- 1 -

CALIBRATION METHOD AND CALIBRATION DEVICE OBTAINABLE BY

SAID METHOD.

DESCRIPTION

The present invention relates to a method for calibrating a machine for the manufacture of layered three-dimensional objects by exposure of an operating material to a light radiation adapted to solidify said material, where the light radiation is emitted by multiple light sources operating in succession or concurrently with each other.

The invention further relates to a calibration device obtainable by the application of the method to the aforesaid machine.

As is known, there are existing machines for the manufacture of three- dimensional objects by superposition of a plurality of layers obtained by selective solidification of corresponding layers of a liquid, pasty or powdery operating material. The aforesaid machines belong to the technical sector generally known by the term "additive manufacturing".

In the aforesaid machines, each layer of operating material is solidified exposing it to an appropriate light radiation and the various layers thus obtained are made to adhere to each other so as to obtain the complete three- dimensional object.

A particular case of additive manufacturing machines is represented by stereolithographic machines, which use as a basic material a liquid or pasty photopolymer, which is solidified by means of a light radiation, typically in the ultraviolet range, adapted to cause the selective polymerisation of the superficial layer of the photopolymer.

A different type of the aforesaid machines comprises those that use as an operating material a sinterable powder, for example metallic, and as a light radiation a radiation adapted to sinter the aforesaid powder, typically a laser beam of sufficiently concentrated energy. These machines operate according to similar principles to those of stereolithographic machines, with the exception of the way in which each layer of basic material to be solidified is formed, and of the type of light radiation employed.

For the sake of simplicity, the description that follows shall refer to stereolithographic machines. However, this specification should not be construed in a limiting sense, because it will be readily apparent that the description can be adapted, albeit with the due and obvious modifications, also to the other additive manufacturing machines.

It should also be specified that, although the present description refers to a light radiation, this specification should not be construed in a limiting sense. It is readily apparent that the present description could be applied, with the due and obvious modifications, to any machine for the manufacture of layered three-dimensional objects by solidification of a liquid, pasty or powdery basic material, which employs any radiation of a different nature from light radiation, provided it can be selectively conveyed on the operating material.

It is known that the light sources used in stereolithographic machines have limits in the superficial extension that can be exposed, as well as in manufacturing precision and speed.

To overcome the aforesaid limits, a known solution is to provide multiple light sources operating simultaneously on respective different work areas of the surface of the photopolymer. Mutually adjacent work areas are superposed so that the corresponding superposition area can be exposed to both corresponding light sources. This makes it possible to produce three- dimensional objects that extend on two or more adjacent work areas, using different light sources to produce different portions of the object that intersect each other, thereby being rigidly connected.

However, because of the limited installation precision of the aforesaid light sources, it is difficult for the light sources to be assembled so that the respective coordinate systems are perfectly aligned to each other. In particular, the installation precision is markedly lower than the one corresponding to the precision of each light source within the respective work area.

As a consequence of the aforesaid misalignment, a three-dimensional object extending on two different work areas, hence obtained from the cooperation of two different light sources, will have surface flaws at the interface between the two portions.

To avoid the aforesaid surface flaws, it is necessary to calibrate the light sources in such a way as to align their coordinate systems, thereby making them mutually consistent.

According to a known variant, the light sources irradiate on respective work areas that are entirely superposed on each other. This variant makes it possible to solidify different portions of each layer using different light beams, for example laser beams of different diameters, emitted by the different light sources.

Clearly, the variant just described also requires calibrating the light sources. To obtain the aforesaid calibration, one possibility is to produce one or more specimens by means of the stereolithographic machine to be calibrated. A first portion of each specimen is solidified by a light source and a second portion adjacent to the first is solidified by another light source.

In this way, the operator can measure the size of the flaws of each specimen at the junction area between the two portions and, with appropriate mathematical calculations, determine the relative displacements between the work areas that are necessary to align their coordinate systems.

The calibration method described above has the drawback of being inconvenient and poorly accurate, because it entails measuring very small surface flaws by means of instruments that are not specifically designed for this purpose.

In addition, execution of the aforesaid calibration method requires the operator to possess appropriate instruments and specific training, generally available only at the manufacturer of the stereolithographic machine. This entails the additional drawback that the aforesaid calibration method is relatively costly for the user of the stereolithographic machine.

The aforesaid calibration system has the further drawback of being slow, because it entails the stereolithographic production of specimens consisting of a high enough number of layers to allow observation of the surface flaws, where it is known that the duration of a stereolithographic process is proportional to the number of layers of the object.

As stated previously, similar calibration problems can occur in any machine for the production of layered three-dimensional objects that uses a plurality of light sources to solidify a layer of material in different area of a same work surface. The present invention is intended to overcome all the aforementioned drawbacks of known calibration methods.

In particular, an object of the present invention is to provide a calibration method for a machine for the layered production of three-dimensional objects that is provided with multiple light sources, the method allowing to align the coordinate systems of the different sources in such a way as to make them mutually consistent, in order to allow the production of three-dimensional objects that extend on multiple work areas but that are free of surface flaws in the area of intersection between the work areas.

Another object of the invention is to provide a calibration method that is simpler to be carried out than known calibration methods.

In particular, an object of the invention is to provide a calibration method that does not require measuring instruments or a specific training to be used.

Another object of the invention is to provide a calibration method that is faster than known calibration methods.

The aforesaid objects are achieved by a calibration method according to claim 1 .

Further detailed features of the method of the invention are specified in the respective dependent claims.

The objects of the invention are also achieved by a calibration device obtained by means of the aforesaid method, in accordance with claim 19.

Advantageously, the ease of the calibration method of the invention makes it suitable to be applied even by personnel who are not specifically trained, for example directly by the user of the machine instead of its manufacture, hence at lower costs with respect to known calibration methods.

Also advantageously, the rapidity of the calibration method of the invention limits the time necessary to make the machine operational.

The aforesaid purposes and advantages, together with others that will be mentioned farther on, will be readily apparent in light of the following description of some preferred embodiments of the invention, which is given by way of non-limiting indication with reference to the accompanying drawings, where:

- figure 1 schematically shows a plan view of a machine on which the calibration method of the invention is applicable;

- figure 2 is a lateral view of the machine of figure 1 ;

- figure 3 is a plan view of a calibration device obtainable with a machine according to figure 1 by the application of the method of the invention;

- figure 4 is a plan view of a part of the calibration device of figure 1 ;

- figures 5 to 7 are plan views of respective variants of the calibration device obtainable by the method of the invention;

- figure 8 provides a plan view of an operating representation of an additional variant of the calibration device obtainable by the method of the invention; - figure 9 shows the calibration device of figure 8 as it appears to an operator. The method of the invention concerns the calibration of a plurality of light sources belonging to an additive manufacturing machine, i.e. of the type adapted to produce layered three-dimensional objects by selective solidification of corresponding layers of an operating material in a liquid, pasty or powdery state. The operating material defines a work surface and its solidification takes place by selectively irradiating the aforesaid work surface with a light radiation emitted by the light sources. The light sources are adapted to illuminate corresponding work areas belonging to the work surface, the work areas being superposed in pairs at respective superposition areas. For the sake of simplicity, the description that follows will refer to a stereolithographic machine, which uses as the operating material a photopolymer, in liquid or pasty form, and a light radiation suitable to cause the polymerisation of the photopolymer itself, for example ultraviolet light. However, it will be immediately evident that the application of the method can be extended, with appropriate and obvious modifications, to other types of additive manufacturing machines, for example to those of the type using a powdery operating material that is sintered by exposure to a laser beam. Figure 1 schematically shows a stereolithographic machine, indicated in its entirety by the numeral 18, provided with a work surface 2 defined by four work areas 5 of square shape and with mutually equal sides, each of which can be exposed to a light beam 4 coming from a corresponding light source 3. The work areas 5 are indicated in figure 1 with different lines, to make them distinguishable. The profile of the aforesaid work surface 2 is defined by two mutually concurrent sides of each work area 5, each of which is aligned to one side of a corresponding work area 5 adjacent to the first so as to define a square shaped work surface 2. Each remaining side of the work area 5 intersects one of the aforesaid adjacent work areas so as to define a corresponding rectangular superposition area 6, which has an axis of symmetry that is parallel to the aforesaid side and coinciding with one of the two axis of symmetry Z of the work surface.

The following description will refer to the stereolithographic machine 18 that has just been described. However, as will be readily apparent from the description, the method of the invention is applicable to machines with work areas 5 having any shape, even different from each other, and in a number that can be different from four, provided at least two of them are mutually superposed.

In addition, the description that follows and the accompanying drawings refer to an embodiment in which the work areas 5 are only partially superposed on each other, so that the work surface 2 is larger than each work area 5 and, therefore, it is possible to obtain three-dimensional objects of relatively large size.

However, it is evident that the invention is applicable even if one or more work areas 5 are fully superposed on each other, as can take place in the case in which different light sources 3 that emit light beams 4 differing from each other, for example laser beams having different diameters, are used to irradiate the same areas of the work surface 2.

The lateral view of figure 2 shows that the light sources 3 are positioned inferiorly to the work surface 2. In this case, the photopolymer is set down on a support surface, not shown in the figure but known in itself, which is transparent to the light radiation emitted by the sources. In this case the work surface 2 coincides with the support surface and the light sources 3 cause the solidification of the layer of photopolymer immediately superior to the aforesaid support surface.

The method of the invention can also be applied to a variant of stereolithographic machine, not shown in the drawings but known in itself, which differs from the one just described because the light sources 3 are positioned above, instead of below, the work surface 2. In this case, the work surface 2 coincides with the free surface of the photopolymer and the light sources 3 cause the solidification of the layer of photopolymer immediately inferior to said free surface.

With regard to the light sources 3, they can be both of the vector type and of the projector type. In the first case, each source is adapted to direct selectively a light beam, for example a laser beam, on any point of the work area 5 by means of adjustable mirrors. In the second case, represented in figures 1 and 2, each light source 3 can comprise a matrix of mirrors, each movable independently from the others to reflect, or not to reflect, a corresponding portion of a light beam 4 towards a corresponding point of the work surface. In both cases, each light source 3 refers to a respective coordinate system, preferably but not necessarily Cartesian, which uniquely defines each point of the respective work area 5. For each pair of two superposed work areas 5, a corresponding configuration of mutual alignment is defined in which the respective coordinate systems are in a predefined spatial relationship with each other, so that the control system of the stereolithographic machine 18 can express the coordinates of any point of the superposition area 6 indifferently according to the coordinate system of any one of the two corresponding light sources 3. The alignment of the two work areas 5 allows to produce a three-dimensional object that partly extends on one of the two areas and partly on the other combining the action of the two light sources 3 and preventing the object from having discontinuities at the transition area between the two work areas 5. The alignment of the two work areas 5 also allows to use two different light sources 3 to produce different portions of the three-dimensional object positioned on the respective superposition area 6.

Since the installation precision of a machine of the aforesaid types is not sufficient to obtain the alignment of the respective work areas 5, it is necessary to calibrate the light sources 3 so as to obtain the aforesaid alignment.

The calibration method of the invention comprises providing a layer of basic material 1 at the work surface 2.

Preferably, the layer of the basic material is a corresponding layer of the operational material of the machine, which in the case of a stereolithographic machine is the photopolymer mentioned above. Advantageously, use of the operating material of the machine as basic material 1 makes it possible to apply the method of the invention exploiting the normal operation of the machine 18 and, hence, without any need for additional interventions by the operator. Use of the operating material as the basic material has additional advantages that will be readily apparent farther on.

It is evident that the method of the invention is suitable to be used on any other type of basic material, provided it is able to be modified selectively in the points of incidence of the light radiation. For example, the basic material can be any photosensitive material, for example the sensitive layer of a photographic plate or another material on which an image can be impressed by selective exposure to a light radiation, or else a material able to be etched in the points in which it is reached by the light radiation.

It will also be readily apparent that, in variants of the invention, the work surface 2 defined by the basic material may not coincide with the surface on which the light sources 3 project the light beam during the normal operation of the machine 18. In this case, the calibration can nevertheless be carried out, provided the relationship that ties the points of the two surfaces reached by the light radiation emitted by the light source according to the same direction is known.

The method further comprises determining the size of one or more relative displacements between the two work areas 5 of each pair, measured according to corresponding displacement directions X, such as to arrange the two work areas 5 in the corresponding alignment configuration. In other words, each relative displacement represents the misalignment between the two work areas 5 according to the corresponding direction of displacement.

The aforesaid relative displacements can comprise translations parallel to the work surface 2, rotations according to an axis perpendicular to the work surface 2, or combinations thereof.

The relative displacements of each pair of work areas 5 are determined through the following operations.

First of all, a plurality of test images 7 is projected on the work surface 2 of the basic material 1 , in the superposition area 6 corresponding to each pair of work areas 5 so as to obtain, on the basic material 1 , corresponding prints that can be read by an operator, as shown in figure 3 and, in more detail, in figure 4. Each test image 7, represented in detail in figure 5, comprises a first image portion 8, which is projected by one of the two light sources 3 corresponding to the two work areas 5 of the pair, while the second image portion 9 is projected by the other one of the two sources.

The two image portions 8, 9 represent, respectively, a first graduated scale 10 and a reading cursor 11 , or vice versa. The first graduated scale 10 and the reading cursor 11 are configured to indicate, in combination with each other, a corresponding relative displacement between the two portions according to a corresponding direction of measurement X and in a corresponding orientation of measurement.

Therefore, observing each test image 7 impressed on the basic material 1 , the operator is able to identify the corresponding relative displacement with no need for any measuring instrument. In fact, the calibration device comprising the aforesaid test images 7 impressed on the basic material 1 is in itself equivalent to a measuring instrument specifically designed to determine the misalignment between the work areas 5 of the stereolithographic machine 18. In this way, the purpose of providing a calibration method that is simpler than known methods is achieved.

The aforesaid method makes it possible to achieve the additional object of making calibration faster. To impress the test images 7 on the basic material 1 , a limited number of exposures is sufficient. For example, if the basic material 1 is the operating material, the aforesaid method can comprise the production of a reduced number of solid layers whose shape corresponds to the union of the aforementioned test images 7. The result is a laminar object formed by the union of the test images 7 and whose surface is impressed with the first graduated scales 10 and the corresponding reading cursors 11.

Preferably, the method of the invention comprises defining a measurement tolerance to measure the aforesaid relative displacement, and configuring the first graduated scale 10 and the reading cursor 11 so as to reproduce the two scales of a Vernier scale whose tolerance is the aforesaid measurement tolerance.

More precisely, the first graduated scale 10 comprises a plurality of main marks 12, while the reading cursor 11 comprises a plurality of secondary marks 14 that, together, define a second graduated scale 13. The main marks 12 and the secondary marks 14 of the two scales 10, 13 are configured so as to be all intersecting a same measurement straight line Y parallel to the aforesaid direction of measurement X when the pair of work areas 5 is in its alignment configuration. Moreover, a first main mark 12a and a first secondary mark 14a are arranged in such a way that, in the aforesaid alignment configuration, they intersect the measurement straight line Y in a same reference point 15.

It is hereby specified that the reference point 15 is not a point in the geometric sense of the term. The points that can be illuminated individually by the light sources 3 have non-zero minimum dimension and, hence, the marks 12 and 14, including the first marks 12a and 14a, have no smaller dimensions than the aforesaid minimum dimension. Therefore, the intersection of each first mark 12a, 14a with the measurement straight line Y corresponds to a segment and the aforesaid intersection point 15 is to be understood as the median point of the aforesaid segment.

The main marks 12 and the secondary marks 14 are so arranged that, in the aforesaid alignment configuration, to each main mark positioned on one side of the aforesaid reference point 15, separated from the first main mark 12a by a given number of intermediate main marks 12 and positioned at a given distance from the reference point 15, corresponds a secondary mark 14 positioned on the same side as the reference point 15, separated from the first secondary mark 14a by a number of intermediate secondary marks equal to the aforesaid number of intermediate main marks 12 and distanced from the reference point 15, along the measurement straight line Y, by a distance equal to the aforesaid given distance, minus a length equal to the product of the tolerance multiplied times the aforesaid number of intermediate secondary marks.

The graduated scales 10, 13 configured as has just been described enable the operator to determine the displacement between the two work areas 5 according to the aforesaid measurement straight line Y and along each orientation, with a tolerance equal to the measurement tolerance and for a maximum displacement equal to the measurement tolerance multiplied times the number of main marks 12 positioned, with respect to the first main mark 12a, on the side corresponding to the aforesaid orientation.

Therefore, advantageously, the aforesaid configuration makes it possible to measure the displacement with high precision, while remaining simple and immediate to read.

The determination of the aforesaid displacement takes place in a wholly similar manner to the one used with a Vernier scale.

First of all, the main mark 12 is identified along with the secondary mark 14 whose intersections with the measurement straight line Y are the closest to each other. Then, the number of main marks 12 by which the aforesaid main mark 12 is distanced from the first main mark 12a, including the aforesaid main mark 12 but excluding the first main mark 12a, is counted, or else the number of secondary marks 14 by which the aforesaid secondary mark 14 is distanced from the first secondary mark 14a, including the aforesaid secondary mark 14 but excluding the first secondary mark 14a, is counted; lastly, the number thus obtained is multiplied times the measurement tolerance, obtaining the displacement.

Preferably, the main marks 12 are uniformly distanced from each other along the measurement straight line Y by a predefined step P. Consequently, also the secondary marks 14 are uniformly distanced from each other by a distance equal to the aforesaid step P, minus the measurement tolerance.

Advantageously, in addition to simplifying the reading of the misalignment, use of the aforesaid predefined step P makes it possible to utilise the first graduated scale 10 to measure misalignments greater than the maximum displacement described above. The number of main marks 12 between the first main mark 12a and the first secondary mark 14a along the measurement straight line Y, excluding the first main mark 12a, corresponds to the misalignment expressed in number of steps P of the first graduated scale 10. For example, if the first graduated scale 10, 101 , 102 has a step P of 1 mm, the aforesaid number corresponds to the number of millimetres of misalignment.

Preferably, the number of main marks 12 positioned, with respect to the first main mark 12a, on the side corresponding to the measurement orientation, and the number of secondary marks 14 positioned on the same side, with respect to the first secondary mark 14a, are both equal at least to the ratio between the aforesaid step P and the measurement tolerance. This, advantageously, makes it possible to measure any misalignment between zero and the step P of the first graduated scale 10, 101 , 102, according to the measurement orientation, with a tolerance equal to the measurement tolerance.

The definition of the two graduated scales 10, 13 can take place in the following way. First of all, the measurement tolerance is defined, as well as the maximum displacement to be measured with said measurement tolerance. The first graduated scale 10 is defined so as to comprise, on each side of the first main mark 12a, a number of main marks 12 at least equal to the ratio between the aforesaid maximum displacement and the aforesaid measurement tolerance and so that the step P is greater, or preferably equal, to the aforesaid maximum displacement. The second graduated scale 13 is defined so as to comprise, on each side of the first secondary mark 14a, a number of secondary marks 14 equal to the aforesaid number of main marks 12 and so that the distance between the secondary marks 14 is in relation to the step P of the first graduated scale 10 as described above.

Figure 5 shows a calibration device in which the step P of the first graduated scale 10 is greater than the maximum displacement defined above. This calibration device comprises a first graduated scale 10, situated in the top part of the figure, with nine main marks 12 to the right of the one indicated with the digit "0", which is the first main mark 12a. The corresponding nine secondary marks 14 of the second graduated scale 13 include those positioned to the right of the mark indicated with the digit "0", which is the first secondary mark 14a, to the one indicated with the digit "3". The step of the second graduated scale 13 is 29/30 of the step P of the first graduated scale 10, while the measurement tolerance is equal to 1 /30 of such step P and the maximum displacement measurable with said tolerance is equal to 3/10 of the step P. Preferably, the step P is equal to 1 mm, so consequently the measurement tolerance is 1 /30 mm and the maximum displacement is 3/10 mm.

Clearly, in variants of the invention, the step P and the measurement tolerance can be different from those just described, for example to adapt them to a different measurement system from the metric-decimal system.

With regard to the measurement tolerance, it is preferably of the same order of magnitude as, but more preferably smaller than, the minimum dimension of a point that can be individually illuminated by each light source 3 on the respective work area 5. Typically, for vector light sources 3, the aforesaid minimum dimension is of the order of 1 /10 mm or smaller, down to a few tens of microns.

In the variant of the calibration device according to figure 6, the first graduated scale 101 , 102 has thirty main marks 12, while the second graduated scale 131 has a step of 29/30 of the step P. In this way, the measurement tolerance is equal to 1 /30 of the step P and the maximum displacement is equal to the step P.

The calibration device according to the variant of figure 7 is similar to the one of figure 6, except that each graduated scale 102, 132 is divided in two sections, obtained dividing the respective graduated scale of figure 6 at the height of an axis that is perpendicular to the respective measurement direction X. The two sections are arranged stacked on each other according to a direction perpendicular to the measurement direction X and each of them allows to determine, in combination with the corresponding section of the other scale, a corresponding interval of relative displacements. For example, in the case of figure 7, the pair of sections positioned at the top allows to determine relative rightwards displacements of the second graduated scale 131 with respect to the first graduated scale 101 according to the measurement straight line Y1 between zero and 16/30 of the step P, while the pair of sections positioned at the bottom allows to determine relative displacements according to the measurement straight line Y2 between 17/30 and 30/30 of the step P. In the calibration device of figure 5, each graduated scale 10, 13 has a plurality of main marks 12, 14 positioned on each side of the first mark 12a, 14a of the corresponding scale. In this way, advantageously, it is possible to easily read the misalignments in both the orientations of the measurement direction X. This reading is also possible in the variants of figures 6 and 7 described above, with the caveat that, if the first secondary mark 14a is displaced leftwards of the first main mark 12a, the reading must be done considering, as the first main mark 12a, the one positioned immediately to the left of the main mark indicated with the numeral "30", and as the first secondary mark 14a the one indicated with the numeral "30", instead of the corresponding marks indicated with the numeral "0".

Also preferably, on each side of each graduated scale is present the same number of corresponding marks, so that the respective first mark 12a, 14a is at the centre of the scale, so as to allow reading the same maximum displacement on both measurement orientations.

For the same reason, it is preferable to configure the two graduated scales so that, when the pair of work areas 5 is in its mutual alignment configuration, the intersection point between the two first marks 12a, 14a lies on the axis of symmetry Z of the corresponding superposition area.

Preferably, one or both graduated scales comprise a corresponding succession of increasing numerical values 16. For example, in the cases shown in the figures, the first mark of each scale is indicated with a digit "0" and other numerical values 16 are also present, which increase progressively away from the aforesaid first mark.

The numerical values 16 are proportional to the size of the relative displacement corresponding to the alignment between a mark of a graduated scale and a mark of the other graduated scale. For example, in the case of figure 5, the numerical values 16 of the second graduated scale 13 indicate the relative displacement, expressed in tenths of the step P, corresponding to the alignment of the related secondary marks with a corresponding main mark. Preferably but not necessarily, and as shown in figure 5, the numerical values 16 are associated only to some of the secondary marks, to facilitate the readability of the scale.

Preferably, the numerical values 16 are positioned on both sides of the first mark of the respective scale. In the case of figure 5, the numerical values associated to the secondary marks 14 positioned to the left of the first secondary mark 14a are preceded by a "-" sign to indicate negative values, so that the operator can easily discriminate the direction of the relative displacement, in addition to its size.

With regard to the shape of the main marks 12 and of the secondary marks 14, preferably they are respective perpendicular segments to the measurement direction X.

Also preferably, the graduated scales 10, 101 , 102, 13, 131 , 132 are configured so that, when the pair of work areas 5 is arranged in its alignment configuration, all segments of the first scale 10, 101 , 102 are positioned on a same side with respect to the measurement straight line Y, and all segments of the second scale 13, 131 , 132 are positioned on the opposite side, and that furthermore each segment of the two graduated scales has an end lying on the measurement straight line Y. In this way, the two scales have an identical appearance to that of a Vernier scale, with the advantage of making its reading intuitive.

Figure 8 shows a calibration device according to a variant of the invention, in which the two image portions 8, 9 are equal to those of figure 5, with the difference that the two graduated scales 10, 13 are configures so that, when the pair of work areas 5 is positioned in its alignment configuration, the segments of the graduated scale 10 are positioned at least in part on a side with respect to the measurement straight line Y and at least in part on the opposite side, so that the two graduated scales 10, 13 are partially superposed.

The configuration of figure 8 advantageously allows to facilitate reading if the scales are impressed on the basic material 1 negatively, i.e. so as to illuminate all the points of each image portion 8, 9, excepting those corresponding to the main marks 12 and secondary marks 14. In this way, the light sources 3 illuminate all the points of the work surface 2 belonging to at least one of the two portions 8, 9, excepting those in which there is a superposition between the marks 12, 14 of the two graduated scales 10, 13. Since the aforesaid superposition pertains only to the marks of the two scales that are mutually aligned, in the test image 73 obtained on the basic material 1 , represented in figure 9, only one of the marks of each scale remains entirely visible, while the others are not visible, or are only partly visible when there is a partial superposition between two marks, i.e. in the case in which the thickness of the marks is greater than that of the measurement tolerance.

For example in figure 9, which represents the case in which the two work areas 5 are aligned along the measurement direction X, in the test image 73 only the first marks 12a, 14a remain entirely visible, because they are the only ones to be perfectly aligned. The two marks of each scale 10, 13 immediately to the right and immediately to the left of the first marks 12a, 14a are only partly visible, because they superpose only for a portion of the respective thicknesses.

It is evident that the aforesaid test image 73 facilitates reading the displacement, because the operator has to do nothing more than to identify which mark is most visible in the area in which the two scales are superposed among the visible ones, or which mark is positioned at the centre among the visible ones.

Preferably, the two graduated scales 10, 13 are positioned so as to superpose only partly according to the direction orthogonal to the measurement straight line Y, so that at least one part of each mark 12, 14 and the numerical values 16 fall outside the area in which the two scales are superposed. This causes the aforesaid elements to remain always visible, so that they can be used by the operator to easily identify the position of the aligned marks of each scale with respect to the related first marks.

It is evident that the aforesaid configuration can be applied to any one of the embodiment variants described above.

More generally, the negative projection of the test image 7, 71 , 72, 73 of the first graduated scale 10 and of the reading cursor 11 provides an additional advantage, which emerges if the calibration device is obtained through the solidification of a photopolymer by a stereolithographic process and, hence, is in the form of a three-dimensional object. In this case, the marks of the first graduated scale 10 and of the reading cursor 11 correspond to as many recesses on the surface of the three-dimensional object obtained. Colouring the surface of the three-dimensional object, for example by means of a felt-tip pen, with a different colour from the one of the solidified three-dimensional object, and subsequently cleaning the aforesaid surface, for example by means of a cloth or another equivalent device, the dye is removed except inside the aforesaid recesses, which therefore are made well visible, further facilitating the reading of the displacement.

Also preferably, the method of the invention comprises projecting, for each superposition area 6, at least two test images according to any of the embodiments 7, 71 , 72, 73 described above, which are arranged in different positions in the aforesaid superposition area. The aforesaid two test images are configured so that the respective measurement directions X are both incident to a first edge 5a of each of the two work areas 5 which is superposed to the other work area 5.

Preferably but not necessarily, the measurement directions X of the two test images are both orthogonal to the first edge 5a of one of the two work areas 5 when the two work areas 5 are arranged in the corresponding alignment configuration.

The aforesaid two test images allow to determine the relative displacements between the two work areas 5 in the two different positions of the related superposition area 6, so as to be able to calculate a mutual translation and a mutual rotation between the two work areas 5.

Preferably, on each superposition area 6 is also projected a third test image, which defines a respective third measurement direction X which is incident to one or to both the measurement directions of the two previous test images and which, preferably, is parallel to the aforesaid first edge 5a. The aforesaid third test image allows to determine a third relative displacement according to the third measurement direction X which, combined to the two displacements determined previously, allows to obtain a complete alignment between the two work areas 5. Each of the two work areas 5 has three degrees of freedom with respect to the other work area 5, i.e. two translations according to two mutually incident directions and a rotation, and the three measurements of displacement described above allow to define the aforesaid three degrees of freedom.

Clearly, each additional work area 5 beyond the two previous ones adds three degrees of freedom to the system. Therefore, to obtain the mutual alignment between a predetermined number of work areas 5, it is necessary to have a number of displacement measurements at least equal to the number of work areas 5 minus one and multiplied times three.

In the case shown in figure 3, which comprises four work areas 5, at least nine displacement measurements are necessary and, hence, nine of the aforesaid test images. Figure 3 shows, for each superposition area 6, two test images with the respective measurement directions orthogonal to the superposed edges 5a of the two corresponding work areas 5, to which is added a test image on any one of the four superposition areas 6, having measurement direction X parallel to the superposed edges of the corresponding work areas 5. All the test images of figure 3 are equal to each other and correspond to the embodiment of figure 5, but in variants of the invention different test images can be used, for example according to any one of the variants of figures 6-9. Clearly, the number of test images can be higher than the one indicated above, with the advantage of increase the measurement precision.

For example, the version of figure 3 comprises seven test images 7 for each superposition area, four of which have measurement directions X orthogonal to the superposed edges 5a of the respective work areas 5 and the three remaining ones have measurement directions X parallel to the aforesaid superposed edges 5a. What has just been described is observed in more detail in figure 4, which shows the enlarged view of one of the four work areas 5 of figure 3, where the two areas 6 can be seen in which the work area 5 is superposed with the two adjacent work areas 5.

Preferably and as shown in figure 4, each test image 7 whose measurement direction X is perpendicular to the superposed edges 5a of the work area 5 occupies the entire width of the respective superposition area 6, so as to maximise the maximum measurable displacement.

Also preferably, two of the aforesaid test images 7 are positioned adjacent to the two opposite sides of the superposition area 6 and at the maximum possible mutual distance. This, advantageously, maximises the precision with which it is possible to measure the mutual rotation between the two work areas 5 and, hence, allows a more precise correction of their mutual position.

According to the embodiment described hitherto, the two image portions 8, 9 of each test image 7 are projected on the work surface 2 by two different light sources 3, so that the light sources can be calibrated with respect to each other. Differently, according to one variant of the invention the first image portions 8 can be provided on a reference surface facing the work surface 2 and the second image portions 9 are then projected on the work surface 2 by a respective light source 3, so as to be adjacent to the respective first image portions 8.

This variant allows to calibrate each light source 3 with respect to the corresponding reference surface rather than with respect to the other light sources 3, and it is particularly advantageous when, because of a high number of light sources 3, calibrating the light sources in pairs would be difficult.

The aforesaid reference surface can be the surface of a modelling plate of the stereolithographic machine 18, not shown in the drawings but known in itself, in contact with which the object is solidified. Alternatively, the reference surface can be a surface on which the basic material 1 rests.

The provision of the first image portions 8 on the reference surface can take place by the application, on the reference surface itself, of laminar elements bearing the first image portions 8, or printing the first image portions 8 on the reference surface, or else making the reference surface in such a way that it incorporates the first image portions 8 in raised or recessed form.

For the aforesaid variant, all considerations made for the first embodiment apply, with the due and obvious modifications.

Operatively, the calibration of a stereolithographic machine 18 with the method described above takes place as follows.

First of all, the test images to be projected or to be positioned as described above are defined. Subsequently, a numerical model of the portions of the test images to be projected is provided, which can be used by the control unit of the machine 18 to drive the light sources 3.

The basic material 1 is then positioned on the machine 18 so as to define the work surface 2.

Subsequently, the machine 18 is activated so as to project the test images on the work surface 2 thereby obtaining the calibration device.

If the basic material 1 is the photopolymer normally used in the machine 18 for the production of three-dimensional objects, preferably the numerical model is defined so as to comprise a plurality of layers, at least one of which contains the test images. In this way, the activation of the machine 18 will entail the solidification of a succession of layers of the photopolymer corresponding to the layers of the numerical model. The result will be a calibration device having the shape of a set of platelets whose surfaces bear the first graduated scales 10 and the reading cursors 11 , with the respective marks raised or, more preferably, recessed.

Advantageously, the latter variant is particularly simple, because it entails obtaining the calibration device using the normal functionalities of the stereolithographic machine 18. Consequently, the method of the invention is applied with an identical procedure to the one normally followed for the manufacture of any three-dimensional object.

After manufacturing the calibration device, the operator reads the displacements thereon and modifies the position of the work areas 5 accordingly by an appropriate adjustment of the light sources 3. These operations can be carried out automatically by the machine 18 itself, by means of a software algorithm that allows the operator to enter the values indicated by the calibration device and that, based on these values, calculates the displacements and adjusts the light sources 3.

Preferably, the aforesaid adjustment takes place modifying the coordinate systems of the light sources 3. Advantageously, the adjustment in the manner just described simplifies the calibration process, inasmuch as there is no need for the operator to intervene physically on the machine 18. For this type of adjustment it is preferable for the work area 5 to be smaller than the operating area that each light source 3 can actually illuminate, so as to allow the displacement of the coordinate systems of the light sources 3 while maintaining the work areas 5 completely inside the respective operating areas, so as not to reduce the work surface 2.

Clearly, in variants of the invention, the aforesaid adjustment can take place mechanically, i.e. physically displacing the light sources 3. The displacement can be carried out manually by the operator or automatically by means of actuators activated by the stereolithographic machine.

Since, generally, the installation precision of the light sources 3 is far smaller than the measurement tolerance required for the alignment between the work areas 5, it may occur that one or more displacements sensed during the phase described above exceeds the maximum displacement measurable with the measurement tolerance, and this prevents obtaining a precise alignment with a single calibration step. In this case, after carrying out the first calibration, the test images are projected on the work surface 2 again to obtain a second calibration device, which is used to check the alignment and possibly carry out a second calibration.

The operations described above are repeated until obtaining the alignment of the work areas 5.

In view of the above, it is readily apparent that the calibration method described above and the calibration device obtainable by means of this method achieve all set objects.

In particular, the projection of a graduated scale and of a reading cursor on the basic material by means of two different light sources allow to directly determine the misalignment between the respective work areas, with no need to use measurement instruments external to the stereolithographic machine to be calibrated.

Moreover, the graduated scale and the reading cursor are obtainable through a lower number of exposures on the basic material, hence more rapidly, than in known methods.