C. Acht’s research while affiliated with E.ON and other places

Kurzfassung Der Sonderforschungsbereich 570 „Distortion Engineering – Verzugsbeherrschung in der Fertigung“ an der Universität Bremen hat sich als ein Ziel die Bestimmung signifikanter Einflussgrößen und Wechselwirkungen auf den Verzug über die gesamte Fertigungskette von ausgewählten Bauteilen gesetzt, um die wesentlichen Mechanismen der Verzugsentstehung zu identifizieren. In diesem Zusammenhang kommt der Simulation eine große Bedeutung zu, da hiermit die zeitliche Entwicklung der Vorgänge auch lokal im Bauteil analysiert werden können. Hieraus können dann die signifikanten Vorgänge der Verzugsentstehung abgeleitet werden. Für die Simulation muss ein qualitativ hochwertiger Datensatz als Grundvoraussetzung angesehen werden. In der vorliegenden Arbeit wird daher über das Vorgehen sowie die Herausforderungen bei der Erstellung eines Datensatzes für den Wälzlagerstahl 100Cr6 berichtet. Als einschränkende Bedingung soll dieser Datensatz zunächst nur für die Berechnung einer durchgreifenden martensitischen Härtung bei einem definierten Austenitisierungszustand gelten. Neben den thermophysikalischen und -mechanischen Kenngrößen werden die Parameter zur Beschreibung einer martensitischen Umwandlung angegeben. Die Güte des erstellten Datensatzes wird aus dem Vergleich von berechneten und experimentell ermittelten Ergebnissen hinsichtlich Maß- und Formänderungen sowie der Eigenspannungen von Wälzlagerringen nach einer martensitischen Härtung diskutiert.

The Collaborative Research Centres 570 “Distortion Engineering - Distortion Control in the Production Process” focuses on the determination of significant factors and interactions in the overall production process of selected components to identify relevant distortion mechanisms. In this context, computer simulations are more and more important for an answer of distortion problems. With these simulations the temporal and local process development in the component can be analyzed to conclude the significant procedures which control distortion. A high class data set is an essential basic requirement for the simulation. In the present paper the proceeding and the challenge of preparation of a data set for the ball bearing steel 100Cr6 (SAE 52100) are reported. As a restrictive condition, this data set is only valid for a completely through hardened component after a defined austenitizing procedure. In addition to the thermo physical and thermo mechanical values, the parameters of the martensitic transformation are specified. The quality of the presented data set is discussed by the comparison of calculated and experimental results concerning size and shape changes as well as residual stresses of ball bearing rings after a martensitic hardening.

The Collaborative Research Centres 570 "Distortion Engineering - Distortion Control in the Production Process" focuses on the determination of significant factors and interactions in the overall production process of selected components to identify relevant distortion mechanisms. In this context, computer simulations are more and more important for an answer of distortion problems. With these simulations the temporal and local process development in the component can be analyzed to conclude the significant procedures which control distortion. A high class data set is an essential basic requirement for the simulation. In the present paper the proceeding and the challenge of preparation of a data set for the ball bearing steel 100Cr6 (SAE 52100) are reported. As a restrictive condition, this data set is only valid for a completely through hardened component after a defined austenitizing procedure. In addition to the thermo physical and thermo mechanical values, the parameters of the martensitic transformation are specified. The quality of the presented data set is discussed by the comparison of calculated and experimental results concerning size and shape changes as well as residual stresses of ball bearing rings after a martensitic hardening.

This paper presents an ultrasonic method to measure the movement of workpiece surfaces during quenching processes in liquids in order to estimate the time-dependent in-process distortion of the workpiece. The movement is determined by the time-of-flight of the ultrasonic signals from the transducer to the surface and back again (impuls-echo-method). As the simulation of quenching processes depends on several assumptions of the model parameters and almost no experimental intermediate geometry data for quenching processes exist, the in-process-measurements of the surface movement can be used to improve the simulation models and also to extend the knowledge about model parameters. The paper shows an analysis of the ultrasonic data for measurements on cylindrical discs of 20MnCr5 steel and compares the ultrasonic in-process results with simulated data. Additionally, coordinate measurements carried out before and after the heat treatment are presented as a reference and the uncertainty of the ultrasonic measurements is discussed.

In order to detect the most important influencing factors on distortion in carburising processes, it is helpful to use a simulation tool. Additionally, design of experiments (DoE) can help to perform the needed simulations in a well structured, clearly arranged manner.The objective of this work is to detect the influencing factors of carburising processes on the simulated results of the distortion of a disc with a centred hole. Therefore, a full factorial DoE was set up and analysed. The advantage of this procedure is the possibility to find out not only the influence of the main factors, but also the interactions between the different ranges. In this work, the varied factors are the surface carbon content, the carburising depth and as a geometrical aspect the thickness of the disc. The responses under investigation are relative changes in the dimension of the discs.In order to analyse the design of experiments, two different ways were used. At first, the influencing factors were identified by using a probability plot. Secondly, the method “analysis of variance” was used which needs a scattering in the results. Therefore, an artificial scattering was implemented based on experimental scatter bands.

During the important industrial process of case hardening, phase transformations in a steel workpiece are essentially influenced by the (non-homogeneous) carbon distribution (near the surface). Moreover, the temperature course is of great importance. We develop a mathematical model of case hardening which takes into account diffusion of carbon in austenite, heat conduction and possible phase transformations. In this work, mechanical behaviour (thermo-elasticity, classical plasticity and transformation-induced plasticity) is not included in the model. As a result, we obtain an initial-boundary value problem for a coupled system of two parabolic partial and several ordinary differential equations. Finally, we present some numerical simulations.

Der beim Härten in Ölbädern ausgelöste Verzug von einsatzgehärteten Stahl-bauteilen ist abhängig von bauteil-, werkstoff- und verfahrensspezifischen Fak-toren wie z.B. dem Kohlenstoffgradienten in der Randschicht. Außerdem beein-flusst auch der zeitliche Verlauf der Benetzung der Bauteiloberfläche mit dem Abschreckfluid die Form und den Grad des Verzuges. Bei der Simulation der unterschiedlichen Einflüsse stellt sich das Problem, dass zwar die Endzustände bei Raumtemperatur mit experimentellen Daten vergli-chen werden können, der Rechenweg aber nicht überprüfbar ist. Daher besteht großes Interesse daran, neben den Benetzungsverläufen zeitlich aufgelöste Maß- und Formänderungsdaten aus dem Abschreckprozess zu erfassen. Um diese Daten ermitteln zu können, wurde ein bereits entwickeltes Ultraschall-Messsystem zur Benetzungsmessung modifiziert. Mit ihm lassen sich die Ab-stände zwischen den eingesetzten Ultraschall-Transducern und der Bauteilober-fläche durch Laufzeitmessungen nach dem Impuls-Echo-Prinzip bestimmen, so dass auf die Verformung der Oberfläche zurück geschlossen werden kann. Neben dem qualitativen Vergleich der Simulationsdaten mit den ersten experi-mentellen Ergebnissen werden auch die erforderlichen und tatsächlich erreich-ten Unsicherheiten der Verzugsdaten aus der Laufzeitmessung diskutiert.

Dr.-Ing. Thomas Lübben, geb. 1959, studierte Physik an der Universität Bremen. Seit 1986 ist er als Wissenschaftlicher Mitarbeiter in der Stiftung Institut für Werkstofftechnik in Bremen tätig und arbeitet hier auf dem Gebiet der Abschrecktechnik. Seit 2001 ist er Geschäftsführer des SFB 570 und Teilprojektleiter im Projekt A6. apl. Prof. Dr.-Ing. habil. Franz Hoffmann, geb. 1950, studierte Maschinenbau an der Universität Karlsruhe. Er ist seit 1975 in der Stiftung Institut für Werkstofftechnik in Bremen tätig und leitet dort die Abteilung Wärmebehandlung. Abstract In the presented research paper the distortion behaviour of components is analysed. Therefore, a large number of discs with inner bores of 20MnCr5 steel was used as a model geometry for simplification of gears. Altogether, six different variations in dimensions were used, where two different heights and three different outer diameters were combined. Additionally, the influence of carburising depth, surface carbon content and batch structure were included in the analyses. Among gas carburised and oil quenched discs, some discs were low pressure carburised and gas quenched for comparison. Since the experiments were arranged with the aid of a design of experiment, it was possible to identify the significant and highly significant influencing factors on the change of dimensions and shapes. Furthermore, the influence of the disc dimensions on the distortion behaviour was analysed and based on simulations, reasons for certain changes in shapes were found.

In the presented research paper the distortion behaviour of components is analysed. Therefore, a large number of discs with inner bores of 20MnCr5 steel was used as a model geometry for simplification of gears. Altogether, six different variations in dimensions were used, where two different heights and three different outer diameters were combined. Additionally, the influence of carburising depth, surface carbon content and batch structure were included in the analyses. Among gas carburised and oil quenched discs, some discs were low pressure carburised and gas quenched for comparison. Since the experiments were arranged with the aid of a design of experiment, it was possible to identify the significant and highly significant influencing factors on the change of dimensions and shapes. Furthermore, the influence of the disc dimensions on the distortion behaviour was analysed and based on simulations, reasons for certain changes in shapes were found.