Sandvik Mining and Rock Technology has launched the "Crush-Vision" intelligent control system for its 2026 lineup of QH442 Hydrocone mobile cone crushers. In aggregate production, the hardness and size of the raw feed from the primary jaw crusher constantly fluctuate based on the blast pattern and rock face geology. Traditional cone crushers rely on a fixed Closed-Side Setting (CSS) and a mechanical accumulator to protect against uncrushable tramp steel. When feed size increases or the rock becomes harder, the mantle and bowl liner open up to relieve pressure, resulting in oversized end product. When the feed softens, the CSS remains wide, leaving uncrushed fines and reducing the yield of the target specification aggregate.
The Crush-Vision system integrates a "High-Speed Machine Vision Camera" mounted directly above the product discharge conveyor. The engineering breakthrough is the "Gradation Feedback Loop." The camera uses machine learning to analyze the particle size distribution of the crushed aggregate on the belt 30 times per second. The ECU correlates this visual data with the real-time crusher frame vibration and hydraulic pressure.
If the system detects an increase in oversized particles on the belt-indicating a harder or larger feed chunk has forced the bowl liner to lift-the ECU preemptively commands the hydraulic motor to tighten the CSS by 2 millimeters. This compensates for the mechanical lift, ensuring the effective crushing gap remains constant and the product gradation stays within specification. Conversely, if the camera detects excessive fines (over-crushing) during a soft feed phase, the system opens the CSS to reduce energy waste and liner wear. This closed-loop vision and CSS control increases the yield of in-spec aggregate by 18% and extends the manganese liner life by 22% by preventing the destructive high-pressure crushing of soft material.
