Dr Priscilla Freire, Engineer at TUNRA Bulk Solids, discusses some of the competing design criteria that engineers are often faced with when designing bulk material handling systems.
Emeritus Professor Alan Roberts, TUNRA Bulk Solids’ founding director and one of Australia’s most prominent scientists in the field of bulk solids handling, often makes a big claim in TUNRA’s courses: “Bulk Solids Handling isn’t rocket science… it’s harder”. One of the challenges engineers and designers often face is the so-called competing criteria, which often turn a design task into trade-off exercises.
In the design of transfer chutes for instance, rock-box type chutes and/or ledges are often implemented to reduce abrasive wear of the chute wall. However, when handling wet and sticky materials, these solutions may often result in material build-up that can lead to blockages or nonoptimal flow patterns. Other examples of competing criteria include the implementation of deflector plates to guide the material stream and ensure that material is loaded symmetrically onto the receiving belt, however potentially increasing wear of the chute itself. After decades of experience, TUNRA Bulk Solids has developed a set of guiding principles for the design of transfer chutes, which have been presented previously in the November 2020 edition of ABHR.
This article presents a recent project conducted by TUNRA Bulk Solids’ engineers regarding the design review of as-built transfer chutes for handling highly cohesive materials. All chutes were originally designed and fabricated by third-parties and TUNRA was engaged to conduct a flow assessment prior to the commissioning of these chutes and propose conceptual modifications for improvements where required.
Modelling with Discrete Element Method (DEM) was conducted on the as-built geometries with a focus on aspects such as belt tracking, wear of the chute itself and of the receiving belt, propensity to blockages and, in one specific case, propensity to spillage and/or dust generation in a very tall chute. The DEM models were carefully calibrated with material properties through lab tests of all materials handled by the chutes, using the material properties at the worst-case moisture to ensure a conservative analysis.
Considerations on material properties:
The two chutes under analysis have been designed to handle, among other materials, fine tailings with very high bulk strength at its worst-case moisture content. Under these conditions, the material has shown high compressibility, with an increase of almost 50% from an unconsolidated value to 50 kPa of consolidation. It should be noted that compressibility often varies significantly with moisture content. Additionally, wall friction testing results showed moderate to high friction coefficients for the liners at the conditions that were tested.
The adequate calibration of DEM models is paramount to ensure that the flow simulated is actually representative of the expected conditions for the given materials at the relevant moisture contents and consolidation pressures. Although DEM is often applied for the prediction of macro material behaviour, its implementation requires the input of parameters on the micro scale. Some of these parameters cannot be directly measured in a laboratory, and a suite of calibration tests is required to produce a macro characteristic that is dependent on the nature of the bulk material, which are then repeated within the simulation domain, with necessary simplifications, and the micro parameters are optimised to yield the desired macro result/behaviour, as seen in the March 2021 edition of ABHR.
Chute 1:
The first transfer chute transfers material from the feeding to the receiving belt with a 90o angle, and transfers with an angle are often challenging to design to ensure that the material will flow onto the receiving belt in a symmetrical manner. After calibrating the DEM model with parameters obtained from laboratory testwork, the simulations showed a region of stagnant material in the lower section of the chute (Figure 1).
Material slowly builds up on the lateral ledges located in the bottom part of the chute, and then shears onto itself, falling onto the receiving belt in a non-central manner as shown in Figure 2, which may lead to belt mistracking issues during operation. One of the design objectives requested by the client was to eliminate, where possible, all non-central loading cases.
Another design issue identified in Chute 1 was the positioning of the deflector: the simulation of the original geometry showed material spillage at the upper part of the deflector.
TUNRA’s engineers proposed repositioning the deflector both horizontally and vertically, bringing it closer to the discharge pulley, as well as rotating the deflector at approximately 6 degrees in relation to the original position. The other recommendation was the removal of the ledges located at the bottom part of the chute. These combined modifications, when simulated in the DEM environment, showed a significant improvement in the loading profile as material falls more directly onto the receiving belt, as shown in Figure 3.
One of the “side effects” of the proposed modifications, however, is a potential increase in belt wear, caused by the more direct loading of material onto the receiving belt: Figure 4 shows the belt wear profiles with the original chute geometry and the proposed modifications. Given that the wear intensity in both simulations uses the same scales, the darker shades of red in the simulation of the modified geometry indicates a higher intensity than with the original geometry. As such, major modifications of the lower chute will be required to increase the in-line component of the stream velocity at the discharge point, aiming at reducing belt wear. It is important to consider that asymmetric loading is a significant issue that, among other consequences, can cause material spillage, belt damage and segregation with consequent downstream processing issues.
Chute 2:
The second chute is a bifurcated chute handling two materials with very different properties. This article focuses on the left leg, which, similarly to Chute 1, also handles tailings at a high moisture content with cohesive properties.
Due to structural constraints, Chute 2 transfers material from the feeding to the receiving belt at a considerable height of 20 m. Preliminary analysis of the original design indicated that the material freefalls with minimal wall interactions before impacting the discharge chute section, which happens at a relatively high vertical velocity. Material builds up asymmetrically and shears across itself, loading non-centrally onto the receiving belt with an approximate mass distribution of 47% on the left side and 53% on the right side, which may result in belt tracking issues. The flow of material considering the original geometry can be seen in Figure 5.
Furthermore, relatively stagnant regions of material were observed on the receiving belt and along the length of the skirts in the loading region which may lead to increased belt wear, with a propensity of build-up on the skirts anticipated. It was concluded that the stagnant region is primarily due to the reacceleration of material in the direction of the belt, as the majority of the material falls directly on the receiving belt, with little to no material stream velocity in the direction of the outgoing belt. Such high fall with little contact with the side walls may also increase the propensity to dust generation, particularly in the lower section, as a result of the high material velocities due to relatively long periods of freefall of the material stream.
In order to address the potential operational issues anticipated with the original geometry, TUNRA’s engineers proposed the insertion of two impact plates in the top and bottom sections of the trouser leg, and a training plate in the discharge section of the chute to reduce periods of material freefall and improve the loading characteristics onto the receiving belt. The aim of these plates is to “guide” the material stream inside the chute such that it flows with less material dispersion and is loaded onto the receiving belt in a more centralised manner, thus reducing the propensity to dust and belt mistracking issues, respectively. The proposed modifications and resulting flow can be seen in Figure 6.
With the suggested modifications, a significant reduction was observed in the wear of the receiving belt, which is likely caused by the reduction in the velocities at which material hits the belt. On the other hand, an increase in the actual chute wear was also observed, caused by the material stream now being directed onto the chute wall surface, whereas with the original geometry the material stream was freefalling inside the chute. Another point of relevance is that the proposed inserted plates are likely going to experience significant wear due to the direct and ongoing contact with material. Considerations regarding the durability of the plate material as well as ease of access for replacement or maintenance should be made in the implementation of the suggested modifications.
Conclusion
Some of the greatest challenges in the bulk solids handling field is to accommodate competing design criteria and constraints whilst still to ensure smooth flow and meeting operational requirements. An understanding of material behaviour through appropriate characterisation is the first step to shed light on the different challenges that may be encountered, and computational simulation consists of a good tool for flow visualisation and identification of points of concerns. However, it is not always possible to meet all design criteria simultaneously, and trade-offs are often necessary in order to make informed design decisions.