GB2634805A

GB2634805A - Heating method of electron beam powder bed fusion

Info

Publication number: GB2634805A
Application number: GB2406394.3A
Authority: GB
Inventors: James Smith Christopher; Andrew Boone Nicholas; Periane Natarajan Counassegarin Sasidharan; Muslim Talha; Alexander John Knight David; Laidler Ian
Original assignee: Wayland Additive Ltd
Current assignee: Wayland Additive Ltd
Priority date: 2023-10-18
Filing date: 2024-05-08
Publication date: 2025-04-23
Also published as: GB202315966D0; GB202406394D0

Abstract

Additive layer manufacturing and apparatus (1, fig 1) using a charged particle beam, e.g. electron beam (17, fig 1), to fuse metal powder in a powder bed (2, fig 1) by heating a plurality of discrete points (i) in a first group 1, 2, 3 wherein at least some of the points in the first group are adjacent; (ii) positioning the charged particle beam at least one discrete point away from the first group; and repeating steps (i) to (iv) for at least a second group 4, 5, 6 and at least one subsequent first and second group, until the charged particle beam has been positioned at every discrete point in the area forming the product. The points may be in a grid of rows, and the particle beam may be jumped along a row, the current, spot size, and velocity of the beam may be varied. A computer-readable storage device is also included. The point-based scan strategy is designed to maintain a uniform temperature and avoid a high thermal gradient and take account of charge accumulation.

Description

Heating method for electron beam powder bed fusion

Field of Invention

The present invention relates to use of a powder bed fusion apparatus in additive manufacturing, and particularly to a powder heating strategy using an electron beam.

The powder heating strategy can be used for pre-heating and/or melting or fusing of the powder.

Technical Background

One of the most prominent technologies employed for additive manufacturing is powder bed fusion, in which a thin layer of powder -typically metal or plastic -is selectively melted by an energy source such as a laser or electron beam. The melted area of the powder layer forms a cross-sectional part of an article to be built. After the layer has been selectively melted, a new layer of powder is deposited and then also selectively melted so that a complete article is constructed layer-by-layer.

The present invention is primarily concerned with electron beam sources, although extends to other types of charged particle beams. Typically, such electron beam sources are controlled/steered using electromagnetic deflectors. These electromagnetic deflectors allow the electron beam to be scanned across the powder bed, such that a pattern may be scanned or traced over the powder bed.

As the electron beam is scanned over the powder bed, energy is deposited into the powder, raising its temperature. Exposure to the electron beam must be carefully controlled to avoid sudden temperature increases in the powder bed, which can contribute to the generation of defects and results in incorrect formation (anisotropy) of the material microstructure. Moreover, during scanning of the electron beam over metal powder, the metal powder particles become charged and experience an increasing Coulombic repulsion. If the charge density exceeds a critical value, particles are repelled from the powder bed. The particles become mobile, creating a highly mobile powder cloud, destroying the layer-wise additive process instantly and potentially damaging the apparatus.

Electron beam powder bed fusion methods may include pre-heating the powder bed.

During the pre-heating phase the powder is sintered, i.e. small necks are formed between the adjacent powder particles. The temperature of the powder may in this way be raised incrementally up to the point of sintering, at which point the powder particles lightly bond together providing a conductive path to ground for the electronic -2 -charge. This increases the apparent weight of the powder particles, heat transfer and the electrical conductivity, thereby allowing better dispersion of negative charges. The sintering also increases the thermal conductivity of the powder, which is beneficial to keep a uniform temperature and to avoid a high thermal gradient between the melt pool and the surrounding powder. The particular scan strategy used for pre-heating must be carefully considered to ensure a desired temperature distribution across the powder bed, as well as to take account of charge accumulation within the powder.

It is desirable to provide a scan strategy which takes account of these issues.

Summary of Invention

According to an aspect of the present invention, there is provided a method of additive layer manufacturing using a charged particle beam to fuse metal powder within a metal powder bed to form a product layer-by-layer, the method comprising heating a predetermined area of the metal powder using the charged particle beam, wherein the predetermined area is defined by (or comprises) a plurality of discrete points. Heating the predetermined area comprises: (i) positioning the charged particle beam at each point, respectively, in a first group of the plurality of discrete points to heat the metal powder at each point in the first group, wherein at least some of the points in the first group are adjacent; (ii) positioning the charged particle beam at least one discrete point away from the first group; (iii) positioning the charged particle beam at each point, respectively, in a second group of the plurality of discrete points to heat the metal powder at each point in the second group (the second group being the at least one discrete point away from the first group), wherein at least some of the points in the second group are adjacent; and (iv) positioning the charged particle beam at least one discrete point away from the second group. The method further comprises (v) repeating steps (i) to (iv) for at least one subsequent first and second group, until the charged particle beam has been positioned at every discrete point in the predetermined area.

This heating strategy is called herein "point heating" or point-based heating. The point heating can be used for pre-heating and/or melting of the powder.

The at least one subsequent first group may be adjacent to a previous first group and the at least one subsequent second group may be adjacent to a previous second group. -3 -

The predetermined area may be divided into a plurality of sub-areas, and the method may further comprise: performing step (i) in each sub-area of the plurality of subareas; performing steps (ii) and (iii) in each sub-area; and performing steps (iv) and (v) in each sub-area, until the charged particle beam has been positioned at every discrete point in every sub-area of the predetermined area. Optionally, this method may comprise consecutively performing steps (ii) and (iii) in each sub-area, and consecutively performing steps (iv) and (v) in each sub-area.

In another example, the predetermined area is divided into a plurality of sub-areas, and the method may further comprise: performing step (i) in a sub-area of the plurality of sub-areas; performing step (ii) to position the charged particle beam in a different sub-area of the plurality of sub-areas; and performing step (iii) in the different sub-area of the plurality of sub-areas.

The method may further comprise repeating steps (i), (ii), (iii), (iv) and (v) until the charged particle beam has been positioned at every discrete point in the predetermined area at least twice.

The plurality of discrete points may be regularly spaced in a first direction.

The spacing between the plurality of regularly spaced discrete points, beam current, size of the sub-areas and/or the number of points in each group may be controlled or determined so as to provide a defined energy density per unit area across the powder bed. The energy density can be a user definable parameter and can be changed within a layer and/or between layers. The beam current and spacing can affect how quickly the layer is sintered or melted, and the number of points in each group (size of the group or point grouping) can affect the charge dissipation. In other words, the spacing can be adjusted as one of a number of optimisation parameters aimed at maximising throughput whilst maintaining a stable powder bed and a defined energy density per unit area. Other optimisation parameters include the beam current, the size of the sub-areas, and the size of each group.

A first point of the second group may be spaced apart from a last point of the first group by the at least one discrete point.

The plurality of discrete points defining the predetermined area may be arranged in a grid, and the grid may comprise one or more rows. -4 -

The one or more rows may extend in a first direction and each of the one or more rows may be offset from one another in a second direction.

The points in the first and/or second group may be arranged adjacent to each other in a same row. The points in the first and/or second group may be distributed across two adjacent rows. In some first groups all of the points may be arranged adjacent in a same row and in other first groups the points may be distributed across two adjacent rows. Similarly, in some second groups all of the points may be arranged adjacent in a same row and in other second groups the points may be distributed across two adjacent rows. The distribution of the points of each group is dependent on the spacing between consecutive (in time) groups and the number of points within the predetermined area.

Step (ii) of the method may further comprise positioning the charged particle beam a first predetermined number of discrete points away from the first group and step (iv) may further comprise positioning the charged particle beam a second predetermined number of discrete points away from the second group.

Steps (ii) and (iv) of the method may further comprise jumping the charged particle beam at least some of the first and/or second predetermined number of discrete points along a same row, and/or jumping the charged particle beam at least some of the first and/or second predetermined number of discrete points along a next row.

The number of points in each group may be equal and the first predetermined number of discrete points and the second predetermined number of discrete points may be a same predetermined number of discrete points, the same predetermined number being a multiple of the number of discrete points in each group.

The predetermined area of the metal powder may be greater than an area of metal powder corresponding to a layer of the product to be formed.

The method may further comprise varying one or more of the following parameters: a current of the charged particle beam; a spot size of the charged particle beam; a point spacing (e.g. spacing between discrete points).

According to another aspect of the present invention, there is provided an additive layer manufacturing apparatus comprising means for performing the above method. In some implementations, the apparatus comprises a charge mitigation mechanism. -5 -

According to another aspect of the present invention, there is provided a computer-readable storage device storing instructions that, when executed, cause at least one processor of an additive layer manufacturing apparatus to perform the above method.

Brief Description of Drawings

Embodiments of the present invention will be described by way of example only, with reference to the accompanying drawings, in which: Figure 1 shows an additive layer manufacture apparatus according to embodiments of the present invention; Figures 2a, 2b, 2c and 2d depict examples of predetermined areas defined by a plurality of discrete points, according to embodiments of the present invention; and Figure 3 shows a method of operating an additive manufacture apparatus according to embodiments of the present invention.

It will be appreciated that for convenience of explanation, some elements of the drawings are not shown to scale.

Detailed Description

Figure 1 shows an example powder bed fusion apparatus 1 according to embodiments of the present invention. The apparatus 1 shown in Figure 1 is configured for additive manufacture using an electron beam 17 to pre-heat and/or melt metal powder to form a part 3 layer-by-layer. Any suitable powder material may be used, including but not limited to a metal powder or a ceramic powder.

The powder bed fusion apparatus 1 comprises an electron-optical assembly 21 to form, condition and steer an electron beam 17. The electron-optical assembly 21 comprises an electron source 7 arranged to emit electrons. The electron-optical assembly 21 further comprises an electron extraction and focusing element 8 for forming a focused electron beam 17 from the emitted electrons, which travels along what is shown in Figure 1 as the z-axis of the apparatus 1. The electron-optical assembly 21 further comprises an electron deflection system 9 for steering the electron beam 17 to be incident on a powder bed 2 of metal powder, and for moving the electron beam 17 over the powder bed 2 to pre-heat and/or melt powder into a desired additive manufactured part 3. The electron deflection system 9 comprises electromagnetic deflectors arranged around the electron beam 17. -6 -

Operation of the electron optical assembly 21 is controlled via signals derived from a build controller (not shown), such as a suitably programmed computer, in accordance with a "scan file" for the desired part 3, as is known in the art. The operation defined by the scan file can be called a "scan strategy".

The apparatus 1 further comprises at least one hopper 4 operable to dispense powder via a dispensing mechanism (not shown) and a stage 20 to support a build tank 19 positioned to receive the dispensed powder in a volume defining the powder bed 2. The stage 20 is movable in the z-direction via a piston, and the hopper 4 and piston are controlled in conjunction with signals derived from the build controller (not shown).

Additive manufacture is performed under vacuum conditions in embodiments of the present invention. Hence the apparatus 1 further comprises a build vacuum chamber 5 through which the focused electron beam 17 passes before being incident on the powder bed 2. Coupled to the build vacuum chamber 5 is an electron column vacuum chamber 6 containing the electron-optical assembly 21. Vacuum conditions are maintained, as known in the art in powder fusion systems, with vacuum pressures of the order of 1 x 10-3 mbar to 1 x 10-7 mbar.

The hopper 4 dispenses powder so as to deposit a measured quantity of the powder on the powder bed 2 surface. A mechanism such as a scraper or blade (not shown) is used to disperse the powder evenly over the moveable stage 20. The electron-optical assembly 21 forms and steers the electron beam 17 such that the electron beam 17 is moved over the powder bed 2 to pre-heat the powder, and then to melt the powder so as to form a solid layer of a part 3. After each layer of the part 3 has been formed, the stage 20 is lowered in the z-direction to accommodate the increasing height of the part 3 and to allow the next layer of powder to be spread and pre-heated, and then melted.

In some examples, the apparatus 1 further comprises an ion source 11 for generating and emitting ions to be used as a charge mitigation mechanism. The ion source 11 may be a plasma source, as shown in Figure 1. The ion source 11 may be positioned within the build vacuum chamber 5, as shown in Figure 1, or may be positioned in an auxiliary vacuum chamber coupled to the build vacuum chamber 5. The ion flux 16 emanating from the plasma source 11 acts to neutralise negative charge building up on the powder bed due to the incident electron beam 17. The ion source 11 may be activated prior to activation of the electron source 7, to positively charge the powder -7 -bed 2 and build up a supply of ions in build vacuum chamber 5 for charge mitigation during the build (a process referred to herein as "priming"). The ion source 11 may remain active throughout the build. Operation of the plasma source 11 may be controlled via signals derived from the build controller (not shown). The charge mitigation mechanism can act to reduce charge accumulation effects in the powder arising from the application of the electron beam (so called "charge neutralisation"). The use of a charge mitigation mechanism thus allows greater flexibility in the preheating strategy, since the pre-heating strategy does not need to be designed to minimise/avoid charge accumulation in the powder.

As discussed above, electron beam powder bed fusion methods typically include an initial step of pre-heating the powder bed 2. According to the present invention, the pre-heating is performed using the electron beam 17. The pre-heating can be performed according to a predetermined area on the powder bed 2 surface, which defines the area within which the electron beam is moved. The predetermined area can be any regular or irregular shape, e.g., square, rectangle, or may correspond to the shape of the part 3 to be formed (i.e. generally follow the outline or shape of the part). In some embodiments, the predetermined area may be greater than the area 104 of a layer of the part 3 to be formed. For example, the predetermined area may form a margin 102 around the outline of the layer (see Figure 2d). The margin may be at least a 2 mm. Any suitable margin can be used.

As shown in Figures 2a and 2b, once the shape of the predetermined area 100 has been defined, the area 100 may be further defined according to a plurality of discrete points (1, 2, 3, ...). in other words, the area 100 can comprise the plurality of discrete points (shown as circles in Figures 2a, 2b). Each discrete point may correspond to a location on the powder bed 2 at which the electron beam 17 may be positioned, so as to pre-heat and/or melt the powder at that location. The scan strategy for moving electron beam 17 through the plurality of discrete points will be described in more detail below.

The plurality of discrete points may form a grid within the predetermined area 100 extending in a first direction and a second direction. The first direction and the second direction may be represented as an X-Y axis, as is shown in Figures 2a and 2b. The grid may comprise multiple rows spaced apart in the Y direction, with each row comprising a series of discrete points spaced apart in the X direction. The discrete points may be regularly spaced in the X direction. The rows of discrete points may be regularly spaced in the Y direction. Alternatively, the spacing in the X direction may -8 -vary between points, and/or the spacing in the Y direction may vary between rows. The spacing in X may be equal to the spacing in Y, or the spacing in X may be different to the spacing in Y. As such, the points in a row may align with the points in other rows in the X direction, or the points in a row may be offset from the points in other rows in the X direction. The arrangement/alignment of the points can depend in part on the size and shape of the area 100.

The point spacing dx may be determined based on any suitable parameters or factors, optionally including one or more of: the part geometry, electron beam spot size, clock speed of the data points, etc. The row spacing dy may be determined based on any suitable parameters or factors, optionally including one or more of: the part geometry, electron beam spot size, etc. For example, the spot size of the electron beam may be approximately 240 microns, the space between points may be approximately 100 x 240 microns, and the row spacing may be approximately 1.2 mm.

The points and/or rows may be spaced closer together than in previous approaches, due in part to the use of charge mitigation or charge neutralisation mechanisms within the device. This can help to facilitate to quicker and/or more efficient pre-heating. The spacing of the points and/or rows can be chosen or selected to improve the properties of the melt pool for subsequent melting/fusing of the part to be manufactured. For example, the spacing of the points and/or rows can be chosen or selected to provide a more even melt and/or a more even temperature across the melt pool, and therefore improve the porosity, microstructure, and/or material properties of the manufactured part. Moreover, with the approach described herein there is no need to sinter the powder first, which can reduce manufacturing time as compared to previous approaches.

When moving the electron beam 17 through the plurality of discrete points, for preheating and/or melting purposes, the electron beam 17 may be positioned at each discrete point within the predetermined area 100 according to a particular pattern.

Positioning the electron beam 17 at each discrete point may involve jumping or moving the electron beam 17 between a predetermined sequence of discrete points (adjacent or non-adjacent) and may involve holding the electron beam 17 at each discrete point for a specified duration of time (a dwell time) between moves/jumps.

The pattern may consist of a series of groups of discrete points, wherein at least some of the discrete points in each group are adjacent (in the same row), and wherein each group is spaced at least one discrete point away from a previous and/or subsequent group. This pattern (hold, move, hold, move) is distinct from a continual scan, in -9 -which the electron beam is scanned along a path (from a start point to an end point) at a constant scan speed.

The number of points in each group (each group is called a "point group") may be equal, and the spacing between the groups (the spacing is called the "point group spacing") may be based on the number of points within each group -for example, the number of points occupying the space between groups may be equal to the number of points in each group multiplied by an integer. Alternatively, the number of points in each group may vary, and/or the spacing between the groups may vary. The number of points in the groups may be predetermined based on one or more factors.

Optionally, the factors include one or more of: the size of the area 100, the beam velocity, the beam spot size, the powder material, or the beam energy.

For example, as shown in Figure 2a(i), a first (point) group of three exposed points (1, 2, 3) is spaced apart from a second (point) group of three exposed points (4, 5, 6) by nine unexposed points (represented by white circles). The numbering represents the order in which each point is exposed to the electron beam 17. During this first pass, represented by the black circles, the electron beam 17 may visit the points according to this pattern until the final point in the predetermined area 100 is reached. During a second pass, represented by the grey circles, the electron beam 17 may repeat this pattern, starting at the first of the nine unexposed points, to achieve the result shown in Figure 2a(ii). In this example, the first groups of the second pass are adjacent to the first groups of the first pass, and the second groups of the second pass are adjacent to the second groups of the first pass. This process may be repeated until all points in the predetermined area 100 have been exposed to the electron beam 17 (in the simplified example shown in Figure 2a, four passes would be required). In some embodiments, the entire process may be repeated until all points in the predetermined area 100 have been exposed to the electron beam 17 two or more times (in the simplified example shown in Figure 2a, eight passes would be required to visit all points twice, twelve passes would be required to visit all points three times, and so on).

In example of Figure 2a set out above, the second pass starts at the first of the nine unexposed points, but in other examples the second pass may start at a different location, such as the fourth or the seventh of the nine unexposed points. Similarly, in this example, the three points in each group are all arranged adjacent to each other in a same row, but in other examples the three points in each group may be distributed across two rows (e.g. two points at the end of a row, and one point at the beginning -10 -of a next row). As shown in Figure 2a example, the nine unexposed points may also be adjacent to each other in a same row or may be distributed across two rows.

As shown in Figure 2b, the predetermined area 100 may be further divided into a plurality of sub-areas 100a-d, each sub-area being defined according to a plurality of discrete points as described above. These sub-areas may also be called "tiles". In the example shown in Figure 2b, the pattern followed by the electron beam 17 would be the same as that described above in relation to Figure 2a, except that each first group of points (1, 2, 3) in each sub-area (100a-d) would be exposed to the electron beam 17 in sequence (i.e. points of the first group of sub-area 100a, then points of the first group of the sub-area 100b, etc.), and then each second group of points (4, 5, 6) in each sub-area (100a-d) would be exposed to the electron beam 17 in sequence (i.e. points of the second group of sub-area 100a, then points of the second group of subarea 100b, etc.), again leaving a number of unexposed points between the first and second group in each sub-area. In this example, it can be seen that the points (4, 5, 6) in the second group are distributed across adjacent rows, while the points (1, 2, 3) in the first group are adjacent to one another within a same row.

As above, this process may be repeated until all points in every sub-area 100a-d of the predetermined area 100 have been exposed to the electron beam 17. In some embodiments, the entire process may be repeated until all points in every sub-area 100a-d of the predetermined area 100 have been exposed to the electron beam 17 two or more times. In other examples, the pattern may be different. The first point (1) from the first group of points may be exposed in each sub-area or tile (100a-d), then the second point (2) from the first group of points may be exposed in each sub-area, and so on. In other words, each point at a time (e.g. point 1 in each tile, point 2 in each tile, point 3 in each tile) may be exposed to the beam 17.

For each layer of the part 3, during pre-heating and/or melting phases, the scan strategy of Figures 2a, 2b can be used as follows: Step 1. Position the charged particle beam at each point, respectively, in a group of discrete points to heat the metal powder at each point in the group.

Step 2. Position the charged particle beam at least one discrete point away from the group.

Step 3. Repeat Step 1 and Step 2 until the charged particle beam has been positioned at every discrete point within a predetermined area.

Step 4. Optionally, repeat Steps 1, 2 and 3 for another predetermined area (e.g. for another tile, where each predetermined area corresponds to a tile). In this implementation, each sub-area or tile is completely exposed before moving onto the next tile (not shown in Figures 2a, 2b).

Step 5. Optionally, repeat Steps 1 to 4 a predetermined number of times (e.g. for multiple passes over the same area). Optionally, one or more parameters of the scan strategy (such as the number of points, point spacing, number of points in a group, group spacing, etc.) may be changed between each repeat.

In one specific example, for each layer of the part 3, during pre-heating and/or melting phases, Steps 1, 2 and 3 of the scan strategy of Figures 2a, 2b described above can be implemented as follows: Step 1. Position the charged particle beam at each point, respectively, in a first group of discrete points to heat the powder at each point in the first group. A predetermined area is subdivided into a number of sub-areas or tiles. This first group of discrete points is arranged in a first sub-area, or first tile. For example, a first group of points (1, 2, 3) are exposed in first tile 100a. This is step (i) of claim 1 as filed.

Step 2. Position the charged particle beam at least one discrete point away from the first group. In this example, the charged particle beam is positioned in a second sub-area, or tile. This is step (ii) of claim 1 as filed.

This second tile can be tile 100c, 100b or 100d. In some particular examples a chequerboard pattern may be used, and the beam may be moved to tile 100d after tile 100a. However, any suitable tiling strategy can be used to determine which tile to move the beam to.

Step 3a. Position the charged particle beam at each point, respectively, in a second group of discrete points to heat the powder at each point in the second group. This is step (iii) of claim 1 as filed.

-12 -The second group can be points (1, 2, 3) located within tile 100c, tile 100d, or tile 100b, depending on the particular tiling strategy. In some examples, the number of points in the first group corresponds to a number of points in a row of the tile (in the example of Figure 2b, this would be 13 points). In one implementation of this example, where the first group are exposed in tile 100a in Step 1, the beam could be moved to tile 100d in Step 2 so as to allow for at least one unexposed point between the first group and the second group which is exposed at Step 3 (chequerboard pattern). In other implementations of this example, the beam could be moved to tile 100b in Step 2 so as to allow for at least one unexposed point between the first group and the second group which is exposed at Step 3.

Step 3b. Position the charged particle beam at least one discrete point away from the second group. This is step (iv) of claim 1 as filed. In this example, the charged particle beam can be positioned in a third sub-area, or tile, or can be positioned in the first tile again.

Step 3c. Repeat Steps 1 to 3b, until the charged particle beam has been positioned at every discrete point in the various tiles of the predetermined area. This is step (v) of claim 1 as filed.

This implementation is described further with reference to Figure 2c.

In accordance with Step 1, the charged particle beam is positioned at each point, respectively, in a first group of the plurality of discrete points to heat the powder at each point in the first group. In this example, the points in the first group are adjacent and the first group comprises the first row of sub-area or tile 100a (points 113). In other words, Step 1 is performed in a sub-area (sub-area 100a) of the plurality of sub-areas. In other examples (not shown), the number of points in each group can be the number of points in half a row of a tile.

In accordance with Step 2, the charged particle beam is positioned at least one discrete point away from the first group. Step 2 is performed so as to position the charged particle beam in a different sub-area of the plurality of sub-areas. In this example, the beam is moved or positioned to sub-area 100d, which is a different sub-area to the sub-area in Step 1.

-13 -In accordance with Step 3a, the charged particle beam is positioned at each point, respectively, in a second group of the plurality of discrete points to heat the powder at each point in the second group. In this example, the points in the second group are adjacent and the second group comprises the first row of sub-area or tile 100d (points 1 -13). In other words, Step 3a is performed in sub-area 100d, which is different to the sub-area of Step 1. The points exposed in this first iteration are shown in black in sub-areas 100a, 100d.

In accordance with Steps 3b and 3c, the charged particle beam is positioned at least one discrete point away from the second group and Steps 1 to 3b are repeated. In this example, when Step 1 is repeated the beam is moved to a new sub-area or tile, 100c, and a new first group is exposed. In Step 2 the beam is moved to a different sub-area, 100b, and a new second group is exposed. The points exposed in this second iteration are shown in grey in sub-areas 100c, 100b. In this way, a chequerboard pattern is followed.

At Step 3c, Steps 1 to 3b are repeated until a group of points has been exposed in each sub-area of the plurality of sub-areas (i.e. as shown in the illustrative example of Figure 2c). The Steps 1 to 3b are then further repeated until the charged particle beam has been positioned at every discrete point within the plurality of sub-areas or tiles of the predetermined area.

However, this is just one specific implementation and it will be understood that any suitable tiling strategy can be used. For example, in some implementations Steps 1 to 3b can be repeated in any suitable order until the beam has been positioned at each point. For example, tiles or sub-areas 100a, 100d may be filled by repeatedly performing Step 1 in sub-area 100a and Step 3 in sub-area 100d until each point in those sub-area has been exposed. After that, sub-areas 100c, 100b may be filled by repeatedly performing Step 1 in sub-area 100c and Step 3 in sub-area 100b until each point in those sub-areas has been exposed. In some other implementations, it may be that the group of points 1 -13 in tile 100a and the group of points 1-13 in tile 100c are exposed consecutively. In this instance, tiles 100a, 100c can be considered as a single sub-area and the black points 1-13 and grey points 1-13 in these tiles can be considered as a first group (Step 1). After exposing the first rows of 100a, 100c, the charged particle beam may be moved at least one discrete point away from the first group (Step 2), for example to tile 100b.

-14 -The tiles or sub-areas may each be filled so that there are no unexposed points between groups; in other words, upon returning to a sub-area a next group is positioned adjacent to a previous group (e.g. consecutive groups within a sub-area are arranged adjacent to one another). In other examples, such as shown in Figure 2b, there may be unexposed points between consecutive groups within a sub-area.

With reference to Figure 2d, a shape to be manufactured has an associated area 104.

A margin 102 is provided around the shape 104. The margin 102, shown by the dashed line, is a predetermined margin size. Optionally the margin 102 is 2mm, but any suitable margin size can be used. The shape 104 and margin 102 make up the predetermined area 100.

The area of the part(s) 104 (in Figure 2d only one part is shown, but there may be multiple parts per layer) and the margin 102 is discretized or divided into sub-areas (tiles), which are shown by the dotted lines. The size of the tiles can be kept constant (as shown in Figure 2d) or can be variable (variable in the same layer and/or variable between the different layers). The size of the part and the size of the tiles affects the number of tiles/sub-areas. The number of sub-areas can change as the size of the preheat area 100 changes along the length of the part(s) (e.g. there are more tiles across the part 104 at point A than at point B).

The sub-areas can be scanned fully, or partially. The maximum size of the point grouping (number of points in each group) is the number of points in each row of the tile. As discussed above, the point grouping is the number of points the charged particle beam scans in one step. In other examples, a point grouping may be defined as the number of points in half a row of a tile, or as any other suitable number. This can be a fixed parameter. As the area to pre-heat changes, the beam current and point spacing and point grouping can each be adjusted to provide the same, defined, energy density per unit area (even if the tile or sub-area size remains the same). In other words, there can be the same number of points in each tile or sub-area, or a different number of points can be defined for each sub-area independently. The point grouping and point spacing can be precalculated for the tile size for a given layer, but might be changed between layers (e.g. as the part geometry changes layer by layer). In instances where there are multiple parts, the number of tiles and the number of points in each tile can be defined independently for the different parts.

Another parameter that can be defined is the number of meltpools. The number of meltpools is the number of sub-areas the user wants the beam to jump to before -15 -returning back to the first sub-area. The number of meltpools can be defined based on the total number of sub-areas (tiles). The number of meltpools can also be varied (one of the optimisation parameters) as the number of subareas (tiles) changes based on the part geometry 104. The number of meltpools can be the same as the number of tiles, half the number of tiles, or any other number.

The charged particle beam is positioned at each point in a first group in a first sub-area. It stays on each point for a predefined dwell time before moving to the next point in the first group. The predefined dwell time for each point can be the same as for the other points, or may be different. Any suitable dwell time can be used for a respective point. The beam moves the length of the point grouping in the first subarea (e.g. sub-area 100a in Figure 2c). Then the beam jumps to the next sub-area (e.g. sub-area 100d in Figure 2c). The beam can jump to the sub-area adjacent to it in the x or y, or it can skip sub-areas in the middle and even skip multiple sub-areas, depending on the particular tiling strategy. The charged particle beam scans a first point group in that sub-area (e.g. sub-area 100d) and then is positioned or jumped to the next sub-area. This process continues until the beam has scanned a number of the sub-areas equivalent to the number of meltpools defined. Then the beam returns back to the first sub-area and scans the next group (second group in the first sub-area). In other words, the beam scans a first point group in the first (start) tile, and then jumps to the next tile and continues scanning the first point group in each tile until it returns to the start tile, where it will scan the second point group. The second group can be located adjacent (either next to or parallel with the first group, depending on the length of the point grouping). In other examples, the second group can be separated from the first group by unexposed points.

The beam continues scanning until it scans all sub-areas. If the beam scanning is configured in such a way such that the beam is skipping sub-areas, then the beam can scan the next group of sub-areas that it was initially skipping and skip sub-areas that it was initially scanning. For example, the beam might scan a fixed number of tiles in the piecewise manner described above (first group in each tile, second group in each tile) and then move on to a second, different, fixed number of tiles to repeat the process. This will continue until all the points in all the subareas are scanned. This happens in one pass, and then the entire process can be repeated, optionally multiple times.

-16 -The number of tiles, or tile size, affects or influences the number of meltpools, and optionally the point grouping (the number of points of the point group). For example, point groups may be larger when the tiles are larger, and vice versa. Preferably, the number and size of the tiles is fixed within a single layer. However, the number and/or size of tiles can change between respective layers of the build (i.e. change between different layers over the course of a build as the part geometry changes). In addition to changing the number of tiles and optionally the point grouping based on the geometry size, the build controller software can (automatically) scale the beam current and the point spacing of the points of the grid as the part geometry changes; this allows for optimising the time, and hence thermal properties, for the preheat or melt process, whilst maintaining a constant energy density. By reducing both the spacing of the points (i.e. more points per tile) and the beam current, whilst maintaining a constant dwell time, a constant energy density can be delivered over a longer period of time, thereby allowing to optimise thermal dissipation in the surface layers of powder to be preheated or melted. This technique can be used for smaller areas needing preheating or melting. Conversely, for larger areas, the point spacing and the beam current can be increased (i.e. fewer points per tile), at a constant dwell time, to deliver similar optimal thermal dissipation in the surface layers of powder to be preheated or melted.

Other parameters of the tiling strategy that might be changed include: tile rotation, where the beam rotates the direction in which it scans in the tiles; hatch stagger, where the scan direction changes by a specific angle after each pass; and stagger, where the tile grid moves forward by a predefined amount between passes (as the tiles do not overlap in the predetermined area 100, stagger can help ensure that there is no tiny area between adjacent tiles that did not get sintered).

A method of additive manufacture using a powder bed fusion apparatus 1, according to embodiments of the present invention, is also provided. The method is illustrated with reference to Figure 3, and is described in conjunction with the powder fusion apparatus 1 described with reference to Figure 1 and the predetermined areas 100 described with reference to Figures 2a, 2b, 2c and 2d.

The build controller obtains an instruction file for a part 3 to be made at step S10.

The instruction file contains the computer-executable instructions to be followed by the controller to form the part 3, for example electron beam parameters (e.g. beam energy, current, scan strategy, beam speed, spot-size) to be used for each of a preheating stage and a subsequent melting stage of the build, and a sequence of -17 -locations on the powder bed 2 to position the electron beam 17 for each of the preheating stage and subsequent melting stage required for formation of each layer of the part 3. The parameters specified in the instruction file may be constant throughout the build, or may be varied during the build depending on the part geometry, or for adaptive control of pre-heating/melting characteristics. The sequence of locations may correspond to discrete points within a predetermined area 100 of the powder bed 2, such as that described with reference to Figures 2a, 2b, 2c or 2d. As discussed above, the predetermined area 100 may be defined according to the shape of a particular layer of the part 3 being produced. For melting, the predetermined area 100 corresponds to the shape of the layer of the part 3. For pre-heating, the predetermined area 100 may be larger than the layer shape, so as to improve the melt pool around the areas to be melted/fused.

At step S20, the build controller starts the electron source 7 in accordance with a specification of pre-heating parameters and a sequence of pre-heating locations within the predetermined area. The build controller positions the electron beam 17 at the first pre-heating location retrieved from the instruction file. When the electron beam 17 is incident on the powder bed 2, it begins to heat the powder at the first preheating location. The electron beam 17 may remain at this location for a predefined dwell time. The build controller then moves or jumps the electron beam 17 to the next pre-heating location on the powder bed 2 as specified in the instruction file, where the electron beam 17 may again remain for a predefined dwell time. The dwell time may be the same for each location, or may be different. The dwell time can be determined in accordance with any suitable parameters, and can optionally be based on one or more of: the electron beam spot size, the point spacing, and/or the beam current.

The beam current, beam energy (defined by the beam voltage), point spacing, and dwell time defines the energy density across the surface of the powder.

The electron beam 17 is moved over the powder bed 2 in accordance with the pattern described above with reference to Figures 2a, 2b, 2c and 2d (e.g. Steps 1 to 5 above).

Using the example in Figure 2a, during a first pass, the electron beam 17 heats the powder at each pre-heating location ("point") within a first group for a dwell time, moves or jumps to a position at least one point away from the first group, heats the powder at each pre-heating location ("point") within a second group for a dwell time, jumps to a position at least one point away from the second group, heats the powder at each pre-heating location ("point") within a third group for a dwell time, and so on. The electron beam 17 may perform as many passes as necessary to expose all points in the predetermined area to the electron beam 17. When the build controller -18 -determines that pre-heating has been performed for the entire predetermined area of the layer of the part 3 being produced, the method proceeds to step S30. In some embodiments, the pre-heating step is repeated, such that each pre-heating location within the predetermined area is exposed to the electron beam 17 at least twice, incrementally raising its temperature. The temperature of the powder may depend on factors such as the electron beam power/current and dwell time.

At step S30, in one embodiment the build controller resets the electron source 7 in accordance with a specification of melting parameters and re-positions the electron beam 17 at the first melting location retrieved from the instruction file. The melting strategy may correspond to the pre-heating strategy described above (e.g. Steps 1 to 5), or may be performed differently. Where the above-described point-based preheating strategy is used for the melting phase, the predetermined area 100 used in the pre-heating phase may be changed for the melting phase, as may the number of discrete points and/or their distribution or spacing, the size of the point group and/or the size of the point group spacing. As the electron beam 17 moves over the powder bed 2, the electron beam 17 melts the powder to form a layer of a desired additive manufactured part 3.

In another embodiment, steps S20 and S30 may be combined into a single step. In other words, the procedure described in Step 1 to 5 -in which the electron beam 17 heats the powder at each point within a first group for a dwell time, moves or jumps to a position at least one point away from the first group, heats the powder at each point within a second group for a dwell time, jumps to a position at least one point away from the second group, heats the powder at each point within a third group for a dwell time, and so on -may be repeated until the temperature of the powder is raised enough to sinter the powder (when used for pre-heating) or melt the powder and form a layer of the part 3 (e.g. when used for melting). In this example, the pre-heating strategy described with reference to Figures 2a, 2b, 2c, 2d may be adapted throughout the step. For example, as discussed above in respect of Step 5, the predetermined area 100 may be changed after a certain number of passes or repeats, as may the number of discrete points and/or their distribution or spacing, the size of the point group and/or the size of the point group spacing.

When the build controller has performed the series of instructions in the instruction file and all required portions of the layer of the part 3 being produced have been sintered or melted, the electron source 7 is switched off at S40 and the method proceeds to step S50.

-19 -At step S50, the build controller determines or checks whether there are more layers in the instruction file to be processed. If there are no more layers to process (S50-N), the method ends. However, if not all layers have been processed, the method returns (S50-Y) via step S60 to step S20. In step S60, pre-heating and melting parameters and locations for the electron beam 17 for the next layer are retrieved from the instruction file, and the stage 20 is dropped and new powder spread to form the powder bed 2 for the next layer of the part 3. On return to step S20, pre-heating and melting of the newly spread powder layer is performed as described above.

In this way, the electron beam 17 may visit all the pre-heating and melting locations specified in the instruction file for each layer of the part 3 such that the part 3 is formed by additive layer manufacture.

In some embodiments, prior to activation of the electron source 7 at S20, a priming stage is performed in which the build controller controls an ion source 11 in accordance with a received specification of charge mitigation parameters (e.g. a particular ion beam current and energy to optimise a particular build process). The powder bed 2 is irradiated with ion flux 16 to charge the powder bed 2 to a low positive potential of the same magnitude as the energy of the ion flux 16 (for example, positive ions leaving the ion source 11 with a user-specified kinetic energy of 200 eV would induce a +200 V potential on the powder bed 20). By priming the powder bed 2 prior to the start of a build process, and prior to the start of the build of each new layer, a cloud of free ions accumulated above the powder bed 20 are then available for charge neutralisation throughout the build process. The ion source 11 may remain active throughout the build. By making use of such charge mitigation processes or mechanisms, the scan strategy described herein can be configured to improve melting of the powder without regard to charge accumulation effects.

In such embodiments, the ion source is described as providing positive ions, of opposite charge to the electrons of the irradiating electron beam. However, in alternative embodiments, the ion source may provide electrons or negatively charged ions to mitigate charge from a positively charged high energy beam. The same principles of operation apply as those described above.

It will be appreciated that the powder bed fusion apparatus can be configured in a number of different ways, depending on the requirements of a user for a particular build process, and compatible features of different embodiments may be readily -20 -combined. As described above, the various implementations of scan strategy described herein can be used for one or both of pre-heating and melting phases (whether separate phases or combined into a single operation). By making use of a scan strategy as described, melting of the powder material may be improved, which can help to improve the microstructure and material properties of the part being manufactured.

Claims

-21 -Claims 1. A method of additive layer manufacturing using a charged particle beam to fuse powder within a powder bed to form a product layer-by-layer, the method comprising heating a predetermined area of the powder using the charged particle beam, wherein the predetermined area is defined by a plurality of discrete points, and wherein heating the predetermined area comprises: (i) positioning the charged particle beam at each point, respectively, in a first group of the plurality of discrete points to heat the powder at each point in the first group, wherein at least some of the points in the first group are adjacent; (ii) positioning the charged particle beam at least one discrete point away from the first group; (iii) positioning the charged particle beam at each point, respectively, in a second group of the plurality of discrete points to heat the powder at each point in the second group, wherein at least some of the points in the second group are adjacent; and (iv) positioning the charged particle beam at least one discrete point away from the second group; the method further comprising: (v) repeating steps (i) to (iv) for at least one subsequent first and second group, until the charged particle beam has been positioned at every discrete point in the predetermined area.
2. The method of claim 1, wherein the at least one subsequent first group is adjacent to a previous first group and the at least one subsequent second group is adjacent to a previous second group.
3. The method of claim 1 or 2, wherein the predetermined area is divided into a plurality of sub-areas, and the method comprises: performing step (i) in each sub-area of the plurality of sub-areas; performing steps (ii) and (iii) in each sub-area; and performing steps (iv) and (v) in each sub-area, until the charged particle beam has been positioned at every discrete point in every sub-area of the predetermined area.
4. The method of claim 1 or claim 2, wherein the predetermined area is divided into a plurality of sub-areas, and wherein the method comprises: performing step (i) in a sub-area of the plurality of sub-areas; -22 -performing step (ii) to position the charged particle beam in a different subarea of the plurality of sub-areas; and performing step (iii) in the different sub-area of the plurality of sub-areas.
5. The method of any preceding claim, further comprising repeating steps (i), (ii), (iii), (iv) and (v) until the charged particle beam has been positioned at every discrete point in the predetermined area at least twice.
6. The method of any preceding claim, wherein the plurality of discrete points are regularly spaced in a first direction.
7. The method of any preceding claim, wherein a first point of the second group is spaced apart from a last point of the first group by the at least one discrete point.
8. The method of any preceding claim, wherein the plurality of discrete points defining the predetermined area are arranged in a grid, the grid comprising one or more rows.
9. The method of claim 8, wherein the one or more rows extend in a first direction and wherein each of the one or more rows are offset from one another in a second direction.
10. The method of claim 8 or claim 9, wherein the points in each of the first and/or second group are arranged adjacent to each other in a same row.
11. The method of claim 8 or claim 9, wherein the points in each of the first and/or second group are distributed across two adjacent rows.
12. The method of any preceding claim, wherein step (ii) comprises positioning the charged particle beam a first predetermined number of discrete points away from the first group and step (iv) comprises positioning the charged particle beam a second predetermined number of discrete points away from the second group.
13. The method of claim 12, wherein steps (ii) and (iv) comprise: jumping the charged particle beam at least some of the first and/or second predetermined number of discrete points along a same row, and/or -23 -jumping the charged particle beam at least some of the first and/or second predetermined number of discrete points along a next row.
14. The method of claim 12 or 13, wherein the number of points in each group is equal and wherein the first predetermined number of discrete points and the second predetermined number of discrete points is a same predetermined number of discrete points, wherein the same predetermined number is a multiple of the number of discrete points in each group.
15. The method of any preceding claim, wherein the predetermined area of the powder is greater than an area of powder corresponding to a layer of the product to be formed.
16. The method of any preceding claim, further comprising varying one or more of the following parameters: a current of the charged particle beam; a spot size of the charged particle beam; a velocity of the charged particle beam;
17. An additive layer manufacturing apparatus comprising means for performing the method of any of claims 1 to 16.
18. A computer-readable storage device storing instructions that, when executed, cause at least one processor of an additive layer manufacturing apparatus to perform the method of any of claims 1 to 16.