Multi-layer variable setting fusion and deposition method for additive manufacturing

JP7594907B2Active Publication Date: 2024-12-05UNIFORMITY LABS INC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
JP2020519002
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Priority Date
2017-06-15
Filing Date
2018-06-14
Publication Date
2024-12-05
Estimated Expiration
2038-06-14

AI Technical Summary

Benefits of technology

【0015】 被覆(covering)問題: 被覆は、重なり合う物体の立体配置であり、d次元ユークリッド空間又はd次元湾曲空間の小領域を完全に被覆しているものである。ユークリッド空間における周期的被覆(periodic covering)とは、基底と呼ばれるγ個の物体が空間内で周期的に複製される体積vUの単位セル内に含まれることである。単位格子の形状と対称性はd個の格子基底ベクトルb(基底物体と混同しないようにする)で定まる。 物体は、凸型(円、楕円、球、楕円、円柱など)や凹型(例えばダビデの星、十字形、星形多面体、双曲線放物面)など、任意の形状をとれる。多分散被覆(polydisperse coverings)では、物体が異なる大きさ及び/又は形状を有する。二元被覆では、2つの異なる大きさ及び/又は形状を有する。三元被覆では、3つの異なる大きさ及び/又は形状を有する。各物体は、その座標rによって立体配置が定義され、その位置と方向の両方が定まる;例として、3次元では、この座標rは、物体の重心を表す3つの空間座標(x,y,z)と、物体の中心の周りの経度方向及び緯度方向の変位を表す2つの角度座標(θ,φ)とを含む。被覆の基本的な特性は被覆密度(covering density)であり、単位体積あたりの物体の総体積である。周期的被覆(periodic covering)の被覆密度は単位格子内のγ個の基底物体の全体積をvUで割ったものである。効率的な被覆では被覆密度が低い。最高の被覆では全ての被覆の中で被覆密度が最も低くなる。

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000042_0000
    Figure 00000042_0000
  • Figure 00000042_0001
    Figure 00000042_0001
  • Figure 00000042_0002
    Figure 00000042_0002
Patent Text Reader

Abstract

The present invention relates to compositions and methods for designing or manufacturing three-dimensional objects by layering, whereby specific techniques for building regions of a single layer by fusing, material deposition, or other path-based or non-path-based procedures that generate track-like shapes require the modification of print settings for the track or track-like shape or track or track-like geometry. The use of this method can result in reduced production time for the three-dimensional object and / or improved physical properties of the printed object, with these favorable printing results depending on the printing procedure, material properties, and source material properties used during printing. [Selection diagram] Figure 7
Need to check novelty before this filing date? Find Prior Art

Description

[Technology Field]

[0001] The present invention relates to compositions and methods for designing or manufacturing three-dimensional articles layer by layer, and discloses a specific method for constructing a single-layer region by fusion, material deposition, or other path-based or non-path-based procedures that generate a path-like shape, wherein the print setting value of the path or path-like shape, or the path or path-like shape, is varied within the single-layer region or within adjacent multi-layer regions. By using this method, a fusion and deposition method with variable single-layer or multi-layer setting values ​​can be obtained, thereby reducing the manufacturing time of the article and / or improving the physical properties of the article, partly because it targets the distribution of scanning paths that satisfy the overlap and / or dense filling criteria in the coating problem. Furthermore, the present invention discloses methods and compositions for designing or manufacturing three-dimensional articles, which are implemented by setting the thickness of the material layer deposited during printing to a value different from the thickness of the fragment or the net displacement of the stage or the print head for material and / or energy deposition. This method can shorten the manufacturing time of molded products and / or improve the physical properties of printed molded products. These favorable results for printing depend on the printing procedure, raw material properties, and the characteristics of the supplied materials used during printing. [Background technology]

[0002] Additive manufacturing (AM), often called 3D printing, can be described as a procedure for constructing a three-dimensional solid or porous molded object from one or more materials by fusing, joining, or otherwise adhering continuous layers of molded objects to one another. Layers in AM are often represented conceptually, on a computer, or in drawings as a combination of finite-thickness slices of the three-dimensional molded object, where each slice is most commonly defined as an intersection between the molded object and one of a set of stacked finite-thickness Euclidean planes, and a plane is defined as having a normal vector in a direction selected in the assembly of the molded object. Slices can be defined by different combinations of cross-sections other than planes, but in general, the property is maintained that the three-dimensional molded object is reproduced when all the cross-sections of the molded object, as a combination of the above, are joined in a specified order. For example, a combination of non-planar slices may be intersections between the molded object and a number of radially increasing hemispherical surfaces (cross-sections), where the increase in the radius of curvature between each consecutive cross-section is equal to the radial thickness of the previous cross-section. In this case, the direction of the build-up at each point on the fragment can be the direction from the center of the sphere of the cross-section to that point (radially outward direction).

[0003] AM (Additive Manufacturing) methods for manufacturing AM molded objects include fused deposition modeling (FDM), stereolithography (STEREOlithography), selective laser sintering (LAM), binder jetting (Binder Jetting), material jetting (Material Jetting), multi-jet fusion (Multi-Jet Fusion), powder bed fusion (PBL), directed energy deposition (DMD), and laminated object manufacturing (LAM). Each of these methods uses a specific type of press, which is a machine or system of machines that generates a three-dimensional molded object by constructing layers and then continuously fusing, bonding, or otherwise attaching these layers to each other. Generally, the press and its components are controlled by a computer, which stores specific data in its memory representing the commands required for the press's hardware components, and determines how to approximately generate each piece of the three-dimensional molded object and how to fuse, bond, or otherwise attach the layers made up of each piece to the previous and next layers. These instructions may also be generated by (the same or different) computers, the latter of which have a geometric representation based on data of a three-dimensional molded object, the ability (by software) to divide these geometric representations into fragments, and the ability (by software) to generate data representing instructions that can be read by one or more printing presses, thereby approximately generating, fusing, joining, or otherwise adhering to a series of adjacent layers.

[0004] Generally speaking, printing presses can manufacture objects from a wide variety of different materials, including, for example, polymers, thermoplastics, metals, ceramics, glass, and composite materials consisting of any combination thereof. However, each type of printing press is generally intended to produce three-dimensional objects from a narrower range of materials. For example, stereolithography (SLA) printers generally print objects from photosensitive polymers, first depositing a thin layer of the photosensitive polymer onto a surface, and then curing or solidifying the area of ​​the surface representing a fragment of the molded object within that layer by exposing that area to UV or other light. The lamination and exposure procedure is then repeated at least once, and generally multiple times, with each deposited layer creating height in the molded object at every point in the deposition direction, and in each consecutive exposed area, consecutive fragments in the same three-dimensional molded object have an approximate shape. Generally, as shown in Figure 1, the continuous layers constituting the fabricated molded product are placed on a build platform, which is further placed on a precision stage. This precision stage is lowered by a certain distance after each deposition and exposure step, and this distance is approximately equal to the thickness of the photosensitive polymer layer deposited in each step.

[0005] In general printing, the approximate average thickness of each layer is roughly equal to the absolute value of the distance traversed by either: a) the stage used in stereolithography, or b) the location where a material or energy deposition nozzle is used, the distance being the distance between a given point on one fragment or its representative layer n and a point on a consecutive fragment or layer n+1, defined as the overlap between 1) a vector pointing in the layering direction starting from point n and 2) the subsequent layer n+1. This approximation of the thickness of each fragment or layer as a whole, or at each point on a fragment or layer, is generally called the layer thickness; in Figure 1, in the example of printing using a stage, the layer thickness is denoted as LT.

[0006] In material spraying procedures, when a dispenser deposits material to produce a flat layer, the material is usually in the form of a suspension, slurry, liquid, gel, powder, or other, and is applied along aligned, adjacent paths, often in straight lines, so that the rough shape of the area covered by numerous trajectories of this material represents a fragment of the molded product. Material spraying printers can deposit energy in addition to material, and can deposit one or multiple trajectories at a time, for example, by using multiple nozzles, or by using different materials with different nozzles.

[0007] A specific technique used to deposit numerous adjacent material trajectories (and / or energy) to form a rough shape defined by fragments of an object can generally be called a scanning method for those fragments or layers. While the order of trajectory generation within a layer is not necessarily critical to the procedure, faster procedures typically involve sequentially forming the nearest adjacent trajectories, as this minimizes the time spent on each layer when generating trajectories by moving the print head or redirecting the material or energy. The deposited material generally concentrates or solidifies quickly after deposition, forming the rough shape of its fragments. As shown in Figure 2, the height of the deposited trajectories is approximately equal to the layer thickness, with spacing roughly corresponding to their width; the spacing between adjacent trajectory paths is often called the hatch spacing. Similar to stereolithography procedures, precision stages can be used, but in contrast to stereolithography, material can be deposited to create continuous layers by raising one or more nozzles in the layering direction by a distance approximately equal to the layer thickness at each point.

[0008] The binder spraying procedure is similar in some respects to the material spraying procedure, where, for each layer, the trajectories of adjacent materials are deposited in the rough shape of a three-dimensional object fragment representative of that layer, according to one of several scanning methods. However, in the binder spraying procedure, thin layers of powder material are initially deposited over the entire area to be constructed, and the material deposited in adjacent trajectories is a binding agent or glue, which holds the powder particles firmly together through polymerization (cured) or other means. The immature green article produced in continuous layers consists of both powder particles and cured binder.

[0009] Powder bed fusion and laser or other sintering procedures are similar to binder injection procedures in terms of powder deposition, but in these procedures, an energy source melts or sintersects powder particles, forming a solid layer from either the material that re-solidifies after melting or the partially melted (sintered) particles, which are roughly fixed in space by their relative rigidity. Figure 3a shows two types of layers using scanning methods commonly used in powder bed fusion and laser sintering procedures, as well as other printing procedures. In this case, as is often seen in the other procedures described above, the scanning path is defined as a one-dimensional path along which the material is deposited, or as the direction in which the energy source is scanned to form the trajectory, but the direction of the scanning path can give the manufactured molded product desirable physical properties by controlling and rotating it from layer to layer, including, for example, increased isotropy of elastic properties, decreased porosity, or increased print volume / decreased printing cost. Figure 3b is a plan view image of the scanning trajectory in a single layer printed in a powder bed fusion procedure according to the scanning path shown in Figure 3a. Figure 3c is a cross-sectional image of a molded product manufactured using the scanning method shown in Figure 3a. There are also powder bed fusion procedures that are not path-based but pattern-based, similar to stereolithography. In these procedures, the energy pattern is projected over a large area of ​​the constructed surface, while smaller spots with energy that are raster-scanned or scanned on that area are not projected.

[0010] Several past patents and studies have discussed scanning methods, claiming their advantages for printing as improvements to the physical properties of the molded material and to the economic and / or manufacturing considerations of printing. For example, U.S. Patent No. 5,155,324 of October 13, 1992, describes a selective laser sintering procedure in which scanning lines, parallel and antiparallel within a single layer, are rotated by a predetermined amount in each of the consecutive layers. U.S. Patent No. 7,569,174 of August 4, 2009, describes a selective laser sintering procedure in which each layer of the molded material is scanned once with a laser at a higher energy, covering the entire planar cross-section of the material in that fragment, and then scanned at least one more time at a relatively lower energy. The purpose of this procedure is to anneal or further melt each layer of the molded material, with reported strength improvements reaching 100%.

[0011] Furthermore, there are many patents that discuss compositions manufactured using scanning in specific patterns. Compositions are often tied to scanning patterns because parts need to be constructed layer by layer and / or trajectory by trajectory. For example, when a molded product is manufactured with trajectories generated by scanning in only one direction (and / or only antiparallel to that direction), the strength, ductility, conductivity, thermal conductivity, and other properties of the molded product typically increase in that direction, but decrease in other directions, particularly in the two directions perpendicular to the scanning direction, although there is no strict regularity. This change in the elastic, mechanical, and other physical properties of the molded product is due to local phase, bonding, and crystal structure, which are brought about by the elongation of the scanning trajectory and the lamellarity resulting from layer by layer formation. Crystal grain boundaries and other types of chemical / microstructural boundaries tend to form at the ends of the scanning trajectory, and the substantial changes in elastic and transport modes resulting from this can be expressed as a function of the location and concentration of these boundaries. In previous examples, all trajectories are scanned in a single direction, but by moving in the scanning direction (or in a direction opposite to it), the boundaries of the trajectories do not intersect, and the physical properties become substantially anisotropic.

[0012] Also, while there are few patents and research papers discussing a special scanning method that specifies not only the angle but also the relative positions of the scanning lines within consecutive planar layers. In this method, the scanning lines are "interleaved" between the layers, and the scanning lines of layer n+1 are placed at the midpoints of the scanning lines of layer n without rotation, as shown in Figure 4. U.S. Patent No. 6,596,224 discusses applying this method to a material injection or thermal melting deposition procedure, stating that by using such a method over approximately 10 layers near the top in terms of the shape of the molded object, the finish of the surface of the molded object can be improved. U.S. Patent No. 6,677,554 discusses applying this method to a selective laser sintering procedure, stating that since the total number of scanning lines, i.e., the sum of the lengths of all scanning lines, can be reduced by this method, it brings an improvement in throughput in the manufacture of the molded object and further has the potential to improve the strength of the molded object. In the journal article: X. Su and Y. Yang, Research on track overlapping during Selective Laser Melting of powders, Journal of Materials Processing Technology 212, pp. 2074-2079 (2012), it is researching to avoid the non-uniform distribution of the input energy by applying this method to a selective laser melting (powder bed fusion) procedure.

Summary of the Invention

Means for Solving the Problems

[0013] Definition:

[0014] Packing problem: Packing is the configuration (spatial arrangement) of non-overlapping objects within a small region of d-dimensional Euclidean space or d-dimensional curved space. Periodic packing in Euclidean space means that γ objects called bases are periodically replicated in the volume v within the space UThese are contained within a unit cell. The shape and symmetry of the unit cell are determined by the "d" lattice basis vector b (to avoid confusion with the base object). Objects can take any shape, such as convex (circle, ellipse, sphere, cylinder, etc.) or concave (e.g., Star of David, cruciform, stellated polyhedron, hyperbolic paraboloid). In polydisperse packings, objects have different sizes and / or shapes. Binary packings are packings where they have two different sizes and / or shapes. Ternary packings are packings where they have three different sizes and / or shapes. Each object's spatial configuration is defined by its coordinate r, which determines both its position and orientation; for example, in three dimensions, this coordinate r includes three spatial coordinates (x, y, z) representing the object's center of mass and two angular coordinates (θ, φ) representing the displacement in the longitude and latitude directions around the object's center. A fundamental characteristic of packing is the packing fraction, which represents the proportion of space occupied by an object. In periodic packing, the packing fraction is the total volume of γ base objects in a unit cell, which is v U It is calculated by dividing by [a certain factor]. Efficient packing results in a high packing fraction. Best packing results in the highest packing fraction among all packing methods.

[0015] Covering problem: Covering is a three-dimensional arrangement of overlapping objects that completely covers a small region in d-dimensional Euclidean space or d-dimensional curved space. Periodic covering in Euclidean space is a volume v in which γ objects called a base are periodically replicated in space. U It must be contained within a unit cell. The shape and symmetry of the unit cell are determined by d lattice basis vectors b (to avoid confusion with the base object). Objects can take on any shape, such as convex (circles, ellipses, spheres, cylinders, etc.) or concave (e.g., Star of David, cruciform, stellated polyhedron, hyperbolic paraboloid). In polydisperse coverings, objects have different sizes and / or shapes. In binary coverings, objects have two different sizes and / or shapes. In ternary coverings, objects have three different sizes and / or shapes. The spatial configuration of each object is defined by its coordinate r, which determines both its position and orientation; for example, in three dimensions, this coordinate r includes three spatial coordinates (x, y, z) representing the object's center of mass and two angular coordinates (θ, φ) representing the displacement in the longitude and latitude directions around the object's center. A fundamental property of a covering is its covering density, which is the total volume of objects per unit volume. The covering density of a periodic covering is the total volume of γ base objects in a unit cell multiplied by v U It is calculated by dividing by [a certain factor]. Efficient coating results in low coating density. The best coating results in the lowest coating density among all coatings.

[0016] Track Cross Section: This is the average of the shapes determined by surface intersections, most simply a plane, and a single trajectory of a material where intersections can occur at various points along the scanning path of the trajectory; therefore, in this specification, this is defined as a track cross section. Most simply, a track cross section is formed by the intersection of the trajectory with a plane having a unit normal parallel to the direction of the trajectory or scanning propagation at that point, and the average shape is obtained by spatially averaging these multiple intersections at different points along the trajectory or scanning path. For clarity, we will refer to the track cross section thus formed as a path-tangent track cross section, but note that the selection of the track-intersecting surface and its direction is not essential for forming the track cross section; however, consistency in the selection of the intersecting surface and direction is useful when identifying differences between trajectories or designing scanning methods. Figure 5 includes several images of different path-tangential trajectory cross-sections produced by a laser powder bed fusion printer, where the differences in the shape and size of the trajectory cross-sections depend on the printing settings, including, but are not limited to, laser power and scanning speed.

[0017] One or more trajectory sections, fixed or not fixed in the angular direction, can be used as a basis set for the solution of a covering problem. The same applies to geometric approximations of one or more actual trajectories in three dimensions (the simplest being cylinders). More simply, if a packing consisting of one or more objects of the same or different sizes, or more simply (but not limited to these), if a two-dimensional disk and a three-dimensional cylinder function as a basis within a two-dimensional or three-dimensional unit cell, then a trajectory section can be considered a two-dimensional representation of a trajectory, or a three-dimensional approximation of a trajectory, and can be used to create a scanning scheme. Thus, mappings exist as trajectory sections, packings and scanning schemes, and as trajectory sections, solutions to covering problems and scanning schemes.

[0018] Slice thickness, layer thickness, material thickness, and net displacement: Generally, slice thickness, material layer thickness (the powder layer thickness before energy or material deposition in powder bed printing procedures), and the absolute value of the net displacement of the stage or material-(and possibly energy) deposition printhead at each point or for each slice or representative layer are all considered the same concept and are referred to as layer thickness. Basically, slice thickness, material / powder thickness, and net displacement do not need to be the same; in particular, using different values ​​for material thickness and net displacement in printing can bring advantages to the quality of the printed part and the printing procedure, as will be explained in more detail later.

[0019] The term "slice thickness" is used herein to mean the thickness of a slice of a three-dimensional molded product in the vicinity of a given point on that slice, where "slice" is defined above. While the term "slice" is used herein in relation to design, the term "layer" is generally used in reference to molded products. In this application, the material layer thickness and the slice thickness are not necessarily the same, either on average or at a specific point.

[0020] The average thickness of a single layer of material, or the local thickness at a specific point within a layer, can vary substantially and can deviate significantly from the maximum height of the trajectory. In Figure 2, the trajectory height varies locally, and the height variation occurs periodically in an order of trajectory width; this is also true in the case of powder bed fusion printing, as seen in Figures 3b, 3c, and 5, and applies to all scans and trajectory-based printing procedures, or similar printing procedures. Therefore, from a local perspective, the thickness of the deposited material layer or trajectory is not the same as the thickness of the fragment: only the average thickness of the material deposited over many trajectories and many layers is approximately equal to the thickness of the fragment. This latter description is based on the assumption that the design length is approximately replicated by the printing press when physically manufacturing the molded product.

[0021] In powder bed fusion printing and other powder-based printing procedures, the differences in the concepts of fragment and layer thickness also stem from local variations in the thickness of the deposited powder layer, which change laterally by a very small order of particle size depending on the distribution of particle size and shape. Furthermore, in powder bed fusion, the trajectory depth tends to be much greater than the fragment thickness due to the remelting of a pre-molten layer to achieve nearly perfect density in the molded product. Also, as described herein, the material thickness or powder thickness is simply determined as the average thickness of the material deposited near a point (before binding, fusion, melting, sintering, or other printing processes), but does not need to be the same as the average layer thickness or fragment thickness. This is an important consideration from the perspective of the trajectory geometry, and therefore also an important consideration from the perspective of the trajectory cross-sectional shape, because changing the material thickness changes the trajectory and the trajectory cross-sectional shape.

[0022] Typically, the thickness of the material is kept nearly constant across the layers of deposited material. Nevertheless, the thickness of the deposited layers depends on the material properties and the settings of the printing press. For example, when powder is deposited or spread using a powder shuttle, the distance between the blades, wipers, rollers, or other spreading devices on the powder shuttle and the height of the object being built strongly affects the thickness of the powder. This latter distance is often controlled by the stage, but during the printing process, the layers may move several times in the direction of the molded object or in the opposite direction. This net movement of the stage during the process of printing layers is an example of net displacement. In printing procedures using a powder bed and stage, the net displacement at a given point is most similar to the thickness of the fragment at that point. Importantly, since the powder is deposited with any displacement of the stage, the thickness of the powder and the net displacement can be controlled individually, meaning that the net displacement (thickness of the fragment) and the thickness of the powder do not need to be the same. Figure 6 shows an example of this concept, where the fragment thickness and net displacement are constant across the illustrated fragment, with a powder thickness of 35 μm for a net displacement of 50 μm (and a fragment thickness of 50 μm).

[0023] The thickness of the material (or powder) and the thickness of the fragments can vary even in AM procedures with a print head but no stage, or in AM procedures with both a stage and a print head; the average thickness of the material deposited near a point on the fragment does not need to be equal to the net displacement of the print head at that point. For example, a fragment thickness that is constant across the fragment from position x=0 to x=L can be used for layer n, and the average thickness of the deposited material changes linearly over that distance, from the minimum thickness of the material deposited at x=0 to a few microns, and from the maximum thickness of the material deposited at x=L to a few microns.

[0024] Filling methods, contour scanning, support structures, print lattices, and others: In printing presses and printing that use procedures based on paths, scanning, and trajectories, the printing settings can be varied for structurally different regions of the molded object, which are determined according to the size of the molded object. Importantly, a set of scanning methods can be used by using different printing settings within a single layer, whereas different printing settings are not used within the same method, but rather different methods are used across regions of the molded object determined by the (larger) geometry of the target molded object, rather than in a periodic, alternating, or other manner determined by the scale of the (smaller) scanning trajectories.

[0025] Contour scanning is used on the surface of parts for several reasons, but particularly for the following purposes: to improve the surface finish of the part; to ensure greater geometric alignment between the fabricated part and the design, drawings, or computer data of the part to be manufactured; or to reduce the porosity of the surface of the fabricated part. Multiple contour scanning trajectories can be placed side by side, but the number of trajectories placed side by side in this manner is usually limited to one, two, or three.

[0026] There are other reasons for using support structures, but they prevent warping of the molded object, allow heat to dissipate by conduction from thermally isolated areas within the print, and hold the molded object in place on the build plate. Often, support structures are formed in a single scan at equal intervals in the unprinted space between scans, creating a continuous series of thin walls, spikes, or slender columns. However, these support structures are not intended to be part of the finished molded object and are usually cut off, melted down, or removed after printing.

[0027] Unlike the mathematical concept of a grid defined above, print lattices are scaffold-like structures consisting of a single three-dimensional unit or voxel in space, encompassing both printed and unprinted materials, and are repeatedly replicated to cover a region of space spanning numerous voxels. Print lattices represent a structure of struts and nodes, where the struts of the printed material consist of nodes extending in various directions, connecting with other struts at other nodes. These can be generated by a single scan, like a support structure, or each strut can be formed by multiple adjacent, overlapping or nearly overlapping trajectories.

[0028] The filling method is designed to fill space and differs from conventional methods in that, in generating solid regions, it either has a very small porosity or a probabilistic porosity where the size of the voids is approximately equal to or smaller than the spacing between parallel lines. Several examples of filling methods are shown in Figures 3a and 34. The filling method can also be used to fill the struts and nodes of print lattices in a way that these struts are not formed by a single trajectory. The distinction between filling methods and contour scanning is ambiguous, becoming so when a greater number of contour scans are performed near the surface of the printed molded object. Some and others of the filling methods are called skins, or by other terms, meaning that the lateral extent of such filling methods is significantly greater than their thickness.

[0029] Summary of the present invention:

[0030] The present invention relates to scanning methods, and related methods, systems, and compositions for additive manufacturing, as well as methods and systems for making the thickness of the material layer deposited during printing different from the net displacement of the stage or the material and / or energy-depositing print head, and / or different from the thickness of the fragments of the molded product. Useful applications of the present invention include: 1) fused deposition modeling, 2) material jetting, 3) binder jetting, 4) powder bed fusion, 5) selective laser sintering, 6) multi-jet fusion, 7) directed energy deposition, 8) direct metal deposition, 9) electron beam additive manufacturing, 10) arc plasma sintering additive manufacturing, and other additive manufacturing methods that can be generally considered to use scanning paths and trajectories, in which the surface height of the material or energy deposition pattern varies periodically over a wider area, as if it were appearing as a single continuous trajectory. In the following discussion, the trajectory, trajectory cross-section, scanning method, and scanning path include additive manufacturing methods, systems, and compositions that utilize scanning paths and trajectories, as well as those in which the surface of a material or energy deposition pattern over a wider area changes in a manner similar to that of a single continuous trajectory.

[0031] In one embodiment, the present invention intends an additively manufactured composition in which the geometric structure of the trajectory or the geometric structure of the trajectory cross-section is substantially similar to all or substantially all scan trajectories within the area of ​​the printed layer, but different from all or substantially all trajectories within adjacent areas of preceding or succeeding layers. In one embodiment, the present invention intends an additively manufactured composition in which two or more shapes or sizes of trajectories or trajectory cross-sections are used when manufacturing one layer, multiple layers, or portions thereof, and the external shapes of such trajectories or trajectory cross-sections are targeted to differ in coordinated fashion within each of one layer, multiple layers, or portions thereof. In one embodiment, the present invention intends a layered composition in which the outlines of multiple trajectories or trajectory cross-sections of different shapes or sizes within a region of the molded product are located along a crystallographic dot pattern, and these trajectories cover unit cells of such a pattern so that the size of the unprinted area within a cell does not exceed the size of the largest trajectory (or its cross-section), and furthermore, portions consisting of two adjacent cells are present in part of each layer within the region.

[0032] In one embodiment, the present invention intends a additively manufactured composition in which multiple trajectories or trajectory cross-sections of different sizes and / or shapes are used as solutions to a two-dimensional (for trajectory cross-sections) or three-dimensional (for trajectories) covering problem, and the outline of the trajectories or trajectory cross-sections in such solutions is used to define the relative position of a scanning path over a region that includes part or all of one or more consecutive layers in a multilayer scanning method.

[0033] In one embodiment, all or nearly all of the scanning trajectories described in paragraphs 2 and 3 of this "Summary of the Invention" chapter that lie within a region encompassing portions of two or more consecutive layers are parallel or antiparallel. In one embodiment, the scanning trajectories described in paragraphs 2 and 3 of this chapter lie within a region of a given layer and are all or nearly all rotated by a specific angle with respect to scanning trajectories in adjacent regions of the previous layer. In one embodiment, the scanning trajectories described in paragraphs 2 and 3 of this chapter are all or nearly parallel or antiparallel to scanning trajectories in adjacent regions of the previous layer within a region of a given layer, but scanning trajectories in adjacent regions of consecutive layers are rotated by a specific angle. In one embodiment, the scanning trajectories described in paragraphs 2 and 3 of this chapter are ordered chronologically, and in each layer partially or entirely contained within a region of the molded product, either adjacent or non-adjacent trajectories are printed consecutively.

[0034] In one embodiment, the present invention is intended to be a method for designing a scanning method in which the geometric structure of the trajectories or the geometric structure of the trajectory cross-sections are substantially similar for all or substantially all scanning trajectories within a printed layer region, but different from all or substantially all trajectories within adjacent regions of preceding or succeeding layers. In one embodiment, the present invention is intended to be a method for designing a scanning method using two or more shapes or sizes of trajectories or trajectory cross-sections in which the external shapes of such trajectories or trajectory cross-sections are targeted to differ in coordinated fashion from the scanning trajectories within a region of an already printed layer. In one embodiment, the present invention is intended to be a method for designing a scanning method in which the external shapes of multiple different shapes or sizes of trajectories or trajectory cross-sections within a region of a molded object are located along a crystallographic dot pattern, and these trajectories cover unit cells of such a pattern so that the size of the unprinted area within a cell does not exceed the size of the largest trajectory (or its cross-section), and furthermore, portions consisting of two adjacent cells exist in at least a portion of each layer within the region.

[0035] In one embodiment, the present invention is intended to be a method for designing a scanning scheme in which a plurality of trajectories or trajectory cross-sections of different sizes and / or shapes are used as solutions to a two-dimensional (for trajectory cross-sections) or three-dimensional (for trajectories) covering problem, and the outline of the trajectories or trajectory cross-sections in such solutions is used to define the relative position of a scanning path over a region that includes part or all of one or more consecutive layers in a multilayer scanning scheme.

[0036] In one embodiment, all or nearly all of the scanning trajectories described in paragraphs 5 and 6 of this "Summary of the Invention" chapter that lie within a region encompassing portions of two or more consecutive layers are parallel or antiparallel. In one embodiment, the scanning trajectories described in paragraphs 5 and 6 of this chapter lie within a region of a given layer and are all or nearly all rotated by a specific angle with respect to scanning trajectories in adjacent regions of the previous layer. In one embodiment, the scanning trajectories described in paragraphs 5 and 6 of this chapter are all or nearly parallel or antiparallel to scanning trajectories in adjacent regions of the previous layer within a region of a given layer, but scanning trajectories in adjacent regions of consecutive layers are rotated by a specific angle. In one embodiment, the scanning trajectories described in paragraphs 5 and 6 of this chapter are sequential, and in each layer partially or entirely contained within a region of the molded product, adjacent or non-adjacent trajectories are printed in sequence.

[0037] In one embodiment, the present invention intends to be an additive manufacturing method in which the geometric shape and / or average area of ​​the trajectory (and / or cross-sectional trajectory) within a region of the molded work is changed by changing the thickness or net displacement of the fragments of each layer within the region of the molded work. In one embodiment, the thickness of the material of the layers within a region of the molded work is different from (may be greater than or less than) the thickness or net displacement of the fragments, but the thickness or net displacement of the fragments of the portion of each layer contained within the region may be the same as or different from that of other layers within the region. In one embodiment, the thickness of the material deposited in one or more layers within a region of the molded work as described in this paragraph is different from (may be greater than or less than) the thickness or net displacement of the fragments of the portion of the one or more layers within the region, by a method of changing the geometric shape and / or average size of the scanning trajectory within the portion of the one or more layers enclosed by the region. [Brief explanation of the drawing]

[0038] [Figure 1] (a) is a schematic diagram of a conventional stereolithography apparatus, showing a standard layering and exposure procedure. As an example, the build plate is mounted on a precision stage positioned within a bath of photocurable polymer. The build plate is positioned so that a thin layer of unpolymerized polymer equal to the layer thickness (LT) to be built is deposited on its surface, and subsequently exposed to an energy source in a region representing a fragment of the article to selectively cure or solidify the material.

[0039] (b) is a schematic diagram of the build plate of a stereolithography apparatus, which is lowered by a similar length (LT) to generate a new layer of unpolymerized polymer, as shown in Figure 1a. This exposure and layering procedure shown in this figure and Figure 1a is repeated, where the continuity of the exposed region and deposited layer gives height to the molded object in the build-up direction at each point, and these approximate the shape of continuous fragments of the same three-dimensional molded object.

[0040] [Figure 2] This is an explanatory diagram of the deposition of planar layers of a molded product by material injection. One or more nozzles deposit material along adjacent paths adjusted to be spaced apart at a periodic interval called the hatch spacing h. The deposited material forms a trajectory with a width matching the hatch spacing and a height approximately equal to the layer thickness (LT) of the fragmented molded product, with the rough shape of the trajectory region representing the fragment of the molded product. Continuous fragments of the same three-dimensional molded product can be deposited by a combination of these scanning paths, with the operation of a precision stage or dispensing nozzle, while offsetting the spacing in the deposition direction by an amount approximately equal to the layer thickness at that location.

[0041] [Figure 3] (a) is a schematic diagram showing an energy deposition scanning method commonly used in powder bed fusion and laser sintering procedures. In this example, the scanning method consists of a single contour path representing the transverse cross-section of a piece of molded material and a number of parallel adjacent paths separated by a hatch spacing h, which reach (or fill) adjacent areas of the piece of molded material when the powder is solidified. By scanning the energy source along a predetermined path, the powder particles are melted or sintered, and then resolidified to form a solid layer. The scanning method does not need to be the same or similar between layers — rotation or movement with increasing magnitude is performed relative to the scanning path. For example, between layers n and n+1 representing consecutive pieces of molded material, the filling path is at an angle θ n From θ n+1It rotates in a manner controlled by a constant angle increase up to a certain point.

[0042] (b) is an optical microscope image of the top of a single layer of the scanning trajectory of steel manufactured by powder bed fusion, using the same scanning method as shown in Figure 3a. Reprinted from I. Yadroitsev, P. Krakhmalev, and I. Yadroitsava, Hierarchical design principles of selective laser meting for high quality metallic objects, Additive Manufacturing 7, pp. 45-56, 2015.

[0043] (c) is an optical microscope image of an etched vertical cross-section of a multilayer molded product manufactured by powder bed fusion, obtained using the same scanning method as shown in Figure 3a. The cross-section of the re-solidified powder along the scanning trajectory shows a complex layered structure and isotropy, which can be achieved by controlling the rotation of the scanning path between layers. Reprinted from I. Yadroitsev, P. Krakhmalev, and I. Yadroitsava, Hierarchical design principles of selective laser meting for high quality metallic objects, Additive Manufacturing 7, pp. 45-56, 2015.

[0044] [Figure 4] Figure 3a is a schematic diagram illustrating a specific embodiment of the scanning method. In this method, the position of the scanning path is defined within a predetermined layer and in relation to consecutive layers to create an "interleaved" pattern. For example, in layer n, a large number of adjacent parallel scan lines are separated by a hatch spacing h, and the cross-section of the molded product is shown as a shaded area. Similarly, the filled scan lines in the consecutive flat layer, layer n+1, are separated by a hatch spacing h, but are shifted laterally by a distance equal to half the hatch spacing relative to the next layer, so that the scan lines of layer n+1 are positioned at the midpoint of the scan lines in layer n.

[0045] [Figure 5] (a) is an image of the path-tangent track cross section of a scanning line on stainless steel printed by laser floor melting at a laser output of 50 W and a laser scanning speed of 60 mm / s. Reprinted from I. Yadroitsev, A. Gusarov, I. Yadroitsava, and I. Smurov, Single track formation in selective laser melting of metal powders, J. Mater. Process. Technol. 210, pp. 1624-1631, 2010.

[0046] (b) is an image of the path-tangent track cross section of a scanned line on stainless steel printed by laser floor fusion bonding with a laser output of 50 W and a laser scanning speed of 120 mm / s. Compared with Figure 5a, the changes in the shape and dimensions of the cross section with increasing scanning speed are noteworthy. Reprinted from I. Yadroitsev, A. Gusarov, I. Yadroitsava, and I. Smurov, Single track formation in selective laser melting of metal powders, J. Mater. Process. Technol. 210, pp. 1624-1631, 2010.

[0047] (c) is an image of the path-tangent track cross section of a scanning line on stainless steel printed by laser powder-bed fusion at a laser power output of 300 W and a laser scanning speed of 800 mm / s. The cross-sectional shape is more elongated compared to Figures 5a and 5b. Reprinted from C. Kamath, B. El-dasher, GF Gallegos, and WE King, and A. Sisto, Density of additively-manufactured, 316L SS parts using laser powder-bed fusion at powers up to 400W, Int. J. of Adv. Manuf. Technol. 74, pp. 65-78, 2014.

[0048] (d) is an image of the path-tangent track cross section of a scanned line on stainless steel printed by laser powder-bed fusion at a laser power output of 300 W and a laser scanning speed of 1500 mm / s. As can be seen from a comparison with Figures 5a-5c, the shape and dimensions of the cross section of the re-solidified material include, but are not limited to, circularity, contact angle, width, material accumulation, and depth, and depend on printing parameters including laser power output and scanning speed. Reprinted from C. Kamath, B. El-dasher, GF Gallegos, and WE King, and A. Sisto, Density of additively-manufactured, 316L SS parts using laser powder-bed fusion at powers up to 400W, Int. J. of Adv. Manuf. Technol. 74, pp. 65-78, 2014.

[0049] [Figure 6] (a) is a schematic diagram of an example procedure using powder, defining the thickness (MT) of the material used in the AM procedure for the molded product. The build plate is lowered by 35 μm using a precision stage to create a gap in the powder bed, and powder from another source is rolled into the gap. The thickness of the material is defined as 35 μm.

[0050] (b) is a schematic diagram of the movement of the build plate in the AM procedure using powder, immediately after the powder has been deposited as described in Figure 6a. Further displacement of the build plate before the powder solidifies specifies a net displacement (ND) that is greater than, equal to, or less than the material thickness (MT).

[0051] (c) is a schematic diagram of the procedure using powder, where the net displacement (ND) and the thickness of the material (powder) (MT) can be adjusted independently and made different, but this is not necessarily required. Figures 6a and 6b show the net movement of the precision stage when all steps before the powder solidification are combined, specifying the total net displacement, which in this case is 50 μm, while the thickness of the material is 35 μm, as specified by the lamination procedure shown in Figure 6a.

[0052] [Figure 7] (a) is a densely packed unit cell of two types of circles (binary disks), from which a multiple setpoint type scanning scheme can be derived.

[0053] (b) is a solution to a covering problem for three objects of different sizes and shapes, with a fixed angular direction and a ratio of 2:1:1 per unit cell, consisting of two large objects and two small objects (mirror images of each other). In this case, the geometry of the objects is determined by laser sintering, powder bed fusion, or a similar procedure, using AM track geometry parameters {P i}1 and {P i}2 represents path-tangent track cross sections for trajectories generated from two separate sets, and the solution to this coating problem can be applied to any additive manufacturing procedure using scanning or trajectories, or to an additive manufacturing procedure that generates a surface pattern similar to that generated by a procedure using trajectories or scanning. Symbols 1, 2, and 3 represent the chronological ordering for the additive manufacturing, and such ordering is not specific to the solution but rather added afterward. Small mirrored objects are used to reflect subtle differences in geometric structure, which is achieved by printing trajectories at positions on the surface where the height changes asymmetrically (in this case, the center of the top of the previously printed scanning trajectory is offset).

[0054] (c) shows a cross-sectional view of an already printed layer, a powder or other material deposited on top of layer 1 shown in Figure 7b, and the filled dots in the already printed layer represent the scanning path when scanning in a direction perpendicular to the plane of the figure.

[0055] (d) is a diagram showing possible results for printed layer 2, where the scanning path is parallel or antiparallel to the scanning path in layer 1, as shown in Figure 7b. The illustration of remaining unconsolidated powder or other material indicates the possibility of sintered material being present. The illustrated path-tangent track cross sections show scanning directions parallel or antiparallel to the path in layer 1, but the path in layer 2 can be oriented at a specific (or arbitrary) angle from the path in layer 1, or along a contour where the angle with respect to the path in layer 1 changes as the scan progresses.

[0056] [Figure 8] The path-tangent track cross section of the printed portion was created using a single-trajectory scanning method. It consists of repeating layers of thermoplastic material (polylactic acid) printed by material ejection, using the basic scanning path described in Figure Gb. To form each trajectory, the same settings as those described in Figure 12 were used, except for a nozzle scanning speed of 1800 mm / m and an extrusion ratio of 0.35. The settings used to create this portion are the default settings of the Prusa i3, which are optimized to minimize mechanical damage, surface roughness, and geometric deformation of the portion, while also minimizing porosity. The measured porosity of this portion was 8%.

[0057] [Figure 9] This is a cross-sectional view of a portion printed by material spraying of a thermoplastic polymer, using the same scanning method as in Figure 8, but with an extrusion volume ratio setting of 0.52, which is equal to the average extrusion volume ratio setting used for the portion shown in Figure 10. In this example, excess material was extruded, and the printing press could not complete the portion. This is because too much material was extruded (resulting in excessive overlap), resulting in very poor surface finish and geometric accuracy, and the porosity of the entire portion changed, becoming higher as it moved away from the edges of the portion. The average porosity of the portion away from the edges is 7%. It is noteworthy that despite the average extrusion volume ratio of 0.52 used for this portion being much higher than the 0.35 used for the portion shown in Figure 8, the increase in material extruded per unit volume of the printed portion was only about 1% due to the increased pressure in the nozzle with a higher average extrusion volume ratio setting.

[0058] [Figure 10] Figure 16 shows a cross-sectional view of a portion printed by material ejection of a thermoplastic polymer using a two-trajectory geometric structure scanning method based on the solution to the coating problem shown. The trajectories were formed using the same settings as described in Figure 8, but the extrusion volume ratio for the large trajectory was 0.69 and for the small trajectory it was 0.34. The measured porosity of this portion was 0.2%. Compared to the method used for the portion shown in Figure 8, it was clear that the porosity was significantly reduced with this two-trajectory method. It is noteworthy that even though a much larger average extrusion volume ratio of 0.52 was used for this portion compared to 0.35 for the portion shown in Figure 8, the increase in extruded material was only about 8.4% in nozzles with a higher average extrusion volume ratio due to the increased pressure.

[0059] [Figure 11] This is a cross-sectional view of a portion printed by material spraying of a thermoplastic polymer. The same basic scanning path as that used for the two-trajectory geometric structure scanning method shown in Figure 26 was employed, and the same settings as those shown in Figure 10 were used to form small and large trajectories. The measured porosity of this portion was 2.3%. It is clear that a significant reduction in porosity is achieved with these two trajectory methods compared to the method used for the portion shown in Figure 8. It is noteworthy that, despite using a much higher average volume extrusion ratio of 0.47 in this portion compared to 0.35 in the portion shown in Figure 8, the increase in extruded material was only about 6.4% due to the increased pressure at nozzles with higher average extrusion volume ratios.

[0060] [Figure 12] Figure 16 shows a cross-sectional view of the path-tangent trajectory of a portion printed by a scanning method consisting of repeating layers of thermoplastic material (polylactic acid) printed by material jetting, based on the solution to the coating problem explained in detail. In this figure, the press settings were selected to show the differences in the geometric structure of the scanning trajectory, and the porosity and other considerations of the portion were not optimized. Within a single layer, different scanning methods result in trajectory cross-sections with alternating trajectory widths. In this case, the nozzle diameter of the Prusa i3 fused deposition modeling press is 400 μm, the print settings are a hatch spacing and layer thickness of 200 μm, a nozzle scanning speed of 900 mm / min, and an extruder temperature of 190 °C, with narrow trajectories printed at an extrusion volume ratio of 0.195 and wide trajectories printed at an extrusion volume ratio of 0.39. During printing of this section, the small trails being printed tend to adhere to the most recently printed adjacent larger trails, so the two groups of trails observed in the image, consisting of one small trail and one large trail, are spatially separated.

[0061] [Figure 13] This diagram illustrates the replication of unit cells within and across layers to cover an arbitrary geometric space by solving a covering problem. As shown in the figure, by omitting the trajectory from individual unit cells, the entire space to be covered can be more accurately fitted.

[0062] [Figure 14] (a) is an illustration illustrating a method for adjusting scanning trajectories to fit the boundaries of objects by removing scanning trajectories within individual unit cells, in order to fit scanning trajectories to a covering problem derived from a scanning scheme using multiple objects with different AM trajectory geometric structure settings. The image shows a cross section with a normal vector that is a path-tangent to the scanning path of the scheme derived from the covering problem in that cross section; however, such a cross section does not mean that the path constitutes a tangent to the boundary of the molded object in any layer, or a tangent to a contour close to that boundary. In general, the trajectories derived from the solution to the covering problem do not need to be straight or parallel to each other, nor do they need to be contours that traverse different layers.

[0063] (b) shows the addition of several layer-based contour line trajectories (represented by trajectory cross-sections) at the boundary of a molded object designed using a scanning scheme derived from the coating problem. In this figure, the contour line trajectory cross-section at the boundary is the thickness of one layer, but it may be for multiple layers, or it may consist of an additional scanning scheme derived from the coating problem using multiple AM ​​trajectory geometric structure settings. Furthermore, although the contour line trajectory cross-section is simply shown as a path-tangent cross-section, the contour line may move in different directions or angles relative to any cross-section of the molded object with respect to the trajectory of the scanning scheme derived from the coating problem. In relation to Figure 14a, note that one trajectory of the scanning scheme derived from the coating problem near the boundary of the object has been removed to illustrate a simple procedure for fitting the contour to the scanning scheme derived from the coating problem.

[0064] [Figure 15] This diagram illustrates the selection of scanning contours and relative positions of each layer for the same scanning method derived from the covering problem shown in Figure 7. This diagram simplifies the layering method in a system where the net displacement of each layer is constant, by redefining the unit cell shown in Figure 7b to include more objects. The new vectors (b1, b2) relative to the unit cell, expressed in units of fragment thickness, are (3.18, 0.0) and (0.0, 4.0), which in this example are constant for a Euclidean coordinate system where the x-axis points to the right and the z-axis points upward. In the same units and coordinates, the width of the large trajectory (X-axis) is 1.84, and its overall height (Z-axis) is 1.79. The width of the small trajectory is 1.28, and its height is 1.31. In this scanning method, the net displacement of each fragment is constant (however, the material thickness of each fragment is not specified here). The x-axis spacing between larger scan trajectories is 1.59, and the x-axis spacings between smaller scan trajectories are 1.08 and 2.05, respectively. From this figure, it can be seen that different unit cells can be defined for the same covering scheme; however, in this particular solution, unit cells with a base smaller than 4 (one cross-section of two large trajectories and one cross-section of smaller trajectories that are mirror images of each other) cannot be defined.

[0065] [Figure 16] This is a tangential cross-sectional view of a formula derived from a solution to a coating problem with a unit cell, and includes two different objects generated from different AM trajectory geometric structure settings using the procedures of powder bed fusion, laser sintering, or binder injection. The solution to this coating consists of only two trajectories, which are the smallest reference unit cells that the unit cell can take. The reference vector (b1, b2) is constant at (1.98, 0.0) and (1.0, 1.0) when described in units of the fragment thickness, and in the same units, the width of the large trajectory is 1.46 and the height is 1.73, and the width and height of the small trajectory is 1.00. In this figure, the X-axis direction is to the right and the Z-axis direction is upward. The scanning interval between trajectories is 0.96, so the interval between the large trajectories is 1.92 and the interval between the small trajectories is 1.92. Three layers of trajectories are shown, and all of these trajectories are parallel in cross-section. However, it should be noted that in order to slightly change the surface height of the scanning trajectory printed on each layer, it is desirable to rotate the trajectory layer by layer, every two layers, or every n layers.

[0066] [Figure 17] (a) is a plan view of a scanning scheme using the covering problem solution unit cell shown in Figure 16. Layer n consists of a linear scanning path that alternately uses different AM trajectory geometric structure settings. The trajectories of adjacent scans do not necessarily have to be printed in chronological order.

[0067] (b) is a plan view of a scanning method using the covering problem solution unit cell shown in Figure 16, where rotation occurs between consecutive layers. Layer n+1, as shown in Figure 17a, consists of the scanning path and AM trajectory geometric structure settings, similar to the previous layer n, and is rotated in the plane around the z axis relative to layer n.

[0068] [Figure 18] (a) is a plan view of layer n, which is the circular scan strategy in the covering problem solution unit cell shown in Figure 16. The scan path of the set of design values ​​for the geometric structure of the alternating AM trajectories is circularly symmetric with respect to the displayed points.

[0069] (b) is a plan view of layer n+1, which is continuous with layer n shown in Figure 18a, and represents the circular scan strategy in the covering problem solution unit cell shown in Figure 16. The continuous layer n+1 consists of a scan path that overlaps with the previous layer n, but alternates with the design values ​​of the geometric structure of the AM trajectory. The scan path is circularly symmetric with respect to the displayed point and, together with the symmetric point in the previous layer n, constitutes the longitude axis of the cylindrical coordinate system.

[0070] (c) is a plan view of layer n+1 and layer n+2, which are continuous with layer n, as shown in Figures 18a and 18b, respectively, and represents the circular scan strategy in the covering problem solution unit cell shown in Figure 16. The scan path is cylindrically symmetric with respect to the previous layers n and n+1, the design value of the geometric structure of the AM trajectory matches that of layer n, and is alternating with respect to layer n+1.

[0071] (d) is a path-tangent cross-section of three consecutive layers (layers n, n+1, and n+2 shown in Figures 18a-c), along the cross-sectional line shown in Figure 18c, illustrating the application of the solution to the covering problem shown in Figure 16 through the rotation of a unit cell around the longitude axis of the cylindrical coordinate system. However, it should be noted that the scanning path does not need to be circularly symmetric with respect to the longitude axis, and the point of symmetry can be translated between layers, so it is desirable that the surface height of the scanning trajectory printed on each layer changes slightly. Furthermore, the scanning path can be drawn along any contour as long as the paths are approximately parallel.

[0072] [Figure 19] This figure shows the scanning trajectory in a scanning method involving a rotation of approximately 90 degrees every two layers, and is derived from the solution to the coating problem shown in Figure 7b. The rotation angle shown in the figure is approximately 90 degrees, but it can be set to suit the desired properties of the printed molded product, to improve the processing rate at a given porosity, or for other reasons. Furthermore, the rotation does not have to be performed every two layers, but may be performed every layer or every n layers. Moreover, the trajectory within each layer does not have to be a straight line; it is shown that way in this figure only for the sake of simplicity, and may consist of a path parallel to any direction along the surface of the layer.

[0073] [Figure 20] (a) shows scaling performed on a unit cell based on the solution to the coating problem shown in Figure 7, thereby allowing some degree of control over the scanning path through this scaling (and other operations), and thus control over the printing speed (printing press load) of the printed molded product, as well as its physical properties and quality, as measured for example with respect to the porosity of the molded product. The unit cell is reduced in the direction of the hatch spacing, which is achieved by reducing the hatch spacing while fixing the AM trajectory geometric structure settings. This is a similar (angle-preserving) scaling within the unit cell, with a coefficient of 0.87 in the X-axis direction (rightward), and the ratio of the fragment thickness to the hatch spacing, the angle between the centers of the objects, and the ratio of the fragment thickness to the width of the trajectory are also similar, under the assumption that the fragment thickness is constant, as in Figure 15. Other scaling and distortion operations are, of course, also possible. Generally speaking, in binder spraying, powder bed fusion bonding, laser sintering, or similar procedures, increasing the degree of trajectory overlap compared to the trajectory shown in Figure 7b may reduce the porosity and printing speed of the printed molded product to a certain point, after which the porosity may begin to increase again. In material spraying, fused deposition modeling, or similar procedures, a slight reduction in trajectory overlap compared to the trajectory shown in Figure 7b may be necessary to reduce the porosity, but as the overlap decreases further, it eventually disappears completely, and the porosity increases.

[0074] (b) is the unit cell shown in FIG. 20a, showing the change in the AM trajectory geometric structure setting value designed to increase the overlap, which may be useful in reducing the porosity of a shaped object printed by binder jetting, powder bed melt binding, laser sintering, or similar procedures. In this example, as a result of the printing speed (AM trajectory geometric structure setting value) decreasing for small objects, such objects become 17.5% deeper and wider, the total area increases by 38%, and these small objects are repositioned somewhat lower within the unit cell, so that when the printing speed is lower, the shaping by the trajectory is more accurately reflected. By applying other such changes (and changes in position) of the AM trajectory geometric structure setting value in a generally similar manner, the porosity, printing throughput, and other printing shaped object and printing characteristics increase or decrease. The unit cell shown in this figure includes fragments and scanning paths perpendicular to the plane of the layer as in FIG. 7d and a rotation for every two layers as in FIG. 19.

[0075]

Figure 21

[0076] [Figure 22] This is a cross-sectional view of a single-layer path tangent trajectory of a stainless steel part printed by laser powder bed fusion, with a laser power of 250 W, a hatch spacing of 140 μm, a Gaussian beam waist of 55 μm, and a laser scanning speed of 1046 mm / s. This single layer is representative of a single layer in a scanning scheme of single-trajectory geometry; in such schemes in general, the direction of the scanning path of successive layers rotates relative to the previous layer, but this is not shown here.

[0077] [Figure 23] This is a two-layer path-tangent trajectory cross-section of a stainless steel part printed by laser powder bed fusion, with a laser power of 250W, a hatch spacing of 140μm, a Gaussian beam waist of 55μm, and a laser scanning speed of 1210 mm / s. This is an implementation of a specific "interleaved" scanning method described in U.S. Patents 6,596,224 and 6,677,554, and illustrated in Figure 4. The path-tangent cross-sections of the trajectories for each scan are similar in the top layer (second layer) and the layer below it (first layer). Note that the area of ​​the first layer trajectory cross-section covered by the second layer trajectory cross-section is no longer visible; this is a common feature of cross-sectional photographs where trajectory overlap occurs.

[0078] [Figure 24] Figure 16 is a photograph of a single-layer path-tangential path cross-section of a stainless steel part printed by laser powder bed fusion, based on the solution to the coating problem detailed in Figure 16. In this case, two different region / geometric structure path cross-sections are created using different scanning methods. A small-width path was printed with a laser power of 250 W, a Gaussian beam waist of 55 μm, and a laser scanning speed of 2038 mm / s, while a large-width path was printed with a laser power of 250 W, a Gaussian beam waist of 55 μm, and a laser scanning speed of 860 mm / s, with the distance between the centers of the small and large paths fixed at 140 μm.

[0079] [Figure 25] This is a photograph of a two-layer path-tangential trajectory cross-section of a stainless steel part, printed by laser powder bed fusion based on a solution to a coating problem, utilizing two types of trajectory geometric structures. In this example, the initial layer of the wider width scanning trajectory, separated by a 140 μm hatch spacing, is printed from 50 μm thick powder, with a laser power of 250 W and a Gaussian beam waist of 55 μm. A second layer with a net displacement and powder thickness of 50 μm is used to form the narrower width scanning trajectory, where the center point of the cross-section of the first layer is displaced by approximately 70 μm relative to the center point of the cross-section of the wider width scanning trajectory; using the same settings, scanning becomes 1.6 times faster.

[0080] [Figure 26] This is a photograph of a cross-section of the scanning trajectory, with a unit normal tangent to half of the trajectory and perpendicular to the other half, of a part printed on stainless steel by laser powder bed fusion, using a repeating technique of the two types of trajectory geometric structures shown in Figure 25. In this case, after forming two layers as shown in Figure 25, the scanning trajectory is rotated 90 degrees, and the next two layers are formed again in the same manner as shown in Figure 25. In addition to forming these two layers, the 90-degree rotation is repeated as many times as necessary.

[0081] [Figure 27] (a) is a photograph of a trajectory cross-section at a 45-degree angle from the unit normal, showing a portion of aluminum alloy printed using a scanning method with two types of trajectory geometric structures similar to those shown in Figure 26. In this example, the difference is that the laser output is 375W and the hatch spacing is 220μm, so that smaller trajectories are printed with a smaller beam waist and faster scanning speed than those for larger trajectories.

[0082] (b) is a photograph of a trajectory cross-section at a 45-degree angle from the unit normal, showing a portion of aluminum alloy printed using a single-trajectory scanning method, with a laser output of 375W, a Gaussian beam waist of 46μm, a parallel line spacing (hatch spacing) of 220μm, and a scanning speed of 1311 mm / s. In this method, the scanning path rotates 90 degrees for each layer.

[0083] [Figure 28] (a) is a cross-sectional photograph of a portion printed with nickel superalloy by laser powder bed fusion, using a scanning method with two types of trajectories similar to that described in Figure 26. In this example, a laser power of 269 W, a hatch spacing of 140 μm, a net displacement and powder thickness of 50 μm, and a Gaussian beam waist of 55 μm were used, with the scanning speed used to form larger trajectories being twice the speed used to form smaller trajectories. The measured porosity of this portion was 0.12%, which was much lower than the measured porosity of the portions shown in Figures 28b and 28c, although these were printed using the same total deposition energy per unit mass as the portions shown here.

[0084] (b) is a cross-sectional image of a portion printed with nickel superalloy by laser powder bed fusion, using a single-trajectory "interleaved" scanning method similar to that described in Figure 23. In this example, a laser power of 269 W, a hatch spacing of 140 μm, a net displacement and powder thickness of 50 μm, and a Gaussian beam waist of 55 μm were used to form the trajectory, but the total deposition energy per unit mass is the same as that shown in Figures 28a and 28c. The measured porosity of this portion is 0.50%, which is lower than that of the portion shown in Figure 28c, but much higher than that of the portion shown in Figure 28a.

[0085] (c) is a cross-sectional view of a portion printed with nickel superalloy by laser powder bed fusion, using a single-trajectory scanning method similar to that described in Figure 8. In this example, a laser power of 269 W, a hatch spacing of 140 μm, a net displacement and powder thickness of 50 μm, and a Gaussian beam waist of 55 μm were used to form the trajectory, but the total deposition energy per unit mass was the same as that of the portions shown in Figures 28a and 28b. The porosity measured for this portion was 0.69%, which was higher than that measured for the portions shown in both Figures 28a and 28b.

[0086] [Figure 29] (a) is a diagram showing a unit cell consisting of the geometric structure of three distinct objects, formed with distinct AM trajectory geometric structure settings that are most consistent with the printing procedure of powder bed fusion, laser sintering, or binder spraying. In this example, each unit cell contains one large, two medium, and three small objects, which is the smallest possible base. In this configuration, several fragment thicknesses can be used; as a result, the basis vectors (b1, b2) are (1.0, 0.0) and (0.44, 1.03) for the length of the vector parallel to the x-axis. The x-axis is to the right and the z-axis is upward.

[0087] (b) is a single-layer diagram with subsequent layers of material deposited, where the net displacement ND2 is less than the material thickness MT2 in this fragment. The scanning method used is designed for the solution of a coating problem based on the geometric structure and size of a multi-trajectory object shown in Figure 29a, and includes the alternating order of trajectory printing, the differing net displacements between fragments, and the fact that the thickness of multiple materials is greater than or equal to the net displacement within a single fragment in each fragment. The label "1.1" indicates that the smaller trajectory is printed first, followed immediately by the label "1.2" for the larger trajectory. This method can be applied to any printing procedure with some modifications, but in this example, the geometric structure of the trajectory cross-section is drawn to reflect trajectories generated using AM trajectory geometry settings for procedures such as powder bed fusion, laser sintering, or binder injection, in which case, expanding the unit cell beyond what is shown in Figure 32 is advantageous in preventing excessive binder deposition.

[0088] (c) is a diagram of the printed layer after the material deposition shown in Figure 29b, and also a diagram of the material deposition for printing the next layer. In this example, a small trajectory (labeled "2.1") is printed first, followed by a larger trajectory (labeled "2.2"). For the next layer, the material thickness MT3 and the net displacement ND3 are the same.

[0089] (d) is a diagram of the layers printed after the deposition of the material shown in Figure 29c. Similar to the previous layer in this example, the smaller trajectory (labeled "3.1") is printed before the larger trajectory (labeled "3.2"). In the case of molded products using this method, the x-axis component of the shear strength is increased compared to the scanning method where one trajectory is placed on top of the other, because the z-axis crossover of the trajectory within and across the layers is large. This method is considered well suited to binder spraying, where the order in which smaller or larger objects within each layer are printed in reverse is expected to result in higher yield strength along the scanning direction (towards the page) and relatively high shear strength in and around the x-axis. Incorporating rotation with each layer increases the yield strength and shear strength in the x-axis direction, while decreasing the yield strength in the scanning direction.

[0090] [Figure 30] (a) is a cross-sectional photograph of a portion printed on stainless steel by laser powder bed fusion, using a single-trajectory scanning method similar to that described in Figure 8. In this example, a laser output of 260 W, a parallel line spacing (hatch spacing) of 140 μm, a net displacement and powder thickness of 50 μm, a Gaussian beam waist of 91 μm, and a scanning speed of 1004 mm / s were used to form the trajectory. The measured porosity of this portion was 0.48%, which was higher than the measured porosity of the portion shown in Figure 30b.

[0091] (b) is a cross-sectional photograph of a portion printed on stainless steel by laser powder bed fusion, using a single-trajectory scanning method similar to that described in Figure 8. In this example, a laser power of 260 W, a parallel line spacing (hatch spacing) of 140 μm, a net displacement of 50 μm, a powder thickness of 60 μm, a Gaussian beam waist of 91 μm, and a scanning speed of 1004 mm / s were used to form the trajectory. The measured porosity of this portion was 0.21%, which was considerably lower than the measured porosity of the portion shown in Figure 30a. The decrease in porosity is attributed to the increase in powder thickness relative to the net displacement.

[0092] [Figure 31] (a) is a cross-sectional photograph of a part printed with aluminum alloy by laser powder bed fusion, using the same two-trajectory scanning method as described in Figure 27. In this example, a laser output of 650 W, a hatch spacing of 200 μm, and a net displacement and powder thickness of 50 μm were used. To form large trajectories, the scanning speed was set to approximately 1.9 times the speed for forming small trajectories, and the Gaussian beam waist for forming large trajectories was set to approximately twice the size of that for forming small trajectories. The porosity measured in this part was 3.29%, which was much higher than the porosity measured in the part shown in Figure 31b, but these were printed using the same total deposition energy per unit mass as the part shown here.

[0093] (b) is a cross-sectional photograph of the portion printed with aluminum alloy by laser powder bed fusion, using the same two-trajectory scanning method as described in Figure 31a. In this example, a powder thickness of 60 μm was applied to this portion instead of 50 μm. The measured porosity of this portion was 1.79%, which was considerably lower than the measured porosity of the portion shown in Figure 31a. The decrease in porosity is attributable to the increase in powder thickness relative to the net displacement.

[0094] [Figure 32] (a) is a cross-sectional photograph of a portion printed on stainless steel by laser powder bed fusion, using the same two-trajectory scanning method as described in Figure 26. In this example, a laser output of 252W, a hatch spacing of 140μm, a net displacement and powder thickness of 50μm, and a Gaussian beam waist of 55μm were used, with the scanning speed for forming large trajectories being approximately 1.6 times that for forming small trajectories. The measured porosity of this portion was 0.23%, which was much lower than the measured porosity of the portions shown in Figures 32b and 32c, although these were printed using the same total deposition energy per unit mass as the portions shown here.

[0095] (b) is a cross-sectional photograph of a portion printed on stainless steel by laser powder bed fusion, using a single-trajectory "interleaved" scanning method similar to that described in Figure 23. In this example, the trajectory was formed with a laser power of 252 W, a hatch spacing of 140 μm, a net displacement and powder thickness of 50 μm, and a Gaussian beam waist of 55 μm, but the total deposition energy per unit mass was the same as that shown in Figures 32a and 32c. The measured porosity of this portion was 0.56%, which is lower than that of the portion shown in Figure 32c, but much higher than that of the portion shown in Figure 32a.

[0096] (c) is a cross-sectional photograph of a portion printed on stainless steel by laser powder bed fusion, using the same single-trajectory scanning method as described in Figure 8. In this example, a laser power of 252 W, a hatch spacing of 140 μm, a net displacement and powder thickness of 50 μm, and a Gaussian beam waist of 55 μm were used to form the trajectory, but the total deposition energy per unit mass was the same as that of the portions shown in Figures 32a and 32b. The porosity measured for this portion was 1.13%, which was much higher than the porosity measured for the portions shown in both Figures 32a and 32b. [Modes for carrying out the invention]

[0097] Description of the invention:

[0098] The system, procedure, and composition described here will explain how, in principle, by changing the shape, size, and position of the trajectory and trajectory cross-section within a single layer or across different layers, the speed and throughput of the printing press will be increased (reducing printing costs), the porosity of the processed molded product will be reduced, the strength and other necessary physical properties of the processed molded product will be improved, and the reproducibility of the print will be enhanced. The invention intended herein is applicable to AM or 3D printing procedures in which material and / or scanning path trajectories are generated or used, or surface height changes in a layer-by-layer manner as if appearing by a single continuous trajectory, including, but not limited to, fused deposition, material jetting, binder jetting, powder bed fusion, selective laser sintering, multi-jet fusion, directed energy deposition, direct metal deposition, electron beam additive manufacturing, and arc plasma sintering additive manufacturing. Furthermore, the intended use of the invention herein is not limited to single or few materials used in printing procedures and presses based on the aforementioned trajectories and scanning paths or in similar procedures and presses that generate layer-by-layer surface changes; rather, the invention is considered convenient for all or almost all uses of such materials, and also includes the use of composite materials and printing of multiple materials.

[0099] The inventions described herein are more efficient than other methods in terms of the use of deposition energy and / or the time required for deposition, curing, melting, bonding, hardening, sintering, or fusing of materials when using a target scanning method and material deposition method by changing the AM trajectory geometric structure setting value {P} and the feedstock parameter setting value {f}, and the compositions produced by these methods have superior physical properties, including, but not limited to, increases in ductility, strength, isotropy (anisotropy), fatigue resistance, thermal conductivity, electrical conductivity, surface roughness, and others. The AM trajectory geometry setting {P} includes, but is not limited to, the scanning speed of energy deposition along the trajectory path or the scanning speed of the print head depositing material and / or energy, the total amount of energy deposited per unit time or the total mass of material deposited per unit time, the average gradient of the surface on which the scanning trajectory is generated, setting values ​​for controlling the thickness of the powder or material and the volume of material deposited, net displacement, setting values ​​for controlling the compression of the material, the area of ​​the region where energy or material is deposited when generating the trajectory, the shape of the region where energy or material is deposited or the shape of the nozzle, the temperature of the area where printing is performed or the build plate, the temperature at which the raw material is heated before deposition, the velocity of the gas flowing over the surface of the build object, and many others. The feedstock parameters {f} include, but are not limited to, the following: the viscosity of the material depending on temperature, the melting temperature of the material, the flow properties of the material (e.g., shear thinning or shear thickening), the packing density of powdered materials (e.g., tap packing density), the amount of particles in suspension, the particle size, particle shape and surface roughness of powders and particle suspensions, the specific heat of the material, the plasticity of the material, and many others.

[0100] The scanning path describes the energy deposition or material movement or the movement of the print head for energy deposition, and these include the local curvature or angle of the trajectory (with respect to both preceding and succeeding layers and the global coordinate system), the order in which the trajectory is printed, the parallel line spacing, and related directional variables, but it is important to note that these are not included in the set of AM trajectory geometry setting values ​​{P} or raw material setting values ​​{f}. This is because these components do not tend to significantly alter the setting values ​​of the trajectory (or the cross-section of the trajectory), but rather merely indicate where the trajectory is formed. Points to note in some printing procedures, including but not limited to powder bed fusion and selective laser sintering, are that the parallel line spacing within the layers and the order in which the trajectory is printed have some influence on the trajectory geometry, due to the formation of erosion zones and other phenomena; in general, changes in these setting values ​​over the phase spacing of the parallel lines involved in filling tend to induce relatively small changes in the trajectory geometry, if measurable within the error.

[0101] Summary of key differences from the prior art. The following differences are not an exhaustive list, but rather a summary of some of the key points where the present invention differs from the prior art.

[0102] Additive manufacturing using geometric structures based on trajectories is almost always performed layer by layer. By arbitrarily designing the scanning method within a given layer, the desired physical and geometric properties can be given to that layer. However, these are not designed to match the scanning method of preceding or succeeding layers, except when it is necessary to consider the rotation of the entire path direction or the path near the end of the molded work. Even in print lattices, the scanning methods used to form struts and nodes are generally formulated based on information obtained only from data related to that layer, and not on information about the scanning methods of preceding or succeeding layers. This method is particularly suitable for parallel computing (simultaneously calculating the scanning path of each layer) when generating scanning methods on a computer.

[0103] The interleaved layer filling strategies described in U.S. Patents 6,596,224 and 6,677,554 are exceptions, in which the scanning path direction in layer n is the same as the path in either the preceding layer (n-1) or the succeeding layer (n+1), and the position of the path in layer n is midway between the paths of the preceding and succeeding layers, as shown in Figure 4. However, in this case, as in other cases, multiple AM ​​trajectory geometry settings are not used within the same filling strategy; rather, the same AM trajectory geometry settings are used across multiple layers, even if the position of the scanning path in each layer depends on the path in a higher or lower layer.

[0104] From a crystallographic perspective, a scanning method using a single set of AM trajectory geometry settings periodically replicates a unit cell containing only one object. Such a method can be mathematically described by a lattice with base 1. It is well known to those familiar with crystallography that there are only five such lattices in two dimensions and only 14 in three dimensions. In the present invention relating to coating and scanning methods, any scanning method or printed composition that can be described by a lattice with base 1 with respect to the base of the AM trajectory geometry settings is excluded; rather, since the present invention requires the application of different AM trajectory geometry settings within the same method, a unit cell with a base greater than 1 is necessary.

[0105] Specifically, in the present invention relating to coating and scanning methods and compositions, the following methods are required: 1) different AM trajectory geometric structures are generated, typically within a single layer or across multiple consecutive layers, by using different AM trajectory geometric structure settings in trajectories or trajectory cross-sections; 2) the features of trajectories or trajectory cross-sections are defined by unit cells, depending on the associated AM trajectory geometric structure settings or trajectory geometric structures, where the unit cells are described in Euclidean space or other geometric space and have two or more bases; 3) the printed trajectories within a unit cell cover the space of the cell in the sense that the area (volume) of the unprinted material is not greater than the maximum trajectory cross-section (maximum trajectory); and 4) at least two adjacent unit cell portions or all of them are present in each layer within the region to which the method is applied.

[0106] The existence of elements in at least two adjacent unit cells means excluding the application of different single-base grid schemes in different regions within a single layer or across multiple layers during printing. Common prior art uses different scanning schemes in different regions within a layer, for example, a printing grid scheme used in the center of the layer, a contour scanning scheme used in the outer edges of the layer, and a filling scheme used in between. A single unit cell with a large base can describe its layer or substantial portion (portion), and any similar preceding or succeeding layer; however, such elements for at least two cells would not exist within each layer. The trajectory within the unit cell requires the formation of a cohesive solid within the cell's region, and the absence of non-printed material with the largest or larger trajectory, and this is intended to restrict unit cells in a printing grid or similar periodic structure where the majority or main portion of the material is not printed. Broadly speaking, the present invention is applied to filling methods, including contour filling, where the contour consists of printed grids, skins, and other such printed structures having struts that do not consist of adjacent scan trajectories or single scan lines.

[0107] Furthermore, the present invention relating to coating and scanning methods includes the use of AM trajectory geometric structure settings in either the form of print settings or the arrangement of trajectories in space, for application to non-periodic or quasicrystal materials, within a filling region of the print space.

[0108] Regarding layer thickness, prior art does not distinguish between fragment thickness, material thickness, and net displacement, and where such distinctions are made (if any), they are considered to have a negative or neutral effect on the quality of the processed molded product, the printing press speed, or other printing or molded product characteristics. The invention described herein describes the perspective of additive manufacturing, in which these concepts are considered distinct, and describes the method of forming layers, in which some or all of these three are intentionally and controlledly staggered, often resulting in superior printing results. In particular, the material thickness and net displacement are used as independent AM trajectory geometric structure settings, and since these settings are changed independently by the method described herein, superior printing results are achieved.

[0109] <Scanning paths derived from multiple trajectory cross-sectional shapes and / or sizes.> With regard to melting, sintering, fusing, or otherwise binding powder particles to one another (hereinafter, unless otherwise specified, all of these shall be included in "fusing" or "fused"), and with regard to solidifying or hardening (hereinafter, unless otherwise specified, all of these shall refer to "fusing" or "fused" for convenience unless otherwise specified) suspensions, slurries, liquids, gels, polymerizing polymers or other media (hereinafter, unless otherwise specified, all of these shall be included in the term "slurry") by directed energy, binders, or material deposition, the re-melting of particles or slurries that melted during the processing of a preceding fragment. Re-sintering, re-fusing, or re-binding (hereinafter, unless otherwise specified, all of these are included in "re-fusing" or "re-fusing") is necessary to produce molded products with low porosity, or to ensure that no cavities consisting of unmelted, unsintered, unfused, unhardened, unsolidified, or unbound (hereinafter, unless otherwise specified, all of these are included in "unfused") particles or slurries remain after printing. There are at least two general reasons why this is necessary. The first reason is that scanning or trajectory-based printing procedures are generally designed to generate trajectories that do not tile the space, and in scanning methods that do not employ sufficiently overlapping trajectories, overlapping trajectories in the intersection regions of multiple trajectories result in re-fusion, creating voids in the space between trajectories where unfused particles or slurry, or particles or slurry are absent or partially depleted, are created.The second general reason is that scanning and trajectory-based printing generate trajectories whose local geometric structures change probabilistically, meaning that the trajectory width, height, depth, and surface structure vary substantially along the length of the trajectory, and the degree of this probabilistic variation depends on both controlled and uncontrolled variables, including the printing procedure used, printing settings, raw material settings, and other variables. As a result, if the scanning method is not designed with sufficient overlap, local changes in trajectory shape near the trajectory boundary at specific points along the trajectory length can lead to a lack of particle or slurry fusion.

[0110] Overlap is necessary to reduce the volume of unfused spaces within the molded product and to form areas with low porosity and almost no voids. However, overlap is inefficient, as energy or time is spent re-fusing particles or slurries that have already been fused. Therefore, a scanning method that reduces or eliminates areas of overlap that are not necessary to ensure the fusion of particles or slurries (excess overlap) is more efficient in terms of energy and time utilization. In summary, compared to using a trajectory or trajectory cross-section formed by only one geometric structure, a scanning method that uses trajectories or trajectory cross-sections with different geometric structures (size, shape, surface condition) can reduce excessive overlap.

[0111] An important point to note is that when generating fused trajectories considering the AM trajectory geometric structure settings and the space of the raw material (powder or slurry), neither the average geometric structure nor the average volume has a precisely linear relationship with the amount of energy or time spent in forming the trajectory. Therefore, for each printing press, the scalar trajectory efficiency function {Tj(P jIt is effective to define the combination of {P}, and this definition is made in units of trajectory volume printed per unit time, trajectory volume printed per unit energy, trajectory cross-sectional area printed per unit time, or trajectory cross-sectional area printed per unit volume, and within this, there are different combinations of functions for each printing press, which relate to the raw material (powder or slurry) represented by "j", and also to the AM trajectory geometric structure setting value specific to the capabilities of the printing press and the raw material {P j i},i=1...I j , the number of raw materials unique to “I j This is based on the following. In summary, the characteristic of size representing the geometric structure of the trajectory or trajectory cross-section is determined by a predetermined combination of printing equipment and raw materials, and these characteristics also depend on the AM trajectory geometric structure setting values. For an example of defining the size characteristic as a function, please refer to the journal article: C. Kamath, B. El-dasher, GF Gallegos, and WE King, and A. Sisto, Density of additively-manufactured, 316L SS parts using laser powder-bed fusion at powers up to 400W, Int. J. of Adv. Manuf. Technol. 74, pp. 65-78. In 2014, the trajectory of 316 stainless steel was formed by changing AM trajectory geometric structure settings such as laser speed and laser power. Other fixed settings included a layer thickness of 30 microns and a laser spot size of approximately 63 microns. Within this, the trajectory height, trajectory width, and trajectory depth were measured as two variable settings.

[0112] The efficiency function, which relies on the AM trajectory geometry settings described above, allows for a direct comparison of single-layer or multi-layer scanning methods, expressed in three dimensions with respect to trajectory overlap or in two dimensions with respect to trajectory cross-section, based on energy or time efficiency. Furthermore, since excessive overlap is undesirable in terms of efficiency, determining specific combinations of printing equipment, raw materials, and AM trajectory geometry settings to generate one or more trajectory geometry structures, thereby improving the solution to the coating problem for these trajectory geometry structures, tends to lead to more time and energy efficient scanning methods. Moreover, for various methods with higher density for filling objects, where the geometry of such trajectories is identical or similar to that used in the coating problem solution, the density in such filling methods tends to positively correlate with the time and energy efficiency of the scanning methods. Figure 7 shows an example of an efficient scanning method based on the solution to the coating problem derived from high-density two-component filling of a disk. In this example, the cross-sectional area of ​​the object's trajectory is an approximation of the cross-sectional area of ​​the trajectory generated using laser powder bed fusion.

[0113] Considering the specified printing press and the “J” supply material (where an integer J>1 represents printing with multiple materials or substances), and a set of set values ​​{P j i} k , k=1...K j Each numerical value in “K” j For the function value Tj({P j i} k ) and lattice basis vector b and volume v U (b) and ΣK j It can be used as a weighting in a coverage problem that occurs in a variable unit cell having a base of objects counted as individual trajectories or trajectory cross-sections (where the sum Σ is j = 1...J), in which each object of the trajectory or trajectory cross-section has a set value {P j i} k Each object O has a corresponding geometric structure. j,kBy identifying it using the index "j" of the raw material and the index "k" of the combination of set values, the coordinates r of the three-dimensional configuration of the object are determined. j,k and area or volume V j,k We can define the spatial arrangement r, where j,k The simplest position component corresponds to the scanning path of the material or energy source of the print head or other AM printing machine that deposits material and / or energy. This weighted coating problem can be described as a constrained optimization problem expressed by the following objective (function):

[0114] [Math 1]

[0115] Here, the three-dimensional arrangement of the object r j,k And minimizing the objective function for basis vector b is performed when the total volume of the unit cell defined by basis vector b is that of object O j,k The constraint is that it is covered by a multiple covering. The covering may also be double, triple, or more, in which a unit cell is covered an integer number of times by the object, and other constraints may be imposed as well. For example, the precise geometric structure of the object (and area or volume V j,k ) is other setting values ​​of other objects within the unit cell {P j} k or it may depend to some extent on their relative spatial arrangements; if such dependence on spatial arrangements is known, it may be effective to constrain the spatial arrangement of the object to the known topological space. However, it is important to note that even if such knowledge may lead to improvements in the scanning method, it is not necessary to know about the dependence on relative spatial arrangements (or other dependences) for the development of a viable scanning method.

[0116] It may be preferable to measure the average geometric structure of the trajectories of a given feed material in an environment similar to actual printing conditions. This method may include, but is not limited to, a) measuring the average geometric structure of trajectories, including trajectories printed on top of other trajectories, rather than trajectories where the material is deposited on a smooth surface; b) when multiple feed materials are used in printing, generating trajectories of individual feed materials printed on layers with composition ratios in a range similar to the target composition in the printed molded product, thereby printing with conditions for elastic stress, thermal conductivity, and, where applicable, electrical conductivity and other physical properties similar to those of the printed molded product; c) when the printing order of trajectories is not chronological due to adjacent trajectories, measuring the average geometric structure of the trajectories in the order in which they are used; and d) when trajectories with different average geometric structures exist in the printed molded product, a predetermined trajectory geometric structure setting value {P j} k When measuring the geometric structure of each combination of trajectories generated by the combinations, the measurement should be carried out in a manner as similar as possible to the state of the projected print, so as to be close to the geometric structure of other trajectories (with different "j" or "k" indices).

[0117] For example, when considering two trajectory shapes corresponding to the path-tangent trajectory cross-sections shown in the unit cell represented in Figure 7, it is advantageous to measure the average shape of a larger trajectory by printing these trajectories on top of a layer (or multiple layers) of smaller trajectories, and vice versa. This may be particularly relevant in powder bed fusion or selective laser sintering procedures; if the material does not completely melt during the printing of layer n, for example, if powder remains between the shown path-tangent cross-sections as seen in Figure 7d, the remaining material may still be sintered or partially sintered, thereby altering the physical properties of the material and, consequently, the trajectory shape of the trajectories in layer n+1. This also applies when the nearest adjacent trajectories are not printed in order, resulting in even-odd parity ("alternating trajectories") or in other orders. As an example of a general printing procedure, the shape of the trajectory may depend on the surface on which the material is deposited. In the example in Figure 7, the surfaces of the smaller and larger trajectories are different, which potentially alters the geometric structure of the trajectories printed on these surfaces, and further alters the trajectories of successive layers.

[0118] An important point to note is that in procedures where material is deposited on a trajectory, such as material spraying, fused deposition modeling, multi-jet fusion, arc plasma sintering, some binder spraying procedures, and similar procedures, the geometry of the trajectory of the printed molded product may, in some cases, depend heavily on the surface on which the trajectory is printed, and that surface is likely to contain fully or nearly solidified, hardened, or cured material from the printing of the trajectory in the previous layer. In binder spraying or similar procedures, if a slurry, liquid, solvent, polymer, suspension, or other flowable material penetrates a layer of microparticles by capillary action or adhesive force, forming a trajectory consisting of both the sprayed material and the microparticles, the bottom shape of the trajectory printed on top of the solidified, hardened, or polymerized material originating from the previously printed trajectory may coincide with the top of the previously printed trajectory. Such trajectory shapes, while printed with the same AM trajectory geometric structure settings, are substantially different from trajectories on a flat substrate with a deep (relative to trajectory height) layer of microparticles or a thinner layer of microparticles. In material jetting, multi-jet fusion, arc plasma sintering, fused deposition modeling, or similar procedures, the geometry of the surface on which the trajectory is printed can also affect the trajectory shape; in such cases, the degree of this effect depends more on the viscosity of the printed material than on the polymerization, drying, or curing time, with high-viscosity, fast-curing materials maintaining a more independent geometry, while low-viscosity, slow-curing materials tend to conform to the geometry of the underlying surface.

[0119] Nevertheless, the solution to the overlap coating problem has a trajectory shape obtained from printing on a fine particle layer (binder spraying) or a relatively smooth surface (material spraying, multi-jet fusion, arc plasma sintering, fused deposition modeling), as shown in Figure 7, and favorable effects can be obtained by using it in these material deposition methods. This is because the volume of the printed trajectory remains almost constant, which depends almost entirely on the AM trajectory geometric structure setting value, is independent of the geometry of the surface on which the trajectory is printed, and does not change even if the shape of the trajectory is changed. Therefore, for these printing procedures, the ratio of the volume (area) of the trajectory (trajectory cross-section) of the overlap to the total volume (total area) of the trajectory in a unit cell of the solution to the coating problem can be considered a measure of the excess amount of solution having binder and / or material that needs to be deposited to fill the cell space without voids.

[0120] Capillary action and adhesive forces during trajectory deposition allow for partial filling of the spaces between previously printed trajectories, but as mentioned above, the bottoms of the printed trajectories do not perfectly coincide with the surface on which they are printed. Consequently, designing scanning trajectories using a good solution to the coating problem that utilizes, but does not utilize, overlap between trajectories, still tends to result in a higher ratio of particles (e.g., powder) to solution in immature articles (green articles) with very low porosity in binder spraying and similar procedures, or a reduction in void space in material spraying, fused deposition modeling, arc plasma sintering, multi-jet fusion, and similar procedures.

[0121] In binder spraying, a sufficient solution containing the binder must be deposited to fill some of the spaces between particles, but not necessarily all of them. The amount of solution deposited for a given geometric trajectory can be directly proportional to the volume (surface area) of the trajectory (trajectory cross-section) in the solution of the coating problem. However, excess solution can cause problems during any binder removal step, which is part of the sintering process, as well as reducing the density of the microparticles in the immature article (green article), thus increasing shrinkage during sintering or increasing the porosity of the sintered part.

[0122] As a result, a scanning method that uses overlap is necessary (for the reasons mentioned above) to ensure that the spaces between particles are sufficiently filled, but if the scanning method is designed from a good solution to the coating problem, the overlap can be reduced to the minimum necessary, thereby reducing the shrinkage of the parts during sintering and / or the porosity of the sintered parts. In material jetting, fused deposition modeling, multi-jet fusion, arc plasma sintering, and similar procedures, overlap is again required, in this case helping to ensure a reduction in voids in the printed molded product (both the immature (green) and the sintered parts that are part of the procedure), but excessive overlap can cause other problems similar to those described for material overlap and / or binder jetting.

[0123] Figures 8 and 9 illustrate a clear and typical example of the problems that can arise when excessive overlap is used in the creation or use of scanning methods. Figure 8 is a cross-sectional image of a cube printed using a material injection procedure (fused deposition modeling) with thermoplastic polymers, where single-shape trajectories of layer n are arranged at 400 μm intervals, and in layer n+1, the same method and trajectory shape are used, but the trajectory is rotated 90 degrees relative to layer n. Figure 8 is an image of a printed portion using machine settings such as nozzle speed, temperature, and material extrusion speed, which are optimized to print the portion on the press without causing mechanical problems, produce a geometrically relatively accurate portion, and minimize surface roughness. With these settings, the porosity of the printed cube is approximately 8%. Figure 9 is a cross-sectional image of the same digital cube (the image is rotated 90 degrees relative to Figure 8), but printed with one setting change, namely an increase in the volume of extruded material, resulting in larger, and sometimes non-cylindrical trajectories. The printing failed because too much material was extruded, causing the nozzle to rub against the molded object. From a coating perspective, this can be seen as excessive, or "over-over-over" overlap. As can be seen in Figure 9, the edges and surface of the section deviated significantly from the specifications from both a geometric and surface roughness standpoint, and the section showed an average void ratio of 7% away from its edges.

[0124] These problems are mitigated when scanning trajectories of multiple geometric structures are used, as described in Examples 1 and 2. Using the same print settings as in Figure 9, except for using a scanning method that utilizes two trajectory geometric structures, a substantial reduction in porosity can be achieved without print failures, loss of geometric tolerances, or increased surface roughness. Figures 10 and 11 are cross-sectional images of the same digital cubes described in Examples 1 and 2 and shown in Figures 8 and 9, printed using two different scanning methods with two trajectory geometric structures, respectively. The trajectories shown in Figure 12 represent the trajectory geometric structures arranged in Figures 10 and 11 (however, in Figure 12, the volume extrusion rate of the material for each type of trajectory is proportionally reduced to produce an image in which the trajectory geometric structures are clearly identifiable). The porosities measured for the cubes shown in Figures 10 and 11 were 0.2% and 2.3%, respectively. The geometric accuracy and surface roughness of the cubes relative to the digital image were improved compared to the cube with 8% porosity shown in Figure 8, clearly demonstrating the advantages of the method using the geometric structure of multiple trajectories.

[0125] The coating optimization problem discussed so far can be extended, for example, so that the geometry and size of the objects, which depend on the AM trajectory geometry setpoint, are variable rather than fixed, and / or the number of objects in the base combination and the supplied raw materials can be changed. Such an extended problem reveals not only preferred or optimal scanning paths, but also preferred or optimal AM trajectory geometry setpoints and the relationship between the two. The extended problem also involves additional constraints, such as the capabilities of the printing press and the setpoints {P} of the particles or slurry to be fused according to their physical properties. j This may involve constraints that limit the range of materials, or constraints that fix the ratio of different materials in a problem involving multiple raw materials. In such problems, knowledge of the trajectory geometry of each supply material is important with various values ​​of the AM trajectory geometry setting.

[0126] Once an arbitrary solution to the problem is determined, a scanning method can be selected from that solution based on the object's geometry (trajectory or trajectory cross-section) and its position. To select the scanning method, the printer generates the trajectory using a combination of set values ​​{P j k} and among the solutions having the combination of said setting values, object O j,k It is necessary to print using the corresponding method. The scanning path is defined by where the print head must be positioned and / or where energy and / or material deposition must occur, thereby object O j,k The formation (or extrusion, in the case of a trajectory cross-section) of something approximating this can be carried out at that position in the solution. In simple terms, energy deposition can be defined as the center point of the object, but in the case of both energy deposition and material deposition, it can vary in a specific position, especially in the layering direction, and will be explained further below. As an example, it is necessary to form a trajectory by positioning the print head that deposits the material at a height of several trajectories on the surface of the molded object to be manufactured, which is shown as an example in Figure 2.

[0127] The solution can be extended over any size of space by duplicating unit cells to cover a desired space. This concept is demonstrated in the covering solutions shown in Figures 7 and 13. With respect to the outer surface of the molded object, the space covered by the extended solution can be bordered by determining the scanning of an arbitrary contour, meaning that the solution does not depend on the geometry of the molded object or fragments, as long as the lateral extent of the molded object or fragment within a given layer is at least the same size as one of the objects within the unit cell. This concept is shown in Figure 14.

[0128] Defining layers from the solution offers flexibility, as not all objects within a unit cell are formed within a single layer. More specifically, it is not necessary to define layers, and in some cases, the net displacement may differ between the upper and lower layers, which is the case for the spatial configuration r. j,kThis may be done by assigning a new layer to the spatial position of the object that has fallen in with the same displacement in the stacking direction for all objects having the same displacement. Examples of layer definitions for the covering solution shown in Figures 7 and 13 enable a constant fragment thickness scanning method (where the AM trajectory geometric structure setting value represents a constant net displacement), as shown in Figure 15. In this example, the implementation of the scanning method is simplified by redefining the unit cell; the covering solution is not changed.

[0129] Furthermore, in the solution to the coating problem, the scanning trajectories within consecutive layers do not need to follow the orientation suggested by the object in the trajectory cross-section. For example, Figure 16 depicts the path-tangent cross-section of a scanning method derived from the solution to the coating problem, showing a unit cell containing two different objects generated from different AM trajectory geometric structure settings using a powder layer fusion or laser sintering procedure, and showing three layers of trajectories, all of which are parallel in the cross-sectional plane. However, in this case, it is desirable to rotate each layer, every two layers, or every n layers, for reasons such as small variation in the surface height of the scanning trajectories printed on each layer. Figure 17 shows a plan view of two consecutive layers where a rotation occurred in layer n+1 relative to layer n. For the solution to the coating problem shown in Figure 16, it is not further necessary to print adjacent scanning trajectories in chronological order; in many procedures, it is preferable to print the smaller trajectory in one layer first, and then the larger trajectory. Moreover, as shown in Figure 17, the trajectories do not need to be straight lines; any contour or path can be used. For example, Figures 18a-c show a method in which three layers of scanning are printed in a circularly symmetric manner around a central point, and Figure 18d is a path-tangent cross-sectional image of the three layers along the dashed line shown in Figure 18c, demonstrating that the method shown in Figures 18a-c, as shown in Figure 29, can be derived by actually using the solution to the same covering problem.

[0130] While it is sometimes necessary to define layers by selecting the angular direction of a unit cell, the angular direction of an object can be fixed according to the relative position of the print head and / or the direction of energy and / or material deposition, thereby fixing the angular direction of the unit cell. The angular direction of a cell is fixed in the case of covering problems solved by non-Euclidean geometric structures, or in other cases as well. Nevertheless, even when directional constraints are imposed in the solution of the problem, there may be choices in which layers a particular object is printed on, and this choice may differ for the same base object appearing in different cells.

[0131] What's important is the different geometric structures of objects O j,k This occurs when objects are displayed within the same layer, or when a particular layer is composed entirely of objects of a single geometric structure. In the example shown in Figure 13, the angular direction of the objects is fixed according to the laser position, and the unit cell most simply contains objects from three different layers, which are vertically stacked objects at the center of the cell (in Figure 15, five layers are defined to more easily define the scanning method); alternating layers contain only one type of object or two different types of objects in the trajectory cross-section; and smaller objects are objects exhibiting two types of trajectory cross-sections, which are mirror images, and in this example, they are due to different surface geometric structures on which they are printed.

[0132] In scanning methods that use multiple combinations of AM trajectory geometric structure settings to print objects, the trajectory direction can be rotated between layers or every 2, 3..., or n layers. As an example, Figure 19 shows a schematic diagram of the scanning trajectory in a scanning method that rotates approximately 90 degrees every two layers, which is derived from the solution to the coating problem shown in Figure 7b. The rotation angle can be selected to suit the desired properties of the printed object, to improve the processing rate while maintaining a predetermined porosity, or for other reasons. Furthermore, the trajectory does not have to be a straight line as shown in Figure 19, but may be a parallel path in any direction within the surface of the layer, including, for example, a circular pattern as shown in Figure 18.

[0133] An overall solution to a coating problem is not necessary to generate a preferred scanning method; a locally optimal solution, or a coating derived from a locally optimal solution, may be even more preferable depending on the target mechanical properties or other requirements of the molded product. For example, locally optimal solutions to coating problems often involve regions or points in space near the boundaries of an object that are coated by only one object. Because scanning and trajectory-based printing methods tend to generate trajectories that vary probabilistically in their local geometry, regions or points in space near or on the boundaries of an object that are coated only once are more likely to be regions or points where powder or slurry fusion failures may occur during printing. Therefore, it may be desirable to increase the overlap in these regions if fewer such fusion-deficient regions are desired, or similarly, if a lower porosity of the molded product is desired.

[0134] This is easily achieved by simply scaling the unit cell, or by a combination of scaling and slight readjustment of the object's position, or by increasing the area or volume of the object in a known way, thereby setting the trajectory geometric structure {P j i} k This can be done by changing the settings, or by all or a combination of these methods. Figure 20 shows such a technique for the example shown in Figure 7. In Figure 20a, the unit cell is reduced in the parallel spacing direction by decreasing the parallel spacing while keeping the AM trajectory geometry setting fixed. In Figure 20b, the printing speed (AM trajectory geometry setting) is lowered for smaller objects, resulting in such objects becoming deeper and wider, and these smaller objects are repositioned somewhat lower within the cell, thus more accurately reflecting the formation of a given trajectory with a predetermined AM trajectory geometry setting and object relative position.

[0135] Cells can be scaled like objects, and this is particularly effective when the geometric structure of the space in which the grid is defined is curved or scaled. Figure 21 is an example of a cross-sectional image of a printed trajectory, following a solution to a coating problem derived using unit cells in a curved and scaled space with a spherical geometric structure, using material jetting, fused deposition modeling, arc plasma sintering, multi-jet fusion, binder jetting, or similar procedures. In the figure, each unit cell is curved according to the radial distance r from the center point, and unit cells further from the center of the sphere have a larger area (or volume) than unit cells closer to the center point. This implies a spherical geometric structure similar to fragments or layers. In this case, the trajectory can be scaled in size according to the radial distance, similar to cells (or potentially) (requiring changes to the geometric structure setting of the AM trajectory), while keeping the size and geometry constant. Furthermore, the trajectory may follow a curved or straight contour, for example, around an axis passing through a central point and reaching its highest point, or along a straight line intersecting the page, but along the arc of the sphere's surface in the stacking direction.

[0136] Scanning methods that incorporate trajectories with average targets of different geometric structures are created, for example, by changing the printing press, AM trajectory geometry, feed material, and other settings, resulting in molded products with fundamentally different compositions than those produced using a single average target of geometric structure. From a materials perspective, scanning trajectories are diverse objects, and they tend to exhibit different crystallographic (or disordered and linked) structures and chemical compositions based on the vicinity of the trajectory surface, due to differences in cooling rate, molecular diffusion rate, polymer bonding properties, gas absorption rate, etc. Furthermore, because their surfaces are exposed to the atmosphere of the printing press (or near vacuum), the trajectory surface tends to form more oxides, nitrides, carbon compounds, and impurities compared to the center of the trajectory, which also occurs in relatively inert atmospheres, such as 99.99% N2, Ar, or other inert gases, and near vacuum atmospheres. As a result of these, and other differences, trajectories with different trajectory cross-sections result in differences in the mechanical, thermal, electrical, and other physical properties of the molded product.

[0137] The differences in microscopic and macroscopic structures generated by different scanning methods and trajectory geometry are visually shown in Figures 12 and 22-27 for various materials printed according to several procedures. The path tangential cross-sectional image in Figure 22 is of a single layer of steel printed using a laser powder bed fusion procedure, where the crystal grain and trajectory boundary structure was revealed by etching with acid. A laser power of 250 W, parallel line spacing of 140 μm, and laser scanning speed of 1046 mm / s were used, with a beam waist of approximately 55 μm, having a nearly Gaussian beam intensity profile, and formed by an F-theta lens. This can be compared to Figures 23–25, which are printed using the same powder and procedure but employ different scanning methods; Figure 23 is a two-layer image using the “interleaved” method described in U.S. Patents No. 6,596,224 and 6,677,554; Figure 24 is a single-layer image using a method based on the solution to the covering problem described in Figure 16; and Figure 25 is a two-layer image using a method that alternates between layers of larger and smaller trajectories, where the trajectories of each successive layer are printed almost between the trajectories of the preceding layer.

[0138] The direction and magnitude of the trajectory significantly affect the macroscopic properties of the molded product. For example, comparing a molded product with a trajectory that always has a scanning path in the X direction or negative X direction to a molded product with a trajectory that has a scanning path that rotates between the X, Y, negative X, and negative Y directions, the former molded product will differ from the latter in that it exhibits significantly different mechanical properties in the X direction compared to the Y direction, including, but not limited to, ductility and tensile strength. The same applies to thermal and electrical properties.

[0139] Scanning schemes incorporating trajectories with different average geometric structure targets may be designed to influence the mechanical, electrical, thermal, and other physical properties of a molded product. More generally, using such a scanning scheme can yield molded products with fundamentally different compositions from a metallurgical and microstructural standpoint, thereby influencing the physical properties of the molded product. Such variations in composition can be desirable. For example, in Figures 7, 16, and 29, using trajectories with trajectory cross-sections of varying heights for each layer results in a molded product with improved shear strength in the Y direction compared to a molded product with the same geometric structure trajectory. This is because, in the latter case, the clear contrast between layers in the Z direction creates shear surfaces between each layer, whereas in the former, the contrast is reduced and therefore absent.

[0140] As an example, Figure 26 shows a cross-sectional image of a section tangent to half of the trajectory and perpendicular to the other half, which is a portion printed with steel powder and has a trajectory setting similar to that which formed the trajectory shown in Figure 25. In this case, two layers are printed with parallel (or antiparallel) trajectories, with the first layer having a larger trajectory generated by a slower scanning speed and the second layer having a smaller trajectory generated by a faster scanning speed; in a similar manner, the trajectories are rotated 90 degrees relative to the first two layers and two more layers are printed. This procedure is repeated. Example 3a describes the difference in porosity, which is achieved by using three different scanning trajectory schemes, while printing with equal laser power, parallel line spacing, beam waist, and total deposited energy per unit mass, and shows that the porosity is substantially reduced by using a multi-trajectory geometric structure scheme, the microstructure of which is shown in Figure 26. Figure 32 consists of images of portions printed using the three different schemes, visually illustrating the difference in porosity. Example 3b illustrates the difference in mechanical properties between coupons printed according to the single-trajectory method described in U.S. Patents 6,596,224 and 6,677,554, and coupons printed according to the multi-trajectory method shown in Figures 25 and 26. Despite similar porosity between coupons printed using different methods, the use of a method with varying trajectory geometry results in a substantial increase in yield strength, ultimate tensile strength, and elongation at break. Figures 26 and 25 clearly show the composition of materials produced by the method with variable trajectory geometry, illustrating the differences in the macroscopic and microscopic structure of the material, compared to portions produced by the single-trajectory geometry method, and Example 3b clearly demonstrates the difference in mechanical properties between the variable trajectory geometry and the single-trajectory geometry.

[0141] Example 4 describes the reduction in porosity observed when printing with a powder layer melting machine, using a multi-trajectory method derived from the solution to the coating problem shown in Figure 16, and comparing it with printing using a single-trajectory method. Figure 27a is an image of a composition formed in an aluminum alloy using a method similar to the two-trajectory geometric structure method used in Figure F, and is compared with Figure 27b. It was formed using the same laser power, parallel line spacing, and total deposition energy per unit mass as the composition shown in Figure 27a, but the beam waist and scanning trajectory are different, further visually illustrating the differences in macroscopic and microscopic structures between the single-trajectory and multi-trajectory methods. Example 6 describes the advantages of using the multi-trajectory method shown in Figure 27a for mechanical properties and porosity in an aluminum alloy, comparing it with the single-trajectory method shown in Figure 27b. Figure 28 visually illustrates the differences in porosity, macroscopic structure, and microscopic structure between two single-trajectory methods and one multi-trajectory method for printing on nickel superalloys. Again, the laser power, parallel line spacing, beam waist, and total energy deposited per unit mass are the same, but the scanning trajectories are different. As shown in Figure 28, the difference in porosity in Example 5 is explained.

[0142] Considering multiple printing procedures, material sets, and single-trajectory and multi-trajectory scanning methods, manufacturing a composition using multiple trajectories offers various advantages compared to manufacturing using a single-trajectory method. Manufacturing a multi-trajectory composition by changing printing settings results in improved physical and manufacturing properties. For example, it allows for faster manufacturing with equivalent porosity, increasing machine processing capacity, which is clearly a desirable method for manufacturing procedures based on various layers and trajectories.

[0143] <Using a different material thickness than the net displacement.> As mentioned above, material thickness and net displacement are generally considered to be the same concept, and together with the thickness of the fragment, they are called layer thickness. However, by properly controlling various printing presses, it is possible to create devices that can individually control the thickness of the material (powder) and the net displacement. There are many examples of using different material thicknesses for the net displacement, which may be advantageous for the properties of the molded product being manufactured, and these exist across various printing methods. Some such examples of preferred use are listed below, but this is not an exhaustive list.

[0144] Consider powders that are not fully solidified when deposited in a layer during binder spraying or a similar procedure. Due to factors involved in the consolidation of the binder-containing solution upon deposition, if the thickness of the deposited powder is greater than the net displacement, immature parts (green parts) with a greater proportion of powder to solution will occur. Such consolidation after deposition occurs, for example, when the surface tension of the solution tends to pull the powder particles toward the center of the trajectory as the solution is deposited on the powder particles to form a trajectory. Particle aggregation occurs due to surface tension pulling the particles toward the center of the trajectory during partial evaporation of the solution within the trajectory, or due to heating or other energy treatment of the trajectory, or for other reasons. If the amount of solution deposited within the trajectory is fixed and the thickness of the powder increases relative to the net displacement, the concentration of powder particles relative to the solution will increase in the powder layer containing particles that will further solidify after suspension.

[0145] In powder bed fusion or similar procedures, the trajectory geometry (and trajectory cross-section) depends on the powder thickness. This can also be the case in material spraying procedures, where the raw material hardens, solidifies, or hardens or solidifies upon exposure to an energy source or other material agent. Therefore, by changing the powder thickness, the trajectory geometry can be altered without significantly changing the printing speed. Furthermore, voids result from a lack of particles in specific areas of the layer, which may be due to adjacent trajectories, uneven lamination, or other reasons, but can be mitigated by simply adding powder to obtain a powder layer thicker than the net displacement. That is, by utilizing a powder layer thicker than the net displacement, pores that occur (probabilistically) due to uncontrolled variables in trajectory formation can be reduced.

[0146] Figure 29 is an explanatory diagram of a scanning method designed from a solution to a coating problem, which involves multiple geometric structures and sizes of objects in the trajectory, multiple net displacements, alternating the order of trajectory printing, and multiple material thicknesses greater than the net displacement within a single layer, depending on the layer. This method can be applied to any printing procedure with some modifications, in which case the geometric structure of the trajectory cross-section will reflect the trajectory formed with AM trajectory geometric structure settings used in powder bed fusion, laser sintering, or, if possible, binder spraying procedures.

[0147] In powder bed fusion bonding or similar procedures, a thinner powder layer relative to the net displacement may be preferable. The geometric characteristics of the trajectory (and trajectory cross-section) depend on the geometric structure of the AM trajectory and the set values ​​of the supplied raw materials. By using these set values, it is often possible to obtain a trajectory that is deeper and narrower in width compared to a powder layer with a thickness equal to the thickness of the net displacement by using a thinner powder layer.

[0148] When the spacing between parallel lines is sufficiently small, but interlayer fusion and re-fusion are insufficient to ensure that continuous layers are completely fused, it is preferable to increase the trajectory depth while decreasing the trajectory width in order to reduce the void ratio.

[0149] Figure 30 compares cross-sectional images of areas printed with iron using the same single-trajectory scanning method; in the area shown in Figure 30a, both the material thickness and net displacement are 50 μm, while in the area shown in Figure 30b, the net displacement is 50 μm, but the powder thickness is 60 μm. Improvements in porosity, macroscopic structure, and microscopic structure are clearly observed: Example 7 quantitatively explains the improvement in porosity. Figure 31 compares cross-sectional images of areas printed with aluminum using the same multi-trajectory scanning method; in the area shown in Figure 31a, both the material thickness and net displacement are 50 μm, while in the area shown in Figure 31b, the net displacement is 50 μm, but the powder thickness is 60 μm. Improvements in porosity, macroscopic structure, and microscopic structure are observed: Example 8 quantitatively explains the improvement in porosity.

[0150] example:

[0151] In all the following examples, the standard deviation used is calculated for multiple parts or coupons printed at different locations on the build plate during a single build, unless otherwise specified.

[0152] Example 1. In this embodiment, material ejection is deployed via a fused deposition modeling (FDM) machine having a single 400 μm nozzle and a thermoplastic polymer feed wire primarily composed of polylactic acid to print three 20 mm cubes, each with different print settings. Following the scheme shown in Figure 26, single-track scan strategies are used for two cubes, having the same machine settings except that each cube is printed with a different extrusion volume ratio setting. A multi-track scan strategy is used for one cube, having the same setting settings as the single-track strategy, except that the alternation between the two extrusion volume ratio settings is used to print a continuous trajectory, resulting in two separate scan trajectory geometric structures. In this example, only one of each type of cube is printed; therefore, the standard deviation of porosity is measured across multiple images taken at different locations within the cube. The standard deviation given for mass is a measure of the reproducibility of the precision scale.

[0153] Single-track cubes:

[0154] Figures 8 and 9 are cross-sectional images of single-track geometry cubes printed on a Prusa i3 machine using the following printer settings (and others not listed herein):

[0155] Floor temperature: 40℃ Extruder temperature: 190°C Fan output: 100% Feed rate (i.e., head movement speed): 1800 mm / second Parallel line spacing (hatch spacing): 400μm Nozzle diameter: 400 μm Layer thickness: 200μm Extrusion volume ratio of cube 1: 0.35 Extrusion volume ratio of cube 2: 0.52

[0156] The measured porosity of cube 1 was 8% ± 0.6%, with a total mass of 8.704 g ± 0.005 g; the porosity of cube 2 was 7% ± 1.3% away from the edges, with a total mass of 3.071 g ± 0.005 g. The main reason for this was that the printing was not completed due to a printing failure, and only about 36% of the "cube" was printed.

[0157] Multi-Track Cube:

[0158] Figure 11 is a cross-sectional view of a multi-track geometry cube printed on a Prusa i3 machine using the following printer settings (and others not listed herein):

[0159] Floor temperature: 40℃ Extruder temperature: 190°C Fan output: 100% Feed rate (i.e., head movement speed): 1800 mm / second Parallel line spacing (hatch spacing): 400μm Nozzle diameter: 400 μm Layer thickness: 200μm Extrusion volume ratio of the smaller trajectory: 0.31 Extrusion volume ratio of the larger trajectory: 0.63

[0160] The measured porosity of the multi-trajectory cube was 2.3% ± 0.2%, with a total mass of 9.260 g ± 0.005 g. As is clear from the data, the amount of material extruded for the trajectories is not proportional to the set extrusion volume ratio; however, the results of implementing the multi-trajectory method are clear: the porosity was substantially reduced without sacrificing the precision of the geometric structure or increasing surface roughness.

[0161] Example 2. To print three 20 mm cubes, each with different print settings, material ejection is deployed via a fused deposition modeling (FDM) machine with a single 400 μm nozzle and a thermoplastic polymer feed wire primarily composed of polylactic acid. Following the scheme shown in Figure 16 and the example in Figure D, a single-track scan strategy is used for two cubes, having the same machine settings except that each cube is printed with a different extrusion volume ratio setting. A multi-track scan strategy is used for one cube, having the same setting settings as the single-track strategy, except that the alternation between the two extrusion volume ratio settings is used to print a continuous trajectory, resulting in two distinct scan trajectory geometric structures. In this example, only one of each type of cube is printed; therefore, the standard deviation of porosity is measured across multiple images taken at different locations within the cube. The standard deviation given for mass is a measure of the reproducibility of the precision scale.

[0162] Single-track cubes:

[0163] Figures 8 and 9 are cross-sectional images of single-track geometry cubes printed on a Prusa i3 machine using the following printer settings (and others not listed herein):

[0164] Floor temperature: 40℃ Extruder temperature: 190°C Fan output: 100% Feed rate (i.e., head movement speed): 1800 mm / second Parallel line spacing (hatch spacing): 400μm Nozzle diameter: 400 μm Layer thickness: 200μm Extrusion volume ratio of cube 1: 0.35 Extrusion volume ratio of cube 2: 0.52

[0165] The measured porosity of cube 1 was 8% ± 0.6%, with a total mass of 8.704 g ± 0.005 g; the porosity of cube 2 was 7% ± 1.3% away from the edges, with a total mass of 3.071 g ± 0.005 g. The main reason for this was that the printing was not completed due to a printing failure, and only about 36% of the "cube" was printed.

[0166] Multi-Track Cube:

[0167] Figure 10 is a cross-sectional view of a multi-track geometry cube printed on a Prusa i3 machine using the following printer settings (and others not listed herein):

[0168] Floor temperature: 40℃ Extruder temperature: 190°C Fan output: 100% Feed rate (i.e., head movement speed): 1800 mm / second Parallel line spacing (hatch spacing): 400μm Nozzle diameter: 400 μm Layer thickness: 200μm Extrusion volume ratio of the smaller trajectory: 0.34 Extrusion volume ratio of the larger trajectory: 0.69

[0169] The measured porosity of the multi-trajectory cube was 0.2% ± 0.06%, with a total mass of 9.440 g ± 0.005 g. As is clear from the data, the amount of material extruded for the trajectories is not proportional to the set extrusion volume ratio; however, the results of implementing the multi-trajectory method are clear: the porosity was substantially reduced without sacrificing the precision of the geometric structure or increasing surface roughness.

[0170] Example 3. In this example, the porosity and mechanical properties of parts printed by powder bed fusion are shown for two 316L stainless steel powders: one with a d50 of 18 μm in Example 3a and another with a d50 of 45 μm in Example 3b. For each of the three methods (two single-trajectory types and one multi-trajectory type), four 1 cm porosity cubes are printed using the printing settings described in Example 3a. For each of the two methods (one single-trajectory type and one multi-trajectory type), three 7.5 cm (2.5 cm gauge) dogbone tensile coupons are printed using the printing settings described in Example 3b. In the single-trajectory method of Example 3a, the scanning method shown in Figures 32c and 32b is deployed, and in the multi-trajectory method, the scanning method shown in Figure 32a is deployed. The scanning paths for the single and multiple trajectory methods in Example 3b are the same as those shown in Figures 32c and 32a, respectively.

[0171] Example 3a. Porosity

[0172] Single-track cubes:

[0173] Laser output: 252 W Parallel line spacing: 140μm Gaussian Beam Waist: 55 μm Net displacement: 50 μm Powder thickness: 50 μm Scanning speed: 1190 mm / second

[0174] The average porosity measured for cubes printed according to the method shown in Figure 32b was 0.56% ± 0.073%, and for cubes printed according to the method shown in Figure 32c, it was 1.127% ± 0.147%.

[0175] Multi-Track Cube:

[0176] Laser output: 252 W Gaussian Beam Waist: 55 μm Parallel line spacing: 140μm Net displacement: 50 μm Powder thickness: 50 μm Scanning speed ratio: 1.6

[0177] The average porosity measured for cubes printed according to the method shown in Figure 32a was 0.231% ± 0.030%, which is significantly lower than the porosity of the single-trajectory method. In this case, since the energy deposited per unit mass of printed cubes (and total printing time) is the same for all three methods, it is clear that the multi-trajectory method can provide substantially lower porosity for the same energy input and printing time. From another perspective, the energy input and the resulting printing time could be reduced by using the multi-trajectory method described above, by making the porosity of the generated cubes equal to the porosity of one of the single-trajectory methods. In this case, the same porosity is obtained at a considerably faster printing speed, so it may be compared between the multi-trajectory methods, but the processing capacity of the multi-trajectory method is substantially increased.

[0178] Example 3b. Mechanical properties

[0179] Single-trajectory tension bar:

[0180] Laser output: 252 W Parallel line spacing: 140μm Gaussian Beam Waist: 55 μm Net displacement: 50 μm Powder thickness: 50 μm Scanning speed: 1066 mm / second

[0181] Multiple-trajectory tension bar:

[0182] Laser output: 252 W Parallel line spacing: 140μm Gaussian Beam Waist: 55 μm Net displacement: 50 μm Powder thickness: 50 μm Scanning speed ratio: 1.6

[0183] The energy deposited per unit mass (and total printing time) was the same for both the single-trajectory and multi-trajectory bars printed. All bars in the build chamber were oriented in the same direction, and the scanning path direction before rotation was the same. However, the total energy deposited per unit mass was greater for the bars printed in Example 3b compared to the cube printed in Example 3a. This helps to eliminate the difference in porosity between the single-geometric scanning method and the multi-trajectory scanning method so that the tensile properties of the bars can be compared based on nearly equivalent porosity. In addition to this purpose, the change in powder prepared for printing in Example 3b resulted in a further reduction in porosity compared to printing the cube in Example 3a. Bars printed using the "interleaved" method described in U.S. Patents 6,596,224 and 6,677,554, with a single geometric structure, exhibited an average ultimate tensile strength of 777 MPa ± 10.5 MPa, a yield strength of 621 MPa ± 12.9 MPa, and an elongation at break of 22% ± 1.7%. Bars with multiple trajectories exhibited an ultimate tensile strength of 802 MPa ± 4.0 MPa, a yield strength of 645 MPa ± 16.5 MPa, and an elongation at break of 36% ± 1.7%.

[0184] The measured mechanical properties revealed that the multi-trajectory method, even with equivalent porosity, yields a material composition superior to that of the single-trajectory method for multiple mechanical properties. Welch's t-test, performed to calculate confidence levels, showed that the ultimate tensile strength was greater for the multi-trajectory printed bar at 96.94% confidence, the yield strength was greater at 92.93% confidence, and the elongation to fracture was greater at 99.97% confidence. Therefore, the macroscopic and microscopic structures generated by the multi-trajectory method, as measured by mechanical properties, produce a different composition than those generated by the single-trajectory method.

[0185] Example 4. This shows the porosity measurement results for parts printed using powder bed fusion printing from the same 316L stainless steel powder used for printing the bars in Example 3b. For each of the two methods (one single-trajectory type and one multiple-trajectory type), four porosity cubes of length 1 cm were printed using the printing settings described below. The scanning path for the single-trajectory method is the same as that described for Figure 32c, and the multiple-trajectory method is based on the solution to the coating problem shown in Figure 16 and is the same as that described for material injection in Figure 12.

[0186] Single-track cubes:

[0187] Laser output: 252 W Parallel line spacing (hatch spacing): 120μm Gaussian Beam Waist: 55 μm Net displacement: 50 μm Powder thickness: 50 μm Scanning speed: 1161 mm / second

[0188] Multi-Track Cube:

[0189] Laser output: 252 W Parallel line spacing (hatch spacing): 120μm Gaussian Beam Waist: 55 μm Net displacement: 50 μm Powder thickness: 50 μm Scanning speed ratio: 1.5

[0190] The energy deposited per unit mass (and total printing time) was the same for both single-trajectory and multi-trajectory printed cubes. The average measured porosity of cubes printed using the single-trajectory method was 0.844% ± 0.110%, while that of cubes printed using the multi-trajectory method was 0.580% ± 0.076%. Welch's t-test showed with 99.46% confidence that cubes printed using the multi-trajectory method had reduced porosity compared to cubes printed using the single-trajectory method; this result is consistent with other results measured for printing for comparison between these two methods.

[0191] Example 5. This example shows the measurement results of the porosity of parts printed using powder bed fusion printing from nickel superalloy powder with a d50 of 21 μm. Four porosity coupons measuring 1 × 1 × 0.5 cm are printed using three methods (two single-trajectory methods and one multi-trajectory method) based on the printing settings described below. The single-trajectory method is unfolded using the scanning method shown in Figures 28b and 28c, and the multi-trajectory method is unfolded using the scanning method shown in Figure 28a. For all cubes, the deposition energy per unit mass and the total printing time were the same.

[0192] Single-track cubes:

[0193] Laser output: 269 W Parallel line spacing: 140μm Gaussian Beam Waist: 55 μm Net displacement: 50 μm Powder thickness: 50 μm Scanning speed: 1071 mm / second

[0194] The average porosity measured for cubes printed according to the method shown in Figure 28b, the "interleaved" method, was 0.497% ± 0.065%, while that for cubes printed according to the method shown in Figure 28c, it was 0.691% ± 0.092%.

[0195] Multi-Track Cube:

[0196] Laser output: 252 W Parallel line spacing: 140μm Gaussian Beam Waist: 55 μm Net displacement: 50 μm Powder thickness: 50 μm Scanning speed ratio: 1.6

[0197] The average porosity measured for cubes printed according to the method described in Figure 28a was 0.117% ± 0.015%, which is significantly lower than that of the single-trajectory method. Given that the total printing time and energy used were the same, the conclusion that the multi-trajectory method results in improved structure and efficiency is clear. From another perspective, by making the porosity of cubes produced using the multi-trajectory method equal to that of one cube produced using a single-trajectory method, energy input and consequently printing time could be reduced. In this case, the same porosity could be achieved at a considerably higher printing speed, thereby resulting in a substantial increase in the processing capacity of the multi-trajectory method, but with a comparable portion of the overall method.

[0198] Example 6. The porosity and mechanical properties of parts printed using powder bed fusion bonding from two aluminum alloy (AlSi10Mg) powders, with a d50 of 32 μm in Example 6a and a d50 of 41 μm in Example 3b, are shown. Using the printing settings shown in Example 6a, four 1 cm porosity cubes were printed for each method, one single-trajectory method and one multi-trajectory method. Three dogbone-shaped tensile coupons with a length of 7.5 cm (with a 2.5 cm gauge) were printed using the printing settings shown in Example 6b, in two similar methods (one single-trajectory method and one multi-trajectory method). The single-trajectory methods of Examples 6a and 6b are unfolded along the scanning path shown in Figure 27b, where Figure 27b is a cross-sectional image of one cube printed as Example 6a. The multiple-trajectory methods of Examples 6a and 6b are unfolded along the scanning path shown in Figure 27a, where Figure 27a is a cross-sectional image of one cube printed as Example 6a.

[0199] Example 6a. Porosity

[0200] Single-track cubes:

[0201] Laser output: 375 W Parallel line spacing (hatch spacing): 220μm Gaussian Beam Waist: 46 μm Net displacement: 50 μm Powder thickness: 50 μm Scanning speed: 1311 mm / second

[0202] Multi-Track Cube:

[0203] Laser output: 375 W Gaussian Beam Waist Ratio: 1.3 Parallel line spacing (hatch spacing): 220μm Net displacement: 50 μm Powder thickness: 50 μm Scanning speed ratio: 1.4

[0204] The average measured porosity of cubes printed according to the single-trajectory method shown in Figure 27b was 1.211% ± 0.158%, while the average measured porosity of cubes printed according to the multi-trajectory method shown in Figure 27a was 0.663% ± 0.086%, both significantly lower than that of the single-trajectory method. The energy deposited per unit mass (and total printing time) for cubes printed using both methods was the same, clearly demonstrating that the multi-trajectory method results in substantially lower porosity for the same energy input and printing time. In other words, by making the porosity of cubes obtained using the multi-trajectory method equal to that of the single-trajectory method, energy input and consequently printing time could be reduced. In this case, the same porosity can be obtained even at fairly high printing speeds, thereby providing a comparable portion between the multi-trajectory methods, although the processing load of the multi-trajectory method increases substantially.

[0205] Example 6b. Mechanical properties

[0206] Single-trajectory tension bar

[0207] Laser output: 375 W Parallel line spacing (hatch spacing): 180μm Gaussian Beam Waist: 88 μm Net displacement: 50 μm Powder thickness: 50 μm Scanning speed: 1389 mm / second

[0208] Multi-Track Cube:

[0209] Laser output: 375 W Parallel line spacing (hatch spacing): 180μm Gaussian Beam Waist Ratio: 1.3 Net displacement: 50 μm Powder thickness: 50 μm Scanning speed ratio: 1.4

[0210] In this case, the deposition energy per unit mass and total printing time were 3.3% lower in the multi-trajectory printing method compared to the single-trajectory method. All bars were oriented in the same direction within the build chamber, and the scanning path direction was the same. Bars using the single-trajectory method had an average tensile strength of 386.6 MPa ± 7.2 MPa, a yield strength of 238.8 MPa ± 2.9 MPa, and an elongation at break of 3.63% ± 0.05%. Bars using the multi-trajectory method had an ultimate tensile strength of 373.9 MPa ± 15.3 MPa, a yield strength of 234.4 MPa ± 11.5 MPa, and an elongation at break of 8.33% ± 3.37%.

[0211] The measurement results of the elongation at break indicate that the multi-trajectory method produced a material composition that exceeded the elongation at break of parts printed using the single-trajectory method, even though the deposited energy and total printing time were reduced. Using Welch's t-test to calculate confidence, the authors found that the ultimate tensile strength and yield strength of bars printed using the different methods fell within one standard deviation of each other, suggesting they were indistinguishable; however, it was found with 93.1% confidence that the multi-trajectory printed bars had a greater elongation at break. Therefore, it became clear that the macroscopic and microscopic structures generated by the multi-trajectory method, as measured by mechanical properties, produce a preferred composition different from that produced by the single-trajectory method. This result distinguishes the mechanical properties between the composition and the parts produced by the single-trajectory method (the deposited retention energy and printing time per unit mass are approximately equivalent), and the favorable results obtained in the multi-trajectory parts are also seen in other comparative tests. This is because the topological space of optimization available in methods using multiple trajectory geometric structures is far larger than the topological space of optimization available in methods using a single trajectory geometric structure.

[0212] Example 7. This shows the measurement results of the porosity of parts printed using powder bed fusion printing from 316L stainless steel powder similar to that used for printing the cube in Example 3b. For each of the two methods (single-trajectory method where the net displacement and powder thickness are equal, and single-trajectory method where the net displacement and powder thickness are different), four 1 cm porosity cubes were printed based on the following printing settings. The scanning method is as shown in Figure 30, and the deposition energy per unit mass (and total printing time) was the same for both methods.

[0213] The net displacement is equal to the thickness of the powder:

[0214] Laser output: 252 W Parallel line spacing: 140μm Gaussian beam waist: 91 μm Net displacement: 50 μm Powder thickness: 50 μm Scanning speed: 1004 mm / second

[0215] The net displacement and the thickness of the powder are different:

[0216] Laser output: 252 W Parallel line spacing: 140μm Gaussian beam waist: 91 μm Net displacement: 50 μm Powder thickness: 60 μm Scanning speed: 1004 mm / second

[0217] The average measured porosity of cubes printed using a method where the net displacement and powder thickness were equal was 0.478% ± 0.062%, while that of cubes printed using a method where the net displacement and powder thickness were different was 0.206% ± 0.027%. Applying Welch's t-test, the confidence level for the decrease in porosity of cubes printed using a method where the net displacement and powder thickness were different compared to cubes printed with equal net displacement and powder thickness was 99.94%.

[0218] Example 8 shows the measurement results of the porosity of parts printed using powder bed fusion bonding from aluminum alloy powder with a d50 of 27 μm. For each of the two methods (a single multiple-trajectory method where the net displacement and powder thickness are equal, and a single multiple-trajectory method where the net displacement and powder thickness are different), four 1 cm porosity cubes were printed based on the following printing settings. The scanning method is as shown in Figure 31, and the deposition energy per unit mass (and total printing time) was the same for both methods.

[0219] The net displacement is equal to the thickness of the powder:

[0220] Laser output: 650 W Parallel line spacing: 200μm Gaussian Beam Waist Ratio: 1.4 Net displacement: 50 μm Powder thickness: 50 μm Scanning speed ratio: 1.9

[0221] The net displacement and the thickness of the powder are different:

[0222] Laser output: 650 W Parallel line spacing: 200μm Gaussian Beam Waist Ratio: 1.9 Net displacement: 50 μm Powder thickness: 60 μm Scanning speed ratio: 1.9

[0223] The average measured porosity of cubes printed using a method where the net displacement and powder thickness were equal was 3.29% ± 0.43%, while that of cubes printed using a method where the net displacement and powder thickness were different was 1.79% ± 0.23%. Applying Welch's t-test, the confidence level for the reduction in porosity of cubes printed using a method where the net displacement and powder thickness were different compared to cubes printed with equal net displacement and powder thickness was 99.82%. This was the case when using a multi-trajectory method with aluminum alloy powder, but the same result was obtained even in the presence of the single-trajectory method using steel powder in Example 7.

Claims

1. printed using a scan- or trajectory-based procedure, or any additive manufacturing procedure that produces layers with a similar surface geometry to those produced using any scan- or trajectory-based procedure, in which within the area of ​​said printed object: a. The trajectory or trajectory cross-section has multiple geometries (shapes and / or sizes) within a single layer or across multiple successive layers contained within the region; b. The contour or cross section of a trajectory can be characterized by a unit cell described by the average of the trajectory or the average of the cross section of a trajectory in Euclidean space or other geometric space, having two or more bases; c. The trajectories printed within a unit cell are connected within said cell in the sense that there is no connected area (volume) or space of unprinted material within the cell that is greater than the area (volume) of the largest trajectory cross section (trajectory); d. Part or all of at least two adjacent unit cells are present in each layer that are partially or completely contained within said region; Additively manufactured compositions.

2. Within the area of ​​the printed object, the trajectory contour and average geometry or trajectory cross section is a solution or approximation of a solution to a covering problem that includes a plurality of unit cells in the portion of each layer that is included within the area.

10. The composition of claim 1.

3. the trajectory geometries or trajectory cross-sections are all or nearly all substantially similar to one another within a region of layer n, and in adjacent regions of successive layer n+1, the trajectory geometries or trajectory cross-sections are also all or nearly all substantially similar to one another but different from those within layer n; 10. The composition of claim 1.

4. Multiple trajectory geometries or trajectory cross-sections are used within or across regions of at least one layer, and these trajectory geometries or trajectory cross-sections vary coordinately within regions or layers of one or more layers that use them; 10. The composition of claim 1.

5. the trajectories are all parallel or anti-parallel; 10. The composition of claim 1.

6. The trajectories in the region of a given layer are rotated by a non-zero angle relative to the trajectories in the previous layer; 10. The composition of claim 1.

7. the locus within a given layer or layers region follows one or more non-linear contour lines; 10. The composition of claim 1.

8. The layer on which the traces are printed in the area of ​​the molding is non-planar; 10. The composition of claim 1.

9. The printing sequence of the tracks within the area of ​​one or more layers is designed so that adjacent or non-adjacent tracks are printed one after the other; 10. The composition of claim 1.

10. The printing sequence is designed so that all traces of one geometric structure are printed before traces of a different geometric structure are printed. The composition of claim 9.

11. the geometry (size and / or shape) and / or contour of the printed tracks is non-periodic or quasi-crystalline; 3. The composition of claim 2.

12. a scan- or trajectory-based procedure, or any additive manufacturing procedure that produces or is designed to produce layers having a surface geometry similar to that produced by any scan- or trajectory-based procedure, in the region of the molded object, a. The trajectory or trajectory cross-section is targeted to be generated using multiple AM ​​trajectory geometry settings within a single layer or across multiple consecutive layers contained within the domain; b. The track contours or track cross sections are designed according to their associated additive manufacturing track geometry parameters such that they are characterized by a unit cell described by an average of the track or average of the track cross sections in Euclidean space or other geometric space having two or more standards; c. The loci or locus geometries within a unit cell are connected within the cell in the sense that there is no connected area (volume) or space of unprinted material within the cell that is greater than the area (volume) of the largest locus cross section (loci); d. Part or all of at least two adjacent unit cells are present in each layer that are partially or completely contained within the region; A method for additive manufacturing or a method for designing a scanning procedure for additive manufacturing.

13. The designed trajectory or trajectory cross-section has a contour that solves the covering problem in two dimensions (for trajectory cross-section) or three dimensions (for trajectory); The method of claim 12.

14. the trajectories or trajectory cross-sections are all or nearly all substantially similar to one another within a region of layer n, and in adjacent regions of successive layer n+1, the trajectories or trajectory cross-sections are also all or nearly all substantially similar to one another but different from those within layer n; The method of claim 12.

15. Multiple trajectories or trajectory cross-sections are used within or across regions of at least one layer, and these trajectories or trajectory cross-sections vary coordinately within regions or layers of one or more layers that use them; The method of claim 12.

16. the trajectories are all parallel or anti-parallel; The method of claim 12.

17. The trajectories in the region of a given layer are rotated by a non-zero angle relative to the trajectories in the previous layer; The method of claim 12.

18. the locus within a given layer or layers region follows one or more non-linear contour lines; The method of claim 12.

19. The layer on which the traces are printed in the area of ​​the molding is non-planar; The method of claim 12.

20. The printing sequence of the tracks within the area of ​​one or more layers is designed so that adjacent or non-adjacent tracks are printed one after the other; The method of claim 12.

21. The printing sequence is designed so that all traces of one geometric structure are printed before traces of a different geometric structure are printed.

21. The method of claim 20.

22. the geometry (size and / or shape) and / or contour of the tracks designed to be printed is non-periodic or quasi-crystalline; The method of claim 13.

23. Within a region of the molding that includes all or part of each of one or more layers of the molding, the thickness of material in a particular layer or layers is varied (greater or smaller) than the thickness or net displacement of a section of the same layer or layers; A method for additive manufacturing or a method for designing a building sequence for additive manufacturing.

24. the thickness of material deposited for the region containing, or each portion of, one or more layers of the molding is such that the net displacement or thickness of the fraction of the same layer or layers is varied in such a manner as to change the average geometry and / or size of the scanning trajectory (and / or trajectory cross-section) within the portion of the layer or layers encompassed by the region; 24. The method of claim 23.