the same checkerboard pattern is displayed, but now the scan order for the individual checkerboard squares is randomized. Regardless of the specific scan pattern or scan strategy used, the laser-based process can obviously involve short, discrete scan lengths with a start, a stop, and a path length.

[0031] If a scan pattern, similar to those in the Fig. 2D or Fig. 2E shown scan patterns, with an Eulerian photodiode such as the first stationary sensor 110 (see Fig. 1A) When monitored, the data returning to the photodiode comprises many individual signals, each representing a specific scan over a particular path length. In some embodiments, the individual signals can be filtered according to their path length. The apparent intensity of the signal, as observed by the photodiode, is a function of this path length. This is because, at the beginning of the scan, the photodiode intensity is zero or very low, as the laser has just been switched on. During the scan, the laser generally heats a portion of the powder bed, and the heated portion becomes a region of molten metal that emits more energy in the form of light. In response to the energy emitted by the region of molten metal, the intensity measured by the photodiode increases. As further shown from Fig. As can be seen in Figure 2C, the intensity generally reaches a maximum value when the energy source, which is guided along the scan pattern 216, is within the field of view 200. The highest temperature can generally be assigned to the scan with the longest dwell time in the field of view 200, such as scan 238. Fig. 2C. In other embodiments, a change in beam power or beam traverse speed can cause the highest temperature to be assigned to a scan with a shorter residence time. In some embodiments, the raw signal from the photodiode may exhibit natural fluctuations and scattering because the light intensity changes throughout the process. This is due to the highly chaotic nature of various physical processes, such as the interactions between the laser energy source and the powder, the chaotic movement of the molten metal, the changing view factor from this small hotspot to the photodiode, and the solidification of the powder bed after the energy source has traversed along the scan pattern 216.

[0032] An observed temperature can be calculated from sensor data generated by the stationary and / or axial sensors. In some embodiments, the stationary sensor can be a pyrometer with a field of view significantly larger than the heated zone. This significantly larger field of view can cause the temperature of the heated zone observed within the field of view to deviate from the actual temperature of the heated zone and / or the weld pool. In some embodiments, the various aspects relating to the Fig. 1A and Fig. The sensors described in Section 1B are used to normalize the observed temperature to an actual temperature. A further description dealing with the determination of the actual temperature is contained in U.S. Patent Application No. 14 / 945,247, filed on November 18, 2015, the disclosure of which is hereby incorporated by reference in its entirety for all purposes.

[0033] Fig. Figure 3A shows an exemplary signal profile measured by a pyrometer according to embodiments of the present invention. The signal profile represents the raw temperature data of a fixed-field-of-view (Euler) pyrometer when multiple scans of a heat source traverse the field of view 200. As the moving heat source approaches the field of view of the temperature sensor, there is an increase in the background temperature 302. The increase in the background temperature 302 may be associated with scans such as 232 or 234, which are in Fig. 2C are shown and run near the field of view 200, but do not enter it. At some point, the moving heat source is completely within the field of view 200, and higher temperatures occur, short-term thermal spikes 304. The short-term thermal spikes 304 can be associated with scans such as 236, 238, 240, 242, which are shown as in Fig. 2C is shown in the field of view 200. When the heat source then disappears from the field of view and moves to other areas of the material, a more slowly decaying thermal transient 306 occurs. The transients of Fig. 3A are also associated with phase changes, but these are only visible or recognizable upon closer inspection of the faster, thermal transients 304 with higher temperature.

[0034] The profile of Fig. 3A arises from the fact that the pyrometer's field of view 200, through which the heat source scans, is circular. During a scan, the heat source enters the field of view 200 at or near an initial point 308. As the heat source approaches the center of the field of view, the path length of the heated material within the field of view increases, and the temperature data output by the pyrometer responds accordingly. As the heat source passes through the field of view, the temperature continues to increase, reaching a maximum 310 while the heat source is within the field of view 200. This maximum 310 generally occurs at or near a time , where time is represented by the x-axis 312, corresponding to the time the heat source leaves the field of view 200 after passing through a central section of the field of view 200 (see, for example, scan 238, as in Fig. (2C shown). When the scanning beam leaves the field of view 200, the intensity of the thermal signature decreases again.

[0035] The Fig. 3B and Fig. Figures 3C show the temperature profile measured by a photodiode and a pyrometer, respectively, for a single scan of the heat source through the respective sensor's field of view. The temperature profile measured by both sensors shows the increase in the background temperature 302, which is associated with the energy source as it moves along the scan pattern close to the field of view. Both temperature profiles contain transient thermal spikes. A difference in the physical properties of the individual sensors leads to differences in the magnitude of these spikes. The maximum 310 is reached at approximately the same time, with time represented by the x-axis 312. Furthermore, both temperature profiles show the more slowly decreasing thermal transient 306 as the heat source moves further away from the field of view.Finally, both temperature profiles show a thermal holding region or inversion region 314, which is associated with a phase transition in the material, which in the case of the cooling curve is the transition from liquid to solid.

[0036] Fig. Figure 4 shows an enlarged view of the temperature profile 400 near the point in Fig. 3 maximums shown 310. Fig. 4 shows that T PEAK The peak temperature is 402. In some embodiments, a limit temperature T may be specified. H 404 is half the peak temperature. The limit temperature 404 is illustrated by line 406, which intersects the temperature profile 400. In other embodiments, the limit temperature differs from half the peak temperature and can be determined by T PEAK Dividing by e is shown in Eq. (1) below: TH=TPeake where e represents Euler's number: e ≈ 2.71828. In other embodiments, the denominator e can be replaced by a predetermined value to reduce the peak temperature to the limit temperature of 404. For example, desired process parameters, other IPQMs, and peak temperature data can be analyzed using machine learning or statistical techniques to determine an ideal limit temperature of 404.

[0037] In alternative embodiments of the present invention, the limit temperature 404 can be modified to adapt to specific types of additive manufacturing processes. For example, high-performance applications can utilize a half-value width that allows for higher energy and / or a higher traverse speed of the heat source. In further embodiments, the half-value width can be adjusted according to the requirements of a specific additive manufacturing process.

[0038] Fig. Figure 5 illustrates a temperature profile 500 for a single peak temperature, such as the one in Fig. The peak temperature shown is 402. The x-axis (504) represents time. Line 406 intersects the temperature profile at the limit temperature (404). Fig. Figure 5 shows two times at which line 406, representing the limit temperature, intersects temperature profile 500. A first time t HR is related to the heating rate. The first time t HR This is illustrated by line 502, which extends from the intersection point 508 of the limit temperature line 406 with the temperature profile 500 to the x-axis 504. A second time t CR is linked to the cooling rate. The second time t CR This is illustrated by line 512, which extends from the second intersection point 514 of the limit temperature line 406 with the temperature profile 500 to the x-axis 504. In some embodiments, using t HRand t CR A metric can be calculated. According to embodiments of the present invention, the metric can be expressed as the HALF-WIDTH t, as specified in equation (2). HPBW to be defined: tHPBW=(tCR−tHR)

[0039] In some embodiments, in addition to the limit temperature, fit lines representing the heating and cooling rates can also be determined. In some embodiments, the fit lines can be calculated using a subset of the data points from the temperature profile 500. This subset can be determined, for example, by a minimum and maximum limit temperature or limit times corresponding to a predefined position of the heat source along the scan pattern. The fit lines can be determined using a simple procedure by connecting the minimum and maximum temperatures. In some embodiments, a more accurate method for determining the best fit line, such as the least squares method, can use all selected data points from the temperature profile 500 to determine a fit line. Many variations, modifications, and alternatives are familiar to the average person skilled in the art.

[0040] Fig. Figure 5 further illustrates a heating rate fit line 516 and a cooling rate fit line 522. The heating rate fit line 516 can be used to determine the first time t HR to determine the time associated with the heating rate. The time associated with the heating rate can be calculated based on the first fit line intersection point 518 of the heating rate fit line 516 with the limit temperature 404. The cooling rate fit line 522 can be used to determine the second time t. CR to determine the time associated with the cooling rate. The time associated with the cooling rate can be calculated based on the second fit line intersection point 522 of the cooling rate line 522 with the limit temperature 404. The half-width t HPBW can be calculated using the relationship given above. In some embodiments, the use of a fit line relative to the temperature profile 500 may result in a different half-width t. HPBWlead to, but the unified use of both models should provide a metric that indicates changes in process parameters that affect the quality of a finished part.

[0041] The full width at half maximum (FWHM) can be used to predict, measure, and control process parameters of additive manufacturing. Process parameters include, for example, beam power, beam feed rate, path spacing, etc. To illustrate this, an increase in the FWHM can be used to predict, measure, and control process parameters of additive manufacturing. HPBW This indicates that the blasting power has increased. For example, if the blasting power increases, the peak temperature rises, and consequently, so do the rise time to the peak temperature and the fall time. Accordingly, the associated width of the thermal profile will increase, as it takes longer for the powder bed to reach the higher peak temperature and longer to cool from that higher peak temperature.

[0042] In other embodiments, changes in the t HPBW to indicate changes in the blast feed rate. For example, if the blast feed rate is increased, the t decreases. HPBW , because the associated heat cycles are faster and the overall heat intensity in J / mm² decreases. As a consequence of this decrease, the width of the thermal profile shrinks. Additionally, t HPBW Display changes in track spacing. For example, the t decreases. HPBW This occurs when the path spacing increases. As the path spacing increases, the time until the next laser pass also increases, allowing more time for the previous laser pass to cool down. In this case, the thermal profile shrinks because the ambient temperature is lower and therefore the cooling rate is faster.

[0043] The half-width t HPBWIt can predict important process changes. The process parameters form the physical basis for fluctuations in the full width at half maximum (FWHM) t. HPBW Since the full width at half maximum (FWHM) t HPBW By reacting to fluctuations in process parameters, the metric can accurately capture the status of the additive manufacturing process. In some implementations, changes in process parameters that lead to errors in the additive manufacturing process can be detected using the full width at half maximum (FWHM). In response, the process can adjust the corresponding process parameter to prevent the formation of defects at the build level.

[0044] In some embodiments, thermal profiles associated with more than one scan can be used to generate a process quality metric based on the full width at half maximum (FWHM) t. HPBWto determine. For example, all peaks associated with the energy source being guided directly through the equator of the field of view can be used to determine a mean full width at half maximum (FWHM) t. HPBW(avg) be used. Fig. Figure 6 shows an embodiment in which the mean full width at half maximum (FWHM) t HPBW(avg) It is used to monitor process parameters. The mean full width at half maximum (FWHM) can be calculated from a number N of peaks centered around a highest peak 3. Fig. Figure 6 illustrates a temperature profile 600 with N = 2. Fig. 6. For each peak 1-5, the half-width t is calculated. HPBW definitely. t HPBW Peak 1 is represented by line 602, t HPBW from Peak 2 through line 604, t HPBW from Peak 3 through line 606, t HPBW from Peak 4 through line 608 and t HPBW from peak 5 through line 610. The mean of the five peaks can be calculated using equation (3): tHPBW(avg)=tHPBW(1)+tHPBW(2)+tHPBW(3)+tHPBW(4)+tHPBW(5)5

[0045] Furthermore, the mean full width at half maximum (FWHM) can be a meaningful metric with less susceptibility to errors associated with measuring a single scan.

[0046] In addition to the process parameters, the distance between the peaks can be determined based on the full width at half maximum (FWHM) t. HPBW The distance between the peaks can be determined. A time interval associated with the peak spacing can be used to determine the relative position of the pyrometer's field of view with respect to the path pattern. The distance can be used to determine the laser scan pattern and path size. Fig. 7A and Fig. Figure 7B illustrates an embodiment in which the field of view is centered on the orbits and the distance between all peaks is approximately the same. Fig. Figure 7A illustrates a heat source scan pattern 750 on a build plane 752. Only a portion of the build plane 752 is covered by the field of view 702 of a sensor. The heat source moves along the scan pattern 750.

[0047] In some embodiments, the heat source can traverse the field of view 702 six times, 704, 706, 708, 710, 712 and 714. Each passage through the field of view is accompanied by a peak of Fig. 7B linked. For each peak, a full width at half maximum (FWHM) t can be determined. HPBW The distance between the respective half-widths t can be determined. space can be done using the cooling rate t CR and the heating rate t HR can be calculated from two adjacent peaks. For example, the first distance t can be calculated. s1 716 using the t belonging to peak 704 CR and of the t belonging to peak 706 HRThe second and subsequent distances can be calculated similarly, resulting in a sequence of times corresponding to the intervals between heat source scans (t). s1 , t s2 , t s3 .... t sN ) are linked. The field of vision is located in the center of the orbits when t s1 ≈ t s2 ≈ t s3 ≈ t sN .

[0048] Fig. 3A and Fig. Figure 8B illustrates an alternative embodiment in which the field of view is not located in the center of the tracks. Fig. Figure 3A illustrates a heat source scan pattern 750 on a build plane 752. Only a portion of the build plane 752 is covered by the sensor's field of view 702. Furthermore, the sensor's field of view 702 is not centered on the heat source scan pattern 750.

[0049] In some embodiments, the heat source can traverse the field of view 702 six times: 720, 722, 724, 726, 728, and 730. There are six traverses because the heat source does not emit energy in the sections of the heat source scan that extend beyond the boundary 754 of the assembly plane 752. Each traverse of the field of view is characterized by a peak of Fig. 8B linked. For each peak, a full width at half maximum (FWHM) t can be determined. HPBW The distance between the respective half-widths t can be determined. space can be done using the time t associated with the cooling rate CR and the time t associated with the heating rate HR calculated from two adjacent peaks. The times associated with the cooling rate and the heating rate were determined using Fig. 5 explained in detail. As from Fig. As can also be seen in 8B, the first distance t s1 732 using the t belonging to peak 720 CRand the t belonging to peak 722 HR to be calculated. The second distance t s2 734 can be calculated in a similar way, but is measurably larger than t due to the time the heat source needs to move along the heat source scan pattern 750 and return to the field of view. s1 The subsequent intervals can be calculated similarly, resulting in a sequence of times linked to the intervals between heat source scans (t s1 732, t s2 734, t s3 736, t s4 738 and t s5 740). The sequence of times linked to the intervals between the peaks shows that the visual field 702 is not centered on the orbits when ts1≈ts3≈ts5 <ts2≈ts4.

[0050] Features of the additive manufacturing process, such as the heating rate, cooling rate, and peak temperature described above, can represent the most important metallurgical characteristics. The full width at half maximum (FWHM) can be used to track these features scan by scan and layer by layer. These features could be used to establish a statistical process control method and answer the question: “Is the process under control?”

[0051] The following three types of process control can be applied to the described procedures, which are at least partly based on the principles outlined in the Fig. 5 and Fig. The 6 shown half-width or mean half-width are based on: First, process intervention or stopping or interrupting a process because one or more critical process characteristics are outside a certain range; second, inter-layer process control or changing process parameters between layers in an additive manufacturing process based on measurements taken during the previous layer, quality or feature metrics calculated from such measurements, and a decision algorithm that decides whether these features are within certain ranges and, if not, how adjustments can be made to process parameters such as heat source power and traverse speed to bring the half-width or other quality metric back into the specified ranges.The third type of process control that can be applied is intra-layer or scan-level process control, in which power, feed rate, or other process parameters can be changed so that certain quality metrics or quality characteristics remain within certain ranges.

[0052] The third type of process control is the fastest and requires the fastest control loop. The first form of process control can be considered an open-loop control system with only one outcome; that is, the process is stopped if the conditions deviate too far from the setpoint. The second type represents a slower form of real-time control and adjusts the parameters only from layer to layer.

[0053] Fig. Figure 9 shows a flowchart for a high-level process control system utilizing the previously described process features. The diagram illustrates the process flow for intra-layer or scan-by-scan control. With this type of control, a single scan is performed, calculations are carried out, and adjustments are made as needed before the next scan. This is intended as a fast control loop that makes changes in a millisecond or even less. At Figure 800, thermal measurements are taken using one or more Eulerian sensors. In some embodiments, the thermal measurements are obtained as voltage data. At Figure 901, the thermal measurements can be corrected to reflect an actual temperature.At 902, features such as those discussed above are extracted, which could include, but are not limited to, features like the full width at half maximum (FWHM), the mean FWHM, and the time between a first FWHM and a second FWHM. These are features that are metallurgically relevant to the material and the deposited layer of the additive manufacturing process.

[0054] At step 903, it is determined whether these characteristics are within the prescribed ranges known to correspond to normal process behavior and to produce acceptable parts. If the answer is yes, the process continues at step 904 with the next scan using the same process variables / parameters. It should be noted that there can be hundreds or thousands of scans within a single layer of an additively manufactured part, and thousands of such layers per part. If the result of the query in step 903 is no, the process flow continues at step 905 to a decision at step 906. At step 906, a methodology is applied that can make a decision based on the magnitude and direction of the observed deviations.This decision logic can be a reduced-order process model, a lookup table or database, a heuristic such as a neural network, or any other algorithmic system that decides which process variables or parameters should be changed, by how much, and in which direction (increase or decrease). For example, a change in process variables or parameters could take the form of changes to the heat output power of the heat source, the feed rate, and the scan pattern of the heat source, thereby changing the amount of energy introduced into one or more layers of a part.

[0055] Then, at step 907, these new process parameters are used to perform the next scan based on the data from the previous scan, and the process is repeated until the layer and ultimately the part are complete. Generally, higher power and a lower traverse speed of the heat source result in more heat being delivered to the component. Delivering larger amounts of heat effectively reduces the solidification rate. Therefore, to correct a condition where solidification occurs too quickly, additional heat can be added to the system. Conversely, if the material solidifies too slowly, less energy can be delivered to the part, thereby increasing the solidification rate. In general, the material's solidification rate is very important, as excessive cooling rates negatively impact the quality of the finished part.Another way to adjust the amount of heat delivered to a specific layer or area is to modify the scan pattern. For example, a scan pattern with closely grouped passes would deliver relatively more heat to the component than another laser using the same settings but with a wider scan pattern. INTEGRATION OF HALF-WIDTH INTO AN AGGREGATION-BASED ANALYSIS SYSTEM

[0056] Fig. Figure 10 shows a process in which data recorded by an optical sensor, such as a non-imaging photodetector, can be processed to characterize a build process in additive manufacturing. Figure 1002 receives raw sensor data, which can include both build-plane intensity data and energy source control signals that are correlated with each other. Figure 1004 allows individual scans to be identified and located within the build plane by comparing the control signal and the build-plane intensity data. In general, the energy source control signal provides at least a start and an end position from which the area covered by the scan can be determined. Figure 1006 allows the raw sensor data, which is associated with an intensity or power of each scan, to be bound into corresponding X and Y grid regions.In some embodiments, raw intensity or power data can be converted into energy units by correlating the dwell time of each scan within a specific raster area. Each raster area can be sized to correspond to one or more pixels of an optical sensor monitoring the assembly plane. It should be noted that various coordinate systems, such as polar coordinates, can be used to store raster coordinates, and that coordinate storage is not limited to Cartesian coordinates. In some embodiments, different scan types can be captured separately, allowing analysis to be performed on only specific scan types. For example, an operator might want to focus on contour scans if these scan types are highly likely to exhibit undesirable variations.At 1008, the energy input at each raster area can be added up so that the total amount of energy received in each raster area can be determined using equation (4). Epdm = ∑n = 1 pixel measurement values ​​in raster cell Epdn

[0057] This addition can be performed shortly before adding a new powder layer to the build plane, or alternatively, the addition can be delayed until a predetermined number of powder layers have been deposited. For example, the addition could be performed only after five or ten different powder layer areas have been deposited and melted during an additive manufacturing process. In some embodiments, a sintered powder layer can contribute approximately 40 micrometers to the thickness of a part; however, the thickness varies depending on the type of powder used and the thickness of the powder layer.

[0058] At 1010, the standard deviation is determined for the acquired measurements assigned to the individual grid areas. This allows for better identification of grid areas where the energy values ​​vary to a greater or lesser extent. Fluctuations in the standard deviation may indicate problems with sensor performance and / or instances where one or more scans are missing or the performance level is far outside the normal operating parameters. The standard deviation can be determined using equation (5). Epdsm=1#Measurement-in-Position-1∑n=1Measurement-in-Pixel(En-E¯)2

[0059] At 1012, the total energy density received in each raster area can be determined by dividing the power values ​​by the total area of ​​the raster area. In some embodiments, a raster area can have a square geometry with a length of approximately 250 micrometers. The energy density for each raster area can be determined using equation (6). ERasterposition=∑n=1Measurements−in−PositionEpdnARasterposition

[0060] In 1014, when more than one part is built, different grid areas can be assigned to different parts. In some embodiments, a system can include stored part boundaries that allow each grid area and its associated energy density to be quickly linked to its respective part, using the grid area coordinates and the boundaries assigned to each part.

[0061] Since the boundary of internal cavities contributes to energy loss, substantial parts of the layer must not receive energy. Therefore, the affected area can be calculated by summing only the grid regions that have been determined to receive energy from the energy source. At 1018, the total amount of power received by the grid regions within the section of the layer associated with the part can be summed and divided by the affected area to determine the energy density for that layer of the part. Area and energy density can be calculated using equations (7) and (8). APart=∑n=1Part−Pixel1(Epdn>0) IPQMPartLayer=∑n=1Part−GridPositionsEpdnAPart

[0062] At 1020, the energy density of each layer can be added to obtain a metric indicating the total amount of energy received by the part. The total energy density of the part can then be compared to the energy density of other similar parts at the build level. At 1022, the total energy of each part is summed. This allows for high-quality comparisons between different built elements. Build comparisons can be helpful in identifying systematic differences such as powder variations and changes in overall output power. Finally, at 1024, the summed energy values ​​can be compared with other layers, parts, or build levels to determine the quality of those other layers, parts, or build levels.

[0063] Other metrics can be used in the Fig. The metrics described in section 10 can be added or stored. For example, the width at half maximum (WHM) can be one of these metrics, and the WHM analysis procedure described here can be used to gain insights into the additive manufacturing process. The collected WHM data can be used in a variety of ways. For example, the WHM values ​​can be averaged for each screen area, or the highest maximum temperature can be associated with a particular screen area. Alternatively, the mode of the top 5 or 10 percent of the WHM values ​​can be associated with the screen area to prevent a single occurrence of a higher WHM measurement from overshadowing an otherwise nominal WHM value.

[0064] It is pointed out that the in Fig. The specific steps described in Figure 10 represent a particular method for characterizing a build process of an additive manufacturing process according to another embodiment of the present invention. Alternative embodiments may also involve other sequences of steps. For example, in alternative embodiments of the present invention, the steps described above may be performed in a different order. Furthermore, the steps described in Figure 10 may be performed in a different order. Fig. The 10 individual steps depicted comprise several sub-steps that can be performed in different sequences depending on the respective step. Furthermore, additional steps can be added or removed depending on the application. Many variations, modifications, and alternatives are recognizable to the average professional.

[0065] The Fig. 11A - Fig. Figure 11D shows illustrative representations that demonstrate how multiple scans can contribute to the energy input into the individual raster areas. Fig. 11A represents a raster pattern consisting of several raster areas 1102 distributed over an area of ​​a part built up using an additive manufacturing system. Fig. Figure 11A further shows an initial pattern of energy scans 1104 extending diagonally across the grid areas 1102. The initial pattern of energy scans 1102 can be generated by a laser or other powerful source of thermal energy scanning across the grid 1104. Fig. Figure 11B shows how the energy supplied via the part is represented in each of the raster areas 1102 by a single shade of gray, which is representative of a certain amount of energy received therein, with darker shades of gray corresponding to greater amounts of energy. It is noted that in some embodiments the size of the raster areas 1102 can be reduced to obtain higher-resolution data. Alternatively, the size of the raster areas 1102 can be increased to reduce the storage and processing power requirements.

[0066] Fig. 11C shows a second pattern of energy scans 1106 that overlaps with at least some of the energy scans of the first pattern of energy scans. As described in the text regarding Fig. Figure 8 explains that the grid areas where the first and second energy scan patterns overlap are shown in a darker shade to illustrate how the energy from both scans increased the amount of energy received at the overlap of the scan patterns. Of course, the method is not limited to two overlapping scans and can include many more additional scans that are added together to fully represent the energy received in each grid area.

[0067] The Fig. 12A - Fig. Figure 12F illustrates how a raster can be dynamically created to characterize and control an additive manufacturing process. Fig. Figure 12A shows a top view of a cylindrical workpiece 1202 located on a build plane 1204. The workpiece 1202 is shown during the additive manufacturing process. Fig. Figure 12B shows how the cylindrical workpiece 1202 can be divided into several tracks 1206 along which an energy source can melt powder distributed on one side of the cylindrical workpiece 1202. In some embodiments, the energy source can change directions 1206 as shown, while in other embodiments, the energy source can only move in one direction. Furthermore, the direction of the tracks 1206 can vary from layer to layer to further randomize the orientation of the scans for building the workpiece 1202.

[0068] Fig. Figure 12C shows an example of a scan pattern for the energy source during the formation of part of the workpiece 1202. As indicated by the arrow 1208, the direction of movement of an exemplary energy source across the workpiece 1202 is diagonal. Individual scans 1210 of the energy source can be oriented in a direction perpendicular to the direction of movement of the energy source along the path 1206 and extend completely across the path 1206. The energy source can be briefly switched off between successive individual scans 1210. In some embodiments, the duty cycle of the energy source when crossing each path 1206 can be approximately 90%. By using this type of scan strategy, the energy source can cover a width of the path 1206 while moving across the workpiece 1202. In some embodiments, the strip 1210 can have a width of approximately 5 mm.This can significantly reduce the number of paths required to build the workpiece 1202, since in some embodiments the width of a melt pool generated by the energy source can be on the order of about 80 micrometers.

[0069] The Fig. 12D - Fig. Figure 12E shows how raster areas 1212 can be dynamically generated along each path 1206 and dimensioned to accommodate the width of each individual scan 1210. The system can predict the exact position of subsequent scans by referring to energy source control signals located along the path to the energy source. In some embodiments, the width of the raster areas 1212 can correspond to the length of the individual scans 1210 or be within 10% or 20% of the length of the individual scans 1210. Again, the scan length of the individual scans 1210 can be predicted by referring to the energy source control signals. In some embodiments, the raster areas 1212 can be square or rectangular. A thermal energy density can be determined for each of the raster areas 1212 as the energy source progresses along the path 1206.In some embodiments, thermal energy density measurements within grid area 1212-1 can be used to adjust the output power of the energy source in the next grid area, in this case, grid area 1212-2. For example, if the thermal energy density values ​​generated by individual scans 1210 within grid area 1212-1 are significantly higher than expected, the energy source output power can be reduced, the speed at which the energy source is traversed across individual scans 1210 can be increased, and / or the spacing between individual scans 1210 within grid area 1212-2 can be increased. These adjustments can be made within a closed-loop control system. Even if in . Fig. While 12E clearly represents only five individual scans 1210 within each grid area 1212, this serves only as an example, and the actual number of individual scans within a grid area 1212 may be significantly higher or, in some cases, lower. For example, if the melt zone generated by the energy source is approximately 80 micrometers wide, about 60 individual scans 1210 are required for all the powder within a 5 mm square grid area 1212 to fall into the melt zone.

[0070] Fig. Figure 12F shows an edge region of the workpiece 1202 after the energy source has completed traversing the pattern of paths 1206. In some embodiments, the energy source can continue to supply energy to the workpiece 1202 after a large portion of the powder has melted and resolidified. For example, the contour scans 1214 can run along the periphery 1216 of the workpiece 1202 to perform a surface treatment on the workpiece 1202. It should be noted that the contour scans 1214, as shown, are significantly shorter than the individual scans 1210. For this reason, the raster areas 1218 can be considerably narrower than the raster areas 1212. It should also be noted that the raster areas 1218 are not truly rectangular, since in this case they follow the contour of the periphery of the workpiece 1202.Other situations that can lead to varying scan lengths arise when a workpiece has walls of varying thickness. A wall with variable thickness can cause the scan length to vary within a single grid area. In such a case, the area of ​​each grid area can be kept uniform by increasing the length of the grid area while simultaneously decreasing its width to compensate for changes in the length of individual scans.

[0071] Fig. Figure 13 shows an example of a control system that incorporates a feedback control loop 1300 to establish and maintain control of an additive manufacturing process. Block 1302 inputs a baseline thermal energy density value for the next raster region that the energy source is to traverse into the control loop. The baseline thermal energy density value can be determined using modeling and simulation programs and / or from previously conducted experiments / test runs. In some embodiments, the baseline thermal energy density data can be adjusted by the energy density adjustment block 1304, which includes energy density values ​​for different raster regions recorded in the preceding layers. The energy density adjustment block 1304 can include an adjustment to the energy density baseline value block if previous layers received too much or too little energy.For example, if optical sensor readings in a workpiece indicate a thermal energy density below the target values, energy density adjustment values ​​can increase the baseline energy density value in grid areas that overlap grid areas with readings below the target thermal energy density. This allows the energy source to additionally melt powder that was not completely melted in the preceding layer(s).

[0072] Apart from energy density, other metrics can be used in the Fig. 12A - Fig. The metrics described in Section 12F can be summed or stored. One such metric might be the half-width (WWH), and the WWH method described herein for performing a WWH analysis can be used to gain insights into the additive manufacturing process. The collected WWH data can be used in a variety of ways. For example, the WWH values ​​can be averaged for each screen area, or the highest maximum temperature can be associated with each screen area. Alternatively, the mode of the top 5 or 10 percent of the WWH values ​​can be associated with the screen area to prevent a single occurrence of a higher WWH measurement from overshadowing an otherwise nominal WWH value.

[0073] Fig.Figure 13 further shows how the inputs of blocks 1302 and 1304 interact to generate an energy density control signal, which is received by the controller 1306. The controller 1306 is configured to receive the energy density control signal and generate heat source input parameters configured to produce the desired thermal energy density within the current raster area. The input parameters can include power, scan speed, web spacing, scan direction, and scan duration. The input parameters are subsequently received by the energy source 1308, and any changes to the input parameters are applied by the energy source 1308 for the current raster area. Once optical sensors measure the scans from the energy source 1308 that constitute the current raster area, the thermal energy density for the current raster area is calculated at block 1310 and compared with the energy density control signal.If the two values ​​are the same, no change is made to the energy density control signal based on the optical sensor data. However, if the two values ​​are different, the difference is added to or subtracted from the energy density control signal for scans in the next raster area.

[0074] The various aspects, embodiments, implementations, or features of the described embodiments can be used individually or in any combination. Different aspects of the described embodiments can be implemented by software, hardware, or a combination of hardware and software. The described embodiments can also be realized as computer-readable code on a computer-readable medium for controlling manufacturing processes or as computer-readable code on a computer-readable medium for controlling a production line. The computer-readable medium can be any data storage device capable of storing data that can subsequently be read by a computer system. Examples of computer-readable media include read-only memory, random-access memory, CD-ROMs, hard disks, DVDs, magnetic tapes, and optical data storage devices.The computer-readable medium can also be distributed across network-connected computer systems, so that the computer-readable code is stored and executed in a distributed form.

Claims

[1] Additive manufacturing process comprising: Monitoring the temperature of a portion of a build plane during an additive manufacturing process using a temperature sensor while a heat source passes through the portion of the build plane; detecting a peak temperature associated with one or more passes of the heat source through the portion of the build plane; Determining a limit temperature by reducing the peak temperature by a specified amount; Determining a time interval during which the monitored temperature exceeds the limit temperature; Identifying a change in manufacturing conditions that is likely to result in a manufacturing defect based on the time interval; and Changing a process parameter of the heat source in response to a change in manufacturing conditions. [2] The additive manufacturing method of claim 1, wherein determining the limit temperature comprises setting the limit temperature based on a requirement of a particular additive manufacturing process. [3] The additive manufacturing method of claim 1 or 2, wherein changing the process parameter in response to the change in manufacturing conditions further comprises changing a second process parameter. [4] Additive manufacturing method according to claim 1, 2 or 3, wherein the process parameter comprises the beam power and / or the beam travel speed and / or the track distance. [5] Additive manufacturing process according to any one of claims 1 to 4, wherein the temperature sensor comprises a pyrometer. [6] Additive manufacturing process according to any one of claims 1 to 5, further comprising: Determining a plurality of time intervals, each time interval of the plurality of time intervals being associated with a region of the portion of the build plane associated with the temperature exceeding the threshold temperature; and determining an average time interval associated with the plurality of time intervals. [7] Additive manufacturing process according to any one of claims 1 to 6, further comprising: Determining a plurality of time intervals, each time interval of the plurality of time intervals being associated with a region of the portion of the build plane associated with the temperature exceeding the threshold temperature; and determining a path distance parameter associated with the plurality of time intervals. [8] An additive manufacturing process according to any one of claims 1 to 7, wherein the temperature is an observed temperature. [9] An additive manufacturing process according to any one of claims 1 to 7, wherein the temperature is an actual temperature. [10] Additive manufacturing process comprising: Performing an additive manufacturing process using an energy source; receiving voltage data associated with a stationary photodiode during a scan of the energy source across a portion of a powder bed to produce a part; Detecting a peak voltage associated with one or more passes of the energy source through the portion of the powder bed; Determining a cutoff voltage by reducing the peak voltage by a predetermined amount; Determine a time interval in which the voltage data exceeds the limit voltage; Identifying a change in manufacturing conditions that may lead to a manufacturing defect based on the time interval; and Changing a process parameter of the energy source in response to a change in manufacturing conditions. [11] The additive manufacturing method of claim 10, wherein determining the limit voltage comprises setting the limit temperature based on a requirement of a particular additive manufacturing process. [12] The additive manufacturing method of claim 10 or 11, wherein changing the process parameter in response to the change in manufacturing conditions further comprises changing a second process parameter. [13] Additive manufacturing method according to claim 10, 11 or 12, wherein the process parameter comprises the beam power and / or the beam travel speed and / or the track distance. [14] An additive manufacturing process according to any one of claims 10 to 13, wherein the stationary photodiode comprises a pyrometer. [15] Additive manufacturing process according to any one of claims 10 to 14, further comprising: Determining a plurality of time intervals, each time interval of the plurality of time intervals being associated with a region of the portion of the build plane associated with the voltage data exceeding the threshold voltage; and Determining an average time interval associated with the multiple time intervals. [16] The additive manufacturing method of any one of claims 10 to 15, further comprising: determining a plurality of time intervals, each time interval of the plurality of time intervals being associated with the voltage data exceeding the threshold voltage; and determining a path distance parameter associated with the plurality of time intervals. [17] Additive manufacturing system comprising: a heat source; a non-imaging optical sensor configured to measure temperatures across a surface of a material during an additive manufacturing process in which the material is heated by the heat source; and at least one processor configured to receive temperature data from the optical sensor, wherein the at least one processor is further configured to: Monitoring a temperature of a portion of a build plane during an additive manufacturing process using the non-imaging sensor when a heat source passes through the portion of the build plane; detecting a peak temperature associated with one or more passes of the heat source through the portion of the build plane; Determining a limit temperature by reducing the peak temperature by a predetermined amount; Determining a time interval during which the monitored temperature exceeds the limit temperature; Identifying a change in manufacturing conditions that is likely to result in a manufacturing defect based on the time interval; and Changing a process parameter of the heat source in response to a change in manufacturing conditions. [18] The additive manufacturing method of claim 17, wherein determining the limit temperature comprises setting the limit temperature based on a requirement of a particular additive manufacturing process. [19] The additive manufacturing method of claim 17 or 18, wherein changing the process parameter in response to the change in manufacturing conditions further comprises changing a second process parameter. [20] Additive manufacturing method according to claim 17, 18 or 19, wherein the process parameter comprises the beam power and / or the beam travel speed and / or the path distance.

Images