الفهرس 
Introduction & Literature ReviewOnly 14 pages are availabe for public view
 |  | from | 78 |  |  |
| | | from | 78 | | |
Abstract
Various dimensional measurements are carried out to insure appropriate performance and functionality of different mechanical elements and engineering components. There is a large variety of well-established and special purpose conventional and contact measuring instruments designed for different dimensional measuring tasks. However, conventional and contact measuring equipment have their own limitations such as unavailability, vulnerability or sensitivity to environmental conditions, time consumption, and cost. The measured parts frequently impose their own measurement difficulties using conventional instruments due to their size, complexity, and fragility limitations. In this work a new approach is introduced to overcome the limitations of conventional and contact measuring instruments while maintaining their advantages. Conventional measuring techniques were simulated through novel algorithms and mathematical models. The developed algorithms were applied to the coordinates of a set of discrete points extracted from a 3D model reconstructed from the measurement of the actual part by photogrammetry. Among the various types of 3D reconstruction acquisition systems, photogrammetry was chosen due to its unique advantages and desirable characteristics. It is one of the most widely used techniques due to its flexibility, versatility, portability, competitive accuracy, and costeffectiveness. An imaging setup was used to capture successive images of the part of interest. Several image processing techniques were tested to enhance the captured images quality and, consequently, achieve a more accurate 3D reconstructed model. The 3D reconstructed model reflects the shape and dimensions of the actual part. Mathematical models were developed to simulate the function performed by the conventional measuring instruments taking into consideration the necessity of the coincidence between the direction of measurement and the dimension to be measured in any measuring process. The developed algorithms were applied to the 3D reconstruction of the actual part to evaluate the same dimensions measured by the conventional instruments. The proposed approach can be applied to a wide range of conventional measurements. In this work it was applied to the measurement of spur gear dimensions due to their high importance and wide utilization. Involute gears of all types and sizes are widely utilized in most machinery. They are considered as key components in most engineering equipment. A 3D model of the spur gear of interest was reconstructed using photogrammetry. The coordinates of the discrete points presenting the gear teeth flanks and the bore periphery were extracted. The bore center was determined based on the extracted bore periphery points coordinates. Different algorithms and mathematical models were developed to simulate the measurement of tooth thickness, tooth span by base tangent method, and over rollers. Performing measurements on the 3D reconstruction of the actual gear instead of the actual gear itself encouraged the development of mathematical models to evaluate dimensions not measurable by traditional techniques. A novel approach was developed to directly evaluate the base circle radius and geometrical center using only the actual teeth profiles. Accurate generation of each tooth flank involute profile from the nominally required base circle is most vital in gear manufacturing, inspection, and performance. However, the base circle is not a commonly used inspection parameter despite its vital importance in actual gear performance. The base circle radius and geometrical center cannot be obtained by vii conventional direct measurement instruments. Thus, an algorithm was developed to evaluate the base circle dimensions from its tangent lines. Prior knowledge of any gear dimension is not required, the developed mathematical models require only the measurement of the teeth profiles. A mathematical model was then presented to evaluate the base pitch, pitch circle radius, hence pressure angle, and concentricity error of the gear. The proposed algorithms were first applied on an ideal gear model constructed by solid modeling to check its accuracy and validity. The ideal gear model was designed with the same nominal parameters as the actual test gear. The proposed algorithms were then applied to the 3D reconstructed model of an actual manufactured spur gear obtained by photogrammetry. The gear dimensions evaluated using the proposed technique were also measured using conventional instruments. The gear dimensions obtained by the proposed simulated measurements were compared to their corresponding conventional measurements and nominal values. They were all found to be in good agreement. The proposed method provides a non-contact measurement technique well suited for inspection, reverse engineering, and in-situ measurements of part dimensions with diverse sizes and complexities.