Ce travail traite de la thermohydraulique d’un bain de melt couplée à la physicochimie pour ladescription du comportement de mélanges de matériaux (non-eutectiques).On décrit le transitoire d’établissement de température dans un liquide avec dégagement de puissancevolumique en présence de solidification sur une paroi refroidie. Le modèle développé à cet effet estvalidé par rapport aux résultats des essais LIVE réalisés à KIT. Dans les conditions de ces essais onmontre que la température d’interface suit la température liquidus (correspondant à la composition dubain liquide) pendant le transitoire d’établissement de la température dans le bain et des croûtessolides.Par ailleurs, on propose un modèle d’interaction entre un liquide non-eutectique (soumis à dissipationvolumique de puissance) et une paroi fusible dont la température de fusion est inférieure à latempérature liquidus du bain. Les prédictions du modèle sont comparées aux résultats des essaisARTEMIS 2D. On en déduit une nouvelle formulation de la température d’interface (inférieure àliquidus température) entre le liquide et la couche pâteuse en paroi. / This work deals with the thermal-hydraulics of a melt pool coupled with the physical chemistry for thepurpose of describing the behaviour of mixtures of materials (non-eutectic).Evolution of transient temperature in a liquid melt pool heated by volumetric power dissipation hasbeen described with solidification on the cooled wall. The model has been developed and is validatedfor the experimental results given by LIVE experiment, performed at Karlsruhe Institute ofTechnology (KIT) in Germany. Under the conditions of these tests, it is shown that the interfacetemperature follows the liquidus temperature (corresponding to the composition of the liquid bath)during the whole transient. Assumption of interface temperature as liquidus temperature allowsrecalculating the evolution of the maximum melt temperature as well as the local crust thickness.Furthermore, we propose a model for describing the interaction between a non-eutectic liquid meltpool (subjected to volumetric power dissipation) and an ablated wall whose melting point is below theliquidus temperature of the melt. The model predictions are compared with results of ARTEMIS 2Dtests. A new formulation of the interface temperature between the liquid melt and the solid wall(below liquidus temperature) has been proposed.

The work is particularly aimed at metallurgical problems connected with manufacture of castings of weight about 1 up to 40 tons from high-alloy austenitic nickel cast irons also known under the trade name Ni-resist. The main aim is to check and complement the knowledge about manufacture of heavy castings from austenitic nickel cast irons in a field of control of tapping and casting temperature, carbon content control with regard to the prediction of eutectic composition and in a field of setting a suitable chemical composition of cast iron with regard to the formation of chunky graphite and stability of the austenitic matrix and further on also possibilities of cast iron filtration during pouring the castings. Based on done experiments the formulae were determined that enable to calculate eutectic and liquidus temperatures on the basis of chemical composition of the cast iron. The use of several formulae given in the literature for calculating the carbon equivalent of austenitic nickel cast irons was compared and evaluated. For utilization of determined results in practice a calculating programme has been compiled making possible to be used in operational practice. This programme is named Calculation of casting temperature of Ni-resist cast irons. After the entry of the melt chemical composition the programme calculates the all important parameters necessary to know and to be controlled for manufacture of high quality castings from austenitic nickel cast irons. Eutectic temperature, liquidus temperature and then based on it the tapping and casting temperatures are included among those parameters. Further on there are eutectic concentration of carbon, carbon equivalent, nickel equivalent, and a criterion evaluating a danger of chunky graphite formation.