Influence of Melt Feeding Scheme and Casting Parameters During Direct-Chill Casting on Microstructure of an AA7050 Billet

Abstract

Direct-chill (DC) casting billets of an AA7050 alloy produced with different melt feeding schemes and casting speeds were examined in order to reveal the effect of these factors on the evolution of microstructure. Experimental results show that grain size is strongly influenced by the casting speed. In addition, the distribution of grain sizes across the billet diameter is mostly determined by melt feeding scheme. Grains tend to coarsen towards the center of a billet cast with the semi-horizontal melt feeding, while upon vertical melt feeding the minimum grain size was observed in the center of the billet. Computer simulations were preformed to reveal sump profiles and flow patterns during casting under different melt feeding schemes and casting speeds. The results show that solidification front and velocity distribution of the melt in the liquid and slurry zones are very different under different melt feeding scheme. The final grain structure and the grain size distribution in a DC casting billet is a result of a combination of fragmentation effects in the slurry zone and the cooling rate in the solidification range.

A Numerical Study of 3D Turbulent Melt Flow and Solidification in a Direct Chill Slab Caster with a Porous Combo Bag Melt Distributor

“…The shape and location of the cliff-shaped mushy region are seen to be dependent on the strength of the flow there and this region is moving downward with the increase of the casting speed. A similar cliff-shaped region is reported in a recent paper by Zhang et al [26] in their 2D modeling of a round billet caster with a vertical melt-feeding system through a delivery nozzle for an aluminum alloy AA-7050. Figures 5a-5c through Figures 8a-8c illustrate longitudinal 2D views of the temperature and velocity profiles at the wide symmetry plane (z ¼ 0) and parallel to the wide symmetry plane at z ¼ 62.5 mm and z ¼ 312.5 mm for the aforementioned four cases.…”

Mechanisms and Control of Macrosegregation in Dc Casting

“…a semihorizontal scheme when the melt is directed through four outlets from the centerline to the sides of the billet and a vertical scheme when the melt is directed downward along the centerline. The experimental and computer simulation details are reported elsewhere [15]. The obtained macrosegregation patterns are distinctly different as shown in Fig.…”

“…Figure 6 shows the simulated flow patterns and sump profiles. The uneven change in the sump profile and distances between solidus and liquidus in the vertical feeding scheme cause sharp increase in the cooling rate in the center followed by much lower cooling rate at about 40 mm off center [15]. A combination of melt flow velocity and cooling rate affects the grain structure formed with local grain refinement in the center of the billet in the vertical scheme compared to grain coarsening in the semi-horizontal scheme.…”

“…While the semi-horizontal scheme produces the macrosegregation typical of DC cast billets, the vertical scheme results in solute enrichment in the center of the billet followed by deep negative segregation troughs. In order to understand the formation of such macrosegregation patterns, the structure of the billets was examined and the sump shape and melt flows were simulated [15]. It appears that the vertical feeding scheme results in significant sump depression in the center of the billet with very high flow velocities close to the melt outlet, while for the semi-horizontal scheme the maximum flow velocity occurs close to the billet surface.…”

Jet Mixing in Direct-Chill Casting of Aluminum: Crater Effects and its Consequence on Centerline Segregation

“…We then introduce the following non-dimensional variables for vertical and radial coordinates and the pressure and velocity fields (14. ) This pressure field satisfies Laplace's equation and if one assumes that the surface pressure distribution is localized to the point of impingement, we can derive the axisymmetric pressure distribution by solving equations (9)(10)(11)(12), which reduce to:…”