R. A. Harding

A study of refractories as crucible and mould materials for melting and casting γ-TiAl alloys

An assessment has been made of the potential use of mullite (3Al2O3.2SiO2), zirconia (ZrO2), alumina (Al2O3) and yttria (Y2O3) for making crucibles and ceramic moulds for γ-TiAl alloys. Small refractory crucibles have been made by slip casting or pressing techniques and used to melt a γ-TiAl (Ti-48at%Al-2at%Nb-2at%Mn) alloy in a carbon resistance furnace. The effects of different refractories, volume fractions of open porosity and melting time on interactions and microstructures have been established. The width of the interaction region versus the holding time at 1550 °C can be described by a linear law for the mullite crucible and by parabolic laws for the zirconia, alumina and yttria crucibles. Based on these small-scale tests, it can be concluded that mullite and zirconia are unsuitable as crucible and mould materials for melting and casting γ-TiAl alloys, whereas both pure alumina and yttria show promise.

Thermophysical properties of a Ti–44%Al–8%Nb–1%B alloy in the solid and molten states

Abstract The families of titanium aluminide intermetallic alloys have attractive high temperature mechanical properties which make them potential candidate materials for a wide range of applications, particularly in the aeronautic and automobile sectors. The development of appropriate manufacturing techniques is an essential stage in the engineering exploitation of these materials, e.g., Induction Skull Melting is one of the techniques which needs to be optimised for the casting of titanium aluminides. Research is underway to develop a computer model of this process but data are required for the key thermophysical properties. Pulse-heating techniques have been used to measure properties for the Ti–44Al–8Nb–1B system. Rectangular samples have been prepared and are resistively heated as part of a fast capacitor discharge circuit. Time-resolved measurements with sub-μs resolution of currents through the specimen were made with a Pearson probe current monitor using the induction principle. Voltages across the specimen were determined with knife-edge contacts and voltage dividers, and radiance temperatures of the sample were measured with a pyrometer. These measurements allow the calculation of specific heat and dependencies between enthalpy, electrical resistivity and temperature of the alloy up into the liquid phase. Data for thermal diffusivity have been obtained by using the Wiedeman–Franz relation. The results are compared with those obtained using DSC and the four-probe method to measure the temperature dependence of the resistivity.

Entrained oxide films in TiAl castings

Abstract Tilt pouring of TiAl alloy from an induction skull melting (ISM) furnace into a ceramic shell mould has been carried out in this research. Microstructures and casting defects entrained by surface turbulence have been examined using scanning electron microscopy and EDX analysis. Oxide films were found draped over the dendrites of shrinkage porosity and the inner surface of bubbles in the experimental TiAl casting. The entrained oxide films act as heterogeneous nucleation sites, contributing to the formation of centre-line shrinkage porosity in the cast TiAl bars.

Effect of top and bottom filling on reliability of investment castings in Al, Fe, and Ni based alloys

Abstract The effect of bottom and top filling running systems on the properties of four investment cast alloys susceptible to contamination by oxide films during filling was studied. The alloys were air cast 2L99 Al–Si–Mg alloy and 254-SMO super duplex stainless steel and vacuum cast IN939 and IN738LC nickel based superalloys. The Weibull moduli for the tensile strength of investment cast bars produced using top and bottom filling were compared as indicators of casting reliability and of oxide damage produced by the running systems. A Weibull modulus of 18 was obtained for top filled 2L99 castings; this was increased to 34 when a correctly designed bottom filling system with a filter was used, thus reflecting the decreased scatter in properties. However, a similar effect was not found for the stainless steel. The use of improved running system designs led to minor increases in the Weibull modulus of the IN738LC and IN939 Ni alloys.

Optimised running system design for bottom filled aluminium alloy 2L99 investment castings

Abstract This research has evaluated the effect of surface turbulence introduced during the filling of moulds on the reliability of Al-7Si-Mg alloy (2L99)investment castings. Four different running systems were designed to provide different amounts of surface turbulence: top filled, uncontrolled bottom filled, and controlled bottom filled with and without a filter. Computer modelling has been used to simulate the filling of the different designs and the results have been compared with the actual flow behaviour established using real time X-ray radiography. Castings have been produced using the different running systems, tested in four point bending, and the results analysed using the Weibull statistical technique. It has been shown that top filling produces significant turbulence, which results in the least reliable castings (Weibull modulus of 25.5). There was no significant difference between this and the Weibull modulus of 27 found for uncontrolled bottom filled systems. This indicates that a poorly controlled bottom filled system is no better than a top filled system. Both were inferior to the controlled bottom filled systems without and with filters, which had Weibull moduli of 38 and 54 respectively. SEM examination and oxygen analysis of representative fracture surfaces has provided supporting evidence for the important role that oxide films played in reducing properties.

The tilt casting process

Tilt casting is a process with the unique feature that, in principle, liquid metal can be transferred into a mould by simple mechanical means under the action of gravity, but without surface turbulence. It therefore has the potential to produce very high quality castings. Even so, the process is not often optimised in the industrial environment. This investigation represents an attempt to investigate some fundamental problems associated with the process. A computer controlled, programmable roll-over casting wheel with a diameter of 1 m was used to produce sand castings in an Al-4.5% Cu alloy. The filling of the mould was studied using realtime X-ray radiography. Real-time X-ray radiography revealed that the molten metal could exhibit tranquil, turbulent or chaotic flow into the mould during tilt casting, depending on (i) the angle of tilt of the mould at the start of casting, and (ii) the tilting speed. Essentially horizontal transfer of the melt could achieve tranquil filling of the mould with minimum surface turbulence by a tilt starting position above the horizontal. The tensile properties of castings made using various starting conditions and rotation rates were measured and the results analysed using Weibull statistics to quantify reliability. Results are summarised on a map of the various operating regimes for tilt casting. An operational window for the production of reliable castings has been defined for the first time.

Temperature measurements during induction skull melting of titanium aluminide

Abstract An immersion thermocouple and a twin wavelength infrared radiation pyrometer have been used to measure the temperature of titanium aluminide during melting in an induction skull melting furnace. A molybdenum-alumina cermet thermocouple sheath has been found to survive for long enough to enable about three reliable measurements to be made. A significant drift occurred between the temperatures measured using the pyrometer and the thermocouple when melting under a partial pressure of argon, and has been attributed to the generation of fume inside the melting vessel. The drift did not occur when melting was carried out in a continuously pumped vacuum. The superheat obtained increased from ~33°C when melting 4·5 kg of TiAl with 200 kW in a vacuum to ~44°C and ~62°C when melting under argon partial pressures of 20 and 80 kPa respectively. There appears to be little benefit from increasing the applied power from 200 to 350 kW.

Examination of defects in gamma titanium aluminide investment castings

Induction skull melting (ISM) and vacuum induction rollover (VIR) furnaces have been used to melt a γ-TiAl alloy prior to casting into ceramic shell moulds. Castings produced by ISM contained only a slightly higher oxygen content than the charge material, but the melt was cast at a relatively low superheat resulting in numerous turbulence-induced gas bubbles in the castings. Although higher superheat could be achieved by VIR, the use of refractory crucibles resulted in castings with a high oxygen content. Castings poured turbulently under an argon atmosphere were liable to misrun and contained gas bubble defects. These defects were eliminated by pouring the melt into the moulds in a quiescent manner. Casting in a vacuum helped to reduce misrun even when the mould was filled quickly in a turbulent manner. Zirconia cannot be used as a crucible material for melting γ-TiAl alloys, whereas high purity (>98%) calcia is more suitable.

Calculation of the free surface shape in the electromagnetic processing of liquid metals

Abstract An iterative numerical procedure has been developed to predict the free surface shape of the molten metal pool in electromagnetic processing techniques such as induction melting and semi-levitation melting. This required the calculation of the free surface shape to be coupled with the calculation of the e.m. field. The finite element method has been used to calculate the electromagnetic field and has included consideration of the skin effect in all conductors. Special care was taken in discretizing the billet in order to obtain a good representation of the change in the shape at the top. The pinch effect of the coil current on the free surface shape has been clearly shown. The software has been used to predict the shape of the melt in a conventional careless medium frequency induction melting furnace, and satisfactory agreement has been reached between the calculated and the experimentally measured shapes. The model has also been used to predict the meniscus shape in a semi-levitation melting furnace.

Numerical modelling of tilt casting process for γ-TiAl alloys

Abstract The tilt casting method is used to achieve tranquil filling of γ-TiAl turbine blades up to 400 mm long. The reactive titanium alloy is induction melted in a cold crucible, and the crucible with the attached mould is then rotated through 180° to transfer the metal into the mould. In the cold crucible, heat losses to the water cooled copper walls and base limit the superheat available, increasing the risk of premature freezing during mould filling. A compromise is required between fast and slow rotations to minimise the casting defects, such as misruns or gas entrainment. Simulations are presented using the authors’ Computational Fluid Dynamics code with several novel developments in front tracking, heat transfer algorithms and turbulence model adaptation, which accounts for an advancing solid front. The computational results are validated against prototype castings produced at the University of Birmingham, and the model is then used to optimise the tilt casting process.