Soma Gold Corp is advancing its mineralized material upgrading strategy at its El Bagre operation in Colombia with TOMRA Mining’s sensor-based sorting technology. This system combines XRT technology with advanced artificial intelligence through OBTAIN™ and CONTAIN™, according to TOMRA.
After positive test results at the TOMRA Test Center in Germany, Soma decided to install a COM Tertiary XRT 1200 sorter to boost process efficiency and maximize the value of its mineralized materials. Soma, focused on gold exploration, development, and production in Colombia, operates mainly in Antioquia, aiming to optimize performance, maintain strong operational efficiency, and utilize resources effectively.
El Bagre processes gold-bearing material from Soma’s own mines and third-party sources, requiring flexible and efficient handling for varied material types and grades. Like other gold operations, Soma must manage variable mineral characteristics to keep plant performance stable and efficient, especially reducing non-valuable material entering energy-intensive stages like crushing and grinding.
By removing waste early, Soma aims to improve feed consistency, optimize downstream capacity, and reduce unnecessary wear and energy consumption. Mark Bren, VP Operations at Soma, stated: “Sensor-based sorting can play a key role in managing our mineralized material more efficiently. TOMRA showed consistent and reliable results under modeled conditions during our tests. By concentrating valuable material and removing waste early, we see clear potential to improve overall plant performance.”
The core solution is TOMRA Mining’s X-ray Transmission (XRT) technology, which differentiates material based on atomic density for precise separation. The system at Soma uses dual-energy XRT sensing and advanced AI algorithms for accurate particle classification.
TOMRA’s innovations, OBTAIN and CONTAIN, enhance this capability. OBTAIN applies deep learning for high-precision sorting at the individual particle level even at high throughput, ensuring consistent performance regardless of belt occupancy. CONTAIN leverages advanced neural networks to detect valuable mineralization embedded within host rock, addressing challenges that conventional methods struggle with.
Bren added: “The real value lies not just in upgrading material but also in improving overall plant operation. If we reject non-valuable material sooner, we optimize downstream capacity and process mineralized material more effectively.”
To evaluate the technology, Soma conducted benchmark tests at the TOMRA Test Center using material from their Colombian operations. Results confirmed a clear upgrade of mineralized material, strongly validating the technology.
Tests covered different size ranges and material types, confirming robust performance across varying feed conditions. Special attention focused on challenging scenarios like small particles and high belt occupancy, where the system excelled at detection and sorting.
Bren said: “What stood out was not only the upgrade level achieved but also the consistency across material types and sizes. That gave us confidence to proceed with implementation.”
Fernando Romero of TOMRA Mining noted: “The test work showed that sensor-based sorting efficiently separates lower-value material, enabling more controlled and optimized processing.”
A key advantage observed was the sorter’s ability to maintain precision under demanding conditions, performing well with both coarse and fine particles and dense material flows.
Integrating TOMRA’s high-speed TS100 ejection module reduces compressed air consumption by up to 70% while maintaining reliable separation, enhancing operational efficiency and reducing costs.
Following these results, Soma placed an order for a COM Tertiary XRT 1200 sorter to be integrated into its main processing flowsheet, improving material handling and plant efficiency. The sorter’s integration enables more efficient handling and better plant performance, stabilizing material flow within the main process, TOMRA reports.
The project is developed in close partnership with DISMET, TOMRA’s local integrator, who supports design and implementation tailored to El Bagre’s specific needs.
One major advantage of TOMRA’s sensor-based solutions is improved process control and material value. By removing non-valuable material early, operations can better manage feed volume and consistency while increasing processed stream value.
Romero commented: “Rejecting waste early contains variability and processing load, obtaining a higher-value stream. This directly impacts plant efficiency.”
Combining TOMRA’s XRT sensing, deep learning, and precise ejection tech, mining operations can stabilize processing, reduce unnecessary throughput, and unlock value from marginal materials.
As Soma moves toward full implementation at El Bagre, sensor-based sorting is expected to become central to its operational strategy. By combining validated test work with advanced XRT and AI-driven technologies, Soma aims to further enhance performance and efficiency across its Colombian sites.
