Automotive Industry
Industry: Email Alert RSS FeedDesigning things right: here are three new CAD tools. One focuses on part assemblies. Another focuses on displaying CAD models. The last focuses on making CAD available everywhere for everybody
Automotive Design & Production, Sept, 2007 by Lawrence S. Gould
Every manufactured part is not produced the same. There are slight deviations from the nominal design. Software can account for these deviations, even simulate the buildup of these deviations that lie within the tolerances inherent in the assembly process and in the piece parts themselves. A harder problem is simulating the accumulation of tolerances in assemblies composed of non-rigid parts, such as those made of sheet metal, rubber, and plastic. When dealing with individual parts, tolerance buildup is not an issue. But when it comes to assemblies--seat assemblies, dashboard assemblies--it is of vital concern.
- Most Popular Articles in Autos
- Service Slants
- 2007 utility vehicle buyer's guide: Side-By-Sides are popular; here's who ...
- Transmission considerations: beyond the manual gearbox
- Buell Motorcycle engineering, innovation, & dedication: in an industry ...
- 100 + 10: America's oldest automotive magazine celebrates its 110th year ...
- More »
A few years ago, Dimensional Control Systems Inc. (DCS; Troy, MI; www.3dcs.com) was awarded funding from the National Institute of Standards and Technology (NIST) to develop SOVA: stream of variation analysis. SOVA is a mathematical system for predicting the accumulated variations in assembled rigid and flexible parts influenced by material deformation. It models all the tolerances, simulates thousands of assembly builds, puts them all together, and shows for any group of parts in an assembly the gaps and variations between those parts, as well as the major contributors to a given measurement. By removing the trial-and-error in estimating product quality and in identifying assembly problems from out-of-tolerance parts, the time for product development shrinks, especially in downstream assembly operations. It also lets engineers maximize part tolerances while controlling dimensional assembly requirements, according to John Sienkowski, new business development, for DCS. This leads to a reduction in tool rework, scrap, and warranty defects--all leading contributors to manufacturing costs.
SOVA is incorporated in 3DCS, the modeling and simulation system from DCS. 3DCS lets users model the effect of variation on an assembly, determine the robustness of the design, and test alternative tolerancing methods. (The tool can run standalone or integrated as another workbench within Catia V5 CAD.) Users statistically simulate producing "virtual assemblies." They can then actually see the variations in the virtual assemblies. Tolerance analysis requires inputs such as geometry (from the solid model), moves (locating strategy and assembly order), tolerances (piece part and assembly), and the dimensional requirements to be analyzed. The flexibility of such non-rigid parts and assemblies are captured by FEA, which 3DCS incorporates into a Monte Carlo simulation of the assembly builds. After analysis, the outputs include standard deviation, percent out of specification, tolerance ranges, Cp, Cpk, and a histogram that displays the distribution for each measurement in the simulated build model (including whether it is in/out of spec), nominal distance of overlap, the range of variation, the specification limits of min/max variation, and the min/max variation from the simulated runs.
[ILLUSTRATION OMITTED]
Engineers can click on individual measurements to see where those measurements appear in the solid model. Out-of-spec measurements, including clashes, show up in red. 3DCS can identify the key contributor for every measurement in a model, including the tolerance and assembly process responsible for variation. The program also gives engineers a statistical prediction as to whether the design will meet build requirements when it goes into production. By holding all tolerances in the model perfect, an engineer can move one tolerance at a time to its min/max, thereby seeing the effect of one tolerance on the whole assembly. Alternatively, 3DCS Advanced Analyzer & Optimizer uses an equation to represent an entire assembly model. The module basically outputs a spreadsheet that lets users assess how one tolerance might affect other measurements in an assembly. Any change in a tolerance yields immediate feedback regarding the effect of that change. (The time-consuming alternative is to run and rerun the Monte Carlo simulation.) The module's optimizer lets users apply a cost to each tolerance. An engineer can then determine the level of quality in an assembly produced on a fixed budget. Conversely, given a set of objectives, an engineer can use the optimizer to determine how much an assembly will cost. In the U.S., one seat for the base 3DCS module costs $19,000; a floating license is $24,700.
Ray of truth
Ray tracing is a 3D rendering technique that generates physically correct simulations of lighting, such as reflections and refractions. One problem: Ray tracing calculations are time-consuming. Clusters of workstations, grid computing, and multiprocessor workstations significantly reduce that compute time. So does specialized graphics processing hardware. So does optimized ray-tracing software. In all of these approaches, the goal is the same: Produce high-fidelity, highly accurate scenes in real time. From this, as with all digital design development tools, automotive engineers can better analyze product designs throughout product development with fewer prototypes. This capability becomes particularly enticing in hard-to-simulate scenes involving, say, new headlight designs or the potentially troublesome reflections from a windshield.