EXPERIMENTAL AND NUMERICAL-ANALYSIS OF A SEDIMENTATION FORMING COMPRESSIBLE COMPACTS

Batch sedimentations of the mineral talc suspended in water at various initial concentrations resulted in compacts that displayed compressio n, and compression with channel formation. During the experiments the local concentration was deduced by means of local electrical resistanc e measurement. The technique provided concentrations that integrated t hroughout the vessel to give masses that matched the known initial mas s employed to within +/- 5%. Two types of channel zones were observed: soft and hard, the former appeared to be due to the liquid inertia of water discharging from the latter. The region within and above the so ft channel zone diluted from the initial concentration, and this cause d the visible interface between the suspension and the supernatant to accelerate. The top of the hard channel zone followed the line of cons tant solids concentration representing the first significant increase in concentration over the initial suspension. A finite difference nume rical model of sedimentation matched the experimental data, including the data determined below the visible interface, with very high precis ion for the talc suspensions exhibiting compression with insignificant channeling. The implicit model was implemented on a conventional comp uter spreadsheet package and rapidly converged. The model did not empl oy a function for hydraulic permeability, instead a linear function be tween the so-called Kozeny ''constant'' (or coefficient) and concentra tion was used. In order to provide an accurate numerical model for com pressible sedimentation with significant channel formation, the hydrau lic permeability needs to be augmented, or the Kozeny coefficient redu ced, and the dilution above the channel zone must be predicted. These should be achieved in a way that is general to all sedimentations of a given type of material, and not specific to only one starting concent ration. Experimental and numerical results also indicate that the buoy ancy force experienced by the solids is adequately described by the de nsity difference between the solids and the suspending liquid, and not the density difference between the solids and the suspension.