Regression analysis using indicator variables has been applied to interpret results obtained in a designed experiment, conducted to investigate the influence of carbon, silicon, phosphorus, manganese and sulfur on liquidus and eutectic arrestparameters of base and inoculated gray iron cooling curves. No significant difference was found in base iron austenite liquidus temperature as determined from tellurium and non-tellurium cup cooling curve data, with carbon, silicon and phosphorus beingthe main chemistry effects.Similarly, no significant difference was found to exist between base and inoculated iron TAL determined from nontellurium cooling curve data, with sulfur appearing as an additional chemistry effect in the model generated from the combined data set The TAL relationships determined in the present study are similar to those previously presented in the literature but exhibit distinctive features related to the melting conditions employed in the present study.Application of austenite liquidus (TAW) and eutectic arrest (TEW) tellurium cup parameters to the development of regression model chemistry prediction Equations yielded a robust prediction Equation for carbon, but not for silicon. In the case of silicon,a phosphorus correction is required to yield an accurate silicon estimate, whereas the phosphorus term fortuitously drops, leaving the carbon estimate insensitive to variation in phosphorus content Regression relationships giving the chemistry dependenceof minimum eutectic undercooling temperature, TEU, maximum recalescence temperature, TER, magnitude of recalescence, TER-TEU and the difference parameter, TEU-TEW, are also presented.
展开▼
