A solar project can have a strong resource, a signed PPA and a credible EPC price and still carry too much debt. Project finance lenders are repaid from the project's own cash flows, so the model has to convert operating performance into a transparent debt-service capacity.
The central bridge is cash flow available for debt service (CFADS). A simplified sequence is revenue, less operating costs, taxes and the project-level cash items that sit ahead of debt service. The exact definition must match the financing documents, but the principle is stable: CFADS is the cash pool from which scheduled principal and interest are paid.
1. Start with revenue that can actually be collected
For a contracted solar IPP, the revenue schedule should connect generation to the PPA rather than treating annual revenue as a hard-coded line. At minimum, model net export, tariff, escalation or indexation, curtailment treatment, degradation and any applicable losses. If the project has multiple buyers or merchant exposure, separate the revenue streams so each can be stressed independently.
That matters because the debt schedule will often look safe if revenue is overstated by only a few percentage points. A lender-style model therefore makes the chain from energy yield to cash collection auditable.
2. Build CFADS consistently
A useful model shows the path from EBITDA-like operating cash flow to CFADS rather than burying it inside a long formula. The model should make clear which taxes, working-capital movements, reserve funding, maintenance items or other project cash flows sit before scheduled debt service.
Once CFADS is clear, the basic coverage calculation is straightforward:
DSCR = CFADS / scheduled debt service
A DSCR above 1.00x means the project produced more cash than the scheduled principal and interest due in that period. But a financing decision is not based on the mathematical minimum. Lenders set covenant and sizing requirements according to technology, contract quality, offtaker risk, resource uncertainty and the wider financing structure.
3. Use debt sizing as an output, not an assumption
A common modelling weakness is to assume a debt percentage first and only then check whether the project can service it. A stronger approach tests the maximum debt supported by the agreed sizing constraints. The resulting debt quantum can then be compared with other constraints such as a maximum gearing ratio, lender appetite or construction funding requirements.
Debt capacity is therefore a function of several linked variables: CFADS, required coverage, interest rate, tenor, repayment shape and reserve requirements. A project with lower revenue volatility may support more debt even if its headline project IRR is similar to another project.
4. Repayment profile changes the answer
Equal principal, annuity-style debt service and sculpted repayment profiles produce different coverage patterns. Sculpting aligns scheduled debt service more closely with forecast CFADS, subject to the lender's target coverage and other constraints. That can increase debt capacity, but it should not be used to hide weak economics or unrealistic back-ended cash flow.
The model should show annual debt service, interest, principal, closing balance and coverage ratios together. If the repayment schedule is difficult to trace, the financing case is difficult to trust.
5. DSCR is not enough on its own
Period-by-period DSCR answers whether the project can meet scheduled debt service in each forecast period. Longer-horizon ratios such as LLCR and PLCR answer different questions by comparing the present value of future project cash flows with outstanding debt. They are useful cross-checks, especially when cash flows change materially over the tenor.
For investment decisions, the debt case also needs to connect to equity cash flows. More leverage can raise equity returns in the base case while reducing resilience in downside cases. The model should make that trade-off visible rather than treating leverage as automatically beneficial.
6. Stress the variables that can break debt service
A base case is only the starting point. Solar debt capacity is usually sensitive to generation, tariff, COD delay, CAPEX, operating costs, interest rates and contract-specific revenue deductions. Combined cases matter because real projects rarely experience one isolated downside at a time.
I prefer to read the sensitivity table alongside the debt schedule: which year becomes the minimum DSCR, why that year is tight, and whether the issue is temporary timing or a structural weakness in the project economics.
What a lender-style model should let you answer quickly
- Where does revenue come from and which assumptions drive it?
- What is the exact bridge from operating cash flow to CFADS?
- What limits the debt quantum?
- Which year produces the minimum DSCR and why?
- How do tenor and repayment shape affect coverage?
- How much headroom remains under plausible combined downside cases?
- What happens to equity returns when leverage changes?
The point is not to maximise the number of ratios in the workbook. It is to make the financing logic easy to audit and hard to misunderstand.