Towards a New Concept of In-Line Crankshaft Balancing by Contact Less Measurement: Process for Selecting the Best Digitizing System

Abstract

Controlling the part’s balance of crankshafts are important issues for automotive manufacturers. Unbalance measurement is usually carried out using high-precision mechanical machines. The main objective of the present work is to replace mechanical measuring systems by a non-contact digitizing system, which permits the acquisition of the crankshaft surface. As the geometry to be measured presents a large variety of shapes and textures with accessibility issues, the definition of the best-suited scanning system related to geometrical and industrial constraints is a major issue. In this direction, the paper deals with the definition of a protocol based on quality indicators associated to the collected data to compare various digitizing systems. Those quality indicators are assessed thanks to simple artifacts measurement according to a specific procedure. The comparison protocol is applied to evaluate three triangulation based digitizing systems: Results allow us to identify well-adapted digitizing systems in relation to crankshaft balancing requirements.

A knowledge database of qualified digitizing systems for the selection of the best system according to the application

“…If such an approach seems interesting, the list of all the rules used is not clearly detailed. Moreover, the approach is only implemented for a laser scanner and a touch probe.For a specific application of crankshaft balancing, Zuquete et al [6] propose to select the best-suited system by assessing optical digitizing systemsthanks toindicators. Those indicators -noise, trueness, accessibility, and measured area, are obtained through a specific protocol based on the measurement of simple artefacts.Audfrayet al|7] extend this work by defining an assessment protocol applicable to a large range of digitizing systems and suggest storing information in databases.…”

Voxel-based Path Planning for 3D Scanning of Mechanical Parts

“…This value ensures a noise less than 3 µm for the ATOS system and around 15 µm for the KZ25 sensor. In these conditions, the trueness can be evaluated at 3.5 µm for both digitizing systems [11]. As expected the number of points of view is reduced for the large FOV sensor.…”

“…The choice of the admissible angle of view is given thanks to a first sensor assessment process to evaluate its quality in terms of trueness and of noise [11]. Results are shown in figure 6.…”

Conversion of 3D scanned point cloud into a voxel-based representation for crankshaft mass balancing

“…Several researchers had studied the sources of inaccuracy on 3D scanning systems and how to evaluate them [4][5][6]9].…”

“…The quality of the chosen scanning system will allow an error of less than 5% in relation to the estimated total mass and reference unbalance vectors. The quality of digitization is estimated by ZuqueteGuarato [9]. Finally, the scanning of three different models of crankshafts allowed validating the Voxel2inertia algorithm.…”