JPH0491929A - System and method for three-dimensional formation - Google Patents

Abstract

PURPOSE: To mold a sterolithographic part quickly without sacrifice of accuracy by providing a shutter disposed along an optical path while intersecting a beam and blocking passage of the beam selectively in the surface region of an object not to be converted. CONSTITUTION: A system for building a three-dimensional object, especially a large or complex part, quickly and accurately by stereolithography employs a more powerful laser 11 and a high speed dynamic mirror 15 in order to generate a beam for hardening a polymerizable substance and to write a pattern thereon. A controllable shutter 20 is disposed on the optical path of laser beam in order to block irradiation of a selected surface region of the substance with laser beam. It is performed through an appropriate feedback loop for positioning the dynamic mirror 15 accurately based on a known shift from a desired position.

Description

[Detailed description of the invention]

(Field of Industrial Application) The present invention generally relates to improvements in a three-dimensional modeling method and apparatus for producing five-dimensional objects, and more specifically, to two-dimensional objects, especially large or complex objects. This invention relates to improvements in producing objects quickly without sacrificing accuracy and eliminating obstacles in producing such objects. (Problem to be solved by the invention) Three-dimensional modeling 3D reproduction is the process of constructing an object layer by layer, with layers of the object being successively formed on top of each other until the entire object is reproduced or completed. This process is described in detail in U.S. Pat. No. 4,575,330. , 3D modeling device (1”'S1..^
”) means that objects can be recreated through a three-dimensional modeling process! 1- It is a device that performs An example of an SLA includes a four-acting stimulation means, such as a LiVL-subbeam, a photocurable liquid resin placed in a hat, and an elevating means. The three-dimensional molding device forms each layer of the part by exposing the liquid resin surface (1) to sufficient light to harden the liquid resin to a predetermined depth using a UV laser beam, thereby drawing a horizontal pattern. do. The four rising stages indicate which parts are to be stacked on top of each other. The first layer is attached to and supported by the cured resin in the form of a web or the like, known as a base or support. is deposited directly on the lifting means.The layer formed is then stacked on top of the first layer of soil.As the sections are stacked, the c1 rise stage f descends and sources successively into the hive of liquid resin. In each step of the method, after a layer is formed, the next layer (along with all previously formed layers) is lowered into a vat of liquid resin. Specifically, the lifting means is immersed into the liquid resin deeper than the thickness of the next layer so that the liquid resin flows quickly over the previous layer. The lifting means is then raised to remove excess resin until the coating thickness is equal to the desired thickness of the next layer. One or more techniques for reducing the thickness are implemented, at some point in the method the upper surface of the previously cured cross section is reduced to a depth equal to the desired thickness of the next layer of liquid resin surfacing. This prepares the surface of the resin and the position of the previous layer for exposing the next cross section and attaching it to the previous layer.Excess resin thickness Various methods for thinning are disclosed in the U.S. Patent Application and the International Patent Application No. F. U.S. Patent Application: April 1989, 171, Application Tube 339
, No. 246, March 31, 1989 [co-application no. 331,6
No. 84, April 1988] 8 ("Application No. 183,015
No. 182.801, April 18, 1988.
1988 1iJJ8F+ Application No. 268.429, 1
November 988 81-” Application Tube No. 268.428, 1
November 8, 1988 Application No. 288.408, 1988
November 81, 1988, No. 268, No. 816, II, 8, 1988 [(Application No. 268,907, November 1988)
Filing No. 268.837, September 8th, 1988, 9J12
6th Application No. 268.399, July 12, 1989 Application No. 365.444, October 31, 1988. E1 Application No. 2B5,039, filed September 26, 1988, No. 2
No. 49.399, and the three-dimensional modeling device f (STEREOLITHGRAPHIC) filed at the same time as this application.
APPARATIJS) J. 1989 97126B International Patent Application No. No. 0.188/
189PcT. In this example of stereolithography, the HeCd laser
The wavelength is approximately 325ni, and the maximum output is 50+akl lJ
A V laser beam is typically emitted. The problem of curing unwanted liquid resin when the laser beam is positioned between the vectors (across a region that should remain unpolymerized) results in higher peaks. The laser was not considered to be effective (intended to harden only in locations corresponding to vector delta). The laser is typically static relative to the vat, and the optical path from the laser to the surface of the beam or resin is directed by an optical system that includes a pair of rotating dynamic mirrors. is traced, after which Table 1 of the liquid resin
A partial layer of TijY is traced by controlled rotation of the dynamic mirror. In this item j, the dynamic mirror has two approximately centroid axes ζJ? A-)-C movement along the X and Y axes of the resin surface provided at a fixed distance from the mirror;
It is normal that two-wheel gauging = one scanning head. The curing depth of the resin in a particular area of the resin surface −t is
Logarithmically dependent on the laser beam exposure in that area. The exposure depends on the power and intensity distribution of the beam, and on the scanning speed J of the dynamic mirror as it passes through the area. b - Will the strength become more powerful (
+, -, the dimming becomes stronger), or as the movement of the mirror slows down, the exposure becomes stronger. Currently, as mentioned at the beginning, the maximum output of the HeCd laser used in three-dimensional modeling is 50 i1i, and the speed of the dynamic mirror is about 0 to 7
6(1) controllably exposes the surface for a period of time in the range of 7 seconds (approximately (]-30 in) seconds. Position the L-sample by passing the laser beam to the area where the resin will be cured to an appreciable amount (sufficient to gel). This is sometimes done, in the example above, by limiting the power of the laser so that at the maximum speed of the mirror, the exposure is not sufficient to cure an appreciable amount of resin. Is this it?

[/・- This is the reason why the maximum output of the device is limited to 5011W. This previous limitation on laser power was based on the ability of the scanning mirror to jump the laser beam at L in the area of the resin surface that is not intended to cure.1. It's ours. Another reason for limiting the power of the laser beam is based on the highest controllable speed of scanning that can be effectively used to expose the selected area. For example, if λ is exposed in the manner 1ζ, -1. 'Curing and Curing': When a very thin film of material is desired, the film is exposed to a series of approaches 1. 8.5) Determine whether the contour or vector of the strength of the beam formed from the hexagonal beam should be as close as 8.5). 1. −Subbeam output and
Hair to obtain the desired hardening depth (zo 0) @ hardening depth; 4-
The required scanning speed is determined by the combination of the dew tip amount) and the contour of the vector i. If the laser power becomes too high, the required scanning speed in this example situation will easily exceed the maximum allowable scanning speed, thereby imposing a limit on the maximum measured laser power. In this discussion, maximum laser power refers to the limiting factor when in fact the maximum speed is the limiting factor. Power and intensity will be used interchangeably in this description, since we specifically use the smallest beam point possible in order to maintain high resolution in the x-y dimensions. 5
Therefore, a small, relatively fixed beam point size /j
For 4, the strength (outer cuff, 'unit area) and the outgoing horn cannot be easily separated. In addition, -1-11ecd ly used for seven body modeling
- The area cannot be adjusted. That is, the lane is either Hon or 4' (:, EXi capable maximum 1 i 1 force Ray. Generates subbeam, 1) - Zabi ~ l, generates [7? It is one of the following. [, Taka-) So, one of them
When using Nona, exposure constraints (7 to 12) It was not possible to vary the beam output. However, the problem with limiting the laser output is that the part formation or the more powerful Compared to using a laser,
That is, it is sufficiently late. Larger items, typically about 127 (punishment x about 17.8eniX about 15.2r
The problem becomes even greater where m(5in, X7)n, X6in, ) is constructed. This is because these parts already take a considerable amount of time to construct due to their dimensions. The problem is significantly worse when using lasers with lower power. However, as mentioned earlier, a more powerful laser beam is used due to unnecessary resin curing, the highest i'-power scanning speed, and the fixed nature of the laser beam's power. It was not possible to use the more powerful 1/
Thin, i.e., less than about 0.13 in (about 5 mils)
This is because it is not possible to form a cross section of F. Faster dynamic mirror IF, accuracy is about 76cs, /
The speed becomes extremely low at speeds of less than 30 isots/sec (approx. J This is not a general solution. For that and other reasons, constructing parts by body molding is now typically quite time consuming, especially for constructing large or complex parts. It takes time. Small, e.g. nominal 12.7 (7) x about 17.8eix about 15
．． 2cm (5in, X7in,
It takes 48 hours and the size is approximately 51.8cm>(approximately 51.8em
Approximately 61ea (20in, x 20in, x 24i
n,) requires longer time, up to 2 to 10 days. Another problem with the above example of stereolithography is that there are some impediments to the flow of data throughout the system. When constructing detailed or complex parts,
These obstacles can be very severe and can reach 5 degrees. The reason for this is that (4) in those cases, we need t bits of data to represent the part,5 and this data, IJ for large parts, is typically about 3 GB (for large parts) How many megabytes do you manage?2
, part, "1) 4. During the construction, the various parts of the stereofabrication machine must be sent to the various parts. <7
It's true. For example, λ, the V4 harm of 1-) is a dynamic mirror, and the damage to draw is 1
Jll Hector 8i. [1- There is a need to control the 10W device and other related equipment in real time, and as a result, the control computer required to perform all of these tasks is AD/CA
This is due to the unavailability of other important data operations, such as data conversion, which may not be necessary in real time. ] or [, Takeko simply provides another combination, Zuta, to perform these support tasks, especially for large parts 1.
However, it is not very effective for IJ or complex component configurations. For example, in order to buffer the data flow of one day's worth of data in two II p-bicoators, an additional storage device of f=1 must be added, but this is not the case for large or complex parts. The equipment requirements, typically 3, make this equipment prohibitive and expensive. Another obstacle in this example of stereolithography equipment is the phase 1- The control computer determines the desired hardening depth (layer thickness may vary depending on several factors) on the part before starting construction of the part. The computer must interact with the user in order to obtain parameters to control the configuration and parameters to control the physical movement of the device and associated equipment. are collectively known simply as the construction parameters. 1-Ser and phase, 13 Various ways have been tried to interact with each other. In one spear, C is 1-Ser and ~ is the entire part. Kneads can be difficult to handle, especially for large or complex parts. In another approach, the user can Allows you to decompose a range into layers and specify different configuration parameters for each range. However, for large or complex parts Emphasizing various qualities such as strength, genius, speed, or ease in・If you seek to reduce the layers in those areas to the same range, this will be hindered.3 However, these methods are difficult to achieve if you are trying to reduce the number of layers in those areas to the same range. Success is inevitable (and has not always been proven. Dynamic mirror and ■-Noboru 2 position.1
-, the physical properties of the laser and the details of each realization - (must be discussed.As a result,
A novice user may be able to operate the system effectively in minutes without being given time-consuming instructions, especially with delicate or complex parts. 1 This makes the construction of parts even slower. (Problem to be solved by the invention) Therefore, the purpose of the present invention is to solve the problem by sacrificing accuracy.
It is an object of the present invention to provide a means for rapidly constructing three-dimensional molded parts, especially large or complex parts, without any problems. (Means for Solving the Problems) The present invention is directed to a system for rapidly constructing dimensional objects, particularly large or complex parts, by three-dimensional modeling. To this end, more powerful 1-nosers and faster dynamic mirrors are employed to generate beams that can cure polymerizable materials and quickly write patterns on them. . However, it is often necessary for a laser beam to scan the surface of a polymerizable material without curing the material to any appreciable extent. use increases the risk of such undesirable hardening occurring. In addition, with a high-speed dynamic mirror, the positioning of the mirror is unstable, and there is a risk that the error in one scan of the beam will be affected. Therefore, in the system, these risks should be minimized or solved.The present invention provides a method for directing the laser beam onto a selected region of the material surface during the optical path of the laser beam. )″
It is achieved by a controllable shutter to stop the beam from being exposed, and by a suitable feedback control loop to accurately position the dynamic mirror based on known deviations from the desired position. This is what we do. Due to the shutter response time, the shutter command is fired before the desired response, so that the shutter response coincides with the moment the shutter response passes through the selected area. Told. Obstacles in data flow and interaction with users also limit the speed with which large or complex parts can be assembled (Z). For this purpose, a multi-build processor architecture is adopted (for example, a dynamic mirror a!l function and control, and a dedicated block (-7 for the time device management f'i). 11. Power / Assignment 4 / h, External +... Reception of tD/CAM data and back-end work such as conversion performed by three-dimensional modeling sends the three-dimensional modeling data to the dedicated processor. In order for that dedicated processor to be assigned to another more general-purpose processor, which is coupled to the connector, the dedicated processor and the general-purpose processor typically operate in parallel. In the case of components, the flow of data between processors can be impractically time-consuming and expensive. In, CAD/
CAM stereolithography data conversion is employed. This three-dimensional structure delta conversion method accurately expresses the three-dimensional structure necessary for the 15th! Km )'～ri part I is generated by the general purpose professional tree, and the rest of the data is by the dedicated professional...
4. Generated by j. For this purpose, in the present invention, the boundary data of layers suitable for three-dimensional modeling is sent to the general purpose printer from January 7th, and in order to represent the area 4 surrounded by the boundary gutter, 1°8 The remaining three-dimensional modeling data is generated by the 1.4 dedicated process. The system of the present invention also provides a model for organizing, displaying, and inputting data related to three-dimensional molded parts. For large or complex parts, the user may wish to decompose the part conceptually into components and then decompose each component into layers. The layers are separated by 91 layers (along a particular axis) in an order of 11 to 91 to be finally assembled in a molding system.
Construct each layer range of the component that emphasizes various desired characteristics, such as 1. The physical devices used, such as dynamic mirrors and lifting devices, do not require detailed knowledge of the physical equipment used (5), such as dynamic mirrors and lifting devices.
, Transfer to the three-dimensional construction q) equipment that should be assembled, J.
, -, R can also be requested. To do this, we need to arrange cells in a grid at the intersections of the JML columns (1) to arrange the external C^11/CAM data that describes the structure of the parts into columns, A component having the ability to serve 4 sets of rows △ in the range of 4 adjacent layers of the component separated by 7 along a particular axis in order of layers or layers on the surface of Once this has been done, the system applies a selection of a predefined library of style data to the expanded sheet model. Each cell IJMi has the property of combining. 7 Each factory is assembled to the desired part characteristics, and is assembled into a built-in system).
1-Use-4 to command the L7 implementation. During the realization of the special nailing process, the part is layered onto the surface of a hardenable material using a layer of JIJ structure I1, the internal structure of the layer is made up of elements, Overview of the S-stem Figure 1 (here is an example of Sl, A that implements this QIIJ theme). There is also Figure 15.
), Sl, A 17) l (spirit component force is on control - call (i), % science anil (2), service 't: Z-
(:(),,!',',,'Noksteni r > (Fig. C1, not drawn) or 1゛,! depends on. - The power supply, - ＼ Soto heat insulation: ii There is a part of the main part. This is to prevent vibrations and electromagnetic field noise from reaching the duct during processing and to suppress the function f1 as much as possible.
η, Circulate Qi. The air is inside the duct in the area near the electrical equipment or IJI:] iA.
, :L - Liquefied resin by full control, Kokanin 15, liquefied 1. Is the 4° processing Ei゛v-le a Kameko circuit? Senguri, processing buya L
...-1 The electronic circuit, which consists of a collection system, is housed in a standard type 19F, and is controlled by JL, Viata, 1 and 1'.・＼Circuit and power supply are stored in full-〇 control 3:,, coater is 14-:,, l-(No.1 then1t!L
) Do not communicate with a network or equivalent product. The configuration of the control unit is as follows. 80386 Microbe II SE--No. 80387
Numerical performance p-use 2] BU (fuse sosa l RAM:gmenob・)ito 140 mechanism
2 Mezzo J Height Flosso Bee Disk Tra/R 1 Standard 4-Bold (101 keys ~) VGA Driver Board and Color T; , John → - 1 disk, 0 - board 1 printer driver board 1 7 4 modules 1 digital processor 54 2 processors (DSP) 1 processing channel As shown in Figure:?
The chamber includes a removable bath I"' (4), a plunger (5), a plate hoe (6), a recoater blade (7), and a liquid level 1/bed detector (visible and not visible in the figure).
, B~MP1 Cofa f-"shi (8), Elm Beta (
9), L/Belling Jatu 1: (lO) 1-12 (
,゛(There is 1, there is almost 07 meb D in the bat>(254
Liquefaction of Little Little) 1.・J'1. There is enough storage space inside...! <, rL Senkaku (51) 8mm
)"C height 24(;su(6([:, imm IyoZ'
You can make ＼-tsu.・Liquification 1 no, :・ 1 unit in the site is liquid level ・ ＼ru detector or monitoring function 1, yes to this =) i elm sunia otsu l :: y -1 Yes (buy, you can buy the equivalent item 3)
, check the liquid level I)
Compare with L and F settings. , brushes, and rollers are controlled by the two-wheel beator, and the forward screw J' driven by the sphere knob motor is controlled to keep the liquid level at 1. ~
゛1. - coincides with the focal plane of 4.) Setting 1/c.,
Based on the comparisons made by the 3. H1 tester used to keep the liquid level (-11, 11, 1, -j), if the liquid level is too low, press the Also 11,
. When the liquid level becomes appropriate: expel 1 nozzle on the table, perform 3 degrees and 1/bel or 4 too high, the control switch and the computer will move the plunger.5. Subtract 1. Liquid 1f + 1 is proper Saga, L is about 4゛. iti definite) j ~-ノ, 1. ： vs. (temperature J, f liquid IR) I//＼＼ノL・kept 1
An example of how to use Nori is 1, ζ is IJ > 1x-9 control and -1...Cistano, 4 is used 7<:T!・Itself is 1-5■・[5, jCI"1., 1iii The range can be widened. When a part is manufactured, it is attached to and supported by the Bratbohm. In addition, Blattbohm is attached vertically to the electric motor.
．． This is controlled by a precision advance screw driven by a control computer or a control step motor. After each layer of the bar 4. ) into the liquefied resin deeper than the required thickness so that the liquefied resin flows quickly over the top of the scrap. Later, when the liquefied resin has covered the previous layer, place it in the recoater. Wipe parallel to the blade or resin surface and at a constant distance until the platform is at the [arm] of the coater blade, which is equal to or greater than the pre-1tPI thickness of the previous or next layer. Yes.Liquification register〉・is 1
)・°f remains on the previous layer due to the surface tension. At this point, use a recoater blade to scrape over the curved layer until the liquefied resin surface is sufficiently smooth and the desired thickness is reached. The platform descends until it reaches the level of the liquid level.The number of times it swabs is controlled by the position, speed, and recoater blade.
The computer-controlled steps are determined by Tanabata. The distance from the blade to the parts is controlled by Erehe-Yu. The speed of the plate and the number of wipes are precisely controlled.
Forms a smooth liquefied resin layer without creating bubbles, minimizing the number of times required. Details regarding recoatinig and surface leveling with plates can be found in International Patent Application No. 1, Yon & I, Yon Doeket
No. 11'187189 PCT, and U.S. Patent Applications No. 265.039 and 249,399 and 1;
The elevator, recoater blade, and blunt gloss are intentionally not mechanically bonded to the butt.This allows for various changes such as color and conductivity during part formation. It can easily be exchanged with a vat containing a resistor with different characteristics, and the material can be maintained during the forming of parts. If it is connected to the baler 1fS, it is possible to exchange it to the middle part.
Fill the new assembly with J/7 (W) (residues from the site and need to be completely removed, adjust various ratios of each composition (I) before proceeding with baler formation. This is necessarily to mark the cow. 2-) The B L profiler is placed exactly at the corner of the diagonal line of the hat, and the profiler of TORA is also covered with an ultraviolet detector. After each layer has been formed, the control combicoder moves the nose beam 1, then -8 to the position where the first profiler is detected. At this point, the control computer detects each filler beam or I-server beam, and records the angle of the dynamic mirror at position C [1]. t. Compare with the value. If any difference is observed, this is considered to be due to the 1-drift J of the dynamic mirror. One of the well-known processes for correcting the drift is control': The viewer continuously compensates for the dynamic mirror tree. Details regarding drifting are previously filed in U.S. Patent Applications No. 2FI8.907, No. 268.837, and No. 268.816. The drawing subsystem is also shown in Figure 3 (5.
, safety shutter (12), focusing system (13)
), fixed mirror (14), dynamic mirror (1.
5) , , 1-1-Miller (1B) and Hunt (17
). The laser is attached to the back of the l〈shito・ and emits 17-servi 1・ from the opening 11. There are two manufacturers, Spectra-PhysiCs and C.
oherent company, Izuminori. This sensor is water-cooled and has a maximum output of 400 milliwatts at ultraviolet wavelengths from 351.1 to 3G 3.8 nanometers. The output actually used is controlled by the combi coater. Example of SLA usage explained earlier - Comparison with the laser used (. Here, the light intensity is large and has more than 11 wavelengths.
The laser used in the example typically uses a beam of 20+'ffi to 25mm' knots and a diameter of about 10mm.
Therefore, the light intensity is 50 watts/cm. -4j The subject laser has a maximum power of 400 milliwatts and a beam diameter of 5 to I mils. This corresponds to a light intensity of 32·no·soto/′ square mm. If the racer is operated at less than maximum power, for example around 250 milliwatts, a 5 to 10 millimeter beam will provide an intensity of 20 watts per square millimeter. The thick broken line in FIG. 3 represents the course of the laser beam. As seen here, the beam is fixed mirror (14)
, dynamic mirror (15) and main mirror (1B
) to take the optical path to the resin liquid level in the vat. A safety shutter (12) and a focusing system (13) are also placed in the optical path. The safety shutter is solenoid operated (located just in front of the nocer aperture) and blocks the beam under computer control. The beam exiting the aperture and passing through the safety shutter is guided by a fixed mirror into the focusing system. The system spreads out the plane of the resin surface and determines the focal point A. Attenuator and high-speed shutter ((1, no X in Figure 3) are also part of the focusing system. The attenuator is a computer Used to change the laser output at high speed through control, harden thin layers that cannot be withstood by a 400 milliwatt laser, and change the hardening depth within the layer1-
There's a chance. The output of the racer is computer controlled and
Therefore, it is theoretically possible to directly change the output of the laser to reach the target. However, it usually takes about 20 to 30°C until the change in laser output causes a white effect.
It takes several minutes, which is not fast enough. The high speed shutter is 1, S series high performance programmer shutter, nm La5er Produet.
, Inc. (Sinnyvale, Calif.). Other than this, V i nee former As5ociates
Mike L7 Blitz (Rochester, New York)
A Model 12X programmable shutter is also available. The high-speed shutter is used to prevent this by shielding resin surfaces such as hollow parts of parts that do not want to harden. Without a shutter or shielding the beam, even if the digematic mirror could move at the highest possible speed, it would harden the plastic in areas where it is not needed. Currently, the dynamic mirror has a Grayha
Scanning Mirror Stem 1.13311-00 from vk Systems (Milvitas, California)
1 is used, which can operate at speeds from 0 to 100 inches per second. This servo can be used at any high speed:
The Bontrol system is added to accurately control the mirror position without sacrificing accuracy (more on this later). In fact, by using a high-power laser in conjunction with a high-speed servo-controlled mirror, they were able to dramatically increase the speed of part creation by more than 20 times, without sacrificing accuracy. A high speed shutter requires a certain amount of time to open or close. Therefore, to ensure the opening or closing of the shutter at the correct time, Y-measure the time to open or close the shutter using the equation of time,
Send the appropriate command to the shutter at the appropriate moment. There is a debate in history regarding this h-character of the shutter. 5iliCon Graphi on the workstation
As standard, it is connected to an external CAD/CAM system via a local network connection using a Personal 1R4S computer manufactured by CS Corporation. CAD/CA
The M system provides the surface shape of an object2.
The workstation pre-processes and controls this data.”
Give it to Shibuta 1 and start making parts there. Workstation equipment includes: RAM 8 MB old 70 MB Bart disk] Cartridge tape device 1 High precision color = Monitor 1 Keyboard 1 Mouse 1 Ethernet board. FIG. 4 more clearly illustrates the optical system of the imaging subsystem. As shown in the figure, the first eight elements in one string of light are the racer C11), the main shutter (12), the refracting mirror/beam splitter (14), the beam, the mirror, and the goo (18). Beam Exbany/Dar No.
l-no-shiz (19), beam expander second l-roose (121), high-speed/〉yanota- (20), attenuator (
22), Dynako Tsukumirror (15), Window (23)
), straight j part mirror <16) di-varnish/litter (2
4), (5) and the quad cells (25 and 26). The beam (27) emitting from the racer is first
・Go through Yatta. The safety shutter is normally open, allowing the beam to pass through without attenuation, or closed in an emergency, blocking the beam. After this, the deflection ζ9-(14) occurs, and this becomes the beam streak 2 knock, resulting in division. rmuh. Beam splitter and one mirror or beam (27
) into beams (28) and (29). Bim (28
) accounts for nearly 9900 of the incident light, (29)
The beam (28) then bends the direction of (28) by approximately 90 degrees when viewed from the incident light beam (28).・Expansion 1
, )·λ7 (19), and the first expansion 1/ns (21) enters the exbender (18). The -l/〉 lens expands the beam (1, the second lens is either at the same point as the resin liquid level, etc.) and focuses it to a spot of the required size. ) is placed at the focal position of the first lens (19), and is used to selectively block the beam under the control of the control unit as described above.The seventh lens (21) After passing [7], the beam then enters the attenuator (22), which operates in the mode of 1!-]. One is the 959o cutoff mode.Control ■Reviewer 34
, At a certain moment, you can select either mode or 3.
, the substantially transparent shade is less than the usual 5 to 40 degrees, and the 95% cut-off mode is less than the desired hardening depth, and there is considerable overlap with each other. Used when exposing along or along a vector. This is because the attenuator fills the vector with a thin epidermis for example λ. )%, the attenuation is made and the beam becomes about 596 at the tip of the hem. After passing through the attenuator, the beam is then injected into the rays (15), 1, 22C', and passes through the seal at the periphery of the optical system and the LT working window (23). . The second beam directly hits the two-part mirror (1B), and then heads toward the hardened bamboo surface bent over the pigeon 1 (17). ).This 1: The beam (29) first reaches the beam splitter (2, I4), where the beam (28) is connected to the beam (30) and ( 31) and 50% of the beam (29).The beam splitter also redirects the beam (30) to 12''
The beam is bent in a direction of approximately 90 degrees with respect to the path of the incident light, and the beam enters the quadceno+ (26). -Strength (3j)
Head to Beam r7 Tocell (25). The quad cell is disclosed in U.S. Patent Application No. 265.0', (9+3) (similar to the cell that is l,'' but only in the two cell C-=- dimension directions). l,
Instead of using bicells, 24 cells can be used in parallel in two-dimensional directions. 7) Full is the position of the incident light, Detection 2J, H1,
, and compare it with a preset position.). J-z-
By essentially adding (31) and (26) to the setting device, high-speed alignment of the laser (1, 1) is possible, and alignment of other parts of the optical system is unnecessary, so Use these utilities. As a result, the racer can be replaced in the cylinder, a new one can be installed, and the
- Ther also beams it 4, and it is a substantial ally with the other thing〉
The position can be adjusted until the ment is removed. : Overview of computer architecture S1. The overall picture of the architecture used in A is shown in Figure 5.As can be seen in the figure, three types of computers are used. Sheet Al) / CAM Stem (32), Two =
Couste John (33) and control duo 71
(1). Workstation and Control = J
It is offered in ~ part and 17. CAD/CA
M system is Sl, ^external, CAD/CA
It is responsible for converting M files into an appropriate format for use in ``81-5^''. This file is known as a step-17 lithography file, or for short, an .stl file. Each C
An AD/CAM file may specify only a single part or, in the case of complex or large parts, only the component parts of that part. Separate st for each CAD/CAM file
Create l file using CAD/CA, M system. Thereby, complex or one-large parts can be broken down into component parts, each of which can be represented in its own CAD/CAM and later in an sLI file. There are several reasons why a medium-sized CAD/CAM file should be split into multiple-part files. This may be necessary for large parts simply because the file of the large part cannot be managed within the CAi)/CAM system. Additionally, it may be desirable to specify different construction parameters for different elements of a part, emphasizing part creation strength and speed in some areas, aesthetic criteria, and accuracy in others. . By dividing the parts file into elements, preparations for executing them are completed. CAD/CAM system to workstation, s
local area network to transfer files.
・Connect to the Rone station using Norn, or use another H
As a method, you can use Kartlin Dave or C on a floppy disk.
It is also conceivable that one node receives and receives data from the AD/CAM system. Each St,l file is made up of a series of adjacent surfaces called facets. It is assumed that this completely covers the entire surface of the component, and that its intersection intersects the vertex of one adjacent I-gon. For more information regarding formats of [stl files], please refer to U.S. Patent Application No. 331.664 and 1. The vector unit standard of a rectangle is also 1. It is assumed that m refers to a point far from the inner surface of the object that is submitted (2. Also, its conformal angle or covering) is a point far from the inner surface of the object. An example of a .stl file is shown below. 5aIllpls, ETL Fi1m diagram ζJ small, C
As you can see, each ゛, 77 inches, tsuto is represented by three vertices and a median standard, and each vertex corresponds to a set of x, , Y
It is expressed in SZ coordinates. Once the elements of the stl file have been transferred to the workstation 3', start up the software on that workstation and follow a series of steps to create the stl file. convert the file into a build file or a file for that part, called an r file. The software executed in the 5LA Part Manager is called the 5LA Part Manager, or Part Manager for short. ,, stl file to b f f-) file, Part manager will convert it to "- as 5"
Realize the functions of the 1Z). First, Coop gives instructions to divide the 811 file into layers1, and then groups each layer into ranges. Next 11, request each range JJ (, some form vane meter 蓓 user or "3z9". Third, each range, S
tl file decomposition into multiple layers (5, slil Mitsuair E
NDSOIJD Create a file that will be released. It contains data in hector format that describes the edges of each layer (see U.S. Pat. 9
Teru). Fourth, combine the formation parameters of all elements such as edge vector data and create bffff for that part.
Generate the file 1. If there is any discrepancy between the formation parameters of each component in the same layer, select Part M.
anager regulates this. Finally, pass the build file to the control combicoter. In FIG. 6, details are described for the settings of the control collector. As shown in the figure, in this setting, two computers are controlled. Stereo I) SP (34
), servo DSP (35) and encoder (3
7) and a dynamic mirror interface (3) consisting of a DAC (38). The control computer is electrically connected to the stereo DSP. ! It is connected to the t-bo DSP. The servo DSP is also connected to the DAC, which is also connected to the dynamic mirror. Control 7 complete; Data flows from the computer to the 1) SP, from the servo DSP to 1) AC, and further to the dynamic mirror. The control mirror is connected to the encoder, and from the encoder there is a feedback connection to the servo DSP, thereby forming a servo control feedback loop. As will be discussed in more detail later, the servo control feedback loop allows accurate mirror position even at high speeds. Is each DSP fast AT&T? The aWE DSP32C digital signal processor is a programmable circuit with a 32-bit floating point unit, 1.6-/24-bit fixed point unit, on-chip memory, and flexible serial and parallel 110 ports. As such, the DSP32C can be programmed to support a wide range of computing applications. For more information about DSP, please refer to AT&T WE DSP32C Digital Signal Processor Information.
Manual C (December 1988) and related manuals are available. These are comprehensive instructions, referred to throughout the text, and given comprehensive instructions. XY, 1:, Korg is a Canon radar rotary: I-radar 7-dar---K-1 ultra high resolution (81,000 barnos/rotation) ultra high precision rotation each Such as Sen-go. Figure immobile. As shown, the control 'nil-piuter is the elevator (9) 1. Wiper (7), Branniya (5),
Racer (fl) Attenuator (22), Leveler (39)
) and beam profiler (8), and controls these and in some cases also monitors them. As shown in the figure, the position of the dynamic mirror is
It is directly controlled by SP. Historically, in order to position a mirror in a particular direction, the servo DSP first determines the current position of the mirror through an encoder and numerically compares it with the desired destination. Then send the appropriate command. According to the above-described method, the mirror is moved to the desired position in a linked manner. Also connected to servo DSP

The high-speed shutter controlled by this is also shown in Figure 6.1. It is. As will be described later, the servo DSP sends commands for opening and closing the shutter and at the same time sends commands for positioning the dynamic mirror (that is, the laser beam). The shutter takes a certain amount of response time. Therefore, when the beam starts to enter the hollow part from the solid part of a certain layer, the servo DS)) will send out a shutter closing command or
The timing of sending the command is such that the closing of the four-wheeler and the entry into the hollow part occur at the same time. Conversely, when the beam enters the hollow section into the same section, the shutter receives or accepts an opening command, and the sending of the command takes into account the response time of the shutter. A control computer receives a build file for a part and generates hatching and fill vectors from internal edge vectors. The details of these vector formats have been described in US patent application Ser. No. 331.664, but they will be briefly summarized here. The vectors describe the movement of the laser beam along the resin surface L5, and each vector has a starting point and an ending point. For each vector, the laser beam traverses the resin surface starting at the beginning of the vector and stopping at the end point. Vectors are generated for each layer. The edge vector describes the edge of each layer, the ノ\・yf song vector describes the interior of each layer, and the skin vector describes all the outer surfaces of that part that fall into that layer.
I'm here. The edge, )\-soching, and fill vectors are transferred to the stereo DSP, where laser jump and movement commands are generated. For each command
/- Specify the desired X and Y axis positions with the laser beam. The movement and jump commands will be explained in detail later. The Stereo L/O DSP also performs movement commands that include the geometry at the desired position, followed by four lines of drift correction (drift correction is based on the X, Y coordinates of the jump vector). The above is not executed, and the geometry-compensation is executed only at the end point). The corrected X,
A command using the Y position is sent to the servo DSP, which moves the mirror to the desired position after first detecting the current position. Sir, Bo DSP also sends an appropriate command to the high-speed shutter, calculates the constant response time of the jack, and synchronizes it with the C Mira command. , the control computer, and each DSP are discussed below. Part Manager Details The SLA Part Manager or PartManager is software that runs on a workstation.
The input to er is all the component files related to a certain part, and the build file related to that part is 4,
to be accomplished. The build file is then transformed and used by a controlling computer to form the part layer by layer. During build file generation, Part and Manager generate many other files including style files, slice files, and parts files. Part Manager for each component file of a part
reminds the user to split the file into a range of layers. You will then be prompted to define build parameters for each range. Build parameters are preset and saved in files called style files, and for each style file you can select a pill type and set different attributes, such as strength, aesthetics, and part formation speed. , or the accuracy of building another style of library. After this, in PartManage+, the user only needs to associate a certain style file with each configuration range.The style file is preconfigured, so the user does not need to be familiar with the details of SLA implementation. . Even beginners can use Pa without much preliminary training.
, can use pt Manager 0 I will explain later,
Pa't Manager implements the above-mentioned functionality in a Spreadsheet-style screen display. Once you have specified all the style files for all the ranges of each component of a part, , na
ger is attached to the part and combines the styles and data from the component files to generate the part file. Spreadsheets provide a visual representation of the organization's relationships among files, ranges, and styles. A parts file is basically a data set of a spreadsheet at the time of 1.7, and it is possible to visually represent the data in Spreadsheet 1. Therefore, if the user decides to stop preparing a part at an unfinished time, that part file will be used to regenerate the unfinished sub-L knot sheet at a later time. After the build file is created, Part Manag
er slices each component file, creates a slice file of edge vector data that describes the edges of each layer of the component, and combines all style files with build parameters from the style files to create a single build file. summarized in. The build file is composed for each layer, and the data for each layer is related to that layer (a combination of vector data from all slice files for each component IJ and build parameters from the style file for each component). Part, Mar
The +ager also accommodates conflicts between the build parameters associated with the -) layer, which is made up of various components. The overall picture of the file management process is shown in Figure 7. Illustration 12. As per above, independent component file 4Oa
A range is specified in S40b, 40c (and later, a corresponding style file (41) is selected for each range. The components and style files are now combined to generate a part file 42. Individual files from that part are The component files are sliced into slice files 43a, 43h, 43e, 4 using Part Manager.
Generate 3d. The slice file is further combined with build parameters from the style file to create a build file (47). Each slice file (also known as an sN file) is organized in layers. Edge vectors are written in each layer and are grouped in edge vector format. Here, the format of the edge vector and its associated memory are as follows. 21 Monique Description LB Edge FDB of layer Edge FUB facing the bottom surface
The edge NFDB facing the top surface is the edge NFUB facing the bottom surface.
The vector of each edge that mostly faces the soil surface is 4.
The first two points indicate the X and Y coordinates of the vector origin, and the last two points indicate the X and Y coordinates of the vector's end point. , let's go through the slil Mitsuair example next. / 567 npls, SL Engineering File L 8000 T, R 7 only * Ne * Other file formats, such as those of bar, rough file, style file, build-file, will be discussed in the next chapter. As mentioned above, each component file conforms to the format of the S profile file. For each component file, user range information and style files in that range are specified using the spreadsheet shown in FIG. Each spreadsheet is associated with a part, and the columns of the spreadsheet represent the components of that part. The rows also indicate the range of layers. If you consider Figure 8, the spreadsheet is divided into six parts. ~Header bar in Part 1 of the spreadsheet = (45
) displays the program name and version number,
This is followed by the file name of the current sub 1/sheet. Ne if the file name is not specified
The name wPa, rt, is displayed. -The menu bar (47) at the top of the header shows Part, Component, View, and P.
repair, Config. Part M including FileMan and He1p
a[lag-': The function name of I-r is displayed. The column to the left of the spreadsheet cell (45) is the range of values. If you move away from the top cell in this column containing Z 5pace, this column will be blank when a new spreadsheet is first displayed. The most recent rows (40) contain names of constituent elements, except for the first cell. These cells are new and will be white when you first view the new spreadsheet. The remaining cells in the spreadsheet are for data, with a 1° line and 1 knot horizontal scroll bar (51
) and the vertical crossbar (50) at the right end of the spreadsheet are used to access cells that are not currently displayed. 8), the second, third, fourth, and fifth columns 40a, 40b, 40e, and 40d are the constituent elements
(the first column (45) is not used to display range information). The 8th column displayed on the screen is 22
It can be larger than a row. It's a month-long window in a spreadsheet that may be as large as these rows and columns, or even larger. Control the red χ medium cursor position with an optical mouse or
You can use the mouse pad to move the cursor and align it with any point on the screen. Part
To control the Manager's operations, the user can move the red arrow cursor to the menu name or item [1] and press one of the three buttons on the mouse. There are three ways to operate mouse buttons. They are called clicks, presses, and drags. The expression "click a button" means to speak immediately after pressing the button. ``Press a button'' means to press a button and then continue to press it until some response appears on the screen. 1. ``Drag'' refers to moving the mouse while holding down the mouse button. Mouse 1. r has three buttons. These buttons are used in the following "I-button", - The human mouse button is Menu, and is used to select item 1-1. ,A.,In most cases, menu item 1] is the menu! Point to the menu name in <- and select 1 or 2, then press the button until the desired menu is highlighted in red in red letters on a white background [and drag down again]. If you release the mouse button at this point, the menu items will become white. Another use of the left button is to select a simulated button on screen 1-. In most cases, a dialog box appears on the screen asking you to select the menu item. As shown in Figure 9, most dialog boxes include
5elect (48) and Car+eel (49) here
) button. The user can select the button by moving the mouse so that the tip of the red arrow is within the outer periphery of the button and clicking the left button. It is best to release the mouse button when the tip of the pointer is inside the button. Yet another useful use of the mouse button is with respect to the scroll bars along the right-hand side and the bottom of the main screen. In FIG. 8, the scrolls to the right and C are indicated by 41 reference numbers 50 and 51, respectively. If there is more data than can be seen on the screen, you can use the scroll bar to move the display to other parts of the screen. To move mu from right to left, move the red arrow pointer so that it aligns with the red square on the horizontal scroll bar at the bottom of the Zub1/Tsudo sheet. If you drag the mouse 4 to the left or right while holding down the button, the field of view of the spray/sheet will also move horizontally by 1J. Similarly, touch the vertical scroll square on the side of the spreadsheet to view the spreadsheet screen! t"-, you can do it without moving. At any time, the position of the red square in the scroll bar indicates the position of the first row and column of the spreadsheet section.

[,2
ing. Some dialog boxes (Figure 9) display a list of file names. There may be so many names that the list cannot all be displayed in the limited space available. In that case, when moving the spreadsheet up or down on the screen, use the method described above (using the vertical scroll bar C52 on the right side of the dialog box) to move the list of file names up or down. I can do it. Middle Button Another way to use the middle mouse button is to move the entire window on the screen. Paft Manager
The main window is initially located in the center of the workstation screen. You can also move the entire window to another location on the screen by pointing the red arrow pointer to the white title bar at the top of the spreadsheet, holding down the center button, and dragging the mouse. The magenta rectangle moves on the screen 1.2 to display the new position of the window. The window moves to its new position when you release the mouse button. Another example of moving a window is
When an image of a component or part is displayed using the View function, this will be described later. In some cases, it appears in a dialog box or screen, on a spray sheet, or on a part or component. You can use the technique described in item 1 to move the dialog box inserted on the screen 7 and see hidden spreadsheets or images. The middle mouse button is also used to move columns in a spreadsheet. With the spreadsheet displayed, move the red arrow pointer to the column you want. When you press the center button, the selected column will be outlined in magenta. Then move the red arrow to the right or left to move to that column or to a new position. When you release the mouse button, the column appears in its new position. Other columns are moved to fill the empty column positions. Right button Under certain conditions, it is possible to display two or more windows at the same time. In most cases, a small window closebox will appear in the top right corner of all but the first window. For windows with a window close box, point the red arrow pointer inside the box 2.
Click the right mouse button to close it. When a user first accesses Part Manager, the main spreadsheet screen is displayed. When the spreadsheet appears as shown in Figure 10, a dialog box (52) or spreadsheet 1 appears. There are three mock buttons (53, 54, and 55) in the dialog box that allow the user to choose between creating a new part, opening an existing part, or requesting help. I can do it. To create a new part, the user points to the New button (53) and clicks the left mouse button. The workstation responds by displaying a blank spreadsheet followed by a dialog box. Figure 1
1, the dialog box contains a list of all component files in the current directory (5G).
is displayed. The list that appears in the dialog box is the available components, and the two buttons (57, 58) each
These are the Iect and Canca+ buttons. If there are more component files than can be displayed, use the Goal bar on the right side of the screen to
You can scroll through file names. To select a file name, select it and click the left mouse button. With this action, the file name changes from black to red and is displayed in the square area directly above the Copy File Name button. To move selected files to the spreadsheet, point to the Select button and click the left mouse button. Note that multiple components can be selected here. To display the previous spreadsheet (without the component-7 isle name in the cell) as is, point to the Cancel button and click the left mouse button. ComponentsThe components of a certain part can be selected from the file selection dialog box by pointing to one file name and clicking the left mouse button (FIG. 11). The file name changes from black to red to indicate that it has been selected. You can select as many file names as you like using this method. If you accidentally select the 7th isle, you can deselect it by pointing to the 7th isle and clicking the mouse button. After this, the color of the -7 isle name returns to black again. Once all the necessary file names have been selected in this manner, pointing to the 5 select button at the top of the dialog box and clicking the left mouse button will load the selected component files into the spreadsheet. Assuming that the spreadsheet is blank, Figure 11
If component file 3ii logo, st I is selected, a column (59) for this file is created in the spreadsheet as shown in FIG. The first cell at the end of this column displays the component file name. Left column (60
) shows the positions of part of the component 1 and part F in the continuation direction. Initially, only three cells in the component column are used. The top cell (59a, ) indicates the component name. The second [-] cell (59b) is entered with * * T OP and indicates the top layer of the component. The last cell (59c) contains **Default,
It indicates the lowest layer of the component, and the default style file name is l', )efault, which indicates that it is applied to all layers of this component (this will be explained later). The concept of styles and style files will be explained in more detail below. As shown above, let's assume that we are defining some style or build parameters, such as the thickness of a certain layer and the details of how it will be formed. Since it is selected from and added to the spreadsheet, an additional column or column is created for each calibration element.In Figure 13, as an example,
Component file ey1. stl, eyll, and stl are selected, and columns (61) and (62) are created for each. A new row is also created for the upper layer of each component. As shown in FIG. 12, initially 3i1.10g0. S
After t and l are selected, rows (60a) and (6,O
Only b) was made. When two more components are added, rows (60c) and (60e) are also created, corresponding to eyl, stl and eyll, stl, respectively.
ing. The row (BOd) was created in the F part of cyl and stl. The T' parts of c, yl, and 5til coincide with the F part of 3dlogo and st1, so there was no 8th part to make another i in this layer. As shown in the figure, these rows are arranged so that they match vertically. After adding many components to a spreadsheet, it is quite convenient to reposition the columns so that related components are adjacent. Press the center button to create a new column 7
You can move it by dragging it to a new position. Once you have assembled the components of the parts in Spreadsheet 1, you can use the View function to check the parts. The View function will be explained later. When selecting a component in a style file, the default style file, or default style, is initially attached to this IJ.
Default character is the name component button 1. . I'm very careful to put it in the corresponding cell. However, if the default style is not suitable, you can change it to another preexisting style, or create a new style for the component. To change a style file that already exists, point to the cell where the current style is defined and click the left mouse button.For example, to change the style file for a component of 3dl ogo: , the user is **
Point to the cell in the 3d1 ogo column containing Default and click the mouse button. Here, the range pop-up menu (63
) appears. When the range pop-up menu (63) appears, the user points to Add with the mouse and clicks the button 7 to display the style dialog box on the spreadsheet. The style dialog box (64) is shown in FIG. The upper right corner (65) of this dialog box shows the location of the top layer of the component.Kone's "10" has its directory list (66), which includes the 1J style file. As shown in the figure 11, Paft resistance a
nager originally had 2 style-7 isle directories.
Set in the location (5 is 4, directory /st, yl
es stores style files created by system users. Directory /usr/3d/11
b/5styles has a style file provided by Part Manager or someone else. When the dialog box first appears, the directory containing the user's style files is selected and the style file names (68
) is displayed 1, and the current style name is displayed in a box (67) near the bottom of the dialog box. The window (66) displays style directory names, and the window (B8) displays file names, and these display as many file names as can be displayed at one time. Your window is also good. Use the side scroll bars (68a and 66b) to view the current view and display missing data/directories and file names. . To change the display name to another one, point the user at the desired date name (7) and click the left mouse button.Similarly, to select a style, click the style name. Click the mouse with your finger (,). The selected style name will be displayed in the box (69) near the dialog box. To use the selected style, click the button at the bottom of the dialog box. Click the Done button (71) and click the 'r left mouse button.6 As shown in Figure 16,
) -- Horn 9' Ia ogbozocus, Replac
e Range hox (70) appears on the screen 3
, here is the current n- style, this example is **I) ef
It also provides the option to delete the ``fault'' and replace it with a new style of your choice. Normally, select the user or delete button by pressing the Yes button. and click the mouse button. By doing this, a double-edged dialogue box disappears from the screen. Choice 1. The style of the selected component will be replaced with the newly selected one. As shown in FIG. 17, the cell (73) is style **I)
Changed from efault to **Std 10. Me1Ii) Then, t? eplaec Ran) He 9
'-When Iaogbo-nksu (70) is displayed, select the new 7i/mita-iru 1st order [2], so press the No button in this dialog box C74
) Point to G and click the left mouse button to enter +l Replay Rangf,! When the dialog box is not displayed, the new 1. To cancel the selection of new styles, press 6Add R as shown in Figure 15.
Canc, e at the bottom of the ange dialog box
Point to the l button (75) and click. A style is a set of build parameters that define how a range of objects is created. [The styles used are the style files on disk 1 and 1. You can save it with . PartMariager contains all of this parameter in a certain style file. You can control how each layer in that range is formed. If desired, you can edit the style of the border to change the selected barometer, as described below. When editing the style and also supplying the resulting light 4)
, the style name used for the maid is saved on the disk, and the style name is displayed in the same style.However, it is displayed in one of the different tiles, or the tile is the same name. Occurrence can be avoided. If the changed style parameters or other components "C" are also used, the parameters can be written as a new style file. Point to the style name of 1 and click the left button.
Hover your mouse over cell (73) and select this style.
,ing. A pop-up menu similar to (63) in FIG. 14 appears. If the pop-up menu is currently the first one, -
7- The user points P, d it (C click the mouse,
Edit dialog box (7B) in Figure 18 t = Sub [Display on Radjit. Each field in the style editing dialog box contains build parameters. The meaning of each field and its associated built-in meter in the edit dialog box is explained below. 5style - The style field (77) contains the current style name. If you change the current style, you will usually change this name as well. However, if you create a new style and keep the current style, you must change the style name to a name that does not previously exist in the current style directory. To change the style name, point to the box containing the current style name, box (77) in the figure, and click the left button. An edit cursor appears to the right of the -7 file name. When you type the first letter of your new name, your old name disappears from the screen and the new letter appears. Subsequent characters appear on the screen exactly as you type them. The Baekspaee key is used to delete misspelled letters. Press the Enter key to skip typing the new name. The edit cursor now moves to the Ricoh style name (76). T? eeoaj: A recoat style is associated with its own style file (and thereby with each style on the spreadsheet). The edit dialog box initially displays the name of the recoat style associated with the current style. .In Figure 18,
A dialog box (76) appears that is associated with the Std to style or the coat style [)efault. If you want to keep the recourt style unchanged and leave it at 2, you don't need to change the court field. To use a different recoat style, if the recoat file already exists, point to the 1...o ad button (79) at the bottom of the recoat field and click the mouse button. The box in Figure 15 (68
) will appear, and the available code file names will be displayed. point to the desired recoat file name and click the mouse button. To confirm that you have selected the correct recoat file, type the recoat style name in the field below the list of recoat styles, and click Don
Point to the e button with the mouse and click. Edit the current Ricoh style and create a new one.
7-1. You can also create a style. This will be explained later,
It's dawning. End Z: The initial value appearing in the end Z field (8G) is that at the top of the current range. You can change this value by pointing to it, clicking the C mouse button, and typing a new number. 5tart Z: The initial value appearing in the Z start field (80) is that of the top of the current range. ”, Record Z
You can change the second value in the same way as the final r (80) value. Layer Th1ckness Layer I! The initial value appearing in the no yield (82) of I is that of the current layer J9 only. This value can be changed in the same seven ways as the Z end (80) value. Layer thickness is usually 0.00! 0.020 from i
It is in the range of 1/2 mm (0.13 to 0.51 mm). Description The sentence in F (-) (83) is a simple explanation of its style.The current n style includes all explanations in y-) (, s or originally). Commentary editing follows the new style. There is space for up to 53 characters. MSA In this °5 field, the current minimum surface angle is first measured in Table 4). Change this value in the same way as the Z end (80) value -C. The usable adhesive A value is determined by the layer thickness and -12 width. , iJ! Uniform MSA
The values are as shown below. Hirothickness MSA. 0.020 inches 60 0 , 0 ], 01 inches 50 0 , (+50 inches 43 +1atch TyT) ρ The hatch format defined in the current style is initially entered in the hatch format field (84). To change the hatch format, select the current hatch format 4 Definition 6 Select the word you want (Run with your left hand, press 1 or press 1 and press and hold the button to spell. Then drag the mouse l, 7.
Don't release the taran button 5. Available hatch formats are ↓■
'-' angle hp, box, diagonal, kobi castano. The first three of these hatch types (! have already been set and no further input is required. The hatch types for regular-unigons are regular squares on the surface of 2'C, square 7\- \It defines where to trace the line, and where to trace the hook in the hook hatch format.
The diagonal format also defines where to trace to the diagonal. Details regarding these hatch types were previously filed in US Pat. No. 331,664 and 12. If you select a custom hatch format, a dialog box will appear that allows you to manually adjust the spacing and angle of the hatch. Fitl Type - The fill type that is Defined I in the current style is initially displayed in the Fill Type field (85). To change the fill format, point to the word that defines the current fill format and hold down the left mouse button. After 7, drag the mouse down and release the mouse button when the desired filling format is displayed. Use i-
Possible filling formats are X filling, Y filling, and X&Y filling. For more information on these, see U.S. Patent Application No. 331
．． 664 has already been submitted 1,=. After selecting one of these, a dialog box will appear allowing you to enter the fill spacing. 0vereure Amounts: The amount of overcure defined by the current style in nine fields (86). The amount of excess hardening can be defined for each valley/ctor format. A series (7) 7 (-rule)- such as 88a, 86b, 8fie, etc. is displayed. To measure any or all of these excess hardening amounts, see Chapter 11 Z.
Use the final r value and the 1, h method. '74 with excessive hardening amount
- The edge of the LB & LH layer and the layer! Inch - NFDB - Downward J' Near flat edge N F D H - T " Direction near flat 7X Sochi NFDF
- Downward flat tQ (near filling - F U B & N
F U B ~ Flat 1-facing edge and nearly flat upward edge FDB - Flat downward edge FDH - - Flat downward hatch FDF ~ Flat downward filling FUF & NFUF - - Flat upward filling! An example of 2 types is as follows: / Sample 5 style Fi, 1 eocNFD
B = 0.006 ocNFDH = 0.006 ocNFDF = 0.003 ocFDB = 0 ocFDH = 0 ocFDF = 0 ocFUB NFUB -0,006 Each ゛ field is self-explanatory. Yes, there are some exceptions that I will explain below.The Notes field is
Corresponding to Feepard [1, [Jnjts], yield indicates the middle numerical value (inches) in the file. NuIIHatc, h is 2〕＼ and 1 if there are three hatch lines to form the 11-heptagonal hatch form.
, Hatchangies shows that the lines of kneading are placed at 60 degrees, 120 degrees, and 0 degrees from the water, respectively. Also) 1. For alc and hspaees, the spacing between these lines is 0.05 inch. Nu+vFi l
l field uses one filling fin to refine the filling.
It is my pleasure to be your envoy.
s indicates that it is placed at an angle of 0 degrees. Also, Fi l l5 paees is line spacing or 0. (lr)
3 Ishiji. Reeoav 5styles: As mentioned above, Reeoav 5styles are associated with each style file. Defined in the 21-tostyle file is a parameter C that controls the lineage and 7th processing that is performed before forming a part or part. When editing a style, save the existing RJ style shown in the style file, and then select the existing 7J style [, or, some,
It is also possible to create a new coat style by editing an existing style. The front 2 right lotus C is explained as °C. For the following explanation, please refer to the explanation below. click. The Lineat Style Edit dialog box (88), as shown in Figure 19,
The fields in box (18) are related to neat parameters.The meaning of fields in each field will be explained in detail. : The initial value shown in Yield (89) is the waiting time (in seconds after the previous layer formation) required for the surface of the liquefied resin in the vat to reach a perfect level before the start of the renealing process. It is expressed as a number. To change this 1st shift, change the number in the petite group, dialog box, or line in 1st order like L.). Z Dip Delay: 7 (-LD(,9[1
The initial values shown in ) are the surface of the liquid 1, and the next layer is created from the surface of the liquid with a linear tab blade. It is the number of seconds. To change this number, follow the same procedure as changing the numerical value in the style dialog box (i.). S:Field (91) The initial value represents the same number of plate wipes performed in 8 layers. To change this value, follow the same steps as changing the value in the style dialog box. The numerical range is from 0 to 7 (not wiped). Z Dip Dista, r+ee: Field (9
The initial value shown in 2) is the distance below the surface of the liquefied resin that the part is immersed in by the elevator, and is necessary for a high velocity fresh resin to flow over the previous layer. This value is defined in the associated style file and written using the following method: To change it, use the same procedure as changing the value in the style dialog box. Ruto (9
3) Indication 1 Initial value 1.During processing, there is a speed I5t'' at which the rehake moves the gear. The initial value shown in field (94) is the same as changing the value in the style dialog box. - is the acceleration when changing the velocity. This value is defined in the associated style file and is written in the ``t'' method, and can be changed using the same method as changing the value in the style dialog box. Do it in steps. Blade Gap, Velocity, and
Delay: The lineart style can be defined with seven wipes, and the blade gap, speed, and delay time can be defined for each. This field allows you to set these parameters for each wipe. The blade gap, velocity, and delay time fields are each (95
), (96), and (97) are numbered. The wipe gap is the gap between the blade curve and the previous layer, the blade speed is literally the blade speed, and the delay time is the delay time until the next wipe, if any. The initial value is the wipe parameter of the current glue coat style. As explained above, to change any of these parameters, follow the same steps as changing the values in the Style dialog box. The fields of the second file, Sample Recoat 5style File, which shows an example of the linet style below, do not require explanation and correspond to the fields explained so far. Some exceptions will be explained below. The Notes field is the previous Descr.
Corresponds to iption 2, travelVel l-7
(Linet is the velocity of the plate in this area in the vat until it touches the part being made. When it hits the part being made, define the plate velocity in the bladeVell field. For more information on lineat-C U.S. Patent Application No. 265.039, International Patent Application Lyon
& Lyon Doeket No. 188/189 PCT
It has already been submitted. Saving style parameters To write a new style as a file: Sa, v
Point the e button (98 in FIG. 18) on the style editing dialog box and click the mouse button. The file will be saved with the name currently displayed. If a file with that name already exists in the current directory, a dialog box will appear giving the user the chance to overwrite the already existing file or cancel. After canceling, the user can choose a different file name to avoid erasing the already existing file. When a component is first added to a spreadsheet, there is a range of layers that extend from the bottom of the component's layers to the top. For example, in Figure 12, component 3d is added to the spreadsheet.
When adding logo, only one range, o, ooo
A height of 1.4628 inches was made from inches. It is often better to divide the layer of a component into several ranges, so that each range can be associated with a different style (a tori coat style) that is optimized for that section of the component. To set a new range for a component, point to the column representing the component and click the left mouse button. The range pop-up menu shown at 63 in FIG. 14 will appear, so point to Add and click with the mouse, and the range addition dialog box will appear on the screen. In FIG. 20, the Add Range dialog box is referenced 98. In this dialog box, point to the name of the style you wish to use for the new range from the list of available styles (99) and click the mouse button. In this example, the style name of the new range is 5td10. From this dialog box, you can also change the Z end and 2 start fields (100) to make the range's start and end γ points correspond to the new range. In this example, the new range starts at 0.7000 inches and ends at 0.800 inches. After making these changes in the dialog box, click Do.
Point to the ne button (lot) and click the mouse button. The dialog box disappears and a new range is added to the spreadsheet. In the example, before adding a new range, start from 0.0000 inches and 1. ．． 462
5td for 3dlogo components ending in 6 inches
I was using a style called 1.0. Adding a new range, the Std 10 style is now Std 5 which covers a range of 0.7000" to 0.8000"
It has been replaced by style. In FIG. 21, the style of cell (1, 02), which was previously Std 10, changes to 5td-I') in response to the change. In the second example, add one range I) D l = ta 1j,
In reality, as many ranges as necessary τi can be added IJ11+. Each new range ζ can replace the previously defined style in its range C. /Additional Features Users must first start running a program by pressing F3 to create a new component, open an existing component, or request help. This is illustrated in Figure 1. To open up an existing part and place it on the strip/rad sheet so that its structure can be viewed and perhaps modified. , 7.-Sir indicates 0pen button 54 7.
Move the left mouse button. The workstation responds by displaying a dialog box that lists existing part files, files, and files with names that have part extensions. FIG. 21 shows such a dialog box. The dialog box is similar to that showing component file names and is illustrated in FIG. 115. Figure 22 shows a dialog box explaining the component file name.
Select the part of part SOI/L side t Jd, or, ---1), or remove without selecting the part/I'11
, or. However, one difference is that although a large number of components can be selected, only one part can be selected at a time. Additional functions can be added by selecting item -7 from the menu displayed at the top of the screen (designated by number 46 in Figure 8). The main features are neip
It is a function. He1p can be accessed from the main screen by pointing to elp at the right end of the menu bar and moving the left mouse button. In some cases, He1p can also be accessed from a dialog box. This can be accessed from the dialog box by pointing to the He1p button and moving the left mouse button.
The elp screen will be displayed. FIG. 23 shows the ttelp screen. The Help screen that is superimposed on the existing screen is composed of a list of headings 103 on the left and information 104 regarding these headings on the right. The information is arranged in the order of representative 1 - sir or μ - month). 4. To see the information on the special section, please look at the heading in the index on the left. , move the mouse button on the - side, and at the beginning of 10,
The headings and lists on the side are arranged in the same order as the more detailed information on the right. The list on the left is 5o on the left Y of the screen.
rted Topie, point to s box 105,
You can change the alpha head order by moving the mouse button. There are two ways to cancel the Hell's orientation and return to the previous display. One is to specify the Done button 106 at the bottom of the Help screen and move the left mouse button, and the other is to specify the window close box 107 of the He1p screen and move the right mouse button. It is. Managing Component File Returning to Figure 10, the user opens the new component file.
(by pressing the w button), new parts are created by adding components to the spreadsheet. If the user decides to modify the existing part file instead (by pressing the 0pen button), add the component on the spreadsheet,
Existing parts are modified by deleting or modifying them. The component menu provides access to these functions. To access the component menu, the user points to the word Component in the top menu bar of the spreadsheet (the top menu bar is item 47 in Figure 8) and presses the left mouse button. , until the component menu appears. Component menu 107 is shown in FIG. To select a function on a menu, the user pulls the mouse down until the desired menu item becomes bright, then releases the mouse to access that item. Each item [ ] of the component menu will be explained individually below. Add: When you select Add from the menu, components A, c-1d, dialog, etc. are added to the spread note as shown in Figures 11 and 12.
Box or superimposed. This dialog box is used to add components to the part, as described above. Delete To delete a component from the part, the user points to the spreadsheet row corresponding to the component and moves the left mouse button. The user then pulls down on the component menu until Delete becomes bright and two buttons appear in the dialog box. User Del
, ete button and move the mouse button to delete the component. Alternatively, the user may indicate the Cancel button and move the mouse button to stop deleting the component. 5tatus: Click on the spreadsheet line that represents the part, and select S on the component menu.
When you access t a, v, u s i, the status of the component is displayed. At that time, the status of the component is displayed as shown in FIG. As shown, the displayed state indicates the length of the part in the X, Y, and Zh directions. When the status is displayed, initially the three slice axis buttons 1
One of 08a, 108b and 108c is lit to indicate the currently selected slice axis. The currently selected slicing axis is the axis along which the part is cut in layers. The slice axis can be changed by pointing the appropriate button and clicking the mouse button. In the case of changing the slice axis, the change is made by instructing the OK button 109 and clicking the mouse button. Alternatively, the user may click the cancel + button 110 to return to the state before the change. Colbine: Component files created with some CAD systems may not represent components with closed surfaces. However, at the part manager's request, the component file must represent the part of the part that has a sealing surface. To be able to do this, component files may have to be combined to obtain a component that is combined with a sealing surface. If the imported CAD file represents a component with a closed surface,
There is no need to combine files. If files need to be combined, the user selects Collbine from the component menu. As shown in FIG. 26, at this point a combination component dialog box 111 is displayed with a selection file window 112 overlaid. Next, the user selects files to be combined from the selection file window 1.12. Once all files have been selected, the user selects a name for the combined file and edits the name in the file name field 113 so that the selected name is displayed. Finally, the user can
Click on the mbine button 114 to combine the files. Translate: Translate (movement) C. This is the operation of moving a component from one place to another in a resin tank. Translation sometimes involves orienting the part appropriately to eliminate trapped volumes of liquid resin, locating flat surfaces on a horizontal XY plane, reducing the number of unsupported surfaces, etc. is necessary. Individual components can be moved, or all components of a part can be moved together. Individual components or multiple components can be moved a predetermined distance in the X, Y, and Z directions. Alternatively, they can be moved to the center of the resin tank, or to the origin if the origin is not the center. To move a component, the user selects Translate from the component menu. Then, as shown in FIG. 27, a translate component dialog box is displayed on the screen. The center panel 115 of the dialog box shows the name of the component on the current spreadsheet. The All button 116 at the top right allows selection of whether to move individual :1 components or all components of the part together. Initially, the button is centered in green and the word Pavt, followed by the current part name, is in the top left of the dialog box. This indicates that the entire piece consisting of all components of the current spreadsheet is selected. To select the movement of individual components, the user points to button 117 at the top right of the dialog box and clicks the mouse. The center of the button turns red,
The word Component is displayed at the top left of the dialog box. To select an individual component, the user points to the component name in the center panel 115 and clicks the left mouse button. The selected component name changes from black to red, and the component name is displayed at the top center of the dialog box. To move a component or part a predetermined distance in the X direction, the user points to the field to the right of the
Move the left mouse button until the edit cursor appears to the right of the current value. The user types a new number representing the distance traveled. Positive numbers represent movement to the right, negative numbers represent movement to the left. If necessary, use the Status function described above.
+r+etfor+) to display the position of the component or part after it has been moved to confirm the movement. Without specifying a predetermined distance to move the component or part, click the Orfglr+ button 120 to move the component or part to the origin of the axis, or click the Cer+ter In vat button 121 to move the component or part to the resin tank. It is also possible to move it to the center. Rotate: The rotate function is similar to the translate function. The latter moves the part or component linearly, while the former rotates the part or component about the X, Y or Z axis. The Rotate dialog box (Figure 28) is similar to the Translate dialog box, but instead of units such as inches for a given linear distance, rotation values about the X, Y, or two axes are expressed in degrees (degrees). The difference is in how they are specified (angle). 5cale: Magnification al! The functions are similar to the translate and rotate functions. 5cale can be used to select a component of a part and scale it uniformly in all directions, or by different predetermined factors in the x, y, and z directions. The top of the Magnification dialog box (Figure 29) is similar to the top of the Translate and Rotate dialog boxes. The button 122 labeled 5eaHAll in the bottom half of the magnification dialog box is initially set to scale the two components or parts by the same magnification in all directions. Field 1 below the button
There is an initial magnification value in 2 and 3, and an edit cursor on the right side. To change the magnification, the user replaces the 1 with the desired value. To be able to scale the component or part by different factors in the X, Y, and Z directions, the user clicks on the 5cale All button. The edit cursor disappears from the field below the button and appears in the X field. The cursor is
By pointing to that field and clicking the mouse button, the Y field dot can be moved to the Z field. The user selects the appropriate field and sets the initial value.
Change it to the desired value. Once the enlargement or reduction is complete, the user clicks the SCa, 11 button at the bottom of the dialog box. Scale right hand (right hand)
ed) CAD/CAM files to the left hand (left, -handle) according to the request of the part manager.
d) Can be used to convert files. Light no＼
A user can convert a left-hand CAD/CAM file to a left-hand format by applying a -1 magnification only in the Z direction. The Part item 1-1 in the main menu at the top of the component file management screen provides access to functions for processing component files. These functions can be accessed by displaying the part 75 (2) and clicking the mouse button. At this point, the parts menu shown in FIG. 30 is displayed. Each component menu will be explained below. New: The first item in the parts menu is ~eV. If you select this item, it indicates that you want to insert a new part file, and an empty spreadsheet will be displayed. 0pen: second term] is 0pen. If this item is selected, it indicates that an existing component file will be opened, and a list of component files in the current directory (see FIG. 9) will be displayed. After a part file is selected to be opened, the spreadsheet shows the components of that part. Llriits: [InHs menu item is C
It is possible to specify the unit in which AD/CAK1 data is specified and the unit in which the part manager value is displayed. If this item is selected, in (inch) is used as the unit for both CAD/CAM and Part Manager, ■11
(1/1000th of an inch), IIIm (millimeter), cm (centmeter), m (meter)
Or tt (feet) can be selected. This is important because the user may want to use a different unit in the part manager than the one used in the CAD/CAM system. 5ave: A 5ave menu item saves the current component in the current directory using the current name. 5ave As: The 5ave As menu item also saves the current component in the current directory. L, key, gives you the opportunity to use =7 file names other than the current file name. 5tatus: 5tat, us menu item (Fig. 31) ('', indicates the range of related information such as the overall method of the part and the preparation of files to be sent to the control combicoater for manufacturing the part. Quffit: Quit menu is used to quit the neat manager program.If any file has been changed and not saved, the user can save the changed file or The format of the part file can be summarized as follows: 1. Part file version number 2. Part data Part area unit manufacturing parameters Basic parameters Final spreadsheet display position and magnification 3. Base φ style data 4, bass lute style data 5,...
Multiple copy offset 6 Number of components 7, For each component, te a, Component data stl File name Explanation area Slice axis b, Number of styles C1 For each style slil mistile Aligned style data According to the above format An example of a parts file is shown below. Parts cycle, parts file version 1.090 Description area X: 0.0Of) 0. '-3,2400Y: 0
．． 0000.3.2400 Z: 1.0000, 3.8200 Slice decomposition - 8000.000000 CAD data unit is rinJ (inch) Manufacturing parameter X, 5cale -1,, 000000, Y 5
caie-1,000000, Z 5cale = 1
．． 000000 Overcure factor 1.000000 Speed factor 0.50000O Nw Copies -0, gin Dfst -0,5
00000° Beam Comp O Component pot cycle area ．． 000 to 3.820, I
Th1ck -0,020 Desc + default style hatch Type -0, null 1 (aLcll
= 30:hAngle -60,00,hspac
e -0,05゜1 :h Angle -120゜D
O,h 5pace -0,05゜2:hAngle-
0,00,hspace-0,05rth I Type
-0, ton Fill-10: f Angle -
80,00゜1, : r Angle -90,00
゜2: r Angle = 0.00゜Overcured 0CLB-LH CNFDB CNFDH CNFDF CFDB CFDH OCF D F f 5pace -0,00゜r 5pace = 0. 00, r 5pace -0,00 ooo. Ten, ten. Ten, ten. Ten. Ten. Ten. 0CFUB NFUB-0,00600CFUF
NFUF-0,0060 Recoat style: When omitted Desc: When omitted Recoat style zDip= 0
．． 750. zDip Vel -1,000゜zD
fip Aeeel -1,000,Po5t Dip
Delay-〇, 00 nus Sweeps-i, . # O-(blade Gap -1,00,00° Speed -3,00, Delay -0,00) z Levelνait -15,00 Most of the above fields have already been discussed or are self-explanatory and do not require explanation. belongs to. One exception is slice resolution, which indicates the unit of representation for component sli files. In the above example 8000
The slice unit is −1 inch. The Prepafe item on the menu bar at the top of the parts production preparation screen is selected to further prepare the part file for part production. 1) "(Pre
The pace menu (32nd page) is displayed and the functions that must be performed to prepare the part are restored. Each of these functions will be explained below. Make Ba5p: The Make Ra5e menu displays the dialog box shown in Figure @3'3. This item allows the user to create the basis for the parts to be manufactured. Foundations should not be confused with component supports. The base is a platform located at F of the main part of the part.
This is to keep the parts at a constant distance from each other, and to allow the parts to be easily removed from the stand after they have been manufactured. On the other hand, support is selectively increased in those parts of the parts that are twisted or distorted. For example, in the case of a coffee cup, a support may be necessary to secure the handle of the cup. Because it adheres to the body of the cup during manufacturing, °ζ゛14. This is because it will happen. The base will keep the body of the cup at a constant distance from the base. A dialog box displays the file number. Each -7 isle also has default parameters, which can be left as is or changed by 76-, sir,-. The linet style must be related to the base as it is related to the rest of the part. If you need a glue coat style other than the default style, you can load it from a file if it is available, or edit the current glue coat style if it is available in 1 or 9.・
The above discussion of editing files is applicable to loading and editing base+1 coat style files. Other fields that are visible and editable include the height of the base, the extent of overlap with the part, the extent of the base, the grid base for the base in the X and Y directions, the base layer Includes thickness, and trough settings. Edit any of these fields1
．． After that, to accept the displayed parameters, the user must press the Mak+≧Ba5e button.
Indicate 4 and click the ove Ra5c button. Build Pa "ms: 13ujld Pa
The es menu items allow the user to scale the part in any of the X, Y, or Z directions to offset shrinkage, adjust the overcure factor, specify precision and speed priorities, and copy the part to be fabricated. Displays a dialog box (Figure 34) that allows you to specify the number of parts and, in the case of a large number of parts, the minimum spacing between parts. This is accomplished by editing the field items. Most field items are self-explanatory and require no explanation. The overcuring factor field 126 contains all vector types in all style files with one Provides a method to globally scale up or down the overcure parameters specified in the Accuracy vs. Speed field 12
7 is 0.0-1. Make it possible to trade speed and accuracy on a scale of 0. Replacement is necessary if the positioning accuracy of the guide mirror or the speed of the mirror is reduced by 4'. Do Prepa'e: Do P'prepare starts the slicing process of the part file and subsequently creates a production file. If the part has not been saved to a file yet, you must save the part. A dialog box will appear asking you to specify the name of the part. The normal response in this case is to click the 5ave button. On the next screen, accept the abbreviated part name or type in a new part name. Select 1. If you want to overwrite the current directory with the specified name, choose whether to overwrite the old file or choose a different part name.If you accept the part name, , the dialog box shows the list of components to slice, low (lov),
Select medium (++sdium) or high (high) priority. For complex parts, this choice is made because slicing is a processor-intensive task that occupies the workstation for a considerable amount of time. By preferentially selecting low for slicing, the workstation can be used for other purposes while slicing. On the other hand, if high is preferentially selected for slicing, the time required to complete the slice will be reduced, but there will be almost no opportunity to use the work station for other purposes. Selecting one of these will start the slicing process. A window will automatically appear on the screen so that you can monitor the process. The Abof+ button at the bottom of the window can be used to avoid problems in the slicing process. As each component begins slicing, another message appears on the screen. Once all components have been sliced, you will receive a message that convergence has started. In the convergence stage, the bat manager merges all τ slice files with associated style information.1. , resolve conflicts between styles, and create a C fT-7 file for the data. Finally, the window (
Figure 1b) shows that the convergence is complete (2) and prompts the user to select a message file. Messages, files, and files are created for each component of the part. You can select a message file by clicking on its name. The message file displayed for the component file (Figure '36) shows the number of squares of each type loaded from the component file and displays a warning message indicating possible problems. Also shown. If there are more one-color messages than can be displayed on the screen, you can use the scroll bar to scroll through hidden messages. Slice quoted earlier

[7
U.S. Pat. During the fabrication process, the control computer is accessed. The format of the fabrication file can be summarized as follows: 1. Total 14 sections 2. Overall component section 4. Wide section 86. Layer section 0. ) Component [) Section block 1 Block 2 Block 2 Block P 2) Component ] Section block] B [Gonk 2 block q block ff'1 Component block] Block 2 1 Section b, layer section ] ]) Component O section block] End As shown in Figure 1), the file is composed of the overall cross section of the part as a whole, the overall cross section for each = 1 component, and the cross section of each layer. , and each component has a vector block. An example of the production file format is shown below. Example of a BFF file: BFF file: eyl, bfT Riyo Mikawa: -B F F 7 file: , /e
y1. b "rGIOBAL Cross-section Global Cross-section attribute, 84 CNAME = 'eyl' [3
] <str>Deaf85 CNUM -0<Int2> $813 Ca1nx -0,1)00000
<FIoat4>11.87 CMAXX -3,
240000<FIoat4>1E88 CMINY
-0,000000-: FIoat4>Deaf 89 C
maxy -3,240000<[Ioat4>$90
C') Lt~Z -1, 1) Ofl+0f) 0
-:FIoat4>$91 CMAXZ -3,8
20000cFloat4>$92 NCC0PY −
1<1nt2>$93 CROTO= 0.00000
0 <FIoat4> Answer 103 C'/desired a-1,,
0Ω0000 <FIoat4>$113 CX5C
AI, FO-1,000000 FIoat4>Answer 12
3 CYSCAJ, EO-1,000000 FIo
at4 > Lou133 CX0FFC) -0,000000
<FIoat4>1t143 CYOFO-0, (1
000f) (1<FIoat4>Layer 5ect
ion: [1,000000] Attributes
f'or Layer 5ection referee 153 L
THICK -0,020000<FIoat4> Arithmetic 1
54 ZWAIT -15, [100000 FIo
at4〉$155 ZDIP -0,750 Roro 0
FIoat, 4>ll158 ZVEL -1,
000000 <FIoat4> Umpire 157 ZACCEI
--1,000000<FIoat4>1151!
PDDEIJ −0,000000<Float, 4
>Answer! , 59 N5WEEPS -1<Int2>$1
60 8TVEI-1, -4,000000Floa
t,4>寥1675SPO81,-0,0000[10
<Floatr$1745EPO3I -0,000
000<FIoat4> Trial 181 BVEl,,,l
= 3.000000 <FIoat4>$1
88 8G 8 yen -100,000000<Ploat
4b$I TOMll,,,S -1000,000
000<White oat 4>$2 LINAME -
'in [2k Stv>Calculation 3 TITLE
-'CYL' [3] cst '> gain 4 CDAT
E-'Non Oct 2317:52:30198
9tf5 MINBHCD -0,020000F
Ioat4>$6 MAXBHCD - (1,026
000<FIoat4>son7)41N1℃D
-0, []20000 kuPiat, 4>tt8
MXFCD -0,026000 FIoat4
>$9 0CFAC-1,000000<FIoat4
＞偁12 X5CAIJ -1,000000＜FIo
at4>$13 YSCALE -1,000[1
00<FIoat4>114 ZSCAI, E = 1
．． 000000 <Ploat, 4> Minoh 15 5PEE
DF -0,500000 FIoat4>$16 N
COMPS-1(Int2〉$17 PMINX-0,
01)0000 <Floaz4>#, 18 PMA
XX -3,24QOOOOKUFIoat4>ttl,
9 PMINY -0,000000<FIoat
4> 20 P) tAXY -3,240000<FI
oat4>121 PMINZ = 1.0000
00 <FIoat4>$22 PKAXZ -3,8
20000<FIoat4>$23 NPCOPY -
1 (Int2>Storage 24 PRffrO-0,0000
00<FIoat4>$34 P73::All:0
= 1.000000 <Pioat4>1144
PXSCALEO-1,000000<FIoat4>
$54 PYSC^I, EO= 1.000000
<FIoat4>184 PXOFPO-[1,
000000KUFIoat4>寥74 PYOFPO
−[1,000000<FIoat4>[241<st
, r> (J)BAI, Component 5ecytio
n O[01Attributes for Gl
obal Component 5ection
t195 ['5DEIJl -0,00QOOO
FIoat4>Coaponent 5ection
: 0Attributes for Compone
nt 5ectiontt84 CNA) 4E -
'eyl' [3] Kustr>son85 CN1
1M-D <jnt2>$86 0MINX -0,0
00000<FIoat4>tf87 CMAXX
-3,240000 (FIoat4>$88 CMI
NY -0,000000<FIoat4>$89 0
MAXY -3,24000[] <Float4
＞$90 CMINZ -1,,oooooo
<FIoat4> Storage 91 Cl1tAXZ -3,82
0000<Float, 4>[2NCC0PY-1,<
Int2> $93 CROTO= 0.000000 F
Ioat4>$103 CZSCALEO-1,00
0000kuFloat4bltl13 CX5CALF
O-1,000000<FIoat4>$123 CY
SCALEO-1,000000cFloat4>$1
33 CX0FFO” (1,00(1(lOO<FI
oat4>ll143 CYOFFO-0,00000
0<FIoat4>1!202 NLI)4)1-3
<1nt2> Book 203 HANGl -80,00
0000<FIoat4>$20411ANG2-12
0.000000 <FIoat4>Deaf 205 HAN
G3-0.000000 <FIoat4> Lou 207
H8APCEI -0,050000 FIoat
4>Neck 208 H8PACE2 - 0.05000
0 <Ploa “4> Banquet 209 H8PACE3
- (1,05(1000ku FIoat4>Fog 21.
I H8TEP -0,000000<FIoat4>
1211 H3TEP = 0.000000
KU FIoat4> 212 BCD -0,026
000<FIoat4>$213 1(CD -0,0
26000kuFIoat4>$214 UPIICD
-0,026000<Float4>$215 DNH
C+)su], 02GOOO<white oat4>$218
To XFSP8CE -0,0D3flOO<Float
4>$219 YFSPAi King-0000flOfl
<Float4>$220 PCD -0,0230
00<Float4. '-$211 [JPFCD
-0,028000<Float4>It222 D
N[”CD −0,020000<Float, 4>
Block! -x, t, ion: Block Type: IJ (Bloek fn Vecytor Foniat, 1
20~In=tors], 504000. 0.5
111i1251.620000.0.5+, 0000
1.620000, 0.5100001.73600
fl, 0.5161.251.73BOOO, 0,51
B1251.850750.0.5342501.85
0750- 0.5342501.9133000.
0.5643751,. 983000. 0.5B4
3752.071500.0.6060002.0?1
500. 0.6060002.175[100,0
,6587502,175000,0,6587502
,272500,0,7220002,272500° 0.7220002.362750.0.7951
252.3G2750. 0.7951252.44
4875.0.8772502.444875□ 0
゜8772502.5+, 8000.0.967500
2.518000. 0.9675002.5B12
50.1.0850002,. 581250. 1.
0B50002.834000.1.1B8500Bl
ock 5ection Block Type: PI] 3 BIoc, k in Vcet, or Fofm
at, 120 Veetocs2.352750.
0.8062502.433750.0.88725
02433750゜2.505875 2.568: 柘0゜2, (i20375 2.661375゜2.691125 2.709000 2.71.5000. 2.709000 2.69], 1.25 0.8872502.505875 0. 9763752.568250 1.0?25flo 2.62[]3751.1.74
6252.661375 1.281G252691125 119 Niji 3752.709110111,. 5055
002.715000゜1.82000112.709
00[11,,7345002,691125 1,847[i252.66]:3750.971 Introductionγ
j] 1,. 1,281・↓-#143 consists of the entire surface of "1"/component [-]. This is the only component in the example. Field #], 53-1
96 consists of a layer cross section, and is the layer 8 in the example. The layer cross-section dictates the reneeting parameters of the layer phase. Fields #202-#222 indicate manufacturing parameters specific to the component. (Note that the overall component field #8.ll #143 is duplicated in the component section! This allows each component section to be run directly by the control computer independently of the rest of the fabrication file. ) are listed by vector or vector type after the component-specific parameters. The fields of the entire cross section are explained below. Field Description TOMILS Number of mils in each unit (see IJNA)tE below) UNAME Unit representing data TlTl Top Part title CDATTSAMI NBHCD MAXIIICll MINI"CD AXFCI) 01 'AC XSCAL, YSCALI' pumping : U-C PEEDF Production 1] A (j Minimum hatch hardening depth, full, Bohne; G, full thickness maximum hatch hardening method, full, ■:・Boh=Neret, full thickness minimum filling hardening depth, above , Component・, Minimum filling hardening depth for all layers, f:: Sibone: 71・, Full layer overcuring factor for the entire layer/no-after.Typical ones are shrinkage cancellation speed and viscosity exchange, typical It takes one officer to select the mirror speed that corresponds to one of the output settings, L-U, and one of the laser output settings. Part copy number PRω'O,
PZSCALEO, rotation, scale and PXSCAI positive 0 for each copy. PYSCl), offset PXOF
PO, PYOFFO The fields of each overall component cross section are explained below. CNAME Component name C~IJM
Number of components CM I NX,
CM, AX, X, Near 7 component 0)
Region CMINY, CMAXY CMINZ, CMAXZ NCCOP)' Component copy number (J?OTO, CZSC^I 0. Rotation, scale, and CX5CALF, 0.CYSCA for each copy
LIEO, Offset CX0FPO, CYOFPO A full field description of the layer cross section is given below. 1,, THIcK layer thickness ZWAIT
ZDIP waiting for level after cleaning
Brat Home Immersion Volume I7 Elevator Speed ZACCEI - Elevator Acceleration PDDEI Request Y Bost Immersion Delay NSV SE Recorder Number of Sweeps 5 TVEI, 1
Blade setup speed, 5SPoS1
Wait (see below) Sweep start position *PO3l 5EPO8I Sweep end r position BVEi
, 1 Blade speed in BGA
Blade Gap PSDI Mi1. ．． A Y l
Boss [...After sweep delay i, 27 of layer cross section
The component special field 5 field (other than the entire component field described above) will be explained. NIJMI (number of hatch lines HANGI, HANG2, NU) 41 (spacing between each hatch line specified by) tANG3 H8TEP Angle to slightly move the hatch vector to the nearby layer BCI) Boundary hardening depth tJP) ICD Upward hatch hardening depth DN) ICD Downward hatch hardening depth XI
"3PACE, YFSPACE X and Y fill spacing POD Fill hardening depth IJPI" CD Upward filling hardening depth DN
PCI) Downfill Cure Depth As previously mentioned, when combining individual sliced component files into a fabrication file, the part manager may sometimes have to adjust the coating parameters associated with different components going into the same layer. 1ii1 must be resolved. To do this, the part manager currently makes the following selections among all the components: Reneeding Parameters ZDIP VEL ZACCEI, PDDELAY ZWAIT NSI+1EEPS VEL cAP Selection Method from All Components Maximum Minimum Maximum Maximum Maximum Number of Sweep Components: In case of tie, take the smallest of these. For component ties with maximum number of sweeps, take the largest of these. PSDELAY Component with maximum number of sweeps, for ties, take the largest of these. For example, according to the table above, for a particular layer, the part manager will select the maximum value of the reneeting style ZDIP from all the components of that layer. As another example, the part manager has NSνE equal to the maximum value for all components of its layer.
It would set EPS and set BV^L for each sweep to the value specified for that component. The specific solutions described above for each parameter were designed to provide the safest possible solutions. For example, for the first five parameters for an elevator, it is safest to select the maximum duration and immersion distance, and the minimum force and velocity and acceleration. Regarding the following four parameters regarding the recorder, the safest method is to select parameters that are related to components that have several sweep numbers. It is always safe to tie, if any, to resolve conflicts for maximum delay and plate ganging and minimum velocity. Beamwidth Correction The Bart Manager also performs beamwidth correction of boundary vectors that end up in the production file. The purpose of beam width correction is to correct for the limited radius of the laser beam when the boundary of a given layer is a trace. The part manager first determines the beam width correction amount. Typically, the curing radius is determined either manually by the user or automatically in real time from a beam profiler. For each layer, the part manager first sets the flat tip and
Down layer and mostly flat. Collect all polygons of special boundary types such as up and dovn layers. For each vertex of the polygon,
The part manager calculates a distance d along the inner bisector of the first vertex to the center of a circle with a diameter r that is tangent to the two sides forming the first vertex, and points jT1 of the polygon. ] - Tsuru.Connect the first vertex of M and the moved vertex, length d
The hector of is called the hector of displacement. Move some or all of the vertices of the polygon (from C, the part manager has moved) n
Redetermine the boundary vectors associated with the points and deploy them to the production file. Explain the exception case. 1) If λ exceeds d or 5qrl (2) x r,
It did not decrease to 5qrt(2)Xr. 2) If the displacement vector is doubled and the edges overlap, or if it crosses the boundary of the first polygon, change the length of the displacement vector to a double vector or J-magic boundary line vector. Shorten it until it touches. 3) If two displacement vectors from different vertices intersect, shorten their lengths to the intersection. 4) If the displacement vector intersects the redrawn boundary of the correction polygon, place the redrawn boundary line just touching the entered displacement vector (without reducing the length of the corresponding displacement vector). It can be retracted until. The redrawn boundary line should be as close to the original boundary line as possible.
Make it. After adjusting the displacement vector, redraw the boundary line by connecting the moved vertices and send it to the production file. Further information on beam correction can be found in the above-cited US patent application Ser. No. S,N'331,664. The Conbig Menu The Conbig Menu Figure 37 provides a way to establish default conditions. Each field in the menu represents a default condition2, and the contents of the field can be set or changed by typing new contents. An explanation of each field is given below. orkDir: The work directory contains intermediate files as well as component and part files. This field contains the default work directory. BfT Dir: Production files created with Part Manager are written to this directory. This field specifies the default Mf directory. 5style Dir: This directory contains style files. This field has a default style, Directory. 5style: This field contains the default style that initializes components or soubrets when they are added to a tont. CAD Units: This field is default) The CAD units are in (inch), 1tll (1/1000th of an inch), ll1 (mm>,
ec ccm), m (meter) or ft
(feet). 5lice Axis: This field specifies the default slice axis. CAD Hancl This field is CAD
Specifies the system's default h direction as right hand or left hand. The Fi leman Menu: Fi I
The eMan (file management) menu (Figure 38) provides various 7-aisle management functions. I would like to explain 1 to the ancient tomb of Mego. 1) Cop>・Copy also has the ability to copy files. Clicking Copy brings up a dialog box that you can use to specify the appropriate directory and file. The top portion of the dialog box is used to select the directory containing the files to copy. In the center part 11 of the screen, files in the selected directory are stored. To select a file to copy, the user clicks on the file name. The name changes from black to red and appears in a box below the filename or list of filenames. Uther then clicks on the 5-select button to accept the selection and appears in the box near the edit section or section of the log box. The user uses this box to create a copy of the file. You can select one file or many files. Finally, the user clicks the Cop button 4 to copy the file. Rather than copying individual files, you can also select all files in a directory to be copied by clicking the ^11 button. 2) Rename Used to change the name of the Rpnamc file management utility file management file. A dialog box similar to the Copy dialog box is used to select a directory and file and type a new name for the file. Used to delete a large number of files.Cop) dialog
A box-like dialog box used to select directories and files to delete. , 4) Backup The Backup file management utility 1 is required to perform a backflash to the Phi/L save cartridge. (',,
Op y dialogue pock 41J class a)
Used to select files to trim. During the hack T knob process, the 6th or previous screen of the ``cutup'' file will be displayed. 5) Re5tore is used to restore files to a disk that was previously hacked up to a tape cartridge. Clicking on Re5tore displays a message reminding the user to insert the tape cartridge with the files or backups into the cartridge drive. Files are automatically restored to the first directory. 6) 5pace The 5pace file management utility displays the amount of unused disk space. 7) C1eanup The C1eanup file management utility deletes intermediate files created by Part Manager. Also deletes other named files. 8,) 1lnix UNIX gives access to the IRIs window. The View function, accessed from the main menu, allows you to view constituent elements and parts in the A position of the workpiece display, and view them from any angle without changing the magnification. Make it possible to do things. When working with views, it is useful to clearly distinguish between what you are viewing and what you are viewing. Hereinafter, the component or part to be viewed will be referred to as an object, and the device that performs pico will be referred to as a camera. The first step in using the view is to open a blank pico-window on the screen. Next, each appropriate component, slice or fabrication file (respectively, stl, , sl as file type)
i or bff), the object to be viewed is displayed on the screen. In this way, part manager 7ya (parts management blog 2) can display any of these 1-isle types. When an object is first displayed on the screen, The object is displayed as if viewed from the front, oriented as defined in the file.The object can be moved by changing the object's position, by changing the camera's position, or by changing the position of both the object and the camera. The spreadsheet can be viewed from different angles and from different distances. To access the views, move the cursor to the View section 1 in the main menu and click the mouse button. After a few seconds, the spreadsheet will be removed from the screen. The sheet disappears and is replaced by a blank view window with a horizontal menu at the top (Figure 39). The view menu is identified by reference numeral 128. The view window also displays a wireframe representation of the object. The functions of each item on the menu are described below. File - Includes the following functions: 1) New Creates a new view window that is superseded by the window that is already displayed. 2) Load: A drop-down menu to select the component or part to view.1. - is displayed. From this menu, component (,5tl) files, slice (,5li) files, and fabrication (,bff) files can be accessed. '3) Save Save the currently displayed file with its current file name. 4) Save As Saves the file currently in view after giving the user a chance to assign a file name. 5) Clear, clear the view screen. 6) Quit Pico - Turn off the screen L5 and display the previous spreadsheet again. View, including the following features: ]) Zoom, enlarge a part of the displayed image. 2) Chain 7 Pa Converts the current camera position. For more details, please refer to the explanation of Qi Dingji dialogue box below. Light. 3) Restore all conversions done for the reset 7 view to their initial state. . 4) Select, select perspective view, front view, back view, or left side view. 5) Load - Load a file containing predefined view conditions. 6) Save: Save the current view conditions as a file. 7) Copy: Copy the file containing the view conditions. Lights: Lights include functions that control the illumination with which objects in a view are displayed. Object: Contains functionality that affects the view or state of an object. These functions are as follows. ]) Select, select objects by name. 2) Draw Mode: Makes selections that determine how the component or part is drawn on the screen, i.e. the wire frame representation. '3) Orient Change the orientation of the part. For details, please refer to the explanation of ChiJno〉/Gekwa Box below. 4) Layers: Select a specific range of layers to view. For some details, please refer to the explanation of Chijinji dialogue hox below. Changing View Orientation In the View menu, you can change the camera position by selecting/functioning. The object menu allows you to change the position of objects using the orientation function. Both functions display the dialog box shown in Figure 40. The area at the top of the dialog box] 29 is used to transform (move) the camera position along the X, Y and Z axes. The middle section 130 rotates the camera about each axis. The lower area [3] sets the distance between the camera and the object. The temperature for transfer is represented by a scale with a pointer indicating the current status. Furthermore, the current r■:
For the status of , the numerical value C is also displayed in the 10 of the screen. The state in each column can be changed numerically or graphically. To change the state numerically, click the mouse button on the number that represents the current state of the movement you want to change. The white edit cursor or is displayed next to the value. Enter the new value manually [62, press the enter key. Then, that field will already assume the new value, and use the pointer to reduce the value graphically in the ○ field. The state can be changed graphically in a manner similar to that used to change the visible screen area using scroll bars - one method is 1. Move the mouse over the scale pointer and press the mouse button. Press, then drag,
Another way to move the pointer is to move the pointer to a new (7) position. moves towards squares 1 and 7 on the active side6. There is a red square at the left end of each square. By clicking the button, it will turn green. When the square turns green, either the button or the ink will move.
, the camera moves continuously during the ride. If it's square or red, move the pointer to a new position, release the mouse button, and then the camera will move.The green mote helps you see the result of moving the camera, and the red mote moves the camera rapidly from one position to another. In the previous section, we explained how to use the dialog box to control the camera position (5). A similar dialog box controls the object position. Contains the term ``Layers'' used to select the illumination/image that performs the same operation.Selecting this term 1) selects the layers used to control the position of the camera and objects. 3 insigata similar to Uichita, ri-2 dialogues are displayed (Fig. 41).
Previously - similarly controlled. Displayed as 5cale, box 1.32 is used to adjust the interlayer spacing during display. Term 1 labeled M i n and Max, respectively.
113'3 and 134 other 2C a), zip
In the evening, I did not select the layer I5/C to display. The description of the Bar, I Money 7 Ya is in ten parts. --- The tool provides a convenient way to organize and record the pigs needed to make parts, especially large or complex parts. Other advantages of the Tort Manager are that it facilitates its use, especially by novice users, and that it allows users to easily understand implementation details and production parameters. Many of these parameters can be specified in a library of style files and linet style files, and then the individual components and beginner names according to Sujishi Tsuton~1. Sequential relationship = 5 is entered by the user.
The distance is shown in Figure 42. -(-Tomahuja 1341
．． stl' file or CAD/CAN1 system 3'',
), the 3° part that starts the operation after being exposed to lj-me enters step 136'' and specifies the part to be worked on in step 1: S7. As indicated by -1138, the user is given the option of working on a new part or resuming work on the current F part. In step 13,
Specify the components related to the part, such as 1 thinning,
Each component has its own, s? , after being associated with one file, the Konera-no-tani file has the parts of its parts = 7
associated with isle. As indicated in item 148, the user may also choose to convert the component file or preslice the file, so that the Part Manager will completely cut the triangles or boundaries of the represented component. Inspections are conducted to confirm that the standards are met. Next, in step 140, the user specifies receipt information for each component. As indicated in item 141, this involves adding 1/digits to the component and specifying each 1.:S file, tile, and linet style file. Steps 1, 1'', and 2 allow the Fuser to view the part and selectively move or scale it in space, such as in item 143 The part to be manufactured is 7K, ijl"'(: cut. In step 144, the second part is the term e. In 145, the base (y) of the part '\ is added, and some input areas are added. A final step 140 before fabrication is performed, including the primary step 4, such as determining the fabrication parameters and forming the parts. , -1-the is the term [-1147 and small O L
It is also possible to perform certain tasks such as 7) managing 7) files. Finally, there are 7 aisles for parts production:
,) o -ru - ■゛, passed to P2-Y1. Each of the above 7. Hale's file format was discussed on page I7. CFAI Uno explanation made f'+ file?
= After being transferred in the evening, this low-speed model can start running (:FAB and 1, -ζ"Known parts fabrication")"Fut x -j". 11, Similarly, e F A s also has a knee and a menu that is snowy.
Start A'El:m command, CFAB is D S
In cooperation with i), start the group [] process.

['-7-note 125hi'' users and 2,) (7) shows the overall distribution of tasks among the DSPs. Execute 18 programs, BEAM, EXE
CFAB, l':XE and BETWEEN, EX
E, soh so' tz number 'i'], 51,
150 b Yohi 1.52 TeJJlSil,, (-
I'm coming. Both of these programs are 1/-]]-2
Same as 5.→S8, L1, 7, 7, 7, and A7.1 performs stage 1, which controls the operation of motor 22. Furthermore, the coat
L7 Le: ″ll;
and filling hector σ) also produced? lj'! 61-c
s;su, :: (:”+ 9.7..7 c number 15
It is identified by 7 and there is 1. , the Areo DSP 34 is responsible for the geometrical correction of vectors and the execution of the tasks of Section II-7P Supplement 11. I'm here. Nio 1. These tasks and 17 related tasks are:
Number 16', a to 16 to 3.167] 6 (identified as 9. → Novo DSP 5 is in charge of controlling the controller 20 and controlling and monitoring the position of the dynamic mirror. The control is by F(i-tal, analog controller...
(I) AC) 38i This is done by giving a positioning command to the minnow. The mirror position is monitored by Ding and Shifuda 3'7, and the mirror's current position 1,
- Continuously related 1 - Providing updated information 1
, “Performed by: Bon Trolco; B,,, and 2 DSP
A task that is executed by [1, is executed. 1 (5) This investigation is based on this M device configuration 1.Limited 4N or L. J is small, other device configurations are 3,
Or J! For example, if several tlfE rows are processed i'
I used ↓・guu [2 Luzorose・l+1 for iJ Noh. Narusubu I7 SE/S+J configuration is possible. It is also possible to consider a configuration using a single integrated circuit consisting of several parallel processing capable G8 circuits. moreover,
Nunguruburoyasoza+Xsei using multi-task function also
It is possible to include an arrangement that allows each scale task to be executed in a time-divided manner according to the gross set of the medium. Sub-) c, The above explanation is in no way limiting.7, Next 1. We will explain in detail the structure of the two processors and the tasks they will perform. Figure 43 shows the entire contents of the bff file 49.
This file is accepted by CFAB Zorograd and the data is retrieved. 41 number 1.56 T used to control each device is passed to BETWEEN, EXE program, and those production buff meters regarding the control of
After the identified boundary lines 71 are extracted, they are used to continuously crosshatch and fill the fill vectors J1, as described in step 15"l". Creation of Crosshatch and Filling Vectors The algorithm for calculating the losshatch and filling vectors is shown in chart 70 of FIG. As shown, in step 177, the ° algorithm first calculates the lines of intersection for the crosshatch or fill vectors. These straight lines describe the type and spacing of the filling vector.The straight lines are calculated from the information entered in the bff-fill.Then, the straight lines are (overlaid on top of the layer) and begins the process of generating crosshatch or fill vectors for that layer. The important requirements (for reasons discussed below) are straight lines or
It is parallel to the first power coordinate (usually the Y coordinate) of one of the coordinate spaces. The straight line is placed 1 parallel to the Y-axis. Figure 45 shows
86 is shown with a crosshatch superimposed at intervals on a particular skin boundary vector 187.The straight line for a continuous line is numbered 86. 1.86 al 86 b,
], 86 c, . The consecutive border vectors are numbered 187a187b,
187c, denounced by. The spacing and direction of the straight line for crosshatch/filling solid are 1,
Determined by data obtained from the bff file. Returning to FIG. 44, the intersection point of the straight line for filling vector and the boundary vector J is determined. To facilitate data storage, the corresponding intersections are sorted by lines for filling nodes (instead of all intersections for a given boundary vector are connected together). Example, B, 4th +) Figure - Go, the intersections 188a, 188b and 188C are ■! , Kaisen ＼ritor 1-87 a 1. : I can't relate. Next, the suite y 'f 1-2"0 in Figure 144 is
Individual glue 111111s・\・, f, exemption ha, tato!
[Usage] 1 difficult "J's all intersections 1, K pot 31 knee z,) are classified B-1, sea urchin, (point or reclassified. Konepa, 2 ton is ・cho's After that, it is ordered again along the direction of the straight line for filling.Example 2
- For example, in Figure 45, the intersections 188b and 188
d are related together -) after being kicked, point ] 88d or ahead, ],
88b (assuming that the direction of the straight line 186b for the filling vector is directed in one direction of the vane). Next, No. In the steer-i 180 of Figure 4.4, a quantitative analysis is performed on the boundary line /\, kutle. In the quantitative analysis, each boundary line,
The direction of the line is compared with the direction of the line for filling the filler line, and it is determined whether the boundary vector increases or decreases. This hankan Y is riros・＼2・masu、・′filling・＼ril
...The coordinate perpendicular to the direction of the straight line, C, ζ- in Figure 45, is based on the X coordinate [, and E15. To the boundary line, off the edge of Kutle - a beautiful 丙 (vote (X), or 2+e is the first, compared with the X coordinate of the intersection. The boundary line hek(・le is considered as increasing 4-ru(,), and if it is small 1j, the decreasing one is ξt. Based on this comparison, each boundary line ・\2 l
・A fixed number is assigned to the key. In making this comparison, we assume that all boundary lines follow the "left-hand modulus fi11"), and accordingly, the direction of successive boundary vectors is layered in the clockwise direction. Surrounding the interior of is preserved. For example, in FIG. 45, the continuous boundary line ＼RI1~RU187a. ], 87b, ], 87c, , surrounds the interior 189 of the layer in the clockwise direction. If the boundary vector increases [,C, then the vector is
If 2 is assigned and decreased, then the number 2 cannot be assigned. Therefore, in FIG. 45, the boundary lines 87d and 87C are assigned a quantitative number of +2, and vectors 187a and 1.87b are assigned a quantitative number of -2. The end point of each increasing vector is assigned the number 71, and the ending point of each decreasing vector -1 is assigned. sTi - 1 is assigned to the end point of each decreasing vector. Parallel boundary vectors and their end points are assigned the number 0. In FIG. 44, each cross hatch 7.゛On the straight line for filling hectares -), the existing irises of the quantitative number of intersections Jl
It is very well calculated. First, the current tone is [
−). , After each point, the valley intersection is added to the current sum = -1 for each point in order. [Suitetsubu] 83, current situation 1) Is it 1 or 0; >r-'
If the number changes to L, crosshatch 2' fills 21/sword 5! ! The current intersection point is the starting point of the vector and 1. ζ - Directed. In step 182, the current n
If it changes from the number of stops to 0, then ri 0 shat 7, ・'Il composition of the filled bettor is the end E' L,, the intersection of the current 6 (with the end point of the obetator (indicated by , Z).The current total? If you don't have a cello and correct J-natte, continue it 14'l, #-T
'il: Started the 1st v or 1j su soji 'filling/\(7 tor generation or continued, current situation, -1゛?(
・1 completed, I, na-ζ-1 If you continue '6, the JK formation of cross--Sochi 7-'filling \ri]-le will be continued in the same way. For reasons that will become clear later, it is currently difficult to read the lilies in their entirety. Straight line 186 b 1- Q intersection 188d
Then, →2 or the current sum = 1 is incremented by 4. At this point,
The current n851 changes from 0 to →', 2, and the coordinates of the intersection point are designated as the starting point of the cross \titch/' filling hector. Next, at the intersection 188b, C-? or currently F−,2,,,j
-1, added, 171. yo〕“-This is what it looks like LJ
+: From a (from -1. Below -1 and before 11 1.9
44 Returning to -), in step 184, the remainder of C/, ), f, -r points are deleted, and all the points for each Cuegos Hazza 7' filling hector H-1 line are calculated.
j sl-soza ′filling・\rittle(,ha,zo(〕゛)4
In order of F, ＼＼ﾂﾂﾂ、＼＼ﾂﾂﾂ、＼ﾂﾂﾂﾂﾂﾂ、＼ﾂﾂﾂ、
ZN, those hectors are that 1& , ,';1-j
At 118''), once is the area above the liquid surface where the hardening is done. :: Seki 1/Goshi Zabimu's・1・；zo to a minimum of 18
So 1 order is 0re・ζ). This is related to the more powerful Envoy of Reza III.
In the order of 1st floor, in addition to the seven of each, Ku / Day, 1, Ko, ＼, Cuttle, in addition to the list of drawing, drawing, it is a continuous batch) and continues. -)C
゛Skilled in action. After a batch of tbego or 4 thoughts, tl.
Once considered in reverse order, a list of 6° hectares is created in the order in which they are removed from the batch, in the order in which they are drawn. The vector 7) 25° direction is adjusted so that the vector points in the alternating direction. The ordering of pect/l will be explained in more detail in Section 9, ↓6]. Figure 416 1, - is number 190a-1, .
90 k identified a)ri1! The input Hatsuji vector is shown. These vectors are 1- of the boundary y12 line vector of the layer 1 constituting the hollow solid upper frame, and Q 1 a -1
93, G or more self-cum, created, - mother) ('
,',7)-This is it. The frame of the solid part has a 1:1 diagonal shadow.
2 reference number 1 (at 92, J(t, +, "January's part is 2! M number 1q 3-C+,, i, Sane-Koiru,
. 4-4 If we follow the 7-, 'l algorithm, in Sutenobu 184, ``, these lines 1st \ Sochihector are each 2 [-1be / \tsu square vector, line 5 and -L sword like 1゛) -) Kicked '\,, put into chi. !\・!もstraight line ・1 sof) J direction I Q 1 a I Q Oa
l-191, b 1. , 90b
-, 111-11c 1≦) Oc
, 1 ('j Ok , 111, 91 d
1.90d, 1. , 90 j ,
,, tl 91 e 'l 90 e , 1
90 i,,,,I 91f 1
() (1) f, IQOh Jl
(Q 1 g 1.9 0 g
1 At the beginning, 11.
The direction of the straight line for the hector is 1. Facing Z)
L. zu,'5''-;rg' 185 i,,:oi'r,
9 million of these, 71 to J1 Tuna, minimize the distance (,'2 to 5,), low.
The 4, this, last, first, first, first and second letters arranged in the order of ``-'' are taken out in order from consecutive hatches and added to the list. First, consider 7<Sochi] 91a to 191g in order of 1, then Hatsuchi or the remaining 1. As such, those batches are considered old, but now in reverse order. After that, there are no more vectors left in the hatch. Then, the consecutive vectors in the list are considered and each force direction is adjusted so that they point in alternating directions. Once this is complete, the ordered list of vectors in Figure 46 will look like this: vector direction 190a on 190b). 190e,,,h ko 9 (] d
90f
Lower 190 g,,,J-1gO1'
l lower 19 [1 i upper 190 j 1” stiffness OK
,,IX As a result of the above, the laser beam does not need to cross the hollow portion 183 at all. An important aspect of the present invention is the ability of the algorithm of FIG. 44 to detect at tS when a cross/filling bed straight line enters or exits the solid portion of the bed. The algorithm does this using a quantitative number, 6H, that is related to a straight line. Current total or cell
: If it is not thicker than I, the straight line is considered to be in the solid part and not out. If the current sum is zero, the straight line is considered not to have entered the solid part, or to have entered and then exited. This algorithm works at point 188at or 18 in Figure 45.
A case where a crosshatch curve intersects a vertex where two boundary line vectors intersect, such as 8e, can be easily handled. Point 1
88 a, the intersection of the curve 186a and this point is already at U) in the solid part of the layer, so there is no need to generate the cross l\-notch or filling vector. 1. Don't go there again. This is the intersection of a point and a curve, or a boundary line..., vectors 187a and 187
d and the straight line 41- are treated as simultaneous exchange cooling!
j! It is. 4 ゛Ware, \ vector 18 n'cN stuck and 1-1 or 1 river is contributed (Ll[ci1, -1 or is added for vector 18/a. Follow), and 1 in the current conjunction is () remains unchanged. Point 188 e -) and (4, the straight line that intersects this point is about vector]87d)1.
Since the solid state starts at this point, if appropriate, 1, cross, 1〉, chi/filling vector will be generated. Then, when the vector ], 87 b 1-(7) 11 row line appears, -2 is subtracted from the current sum J1, thereby setting it to 0. 'The generation of the filled betatle is completed. This algorithm also uses crosshatch, line for filling vector:, intersects boundary line hector parallel to 11 and φ“
We will also handle these cases appropriately. This situation is shown in Figure 47. In this diagram, ゛ is a cross! 1/' filling hex[*rut (]straight line 194a!'; and 194b, respectively, *10 line beta*ru 1.96a, intersects with 196b). When the straight line 104a intersects the point 195a, it intersects with the end point of the vector ]97a; From this point, "7"
Formation of the first 7' fill hectares begins. This vector is that cross! ＼lも、・If the straight line for filling hectares is flat, addition will not be made if it intersects with 196a. Also, points 195 b-c are hector 1
At the end of 97b, +1 or 1 is added to the current total by the intersection of l-, and the generation of crosshatch 7·' filling/\ctor C continues. Then, when the straight line for the ri[7shunch/'filling vector exits the isobody part at point 195d, 2 is added to the current sum, and the lt, formation of the ri[7shunch/'filling vector, if appropriate, ends r. . Regarding the straight line '1.94 b, ζ- is C at point 195C.
, the intersection with the end point of the vector 197b adds 1 to the current sum, and the vector 196b is parallel to the crosshatch straight line C, which affects the current rI-, the sum 1. do not have. Then, the generation of C1 to L7 hatch/' fill hectares is started. Point 195eTi! , the intersection of vector 197c with the end point adds -1 to the current r1 sum, so that the current sum is set to zero. Therefore, at this point, the generation of crosshatching/filling vectors is terminated, if appropriate. An important aspect of the present invention is that components such as tea bowls and tea bowl handles in which certain areas overlap each other, as illustrated in FIG. It is the ability of an algorithm to process C. As illustrated, the handle 99 intersects the side surface 98 of the main body of the bowl. The intersection regions are identified by numbers 200a and 200b. The problem that can occur with overlapping components can be illustrated by FIG. 49, which is a plan view of FIG. 48. The straight line 201 for crossmetal/filling pectel is 2
02a, 202b, 202c and 202d intersect the component layers. The problem begins at point 202C, where the crosshatch/fill vector straight line exits the region of overlap 200a of the two components. At this point, the crosshatch/filling vector generation algorithm continues generating until point 2 ('12d or 0) and stops generating the crosshatch/filling vectors even though it should have stopped. The algorithm of the present invention easily handles this problem because the generation of the crosshatch/filling vector is
If it starts with 02a, +2 will be added to the current tone at this time, 15 points 202b, and when it continues, +2 will be added to the current &=
This is because -2 is added to the current n-total at point 202c and the 1st C is still continued. Since the sum is now positive at this point, the generation of the crosshatch 7/fill vector continues 1'j. The generation of the crosshatch/filling vector ends when -2 is added at point 202d and the current sum is reset to 1 or 0. Returning to FIG. 43, after continuously generating the CFAB or crosshatch and fill vectors, the five-year-old vectors are stored together with the boundary vectors in step 58). I can do it. , later, the CFAB receives these vectors and splits them into move commands and commands. The jump command moves in a straight line to the desired position with the laser and during this period it irradiates and cures the resin. In the jump command, the individual path the laser follows is not important.
3 is also extracted. Curing depth is a component, composition. #Can be changed depending on the internal range and the type of vector within the range. The desired hardening depth is BEAM,
It is coupled with the laser power 154 and used to calculate the scan rate 55 as determined by the EXE program. Cure depth is also used by the CFAB to control the attenuator. The attenuator is normally in the off state, or is attenuated due to a small hardening depth or ζ.
When curing to 1, it turns on and cuts off part of the laser beam (-1 to ensure an appropriate hardening depth.) 1. Responsible for processing the executed move commands.As previously mentioned, the move:'J commands and jump commands have the desired X, Y position of the laser as arguments. Move command processing is step 1
Starts in 62. The purpose of this snap is to change the parameters of the bucommanito, i.e., the desired X, Y position of the laser beam, from the middle of the resin tank to
The plate is to be converted to medium. By converting to units like this, the calibration table indexing that must be performed subsequently in the geometric correction step 164 is made faster. It is determined by the position of the pores.The pores of the calibration plate are arranged in a grid pattern in the Xh direction and the Y direction, and the continuous pores in each axis direction are /'2 inch. In the future, it is envisioned that the pore spacing will be in increments of 74 inches. At 1/4 inch spacing, the holes will be spaced in increments of 83 x 8'3.
The result is a lattice with micropores of 1. ...with a spacing of 2 inches, a grid of 41 x 41 micropores. For a grid with 1/'2 inch spacing, the microholes now start at the left corner coordinate (128° 128) and start at the right -1 corner coordinate (168°
]68), in positive X-Y space, and τ
I don't know if it's a numerical value or not. To perform the transformation of this new coordinate system, a known matrix C is used as the D'', ? matrix. The D2 matrix is a 2×'B matrix that transforms the coordinates by matrix multiplication. As is known, the coordinates (U V) of another coordinate system (
Any conversion to x, y) coordinates can be performed by matrix multiplication as follows. /' Can be used to convert the resin tank plate coordinate system when using 24 inch spacing. To perform these transformations, in step 162, the XY position of the target relative to the knob command is transformed by the D2 matrix. In step 163, the vector argument of Moosony f-mant is the mirror's! 4' - Divided into microvectors in preparation for the geometric capture 8 of step 164, necessitated by the radial movement. In step 164, the distortions caused by the microvectors are individually corrected and t' is performed. Said U.S. Special Application No. 331,64
4 and No. 268837),
The red path of the laser is made as straight as possible, so it is a straight move, so before executing [CIII scientific supplement 11], first set each . Is there a need to divide it into parts? If the vector is not divided into microphones, the movement of the laser beam will be non-linear, and the distance will be very small, especially in the resin tank. Micro solid!・Le has something to predict μ, lj °C. The length of the microvector can be any number equal to the distance between the calibration plates. "The length of the gobetator is 1./4 inch or preferably 1.7".]7
If it is between 22 inches, the length of the microvector is 1/
I prefer 2 inches. The servo control loop time, ie, between successive DAC outputs, is called a tick. In each cycle,
1-D37 First, update the current mirror position using servo Ds.
p+: 5, later in the cycle, the servo DSP usually gives the DAC a mirror control near-83, and currently one tick is defined as 35 microphones [3 seconds]. So, the number of decks in Tani Microno\ and Kutle is CF
By AB! It is the number of ticks required to traverse the length of the τ microvector in the no-name speed epidemic 155. A stopper plate is used to perform geological correction of microvectors. For each aperture in the bullet, the position of the centroid of the beam in each aperture is detected by a sensor positioned in the aperture and is stored in the naso daple along with the corresponding radius of the aperture. Ru. The reference table is then used to determine the 1' radial position of the mirror needed to apply the L nosa beam to a particular device on the resin surface. For the beam position, bilinear interpolation or the like can be used individually to determine the radial position with an appropriate response based on consideration of the immediately surrounding micropores. , the first one is calculated periodically every few months or so,
There is no need to calculate = every time SLA is activated. The angular position of the mirror is preferably ('J1, expressed as a number in the range 64 K x 64 K, with the origin at midpoint and 17. In step 165, the microvectors are corrected for dynamic mirror drift, as described above. As described in U.S. Patent Application No. 268,907, the drift compensator 11'' is intended to correct the drift of the dynamic mirror over a period of time. It is located on both sides of the resin tank. The dynamic mirror is positioned until the laser beam is sensed by each profiler at an angle of 7> when the geometric complement (8) look-up table is calculated. Once sensed, the corresponding position of the dynamic mirror for each sensor is saved along with a reference table.Then, in drift correction, the positioning of the dynamic mirror for each profiler is determined again; The two tl are compared with the positions stored in the table for the X direction and Yh direction.
1 is used to calculate 3f of the "gain" and "autoset" in the direction.The gain is the product of the distance between the main end of the brush 7 filer and the distance at that time in the Xjj direction and Yh direction. The offset is equal to the average of the two profilers of the difference between the original distance of the profiler and the current distance from the center of the resin tank in the X 'llj direction and the X direction. ,ha,XI
After first multiplying the seating post and )' coordinates by X cane and Y 'y'in, respectively, then
This is done by adding each set. Details regarding tree correction are described in the aforementioned US patent application Ser. No. 268,907. Processing of the jump command begins at step 167. In step 167, as shown in the figure, the argument of the jump command is converted into C using the I)2 matrix in the same manner as described above. In step 168, geometrical calculations 1)-' are performed in the same manner as described above. The first thing to do is 5 degrees Celsius, not with respect to microhectors, but with respect to the end point of the vector. Move-Unlike 7 Mayudo, jump command-
C is 1 Nosa's path is not baking soda, but 1 or 1 is cut off by L/Sabi-Noa-Yatta, or it moves at a fast enough speed to not cause hardening. Since there is no need to divide the vector into microvectors (9), step 169 the number of ticks awarded to perform the gloss is calculated. 19. The number of ticks required may exceed the maximum overflow value of 1" determined by the field size limitations, especially for jump commands at slow mirror speeds;
In some cases, it may be necessary to split the in this case,
Since the number of ticks required may exceed the range of the tick counter, the problem is solved by splitting the command into single commands with smaller tick values. In step 170, a de-kuva now is generated from the jump:7 command and the move command. These data buffs are items passed to the servo DSP for controlling the movement of the dynamic mean. 71-This will either open the shutter that was closed (j, 1″) with the move command, or close the shutter that was open with the nyabu command. Because it was corrected, the +f three formulas of the datapano 7t are as shown in Figure 50. 203-208
Or I? Consists of N. -71-root 2 n-) and 2
04 defines the mirror's X and Y assumption start ( ) effect ρ. -Field 205 defines the number of f-skus required for one nori's web or movement, and field 200 defines the command number.Fields 207 and 2 o
8 defines the desired incremental movement of each laser in the There is.
2 C is recognized. 840 V E JUMI) MOVE&5HCI, 03E D E 1. ,,,, AY to 17', to and to Yashibukoma, ・1, ni-゛) should be explained to C], ta 1, hio~...' F
i & S 14 CLOS)・”=7 The command is the cloak for closing the shutter after moving the laser. The DELAY command is, as will become clear below, "Filter" =] Brassed because it requires 5 cloaks. The data packets for the move and jump commands are then placed in the queue 172. Next, the frame sequence is analyzed to take into account the finite time it takes for the shutter to respond.
Any necessary command is placed in the shutter command. Currently, when closing the shutter, :T
It takes about 600 microseconds from when the cloak is released until it actually closes, and when it is opened, it takes about 900 microseconds from when the cloak is released until it actually opens. Therefore, the move and jump commands must be arranged so that the shutter opens and closes at the appropriate times. Sometimes you may need to split a move command or a jump command.9. Example 7: / To open a jump command after a jump, split the jump command into two and open a jump command. You may need to insert a near 7 command. Similarly, you may need to split the move command in order to close the shutter after the move command is executed. Issuing a shutter command to the shutter in advance of the command's time to act in order to offset the time it takes is a concept known as feedforward. etc.4
Can be leveraged with any of the other Sundem components, including: The following example shows a case where a jump command of 52 ticks is divided into =] cloaks of 27 ticks and 25 ticks. 2
The open shirt command cannot be placed before the 5 tick command. Second jump: 7 mands (25 ticks, about 90
The time required to realize jT (0 μsec) is 1 minute, which is the time it takes to open the shutter. JP:64 d2.・'2951.7 #-27d
-(',,), 024, -5. OOC)
E-64,8'2816,'7 SP
: 25 (3, Oir+ ,,,,,/s
e eCOMPllAND: 0PEN 5
HUTT'ER (SPLIT) JP: 64.2 i' 295 ], , 7
# = 2')d” 0. Cal 24
/ -5,000E -64,8,・'2816,7
SP: 250. Oin/5eeF each J
The P command follows the format 71 shown in FIG. The first two numbers are the X and Y starting coordinates, the next number is the tick number, and the next two numbers are the desired X and Y incremental movement. The other numbers are the first two numbers being the desired X and Y end positions and the last number being the desired scan speed. Similarly, the example below splits the move command so that the shutter has 17' ticks (approximately 600 microphones [11 seconds)] of time to close. MV: 1908.6/2698.1 $=197d to 3.689.10.010 E-2635,4y・' 27 (10, OS P84
．． 5 In/sec COMMAND: CLO5E 5HUTTE
R(SPLIT”) MV 2635.4/'27f-)0.On
-17d=3. 68910. 010 E-26
98,1/”27Q0.2 SP:184.
5 i ny' e C Each move piece 7.1 is the jump mentioned above:
It follows a similar format to 7 Manto. The corresponding data valid formats for 5HUTTER or DEL, AY7 personnel are shown in detail in Figure 51. As shown, the format is similar to Figure 5 except that field 209 is set to 0 and field 216 is always set to 1000. This causes the servo The DSP can easily distinguish between jump or move commands and other commands. Finally, field 21] is used to indicate the type of command and may include a shirt open, shirt closed or delay command. When Shack commands and Day 1/E 17] commands are inserted into the stereo queue, these commands are then transferred to servo queue 1 by DMA transfer 114.
73. The commands in the servo queue 11 are later parsed by the servo DSP in step 7"6, and the shutter command is sent to the shutter 20, and the mirror command (move command or jump command) is sent to the DAC 38. (Step 17)
In 6, the servo DSP also uses the input from encoder 37 to determine the mirror position of the current 6, and 4 for use in formulating the two T mantles to the dynamic mirror. , Ill have to. This will be explained in detail below. ) If the servo queue is empty for a long time, servo D
The SP may run out of commands. In this case, send 1 piece of the beam to the predetermined body + 1 position outside the resin tank. By doing this, you can open the shutter to prevent the laser from burning through the shutter, which can occur if you use a powerful laser and leave the shutter open and closed for a long time. The delay command is used to keep the servo queue full so that the L-Hoop does not have to move to the body position. The position delay is used to indicate the number of ticks. The servo DSP operates in 35-microphone [7 second cycles. During each cycle, the servo DSP first transmits the signal from Korg to the mirror. It reads the current position and outputs an amplified error value to the DAC based in part on the difference between the desired mirror position and the current mirror position. ``Discrete Time Control System'' (``DISC
R,ETETIME CON"1'ROL SY'
STEMS-Katsuhiko Ogata. Prentice Hall, Inc. This is a proportional-integral-derivative (PID) term that is calculated according to the well-known principle described in (1987) and others. 1. :
- Korg gives a reading of the current mirror position of approximately 0.95.
It is given with a resolution of ] arcseconds, corresponding to 4E. Step 176 ``''C is executed for data packets. Each count requests the number of cycles to be executed as specified in the ``Currant'' field.The value is initialized to 0. In each cycle, errors related to position, velocity, and integral are determined.Furthermore, in the last six cycles, the acceleration term is determined by 1+.After being multiplied by the corresponding gain, they are combined. i31 Form the D error term. The position error of one cycle is the difference between the desired position and the current position. The current position is equal to the encoder reading. The desired position is Value (item 207 in Figure 50)
and 208) to the starting X value and starting Y value (items 203 and 204 in the figure). Speed error is the difference between the desired speed and the current speed. X
The desired velocity in the and Y direction is determined by dividing X and Y by the number of ticks, respectively. The current speed is
It is equal to the value obtained by subtracting the [Shishiboda] reading in the 11th cycle from the En:7-da reading in the current cycle, divided by 35 microseconds (1 tick). Each acceleration error term is calculated by 1 at branch 6 of the command and requests prediction for the next data packet. This term is calculated by taking X and Y from the next bet and subtracting X and Y from the current bet (dividing by 7,6 nuts. As shown in Figure 1152, the angle is In the close case, the acceleration error term helps to ensure that the laser path follows the angle as closely as possible, with only a small fraction of the difference.The integral term also The integral term is used to help control overshoot and to provide mirror stability. It is necessary in the equation. Addition of integral term 14, the beam beats on the trajectory AC
Change the output value by 1. If this is insufficient,
A shutter may be used to shut off the laser for a period of overshoot. The integral term is equal to the net sum of all position errors to date. At each cycle, once the error terms and integral terms for position, velocity, and acceleration are calculated, they are each multiplied by the appropriate gain term before being sent to the DAC. 52nd
As shown in the figure, it is desirable that the actual path of the laser be as close to a right angle 212 as possible. The gains are selected to do this. It should be noted that looking ahead to the next data packet in order to predict the angle is another example of the feedforward concept described above. In this particular situation, the use of feed-forward techniques helps anticipate sudden changes in mirror position and ensures that the current command is followed by the mirror as closely as possible to the desired path. be sent to the other party. Returning to the stereo DSP, it should be noted that the vectors used in the jump command are not divided into microvectors. As a result, the possible error margin for the Nio et al. vector is about 12-14 mils. Therefore, Souvret, tDsP usually compensates for any rounding errors.
Generate a packet at the end of each jump command to do this. Dialogue tool F with users of CF A and B describes how to talk with users of CF AB. The control computer's display displays the main screen when the system is first started (Figure 53). This screen consists of the following four areas. - Screen-F section 215 displays the program and its hardware number.・The lower part of the screen 2]-6 displays the function of the program that is started by pressing the function key on the computer keyboard.・The large area 217 on the left center of the screen is used for the current n rib directory! 111J Capable Production - Display a list of files. By using the PgDn key and the I) g[+p key, you can scroll the C1 list to see file names outside the display area. The current state of n and the temperature of the resin tank are displayed.Furthermore, the current 1-34-) and time are displayed in the y section of the picture. When the F1 key is pressed, a help screen is displayed superimposed on the main screen (Figure 54). As mentioned above, the main function of CFAB is to
, bff) to control the production of parts specified by the data in the file. Other functions performed by the CFAB include:・Laser on/off control and scale 1, -ring of laser on/off time for combination control ・SLA hardware initialization ・Adjustment of liquid resin level in resin tank ・Liquid resin in resin tank When the main menu is displayed, use the cursor control keys on the combi-coater keyboard to display the information in the center of the C1 screen. Choose one of the part names -----)i and tell your fortune. The selected file name is highlighted and displayed in the center of the screen along with the "7 isle size" and the date and time of last edit. In this example (Figure 53),
The files 5TD2X and BFF were selected. F4
You can access another filename by pressing the key (7) and then typing the bus name of the required directory. Pressing the enter key will display a pop-up menu 219 of production options. (Figure 5055). The name of the part currently selected for fabrication is displayed at the top of this menu. You can use the cursor control keys to highlight and select any of the menu items. The F4 and F5 keys It must be noted that the functionality of ~ is different from the previous functionality in Figure 53. If the Crafting Options pop-up menu is displayed,
The F4 key 220 is used to save the current build f'1 options, and the F5 key 221 is used to restore previous build options. Unless otherwise specified, all layers of the part are created during fabrication. However, it is possible to select the layers of the part that start and stop the fabrication process. In this pop-up menu, select the “Start Layer” row 2.
If you highlight 22 and press enter to select it (Figure 56), another pop-up menu will appear. The current starting layer value will be displayed, so you can enter a new value and press the Enter key as many times as you like. When the enter key is pressed, the new starting layer value indicated at 22' will be displayed in the build options menu (Figure 57). The value of the end layer specified in step 224 can be changed and displayed in the production option 9 menu in the same way (fifth
Figure 8). The menu items 225, ``Brezny Graph'' and ``Zoom,'' shown in FIG. 59, determine the display shown on the display during part fabrication. ``When the grab 226 is turned on, the display device sequentially displays a simulation of the entire back surface of the resin sash while the part is being manufactured.Furthermore, when the ``zoom'' 227 is turned on, the display device sequentially displays a simulation of the entire back surface of the resin sash while the part is being manufactured. Accordingly, the display screen is enlarged to show the part being built in as much detail as possible. When “Preview=” 228 is turned on, the display displays a simulation of part fabrication without actually fabricating it. are displayed in sequence. By pressing the F2 key, you can access the user utility. By pressing the F2 key, the utility list 229 shown in FIG. 60 is displayed. When the laser utility main screen is displayed, , the status of the laser is displayed on the right side of the screen.In Fig. 60, it is divided by number 232.As mentioned above, this status can be detected by pressing the F2 key.
Changes can be made by displaying ``L-The Utility'' and highlighting the item ``L-The Utility'' and pressing the Enter key. After that, the laser control function 2
-31 is displayed. The F key can be used to view a summary of laser control functions. This description is shown in FIG. 61 and designated by the number 230. If the laser is off, it can be turned on by selecting the ``Laser On'' line (number 233 in Figure 62) in the menu and pressing the Enter key. On the right side of the display screen, a power-up countdown is displayed. To indicate that the countdown is in progress, the lower right side of the display screen will display the remaining time until the 1st nose is turned on.When the countdown is complete, the right side of the display screen will show 1st time. The display will also change to indicate that the switch is "ready for use". The display will also show the current power output from the switch. You can turn it off by selecting a row and pressing Enter.The right side of the display screen will now show the current countdown to shutoff and the elapsed time until turnoff occurs. Change A. You can interrupt the laser shutdown by selecting “Interrupted Shutdown” and pressing the Enter key. In addition to turning the laser on/off directly, you can also turn the laser on/off. You can also set the scheduled time every time.This can be done by selecting “Laser Schedule” from the menu.
, done by each press of the enter key. The pop-up screen 234 is currently scheduled...
The off time is displayed (Figure 63). The time on this screen is
You can change them by simply adding ``r'' to them. After pressing the keys for the required time, the on time can be set by pressing the F2 key, and the off time can be set by pressing the F4 key. Pressing any of these keys will give you the opportunity to overwrite or not overwrite the current time. The on time can be determined by pressing the F3 key, and the off time can be deleted by pressing the F5 key. While the temperature setting main screen is displayed, you can access chamber temperature control by pressing the F3 key or selecting the ``Thermo Settings'' function in Figure 60 (Figure 64). Changes can be made by pressing the Enter key over the current thermostat setting.Maintenance FunctionsIn addition to the user functions described above, the CFAB includes diagnostic and calibration routines used by maintenance technicians. To access these routines, press the F5 key and then enter the required password.Making the PartThe part manufacturing process can be accessed by pressing the F9 function key, then selecting the "Reit Options Menu" line in the Manufacturing Options Menu. Start by selecting (Figure 56). While the part is being manufactured, the display screen changes to reduce the manufacturing status (Figure 65). This completes the explanation of the system. Therefore, as is clear from the foregoing description, a method and apparatus for stereolithography to increase part production speed and remove fabrication obstacles without sacrificing accuracy, especially for dolls or complex parts. Certain improvements have been explained. The first improvement is the high-speed mirror and coordination 1. . In this case, the output of the laser can be selectively cut off by a high-speed shutter, an attenuator, etc. to prevent erroneous curing of the resin. Mirror accuracy, even at high speeds, is maintained by a servo control system with gain parameters that represent an acceptable trade-off between accuracy, speed, and stability. The second improvement is the use of a SBR-style conceptual model to facilitate data management and subsequent part fabrication. A third improvement is the use of a dedicated block setter for real-time device management tasks such as dynamic mirror control. A fourth improvement is to generate crosshatch and fill vectors without pause, thereby eliminating the need for large or expensive storage devices, especially for the production of dolls or complex parts. Although the present invention has been described with reference to embodiments and applications, it will be apparent to those skilled in the art that many other modifications may be made without departing from the inventive concept of the present application. , the invention is not to be limited in any way except in the spirit of the appended claims.

[Brief explanation of drawings]

Figure 1 is a perspective view of the stereolithography system, Figure 2 is a perspective view of the process chamber of the stereolithography system, Figures 3 and 4 are perspective views of an embodiment of the drawing subsystem of the stereolithography system, and Figure 5. Figure 6 is a diagram showing the architecture of all processors in the stereolithography system; Figure 6 is a diagram showing the equipment configuration of the call computer in the stereolithography system; Figure 7 is a diagram showing the sequence in which files are generated in the stereolithography system; Figures 8 to 41 are diagrams showing the display screens generated by the part manager, Figure 42 is a diagram showing the order of various tasks normally performed by the part manager, and Figure 43 is A detailed diagram showing the data flow and task distribution within the control combicoater; FIG. 44 is a flowchart of the algorithm for crosshatch and filling hector generation executed continuously by the control computer; FIGS. 45 to 49 are crosshatch 7 /Figures 50 and 51 more fully illustrate the operation of the filling vector generation algorithm for stereo DSP and servo DS.
Figure 52 is a diagram showing the actual path of the trajectory of the same with respect to an acute angle; Figures 5'3 to 65 are diagrams showing the format of data packets passed between FIG. 1. Control more annealing, 2. Process more annealing, 3. Service module, 4. Resin tank, 6. Platform, 8. Beam profiler, 9
... Engineering [Noheta] 1... Laser, ] 3... Focusing sensor 4.15... Dynamic mirror 16
...Main mirror,]7 Resin tank, 18...Beam extractor, 20.High-speed shutter, 22.
Attineta.

Claims (1)

[Claims] 1. A container of material capable of undergoing a selective physical transformation when exposed to an active stimulus, and having a surface, and having an intensity of at least about 20 W/mm^2; means for providing a beam of active stimulus that follows an optical path to the surface of the material; capable of rotating at a speed and adapted to direct a beam toward said surface;
a rotatable movable mirror positioned along the optical path, the mirror being adapted to sequentially position a beam across the surface to transform material at the surface and form successive layers of the part;
A three-dimensional fabrication system comprising: a shutter that intersects the beam and selectively blocks passage of the beam in areas of a material surface that are not to be transformed. 2. The system of claim 1, wherein said means produces a beam having an intensity of at least about 32 W/mm^2. 3. The system of claim 1, wherein the movable mirror is rotatable at a speed of at least about 100 inches. 4. The system of claim 1, further comprising an attenuator placed along the optical path and intersecting the beam to selectively attenuate the beam typically at regions of the material surface where a thin layer is desired. 5. A container of material capable of undergoing selective physical transformations and having a surface when exposed to an active stimulus, and means for providing a beam of said active stimulus that follows an optical path to the surface of the material. a movable dynamic mirror positioned along the optical path for directing a beam toward the surface and sequentially positions the beam across the surface of the material to form successive layers of the material in response to generation of sequential commands; A first step is coupled to the mirrors and sequentially generates the commands to the mirrors in order to make the three-dimensional modeling data describing the layers.
processor means coupled to the first processor means for receiving external CAD/CAM data describing a surface of a portion, and thereafter receiving external CAD/CAM data describing a first portion of the at least one layer; a second portion of the stereolithography data, the second portion of the stereolithography data describing the second portion of the layer; A stereolithography system comprising: a processor means; 6. The system of claim 5, wherein said second processor means converts said CAD/CAM data into said first portion of said stereolithography data describing boundaries of said layers; Means is a system for generating the second portion of the stereolithography data describing an area within the boundary of the layer. 7. An expanded sheet containing a lattice of cells at the intersections of substantially parallel rows and columns separated by a container of matter capable of undergoing selective physical transformations when exposed to an active stimulus. means for displaying a model of a part; and combining at least one CAP/CAM file describing a surface of at least a component of a part into a first selected of said rows and said columns of said expanded sheet model. a second selection of said rows and said columns of an expanded sheet model, said second selection intersecting said first selection in a cell; means for assembling all ranges of layers; means for providing a style file containing stereolithography process data; means for associating the style file with the cells of the expanded sheet model; and the CAD/CAM file. means for converting the 3D model into 3D modeling layer data describing the constituent elements within the range of layers; and means for configuring the layer of the component of the part by sequentially exposing the material to the action stimulus according to the stereolithography layer data and the stereolithography process data. 8. The system of claim 7 further comprising means for sequentially forming layers of material on top of previously formed layers of the portion in accordance with the stereolithography process data. 9. An extended process involving a substance capable of undergoing selective physical transformations when exposed to an active stimulus and a lattice of cells at the intersections of nearly parallel rows and columns separated by displaying a model of a sheet; and assembling at least one CAP/CAM file describing a surface of at least a component of a part into a first selected of said rows and said columns of said expanded sheet model; a second selection of the rows and columns of the expanded sheet model, the second selection intersecting the first selection in a cell; providing a style file containing stereolithography process data; assembling the style file to the cells of the expanded sheet model; converting a file into stereolithography layer data describing said components within said range of layers; and utilizing the expanded model to obtain stereolithography layer data and stereolithography process data associated with cells. and then configuring the layer of the component of the part by sequentially exposing the material to the action stimulus in accordance with the three-dimensional modeling layer data and the three-dimensional modeling process data.