A novel tool monitoring approach for diamond wire sawing
At a Glance
Section titled âAt a Glanceâ| Metadata | Details |
|---|---|
| Publication Date | 2021-11-10 |
| Journal | Production Engineering |
| Authors | Berend Denkena, Benjamin Bergmann, Björn-Holger Rahner |
| Institutions | Leibniz University Hannover |
| Citations | 2 |
| Analysis | Full AI Review Included |
Executive Summary
Section titled âExecutive SummaryâThis research introduces a novel, robust, and real-time monitoring system for detecting tool failure in mobile diamond wire sawing, addressing critical safety and productivity gaps inherent in manual inspection.
- Core Problem Solved: Eliminates the need for intermittent manual inspection (which risks wire breaks and catastrophic failure due to segment displacement) by providing continuous, in-process tool monitoring.
- Sensing Methodology: Utilizes a non-contact, inductive (eddy current) measuring principle, selected for its high sampling rate capability (> 100 kHz) and robustness against harsh environmental conditions (water, mud, dust).
- Monitoring Feature: Tool condition is monitored using time-based features derived from the sensor signal, specifically the time interval between two signal minima (tmin).
- Performance Achievement: The tmin feature proved highly robust, enabling reliable detection of displaced grinding segments starting from a minimum displacement of 2 mm on a wire moving at 20 m/s.
- Process Robustness: The monitoring approach is independent of the toolâs bonding matrix (sintered, galvanic, vacuum-brazed) and the wire velocity (tested up to 20 m/s).
- Advanced Logic: A dual-sensor approach is proposed to logically distinguish between actual segment displacement (tool failure) and wire slip (process defect), preventing false alarms.
Technical Specifications
Section titled âTechnical Specificationsâ| Parameter | Value | Unit | Context |
|---|---|---|---|
| Typical Cutting Speed (vs) | Up to 30 | m/s | Mobile diamond wire sawing |
| Inductive Sensor Type | Eddy Current | - | Displacement sensor |
| Inductive Sensor Range | 4 | mm | Measuring range on flat ferromagnetic targets |
| Data Acquisition Rate | 50 | kHz | Sampling rate used in experiments |
| Minimum Required Sampling Rate (f) | > 15 | kHz | Calculated requirement for 4 mm segment at 30 m/s |
| Test Wire Tension (Fs) | 1.000 | N | Applied during stationary and moving tests |
| Test Sensor Distance (s) | 1.5 | mm | Optimal distance for feature extraction |
| Segment Length (ls) - Tool 1 | 6.8 ± 0.37 | mm | Sintered bond tool |
| Segment Distance (ll) - Tool 1 | 25.1 ± 0.24 | mm | Sintered bond tool |
| Segment Density - Tool 2 | 53 | segments/meter | Vacuum-brazed tool (highest density tested) |
| Reliable Detection Threshold (ldisp) | 2 | mm | Minimum displacement detectable using tmin on moving wire |
| Maximum Standard Deviation (tmin) | 46.8 | ”s | Maximum deviation (±3Ï) on moving wire (20 m/s) |
| Maximum Standard Deviation (tt) | 63.8 | ”s | Maximum deviation (±3Ï) on moving wire (less robust than tmin) |
Key Methodologies
Section titled âKey MethodologiesâThe monitoring approach was developed and verified through a structured process involving sensor selection, two test rig setups, and feature robustness analysis:
- Measuring Principle Selection: Inductive measurement was chosen over optical or capacitive methods because the metallic grinding segments are electrically conductive, and the principle is inherently robust against common environmental contaminants (water, mud, dust) present in wet wire sawing.
- Test Rig 1 (Stationary Wire, Moving Sensor): Used for initial functional verification. This rig allowed the sensor to be moved along the wire tool with defined boundary conditions, preventing influences from wire vibration or velocity changes. It confirmed that the signal curve was comparable across all three common bonding matrices (sintered, galvanic, vacuum-brazed).
- Feature Definition: Two time-based features were defined based on the correlation $l = t \cdot v_s$:
- Segment Duration (tt): Time difference between the beginning of two consecutive segments (determined by threshold crossing).
- Time Interval Between Minima (tmin): Time difference between the respective extremes (minima) of the signal curve.
- Feature Robustness Analysis (Rig 1): The features tt and tmin were compared for robustness against varying sensor distances (0.5 mm to 2.0 mm) and threshold values. It was found that tmin had a standard deviation 22% lower than tt and was independent of the sensor distance.
- Test Rig 2 (Moving Wire, Stationary Sensor): Integrated into a CNC surface grinding machine to simulate real-world high-speed operation (up to 20 m/s). This rig confirmed that the sensor signal was independent of the tool velocity.
- Minimum Displacement Detection (Rig 2): A prepared diamond wire tool (Tool 2) was used where one segment was movable. Systematic variation of the displacement (ldisp) confirmed that tmin reliably detects displacement starting at 2 mm, making it the preferred feature for monitoring.
- Monitoring Logic Implementation: The final approach compares the actual feature value (tt or tmin) against a moving average with a tolerance band. A second, offset eddy current sensor is required to differentiate between segment displacement (where both sensors show deviation) and wire slip (where results do not match).
Commercial Applications
Section titled âCommercial ApplicationsâThe developed monitoring technology is crucial for improving safety, reliability, and productivity in high-risk and high-value cutting operations:
- Nuclear Decommissioning: Used for the dismantling of nuclear facilities where reliable, remote cutting of heavily reinforced concrete or steel is required.
- Renewable Energy Infrastructure: Dismantling of large steel constructions, such as wind turbine components.
- Construction and Civil Engineering: General cut-off grinding applications, including wall sawing and separation of large concrete/steel structures.
- Quarrying and Natural Stone Industry: Improving the safety and efficiency of rock extraction processes.
- Machine Tool Integration: The low-computation monitoring logic (based on simple time difference calculations) is suitable for real-time integration into standard machine tool controls (e.g., Siemens 840d sl).
- Safety Enhancement: Direct mitigation of the âwhip effectâ risk associated with segment detachment and subsequent wire breakage, leading to serious or deadly accidents.