Although a properly conducted and anlaysed sampling regime can provide comprehensive performance data around dense media separation, its not always practical to take this approach. The sampling process presents following disadvantages.
Sampling difficult (extremely difficult for large modern cyclones and jigs
Sampling is time and labour consuming
Size & float-sink analysis expensive
Data takes long time to produce (can be months)
Data often poor due to sampling/analysis errors
Very large samples may be required if sampling baths treating coarse coal
The high density liquids available for densimetric analysis are toxic, and unpleasant to use. The conventional procedure for analysis with heavy liquids may also be criticised on other grounds:
Representative samples are frequently difficult to obtain from high tonnage flows. The samples must also be of statistically acceptable size and where an important density component is in low concentration they may have to be very large, rendering the analysis more difficult, sometimes prohibitively so.
The analysis procedure is tedious, time-consuming and prone to error.
The final partition numbers represent the mean of the two bounds of the density interval (the density of the two liquids used in each analysis), so a degree of uncertainty is inevitably associated with each point. The problem is usually alleviated by making separations in a large number of liquids, thus reducing the range of each interval, but this results in a considerable increase in the required volume of work and the cost of the analysis.
Although a properly conducted and anlaysed sampling regime can provide comprehensive performance data around dense media separation, its not always practical to take this approach. The sampling process presents following disadvantages.
The high density liquids available for densimetric analysis are toxic, and unpleasant to use. The conventional procedure for analysis with heavy liquids may also be criticised on other grounds: