Difference between revisions of "Laser Powder Bed Fusion - Metal"

From MIT Technology Roadmapping
Jump to navigation Jump to search
 
(20 intermediate revisions by 2 users not shown)
Line 1: Line 1:
=Technology Roadmap Sections and Deliverables=
=Technology Roadmap Sections and Deliverables=


The clear and unique identifier for this technology roadmap is:  
The clear and unique identifier for this technology roadmap is:
* '''2LPBF-M - Laser Powder Bed Fusion - Metal'''
 
This indicates that we are dealing with a “level 2” roadmap at the specific implementation level, where “level 1” would have indicated the over-arching roadmap and “level 3” or “level 4” would have indicated an individual technology roadmap.
* '''3LPB - Laser Powder Bed Fusion - Metal'''
This indicates that we are dealing with a “level 3” roadmap at the specific implementation level, where “level 1”and "level 2" would have indicated the over-arching roadmap and “level 4” would have indicated an individual technology roadmap.
 
In select sections of the roadmap, "3LPB" signifies a fictional company seeking to develop an improved LPBF-M machine.
 


==Roadmap Overview==
==Roadmap Overview==
Line 9: Line 13:
The working principle and architecture of Laser Powder Bed Fusion - Metal is depicted in the image below.  
The working principle and architecture of Laser Powder Bed Fusion - Metal is depicted in the image below.  


[[Image:Overview_Image.png|center|1000px]]
[[File:M400.png|center|600px]]
 
Figure: EOS M400-4 LPBF-M machine (source: EOS)
 
[[File:PD2.png|center|600px]]
 
Figure: LPBF-M Process (source: CustomPartNet)
 
Laser Powder Bed Fusion - Metal (LPBF-M) is an additive manufacturing (AM) technology that enables the creation complex metal components directly from a digital 3D model without any tools. While the number of available materials is still limited compared to other milling and injection molding processes, LPBF-M utilizes various metals such as cobalt chrome, nickel alloys, titanium, aluminum, stainless steel and copper to generate robust functional prototypes and production parts that are capable of surviving in severe use cases. The LPBF-M technology offers comparable quality to parts made with traditional manufacturing methods, with material properties surpassing those of cast parts and approaching those of billet. It is most widely used in aerospace, medical, motorsports, energy, and molding. Beyond highly-stressed functional components, LPBF-M can be used for producing parts in cosmetic applications, manufacturing aids, small integrated structures, dental components, surgical implants.
 
The process shares similarities with many other layer-wise additive manufacturing technologies. A program utilizes 3D model data and mathematically slices it into 2D cross-sections. Each section will act as a template that lets the LPBF-M machine know where to precisely create perimeters and cross-hatching. The data is transferred to the LPBF-M equipment. Subsequently, a "recoater" spreads a 25-120µm thick layer of powdered metal to produce a uniform layer over the solid metal build plate. A laser then draws a 2D cross-section on the powdered material, fusing the substrate into a solid. Once a layer is complete, the base plate is lowered enough to make room for the next layer. More material is collected from the reservoir and recoated evenly on the previously sintered layer. The LPBF-M machine continues to create layer upon layer building from the bottom up as the part is built. Support structures are added to part to provide a reinforcement for overhangs, and to prevent thermally expansion from negatively impacting dimensional accuracy. Following printing, the unfused powder is vacuumed from the chamber and the build plate is removed from the machine. Thereafter, the components typically undergo a stress relief and/or heat treatment process to remove residual thermal stresses and improve material properties. Following heat treatment, the next steps are support removal, detachment from the build plate (via bandsaw or wire EDM), and media blasting for final powder removal.
 
 
[[File:INDST.png|center|600px]]


Laser Powder Bed Fusion - Metal (LPBF-M)is an additive manufacturing technology that enables the creation of more or less complex metal components directly from a 3D model without any tools. While the number of available materials is still limited compared to other milling and injection molding processes, LPBF-M utilizes various metal and alloy materials such as stainless steel and cobalt chrome to generate substantial and durable parts, functional metal prototypes, high-temperature applications, and end-use parts. The LPBF-M technology offers comparable quality to parts made with traditional manufacturing methods. LPBF-M can be used for producing parts in highly cosmetic applications, manufacturing aids, small integrated structures, dental components, surgical implants, and aerospace parts.
Figure: Industry breakdown for installed metal powder bed fusion machines. This figure includes non-laser powder bed machines but 85-90% of these unit sare LPBF-M. (source: Dr. Maximilian Munsch, AMPOWER 2021 Additive Manufacturing Report)


The process is almost the same as other layer additive manufacturing technologies. A program utilizes 3D model data and mathematically slices it into 3D cross-sections. Each section will act as a draft that lets the LPBF-M machine know where to center the metal material precisely. The data is transferred to the LPBF-M equipment to assembly powered metal material from the powers to produce a uniform layer over the base plane. A layer then draws a 2D cross-section on the build material, fusing the material using a laser beam or an electron beam. Once a layer is complete, the base plate is lowered enough to make room for the next layer. More material is raised from the cartridge and recoated evenly on the previously sintered layer. The LPBF-M machine continues to center layer upon layer building from the bottom up as the part is built. Support structures are added to gives supplemental strength to find features and overhanging surfaces. The completed part is removed from the base play and treated with an age-hardening heat process to harden the part further.
There is also an environmental advantage in the raw material efficiency of the processes. While the scrap rate for many complex milled parts is over 90%, the scrap rate for LPBF-M parts is typically less than 5%. While milled chips can be recycled, they must be processed offsite using considerable energy. In many cases, the LPBF-M powder can be immediately returned to a machine reservoir following sieving.


The advantage is the excellent material efficiency of most additive manufacturing processes. While the scrap rate for many complex milled parts is over 90%, the scrap rate for LPBF-M parts is typically less than 5%.  
[[File:A350.png|center|600px]]


With decreasing available raw materials and rising costs, this material efficiency will be a significant advantage in the long run. In the future, the introduction of high-speed systems with more powerful lasers and larger build chambers is expected to increase the share of LPBF-M in the production process, and a significant number of materials will become compatible with LPBF-M.
Figure: A350 XWB cabin brackets (source: Airbus)


The primary deficits of the technology in its current form are slow production rates, expensive machines, and high powder costs due to low volume requirements. The combination of these three challenges results in a high part piece cost. However, with improved performance, elimination of tooling, and reduced lead-time, there are many positive business cases for LPBF-M. In the future, the introduction of high-speed systems with more powerful lasers and larger build chambers is expected to increase the number of viable economically-viable applications that will drive increased demand and scale for LPBF-M systems.


[[File:COST2.png|center|600px]]


==Design Structure Matrix (DSM) Allocation==
Figure: Cost comparison for conventional manufacturing (CM), additive components, and AM-enhanced tooling. The freedom provided by AM methods like LPBF-M provides entry to design spaces that would otherwise be impractical or impossible to access. (source: Modified from John Hart and Haden Quinlan, MIT xPro "Additive Manufacturing for Innovative Design and Production course," MIT)


[[Image:LPBF-M_DSM.png|center|750px]]
[[Image:LPBF-M_decomposition.png|center|750px]]


The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here. The sentence will come here.
 
The LPBF-M market is maturing with breakthrough applications across multiple industries that provide the justification and institutional know-how to continue growing the technology.
 
[[File:TRND.png|center|600px]]
 
Figure: Metal powder bed fusion machines expected sales show a predicted 6000-7500 units by 2025. This figure includes non-laser powder bed machines but 85-90% of these sales are LPBF-M, bringing the predicted totals to 5100-6750. (source: Dr. Maximilian Munsch, AMPOWER 2021 Additive Manufacturing Report)
 
 
[[File:SALES.png|center|600px]]
 
Figure: Metal powder bed fusion is responsible for 39% of sales revenue of all AM machines. While significantly more expensive than polymer systems (>2X), LPBF-M sales revenue is reflective of the technology's increasing foothold. (source: AMPOWER 2021 Additive Manufacturing Report)
 
Finally, LPBF-M is the leading metal AM technology from an industrialization and technology maturity standpoint, yet it still has significant room for improvement, making it the ideal technology for roadmapping.
 
[[File:MATR.png|center|600px]]
 
Figure: LPBF-M leads industrialization and technology maturity indices. (source: AMPOWER 2021 Additive Manufacturing Report)
 
==Design Structure Matrix (DSM) Allocation[edit]==
 
The classification tree below shows us that LPBF-M is part of the larger Additive system. The DSM and tree both show that LPBF-M requires the following technologies at subsystem level 4: Laser (4L), Powdered Metal (4PM), Scanner (4S), Recoater (4R), and Build Chamber (4BC). Each level 3 subsystem also requires enabling technologies shown as level 4 systems.
 
 
[[File:LPBF-M_DSM_allocation_2.png|center|1000px]]
 


==Roadmap Model using OPM==
==Roadmap Model using OPM==
We provide an Object-Process-Diagram (OPD) of the 2LPBF-M roadmap in the figure below. This diagram captures the main object of the roadmap (Laser Powder Bed Fusion - Metal), its various instances including development projects, its decomposition into subsystems, its characterization by Figures of Merit (FOMs) as well as the main processes.
An Object-Process-Diagram (OPD) of LPBF-M is provided in the figure below. This diagram captures the main object of the roadmap (LPBF-M), its various instances including development projects, its decomposition into subsystems, its characterization by Figures of Merit (FOMs) as well as the primary processes.
 
[[File:LPBF-M OPM.png|600px]]


[[File:LPBF-M_OPM.png|center|1000px]]
Figure: Object-Process-Diagram for LPBF-M


An Object-Process-Language (OPL) description of the roadmap scope is auto-generated and given below. It reflects the same content as the previous figure, but in a formal natural language.


An Object-Process-Language (OPL) description of the roadmap scope is auto-generated and given below. It reflects the same content as the previous figure, but in a formal natural language.


[[File:LPBF-M_OPL.png|center|1000px]]
[[File:LPBF-M OPL.png|600px]]
 
Figure: Object-Process-Language for the LPBF-M OPM


==Figures of Merit==
==Figures of Merit==
The table below shows a list of FOMs by which in-space additive manufacturing can be assessed. Most of the FOMs are related to the performance of the printer such as build rate, resolution, build volume, lead time and mean time to failure. Feedstock mass, as well as cost, relates to the manufacturing efficiency when creating the printer itself. Finally, peak power, average power, and density relate to the energy consumption and efficiency of printing.


[[Image:FOM Table 1.png|800px|center]]
The table below shows a list of FOMs by which Laser Powder Bed Fusion – Metal, LPB can be assessed. For LPBF-M, the key FOMs for increased productivity are build rate (also called printing speed), build chamber volume, and layer thickness. Additional FOMs such as laser power, laser count, and laser scan speed play a critical role in productivity as well. FOMs on this list can be categorized into four types, productivity, performance, efficiency, and competitiveness. Most of these, including key FOMs, are explicitly related to productivity. Thus, productivity as a FOM would be a critical indicator to evaluate technological progress.
[[Image:Table of FOM Equations.png|800px|center]]
 
 
[[File:FOMCHARTV2.jpg|600px]]
 
Figure: Primary figures of merit for the LPBF-M technology
 
 
[[File:BRV5.jpg|600px]]
 
Figure: Build rate change over time. Printing speed has grown step-wise over the brief timeline of LPBF-M. Today's fastest commercial machines contain up to 12 lasers, have automation provisions for rapid changeovers, and are modularized for quick repairs. In addition to machine type, build rate varies by material, layer height, and machine parameter settings so the plotted data points are representative values. (Source: Machine manufacture datasheets)
 
 
[[File:NXG.png|600px]]
 
 
[[File:CLML.png|600px]]
 
Figure: GE Additive's Concept Laser M Line system is highly modularized and scalable. (Source: GE Additive)


==Alignment with Company Strategic Drivers==
Alignment with Company Strategic Drivers[edit]
In select sections of the roadmap, "3LPB" signifies a fictional company seeking to develop an improved LPBF-M machine.


The table below shows an example of strategic drivers and their alignment with the Zero-Gravity Additive Manufacturing (2ZGAM) roadmap.
The table below shows an example of strategic drivers and their alignment with 3LPB's Laser Powder Bed Fusion - Metal roadmap.


[[Image:Company Strategic.png|center| 800px]]
3LPB's roadmap is focused on meeting a wide variety of 3D printer needs, such as rapid prototyping and serial production applications, different surface qualities, and cost reduction by decreasing production time. By balancing the trade-off between surface roughness and time-to-build, and reducing production time and material costs on the part of the user, 3LPB hopes to meet the needs of a large number of users while achieving significant cost efficiencies and productivity improvements. As a result, 3LBP will be able to increase its penetration rate and significantly impact the productivity of each market on a national scale, contributing to stable and sustainable industrial growth regardless of labor shortages and economic conditions.


The list of drivers shows that the company wants to develop an in-space manufacturing capability that can fabricate at least 30% of spare parts for future long-endurance human exploration missions by 2035. The company wants to develop this in-space manufacturing capability within the size, weight, and power (SWaP) allocation for future space missions. Both of these aforementioned strategic drivers are aligned with the technology roadmap at present. As a third strategic driver, the company also wants to ensure that future spacecraft component design incorporates principles of design for in-space manufacturing (DFISM) to facilitate the adoption of ISM and realization of benefits. However, this driver is not aligned with the roadmap as currently scoped. While there is a clear interplay between the design of components and the required manufacturing capability for their fabrication, this roadmap makes an assumption of fixed component designs to focus specifically on the manufacturing technology development alone.


A target of 30% parts was determined as a feasible target which can be achieved by 2035. A high level of resolution is required for the manufacturing of spare parts for space systems due to the sensitivity of the systems involved and the complexity of the parts to be manufactured. However, the resolution is not the only critical factor as AM technologies that have very high resolution but low build rates would compromise on the time-sensitivity of the required spare parts. Hence it is essential that we are able to manufacture a wide variety of parts that are precise in their structural composition. If we are not able to manufacture at least 30% spare parts, then there would be minimal additional value for the launching organizations or government agencies to invest in the same and they would continue with the present process of terrestrial manufacturing and deployment to space.
[[File:Alignment-table.png|600px]]
 


==Positioning of Company vs. Competition==
==Positioning of Company vs. Competition==


We compared and organized the FOM from related/competitive companies. Some key data/info is confidential, therefore we will keep it blank.
In select sections of the roadmap, "3LPB" signifies a fictional company seeking to develop an improved LPBF-M machine.
 
The following two figures are a table of the current competitive situation with key FOMs and a representation of our position and the situation of other companies in promoting the above strategy of 3LPB. Some key data/info is confidential, therefore it will be kept blank.
 
Table: Fictional company 3LPB's technical market positioning. (Source: Machine manufacture datasheets)
 
 
[[File:POSI.png|600px]]
 
 
The following figure is the Pareto Front of Surface Roughness vs. Build Time for Various Additive Manufacturing Laser Powder Bed Fusion (Metal). Currently, 3LPB is right on the Pareto Front of the two FOMs; however By 2036 the company would like to achieve a surface roughness of less than 6 µm and a build time of 3 hours per 1000 cm³ part.
 
 
[[File:PF.png|600px]]
 
Figure: Current Pareto Front and fictional company 3LPB's intended destination. (Source: Machine manufacture datasheets)
 
Technical Model[edit]
In order to assess the feasibility of technical targets at the level of the 3LPB roadmap it is necessary to develop a technical model. The purpose of such a model is to explore the design tradespace and establish what are the active constraints in the system. The first step can be to establish a morphological matrix that shows the main technology selection alternatives that exist at the first level of decomposition, see the figure below.
 
Table: Technical Model: Morphological Matrix and Tradespace. (Source: Machine manufacture datasheets)
 
 
[[File:1.7 3LPB.png|600px]]
 
 
[[File:4LSR.png|600px]]
 
Figure: EOS M400-4 four laser system. The lasers can either be isolated to one laser per part (i.e. several small parts) or combined into multiple lasers per part (i.e. a large part). Use of more than one laser per part creates a visible seam line that can result in the need for real or perceived additional component validation. (Source: EOS)
 
 
The relationship between the parameters of laser power, scan speed, and spot size, and outputs of porosity and surface roughness is viewed as a primary physics-based limiting factor (Zhao 2020, Gee 2020). This is investigated further below.
 
 
[[File:PWR-SPD.png|600px]]
 
Figure: An exponential boundary is created for the threshold between stable melting and keyhole porosity in the Zhao study. Maximum power and speed--without increasing surface roughness beyond the acceptable specification--is the preferred operating window. (source: Zhao 2020)
 
 
[[File:GEE.png|600px]]
 
Figure: Tradeoffs (or operating windows) for build rate as a function of laser power, spot size, scan speed. The color gradient is build rate, the black lines indicate scan speed, and the white lines can be ignored for this discussion. (source: Gee 2020)
 
 
 
==Key Publications, Presentations and Patents==
 
Fundamental Patents[edit]
 
US Patent 4863538
 
Date: 1989
 
Assignee: University of Texas
 
Method and apparatus for producing parts by selective sintering
 
The original utility patent for laser sintering powder bed polymer which expired in 2007 kicking off the revolution of 3D printing technology advancement.
 
[[File:3DS.png|400px]]
 
US Patent 5460758
 
Date: 1995
 
Assignee: EOS / 3D Systems
 
Method and apparatus for production of a three-dimensional object
 
The first metal laser sintering patent enabled by the work or EOS and the Fraunhofer Institute. As laser power increased, this would transform from sintering to fusing (welding). Expired in 2013, creating increased competition.
 
 
[[File:EOSP.png|400px]]
 
==State of the Art Patents==
 
US Patent 10919090
 
Date: 2021
 
Assignee: Vulcanforms Inc.
 
Additive manufacturing by spatially controlled material fusion
 
This patent is for a LPBF-M machine with a plurality of modulated line-shaped laser systems that provide the potential to rapidly speed production and improve fusion.
 
[File:LINE4.png|600px]]
 
US Patent 10195693
 
Date: 2019
 
Assignee: Velo3D Inc.
 
Apparatuses, systems and methods for three-dimensional printing


[[Image:FOM Table And Description.png|center| 800px]]
This patent describes low-impact recoating mechanisms like those used in the Velo3D Sapphire machine that prints with no or minimal supports, thereby reducing post-processing time and increasing productivity.


The following figure is the Pareto Front of Resolution vs. Build Rate for Various Additive Manufacturing Processes
[[File:VELO.png|600px]]


[[Image:FOM Chart 2.png|center| 800px]]
US Patent 10974456


The utopia point for the chart above is on the bottom right corner and at-present the Pareto front for Terrestrial SOA is closer to the utopia point than for ISM SOA. The parts which we will be manufacturing in space need to be of high quality and should have a reasonable manufacturing time. If the parts produced in space have a much lower resolution (higher micrometers) and have a low build rate (and consequently high build times and mean times to failure), then the manufactured parts may be unusable and provide minimal advantage over manufacturing on earth. The focus here is to shift the Pareto front for ISM SOA closer to the utopia point which can be achieved by maximizing resolution (lower micrometers) and build rate within reasonable limits. Working on improvements in both areas align with our strategic drivers and will help us achieve our goals within the target dates specified in the Technology Strategy Statement mentioned in the latter section.
Date: 2021


While there are many technically feasible approaches to additive manufacturing of parts and systems, not all are adaptable to space. Present limitations for in-space manufacturing which contribute to the differences in the Pareto front for resolution and build rate include -
Assignee: Lawrence Livermore National Security LLC.


# Space systems engineering is a complex discipline with very mature methods of certification which results in the deployment of extensively tested apparatus. Due to the extensiveness of testing before sending it to space, AM technologies have been progressing faster for terrestrial applications.
Additive manufacturing power map to mitigate defects
# Current additive manufacturing systems deliver accurate and precise results in a 1 g, thermally controlled environment and are well understood. The same level of information is not available for ISM as there is much more testing for terrestrial AM due to less stringent certification methods.
# Current processes such as photolithography can create electric components at scales of 35 nanometers as compared to common AM resolution (>50 microns) which would produce components 1,000 times larger than the physical size of currently available parts. This disparity in performance discourages investment by organizations to conduct research in this area.
# Physics-based models of in-space additive manufacturing processes are needed to understand and predict material properties and help optimize material composition. 
# The thermal effects of energy source and energy density in space has not yet been extensively researched. 
# There is not enough investment in systems that produce open-system design, planning, simulation, and analysis tools for ISM. 
# Design for and construction of objects in space will likely require much less mass, due to the reduced gravity, but it is difficult to predict the overall mass reduction and corresponding impact on construction time without knowing the resolution required and the impact of other environmental effects on the process.
# The impact of vacuum and thermal environments on the AM technology is not holistically understood due to a lack of data.


'''References:'''
This patent includes methods for mapping laser power to select geometry for improved print quality via defect reduction. When implemented in-situ, this type of control mechanism offers significant potential in quality and productivity improvement.


https://www.nap.edu/read/18871/chapter/5#60
[[File:LINL.png|600px]]


==Technical Model==
US Patent 10583529


In order to assess the feasibility of technical (and financial) targets at the level of the 2ZGAM roadmap it is necessary to develop a technical model. The purpose of such a model is to explore the design tradespace and establish what are the active constraints in the system. The first step can be to establish a morphological matrix that shows the main technology selection alternatives that exist at the first level of decomposition, see the figure below.
Date: 2020


''Technical Model: Morphological Matrix and Tradespace''
Assignee: EOS Of North America Inc


[[Image:Morphological Matrix.png|center | 800px]]
Additive manufacturing method using a plurality of synchronized laser beams


[[Image:Morphological Matrix 2.png|center | 1000px]]
This patent provides methods of using multiple laser beams and scanners in unison. Packaging and coordinating the optical systems is one of the greatest challenges of multi-laser systems. Intelligent packaging and control systems like that included in this patent will increase both quality and productivity.


''Mathematical Equations''
[[File:EONA.png|600px]]


For this problem, we will focus on two figures of merit: (1) the ratio of build volume to overall manufacturing equipment volume, and (2)  build rate [cm^3/hr]. Whereas the volume ratio will be determined using statistical regression techniques from available data about existing manufacturing equipment, the build rate will be computed from the fundamental physics of the manufacturing process (although it can also be validated with experimental results).


By looking at analogous Earth-based manufacturing equipment with plenty of available data on manufacturing equipment build volume and overall footprint, it is possible to derive a statistical relationship between build volume and overall volume of manufacturing equipment. The data plotted below represents the volumes for typical commercially available additive manufacturing equipment, particularly for FDM and SLS. With an initial y-intercept offset, we see that the ratio of build volume to overall volume is typically about 1.9% (or 1/50.362). The trendline can be used to predict the expected overall volume of manufacturing equipment to provide a desired build volume. However, it should be noted that for in-space manufacturing applications, where volume is at a much higher premium than for terrestrial manufacturing equipment, it is likely possible to improve the build volume to overall volume ratio through dedicated design and technology development efforts.
==Publications and Ongoing Research==


[[Image:Graphic for Equation 2.png|center|600px]]
These recent publications outline work being done at the forefront of LPBF-M technology.


The achieved build rate for fused deposition modeling (FDM), as well as for the deposition phase in Bound Metal Deposition (BMD), can be computed analytically. The build rate is computed as the most limiting case of the following factors: heat transfer to/from the feedstock, maximum extrusion force, and traverse speed of manufacturing head and nozzle diameter. For these equations, consider the schematic shown below representing the process.


[[Image:Graphic for Technical Model.png|1000px|center]]
[[File:LR3.jpg|600px]]


If we assume that the heating element is maintained at a temperature, <math>T_s</math>, that is equal to the glass transition temperature of the polymer being extruded, then the achieved rate of material extrusion, <math>\dot{V}_{thermal}</math>, can be computed as shown below, where <math>L</math> is the length of the heating element and <math>\alpha</math> is the thermal diffusivity of the material:
:<math>\dot{V}_{thermal} = 2 \pi L (k / \rho c) = 2 \pi L \alpha</math> (Eqn 4-1)


To ensure that the material is deposited smoothly and continuously, the deposition rate should also satisfy the relation shown below, where <math>d</math> is the nozzle exit diameter and <math>v</math> is the maximum traverse velocity of the manufacturing head:
==Review of a Published Roadmap==
:<math>\dot{V}_{traverse} = 0.25 \pi d^2 v</math> (Eqn 4-2)
Because there are no other publicly available roadmaps for LPBF-M, this review focuses on a straightforward project record or "roadmap" for additive manufacturing of HAYNES® 282® superalloy by laser beam powder bed fusion (PBF-LB) technology.


Because the lesser of these build rate values will be the limiting factor, we know that heat transfer to the filament will be the limiter of build rate when <math>\dot{V}_{thermal} < \dot{V}_{traverse}</math>, and vice versa. It is also very convenient to compute the critical nozzle exit diameter, <math>d</math>, which also defines the print resolution, at which the build rate due to heat transfer is exactly equal to that due to traverse speed. At this resolution, both build rates are the active limiting constraint and give the same result. We thus find that <math>\dot{V}_{thermal} = \dot{V}_{traverse}</math> when <math>d_{crit} = \sqrt{8 \alpha L / v}</math>
Source: ScienceDirect Materials and Design, Volume 204, 2021 Authors: Otto, et al


So, let us assume nominal values for the key parameters in the calculation of this build rate FOM:
Subject: As noted, there is an absence of publicly-available roadmaps for LPBF-M. They certainly exist within the corporate knowledge banks of the machine OEMs and heavy users of the technology (i.e. Airbus, General Electric). In the absence of an exact math, a roadmap for the development of a difficult-to print nickel alloy (Haynes 282) has been selected. This superalloy has desirable mechanical and thermal properties but also has challenges that must be systematically overcome before deploying printed parts into the field. Challenges include surface crack, bulk cracks, porosity, and crystalline microstructure.


[[Image:Graphic for Equation 4 Table.png|center|600px]]
CAD model of fusion zone used for [[File:CAD model of fusion zone used for FEA.png|600px]]  


We can now plot the build rate vs. resolution for the parameter values shown in the table above. We see that the particular feedstock material affects the critical nozzle exit diameter (i.e. resolution) at which the active constraint on build rate switches from the traverse speed of the print head to be now limited by the heat transfer from the nozzle. The thermal diffusivity, , of the feedstock material also affects the maximum build rate achievable. However, when traverse speed is the limiting factor, the build rate is independent of material and only depends instead on the desired resolution.
Key Inclusions:


[[Image:Graphic for Equation 5.png|center|600px]]
Background on: general LPBF-M, past efforts via other printing technologies (EBM, DED), and the nickel alloy material
Detailed descriptions of the physics-based parameters that affect build quality (hatch pattern, scan speed, chamber temp., etc.)
Full list of the inspection equipment used to evaluate the printed specimen quality
Standards used within the development process (i.e. ISO/ASTM XXXX)
Detailed experimental setup
Figures of merit specifically mentioned (volumetric energy density (J/cc)
Digital modeling to predict material behavior
Key publications used to generate the roadmap
Graphical comparison of alternate possible solutions
Overall roadmap structure. This document is less of a roadmap and more of an experimental report. While the methods used can be applied to other similar materials as a means of material parameter development, it is not sufficiently generalized to be considered useful in a broader sense. * Perhaps it could be considered a roadmap for the path traveled.
OPM, DSM, other forms of system interaction
Diagrams or photos for experimental setup. There are detailed text descriptions, but an image would convey the information without ambiguity.


In general, the equations above apply not only for FDM and BMD process, but more generally for any process that carries out serial deposition of a feedstock material made molten by thermal conduction from a heater element in the manufacturing head. Using the morphological matrix show later in this answer to describe different manufacturing processes, we know that each process that shares the same pairing of energy source and material phase will have the same fundamental equations describing the manufacturing process, but simply with different parameters. However, different combinations of energy source and material phase will have different fundamental equations. For example, using electrical energy to deposit molten material describes both wire-fed directed energy deposition processes such as laser DED and electron beam freeform fabrication (EBF3). The description of these processes will differ in the input parameters, but their model will be similar, albeit different from the model presented for FDM.


''Sensitivity Analysis''


We can compute the sensitivity of build rate to the four key parameters identified: thermal diffusivity (<math>\alpha</math>), heated nozzle length (L), nozzle exit diameter (d), and print head traverse speed (v). These sensitivities are given by taking partial derivatives of Eqn 4-1 and Eqn 4-2. For convenience we will evaluate the sensitivities at the critical nozzle exit diameter (dcrit) where the build rate achieved due to traverse speed and due to heat transfer is identical. The values reported below are for Ultem 9085, a high-temperature aerospace grade polymer. This gives:
Correlated Simulation and Benchmark [[File:Analysis.png|600px]]
Summary:


[[Image:Graphic for Equation 6.png|center|400px]]
This was a highly-effective, well-executed experimental report. It provided clear and concise explanation of any required background information; and included rigorous technical detail and analysis.
This could be used as a map or guide to developing future similar alloys, especially by the group conducting the experiment because the FE models they developed are correlated and ready to use WLOG.
A high degree of care was taken to ensure the results were valid—with deep consideration on the DoE and testing methods.
FOMs were specifically mentioned, with the focus being on laser volumetric energy density .png
Financial Model[edit]
In this section of the roadmap, "3LPB" signifies a fictional company seeking to develop an improved LPBF-M machine.


''Normalized Tornado Chart''
The financial model calculates the net present value (NPV) of the entire 3LPB project over 15 years. Since the financial information for this project does not already exist, the project direction from internal discussions is to use the financial information of Stratasys, a plausible case study within the industry, as a reference to create a parameterized financial model. Wherever possible, the inputs to the model are based on publicly available information, sometimes using as-is values, sometimes estimating the variability of parameters using approximate formulas, and in some cases making unsubstantiated estimates. 3LPB will analyze the impact on the financial model of focusing on R&D by preparing two cases: one that will serve as a baseline for the 3LPB to proceed with its financial and technological strategies as before, and the other in which the company invests more through R&D. Also, in order to perform this analysis in a more focused manner, some financial parameters are not considered.


We now normalize the above sensitivities to represent the percent change in system performance, i.e. build rate, for a one percent change in the underlying parameter. In this way,  we produce the normalized tornado chart shown below. The tornado chart indicates that changes nozzle exit diameter (d) results in twice an improvement of system performance as compared to print head traverse speed (v), heated nozzle length (L) and thermal diffusivity (<math>\alpha</math>).  
==Baseline Case==
The assumption parameters used for 3LPB baseline case model are as below.


[[Image:Tornado Chart.png|600px|center]]
Revenue


==Financial Model==
- Initial sales is set as 521 million USD which is the sales of Stratasys in 2020.


Hypothesis and Assumption
- Sales growth rate increases 17% for first 5 years, 12% for next 5 years and 8% for last 5 years by considering the slowing growth of the technology s-curve in the market.
# The major reference of the data is Figure 7 (P13) - POS (purple color in the diagram) from the paper “Feasibility Analysis of Commercial In-Space Manufacturing Applications” by Alejandro E. Trujillo, Matthew T. Moraguez, Samuel I. Wald, Andrew C. Owens, and Olivier L. de Weck, 2017.
# Since we use POS=0.9999, the revenue of $2,838,000/yr is derived from $7 M launch cost savings over 900 days.
# In the context of the project, the team assumes the company can print 30% of ECLSS spares in space by additive manufacturing.
# In the context of the project, the team assumes the company will put the upfront investment e.g. money, people training, and other built/manufacturing material in the first four years as the main investment. Then a first manufacturing unit is built and launched (250 kg at $10,000/kg).
# In the context of the project, the team captures and documents the data of 15-year including cash flow, discounted cash flow, launch cost-saving (which translates into revenue), recurring costs, and non-recurring costs.
# In the context of the project, the team assumes we have the expense of staff of three FTEs during operational years, as well as the launch of feedstock resupply.
# In the context of the project, the net present value is $1.15M assuming a discount rate of 12%.


[[Image:Figure 7.png|center|800px]]
- Sales in each year increase based on sales growth rate.


Costs
- R&D spending is set as 84 million USD which is the sales of Stratasys in 2020.


[[Image:NPV 01.png|center|700px]]
- R&D Spending is calculated as sales multiplied by R&D spending per sales. The average investment is 16.1% of sales based on Stratasys data.


- As for the baseline case, R&D spending growth rate is not considered, thus this is set as 0%.


[[Image:NPV 02.png|center|700px]]
- SG&A expenses is set as 205 million USD which is the value of Stratasys in 2020. This expense in each year is calculated by multiplying SG&A expenses by SG&A expenses rate.


- SG&A expenses rate is set as -2.1% defined based on Stratasys data.


[[Image:NPV Table.png|center|800px]]
- Operating costs is set as 591 million USD which is the value of Stratasys in 2020. This cost in each year is calculated by multiplying operating costs by operating costs rate.


==List of R&T Projects and Prototypes==
- Operating costs rate is set as -1.7% defined based on Stratasys data.-


The overall goal for technology development under this roadmap, as defined in Strategic Driver #1, is to enable in-space manufacturing of at least 30% of spare parts for long duration human exploration missions by 2035. Ensuring that the ISM equipment can fabricate the necessary spares requires that technology is developed for the manufacturing of the required materials with appropriate resolution, build volume, build rate, and system SWaP.
- A discount rate is defined as 8.76% based on Stratasys data.


According to Johnston et al., of the components that require repair and replacement on the ISS, 28.6% are plastics or composites, and 18% are metallics. The remaining spares are either electronics or ceramics/glass. Thus, it becomes immediately clear that to achieve the target of 30% of spares manufacturable by ISM, the technology developed under this roadmap will need to be able to fabricate in both metal and polymers.  
The detailed free cash flow balance sheet for the baseline case and the resulting performance financial model are shown below.


[[File:Spares Breakdown by Material.png|center|500px]]


However, while it is necessary to fabricate in these multiple materials, simply incorporating the ability to fabricate new materials is insufficient to meet desired target performance. This is because there are other active constraints, such as the build volume, build rate, and resolution that limit which of the metallic and polymer components can actually be fabricated. Additionally, the new metal manufacturing capability, which has as yet not been demonstrated in orbit, will require SWaP reduction development activities to meet the required target.
[[File:Baseline case 3LPB.png|600px]]


[[File:Additional Materials Project.png|center|500px]]


Realizing and demonstrating the ability to fabricate with a build volume large enough to accomodate over 30% of the spares is a second key technology development effort. Based on available data from NASA’s Baseline Values and Assumptions Document (BVAD), we see that a build volume of 8” x 8” x 8” is required in order to accomodate 50% of spares, as shown in the figure below. (Note: The 50% value is used here for margin, because some of those components which are considered manufacturable simply based on build volume, may not be manufacturable for other reasons, such as not being of the right material or having too tight of tolerances to be manufactured.)
According to the baseline financial performance model that does not enhance R&D, overall cash flow will improve because it does not take into account the investment rate for R&D expenses. However, product quality and product performance cannot be expected to improve without higher investment in R&D. As a result, the NPV is forecasted to be $2,806 million, with the present value and cumulative present value being negative until the fourth and eighth years, respectively.


[[File:Cumulative Spares by Volume.png|center|500px]]
Improved Performance Through R&D Case[edit]
The same financial model is used to analyze the effectiveness of investment in R&D by comparing NPV. The following changes to model inputs from the baseline financial case:


Because the overall size, weight, and power of our ISM system will be severely constrained, as is common in space systems engineering, a potential development project involves achieving improvements in the ratio of build volume to overall system volume. Terrestrial additive manufacturing equipment is not nearly as volume constrained as an in-space manufacturing unit would be. Thus, a concerted development effort would be necessary to achieve improvements in this packaging efficiency of the build volume within the unit. This could take the form of miniaturization of the ancillary manufacturing support equipment, or of re-design of the build tray and gantry system to more efficiently utilize available volume. A development project to simply reduce the mass of existing ISM capability is not enough to reach the desired target, as shown below.
Revenue
Sales growth rate is affected by the increase of R&D spending and 5% added improved sales growth rate is used.
Costs
R&D Spending per sales in each year increases 5% as R&D spending growth rate from the previous year.


[[File:Mass Reduction Project.png|center|500px]]
The detailed free cash flow balance sheet for the R&D-enhanced case and the resulting improved performance financial model are shown below.


Financial Model Results in Baseline Case


[[File:R&D enhanced case 3LPB.png|600px]]


Build rate is also identified as a key contributor to determining which spares will be deemed manufacturable (see Moraguez & de Weck 2019). The ISM system must be able to fabricate components quickly enough to produce the spare before the effects of the non-operational system lead to hazardous conditions for the spacecraft or crew. It has previously been shown in this roadmap that build rate is directly tied to resolution. However, while build rate can be traded for resolution, either through process modifications or a change of manufacturing process, it is not desireable to sacrifice resolution for build rate because that can reduce the number of spares that can be fabricated within the required tolerance. Recall that in-space manufacturing equipment typically has reduced build rate and resolution over its terrestrial counterparts due to the limitations on size, weight, and power. Thus, technology development efforts for improvement of build rate to approach terrestrial values without sacrificing resolution are desired.
According to the R&D-enhanced financial performance model, an increase in the investment rate for R&D expenses will worsen the overall cash flow. However, the sales growth rate is expected to increase due to the improvement in product quality and product performance resulting from the higher investment in R&D. The annual increase in sales is calculated by adding 5% to the sales growth rate. In this way, the increase in sales will exceed the decrease in cash flow, which has deteriorated, and the rate of increase in present value will improve, resulting in a faster positive cumulative present value and an improved NPV. As a result, NPV is expected to increase to $5,276 million, almost double the $2,806 million in the baseline case. The increase in NPV due to higher investment in R&D as a percentage of sales confirms the company's financial value.


The figure below presents the performance frontier for ISM technology in 2035 both with and without the proposed development projects. Currently, ISM systems are limiting to additive manufacturing of thermoplastics (i.e. polymer FDM). While these systems have demonstrated the ability to fabricate components on the ISS, they are extremely limited in application for fabrication of actual spare parts. Business as usual will carry these systems to the point where they can begin to be adopted as either a supplemental or baseline approach for spares for some low-risk components. However, without a concerted technology development effort these systems will not develop beyond fabrication of polymers. This material constraint places an upper limit on the percent of spares that can be fabricated regardless of the mass allocation for the ISM system. Thus, one key technology development project is to extend ISM into fabrication of metals, such as aluminum 6061 T6 and titanium 6Al-4V. Based on the build rate and resolution targets set forth in this roadmap, bound metal deposition is likely to be the process of choice for this development effort. Significant effort will be placed in improved prediction of part shrinkage to enable the required tolerances to be achieved with this process. By demonstrating metals manufacturing through this development, the upper limit on the percent of spares that can be fabricated by ISM will be raised to about 50%, assuming sufficient mass is allocated to the ISM system. As this fraction of manufacturable spares increases, it becomes increasingly important to have a high enough build rate that the ISM system can keep up with component failures. Thus, technology development focused on improved build rate without sacrificing resolution is proposed. This involves improved heat transfer to the filament and/or additional energy sources (i.e. directed laser energy or heated build volume). The final key technology development focus is to reduce the overall ISM system volume and mass. The improved volume efficiency enables the required build volume for the increased spares fraction to be achieved within volume constraints. The reduction in mass leads to a leftward shift in the performance frontier, where a given percentage of spares can be fabricated with less ISM mass allocation. Thus, there is a reduction in the minimum mass allocation for a system that can at least fabricate a single spare. Nevertheless, there is still an upper limit on the maximum mass allocation for an ISM system, which is equal to the mass of carry-along spares that the ISM system seeks to replace. We also note that the performance frontier is initially characterized by a steep slope as the easiest to manufacture spares are able to be fabricated by minimally capable ISM systems. As additional capability is added, in terms of mass allocation for the ISM system, the fraction of manufacturable spares continues to increase, albeit more slowly due to those spares being more challenging to manufacture. This challenge of diminishing returns are largely what drove the target performance to be set as it is.
==List of R&D Projects==
===R & D Projects: Advances in Laser Fusion===


[[File:ISM Performance Frontier.png|center|700px]]
*Fraunhofer Institute of Laser Technology (ILT) Ultrashort Pulse Processing (2019)
**Laser tech allowing 64-way beam splitting
**Each beam capable of being modulated independently
**Holds promise for faster production in future machine architecture


'''References:'''


Johnston, M. M., et al., “3D Printing in Zero-G ISS Technology Demonstration,” AIAA Space, San Diego, CA, Aug. 2014.
[[File:UPP.jpg|600px]]
Hanford, A.J., ed., “Advanced Life Support Baseline Values and Assumptions Document,” Houston, TX, NASA/CR-2004-208941, Aug. 2004.
Moraguez, M., and de Weck, O., “In-Space Manufacturing Production Rate and Reliability Targets for On-Demand Fabrication of ECLSS Spares,” 49th International Conference on Environmental Systems, Boston, MA, Jul. 2019.


==Key Publications, Presentations and Patents==
(source: Fraunhofer Institute ILT)
 
*Fraunhofer ILT Increasing Quality Through Adaptive Process Control (2020)
**Pulsed wave laser applied to part perimeter enables a smooth surface and increased accuracy
**There is significant technology improvement that can be realized by increasing execution quality of existing hardware (AJ Hart 2021)
 
 
[[File:APC.jpg|600px]]
 
==R & D Projects: Advances in Materials Processing==
*Equispheres Highly-Spherical Aluminum Powder for Fast and Low-Porosity Production (2021)
**Allows use of higher-power lasers at standard scan speeds without the introduction of porosity
**Improved powder flowability enables reduced keyhole pore formation by minimizing the drag and effect of acoustic waves as described in Zhao's 2020 paper
**The net result is parts with the same mechanical properties produced in up to 60% less time
 
[[File:EAP2.png|600px]]
 
*AL3D-Metal 200 Printer with Fully-Contained and Swappable Multi-Metal Cartridges(2021)
**Allows for multi-material consolidated components
**Enables rapid material changeover
**Increases operator and facility safety through reduced potential for airborne particulate contamination
 
[[File:AL3D.jpg|800px]]
 
(source:AL3D)
 
 
==R & D Projects: Advances in Automation==
 
*EOS and Grenzebach Expanded Factory Automation Development Project (2020)
**Nearly full-factory automation
**All bill of process stages managed including setting-up, loading, unloading, unpacking, cleaning, and inventorying
**Automation will drive greater process control, a prime area of opportunity in AM


We will focus it on Additive Manufacture in the space as the main direction to search and do our paper and patent analysis
[[File:EGA3.png|800px]]


[[Image:Key Papers.png|center | 800px]]
(source: EOS)
*Additive Industries MetalFABG2 Self-Contained Production Line (2021)
**One unit with automated prep, fuse, clean, and heat treat modules that enable continuous production
**This second-generation machine should improve productivity and integration
**Aligning with industry trends and demands for diagnostics and process control, this machine contains increased in-situ meltpool monitoring


[[Image:Key Patents.png|center| 800px]]


[[Image:Key Patents 2.png|center | 600px]]
[[File:MFG2.jpg|600px]]


(source: Additive Industries)
==Technology Strategy Statement==
==Technology Strategy Statement==


'''Our target is to develop an in-space manufacturing facility that can fabricate 30% of spare parts (in the required geometries and with sufficient responsiveness) in both high-temperature thermoplastics and metals with resolution of 0.15 mm and build rates of at least 25 cm^3/hour. Additionally, we want to achieve the envisioned NASA FabLab allocations of 0.45 m^3, 260 kg, and 2 kW by the target year 2035. To enable this we will invest in two parallel R&D projects with immediate effect. The first project would be a joint project with Airbus to achieve the 0.1 mm resolution by Laser Metal Deposition achieved by Airbus Metal3D by the year 2025 for technology demonstration on the ISS. This demonstration would prove our capabilities with the fabrication of actual representative spares to pave the way for becoming a baseline sparing approach for future commercial space station customers in 2035. The second project would be to maximize build rates without compromising on resolution, such as by using a longer heating element, using multiple print heads, using improved materials or external sources of heating. By 2025, if we see that the expected build rate cannot be achieved without compromising on resolution, then we will trade-off resolution for increasing the build rate so that minimal machining can help us achieve the same resolution by 2030. A combination of improvements in both projects will help us reach our technical and business targets by 2035. In case time permits, we will invest in an additional project to minimize the volume and mass of the printer without compromising on the range of possible parts which can be generated.'''
The strategy laid out in this roadmap will focus on improving the overall production costs associated with LBPF-M by reducing the build time and post printing processes. This will be accomplished by reducing the surface roughness to an Ra of 6 µm and decreasing the build time to 3 hours per 1,000 cm³ part.
 
*Current and future work to reduce the build time will include some or all of the following:
**Increased laser unit count
**Increased beam splits
**Beam shapes and intensity distributions (for example linear distribution)
**In-situ monitoring and controls to optimize power, scan speed, and surface roughness
 
*Additional ongoing work that will increase overall manufacturing time includes:
**Automation within additive machines such as the MetalFABG2 printer presented earlier
**Automation between machines such as the work being done by EOS
**Elimination of supports to reduce post-processing steps
 
 
[[File:Swoosh_chart_3LPB.png|600px]]
 
Figure: Swoosh chart for future technology development.


[[Image:Screen Shot 2019-12-03 at 1.18.53 AM.png|center|1100px]]
Retrieved from "http://roadmaps.mit.edu/index.php?title=Laser_Powder_Bed_Fusion_-_Metal&oldid=113982"

Latest revision as of 18:11, 11 September 2022

Technology Roadmap Sections and Deliverables

The clear and unique identifier for this technology roadmap is:

  • 3LPB - Laser Powder Bed Fusion - Metal

This indicates that we are dealing with a “level 3” roadmap at the specific implementation level, where “level 1”and "level 2" would have indicated the over-arching roadmap and “level 4” would have indicated an individual technology roadmap.

In select sections of the roadmap, "3LPB" signifies a fictional company seeking to develop an improved LPBF-M machine.


Roadmap Overview

The working principle and architecture of Laser Powder Bed Fusion - Metal is depicted in the image below.

M400.png

Figure: EOS M400-4 LPBF-M machine (source: EOS)

PD2.png

Figure: LPBF-M Process (source: CustomPartNet)

Laser Powder Bed Fusion - Metal (LPBF-M) is an additive manufacturing (AM) technology that enables the creation complex metal components directly from a digital 3D model without any tools. While the number of available materials is still limited compared to other milling and injection molding processes, LPBF-M utilizes various metals such as cobalt chrome, nickel alloys, titanium, aluminum, stainless steel and copper to generate robust functional prototypes and production parts that are capable of surviving in severe use cases. The LPBF-M technology offers comparable quality to parts made with traditional manufacturing methods, with material properties surpassing those of cast parts and approaching those of billet. It is most widely used in aerospace, medical, motorsports, energy, and molding. Beyond highly-stressed functional components, LPBF-M can be used for producing parts in cosmetic applications, manufacturing aids, small integrated structures, dental components, surgical implants.

The process shares similarities with many other layer-wise additive manufacturing technologies. A program utilizes 3D model data and mathematically slices it into 2D cross-sections. Each section will act as a template that lets the LPBF-M machine know where to precisely create perimeters and cross-hatching. The data is transferred to the LPBF-M equipment. Subsequently, a "recoater" spreads a 25-120µm thick layer of powdered metal to produce a uniform layer over the solid metal build plate. A laser then draws a 2D cross-section on the powdered material, fusing the substrate into a solid. Once a layer is complete, the base plate is lowered enough to make room for the next layer. More material is collected from the reservoir and recoated evenly on the previously sintered layer. The LPBF-M machine continues to create layer upon layer building from the bottom up as the part is built. Support structures are added to part to provide a reinforcement for overhangs, and to prevent thermally expansion from negatively impacting dimensional accuracy. Following printing, the unfused powder is vacuumed from the chamber and the build plate is removed from the machine. Thereafter, the components typically undergo a stress relief and/or heat treatment process to remove residual thermal stresses and improve material properties. Following heat treatment, the next steps are support removal, detachment from the build plate (via bandsaw or wire EDM), and media blasting for final powder removal.


INDST.png

Figure: Industry breakdown for installed metal powder bed fusion machines. This figure includes non-laser powder bed machines but 85-90% of these unit sare LPBF-M. (source: Dr. Maximilian Munsch, AMPOWER 2021 Additive Manufacturing Report)

There is also an environmental advantage in the raw material efficiency of the processes. While the scrap rate for many complex milled parts is over 90%, the scrap rate for LPBF-M parts is typically less than 5%. While milled chips can be recycled, they must be processed offsite using considerable energy. In many cases, the LPBF-M powder can be immediately returned to a machine reservoir following sieving.

A350.png

Figure: A350 XWB cabin brackets (source: Airbus)

The primary deficits of the technology in its current form are slow production rates, expensive machines, and high powder costs due to low volume requirements. The combination of these three challenges results in a high part piece cost. However, with improved performance, elimination of tooling, and reduced lead-time, there are many positive business cases for LPBF-M. In the future, the introduction of high-speed systems with more powerful lasers and larger build chambers is expected to increase the number of viable economically-viable applications that will drive increased demand and scale for LPBF-M systems.

COST2.png

Figure: Cost comparison for conventional manufacturing (CM), additive components, and AM-enhanced tooling. The freedom provided by AM methods like LPBF-M provides entry to design spaces that would otherwise be impractical or impossible to access. (source: Modified from John Hart and Haden Quinlan, MIT xPro "Additive Manufacturing for Innovative Design and Production course," MIT)


The LPBF-M market is maturing with breakthrough applications across multiple industries that provide the justification and institutional know-how to continue growing the technology.

TRND.png

Figure: Metal powder bed fusion machines expected sales show a predicted 6000-7500 units by 2025. This figure includes non-laser powder bed machines but 85-90% of these sales are LPBF-M, bringing the predicted totals to 5100-6750. (source: Dr. Maximilian Munsch, AMPOWER 2021 Additive Manufacturing Report)


SALES.png

Figure: Metal powder bed fusion is responsible for 39% of sales revenue of all AM machines. While significantly more expensive than polymer systems (>2X), LPBF-M sales revenue is reflective of the technology's increasing foothold. (source: AMPOWER 2021 Additive Manufacturing Report)

Finally, LPBF-M is the leading metal AM technology from an industrialization and technology maturity standpoint, yet it still has significant room for improvement, making it the ideal technology for roadmapping.

MATR.png

Figure: LPBF-M leads industrialization and technology maturity indices. (source: AMPOWER 2021 Additive Manufacturing Report)

Design Structure Matrix (DSM) Allocation[edit]

The classification tree below shows us that LPBF-M is part of the larger Additive system. The DSM and tree both show that LPBF-M requires the following technologies at subsystem level 4: Laser (4L), Powdered Metal (4PM), Scanner (4S), Recoater (4R), and Build Chamber (4BC). Each level 3 subsystem also requires enabling technologies shown as level 4 systems.


LPBF-M DSM allocation 2.png


Roadmap Model using OPM

An Object-Process-Diagram (OPD) of LPBF-M is provided in the figure below. This diagram captures the main object of the roadmap (LPBF-M), its various instances including development projects, its decomposition into subsystems, its characterization by Figures of Merit (FOMs) as well as the primary processes.

LPBF-M OPM.png

Figure: Object-Process-Diagram for LPBF-M

An Object-Process-Language (OPL) description of the roadmap scope is auto-generated and given below. It reflects the same content as the previous figure, but in a formal natural language.


LPBF-M OPL.png

Figure: Object-Process-Language for the LPBF-M OPM

Figures of Merit

The table below shows a list of FOMs by which Laser Powder Bed Fusion – Metal, LPB can be assessed. For LPBF-M, the key FOMs for increased productivity are build rate (also called printing speed), build chamber volume, and layer thickness. Additional FOMs such as laser power, laser count, and laser scan speed play a critical role in productivity as well. FOMs on this list can be categorized into four types, productivity, performance, efficiency, and competitiveness. Most of these, including key FOMs, are explicitly related to productivity. Thus, productivity as a FOM would be a critical indicator to evaluate technological progress.


FOMCHARTV2.jpg

Figure: Primary figures of merit for the LPBF-M technology


BRV5.jpg

Figure: Build rate change over time. Printing speed has grown step-wise over the brief timeline of LPBF-M. Today's fastest commercial machines contain up to 12 lasers, have automation provisions for rapid changeovers, and are modularized for quick repairs. In addition to machine type, build rate varies by material, layer height, and machine parameter settings so the plotted data points are representative values. (Source: Machine manufacture datasheets)


NXG.png


CLML.png

Figure: GE Additive's Concept Laser M Line system is highly modularized and scalable. (Source: GE Additive)

Alignment with Company Strategic Drivers[edit] In select sections of the roadmap, "3LPB" signifies a fictional company seeking to develop an improved LPBF-M machine.

The table below shows an example of strategic drivers and their alignment with 3LPB's Laser Powder Bed Fusion - Metal roadmap.

3LPB's roadmap is focused on meeting a wide variety of 3D printer needs, such as rapid prototyping and serial production applications, different surface qualities, and cost reduction by decreasing production time. By balancing the trade-off between surface roughness and time-to-build, and reducing production time and material costs on the part of the user, 3LPB hopes to meet the needs of a large number of users while achieving significant cost efficiencies and productivity improvements. As a result, 3LBP will be able to increase its penetration rate and significantly impact the productivity of each market on a national scale, contributing to stable and sustainable industrial growth regardless of labor shortages and economic conditions.


Alignment-table.png


Positioning of Company vs. Competition

In select sections of the roadmap, "3LPB" signifies a fictional company seeking to develop an improved LPBF-M machine.

The following two figures are a table of the current competitive situation with key FOMs and a representation of our position and the situation of other companies in promoting the above strategy of 3LPB. Some key data/info is confidential, therefore it will be kept blank.

Table: Fictional company 3LPB's technical market positioning. (Source: Machine manufacture datasheets)


POSI.png


The following figure is the Pareto Front of Surface Roughness vs. Build Time for Various Additive Manufacturing Laser Powder Bed Fusion (Metal). Currently, 3LPB is right on the Pareto Front of the two FOMs; however By 2036 the company would like to achieve a surface roughness of less than 6 µm and a build time of 3 hours per 1000 cm³ part.


PF.png

Figure: Current Pareto Front and fictional company 3LPB's intended destination. (Source: Machine manufacture datasheets)

Technical Model[edit] In order to assess the feasibility of technical targets at the level of the 3LPB roadmap it is necessary to develop a technical model. The purpose of such a model is to explore the design tradespace and establish what are the active constraints in the system. The first step can be to establish a morphological matrix that shows the main technology selection alternatives that exist at the first level of decomposition, see the figure below.

Table: Technical Model: Morphological Matrix and Tradespace. (Source: Machine manufacture datasheets)


1.7 3LPB.png


4LSR.png

Figure: EOS M400-4 four laser system. The lasers can either be isolated to one laser per part (i.e. several small parts) or combined into multiple lasers per part (i.e. a large part). Use of more than one laser per part creates a visible seam line that can result in the need for real or perceived additional component validation. (Source: EOS)


The relationship between the parameters of laser power, scan speed, and spot size, and outputs of porosity and surface roughness is viewed as a primary physics-based limiting factor (Zhao 2020, Gee 2020). This is investigated further below.


PWR-SPD.png

Figure: An exponential boundary is created for the threshold between stable melting and keyhole porosity in the Zhao study. Maximum power and speed--without increasing surface roughness beyond the acceptable specification--is the preferred operating window. (source: Zhao 2020)


GEE.png

Figure: Tradeoffs (or operating windows) for build rate as a function of laser power, spot size, scan speed. The color gradient is build rate, the black lines indicate scan speed, and the white lines can be ignored for this discussion. (source: Gee 2020)


Key Publications, Presentations and Patents

Fundamental Patents[edit]

US Patent 4863538

Date: 1989

Assignee: University of Texas

Method and apparatus for producing parts by selective sintering

The original utility patent for laser sintering powder bed polymer which expired in 2007 kicking off the revolution of 3D printing technology advancement.

3DS.png

US Patent 5460758

Date: 1995

Assignee: EOS / 3D Systems

Method and apparatus for production of a three-dimensional object

The first metal laser sintering patent enabled by the work or EOS and the Fraunhofer Institute. As laser power increased, this would transform from sintering to fusing (welding). Expired in 2013, creating increased competition.


EOSP.png

State of the Art Patents

US Patent 10919090

Date: 2021

Assignee: Vulcanforms Inc.

Additive manufacturing by spatially controlled material fusion

This patent is for a LPBF-M machine with a plurality of modulated line-shaped laser systems that provide the potential to rapidly speed production and improve fusion.

[File:LINE4.png|600px]]

US Patent 10195693

Date: 2019

Assignee: Velo3D Inc.

Apparatuses, systems and methods for three-dimensional printing

This patent describes low-impact recoating mechanisms like those used in the Velo3D Sapphire machine that prints with no or minimal supports, thereby reducing post-processing time and increasing productivity.

VELO.png

US Patent 10974456

Date: 2021

Assignee: Lawrence Livermore National Security LLC.

Additive manufacturing power map to mitigate defects

This patent includes methods for mapping laser power to select geometry for improved print quality via defect reduction. When implemented in-situ, this type of control mechanism offers significant potential in quality and productivity improvement.

File:LINL.png

US Patent 10583529

Date: 2020

Assignee: EOS Of North America Inc

Additive manufacturing method using a plurality of synchronized laser beams

This patent provides methods of using multiple laser beams and scanners in unison. Packaging and coordinating the optical systems is one of the greatest challenges of multi-laser systems. Intelligent packaging and control systems like that included in this patent will increase both quality and productivity.

EONA.png


Publications and Ongoing Research

These recent publications outline work being done at the forefront of LPBF-M technology.


LR3.jpg


Review of a Published Roadmap

Because there are no other publicly available roadmaps for LPBF-M, this review focuses on a straightforward project record or "roadmap" for additive manufacturing of HAYNES® 282® superalloy by laser beam powder bed fusion (PBF-LB) technology.

Source: ScienceDirect Materials and Design, Volume 204, 2021 Authors: Otto, et al

Subject: As noted, there is an absence of publicly-available roadmaps for LPBF-M. They certainly exist within the corporate knowledge banks of the machine OEMs and heavy users of the technology (i.e. Airbus, General Electric). In the absence of an exact math, a roadmap for the development of a difficult-to print nickel alloy (Haynes 282) has been selected. This superalloy has desirable mechanical and thermal properties but also has challenges that must be systematically overcome before deploying printed parts into the field. Challenges include surface crack, bulk cracks, porosity, and crystalline microstructure.

CAD model of fusion zone used for CAD model of fusion zone used for FEA.png

Key Inclusions:

Background on: general LPBF-M, past efforts via other printing technologies (EBM, DED), and the nickel alloy material Detailed descriptions of the physics-based parameters that affect build quality (hatch pattern, scan speed, chamber temp., etc.) Full list of the inspection equipment used to evaluate the printed specimen quality Standards used within the development process (i.e. ISO/ASTM XXXX) Detailed experimental setup Figures of merit specifically mentioned (volumetric energy density (J/cc) Digital modeling to predict material behavior Key publications used to generate the roadmap Graphical comparison of alternate possible solutions Overall roadmap structure. This document is less of a roadmap and more of an experimental report. While the methods used can be applied to other similar materials as a means of material parameter development, it is not sufficiently generalized to be considered useful in a broader sense. * Perhaps it could be considered a roadmap for the path traveled. OPM, DSM, other forms of system interaction Diagrams or photos for experimental setup. There are detailed text descriptions, but an image would convey the information without ambiguity.


Correlated Simulation and Benchmark File:Analysis.png Summary:

This was a highly-effective, well-executed experimental report. It provided clear and concise explanation of any required background information; and included rigorous technical detail and analysis. This could be used as a map or guide to developing future similar alloys, especially by the group conducting the experiment because the FE models they developed are correlated and ready to use WLOG. A high degree of care was taken to ensure the results were valid—with deep consideration on the DoE and testing methods. FOMs were specifically mentioned, with the focus being on laser volumetric energy density .png Financial Model[edit] In this section of the roadmap, "3LPB" signifies a fictional company seeking to develop an improved LPBF-M machine.

The financial model calculates the net present value (NPV) of the entire 3LPB project over 15 years. Since the financial information for this project does not already exist, the project direction from internal discussions is to use the financial information of Stratasys, a plausible case study within the industry, as a reference to create a parameterized financial model. Wherever possible, the inputs to the model are based on publicly available information, sometimes using as-is values, sometimes estimating the variability of parameters using approximate formulas, and in some cases making unsubstantiated estimates. 3LPB will analyze the impact on the financial model of focusing on R&D by preparing two cases: one that will serve as a baseline for the 3LPB to proceed with its financial and technological strategies as before, and the other in which the company invests more through R&D. Also, in order to perform this analysis in a more focused manner, some financial parameters are not considered.

Baseline Case

The assumption parameters used for 3LPB baseline case model are as below.

Revenue

- Initial sales is set as 521 million USD which is the sales of Stratasys in 2020.

- Sales growth rate increases 17% for first 5 years, 12% for next 5 years and 8% for last 5 years by considering the slowing growth of the technology s-curve in the market.

- Sales in each year increase based on sales growth rate.

Costs - R&D spending is set as 84 million USD which is the sales of Stratasys in 2020.

- R&D Spending is calculated as sales multiplied by R&D spending per sales. The average investment is 16.1% of sales based on Stratasys data.

- As for the baseline case, R&D spending growth rate is not considered, thus this is set as 0%.

- SG&A expenses is set as 205 million USD which is the value of Stratasys in 2020. This expense in each year is calculated by multiplying SG&A expenses by SG&A expenses rate.

- SG&A expenses rate is set as -2.1% defined based on Stratasys data.

- Operating costs is set as 591 million USD which is the value of Stratasys in 2020. This cost in each year is calculated by multiplying operating costs by operating costs rate.

- Operating costs rate is set as -1.7% defined based on Stratasys data.-

- A discount rate is defined as 8.76% based on Stratasys data.

The detailed free cash flow balance sheet for the baseline case and the resulting performance financial model are shown below.


Baseline case 3LPB.png


According to the baseline financial performance model that does not enhance R&D, overall cash flow will improve because it does not take into account the investment rate for R&D expenses. However, product quality and product performance cannot be expected to improve without higher investment in R&D. As a result, the NPV is forecasted to be $2,806 million, with the present value and cumulative present value being negative until the fourth and eighth years, respectively.

Improved Performance Through R&D Case[edit] The same financial model is used to analyze the effectiveness of investment in R&D by comparing NPV. The following changes to model inputs from the baseline financial case:

Revenue Sales growth rate is affected by the increase of R&D spending and 5% added improved sales growth rate is used. Costs R&D Spending per sales in each year increases 5% as R&D spending growth rate from the previous year.

The detailed free cash flow balance sheet for the R&D-enhanced case and the resulting improved performance financial model are shown below.

Financial Model Results in Baseline Case

R&D enhanced case 3LPB.png

According to the R&D-enhanced financial performance model, an increase in the investment rate for R&D expenses will worsen the overall cash flow. However, the sales growth rate is expected to increase due to the improvement in product quality and product performance resulting from the higher investment in R&D. The annual increase in sales is calculated by adding 5% to the sales growth rate. In this way, the increase in sales will exceed the decrease in cash flow, which has deteriorated, and the rate of increase in present value will improve, resulting in a faster positive cumulative present value and an improved NPV. As a result, NPV is expected to increase to $5,276 million, almost double the $2,806 million in the baseline case. The increase in NPV due to higher investment in R&D as a percentage of sales confirms the company's financial value.

List of R&D Projects

R & D Projects: Advances in Laser Fusion

  • Fraunhofer Institute of Laser Technology (ILT) Ultrashort Pulse Processing (2019)
    • Laser tech allowing 64-way beam splitting
    • Each beam capable of being modulated independently
    • Holds promise for faster production in future machine architecture


UPP.jpg

(source: Fraunhofer Institute ILT)

  • Fraunhofer ILT Increasing Quality Through Adaptive Process Control (2020)
    • Pulsed wave laser applied to part perimeter enables a smooth surface and increased accuracy
    • There is significant technology improvement that can be realized by increasing execution quality of existing hardware (AJ Hart 2021)


APC.jpg

R & D Projects: Advances in Materials Processing

  • Equispheres Highly-Spherical Aluminum Powder for Fast and Low-Porosity Production (2021)
    • Allows use of higher-power lasers at standard scan speeds without the introduction of porosity
    • Improved powder flowability enables reduced keyhole pore formation by minimizing the drag and effect of acoustic waves as described in Zhao's 2020 paper
    • The net result is parts with the same mechanical properties produced in up to 60% less time

EAP2.png

  • AL3D-Metal 200 Printer with Fully-Contained and Swappable Multi-Metal Cartridges(2021)
    • Allows for multi-material consolidated components
    • Enables rapid material changeover
    • Increases operator and facility safety through reduced potential for airborne particulate contamination

AL3D.jpg

(source:AL3D)


R & D Projects: Advances in Automation

  • EOS and Grenzebach Expanded Factory Automation Development Project (2020)
    • Nearly full-factory automation
    • All bill of process stages managed including setting-up, loading, unloading, unpacking, cleaning, and inventorying
    • Automation will drive greater process control, a prime area of opportunity in AM

EGA3.png

(source: EOS)

  • Additive Industries MetalFABG2 Self-Contained Production Line (2021)
    • One unit with automated prep, fuse, clean, and heat treat modules that enable continuous production
    • This second-generation machine should improve productivity and integration
    • Aligning with industry trends and demands for diagnostics and process control, this machine contains increased in-situ meltpool monitoring


MFG2.jpg

(source: Additive Industries)

Technology Strategy Statement

The strategy laid out in this roadmap will focus on improving the overall production costs associated with LBPF-M by reducing the build time and post printing processes. This will be accomplished by reducing the surface roughness to an Ra of 6 µm and decreasing the build time to 3 hours per 1,000 cm³ part.

  • Current and future work to reduce the build time will include some or all of the following:
    • Increased laser unit count
    • Increased beam splits
    • Beam shapes and intensity distributions (for example linear distribution)
    • In-situ monitoring and controls to optimize power, scan speed, and surface roughness
  • Additional ongoing work that will increase overall manufacturing time includes:
    • Automation within additive machines such as the MetalFABG2 printer presented earlier
    • Automation between machines such as the work being done by EOS
    • Elimination of supports to reduce post-processing steps


Swoosh chart 3LPB.png

Figure: Swoosh chart for future technology development.

Retrieved from "http://roadmaps.mit.edu/index.php?title=Laser_Powder_Bed_Fusion_-_Metal&oldid=113982"

Retrieved from "https://roadmaps.mit.edu/index.php?title=Laser_Powder_Bed_Fusion_-_Metal&oldid=111718"