京都大学 / 0048 / 新制・課程博士 / 博士(理学) / 甲第19479号 / 理博第4139号 / 新制||理||1595(附属図書館) / 32515 / 京都大学大学院理学研究科物理学・宇宙物理学専攻 / (主査)准教授 戸塚 圭介, 教授 川上 則雄, 教授 石田 憲二 / 学位規則第4条第1項該当 / Doctor of Science / Kyoto University / DFAM

Detta examensarbete är på grundnivå som har genomförts under en period av 20 veckor undervåren 2023 och motsvarar 15 högskolepoäng. Projektet har haft som syfte att utveckla ett konceptför en kontaktor och tillämpa additiv tillverkningsteknik som en tillverkningsmetod för konceptet.Uppdragsgivaren för projektet har varit ABB Control Products i Västerås och de har identifieratbehovet av en lämplig lösning för limproblemet i släckpaketet för kontaktorer av storlek nio. Dennuvarande användningen av lim i kontaktorerna ansågs vara ohälsosam och uppfyllde intemiljökraven. Inom ramen för projektet genomfördes en konceptutveckling för att lösa det identifieradeproblemet. Dessutom undersöktes möjligheten att tillämpa additiv tillverkningsteknik ochgenomföra en förenklad kostnadsbedömning för att visa skillnaden i kostnad mellan den nuvarandetillverkningsmetoden och additiv tillverkningsteknik. Produktutvecklingsmetoder användes för attgenomföra projektet, vilket resulterade i ett fungerande teoretiskt koncept. För att sålla ut idéeroch koncept användes en prioriteringsmatris och ett poängsystem. Det resulterande konceptetrepresenteras av CAD-modeller som består av metallplåtar och två delar av det utveckladekonceptet. Konceptet möjliggör en ny design utan användning av lim vid montering. Det har ocksåvisat sig att det är möjligt att ändra tillverkningsmetoden genom att använda additivatillverkningstekniker. Även om prototyperna ännu inte har testats, anses de teoretiska lösningarnavara fungerande.

Les procédés de fabrication additive sont aujourd'hui de plus en plus utilisés dans l'industrie. Parmi les différentes technologies existantes, les procédés additifs métalliques, et notamment les procédés en couches, sont les plus prometteurs pour la conception de produits mécaniques. Des travaux ont été menés sur la thématique de la conception de produits réalisés par ces moyens, il traitent principalement du choix du procédé le plus adapté, de l'optimisation de formes ou présentent des cas de reconception. Il n'existe cependant pas de démarche globale de conception de produits qui permettent de prendre en compte les spécificités des procédés additifs en couches, notamment leurs contraintes de fabrication.Lors de ce travail de thèse, les changements que ces procédés introduisent dans le domaine des possibles en conception de produits ont été montrés et illustrés par des pièces réalisées par EBM. De nouvelles opportunités s'offrent au concepteur, comme l'accès à l'ensemble du volume de fabrication, la facilité de réalisation de pièces complexes, la possibilité de réaliser des treillis tridimensionnels et la capacité de produire des mécanismes sans assemblage. Les contraintes de fabrication de ces procédés sont spécifiques. Les phénomènes thermiques lors de la fabrication ont une incidence sur la fabricabilité et la qualité des pièces. La phase de retrait de poudre impose quant à elle des contraintes d'accessibilités. Pour prendre en compte cette évolution, il est nécessaire de concevoir spécifiquement les pièces pour la fabrication additive.Le procédé EBM est au centre du travail réalisé. Il s'agit d'un moyen de fabrication additive en couches, par fusion, à l'aide faisceau d'électrons. Les phénomènes thermiques, qui peuvent causer déformations et mauvaise intégrité de la matière, l'opération de dépoudrage et la problématique de la qualité des pièces fabriquées par EBM ont fait l'objet de caractérisations expérimentales. La durée de fabrication et le coût de revient technique des pièces réalisées par EBM ont également été étudiés, afin d'établir la relation entre durée, coût et géométrie des pièces.Pour de prendre en compte les contraintes explicitées auparavant, et pour bénéficier des importantes libertés que ce procédé offre aux concepteurs, une démarche de conception a été proposée. Cette démarche consiste à générer une ou plusieurs géométries initiales, soit directement par le concepteur, soit par l'utilisation d'outils d'optimisation topologique, à partir de données extraites du cahier des charges. Une fois le balançage de la pièce choisi (en prenant en compte les contraintes de fabrication, le tolérancement de la pièce et la productivité de la fabrication), la pièce est modélisée en incluant un jeu de paramètres pour effectuer une optimisation paramétrique. Cette optimisation permet de dimensionner la pièce, tout en prenant en compte les contraintes de fabrication. A l'issue de cette phase d'optimisation, la géométrie finale est obtenue en prenant en compte les exigences des opérations de parachèvement éventuelles et en définissant les supports, s'ils sont nécessaires. Cette démarche a été illustrée par la reconception de deux pièces mécaniques qui répondent aux exigences de leur cahier des charges fonctionnel, sont fabricables à l'aide du procédé EBM et offrent des gains de masse importants.Enfin, un chapitre particulier est consacré aux perspectives mises en évidence (et ayant parfois fait l'objet de travaux préliminaires) à l'occasion de ce travail de thèse. / Nowadays, the use of Additive Manufacturing processes keeps growing in the industry. Among the numerous kinds of AM processes, metallic additive manufacturing processes, and metallic Additive Layer Manufacturing in particular, are the most interesting from a mechanical designer point of view. Several research studies have been conducted on the topic of Design For Additive Manufacturing, mostly discussing the choice of AM processes or presenting the redesign of parts. There is no specific design methodology for ALM processes that takes their specificities into account.During this PhD thesis, the changes that ALM processes bring to the design space were investigated. The designer has the opportunity to easily manufacture thin parts, complex parts, lattice structures or mechanisms that don't need any assembly. These processes also have specific manufacturing constraints compared to conventional processes. The heat dissipation is the most important factor since it can cause distortions and porosities. Powder removal, surface and geometrical quality also need to be considered during design. A specific design for additive manufacturing methodology is necessary to take these changes into account.This work focuses on the Electron Beam Manufacturing process. Experiments were conducted and analyzed to assess the manufacturability regarding the thermal phenomena (during melting), the powder removal and the quality of the parts produced by EBM. The impact of the part geometry on manufacturing duration and manufacturing cost was also established.In order to use allow designers to use these pieces of information, we suggested a designing methodology. From the requirements of the parts, one or several parts are generated by the designer or by using topological optimization tools. The orientation of the part inside the manufacturing space is set before designing a refined parametric geometry. This parametric geometry is optimized in order to meet the user requirements as well as the EBM requirements. The last step is the modification of the geometry to comply with the finishing operations (machining allowances for example) and the placement of supports, if needed. This methodology was illustrated with the redesign of two example parts and showed important mass savings from the parts (while meeting user and process requirements).The prospects discovered and highlighted during this work, some of which were preliminary investigated, are presented in a specific chapter.

Additive Manufacturing (AM), commonly known as 3D printing, represents manufacturing technology that creates objects layer by layer based on 3D model data. AM technologies have capabilities that provide engineers with new design opportunities outside the constraints of traditional subtractive manufacturing. These capabilities of AM have made it attractive for manufacturing components in the space industry., where parts are often bespoke and complex. In particular, Laser Powder Bed Fusion (LPBF) has attracted attention due to its ability to produce components with the part properties required for space applications. Additionally, the precision of the laser enables the production of innovative near-net shape and low-weight part designs.  However, due to the powdered metal material, the LPBF process is categorised with rough surfaces in the as-built state. The extent and effect of surface roughness are closely linked to geometrical design variables, including build orientation, overhangs, support structure, and build parameters; hence the more intricate the design, the more difficult the removal of this roughness. Consequently, the as-built surface for most applications is too rough and could adversely affect proprieties, i.e., fatigue. Hence, practical Design for AM (DfAM) supports should be developed that understand how design factors, such as surface roughness, will impact a part’s performance. This thesis therefore presents literature reviews on research related to LPBF surface roughness and design support, exploring the trends in managing surface roughness and investigations on the characteristics of design support. Additionally, through a space industry case study, a proposed process involving additive manufacturing design artefacts (AMDAs) is considered to investigate and describe the relationship between design, surface roughness, and performance. The review found that, in general, research focuses on the relationship between surface roughness and LPBF build parameters, material properties, or post-processing. There is very little support for design engineers to consider how surface roughness from an AM process affects the final product (less than 1% of the review articles). In investigating surface roughness, the AMDA process identified characteristics that impact roughness levels and geometric adherence to part design. Additionally, twelve characteristics of design support were identified and considered to review the AMDA process. The process aided the evaluation of design uncertainties and provided indications of part performance. However, iterations of the process can be required to clarify product-specific design uncertainties. Though, the designer obtains a better understanding of their design and the AM process with each iteration. The inclusion of the requirement to set evaluation criteria for artefacts was recommended to develop the AMDA process as design support.

Additive Manufacturing (AM) using the Powder Bed Fusion (PBF) is a relatively new manufacturing method that is capable of creating shapes that was previously practically impossible to manufacture. Many think it will revolutionize how manufacturing will be done in the future. This thesis is about some aspects of when and how to Design for Additive Manufacturing (DfAM) when using the PBF method in metal materials. Designing complex shapes is neither easy nor always needed, so when to design for AM is a question with different answers depending on industry or product. The cost versus performance is an important metric in making that selection. How to design for AM can be divided into how to improve performance and how to improve additive manufacturability where how to improve performance once depends on product, company and customer needs. Using advanced part shaping techniques like using Lattices or Topology Optimization (TO) to lower part mass may increase customer value in addition to lowering part cost due to faster part builds and less powder and energy use. Improving PBF manufacturability is then warranted for parts that reach series production, where determining an optimal build direction is key as it affects many properties of PBF parts. Complex shapes which are designed for optimal performance are usually more sensitive to defects which might reduce the expected performance of the part. Non Destructive Evaluation (NDE) might be needed to certify a part for dimensional accuracy and internal defects prior use. The licentiate thesis covers some aspects of both when to DfAM and how to DfAM of products destined for series production. It uses design by Lattices and Topology Optimization to reduce mass and looks at the effect on part cost and mass. It also shows effects on geometry translation accuracies from design to AM caused by differences in geometric definitions. Finally it shows the effect on how different NDE methods are capable of detecting defects in additively manufactured parts.

This thesis investigates the differences in the design process when developing a product for additive manufacturing (AM) compared to traditional manufacturing methods, such as CNC machining. In recent years, additive manufacturing (AM), including metal-based laser powder bed fusion (L-PBF), has gained popularity, leading to increased adoption by companies. The design process for AM, particularly in the context of metals, differs compared to for traditional manufacturing methods. L-PBF, being a method based on highly concentrated laser beam fusion, offers a higher level of design freedom, enabling the creation of intricate shapes, internal structures, and varying wall thicknesses. In contrast, traditional manufacturing methods based on subtractive processes impose limitations on design possibilities due to tooling and machining constraints. Adapting to L-PBF requires designers to reconsider, re-think and redesign parts specifically for AM, taking into account factors suchas cost, knowledge requirements and build volume limitations. The application of L-PBF extends to various industries, including aerospace and performance automobiles. Designing for L-PBF opens up new possibilities for product development by leveraging the advantages of AM, such as design flexibility and topology optimization. Topology optimization allows for the creation of lightweight components while maintaining structural integrity. However, transitioning from traditional manufacturing to L-PBF presents challenges, requiring designers to navigate the unique considerations and constraints associated with AM. This research aims to enhance the understanding of the design process for AM, with a specific focus on L-PBF, and its implications for product development. By exploring the differences between AM and traditional manufacturing methods, this study contributes to the broader adoption and effective implementation of AM technologies in various manufacturing sectors. / Detta arbete undersöker skillnaderna i designprocessen vid utveckling av produkter för additive manufacturing (AM) jämfört med traditionella tillverkningsmetoder, såsom CNC bearbetning. På senare år har additiv tillverkning (AM), inklusive Laser Powder Bed Fusion (L-PBF), blivit populärt och allt fler företag använder sig av tekniken. Designprocessen för AM, skiljer sig jämnfört med för traditionella tillverkningsmetoder. L-PBF erbjuder en hög grad av designfrihet och möjliggör avancerade former, interna strukturer och varierande väggtjocklekar. I kontrast begränsar traditionella tillverkningsmetoder, som bygger på subtraktiva processer, designmöjligheterna på grund av verktygs- och bearbetningsbegränsningar. Att anpassa sig till L-PBF kräver att designers omprövar och omdesignar delar specifikt för AM och tar hänsyn till faktorer som kostnad, kunskapskrav och begränsningar i byggvolymen. Användningen av L-PBF sträcker sig till olika branscher, inklusive luft- och rymdindustrin samt prestandabilar. Att designa för L-PBF öppnar upp nya möjligheter för produktutveckling genom att utnyttja fördelarna med AM, såsom designflexibilitet och topologioptimering. Topologioptimering möjliggör skapandet av lätta komponenter samtidigt som den strukturella integriteten bibehålls. Övergången från traditionell tillverkning till L-PBF innebär dock utmaningar och kräver att designers hanterar de unika övervägandena och begränsningarna som är förknippade med AM. Denna forskning syftar till att förbättra förståelsen för designprocessen för AM, med särskilt fokus på L-PBF, och dess implikationer för produktutveckling. Genom att utforska skillnaderna mellan AM och traditionella tillverkningsmetoder bidrar denna studie till en bredare användning och effektiv implementering av AM-teknologier inom olika tillverkningssektorer.

The use of drones is becoming increasingly common in the industry due to their efficiency and safety in environments that are difficult to access. The development of industrial drones has created a new need for specialized drones with unique functions. This has led to a growing interest in additive manufacturing as a production method. Additive manufacturing, previously primarily used for prototyping, is now emerging as a viable manufacturing method. This evolution has in turn opened new design methods intended for additive manufacturing, such as topology optimization. The purpose of this project was to redesign a drone to increase its strength, reduce its weight and improve water resistance and appearance. This was achieved using a classic product development process where a concept was developed and refined using simulation tools and a product requirement specification derived from interviews and observations of the existing drone. The product was developed using SolidWorks tools such as Topology Optimization, The Finite Element (FEM) and Computer Aided Design (CAD). The result of the work is a detailed design and a prototype developed using Topology simulations based on the product requirement specification. This project lays the foundation for continued production and development of the drone. The conclusion drawn from the work is that the new product is an improved version in several aspects compared to the previous product.

Over the past few decades, additive manufacturing (AM) evolved from rapid prototyping (RP) into mature manufacturing technologies for complex high-performance end use products. The transition in the industrial application of these technologies necessitates a design methodology tailored to the potentials and restrictions of AM. This contribution presents a new application system for opportunistic and restrictive component design, which can be integrated into design for additive manufacturing (DfAM) methodologies. For the first time, methods for AM part design are being drawn from standards and integrated into an application system, supporting professionals in the practical implementation of design for additive manufacturing methods.

Inventée en 1983, comme procédé de prototypage rapide, la fabrication additive (FA) est aujourd’hui considérée comme un procédé de fabrication quasiment au même titre que les procédés conventionnels. On trouve par exemple des pièces obtenues par FA dans des structures d’aéronef. Cette évolution de la FA est due principalement à la liberté de forme permise par le procédé. Le développement de diverses techniques sur le principe de fabrication couche par couche et l’amélioration en quantité et en qualité de la palette de matériaux pouvant ainsi être mis en forme, ont été les moteurs de cette évolution. De nombreuses autres techniques et matériaux de FA continuent de voir le jour. Dans le sillage de la FA (communément appelée impression 3D) a émergé un autre mode de fabrication : l’impression 4D (I4D). L’I4D consiste à explorer l’interaction matériaux intelligents (MIs) – FA. Les MIs sont des matériaux dont l’état change en fonction d’un stimulus ; c’est le cas par exemple des matériaux thermochromiques dont la couleur change en réponse à la chaleur ou des hydrogels qui peuvent se contracter en fonction du pH d’un milieu aqueux ou de la lumière. Les objets ainsi obtenus ont – en plus d’une forme initiale (3D) – la capacité de changer d’état (en fonction des stimuli auxquels sont sensibles les MIs dont ils sont faits) d’où la 4e dimension (temps). L’I4D fait – à juste titre – l’objet d’intenses recherches concernant l’aspect fabrication (exploration de nouveaux procédés et matériaux, caractérisation, etc.). Cependant très peu de travaux sont entrepris pour accompagner les concepteurs (qui, a priori, ne sont ni experts FA ni des experts de MIs) à l’utiliser dans leurs concepts. Cette nouvelle interaction procédé-matériau requiert en effet des modèles, des méthodologies et outils de conception adaptés. Cette thèse sur la conception pour l’impression 4D a pour but de combler ce vide méthodologique. Une méthodologie de conception pour la FA a été proposée. Cette méthodologie intègre les libertés (forme, matériaux, etc.) et les contraintes (support, résolution, etc.) spécifiques à la FA et permet aussi bien la conception de pièces que celle d’assemblages. En particulier, la liberté de forme a été prise en compte en permettant la génération d’une géométrie minimaliste basée sur les flux fonctionnels (matière, énergie, signal) de la pièce. Par ailleurs, les contributions de cette thèse ont porté sur la conception avec les matériaux intelligents. Parce que les MIs jouent plus un rôle fonctionnel que structurel, les préoccupations portant sur ces matériaux doivent être menées en amont du processus de conception. En outre, contrairement aux matériaux conventionnels (pour lesquels quelques valeurs de paramètres peuvent suffire comme information au concepteur), les MIs requièrent d’être décrits plus en détails (stimulus, réponse, fonctions, etc.). Pour ces raisons un système d’informations orientées conception sur les MIs a été mis au point. Ce système permet, entre autre, d’informer les concepteurs sur les capacités des MIs et aussi de déterminer des MIs candidats pour un concept. Le système a été matérialisé par une application web. Enfin un cadre de modélisation permettant de modéliser et de simuler rapidement un objet fait de MIs a été proposé. Ce cadre est basé sur la modélisation par voxel (pixel volumique). En plus de la simulation des MIs, le cadre théorique proposé permet également le calcul d’une distribution fonctionnelle de MIs et matériau conventionnel ; distribution qui, compte tenu d’un stimulus, permet de déformer une forme initiale vers une forme finale désirée. Un outil – basé sur Grasshopper, un plug-in du logiciel de CAO Rhinoceros® – matérialisant ce cadre méthodologique a également été développé. / Invented in 1983, as a rapid prototyping process, additive manufacturing (AM) is nowadays considered as a manufacturing process almost in the same way as conventional processes. For example, parts obtained by AM are found in aircraft structures. This AM evolution is mainly due to the shape complexity allowed by the process. The driving forces behind this evolution include: the development of various techniques on the layer-wise manufacturing principle and the improvement both in quantity and quality of the range of materials that can be processed. Many other AM techniques and materials continue to emerge. In the wake of the AM (usually referred to as 3D printing) another mode of manufacturing did emerge: 4D printing (4DP). 4DP consists of exploring the smart materials (SM) – AM interaction. SMs are materials whose state changes according to a stimulus; this is the case, for example, with thermochromic materials whose color changes in response to heat or hydrogels which can shrink as a function of an aqueous medium’s pH or of light. The objects thus obtained have – in addition to an initial form (3D) – the capacity to shift state (according to the stimuli to which the SMs of which they are made are sensitive) hence the 4th dimension (time). 4DP is – rightly – the subject of intense research concerning the manufacturing aspect (exploration of new processes and materials, characterization, etc.). However, very little work is done to support the designers (who, in principle, are neither AM experts nor experts of SMs) to use it in their concepts. This new process-material interaction requires adapted models, methodologies and design tools. This PhD on design for 4D printing aims at filling this methodological gap. A design methodology for AM (DFAM) has been proposed. This methodology integrates the freedoms (shape, materials, etc.) and the constraints (support, resolution, etc.) peculiar to the AM and allows both the design of parts and assemblies. Particularly, freedom of form has been taken into account by allowing the generation of a minimalist geometry based on the functional flows (material, energy, and signal) of the part. In addition, the contributions of this PhD focused on designing with smart materials (DwSM). Because SMs play a functional rather than a structural role, concerns about these materials need to be addressed in advance of the design process (typically in conceptual design phase). In addition, unlike conventional materials (for which a few parameter values may suffice as information to the designer), SMs need to be described in more detail (stimulus, response, functions, etc.). For these reasons a design-oriented information system on SMs has been developed. This system makes it possible, among other things, to inform designers about the capabilities of SMs and also to determine SMs candidates for a concept. The system has been materialized by a web application. Finally, a modeling framework allowing quickly modeling and simulating an object made of SMs has been proposed. This framework is based on voxel modeling (volumetric pixel). In addition to the simulation of SMs behaviors, the proposed theoretical framework also allows the computation of a functional distribution of SMs and conventional material; distribution which, given a stimulus, makes it possible to deform an initial form towards a desired final form. A tool – based on Grasshopper, a plug-in of the CAD software Rhinoceros® – materializing this methodological framework has also been developed.

The following report consists of a master thesis (30 credits) within product development. The thesis is written by Philip Ragnartz and Axel Staffanson, both studying mechanical engineering at Mälardalens University. Developing new components for a production line is costly and time consuming as they must be made from manual measurements and must go through all the conventional manufacturing (CM) steps. Eventual design mistakes will be discovered after the component have been manufactured and tested. To fix the design a completely new component must be designed and therefore double the overall lead time. The purpose of this thesis is to establish how additive manufacturing (AM) can best be used to minimize the cost and lead time in the development of new components. The study was performed by looking at the current product development process in the automotive industry at a large company, here by referred to as company A. 56 components already manufactured at company A´s own tools department was examined and compared to different AM methods. The aim of this was to get a larger pool of data to get an average on production time and cost and see how this differ to the different AM methods. Additionally, two work holders were more closely examined in a case study. Work holder one is a component in the production line that occasionally must be remanufactured. It was examined if this problem could be solved with a desktop plastic printer to hold up for a production batch. Work holder two was the development of a new component, this was to examine the use of printing the component in an early stage impact the development process. The findings from this study is that AM can today not be used in a cost efficient way in manufacturing or development of simple components. This is due to the cost of a metal 3D-printer is still very high, and the building material even higher. This results in components that gets very expensive to make compared to producing them with CM. For design evaluation to be cost efficient there will have to be a design fault in over 12 % of the newly design components for it to be cost effective to print the design for validation before sending it to be manufactured. There are however a lot bigger potential savings in the lead time. Producing the end product with a metal 3D-printer can cut down the lead time up to 85 %. This is thanks to the fact that the printer will produce the component all in one step and therefore not get stuck in between different manufacturing processes. The same goes for design evaluation with printing the component in plastic to confirm the design and not risk having to wait for the component to be manufactured twice. Despite the facts that it is not cost efficient to use AM there are other factors that play an important role. To know that the designed components will work will create a certainty and allow the development process to continue. In some cases it will also allow the designer to improve the design to function better even if the first design would have worked. As AM is expanding machines and build materials will become cheaper. Eventually it will become cheaper to 3D-print even simple components compared to CM. When this occurs, a company cannot simply buy a 3D-printer and make it profitable. There is a learning curve with AM that will take time for the designers to adapt to. Therefore, it is good to start implementing it as soon as possible as it allows for more intricate designs and require experience to do so.