Method for layer-by-layer production of at least one object using stepwise update of coordinate transformation of scanner

By updating the programmatic coordinate conversion of the scanner multiple times on the construction platform, the object structure defects caused by the change in the scanner orientation are solved, the object quality and production efficiency are improved, and non-production time is reduced.

CN119013111APending Publication Date: 2024-11-22TRUMPF LASER & SYSTEMTECHNIK GMBH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202380032663.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2022-04-07
Filing Date
2023-03-13
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

When producing three-dimensional objects on the construction platform, due to the relative orientation changes of the scanner relative to the construction platform, the sub-faces of different high-energy beams may be misaligned relative to each other, causing defects in the structure such as the joints of micropores or local density in the structure Fluctuations affect object quality. The existing technology supplementary adjustment programming coordinate conversion requires non-production time, and the update frequency is too high, which may introduce offsets.

Method used

Between the two successive determinations of the current actual coordinate conversion, the programmed coordinate conversion of at least N-1 scanners is updated multiple times to gently supplement the adjustment of relative bias, decompose the accumulated bias to a smaller separate correction, reduce non- Production time to avoid structural defects.

Benefits of technology

It realizes the production of high-quality objects on the construction platform, reduces non-production time, avoids structural defects, improves productivity, and ensures the geometric accuracy and density uniformity of objects.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119013111A_ABST
    Figure CN119013111A_ABST
Patent Text Reader

Abstract

The invention relates to a method for the layer-by-layer production of at least one object (2) on a construction platform (6) by locally solidifying a powdered material (3) in each layer (7), in which: N high-energy light beams (8a, 8b) are used at least temporarily at the same time with N scanners, where N > = 2, at least in a plurality of layers; each scanner is provided with a scanner coordinate system; the control device (10) provides (102) exposure data of the processing pattern in a reference coordinate system for each scanner for exposure of each layer; transforming (103) the exposure data in the reference coordinate system into exposure data in a scanner coordinate system by means of programmed coordinate transformation (PKT), and assigning (104) the obtained exposure data in the scanner coordinate system to the associated scanner such that the scanner exposes (105) the processing pattern in the layer on the build platform; measuring is repeated during the layer-by-layer production of the at least one object, by means of which a current actual coordinate transformation (MTKT) of each of the at least N-1 scanners is determined; producing M layers, where M > = 2, between two consecutive determinations of the current actual coordinate transformation; and, during the production of the at least one object, updating (205.1, 205. J, 205. A; 305.1, 305. J, 305. A) for the programming coordinate transformation of the at least N-1 scanners. The invention is characterized in that the current actual coordinate transformation is determined between two consecutive determinations (201, 206; a plurality of updates of the programming coordinate transformations of the at least N-1 scanners are performed between the plurality of scanners (101, 301, 306). The present invention provides a method that makes it possible to produce at least one object layer by layer on a construction platform, such that the quality of the object is high while reducing non-production time.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a method for producing at least one object layer by layer on a building platform by locally solidifying a powdered material in individual layers,

[0002] wherein N high-energy light beams are used at least temporarily simultaneously with N scanners, at least in a plurality of layers, wherein N ≥ 2,

[0003] Each scanner is assigned a scanner coordinate system.

[0004] The control device controls the exposure of each layer for each scanner.

[0005] -Provide exposure data of the processing pattern in the reference coordinate system;

[0006] - transforming the exposure data in the reference coordinate system into exposure data in the scanner coordinate system by means of a programmed coordinate transformation; and

[0007] - assigning the obtained exposure data in the scanner coordinate system to the associated scanner so that the scanner exposes the tooling pattern in a layer on the building platform,

[0008] In this case, during the layer-by-layer production of at least one object, measurements are repeatedly performed, with the aid of which the current actual coordinate transformation of at least N-1 scanners is determined in each case, wherein between two successive determinations of the current actual coordinate transformation, M layers are produced, wherein M≥2,

[0009] And wherein, when producing at least one object, the programmed coordinate transformation for the at least N-1 scanners is updated taking into account the current actual coordinate transformation. Background Art

[0010] Such a method is known from EP 3 907 021 A1.

[0011] By producing the object layer by layer by locally solidifying a powdered material using a high-energy light beam (usually a laser beam or an electron beam), three-dimensional objects can be produced relatively easily and quickly. In this respect, the geometric limitations of conventional production processes such as milling or injection molding can be overcome. Layer-by-layer production is often used for prototypes or for objects that are only produced in small quantities or even only once (e.g. dental crowns).

[0012] In order to achieve particularly fast production, it is also possible to use multiple high-energy beams simultaneously on the building platform. For each high-energy beam there is a scanner (also called scanner system), which, for example in the case of laser beams, can include mirrors that can be adjusted by means of piezoelectric actuators.

[0013] In order to produce large objects on a building platform, multiple high-energy beams are usually used simultaneously on the same object. In each layer, the surface of the layer to be processed (cured) determined by the object is divided into sub-surfaces, which are each produced using one of the high-energy beams. The sub-surfaces to be processed by different high-energy beams touch at the boundary line.

[0014] The different scanners for the high-energy beam are controlled by an electronic control unit. Depending on the desired object to be produced, an exposure pattern is generated for each scanner for each layer to be produced based on its 3D data, which exposure pattern is initially given in a reference coordinate system. Each scanner is assigned a scanner coordinate system, and the respective scanner receives from the control unit the coordinates (exposure data) to be reached with the high-energy beam with reference to its scanner coordinate system. For this purpose, the exposure data for the scanner are transformed from the reference coordinate system into the respective scanner coordinate system using a programmed coordinate transformation for the respective scanner. The reference coordinate system is usually associated with the processing machine (machine coordinate system), but the scanner coordinate system of a selected scanner ("guide scanner / master scanner, Führungsscanner") can also be selected as the reference coordinate system.

[0015] As stated, for example, in EP 3 907 021 A1 or in WO 2018 / 086996 A1 or in WO 2019 / 173000 A1 or in DE 10 2018 205 403 A1, the coordinate transformation of a scanner, i.e. the relationship between a reference coordinate system and its scanner coordinate system, can be measured experimentally. Typically, in this respect, a scanner is used with its high-energy beam to generate one or more measurement points (e.g., reflection or fusion test patterns) on a building platform or calibration object, the positions of which are determined relative to the reference coordinate system.

[0016] However, when producing a three-dimensional object on a building platform, the relative orientation of the scanner relative to the building platform will change over time. This can be caused by, for example, temperature changes, changes in air pressure and changes in humidity. Different scanners are usually affected by orientation changes in different ways. In other words, compared with the programmed coordinate transformation for the scanner, the current actual coordinate transformation will change (except for the guide scanner if a guide scanner is provided). Therefore, the sub-faces on the building platform processed by different scanners or their high-energy beams may be misoriented relative to each other. This will cause defects in the structure of the produced object in the region of the boundary lines of the sub-faces of each layer, such as seams with micropores or local density fluctuations or even geometric defects. This affects the quality of the produced object.

[0017] A variant is disclosed in EP 3 907 021 A1, in which during the production of an object on a building platform, after every twelve produced layers, the current actual coordinate transformation of the other scanners is newly determined relative to the guide scanner and the programmed coordinate transformation of the other scanners is set to the most recently determined current actual coordinate transformation in an update and used for the production of the next twelve layers.

[0018] This procedure allows the programmed coordinate transformation to be adjusted and adapted to changes in the orientation of the scanner. However, it should be noted that each determination of the current actual coordinate transformation generally requires time, which is not available for the production of objects (so-called non-productive time). In addition, there is the risk that between two successive determinations of the coordinate transformation, the current actual coordinate transformation changes so strongly that the updating of the programmed coordinate transformation itself will itself lead to structural defects in the object to be produced (e.g., visible offsets, also called "steps") in the object. Summary of the invention

[0019] Purpose of the Invention

[0020] The object of the present invention is to present a method for producing at least one object layer by layer on a building platform, in which method the produced object is of high quality while keeping non-productive times low.

[0021] Description of the invention

[0022] This object is achieved according to the invention by a method of the type mentioned at the outset, which is characterized in that a plurality of updates of the programmed coordinate transformations of the at least N-1 scanners are carried out between two successive determinations of the current actual coordinate transformation.

[0023] The invention proposes that between two successive determinations of the coordinate transformations of the at least N−1 scanners the programmed coordinate transformation is supplemented for each scanner not with a single update but with a plurality of updates.

[0024] This makes it possible to divide the accumulated or expected relative offsets of the individual scanners, which are caused by disturbing influences such as temperature fluctuations (e.g. due to solar radiation), pressure fluctuations or humidity fluctuations (e.g. due to weather influences), into smaller individual corrections and to perform multiple updates with these corrections. This makes it possible to divide the programmed coordinate transformations of the scanners into smaller individual corrections, and to perform multiple updates with these corrections. This relative offset manifests itself in particular as a deviation between the previous programmed coordinate transformation and the current actual coordinate transformation. This enables a gentle supplementary adjustment of the programmed coordinate transformations of the scanners. The invention makes it possible to minimize or completely avoid defects at the boundary line between two sub-surfaces produced using different scanners in a layer in the structure of at least one object to be produced on a building platform.

[0025] Structural defects or the like due to relative offsets of the scanners relative to one another and / or relative to the building platform can be minimized or completely avoided by supplementary adjustment of the programmed coordinate transformations of the at least N-1 scanners during the processing of the building platform. In particular, microporous seams or density fluctuations or general geometric defects and offsets are avoided in the region of boundary lines in the layers or in the region of corresponding cross-layer interfaces, which would be visible on the surface.

[0026] Since even larger accumulated relative offsets can be supplemented and compensated if necessary without introducing significant defects into the structure of the object to be produced, the determination of the current actual coordinate transformation can be performed relatively infrequently. This can reduce the amount of non-productive time and thus increase productivity.

[0027] The determination of the current actual coordinate transformation of the scanner (or scanner system) is usually based on the measurement between the production of two layers; however, the determination can also be based on multiple sub-measurements distributed over the production of multiple layers. Measurements that can determine the current actual coordinate transformation of the scanner are known in the prior art and can be applied in the context of the present invention; in particular, the method described in EP 3 907 021 A1 or WO 2018 / 086996A1 can be used as part of the method described by the present invention; the contents of these documents are hereby incorporated by reference into the disclosure of the present invention. For example, one or more test structures can be set at a defined position on the building platform, and the defined structure can be introduced into the powder or introduced into the already welded area by remelting with all participating high-energy beams or scanners in the powder; these structures can then be detected using a camera device (these structures then represent detectable measurement points). Alternatively, such a structure can also be placed in a separate area (with its own building cylinder) outside the actual building range / building platform. It is also possible to use all scanners to scan fixed marks or known objects with laser beams and to obtain their latest poses, for example, by detecting backscattered, backreflected or generally emitted lasers and associating them with the position imprint of the scanners. In particular, a sapphire ball can be attached as a quasi-retroreflector to a feeder (which applies powder to the build platform) and is incidentally introduced into the build space and scanned by the feeder. Based on a certain number of independent measurement points (for each scanner), the relative position (at one or more measurement points) or orientation / rotation position (at two or more measurement points) of the scanners relative to each other or to the machine tool coordinate system can be obtained. In the case of determining the offset of N scanners relative to each other, that is, for N-1 scanners relative to the guide scanner (the scanner coordinate system of the guide scanner is then permanently defined as the reference coordinate system or defined as having a fixed relationship with the reference coordinate system), the update of the programmed coordinate transformation is usually performed for the N-1 scanners, and in the case of determining the offset of the N scanners relative to the machine tool coordinate system, the update of the coordinate transformation is usually performed for all N scanners.

[0028] The supplementary adjustment of the programmed coordinate transformation in the multiple updates between two determinations of the current actual coordinate transformation of the at least N-1 scanners takes into account at least the most recently determined measured current actual coordinate transformation and may also take into account measurement results determined earlier. In particular, algorithms from classical controllers can be used for the supplementary adjustment, in particular in order to limit the control variables (changes in the programmed coordinate transformation) in the individual control steps (updates).

[0029] It can be provided within the scope of the invention that the supplementary adjustment per update of the programmed coordinate transformation is limited to a maximum value (in terms of displacement and / or rotation), in particular to a certain maximum displacement, for example, for a spot diameter of about 80 μm to 100 μm, the supplementary adjustment per update is limited to 5 μm or 10 μm or 15 μm; for larger spot diameters, a larger maximum displacement can be selected, and vice versa. In most cases, the maximum displacement of the supplementary adjustment per update is in the range of 5% to 15% of the spot diameter of the associated high-energy beam. For example, the spot diameter can be determined according to the 86% criterion so that 86% of the beam power is located in a circle with the spot diameter.

[0030] Between two determinations of the current actual coordinate transformation, A updates are performed, where A≥2, preferably A≥5, and particularly preferably A≥10. Between two determinations of the current actual coordinate transformation, a total of M layers are produced, where M≥2, preferably M≥5, particularly preferably M≥10, and most preferably M≥20. Note that M≥A. The respective update (or the then newly stored set of programmed coordinate transformations) is then applied to the subsequent production of one or more layers.

[0031] Preferred variations of the invention

[0032] A variant of the method according to the invention is particularly preferred according to which

[0033] After each latest determination of the current actual coordinate transformation for the at least N-1 scanners, for each of the at least N-1 scanners,

[0034] - determining the target coordinate transformation taking into account the latest determined current actual coordinate transformation,

[0035] - and the programmed coordinate transformation is gradually transformed to the target coordinate transformation by means of multiple updates of the programmed coordinate transformation between the latest determination and the next determination. By gradually transforming the programmed coordinate transformation to the target coordinate transformation, the accumulated or expected offset of the scanner is softly compensated and the defects of the structure of at least one object to be produced on the building platform are minimized. Then, the new programmed coordinate transformation with the corresponding update (corresponding supplementary adjustment step) is applied to the production of one or more layers until the next update (next supplementary adjustment step). Between the latest determination and the next determination, after the target coordinate transformation is reached, the programmed coordinate transformation generally no longer changes until the next determination of the current actual coordinate transformation. If desired, the maximum change (offset and / or rotation) of the programmed coordinate transformation allowed in a single update can be defined for each scanner, typically between 2.5 μm and 25 μm for the maximum value of the offset and between 0.05° and 1.0° for the maximum value of the rotation in the plane of the building platform.

[0036] According to a preferred development of this variant, after the respective latest determination of the current actual coordinate transformation for the at least N-1 scanners, for each of the at least N-1 scanners,

[0037] - determining the initial deviation between the target coordinate transformation and the programming coordinate transformation on which the latest determination is based;

[0038] - dividing the determined initial deviation into a plurality of deviation parts;

[0039] - and changing the programmed coordinate transformation stepwise over a plurality of updates for the layers produced after the last determination of the current actual coordinate transformation until the next determination,

[0040] In this case, with each update, the additional deviation portion is added to the last programmed coordinate transformation applied by the control device. This procedure is simple and effective in practice.

[0041] The following sub-variant of this development is advantageous: the determined initial deviation is divided into deviation parts of equal size. This allows a particularly gentle adjustment of the programmed coordinate transformation of the individual scanners to the associated target coordinate transformation. In this sub-variant, multiple updates between two successive determinations are usually carried out after the production of an equal number of layers.

[0042] Alternatively, unequal deviation portions can also be selected. In particular, earlier updates in a determination interval can have a larger deviation portion than later updates. It is also possible to select the first deviation portion in a determination interval in accordance with the maximum permissible deviation portion (with respect to offset and / or rotation) and to compensate the remaining remainder of the initial deviation in the last update.

[0043] In a further sub-variant, it is provided that the respective deviation portion is limited by a maximum deflection and / or a maximum rotation. Defects in the produced object (e.g. microporous seams) can thereby be avoided very reliably. Typical maximum deflections per update are usually between 2.5 μm and 25 μm, and maximum rotations per update are usually between 0.05° and 1.0° in the plane of the building platform.

[0044] The following sub-variant of the above-mentioned development is also preferred: A updates are performed between two determinations, A=M being selected and the initial deviation is divided into M equal deviation parts. In other words, one update of the programming coordinate transformation is performed for each slice and the initial deviation is evenly distributed over these updates. This also makes it possible to adjust the programming coordinate transformation of the individual scanners particularly smoothly to the associated target coordinate transformation.

[0045] Particularly preferred is the following extended scheme, in which the target coordinate transformation corresponds to the latest determined current actual coordinate transformation. Until the next determination of the current actual coordinate transformation, the programming coordinate transformation of each scanner is supplemented and adjusted to the latest available experimental value of the current actual coordinate transformation. In most applications, this has made the quality of the at least one object produced good and is also easy to implement. No predictive calculation or prediction model is required, and there is no need to use or store the previously determined results of the current actual coordinate transformation. This approach is particularly useful for changes that cannot be predicted or can only be predicted with large uncertainty in the orientation of the scanner relative to the building platform. If the current actual coordinate transformation cannot be implemented as the target transformation within the planned determination interval, for example because the initial deviation is too large for the planned number of updates of the programming coordinate transformation and the maximum offset per update, then a coordinate transformation between the current actual coordinate transformation and the coordinate transformation based on the latest determination can be selected as the target transformation.

[0046] Also preferred is the following extension, in which the target coordinate transformation is determined as a predicted coordinate transformation taking into account the latest determined current actual coordinate transformation and at least B previously determined current actual coordinate transformations, where B≥2. As a result, the quality of the at least one object produced can be further improved in many applications. For each layer, the programming coordinate transformation of each scanner can be kept relatively close to the current actual coordinate transformation; usually, at each determined time point, the programming coordinate transformation is then also particularly close to the current actual coordinate transformation. By observing the current actual coordinate transformation at a large number of (known) recent time points, in many cases the coordinate transformation expected at a specific time point in the future ("predicted coordinate transformation") can be determined with good accuracy by model calculation. By converting the programming coordinate transformation into a predicted coordinate transformation in the period up to the aforementioned specific time point, the future development of the scanner's offset can be taken into account in advance. In particular, the predicted coordinate transformation can be determined using a model, for example a model known from a so-called state observer in control engineering. B is usually selected as B≥3 or B≥5 or B≥10. If the predicted coordinate transformation cannot be reached as the target transformation within the planned determination interval, for example because the initial deviation is too large for the planned number of updates and the maximum offset per update of the programmed coordinate transformation, then a coordinate transformation which lies between the predicted coordinate transformation and the coordinate transformation on which the latest determination is based can alternatively be selected as the target transformation.

[0047] According to a sub-variant of this embodiment, the predicted coordinate transformation is determined by means of a trend analysis. In this way, slow but uniform changes in the scanner orientation, such as usually occur due to heating processes during the processing of the load on the building platform, can be compensated quite accurately.

[0048] It is preferably provided that the trend analysis comprises a regression of the most recently determined current actual coordinate transformation with the at least B previously determined current actual coordinate transformations, where B≥2. The regression is particularly easy to perform. In particular, the regression can be a linear regression, a polynomial regression or an exponential regression. In the case of a polynomial regression, four times (fourth degree polynomial) or less (per coordinate direction) is usually sufficient, and three times or less is often sufficient, and in some cases even two times is sufficient.

[0049] In another sub-variant, it is provided that the target coordinate transformation is determined as the average value of the most recently determined current actual coordinate transformation and the at least B previously determined current actual coordinate transformations, where B≥2. In other words, it is assumed that the actual coordinate transformation fluctuates around the average value and that, as a prediction, the actual coordinate transformation will again approach the average value in the future. This approach is relatively simple. This approach produces good results if the current actual coordinate transformations involved fluctuate mainly statistically (for example due to non-systematically and rapidly changing external conditions or due to statistical measurement errors involved). It should be noted that a separate average value is usually determined for each of the at least N-1 scanners.

[0050] Particularly preferred is a variant in which the multiple updates between two successive determinations are each carried out after the production of an equal number of layers. This contributes to a particularly gentle supplementary adjustment of the programmed coordinate transformation and to a good quality of the produced objects.

[0051] Also preferred is the variant in which the multiple updates between two successive determinations are each carried out after the production of exactly one layer. In other words, the updates are only applicable to a single layer to be produced. This facilitates particularly precise post-regulation.

[0052] Further preferred is a variant in which the multiple updates between two successive determinations are evenly distributed over the layers produced between two successive determinations. In other words, the entire period between two successive determinations is used for supplementary adjustment of the programming coordinate transformation, and the updates are performed at equal (at least approximately equal) intervals. This can also contribute to making the supplementary adjustment particularly gentle. Preferably, A updates are performed between two determinations, where A≥2, wherein an update is performed after the production of M / A layers, respectively, where M / A is an integer. Usually, an equal (increasing) deviation portion is also provided per update for the correction of the programming coordinate transformation.

[0053] Particularly preferred is a variant in which the number M of layers produced between two successive determinations of the current actual coordinate transformation during the production of the at least one object is variable. This enables the frequency of the determination of the current actual coordinate transformation to be adapted and optimized to the requirements of the specific application, in particular with regard to the (predetermined or best possible) production accuracy and the (lowest possible or predetermined) ratio of non-productive time for measurement. If desired, M can be selected as a function of the temporal variation of measured values ​​(e.g. pressure or temperature) in the 3D printing system and / or in its environment; for example, if strong variations of the measured values ​​are observed (which are usually accompanied by a significant offset of the scanner), M should be selected to be smaller than in the case of small variations.

[0054] In a further preferred development of this variant, the number of layers M produced or the moving average Mgld of the number of layers M produced between two successive determinations at the start of the layer-by-layer production of the at least one object is selected to be lower than in the subsequent course of the layer-by-layer production of the at least one object. Experience has shown that at the start of production (i.e. after a new building platform is provided in the feed chamber) a rocking process first still occurs, which can be expected to cause significant changes in the orientation of the scanner, so it is advisable to control the coordinate transformation more frequently for accurate supplementary adjustment. In the subsequent course of production, usually when a predetermined proportion of layers (e.g. selected as between 10% and 35% of the layers) or a predetermined minimum number of layers (e.g. selected as 100 or more, preferably 200 or more) has been produced or layers have been produced on the building platform for a predetermined minimum time (e.g. selected as between 1 h and 3 h) since the start of production, only small changes still occur, which can be supplemented well by less frequent control of the coordinate transformation. The moving average Mgld can be formed, for example, over 4 or more, preferably 8 or more, determination intervals and further, for example, over 24 or fewer, and preferably 12 or fewer, determination intervals (i.e., values ​​of M, typically the most recent values ​​of M).

[0055] Also preferred is an embodiment in which the number M of layers to be produced between the last determination of the current actual coordinate transformation and the next determination is selected as a function of how large the respective adjacent deviations are between the last determined current actual coordinate transformation and the last previous determination of the current actual coordinate transformation for the at least N-1 scanners. This makes it possible to react in a flexible manner to disruptive influences occurring during production.

[0056] In a preferred sub-variant of this development, the number M of layers to be produced is selected to be smaller for the at least N-1 scanners, the greater the adjacent deviations. The proposed procedure makes it possible to keep the changes in the programmed coordinate transformations per determined interval small without selecting the determined intervals to be unnecessarily short. Typically, the numerical adjacent deviations of the coordinate transformations for the at least N-1 scanners are averaged, or the respective largest deviation occurring in the N-1 scanners is determined and used for the size estimation.

[0057] Preferred is the following variant, in which

[0058] Select one of the N scanners as the guide scanner,

[0059] The scanner coordinate system of the guide scanner or another coordinate system having a fixed relationship with the scanner coordinate system is selected as the reference coordinate system.

[0060] And between two successive determinations of the current actual coordinate transformation, only the remaining N-1 scanners are repeatedly updated with the programmed coordinate transformation. For the guide scanner in the first alternative, the exposure data in the scanner coordinate system simultaneously corresponds to the exposure data in the reference coordinate system, so that the associated programmed coordinate transformation of the guide scanner is confirmed to remain the same by the selection of the guide scanner. In the second alternative, the scanner coordinate system of the guide scanner is in a fixed relationship (for the feed of the build platform) with another coordinate system (reference coordinate system), so that the exposure data from the other coordinate system is transformed into the scanner coordinate system using a fixed programmed coordinate transformation. For example, the other coordinate system may include a machine tool coordinate system or be the same as the machine tool coordinate system. The programmed coordinate transformation of the guide scanner is not updated between the two determinations (e.g., remains the same). The coordinate transformations of the remaining N-1 scanners are defined relative to the guide scanner or relative to the scanner coordinate system of the guide scanner, and the coordinate transformations of the remaining N-1 scanners are respectively updated multiple times between the two determinations. This approach is particularly simple and reliable. Perturbations of the measuring system generally affect the measuring points of all scanner systems (guide scanner and other scanners) equally and therefore have no effect on the (relative) supplementary adjustment of the remaining N-1 scanners. However, it should be noted that in the case of this approach, possible changes in the relative orientation of the guide scanner with respect to the building platform during the production of the at least one object remain unknown and no supplementary adjustment is performed. It should also be noted that in a 3D printing system, within the scope of the present invention, multiple groups of N scanners (where N≥2, where N can be different for different groups) can also be provided, where each group includes its own guide scanner and the coordinate transformations of the remaining scanners of each group are supplementarily adjusted relative to the associated guide scanner or relative to the reference coordinate system of the associated guide scanner.

[0061] In an alternative variant, the reference coordinate system is the machine coordinate system of a processing machine comprising N scanners and a building platform, and multiple updates of the programmed coordinate transformations of the N scanners are performed between two successive determinations of the current actual coordinate transformation. In the case of this approach, orientation changes of all scanners can be identified and compensated by updating the programmed coordinate transformations of the scanners. However, in this case, the measuring system on the processing machine should be particularly stable and should not be subject to significant disturbances that could distort the determination of the measuring points in the machine coordinate system. The machine coordinate system is usually defined directly or indirectly by the position of a machine tool component, wherein the machine tool component is stationary on the machine tool or is at least moved to a defined (usually always the same within the range of positioning accuracy) position for the corresponding determination, for example using a feeder. In particular, the machine coordinate system can be defined via a structure scanned by a scanner during measurement, in particular via a retroreflector (e.g. a reflector sphere), or also via a measurement structure generated by the machine tool component, such as a light cross of an alignment laser.

[0062] Preferred is a variant in which the programmed coordinate transformation and the current actual coordinate transformation only include shift information in two orthogonal directions. The two orthogonal directions (x, y) usually lie in the plane of the building platform. The determination of the shift information can be realized relatively easily and is usually sufficient in practice, since in practice a rotation of the scanner usually does not occur at all or only to a small extent. A shift is also called an offset.

[0063] In an alternative variant, the programmed coordinate transformation and the current actual coordinate transformation include shift information in two orthogonal directions and rotation information in a plane spanned by the two orthogonal directions. The two orthogonal directions (x, y) are usually located in the plane of the building platform, and the rotation occurs about a direction (z) orthogonal to the two orthogonal directions. By taking into account the rotation of the scanner, the programmed coordinate transformation can be supplemented with particular accuracy.

[0064] Also within the scope of the invention is a system for producing at least one object layer by layer on a building platform by locally solidifying a powdered material in individual layers, the system comprising: a building platform; N scanners, where N ≥ 2; and a control device, the system being designed to carry out the method according to the invention as described above. With the system according to the invention, the at least one object can be produced with high quality and short non-productive times in the corresponding feed of the building platform using a plurality of scanners or associated high-energy beams.

[0065] Also within the scope of the invention is a computer program product which, when used on a system for producing at least one object layer by layer on a building platform by locally solidifying powdered material in individual layers, performs the method according to the invention as described above.

[0066] Further advantages of the present invention are apparent from the description and the accompanying drawings. Similarly, the features mentioned above and the features yet to be described can all be used alone or in any desired combination according to the present invention. The embodiments shown and described should not be understood as an exhaustive list, but rather as exemplary features for describing the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] The invention is explained in more detail based on exemplary embodiments and is shown in the drawings.

[0068] Figure 1 shows a schematic cross-section of an exemplary system on which the method according to the invention can be performed and with which an object can be produced on a building platform;

[0069] Figure 2 The production of a layer of the object to be produced is explained for the present invention with the aid of a flow chart;

[0070] Figure 3 A flow chart is shown for a first variant of the method according to the invention, in which the target coordinate transformation corresponds to the most recently determined current actual coordinate transformation;

[0071] Figure 4 A flow chart showing a second variant of the method according to the invention, in which the target coordinate transformation corresponds to the predicted coordinate transformation;

[0072] Figure 5 Four further variants of the method according to the invention are explained with the aid of four layer diagrams, which relate to variants of the determination after how many layers have been produced and to variants of the number and distribution of updates between two determinations;

[0073] Figure 6 The execution of a seventh variant of the method according to the invention is shown in an exemplary manner with a layer diagram, an example calculation and a selection table;

[0074] Figure 7a The determination of the predicted coordinate transformation by means of polynomial regression is shown by means of a graph for an eighth variant of the method according to the invention;

[0075] Figure 7b The determination of the predicted coordinate transformation by means of linear regression is shown by means of a graph for a ninth variant of the method according to the invention;

[0076] Figure 8 The determination of the predicted coordinate transformation by means of averaging is shown with the aid of a diagram for a tenth variant of the method according to the invention. DETAILED DESCRIPTION

[0077] Figure 1 A system 1 according to the invention is schematically shown, comprising a processing machine 1 a and a control device 10. The system 1 is used to produce a three-dimensional object 2 from a powdery material 3. A computer program product for executing the method according to the invention can be executed on the system 1 .

[0078] The processing machine 1a comprises a processing chamber 4 in a housing. With respect to the housing, a cover 5a with further components (see below) and a base plate 5b are schematically shown here, which delimit the processing chamber 4. Here, the processing chamber 4 is also delimited by side walls and a rear wall (not shown in detail). The processing chamber 4 can be accessed via an access door (not shown in detail).

[0079] The production process takes place on a building platform 6. The building platform 6 can be lowered relative to the base plate 5b along an axis in the Z direction in order to gradually place new layers of powdery material 3 on the building platform 6 for the layered production of the object 2; the application of the powdery material 3 is performed by a feeder not shown in detail. Figure 1 , the building platform 6 has been lowered relative to the base plate 5b.

[0080] In order to produce an object 2 (in this case a prototype 2a), the uppermost layer 7 of the powdery material 3 is irradiated with two high-energy light beams 8a, 8b (in this case laser beams), which are directed onto the uppermost layer 7 by scanners 9a, 9b (in this case two scanners). The scanners 9a, 9b here comprise flippable mirrors (not shown in detail). The high-energy light beams 8a, 8b are directed to predetermined positions in the powdery material 3. The high-energy light beams 8a, 8b are absorbed by the powdery material 3. The powdery material 3 melts and solidifies again when it is no longer irradiated. The solidified material then forms a further part of the object 2 to be produced.

[0081] The two scanners 9a, 9b are connected to a control device 10. To expose the corresponding layer 7, the control device 10 provides each scanner 9a, 9b with exposure data of the processing pattern in a reference coordinate system. The control device 10 can determine the exposure data based on the CAD data of the object 2 to be produced. The scanners 9a, 9b are each assigned a scanner coordinate system. Figure 1 In FIG. 1 , the reference coordinate system corresponds to the scanner coordinate system of the scanner 9 a. The scanner 9 a is then referred to as a guide scanner 11. Alternatively, the machine coordinate system of the processing machine 1 a may also be selected as the reference coordinate system.

[0082] In the control device 10, the exposure data in the reference coordinate system is transformed into the exposure data in the scanner coordinate system of the corresponding scanner by means of a programmed coordinate transformation. For the scanner coordinate system of the guide scanner 11, the exposure data in the reference coordinate system and the exposure data in the scanner coordinate system are the same. For the scanner coordinate system of the scanner 9b, the exposure data in the reference coordinate system is different from the exposure data in the scanner coordinate system. The exposure data in the corresponding scanner coordinate system is transferred to the associated scanners 9a, 9b. Then, the scanners 9a, 9b expose the processing pattern on the building platform 6 in the layer 7.

[0083] exist Figure 1 In the embodiment, the processing machine tool 1a also includes a monitoring device 12 (in this case a camera device 12a), which is used to monitor the production. The monitoring device 12 is connected to the control device 10. When producing the object 2, measurements are repeatedly performed using the camera device 12a. These repeated measurements are used to repeatedly determine the current actual coordinate transformation of the scanner 9b. Here, the current actual coordinate transformation of the scanner 9b may deviate from the programming coordinate transformation on which the associated determination is based, because during production, the environmental conditions and influences in the processing chamber 4 may change (for example, the processing chamber 4 may heat up during processing, or there may be changes in pressure differences or humidity). This may cause the scanner 9b or its scanner coordinate system to shift or rotate relative to the guide scanner 11. In order to correct these shifts or rotations, the programmed coordinate transformation of the scanner 9b is updated taking into account the current actual coordinate transformation of the scanner 9b.

[0084] It should be noted that in case the reference coordinate system is the machine tool coordinate system, the current actual coordinate transformation of all scanners (9a, 9b) is repeatedly determined and updated by repeated measurements.

[0085] Figure 2 A flow chart is shown which illustrates how a single layer of an object to be produced is produced in the context of the present invention. The production is controlled by means of a control device.

[0086] Initially, in step 100, the exposure of a layer of powdery material is planned in order to produce a layer of an object. To this end, it is determined which surface areas of the entire layer of powdery material need to be exposed to the high-energy beam of the scanner in order to produce the corresponding layer of the object (layer processing surface). The layer processing surface is essentially derived from the CAD data of the object and the position of the current layer in the object.

[0087] In the next step 101, the layer processing surface is distributed to the N scanners involved in producing the object. Therefore, N processing patterns are obtained for the N scanners. Usually, the N processing patterns are divided so that the processing time of each scanner is as long as possible. The following steps 102, 103, and 104 are performed separately for each of the N scanners.

[0088] In step 102 , first, exposure data of a corresponding processing pattern of a corresponding scanner in a reference coordinate system (target positions of beam points of a high-energy beam at multiple time points) are provided.

[0089] In the following step 103, the exposure data in the reference coordinate system are transformed into exposure data in the scanner coordinate system of the corresponding scanner. The transformation is performed using the programmed coordinate transformation of the scanner.

[0090] In step 104 , the exposure data in the scanner coordinate system is assigned to the corresponding scanner (transmitted for execution).

[0091] Then, in step 105, the layers on the building platform are exposed using N scanners, and the corresponding layers of the object are produced.

[0092] After the exposure is completed, production continues with the next layer (the process starts again at step 100).

[0093] In the context of the present invention, programmed coordinate transformations are updated in a specific manner, as described below.

[0094] Figure 3 A flow chart of a first variant of the method according to the invention is shown. If the reference coordinate system is defined as the machine tool coordinate system, the process is used for all N scanners. If the reference coordinate system is defined as the scanner coordinate system of the guide scanner, the process is used for N-1 scanners (i.e. all scanners except the guide scanner).

[0095] In step 200, in the variant shown, a layer has already been produced on the building platform and accordingly the object to be produced has already been partially produced. This can be described as "production up to now".

[0096] In the next step 201, the current actual coordinates of the scanner are transformed into MTKT AB In this case, the scanner is used in a measurement, wherein one or more measuring points are generated by means of a high-energy light beam, wherein known coordinates in a reference coordinate system are predetermined and are converted using a programmed coordinate transformation PKT on which the latest determination is based. AB The coordinates of the scanner coordinate system are transformed and assigned to the scanner. The actual position of the one or more measurement points ("actual coordinates") is then measured in the reference coordinate system. The programming coordinate transformation PKT can then be determined based on the newly determined AB Add the deviation (if any) between the given known coordinates and the measured actual coordinates to obtain the current actual coordinate transformation MTKT ABTypically, during production, due to changing ambient conditions in the processing room and the resulting offsets (displacements and possibly rotations) of the scanners relative to each other or also relative to the processing machine, the current actual coordinate transformation MTKT AB Converts coordinates different from those previously programmed to PKT AB .

[0097] In order to correct these offsets, a target coordinate transformation ZKT is determined in step 202. In the variant shown here, the target coordinate transformation ZKT is related to the most recently determined current actual coordinate transformation MTKT AB Therefore, the following formula applies:

[0098] ZKT=MTKT AB . (Formula 1)

[0099] It should be noted that a coordinate transformation includes the value of each coordinate direction (x, y) and may also include a rotation They can be considered separately, but for simplicity they are combined here into one formula notation.

[0100] Then, the initial deviation AUS is obtained in step 203. The initial deviation AUS is obtained by the following formula:

[0101] AUS=ZKT-PKT AB (Formula 2)

[0102] And explained the target coordinate transformation ZKT and programming coordinate transformation PKT AB How big is the deviation.

[0103] In the following step 204, the determined initial deviation AUS is divided into deviation parts ABW for A updates: i , where i = 1, ..., A and where i = update index. Then the following applies:

[0104]

[0105] That is, all deviation parts ABW i The sum of the initial deviation AUS results. Furthermore, the M layers to be produced until the next determination of the current actual coordinate transformation are divided into A updates, each with m i A production blocks of layers. Then the following applies:

[0106]

[0107] In step 205.1, after the first update j=1 (where j=1, ..., A, and where j: index of the current update), a new programming coordinate transformation PKT1 is determined. The programming coordinate transformation PKT1 is obtained as follows:

[0108] PKT1=PKT AB + ABW1. (Formula 5)

[0109] The new programmed coordinate transformation PKT1 replaces the last valid programmed coordinate transformation (here PKT AB ) and stored in the control device. The programmed coordinate transformation PKT1 updated in this way is then used to produce the m1 layers belonging to the first updated production block.

[0110] The corresponding approach is used for further updates. Step 205.j shows the determination of the programming coordinate transformation PKT after the jth update. j And produce m j The general situation of the production block of each layer. The programming coordinate transformation PKT is obtained as follows j :

[0111] PKT j = PKT j-1 + ABW j (Formula 6)

[0112] This is equivalent to

[0113]

[0114] Programming coordinate transformation PKT j is saved, and then produces m belonging to the jth updated production block j is applied when a layer is created.

[0115] In step 205.A, the programming coordinate transformation PKT is determined after the last update j=A A . The programming coordinate transformation PKT is obtained by the following formula: A :

[0116] PKT A = PKT A-1 + ABW A (Formula 8)

[0117] This is equivalent to

[0118]

[0119] Programming coordinate transformation PKT A is saved and in the production of m of the production block belonging to the last update j = A A is applied when a layer is created.

[0120] This is followed by step 206, in which the next determination of the current actual coordinate transformation is performed. Subsequently, further layers can be produced and the programmed coordinate transformation can be updated as described in steps 202 to 205.A, and further determinations of the current actual coordinate transformation can be performed, etc. This is summarized in step 207. Production continues, i.e. steps 201 to 205.A are repeated, until the object is completed. In step 208, the processing of the object to be produced on the building platform is ended.

[0121] During the flow of the first variant of the method according to the invention, a plurality of updates (number A) of the programmed coordinate transformation PKT are performed between two consecutive determinations. This divides the accumulated offsets of the scanner into small individual corrections and allows a gentle supplementary adjustment of the programmed coordinate transformation PKT of the scanner. This can improve the quality of the objects to be produced. The most recently determined current actual coordinate transformation MTKT is made AB Equating with the target coordinate transformation ZKT is easy to implement and results in objects to be produced with good quality.

[0122] Figure 4 A flow chart of a second variant of the method according to the invention is shown. If the reference coordinate system is defined as the machine tool coordinate system, the process is used for all N scanners. If the reference coordinate system is defined as the scanner coordinate system of the guide scanner, the process is used for N-1 scanners (i.e. all scanners except the guide scanner). The process of the second variant largely corresponds to the process of the first variant (see Figure 3 ), so only the main differences are described here.

[0123] In step 300, a layer has been produced on the build platform and the object to be produced has been partially produced. This can be described as "production so far". During the production so far, a number of current actual coordinate transformations of the scanner have been determined, see step 300.a. The previously determined B (where B≥2) current actual coordinate transformations MTKT ZB,k Together with the time points at which they were determined, B is stored in the control device, where k=1, . . . , B and k: counting index.

[0124] In the next step 301, the current actual coordinate transformation MTKT of the scanner is performed. AB The latest determination.

[0125] In step 302 , the target coordinate transformation ZKT is determined. In the variant shown here, for this purpose, a predicted coordinate transformation PNSKT is first determined. The predicted coordinate transformation is the most recently determined current actual coordinate transformation MTKT AB and other previously determined current actual coordinate transformations MTKTZB,k A function of k = 1, ..., B (previous step 300.a). Therefore, the predicted coordinate transformation PNSKT is determined as follows:

[0126] PNSKT = f(MTKT AB ,MTKT ZB,1 ,…,MTKT ZB,B ). (Formula 10)

[0127] In this respect, f denotes a prediction function. The prediction coordinate transformation PNSKT can be performed, for example, with the aid of a trend analysis (e.g. regression, see Figure 7a , Figure 7b ) or by determining the mean value (see Figure 8 ) is obtained. Then the target coordinate transformation ZKT is equal to the predicted coordinate transformation PNSKT. Therefore, the following formula applies:

[0128] ZKT = PNSKT. (Equation 11)

[0129] The target coordinate transformation ZKT determined in this way can then be used until the next determination of the current actual coordinate transformation 306 in the programming coordinate transformation based on the value PKT j When the Ath update is made, the production is respectively j A production blocks consisting of A layers, where j=1, ..., A, as described in steps 303 to 305. A. Steps 303 to 308 correspond to steps 203 to 208 (see above).

[0130] During the flow of the second variant of the method according to the invention, multiple updates of the programmed coordinate transformation PKT are performed between two successive determinations. This allows the accumulated offsets of the scanner as well as the expected offsets to be divided into small individual corrections and the programmed coordinate transformation PKT of the scanner to be adjusted softly and supplementally. This can improve the quality of the objects to be produced. Equating the predicted coordinate transformation PNSKT with the target coordinate transformation ZKT allows effective countermeasures to be taken against future developments of the scanner offsets. A good quality of the objects to be produced can be achieved.

[0131] Figure 5 A further variant of the method according to the invention is shown, which uses a different layer map for a single scanner. The layer map is plotted against time t and shows the index number (Arabic numeral) of the corresponding layer produced between two determinations of the current actual coordinate transformation (determinations are shown as dashed lines) and the number and distribution of updates of the programmed coordinate transformation performed between the two determinations (updates are shown as dotted lines).

[0132] It should be understood that only characteristic determined intervals are shown in each case, rather than the entire production in the feed of the building platform.

[0133] Sub-figure a) A third variant of the method according to the invention is shown. The current actual coordinate transformation is determined at the time points t1a, t2a and t3a. Between the time points t1a and t2a and between the time points t2a and t3a, five layers (M=5) of the object to be produced are each produced. An update is performed after each produced layer; therefore, five updates are performed between the time points t1a and t2a and between the time points t2a and t3a.

[0134] After the current actual coordinate transformation has been determined at time t1a, the initial deviation is determined and divided into five deviation parts of equal size, which are divided into the next five updates determined up to time t2a. The first update of the programmed coordinate transformation is then carried out and layer 1 is produced. Between layer 1 and layer 2, the programmed coordinate transformation is updated again and layer 2 is produced. Layers 3 to 5 are produced in the same way. After layer 5 has been produced, the current actual coordinate transformation is determined at time t2a.

[0135] The production of the first to fifth layers between the time points t2a and t3a is performed in the same manner as the production of the first to fifth layers between the time points t1a and t2a.

[0136] In this variant, the updating of the programmed coordinate transformation takes place at equal intervals with particularly small steps.

[0137] Sub-figure b) A fourth variant of the method according to the invention is shown. The current actual coordinate transformation is determined at the time points t1b, t2b and t3b. Eight layers (M=8) are produced between the time points t1b and t2b and between the time points t2b and t3b. Here, the layers are respectively divided into production blocks with two layers per production block. An update is performed after the production of a production block; therefore, four updates are performed between the time points t1a and t2a and between the time points t2a and t3a.

[0138] After the current actual coordinate transformation is determined at time t1b, the initial deviation is determined and divided into four deviation parts of equal size, which are divided into the next four updates determined until time t2b. The first update of the programmed coordinate transformation is then carried out and layers 1 and 2 are produced. Between layers 2 and 3, the programmed coordinate transformation is updated again and layers 3 and 4 are produced. Layers 5 to 8 are produced in the same way. After the production of layer 8, the current actual coordinate transformation is determined at time t2b.

[0139] The production of the 1st to 8th layers between the time points t2b and t3b is performed in the same manner as the production of the 1st to 8th layers between the time points t1b and t2b.

[0140] In this variant, the updates are also performed at the same intervals, but less frequently than in the variant in sub-figure a).

[0141] Sub-figure c) A fifth variant of the method according to the invention is shown. In the variant shown here, the production of the object has just begun. The current actual coordinate transformation is determined at the time points t1c, t2c, t3c and t4c. Five layers are produced between the time points t1c and t2c and between the time points t2c and t3c, respectively. Eight layers are produced between the time points t3c and t4c. An update is performed after each produced layer; five updates are performed between the time points t1c and t2c and between the time points t2c and t3c, respectively, and eight updates are performed between the time points t3c and t4c.

[0142] After the current actual coordinate transformation has been determined at time t1c, the initial deviation is determined and divided into five deviation parts of equal size, which are divided into five updates until the next determination (determination interval BI1) at time t2c. The first update of the programmed coordinate transformation is then carried out and layer 1 is produced. Between layer 1 and layer 2, the programmed coordinate transformation is updated again and layer 2 is produced. Layers 3 to 5 are produced in the same way. After the production of layer 5, the current actual coordinate transformation is determined at time t2c.

[0143] The production of the 1st to 5th layers between the time points t2c and t3c (determination interval BI2) is performed in the same manner as the production of the 1st to 5th layers between the time points t1c and t2c.

[0144] In the further course of layered production, between the times t3c and t4c (determination interval BI3), the number of produced layers is increased to eight layers. After the current actual coordinate transformation is determined at the time t3c, the initial deviation is determined and divided into eight deviation parts of equal size, which are divided into the next determined eight updates up to the time t4c. The first update of the programmed coordinate transformation is then carried out and the 1st layer is produced. Between the 1st and 2nd layer, the programmed coordinate transformation is updated again and the 2nd layer is produced. Layers 3 to 8 are produced in the same way. After the 8th layer has been produced, the current actual coordinate transformation is determined at the time t4c.

[0145] Here, determination intervals BI1 and BI2 represent early (first processed), shorter determination intervals, and determination interval BI3 represents late (later processed), longer determination intervals. Processing the entire feed height of the building platform typically involves at least 20 determination intervals, and often involves at least 40 determination intervals.

[0146] In this variant, a more accurate and faster supplementary adjustment of the programmed coordinate transformation can be achieved at the beginning of the production of the object. Usually, at the beginning of the production, larger changes in the orientation of the scanner can be expected, which is why determinations are made more frequently at the beginning in order to control these changes and to offset them as quickly and accurately as possible. In most cases, the later determination interval is at least 1.5 times longer than the earlier determination interval, and usually at least 2 times longer or even at least 3 times longer than the earlier processing interval.

[0147] Sub-figure d) A sixth variant of the method according to the invention is shown. The current actual coordinate transformation is determined at the time points t1d and t2d. 16 layers are produced between the time points t1d and t2d. An update is only performed at the beginning, that is, before the production of the first to fifth layers, respectively, that is, a total of five updates.

[0148] After the current actual coordinate transformation is determined at time t1d, the initial deviation is determined and divided into five deviation parts of equal size, which are divided into five updates. The first update of the programmed coordinate transformation is then carried out and layer 1 is produced. Between layer 1 and layer 2, the programmed coordinate transformation is updated again and layer 2 is produced. Layers 3 to 5 are produced in the same way. After the production of layer 5, layers 6 to 16 are produced without further updates before the next determination of the current actual coordinate transformation is carried out at time t2d.

[0149] In all the variants shown above, the initial deviation between the two determinations is always divided into deviation parts of equal size. Alternatively, it is also possible to select deviation parts of different sizes between the two determinations (not shown in detail). It should be noted in this regard that the deviation parts must not be selected to be too large, so as to avoid the formation of steps or offsets when producing the object to be produced.

[0150] Figure 6 The execution of a seventh variant of the method according to the invention is shown in an exemplary manner. In this variant, a correspondingly determined current actual coordinate transformation is made equal to a target coordinate transformation.

[0151] Sub-figure a) shows a single scanner layer map; the diagram is similar to Figure 5 (See above).

[0152] The current actual coordinate transformation is determined at the time points t1, t2, t3 and t4. Ten layers are produced between the time points t1 and t2 and between the time points t3 and t4, respectively, and five layers are produced between the time points t2 and t3. An update is performed after each produced layer; therefore, ten updates are performed between the time points t1 and t2 and between the time points t3 and t4, respectively, and five updates are performed between the time points t2 and t3.

[0153] In the variant shown here, the number of layers produced between the determinations (of the current actual coordinate transformation) is determined using the adjacent deviations of the current actual coordinate transformation between adjacent time points t1, t2, t3, t4. In this respect, the adjacent deviations are determined using the offset values ​​of the coordinate transformation. The detailed approach for this is described in more detail in sub-figures b) and c).

[0154] Sub-figure b) The currently determined actual coordinate transformation and the programmed coordinate transformation based on which the determination is based are listed for the selected example at the time points t1 to t4, respectively. Also listed are the corresponding adjacent deviations and offset values ​​between the time points t1 and t2, t2 and t3, and t3 and t4, respectively. In the variant shown here, the corresponding adjacent deviations NABW tg / h It is determined as follows:

[0155] NABW tg / h = MTKT th – MTKT tg (Formula 12)

[0156] where tg and th are adjacent time points, and where MTKT tg and MTKT th : Determination of the current actual coordinate transformation at time points tg and th; h and g are time indices, where h in this case is: 2,…,4 and g is: 1,…,3.

[0157] In the variant shown here, the coordinate transformation or the associated shift information in the direction x and in the direction y as well as the adjacent deviations are taken into account. The x and y directions are orthogonal to each other and lie in the plane of the building platform. The offset value BV of the adjacent deviation between adjacent time points tg and th is tg / h is determined as follows (using the Pythagorean theorem):

[0158]

[0159] Sub-figure c)A table showing the selection criteria for the selected examples is presented. The determination interval is adapted according to the offset value, that is, the interval between the layers is determined. Here, the selection criteria are chosen such that: for BV ≤ 5 μm, 15 layers are produced between two determinations; for 5 μm < BV ≤ 15 μm, ten layers are produced between two determinations; and for BV > 15 μm, five layers are produced between two determinations.

[0160] Three examples will be used to further illustrate this variant of the method according to the present invention.

[0161] Example 1:

[0162] Between time points t1 and t2, ten layers have been produced and ten updates of the programmed coordinate transformation have been performed, as can be seen in subfigure a); this was confirmed without applying the table of subfigure c). The current actual coordinate transformation MTKT at time point t1 t1 and the current actual coordinate transformation MTKT at time point t2 t2 yield an adjacent deviation NABW with an x - coordinate of x = 12 μm (such as 17 μm – 5 μm = 12 μm) and a y - coordinate of y = 18 μm (such as 25 μm - 7 μm = 18 μm). t1 / 2 Then the offset value BV between time points t1 and t2 t1 / 2 is BV t1 / 2 = 21.6 μm (such as ). This offset is rather large. According to the selection criteria in the table of subfigure c), this means that the interval between the layers produced between time points t2 and t3 is chosen to be only five layers. This allows for earlier further determinations. The determination interval of M = 5 layers is relatively small, such that only a very small offset may be accumulated before the next determination.

[0163] Example 2:

[0164] Between time points t2 and t3, five layers have been produced as previously confirmed and five updates of the programmed coordinate transformation have been performed, as can be seen in subfigure a). The current actual coordinate transformation MTKT at time point t2 t2 and the current actual coordinate transformation MTKT at time point t3 t3 yield an adjacent deviation NABW with an x - coordinate of x = - 8 μm (such as 9 μm – 17 μm = - 8 μm) and a y - coordinate of y = - 7 μm (such as 18 μm – 25 μm = - 7 μm). t2 / 3 Then the offset value BV between time points t2 and t3 t2 / 3 is BV t2 / 3 = 10.6 μm (such as ). According to the selection criteria in the table in sub-diagram c), this means that the interval of the layers produced between the time points t3 and t4 is selected to be ten layers. This means that more layers are produced than in the first example before further determinations are made.

[0165] Example 3:

[0166] Between the time points t3 and t4, ten layers have been produced as previously confirmed and ten updates of the programmed coordinate transformation have been performed, as can be seen in sub-figure a). The current actual coordinate transformation MTKT at the time point t3 t3 Transform MTKT with the current actual coordinates at time point t4 t4 The comparison of the adjacent deviations NABW with x coordinate x = 2 μm (e.g. 11 μm–9 μm = 2 μm) and y coordinate y = -3 μm (e.g. 15 μm–18 μm = -3 μm) t3 / 4 The offset value BV between time points t3 and t4 is t3 / 4 BV t3 / 4 =3.2μm (such as ). According to the selection criteria in the table in sub-figure c), this means that the interval of the layers produced between time t4 and a later time (not shown in greater detail) is selected to be 15 layers.

[0167] This variant of the method according to the invention allows flexible adaptation to different situations when producing the object to be produced. In the case of larger adjacent deviations, the determination interval is shortened so that the production can be monitored more accurately; and in the case of smaller adjacent deviations, the determination interval is increased so that the measurement requires less time and the production can be completed more quickly.

[0168] Figure 7a The determination of the prediction coordinate transformation PNSKT by means of polynomial regression is shown for an eighth variant of the method according to the invention.

[0169] In the figure, the x coordinate of the current actual coordinate transformation MTKT is plotted against the number of layers produced (the number of layers produced corresponds to the time coordinate). The figure shows three determination points or measurement points (solid circles), i.e. two previously determined current actual coordinate transformations MTKT. ZB1 and MTKT ZB2 And the latest determined current actual coordinate transformation MTKT AB Here, the number of produced layers is the same between all measuring points. A polynomial regression is performed and a second-degree polynomial is fitted to the measuring points as a regression curve. The regression curve can be used to determine a predicted coordinate transformation PNSKT (see open circles) at a future point in time or after a future production layer SZ.

[0170] It is noted that the determination of the predicted coordinate transformation PNSKT applies not only to the x component but also to the y component and, respectively, to the possible rotation component of the current actual coordinate transformation MTKT (not shown in detail).

[0171] Figure 7b The determination of the prediction coordinate transformation PNSKT by means of linear regression according to a ninth variant of the method according to the invention is shown. Figure 7a The main differences in the variants.

[0172] A linear regression is performed and a regression curve in the form of a regression line is fitted to the measurement points. This allows the predicted coordinate transformation PNSKT to be determined at a selected future time or after the future production of a selected slice SZ.

[0173] Figure 8 The determination of the predicted coordinate transformation PNSKT by means of averaging is shown according to a tenth variant of the method according to the invention. Figure 7a The main differences in the variants.

[0174] From the three measuring points, an average value is determined: the average value is marked with a horizontal straight line. It is assumed that the current actual coordinate transformation will fluctuate around this average value or return to this average value. From this, a predicted coordinate transformation PNSKT (see open circle) can be determined at a selected future time point or after the future production of a selected slice SZ.

[0175] Reference numerals list

[0176] 1 System

[0177] 1a Processing machine tools

[0178] 2 Objects

[0179] 2a Prototype

[0180] 3 Powdered materials

[0181] 4 Processing room

[0182] 5a Cover

[0183] 5b Base Plate

[0184] 6. Build the Platform

[0185] 7th floor (top floor)

[0186] 8a, 8b High Energy Beam

[0187] 9a, 9b Scanner

[0188] 10 Control Device

[0189] 11 Guide Scanner

[0190] 12 Monitoring Devices

[0191] 12a Camera device

[0192] A is the number of updates between two determinations

[0193] AUS Initial Deviation

[0194] ABW i The i-th deviation part

[0195] B The number of previously determined current actual coordinate transformations used for prediction

[0196] BI1-BI3 Determine the interval

[0197] BV tg / h The offset value of the current actual coordinate transformation between adjacent time points tg and th

[0198] g Index

[0199] h Index

[0200] i Index

[0201] j Index

[0202] k index

[0203] M is the number of layers produced between two determinations.

[0204] m i The number of layers in the production block after update i / after the i-th update

[0205] MTKT current actual coordinate conversion

[0206] MTKT AB The most recently determined current actual coordinate transformation

[0207] MTKT tg Current actual coordinate transformation at time point tg

[0208] MTKT th The current actual coordinate transformation at time point th

[0209] MTKT ZB,k The current actual coordinate transformation previously determined

[0210] N Number of scanners and number of high-energy beams

[0211] NABW tg / h Time point t g With t hThe adjacent deviation between the current actual coordinate transformation

[0212] PKT Programming Coordinate Transformation

[0213] PKT1 Programming coordinate transformation after 1 update

[0214] PKT A Update the programmed coordinate transformation after A times

[0215] PKT AB The programmed coordinate transformation on which the latest determination is based

[0216] PKT j Update the programmed coordinate transformation after j times

[0217] PNSKT Prediction Coordinate Transformation

[0218] SZ The layer selected for future production

[0219] t time

[0220] Time points t1-t4

[0221] t1a-t3a time point

[0222] t1b-t3b time point

[0223] t1c-t4c time point

[0224] T1d, T2d time points

[0225] tg time point

[0226] th time point

[0227] x direction

[0228] y direction

[0229] Z direction

[0230] ZKT Target Coordinate Transformation

Claims

1. A method for producing at least one object (2) layer by layer on a building platform (6) by locally solidifying a powdered material (3) in individual layers (7), in, At least in a plurality of layers (7), N high-energy light beams (8a, 8b) are used at least temporarily simultaneously by N scanners (9a, 9b), wherein N ≥ 2, Each scanner (9a, 9b) is assigned a scanner coordinate system. The control device (10) controls the exposure of each layer (7) for each scanner (9a, 9b). - providing (102) exposure data of the processing pattern in a reference coordinate system; - transforming (103) the exposure data in the reference coordinate system into exposure data in the scanner coordinate system by means of a programmed coordinate transformation (PKT), and - assigning (104) the exposure data obtained in the scanner coordinate system to the associated scanner (9a, 9b) so that the scanner (9a, 9b) exposes (105) the machining pattern in the layer (7) on the building platform (6), wherein during the layer-by-layer production of the at least one object (2), measurements are repeatedly performed, with the aid of which the current actual coordinate transformation (MTKT) of at least N-1 scanners (9a, 9b) is determined in each case, wherein M layers (7) are produced between two successive determinations of the current actual coordinate transformation (MTKT), wherein M≥2; and wherein, during the production of the at least one object (2), the programmed coordinate transformation (PKT) for the at least N-1 scanners (9a, 9b) is updated (205.1, 205.j, 205.A; 305.1, 305.j, 305.A) taking into account the current actual coordinate transformation (MTKT), It is characterized in that Between two successive determinations (201, 206; 301, 306) of the current actual coordinate transformation (MTKT), a plurality of updates of the programmed coordinate transformation (PKT) of the at least N-1 scanners (9a, 9b) are performed.

2. The method according to claim 1, It is characterized in that In the current actual coordinate transformation (MTKT) for the at least N-1 scanners (9a, 9b) AB ) after each latest determination (201; 301) of the at least N-1 scanners (9a, 9b) - taking into account the current actual coordinate transformation (MTKT) of the most recently determined (201; 301) AB ) to obtain (202; 302) the target coordinate transformation (ZKT), - and utilizing the multiple updates of the programming coordinate transformation (PKT) between the latest determination (201; 301) and the next determination (206; 306), gradually transforming (205.1, 205.j, 205.A; 305.1, 305.j, 305.A) the programming coordinate transformation (PKT) to the target coordinate transformation (ZKT).

3. The method according to claim 2, It is characterized in that In the current actual coordinate transformation (MTKT) for the at least N-1 scanners (9a, 9b) AB ), for each of the at least N-1 scanners (9a, 9b), - Determine (203; 303) the target coordinate transformation (ZKT) and the programming coordinate transformation (PKT) on which the latest determination is based AB ) between the initial deviation (AUS); - dividing (204; 304) the determined initial deviation (AUS) into a plurality of deviation parts (ABW i ); - and changing stepwise (205.1, 205.j, 205.A; 305.1, 305.j, 305.A) the programmed coordinate transformation (PKT) in said plurality of updates for the layers (7) produced after said most recent determination (201; 301) of said current actual coordinate transformation until said next determination (206; 306), In which, with each update, the additional deviation part (ABW i ) is added to the last programmed coordinate transformation (PKT) applied by the control device (10).

4. The method according to claim 3, It is characterized in that The initial deviation (AUS) is divided into deviation parts of equal size (ABW i ).

5. The method according to any one of claims 3 or 4, It is characterized in that Each deviation part (ABW i ) is subject to maximum offset and / or maximum rotation limits.

6. The method according to any one of claims 3 to 5, It is characterized in that A updates are performed between two determinations (201, 206; 301, 306), wherein A=M is ​​selected, and Divide the initial deviation (AUS) into M equal deviation parts (ABW i ).

7. The method according to any one of claims 2 to 6, It is characterized in that The target coordinate transformation (ZKT) corresponds to the current actual coordinate transformation (MTKT) of the most recently determined (201) AB ).

8. The method according to any one of claims 2 to 6, It is characterized in that Considering the current actual coordinate transformation (MTKT) newly determined (301) AB ) and the current actual coordinate transformation (MTKT) of at least B previously determined (300.a) ZB,k ) obtains the target coordinate transformation (ZKT) as the predicted coordinate transformation (PNSKT), where B≥2.

9. The method according to claim 8, It is characterized in that The predicted coordinate transformation (PNSKT) is determined with the aid of a trend analysis.

10. The method according to claim 9, It is characterized in that The trend analysis includes the current actual coordinate transformation (MTKT) of the latest determination (301) AB ) and the current actual coordinate transformation (MTKT) of the at least B previously determined (300.a) ZB,k ), where B ≥ 2.

11. The method according to claim 8, It is characterized in that The target coordinate transformation (ZKT) is determined as the current actual coordinate transformation (MTKT) of the most recently determined (301) AB ) and the current actual coordinate transformation (MTKT) of the at least B previously determined (300.a) ZB,k ), where B≥2.

12. The method according to any one of the preceding claims, It is characterized in that The multiple updates between two successive determinations (201, 206; 301, 306) each produce an equal number (m i ) is performed after layer (7).

13. The method according to any one of the preceding claims, It is characterized in that The multiple updates between two successive determinations (201, 206; 301, 306) are each performed after the production of exactly one layer (7).

14. The method according to any one of the preceding claims, It is characterized in that The multiple updates between two successive determinations (201, 206; 301, 306) are evenly distributed over the layers (7) produced between the two successive determinations (201, 206; 301, 306).

15. The method according to any one of the preceding claims, It is characterized in that During the production of the at least one object (2), the number M of produced layers (7) between two successive determinations (201, 206; 301, 306) of the current actual coordinate transformation (MTKT) can be changed.

16. The method according to claim 15, It is characterized in that At the beginning of the layer-by-layer production of the at least one object (2), the number M of layers (7) produced between two successive determinations (201, 206; 301, 306) or the moving average Mgld of the number M of layers (7) produced is selected to be lower than in the subsequent process of the layer-by-layer production of the at least one object (2).

17. The method according to any one of claims 15 or 16, It is characterized in that The number M of layers (7) to be produced between a most recent determination (201; 301) and a next determination (206; 306) of the current actual coordinate transformation (MTKT) is selected according to the following: for the at least N-1 scanners (9a, 9b), the most recent determination (201; 301) of the current actual coordinate transformation (MTKT AB ) and the current actual coordinate transformation determined before the latest determination (NABW) tg / h ) is big.

18. The method according to claim 17, It is characterized in that For the at least N-1 scanners, the adjacent deviation (NABW tg / h ) is larger, the smaller the number M of layers (7) to be produced is chosen.

19. The method according to any one of claims 1 to 18, It is characterized in that selecting one of the N scanners (9a, 9b) as a guide scanner (11), The scanner coordinate system of the guide scanner (11) or a coordinate system having a fixed relationship with the scanner coordinate system is selected as the reference coordinate system. And between two successive determinations (201, 206; 301, 306) of the current actual coordinate transformation (MTKT), a plurality of updates of the programmed coordinate transformation are performed only for the remaining N-1 scanners (9a, 9b).

20. The method according to any one of claims 1 to 18, It is characterized in that The reference coordinate system is a machine tool coordinate system of a machining center (1a) comprising the N scanners (9a, 9b) and the building platform (6), and multiple updates of the programmed coordinate transformation of the N scanners (9a, 9b) are performed between two successive determinations (201, 206; 301, 306) of the current actual coordinate transformation (MTKT).

21. The method according to any one of claims 1 to 20, It is characterized in that The programmed coordinate transformation (PKT) and the current actual coordinate transformation (MTKT) only include shift information in two orthogonal directions (x, y).

22. The method according to any one of claims 1 to 20, It is characterized in that The programmed coordinate transformation (PKT) and the current actual coordinate transformation (MTKT) include shift information in two orthogonal directions and rotation information in a plane spanned by the two orthogonal directions (x, y).

23. A system (1) for producing at least one object (2) layer by layer on a building platform (6), the production being carried out by locally solidifying a powdered material (3) in each layer (7), the system comprising a building platform (6), N scanners (9a, 9b) and a control device (10), wherein N ≥ 2, The system is designed to perform a method according to any one of claims 1 to 22.

24. A computer program product, which, when used on a system (1) for producing at least one object (2) layer by layer on a building platform (6), performs a method according to any one of claims 1 to 22, wherein the production is carried out by locally solidifying powdered material (3) in individual layers (7).