Designing Capable and Reliable Products Episode 1 Part 4 pot

Figure 2.4 Geometry to process risk chart, g p Notes ¬ Allowances should be made for tolerances across the parting line. Allowance may also be required for mismatch of the dies/moulds in some casting and forging processes. Flash thickness allowances may also be required in closed die forging. Note, this only applies to casting, moulding and forging processes.  Refers to the number of orthogonal axes on which the critical characteristics lie, and which cannot be achieved by processing from a single direction. ® The processing of components that are on the limits of technical feasibility is likely to result in out of tolerance variation. High forces and ¯ow restriction in metalworking and metal cutting processes can lead to instability. Also, material ¯ow in casting processes, where abnormal sections and complex geometries are present, can lead to variability problems and defects. ¯ Slender unsupported regions with large length to thickness ratios are highly liable to distortion during processing. ° Repetitive, irregular or non-symmetrical features require greater process control and complex set-up or tooling requirements. This can be an added source of variability. ± There is all increased risk of variation each time a new set-up operation is required due to changes in the orientation of the part or tooling. Component Manufacturing Variability Risks Analysis 45 Figure 2.5 Tolerance to process risk chart, t p 46 Designing capable components and assemblies Figure 2.6 A sample set of process capability maps Component Manufacturing Variability Risks Analysis 47 Validation studies in manufacturing businesses and discussion with experts led to the view that knowledge used to de®ne m p , g p , t p and s p , could be structured such that q m may be formulated as: q m  t p  s p 2:2 where: t p  f design tolerance; m p ; g p 2:3 (See Figure 2.5 for the complete formulation of t p .) A link between the material used and the geometry of the component is com- pounded in the formulation for the tolerance to process risk as shown in equation 2.3. It is recognized that increasing material incompatibility and geometry complexity Figure 2.7 Surface roughness to process risk chart, s p 48 Designing capable components and assemblies has the eect of increasing the variability associated with achieving the dimensional tolerance requirement. The above equation relates this notion to the tolerance process risk by dividing the design tolerance for the characteristic, as stated by the designer, by the product of m p and g p . The risk index for surface roughness, s p , usually defaults to unity unless a surface roughness requirement is considered critical, for example a valve seat or lubricated surface. Figure 2.8 Surface roughness risks for a number of manufacturing processes Component Manufacturing Variability Risks Analysis 49 The underlying notion of the Component Manufacturing Variability Risk Index, q m , is that an ideal design exists for a component where the risk index is unity, indicating that variability is in control. Risk indices greater than unity exhibit a greater potential for variability during manufacture. The resulting value of q m indicates the risk of out of tolerance/surface roughness capability when compared to an ideal situation. For the ideal design of a component and processing route, each of the quantities is unity and therefore in all cases q m ! 1. Tolerance, surface roughness, material and geometry designed into a component, which is not matched with the ideal, have an eect on variability. For example, in die casting there is a higher risk when processing copper alloys compared with the tolerance capability resulting from processing zinc-based materials, largely due to temperature eects. Additionally, there are other manufacturing processes (for example, tempering and nitriding) that must be considered in the analysis if used in the product's manufactur- ing route. These processes are carried out after the primary/secondary processes have been used to manufacture the component and are treated as post-manufacturing processes. The potential for variability in the ®nal component due to these processes is great, due to the possible combination of high temperatures and unsymmetrical sections, which are particularly likely to cause out of tolerance variations. This introduces an additional factor to consider, based on: . Surface engineering processes (bulk and surface heat treatment/coating processes). Surface engineering processes are usually performed after the primary shape genera- tion of the component, or post-manufacturing, therefore q m defaults to k p when it is considered in the manufacturing route: q m  k p 2:4 where: k p  surface engineering process risk: Figure 2.9 shows the surface engineering process risk chart, k p . It includes the key variability issues related to these types of process. Validation of the predictions for process capability through the use of the compo- nent manufacturing variability risks analysis, q m , is given later. 2.2.1 Process capability maps As can be seen from the above, central to the determination of q m is the use of the process capability maps which show the relationship between the achievable tolerance and the characteristic dimension for a number of manufacturing processes and material combinations. Figure 2.6 shows a selection of process capability maps used in the component manufacturing variability risks analysis and developed as part of the research. There are currently over 60 maps incorporated within the analysis covering processes from casting to honing. The full set of process capability maps is given in Appendix IV. Data on the tolerance capability of the manufacturing processes covered were com- piled from international standards, knowledge engineering in specialist businesses 50 Designing capable components and assemblies and engineering texts. A selection of references used to generate the maps is given in the Bibliography. The data used in the creation of the maps usually comes in the form of tables, such as that given in Figure 2.10 for machining using turning and boring. International Organization of Standards (ISO) tolerance grades are com- monly used as a straightforward way of representing the tolerance capability of a Figure 2.9 Surface engineering process risk chart, k p Component Manufacturing Variability Risks Analysis 51 manufacturing process. The lower the tolerance grade, the more dicult the attain- ment using the particular manufacturing process. The tolerance grades are interpreted using standard tables (BS EN 20286, 1993) for conversion into dimensional tolerances. However, the tolerance grades do not take into consideration dierent materials machined or the complexity of the component being processed. Both unilateral and bilateral tolerances are encountered in practice. A unilateral tolerance permits variation in only one direction from a nominal or target value; a bilateral tolerance permits variations in both directions. Most tolerances used, unless stated otherwise, in the generation of the maps are bilateral or Æt, where `t' is half of the unilateral tolerance, T. Bilateral tolerances are a common way of repre- senting manufacturing process accuracy, although some processes are more suited to other tolerance representations. For example, forging requires that the total tolerance or unilateral tolerance is divided  2 3 T, ÿ 1 3 T, and drilling has a positive tolerance only, T, the negative tolerance from target being negligible. This is catered for in the representation of the tolerance data in the process capability maps. After plotting the tolerance data, it is useful, in the ®rst instance, to set the bound- ary conditions as A  1 corresponding to a dimension/tolerance combination that is of no risk, and A  1:7 on the interface of acceptable/special control region. The data used in the creation of the maps spans these two conditions, that is, the region where the process consistently produces the required tolerance. This is shown in Figure 2.11 for the turning/boring data taken from the ISO tolerance grades and many other references. The risk index A  1:7 was taken from initial work in this area, where the empirical values for the component manufacturing variability risks determined were compared to historical c pk data (Swift and Allen, 1994). The intermediate values for `A' are derived from the `squared' relationship that is analogous to that of the relative cost/diculty trend exhibited by manufacturing pro- cesses and their tolerance capability (see Figure 2.12). A target process capability value, C pk  1:33, is aligned to the risk value at A  1:7. Values for `A' greater than 1.7 indicated on the maps continue with the squared Figure 2.10 ISO tolerance grades for machining processes (adapted from Green, 1992) 52 Designing capable components and assemblies relationship, therefore 1:7 2  3and3 2  9. It follows then that risk indices of `A' greater than 1.7 would not be process capable. In essence, the spacing of the lines A  1toA  9 represent decreasing percentages of the tolerance band at any given dimension as the value of `A' increases. However, the log±log axis as used on the maps show the dierence as a linear step. A further development of the use of the maps is that the `A' values can be interpolated between A  1 and A  9 values bounded on the map. This ultimately improves the accuracy in determining the risk value. Therefore, to determine the tolerance risk value `A', look along the horizontal axis until the characteristic dimension is found, and locate the adjusted tolerance on the vertical axis. Read o the `A' value in the zone at which these lines intersect on the map by interpolating as required between the zone bands, A  1toA  9. The knowledge contained in the maps is also useful in determining the tolerance requirement at an early stage in the detailed design process. In this capacity, the region of process capable tolerance is bounded by two bold lines at A  1 and A  1:7 on the maps. Of course, this does not take into consideration the material and geometry eects initially, for example parting line allowances. Reference to Swift and Booker (1997) can be made for approximate parting line allowances. In most cases, guidance is also given on the maps for the need of a secondary process if the dimension/tolerance combination de®ned gives a risk index greater than 3 (which is considered to be out of manufacturing control). Constructing process capability maps from manufacturing data The procedure shown in Figure 2.13 can be used by a company to construct their own process capability maps. It is necessary only when a new or specialized manufacturing process is to be used, which is not contained in Appendix IV, and when data from the Figure 2.11 Employment of the tolerance data in the generation of a process capability maps Component Manufacturing Variability Risks Analysis 53 machine tool supplier is not available. Of course, this activity may become prohibitive when the costs involved and/or time in performing such studies are high. The data used to generate the maps is taken from a simple statistical analysis of the manufacturing process and is based on an assumption that the result will follow a Normal distribution. A number of component characteristics (for example, a length or diameter) are measured and the achievable tolerance at dierent conformance levels is calculated. This is repeated at dierent characteristic sizes to build up a relationship between the characteristic dimension and achievable tolerance for the manufacture process. Both the material and geometry of the component to be manu- factured are considered to be ideal, that is, the material properties are in speci®cation, and there are no geometric features that create excessive variability or which are on the limit of processing feasibility. Standard practices should be used when manufacturing the test components and it is recommended that a number of dierent operators contribute to the results. 2.2.2 Surface roughness chart Figure 2.8 shows the range of surface roughness values likely for various manufac- turing processes. The ranges determined are bounded within the risk index, A,in the same way as the process capability maps, because a similar cost±surface ®nish relationship exists, as suggested for tolerance and cost. This is shown in Figure 2.14 for several machining processes. The ®ner the surface ®nish required, the longer the manufacturing time, thereby increasing the cost (Kalpakjian, 1995). Figure 2.12 Modelling tolerance risk using a `squared' relationship 54 Designing capable components and assemblies [...]... Figure 2 .13 Process capability map construction ¯ow chart 55 56 Designing capable components and assemblies Figure 2 . 14 Relative manufacturing time as a function of surface roughness for several machining processes (BS 11 34, 19 90) For a given manufacturing process, shown on the vertical axis in Figure 2.8, and design surface roughness, shown on the horizontal axis, the risk index `A' on the shaded band... Reduce part count and types Modularize the design Assembly capability Strive to eliminate adjustments (especially blind adjustments) Design parts for ease of handling (from bulk) Design parts to be self-aligning and self-locating Ensure adequate access and unrestricted vision Design parts that cannot be installed incorrectly Use ecient fastening or ®xing techniques Minimize handling and reorientations... trying to 61 62 Designing capable components and assemblies Figure 2 .16 Assembly sequence diagram for a castor wheel Component Assembly Variability Risks Analysis balance the assembly line In turn, the choice of assembly sequence and the identi®cation of potential subassemblies can a€ect or be a€ected by product testing options, market responsiveness and factory ¯oor layout (Baldwin et al., 19 91) , as well... coecient, r ˆ ÿ0:956, and the corresponding power law is given by: 3:9 81 …2:7† Cp ˆ 1: 332 qm This can be approximated by the following equation: Cp % 4 4=3 qm …2:8† It is possible, therefore, to determine an estimate for Cpk from the formulation given above in equation 2.6, within some con®dence It is assumed that the Cpk values given 57 58 Designing capable components and assemblies Figure 2 .15 Empirical relationships... et al., 19 94) Computer Sciences Corporation's (CSC) DFA/Manufacturing Analysis (MA) (CSC Manufacturing, 19 95) Hitachi's Assembly Evaluation Method (AEM) (Shimada et al., 19 92) In fact, the use of DFA techniques is now mandatory in some companies, such as Ford and LucasVarity (now TRW) (Miles and Swift, 19 98) These techniques o€er the opportunity for a number of bene®ts, including: Reduced part count... application and systematic approach is essential as there are many subjective processes embedded, but many companies have found them to be pivotal techniques in designing cost-e€ective and competitive products (Miles and Swift, 19 98) An overview of each of the main commercial methods can be found in Huang (19 96), but in general, a number of design for assembly guidelines can be highlighted (Leaney, 19 96b):... or the single characteristic are capable By identifying components with high assembly risks and potentially high failure costs, further design e€ort is highlighted and performed in order to identify the associated tolerances, for example clearance for the optimal ®t and function of the components 2.3 .1 Design for assembly techniques Early work looking at designing products for mechanized assembly started... costing and process selection Lower component and assembly costs Standardized components, assembly sequence and methods across product `families' leading to improved reproducibility Faster product development and reduced time to market Lower level of engineering changes, modi®cations and concessions Fewer parts means: improved reliability, fewer stock costs, fewer invoices from fewer suppliers and possibly... approximated to: 4: 093 q2:0 71 m …2:5† 4 q2 m …2:6† Cpk % Note that the `squared' relationship which was initially used to model the degree of diculty in obtaining more capable tolerances for a given manufacturing route and product design is being returned by the power law Similarly, a relationship between the process capability index Cp and qm for the components analysed is shown in Figure 2 .15 (b) The data... least potential variability problems or least failure cost should be chosen for further development The indices should not be taken as absolutes as assembly variability is dicult to measure and validate The component assembly variability risk, qa , as determined by CA, attempts to better understand the a€ects of the assembly situation on variability by quantifying 63 64 Designing capable components and . for several machining processes (BS 11 34, 19 90) 56 Designing capable components and assemblies of collaborating companies which had produced the components and had measured a critical characteristic. sheet Figure 2 .15 Empirical relationships between (a) q m and C pk and (b) q m and C p (with 95% con®dence limits) 58 Designing capable components and assemblies metal, pipes, wires and a variety. squared Figure 2 .10 ISO tolerance grades for machining processes (adapted from Green, 19 92) 52 Designing capable components and assemblies relationship, therefore 1: 7 2  3and 3 2  9. It