JPH07125078A - Data processing method - Google Patents

Abstract

PURPOSE:To provide an optical shaped article excellent in dimensional accuracy by enabling data processing automatically detecting an overhang part and the bottom surface of a shaped article in surface shape data to automatically correct dimensional deviation generated by excessive curing due to laser transmitted beam. CONSTITUTION:The apexes of a polygonal patch constituting surface shape data are replaced with nodes and the nodes constituting the polygonal patch positioned on the bottom surface of an overhang part are moved to alter the shape and position of the polygonal patch. By this method, dimensional deviation is automatically corrected in surface shape data and shaping is performed on the basis of the data. By this constitution, an optical shaped product high in dimensional accuracy is obtained. It is unnecessary to correct dimensional deviation by the hand while returning to three-dimensional CAD and processing can be automatically performed within a short time.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a stereolithography method and apparatus for curing an ultraviolet curable resin by laser irradiation to create a three-dimensional model, and particularly by correcting dimensional deviation due to excessive curing by laser transmitted light. The present invention relates to a data processing method for providing a stereolithography object having excellent dimensional accuracy.

[0002]

2. Description of the Related Art Conventionally, a stereolithography method and apparatus have been described in Marutani et al .: Stereolithography: Convert three-dimensional shape model data into contour line data as described in Nikkan Kogyo Shimbun, and sequentially stack according to the cross-sectional shape of each contour line. It is known as a method of creating a three-dimensional model.

Regarding measures for improving accuracy, Japanese Patent Publication No. 5-339.
No. 00, 33901.

Also, at the RP & M / SLA Special Seminar: Japan 3D System Sponsor: '92 .10.29-30,
It is described that the curing depth at the time when the resin layer is scanned by laser irradiation once and the curing depth at the time when the laser intersects is about 1.7 times deeper at the intersecting portion. .

[0005]

In the stereolithography technique, there is a problem in that the bottom surface of a horizontal plate or the bottom surface of an overhang portion is excessively hardened due to the accumulation of leakage light of a laser which has passed through a hardened material during lamination. This problem appears as dimensional deviation and is an essential problem of stereolithography.

The above-mentioned prior art does not consider the correction of the excess cured thickness by the laser transmitted light on the bottom surface of the overhang portion. Therefore, in the lower part of the portion, the uncured resin is cured by the laser transmitted light and the thickness exceeds the design thickness
There was a problem of reducing the dimensional accuracy.

Further, in order to make the dimensions of the modeled object equal to the design value, a secondary machining step such as shaving is required, but in some cases it is impossible to shave in a narrow gap, and it is necessary to correct the dimensional deviation. It was

A three-dimensional CA is used as a method for correcting the dimensional deviation.
There is also a method to modify the design dimensions in D, but C
There is a problem that the dimension correction by AD is not easy and takes time.

An object of the present invention is to solve the above problems and provide a method for automatically detecting the bottom surfaces of a modeled object and an overhang portion and automatically correcting the dimensional deviation which is an essential feature of the optical modeling technique. Especially.

[0010]

In order to achieve the above object, the vertices of polygon patches forming the surface shape data are replaced with nodes, and the nodes forming the polygon patches located on the bottom surface of the overhang portion are moved. By changing the shape and position of the polygonal patch, the dimensional deviation is automatically corrected in the surface shape data, and the modeling is performed based on the data.

[0011]

In the present invention, it becomes possible to automatically correct the dimensional deviation which is an essential feature of the stereolithography technique on the surface shape data, and by performing the shaping based on the corrected surface shape data, the stereolithography object is produced. Dimensional accuracy is improved. Further, since the above-mentioned correction processing can be automatically performed, the efficiency of dimensional deviation correction is improved. Further, a secondary processing step such as shaving is omitted.

[0012]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a PAD diagram of a processing method for correcting a dimensional deviation due to excessive hardening of the bottom surface of an overhang portion according to a first embodiment of the present invention by using surface shape data.

First, a surface shape data file for correction is designated. The surface shape data represents the surface of the three-dimensional shape model by a set of polygonal patches having at least three vertices, and this embodiment shows a triangular patch. FIG. 2 shows the format of the surface shape data file. The normal vector and the three-dimensional coordinates of the three vertices forming the patch are described, and the data of each patch are distinguished by a delimiter mark. Next, the coordinates and normal vectors of the three vertices of the patches in the surface shape data file are sequentially read, and the total number of patches is counted. Next, a node number is added to the vertex coordinates that have been read in, and the same node number is assigned to the one that matches the vertex coordinates of the patch that was previously read. Further, it stores which node each patch is composed of. Next, in order to detect the overhang portion, whether the Z component of the normal vector is positive or negative is determined. When the Z component is negative, that is, when the patch surface faces downward, the patch is located at the bottom surface of the overhang portion or the bottom surface of the modeled object.
Set a flag to indicate that it is located on the bottom of the overhang. Repeat this for all patches. The above operation is the same as the entire processing even if it is performed at any time while reading the patch data.

Next, a surplus cured thickness α is set as a correction amount. The method for obtaining the surplus cured thickness α will be described later. Next, the flags of all the nodes are checked, and for the node having the flag set, the correction amount is added to the Z coordinate of the node, and the corrected Z coordinate value is replaced as the new coordinate value of each node. Rewrite the vertex coordinates that make up the patch with the new coordinates of the replaced node. Here, since the surface inclination of the patch is changed, the normal vector of the patch is calculated using the corrected Z coordinate. Create a new surface shape data file in the same format as the read data format from the corrected vertex coordinates and normal vector.

Regarding the setting of the correction amount, the overall processing is the same regardless of which stage is performed before the Z correction.

Here, the reason why node numbers are assigned to the vertices of the patches and the Z coordinate correction is processed by the nodes will be described. FIG. 3 is a diagram of the triangular patch of the surface shape data. FIG. 3A shows two adjacent patches A and B. The patch A is located on the bottom surface of the overhang portion, and the patch B is not the overhang portion.
The vertices of each patch are a1, a2, a3, b1, b2, b
3, Here, if the coordinate values of the vertices a1, a2, and a3 are moved when changing the shape of the patch A on the bottom surface of the overhang portion, the shape of the triangular patch A is changed, but the patch is separated from the patch B, and the patch is separated. The relationship between A and patch B is not maintained, resulting in incorrect data. A patch of normal surface shape data is composed of at least three vertices, but it is not recognized that adjacent patches share these vertices. In FIG. 3B, the vertices shared by the triangular patches A and B are replaced with nodes n1 and n3,
The shape of the patch A is changed by moving 1, n2 and n3. As a result, the shape of the patch B is also changed at the same time, the patches A and B are not separated, and the relationship between both patches is maintained and correct data is obtained. As described above, in the triangular patch representing the surface of the model, the vertices shared by the adjacent patches are replaced with nodes, and the nodes are moved, so that the shapes and positions of the patches are changed without separating or intersecting. be able to. Here, an inclined overhang bottom surface is shown as an example, but the same applies to a horizontal case.

Next, a method of determining the patch located on the bottom surface of the modeled object or the bottom surface of the overhang portion by the normal vector will be described. FIG. 4 shows a triangular patch located on the bottom surface of the overhang portion and its normal vector. Let the stacking direction of the model be the positive direction of the coordinate Z-axis, and let the normal vector that is perpendicular to the plane defined by the patch and goes to the outside of the model be a
Assuming that (Xa, Ya, Za), if the Z component is Za <0, it can be determined that the patch is located on the bottom surface of the overhang portion because the patch faces downward. In this way, it is possible to automatically detect the patch located on the bottom surface of the overhang portion by determining whether the normal vector Z component is positive or negative.

With the above processing method, the bottom surface of the modeled object and the rolling surface of the overhang portion can be automatically detected, and the dimensional deviation can be efficiently corrected on the surface shape data, and the modeling is performed based on this data. This makes it possible to obtain a stereolithography product with high dimensional accuracy.

In this embodiment, the triangular patch is shown, but the same effect can be obtained with a polygon patch.

The surplus cured thickness will be described below.
FIG. 5A shows the principle of surplus curing by stereolithography. Excessive curing is caused by the leakage light of the laser which has passed through the bottom surface of the overhang portion of the molded article and cures the uncured resin on the bottom surface of the overhang portion. Next, FIG. 5 shows how to determine the surplus cured thickness α.
An explanation will be given using (b).

P is the stacking pitch, D is the depth from the upper surface of the first cured layer, and D is the number of stacked layers. N is the light absorption coefficient of the resin.
(D, N) is En (D, N) = EXP (−k (D + (N−1) P) /
λ).

Here, since the transmitted light is irradiated many times in the lower part of the first layer due to the stacking, the accumulated energy is E
If total (D, N) is given, Etotal (D, N) = EXP (−kD / λ) + EXP (−k (D + P) / λ) + ... + EXP (−k (D + (n−1) P) / λ) where A = EXP (-kD / λ) B = EXP (-kP / λ) Etotal (D, N) = A (1 + B + B ^ 2 + ... + B
^ (N-1)) Taking the logarithm of both sides, 1n (Etotal (D, N)) =-kD / λ + 1n (C) where c = (1-B ^ n) / (1-B) Therefore D = -Λ / k (1n (Etotal (D, N) -1n
(C)) D is the depth from the top surface of the first layer, and Etota
Assuming that l (D, N) is the critical curing power (ratio to irradiation power), the surplus curing thickness α when N layers are stacked is

[0024]

## EQU1 ## α = DP (n-1).

FIG. 6 shows the relationship between laminated thickness and dimensional deviation.
This data is a horizontal overhang

The results are for the bottom surface. Than this,

It can be seen that the calculation result of [Equation 1] is in good agreement with the measured value.

FIG. 7 is a PAD diagram when the correction amount is a value depending on the inclination of the bottom surface of the overhang portion in the first embodiment of the present invention. After detecting the patch located on the bottom surface of the overhang portion, a value that depends on the inclination of the patch is set as a correction amount in the nodes that form the patch. Figure 8
(A) is a cross-sectional view of model data obtained by performing the above correction with a correction amount of 1 on the surface shape data of a model having a hole with a radius r = 3.0. However, the surplus hardening thickness depends on the inclination of the bottom surface of the overhang part, so if stereolithography is performed based on this data, the dimensional deviation will not be uniform, and a distorted stereolithography model as shown in the right figure will result. . FIG. 8B is a cross-sectional view of the model data on which the above-described correction is performed with the correction amount as a value depending on the inclination of the bottom surface of the overhang portion. A hole shape is created, and a stereolithography model with high dimensional accuracy can be obtained.

The setting of the correction amount here is the same as that of the whole process regardless of the stage after the detection of the patch located on the bottom surface of the overhang portion.

By correcting the surface shape data by the above processing method and performing modeling, an optical modeling product with high dimensional accuracy can be obtained.

FIG. 9 shows the PAD when the normal vector is calculated from the vertices of the polygon patch in the first embodiment of the present invention.
It is a figure. The coordinates of the vertices forming the triangular patch included in the surface shape data file are read, and the normal vector of the patch is calculated from these vertex coordinates. A patch located in the overhang portion is determined using the normal vector. If the calculation of the normal vector is at a stage prior to the determination of whether the Z component is positive or negative, the overall processing is the same at any time.

By correcting the surface shape data by the above processing method and performing modeling, an optical modeling product with high dimensional accuracy can be obtained.

FIG. 10 shows a second embodiment of the present invention, which is a procedure for performing correction processing on the above surface shape data to carry out stereolithography. A three-dimensional CAD is used to create a shape model, which is converted into surface shape data for stereolithography, and the data is subjected to correction processing such as automatic detection of overhangs and automatic correction of dimensional deviation to perform stereolithography. FIG. 11 is a cross-sectional view of a model having a hole with a design value of diameter D = 10.0. FIG. 11 (a) is a cross section of a model without correction, and FIG. It is a figure. As described above, by performing the modeling according to the procedure of the present invention, it is possible to obtain an optical modeling product with high dimensional accuracy.

[0033]

According to the present invention, it is possible to automatically detect the bottom surface of a modeled object and the bottom surface of an overhang portion, and correct the essential dimensional deviation of the optical modeling technique generated on the bottom surface on the surface shape data. Therefore, there is an effect that a stereolithography product with high dimensional accuracy can be obtained.

Further, the correction of the dimensional deviation does not have to be manually corrected by returning to the three-dimensional CAD, and there is an effect that it can be automatically processed in a short time.

[Brief description of drawings]

FIG. 1 is a PAD diagram of a processing method for correcting a dimensional deviation due to excessive hardening of a bottom surface of an overhang portion according to a first embodiment of the present invention using a surface shape model.

FIG. 2 is a format of surface shape data in the first embodiment of the present invention.

FIG. 3 is a diagram showing triangular patches and nodes according to the first embodiment of this invention.

FIG. 4 is a diagram showing a normal vector of a patch of an overhang portion in the first embodiment of the present invention.

FIG. 5 is a diagram showing a surplus cured thickness in the first embodiment of the present invention.

FIG. 6 is a graph showing the relationship between laminated thickness and dimensional deviation in the first embodiment of the present invention.

FIG. 7 is a PAD diagram when the correction amount is a value depending on the inclination of the bottom surface of the overhang portion in the first embodiment of the present invention.

FIG. 8 is a cross-sectional view of a model when the correction amount in the first embodiment of the present invention is a value depending on the inclination of the bottom surface of the overhang portion.

FIG. 9 is a PAD diagram when a normal vector is calculated from the vertices of a polygon patch in the first embodiment of the present invention.

FIG. 10 is a diagram showing a procedure for performing optical shaping by performing surface shape data correction processing according to the second embodiment of the present invention.

FIG. 11 is a cross-sectional view of a model having a hole having a design value D = 10.0 of diameter.

[Explanation of symbols]

 a ... normal vector, a1, a2, a3, b1, b2, b3 ... vertex, n1, n2, n3, n4 ... node, Za ... normal vector Z component, α ... surplus hardening thickness.

 ─────────────────────────────────────────────────── --- Continuation of the front page (72) Inventor Kentsugu Okubo 4, 6 Kanda Surugadai, Chiyoda-ku, Tokyo Stock company Hitachi Ltd. (72) Inventor Toshiro Endo 292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa Hitachi, Ltd. AV Equipment Division

Claims (5)

[Claims] 1. In data processing for forming a three-dimensional shape model using surface shape data representing the surface of a three-dimensional shape model with a set of polygonal patches having at least three vertices, a normal line of the patch surface for each patch. The direction of the vector is used to identify whether the patch is located on the bottom of the model or the bottom of the overhang portion, and when the patch is identified to be located on the bottom, the coordinates of the vertices forming the patch are set in a predetermined direction. A data processing method characterized by moving a predetermined value. 2. In identifying and locating coordinates at the bottom, the vertex of the patch is replaced with a node, and the polygon patch located at the bottom of the model or the bottom of the overhang is Z of the normal vector of the patch. Judging by the positive or negative of the component, the Z component is negative, that is, for the patch located at the bottom of the overhang portion, a flag indicating that the Z component is located at the bottom of the overhang portion is set, all nodes are checked, and the flag of the flag is set. 2. The surface shape model data is corrected by a processing method in which the Z coordinate value of a standing node is corrected and the shapes and positions of all polygon patches having the node as a vertex are changed.
Described data processing method. 3. A correction amount applied to a Z coordinate value of a node forming a polygon patch located on the bottom surface of the modeled object or the bottom surface of the overhang portion depends on the inclination of the bottom surface of the overhang portion. The data processing method according to claim 1 or 2. 4. A normal vector, which is perpendicular to a plane defined by the polygon patch and extends toward the outside of the model, is calculated from the coordinates of the vertices forming the polygon patch, and the positive / negative of the normal vector determines the position on the bottom surface of the overhang portion. The polygonal patch to be determined is determined, claim 1, claim 2 or claim 3
Described data processing method. 5. An optical modeling method for producing an optical modeling model by irradiating an ultraviolet curable resin with a laser to cure it and stacking the cured products, and the apparatus is used. The data processing method according to claim 2, claim 3, or claim 4.