Avoiding the problems of FIRST IN, LAST OUT

To some degree, all powders and bulk solids exhibit different flow properties. These are a function of a variety of factors such as particle size distribution, bulk density, moisture content, particle shape, surface properties, chemical composition or electrostatic charging. Variations in any of these common characteristics within a production run may result in a material that flows irregularly.

For a material to be handled efficiently, its `worst case' flow characteristics need to be known and the equipment designed accordingly. In a large number of plants, however, there is seldom the luxury of dedicating machinery to just one product. In most cases campaign production is undertaken to optimise machinery use, and a wide variety of blends and powders will be handled through the same machine. This necessary compromise can result in unreliable flow and poor filling performance, unless there is a good understanding of material flow and behaviour through the machine.

Most hoppers, whether on packing machines or further upstream in the production process, discharge in core (or funnel) flow. As can be seen from Fig 1, a core flow hopper will always draw preferentially from a central channel which extends to the top surface of the stored material. Segregation effects that occur during filling operations result in more fines within the central zone of the inventory, with an inverse size distribution deposited at the periphery of the inventory.

So, with core flow discharge, material with a disproportionately high level of fines will be discharged first from the hopper, whilst the level of fines will reduce steadily to below average as inventory falls; the last out may have a high fines content again. This distribution of particles is typical of a centrally filled hopper, but the segregation pattern is reversed if the hopper is filled using a pneumatic system that has a tangential entry (i.e. fines at the outside and coarse in the centre).

In addition to exaggerating segregation effects, the material which eventually passes through the outlet will have sloughed off from the top surface of the inventory and passed down through the central flow channel. This involves continuous (but variable) shear against static or slow moving particles, resulting in erratic flow. These two factors alone can lead to substantial difficulties when trying to implement accurate instrumentation and control techniques into a packing process due to the inconsistency of flow and bulk density.

Of course, the above description of core flow discharge assumes that the material will actually come out at all. Many poorly designed hoppers can also exhibit mechanical arching, cohesive arching or even `rat holes' - all of which would make an article on their own.

An alternative to core flow discharge, however, is the mass flow `design' of a hopper. The word `design' is emphasised here because unless a hopper has been specifically designed using calculations based upon the characteristics of the materials to be handled, mass flow is unlikely to be achieved.

The key features of mass flow are shown in Fig. 2. From a process control stand point, the most important aspect is that the material is drawn down evenly across the entire cross section of the hopper, the major proportion of shear being that between the sliding material and the hopper wall. This minimisation of shear enables a very even and consistent flow of material to develop at the outlet, introducing a degree of homogeneity into the material as the fines, medium and coarse particles are recombined at the outlet.

This `first in, first out' characteristic (as opposed to the `first in, last out' of core flow) also enables particle maturing, conditioning or de-aerating to take place, all of which contribute to maintaining a consistent product quality for downstream processes. The fact that maturing of material is possible is very important, particularly when dealing with materials which have a tendency to cake. If this type of material is put through a core flow regime, moisture migration within the packaged material can often result in end users receiving `tombstones' or lumps instead of a flowing material.

Designing a hopper to give mass flow is just the first stage in maintaining an even flow of material through a process. Having carefully designed the hopper, it is very easy to undo all the good work by incorrectly interfacing the feeder to the bottom of it. Some of the more commonly encountered mistakes involve the installation of screw feeders. If a mass flow pattern of discharge is to be allowed to develop within the storage vessel, it is essential that the whole of the outlet area is activated by the feeder. Therefore an increasing capacity for material to be drawn off is required along the inlet to the screw. Commonly installed items such as constant pitch screws, rotary valves and parallel outlets above belts will not achieve this.

Preferential draw

Taking screw feeders as an example, it can be appreciated that once the first parcel of material has fallen into the void created between the end pitch and the rear face of the trough, it is carried forward towards the outlet of the screw. This initial filling and transportation from the end of the screw prevents any material being extracted from points forward along the outlet. Thus, a preferential draw of material is created from the far end of the outlet whilst the material at the near end remains immobile and forms a shear face for the material drawn into the screw. In effect the hopper is now discharging in core flow (see Fig. 3a). The same effect happens if the screw is mounted in a trough with a `V' profile. The screw can only draw material from directly above, but the `V' trough presents material into the sides of the screw for which there is no conveying capacity. In these instances the static material can build up and extend in the hopper - again creating a core flow discharge pattern. The solution to these problems is to install a screw which has varying capacity along its length (achieved by either varying the pitch or shaft diameter - or both), and installing it into a `U' trough so that material is only presented above the screw (Fig 3b).

Similarly, the interfacing of rotary valves between hoppers and pneumatic conveying systems can (by the preferential draw characteristic described above) also introduce core flow. An effective way around this problem is to install a variable pitch screw to the outlet of the hopper which feeds material into the rotary valve. This gives the process engineer far more control over the quantities of material introduced into the pipeline, due to the screw acting as an effective metering device and the rotary valve performing the role of an air lock into the conveying pipeline. Although the initial costs associated with this type of approach are fairly high, the benefits in terms of improved product quality and reduced `give-aways' as a result of more accurate packaging, can quickly justify the expenditure.

Erratic flow can also result from an increase in the fines content which easily key into surface irregularities and imperfections in chutes and machinery. These self-supporting planes of material present shear faces over which moving material must pass - often depositing larger particles that can build up to substantial ledges of material.

While the above discussion has focused on gravity discharge hoppers specifically, for example those which feed into filling machines, it is important to realise that the same issues apply to the flow channels within the actual filling heads themselves. It is therefore important to ensure that the same principles of mass flow are applied throughout the solids flow route associated with the packing machines, from reception of the solids to the nozzle where material enters the pack.

Richard Farnish is a consulting engineer at The Wolfson Centre for Bulk Solids Handling Technology, University of Greenwich.