When the Turbine Maker Fails: Warranty and Asset Value in Wind
Supporting hypothetical analytical model data

All inputs are author-selected hypothetical assumptions. All financial results are deterministic calculations; no observed project, transaction or market outcome is represented. These files provide the machine-readable data accompanying the research. They are not investment, lending, accounting or legal advice.

Contents
- model-results.json: assumptions, annual cash flows, sensitivity results and calculated values.
- model-tables.json: display tables with their stated rounding.
- README.txt: this explanation.

Units and timing
Cash amounts are real USD, capacity is MW, generation is MWh, availability and recovery fractions are unit fractions, and DSCR is a dimensionless ratio. Sensitivity keys explicitly labelled percentage_points use percentage points below normal availability. The table headings identify USD millions and percentages where displayed. All operating cash flows occur at annual year end. The remaining years, real discount rate, price and debt-service assumption are disclosed in common_inputs.

Calculation perimeter
Energy equals capacity times the resource capacity factor before technical availability times annual hours times technical availability. CFADS equals electricity revenue less baseline operating expense, incremental expense and component/mobilisation spending; it excludes warranty receipts. Asset NPV discounts CFADS plus the actual assumed warranty receipt in its specified year, deducting the inception continuity-package price once. Debt service is excluded from asset NPV and included in DSCR and the liquidity test. Warranty face is never booked in full as a separate asset or impairment charge. Terminal value is zero.
DSCR equals CFADS divided by the illustrative annual debt service. Opening reserve equals the greater of zero and the negative minimum cumulative annual cash after CFADS, actual receipt and debt service, including the zero starting cumulative balance. The reserve earns no interest, all surpluses are retained and distributions are suspended. Inception package price is outside this reserve and is added separately to initial package-plus-reserve cash needs. This annual test does not estimate intrayear liquidity.
The package break-even ceiling equals the present value of the assigned pre-funded operating/recovery pathway before its package price, less the comparator asset NPV. The sensitivity grid varies the stated first-year downtime and recovery fraction while retaining the remaining severe-case assumptions; it assigns no probabilities.

Display precision
Calculations use Python float precision without display rounding. JSON retains native numeric precision. Tables show USD millions to 2 decimals, DSCR to 2 decimals and percentages to 0 or 1 decimal as shown; figures show 1 or 2 decimals. Rounded components may not sum to rounded totals.

Table numbering
The model table numbers map to the complete research as follows: 1 to 8, 2 to 9, 3 to 10, 4 to 11, 5 to 12.

Limitations
This deterministic annual model contains no tax, inflation, refinancing proceeds, debt-balance valuation, reserve-release value, failure probabilities or transaction-specific enforceability assessment. Project application requires verified operating, contract, component, debt and recovery evidence.
