Shading is one of the most significant and most underestimated sources of energy yield loss in solar projects. A single obstruction that covers 5% of an array can cause 15–25% production loss if the string architecture is not designed correctly. Accurate shading analysis is not optional — it is a prerequisite for reliable energy modelling, bankable performance guarantees and correct stringing decisions.
Types of shading losses
Solar shading losses fall into two categories:
- Near shading (architectural shading): shading from objects in the immediate vicinity of the array — parapets, rooftop equipment, chimneys, adjacent buildings, trees. These create geometric shadows that move across the array as the sun moves. Near shading losses are typically 2–15% for urban commercial rooftops.
- Far shading (horizon shading): shading from distant obstructions — mountains, forests, high-rise buildings at the horizon. Far shading affects only low solar angles (early morning, late afternoon) and is typically 0.5–3% for most sites.
Additionally:
- Self-shading (row-to-row shading): shadow cast by one row of panels on the next row behind it. This is the primary shading loss for ground-mount and low-tilt flat-roof systems and is controlled by row spacing (pitch-to-height ratio). See our article on row layout shading losses for detailed methodology.
- Module mismatch from partial shading: when a shadow covers part of a string, the shaded cells become a current bottleneck; without bypass diodes operating correctly, the entire string output drops. MLPE (module-level power electronics) significantly reduces this effect.
Shading analysis methodology
A rigorous near-shading analysis process:
- 3D model the site: import building geometry, obstructions and proposed array into the simulation software; verify model against site measurements or LiDAR data
- Define shading objects: parapet heights, HVAC units, skylights, chimneys — include any object above the array plane within ~100 m
- Run irradiance map: calculate annual irradiance (kWh/m²) at each point on the array plane, accounting for direct, diffuse and reflected irradiance blocked by shading objects
- Apply electrical shading losses: account for string-level voltage reduction from partial shading; this requires string layout definition in the model
- Compare shading loss with and without MLPE: quantify the benefit of adding optimisers or microinverters to shade-affected strings
PVsyst vs HelioScope vs Aurora Solar
| Feature | PVsyst 7.x | HelioScope | Aurora Solar |
|---|---|---|---|
| Primary use case | Engineering/bankable studies | Sales + engineering | Sales + design automation |
| 3D shading model | Detailed (parameterised objects) | Google Maps/LiDAR import | LiDAR + Google Maps + aerial imagery |
| MLPE modelling | Via module mismatch factor | String + MLPE mode | Module-level simulation |
| Weather data | Meteonorm, NASA SSE, custom TMY | NREL, Solargis, custom | NREL, Solargis |
| Bankability | Industry standard for lenders | Accepted by most lenders | Growing acceptance |
| Learning curve | High (3–6 months to proficiency) | Low–medium (days) | Very low (hours) |
| Price (approx.) | ~$1,500/year | ~$2,000–5,000/year | ~$3,000–8,000/year |
| Best for | Utility scale / engineering firms | C&I commercial sales | Residential + C&I sales teams |
PVsyst: industry standard for bankable studies
PVsyst remains the reference tool for bankable energy assessments. Lenders and independent engineers accept PVsyst reports because the tool's irradiance models, loss factor framework and simulation engine are well-documented and auditable. Key strengths:
- Detailed loss cascade (horizon, near shading, IAM, soiling, mismatch, wiring, inverter) with each loss quantified separately
- Monte Carlo uncertainty analysis for P50/P90 production estimates
- Extensive inverter and module database
- Can model bifacial rear-side gain with albedo inputs
Limitations: steep learning curve; 3D obstacle modelling is parameterised (manually defined shapes) rather than LiDAR-based; less suited to rapid sales proposals.
Shading loss thresholds and design decisions
Rule of thumb thresholds for design decisions based on shading analysis:
- <2% near shading loss: no special action required; standard string inverter acceptable
- 2–5% near shading loss: consider MLPE (optimisers) for affected strings; review string orientation
- 5–15% near shading loss: MLPE strongly recommended; redesign string layout to keep shaded and unshaded modules on separate strings
- >15% near shading loss: reconsider array layout; some zones may be uneconomic to populate