When evaluating a wind farm for acquisition, two components dominate the risk conversation: gearboxes and generators. They're the most expensive to replace, the most disruptive when they fail, and the hardest to assess without structured, component-level data.
But they fail differently. Understanding how — and what that means for your deal model — is the difference between a reserve that holds and one that doesn't.
Gearboxes: higher frequency, more predictable
Gearbox failures are the most common major component event in wind turbines. Industry data consistently shows gearboxes accounting for 15-20% of total turbine downtime. For doubly-fed induction generator (DFIG) configurations — which describes most of the installed fleet in the 1.5-3.0 MW range — gearbox failure rates run roughly 5-8% per turbine per year after year 10.
That sounds high. It is. But gearbox degradation is also relatively predictable. Oil analysis catches metal particle contamination early. Vibration monitoring detects bearing wear months before failure. A structured DD assessment that includes oil analysis history and vibration trends can identify at-risk gearboxes with reasonable confidence.
What to look for during DD
- ✓ Oil analysis trends. Rising iron, copper, or tin particle counts indicate internal wear. A single bad sample isn't conclusive. Three consecutive samples with increasing contamination is a trajectory.
- ✓ Vibration data. High-speed shaft and intermediate shaft bearing frequencies. Look for amplitude increases over time, not just snapshot values.
- ✓ Replacement history. How many gearboxes in the fleet have already been replaced? On what schedule? Replacement patterns across a fleet tell you more than individual condition scores.
- ✓ OEM and model-specific issues. Some turbine models have known gearbox design defects. Check whether this fleet is affected and whether retrofits have been applied.
Cost profile
Full gearbox replacement: $150,000-$250,000 per turbine, including crane mobilization. That's the headline number. But the real cost includes 2-4 weeks of downtime per turbine, which at typical capacity factors represents $30,000-$60,000 in lost revenue. A fleet with five gearbox replacements in the pipeline isn't a $1M problem. It's a $1.3-$1.5M problem.
Generators: lower frequency, higher consequence
Generator failures happen less often than gearbox failures. Annual failure rates for generators in the same fleet age range run 2-4% per turbine per year. But when a generator fails, the consequences are often more severe.
Generator issues split into two categories: electrical failures (winding insulation breakdown, stator faults) and mechanical failures (bearing collapse, rotor damage). Electrical failures can sometimes be addressed with a rewind ($80,000-$120,000). Mechanical failures typically require full replacement ($200,000-$350,000 for large DFIG generators).
What makes generators harder to assess
Unlike gearboxes, generator degradation is harder to detect incrementally. Oil analysis doesn't apply. Vibration monitoring helps for bearing condition but doesn't catch winding insulation breakdown. The best predictive indicators are:
- ✓ Winding resistance and insulation testing. Megger tests and polarization index measurements reveal insulation degradation before failure.
- ✓ Bearing temperature trends. SCADA data showing rising bearing temps over time, especially after bearing replacement, indicates alignment or loading issues.
- ✓ Operating environment. Coastal sites, high-humidity locations, and sites with frequent grid events put generators under more stress.
Many DD campaigns under-assess generators because the diagnostic data is less accessible than gearbox data. This is a mistake. Generator replacements are more expensive, take longer, and have longer lead times for parts.
How this shapes your deal model
The gearbox vs. generator distinction matters for your CapEx reserve in two ways:
- 1. Gearbox reserves should be higher in volume but more predictable in timing. You'll likely need more gearbox replacements across the fleet, but you can forecast them with reasonable accuracy using oil analysis and vibration trends. Budget accordingly: more events, better timing.
- 2. Generator reserves should account for higher per-event cost and less predictable timing. You'll have fewer generator events, but each one is more expensive and harder to see coming. Your model needs a larger contingency per event.
A fleet of 25 turbines at year 12 might need 3-4 gearbox replacements over your five-year hold period (estimable from condition data) and 1-2 generator interventions (harder to predict, higher cost each). The total CapEx reserve might be similar, but the risk profile is fundamentally different.
Why component-level tracking matters
Fleet-level failure statistics give you a starting point. They don't give you a plan. Knowing that "gearboxes fail at 6% per year" across the industry doesn't tell you which five gearboxes in your 25-turbine fleet are the problem.
That's where component-level DD changes the conversation. When every gearbox and generator has individual condition data — oil analysis, vibration history, inspection scores, replacement records — you can identify the specific units at risk and budget for those specific units.
A structured risk register translates these individual assessments into financial terms: which component, what severity, what estimated cost, and what timeline. Your investment committee doesn't need to understand vibration spectra. They need to know that turbine 7's gearbox is $180,000 away from replacement and turbine 14's generator needs monitoring but isn't critical yet.
The bottom line
Gearboxes fail more often. Generators fail more expensively. Both are assessable during DD if you have the right data at the right level of detail.
The teams that build credible CapEx reserves — reserves that hold up through a five-year hold — are the ones doing component-level DD with consistent scoring, traceable evidence, and a direct link from condition findings to financial projections.
That's not something a spreadsheet can sustain across 25 turbines and 14 component categories. It's something a database-first approach was designed for.
Assess every component. Price every risk.
Ridgeline gives you turbine-level condition data that translates directly into your deal model.
Start a DD Sprint