Materials Selection in Mechanical Design Michael Ashby, Lecture notes of Design

Download Materials Selection in Mechanical Design Michael Ashby and more Lecture notes Design in PDF only on Docsity! Materials Selection in Mechanical Design Michael Ashby Chapter 1. Introduction Mechanical components have mass, they carry loads, they conduct heat and electricity, they are exposed to wear and to corrosion, they are made of one or more materials; they have shape; and they must be manufactured. We need to understand how these activities are related. Materials have limited design since man first build shelters, made clothes and undertook human conflict. But right now materials and the processes to shape them are developing faster than at any time in history; the challenges and opportunities are therefore greater than ever before. This book and this course are about strategies for exploiting materials for design. Normally the choice of material is dictated by the design, but sometimes it is the other way around, the development of a new material changes the way something is designed. You can all think of examples in sporting equipment where for example the development of fiber-reinforced composites changed the pole vault or modified the fishing pole or the golf club. Today the number of materials available to the designer is vast, more than 50,000 are available. While some materials are standardized, removing some close options, new materials are developed that expands the list. The process for the selection 1 of the material depends on the stage of the design, with initial design suggesting the consideration of a wide range of material, but final design requiring more accurate information to choose between a few materials and to make the final accurate design. The choice cannot be made independently of the forming and finishing processes for the materials for these significantly impact on properties and cost. Another factor that must be considered is the aesthetics of the final product. Here function overlaps marketing. Evolution of engineering materials Throughout history materials have limited design. The prehistoric ages were named for the chief materials that man employed: the stone age, the bronze age and the iron age. The evolution is shown schematically in Figure 2 in which the time scale is not linear. We see the early history of materials was dominated by the development of metals – usually produced by metallurgical processes, which replaced the naturally occurring materials like wood, skins and flint. Laterally, a range of high performance polymers, composites and ceramics has been reversing the process away from metals and now metals are just one of the options. Although the expansion of the steel industry has slowed or even started to contract, it should never be forgotten that 90% of all metal used in technology is iron based and it would be impossible to replace this versatile, inexpensive, strong and stiff material in most of its applications. 2 This way of thinking is most useful in analysing an existing design and perhaps refining it by better materials selection. Ashby suggests that a systems approach may be more useful in innovative design as suggested in the next figure. The process is to consider what has to be accomplished and to look for different ways of doing it. In the design of a car such traditional thinks as the motor, the transmission, the braking system etc., could be looked at by considering a different ordering and structuring of components, so the braking system can be hooked up to use that energy to help drive the car forward – an innovative design that is used in some energy efficient cars. 5 Types of Design Innovative design (a new approach to the solution of the need) such as the change from a record player for sound to the tape recorder or the CD player are the most complete design steps. New materials like semiconductors or high purity glass may permit breakthrough original designs. Adaptive or development design takes an existing design and improves it through better materials selection. We see this in new sports equipment for example. The final variation is so called ‘variant design’ where a change of scale or dimension necessitates a change of materials. Small boats can be made of fibreglass most efficiently, but large tankers can only be made from steel. At each stage of design from conceptual to detailed, materials data is needed, but the type of data is different and gets more precise and less broad as the final design is approached as shown in the next figure. Final design is often done with manufacturer’s specifications and may require in-house testing of critical properties. 6 The final insight is the material selection is three fold as suggested in the following diagram. Specifically if we want a beam to support a load in one direction, then the I beam is an ideal shape and becomes part of the design. But while both steel and wood have similar properties per unit weight as beams, steel is readily shaped by hot rolling into an I beam, but it is not so simple to make an I beam out of wood in an efficient manner, so steel may become the material of choice. There is a more complicated example of wine cork removing devices in the text in Chapter 2 to which you are referred. Chapter 3. Engineering Materials and Their Properties This chapter reviews the important classes and properties of materials. The material classes are shown in the figure 3.1. You are likely already familiar with these materials. The materials properties used in design are covered in Table 3.1, again, most of these properties you will have covered. If in your design you come up against a property that you are not totally familiar, this chapter is a place to start. But in design we need to escape the blindfold of thinking only of one class of materials in a given context and think instead of a materials as a certain property profile. In the next chapter we will look at an innovative way of considering these property profiles, which will aid us in thinking about design. 7 An example of the application of this chart is to find material with high elastic wave velocity. It is known that ( )/ ρEV = . So log V = 0.5 log E – 0.5 log ρ or lo =log ρ +2 log V. Thus constant wav ill occur when E and ρ change by equal amounts (i.e. a line of slop t). Thus highest wave velocity material types are engineering composites, glasses and engineering ceramics. (see Fig. 4.3 for details). In polymers and elastomers the elastic wave velocity is as much as a factor of lower. The eng 5.0 g E e velocity w e 1 on a log log plo 30 ineering properties of materials are usefully displayed as materials selection harts. The charts are accessible and summarize the range of properties and the st r , rences ’ll . c appropriate classes of materials for many different applications of materials. It is mo striking how material classes are so clearly clustered on the charts. The first orde difference in most material properties are controlled by the nature of the atomic bonding the mass of the atoms involved and the nature of the packing. Microstructural diffe that are stressed in metallurgy while significant are clearly second order in the overall context of differences. Charts can serve to estimate and validate material data. They provide hints as to the potential application of new materials. Most significantly as we see they serve as the basis for a very useful procedure for selecting materials for design 10 Chapter 5 Materials Selection - the Basics s required in the application, we will derive scheme for materials selection following the Ashby method. At the beginning we only and From the material properties and the propertie a need the kind of property information that is available in handbooks. In this course we will use the materials database CES, with which you have had some experience already in MATE 452. Using this program we have access to a large amount of detailed and authenticated information about materials in all classes. As will be made clear, the selection of material will actually involve a bit more than the property profile. Shape the ability to shape the material in desired form also can have a significant impact in 11 some applications. Thus the full package in selecting a material can be illustrated as i Figure 5.1. n simple example of the importance of shape and processing can be seen in the selection e t hus we must consider a combination of the material properties, how it can be processed, A of materials for constructing the structural members of a bicycle. One of the important attributes that these members must possess is the ability to resist bending (a stiffness property) and at the same time be light. When it comes to bending the resistance to bending of different shapes can be found using the second moment of inertia. A shap that has a high second moment of inertial for bending in all directions is the cylindrical tube. (The I beam is good in resisting bending in a gravitational field). So if we want to use a material for the members of a bicycle it is advantageous if that the material can be readily shaped into tubes. Thus, steel and aluminium and titanium can be considered, bu wood, which has some very good specific properties, cannot be readily shaped into a tube (it can be drilled out, but this weakens it and is expensive). So wood becomes much less attractive in this application. The development of better techniques for making tubes out of composites and of joining them would contribute to making these materials more desirable for this application. T what the function of the material is in the application and what the desirable shape of the material is to perform that function. In the CES database there is information about material processing as well as information about all its properties. 12 criterion is just met. Hence, we can find A in both equations to obtain: m/lρ =(Fl3/4δE)0.5 or 5.0 5.1 5.0F ρ 4 E lm δ ⎟ ⎠ ⎞ ⎜ ⎝ ⎛ dimensions and the given constraints (the allowable deflection for load F). ≥ . Here we’ve arranged the last term to contain the material properties, so it becomes the materials index. The other terms are the geometric shby generalizes this by noting that a design requirement can be written as: A where p is the performance. nd M, then the problem is greatly simplified since we can e. When we can separate F,G, a select the material to maximize performance irregardless of the details of the problem, i. when : And the f1, f2 and f3 are separate functions, which are simply multiplied together. summary, the design requirements of a component, which performs mechanical, such arts of hape, rts hapter 7 Selection of Material and Shape s indicated briefly already, shaped sections like tubes and I beams carry bending, d. e In thermal or electrical functions, can be formulated in terms of one or more functions as those above. Often the equations will yield a group of materials problems (the material index that can be maximized or minimized in selecting a material. The ch these properties then allow the selection of the material using the methods already outlined. Sometimes this is independent of shape, sometimes we need to consider s which will be the topic of Chapter 7. A large number of design problems have been considered using this approach and these are covered in Chapter 6. The selection cha appear in Appendix C of Ashby’s book. C A torsional and axial compressive loads more efficiently than solid sections. Here by efficiency we mean that the section uses less material and mass to carry the same loa This efficiency can be extended even further with sandwich panels (thin load bearing skins bonded to a foam or honeycomb interior). We will look at some aspects of shape here. The scheme for design is now shown in Fig. 7-1. In the example mentioned befor of bicycles, the forks are loaded in bending. What are the pros and cons of using beams of wood or steel in this case? It is interesting to note that early bicycles were made of wood. However as steel became available in tubular sections wood disappeared. It turns out that wood is better in solid rods but steel in tubes is better than wood in a rod shape. What about aluminium or titanium tubes? This chapter outlines a method for answering this question. We will not look at all the details in this chapter, but will only consider bending and compression, if you are interested in torsion, you should refer to the text. 15 hape Factors he modes of loading and the shapes that resist them well. Only the last material has properties but no shape, when we shape the material such as the I beam in lastic bending ness is given by SB, the force per unit displacement as in the next S Fig. 7.2 shows t three are affected by shape and we will only deal with bending and a little about compression. To deal with these we need to define shape factors, which measure the efficiency of a shaped section for each mode of loading. A Fig. 7.3 then we need to introduce the shape factor, φ, to describe the performance of the material in that application. All shape factors are described in comparison to a circular cross section rod which is taken as having φ of 1. We consider φe B and φf B for elastic bending of beams and failure of beams respectively. E The bending stiff equation: 3 1 l EICSB = , where C1 is a constant that depends on end constraint (see Appendix A p. 381), I is the second moment of area about the axis of bending, l is the span of the beam and E is Young’s modulus. The I’s can be found in Table 7.1 on p. 164. By 16 comparison to a round beam of solid section we can then define the shape factor as SB for 17 Material indices that include shape. If we again consider the problem of a beam of minimum weight to support a given load as was done in chapter 5. The material index to be maximized, M1, is just ρ 5.0E . If materials can have equivalent shape factors then this material index will suffice. However, if we wish to select a material shape combination for the beam then we must include the material factor φe B, and the factor M1 becomes ( ) ρ φ 5.05.0 e BE . We can use the factors in Table 7.3 for this purpose. An example of this calculation for four materials is given in Table 7.4. We see that without the shape factor (column 5) wood is the best material for a beam (we use it to build houses). However, when maximum shape factors are included Al becomes the best material (hence its use in aircraft). Steel is not as good but is more than 50% better than wood and when cost is also considered steel is found to be quite competitive. Other factors such as durability in the environment and construction costs can then become decisive. A graphical method is illustrated in Fig. 7.12. The shape factor is introduced as moving the point along a line of slope 1 by the value of the shape factor hence increasing its performance. For more details see section 7.7 p. 186. 20 Table 7.4 The selection of material and shape for a light, stiff, beam Material p E Prax EM (PmacE )Y? Mat GPa Pp p 1020 Steel 7.85 205 65 18 14.7 6061-T4 Al 2.7 70 44 3.1 20.5 GFRP (isotropic) 1.75 28 39 2.9 19.0 Wood (oak) 0.9 13.5 5 41 OL *“deyax Means the maximum permitted value of ¢ from Table 7.3. 1000 Modulus - Density 100 Young's Modulus, E (GPa) 0.1 0.01 0.1 1.0 Density, p (Mg/m®) Fig. 7.12 Schematic of Materials Selection Chart 1: Young’s modulus plotted against density. The best material-and-shape for a light, stiff beam is that with the greatest value of £1? /p, The structured material behaves in bending like a new material with modulus E* = E/ and density p* = p/@ (where @ means 3) and can be plotted onto the charts. All the material-selection criteria still apply. A similar procedure is used for torsion. 21