Courtyard daylight: the width to height ratio that works

In a perimeter block the courtyard daylight ratio does most of the work. Once the width of the well and the height of the surrounding facades are fixed, no amount of glazing, balcony redesign or pale render will rescue the ground-floor units looking into it. A courtyard narrower than about twice the height of the enclosing elevations will produce lower-floor rooms that fall short of the daylight recommendations in BS EN 17037 and its UK National Annex, and the only reliable fixes are geometric.

That is worth saying early because courtyard proportion is usually settled in the first week of massing and revisited in month nine, when the only remaining lever is the unit mix.

Why courtyard proportion decides the result

Daylight in an enclosed well comes from two places. Direct skylight through the patch of sky visible from the window, and light reflected off the facades opposite. The first is controlled entirely by geometry. The second is a modest addition that depends on the first, because a surface cannot reflect light it never receives.

The BRE site layout guidance frames this as an obstruction angle. Where the angle from the centre of a window to the top of the obstruction in front of it is no more than 25 degrees above the horizontal, enough skylight normally reaches the window for conventional design. Convert that to a plan dimension and a 25-degree angle corresponds to a horizontal distance of roughly 2.1 times the height of the obstruction measured above the window.

In a courtyard the obstruction is the facade opposite, and the window being tested is usually at ground or first floor. So the height that matters is not the building height but the height of the opposing facade above that particular window. A ground-floor window in a six-storey block faces almost the whole elevation. A fourth-floor window in the same block faces two storeys and performs well.

Ratios that tend to work

The table below sets out the plan width a courtyard needs if the lowest residential window is to sit at or inside a 25-degree obstruction angle, assuming a floor-to-floor of 3m and the window centre around 1.5m above its own floor level.

Those figures are illustrative arithmetic from the 25-degree benchmark rather than a policy standard, and they explain why large courtyards appear on tall perimeter blocks and why a tight well with six storeys around it will not perform. Treat them as the sanity check you run before the first massing review.

Two qualifications. A courtyard that misses 25 degrees is not automatically unacceptable, because BS EN 17037 is assessed on illuminance over the working plane rather than on an angle, and a generous window head height, a shallow room depth or a corner position can recover a room that the angle test flags. And the angle test says nothing about sunlight, which depends on orientation rather than on symmetry of proportion.

Corner units and the rooms that fail first

The pattern of failure in a courtyard is consistent. Single-aspect units on the lowest two floors with their only window facing into the well do worst. Next come corner units where the window sits close to the internal angle of the courtyard, because such a window sees sky in a narrow wedge between two tall returns rather than across an open frontage.

Deep plans compound it. A room 6m deep lit by one window into a courtyard will have a no-sky line close to the window, so the back half of the room gets almost nothing. Where the brief pushes room depth, the courtyard has to get wider or the lowest floors have to stop being flats.

Three moves that change the numbers:

1. Give the lowest one or two storeys to non-residential use, cycle stores, plant or dual-aspect maisonettes that also take light from the street frontage.
2. Step the upper storeys back from the courtyard edge, which raises the effective obstruction angle for every window below without losing the floor area at the top.
3. Chamfer or open the internal corners so no window is trapped in a right-angled return.

Testing individual rooms rather than whole floors is what the method in internal daylight assessment for proposed dwellings is for, and the room-by-room output is what an officer will read.

Reflectance inside the courtyard

Inter-reflected light does help in an enclosed well, and it is routinely overstated. The standard assumption for an external obstruction in daylight modelling is a reflectance of around 0.20, which is a reasonable average for brick, dark render and glazing with reveals and shadows. Specifying a pale finish at 0.40 or above can lift the figures for the lower floors, because those are the windows whose view is dominated by the facade opposite rather than by sky.

The improvement is real and small. It is worth perhaps a modest percentage uplift in illuminance on the worst rooms, not the difference between failing and passing a courtyard that is proportionally wrong. It also has to survive the value engineering and the next twenty years of weathering, which is why authorities sometimes condition the finish. The assumptions behind this are set out in reflectance assumptions and daylight results.

Do not model a courtyard with light surfaces and then detail it in dark brick. That mismatch between the submitted assessment and the approved drawings is the thing a neighbouring surveyor looks for.

Gaps, undercrofts and openings that admit light

A fully enclosed courtyard is the worst case. Breaking the perimeter changes the geometry for a surprising number of windows.

An undercroft or pend through one side admits light at ground level along the line of the opening and gives the courtyard a route for air as well. A full-height slot between two blocks lets sky into the corner that would otherwise be the worst in the scheme. Dropping one side of the perimeter to two or three storeys, typically the southern side where sunlight matters most, raises the obstruction angle for every window facing it.

These moves cost area. The honest trade is that area lost at the perimeter buys lettable, consentable flats on the lower floors, and that trade is far easier to make in a massing study than in a committee report.

Testing the geometry while it is still easy to change

A full assessment on a massing model takes very little time and answers the question that matters: which rooms, on which floors, in which positions, miss the recommendations. Run it at the first credible layout, before the facade is designed and before the unit schedule is sold internally.

What to test at that stage:

• The lowest residential window on each courtyard elevation, taken at the plan position closest to the internal corner.
• A mid-courtyard window on the same floor, for comparison.
• Sun hours on the courtyard surface itself on 21 March, which is what tells you whether the amenity space is usable rather than whether the flats are lit.

The courtyard floor is a separate question from the windows around it. If the space is sold as communal amenity space, the sunlight it receives is a planning matter in its own right and belongs in an overshadowing study as well as in the daylight work.

Reporting the courtyard as its own case

Officers read courtyard results badly when they arrive buried in a scheme-wide pass rate. A block of 180 flats where 94 per cent meet the recommendation looks fine until you notice that every failure sits on two floors of one courtyard elevation.

Report the courtyard separately. Give the proportion as a stated ratio, give the obstruction angle from the worst window, list the rooms that fall below the recommendation with the figures they achieve, and say what was done about them. Where units remain below the recommendation, explain the design reason and the compensating factors, because the National Design Guide and the NPPF both expect a judgement about living conditions rather than an arithmetic pass mark. A daylight and sunlight assessment that admits its own worst case is far more persuasive than one that averages it away.

And put the ratio in the design and access statement as well as the technical appendix. The planning committee will never open the appendix.

This is general guidance, not legal advice. Daylight assessment for a scheme of any size should be carried out by a suitably qualified consultant, and councils usually expect a named author for this work.

To see which daylight, sunlight and amenity documents a scheme is likely to need, enter your postcode on PlanScope and work through the guided questions. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.