METHOD OF GENERATING PRINTHEAD ACTUATION DATA FOR PRINTING A 3-D OBJECT

Abstract

A method of generating printhead actuation data for printing a 3-D object, the method comprising the steps of: slicing, for a given 3-D object having 3-D object data corresponding thereto, the 3-D object data into a series of layers; and generating a 2-D vector graphics image associated with each layer, wherein the colour and/or the density of colour within the 2-D vector graphics image at a given point is used to determine at least one property of the material to be ejected, during printing, at that given point in the image.

Claims

1. A method of generating printhead actuation data for printing a 3-D object, the method comprising the steps of: slicing, for a given 3-D object having 3-D object data corresponding thereto, the 3-D object data into a series of layers; and generating a 2-D vector graphics image associated with each layer, wherein the colour and/or the density of colour within the 2-D vector graphics image at a given point is used to determine at least one property of the material to be ejected, during printing, at that given point in the image.

2. A method according to claim 1, wherein the property of the material to be ejected includes one or more of: the volume of material to be ejected, the depth of material to be ejected, the pattern in which the material is to be ejected, the printing material itself, the mass density of a binder fluid within the material to be ejected.

3. A method according to claim 1, further comprising the step of obtaining the 3-D object data corresponding to the 3-D object.

4. A method according to claim 1, further comprising the step of saving the 2-D vector graphics images as a 2-D vector graphics file.

5. A method according to claim 1, further comprising the step of storing, within metadata associated with the 2-D vector graphics file, information that relates to the layer thickness.

6. A method according to claim 1, further comprising the step of storing, within metadata associated with the 2-D vector graphics file, information that relates to the total height of the 3-D object.

7. A method according to claim 5, further comprising the step of analysing the metadata associated with a plurality of 2-D vector graphics files when determining how to nest or stack the objects to be printed within a print box.

8. A method according to claim 1, wherein the colour and/or the density of colour determines the pattern of material to be ejected within a given region.

9. A method according to claim 1, wherein the pattern includes at least one repeated pattern.

10. A method according to claim 9, wherein the repeated pattern is a 2-D tile pattern.

11. A method according to claim 10, wherein multiple adjacent 2-D tile patterns produce a contiguous pattern.

12. A method according to claim 10, wherein the 2-D tile pattern has at least one pair of opposing edges which match to permit a contiguous pattern to be formed when adjacent tiles are aligned.

13. A method according to claim 1, wherein the pattern of material is a honeycomb having regions of deposited material and void regions in which no material is deposited.

14. A method according to claim 1, further comprising the step of manipulating one or more of the 2-D graphics images to alter the form of the final 3-D object to be printed.

15. A method according to claim 14, wherein the manipulating step includes one or more of: rotation, scaling, changing the orthogonality of the image.

16. A method according to claim 1, wherein the at least one property of the material which is determined is a non-colour property of the material.

17. A system comprising a processor configured to carry out the method of claim 1.

18. A system according to claim 17, further comprising a screening engine wherein the screening engine is configured to convert the 2-D graphics vector file into raster data for inkjet printing a layer of a 3D structure.

19. A system according to claim 18, wherein the screening engine is configured to, in dependence upon the colour, tone, shading or other property of different parts of the image, alter the printing instructions to vary the printing within the different parts of the image.

20. A system according to claim 18, wherein the screening engine is configured to allow further editing of the 2-D graphics vector file pages prior to raster image processing (RIPping) so as to adjust for post-processing effects on the shape of the printed object.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0041] The present invention will now be described by way of example with reference to the accompanying drawings.

[0042] FIG. 1. schematically illustrates the workflow of existing 3D inkjet printers. 2D vector images are manipulated to define shell and infill structures.” This should be in done in step 2 of FIG. 1, and step 3 of FIG. 2.

[0043] FIG. 2. schematically illustrates the workflow of the invention. Note that [0044] a. step 3 “manipulated as 2D vector images and/or archived” does not exist in FIG. 1. [0045] b. Step 4 “ . . . using unique screening” does not exist in FIG. 1 and is where material qualities including in-fill microstructures are added to the image.

[0046] FIG. 3. schematically illustrates some of the 3D material characteristics that can be encoded as greyscale colour in our invention.

[0047] FIG. 4. schematically illustrates a screen tile created according to this invention. The dark lines are where fluid will be jetted to produce a structure element and the gaps are where none or little fluid will be jetted to save weight and materials.

[0048] FIG. 5. schematically illustrates a 2D arrangement of screen tiles created according to this invention. The dark lines of each tile match up to join those of neighbouring tiles to produce contiguous elements.

[0049] FIG. 6. schematically illustrates different tile arrangements in layers of tiles created according to this invention so that the detailed structure is contiguous with the detailed structure of adjacent layers above and below when overlaid. Adjacent layers can have different wall thicknesses, such that, whilst the walls are contiguous, i.e. joined to each other, the “white space” between the walls changes in size. In this way, the “white space” between the walls may form a tapered volume, such as a pyramidical 3d volume for the square tiles.

[0050] FIG. 7. A continuous tone image

[0051] FIG. 8. A half-tone image

[0052] FIG. 9. Magnified section of a half-tone image showing detailed pixilation normally too small to notice.

[0053] FIG. 10. Schematic interpretation of one preferred embodiment.

[0054] FIG. 11. Examples of in-fill structures, along the form shown in FIGS. 3, 4 and 10 comprising a structure with a shell, solid sections defined by the 3D model and large in-fill sections defined only by repeated screen tiles.

[0055] FIG. 12. Example of a 3D shape that can be produced as a homogenous solid using the present invention depending on the screen tile used.

[0056] FIG. 13. One 2D slice in vector format through the shape of FIG. 12 showing a solid interior body (lighter internal colouring) formed with a stronger outer skin (darker outer line).

[0057] FIG. 14. Example of a rasterized slice through the shape of FIG. 12 showing a first infill pattern that could be deposited depending on the screen tile used.

[0058] FIG. 15. Example of a further rasterized slice through the shape of FIG. 12 showing a second, more open, infill pattern that could be deposited depending on the screen tile used.

[0059] FIG. 16. Schematic of a 3D printer comprising a build box within which objects may be printed.

DETAILED DESCRIPTION OF THE INVENTION

[0060] In 2D printing, continuous tone grey levels in a 2D file are used to control and represent visual density, or luminance, see FIG. 7 as an example. Halftone screening is used with a binary palate of black and white to emulate varying shades of grey. The visual appearance of this effect is accomplished by varying the ratio of black to white dots within a specific area as shown in FIG. 8.

[0061] In the invention, a 2D vector file is created by slicing through a 3D object file to create a series of images representative of different heights within the 3D object. Each image equates to a slice of the 3D object, with adjacent images being equivalent to adjacent layers in the object. Instead of using colour and/or grey levels to show visual density variableness as in their original intended purpose of that 2D file format, the colour and/or greyscale levels within those images are used to represent the desired specific non-colour characteristics of certain areas within the 3D object to be printed. These non-colour characteristics relate to the nature of the 3D inkjet printing process such as the mass density of binder fluid deposited to produce strength, or the pattern of microstructural elements within an area or volume to produce strength but little weight.

[0062] This is shown schematically in FIG. 10 where 4 different drop sizes (including no drop) available from a selected printhead in 3D printing mode are encoded into 2 bpp (bits per pixel) grey levels as if they were greyscale colours and used to interpret a desired 3d image file to produce a solid structure with a stronger skin. As shown, the left hand side of the Figure is an image from a 2-D vector file, which has different greyscale levels used in different regions of the image. Those greyscale levels are converted using the 2 bpp levels in the middle part of the Figure such that the layer of the 3D object on the right hand side of the Figure that corresponds to the 2D image is printed with the correct detailing, such that the darker “edge” of the 2D image is 3D printed with a large drop so as to create the desired shell. The lighter greyscale is interpreted as a light drop, which can either be deposited as a less strong/lower density inner material.

[0063] The lighter greyscale level may be interpreted as shown in FIG. 11 as one of a pattern of different infills. Each infill may have a different density of material, or may have a particular characteristic in a particular direction, i.e. a series of infill lines in particular direction would provide increased strength in that particular direction. FIG. 11 shows how the outer portion of each part has been formed with a “solid” shell corresponding to a darker outer within a 2D image, with the centre part being less dense, i.e. by virtue of the infill pattern rather than solid material, and having been represented in the 2D images by a lighter greyscale.

[0064] The grey levels in vector slice files thus may be used to control and select infill patterns that produce microstructural elements in 3D objects. Each slice in vector format having a gray level to specify the area to be printed with an infill pattern. The exact pattern and location of these microstructures is defined by “screen tiles” that may not be specific to the actual 3D shape being printed. They may be determined by the greyscale within the 2D images and/or may be determined by other means, such as the infill pattern, i.e. triangular or cross hatching. These screen tiles produce structures, often called honeycomb structures even though that is but one example, that give strength but little weight. In some cases, they can be of temporary purpose, such as supporting the structure during printing, and quickly removed afterwards by mechanical or chemical means to reveal the intended 3D structure. In some cases, the microstructural elements are an intended structural component of the finished product, albeit that their exact positioning and pattern is not shown in the original 3D model; such model shows merely the 3D volumes in which such a structure is to be printed.

[0065] The honeycomb structure may be square or rectangular, offset alternating square or rectangle, triangular, hexagonal or diamond.

[0066] However, being able to draw such 3D volumes without having to draw each honeycomb element is a great time saving for the designer of the 3D shape. Likewise being able to quickly add supporting structures as undetailed volumes within the 3D model, without having to detail them in 3D, is a great time saving for the engineer whose task it is to make the 3D model suitable for the 3D printing process. Moreover, it can be foreseen that gravity can affect the choice of the temporary structure. The orientation of the 3D object within the 3D printer is often not known until a final stage of nesting products within the printer, so creating the temporary structure as part of the final RIPping of the nested image is advantageous in ensuring that the microstructure is always aligned to gravity irrespective of the orientation of each part.

[0067] An example of the use of 2D tiles is shown in FIGS. 3 to 6, in which FIG. 3 illustrates a structure having inner and outer solid shell walls, between which a less dense material is located. This could be represented in 2D image files using for example black lines for the solid shells and a mid tone grey scale for the less dense material therebetween. The infill pattern inside the inner shell would however be complex to draw as a series of solid lines, but by using a different greyscale level, e.g. a light grey tone, this complex in fill material can be rendered as a simple greyscale in the 2D images. The greyscale level of the infill material however is representative of a particular infill pattern made up of repeating tiles as shown in FIGS. 4 to 6. Thus, the 2D images can be manipulated as required and the only thing that changes is the shape of the light greyscale region. When it comes to actually RIP the 2D image into the printing commands, the area of the light grey tone is then translated into a series of identical tiles, such as shown in FIG. 4, which can be used repeatedly, as shown in FIG. 5 to fill the space defined by the light grey tone in the images. Then the desired infill pattern can be printed, so as to result in the desired infill shown in FIG. 3. FIG. 6 illustrates different arrangements of tiles that can tesselate, so as to create an infill pattern.

[0068] A further advantage of the use of 2D vector image files is that rasterization and pattern generation are no longer required in the slicing application. This makes the computational task of the slicer easier and faster. It is also more flexible since it is in vector format. Changes to infill pattern and small highly accurate re-sizing can be accomplished post slicing due to the use of vector format for the slices.

[0069] A screening engine in a standard 2D graphics RIP can be used to create the repetitive infill patterns. The screening engine may be a lattice pattern generating engine. Controls of pattern shape, wall thickness (for example, by controlling the ejected droplet size), cell size (that is, the size of the less dense material area between walls or boundaries of more dense material) and pattern size can be specified in the RIP. The screening pattern will maintain precisely the same point of origin for each successive Z axis slice to ensure X and Y axis alignment of the infill walls. The pattern point of origin will be in relation to the build box X and Y dimensions, not the object, so if multiple objects are to be temporarily joined during printing by a supporting structure then this too is easily achieved. The walls within a lattice pattern may be non-uniform, that it they may have a differing thickness at different depths of the wall. For example, a lower portion of a wall may be thicker than the top part of the wall. The change in thickness may be gradual, such as a sloped surface or may include one or more step changes.

[0070] Grayscale data in a vector file may produce internal structural patterns. Variation of the tonal values can be used to alter the characteristics of the structural pattern as mentioned above. For example, an increase in grayscale value may cause an increase in wall thickness of an internal geometric structure and/or a change in pattern. This allows for an overall decrease in fluid volume in a given area without sacrificing the strength of the object. An advantage of this method is the reduction in drying time of the fluid and, or, reduction in energy required to cure the fluid.

[0071] Whilst each slice of the 3d structure is a 2D vector file, it can be easily and accurately manipulated to make minute independent adjustment of X and Y scaling factors. This is necessary in some binderjet sintering processes in which the high heat used in the sintering process causes shrinkage of the printed object. Small adjustments to Z scaling can also be achieved by removing or duplicating pages (ie layers). Since the adjustments are made prior to rasterization, they can be accomplished with great accuracy and speed.

[0072] Additional manipulations such as rotation and orthogonality can also be made prior to rasterization to realize greater accuracy and speed for these types of adjustments to the final printed object.

[0073] A further disclosed innovation is the use of metadata in the 2D file format to store the height of the 3D object or the slice thickness that was used in creating the 2D file from the 3D file from which the height of the 3D structure can easily be calculated by the printer software by multiplying the number of pages by the slice thickness stored in the meta data. The printer software can then easily pack a print job by nesting individual jobs within the operating volume of the printer. X and Y dimension of each object can be obtained from the X and Y dimensions of the pages and the Z dimension of each object can be found, or calculated from the information placed in the metadata of the 2D file.

EXAMPLE EMBODIMENTS

[0074] In a first example embodiment, namely 3D binderjet printing, it is desirable to have a stronger, higher density shell around, but within, the object. This can be accomplished by ejecting larger inkjet droplets of binder fluid in the shell areas of the object. FIG. 10 shows how this can be controlled as described above. A Raster Image Processor (RIP) is used to interpret the PDF file and convert it to a raster TIFF file which is necessary to drive the inkjet printheads. Grayscale printheads have the ability to eject variable sized droplets which is controlled by a multiple bit per pixel (bpp) format TIFF file, of which 2 bpp and 4 bpp are commonly used. When ejecting binder fluid into sand layers, larger droplets will create stronger bonds between these grains of sand.

[0075] In a second example embodiment, namely additive manufacturing (AM) methods of 3D printing, it is common and desirable to use infill patterns in the inner spaces of objects. This provides lighter weight and stronger objects as well as lower cost due to less material usage. This can be seen in FIG. 11 which demonstrates a variety of different infill patterns. Each infill pattern may be associated with a specific colour or colour density in the 2D vector files.

[0076] In this invention STL or OBJ format CAD files are sliced to vector format and stored in Adobe PDF file format. One preferred embodiment of this method uses a single PDF file containing multiple pages where each slice of the 3D object is represented by a corresponding individual page in the multi-page document. Although this was not the intended use for Acrobat PDF files it provides many benefits which are significant to the 3D print systems which require rasterization for imaging. [0077] PDF is easily viewable with many open market tools for inspection of the job before printing. [0078] Since the files are in vector format, they are smaller in data size as compared to conventional raster slice files. Smaller file size makes them easier to archive or share via electronic networks. [0079] Vector PDF files can be edited or manipulated prior to rasterization. This provides capabilities such as scaling, infill pattern selection, geometric alterations, etc. [0080] PDF formats provide security encryption methods to protect IP content. [0081] PDF format is commonly used by many open-market Raster Image Processors (RIPs) which can be modified to adapt to the specific needs of 3D and AM inkjet printing.

[0082] As shown schematically in FIG. 16, each object of group of objects are shown here as having been nested in X and Y into a “slice pack” that fits the dimensions of the build box. The height of the slice pack is identified from the meta data of the 2D files and used to determine if another slice pack may be loaded above it. Here, two copies of one slice pack are shown and a second slice pack added above, all fitting within the internal volume of the build box. in one preferred embodiment of this invention, the items to be printed are nested by the printer software in a horizontal plane, according to their X and Y dimensions and then another layer is placed above on a virtual floor placed just above the highest Z height of the items in the lower layer until the predicted height of an item would be above the maximum achievable by the printer. This may be based on metadata stored in the 2D files in which at least the height (Z) of each “slice”, i.e. the 2D vector file, is stored. Other data may also be stored such as X and Y dimensions. For automated loading and unloading of a 3D printer, it can be seen that such automatic nesting by software removes the need for an operator: multiple jobs of any number of copies may be queued for unmanned production without regard to their actual 3D shapes.

[0083] The principle of storing the height of each slice in the metadata of the respective 2D vector file has various uses including: [0084] (1) Using a PDF file to describe the X-Y-Z volume dimensions of a 3D print job. [0085] (2) Using modified PDF files which describe X-Y-Z volume dimensions to efficiently manage 3D printing.

[0086] Each multipage PDF file where each page describes a slice of a 3D object can be used to determine the precise volume of its printed space. This can be achieved by using the following formula:

(X page dimension)*(Y page dimension)*((sum of all pages*slice thickness)=Z dimension)

[0087] The above formula will calculate the total volume in cubic inches or meters of each PDF 3D print file.

[0088] All of the variables in the above formula are known at the time the PDF 3D print file is produced. [0089] Page width=X dimension [0090] Page height=Y dimension [0091] Page count [0092] Slice thickness

[0093] When the slicing algorithm creates the PDF file additional data can be written into the metadata of the PDF file to provide the user, or software controlled by the user, the X-Y-Z value of the PDF 3D job. [0094] PDF job header contains X-Y dimensions of the PDF 3D print job. [0095] PDF metadata contains slice thickness for each layer (page). [0096] PDF metadata can also contain pre-calculated Z dimension of the PDF 3D print job (slice thickness*page count).

[0097] When each PDF file's X-Y-Z dimensions are known, a user can manually, or with the aid of a computer algorithm, programmatically select PDF 3D print jobs to make the most efficient usage of the 3D printer's build box volume.

[0098] For the purpose of the graphic examples shown each PDF 3D print job is referred to as a PDF “Slice Pack”

[0099] The applicant hereby discloses in isolation each individual feature described herein and any combination of two or more such features, to the extent that such features or combinations are capable of being carried out based on the present specification as a whole in the light of the common general knowledge of a person skilled in the art, irrespective of whether such features or combinations of features solve any problems disclosed herein, and without limitation to the scope of the claims. The applicant indicates that aspects of the present invention may consist of any such individual feature or combination of features. In view of the foregoing description, it will be evident to a person skilled in the art that various modifications may be made within the scope of the invention.