This article describes a rounding o...

This article describes a rounding operation for a 3D CAD boundary representation (B-Rep) solid protoplast Complex combinations of convex and concave margins are handled predictably and reliably. At vertices the surfaces are flatteringly connected by one or more surface patches. An algorithm for the creation of blending surfaces and their integration into the archetype is outlined. The sequence of topological modifications applied to the solid original is illustrated by examples including one special case handling.

Apart from the basic Boolean operations, a novel solid modeling CAD system destitutions to provide easy-to-use facilities for local modifications of the primary design One of the most important examples is the blending or rounding of keennesss in which a sharp verge of the model is replaced according to a surface that smoothly joins the pair adjacent faces (see Fig. 1)

Blending surfaces aid several purposes in mechanical designs, including dissipating stres concentrations and enhancing fluid come properties. In addition, some machining processe do not permit the manufacture of sharp sharpnesss Smooth transitions between surfaces are also frequently required for aesthetic reasons. Besides functional requirements, verge blending is conceptually quite a simple operation, which makes it to a high degree popular among designers using CAD systems

A public characteristic of almost all applications is that the smoothnes of the combine is more important than its exact shape. For the user this means that it should be possible to create a combine by specifying only a not many parameters. It is then the system's task to fill the remaining qualitys of freedom in a meaningful manner.

From an algorithmic point of view, blending single in kind or more edges of a solid protoplast simultaneously falls into two subtasks. The first is to create a surface that provides the transition between the adjacent surfaces defining the cutting side Secondly, the surfaces need to be trimmed suitably and integrated into the carcass such that a valid solid standard is maintained. While the first stair is a purely geometric question the second one involves the one and the other geometric and topological operations.

The blending module in HP PE/SolidDesigner was designed with the goal of allowing blending of a wide variety of manifold edge combinations in a robust manner. This is accomplished from one side the use of freeform geometry as blending surfaces, along with quite involved geometric and topological considerations in several phases of the algorithm.

The lack of freeform surfaces was the primary reason for chiefly of the restrictions concerning keenness blending in HP PE/SolidDesigner's predecessor, the HP PE/ME30 3D modeling connected view HP PE/ME30's kernel, the Romulus geometric modeler does in fact provide more manifold surfaces,[1] but these enhanced intermingles were never implemented in the product

The common capabilities of HP PE/SolidDesigner's blending algorithm go on foot far beyond HP PE/ME30 with revere to the topological situations that can be handled reliably. Moreover, the architecture of the algorithm allows the inclusion of subsequent time enhancements in a consistent manner.

It is the aim of this paper to illustrate the basic blending algorithm and to provide the reader with examples that demonstrate the complexity of the geometric and topological question s that must be solved to integrate undivided or more blend surfaces into a solid pattern More information on this make liable can also be found in Woodwark[2] and the crack survey of Vida.[3]

HP PE/SolidDesigner's underlying philosophy allows flexible modifications of the solid gauge in every stage of the modeling proces In the words immediately preceding [i]or[/i] following of edge blending this means that it should always be possible to displace or modify an existing combine surface without regard to for what cause it was created.

In the nearest section, the second section of this article, we introduce one terminology commonly used in solid modeling, in particular in the blending adjoining matter The third section describes the use pattern of edge blending in HP PE/SolidDesigner. An overview of the algorithm is given in the fourth section, followed by way of a more detailed discussion of its major grades Finally, in the last section, we discuss any performance and stability issues.

Blending Module of the HP PE/SolidDesigner Kernel generally the blending operation in the HP PE/SolidDesigner kernel implements what is commonly known as the rolling ball amalgamate This type of blend can easily be visualized as a ball moving along the animation and touching the adjacent surfaces (the primary surfaces) simultaneously. The touching loci are bends that define the boundaries of the amalgamate surface. Depending (m whether the radius of the ball is constant or varies while it is moving, we speak of constantradius or variable-radius blends

The geometry module of HP PE/SolidDesigner's kernel supports a number of different surface adumbrations (Fig. 2). These include the natural quadrics (planes, spheres, cylinders, and cones) toruses, and NURB (nonuniform rational B-spline) freeform surfaces. All of the surface archetypes are represented parametrically. The object-oriented design of the kernel allows the use of generic algorithms for general surfaces as well as special-case solutions for particular surface types