SAMPLING FROM SLURRY PIPELINES

Samples must frequently be removed from liquid-solid handling systems so that the proper particle size or concentration can be established. Serious errors in concentration or size distribution measurements can result from improper sampling. The object of this thesis was to study these errors for various sampling techniques and to recommend procedures for correct sampling.; The equations of motion for each phase were combined using the mixture mean mass velocity. Assuming constant solids concentration, the mixture velocity field was obtained from a potential flow solution using a finite difference method. The particle equation of motion was then solved to predict particle trajectories. The sampling efficiency was obtained from the limiting particle trajectory.; Experimental work was done with a vertical pipeline. Water-sand slurries with particles of surface mean diameter between 0.165 and 0.72 mm were used at discharge concentrations to 25% by volume. The effects of concentration, velocity and particle size on the sampling efficiency were studied for both probe and wall sampling.; Satisfactory agreement between predicted and measured concentration for different particle sizes has been achieved for thin and tapered L-shaped probes. For blunt probes particle rebound has been shown to be an important source of error and an empirical correction procedure for predicting sampling efficiencies has been obtained.; Side wall sampling was shown to produce substantial errors in many circumstances. A two-dimensional model provided only a qualitative explanation of these errors because the flow is in fact three-dimensional. Simple devices to improve wall sampling were examined experimentally.; The study shows that measuring the sampling velocity is necessary if reliable concentration measurements are to be taken from slurry flows.