Shading analysis without guesswork

Ask ten EPCs how they handle shading and most will describe a version of the same thing. Someone stands on the roof, looks around, decides it is fine, and the proposal goes out with a generation figure taken from a per kW rule of thumb.

That works until it does not, and when it fails it fails in the most expensive way available: after commissioning, in front of a customer holding your generation estimate.

Indian rooftops are unusually good at this. Parapets tall enough to be legal, water tanks on stilts, staircase head rooms, vent pipes, dish antennas, a neighbour building a second floor next year.

Why a small shadow costs more than it looks like it should

The intuition most people carry is that shading ten percent of an array costs ten percent of the output. It does not, and understanding why changes how you lay panels out.

Modules in a string are wired in series, so they share one current. A module in shade produces less current, and in a simple series string it drags the rest toward its own level. Bypass diodes limit the damage by routing current around the affected sub section, which is why a module typically carries three of them, and why the Central Electricity Authority's 2026 technical standards now write bypass diodes into the requirements for solar plants.

The practical consequence is that shade on a single module in the morning can cost you a third of that module's string, not a tenth of one module. Two panels shaded in the wrong places on two different strings is a real loss on a real bill.

The measurement that replaces the eyeball

You need three things off the site, and none of them are a photograph.

1. The height of every obstruction above the plane the modules will sit in. Not the height above the roof, the height above the module surface, because a 600 mm mounting structure changes the answer.
2. The horizontal distance from each obstruction to the nearest proposed module row.
3. The direction each obstruction sits in relative to the array, because an object due north of a roof in India casts almost nothing onto it, while the same object to the south west will take your evening.

With those, shadow length follows from geometry rather than opinion:

shadow length = obstruction height divided by the tangent of the sun's elevation angle

The elevation angle is the part people skip, so here is the shortcut. At noon on 21 December, the sun's elevation is roughly 90 degrees minus your latitude minus 23.4. In Delhi at 28.6 degrees north that is about 38 degrees. In Chennai at 13 degrees north it is about 53 degrees.

Run the arithmetic for a one metre parapet in Delhi at solar noon in December and you get a shadow about 1.3 metres long. That is the best case of the worst day. At nine in the morning the sun is much lower and that same parapet is throwing several metres.

The nine to three rule, and why December decides it

The convention worth adopting is simple: design so that modules are clear of shade between 9 am and 3 pm on the winter solstice.

Two reasons. December is the worst geometry of the year, so a layout that survives it survives everything else. And those six hours carry the large majority of the day's energy, so shade outside that window costs far less than shade inside it.

This is also the honest answer to the customer who asks about the tree. The question is not whether the tree shades the roof. It is whether it shades the array between 9 and 3 in December, and how much of it.

Inter row spacing is the same calculation

On a flat roof with tilted rows, each row shades the one behind it, and the spacing between rows is not a matter of taste. It is the shadow length of a row at your design hour on your design day.

Tighten the spacing and you fit more kilowatts on the roof, then lose part of them every winter morning. Widen it and you fit fewer. Where that trade lands depends on the tariff, the roof and whether the customer is optimising for peak capacity or annual units, which is a conversation worth having explicitly rather than settling with a default number copied from the last job.

When electronics are the answer, and when they are not

Module level electronics, whether optimisers or micro inverters, genuinely help when shading is unavoidable and irregular. They also cost money and add failure points on a roof.

A reasonable order of preference:

• Move the array. Free, and usually possible on at least part of the roof.
• Restring it. Group the modules that get shaded together on the same string or the same MPPT input, so the loss is contained instead of spread across the system.
• Then consider electronics, for the residual shading you cannot design away.

Reaching for optimisers first is how a proposal gets expensive without getting better.

What to put in front of the customer

The strongest thing you can show is not a number, it is a picture of their own roof with the shadow on it at 9 am and 3 pm in December, next to the same roof at noon in June.

It ends the argument about the tree, it justifies why the array is where it is rather than where they imagined, and it makes the generation figure look like the output of a method instead of a guess. It also separates you immediately from the competing quotation, which will have a rule of thumb per kW and a photograph.

None of this can be done properly from a rule of thumb. It needs the real roof, the real obstruction heights and a model that can be re-run when the customer asks what happens if the water tank moves.

The short version

• Shade does not cost output in proportion to the area it covers, because series strings share a current. Bypass diodes limit the damage, and the CEA's 2026 standards now require them.
• Measure obstruction height above the module plane, horizontal distance and direction. A photograph is not a measurement.
• Shadow length equals obstruction height divided by the tangent of the sun's elevation. At noon on 21 December, elevation is roughly 90 minus latitude minus 23.4.
• Design for no shade between 9 am and 3 pm on the winter solstice. That is the worst geometry and the bulk of the energy.
• Move the array, then restring it, then consider module level electronics. In that order.
• Always state a shading loss figure in the generation estimate.

Sources

• CEA notifies Technical Standards for Construction of Electrical Plants and Electric Lines (Amendment) Regulations, 2026, Power Line
• CEA Notifies Technical Standards for Solar, Wind, Floating Solar and BESS Projects, Energetica India