.png)
Steam turbine blade failures are the single largest source of unplanned outage hours in fossil and nuclear plants — a single dropped LP last-stage blade can sideline a 600 MW unit for 14 days and cost upward of $3.5 million in lost generation alone. Effective steam turbine blade inspection combines visual mapping, dimensional erosion checks, eddy-current and penetrant NDT, and metallurgical replication timed to ASME PTC 6 intervals and OEM lifecycle curves. This guide consolidates inspection cadence, erosion measurement thresholds, cracking acceptance criteria, and replacement economics so reliability engineers can defend outage scope and stretch blade lifecycles with confidence. For teams ready to digitize blade records and CMMS-linked lifecycle tracking, you can Start Free Trial and centralize every NDT report in one asset register.
Blade Reliability · Inspection Intelligence
Are your last-stage LP blades one outage away from a destructive failure?
Erosion, fatigue cracking, and FOD accumulate silently between outages. A disciplined NDT program catches 92% of blade defects before they propagate to failure — yet most plants still rely on spot-checks rather than full-blade mapping. Build a defensible inspection workflow in days, not outage cycles.
Inspection Cadence · Outage Timing
When to inspect: the blade lifecycle timeline
A 600 MW steam turbine typically runs 6,000–8,000 hours per year. OEM lifecycle curves and ASME PTC 6 guidance define inspection windows by operating hours, not calendar time — and missing a window doubles the probability of in-service blade loss.
Visual & borescope sweep
Non-intrusive borescope of HP/IP/LP blade paths during minor outage. Targets FOD, leading-edge erosion pitting, and deposit buildup. Detection threshold: surface defects ≥0.5 mm.
LP blade eddy-current scan
Last-stage and next-to-last-stage LP blades scanned with eddy-current array probes. Detects surface-breaking fatigue cracks at blade roots and lacing wires down to 0.2 mm depth.
Full NDT + dimensional check
Major overhaul: UT of blade roots, dye-penetrant on shrouds and tenons, chord-length erosion measurement, metallurgical replication on suspected creep zones. ISO 55000-aligned evidence package.
Lifecycle & replacement decision
Cumulative erosion, crack density, and operating-hour data reviewed against OEM curves. Decide on blade re-tipping, full LP disc replacement, or life extension with enhanced monitoring.
Erosion Quantification · Leading-Edge Loss
Erosion measurement: formulas, thresholds, and action triggers
Moisture-impingement erosion on LP last-stage blades is unavoidable — but it is measurable and predictable. Track chord loss against OEM thresholds and you can stage replacements months before efficiency collapse.
C₀ = original chord length (mm) · Cₙ = measured chord after n operating hours. Action threshold: 6% CL on L-0 blades, 8% on L-1.
Baseline for Ti-6Al-4V L-0 blades: 0.18 mm/kh. For 12Cr stainless: 0.42 mm/kh. Rate spikes indicate moisture separator drift or load-following abuse.
Empirical correlation from ASME PTC 6 testing. A 10% chord loss on L-0 ≈ 0.35% stage efficiency drop ≈ $180K/yr on a 600 MW unit at $40/MWh.
A 500 MW coastal plant measured 7.2% chord loss on L-0 blades after 38,000 hours. At ER = 0.19 mm/kh, the OEM's 9% replacement trigger would be hit at ~44,000 hours — well within the next outage window. By staging replacement parts 9 months ahead, the plant avoided a forced outage that would have cost $2.1M in replacement power and lost availability. Capturing this in a CMMS-linked lifecycle register means the next engineer inherits the decision logic, not a spreadsheet graveyard.
Method Selection · NDT Playbook
Which NDT method catches which defect — and where
No single NDT technique covers every blade failure mode. Match the method to the defect type, blade location, and access constraints during the outage window.
| NDT Method | Target Defect | Blade Location | Detection Limit | Outage Time |
|---|---|---|---|---|
| Eddy Current (ET) | Surface fatigue cracks | LP root, lacing wire | 0.2 mm depth | 4–6 hrs / blade row |
| Dye Penetrant (PT) | Surface-breaking cracks | Shrouds, tenons, trailing edge | 0.05 mm width | 2–3 hrs / stage |
| Ultrasonic (UT) | Subsurface & root cracks | Fir-tree root, disc steeple | 1.0 mm length | 8–12 hrs / disc |
| Magnetic Particle (MT) | Surface & near-surface | HP/IP blade shrouds | 0.1 mm width | 1–2 hrs / stage |
| Replication | Creep cavitation, microstructure | HP blade root neck | Grain-level | 3–4 hrs / location |
| 3D Laser Scan | Erosion profile, chord loss | L-0, L-1 full airfoil | 0.02 mm accuracy | 2 hrs / blade |
Acceptance Criteria · Go / No-Go
When to replace: blade acceptance thresholds
Replacement decisions hinge on documented thresholds — not gut feel. Use these industry-benchmarked go/no-go criteria to defend outage scope and avoid both premature spend and in-flight failure.
Erosion — Replace
- Chord loss exceeds 9% on L-0 blades
- Erosion rate trending above 0.35 mm/kh
- Leading-edge pitting covers >30% of airfoil height
Cracking — Replace
- Any crack >2 mm at blade root or fir-tree
- Multiple cracks <2 mm in same root section
- Lacing wire cracks at >2 adjacent blades
Monitor — Re-inspect
- Chord loss 6–8% with stable erosion rate
- Isolated crack indications <1 mm, non-propagating
- Replication showing Stage 2 cavitation only
Cost of Inaction · Replacement Economics
What delayed blade replacement actually costs
A planned LP blade replacement during a scheduled major outage costs $450K–$800K. A forced outage from a dropped blade costs 4–7× more — plus collateral damage to the condenser, diaphragms, and downstream stages.
"We extended L-0 blade life by 18 months on three units using structured erosion-rate trending in Oxmaint. That deferred $2.4M in replacement spend and let us stage the parts order across two fiscal years."
Turn blade inspection data into defensible outage decisions
Centralize NDT reports, erosion trends, and replacement thresholds in one CMMS-linked register. Start in under an hour.
FAQ · Blade Inspection & NDT
Steam turbine blade inspection — frequently asked questions
How often should steam turbine blades be inspected?
HP and IP blades require borescope inspection annually during minor outages, with full NDT every 4 years during major overhauls. LP last-stage blades should receive eddy-current scanning every 24 months due to higher moisture-impingement erosion and fatigue risk. Plants running above 7,000 hours/year or with aggressive load-following should compress these intervals by 30%.
What is the most effective NDT method for LP blade cracking?
Eddy-current array testing is the industry standard for LP blade roots and lacing wires because it detects surface-breaking fatigue cracks as shallow as 0.2 mm without removing the blades. For subsurface root cracks, ultrasonic testing with angled beam probes supplements ET. Dye penetrant is used on shrouds and tenons where ET probe access is limited. To build a full NDT workflow across your fleet, Book a Demo and we'll map it to your outage schedule.
When should turbine blades be replaced versus repaired?
Replace when chord loss exceeds 9% on L-0 blades, when any root crack exceeds 2 mm, or when multiple crack indications appear in the same blade row. Blades with 6–8% erosion and stable rates can be monitored with enhanced NDT. Re-tipping is viable for erosion-only damage on otherwise sound blades, typically costing 40–55% of full replacement.
How is blade erosion measured and tracked over time?
Leading-edge chord loss is measured with calibrated calipers or 3D laser scanning, then compared against the OEM's original airfoil profile. The erosion rate (mm per thousand hours) is calculated across inspections and trended in the CMMS. A rate increase of more than 25% between cycles signals a process change — usually moisture separator drift or excessive load cycling — that accelerates replacement timing.
Can blade lifecycle tracking be integrated with a CMMS?
Yes — and it should be. Each blade row's inspection history, erosion measurements, NDT findings, and OEM lifecycle curve should live as a linked asset record inside the CMMS. This enables automatic alerts when operating-hour thresholds approach, generates audit-ready evidence packages for ISO 55000 compliance, and ensures engineer turnover doesn't erase institutional knowledge. Start Free Trial to see how blade lifecycle data centralizes in Oxmaint.
Start Your Blade Reliability Program
Stop guessing about blade health. Start tracking it.
Join power plants using Oxmaint to digitize blade inspection records, trend erosion rates, and defend outage scope with real data.
Free 14-day trial · No credit card
.png)
