Professional Documents
Culture Documents
Rapid Prototyping
Rapid Prototyping
Rapid Prototyping
In recent years the competitiveness of manufactured products in the world market has
intensified tremendously. In addition mass customization, accelerated product
obsolescence and continued demands for cost savings are forcing companies to look for
new ways to improve their business processes. Fast and cost effective development of high
quality products with a swift response for market demands regarding rapid changes in
design and functionality is the requirement of today’s global economies. It is extremely
important for new products to reach the market as early as possible and that too before the
competitors. To achieve this, the processes involved in the design, test, manufacture and
market of the products require to be squeezed, both in terms of time and material
resources. This therefore requires new tools and approaches to tackle these issues, and
many of these tools and approaches have already evolved or are being evolved. These are
mainly technologically-driven, usually involving computers. The rapid development and
advancement in such technologies over the last few decades has made it possible to
achieve such results.
In product development, the pressure of time is a major factor that drives the direction of
development and success of new methodologies and technologies for enhancing its
performance [1]. To meet these challenges of decreased time to market for mass products
and decreased delivery times for low production parts a variety of measures can be taken. In
response to these challenges, the industry and academia have worked on a number of
technologies that help to develop new products and broaden the number of product
alternatives. Examples of these technologies include feature-based design, design for
1
manufacturability analysis, simulation, computational prototyping, and virtual and physical
prototyping.
Rapid Prototyping (RP) is an early design exploration and validation process, in which
executable models of a system are created that reflect some subset of properties of interest
[2]. It is a process that automatically creates a physical prototype from a three-dimensional
computer aided design (CAD) model, in a short period of time [3,4]. It is an indispensable
tool for shortening product design and development time cycles [5,6]. Furthermore, it is a
powerful communication tool that bridges design, marketing, process planning and
manufacturing, and can facilitate the implementation of concurrent engineering [7,8]. Most
designers are of the idea that “getting physical fast” is critical in exploring novel design
concepts. The sooner designers experiment with new products, the faster they gain
inspiration for further design changes. The evaluation of a design is required so as to
quickly uncover design errors or undesirable properties and this helps refine the
requirements. The main idea is to quickly create prototypes to reveal any shortcomings in
the product and with less effort than required for full system implementation and
production. If prototype models are available for testing and visualization the time needed
to design a product can considerably decrease.
Since RP refers to a class of technologies that can automatically construct physical models
from Computer-Aided Design (CAD) data. Therefore CAD/CAM systems are required that
can quickly produce physical objects directly from CAD models. The main idea is to
decompose the 3-D computer models into thin cross-sectional layers, followed by
physically forming the layers and stacking them up “layer by layer”. The basic RP process
chain is presented in figure 1.1.
The idea of creating 3D objects in a layered fashion is almost as old as human civilization.
The earliest constructions such as the Egyptian pyramids were built block by block, layer
by layer. Stacking up of individually shaped material layers also has a long tradition in a
number of manufacturing applications such as tape casting and shape melting [9]. In the
present form RP has its beginning in the mid-1980s with the debut of 3D Systems Inc.’s
(Valencia, CA) SLA-1 [8]. In its early days of development, the automotive and aerospace
industries were the leading users of RP applications [10]. This
2
CAD
3-D Modelling
STL file conversion
RP Fabrication
Layer-by-layer
buildup
Post Processing
Cleaning, hardening
Finishing
3D Physical Part
3
however, is no longer the case as the use of RP has spread into many other industries. In
the frame of research a few space programs have investigated the possibility to use RP and
rapid manufacturing techniques during space instrument development [11]. At present
there are about 28 manufacturers worldwide offering a total of more than 56 different RP
systems to meet the diverse demands of end-users [12]. This shows how fast the RP
technology is developing.
In the additive and/or incremental processes, thin layers with distinct shape are
stacked one upon other to build the models. The shaping method however, of each
layer varies for different processes. Most of the commercial RP systems belong to this
category. Such processes are also called layered manufacturing (LM) or solid
freeform fabrication (SFF). LM implies that the artifacts are built in layers which
greatly simplifies the processes and enables their automation. Dutta et al. [13] has
given a comprehensive review of LM processes and related process planning issues. An
important feature in LM is the raw material, which is either one-dimensional (e.g.
liquid and particles) or two-dimensional (e.g. paper sheet) stocks. This is in contrast
with the three-dimensional raw material stocks in case of subtractive RP processes.
Stereolithography apparatus (SLA), selective laser sintering (SLS), three dimensional
printing, fused deposition modeling (FDM), contour crafting (CC) etc. are few
examples of LM.
In the subtractive or material removal (MR) processes models are build by cutting of
excessive material from the raw material stocks. There are not as many subtractive
prototyping processes as that of additive processes. A good example of the
subtractive process is the sculpturing robot (SR) system at the Delft University of
Technology [14]. A commercially available system is DeskProto [15], which is a
three-dimensional computer aided manufacture (CAM) software package for RP and
manufacturing. Another commercially available system is the Roland's SRP process
that allows designers to create precise, smooth prototypes from a wide
4
choice of non-proprietary, inexpensive plastics [16]. The major feature of pure
subtractive RP processes is that a model is made from a single stock, thus fully
compact parts of the same material as per actually required for end use is possible.
Hence the idea of form fit and functional parts is possible. The added advantages are
the accuracy of the part dimensions and the surface quality that can be achieved by
the subtractive machining approach. However the MR processes are limited in
geometric complexity as compared to the LM processes. The cutting methods used
include computer numerical control (CNC) milling, water-jet cutting, laser cutting
and the Millit [17] package.
The pros and cons of LM and MR processes are quite obvious. The constraint of the
geometric complexity of models is relaxed to a significant extent due to the layer by
layer buildup in LM. Many features (such as deep cavities and freeform internal
surfaces), which are difficult or not possible to fabricate by conventional MR
processes, can be easily built by LM. Thus additive processes have great potential for
hands-off free form production of very complicated parts. However the geometric
accuracy and surface quality of the part built in LM is far less than that in MR. One of
the main error sources arises due to the staircase effect: a model made by LM is a
zeroth-order approximation of the design model. The current LM processes are also
restricted due to the limited build envelope and choices of raw materials. To obtain
the benefits of both the LM and MR processes, there have been increasing efforts in
the recent years for the integration of the two. This integration leads to a hybrid RP
system in which the staircase phenomena can be eliminated and a better surface
quality can be acquired without losing the manufacturability for complex features.
The common ground of the hybrid RP systems is to build models in additive ways,
while shaping each layer in subtractive ways. Thus complex geometries can be
produced due to layered buildup and accuracy and surface quality achieved due
intermittent machining. Typically such systems include shape deposition
manufacturing (SDM) [19], Computer-aided manufacture (CAM) of
5
laminated engineering materials (LEMs) [20], laminated object manufacturing or
cured at temperatures which do not (LOM) [21], hybrid plasma deposition and milling
(HPDM) [22,23], Paper lamination technology (PLT), Solid ground curing (SGC),
Solidscape and the solvent welding freeform fabrication technique (SWIFT) [24].
In most additive SFF processes three dimensional shapes of arbitrarily complex geometries
are created by incremental material deposition of 2 1/2 dimensional, cross-sectional layers
embedded in complementary shaped, sacrificial support material (as shown in figure 1.2 for
the case of SDM). This support material is required during buildup to accommodate the
geometries with overhanging features. In addition these are useful for delicate features such
as islands, internal cavities and thin-walled sections. Since pieces of a part can only be
located by being built in contact with previously deposited material or the support platform,
special procedures are required to build layers with islands or overhanging features. In
these cases, special support structures are built upward from the platform to support these
objects. The previous layer, embedded in the support material, provides the necessary flat
surface for the overhangs, thus allowing manufacturing without any change in strategy.
This approach facilitates the automatic planning and execution of fabrication by eliminating
the need for part specific tooling or fixturing. The support structures are designed to be only
as large or as strong as necessary for support since they are removed once the part is
completed.
In some SFF processes the support material is the same as part material which remains
unused / unprocessed during the buildup (e.g. as in SLS, LOM etc.). However, in other
processes the two materials are different (e.g. as in SDM, FDM etc.). In the second
category as each part is encased in a sacrificial support material the process has to deal with
more than one material even for homogeneous parts. The basic requirement is the
compatibility of part and support materials, subject to the constraints of the deposition and
removal processes. Part and support materials must be physically and chemically
compatible. These materials must be deposited or cured at temperatures which do not
damage previously shaped features. Also the part and support materials must have minimal
intersolubility, and neither material may inhibit curing processes of the other
6
Deposit Remove Part with Support Remove Support
Material Material
7
[25]. There must be a suitable degree of adhesion between the two materials without the
formation of any defects, e.g. internal voids.
In the industry several RT technologies are now commonly available. Some of these
technologies produce the tool directly from the RP process. However, the majority of RT
8
technologies use the model created by the RP process in a secondary process to produce
the tool.
A variety of tooling can currently be produced using different RP technologies. The tooling
is basically classified as hard or soft tooling and also as direct or indirect tooling [34]. For
short manufacturing runs the tooling used is often known as soft tooling. As the name
suggests these tools are often made from softer materials such as silicon rubber, epoxy
resins, low-melting-point alloys, or aluminium, which are easier to work with than tooling
steels. On the other hand the tooling for longer manufacturing runs is known as hard
tooling and is usually made of hard tooling steels. When the tool or the die is created directly
by the RP process it is known as direct tooling. In injection moulding, for example, the RP
process is used to produce the core and cavity, together with the gating and ejection
system. On the other hand in the indirect tooling, the RP technology is only used to create
the master. This master is used to make a mould out of a material such as silicone rubber,
epoxy resin, soft metal, or ceramic.
RP parts produced till to date are mostly used for prototyping or tooling purposes;
however, in the coming years majority of these may be produced as end products. In this
context the term ‘rapid manufacturing’ (RM) is introduced in which RP technologies are
used as processes for the production of end products. In this everyday changing world
there is an increased and ongoing demand for new and improved products. This demand
derives the way for newer versions of the product with lesser product development time
and therefore smaller production volumes. The application of RP technologies for
production rapid manufacturing may therefore grow considerably as the production
volumes decrease. The RM operation can greatly improve the competitive position of
companies adopting it. It is a natural extension of RP. The key enabling technologies of
rapid manufacturing are RP and RT [35]. RM will never completely replace the
conventional manufacturing techniques, especially in large production runs where mass-
production is more economical. However, due to no tooling requirements for short
production runs, RM is much cheaper [36]. It also allows geometric freedoms such as
variable wall thickness. Furthermore, manufacturing without tooling also permits
distributed manufacture so that parts may be made in or near the location where they are
9
required. An example of this may be seen by NASA’s adoption of fused deposition
modeling (FDM) to make spare parts on the international space station [37].
Most systems developed so far mainly have a single function and have one material on one
stage. A new concept of functional prototype development (FPD) is newly proposed that
provides the necessary prototype functions such as mechanical, optical, chemical and
electrical properties in order to meet the broad requirements of the industry [38].
1.3 Ra pi d P r oto ty p i ng Pr os an d Co ns
1.3.1 Advantages of RP
RP implies that complex shapes are as easy to build as simple shapes, since the planning
and manufacturing processes are automated. RP machines are actually "three dimensional
printers" that allow designers to quickly create tangible prototypes of their designs, rather
than just two-dimensional pictures. The main benefits of RP are:
Rapidly Prototyped parts show great time, cost and material savings.
10
Minimize time consuming discussions and evaluations of manufacturing
possibilities.
Minimize time and cost for design, manufacturing and verification of tooling. Since
RP encourages creation of detailed design at the b eg i n n i ng of a product’s
development. This can lead to a more usable product in a shorter period of time. In
fact, development teams that prototype, require 45% less effort to produce systems
than development teams that did not prototype [39].
Reduce the labor content of manufacturing, since part specific setting up and
programming are eliminated, machining/casting labor is reduced, and inspection
and assembly are also consequently reduced, however more manual finishing is
required for the RP parts.
Allows early customer involvement. RP parts make excellent visual aids for
communicating ideas with co-workers or customers. There is better communication
between the user and designer as the requirements and expectations are expressed in
the beginning itself.
Research has shown that the total costs for new products can be reduced by as much
as 30 to 60% [40]. Therefore consumer can buy at lower prices, since the
manufacturers savings will ultimately be passed on.
1.3.2 Disadvantages of RP
Rapid prototyping also poses some special challenges that are usually worth overcoming.
Some of the disadvantages of rapid prototyping are described below.
Some people are of the opinion that rapid prototyping is not an effective model of
instructional design because it does not replicate the real thing. It could so happen
11
that many important steps of instructional design are forfeited for a faster, cheaper
model.
Many problems may be overlooked that results in endless rectification and revision.
The user may have very high expectations about the prototype’s performance and
the designer is unable to deliver these.
The system could be left unfinished due to various reasons or the system may be
implemented before it is completely ready.
The producer may produce an inadequate system that is unable to meet the overall
demands of the organization.
Too much involvement of the user might hamper the optimization of the program.
St er eolithography ( SL)
Patented in 1986, stereolithography started the revolution in rapid prototyping. This process
is based on a photosensitive liquid resin which when exposed to ultraviolet (UV)
12
light solidifies and forms a polymer. SL systems consist of a build platform (substrate)
which is mounted in a vat of resin and a UV helium– cadmium or argon ion laser [44]. The
information obtained from the three-dimensional solid CAD model is used by the laser
system to image the first layer of the part on the resin surface. Once the layer is hatched
after scanning the contour, the platform is lowered and a new layer of resin is applied. The
next layer is then scanned. After the completion of the part, it is removed from the vat, the
excess resin drained and supports are broken off. The ‘green’ part is then placed in a UV
oven for post-curing. The SL approach is depicted in figure 1.3.
Stereolithography Apparatus (SLA) machines have been made since 1988 by 3D Systems of
Valencia, CA [45, 46]. Since it was the first technique, SL is regarded as a benchmark by
which other technologies are judged. In the beginning SL prototypes were fairly brittle and
prone to curing-induced warpage and distortion, but recent improvements have largely
corrected these problems. To broaden the application area of SL, research and technology
development efforts are being directed towards process optimization [47, 48].
1. 4.1 .1 .2 So l i d g r ou n d cu r i n g ( SGC )
This system again utilizes photopolymerizing resins and UV light. This is a complicated
process developed by the Cubital Corporation in which thermoset part material is imaged
using an ultraviolet light to form layers of a part [49, 50]. To build a layer, a thin coating of
ultraviolet-curable photopolymer is spread over the bottom of the build chamber. An
electrostatically-charged roller is used to apply toner to a mask. This will shield all the
undesired areas of photopolymer. The mask is then placed over the resin surface and the
entire layer is illuminated with a powerful UV lamp to cure the desired portion of this
cross-section of the part. The uncured photopolymer is then removed and hot wax is filled
in areas where cured part material is not present. A cooled platen is then used to solidify
the wax, which acts as a support material for undercut features. Finally, the unwanted wax
and photopolymer on the top of the layer is planed by a milling cutter to a desired height.
This cycle is repeated for each layer to ultimately build the entire part. The part is taken out
from the chamber at the end of this process, and the support material is removed manually
by heating. The company plans to develop a reverse-cycle process,
13
Beam delivery
(e.g., galvo-mirrors)
Laser
Elevator
Recoating
mechanism
Solider
Cubital, Inc. (Israel/Germany)
14
in which photopolymers would be used as support materials while other polymer, metallic,
or ceramic materials are used as part materials [51].
A distinct feature of this technique is the capability to produce numerous parts in a timely
fashion in a single batch. Because the processing time for each layer is independent of the
part size or geometry, multiple parts packed into a single batch (e.g., the four replicates
next to each other in figure 1.4) can be fabricated in the same time as is required to build a
single part. This reduces the average cost and time required to build parts. This machine was
a very complicated rapid prototyping machine, with a reputation of requiring a lot of
maintenance [52]. However this machine is not available commercially since 1999.
In place of layer by layer solidification, a holographic image is projected into the resin,
causing an entire surface to solidify. The imaging data are still obtained from the CAD
model, although not in the form of slices [41]. There are no commercial systems available
yet.
15
1.4.1.1.6 Objet Quadra process.
This technique employs 1536 nozzles for spreading layers of photosensitive resin while
building a part. After spreading the resin each layer is cured by using two UV lights. The
intensity of the lights and the exposure are controlled in such a way that the parts produced
do not require post-curing. A second material is also deposited for support overhanging
areas and undercuts that can be easily separated later on from the model [54].
A conductive material such as aluminium is used as electrodes on to which the layer cross-
section is printed. After printing of all the layers, they are stacked onto each other,
immersed in a bath of electrosetting fluid and energized. As a result the fluid in between
the electrodes solidifies to form the part. The composite is then removed and drained. The
unwanted aluminium is trimmed from the part. This technology is advantageous due to the
control available over certain material properties. This is done by adjusting the voltage and
current applied to the aluminium electrodes which may result in the control of part density,
compressibility, hardness and adhesion. Silicon rubber, polyester, polyurethane or epoxy
parts can be made by this technique. The hardware for such a system may be inexpensively
bought off the shelf [55].
16
Figure 1.5: Ballistic Particle Manufacturing BPM, Inc. (U.S.)
17
the support structures were made from the same material. Commercial systems based on
this process were available until 1998, which has since gone out of business.
In this technique, filaments of heated thermoplastic are extruded from a tip that moves in
the x-y plane [60] (figure 1.6). The extrusion head deposits very thin beads of material in a
controlled fashion onto the build platform (like a baker decorating a cake) to form a layer.
The material is heated just above its melting point so that it solidifies immediately after
extrusion and cold-welds to the previous layers [61]. The platform is kept at a lower
temperature, so that the thermoplastic quickly becomes firm. The deposited layers can be as
fine as 0.005 inch (0.127 mm) thick. After deposition of a layer the platform lowers and the
extrusion head deposits the next layer on the previous. Supports are built alongside the
deposition where required. These are fastened to the part either with a second weaker
material or with a perforated junction.
1.4.1.3.3Three-dimensional welding ( 3D W)
This technique is still in its development stages. Molten metal is sprayed on to a substrate
in a near-net shape and then unwanted material is removed via numerically controlled (NC)
operations [63, 64]. The support material is added in the same way; however this is done
either before or after the deposition of the prototype material. This is done
18
depending on the fact whether the layer contains undercut features or otherwise (Figure
1.2). If the features of the layer are complex, it may be required to add the support material
both before and after the prototype material. To remove residual stresses, each layer is
shot-peened. The process is then repeated layer after layer. The prototype is shifted from
one station to another using a robotized pallet system. Typically stainless steel parts have
been produced by this technique using copper as support material. Copper can be removed
afterwards by immersion in nitric acid. The prototypes produced by this technique have a
structure similar to cast or welded parts and the accuracy of NC milled components.
Selective laser melting is a generative fabrication process for the manufacture of metallic
components (Figure 1.7). The parts are built layer by layer with the metal powder being
melted locally by an intensive laser beam that traces the layer geometry. The layerwise
design may be compared in principle with that of stereolithography. On a building
platform, a thin layer of metal powder is applied, which is locally melted by the laser beam
energy. The building platform is lowered by a defined layer height, coated with the metal
powder, and melted again. The process is repeated until the component is completed.
The process builds up to 100% dense metal parts direct from design data using customary
metal powder. Very fine details are achievable such as thin vertical walls of less than 100
µm thickness. Metals like stainless Steel, Titanium, Tool Steel, Cobalt Chrome and Inconel
can all be processed. Unlike SLS, the SLM process is hard to control, due to the large
thermal stresses and the presence of a liquid pool around the laser spot. Recently Elsen
et.al. made an effort to describe the SLM process by making the parameters dimensionless
which helped in comparing the results of different research groups [65]. SLM technology is
commercially available from Staffordshire based MCP Tooling Technologies [66].
Electron Beam Melting (EBM), patented by Arcam AB (Mölndal, Sweden) [67] and
distributed in the U.S. by Stratasys, Inc. (Eden Prairie, MN) offers another method for
19
Figure 1.7: Selective Laser Melting (SLM).
Electron Beam
Powder Container
Vacuum Chamber
Building Table
20
rapid manufacturing and prototyping metal components (Figure 1 .8).It can produce fully-
dense metal components with properties similar to or better than that of wrought materials
[68]. In this process, powder metal is spread over a vertically adjustable surface. A
computer guided electron beam then traces the cross section of the modeled part, first
heating then melting and forming the first layer. The surface is then lowered and the
process repeated for each successive layer, forming the three-dimensional object modeled.
The process takes place in a vacuum to prevent loss of energy that would be caused by the
fast-moving electrons colliding with air or gas molecules. Added advantages of vacuum
are clean environment (resulting in excellent material characteristics) and higher energy
efficiency (lower operating cost and faster build times—five to 10 times greater that laser
technology.
21
Molten
plastic
(a) (b)
Figure 1.9: (a) Multi-Jet Modeling 3D Systems, Inc. (U.S.) (b) Multi jet modeling head
22
Processes involving discrete particles
Fusing particles by laser
Selective laser sintering (SLS)
Selective laser sintering (SLS) is a process developed at the University of Texas. It is an
indirect process for manufacturing metal parts and molds. It was initially available
commercially from DTM Corporation but was later bought over by 3D Systems in 2001. In
SLS a fine powder is heated with a CO2 laser which causes sintering of the powder particles
and as a result they are mutually bound. The heating is done to a level so that the surface
tension of the particles is overcome and they fuse together. In order to minimize thermal
distortion and facilitate fusion to the previous layer the entire bed is heated to a temperature
just below the melting point of the material and this is done prior to the commencement of
sintering [71]. The laser is modulated in such a way that only the grains which are directly
exposed to the beam are affected. After the laser beam has scanned a layer on the powder
bed, the bed is lowered. The powder-feed cartridge is then raised and the counter-rotating
roller evenly spreads the next layer of powder over the build area. The sintered material
forms the part while the un-sintered powder remains in tact as support structure which is
cleaned away and recycled after the completion of the part (Figure 1.10). The laser
sintering technology (LST), is another process based on the same physical principles.
Presently such dual laser systems are available for the processing of thermoplastics and
sand. To widen the range of applications of SLS and LST a significant amount of
development efforts have been directed towards process optimization [44,71–74].
To receive fully dense ceramic parts a French company ‘Phenix’ has developed a system
that completely melts the powder. The so called PM 250 is designed as a high temperature
laser sintering device with a fibre laser. According to the company all ceramic powders
(and even metallic powders) can be processed. Ceramic powders must be sintered in a
subsequent oven process.
This process involves simultaneous feeding of powder, through a nozzle, and fusing it with
a laser on the part bed (Figure 1.11) [75]. The nozzle can be on a side of the bed or
23
Z Axis Positioning
of Focusing Lens
Figure 1.12: Three Dimensional Printing MIT, and Soligen, Inc. (U.S.)
Figure 1.13: Multiphase Jet Solidification Fraunhofer Institutes IFAM and IPA (Germany)
24
coaxial with the laser beam. If it is on a side, the solidified sections cause shadowing areas
for the sections to be built. To prevent this, a constant orientation to the part creation
direction must be maintained. If the powder feeder is coaxial, inaccuracies may occur in the
geometry of the part and the layer thickness if the beam and the powder feeder move out of
alignment. Some other systems based on the same principle have also been developed e.g.
direct metal deposition (DMD) [76] (commercially available by POM and Trumpf) and
AeroMet laser additive manufacturing [77]. Another technique uses wire based alloy in
place of metal powder [78]. In a latest attempt laser is used in conjunction with plasma,
thus a new direct metal prototyping technology is developed by using hybrid plasma-laser
deposition manufacturing (PLDM) for obtaining near net shape full density and good
surface intractable high-temperature alloy prototype [79].
In gas phase deposition process, a laser is used to produce a solid by decomposing the
molecules of a reactive gas [44]. The resulting solid adheres to a substrate to form the part.
Three slightly different methods to construct a part have been investigated. The first
method is known as the selective area laser deposition (SALD). In this technique only the
solid component of the decomposed gas is used to form a part. It is possible by the SALD
technique to produce parts made of carbon, silicon, carbides and silicon nitrides. In the
second method called the laser-assisted chemical vapour deposition (LCVD), a thin
covering of powder is spread in each layer and the spaces between the powder grains are
filled in by the decomposed solids. The third method is called the selective laser reactive
sintering (SLRS). In this technique the laser initiates a reaction between the gas and the
spread layer of powder. As a result a solid part of silicon carbide or silicon nitride is
formed. As yet there are no GPD based systems available commercially.
This technique refers to an entire range of machines that employ ink-jet methodology. The
first equipment of its type (Three-Dimensional Printing or 3DP) [80] was developed at MIT
and licensed to Soligen Corporation, Extrude Hone, and others (figure 1.12). In 3DP a layer
of powder is spread onto a substrate and the cross-sectional pattern of the
25
layer is scanned by spraying a binder material into the powder using ink jets. Another layer
of powder is then added onto the preceding one and the process is repeated. The
unbounded powder remains in its position and acts as a support material. Upon completion
the unbounded powder is removed and the “green” part is sintered. In best cases the
porosity of the “green” and debindered parts lays between 20 and 30% [81]. Typical
powders used are ceramic powders such as aluminum oxide, silica, zirconia and silicon
carbide. Experiments with stainless steel resulted in a low density of the part (around 78%)
and significant shrinkage during the sintering process [82]. Ceramic cores and shells
produced by 3DP are successfully being used for investment casting. Soligen, Inc. has
commercialized the process for ceramic core and mold applications under the trade name
Direct Shell Production Casting. The Z Corp. 3D printing is the fastest additive technology
commercially available on the market. In place of single-jet technology it uses inkjet print
heads with a resolution of 600 dpi (dots per inch), that focuses on a drop-on-demand
approach [83].
26
visible light on to the photocurable slurry [86]. DPS based system is commercialized by
EPFL in Switzerland.
1 . 4 . 1 . 4 . 2 . 4 S P A T I AL F O R M I N G ( S F)
This technology was developed for prototyping specialized, high precision, metallic,
medical equipment within a small build envelop of 2 x 2 x 300 mm. It was conceived and
demonstrated as a method of manufacturing parts for cardiac catheter systems [87]. The
process combines several technologies to produce solid metallic microstructures from fine
powder. Cross section data from computer solid models are used for patterning of a chrome
mask which images a lithographic printing plate like those used in the publishing industry.
An offset printing press is then used to print “negative” material (the space around the
parts) with ceramic pigmented organic ink (averaging 0.5 µm thick) on a ceramic substrate
in multiple registered layers. Each layer is successively cured with UV light. After a period
of approximately 30 layers, the positive space (non-image voids) left by the printing,
(corresponding to the part cross section) is filled using another ‘ink’ which contains metal
particles. This material is also UV cured and the surface milled flat. The entire process is
repeated until the desired thickness is reached. The semi-finished part is then debinderized
to remove organic ink components, and sintered in a furnace with nitrogen atmosphere to
join the metal particles. The ceramic negative material either crumbles or removed in an
ultrasonic bath. This also results in the separation of the finished part from the substrate.
27
1.4.1.5 Joining solid sheet
1.4.1.5.1 Laminated object manufacturing (LOM)
This technique employs the same basic building approach as that used for the LOM
machine. However, the Kira Solid Center (SC) machine [92] (Figure 1.15) is operated in a
significantly different manner. A conventional laser printer is used to feed standard
printing paper to the SC machine. An adhesive-based toner is used by the printer to print
the outline of the cross-section. In addition a cross-hatched bonding pattern is also printed on
each piece of unwanted paper. A hot plate then laminates the paper to the previous layers.
A carbide knife, mounted on a swivel base, then cuts the cross-sectional outline.
Additionally, to facilitate the removal of the support material, segments of “parting-plane”
sections are also cut.
In this technique semi-polymerized foils, soluble in monomer resin, are used to build up a
part. The foil solidifies and bonds to the previously deposited layer on exposure to UV
28
Figure 1.14: Laminated Object Manufacturing Helisys, Corp. (U.S.)
29
Figure 1.15: Solid Center Kira Corp. (Japan)
30
light. Once the cross-section has been exposed, a new foil can be applied. The unexposed
areas of foil acts as support structure and are later on easily removed by dissolving in the
resin [41, 93]. There are no commercial systems available as yet.
In this section, the major developments in machining (MR) based RP approaches are
briefly reviewed.
This technique, at the Delft University, employs a six-degree of- freedom industrial robot in
conjunction with a three-axis milling to build CAD models [14]. The raw material stock is
mounted on a turntable. A milling device is mounted on the end effector of the robot. This
device constitute of a milling tool and a tool holder. A six-grid voxel data structure is used
to generate the tool path. An interference-free milling tool path is generated by applying
the Minkowski operations [94]. An interactive simulation of the milling process is also
implemented [95]. The SR is a subtractive RP system without the involvement of any
layer. It machines a part from a single piece of stock.
The CAM of LEMs (CAM-LEM Inc., OH) is a commercialized hybrid RP process [20]. It
is capable of fabricating metal or ceramic tooling using pre-formed stainless steel or
ceramic sheet material. The sheet material is made from steel or ceramic powders, held
together using a binder. Part slices are cut from the stock sheets using a high-power laser
with five degrees of freedom. These slices are then assembled together with the help of a
selective-area gripper and registration system (pins and holes) in a correct relative position.
To obtain a monolithic structure the laminated model is fired after assembly. The CAM of
LEMs parts have a relatively low accuracy. The error is mainly caused by two reasons: one
is the post-assembly process; and the other is unpredictable shrinkage due to the firing
process. The typical shrinkage rate is as high as 12-18 per cent. The potential source of
error can be the registration system which destroys the continuity of
31
material properties within the final part. The build envelope of the available CAM of
LEMs machines is also limited to a cubic space of side 150 mm.
This process is based on solvent welding and CNC contour machining [24, 95]. As is the
case with most RP processes, SWIFT builds parts one layer at a time. A laser printer
applies a thin film and prints a high-density polyethylene (HDPE) image for each layer.
The HDPE image acts as a solvent mask and specifies areas of the downward surface for
welding to the previous layer. Acetone solvent is then applied to the bottom side of the
sheet and stacked onto the previous layers. A pressing force is then applied which welds
the current sheet to the previous. The current sheet is then machined to the required shape
with a 3-axis CNC machine. The part is completed by repeating the above cycle for all the
layers. SWIFT has cost, accuracy and speed advantages over most commercially available
RP processes. Although, a large geometric error exists due to the uniform stock layer
thickness which is limited by the feeding system in SWIFT and the zeroth-order or first-
order approximation to the design surface.
Millit [17] is a commercial software package which generates a numerical control (NC)
tool path from STL or 3DS (three-dimensional slice) files. With the help of a build
orientation, it decomposes a model into layers free of undercuts, and these layers are called
‘components’. Dual-sided milling can be employed to mill each component from the base
material. Post-assembly operations are then facilitated by adding borings to the
components’ mating surfaces. All the components are laid out on the raw material sheets
with user-defined thickness. The leftover sheet material is called a ‘frame’. The frames are
connected to the components via bridges, which also act as fixtures during the milling
process. To ensure the accuracy in dual-sided milling, positioning borings are added to
frames. The components are removed from frames, after milling, by breaking the bridges.
These components are then assembled by inserting fitting pins into the borings. To
complete the final physical model, material-specific bonding is needed.
32
of frames during dual-sided milling, and the extraction and post-assembly of components. In
addition, the registration of frames and components and the post-assembly operation are
potential error sources.
The TLOM project was carried out in order to achieve G1 (tangential) or G2 (curvature)
surface continuity in building large-volume models [96]. The main feature of TLOM is the
cutter, which is a deformable blade and is known as flexible blade tool. A pair of
electromechanical rollers controls the blade. The blade profile can be adjusted
instantaneously to fit the local shape of the model features. The layered manufacturing
process is called hot blade cutting. First of all the model is sliced into various thick layers.
The blade profile and tool path is then calculated for each layer and the layer is shaped
accordingly. After all layers are made, they are manually assembled into a physical model
and finishing touches if needed are performed. The complexity of models that can be built
by this technique is limited due to the clumsy tool profile controller assembly. The final
part geometry may have errors due to the introduction of post-assembly.
1.5 Conclusions
The ability to rapidly produce the designs of parts and objects with complex 3D geometry is
a basic necessity in today’s market place. RP has emerged over the last 20 years based on
the principle of creating 3D geometries using computer aided design (CAD), directly and
quickly with little human interaction. It is a technology used in facilitating concurrent
engineering, thus it has started to change the way companies design and build products.
The main goal is to reduce product development, manufacturing costs and lead times,
thereby increasing competitiveness. In the last two decades impressive steps towards this
goal have been made. However the field of RP is still developing, with much effort being
expended on improving the speed, accuracy and reliability of RP systems and widening the
range of materials for prototype construction. Another area for improvement is the costing,
as most RP systems are currently too expensive to be affordable by any but the larger
firms. However, RP technology is available to most companies via the services
33
provided by different bureaus. The future is likely to see more user-owned RP machines
as their costs are reduced.
To date most RP parts are used for prototyping or tooling purposes; however, in future
the focus may be to produce parts as end-use products. In this context the term ‘rapid
manufacturing’ (RM) is used in which RP technologies are employed as processes for the
production of end-use products. The introduction of non-polymeric materials, including
metals, ceramics, and composites, represents a much anticipated development. These
materials would allow RP users to produce functional parts. Today’s plastic prototypes
work well for visualization and fit tests, but they are often too weak for function testing.
More rugged materials would yield prototypes that could be subjected to actual service
conditions. In addition, metal and composite materials greatly expand the range of
products that can be made by RM. In this context the focus of the current work is on
metals; specifically with respect to 3D welding approach, with main emphasis on the
material properties and deformations due to the high temperatures involved in welding. A
more detailed discussion on the welding based systems is presented in the forth coming
chapter.
34