14 Chapter 2 The design process Component 1.1 Sub-assembly Component 1.2 Component 1.3 Component 2.1 Technical Sub-assembly system Component 2.2 Component 3.1 Sub-assembly Com 3.2 Figure 2.2 The analysis of a technical system as a breakdown into assemblies and components Material and process selection is at the component level Input Function Material unctIo Function Information Information Sub-systems Figure 2.3 The systems approach to the analysis of a technical system, seen as transformation of energy, materials and information(signals). This approach, when elaborated, helps structure thinking about alternative designs
Technical system Sub-assembly 1 Component 1.1 Component 1.2 Component 1.3 Sub-assembly 2 Sub-assembly 3 Component 2.1 Component 2.2 Component 2.3 Component 3.1 Component 3.2 Component 3.3 Figure 2.2 The analysis of a technical system as a breakdown into assemblies and components. Material and process selection is at the component level. Energy Material Information Function 1 Inputs Function 3 Function 5 Function 6 Technical system Outputs Sub-systems Function 2 Function 4 Energy Material Information Figure 2.3 The systems approach to the analysis of a technical system, seen as transformation of energy, materials and information (signals). This approach, when elaborated, helps structure thinking about alternative designs. 14 Chapter 2 The design process
2.2 The design process 5 component are drawn up. Critical components may be subjected to precise mechanical or thermal analysis. Optimization methods are applied to com- ponents and groups of components to maximize performance. A final choice of geometry and material is made and the methods of production are analyzed and costed. The stage ends with a detailed production specification All that sounds well and good. If only it were so simple. The linear process suggested by Figure 2.1 obscures the strong coupling between the three stages The consequences of choices made at the concept or embodiment stages may not become apparent until the detail is examined. Iteration, looping back to explore alternatives, is an essential part of the design process. Think of each of the many possible choices that could be made as an array of blobs in design space as suggested by Figure 2. 4. Here C1, C2, . are possible concepts, and El, E2,..., and D1, D2,... are possible embodiments and detailed Market need design requirements Concept c7) E2).( Embodiment Detail Product specification Figure 2. 4 The previous figure suggests that the design process is logical and linear. The reality is otherwise. Here the C-blobs represent possible concepts, the E-blobs possible embodiments of the Cs, and the D-blobs possible detailed realizations of the Es The process is complete when a compatible path form"Need"to"Specification"can be identified. The extreme coupling between the idealized design"stages"leads to a devious path(the full line)and many dead-ends(the broken lines). This creates the need for tools that allow fluid access to materials information at differing levels of breadth and detail
component are drawn up. Critical components may be subjected to precise mechanical or thermal analysis. Optimization methods are applied to components and groups of components to maximize performance. A final choice of geometry and material is made and the methods of production are analyzed and costed. The stage ends with a detailed production specification. All that sounds well and good. If only it were so simple. The linear process suggested by Figure 2.1 obscures the strong coupling between the three stages. The consequences of choices made at the concept or embodiment stages may not become apparent until the detail is examined. Iteration, looping back to explore alternatives, is an essential part of the design process. Think of each of the many possible choices that could be made as an array of blobs in design space as suggested by Figure 2.4. Here C1, C2, ... are possible concepts, and E1, E2, ... , and D1, D2, ... are possible embodiments and detailed Market need: design requirements Product specification C1 C2 C6 C4 C7 C3 E3 C5 E1 E6 E2 E4 E7 E8 D3 E5 D2 D1 D4 D6 D5 Embodiment Detail Concept Figure 2.4 The previous figure suggests that the design process is logical and linear. The reality is otherwise. Here the C-blobs represent possible concepts, the E-blobs possible embodiments of the Cs, and the D-blobs possible detailed realizations of the Es. The process is complete when a compatible path form ‘‘Need’’ to ‘‘Specification’’ can be identified. The extreme coupling between the idealized design ‘‘stages’’ leads to a devious path (the full line) and many dead-ends (the broken lines). This creates the need for tools that allow fluid access to materials information at differing levels of breadth and detail. 2.2 The design process 15
16 Chapter 2 The design process laborations of them. Then the design process becomes one of creating paths, linking compatible blobs, until a connection is made from the top("market need")to the bottom("product specification"). The trial paths have dead-ends, and they loop back. It is like finding a track across diffcult terrain -it may be necessary to go back many times if, in the end, we are to go forward. Once a path is found, it is always possible to make it look linear and logical (and many books do this), but the reality is more like Figure 2. 4, not Figure 2.1. Thus a key part of design, and of selecting materials for it, is flexibility, the ability to explore alternatives quickly, keeping the big picture as well as the details in focus. Our focus in later chapters is on the selection of materials and processes, where exactly the same need arises. This requires simple mappings of the kingdoms"of materials and processes that allow quick surveys of alternatives while still providing detail when it is needed. The selection charts of Chapter 4 and the methods of Chapter 5 help do this Described in the abstract, these ideas are not easy to grasp. An example will help-it comes in Section 2.6. First, a look at types of design 2.3 Types of design It is not always necessary to start, as it were, from scratch. Original design does: it involves a new idea or working principle (the ball-point pen, the compact disc). New materials can offer new, unique combinations of proper ties that enable original design. Thus high-purity silicon enabled the transistor; high-purity glass, the optical fiber; high coercive-force magnets, the miniature rhone, solid-state lasers the compact disc. Sometimes the new material uggests the new product; sometimes instead the new product demands the development of a new material: nuclear technology drove the development of a eries of new zirconium-based alloys and low-carbon stainless steels; space technology stimulated the development of light-weight composites; turbine technology today drives development of high-temperature alloys and ceramics Adaptive or developmental design takes an existing concept and seeks an incremental advance in performance through a refinement of the working principle. This, too, is often made possible by developments in materials polymers replacing metals in household appliances; carbon fiber replacing wood in sports goods. The appliance and the sports goods market are both large and competitive. Markets here have frequently been won(and lost) by the way in which the manufacturer has adapted the product by exploiting new materials Variant design involves a change of scale or dimension or detailing without change of function or the method of achieving it: the scaling up of boilers, or of pressure vessels, or of turbines, for instance. Change of scale or circumstances of use may require change of material: small boats are made of fiberglass, large ships are made of steel; small boilers are made of copper, large ones of
elaborations of them. Then the design process becomes one of creating paths, linking compatible blobs, until a connection is made from the top (‘‘market need’’) to the bottom (‘‘product specification’’). The trial paths have dead-ends, and they loop back. It is like finding a track across difficult terrain — it may be necessary to go back many times if, in the end, we are to go forward. Once a path is found, it is always possible to make it look linear and logical (and many books do this), but the reality is more like Figure 2.4, not Figure 2.1. Thus a key part of design, and of selecting materials for it, is flexibility, the ability to explore alternatives quickly, keeping the big picture as well as the details in focus. Our focus in later chapters is on the selection of materials and processes, where exactly the same need arises. This requires simple mappings of the ‘‘kingdoms’’ of materials and processes that allow quick surveys of alternatives while still providing detail when it is needed. The selection charts of Chapter 4 and the methods of Chapter 5 help do this. Described in the abstract, these ideas are not easy to grasp. An example will help — it comes in Section 2.6. First, a look at types of design. 2.3 Types of design It is not always necessary to start, as it were, from scratch. Original design does: it involves a new idea or working principle (the ball-point pen, the compact disc). New materials can offer new, unique combinations of properties that enable original design. Thus high-purity silicon enabled the transistor; high-purity glass, the optical fiber; high coercive-force magnets, the miniature earphone, solid-state lasers the compact disc. Sometimes the new material suggests the new product; sometimes instead the new product demands the development of a new material: nuclear technology drove the development of a series of new zirconium-based alloys and low-carbon stainless steels; space technology stimulated the development of light-weight composites; turbine technology today drives development of high-temperature alloys and ceramics. Adaptive or developmental design takes an existing concept and seeks an incremental advance in performance through a refinement of the working principle. This, too, is often made possible by developments in materials: polymers replacing metals in household appliances; carbon fiber replacing wood in sports goods. The appliance and the sports-goods market are both large and competitive. Markets here have frequently been won (and lost) by the way in which the manufacturer has adapted the product by exploiting new materials. Variant design involves a change of scale or dimension or detailing without change of function or the method of achieving it: the scaling up of boilers, or of pressure vessels, or of turbines, for instance. Change of scale or circumstances of use may require change of material: small boats are made of fiberglass, large ships are made of steel; small boilers are made of copper, large ones of 16 Chapter 2 The design process
2.4 Design tools and materials data 17 steel; subsonic planes are made of one alloy, supersonic of another; and for good reasons, detailed in later chapters 2.4 Design tools and materials data To implement the steps of Figure 2.1, use is made of design tools. They are shown as inputs, attached to the left of the main backbone of the de methodology in Figure 2.5. The tools enable the modeling and optimization of a design, easing the routine aspects of each phase. Function-modelers viable function structures. Configuration optimizers suggest or refine Geometric and 3D solid modeling packages allow visualization and create files that can be down-loaded to numerically controlled prototyping and manufacturing systems Optimization, DFM, DFA, and cost-estimation Market need design requirements Design tools Material data Data for ALL materials Concept iability studies and detail Geometric modeling Embodiment ecision and detail Cost modeline Finite-element Data for ONE material modeling(FEM Detail and deta DFM. DFA Product Figure 2.5 The design flow chart, showing how design tools and materials selection enter the procedure. Information about materials is needed at each stage, but at very different levels of breadth and precision Design for Manufacture and Design for
steel; subsonic planes are made of one alloy, supersonic of another; and for good reasons, detailed in later chapters. 2.4 Design tools and materials data To implement the steps of Figure 2.1, use is made of design tools. They are shown as inputs, attached to the left of the main backbone of the design methodology in Figure 2.5. The tools enable the modeling and optimization of a design, easing the routine aspects of each phase. Function-modelers suggest viable function structures. Configuration optimizers suggest or refine shapes. Geometric and 3D solid modeling packages allow visualization and create files that can be down-loaded to numerically controlled prototyping and manufacturing systems. Optimization, DFM, DFA,1 and cost-estimation Data for ALL materials, low precision and detail Data for a SUBSET of materials, higher precision and detail Data for ONE material, highest precision and detail Function modeling Viability studies Approximate analysis Geometric modeling Simulations methods Cost modeling Component modeling Finite-element modeling (FEM) DFM, DFA Market need: design requirements Product specification Embodiment Detail Concept Material data needs Design tools Figure 2.5 The design flow chart, showing how design tools and materials selection enter the procedure. Information about materials is needed at each stage, but at very different levels of breadth and precision. 1 Design for Manufacture and Design for Assembly. 2.4 Design tools and materials data 17
18 Chapter 2 The design process software allows manufacturing aspects to be refined. Finite element(FE)and Computational Fluid Dynamics (CFD) packages allow precise mechanical and thermal analysis even when the geometry is complex and the deformations are large. There is a natural progression in the use of the tools as the design evolves approximate analysis and modeling at the conceptual stage; more sophisticated modeling and optimization at the embodiment stage; and precise("exact but nothing is ever that)analysis at the detailed design stage Materials selection enters each stage of the design. The nature of the data needed in the early stages differs greatly in its level of precision and breadth from that needed later on(Figure 2.5, right-hand side). At the concept-stage the designer requires approximate property-values, but for the widest possible range of materials. All options are open: a polymer may be the best choice for one concept, a metal for another, even though the function is the same. The problem, at this stage, is not precision and detail; it is breadth and speed of access: how can the vast range of data be presented to give the designer the greatest freedom in considering alternatives? At the embodiment stage the landscape has narrowed. Here we need data for a subset of materials, but at a higher level of precision and detail. These are found in the more specialized handbooks and software that deal with a single class or sub-class of materials -metals, or just aluminum alloys, for instance The risk now is that of loosing sight of the bigger spread of materials to which we must return if the details do not work out; it is easy to get trapped in a single line of thinking -a single set of"connections"in the sense described in the last section -when other combinations of connections offer a better solution to the design problem The final stage of detailed design requires a still higher level of precision and detail, but for only one or a very few materials. Such information is best found in the data-sheets issued by the material producers themselves, and in detailed databases for restricted material classes. A given material (polyethylene, for instance) has a range of properties that derive from differences in the ways different producers make it. At the detailed design stage, a supplier must be identified, and the properties of his product used in the design calculations; that from another supplier may have slightly different properties. And sometimes even this is not good enough. If the component is a critical one(meaning that its failure could, in some sense or another, be disastrous)then it may be pru dent to conduct in-house tests to measure the critical properties, using a sampl of the material that will be used to make the product itself It's all a bit like choosing a bicycle. You first decide which concept best suits your requirements(street bike, mountain bike, racing, folding, shopping, reclining, ..) limiting the choice to one subset. Then comes the next level of etail. What frame material? What gears? Which sort of brakes? What shape of handlebars? At this point you consider the trade-off between performance and cost, identifying (usually with some compromise) a small subset that meet both rour desires and your budget. Finally, if your bicycle is important to you, you seek further information in bike magazines, manufacturers'literature or the
software allows manufacturing aspects to be refined. Finite element (FE) and Computational Fluid Dynamics (CFD) packages allow precise mechanical and thermal analysis even when the geometry is complex and the deformations are large. There is a natural progression in the use of the tools as the design evolves: approximate analysis and modeling at the conceptual stage; more sophisticated modeling and optimization at the embodiment stage; and precise (‘‘exact’’ — but nothing is ever that) analysis at the detailed design stage. Materials selection enters each stage of the design. The nature of the data needed in the early stages differs greatly in its level of precision and breadth from that needed later on (Figure 2.5, right-hand side). At the concept-stage, the designer requires approximate property-values, but for the widest possible range of materials. All options are open: a polymer may be the best choice for one concept, a metal for another, even though the function is the same. The problem, at this stage, is not precision and detail; it is breadth and speed of access: how can the vast range of data be presented to give the designer the greatest freedom in considering alternatives? At the embodiment stage the landscape has narrowed. Here we need data for a subset of materials, but at a higher level of precision and detail. These are found in the more specialized handbooks and software that deal with a single class or sub-class of materials — metals, or just aluminum alloys, for instance. The risk now is that of loosing sight of the bigger spread of materials to which we must return if the details do not work out; it is easy to get trapped in a single line of thinking — a single set of ‘‘connections’’ in the sense described in the last section — when other combinations of connections offer a better solution to the design problem. The final stage of detailed design requires a still higher level of precision and detail, but for only one or a very few materials. Such information is best found in the data-sheets issued by the material producers themselves, and in detailed databases for restricted material classes. A given material (polyethylene, for instance) has a range of properties that derive from differences in the ways different producers make it. At the detailed design stage, a supplier must be identified, and the properties of his product used in the design calculations; that from another supplier may have slightly different properties. And sometimes even this is not good enough. If the component is a critical one (meaning that its failure could, in some sense or another, be disastrous) then it may be prudent to conduct in-house tests to measure the critical properties, using a sample of the material that will be used to make the product itself. It’s all a bit like choosing a bicycle. You first decide which concept best suits your requirements (street bike, mountain bike, racing, folding, shopping, reclining, ... ), limiting the choice to one subset. Then comes the next level of detail. What frame material? What gears? Which sort of brakes? What shape of handlebars? At this point you consider the trade-off between performance and cost, identifying (usually with some compromise) a small subset that meet both your desires and your budget. Finally, if your bicycle is important to you, you seek further information in bike magazines, manufacturers’ literature or the 18 Chapter 2 The design process