‘ventaserngcanrcaw | wGeametie Mcsing | 8 Soles © ecw | { cuessiassand lashes besiansiedbassiesbesdlesiessaasuanieay) Chapter 9 Solids 347 Introduction Solid models are known to be complete, vali, and unambiguous representations of objects. Simply stated, x complete solid is one which enables point in space to be classified relative to the object, if itis inside, outside, or on the object. This elassific called spatial addressability. A vali solid is one that doesnot have dangling exiges or faces. An ‘unambiguous solid has one and only one interpretation. Solid modeling achieves completeness, validity, and unambiguity of geometric models. Therefore, the automation of tasks such as interference analysis, mass. property calculations, finite element modeling and analysis, ‘Computer Aided Process Planning (CAFP}, maemine vision, ana IMC machining s possible. CAD systems offer two approaches to creating solid models: primitives and features. The former approach allows designers to use predefined shapes (primitives) as building blocks (like Lego pieces) ro create complex solids. Designers must use Boolean operations 10 eombine the primitives. This approach is limited by the restricted shapes ofthe primitives. The Features are ‘more flexible as they allow the constuction of more complex and elaborate solids than what the primitives offer. Some CAD systems (such as Unigraphies, CATIA, and I-DEAS) offer both approaches, while oshers (such as SolidWorks and ProvE) offer ooky the features approach, ‘Consider the object shown in Figure 9.1 to illustrate the two approaches. We ean create block and subtract six cylinders from it wsing the primitives approach. Or, we can create a rectangle with six cirles inside it in the Top sketch plane and extrude it using the features "approach. The resulting solid is the feature in this case Figure 9.1 Atypical solid model. ‘The features approach is considered by some to he a generaization of the primitives approach. Systems that offer dhe features hide away the Boolean operations from their users These operations must be used to create the final solid model. A ozs command isa form of Boolean union operation, and a cut command isa form of @ Boolean subtract operation ‘This chapter covers the theoretical and practial aspects of solids. Throughout the chapter, ies related to constructing solids on CADICAM systems are coveredTeienseingcaDcawt | Wésemac Med | 8 Sa Tew Ee ae eee EDD 348 Chapter 9 Solids BD Geometry and Topology ‘A solid model of an object consists of bath the tcpological and geometrical data of the object. The completeness and unambiguity of a solid model are attributed to the fat that its database stores both its geometry and its topology. The difference between geometry and topology is illustrated in Figure 9.2. Geometry (sometimes called metre information) is the ‘actual dimensions that define the entities of the object. The geometry that defines the object shown in Figure 9.2 isthe lengths of lines CZ, and L,, the angles between the lines, and the radius R and the center P ofthe half etc, no Se ne” de, Figure 9.2. Difference between geometry and topology. ‘Topology (sometimes called combinatorial structure) is the connectivity and associativity of the object entities. It has to do with the notion of neighborhood: that is, i determines the relational information between object entities. The topology of the object shown in Figure 9.25 can he stated as follows: L shares a vertex (point) with Lp and C), Lp shares vertex with Ly and 1, L; shares a vertex with £3 and C), Land Ly do not overlap, and P, les outside the object. Based on these definitions, neither geometry nor topology alone can completely define objects ‘While solid models are complete and unambiguous, they are not unique. An object may be constructed in various ways. Consider the object shown in Figure 9.3. Using the primitives approach, one ean construct the solid model ofthe object by dividing it into two blocks and a cylinder, We ean add the two blocks frst and then subtract the eylinder (Figure 9.3), or we ean subtract the cylinder from block and ada the other block to the resulting subsoli (Figure 9.3). Figure 9.4 shows two altematives (create different crass sections and extrude them) if we use the features approach. Regardless ofthe order and method of construction, the resulting solid model‘natin cat [Mtwtmaieg | ena beat Chapter 9 Solids 349 ofthe abject is always complete and unambiguous. However, there will always be one way that ‘more efficient than others construct sold models as isthe case with curves and surfaces nore owes matt oe off (Abentve sh al of tet Figure 93 Nonuniqueness of solid model 2° 7 xnapofiesavsbwmanat cre pleB ude Anse and ad cur ple B the primitives approach. Figure 9.4 Nonuniqueness of solid models: the features approach. QD solid Entities ‘The entities we use to ereate solid models depend onthe approach we use. The primitives approach uses primitives and the features approach uses sketches. Many CAD systems provide both approaches to increase their modeling domain. The entities required for sketches are curves and surfaces, which we covered in Chapters 6 and 7, This section covers the basies of primitives.‘ventaserngcanrcaw | wGeametie Mcsing | 8 Soles © ecw | Ee ae eee EDD 350 Chapter 9 Solids Primitives are considered building blocks. Primitives are simple, basic shapes which ean bbe combined by mathematical set of Boolean operations to create the solid. Primitives themselves are considered vali off-the-shelf solids. The user usually positions primitives as required before applying Boolean operations to cansiruct the final solid ‘There is wide varity of primitives available commercially to users. However, the four ‘most commonly used ones are the block, eylinder, cone, and sphere, These ae based on the four natural quadrics: planes, eylinders, cones, and spheres. For example, the block is formed by intersecting six planes. These quadries are considered natural because they represent the most commonly accurting surfaces in mechanical design which can he prodnced hy rolling, mening nilling, cutting, drilling, and other machining operations used in industry Planar surfaces result from rolling, chamfering, and milling: cylindrical surtues ftom turing or fileting: spherical surfaces ftom eutting witha ball-end eutting tool; conical surfaces from turning as well as from drill tps and countersinks. Natural quadies are distinguished by the fact that they are combinations of linear motion and rotation. Other surfaces, except the tons, requite at least dual axis control From a userinput point of view and regardless of a specifie system syntax, a primitive requites a set of location data, a set of geometric data, and a set of orientation data to define it completely, Location data entails a primitive local coordinate system and an input point defining is origin. Geometeical data difers from one primitive to another and is user inpur. Orientation data is piv primitives property «clative tthe MCS or WCS of ly used o¥ the solid model under construction. Primitives are usually translated! andor rotated to position and oriented properly before applying Boolean operations. Following are descriptions of the ‘most commonly used primitives (refer to Figure 9.5): 1. Block, This is @ box or cube whose geometrical data isis width, height, and depth. Its local coordinate system X;¥;Z;, is shown in Figure 9.5, Point P defines the origin of the X,¥,Z system, The signs of W, H, and D determine the position of the block relative to its coordinate system. For example, a block witha negative value of Wis displayed as ifthe block shown in Figure 95 is mirtored about the ¥;Z; plane. Cylinder. This primitive is aright citcular cylinder whose geometry is defined by is radius R (or diameter D) and length #. The length H is usually taken along the direction of the Z axis. H ean be positive or negative 3. Cone. This is aright cccular cone or a frustum of aright cteular cone whose base diame- ter R, top diameter (for truncated cone}, ad height Hare user-defined, 4, Sphere. Ths is defined by its radi R or diameter D and is centered ahout the origin ofits local eoordinate system, 5. Wealge. Ths isa right angled wedge whose height £7, width W, and base depth D form its geometric data 6. Torus, This primitive is generated by the revolution ofa circle about an axis Lying in its plane Z; axis in Figure 9.5. The torus geometry can be defined by the radius (or diameter)‘ventaserngcanrcaw | wGeametie Mcsing | 8 Soles © ecw | { cuessiassand lashes besiansiedbassiesbesdlesiessaasuanieay) Chapter 9 Solids 351 ‘of its body R, and the radius (or diameter of the centerline ofthe torts body Ra, the ‘geometry can he defined by the inner radius (or diameter) Rand outer radius Bo, All shese primitives ean be created using the features approach. They are all 24D objects The block, cylinder, and wedge are uniform thickness. The cone, sphere, and torus are axisymmetric. This explains why some CAD systems such as Pro“E and Solid Works do not offer them — the user can generate them via sketching, This simplifies software development as there {sno need to write separate primitives" funetions, Cote Figure 8.5 Common primitive For ll the primitives, there are default values forthe data defining their geometries, Most CDA systems use default values of 1, In addition, the local coordinate systems for the various primitives shown in Figure 9.5 may change from one system to another. Some systems assume‘natin cat [Mtwtmaieg | ena aie 352 Chapter 9 Solids that the origin, P, of the local coordinate system is coincident with that ofthe MCS or WCS and require the user to translate the primitive tothe desired locaton, thus eliminating the input of point P by the user, ‘Two or more primitives can be combined to form a solid. To ensure the validity of the resulting solid, the allowed combinatorial relationships beoween primitives are achieved via Boolean (or set) operations. The available Boolean operators are union (or +), inerseetion (or and difference (-). The union operator used to combine or add together wo objects or primitives. Interseeting «wo primitives gives a shape equal 10 their common value. The Sifscuce epesator is used te subuact one object fiom the ether and 2esults i a shape equal to the difference in their volumes. Figure 9.6 shows Boolean operations of a block A and a cylinder B. bIGBld (a) Te dneiot aba0 Figure 9.6 Boolean operations.TeienseingcaDcawt | Wésemac Med | 8 Sa Tew { cuessiassand lashes besiansiedbassiesbesdlesiessaasuanieay) Chapter 9 Solids 353, Solid Representation old representation ofan object can suport selinbly and automatically ot est in theory, relued design and manufituring applications due to its informational completeness. Such representation is based fundamentally onthe fact that physical objet divides an n-dimensional Space, into two regions, interior and exterior, separated by he object boundaries. A region i defined a a portion of space E*, andthe boundary ofa region i closed surface, as inthe case of a phere, ora collection of open surfaces connected a proper edges sin he case of box In ferns ofthe above notion, slid model of an object is defined muthenaticaly ax 8 point set § in 3D Euclidean space (£7), If we denote the interior ond boundary of the set by iS tnd bS respectively, we can write sais bs on And if we let the exterior be defined by cS (eomplement ofS). then Weis bs os 02) ‘where W isthe universal set, which i the ease of £* is all possible 3D points. ‘The solid detinition given by Eq, (9.1) introduces che concept of eometrc closure, whieh implies that the interior of the solid is geometrically closed by its boundary. That is, the boundary acts tight “skin” that i entirely in contact with the solid interior. Thus, Bq, (9.1) ean be rewritten as yaks os where kis the closure ofthe solid (point set 8}. Comparing Eqs. (9.1) and (9.3) gives kS as ks= is bs 04) Figure 9.7 shows the geometric explanation of equations (9.1) t0 (9.4, Figure 9.7 Closure of a solid.‘ventaserngcanrcaw | wGeametie Mcsing | 8 Soles © ecw | 354 Chapter 9 Solids ‘The properties tht a solid model should capture mathematically ae: a. Rigidity. This implies thatthe shape of a solid mode! is invariant and does not depend on the model's locaton or its orientation in space +b, Homogeneous three-dimensionality. Solid boundaries must be in contact withthe interior. No isolated or dangling boundaries (see Figure 9.8) should be permite. Finjteness and finite deseribability. The former property means that the size ofthe solids not infinite, while the latter ensures that imited amount of information can Uescribe the solid. This latter property is needed to be able to store slid modes in computers whose storage space is always limited. t should be noted thatthe former property doesnot include the later and vice versa, Foc exaumple, cylinder which may have finite radius and length may be described by an infinite number of planar faces Closure under rigid motion and regularized Boolean operations. This property ensures that manipulating solids by moving them in space or changing them via Boolean operations must produce other valid solids. «Boundary determinism. The boundary of «solid must contain the solid and hence ‘must determine distinctively the interior ofthe slid ‘The mathematical implication of these properties is that valid solid models are bounded, closed, regular, and semianalytic subsets of E°. These subsets ar called sos regularized sets). Intuitively, rsets are “curved polyhedra” with “well-behaved” boundaries. The point set $ that defines a solid model and is given by Eg. (9.1) i always an rset Intuitively, a “closed regular set” means that the set is closed and has no dangling portions as shown in Figure 9. and a “semianalytc set” means tha the set does not oscillate infinitely fast anywhere within the se. ‘The concept of "semianalytc set” is important in choosing equations fo describe surfaces or primitives of solid models. For example, the point set tat satisfies sin(x) <0 is semianalytic set, swhile the set that satisfies sin(/x) <0 is not because the Function sin(1/x) oscillates fast len. x approaches ze, Figure 9.8. Isolated boundaries of a solid.