The entire chain of gear design and gear production contains subsequent steps including the geometry design, process design, production, quality control, and investigation of the running behavior. The performance of a gear is a result of a combination of several process steps from the raw part to the finished gear. In addition to gear macro geometry, the gear micro geometry as well as the surface integrity have a decisive role in the running behavior. Since all production processes directly influence the geometry und surface characteristics of the gear, in a ‘design to make’ approach the entire process chain already has to be considered in the gear design phase.
One way to handle the complexity of gear production is the consistent mapping of all process steps in a digital shadow according to Industry 4.0-related approaches. The architecture required is provided by the Industrial Internet of Things (IIoT) or the Internet of Production (IoP) that has been launched during the last “Aachen Machine Tool Colloquium AWK 2017: Internet of Production for Agile Enterprises” at WZL of RWTH Aachen and Fraunhofer IPT. The digital shadow consists on one hand of suitable simulation models starting at the design phase over the production chain up to the resulting running behavior. On the other hand, process data are accumulated from the machine control and measurement devices and integrated in an intelligent database. The combination of process data with simulative information in real time delivers new approaches for the adaptive control of processes regarding part functionality as well as for the analysis of new cause-effect dependencies not covered by existing models so far.
The enabler for this approach in gear technology is a virtual production chain for the gear set, Figure 1. In order to derive economic tolerance fields, the analysis starts with the requirements in operation and, starting from there, goes back step by step along the manufacturing chain. The idea is not to require fixed IT-tolerances and correct those in a trial and error approach, as it has been done in the past. Instead, functional tolerances for each parameter and each process according to their impact on the next process step and/or the surface integrity determining the operational behavior shall be proposed. Regarding sensitive parameters, tolerances are individually narrowed. Regarding insensitive parameters, tolerances are individually widened. This requires software support for each single step, but allows for an efficient design of manufacturing processes under consideration of the desired running behavior.
The starting point is the macro and micro geometry design. Therefore, extensively validated software solutions for optimization can be used such as µOpt (statistical analysis of variant calculations) and GearGenerator (generation of gear geometries under consideration of manufacturing deviations) in combination with FE-Stirnradkette/STIRAK or ZaKo3D (FE-based tooth contact analysis for running behavior). The simulation programs are either part of the WZL Gear Toolbox, developed in cooperation with the WZL Gear Research Circle, or the FVA-Workbench, a software platform under the umbrella of the German Forschungsvereinigung Antriebstechnik e.V. The optimized gear geometry is determined by means of mass variant calculations. Regarding micro geometry, the manufacturing and assembly tolerances are considered automatically by statistical analysis of results obtained in the variant calculation. Due to the efficient algorithm, the tooth contact analysis allows for the simulation of a couple of million variants in only a couple of hours. The results are proposals for cost-efficient flank modifications that are robust within the entire tolerance band regarding noise, strength, and efficiency.
In the simulations for soft and hard machining processes (SPARTApro, GearGrind3D), tool loads as well as characteristic surface topographies and surface integrity indicators are calculated. With this knowledge, production processes can be improved regarding time, costs and/or quality. In addition, the interactions of the individual steps can be demonstrated, since the simulation output of one process step is the input of the next process step or the tooth contact analysis. A virtual gear production must be supported by a uniform data management. Only in this way suitable proposals can be made for process parameters adjustment. The IoP offers the possibility to link all aspects of the physical as well as the virtual process chain. By matching measured values with expectations from simulation for every individual process step, the process can be optimized with the help of the digital shadow. Likewise, an adaptive adjustment of the process parameters due to a link of process monitoring and simulations is possible; thus increasing the quality and productivity in production and reducing waste. Therefore, a real-time comparison of process simulation and measurement data is performed. The result is an assistance system for gear manufacturing that is part of current research at WZL Gear Technology Department of RWTH Aachen, Figure 2.
In summary, an increase in productivity and part functionality is expected by a coupling of gear design and gear manufacturing in a virtually enhanced production chain. The efficient variant calculation allows for the consideration of tolerance effects in mass calculations and enables robust gear and process designs. By linking simulation data with sensor signals of the machine tool in real-time, the adaptive control of processes is possible and a direct feedback can be provided to the worker in an assistant system. Thus, resulting in less trial and error – and more first part right!
Dr.-Ing. Dipl.-Wirt.-Ing. Christoph Löpenhaus
Prof. Dr.-Ing. Christian Brecher
Prof. Dr.-Ing. Dr.-Ing. E.h. Dr. h.c. Dr. h.c. Fritz Klocke
All: Laboratory of Machine Tools (WZL) of RWTH Aachen, Department of Gear Technology