A PVsyst energy yield report is one of the most consequential technical documents in a solar project's life. It underpins the PPA tariff negotiation, the debt sizing in the financial model, and the lender's confidence in the project's ability to service its debt over a 25-year loan tenor. Yet the majority of yield reports produced in the Indian market treat PVsyst as a software tool rather than an engineering discipline.

This article explains what a bankable PVsyst report looks like — and what lenders actually scrutinise when their independent engineer reviews one.

The difference between a report and a model

A PVsyst output is only as reliable as its inputs. The software will produce a plausible-looking annual energy yield figure regardless of whether the meteorological data, system configuration, and loss assumptions are correct. A report that presents the headline P50 yield figure without documenting every input assumption and validating every loss parameter is not a bankable report — it is a number without a defensible basis.

Meteorological data: the most important input

PVsyst accepts several sources of meteorological data, ranging from satellite-derived datasets (Meteonorm, SolarGIS, NASA SSE) to ground-measured data from on-site stations. The choice of dataset has a direct impact on the yield estimate — differences of 5–10% between datasets for the same location are not uncommon in India.

For a rigorous C&I feasibility report, use at minimum two independent meteorological datasets reconciled against each other, with the lower of the two P50 estimates used as the conservative basis for savings modelling. For large open access projects (above 5 MW), on-site irradiance measurement strengthens the model significantly.

Key loss parameters that lenders check

  • Soiling losses: India-specific soiling assumptions must reflect the actual dust environment of the project site. A coastal Karnataka project has materially different soiling characteristics from a Thar Desert project in Rajasthan. Generic soiling factors of 2–3% — commonly used because they are the PVsyst default — are not defensible for Rajasthan projects where soiling losses of 6–8% are more realistic.
  • DC cabling losses: These must be calculated from the actual cable sizing and routing in the project design, not estimated. A yield report submitted before the electrical design is complete is using assumed cable losses that may be materially wrong.
  • Module degradation: The linear degradation rate and first-year degradation factor must be sourced from the module manufacturer's warranty documentation and validated against independent testing data. Optimistic degradation assumptions are one of the most common sources of yield overestimation in Indian projects.
  • Inverter clipping losses: The DC/AC ratio of the plant design must be explicitly modelled, and clipping losses documented. Aggressive DC/AC ratios (above 1.35) without documented clipping loss analysis are a red flag for lenders' engineers.

P50 vs P90: understanding what you are modelling

The P50 estimate is the energy yield that the project is expected to meet or exceed in 50% of years. The P90 estimate is the yield expected to be met or exceeded in 90% of years — the conservative scenario used to size debt service coverage ratios.

The gap between P50 and P90 is determined by the inter-annual variability in solar irradiance at the project location. In India, this gap is typically 5–8% for ground-mounted projects, depending on location. A financial model that uses P50 yield to size debt service coverage without also demonstrating adequate coverage at P90 will not pass lender due diligence.

A PVsyst model that is built with rigour — documented inputs, validated loss parameters, and independent meteorological data — is a technical asset that supports confident capex decisions and faster finance review. One that is produced quickly to meet a bid deadline is a liability that surfaces during technical due diligence.