🎓 Lesson 3
D2
Failure Physics: Bathtub Curve, Wear-Out Mechanisms & Stress-Strength Interference
Reliability over time follows a bathtub-shaped curve—high early failure, then low steady failure, then rising failure as equipment wears out—and understanding this helps prevent unexpected breakdowns in mining conveyors, crushers, and feeders.
🎯 Learning Objectives
- ✓ Explain the physical origins of each phase of the bathtub curve using real failure modes in conveyor idlers or crusher liners
- ✓ Calculate reliability R(t) using stress-strength interference for a normally distributed stress and strength distribution
- ✓ Analyze field failure data to identify dominant phase (infant, random, or wear-out) and recommend targeted mitigation strategies
- ✓ Design inspection intervals for critical subsystems (e.g., gearboxes in apron feeders) based on wear-out onset predictions
📖 Why This Matters
In underground and open-pit mines, unplanned stoppages in materials handling—like belt conveyor jams or primary crusher bearing seizures—cost $15,000–$50,000 per hour in lost production. The bathtub curve isn’t abstract theory: it explains why a new crusher gearbox fails within weeks (infant mortality), runs smoothly for 2 years (useful life), then suffers accelerating bearing spalling (wear-out). Recognizing which phase you’re in determines whether you need better commissioning, condition monitoring, or proactive replacement—directly impacting MTBF, maintenance budgeting, and safety.
📘 Core Principles
The bathtub curve emerges from competing physical mechanisms: infant mortality stems from latent manufacturing flaws (e.g., microcracks in cast crusher housings), useful-life failures arise from rare random events (e.g., foreign object ingestion into a feeder drive), and wear-out results from deterministic degradation—fatigue, abrasion, corrosion, and thermal cycling. Stress-strength interference models reliability as P(S > σ), where S is material strength (e.g., Brinell hardness of jaw plate alloy) and σ is time-varying operational stress (e.g., cyclic compressive load from ore impact). As wear reduces S(t) and/or increases σ(t), the overlap region grows—reliability declines. In mining, wear-out is rarely sudden; it manifests as progressive loss of dimensional tolerance, increased vibration (>4.5 mm/s RMS), or rising motor current variance (>8% CV).
📐 Stress-Strength Interference Reliability
When stress (σ) and strength (S) are independent normal distributions, reliability R is calculated using the standardized difference δ = (μ_S − μ_σ) / √(σ_S² + σ_σ²), where Φ(δ) gives the cumulative probability. This formula enables quantitative comparison of design margins against actual operating conditions.
Normal Stress-Strength Reliability
R = Φ[(μ_S − μ_σ) / √(σ_S² + σ_σ²)]Reliability when both stress and strength follow normal distributions.
Variables:
| Symbol | Name | Unit | Description |
|---|---|---|---|
| μ_S | Mean strength | MPa | Average tensile/compressive yield strength of component material |
| μ_σ | Mean applied stress | MPa | Average operational stress from load, impact, and thermal cycling |
| σ_S | Strength standard deviation | MPa | Variability in material strength due to casting/heat treatment |
| σ_σ | Stress standard deviation | MPa | Variability in operational stress due to ore size distribution, feed rate fluctuations, and alignment |
Typical Ranges:
Crusher jaw plates (Mn18Cr2): μ_S = 1,100–1,350 MPa; σ_S = 35–55 MPa
Conveyor pulley shafts under cyclic loading: μ_σ = 280–410 MPa; σ_σ = 45–75 MPa
💡 Worked Example
Problem: A gyratory crusher main shaft bearing has strength distribution: μ_S = 1,250 MPa, σ_S = 42 MPa (from material test data). Operational stress distribution under typical ore flow: μ_σ = 980 MPa, σ_σ = 65 MPa (from strain gauge & FEA validation). Calculate reliability R.
1.
Step 1: Compute mean difference: μ_S − μ_σ = 1250 − 980 = 270 MPa
2.
Step 2: Compute combined standard deviation: √(42² + 65²) = √(1764 + 4225) = √5989 ≈ 77.4 MPa
3.
Step 3: Compute δ = 270 / 77.4 ≈ 3.49; look up Φ(3.49) ≈ 0.99975
Answer:
The reliability is 99.975%, meaning ~1 failure expected per 4,000 operating hours — well within target for critical rotating equipment (target R ≥ 99.9%).
🏗️ Real-World Application
At Newmont’s Boddington Mine (WA), a fleet of 12 × 1.8 m wide overland conveyors experienced premature pulley lagging wear after 4,200 hrs—well below the 12,000-hr design life. Vibration and thermal imaging revealed localized slip heating at 3 o’clock position. Root cause analysis showed stress concentration from misalignment increased effective σ by 32%, while aging reduced S by 18% due to UV/ozone degradation of rubber compound. Applying stress-strength interference with updated distributions predicted 92% reliability at 4,200 hrs—confirming wear-out dominance. Mitigation included laser alignment certification (reducing σ variability by 40%) and switching to EPDM lagging (increasing μ_S to 1,420 MPa), extending life to 11,800 hrs.
🔧 Interactive Calculator
🔧 Open Mine Materials Handling System Reliability Calculator📋 Case Connection
📋 Iron Ore Export Terminal Conveyor Reliability Upgrade
Chronic belt splice failures (>22 unscheduled stoppages/yr) causing demurrage penalties and stockpile congestion
📋 Open Pit Gold Mine Stacker-Reclaimer Rail Alignment Reliability Program
Repeated rail misalignment (±8mm lateral deviation) causing slewing gear tooth pitting and emergency shutdowns
📋 Coal Mine Thermal Lagging Failure on High-Temperature Conveyor
Rubber lagging delamination on 120°C discharge conveyor due to thermal cycling (80–120°C), causing slippage and fire ris...