Unit-3 Geometric Modeling REQUIREMENTS OF GEOMETRIC MODELLING Traditionally, product drawings were made together with the prototypes

s for passing the information across. However, in a computerized environment, the information a designer generates can form the basic unit to be accessed by a number of other elements of a CAM system, as explained in the first chapter. Hence, it is important that the geometric model generated should be as clear and comprehensive as possible so that the other modules of the modeling and manufacturing system are able to use this information in the most optimal way. The functions that are expected of geometric modeling are: Design analysis: Evaluation of areas and volumes. Evaluation of mass and inertia properties. Interference checking in assemblies. Analysis of tolerance build-up in assemblies. Analysis of kinematics - mechanics, robotics. Automatic mesh generation for finite element analysis. Drafting: Automatic planar cross-sectioning. Automatic hidden line and surface removal. Automatic production of shaded images. Automatic dimensioning. Automatic creation of exploded views for technical illustrations. Manufacturing: Parts classification. Process planning. Numerical control data generation and verification.\ Robot program generation. Production Engineering: Bill of materials. Material requirement. Manufacturing resource requirement. Scheduling. Inspection and Quality Control: Program generation for inspection machines. Comparison of produced part with design. Therefore, it becomes necessary for the geometric modeling system to provide complete information on all aspects related with the further use of the system and at the same time be simple and in tune with the designer's methods. The following properties to be desired of in any geometric modeling (solids) system1. The configuration of solid (geometric model) must stay invariant with regard to its location and orientation. 2. The solid must have an interior and must not have isolated parts. 3. The solid must be finite and occupy only a finite shape. 4. The application of a transformation or other operation that adds or removes parts must produce another solid.

5. The model of the solid in E3 (Euler space) may contain infinite number of points. However, it must have a finite number of surfaces which can be described. 6. The boundary of the solid must uniquely identify which part of the solid is exterior and which is interior.

GEOMETRIC MODELS The geometric models can be broadly categorized into two types: 1. Two-dimensional 2. Three-dimensional. The two-dimensional models were the first ones to be developed in view of their relatively lesser complexity. However, their utility is limited because of their inherent difficulty in representing complex objects. Their utility lies in many of the low end drafting packages, or in representing essentially two-dimensional manufacturing applications such as simple turning jobs (axi-symmetric), sheet metal punching or flame or laser cutting. Serious CAM applications would be extremely difficult in operation, if they start from the two-dimensional geometric modeling. Hence, hardly any application exists with only a two dimensional geometric modeling. In contrast, the three-dimensional geometric modeling has the ability to provide all the information required for manufacturing applications. There are a number of ways in which the three dimensional representation can be arrived at. The three principal classifications can be; 1. The line model, 2. The surface model 3. The solid or volume model.

GEOMETRIC CONSTRUCTION METHODS

A technical (production) drawing consists of a lot of information besides the simple view or the geometric representation. Examples are - the dimensions, tolerances, material specifications, the processing requirements, assembly requirements, etc. Many alternative views of the details in the object should be presented for clarity. The three-dimensional geometric construction methods which extend from the 2D that is normally used are: Linear extrusion or translational sweep, and Rotational sweep. whereas a true 3D modeling method is the primitive instancing or solid modeling 1. Sweep In linear extrusion, initially a two dimensional surface is generated and then swept along a straight line thus generating third dimensions as in Fig. It is possible to repeat the same technique for generating reasonably complex geometry. The sweep direction can be any three-dimensional space curve and need not be a straight line. The major advantage of the system is that in view of the varied

Another type of construction technique is the rotational sweep which can be utilized only for axisymmetric jobs as shown in Fig.

One of the reasons why sweep is the more used geometric construction in spite of its limitations is that it is the natural extension in which designers or draftsmen work. As a result, further variations in sweep are now being made available for generating more complex geometry. For example, it is possible to sweep in a linear direction with taper along the sweep direction. It is also possible to add twist to the sweeping in the third dimension. Similarly, the rotational sweep can be enhanced by the addition of axial and/or radial offset while sweeping to get helical or spiral objects. 2. Solid Modeling Though, the above techniques are capable of generating surfaces reasonably complex, they are not suited for the purpose of inputting geometry. The best method for the three-dimensional solid construction is the Solid Modeling technique, often called the Primitive Instancing or Constructive Solid Geometry (CSG). In this a number of 3-dimensional solids are provided as primitives. Some typical primitives utilised in the solid modelers are shown in Fig.

Though these are the analytical solid primitives generally used, the modeling will not be restricted only to these. It is possible to use any solid obtained by the other modeling tools in the system as a primitive. From these solid primitives, the complex objects are created by adding or subtracting the primitives as shown in Fig.

For combining the primitives to form the complex solid, the basic set operators, also called boolean operators, are used. They are of the following typesunion, intersection and difference. The effect of these operators on the primitives is shown in Fig. for the simple case of a block and cylinder shown in their 2-dimensional relationship. However, the same is true for 3-dimensional, as well as for any other complex orientation of the primitives and for any size.

3. Sculptured Surfaces There may be surfaces which cannot be defined by any of the analytical techniques available. A few examples are the car bodies, ship hulls, some die cavity surfaces, and decorative surfaces styled for aesthetics. An example of sculptured surface is shown in Fig. The only way these surfaces can be modeled is through a series of control points and other boundary conditions which specify the nature of the surface desired. There are a large number off numerical techniques available in such cases.

4. Skinning or Lofting In this method, a number of two-dimensional arbitrary or regular profiles are placed along an arbitrary three-dimensional space curve (spline) and then skinning is done on all these profiles. This method is extremely useful for modeling engine manifolds, turbine blades, air frames, volute chambers and the like. Such solids are otherwise difficult to produce by any off the previously described modelling methods. This facility is now generally available in all commercial modelling software systems.

5. Miscellaneous Construction Methods A few other geometry construction techniques that are needed to be a part of the modelling system are: Filleting This is the ability to automatically generate the fillet radius between two surfaces, either analytical or sculptured. The radius could be uniform or vary linearly, depending upon the meshing surfaces. If the same fillet is to be generated through the solid modeling approach, a large amount of input is needed in terms of the difference of a number of solids that would form a regular fillet surface. Tweaking This is the ability to alter a model already created using any of the earlier approaches described. In this, a face or a vertex in the model would be interactively moved to see the effect in the modification of the geometry. For example, in the case of sculptured surfaces, a point on the surface can be changed, and the surface modification observed for the purpose of styling. However, when tweaking is applied to CSG models, then certain problems are involved. Since the CSG representations retain the way the job was modeled using the primitives, any tweaking done may have unpredictable results because the effect obtained depends on how the original model was developed. MODELLING FACILITIES DESIRED The total modelling facilities that one would look for in any system can be broadly categorized as follows: The geometric modelling features. The editing or manipulation features. The display control facilities. The drafting features. The programming facility. The analysis features. The connecting features A. Geometric Modelling Features The various geometric modelling and construction facilities that one should expect to have in any good system are as follows: 1. Various features to aid geometric construction methods, such as Cartesian and polar coordinates, absolute and incremental dimensions, various types of units, grid, snap, object snap, layer, etc. 2. All 2D analytical features, such as points, lines, arcs, circles, conics, splines, fillets, chamfers, etc. In each of these features, various constructional features including interactive and dynamic dragging facilities. 3. Majority of the 3D wireframe modelling facilities including 3D lines, 3D faces, ruled surfaces, linear sweep from 2D topology with any sweep direction, rotational sweep, and tapered sweep. General sweep with twist. Revolving about an axis with axis or radial offset for generating helical or spiral shapes. 4. Solid modelling with various basic primitives such as block, cylinder, sphere, cone, prism, torus, pyramid, quadrilateral, along with the ability to apply the boolean operators on any solid that can be constructed using the other techniques available in the modeler. 5. Skinning around regular and arbitrary surfaces. Profiles (cross-sections),both analytical and arbitrary placed across any 3D curve.

6. Sculptured surfaces of the various types like Bezier, Coons and other free form surfaces. 7. Comprehensive range of transformation facilities for interactively assembling the various solid models generated by the modeler with features such as surface filleting and trimming.

B. Editing or Manipulation Features 1. Transformations such as move, copy, rotate, scale, elongate or compress, b mirror or to any arbitrary coordinate frame. 2. The editing features used to alter the already drawn geometric entities, such as stretching, trimming or trimming to any intersection, delete or erase, undo or redo. 3. Symbols in drawing refer to often-repeated set in a number of drawings, which may consist of a number of geometric entities that are grouped together and stored as a symbol. This symbol can be recalled at any scale, at any angle or exploded if necessary to treat all of them as separate entities. C. Display Control Facilities In this range of features are all the facilities needed for interacting with the modelling system so as to obtain the necessary feedback at the right time during the niodelling stage. The facilities required are: 1. Window - to identify a set of entities for any possible display or editing function. 2. Zoom - to change the scale of display of the image selected on the screen. 3. Pan - to move the image on screen without changing the scale at which the drawing is displayed on the screen. 4. Hidden - to remove hidden lines or hidden surfaces for viewing the geometry in proper form 5. Shading - to show the 3D view of the image on screen complete with the light source location and the resultant light and shade as it appears on the image.

D. Drafting Features These facilities refer to the way the model developed can be utilized for the purpose of transmitting the information in hard copy form for other applications, such as part prints onto the shop floor, or maintenance manuals for the equipment. A really large range of facilities are required in this particular category, and it is sometimes treated as a separate module in the modeling system.

A large number of types of views should be obtained from the solid model of the geometry stored in the data base. The types of views required may be , as for display functions, such as perspective views, orthographic views (Fig.), isometric views, and axonometric views.

E. Programming Facility Programming ability (MACRO programming) within a CAD system is going to be a very useful feature. It is well-known that not all kinds of facilities would be available in any general purpose CAD system. Therefore, it is necessary that the CAD system would have to be customised for a given range of application processes specific to the company. For this purpose, if a programming facility exists in a CAD system, it is possible to program specifically for an application, making use of all the features available in the system for either modelling or for any specific application based on the information generated during the modelling. Some such examples are, the GRIP in Unigraphics and GLUE in CAM-X. The availability of such a program helps the user to input the least amount of information for any required design, if the application programs are written well using the programming language. F. Analysis Features In this range, the kind of analysis facilities that are required to be carried on the product models being generated should be considered. The simplest kind to the most sophisticated features may be available under this category. The simplest facilities may be calculating perimeter, area, volume, mass, centre gravity, moment of inertia, radius of gyration, etc. G. Connecting Features

Modelling is only the start of the complete process of a product evolution, and as such the data generated is used directly by the other systems. It is, therefore, necessary that the internal data format in which the data is stored by the modelling system should be well documented and should also have very good connectivity (data interfacing) with other allied modules. Ideally, an integrated data base structure would be useful wherein all the various modules share the common data base. However, this would only be possible if all the modules are developed at a single developer as in the case of Computer vision or Unigraphics for CAD/CAM integration.