Key points
In recent decades, inventory monitoring systems for silos and bunkers have improved in terms of technology and quality. Such equipment types can be employed to offer indications of material volumes held in storage for equipment that cannot be readily inspected visually. Thus such systems can improve Health & Safety (i.e. no staff access to assess inventory) but also allow the adoption of fully automated processes.
Common types of equipment used to monitor storage equipment include radar, ‘dipping’ probes, level probes and load cells. The effectiveness of these approaches is largely dictated by the nature of the installation – with particular emphasis on the discharge behaviour of the silo/hopper/bunker that is being monitored.
The flow pattern that develops within the vessel is key to the level of performance achievable, and in this respect it is important for the reader to understand what can develop.
The two main discharge behaviours
The two main discharge behaviours that can develop in silos and bunkers are as shown in the figures 1 & 2 below, which illustrate these flow patterns in the context of conical silos, but the principles are equally applicable to bunkers.
Fig 1, illustrates a discharge behaviour called core flow, where the bulk particulates discharge preferentially down a vertical channel above the outlet. The bulk particulates are fed into the flow channel from the top free surface and the product around the walls remains static until the level descends to the point where it becomes the top surface and discharges. If this discharge behaviour is present, then it can impact on the process in several ways:
- The vessel will operate on a first-in, last out stock rotation (which can result in the unpredictable appearance of aged or different material on the process line if the inventory is maintained at a high level for a significant length of time).
- An increased variability in bulk density throughout the discharge cycle of the bin may become apparent due to variations in particle packing due to variability in residence time before discharge (recalling that the vessel will draw down freshly introduced material).
- If the material blend is subject to surface effect segregation (i.e. tends to separate in such a way that fines enrichment occurs in the central region – whilst the periphery of the vessel demonstrates a lack of fines), then initial discharges from the store will be fines rich (correlating to high bulk density or blend imbalance) whilst the final stages of the discharge will tend to feature a higher coarse content (low bulk density or blend imbalance).
By default, this discharge behaviour is most likely to be found in most processes and equipment types.
Fig 2 illustrates a discharge behaviour called mass flow. By contrast this type of discharge is arrived at by designing the vessel/equipment to suit the measured flow characteristics of the worst case material to be handled.
If this discharge behaviour is present, then the following benefits can be derived:
- The vessel operates on a first-in, first-out basis and as such maintaining a high inventory level does not affect the residence time of material being discharged.
- Stagnant regions of material are eliminated.
- The inventory is drawn down evenly through the equipment – thus providing a degree of recombination of radially segregated bulk particulates.
- Discharge can be sustained reliably without resorting to discharge aids.
The above (brief) descriptions summarise the principle differences in discharge behaviour between the two types of vessel geometries. It should be borne in mind that although the illustrations show conical type vessels, these discharge characteristics can be found in all other shapes and sizes of equipment.
Considering core flow (as this is the default flow behaviour in process plants) it can be appreciated that accurate monitoring of inventory can be problematic for a variety of reasons.
The characteristic ‘crater’ that forms during discharge can present issues for radar or ‘dipping’ probes, in that the point at which the respective reading is taken may only represent a height at a single surface radius – thus some calibration would need to be undertaken to allow for the actual gradient (and hence volume) within the vessel.
This approach would be further complicated if the material changes (i.e. the gradient may steepen or become more shallow) and introduce an error. Equally if the vessel is prone towards rat-hole formation (i.e. the flow channel empties, but material does not flow into it), then the diameter of the rat-hole and its height would represent missing ‘capacity’.
Further more….
Such problems of inconsistent surface profile can also develop if the vessel is subject to regions of time consolidated material (cliffs) that occur asymmetrically.
Level probes may be considered as an approach to overcome these short comings, but their operation can also be compromised by the characteristic presence of long term resident material on the flanks of the cone into which the probes intrude.
This is particularly an issue if the material being stored is cohesive, in which case ‘tuning fork’ type probes will simply indicate that they are covered – even though the vessel has discharged what contents can be reclaimed by gravity (often leaving substantial regions of ‘dead material’ if discharge aids have not been adequately maintained).
‘Paddle type’ sensors can also be vulnerable to build ups of cohesive materials that have consolidated and can, in some cases, simply excavate a ‘cavern’ resulting in a signal indicating an absence of material – despite quantities still being in place.
Load cells can also issue spurious information in the event that the vessel contains long term resident material that cannot self-drain. In such cases, it has been known that some companies progressively ‘tare’ systems through a year in order to only monitor ‘live’ capacity (thankfully not with a food grade product!).
In contrast if mass flow discharge behaviour is obtained, inventory control becomes a considerably simpler and more reliable proposition. Under such circumstances, the surface profile (the shape of which will be dictated by the bulk materials and the filling method) is likely to be highly repeatable and even throughout the vast majority of the discharge cycle to empty.
This makes the detection of the surface and any subsequent initial calibration of systems much simpler and more likely not to require ‘tweaking’ over time.
The establishment of a full cross-sectional flow will also avoid the potential problems of rat-hole formation and long term stagnant regions. Thus level probes mounted on the walls of the vessel will actually ‘see’ the true inventory as the level reduces.
Similarly, load cells will not need to take account of an unknown and growing mass of long term resident material (all of the inventory will ‘live’ during discharge operations).
Summary
In summary, it is widely recognised that without effective monitoring of inventory levels processes can be extremely vulnerable to short falls in feed stock or spillages through over filing of bulk and day bins.
However, the influence of the storage equipment discharge pattern is seldom taken into account when selecting monitoring systems – and yet a lack of consideration of this factor is often at the root of a range of process problems (of which monitoring is only one!).
