Rapid prototyping issues in the 21st century

Computers in Industry 39 Ž1999. 3–10 Rapid prototyping issues in the 21st century Detlef Kochan a , Chua Chee Kai b b,) , Du Zhaohui b a SFM, Ger...
Author: Branden Simpson
35 downloads 2 Views 229KB Size
Computers in Industry 39 Ž1999. 3–10

Rapid prototyping issues in the 21st century Detlef Kochan a , Chua Chee Kai b

b,)

, Du Zhaohui

b

a SFM, Germany Nanyang Technological UniÕersity, Nanyang AÕenue, 639798, Singapore

Abstract Rapid prototyping ŽRP. technology has been developing very fast in the last 10 years. Some ideas concerning this technology are presented in the paper for predicting the evolution in the future. Shaping science is proposed to research for the improvement of RP process and generation of new processes. Issues including product development tool, direct metal part manufacturing, rapid tooling and RP machine design are discussed. Extension of its application in more areas will be beneficial for the advancement of RP. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Rapid prototyping; Shaping science; Product development; Machine design

1. Introduction Rapid prototyping ŽRP. is a new forming process and principle which is one of the important breakthroughs of recent technological progress in the manufacturing industry. From the emergence of the first RP system ŽSLA. in 1988, 2234 RP systems with about 20 kinds of processes were in operation around the world at the end of 1996. At the end of 1997, system manufacturers had sold a total of 3289 systems around the world w1x. In the early days of RP, the automotive and aerospace industries dominated the RP application. But this is no longer the case as RP has spread into many other industries. It shows how fast RP technology is developing. Digital design ŽCAD. and digital manufacturing ŽRP. have emerged to facilitate and accelerate product creation. Many kinds of technology require two conditions in order to develop quickly. On the one hand, it needs relevant fundamental theories and technologies )

Corresponding author

to support itself. On the other hand, it needs improved capabilities and new application fields. RP technology is no exception. The direction of research and development of RP technology will be analyzed with consideration of these two factors.

2. Major issues facing RP in the 21st century 2.1. Shaping science theory Shaping science w2,3x is the science which studies how to organize materials in sequence to get threedimensional objects with definite shape and certain functions. It is a discipline of basic theory, principles and methods about shaping beyond the practical shaping techniques and will be used to direct the application of these kinds of techniques. There exist various kinds of shaping techniques evolved by human beings for a long time. Although it is a newly developing one, RP process belongs to shaping technologies including casting, metal forming and cutting.

0166-3615r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 3 6 1 5 Ž 9 8 . 0 0 1 2 5 - 0

4

D. Kochan et al.r Computers in Industry 39 (1999) 3–10

It is becoming a research field to study shaping principles, classification and relationships among processes, equipment, software, control and applications at a more systematic level. New shaping theories are beyond the practical shaping processes and will provide the theoretical foundation and scientific principles for the creation of new shaping methods. Chua et al. w4x studied the state-of-the-art of RP technology on the basis of many of the publications on it. Through description of various RP systems with formulas of the important parameters, the qualitative and quantitative assessment is provided so as to highlight the possibility of future improvements for a new generation of RP systems. In shaping science, four general types are included as follows. v Subtracting Shaping. It is a process removing some materials Žor surplus materials. from the basal body, such as latching, milling, planing, drilling, grilling, EDM and laser cutting. v Forced Shaping. In this process a part is shaped under the specific binding force by making use of the deforming ability of material Žsolid or liquid.. For example, foundry, forging, sheet metal forming and plastic injection are all classified under this category which needs molds, dies or other tools.

v Stacking shaping. Emerging in the late 1980s, these technologies are also called RP Žor rapid prototyping and manufacturing, freeform fabrication.. In view of forming, a part can be regarded as a spatial body which is stacked by points and surfaces according to the discretion of its geometric information. v Growth shaping. At present, this type of process only exists in natural systems instead of manmade ones, such as forming of animals’ skeletons. Among the four main shaping types, subtracting and forced shaping are more conventional and mature with individual theories compared to RP and living forming. According to its definition, shaping science should study the most essential forming laws in a higher level beyond the specific techniques. At this level, shaping technology can be as a whole system divided into three main parts: material process, energy process and information process. Each of the main processes consists of a few subelements and each of the subelements has various types of states. The morphological diagram can be constructed for description of shaping system Žas showed in Fig. 1.. Any kind of shaping techniques, such SLA, LOM, or cutting and casting, can be seen as a set of the elements with corresponding states in this diagram.

Fig. 1. Morphological diagram of shaping system.

D. Kochan et al.r Computers in Industry 39 (1999) 3–10

The morphological method also offers one way to classify various kinds of RP techniques with any one or more elements as taxonomy. It is also useful to integrate various kinds of RP processes in one machine as a whole, such as the multifunctional rapid prototyping systems ŽMRPMS. developed at Tsinghua University, China w5x. It is one of the most important problems to study what is the relationship among the material processes, energy processes and information processes and how to improve the interrelated degree of different kinds of processes. This is because the higher the interrelated degree of process is, the more flexibility the shaping technique owns. Enhancing the interrelation among the processes and the evolution of the state of elements will produce the progress of the existing techniques or the generation of new processes. 2.2. Systematic tools supporting product deÕelopment based on RPM The horizons of the industrial world are changing rapidly. Industrial planning, in the past, tended to assume that markets were almost infinite and that whatever was manufactured could be sold if the price was low enough. Now it can be seen that resources and markets are finite. Every aspect of product in the whole life cycle is considered of more importance. A good product development system must enable designers or design teams to consider all aspects of product design, manufacturing, selling and

5

recycling at the early stage of the design cycle, so that design iteration and changes can be made easily and effectively. The more fluent the feedback loop is, the higher possibility of success the system has. Design for manufacturing ŽDFM. and concurrent engineering ŽCE. necessitate that product and process design be developed simultaneously rather than sequentially. The advantages of this kind of system are realized by the integration of reverse engineering, CAD and RP. It provides soft and hard tools for design. During the successful application in various areas in recent years, RP is being used as a communication and inspection tool in the procedure of product development and realization of the rapid feedback of the design information. This kind of product development system which is dynamic, controllable and simultaneous should be realized under the structure shown in Fig. 2. Many of the unit technologies in the system are already mature. But which kind of techniques should be selected and how to integrate those technologies effectively remain difficult problems to solve. For example, among SLA, LOM, SLS, FDM and others, some RP processes need to be chosen with the consideration of material, dimensional precision, surface finish, building speed and cost according to the requirement of designers. It is necessary to develop software to manage and control the whole information and processes. Axiomatic method will be used as a strategy for the effective integration of people, processes, design tools and design data.

Fig. 2. Product development system based on RP.

6

D. Kochan et al.r Computers in Industry 39 (1999) 3–10

2.3. Direct manufacturing of metal parts with RP RP technology has proven successful in many ways as a process which can be easily and rapidly automated and with almost no geometry limitation in parts to be fabricated. But limitations of this kind of technology are mainly the availability of materials. Commonly materials used in present RP systems are polymers, paper and ceramic. Metal parts directly built with RP are rare comparatively. It is obvious that with metal, the technology would be extended to more useful prototypes or parts and fast manufacturing of tools. The most important requirements RP should fulfill are to build metal parts with adequate strength and accuracy. That is the main problem existing now. LOM is an approach with combination of thin andror thick sheet metal. But the major problem is difficulty in controlling both the gluing temperature and ambient temperature and in preventing the sheets from curling during the polymer coating is heated. The most practical approach is to bind metal powder together, thermally Žsuch as SLS. or chemically Žsuch as 3DP.. SLS of metals has been researched for some time and seems to be more possible to reach. The main problem is the control the heat balance between the laser sintering part and the surrounding material. In many cases, heat stresses will occur, which will bring about shrinkage and result in loss of accuracy. In the paper of Killander and Gohlenius w6x, many creative ideas and thoughts had been introduced in this area with the inspiration of theory of inventive problem solving ŽTIPS.. However, there is still much to do. Other perspective ways to build metal parts with RP is a combination of the benefits of material additive process with the benefits of material removal process. Carnegie Mellon and Stanford universities are developing an additionrremoval process, named shape deposition manufacturing ŽSDM. w7x. In SDM, a CAD model is first sliced into 3D layer structure. Layer segments are then deposited as near-net shape and then machined to net shape before additional material is deposited. Within the building of one layer, SDM likes a process with the occurrence of microcasting without any mold and CNC cutting sequentially. The process shows great potential with higher precision and less inner stress.

However, the problem in the procedure of deposition of molten metal droplet, the data slicing and CNC tool planning is very difficult because the outer surface of each layer maintains the 3D geometry of the original model. 2.4. Combination of RP and metal casting, rapid tooling Although direct manufacturing of metal parts with RP is not well developed, indirect methods have been found and shown feasible through the combination of RP and metal casting Žindeed investment casting.. The major impact RP technology has is its ability to fabricate high quality and complex patterns used in investment casting with lower cost and shorter leading times. Although application in mass production is not practical, this kind of technology is very suitable for one case or low volume production. Tooling is one of the best suitable application areas. Many case studies with different RP processes and casting techniques have proven to be successful. Of course, there are many combinations in which the approach can work. Each kind of RP process will fabricate prototypes with the specific kind of materials, which will need the corresponding casting technique to support the conversion from prototypes to metal parts. In fact, it is not always practical that any kind of RP technique can combine with an arbitrary kind of casting process directly. Some paths need transition steps and many of them need improvement of quality. Table 1 shows the most common commercial RP systems and the material field of prototypes produced by them. The application approaches in casting or tooling with different kinds of prototypes are listed in Table 2. Because there exists more than one approach, it is necessary to estimate whether or not a process with

Table 1 RP processes with materials SLA LOM SLS FDM 3DP

Polymer Paper Ceramic, wax and alloy Wax and polymer Ceramic and wax

D. Kochan et al.r Computers in Industry 39 (1999) 3–10 Table 2 Conversion approaches Kind of material

Conversion approach

Polymer Paper Ceramic Wax Alloy Arbitrary

Quickcasting, epoxy pattern, soft tooling Ceramic investment casting, Sand casting Direct ceramic casting mold Lost wax casting Soft tooling Metal spray, EDM

the selected kind of RPM and conversion technique can meet the special order. Which one among the various RP and conversion techniques is the best choice with the consideration of cost, cycle time, physical performance and geometrical accuracy. The issues could be seen as process planning. Fraunhofer Institute for Manufacturing Engineering and Automation ŽIPA. uses a quality function deployment ŽQFD. approach for selecting the most appropriate RP technology w8x. To incorporate multiple factors into the process decision making, a ‘feed-forward’ process planning procedure should be developed Ždata flowing shown in Fig. 3. w9x. During process planning, the designer is faced with a set of alternative RP and casting techniques to produce a finished part or tool. The requirement serves as input to the process planning scheme. According to the ability of RPM and casting techniques, feasible process combinations can be chosen. During the scheme of process selection, the reliable criteria should be followed. Factors including machining time, cost, part quality and environment effect need to be considered. The complexity is further increased due to

7

the fact that the degree of these factors varies from order to order. To meet the requirement of parts or tools with high, a programmable scheme should be built as a rapid intelligent tooling ŽRIT. system. The RP and conversion processes ability matrix are constructed for selecting the feasible process approaches according to the criteria of process reliability. Analytic hierarchy process and weighted process ability scores are introduced to evaluate the priority among factors such as machining cost, time and quality of tools. With these programs, an optimal process approach can be decided from a set of feasible paths which best reflect the engineered need.

2.5. Extension of the application with RP RP technology has the potential to ensure that quality-assured prototypes or parts are developed quickly for two major reasons. There are almost no restrictions on geometrical shapes; and the layered manufacturing allows a direct and simple interface with CAD to CAM which almost completely eliminates the need for process planning, a complex procedure for CNC machining. The technology shows various applications after its emergence Žas shown in Fig. 4.. RP technology has been introduced successfully in the industries of automotive, aerospace, shipbuilding, computer, toy, and consumer products. Microelectronic–mechanical system will become the other important field of application with RP. The height of microelectronic circuits and microelectronic-mecha-

Fig. 3. Data flow of RIT process planning.

8

D. Kochan et al.r Computers in Industry 39 (1999) 3–10

Fig. 4. Applications with RP.

nism is typically between 2 and 10 mm. That means SLA or other processes will be a feasible way of fabricating these systems. Of course, medical application is another important and promising direction for developing RPM technology. Now human organ models can be produced by means of using RPM technology and medical digital imaging systems Žsuch as X-ray, CT and MRI.. These models are very helpful in diagnosis, preparation of complex surgery and recuperation engineering. The organ-shaped models with biomaterial or bio-compatible material will be applied directly for orthopaedic implant and prostheses. SLS will be used to precisely control the fabrication of drug delivery devices. The nature and characteristics of the devices is such that porosity and channels of specific orientations will be built into the devices. The problem is to remove the loose powder from the channels after building the device without compromising on the quality of subsequent layers. 2.6. Machine design

large equipment. The reason is that in this process, hatch lines scanned for removal of excess material is with large space be apart. SLS and 3DP processes can be used to make ceramic casting mold shells directly from a CAD model without pattern but they are not suitable for large-sized castings because of low strength of the shells fabricated and slow building speed. This explains why it is necessary that the process of pattern less casting molding ŽPLCM. was created which can fabricate sand casting mold shell w3x. With curing sand on the top layer by selectively spraying resin and catalyst sequentially, the shell built can be delivered directly to the foundry. The process is very practicable in producing castings with large size and low precision. In other cases, it is necessary to fabricate smallscaled prototypes or parts with high precision, such as in art jewelry and microelectronic–mechanical systems. A machine with some kind of RP process should be designed according to the special requirements.

2.6.1. Larger scale or smaller scale It is common that the sizes of parts that can be built from the presently available RP systems are comparatively small because of the characteristics of such processes. In some cases, parts with large size will be split into a few smaller parts so as to build them separately in RP systems. After building, all parts can be assembled together as a whole. With the consideration of building speed, LOM process is the most suitable one to enhance the working space with

2.6.2. Layer forming quality While forming any layer in RP processes, it always happens that materials are cured with the scanning of laser or materials are cured or delivered from the scanning nozzle. The movement of laser or the movement of nozzle is driven by step motors or servo-motors. In terms of drawing the scan path, the optimum situation is one in which the path drawn is uniform along its length and has the same width near the beginning, middle and end. Of course, it is the

D. Kochan et al.r Computers in Industry 39 (1999) 3–10

best that the laser or nozzle can be opened or closed at the start or end point of paths without a delay time. But the optimum situation can not happen always because there exists the transitional state during acceleration or deceleration of motors and opening or closing of nozzles. Thus match control between laserrnozzle and motion system is of great importance for the quality of forming each layer. Some control units or components in special use will be developed for solving the problem. 2.6.3. Material deliÕery An important characteristic all RP machines have in common is that they need to deliver material. The delivery device is generally a critical part of these technologies and often embodies proprietary development. In SLA, the speed of recoating of resin and the flatness of the resin surface produced are so important that they become the competition goal among the different commercial systems. In FDM, the extrusion nozzle should deliver the material at a strictly controlled temperature for maintaining the shapes. Controlling the width of the material delivered is critical to achieving dimensional accuracy. In SLS or 3DP, the flatness of the powder spread and its density are very important for the realization of processes. In SDM, the deposition nozzle is critical to achieving the desired material properties. Much research and development in those fields needs to be carried on. 2.6.4. Cost, reliability, operation and others With the consideration of extension of markets, the price of commercial RP systems and the material used is too high. This kind of situation always happens at the initial stage of a new high-technology. It can be seen that the price will become lower and lower. According to the 3D system, the MJM system, their desktop RP system running in an office environment, enables the creation of physical models of their concepts almost as easy as printing paper plots. It is possible in the future that the RP system with low price and reliability will come into the family as a tool for making arts or adornments with ideas and CAD data from one’s original or from the Internet. In fact, many issues such as CAD, information processing, material, accuracy, surface finish, speed,

9

pre-processing, post-processing and so on, are very important for RP technology as well. Since the emergency of the RP technology, much research on them has been carried on. Some of them have been solved while others are still unsolved yet. For example, some kinds of part need to add supports while building with certain RP process. In this case, a program should be developed for adding supports at the right place with the right structure. In other RP process, the support is not necessary for some kinds of processes. In this case, removal of the redundant material from the finished parts without damaging is very critical.

3. Conclusion Many issues from the improvement of RP technology to the extension of applications with RP are presented and discussed. It shows that some of them are critically important for this kind of technology and urgently need to be further researched. Although it is not practical to tend to overestimate what will happen in the near future, the next century will see a significant progress in RP technologies.

References w1x T. Wohlers, Rapid prototyping state of the industry: 1997 worldwide progress report, RPA of SME, Dearborn, MI, 1997. w2x Y. Yongnian, Modern shaping-science and RP processes, Proceeding of INCOM’95 Beijing, Oct. 11–13, 1995, pp. 300– 304. w3x D. Zhaohui, Study and development of patternless casting mold manufacturing directly driven by CAD model, PhD dissertation, Tsinghua University, Beijing, China, 1997. w4x C.K. Chua, S.M. Chou, T.S. Wong, Study of the state-of-the-art rapid prototyping technologies, Int. Journal of Advanced Manufacturing Technology 14 Ž1998. 146–152. w5x Y. Yan, R. Zhang, Q. Lu, Z. Du, Study on multifunctional rapid prototyping manufacturing system ŽM-RPMS., will be published in Rapid Prototyping Journal, 1998. w6x L.A. Killander, G. Gohlenius, Future direct manufacturing of metal parts with free-form fabrication, Annals of the CIRP 44 Ž1. Ž1995. 450–454. w7x L.E. Weiss, F.B. Prinz et al., Shape deposition manufacturing of heterogeneous structures, Journal of Manufacturing Systems 16 Ž4. Ž1997. 239–248. w8x R.F. Aubin, Tooling applications, JTECrWTEC Panel Report on Rapid Prototyping in Europe and Japan, RPA of the SME, 1997.

10

D. Kochan et al.r Computers in Industry 39 (1999) 3–10

w9x D. Zhaohui, et al., Study of rapid intelligent tooling system based on RPM technology, Proceedings of the International Conference on Manufacturing Automation ŽICMA’97., April 1997, Hong Kong. Detlef Kochan was born in Germany in 1935. He received his doctorate in engineering in 1971 from the Industrial Manufacturing and Automation Faculty of the University of Technology, Dresden. Since 1970 he has been an associate professor and since 1975 a full professor for Manufacturing Technologyr CAM at the same institute. In 1993, he founded the company SFM-Gesellschaft zur Schnellen Fertigung von Modellen. He is currently manager of this company, honorary member of IFIP TC 5 and Chairman of GFaI LV Sachsen. His research interests are intelligent production system and solid freeform manufacturing. Chua Chee Kai is an associate professor at the School of Mechanical and Production Engineering of Nanyang Technological University, Singapore. He is concurrently the co-director of the Centre for Engineering and Technology Management, and the deputy director of the Biomedical Engineering Research Centre. He received his BEng ŽMech. with first class honours and MSc ŽInd. from the Nation University of Singapore and his PhD from the Nanyang Technological University, Singapore. He is a member of Working Group 5.3 ŽComputer-aided Manufacturing. of the International Federation for Information Processing ŽIFIP.. His current research interests include reverse engineering, CADrCAM, geometric modelling and rapid prototyping.

Du Zhaohui received his Bachelor degree on engineering in 1993 and his PhD in 1998, both from Tsinghua University, China. He is presently a research fellow at the School of Mechanical and Production Engineering of Nanyang Technological University, Singapore. He has worked exclusively in the area of design and manufacturing science and technology. He developed CADrCAM systems for plate heat exchanger and patternless casting mold making technique. His current research interests include rapid prototyping and manufacturing, CADrCAM, reverse engineering, rapid tooling and patternless casting technology.