Growth Simulation of Spheroidized Carbide in the Carbide-Dispersed Carburizing Process
- PDF / 883,366 Bytes
- 10 Pages / 593.972 x 792 pts Page_size
- 74 Downloads / 303 Views
UCTION
THE carbide-dispersed carburizing (CDC) process has been increasingly applied to machine parts in order to improve their durability against high repeated compressive loads.[1,2] In most cases, these parts are carburized in a high carbon potential (CP) atmosphere followed by quenching. The carburized surface is then ground to an appropriate depth after examining the cross-sectional microstructure, hardness, and carbon concentration profile. Surface mechanical properties, such as fatigue strength and wear resistance, largely depend on the size (or mean diameter) as well as fraction of carbide particles exposed. For the development of high-performance steel parts, controlling the carbide size is of great importance, and a theoretical simulation has been desired for predicting the effect of carburizing conditions and alloying elements on carbide size distribution in carburized layers. A planar one-dimensional (1-D) finite difference analysis (FDA) is usually employed for the calculation of carbon diffusion in the austenite (c) phase. For the KOUJI TANAKA, Senior Researcher, HIDEAKI IKEHATA, Researcher, and KOUKICHI NAKANISHI, Research Manager, are with Toyota Central R&D Labs., Inc., Nagakute, Aichi, Japan. Contact e-mail: [email protected] TOMOAKI NISHIKAWA, Representative Manager, is with Aichi Steel Co. Ltd., Tokai-shi, Aichi, Japan. Manuscript submitted September 6, 2007. Article published online April 16, 2008 1248—VOLUME 39A, JUNE 2008
conventional gas carburizing process, the concentration profile of solute carbon is well predicted by this numerical method. In the CDC process of high-carbon chromium steels, however, (1) diffusion in the c matrix dispersed with a large amount of the carbide phase needs to be modeled, and (2) it has to be taken into account that the carbide-c system takes time to reach thermodynamic equilibrium. A multicomponent diffusion simulation code that performs the CALPHAD equilibrium calculation for each grid point in 1-D FDA is DICTRA. If carbides are assumed to form instantly in the equilibrium amount given by local chemical compositions, then the carbide fraction in a carburized layer would be estimated in the discrete process of FDA.[3] The software also takes into account the so-called labyrinth factor[4,5] for the impeding effect of high volume fraction precipitates. However, this kind of dispersed planar model does not provide any information on the carbide particle size. The situation becomes more complex in the case of chromium-bearing steels. These contain a large amount of spheroidized carbide and the addition of Cr makes the dissolution process much slower than that in Fe-C binary steels. Thus, the carbide particles are expected to be half dissolved when heated in a carburizing furnace. As a result, the previous equilibrium assumption usually does not hold, which reflects the difficulty in setting a reasonable status for the system at the start of carburizing. In view of promoting the CALPHAD-assisted FDA, this article presents a simulation method relating the output/input o
Data Loading...
