Shading is the most underestimated loss in a solar plant. A shadow falling on a single cell affects the whole series string, so the result is not proportional to the shaded area. SolarSimPro computes near shading by ray tracing: it takes the sun's position hour by hour through the year, projects every obstacle's shadow onto the module plane, and turns the loss into a number.
Shadows cast by chimneys, parapets, HVAC units, lift overruns, neighbouring buildings, trees and by the module rows on one another. Obstacles are modelled in three dimensions; their height and position determine the shadow.
On land, table rows shading one another around midday in winter. You see the trade-off between row spacing and yield directly.
Shading is computed hour by hour rather than as a single annual percentage, so morning and evening losses are visible separately.
Which modules are affected and by how much is shown on the layout plan, so the decision to move a problem area or remove those modules rests on data.
The classic approach on land is to choose the row spacing that stays shade-free at noon on 21 December. That brings shading close to zero, but fits fewer modules on the same site. The real question is this: which is worth more — the extra square metres or the extra kilowatt-hours?
SolarSimPro puts a number on that decision. Change the row spacing and the module count, the shading loss and the annual yield all update together; at the feasibility step the change feeds straight through to the payback period. So "how many metres should we leave?" is answered by your own project's figures rather than by habit.
On rooftops the loss usually comes from the roof itself. Parapet height, chimneys and ventilation stacks, outdoor HVAC units, lift overruns and neighbouring buildings all enter the model in three dimensions. When the layout is generated, the clearance required around these obstacles is respected; the same obstacles are then used as shadow-casting objects in the shading analysis.
The result is not an assumption such as "85% of the roof is usable", but a plan that can actually be installed and the measured shading loss of that plan.
Shading analysis is not a separate program — it is one step inside the design workflow.
Shading loss is the weighted sum of two components. The direct beam component is computed by ray tracing: a ray is cast from the sun's current position to the module surface, and every table, obstacle and ground undulation along that path is tested, table by table. A sky model is used for the diffuse component — how much of the open sky is blocked enters the calculation. Separating the two matters, because on overcast days most of the yield comes from diffuse irradiance, and a calculation that looks only at beam shading reads those days wrongly.
A fixed "shading factor" read from a look-up table cannot see the site's own topography or the real geometry of neighbouring rows. On sloping ground the shadow length changes with the direction of the slope even when row spacing stays the same; on a north-facing slope the same spacing is not enough. Ray tracing measures that difference directly.
Near shading and far shading are different things. Near shading comes from the site's own features: neighbouring rows, tables, stacks and parapets on a roof, and any trees or buildings that have been defined. Far shading is the horizon — the ridge line around the site cuts the first hours of the morning and the last of the evening, and its effect becomes pronounced in winter. The two are handled by different methods; substituting one for the other is misleading, particularly for sites in a valley or at the foot of a mountain.
The report states modelled and unmodelled effects separately — items such as an undrawn neighbouring building or snow accumulation are listed explicitly as "not included". The reliability of a shading analysis is measured by whether it also tells you what it did not calculate.
The most common mistake in shading work is to treat the shaded area fraction as the loss. That is the linear loss, and it is only part of the real figure.
Electrical loss comes from how a string works: modules in series carry the same current, so shading a single module pulls down the current of the whole string. Bypass diodes limit the effect but do not remove it. SolarSimPro models the string and bypass-diode structure to calculate this additional loss separately, and reports it per string.
The difference is not small: on densely pitched projects the gap between the two calculations is measured in percentage points and feeds straight into annual yield. A study that looks only at shaded area makes the plant look better than it is — and once that is discovered after contract, it is expensive to correct.
Shading is not a single number; it varies with the hour of the day and the day of the year. The report is therefore read at three scales:
The chart also marks the two extremes of the year. 21 December is when the sun is lowest and shadows are longest; row spacing is usually argued over this day. 21 June is the most favourable. Every other day falls between those two curves, so reading the extremes together gives a sense of the whole year.
The analysis measures the loss; the design makes the decision. The levers that work in practice:
Near shading is computed by ray tracing. The sun's hourly position through the year is used; each obstacle's shadow is projected onto the module plane and the affected area is determined per module.
Yes. On ground-mounted plants the shading that table rows cast on one another is computed as a function of row spacing, and updates together with the yield when you change the spacing.
Yes. Surrounding obstacles can be defined with their heights and are included in the shading calculation.
It does. The shading-adjusted yield is used directly in the simulation and feasibility steps; no manual transfer is needed.
Both are calculated. The shaded area fraction is the linear loss; the electrical loss — caused by the current of a whole string falling when one module is shaded — is found separately by modelling the string and bypass-diode structure. On densely pitched projects the gap between the two is measured in percentage points.
Yes. The direct beam component is computed by ray tracing and the diffuse component with a sky model; the annual loss is the weighted sum of the two. Because most of the yield on overcast days comes from diffuse irradiance, this distinction changes the result noticeably.
Usually not. Opening up the pitch lowers the loss but also lowers the capacity that fits on the same site. The decision is made where net yield per hectare is highest; on some projects a few points of loss is cheaper than extra land rent or a longer DC run. The analysis puts a number on that trade-off.
It does. Shadow length varies with the direction and steepness of the slope; on a north-facing slope the same row spacing is not enough. A fixed shading factor from a look-up table cannot see that difference — ray tracing measures it directly.
The report says so explicitly. Undrawn neighbouring structures, snow accumulation and soiling do not enter the shading engine; they appear as separate loss headings or as stated assumptions. Recording what is not modelled matters as much as what is.
Let us run the analysis together on your own roof or site.