Published on: September 13, 2021
The design of mining equipment is a subject of high complexity. Despite immense mass flows and wear-promoting materials, reliability and service life must be maximized. Software for simulating the transport and sorting of rock can make a significant contribution to optimizing such equipment.
Centristic, a Southwest UK-based practical engineering services supplier, teamed with CADFEM and Rocky DEM in a major project to design and install new conveyors, structures and equipment at a large granite quarry near Loughborough, England. The crushed primary product was to be transported from the quarry to a screen house, where fine rock has to be screened off. In the new development of the plant, the boundary conditions of the screen house were critical since the inlet for the rock as well as the two outlets are predefined. The overall system of the screen house consists of two chutes and a screen.
The material – 2400 tons of granite per hour – arrives at the new screening house from the primary crusher on a 1.8-metre-wide conveyor belt. There, after passing through the screen, it is unloaded onto one of two conveyor belts. The high feed rate, the highly abrasive effect of the rock and the dimensions of the equipment required to process the rock presented major challenges. It was particularly difficult to divert and best distribute the rock on the screen to ensure effective separation. Rocky DEM (Discrete Element Method) software was used to analyze the material flow, but the screen was not included in the simulation because it is a supplier part for Centristic.
When developing the inlet chute, Centristic designed the internal design as well as the surrounding supporting structure. Through internal studies, Centristic staff arrived at a chute design that could meet the requirement, but only under optimal conditions. Despite their many years of extensive experience, the material handling specialists were unsure whether they could guarantee the trouble-free operation of the system. To avoid time-consuming and financially costly rework, the design was handed over to CADFEM together with all known details of the rock to be conveyed. There, either the feasibility of the original design was to be confirmed or an alternative, functioning solution was to be found.
The task of the inlet chute is to deflect the material in the direction of the screen. In this case, this means a change of direction of 145° (see picture). In addition, the material must be distributed from a 1.8-metre-wide conveyor belt to the full 2.9 meter width of the screen. This distribution should be as even as possible (uniform distribution), which was required by the screen manufacturer. In addition, the risk of clogging should be kept low.
In order to check these requirements for the system in the classic design, the only option is to build a prototype or a model. However, both are associated with a high expenditure of time and money and were unthinkable for this project, so the virtual prototypes and the simulation with Rocky DEM were the saving alternative.
Within the scope of the project, the initial design and five optimizations were calculated. Stone-shaped polyhedron particles were used, corresponding to the client’s grading curve in a size of 37.5 to 300 millimeters. Fines below 37.5 millimeters were not considered in the simulation, as they increase the calculation time and hardly influence the material behavior. The material was calibrated according to the customer data and then used for the simulation. The calculation time for each design was about 48 hours. The individual optimizations were discussed between the Centristic and CADFEM employees for each variant and the resulting new modifications were implemented within one day.
In order to achieve a uniform mass flow via the outlet to the connected screen, this was determined in six sub-ranges (see figure). Through the optimizations, the deviation from the uniform distribution of -10.5 / + 6.5 mass percent could be reduced to ± 1.8 mass percent.
For the initial design of the input chute and five variants, the calculations were performed with Rocky DEM.
In the initial design, the deflection of the material was not optimal, which leads to an uneven mass flow at the outlet (deviation from the ideal state -10.5 / +6.5 %). In addition, a so-called dead zone is created, so that a great deal of material remains in the chute (see circle).
Stages were added to increase the flow velocity and reduce the dead zone. However, this had no measurable effect on the distribution at the outlet.
Beveling the upper protrusion to redirect the material brought a slight improvement in uniformity.
Reorganization and beveling of the steps to deflect the material indicated a further improvement in the mass flow rate.
The use of beveled ledges on the two protrusions to direct the material from the center to the edges results in a strong shift of the mass flow to the right side. This solution also tended to clog.
With the return to a shortened, angled top protrusion, the goal is achieved: Most uniform material flow with ± 1.8%.
The outlet chute also posed considerable design challenges. The chute has two outlets, both of which have to be flowed through equally well. Here, the small installation space and the small height difference between inlet and outlet pose hurdles. The main problem, however, was the screen. It is a model with two decks, which were necessary in order to screen safely despite the large mass flow. The two mass flows of coarse and fine material resulting from this sieve had to be reliably combined, regardless of which of the two exits the material was let out through.
Furthermore, during operation, it must be possible to open and close the associated slides in front of each outlet. Therefore, blockages and remaining material in the critical areas must be avoided. No solution for this requirement emerged from the empirical values, which is why only the creation of a prototype or a model would have been an option here, too, according to the classic approach.
In the second sub-project, the initial design and five optimizations were calculated with Rocky DEM, whereby the calculation time for each design was around 24 hours. Here, too, the customer’s specified grading curve (37.5 to 300 millimeters) minus the screened-out particle sizes was considered. The biggest problem was found to be reliable mixing, as well as preventing clogging at the remote outlet 2.
Calculations were performed with Rocky DEM for both the initial chute design and five optimized variants.
The initial design resulted in large material deposits with gate 1 closed and blockages of the chute with both gate 1 and gate 2 open.
During revision 1, modifications were made to the lower section and an intermediate level was installed. The aim was to reduce sedimentation and increase the flow velocity to outlet 2. A better flow through outlet 1 was achieved, but no improvement at outlet 2.
During revision 2, an enlargement and relocation of the gate valves to the intermediate level took place. But as before, severe segregation and the risk of clogging was observed at slide gate 2.
Revision 3 included the reconstruction of the outlet area of outlet 2 to increase the flow velocity. However, this did not result in improved flow.
In revision 4, the step after gate 2 was removed to increase the flow velocity. The step served to prevent wear, but its removal did not result in improved flow.
With revision 5, the outlet area of outlet 2 was rebuilt to increase the flow rate. Likewise, the space was enlarged for a higher volume flow. In addition, the slope was increased to reduce resistance and increase the flow velocity. This enabled a reliable flow through the chute for outlet 2 and also ensured the mixing of the two inlet mass flows for both outlets.
Through the Rocky DEM flow and wear analysis, Centristic were able to identify areas which required additional protection and where stone on stone contact could be relied on, making provision of linings efficient. Alongside this element, Centristic, with client input, chose suitable lining media for the various impact locations using site-based experience, meaning the chutes were supplied with readily available products that the site knew would be appropriate for the application.
Compared to classic chute design techniques, Centristic were able to provide fabrications based on real-time flow simulations using client supplied material properties. These parameters allowed the Rocky DEM analysis to show actual flow of the graded material through changes in direction and in flight, where previously, engineers would only be able to see this real information during commissioning.
The use of Rocky DEM proved to be very useful – firstly in the actual design of the chutes and secondly in the communication with the customer.
During the chute design, it was only through the use of Rocky DEM that it was possible to complete the tasks to be done within the tight schedule. The Centristic designers were able to make all basic design decisions based on the geometric constraints and the simulation results of the calculated variants. This ensured, even before commissioning, that the rock would pass through the chutes without any problems and arrive correctly on the screen or conveyor belt.
Furthermore, with the help of the simulations, Centristic was able to convince the client that his specific requirements had been fully taken into account. Especially the specific operating conditions of the process screen, but also those of the entire plant, were well represented by the use of Rocky DEM and formed the basis for the final design of the individual plant components.
This blog post was originally published by CADFEM.
