Regression Analysis of Thermal Conductivity Based on Measurements of Compacted Graphite Irons
- PDF / 709,317 Bytes
- 10 Pages / 593.972 x 792 pts Page_size
- 24 Downloads / 187 Views
ON
IN an engine, components such as cylinder blocks, cylinder heads, and piston rings are exposed to large variations in temperature that can give rise to thermal stresses and result in engine failure. The combination of strength and thermal transport properties of the engine material is therefore very important. Previous cast iron investigations have concluded that graphite morphology generally has the largest influence on the thermal conductivity.[1–3] Gray cast iron, with lamellar-shaped graphite, conducts heat more efficiently than compacted graphite iron (CGI) and the worst thermal properties are found in ductile cast iron, with nodular-shaped graphite. Gray cast iron has therefore been the material of choice in most engine components, due to its ability to remove heat rather than its strength. In the transport industry, there is an increasing demand for environmentally sustainable transport, driven by governmental decrees and customer requirements. Increasing an engine’s peak firing pressure is one solution to meet these demands. With increasing engine temperatures, the combination of strength and thermal properties in gray cast iron are not good enough, due to poor strength at elevated temperatures. Attention is therefore directed toward CGI as a replacement material. Even though the thermal transport property is worse compared to gray cast iron, the improved strength justifies the change. The lower value of the thermal conductivity found in CGI MARTIN SELIN and MATHIAS KO¨NIG, Postdoctoral Students, are with the Department of Mechanical Engineering, Materials and Manufacturing—Casting, Jo¨nko¨ping University, Jo¨nko¨ping, SE - 551 11, Sweden. Contact e-mail: [email protected] Manuscript submitted February 12, 2009. Article published online September 25, 2009 METALLURGICAL AND MATERIALS TRANSACTIONS A
therefore requires further attention before a material with optimum properties can be achieved. Various methods for approximation of the thermal conductivity in cast irons have previously been made by, for example, Helsing et al.[4,5] and Holmgren et al.[2,6] Helsing et al. used an average field approximation to derive models for the thermal conductivity value, while Holmgren et al. solved linear regression models. Thermal conductivity values for all three major graphite morphologies, i.e., lamellar, compacted and nodular, could be calculated using any of the methods mentioned earlier. Both methods have advantages and disadvantages. The average field approximation is flexible and can be applied for a wide range of graphite shapes and microstructures. However, this method requires very accurate values of the thermal properties of the various microconstituents and sometimes rough approximations are needed. A regression model is based on the parameters believed to affect thermal conductivity significantly. These parameters are usually easy to measure and the model is based on experimental measurements, requiring no assumptions such as grain conductivities. One downside with this method is the large amount of data neede
Data Loading...
