Simulation of through‐hardening of SAE 52100 steel bearings – Part II: Validation at industrial scale

Abstract

In this study, the material dataset presented in part I of this article is validated at industrial scale in batch through-hardening of bearing races. The material dataset acquired is implemented in a commercial heat treatment simulation software. Heat transfer coefficients for the oil and salt bath are determined by using a commercial standard quench probe. Zone temperatures and transfer times of the roller-belt furnace are measured directly from the system.Through-hardening of inner ring (IR) of 6813 bearing in oil and salt bath is simulated considering most of the industrial details. Finally, predicted dimensional changes are compared with the coordinate measurement results and a good agreement is achieved. It is concluded that determined material and process data, idealizations and simulation procedure can be considered "validated" for further improvement of the industrial process. In dieser Arbeit wird der Materialdatensatz, der in Teil I dieses Artikels vorgestellt wurde, mit Versuchen zum Durchhärten von Lagerringen im industriellen Maßstab validiert. Der ermittelte Datensatz wird in eine kommerzielle Software zur Wärmebehandlungssimulation implementiert. Zur Charakterisierung der Abschreckung in Öl und Salz wird eine kommerzielle Standardabschreckprobe eingesetzt. Das Temperaturprofil und die Transferzeiten werden direkt am eingesetzten Rollenherdofen ermittelt. Die Durchhärtung des Innenrings des Lagertyps 6813 in Öl und Salz wird unter Berücksichtigung der meisten Details des industriellen Prozesses simuliert. Die berechneten Maßänderungen werden den entsprechenden Ergebnissen von Koordinatenmessungen gegenübergestellt. Dabei wird eine gute Übereinstimmung erreicht und geschlussfolgert, dass die ermittelten Werkstoff-und Prozessparameter, die Idealisierungen und die Simulation an sich als validiert angesehen werden können und für weitere Verbesserungen des industriellen Prozesses eingesetzt werden können.

FEM modeling and experimental validation of quench-induced distortions of large size steel forgings

“…5 h, followed by air cooling (20 °C) for 1.5 h. The quenched block was then tempered at 565 °C for about 48 hours to reduce the level of the residual stresses produced during quenching. While the tempering could very slightly affect the quenchinduced distortions, the dimensional changes are, however, considered relatively negligible in this case, considering the large size of the block[21].…”

“…For example, the size and the shape of the part asymmetrically affect the heat transfer within the part, the phase transformation at different locations, and hence, the distribution of the residual stresses and properties. This can only be circumvented if a 3D modeling approach is used [4,13] and accurate material data (e.g., properties of each phase) are employed [21].…”

Microstructure Evolution and Cooling Characteristics at C−Si−Mn−Cr Steel during Different Quenching Processes

“…[ 10 ] However, there may still be deviations in the prediction of the internal temperature of large‐sized parts during quenching. [ 11 ] The accumulation of temperature data has been limited, and most of the results are predictions without experimental verification. The effect of temperature change at different positions on the microstructure and cooling characteristics of steel during quenching has rarely been reported.…”

On the Accuracy of Finite Element Models Predicting Residual Stresses in Quenched Stainless Steel

“…In this regard, it has been found that the use of static values of HTC in FEM models yields results substantially different compared to models employing temperature-dependent HTC for the purpose of simulating cooling curves and stress profiles both in gas [9,10] and liquid quenching [7]. Therefore, the use of a temperature-independent HTC should be avoided in oil and water quenching baths, although it may be useful in salt baths [11]. There has been a considerable number of papers where the characteristic time-and temperature-dependent HTC curves for a Jominy end quench test [12][13][14][15] have been studied and validated [16,17].…”