Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
DETAILED ACTION
Status of the Claims
Claims 1-20 are pending. Claims 1-8 and 11-18 are the subject of this NON-FINAL Office Action.
Claim Interpretations
The only “melt pool monitor” (MPM) disclosed is a generic “melt pool monitoring system” (MPMS) as shown in Figure 1:
PNG
media_image1.png
828
748
media_image1.png
Greyscale
However, claim 1 requires a “sensor” which is also shown in Figure 1. How are these two distinct entities the same thing? To resolve this conundrum, the Examiner interprets the elected MPM as the MPMS. The specification discloses that the MPMS is a heat sensor (para. 0056). Apparently, this further encompasses a thermal camera (para. 0027). In an attempt to give meaning to the elected MPM sensor, the Examiner interprets it as any thermal/heat sensor.
Another dilemma is that Applicants also elect a melt pool size process characteristic measured by the MPM. However, a thermal/heat sensor is not set up for this. To the extent that a camera is used, this could possibly measure the dimensions of a melt pool. However, in reality, only a very specific high-speed camera, with specific magnification, specific frame rates, and specific optics can do this in DED, PBF, SLM and EBM (see Haley et al, Observations of particle-melt pool impact events in directed energy deposition, Additive Manufacturing, Volume 22, August 2018, Pages 368-374). Applicants have not disclosed any of these necessary highly-optimized parameters. Without this critical information, the elected MPM encompasses any camera.
As to “predictive model data,” the only such model disclosed is “determining a change in the powder feed rate associated with a change in the at least one powder control parameter” (para. 0061). As elected, this means determining a change in the powder feed rate associated with a change in the powder feeder geometry. As explained below, this makes no sense, nor is it disclosed in the specification in enough detail to demonstrate possession.
As a reminder, Applicants elected determining, based on the melt pool size from a melt pool monitor and a predictive model data, powder feeder geometry to achieve a predetermined powder feed rate of the powder stream, and using melt pool capture capability to determine mass flux of powder.
Claim Rejection - 35 USC § 112- Written Description
The following is a quotation of the first paragraph of 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 1-8 and 11-18 are rejected under 35 U.S.C. 112(a) or pre-AIA 35 U.S.C. 112, first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for pre-AIA the inventor(s), at the time the application was filed, had possession of the claimed invention.
The Specification fails to reasonably convey to one having ordinary skill in the art at the time of filing that Applicants were in possession of determining and controlling powder feeder geometry to effect powder feed rate. The claims, as elected, require “determine, based on melt pool monitor measured melt pool size and a predictive model data, powder feeder geometry configured to achieve a predetermined powder feed rate of the powder stream; and control, based on the powder feeder geometry, the energy delivery device and the powder delivery device to deposit a plurality of layers.” Absent a specific definition in the specification, the powder feeder geometry is interpreted as the shape and size of the powder feeder, or hopper (see e.g. He at al, In-situ monitoring and deformation characterization by optical techniques; part I: Laser-aided direct metal deposition for additive manufacturing, Optics and Lasers in Engineering, Volume 122, November 2019, Pages 74-88, Fig. 10). To this end, the claims require the AM system to automatically control or change the feeder or hoper geometry to effect the powder flow rate (see e.g. claim 5, change powder control parameter, or feeder geometry). However, the specification fails to disclose a feeder with adjustable geometry, much less during a print. Stated another way, the feeder is static in geometry, it does not change. There is no disclosure of a dynamic or changing feeder geometry such that it is a “control parameter.” Such a static geometry feeder would not yield any reasonable “control parameter” of feeder geometry.
Furthermore, this feeder geometry is never disclosed in any detail as effecting melt pool size, or vice versa. Feeder geometry is simply mentioned, with no detail.
Previous Response to Arguments
The rejection is maintained because Applicants fail to address the issue. Specifically, the claims are so broad that they read on any geometric change of any part of the powder feeder, none of which is disclosed in ay detail whatsoever in the specification. Applicants hand-wave the actual issue away by arguing that “a component in a system may exhibit a change in geometry in view of commonplace phenomenon or occurrences such as (i) wear of the component, (ii) a deviation in a geometry of the component from nominal specifications arising during manufacturing, or (iii) replacement of the component by another unit or type of the component (replacing a worn component with a new component, replacing one type of component with another, changing a supplier of a component, etc.).” However, even this “commonplace” occurrence is never discussed in the specification. The issue is what the specification discloses, compared to what is claimed. The specification completely fails to disclose any of the above, which is why Applicants never point to any part of the specification- they can’t.
And the amendments only add problems. As to “predictive model data,” the only such model disclosed is “determining a change in the powder feed rate associated with a change in the at least one powder control parameter” (para. 0061). As elected, this means determining a change in the powder feed rate associated with a change in the powder feeder geometry. However, the claim amendment now requires “powder stream is maintained at the predetermined powder feed rate.” If the powder feed rate changes then how can it be maintained? These are contradictory statements, and not disclosed in the specification.
Response to Arguments
Applicants continue to misunderstand the issue. For example, Applicants argue that “the term ‘powder control parameter,’ when properly interpreted in light of the specification, is a parameter that affects control of the powder, not necessarily a parameter that itself is actively controlled” (Reply, pg. 8). However, nothing in the claims or specification require this interpretation. It is only one interpretation. Applicants have not defined the phrase this way anywhere. Thus, it broadly encompasses interpretations provided by the Examiner.
Applicants also assert throughout their arguments that certain elements are not “required.” However, the issue is breadth. In this regard, the broad claims encompass much. For example, he claims broadly encompass “that the system actively or dynamically change the physical geometry of the powder feeder mid-print” (Reply, pg. 8). In fact, Applicants’ election explicitly describes this, or at least one interpretation of it does.
Finally, as previously explained by the Examiner, the mere mention of powder feeder geometry in the specification does not satisfy written description of the claimed invention. There is simply zero detail about powder feeder geometry control in the specification. Applicants’ specification amounts to a wish or plan to later describe it.
New Grounds of Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1-8 and 11-18 is/are rejected under 35 U.S.C. 103 as being unpatentable over He at al, In-situ monitoring and deformation characterization by optical techniques; part I: Laser-aided direct metal deposition for additive manufacturing, Optics and Lasers in Engineering, Volume 122, November 2019, Pages 74-88, in view of Haley et al, Observations of particle-melt pool impact events in directed energy deposition, Additive Manufacturing, Volume 22, August 2018, Pages 368-374, in further view of Mazzucato et al, Monitoring Approach to Evaluate the Performances of a New Deposition Nozzle Solution for DED Systems, Technologies, vol. 5, no. 2, p. 29, May 2017, doi: 10.3390/technologies5020029 and Sheila Moroney, IN-SITU PROCESS MONITORING OF DIRECTED ENERGY DEPOSITION POWDER FLOW TO DETECT ANOMALIES, Thesis, published May 2023, pgs. 1-76 as evidenced by Chen et al., Rapid surface defect identification for additive manufacturing with in-situ point cloud processing and machine learning, Virtual and Physical Prototyping, 16(1), 50–67. https://doi.org/10.1080/17452759.2020.1832695, 10/15/2020, US 20210354397, US20220063199 and US 20080178994.
Applicants are reminded that their claims are very broad. The sensor merely “generate sensor data representative of at least one process characteristic.” This is any representation of any generic characteristic (e.g. powder composition, print speed, print time, defect, humidity, or any other possible “characteristic” of a print process. Similarly, the generic “powder control parameter” that is determined using a generic MLM is any “parameter” affecting the powder (e.g. composition, speed of deposition, temperature, humidity, print speed, or any other possible parameter encompassing innumerable possible variables) using any MLM. At the end, the “the powder stream is maintained at the predetermined powder feed rate” by generically controlling the generic “energy delivery device,” the generic “powder delivery device” and the generic “powder control parameter.” As One can see, the claims are directed to a very generic, broad concept of MLM controlling any aspect of powder-based prints to achieve a powder feed rate maintenance. Applying MLM to achieve this generic outcome using generic data is the epitome of obviousness as explained below.
This rejection is presented in an effort to give reasonable meaning to the elected invention in light of the specification. This is difficult. However, for prior art purposes, the Examiner interprets the claims to require using a camera (e.g. high-speed camera) to determine melt pool size (e.g. geometry, outline, length, width, etc.), which is used to determine melt pool capture/absorption capability/efficiency, which is fed back to control energy device (e.g. laser) and/or powder delivery device (e.g. nozzle).
It would have been prima facie obvious to one having ordinary skill in the art before the effective filing date to use familiar automatic feedback parameters such as melt pool size to calculate melt pool capture/absorption capability/efficiency in order to control energy device (e.g. laser) and/or powder delivery device (e.g. nozzle) in DED, PBF, SLM and EBM with a reasonable expectation of success.
As to claims 1-8 and 11-18, He teaches that high-speed cameras are routinely used in DED, PBF, SLM and EBM to monitor melt pools for closed-loop feedback control of powder delivery components such as nozzles and lasers (Fig. 10, pgs. 80-81). This measurement includes morphology, or size of the melt pool (pg. 81, col. 2). Powder feed rate is also a known process parameter that effects melt pool and its size (pg. 81, col. 2).
As to machine learning, which requires training on data, the prior art is full of motivation to apply this to monitor and adjust 3D printing parameters such as powder feed rates. First, Applicants are reminded that any machine learning implementation (AKA artificial intelligence) requires training on data sets. See GeeksForGeeks, Machine Learning Tutorial, Last Updated: 5 Jun, 2026, avail at https://www.geeksforgeeks.org/machine-learning/machine-learning/, Machine Learning Basic figure. ML is the best-known branch of AI; therefore, any suggestion to apply AI would immediately be recognized by a skilled artisan as a suggestion to apply ML. Thus, any prior art that teaches or suggests to apply AI or MLM to 3D printing parameters inherently requires training on those parameters and inputs that affect them. To this end, He teaches “Based on the big data from the various in-situ monitoring with numerous parameters varied, artificial intelligent is attractive and called for the process control and optimization in the near future” (Summary). Moroney explains that
Process monitoring to assess and improve DED process reliability is broadly recognized as a being a critical step to widespread adoption in industry. To address this need, many researchers have developed methods for in-situ monitoring of various parameters of the DED process to improve reliability. Some of these methods include using a coaxial camera to monitor the melt pool [9]–[13] and optical emission spectroscopy to monitor melt pool vaporization [14][16]. Additionally, using a laser displacement sensor to measure the deposition height has been well established in literature [17]–[19]. Sammons et al. [17] developed a build sensing method that used a Keyence V7200 laser line scanner positioned perpendicular to the deposition path to scan the height of the build after each layer was completed. Chen et al. [18] used a similar displacement sensor with a clustering algorithm to detect and identify build surface defects based on the sensor’s point cloud surface reconstruction. These similar methods using displacement sensors to monitor build height in-situ were very successful and have become common across DED process monitoring efforts
(Chapter 2, Literature Review, pg. 4). Maroney proceeds to provide a solution to “in-situ monitoring of the powder flow rate near the powder focal plane” as explained below. Chen (cite 18 in Maroney, above) uses machine learning/AI to supervise and classify surface defects. See Chen et al., Rapid surface defect identification for additive manufacturing with in-situ point cloud processing and machine learning, Virtual and Physical Prototyping, 16(1), 50–67. https://doi.org/10.1080/17452759.2020.1832695, 10/15/2020. Therefore, Maroney suggests to apply machine learning to the powder feed rate monitoring and adjusting disclosed in Maroney. Finally, US 20210354397 specifically implements automatic monitoring and adjusting DED print parameters including powder deposition amounts and rate, along with solidification/energy adjustments based on the powder amounts in order to maintain desired powder stream (paras. 0067, 0091-92, 0098-99). To accomplish this real-time monitoring and adjustment, “the controller 250 utilizes machine learning or artificial intelligence to actively control the build-powder composition of the build powder 102 as the build powder 102 is being deposited to form the build-powder section 108 of the powder layer 106. Such active control also serves as quality control. Through use of the controller 250, an automated and iterative process is performed, for example, using the lookup tables, to determine both optimized build parameters and material compositions at any moment in time and/or at any point in the build of each one of the object layers” (para. 0179; emphases added). US20220063199 explains that ML is used to monitor and adjust powder-based 3D print parameters such as powder composition, humidity, speed, amounts, etc. in order to maintain certain print outcomes (Abstract, paras. 0036-43). In other words, the prior art as a whole, consistent with society as a whole, demonstrate the immense popularity, and routine application of ML/AI to all facets of industry, including to monitor 3D printing parameters such as powder feed rate, and other powder “parameters.”
He does not explicitly teach to determine pool capture/absorption capability/efficiency for feedback control. However, He suggests as much. Specifically, He points to “Haley et al. [which] have recently and impressively presented a work on the direct observation of interactions between the powder particles and the melt pool using four high-speed cameras (up to 200,000 frames/s), four lens systems, and three illumination systems” (pg. 81, col. 1). He then explicitly states that this work from Haley is directly applicable to “in-situ monitoring of the melt pool temperature characteristics (e.g., maximum temperature and thermal gradient) [which] is underway, and high-speed visual observation techniques are available with different algorithms of digital images developed to characterize the morphology (e.g., length, width and surface area) of the melt pool” (pg. 81, cols. 1-2). In fact, “[a] promising trend is to integrate various inspection techniques into the DLD fabrication equipment for closed-loop control of multiple process parameters ensuring a high part quality, as shown in Fig. 10” (pg. 81, col. 2). “It is also interesting to develop analytical models based on the experimental efforts to better understand the mechanism of the melt pool and the entire process” (id.)
PNG
media_image2.png
390
792
media_image2.png
Greyscale
Thus, He explicitly suggests the claimed invention, aside from determine pool capture/absorption capability/efficiency for feedback control.
Yet, as explained above, He suggests that the work of Haley should be integrated into these feedback techniques. Just like in He, Haley teaches that measuring melt pool size and capture probability are routine (pg. 368, col. 2 (“In order to parse the many physical aspects of the L-DED and other laser based additive manufacturing systems, numerous studies have employed optical in-situ characterization techniques. These measurements tend to focus on two key areas: the trajectories and mass capture probabilities of feedstock particles after they exit the nozzle [2,5,6], and the size, shape [7] and temperature distribution of the melt pool”)). Haley goes on to teach that capture efficiency (AKA capture capability) is a function of melt pool dimensions (Apool) (Section 4.4).
PNG
media_image3.png
446
410
media_image3.png
Greyscale
PNG
media_image4.png
344
404
media_image4.png
Greyscale
PNG
media_image5.png
44
388
media_image5.png
Greyscale
This is well-known and routinely used (see e.g. US 20080178994, para. 0039 (“a larger melt pool is needed to have higher powder capture efficiency, so that the deposit layer has enough material to support the next layer without slumping”)). Haley makes clear that capture efficiency is well-known to be affected by powder feed rate (F) (id.) Thus, He in light of Haley specifically suggests that melt pool capture efficiency, which requires melt pool size and powder feed rate data, which is dependent on feeder type and geometry, should be incorporated into feedback control.
In fact, powder feed rate is well-known to be affected by powder feeder geometry and melt pool size; and capture efficiency by laser power and powder flow rate (Mazzacuto, pg. 2, first para.; Moroney, pgs. 5, 6, 62). Further, capture efficiency is “often used as a metric to determine the quality of powder flow” (Moroney, pg. 5). In other words, the prior art makes clear that a skilled artisan would have been motivated to incorporate melt pool capture/absorption capability/efficiency, including parameters of powder flow rate, powder feeder geometry, and melt pool size, into the familiar feedback control techniques. To this end, a skilled artisan would have been motivated to use familiar automatic feedback parameters such as melt pool size to calculate melt pool capture/absorption capability/efficiency in order to control energy device (e.g. laser) and/or powder delivery device (e.g. nozzle, feeder) in DED, PBF, SLM and EBM with a reasonable expectation of success.
Previous Response to Arguments
The rejection is maintained because Applicants’ core assertions that the prior art does not teach a predictive model or achieving and maintaining a desired powder feed rate are found wanting. First, “predictive model” is so generic that it encompasses any model with any prediction, not just AI (despite the fact that AI as of the effective filing date was incredibly familiar in every field of endeavor). To this end, the Examiner made clear that He explicitly suggests such a model as it is a very familiar technique even in powder-based 3D printing. Applicants proceed to accuse the Office of making “broad assertions,” when in fact the Examiner provided a detailed rationale that any skilled artisan with common sense and knowledge of the art would apply based on the prior art. Applicants will not overcome this art by attacking it piecemeal.
Second, as to maintaining a powder feed rate, this, too was discussed in detail by the Examiner. As explained in detail, all of He, Haley, Mazzacuto, and Moroney discuss how powder feed rate affects melt pool capture/absorption capability/efficiency. In other words, maintaining powder feed rate while other variables change is well within the knowledge and skill of an artisan.
Applicants should take a second look at their “broad assertions” in their claims. The claims very broadly assert to use “sensor data” and “predictive model” to maintain a “predetermined powder feed rate” using a “powder control parameter.” The prior art broadly demonstrates that these broad assertions are at least very obvious as they encompass something as simple as maintaining a powder feed rate based on a sensor that detects amount of powder in a hopper (or any other sensed thing) and a “model” threshold that dictates increasing the powder into the hopper to maintain powder feed. There are innumerable combinations in the art that fall within this very broad claim. Even so, the elected species, which is itself quite broad, is still found within the prior art above, which a skilled artisan would have been motivated from to achieve the claimed, elected invention.
Applicants are very strongly urged to amend their claims to recite the specific sensor, specific “predictive model” (if one is even disclosed in the specification), and specific “control parameter.”
New Grounds of Rejection - Double Patenting- Obvious Type
The nonstatutory double patenting rejection is based on a judicially created doctrine grounded in public policy (a policy reflected in the statute) so as to prevent the unjustified or improper timewise extension of the “right to exclude” granted by a patent and to prevent possible harassment by multiple assignees. A nonstatutory obviousness-type double patenting rejection is appropriate where the conflicting claims are not identical, but at least one examined application claim is not patentably distinct from the reference claim(s) because the examined application claim is either anticipated by, or would have been obvious over, the reference claim(s). See, e.g., In re Berg, 140 F.3d 1428, 46 USPQ2d 1226 (Fed. Cir. 1998); In re Goodman, 11 F.3d 1046, 29 USPQ2d 2010 (Fed. Cir. 1993); In re Longi, 759 F.2d 887, 225 USPQ 645 (Fed. Cir. 1985); In re Van Ornum, 686 F.2d 937, 214 USPQ 761 (CCPA 1982); In re Vogel, 422 F.2d 438, 164 USPQ 619 (CCPA 1970); and In re Thorington, 418 F.2d 528, 163 USPQ 644 (CCPA 1969).
A timely filed terminal disclaimer in compliance with 37 CFR 1.321(c) or 1.321(d) may be used to overcome an actual or provisional rejection based on a nonstatutory double patenting ground provided the conflicting application or patent either is shown to be commonly owned with this application, or claims an invention made as a result of activities undertaken within the scope of a joint research agreement.
Effective January 1, 1994, a registered attorney or agent of record may sign a terminal disclaimer. A terminal disclaimer signed by the assignee must fully comply with 37 CFR 3.73(b).
Instant claims 1-8 and 11-18 are rejected on the ground of nonstatutory obviousness-type double patenting as being unpatentable over conflicting claims 1-20 of U.S. Patent Application No. 17/932959, in view of He at al, In-situ monitoring and deformation characterization by optical techniques; part I: Laser-aided direct metal deposition for additive manufacturing, Optics and Lasers in Engineering, Volume 122, November 2019, Pages 74-88, Sheila Moroney, IN-SITU PROCESS MONITORING OF DIRECTED ENERGY DEPOSITION POWDER FLOW TO DETECT ANOMALIES, Thesis, published May 2023, pgs. 1-76 as evidenced by Chen et al., Rapid surface defect identification for additive manufacturing with in-situ point cloud processing and machine learning, Virtual and Physical Prototyping, 16(1), 50–67. https://doi.org/10.1080/17452759.2020.1832695, 10/15/2020, US 20210354397 and US20220063199.
The instant claims are obvious over the conflicting claims because the conflicting claims teach the same powder monitor and adjust system, and applying ML is the epitome of obviousness. The conflicting claims teach an energy delivery device configured to deliver energy to a build surface of a component to form a melt pool in the build surface of the component (claim 1, the additive manufacturing system is configured to simultaneously deliver energy to form a melt pool in the build surface); a powder delivery device configured to direct a powder stream toward the melt pool (claim 1, and deliver the powder stream to the melt pool); at least one sensor configured to generate sensor data representative of at least one process characteristic (image data of powder stream, claim 1); and a computing device configured to: receive the sensor data from the at least one sensor (claim 1); determine, based on the sensor data and a predictive model data, at least one powder control parameter configured to achieve a predetermined powder feed rate of the powder stream (claims 1 & 2, determine at least one metric associated with the powder stream based on the received image data, determining a mass flow rate or powder distribution within the powder stream); and control, based on the at least one powder control parameter, the energy delivery device and the powder delivery device to deposit a plurality of layers (claims 1 & 7-9, cause the additive manufacturing system to perform at least one action in response to the at least one metric indicating the abnormal state, controlling at least one operating parameter of the additive manufacturing system, wherein the at least one parameter comprises at
least one of a powder feed rate or a carrier gas flow rate, wherein the at least one parameter comprises at least one of a powder feed rate to a selected nozzle of the powder delivery device or a carrier gas flow rate to a selected nozzle of the powder delivery device).
As to ML, as explained above, this was very obvious at the time of filing. As to machine learning, which requires training on data, the prior art is full of motivation to apply this to monitor and adjust 3D printing parameters such as powder feed rates. First, Applicants are reminded that any machine learning implementation (AKA artificial intelligence) requires training on data sets. See GeeksForGeeks, Machine Learning Tutorial, Last Updated: 5 Jun, 2026, avail at https://www.geeksforgeeks.org/machine-learning/machine-learning/, Machine Learning Basic figure. ML is the best-known branch of AI; therefore, any suggestion to apply AI would immediately be recognized by a skilled artisan as a suggestion to apply ML. Thus, any prior art that teaches or suggests to apply AI or MLM to 3D printing parameters inherently requires training on those parameters and inputs that affect them. To this end, He teaches “Based on the big data from the various in-situ monitoring with numerous parameters varied, artificial intelligent is attractive and called for the process control and optimization in the near future” (Summary). Moroney explains that
Process monitoring to assess and improve DED process reliability is broadly recognized as a being a critical step to widespread adoption in industry. To address this need, many researchers have developed methods for in-situ monitoring of various parameters of the DED process to improve reliability. Some of these methods include using a coaxial camera to monitor the melt pool [9]–[13] and optical emission spectroscopy to monitor melt pool vaporization [14][16]. Additionally, using a laser displacement sensor to measure the deposition height has been well established in literature [17]–[19]. Sammons et al. [17] developed a build sensing method that used a Keyence V7200 laser line scanner positioned perpendicular to the deposition path to scan the height of the build after each layer was completed. Chen et al. [18] used a similar displacement sensor with a clustering algorithm to detect and identify build surface defects based on the sensor’s point cloud surface reconstruction. These similar methods using displacement sensors to monitor build height in-situ were very successful and have become common across DED process monitoring efforts
(Chapter 2, Literature Review, pg. 4). Maroney proceeds to provide a solution to “in-situ monitoring of the powder flow rate near the powder focal plane” as explained below. Chen (cite 18 in Maroney, above) uses machine learning/AI to supervise and classify surface defects. See Chen et al., Rapid surface defect identification for additive manufacturing with in-situ point cloud processing and machine learning, Virtual and Physical Prototyping, 16(1), 50–67. https://doi.org/10.1080/17452759.2020.1832695, 10/15/2020. Therefore, Maroney suggests to apply machine learning to the powder feed rate monitoring and adjusting disclosed in Maroney. Finally, US 20210354397 specifically implements automatic monitoring and adjusting DED print parameters including powder deposition amounts and rate, along with solidification/energy adjustments based on the powder amounts in order to maintain desired powder stream (paras. 0067, 0091-92, 0098-99). To accomplish this real-time monitoring and adjustment, “the controller 250 utilizes machine learning or artificial intelligence to actively control the build-powder composition of the build powder 102 as the build powder 102 is being deposited to form the build-powder section 108 of the powder layer 106. Such active control also serves as quality control. Through use of the controller 250, an automated and iterative process is performed, for example, using the lookup tables, to determine both optimized build parameters and material compositions at any moment in time and/or at any point in the build of each one of the object layers” (para. 0179; emphases added). US20220063199 explains that ML is used to monitor and adjust powder-based 3D print parameters such as powder composition, humidity, speed, amounts, etc. in order to maintain certain print outcomes (Abstract, paras. 0036-43). In other words, the prior art as a whole, consistent with society as a whole, demonstrate the immense popularity, and routine application of ML/AI to all facets of industry, including to monitor 3D printing parameters such as powder feed rate.
Instant claims 1-8 and 11-18 are rejected on the ground of nonstatutory obviousness-type double patenting as being unpatentable over conflicting claims 1-19 of U.S. Patent Application No. 17/932945, in view of He at al, In-situ monitoring and deformation characterization by optical techniques; part I: Laser-aided direct metal deposition for additive manufacturing, Optics and Lasers in Engineering, Volume 122, November 2019, Pages 74-88, in view of Haley et al, Observations of particle-melt pool impact events in directed energy deposition, Additive Manufacturing, Volume 22, August 2018, Pages 368-374, in further view of Mazzucato et al, Monitoring Approach to Evaluate the Performances of a New Deposition Nozzle Solution for DED Systems, Technologies, vol. 5, no. 2, p. 29, May 2017, doi: 10.3390/technologies5020029 and Sheila Moroney, IN-SITU PROCESS MONITORING OF DIRECTED ENERGY DEPOSITION POWDER FLOW TO DETECT ANOMALIES, Thesis, published May 2023, pgs. 1-76 as evidenced by Chen et al., Rapid surface defect identification for additive manufacturing with in-situ point cloud processing and machine learning, Virtual and Physical Prototyping, 16(1), 50–67. https://doi.org/10.1080/17452759.2020.1832695, 10/15/2020, US 20210354397 and US20220063199.
The instant claims are obvious over the conflicting claims because the conflicting claims anticipate the instant by teaching an energy delivery device configured to deliver energy to a build surface of a component to form a melt pool in the build surface of the component (claims 1 & 10, further comprising an energy delivery device configured to deliver energy to the build surface of the component to form a melt pool, and wherein the powder delivery device is configured to direct the powder stream toward the melt pool); a powder delivery device configured to direct a powder stream toward the melt pool (claim 1, powder delivery device configured to direct a powder stream toward a build surface of a component, wherein the powder delivery device defines a longitudinal axis oriented toward the build surface); at least one sensor configured to generate sensor data representative of at least one process characteristic (claim 1, powder flow monitoring system comprising: an illumination device configured to illuminate at least some powder the powder stream between the powder delivery device and the build surface; and an imaging device configured to image the illuminated powder at an image
plane that intersects the longitudinal axis, wherein the illumination device and the imaging device are registered to the powder delivery device in a plane substantially orthogonal to the longitudinal axis). This imaging device is almost identical to that disclosed in Moroney (Figs. 3-1 to 3-11), and very similar to those in He (Fig. 10). As disclosed therein, and explained below, these imaging devices are used to monitor powder flows for feedback control of powder flow and laser in DED AM.
Conflicting claims do not explicitly teach a computing device configured to: receive the sensor data from the at least one sensor; determine, based on the sensor data and a predictive model data, at least one powder control parameter configured to achieve a predetermined powder feed rate of the powder stream; and control, based on the at least one powder control parameter, the energy delivery device and the powder delivery device to deposit a plurality of layers.
However, such a process was well-known in the art. For example, He teaches that high-speed cameras are routinely used in DED, PBF, SLM and EBM to monitor melt pools for closed-loop feedback control of powder delivery components such as nozzles and lasers (Fig. 10, pgs. 80-81). This measurement includes morphology, or size of the melt pool (pg. 81, col. 2). Powder feed rate is also a known process parameter that effects melt pool and its size (pg. 81, col. 2).
He does not explicitly teach to determine pool capture/absorption capability/efficiency for feedback control. However, He suggests as much. Specifically, He points to “Haley et al. [which] have recently and impressively presented a work on the direct observation of interactions between the powder particles and the melt pool using four high-speed cameras (up to 200,000 frames/s), four lens systems, and three illumination systems” (pg. 81, col. 1). He then explicitly states that this work from Haley is directly applicable to “in-situ monitoring of the melt pool temperature characteristics (e.g., maximum temperature and thermal gradient) [which] is underway, and high-speed visual observation techniques are available with different algorithms of digital images developed to characterize the morphology (e.g., length, width and surface area) of the melt pool” (pg. 81, cols. 1-2). In fact, “[a] promising trend is to integrate various inspection techniques into the DLD fabrication equipment for closed-loop control of multiple process parameters ensuring a high part quality, as shown in Fig. 10” (pg. 81, col. 2). “It is also interesting to develop analytical models based on the experimental efforts to better understand the mechanism of the melt pool and the entire process” (id.)
PNG
media_image2.png
390
792
media_image2.png
Greyscale
Thus, He explicitly suggests the claimed invention, aside from determine pool capture/absorption capability/efficiency for feedback control.
Yet, as explained above, He suggests that the work of Haley should be integrated into these feedback techniques. Just like in He, Haley teaches that measuring melt pool size and capture probability are routine (pg. 368, col. 2 (“In order to parse the many physical aspects of the L-DED and other laser based additive manufacturing systems, numerous studies have employed optical in-situ characterization techniques. These measurements tend to focus on two key areas: the trajectories and mass capture probabilities of feedstock particles after they exit the nozzle [2,5,6], and the size, shape [7] and temperature distribution of the melt pool”)). Haley goes on to teach that capture efficiency (AKA capture capability) is a function of melt pool dimensions (Apool) (Section 4.4).
PNG
media_image3.png
446
410
media_image3.png
Greyscale
PNG
media_image4.png
344
404
media_image4.png
Greyscale
PNG
media_image5.png
44
388
media_image5.png
Greyscale
Haley makes clear that capture efficiency is well-known to be affected by powder feed rate (F) (id.) Thus, He in light of Haley specifically suggests that melt pool capture efficiency, which requires melt pool size and powder feed rate data, which is dependent on feeder type and geometry, should be incorporated into feedback control.
In fact, powder feed rate is well-known to be affected by powder feeder geometry and melt pool size; and capture efficiency by laser power and powder flow rate (Mazzacuto, pg. 2, first para.; Moroney, pgs. 5, 6, 62). Further, capture efficiency is “often used as a metric to determine the quality of powder flow” (Moroney, pg. 5). In other words, the prior art makes clear that a skilled artisan would have been motivated to incorporate melt pool capture/absorption capability/efficiency, including parameters of powder flow rate, powder feeder geometry, and melt pool size, into the familiar feedback control techniques. To this end, a skilled artisan would have been motivated to use familiar automatic feedback parameters such as melt pool size to calculate melt pool capture/absorption capability/efficiency in order to control energy device (e.g. laser) and/or powder delivery device (e.g. nozzle, feeder) in DED, PBF, SLM and EBM with a reasonable expectation of success.
As to ML, as explained above, this was very obvious at the time of filing. As to machine learning, which requires training on data, the prior art is full of motivation to apply this to monitor and adjust 3D printing parameters such as powder feed rates. First, Applicants are reminded that any machine learning implementation (AKA artificial intelligence) requires training on data sets. See GeeksForGeeks, Machine Learning Tutorial, Last Updated: 5 Jun, 2026, avail at https://www.geeksforgeeks.org/machine-learning/machine-learning/, Machine Learning Basic figure. ML is the best-known branch of AI; therefore, any suggestion to apply AI would immediately be recognized by a skilled artisan as a suggestion to apply ML. Thus, any prior art that teaches or suggests to apply AI or MLM to 3D printing parameters inherently requires training on those parameters and inputs that affect them. To this end, He teaches “Based on the big data from the various in-situ monitoring with numerous parameters varied, artificial intelligent is attractive and called for the process control and optimization in the near future” (Summary). Moroney explains that
Process monitoring to assess and improve DED process reliability is broadly recognized as a being a critical step to widespread adoption in industry. To address this need, many researchers have developed methods for in-situ monitoring of various parameters of the DED process to improve reliability. Some of these methods include using a coaxial camera to monitor the melt pool [9]–[13] and optical emission spectroscopy to monitor melt pool vaporization [14][16]. Additionally, using a laser displacement sensor to measure the deposition height has been well established in literature [17]–[19]. Sammons et al. [17] developed a build sensing method that used a Keyence V7200 laser line scanner positioned perpendicular to the deposition path to scan the height of the build after each layer was completed. Chen et al. [18] used a similar displacement sensor with a clustering algorithm to detect and identify build surface defects based on the sensor’s point cloud surface reconstruction. These similar methods using displacement sensors to monitor build height in-situ were very successful and have become common across DED process monitoring efforts
(Chapter 2, Literature Review, pg. 4). Maroney proceeds to provide a solution to “in-situ monitoring of the powder flow rate near the powder focal plane” as explained below. Chen (cite 18 in Maroney, above) uses machine learning/AI to supervise and classify surface defects. See Chen et al., Rapid surface defect identification for additive manufacturing with in-situ point cloud processing and machine learning, Virtual and Physical Prototyping, 16(1), 50–67. https://doi.org/10.1080/17452759.2020.1832695, 10/15/2020. Therefore, Maroney suggests to apply machine learning to the powder feed rate monitoring and adjusting disclosed in Maroney. Finally, US 20210354397 specifically implements automatic monitoring and adjusting DED print parameters including powder deposition amounts and rate, along with solidification/energy adjustments based on the powder amounts in order to maintain desired powder stream (paras. 0067, 0091-92, 0098-99). To accomplish this real-time monitoring and adjustment, “the controller 250 utilizes machine learning or artificial intelligence to actively control the build-powder composition of the build powder 102 as the build powder 102 is being deposited to form the build-powder section 108 of the powder layer 106. Such active control also serves as quality control. Through use of the controller 250, an automated and iterative process is performed, for example, using the lookup tables, to determine both optimized build parameters and material compositions at any moment in time and/or at any point in the build of each one of the object layers” (para. 0179; emphases added). US20220063199 explains that ML is used to monitor and adjust powder-based 3D print parameters such as powder composition, humidity, speed, amounts, etc. in order to maintain certain print outcomes (Abstract, paras. 0036-43). In other words, the prior art as a whole, consistent with society as a whole, demonstrate the immense popularity, and routine application of ML/AI to all facets of industry, including to monitor 3D printing parameters such as powder feed rate.
Instant claims 1-8 and 11-18 are rejected on the ground of nonstatutory obviousness-type double patenting as being unpatentable over conflicting claims 1-20 of U.S. Patent No. 12333737, in view of He at al, In-situ monitoring and deformation characterization by optical techniques; part I: Laser-aided direct metal deposition for additive manufacturing, Optics and Lasers in Engineering, Volume 122, November 2019, Pages 74-88, in view of Haley et al, Observations of particle-melt pool impact events in directed energy deposition, Additive Manufacturing, Volume 22, August 2018, Pages 368-374, in further view of Mazzucato et al, Monitoring Approach to Evaluate the Performances of a New Deposition Nozzle Solution for DED Systems, Technologies, vol. 5, no. 2, p. 29, May 2017, doi: 10.3390/technologies5020029 and Sheila Moroney, IN-SITU PROCESS MONITORING OF DIRECTED ENERGY DEPOSITION POWDER FLOW TO DETECT ANOMALIES, Thesis, published May 2023, pgs. 1-76 as evidenced by Chen et al., Rapid surface defect identification for additive manufacturing with in-situ point cloud processing and machine learning, Virtual and Physical Prototyping, 16(1), 50–67. https://doi.org/10.1080/17452759.2020.1832695, 10/15/2020, US 20210354397 and US20220063199.
The instant claims are obvious over the conflicting claims because the conflicting claims anticipate the instant by teaching an energy delivery device configured to deliver energy to a build surface of a component to form a melt pool in the build surface of the component (claim 1, the blown powder additive manufacturing process includes adding material via a powder delivery device to a build surface of a component in sequential layers, requires energy source as shown in Figures); a powder delivery device configured to direct a powder stream toward the melt pool (claim 1, powder delivery device); at least one sensor configured to generate sensor data representative of at least one process characteristic (claim 1, powder flow monitoring system- illumination device w/ imaging device). This powder flow monitoring system is almost identical to that disclosed in Moroney (Figs. 3-1 to 3-11), and very similar to those in He (Fig. 10). As disclosed therein, and explained below, these imaging devices are used to monitor powder flows for feedback control of powder flow and laser in DED AM.
Conflicting claims do not explicitly teach a computing device configured to: receive the sensor data from the at least one sensor; determine, based on the sensor data and a predictive model data, at least one powder control parameter configured to achieve a predetermined powder feed rate of the powder stream; and control, based on the at least one powder control parameter, the energy delivery device and the powder delivery device to deposit a plurality of layers.
However, such a process was well-known in the art. For example, He teaches that high-speed cameras are routinely used in DED, PBF, SLM and EBM to monitor melt pools for closed-loop feedback control of powder delivery components such as nozzles and lasers (Fig. 10, pgs. 80-81). This measurement includes morphology, or size of the melt pool (pg. 81, col. 2). Powder feed rate is also a known process parameter that effects melt pool and its size (pg. 81, col. 2).
He does not explicitly teach to determine pool capture/absorption capability/efficiency for feedback control. However, He suggests as much. Specifically, He points to “Haley et al. [which] have recently and impressively presented a work on the direct observation of interactions between the powder particles and the melt pool using four high-speed cameras (up to 200,000 frames/s), four lens systems, and three illumination systems” (pg. 81, col. 1). He then explicitly states that this work from Haley is directly applicable to “in-situ monitoring of the melt pool temperature characteristics (e.g., maximum temperature and thermal gradient) [which] is underway, and high-speed visual observation techniques are available with different algorithms of digital images developed to characterize the morphology (e.g., length, width and surface area) of the melt pool” (pg. 81, cols. 1-2). In fact, “[a] promising trend is to integrate various inspection techniques into the DLD fabrication equipment for closed-loop control of multiple process parameters ensuring a high part quality, as shown in Fig. 10” (pg. 81, col. 2). “It is also interesting to develop analytical models based on the experimental efforts to better understand the mechanism of the melt pool and the entire process” (id.)
PNG
media_image2.png
390
792
media_image2.png
Greyscale
Thus, He explicitly suggests the claimed invention, aside from determine pool capture/absorption capability/efficiency for feedback control.
Yet, as explained above, He suggests that the work of Haley should be integrated into these feedback techniques. Just like in He, Haley teaches that measuring melt pool size and capture probability are routine (pg. 368, col. 2 (“In order to parse the many physical aspects of the L-DED and other laser based additive manufacturing systems, numerous studies have employed optical in-situ characterization techniques. These measurements tend to focus on two key areas: the trajectories and mass capture probabilities of feedstock particles after they exit the nozzle [2,5,6], and the size, shape [7] and temperature distribution of the melt pool”)). Haley goes on to teach that capture efficiency (AKA capture capability) is a function of melt pool dimensions (Apool) (Section 4.4).
PNG
media_image3.png
446
410
media_image3.png
Greyscale
PNG
media_image4.png
344
404
media_image4.png
Greyscale
PNG
media_image5.png
44
388
media_image5.png
Greyscale
Haley makes clear that capture efficiency is well-known to be affected by powder feed rate (F) (id.) Thus, He in light of Haley specifically suggests that melt pool capture efficiency, which requires melt pool size and powder feed rate data, which is dependent on feeder type and geometry, should be incorporated into feedback control.
In fact, powder feed rate is well-known to be affected by powder feeder geometry and melt pool size; and capture efficiency by laser power and powder flow rate (Mazzacuto, pg. 2, first para.; Moroney, pgs. 5, 6, 62). Further, capture efficiency is “often used as a metric to determine the quality of powder flow” (Moroney, pg. 5). In other words, the prior art makes clear that a skilled artisan would have been motivated to incorporate melt pool capture/absorption capability/efficiency, including parameters of powder flow rate, powder feeder geometry, and melt pool size, into the familiar feedback control techniques. To this end, a skilled artisan would have been motivated to use familiar automatic feedback parameters such as melt pool size to calculate melt pool capture/absorption capability/efficiency in order to control energy device (e.g. laser) and/or powder delivery device (e.g. nozzle, feeder) in DED, PBF, SLM and EBM with a reasonable expectation of success.
As to ML, as explained above, this was very obvious at the time of filing. As to machine learning, which requires training on data, the prior art is full of motivation to apply this to monitor and adjust 3D printing parameters such as powder feed rates. First, Applicants are reminded that any machine learning implementation (AKA artificial intelligence) requires training on data sets. See GeeksForGeeks, Machine Learning Tutorial, Last Updated: 5 Jun, 2026, avail at https://www.geeksforgeeks.org/machine-learning/machine-learning/, Machine Learning Basic figure. ML is the best-known branch of AI; therefore, any suggestion to apply AI would immediately be recognized by a skilled artisan as a suggestion to apply ML. Thus, any prior art that teaches or suggests to apply AI or MLM to 3D printing parameters inherently requires training on those parameters and inputs that affect them. To this end, He teaches “Based on the big data from the various in-situ monitoring with numerous parameters varied, artificial intelligent is attractive and called for the process control and optimization in the near future” (Summary). Moroney explains that
Process monitoring to assess and improve DED process reliability is broadly recognized as a being a critical step to widespread adoption in industry. To address this need, many researchers have developed methods for in-situ monitoring of various parameters of the DED process to improve reliability. Some of these methods include using a coaxial camera to monitor the melt pool [9]–[13] and optical emission spectroscopy to monitor melt pool vaporization [14][16]. Additionally, using a laser displacement sensor to measure the deposition height has been well established in literature [17]–[19]. Sammons et al. [17] developed a build sensing method that used a Keyence V7200 laser line scanner positioned perpendicular to the deposition path to scan the height of the build after each layer was completed. Chen et al. [18] used a similar displacement sensor with a clustering algorithm to detect and identify build surface defects based on the sensor’s point cloud surface reconstruction. These similar methods using displacement sensors to monitor build height in-situ were very successful and have become common across DED process monitoring efforts
(Chapter 2, Literature Review, pg. 4). Maroney proceeds to provide a solution to “in-situ monitoring of the powder flow rate near the powder focal plane” as explained below. Chen (cite 18 in Maroney, above) uses machine learning/AI to supervise and classify surface defects. See Chen et al., Rapid surface defect identification for additive manufacturing with in-situ point cloud processing and machine learning, Virtual and Physical Prototyping, 16(1), 50–67. https://doi.org/10.1080/17452759.2020.1832695, 10/15/2020. Therefore, Maroney suggests to apply machine learning to the powder feed rate monitoring and adjusting disclosed in Maroney. Finally, US 20210354397 specifically implements automatic monitoring and adjusting DED print parameters including powder deposition amounts and rate, along with solidification/energy adjustments based on the powder amounts in order to maintain desired powder stream (paras. 0067, 0091-92, 0098-99). To accomplish this real-time monitoring and adjustment, “the controller 250 utilizes machine learning or artificial intelligence to actively control the build-powder composition of the build powder 102 as the build powder 102 is being deposited to form the build-powder section 108 of the powder layer 106. Such active control also serves as quality control. Through use of the controller 250, an automated and iterative process is performed, for example, using the lookup tables, to determine both optimized build parameters and material compositions at any moment in time and/or at any point in the build of each one of the object layers” (para. 0179; emphases added). US20220063199 explains that ML is used to monitor and adjust powder-based 3D print parameters such as powder composition, humidity, speed, amounts, etc. in order to maintain certain print outcomes (Abstract, paras. 0036-43). In other words, the prior art as a whole, consistent with society as a whole, demonstrate the immense popularity, and routine application of ML/AI to all facets of industry, including to monitor 3D printing parameters such as powder feed rate.
Instant claims 1-8 and 11-18 are rejected on the ground of nonstatutory obviousness-type double patenting as being unpatentable over conflicting claims 1-20 of U.S. Patent Application No. 18/593538, in view of He at al, In-situ monitoring and deformation characterization by optical techniques; part I: Laser-aided direct metal deposition for additive manufacturing, Optics and Lasers in Engineering, Volume 122, November 2019, Pages 74-88, Sheila Moroney, IN-SITU PROCESS MONITORING OF DIRECTED ENERGY DEPOSITION POWDER FLOW TO DETECT ANOMALIES, Thesis, published May 2023, pgs. 1-76 as evidenced by Chen et al., Rapid surface defect identification for additive manufacturing with in-situ point cloud processing and machine learning, Virtual and Physical Prototyping, 16(1), 50–67. https://doi.org/10.1080/17452759.2020.1832695, 10/15/2020, US 20210354397 and US20220063199.
The instant claims are obvious over the conflicting claims because the conflicting claims anticipate the instant by teaching an energy delivery device configured to deliver energy to a build surface of a component to form a melt pool in the build surface of the component (claims 1, energy delivery device); a powder delivery device configured to direct a powder stream toward the melt pool (claim 1, powder delivery device); at least one sensor configured to generate sensor data representative of at least one process characteristic (claim 1, powder flow monitoring system comprising optical system). This imaging device is almost identical to that disclosed in Moroney (Figs. 3-1 to 3-11), and very similar to those in He (Fig. 10); computing device configured to: receive the sensor data from the at least one sensor (claim 1, computing device); determine, based on the sensor data and a predictive model data, at least one powder control parameter configured to achieve a predetermined powder feed rate of the powder stream (claim 12); and control, based on the at least one powder control parameter, the energy delivery device and the powder delivery device to deposit a plurality of layers (claim 1).
Claim 7 sates “[t]he additive manufacturing system of claim 6, wherein the computing device is configured to determine, using a machine learning model that takes the determined captured efficiency as an input, the adjustment to one or both of the position of the powder delivery device or the energy delivery device.”
Further as to ML, as explained above, this was very obvious at the time of filing. As to machine learning, which requires training on data, the prior art is full of motivation to apply this to monitor and adjust 3D printing parameters such as powder feed rates. First, Applicants are reminded that any machine learning implementation (AKA artificial intelligence) requires training on data sets. See GeeksForGeeks, Machine Learning Tutorial, Last Updated: 5 Jun, 2026, avail at https://www.geeksforgeeks.org/machine-learning/machine-learning/, Machine Learning Basic figure. ML is the best-known branch of AI; therefore, any suggestion to apply AI would immediately be recognized by a skilled artisan as a suggestion to apply ML. Thus, any prior art that teaches or suggests to apply AI or MLM to 3D printing parameters inherently requires training on those parameters and inputs that affect them. To this end, He teaches “Based on the big data from the various in-situ monitoring with numerous parameters varied, artificial intelligent is attractive and called for the process control and optimization in the near future” (Summary). Moroney explains that
Process monitoring to assess and improve DED process reliability is broadly recognized as a being a critical step to widespread adoption in industry. To address this need, many researchers have developed methods for in-situ monitoring of various parameters of the DED process to improve reliability. Some of these methods include using a coaxial camera to monitor the melt pool [9]–[13] and optical emission spectroscopy to monitor melt pool vaporization [14][16]. Additionally, using a laser displacement sensor to measure the deposition height has been well established in literature [17]–[19]. Sammons et al. [17] developed a build sensing method that used a Keyence V7200 laser line scanner positioned perpendicular to the deposition path to scan the height of the build after each layer was completed. Chen et al. [18] used a similar displacement sensor with a clustering algorithm to detect and identify build surface defects based on the sensor’s point cloud surface reconstruction. These similar methods using displacement sensors to monitor build height in-situ were very successful and have become common across DED process monitoring efforts
(Chapter 2, Literature Review, pg. 4). Maroney proceeds to provide a solution to “in-situ monitoring of the powder flow rate near the powder focal plane” as explained below. Chen (cite 18 in Maroney, above) uses machine learning/AI to supervise and classify surface defects. See Chen et al., Rapid surface defect identification for additive manufacturing with in-situ point cloud processing and machine learning, Virtual and Physical Prototyping, 16(1), 50–67. https://doi.org/10.1080/17452759.2020.1832695, 10/15/2020. Therefore, Maroney suggests to apply machine learning to the powder feed rate monitoring and adjusting disclosed in Maroney. Finally, US 20210354397 specifically implements automatic monitoring and adjusting DED print parameters including powder deposition amounts and rate, along with solidification/energy adjustments based on the powder amounts in order to maintain desired powder stream (paras. 0067, 0091-92, 0098-99). To accomplish this real-time monitoring and adjustment, “the controller 250 utilizes machine learning or artificial intelligence to actively control the build-powder composition of the build powder 102 as the build powder 102 is being deposited to form the build-powder section 108 of the powder layer 106. Such active control also serves as quality control. Through use of the controller 250, an automated and iterative process is performed, for example, using the lookup tables, to determine both optimized build parameters and material compositions at any moment in time and/or at any point in the build of each one of the object layers” (para. 0179; emphases added). US20220063199 explains that ML is used to monitor and adjust powder-based 3D print parameters such as powder composition, humidity, speed, amounts, etc. in order to maintain certain print outcomes (Abstract, paras. 0036-43). In other words, the prior art as a whole, consistent with society as a whole, demonstrate the immense popularity, and routine application of ML/AI to all facets of industry, including to monitor 3D printing parameters such as powder feed rate.
Prior Art
The following prior art is also relevant to the obvious application of ML to powder-based 3D printing: US 20200257933; US 20210354387; EP 3970886 A1; US 11487271.
Conclusion
No claims are allowed.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MELODY TSUI whose telephone number is (571)272-1846. The examiner can normally be reached Monday - Friday, 9am - 5pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Heather Calamita can be reached at 571-272-2876. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/YUNG-SHENG M TSUI/ Primary Examiner, Art Unit 1684
Prosecution Insights