WO1996000422A1

WO1996000422A1 - Programmable mask for producing three-dimensional objects

Info

Publication number: WO1996000422A1
Application number: PCT/US1995/007994
Authority: WO
Inventors: Paul C. Gillette; Herbert T. Conner
Original assignee: Hercules Incorporated
Priority date: 1994-06-27
Filing date: 1995-06-23
Publication date: 1996-01-04
Also published as: AU2908895A

Abstract

A system is provided for generating three-dimensional objects by using a directly programmable mask (34) to selectively control the transformation of a liquid or liquid-like medium (24) into a substantial solid and directly transforms a raw material into a solid object using a general approach not requiring specialized tooling for specific machining operations. This system includes a light source (26), a programmable mask (34), imaging optics (36), and an object building reservoir (22) which may optionally be equipped with a resin layer delivery system (128) and/or liquid medium height control system (122). Computer hardware and software (66) are employed to provide the appropriate means for controlling these various system components. The three-dimensional objects are produced by selectively controlling the liquid:solid solidification of an appropriate reactive liquid medium of thin two-dimensional cross-sectional 'slices', or layers. New layers are formed from a continuous layer of a reactive (i.e. solidifiable) liquid medium adjacent to existing previously fabricated (i.e. solidified) layers of the object.

Description

PROGRAMMABLE MASK FOR PRODUCING THREE-DIMENSIONAL OBJECTS

BACKGROUND OF THE INVENTION

1. FIELD OF THE INVENTION

This invention relates generally to improvements in methods and apparatus for the production of three-dimensional objects. More particularly, the invention relates to systems for producing three dimensional objects from a solidifiable liquid medium environment and, specifically, to a technique for more readily and accurately producing the shape and size of an object being sought.

2. DESCRIPTION OF THE PRIOR ART

Traditional fabrication processes have largely been based on either (1) removing material from a solid block as with conventional machining operations or (2) using a mold as a template to contain a liquid medium in a desired shape which is subsequently transformed into a solid. Extensions of such processes include techniques such as those described in US 4,915,757 (Rando) which make use of a positionable laser beam to ablate material to form the desired object. Conventional machining operations generate material waste, are time consuming, and often require multiple pieces of specialized equipment or different "set¬ ups" to produce a single part. Molding operations necessitate the production of specialized molds and often require a great deal of preparation/experimentation to produce a single satisfactory part.

Over the past century a variety of techniques have been developed for producing three-dimensional objects or relief structures directly from raw materials (e.g. US 774,549 to Baise) . US 2,775,758 (Munz) describes an iterative process for making a three dimensional object from a series of sequentially formed photosolidified thin layers. US 3,428,503 (Beckerle) describes a process which makes use of a sequential series of negatives generated under defined lighting conditions which are used to reproduce a positive or negative representation of a three-dimensional object by etching or applying photoresist in layers. The advent of computer technology led to the development of automated techniques for determining the surface coordinates describing an object (US 3,866,052 to DiMatteo et al.). Early attempts (US 3,932,923 to DiMatteo and US 4,752,352 to Feygin) to utilize such data to produce an object involved cutting thin planar slices corresponding to successive cross-sections of the object and stacking them together to form the desired object. A more complex process involved the selective transformation of material at the intersection of two light beams (US 4,041,476 to Swainson, US 4,078,229 to Swainson et al., US 4,238,840 to Swainson, and US 4,288,861, US 4,333,165, US 4,466,080, and US 4,471,470, all to Swainson et al.). The difficulty of maintaining beam convergence of two beams through a liquid medium as well as fundamental photochemical and diffusion related problems limit the practical implementation of this approach.

US 4,247,508 (Housholder) , US 4,938,816 (Bea an et al.), US 5,006,364 (Fan), and US 5,038,014 (Pratt et al.) describe several processes for building up three- dimensional objects from layers including approaches involving selective particle fusion or welding using a mask or positionable laser beam. US 4,665,492 (Masters), US 4,749,347 (Valavara) , and US 5,121,329 (Crump) describe processes involving controlled deposition of material from a translating nozzle to construct the object. Fabrication of large structures (such as buildings) using a translating nozzle which deposits foam is disclosed in US 3,776,990 ( atkins, Jr.). US 5,059,266 (Yamane et al.) makes use of an ink jet printing head to deposit small droplets of material to construct an object. US 4,801,477 (Fudim) describes a process in which a radiation transmitting device is translated in a container of a photopolymerizable liquid medium to selectively solidify regions in order to form an object. US 4,943,928 (Campbell et al.) sought to eliminate costly lasers and their required scanning systems by making use of a plurality of spot heat sources to selectively cure a thermoset liquid medium. Selective curing of sheets using electro-photographic techniques has also been described (US 5,088,047 to Bynum) . A similar process using selectively hardened precursor sheets is also described in US 5,094,935 (Vassiliou et al.) . US 5,120,476 (Scholz) describes a process in which regions of photopolymerizable fluid are selectively deposited onto a liquid medium supporting surface with continuous irradiation to solidify.

H. Koda a pioneered a technique based upon repeated selective photopolymerization of thin layers corresponding to cross-sections of a three dimensional object. US 4,575,330 (Hull) first introduced the term "stereolithography" to describe:

"... a method and apparatus for making solid objects by successively 'printing' thin layers of a curable material, e.g., a UV curable material, one on top of the other. A programmed movable spot beam of UV light shining on a surface or layer of UV curable liquid is used to form a solid cross-section of the object at the surface of the liquid. The object is then moved, in a programmed manner, away from the liquid surface by the thickness of one layer, and the next cross-section is then formed and adhered to the immediately preceding layer defining the object. This process is continued until the entire object is formed." (Col. 2, line 37 )

Related stereolithography patents include US 4,929,402 (Hull) and US 5,071,337 (Heller et al.). Post-processing with off-peak absorptive wavelengths to achieve a more uniform cure is described in US 5,076,974 (Modrek et al.).

US 4,752,498 (Fudi ) describes a similar process although it makes use of a mask which is in contact with the photopolymerizable liquid medium to form cross-sectional slices. A variety of improvements to stereolithography have been proposed. In an attempt to reduce the amount of photopolymerizable liquid needed to made an object US 5,011,635 (Murphy et al.) surrounds the object with a thin membrane which itself is in a fluid bath. Inherent problems associated with vector scanning have necessitated the development of complex strategies for scanning such as those described in US 5,014,207 (Lawton) which are designed to maintain a constant cure depth at all scanning velocities. Other scanning strategies such as those utilized in US 4,945,032 (Murphy et al.) and in US 5,059,359 and US 5,104,592 (both to Hull et al.) serve to reduce curl and distortion otherwise present in some objects made by stereolithography. US 5,058,988 (Spence) and US 5,059,021 (Spence et al.) describe a method for calibrating stereolithography apparatus. US 4,942,060 (Grossa) and 4,942,066 (Fan et al.) describe compositions which containing radiation-deflecting material which is claimed to improve object resolution. To reduce re-coating time and improve resolution US 5,096,530 (Cohen) proposes that very thin films of liquid photopolymerizable resin supported by surface tension of the liquid are solidified to form cross-sections of the object.

US 4,961,154 (Pomerantz et al.) describes a method which makes use of a second non-solidifiable material which serves a support material.

The use of multiple masks as components of three- dimensional object fabricators has been described by a number of researchers. Kodoma (H. Kodama, "Automatic method for fabricating a three-dimensional plastic model with photo-hardening polymer", Rev. Sci. Instrum. , Vol. 52, No. 11, Nov. 1981, pp. 1770-1773) describes several variations on a system employing UV radiation through masks or scanning optical fibers to generate the solidified cross-sections. According to his description, the masks could either be prepared by hand or by drawing with "black ink on transparent films by using a computer-controlled XY plotter" (p. 1773, col. 2). When irradiating onto the top surface reactive resin container the object is drawn into the reactive liquid. Alternatively one can irradiate through the bottom of the container in which case the object is drawn out of the reactive liquid. Kodoma's method for resin re-coating for the top surface irradiation approach essentially involves lowering the object below the reactive liquid's surface by a thickness corresponding to the target layer thickness. Since he notes: "The fabrication time of a solid model is determined by the product of the solidifying time of a layer by the number of layers" (p. 1772, col. 1), he fails to recognize the significant amount of equilibration time required for resin re-coating for many classes of objects. If short times are used to form fresh reactive resin layers according to his scheme, then only objects with very thin cross-section walls and/or thick layers could be produced rapidly.

US 4,575,330 and US 5,174,943 (both to Hull) disclose the use of an apertured mask placed in close proximity to the working surface and irradiated with UV light. The inventor notes: "Whenever that cross-sectional shape is to be changed, a new mask... for that particular cross-sectional shape must be substituted and properly aligned. Of course, the masks can be automatically changed by providing a web of masks... which are successively moved into alignment with the surface" (col. 10, line 17 in US '330; col. 10, line 28 in US '943) . This description clearly refers to a non¬ programmable mask which would be slower to produce, involve registration problems, and possibly generate waste material associated with the mask. A number of patents refer to systems in which there is contact between an irradiated surface and some other surface. This invariably leads to release problems when a new layer of reactive resin is formed. US 4,752,498 (Fudim) describes a system based upon irradiating "through a transmittent material in contact with the cured photopoly er." (col. 6, line 55). A variety of transparent coatings are disclosed which are designed to prevent the solidified photopolymer from adhering to the contact surface, and the patent continues at col. 5, line 34: "In a preferred embodiment, mask... is a metallic layer on the bottom of plate., and is made of microelectronic mask glass". In US 4,801,477 (Fudim), a radiation guide (e.g. optical fiber bundle) in contact with the photopolymer serves to channel the radiation provided by "a changeable mask or aperture., that has different opacities and thus modulates the irradiation of fibers" (col. 4, line 49) . According to the '477 patent, other masks can be used with the guide including Texas Instrument's chip incorporating a million mirrors, or "an array radiation source., that includes a number of individually controlled radiation sources (similar to LED displays) one per fiber or per cluster of fibers" (col. 4, line 61). US 5,009,585 (Mirano et al.) describes an approach which makes use of a mask provided with slits with irradiation occurring either from the bottom or side of the container.

US 5,171,490 (Fudim) describes a process employing a radiation-transmittent flexible film in contact with the photopolymer. A variety of matrix-type irradiation means are disclosed including "so called flat-panel displays or space light modulators" which "are integrally manufactured microelectronic units having a flat radiation-emitting surface that are based on liquid-crystals (LCD) , light emitting diodes (LED) , lasers, reflecting mirrors, or the like" (col. 7, line 50). US 5,143,817 (Lawton et al.) also describes a process employing a flexible film in which one side of the film is in contact with the photohardenable composition and the other side of the film is in contact with a rigid transparent support plate. The radiation "may be transmitted through any type of variable optical density photomask such as a liquid crystal display, silver halide film, electro-deposited mask etc., or reflected off of any variable optical density device, such as a reflective liquid crystal cell" (col. 6, line 57) .

US 4,961,154 (Pomerantz et al.) discloses several variations on three-dimensional modeling apparatus according to which, in one instance, "exposure through the erasable mask may be line-by-line exposure using an electro-optic shutter, such as a light switching array, or frame-by-frame exposure using a planar array such as an LCD array" (col. 3, line 46). In a related patent (US 5,031,120, also to Pomerantz et al.) from many of the same inventors, brief mention is made of frame-by-frame exposure (col. 3, line 58) although it is not clear how this is actually utilized.

US 5,094,935 (Vassiliou) discloses a variety of imaging techniques including those employing masks (col. 7, line 1) to form images on photohardenable sheets which are subsequently placed on the surface of the partially formed object for additional curing. This technique is undoubtedly cumbersome since it involves precise registering of the sheets.

US 5,135,379 (Fudim) describes a fabrication approach which does not use a container, but rather a constructs a wall surrounding the object from the photopolymer as the object is built. A variety of means for imaging are discussed including laser scanning (col. 5, line 40) , continuous rolls of film (col. 6, line 19) , a matrix of miniature sources (e.g. light emitting diodes, col. 5, line 46), or as a matrix of miniature shutters like flat liquid crystal 8 displays (col. 6, line 32). US 5,139,711 (Nakamura et al.) also deals with a self-growing enclosure technique for developing uniform coating layers. The only apparent discussion of imaging refers to a scanning laser approach (col. 6, line 3). The inventors note that the "thickness is determined by surface tension, viscosity and specific gravity of the liquid resin as well as interfacial tension between the liquid resin., and the base plate.." (col. 6, line 21) . It appears that producing objects with different z-axis resolutions requires a material change. US 5,236,812 (Vassiliou et al.) also describes a similar approach although a reciprocating dispenser serves to deposit successive layers. Scanning laser systems are heavily emphasized although other light sources are suitable: "For example, it may be transmitted through any type of variable optical density photomask such as a liquid crystal display, silver halide film, electro-deposited mask etc. , or reflected off of any variable optical density device, such as a reflective liquid crystal cell" (col. 9, line 64) . The necessity to build a retaining wall with each layer, as disclosed, generates resin waste.

US 5,217,653 (Mashinsky et al.) discloses a complex system which makes use of a liquid crystal matrix (col. 6, line 33) as well as a complex optical fiber with four degrees of freedom which may be directed towards the liquid medium at a predetermined angle equal to the representative slope of the side wall. The inventors employ an anti-adhesion layer such as gelatine (col. 6, line 46) to prevent the photopolymerized liquid from sticking to several key surfaces.

US 5,247,180 (Mitcham et al.) describes a system based upon Texas Instruments' area array deformable mirror or micro- mirror-array. Fresh reactive resin layers may be introduced by either forcing material through a perforated grid supporting the object or with the aid of an "applicator" which sprays a small amount of liquid onto the workpiece to ensure uniform film coverage of the workpiece (col. 3, line 18). The depicted embodiments of the process, however, describe a system in which the distance between the imaging means a working surface varies over the course of object fabrication.

Although US 5,126,529 (Weiss et al.) makes use of masks and builds up an object layer by layer, it differs significantly from the aforementioned prior art. In this case the masks contain apertures through which material is sprayed. Complementary masks are used to enable true cross-sections to be formed that would otherwise not be possible using a single mask. All of the material used to make the masks represents process waste.

The object of the present invention is to provide a rapid, low cost, low maintenance system for producing high quality high quality complex three-dimensional objects directly from raw materials with a minimum of waste.

SUMMARY OF THE INVENTION

Briefly, and in general terms, the present invention provides a new and improved system for generating three- dimensional objects by using a directly programmable mask to selectively control the transformation of a liquid or liquid-like medium into a substantial solid. Unlike most conventional machining processes predicated on material removal to generate the designed shape or the use of molds, the process described herein directly transforms a raw material into a solid object using a general approach not requiring specialized tooling for specific machining operations. The technique is especially useful for transforming three dimensional computer renderings into solid objects. Both equipment and its operation are described. 10

The apparatus of the present invention comprises the following components: a light source, a programmable mask, imaging optics, and an object building reservoir which may optionally be equipped with a resin layer delivery system and/or reactive liquid medium height control system. Computer hardware and software are preferably employed to provide the appropriate means for controlling these various system components. The apparatus of the invention produces three dimensional objects by selectively controlling the liquid:solid transformation (i.e. solidification) of an appropriate reactive liquid medium of thin two dimensional cross-sectional "slices", or layers. New layers are formed from a continuous layer of a reactive (i.e. solidifiable) liquid medium, or resin, adjacent to existing previously fabricated (i.e. solidified) layers of the object. It will be understood that the term "two dimensional layer" is used throughout the disclosure to refer to a layer of reactive liquid medium which has a finite, but minimal, thickness. In actual fact, it is customary for such "two dimensional layers" to have a thickness in the range of approximately 0.004" to 0.030" although both thinner and thicker ranges may be appropriate for certain applications.

The proposed apparatus differs from existing masked-lamp curing in several important respects: Commercially available systems utilize a sequential series of discrete masks produced out of line. Typical of such a commercially available system is the Solider 5600 manufactured by

Cubital Ltd. of Raanana, Israel. In the present invention the mask is readily reprogrammed in place. As a result fewer mechanical components are required. In addition the mask associated with the present invention does not directly contact the reactive liquid medium's surface; rather an image is projected onto the surface of a reactive liquid medium. The approach of the present invention is to eliminate mask-to-object adhesion problems occurring with contact lithography. In one embodiment of the invention, 11 greater flexibility is provided by permitting the user to enlarge or reduce the projected image. This feature permits the user to tailor the operation of the apparatus to fulfill the resolution requirements of the desired object and enable its production in a minimum amount of time. One or more cross-sectional slices of the object may be generated within the two dimensional layer of reactive liquid medium.

A preferred combination of critical system components for a commercial system embodying the invention would include the following:

■ A light source: a high intensity tungsten halogen 650 watt bulb;

■ A programmable mask: minimum 640x480 pixel, high contrast (>100:1) thin film transistor liquid crystal display panel equipped with internal fan cooling, interfaced to a standard personal computer equipped with a VGA (video graphics array) adapter;

■ Imaging optics: prior to programmable mask: parabolic reflector for light source, mirror, and fresnel lens; after programmable mask: objective lens(es) as well as optional aperture stop and/or shutter; and

■ Object building reservoir: equipped with computer controlled stepper motor activated object building platform; suitable liquid medium applicator with liquid level sensordisplacement block liquid medium height control; inert ("oxygen or water free") atmosphere maintained in small headspace above reactive liquid medium may be desirable depending upon the 12 characteristics of the reactive liquid medium.

Other and further features, advantages, and benefits of the invention will become apparent in the following description taken in conjunction with the following drawings. It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory but are not to be restrictive of the invention. The accompanying drawings which are incorporated in and constitute a part of this invention, illustrate some of the embodiments of the invention, and, together with the description, serve to explain the principles of the invention in general terms. Like numerals refer to like parts throughout the disclosure.

BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 is a diagrammatic illustration of a system embodying the present invention for producing three-dimensional objects from a reactive liquid medium;

Figs. 2A, 2B, and 2C are diagrammatic illustrations depicting successive operating conditions for our manner of operating a programmable mask, a component of the system of Fig. 1, operated in accordance with one embodiment of the invention;

Figs. 3A, 3B, and 3C are diagrammatic illustrations, similar to Figs. 2A, 2B, and 2C, depicting another manner of operating the programmable mask of the invention;

Figs. 4A and 4B are side elevational views, respectively, diagrammatically illustrating a reservoir of the system illustrated in Fig. 1 and depicting, successively, two modes of operation of another embodiment of the invention;

Figs. 5A, 5B, and 5C depict complementary masks, 13 respectively, produced by the programmable mask of the system of Fig. 1, according to still another embodiment of the invention;

Figs. 6A, 6B and 6C are diagrammatic side elevation views, in section, illustrating the reservoir of the system of Fig. 1 and depicting one mode of applying a new layer of reactive liquid medium to an uppermost surface of the object being formed;

Figs. 7A, 7B, 7C, 7D, and 7E are all side elevation views, similar to Figs. 6A-6C but depicting another mode of operation of the system for forming a new layer of reactive liquid medium on the uppermost surface of the object being formed;

Figs. 8A, 8B, 8C, and 8D, are side elevation views, in section, illustrating the reservoir of the system of Fig. 1 and illustrating successively, various modes of operation to assure that the height of the surface of the liquid medium remains in a fixed relationship relative to the imaging optical system, another component of the system of Fig. 1; and

Fig. 9 is a perspective diagrammatic illustration of a modified exposure head for use with the system of Fig. 1.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

A more detailed discussion of the specific system components and their associated functions now follows with specific attention, initially, to Fig. 1 which generally illustrates a system 20 embodying the present invention.

The system 20 includes a reservoir 22 filled with a suitable reactive liquid medium 24 in the form of a liquid medium which is curable, that is, solidifiable, when 14 exposed to a source 26 of synergistic stimulation. A surface 28 of the liquid medium 24 is maintained at a constant level in the reservoir 22. The source 26 of synergistic stimulation is positioned above the reservoir 22 in a manner to project its radiation, with the aid of a parabolic reflecting mirror 27 and a condenser lens 27A, via a suitable mirror 30 toward the surface 28 of the liquid medium 24. Radiation from the source 26 is reflected through a fresnel lens 32 and an adjoining programmable mask 34. The fresnel lens 32 and programmable mask 34 are spaced apart and together spaced from (not in contact with) the surface 28 of the liquid medium 24. The fresnel lens is a well known diffractive optic which serves to image the source onto the entrance pupil of a focussing, or objective, lens 42. An imaging optical system 36 is provided intermediate the programmable mask 34 and the surface 28 and, in part, comprises a shutter 38 capable of operation between an open position enabling transmission of an image from the programmable mask 34 onto the surface 28 of the liquid medium and a closed position preventing such transmission. Adjoining the shutter 38 is an adjustable aperture stop 40 which serves to improve the quality and resolution of the projected image. Additionally, a focusing, or objective, lens 42 is utilized in combination with the aperture stop 40 and shutter 38 to insure that a sharp image is projected onto the surface 28. In the event the programmable mask is in relatively close proximity but not in contact with the surface of the reactive liquid medium, the objective lens 42, the aperture stop 40, and the shutter 38 may be dispensed with.

A platform 44 suitably cantilevered from a weir structure 46 is raised and lowered by means of an object elevator 47. To this end, an inboard end of the platform 44 is operatively engaged with an end of a screw shaft 48 so as to move up and down with an end of the screw shaft as a stepping motor 50 is actuated. Thus, the platform 44 has 15 an upper surface 52 which is selectively movable from a location in the plane of the surface 28 to a location adjacent a base 54 of the reservoir 22. It is on the upper surface 52 that a three-dimensional object 53 is constructed from a plurality of successive layers of the solidified liquid medium.

It may be desirable to maintain the surface 28 in an inert atmosphere. To this end, an extension wall 56 extends upwardly from an upper rim of a portion of the reservoir 22 and is capped by a roof member 58 which is of glass or of other suitable transparent material. A head space 60 is thereby defined between the roof member 58 and the surface 28. It may be desirable to introduce into the head space 60, via a port 62, an inert gas such as nitrogen or argon in order to prevent undue exposure to oxygen or water by the liquid medium 24. Indeed, it may be desirable to provide an enclosure 64 to contain all or some of the components just recited for the same purpose. Alternatively, other means can be used to remove or substantially reduce the concentration of undesirable atmospheric components including, but not limited to chemical means of scavenging. In this latter instance, the entire enclosure 64 would be filled with an inert gas instead of only the head space 60.

A suitable computer system 66 is employed for controlling and monitoring operation of the programmable mask 34, the imaging optical system 36, and the stepping motor 50 for raising and lowering the platform 44.

The source of synergistic stimulation 26 comprises an energy source producing radiation capable of inducing chemical reactions which polymerize and/or crosslink appropriate reactive liquid media, such as the liquid medium 24, so as to cause them to solidify. Energy from the source should not substantially effect the performance 16 characteristics of the programmable mask. For example if some types of liquid crystal display devices (LCDs) are employed as masks, prolonged exposure of the mask to high intensity ultraviolet radiation may lead to its unsuitable degradation. Excessive infrared radiation may cause some programmable masks to not function properly. Appropriate filtering of radiation sources may be used to obtain radiation in the desired wavelength region of interest. Either polychromatic or monochromatic radiation sources may be used. The low cost of polychromatic radiation sources makes them especially attractive. A preferred radiation source for use with a TFT LCD programmable mask would use visible light although other radiation sources may be suitable. High intensity light from a tungsten halogen bulb is an example of a preferred light source for use with a TFT LCD programmable mask.

The programmable mask 34 represents a device capable of modulating the projected intensity of individual picture elements, or pixels, in real time. A pixel is a spatial resolution element and is the smallest distinguishable and resolvable area in an image, for example, on a liquid crystal display. As such this device can be used to project a two dimensional representation of a cross- sectional area or portion thereof of an object of interest and is capable of rapidly transforming itself (i.e. changing images) when provided appropriate control signals. Liquid crystal displays, liquid crystal light valves, and area array deformable mirror devices (for example, micro- mirror arrays available from Texas Instruments) are examples of suitable programmable mask technologies. Active matrix thin film transistor liquid crystal displays (TFT/LCD) such as those described by Howard (W.E. Howard, "Thin-film-transistor/liquid crystal display technology - an introduction", IBM J. Res. Develop.. Vol. 36, No. 1, January 1992) , represent a preferred programmable mask technology. High contrast (preferably > 100:1) and fast 17 switching speeds (preferably < 1 sec) are characteristics of suitable preferred programmable masks.

Control signals serving to modulate the image may be generated via a variety of means including, but not limited to, (1) direct computer control using known video signal standards (e.g. VGA) , (2) images stored on magnetic videotape or (3) optical recording disks. To construct large objects and/or improve resolution, it may be desirable to utilize more than one programmable mask in the design of the apparatus. Alternatively, the output from a single programmable mask may be reprogrammed and translated to a different region of the surface to form different regions within a given cross-sectional area at different points in time.

Image translation may be done in a variety of ways including, but not limited to, physical translation of the programmable mask/optics across the surface or translation of the projected image by optical means known to those skilled in the art. Translation can occur in discrete steps or, in the case of very fast programmable masks, in a continuous fashion with synchronized suitable modification of the projected image in real time. Translation of the projected image in increments of less than the distance corresponding to a single projected pixel dimension can be used to improve the apparent resolution of many object features. Depending upon the characteristics of the feature being resolved, multiple steps with different mask exposures may be required. This approach is especially useful when used in conjunction with other imaging techniques.

In this regard, turn to Figs. 2A through 2C. Consider the case of a programmable mask 34 containing an array 68 of ideal projected square pixels 70 wherein each pixel provides a uniform flux of synergistic stimulation within 18 an area enclosed by sides each x units long. With static projection of an object's cross-section onto the reservoir's surface, one is limited to producing objects with a resolution of x units in the x-y plane, or plane of the paper (i.e., features of the object will be reproduced in integral multiples of x units in the x-y plane) . In many cases, however, translation of the projected image can be used to resolve finer features. For example, consider the case where it is desired to produce a feature 3-1/2X units long. In the first step depicted in Fig. 2A, a feature 3x units long is generated. Grid lines 72 illustrate boundaries of the pixels from the programmable mask in its initial position. Only pixels in a darkened region 74 permit the synergistic stimulation to strike the surface 28 of the solidifiable liquid medium 24. Once the first imaging step is completed, the programmable mask is blanked, i.e., all pixels are switched "off" so no synergistic stimulation strikes the surface. Prior to activating an image 76 in Fig. 2B, the output of the programmable mask is translated by a distance equal to l/2x units. This translation can be done by redirecting the projected image optically and/or mechanically. For example, in the latter case, either the optics or the reservoir may be translated by an appropriate amount. Once the programmable mask's output is shifted, another image 76 is projected (Fig. 2B) . The net effect of the two operations, depicted in Fig. 2C, is to produce a structure 78 which is 3-1/2 (x) units long. Although this technique can be used to produce objects with features with detail at resolutions finer than the projected pixel dimension, the smallest isolated feature must have dimensions greater than or equal to the single projected pixel. It is further understood that the translation may occur in either, or both, x and y directions. By appropriate modification of the mask, the imaging may be done concurrently with translation. 19

Until this point, the effect of these operations on solidification in a z direction (reservoir depth) has been neglected. In the preceding example one region l/2x units long was subjected to both exposures. The cumulative effect of this "double exposure" will depend upon the characteristics of the solidifiable liquid medium as well as the geometry of the previously solidified object layers. If the double exposure occurs in a region not overlapping a previously solidified area it is possible that such exposure could lead to excess material being solidified in the z direction. This potential problem is especially acute for solidifiable media which contain bleaching photoinitiators.

One means for avoiding this problem is depicted in Figures 3A to 3C wherein the successive exposures do not overlap in any region. In this case, masks are generated to define a darkened region 80 so as to partially enclose an unexposed region 82 in the x-y plane (Fig. 3A) . For an object geometry in which the adjacent cross-sections overlap this region, it is possible to fully entrap a volume of unsolidified liquid medium. This can be achieved by generating a mask to define another darkened region 84 (Fig. 2B) . the combined darkened regions 80,84 result in a darkened region 86 (Fig. 3C) . The entrapped volume is thereby defined by an unexposed region 88. The liquid medium material so trapped can be subsequently solidified during post processing steps. Alternatively, the effects of double exposure can often be minimized by judicious selection of the area at which the double exposure takes place. In general, it is desirable to have the double exposure occur in a region where previously formed layers are at a maximum thickness.

At any given location on the working surface 28 of the liquid medium 24, the depth of solidification is dependent upon a variety of factors including, but not limited to. 20 liquid medium characteristics, projected light intensity (as a function of wavelength for polychromatic sources) , and time duration of exposure. Systematically varying the exposure conditions by changing time and/or intensity permits derivation of a calibration curve relating solidification depth to exposure. It is further understood that various combinations of time and intensity can be used to achieve the same depth of solidification. Consider the case of a reactive liquid medium in which the solidification depth is directly proportional to the product of the transmitted intensity and time of exposure. In this case the same solidification depth would be achieved by exposing the reactive liquid medium to a relative intensity=l for a relative time=l as would be by exposing with an intensity=2 for time=0.5.

Also, many reactive liquid media are known to undergo shrinkage upon solidification. This shrinkage can lead to part distortions as well as detract from mechanical properties by residual stress concentrations. Programmed panel solidifying may be used to minimize these effects. This technique involves varying the solidification rates in different regions of a given two dimensional slice or layer. For example consider the case of solidifying a solid circular cross-section of the object. Rather than simply exposing all elements within the circle to a uniform intensity for a fixed time, a more sophisticated strategy can be employed. A point in the circle's center is initially exposed to high light intensity which slowly decreases to zero intensity over a time period sufficient to solidify the material to a desired depth. Conversely, material located on the circle's perimeter is initially exposed to low light intensity which is slowly increased to a high intensity over the same time interval. In each instance, material is solidified to the same depth, but for the present example the circle's center solidifies initially more rapidly than material at the perimeter. 21

Other regions within the circle are solidified according to interpolated time:intensity profiles based on their positions relative to the center and perimeter. In this case, programmed panel curing permits liquid medium to be replenished from the edges as shrinkage occurs. An extension of this concept to include more complex imaging strategies in which subregions within individual layers are treated independently will be obvious to those skilled in the art. Just as the technique may be used to relieve residual stresses, it may also be used to intentionally impart stresses within the object to balance other stresses or for other purposes. It is understood that equivalent results can be achieved by varying exposure time as opposed to intensity. An example of this approach for high speed switching programmable masks involves modulating the intensity on:off for different time intervals as a means of varying exposure.

Attached Tables 1A, IB, 1C, and ID are designed to illustrate this "programmed panel solidifying" approach. Consider the case of an object cross-section that is a circle which is solidified using four different masks to control the solidification process so that the center of the circle initially solidifies at a higher rate than points on the circle's edge. Each of the attached tables contains a map of the projected relative intensity on the reactive liquid medium's surface at different points in time for a low resolution programmable mask. The projected intensity may be obtained by either setting the programmable mask 34 to permit the desired amount of light to pass through (i.e. , setting gray-scale intensity) , or by modulating a higher intensity of light on:off to produce the desired exposure for the particular time increment. For the first exposure condition, the center of the circle receives the most intensity with very little intensity at the maximum radius of the circle. (Intermediate intensity values, in proportion to the relative radial position, are 22 projected for regions lying between the center of the circle and its maximum radius.) As time progresses, this relationship is reversed until, by the fourth (and final) time step, very little intensity is projected at the center of the circle, but high intensity is maintained at the maximum radius of the circle. To generate a uniform thickness for the entire cross-section, the total exposure (equal to the sum of the projected intensity values for that point in the array) at any point inside the circle is, however, constant. It is assumed the reactive liquid medium '24 behaves ideally in the sense that the extent of solidification is directly proportional to the total exposure (i.e., tensity x time). Exposure conditions for reactive liquid media exhibiting nonideal behavior may require altering time:intensity relationships to achieve the desired solidification depth. Although the present example contains only 4 time steps, it is understood that the concept can be extended to a very large number of time steps. Other programmed exposure patterns optimized for specific cross-sectional geometries will be apparent to those skilled in the art.

23

24

25

26

Variations in source intensity projected through the programmable mask 34 may be compensated for by using one or more of several techniques depending upon the features of the panel used or other factors. Simply varying the exposure time at each pixel (i.e. pixels passing higher intensities of light are exposed for less time than those pixels passing lower light intensities) represents one means of achieving a uniform solidification. Although not required, use of programmable masks providing graduated individual pixel intensities (as opposed to simple on/off, i.e. bright/dark capability) provides a number of desirable features including, but not limited to, the ability to correct for nonuniformities in light source intensity across the panel's surface and/or the opportunity to vary the solidification rate at each pixel within the slice to minimize shrinkage induced distortions in object geometry.

27

With graduated pixel intensity it is possible to program individual pixels on the panel to transform a nonuniform incident intensity on the panel's illuminated surface into a uniform transmitted intensity. Specific means for panel programming are dependent upon associated electronics. Some panels permit direct modification of color intensities associated with each pixel, others map color indices into a video palette. In the latter case, all pixels having a particular color index may be effectively simultaneously altered by changing the color associated with color index as opposed to re-writing each pixel individually.

Tables 2A-2C below illustrate the application of these concepts for a small llxll programmable mask whose surface is irradiated with a nonuniform intensity across its surface.

28

Table 2B - Panel Time Exposure Scale Factors

11.1 7.2 5.1 4.0 3.5 3.3 3.5 4.0 5.1 7.2 11.1

7.2 4.7 3.3 2.6 2.3 2.2 2.3 2.6 3.3 4.7 7.2

5.1 3.3 2.4 1.9 1.6 1.5 1.6 1.9 2.4 3.3 5.1

4.0 2.6 1.9 1.5 1.3 1.2 1.3 1.5 1.9 2.6 4.0

3.5 2.3 1.6 1.3 1.1 1.0 1.1 1.3 1.6 2.3 3.5

3.3 2.2 1.5 1.2 1.0 1.0 1.0 1.2 1.5 2.2 3.3

3.5 2.3 1.6 1.3 1.1 1.0 1.1 1.3 1.6 2.3 3.5

4.0 2.6 1.9 1.5 1.3 1.2 1.3 1.5 1.9 2.6 4.0

5.1 3.3 2.4 1.9 1.6 1.5 1.6 1.9 2.4 3.3 5.1

7.2 4.7 3.3 2.6 2.3 2.2 2.3 2.6 3.3 4.7 7.2

11.1 7.2 5.1 4.0 3.5 3.3 3.5 4.0 5.1 7.2 11.1

Table 2A reports the projected intensity values measured on the surface of the reactive liquid medium when all pixels in the array are set to pass the maximum amount of light. In this case, maximum energy throughput occurs at the center of the array and falls off in a Gaussian distribution as one moves towards the edges. Assuming the reactive liquid solidifies in direct proportion to exposure, it is possible to compensate for the intensity 29 nonuniformity by simply varying the relative exposure time at each point as defined by the values in Table 2B. The product of corresponding elements in Tables 2A and 2B yields the total exposure at each point which is constant for the values indicated. Alternatively one could program the panel to permit a fraction of the light to pass through in regions that would otherwise be very intense. In this case, uniform intensity reaches the surface at all times in the regions designated by the programmable mask. Table 2C lists the correction factors necessary to obtain a uniform intensity on the surface.

From a practical standpoint, it is desirable to develop intensity correction factors based on the regions in which an image is displayed. Such factors ultimately result in longer exposure times, since light throughput is being limited. Since minimum exposures are desired to minimize fabrication time one should only limit the amount of light based on the "weakest" intensity in a displayed region. Consequently, it is preferred that the specific characteristics of each cross-sectional area be considered when determining the exposure for systems in which the projected source intensity varies with respect to surface position. For example, the projected image for each layer may be evaluated to determine whether solidification is to take place in areas of the two-dimensional layer requiring intensity correction. Where no solidification is intended in low intensity areas, no additional exposure time is required. Conversely, when solidification is desired in lower intensity areas, additional exposure may be given appropriate to those intensity requirements. Practical extension of these concepts to systems in which solidifica¬ tion is not directly proportional to exposure (i.e. reciprocity failure occurs) through the use of appropriate calibration curves is obvious to one skilled in the art.

Turn now to Fig. 4A which diagrammatically illustrates two 30 dimensional layers 90 of an object 53 being formed within the liquid medium 24. More specifically, it illustrates a fully solidified layer 90A and a layer 9OB which is being subjected to incident radiation from the source 26 as depicted by an arrow 92. Thus, layer 90B is in the process of being solidified. Under some circumstances, it is de¬ sirable to effectively form more than one cross-sectional layer 94 of the object 53 within a single layer 90 of reactive liquid medium. This may be desirable in an effort to save fabrication time, to obtain an object 53 having a desired profile, and also to achieve better resolution of the object in its final form. The depth of solidification is controlled at each location corresponding to a projection of a pixel in the programmable mask 32 on the surface 28 of the liquid medium 24.

Each cross-sectional layer 94 within a single layer 90 may be referred to as a "sub-region", that is, some portion of that part of the two-dimensional layer of the liquid medium to be solidified having a thickness less than that of the two dimensional layer. There may be one or more such sub- regions associated with a given two-dimensional layer 90, each of which may be solidified to a different thickness. For constructions such as that depicted in Figure 4A having "downward slopes" this simply corresponds to varying the exposure time and/or intensity so as to vary the depth of solidification. For a reactive liquid medium in which the solidification depth is directly proportional to exposure, this implies exposing an inner region 96 for twice as long as an outer region 98.Of course, it will be understood that Fig. 4A is merely illustrative and that the layer may be profiled in any desirable manner between its peripheral edge or edges and an inner region.

Regions such as those depicted in Figure 4B prove more challenging in the solidification of multiple cross- sections since solidification must be performed below the 31 surface of the liquid medium 24 in the case of an outer region 100. Several approaches are possible: One approach makes use of light activated inhibitors used in conjunction with a suitable source of inhibitor activating radiation as depicted by an arrow 101. Examples of such inhibitors are disclosed in US 3,885,964 (Nacci) , 3,901,705 (Pazos) , 4,029,505 (Nebe) , 4,050,942 (Nacci), and 5,175,077 (Grossa) . Inner region 102 is initially selectively solidified with an appropriate mask. The entire surface would then be exposed with an appropriate source capable of activating the inhibitor to a defined depth. It is not required that the exposure be done through the programmable mask 34, only that it be done in a controlled manner across the surface. The inhibitor activating wavelength would be selected such that initiation of the photoinitiator did not occur. For example it may be desirable to utilize a visibly activated photoinitiator combined with an ultraviolet activated inhibitor.

An additional characteristic of the inhibitor is that it permit penetration of wavelengths to a desired depth sufficient to initiate photopolymerization. Once an inhibiting layer is formed, an appropriate mask for region 100 would be activated and exposed. The activated inhibition layer formed at the surface serves to prevent solidification of the surface. Subsurface solidification beyond the activated inhibition layer would, however, occur. Extension of this concept to substantially infinitely thin cross-sections is straightforward and may involve combinations of the concepts already proposed to achieve optimal results. Programmable masks are especially well suited in this regard since they permit more easily programmed variations in pixel exposure than scanning devices. Other variations on this approach using different exposure techniques to achieve the same ends will be obvious to those skilled in the art. 32

When forming successive layers it is important to solidify to a depth sufficient to insure bonding between the newly formed layer and the surface of previously formed layers. Failure to do so will risk interlayer adhesion failure followed by delamination which will prevent subsequent layers from being formed properly. The temptation is to overexpose layers to account for variations in source intensity or object positioning over the course of fabrication time. Such a strategy would result in many layers being slightly too thick. For reactive liquid media containing initiators which bleach (i.e. those which do not absorb stimulating wavelengths upon decomposition) , there is also a more serious side-effect. In this case, excess exposure energy is not absorbed by previously formed layers, rather it passes through them and can begin to initiate solidification in reactive liquid medium potentially well below the surface.

Although the excess exposure resulting from one layer may be small, the effect is cumulative which can lead to a significant degradation of resolution. One method of minimizing this effect is to make use of interlayer adhesion masks, examples of which are depicted in Figures 5A, 5B, and 5C. With this approach, the overexposure is selectively applied in different regions on successive exposures. Exposure sufficient to induce solidification to a depth slightly less than or equal to the desired target depth is done without the interlayer adhesion mask being applied to the projected image. Each interlayer adhesion mask insures adhesion of the newly formed layer by selective bonding a fraction of the pixels. Referring to Figure 5A ("N=2") , two complementary masks 104, 106 would be alternated between layers. Maximum density, or black, pixels 108 in mask 104 coincide with minimum density, or white, pixels 110 in mask 106. Use of the masks 104, 106 simply requires logically "AND"ing the displayed image with the mask image to yield an image in which half the pixels 33 are set relative to the original image for the case depicted in Fig. 5A.

Commercially available software and hardware provide appropriate means for readily performing these logical operations in video memory using appropriately generated arrays. The thus formed image is then displayed for the excess time to insure adequate adhesion. Use of higher order masks, examples of N=3 (Fig. 5B) and N=4 (Fig. 5C) , serves to reduce excess solidification, although since fewer points are exposed there is a risk of poorer adhesion associated with higher order masks. More sophisticated implementations of this strategy optimized for specific object geometries/cross-sections which insure adhesion of small irregular features and account for other factors will be apparent to those skilled in the art. In general, it is preferable to develop sequential interlayer adhesion masks so that there is minimal overlap of activated corresponding pixels in the z direction.

The imaging optical system 36 is designed so as to provide a crisp representation of the projected image on the surface of the reactive liquid medium. Suitable magnifica¬ tion or reduction of the image by the optics can be used to control resolution and/or object fabrication time. Ideally, the projected image should be sized so as to activate as many pixels in the programmable mask as possible. Depending upon the specific reactive liquid medium formulation being used, it may be desirable, as noted above, to maintain the surface of the reactive liquid medium in an inert atmosphere (e.g. oxygen free for free radical chemistry or water free for cationic systems) . This can be done by including a low reflecting/low radiation absorbing sheet of an appropriate material (e.g. low reflective coated glass) above the reactive liquid medium's surface as embodied by the cover sheet 58. Modification of the atmosphere can be done through the use 34 of chemical reactions designed to scavenge undesired components or by flushing with an_. inert gas which may be introduced through the port 62.

Due to the potential difficulty associated with separating solidified liquid medium from the cover sheet 58, it is preferred that the cover sheet and the surface 28 not directly contact one another. Alternatively, the volume between the surface 28 and imaging optical system 36 may be enclosed, as by the enclosure 64, and the desired atmosphere maintained therein. This approach is advantageous in that it eliminates the potentially reflect¬ ing and/or absorbing cover sheet 58 which would reduce energy throughput. Use of a computer controlled shutter 38 is often desirable to completely eliminate low level light exposure during nonimaging operations since the programmable mask 34, in some instances, may not fully eliminate light transmission when pixels are set "off". Incorporation of the shutter 38 has the added benefit that status information can be displayed on an external display device during nonimaging operations using the same video driver controlling the imaging programmable mask. Without a shutter, the status information would be projected onto the reactive liquid medium '24.

The preceding disclosure presents the components employed for selectively solidifying two dimensional layers 90 of the reactive liquid medium '24 on the surface 28 corresponding to cross-sectional slices of the desired object 53. The following discussion pertains to the steps required to introduce a new layer of the reactive liquid medium covering the most recently created surface of the solidified object to permit formation of a subsequent layer. This can be performed in a variety of ways.

With reference to Figs. 6A through 6C, the approach for forming a new layer of reactive liquid medium originally 35 described by Kodama is presented. As seen in Figure 6A, the upper surface 112 of the partially fabricated object 53 is level with the upper surface 28 of the liquid medium 24. This approach involves lowering the already solidified part of the object 53 from the position illustrated in Fig. 6A a distance such that when the reactive medium flows across surface 112 a layer having a thickness of a single two- dimensional layer 90 is formed. This may be achieved in the manner previously described with reference to Fig. 1. When the object 53 assumes the position illustrated in Fig. 6B, fresh, unreacted liquid medium flows across an upper surface 112 of the previously reacted, now solid, object 53 to form a uniform two dimensional layer 90 (Fig. 6C) of liquid medium capable of undergoing reactive solidifica- tion. This technique is best suited for very low viscosity reactive liquid media or structures with thin wall thicknesses. Unfortunately low viscosity often implies low molecular weight reactive liquid media which require a longer exposure than their higher molecular weight analogues.

Not illustrated are preferred means for maintaining the focal plane at a fixed position. In the event the reactive liquid medium undergoes a volume change upon solidifica- tion, it is preferable to adjust the liquid medium's volume or otherwise alter the shape of the reservoir so as to maintain the surface 28 at a constant distance relative to the imaging optics. A simple variation on this approach is to raise the liquid medium level with appropriate translation of the object building reservoir 22 or imaging optical system 36. In either case, the time required to form a substantially uniform layer of the reactive liquid medium 24 on the upper surface 112 of the previously solidified layer increases with increasing reactive liquid medium viscosity, increasing maximum flow distance across the surface, and decreasing layer thickness. For any point on the interior of a given cross-section, the flow distance 36 is the minimum distance between that point and an edge boundary capable of supplying fresh liquid medium. In most cases the maximum flow distance represents the "worst case" point which will require the longest time to replenish fresh liquid medium or, indeed, drain excess liquid medium. This is often a valuable measurement since it can be used to estimate equilibration times. An example of an edge boundary not capable of supplying fresh medium is an interior trapped volume. After some equilibration time, a uniform layer of fresh reactive liquid medium (Fig. 6C) is formed and is ready to be imaged.

For liquid media having an elevated viscosity or for objects with large maximum flow distances, additional measures are required to form a uniform two dimensional layer 90 within a reasonable time. Turn now to Figs. 7A through 7E. Simple liquid medium flow of a thin layer over a long distance might require too much time for commercial acceptance. Another technique which may be employed involves first submerging the object 53 from the position illustrated in Fig. 7A to a position well below the surface of the reactive liquid medium, as illustrated in Fig. 7B, so as to rapidly induce flow across surface 112. The coated object is then raised above the surface 28 to the position illustrated in Fig. 7C so that the excess liquid medium is above the surface. This movement results in the formation of a dome 116 of liquid medium material overlying the upper surface 112. Thereupon, an appropriately designed doctor blade 114 is provided to remove excess resin when traversed across the upper surface 112, in either direction. As illustrated in Fig. 7D, for example, when moved in a direction of an arrow 118, the doctor blade removes the excess reactive liquid medium, leaving the two dimensional layer 90 atop the previously solidified part of the object 53.

It is understood that excess reactive liquid medium removal 37 may take place in multiple sweeps of the doctor blade 114. A useful technique when large amounts of reactive liquid medium need to be removed involves incrementally indexing the object closer and closer to the doctor blade with removal of a small amount of reactive liquid medium by the doctor blade on each cycle. This prevents accumulation of large amounts of reactive liquid medium on the leading edge of the doctor blade which might generate subsurface distortions of thin solidified layers. Whether excess reactive liquid medium removal is performed in a single pass or multiple passes, the next step requires positioning the object below the surface of the reservoir Fig. 7E, lowering the platform 44 to an operating depth such that the upper surface 112 of the object 53 being formed is below the surface 28 to a distance which is substantially equal to the thickness of that part of the next succeeding two dimensional layer to be solidified. The final levelling step typically involves waiting for the surface to equilibrate prior to exposure to the next mask. Variations on this approach are disclosed in US 5,174,931 (Almquist et al.) and US 5,238,614 (Uchinono et al.).

To avoid unnecessary optical changes during the course of object building, it is preferable that the focal plane, that is, the surface 28 of the liquid medium 24, remain fixed relative to the imaging optical system 36. The height of the reactive liquid medium 24 may change due to thermal fluctuations, liquid medium shrinkage during solidification, changes in displacement of system components in the bath during the course of object building, or other factors. Maintaining a constant focal plane can be accomplished in a variety of ways. For example, as illustrated in Fig. 8A, a feedback loop defining the relationship between a suitable level sensor 120 and a variable displacement block 122 offers a great deal of control and flexibility to compensate for process- induced variations in height of the liquid medium 24. The 38 displacement block 122 is partly submerged in the liquid medium 24 and is suitably supported and movable, as indicated by a double arrow 124, into or out of the liquid medium to the extent necessary to accommodate the volume changes which can occur in the liquid medium.

Level sensor 120 is preferably of a noncontact type which may be based, for example, on reflectance measurements of a laser beam. Alternatively, fiber optic bundles may be employed to accurately monitor surface height. Liquid medium surface height may also be adjusted by varying the shape of the volume containing the reactive liquid medium. Adjusting the position of a syringe plunger containing liquid medium connected to the primary reservoir represents a simple example of this approach. Various other expedients may also be employed. It is also possible to calculate the expected movement of the reactive liquid medium surface during the course of object building. In this instance, control is simplified since the distance sensor 120 would not be necessary.

A far simpler (and somewhat less flexible) means of main¬ taining a constant height involves the use of an overflow spout 126 (Fig. 8B) . In this case, the height of the overflow spout 126 determines the position of the reactive liquid medium's surface in the reservior. Inclusion of a computer activated valve 128 to control flow may be beneficial in some cases to prevent excessive overflow at various points in the levelling sequence. For example, certain object elevator geometries will displace the liquid medium during "deep dunking" (Fig. 7B) which can be undesirable. If the additional displacement volume is constant, then such a valve may not be necessary. Considering the former instance, it might be that when the object elevator 47, and therefore also the platform 44, are at a substantially raised position as illustrated in Fig. 8C, the displacement block 122 is at a substantially 39 elevated position. Conversely, when the object elevator 47, and therefore also the platform 44, are at a substantially lowered position as illustrated in Fig. 8D, the displacement block 122 is at a substantially elevated position. In both instances, the positioning of the displacement block accommodates the displacement of the object elevator 47 to assure that the level of the surface 28 in relation to the imaging optical system 36 remains constant. The reactive liquid medium can, of course, be easily recovered and returned to the primary object building reservoir 22.

Another alternative to maintaining a constant focal length is to translate the entire reservoir 22 or imaging optical system 36. Such designs are generally more cumbersome, however, since they require controlled movement of larger assemblies than the approaches already discussed. In principle, the object elevator 47 can be replaced with an external reservoir translation assembly to maintain the constant focal length. As the reservoir 22 is filled, the entire assembly could be translated by a corresponding amount to maintain the constant focal distance.

Another embodiment of the programmable mask 34 is illustrated in Fig. 9 and will now be described. Fig. 9 is a perspective diagrammatic illustration of an exposure head

130 which may replace the mirror 30, fresnel lens 32, and programmable mask 34 provided in Fig. 1. The exposure head

130 comprises a source of synergistic stimulation 26, as previously, an area array deformable mirror device 132, a pair of lenses 134, 136, and necessary control circuitry would be provided by the computer system 66 of Fig. 1. The source 26 emits radiation, as previously, that is operable to solidify the liquid medium 24. The lens 134 more uniformly illuminates the mirror device 132 than would otherwise occur without it. Lens 136 focuses and magnifies the light reflected off the mirror device 132 onto, or 40 toward, the surface 28 of the liquid medium 24. The mirror device 132 may be an electro-optical device containing a regular n X m array of micro-mirrors of the type manufactured by Texas Instruments, Inc. of Dallas, Texas and as described in US 5,247,180 (Mitcha et al.). Each mirror device may be electronically controlled to reflect incident radiation along one of a plurality of optical pathways. In its preferred embodiment, the mirror device 132 comprises a matrix such that there are two optical pathways for each mirror. The source 26, mirror device 132, and lenses 134 and 136 are positioned such that radiation impinging upon the mirror device 132 from the source 26 may be focused onto the surface 28 if, and only if, one of the two optical pathways is selected. The optical pathway of radiation emitted from the source 26 is depicted by the converging and diverging dashed lines. Each bistable mirror on the mirror device 132 is controlled by circuitry within the computer system 66 which interprets data from a processor also contained therein.

As noted above, the active surface of the mirror device 132 may contain an n X m matrix of individually addressably bi¬ stable mirrors. Each mirror is typically a square or diamond having sides of 12 to 20 microns. This small size allows a single mirror device 132 having a footprint of approximately two square inches to have over two million addressable mirrors in, for instance, a 1920 X 1080 matrix. This small mirror size allows exposure head 130 to solidify a 4 X 8 square inch area in a single exposure interval with the same resolution as achieved by prior x X y scanner/laser exposure head combinations. Typically, these prior exposure heads achieve resolutions of +/- 0.005 inches.

Once all of the layers of the object 53 have been laid down and solidified, the object must be removed from the object building reservoir 22. Depending upon the characteristics 41 of the reactive liquid medium used, it may be necessary to perform cleaning and post-solidifying steps. Unsolidified reactive liquid medium removal is often facilitated by using appropriate washout solutions. Appropriate compositions for washout solutions tailored to reactive liquid media are well known to those skilled in the art. Solvents which cause swelling of the fabricated object should generally be avoided since they may introduce distortion. It is often desirable to spray the washout solution onto the object's surface to mechanically dislodge unreacted liquid medium. Ultrasonic baths can also be used advantageously. Once excess reactive liquid medium is removed, it may be desirable to post solidify the object 22 to fully react any unreacted species or produce an item having enhanced mechanical properties. The most appropriate expedient will depend upon the characteristics of the reactive liquid medium employed. Use of the same solidifying source of synergistic stimulation employed during object manufacture is often sufficient. Inclusion of reactive components not stimulated by the solidifying source of synergistic stimulation employed in the object building apparatus provides a mechanism for enhanced control of object post-solidification. To avoid introducing localized stresses, it is often desirable to solidify the object 53 at a uniform rate. Use of additional reactive components in the formulation provides a greater degree of flexibility in this regard.

Small surface irregularities may be removed using abrasives or smoothed by cutting. Alternatively, the part may be provided with a protective coating either by dipping, brushing, spraying, or otherwise applying an appropriate coating material. Such coatings may serve a number of functions including, but not limited to, enhancing surface finish by smoothing irregularities, imparting color, providing protection against solvents or light induced degradation of the underlying object's material, as well as

42 modifying the surface properties in a variety of ways to enhance the utility of the fabricated object. Additional steps to promote adhesion of coatings (e.g. corona, flame, or plasma treatments) may be desirable prior to their application.

While preferred embodiments of the invention have been disclosed in detail, it should be understood by those skilled in the art that various other modifications may be made to the illustrated embodiments without departing from the scope of the invention as described in the specification and defined in the appended claims.

Claims

43 CLAIMSWhat is claimed is:

1. Apparatus for producing from a liquid medium capable of being solidified a three dimensional object constructed from a plurality of successive layers of the solidified liquid medium when subjected to appropriate synergistic stimulation, said system comprising:

a reservoir containing the liquid medium capable of being solidified;

control means for providing coordinate information to enable definition of a three dimensional object of required configuration;

a source of synergistic stimulation for solidifying at least part of a two dimensional layer at the surface of the liquid medium;

programmable mask means intermediate said source of synergistic stimulation and the liquid medium and spaced from the surface of the liquid medium responsive to said control means for selectively controlling the size and shape of that part of the two dimensional layer of the liquid medium to be solidified; and

replenishment means for providing a new layer of liquid medium overlying the surface of each previously solidified layer thereof.

2. Apparatus as set forth in claim 1

wherein said control means includes a microprocessor.

3. Apparatus as set forth in claim 1 44 including optical means for providing a focussed image on the surface of the liquid medium having the size and shape of that part of the two dimensional layer of the liquid medium to be solidified.

4. Apparatus as set forth in claim 1

wherein said programmable mask means includes at least one liquid crystal display panel providing a plurality of individual pixels capable of defining the size and shape of that part of the two dimensional layer of the liquid medium to be solidified.

5. Apparatus as set forth in claim 1

wherein said programmable mask means includes an active matrix thin film transistor liquid crystal display providing a plurality of individual pixels arranged to define the size and shape of that part of the two dimensional layer of the liquid medium to be solidified.

6. Apparatus as set forth in claim 1

wherein said programmable mask means includes a plurality of mirrors individually responsive to input signals from said control means, said mirrors operable to reflect incident radiation from said source of synergistic stimulation along a plurality of optical pathways, said mirrors forming an n by m matrix; and

wherein said control means is operable to select one of the plurality of pathways along which each of said mirrors will reflect the incident radiation to form on the surface of the liquid medium an image having the size and shape of that part of the two dimensional layer of the liquid medium to be solidified. 45

7. Apparatus as set forth in claim 3

wherein said optical means includes:

an objective lens for focussing onto the surface of the liquid medium an image formed by said programmable mask of that part of the two dimensional layer to be solidified;

shutter means intermediate said objective lens and the liquid medium movable between an open position enabling transmission of the image to the liquid medium and a closed position preventing transmission of the image; and

aperture means intermediate said shutter means and the liquid medium for controlling the amount of radiation transmitted from said source to the surface of the liquid medium and reducing aberration to improve image quality.

8. Apparatus as set forth in claim 1

wherein said source of synergistic stimulation emits visible light.

9. Apparatus as set forth in claim 1

wherein said source of synergistic stimulation emits light; and

including filter means for substantially removing infrared radiation from the emitted light.

10. Apparatus as set forth in claim 1

wherein said source of synergistic stimulation emits 46 ultraviolet light.

11. Apparatus as set forth in claim 1

including:

means for providing a new layer of liquid medium overlying the surface of each previously solidified layer thereof.

12. Apparatus as set forth in claim 1 including:

enclosure means containing an inert atmosphere encompassing said reservoir.

13. Apparatus as set forth in claim 1 including:

elevator means including a platform for supporting the object being formed; and

operating means for moving said platform between a first position at which the object being formed is initially at least partially submerged in the liquid medium and a second position at an operating depth such that the uppermost surface of the object being formed is below the surface of the liquid medium to a distance which is substantially equal to the thickness of that part of the next succeeding two dimensional layer to be solidified, said operating means operable to retain the platform at the operating depth until the liquid medium completely overlies the entire uppermost surface of the object being formed.

14. Apparatus as set forth in claim 1 including:

elevator means including a platform for supporting the object being formed; and 47 operating means for moving said platform between a first position at which the object being formed is initially at least partially submerged in the liquid medium and a second position at an operating depth such that the uppermost surface of the object being formed is below the surface of the liquid medium to a distance which is substantially greater than the thickness of that part of the next succeeding two dimensional layer to be solidified, said operating means operable to raise said platform until an uppermost surface of the object being formed rises above the surface of the liquid medium such that the liquid medium overlying the uppermost surface of the object is thicker at the central regions than at the periphery thereof; and

a doctor blade movable across the upper surface of the object being formed and parallel thereto and at a uniform distance therefrom for removing excess amounts of the liquid medium from the upper surface of the object down to a depth substantially equal to the thickness of that part of the next succeeding two dimensional layer to be solidified.

15. Apparatus as set forth in claim 1

imaging optic means intermediate said source of synergistic stimulation and the surface of the liquid medium; and

means for maintaining the surface of the liquid medium at a focal plane which is a substantially constant distance from said imaging optic means.

16. Apparatus as set forth in claim 1

wherein said source of synergistic stimulation and said programmable mask include means for controlling exposure 48 from the source of synergistic stimulation across that part of the two dimensional layer of the liquid medium to be solidified.

17. Apparatus as set forth in claim 1

wherein said programmable mask means include means for controlling exposure from said source of synergistic stimulation across that part of the two dimensional layer of the liquid medium to be solidified.

18. Apparatus as set forth in claim 1

wherein said programmable mask means include means for controlling exposure across that part of the two dimensional layer of the liquid medium to be solidified so as to achieve uniformity of exposure from the source of synergistic stimulation across the surface thereof.

19. Apparatus as set forth in claim 1

wherein said programmable mask means include means for controlling exposure across that part of the two dimensional layer of the liquid medium to be solidified so as to selectively vary exposure from the source of synergistic stimulation across the surface thereof to vary the rate of solidification within that part of the two dimensional layer of the liquid medium to be solidified.

20. Apparatus as set forth in claim 1 including:

means for controlling exposure across that part of the two dimensional layer of the liquid medium to be solidified so as to achieve uniformity of exposure across the surface thereof. 49

21. Apparatus as set forth in claim 1

wherein said programmable mask means includes:

at least one liquid crystal display panel providing a plurality of individual pixels capable of defining the size and shape of that part of the two dimensional layer of the liquid medium to be solidified; and

means for controlling operation of the individual pixels of said programmable mask means so as to control exposure from said source of synergistic stimulation across the surface of that part of the two dimensional layer of the liquid medium to be solidified.

22. Apparatus as set forth in claim 1

wherein said programmable mask means includes:

means for controlling operation of the individual pixels of said programmable mask means so as to achieve uniformity of exposure across the surface of the two dimensional layer of the liquid medium to be solidified.

23. Apparatus as set forth in claim 1

wherein said programmable mask means includes:

at least one liquid crystal display panel providing a plurality of individual pixels capable of defining the size and shape of that part of the two dimensional layer of the liquid medium to be solidified; and 50 means for controlling operation of the individual pixels of said programmable mask means so as to selectively vary exposure from said source of synergistic stimulation across the surface of the two dimensional layer of the liquid medium to be solidified to vary the rate of solidification within that part of the two dimensional layer of the liquid medium to be solidified.

24. Apparatus as set forth in claim 1

(f) solidifying that part of the two dimensional layer of the liquid medium to be solidified to a depth sufficient to achieve bonding between a newly formed layer and a contiguous surface of a preceding layer at an interface thereof.

25. Apparatus as set forth in claim 1 including:

imaging optical means intermediate said source of synergistic stimulation and the surface of the liquid medium; and

means for maintaining the surface of the liquid medium at a focal plane which is a substantially constant distance from said imaging optical means.

26. Apparatus as set forth in claim 25 including:

platform means for supporting the object being constructed, said platform means being initially at least partially submerged in the liquid medium; and

operating means for lowering said platform means first to an operating depth such that the uppermost surface of the object being formed is below the surface of the liquid medium to a distance which is substantially equal 51 to the thickness of that part of the next succeeding two dimensional layer to be solidified, then for retaining the platform at the operating depth until the liquid medium completely overlies the entire uppermost surface of the object being formed.

27. Apparatus as set forth in claim 25

wherein said maintaining means includes:

means for sensing changes from the normal height of the surface of the liquid medium relative to said imaging optical means; and

means for replenishing the liquid medium in said reservoir when the height of the surface of the liquid medium falls below the normal height and for withdrawing excess liquid medium from said reservoir when the height of the surface of the liquid medium tends to increase above the normal height.

28. Apparatus as set forth in claim 25

platform means for supporting the object being constructed in the liquid medium; and

displacement mass means within said reservoir operable for movement between first and second spaced positions in the liquid medium; and

operating means responsive to movement of said platform means for moving said displacement mass means such that the surface of the liquid medium is maintained at a focal plane which is a substantially constant distance from said imaging optical means.

29. Apparatus as set forth in claim 25 including: 52 means for sensing changes from the normal height of the surface of the liquid medium relative to said imaging optical means;

platform means for supporting the object being constructed in the liquid medium;

30. A method of producing from a liquid medium capable of being solidified a three dimensional object constructed from a plurality of successive layers of the solidified liquid medium when subjected to appropriate synergistic stimulation, said method comprising the steps of:

(a) providing a reservoir containing the liquid medium capable of being solidified;

(b) directing a source of synergistic stimulation toward the liquid medium to be solidified;

(c) providing programmable mask means intermediate the source of synergistic stimulation and the liquid medium and spaced from the surface of the liquid medium, the programmable mask means being operable for selectively controlling the size and shape of that part of a two dimensional layer of the liquid medium to be solidified; and 53

(d) controlling operation of the programmable mask means utilizing coordinate information defining a three dimensional object of required configuration; and

(e) solidifying at least part of the two dimensional layer of the liquid medium.

31. A method as set forth in claim 30 including the step of:

(f) providing a new layer of liquid medium overlying the surface of each previously solidified layer thereof.

32. A method as set forth in claim 30

wherein a microprocessor is used to perform step (d) .

33. A method as set forth in claim 30

wherein steps (b) , (c) , and (d) include the step of:

(f) projecting onto the liquid medium an image having the size and shape of that part of the two dimensional layer of the liquid medium to be solidified.

34. A method as set forth in claim 30

wherein steps (b) , (c) , and (d) include the step of:

(f) projecting onto the surface of the liquid medium an image having the size and shape of that part of the two dimensional layer of the liquid medium to be solidified.

35. A method as set forth in claim 30

wherein the programmable mask means includes at least one liquid crystal display panel providing a plurality 54 of individual pixels capable of defining the size and shape of that part of the two dimensional layer of the liquid medium to be solidified.

36. A method as set forth in claim 30

wherein the programmable mask means includes an active matrix thin film transistor liquid crystal display providing a plurality of individual pixels arranged to define the size and shape of that part of the two dimensional layer of the liquid medium to be solidified.

37. A method as set forth in claim 30

including the steps of:

(g) providing the programmable mask means with a plurality of mirrors individually responsive to input signals from the control means;

(h) operating the mirrors to reflect incident radiation from the source of synergistic stimulation along a plurality of optical pathways, the mirrors forming an n by m matrix; and

(i) selecting one of the plurality of pathways along which each of the mirrors will reflect the incident radiation to form on the surface of the liquid medium an image having the size and shape of that part of the two dimensional layer of the liquid medium to be solidified.

38. A method as set forth in claim 33

wherein steps (b) and (d) include the steps of:

(h) operating shutter means intermediate the source of 55 synergistic stimulation and the liquid medium for movement between an open position enabling transmission of the image to the liquid medium and a closed position preventing transmission of the image.

39. A method as set forth in claim 33

including the step of:

(h) controlling the amount of radiation transmitted from the source to the surface of the liquid medium.

40. A method as set forth in claim 33

wherein the source of synergistic stimulation emits visible light.

41. A method as set forth in claim 33

wherein the source of synergistic stimulation emits light; and

including the step of:

(g) filtering substantially all of the infrared radiation from the emitted light.

42. A method as set forth in claim 30

wherein the source of synergistic stimulation emits ultraviolet light.

43. A method as set forth in claim 30

wherein the two dimensional layer of the liquid medium to be solidified is adjacent the surface thereof. 56

44. A method as set forth in claim 30

wherein the two dimensional layer of the liquid medium to be solidified is spaced from the surface thereof.

45. A method as set forth in claim 31

wherein steps (a) through (f) are all performed in an inert atmosphere.

46. A method as set forth in claim 31

wherein step (f) includes the steps of:

(g) supporting the object being constructed on a platform which is initially at least partially submerged in the liquid medium;

(h) lowering the platform to an operating depth such that the uppermost surface of the object being formed is below the surface of the liquid medium to a distance which is substantially equal to the thickness of that part of the next succeeding two dimensional layer to be solidified; and

(i) retaining the platform at the operating depth until the liquid medium completely overlies the entire uppermost surface of the object being formed.

47. A method as set forth in claim 46 including the step of:

(j) before again performing steps (b) and (e) , allowing the layer of the liquid medium completely overlying the entire upper surface of the object being formed to come to equilibrium so that it has a substantially uniform thickness across the entire upper surface of the object. 57

48. A method as set forth in claim 31

wherein step (f) includes the steps of:

(h) lowering the platform to an operating depth such that the uppermost surface of the object being formed is below the surface of the liquid medium to a depth which is substantially greater than the thickness of that part of the next succeeding two dimensional layer to be solidified; and

(i) raising the platform until an uppermost surface of the object being formed rises above the surface of the liquid medium such that the liquid medium overlying the uppermost surface of the object is thicker at the central regions than at the periphery thereof; and

(j) drawing a doctor blade across the upper surface of the object being formed and parallel thereto and at a uniform distance therefrom substantially equal to the thickness of that part of the next succeeding two dimensional layer to be solidified thereby removing excess amounts of the liquid medium from the upper surface of the object.

49. A method as set forth in claim 48 wherein step (f) includes the steps of:

(k) lowering the platform to an operating depth such that the uppermost surface of the object being formed is below the surface of the liquid medium to a distance which is substantially equal to the thickness of that 58 part of the next succeeding two dimensional layer to be solidified; and

(1) before again performing steps (b) and (e) , allowing the layer of the liquid medium completely overlying the entire upper surface of the object being formed to come to equilibrium so that it has a substantially uniform thickness across the entire upper surface of the object.

50. A method as set forth in claim 33

wherein step (f) includes the steps of:

(g) providing imaging optics intermediate the source of synergistic stimulation and the surface of the liquid medium; and

(h) maintaining the surface of the liquid medium at a focal plane which is a substantially constant distance from the imaging optics.

51. A method as set forth in claim 30

wherein steps (b) , (c) , and (d) include the step of:

(f) projecting onto the liquid medium an image having the size and shape of that part of the two dimensional layer of the liquid medium to be solidified; and

wherein step (d) includes the step of:

(g) controlling exposure from the source of synergistic stimulation across that part of the two dimensional layer of the liquid medium to be solidified.

52. A method as set forth in claim 51 59 wherein step (g) includes the step of:

(h) controlling exposure across that part of the two dimensional layer of the liquid medium to be solidified so as to achieve uniformity of exposure from the source of synergistic stimulation across the surface thereof.

53. A method as set forth in claim 51

wherein step (g) includes the step of:

(h) controlling exposure across that part of the two dimensional layer of the liquid medium to be solidified so as to selectively vary exposure from the source of synergistic stimulation across the surface thereof to vary the rate of solidification within that part of the two dimensional layer of the liquid medium to be solidified.

54. A method as set forth in claim 30

wherein steps (d) and (e) include the step of:

(f) controlling exposure across that part of the two dimensional layer of the liquid medium to be solidified so as to achieve uniformity of exposure across the surface thereof.

55. A method as set forth in claim 51

wherein the programmable mask means includes at least one liquid crystal display panel providing a plurality of individual pixels capable of defining the size and shape of that part of the two dimensional layer of the liquid medium to be solidified;

wherein step (g) includes the step of: 60

(h) controlling operation of the individual pixels of the programmable mask means so as to control exposure from the source of synergistic stimulation across the surface of that part of the two dimensional layer of the liquid medium to be solidified.

56. A method as set forth in claim 51

wherein step (g) includes the step of:

(h) controlling operation of the individual pixels of the programmable mask means so as to achieve uniformity of exposure across the surface of the two dimensional layer of the liquid medium to be solidified.

57. A method as set forth in claim 51

wherein the programmable mask means includes at least one liquid crystal display panel providing a plurality of individual pixels capable of defining the size and shape of that part of the two dimensional layer of the liquid medium to be solidified; and

wherein step (g) includes the step of:

(h) controlling operation of the individual pixels of the programmable mask means so as to selectively vary exposure from the source of synergistic stimulation across the surface of the two dimensional layer of the liquid medium to be solidified to vary the rate of solidification within that part of the two dimensional 61 layer of the liquid medium to be solidified.

58. A method as set forth in claim 30

wherein step (e) includes the step of:

59. A method as set forth in claim 58

wherein step (f) includes the step of:

(g) providing greater exposure to the source of synergistic stimulation of selected regions than unselected regions of that part of the two dimensional layer of the liquid medium to be solidified to provide greater adhesion at the selected regions between the adjacent layers of the solidified liquid medium.

60. A method as set forth in claim 59

wherein the selected regions exposed at one interface between contiguous layers are different from selected regions exposed at an interface between a next adjoining interface between contiguous layers.

61. A method as set forth in claim 58

wherein the programmable mask means includes at least one liquid crystal display panel providing a plurality of individual pixels capable of defining the size and shape of that part of the two dimensional layer of the liquid medium to be solidified; and 62 wherein step (d) includes the step of:

(g) controlling operation of the individual pixels of the programmable mask means so as to selectively vary exposure from the source of synergistic stimulation across the surface of the two dimensional layer of the liquid medium to be solidified and thereby solidify that part of the liquid medium to be solidified to a depth sufficient to achieve bonding between a newly formed layer and a contiguous surface of a preceding layer at an interface thereof.

62. A method as set forth in claim 61

wherein step (g) includes the step of:

63. A method as set forth in claim 62

64. A method as set forth in claim 30

wherein step (e) includes the step of:

(f) solidifying a subregion of that part of the two dimensional layer of the liquid medium to be solidified having a thickness less than that of the two dimensional 63 layer .

65. A method as set forth in claim 64

wherein step (f) includes the step of:

(g) providing a latent inhibitor in the liquid medium capable of being activated by a source of inhibitor activating radiation which does not cause solidification of the liquid medium;

(h) directing the source of inhibitor activating radiation toward the liquid medium to be solidified in order to activate a first sublayer of the inhibitor adjacent the surface of the liquid medium without the inhibitor substantially impairing penetration into a second sublayer adjacent the first sublayer;

(i) solidifying at least part of the two dimensional layer of the liquid medium which comprises the second sublayer.

66. A method as set forth in claim 30

wherein steps (b) , (c) , and (d) include the step of:

(f) projecting onto the liquid medium a plurality of successive images having the size and shape of a plurality of successive subregions of that part of the two dimensional layer of the liquid medium to be solidified.

67. A method as set forth in claim 66

wherein step (f) includes the step of:

(h) controlling exposure from the source of synergistic 64 stimulation onto that part of the two dimensional layer of the liquid medium to be solidified at each of the successive locations.

68. A method as set forth in claim 66

wherein step (g) includes the steps of:

(h) providing imaging optics intermediate the source of synergistic stimulation and the surface of the liquid medium; and

(i) operating the imaging optics to project onto the liquid medium an image having the size and shape of another subregion of that part of the two dimensional layer of the liquid medium to be solidified.

69. A method as set forth in claim 66

wherein the programmable mask comprises a device capable of rapidly modulating the projected intensity of a plurality of pixels;

wherein step (f) includes the steps of:

(g) moving the plurality of successive images to a plurality of successive locations, respectively, on that part of the two dimensional layer of the liquid medium to be solidified in increments smaller than a dimension of a pixel.

70. A method as set forth in claim 33

wherein step (f) includes the steps of:

(g) providing imaging optics intermediate the source of synergistic stimulation and the surface of the liquid 65 medium; and

(h) maintaining the surface of the liquid medium at a focal plane which is a substantially constant distance from the imaging optics;

(i) supporting the object being constructed on a platform which is initially at least partially submerged in the liquid medium;

(j) lowering the platform to an operating depth such that the uppermost surface of the object being formed is below the surface of the liquid medium to a distance which is substantially equal to the thickness of that part of the next succeeding two dimensional layer to be solidified; and

(k) retaining the platform at the operating depth until the liquid medium completely overlies the entire uppermost surface of the object being formed.

71. A method as set forth in claim 70 including the step of:

72. A method as set forth in claim 50

wherein step (h) includes the steps of:

(i) sensing changes from the normal height of the surface of the liquid medium relative to the imaging optics; 66

(j) replenishing the liquid medium in the reservoir when the height of the surface of the liquid medium falls below the normal height as detected in step (i) ; and

(k) withdrawing excess liquid medium from the reservoir when the height of the surface of the liquid medium tends to increase above the normal height as detected in step (i) .

73. A method as set forth in claim 31

wherein step (f) includes the steps of:

(h) maintaining the surface of the liquid medium at a focal plane which is a substantially' constant distance from the imaging optics;

(i) supporting the object being constructed on a platform which is initially at least partially submerged in the liquid medium; and

(j) lowering the platform to an operating depth such that the uppermost surface of the object being formed is below the surface of the liquid medium to a distance which is substantially equal to the thickness of that part of the next succeeding two dimensional layer to be solidified.

74. A method as set forth in claim 73

wherein step (h) includes the steps of:

(k) providing a displacement mass initially submerged in the liquid medium; and

(1) withdrawing the displacement mass from the liquid medium at a mass-rate commensurate with the mass-rate of the platform resulting from step (j) .

75. A method as set forth in claim 73

wherein step (h) includes the steps of:

(k) sensing changes from the normal height of the surface of the liquid medium relative to the imaging optics;

(1) providing a displacement mass initially submerged in the liquid medium; and

(m) withdrawing the displacement mass from the liquid medium at a mass-rate commensurate with changes sensed in step (k) to accommodate changes in displacement of the platform resulting from step (j).

76. A method as set forth in claim 31

(i) raising the platform until an uppermost surface of the object being formed rises above the surface of the 68 liquid medium such that the liquid medium overlying the uppermost surface of the object is thicker at the central regions than at the periphery thereof;

(k) before again performing steps (b) and (e) , allowing the layer of the liquid medium completely overlying the entire upper surface of the object being formed to come to equilibrium so that it has a substantially uniform thickness across the entire upper surface of the object.

77. A method as set forth in claim 31

wherein step (f) includes the steps of:

(h) lowering the platform to an operating depth such that the uppermost surface of the object being formed is below the surface of the liquid medium to a distance which is substantially greater than the thickness of that part of the next succeeding two dimensional layer to be solidified; and

(i) raising the platform until an uppermost surface of the object being formed rises above the surface of the liquid medium such that the liquid medium overlying the uppermost surface of the object is thicker at the central regions than at the periphery thereof; and 69

(j) drawing a doctor blade across the upper surface of the object being formed and parallel thereto and at a uniform distance therefrom in multiple passes successively closer to the object, the sum total of the distance traversed in the multiple passes being substantially equal to the thickness of that part of the next succeeding two dimensional layer to be solidified thereby removing excess amounts of the liquid medium from the upper surface of the object.

78. A method as set forth in claim 77 wherein step (f) includes the steps of:

(k) lowering the platform to an operating depth such that the uppermost surface of the object being formed is below the surface of the liquid medium to a distance which is substantially equal to the thickness of that part of the next succeeding two dimensional layer to be solidified; and