An experimental description of the flow structure of non-Newtonian slurries in the laminar, transitional, and full turbulent pipe flow regimes is the primary objective of this research. Experiments were conducted in a large-scale pipe slurry flow facility with an inside pipe diameter of 51 mm. The transparent slurry formulated for these experiments from silica, mineral oil, and Stoddard solvent exhibited a yield-power-law behavior from concentric-cylinder viscometer measurements. The velocity profile for laminar flow from laser Doppler velocimeter (LDV) measurements had a central plug flow region, and it was in agreement with theory. The range of the transition region was narrower than that for a Newtonian fluid. The mean velocity profile for turbulent flow was close to a 1/7 power-law velocity profile. The rms longitudinal velocity profile was also similar to a classical turbulent pipe flow experiment for a Newtonian fluid; however, the rms tangential velocity profile was significantly different.

No abstract

This paper gives an analysis of the performances of two oil/water separators. One is a laboratory model using a clean oil/water system, and the other is a test separator from an offshore Middle East field. In both cases, we considered the separator's ability to produce clean oil, i.e., the concentration of water in the effluent oil. The analysis of the two separators was based on measurements of the effluent oil quality, inlet water cut, flow rates of water and oil, the overflow rate, and the retention time. For the model separator, we performed detailed batch tests in order to relate the performance of the separator to convenient laboratory measurements. The relationship was made with a mathematical model and a corresponding computer code that was used to predict the separator performance on the basis of the batch test data, and to parameterize the performance data for the two separators. We found that the droplet coalescence rate and the oil/water separation rate were highly nonlinear functions of time, which means that coalescence was an important separation mechanism in both separators. This makes the separator sensitive to changes in the flow rates and the oil pad thickness. The residence time was the critical design parameter for both separators, whereas the overflow rate was found to be of little importance. Due to the importance of coalescence, an increase in the inlet water cut would improve the separator efficiency and actually produce oil with less water.

No abstract

No abstract