CHAPTER 3

HISTORICAL PERSPECTIVE

JOSEPH J. BEAMAN

TECHNOLOGY

The early roots of rapid mechanical prototyping technology can be traced to at least two technical areas: topography and photosculpture.

Topography

As early as 1890, Blanther (1892) suggested a layered method for making a mold for topographical relief maps. The method consists of impressing topographical contour lines on a series of wax plates, cutting the wax plates on the contour lines, and then stacking and smoothing the wax sections. This produces both positive and negative three-dimensional surfaces that correspond to the terrain indicated by the contour lines. After suitable backing of these surfaces, a printed paper map is then pressed between the positive and negative forms to create a raised relief map. This is shown in Fig. 3.1.


Fig. 3.1. Layered mold relief map proposed by Blanther (1892).

Perera (1940) proposed a similar method for making a relief map by cutting contour lines on cardboard sheets and then stacking and pasting these sheets to form a three-dimensional map. Further refinements of this approach were made by Zang (1964), who suggested using transparent plates with topographical detail inscribed on each plate, and Gaskin (1973), who described a three-dimensional geological teaching device. In 1972, Matsubara of Mitsubishi Motors (1974) proposed a topographical process that uses photo-hardening materials. In this process, a photopolymer resin is coated onto refractory particles (e.g., graphite powder or sand), which are then spread into a layer and heated to form a coherent sheet. Light (e.g., from a mercury vapor lamp) is selectively projected or scanned onto this sheet to harden a defined portion of it. The unscanned, unhardened portion is dissolved away by a solvent. The thin layers formed in this way are subsequently stacked together to form a casting mold. In 1974, DiMatteo (1976) recognized that these same stacking techniques could be used to produce surfaces that are particularly difficult to fabricate by standard machining operations. Examples he mentions include propellers, air foils, three-dimensional cams, and forming of dies for punch presses. In one embodiment (Fig. 3.2), contoured metallic sheets are formed by a milling cutter, then joined in layered fashion by adhesion, bolts, or tapered rods. This process has obvious similarity to the earlier 19th century work.


Fig. 3.2. Layered mold of stacked sheets by DiMatteo (1974).

In 1979, Professor Nakagawa of Tokyo University began to use lamination techniques to produce actual tools such as blanking tools (Nakagawa et al. 1979), press forming tools (Kunieda and Nakagawa 1984), and injection molding tools (Nakagawa, Kunieda, and Liu 1985). Of particular note, Nakagawa mentions the possibility of complex cooling channels in injection molds (Nakagawa, Kunieda, and Liu 1985).

Photosculpture

Photosculpture arose in the 19th century in attempts to create exact three-dimensional replicas of objects, including human forms (Bogart 1979). One somewhat successful realization of this technology was designed by Frenchman François Willème in 1860. In his method, shown in Fig. 3.3, a subject or object was placed in a circular room and simultaneously photographed by 24 cameras placed equally about the circumference of the room. The silhouette of each photograph was then used by an artisan in Willème's studio (Fig. 3.4) to carve out 1/24th of a cylindrical portion of the figure.

In an attempt to alleviate the labor-intensive carving step of Willème's photosculpture, Baese (1904) described a technique using graduated light to expose photosensitive gelatin, which expands in proportion to exposure when treated with water. Annular rings of the treated gelatin are then fixed on a support to make a replica of an object, as shown in Fig. 3.5. Similar techniques and improvements were developed by Monteah (1924).

In some of the earliest work in Japan, Morioka (1935, 1944) developed a hybrid process combining aspects of photosculpture and topography. This method (Fig. 3.6) uses structured light (black and white bands of light) to photographically create contour lines of an object. The lines can then be developed into sheets and cut and stacked, or projected onto stock material for carving.


Fig. 3.3. Admiral Farragut sits, late 1860s, for photosculpture
(Bogart 1979; photo courtesy of George Eastman House).


Fig. 3.4. François Willème's photosculpturing studio in Paris, about 1870
(Bogart 1979; photo courtesy of George Eastman House).


Fig. 3.5. Photographic process for the development of plastic objects by Baese (1904).


Fig 3.6. Process for manufacturing a relief by Morioka (Morioka 1935, 1944).

In 1951, Munz (1956) proposed a system that has features of current stereolithography techniques (Fig. 3.7). He disclosed a system for selectively exposing a transparent photo emulsion in a layerwise fashion, where each layer comes from a cross-section of a scanned object. These layers are created by lowering a piston in a cylinder and adding appropriate amounts of photo emulsion and fixing agent. After exposing and fixing, the resulting solid transparent cylinder contains an image of the object. Subsequently this object can be manually carved or photochemically etched out to create a three-dimensional object.


Fig. 3.7. Process by Munz (1956) to reproduce a three-dimensional image of an object.

Early Solid Freeform Fabrication

In 1968, Swainson (1977) proposed a process to directly fabricate a plastic pattern by selective three-dimensional polymerization of a photosensitive polymer at the intersection of two laser beams. Parallel work was conducted at Battelle Laboratories (Schwerzel 1984). The essential features of this process, termed photochemical machining, are depicted in Fig. 3.8. The object is formed by either photochemically cross-linking or degrading a polymer by simultaneous exposure to intersecting laser beams. Although laboratory hardware was constructed for this process, a commercially viable process was apparently not achieved.


Fig. 3.8. Photosculpture process using intersecting laser beams, by Swainson (1977).

A powder process that has more in common with laser surface cladding techniques than photosculpture was proposed in 1971 by Ciraud (1972). This disclosure describes a process for the manufacture of objects from a variety of materials that are at least partially meltable. In order to produce an object, small particles are applied to a matrix by gravity, magnetostatics, or electrostatics, or positioned by a nozzle located near the matrix. The particles are then heated locally by a laser, electron beam, or plasma beam. As a consequence of heating, the particles adhere to each other to form a continuous layer. As Fig. 3.9 shows, more than one laser beam can be used in order to increase the strength of the union between the particles.


Fig. 3.9. Powder laser process proposed by Ciraud (1972).

Hideo Kodama of Nagoya Municipal Industrial Research Institute was the first to publish an account of a functional photopolymer rapid prototyping system (Kodama 1981). In his method, a solid model is fabricated by building up a part in layers, where exposed areas correspond to a cross-section in the model. He studied three different methods for achieving this (Fig. 3.10):


a.
using a mask to control exposure of the UV source and immersing the model downward into a liquid photopolymer vat in order to create new layers
b.
using a mask as in (1), but positioning the mask and exposure on the bottom of the vat and drawing the model upward to create a new layer
c.
immersing the model as in (1), but using an x-y plotter and an optical fiber to expose the new layer

Fig. 3.10. Schematics of three photopolymer systems studied by Kodama (1981).

A second, parallel but independent, effort was conducted by Herbert at 3M Corporation (1982). Herbert describes a system that directs a UV laser beam to a photopolymer layer by means of a mirror system on an x-y plotter (Fig. 3.11). In Herbert's experimental technique, a computer is used to command a laser beam across a layer, the photopolymer vessel is then lowered (~ 1 mm), and additional liquid photopolymer is then added to create a new layer.


Fig. 3.11. Herbert's photopolymer process (1982).

Active Patents

Besides those described above, there are numerous active patents that cover existing commercial processes. Table 3.1 lists the most prominent patents.

Table 3.1
Active Rapid Prototyping Patents

Early Parts

Although very intricate parts produced by rapid prototyping equipment are now common, the first parts out of these types of systems required a good deal of faith that improvements would occur. Shown in Fig. 3.12 are three early parts from different systems. The Housholder part was made from an embodiment that included a grid for separating mold material (concrete and water) from casting material (dry concrete). The Herbert part was created in August 1979. It is not known exactly when the Kodama and Housholder parts were created.


Fig. 3.12. Three early rapid prototyping parts; from left to right by Kodama, Herbert, and Housholder.
[Previous Section][Top of Report][Send Your Comments][WTEC Welcome Page][Next Section]

Published: March 1997; WTEC Hyper-Librarian