Some peculiar and apparent erratic segregation effects can occur with non-mass flow hoppers that are refilled before running empty. One cause is a consequence of the ‘drained cone’ draw down pattern as material is extracted by running down the repose surface into a central ‘core’ flow channel to the outlet through a bed of static product. When the hopper refills before complete discharge has taken place this depression of ‘drained repose cone’ becomes the boundary surface that contains the incoming material. If the product tends to segregate with some fraction deposited at the periphery of the container, there will be preferential deposition on this conical surface as the level increases.

The problem does not arise straight away, but surfaces unexpectedly when the hopper has emptied nearly all the fresh contents and the level in the silo falls towards this original surface. The last portion of the fresh load to exit will possess a high concentration of the segregated component that has accumulated on this draining surface. This dramatic change in product condition will occur during the operation without any preliminary warning and, once this layer is passed, the composition revert to a more normal condition just as ‘mysteriously’. This effect can be even more pronounced if the hopper is discharging whilst it is filling. In these circumstances the accumulating hopper contents tends to have an even higher proportion of the segregated fractions.

It is impossible to investigate, interpret or predict such events without a rigorous record of the sequence and rates of input and discharge to the silo. Some apparently minor changes in the pattern of fill and discharge could therefore be responsible for this change on behaviour.

One clue is that you have visual evidence of segregation, although the extent and form is not clear. This may have a direct outcome, if the extruder finds the excess fines difficult to process or, in some cases, have an indirect effect by instantaneously influencing the rate of discharge from the silo.

Whilst the cause may be unclear, one approach is to address the homogeneity of discharge by fitting segregation regression inserts in the silo base. Details can be supplied on review of the silo drawing sent to lyn@ajax.co.uk

Useful information is given on this subject in: – ‘User Guide to Segregation’ published by BMHB. This book includes a comprehensive review of the forces, mechanisms, processes, flow regimes, and operations that give rise to segregation and includes various tips on its mitigation and correction. Copies are available at £ 25 plus p & p.

Lyn Bates ■