DAYLIGHTING

CHAN Lok Shun Apple (in Chinese: 陳樂舜)

Division of Building Science & Technology, Faculty of Science & Engineering, City
University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong SAR, China.
Tel: 852-27889676
Fax: 852-27889716
Email: [email protected]

1. Solar Radition
2. Sky Conditions for Daylighting
3. Luminance Distribution in Overcast Sky
4. Daylight Calculation
5. Daylight Factor
6. Methods of Daylight Factor Estimation
7. Daylight Factor Contours
8. Models of Artificial Sky
9. Design Consideration
1. Solar radiation
The solar radiation incident on a building is made up of the
direct, diffuse and reflected radiation.
Figure 1 Solar Spectrum
2. Sky Conditions for Daylighting
For daylighting purposes, sky conditions are classified as
either overcast, clear, or partly cloudy. Each classification
has characteristics that influence daylighting design.

(a) Overcast sky (See Figure 2)

Figure 2 Overcast Sky

(b) Clear sky (See Figure 3)

 Figure 3 Clear Sky

(c) Partly cloudy sky (See Figure 4)

Figure 4 Partly Cloudy Sky

3. Luminance distribution in overcast sky
In earlier studies, the luminance of the entire overcast sky
was taken as uniform. This type of sky is termed as uniform
sky and under this condition the illuminance at a point on a
horizontal surface due to an unobstructed hemisphere of
overcast sky can be given by:

(1)
where,
L = sky luminance, Cd/m2
In recent years, the standardised CIE sky was widely
adopted, which represents the diffuse light from a
completely overcast sky when the In recent represents
ground is free of snow. The luminance distribution of the
CIE sky is not uniform. The relative luminance Lq depends
on the angle of elevation q measured with respect to the
horizon, and is given by : (See Figure 5)

(2)
where,
Lz = the luminance of the sky at the zenith
This equation shows that :
(a) Zenith (q = 90o) is the brightest region
(b) The luminance decreases to 1/3 of that of the zenith
towards the horizon
(c) The luminance is independent of the position of the sun,
and therefore the orientation of the windows has no effect
on the illumination of the room.

Figure 5 The CIE Overcast Sky

4. Daylight calculation
Daylight can be handled quantitatively in two ways:
(a) by using luminous quantities (flux, illuminance), i.e., by
a set of outdoor conditions and calculating the resulting
interior illumiances;
(b) by using relative values (the daylight factor) which
compare indoor to outdoor illuminance. For a given
position, this factor is constant under widely varying
outdoor lighting conditions.

5. Daylight factor
The daylight factor is defined as :

(3)
where
Ei = illumiance due to daylight at a point on the indoors
working plane
Eo = simultaneous outdoor illuminance on a horizontal
plane from an unobstructed hemisphere of overcast sky
There are three possible paths along which light can reach a
point inside a room through glazed windows (See Figure
6). They are :
(a) light from the patch of sky visible at the point
considered, expressed as the sky component (SC),
(b) light reflected from opposing exterior surfaces and then
reached the point, expressed as the externally reflected
component (ERC),
(c) light entering through the window but reaching the
point only after reflection from internal surfaces, expressed
as the internally reflected component (IRC).
The sum of the three components gives the daylight factor:

(4)

Figure 6 Components of the Daylight Factor

6. Methods of Daylight Factor (DF) Estimation
The above three components can be determined by various
approaches. Both SC and ERC can be found from the
geometry of the visible sky or external reflected surfaces,
where as based on inter-reflection theory, IRC can be found
from formula, nomogram, or tables. Some common
methods adopted in UK are:
(a) SC and ERC: Waldram diagram; ERE protractors; BRE
tables; and Pilkington 'pepper pot' diagrams.
(b) IRC: Formula; BRE tables; and ERE nomograms.
The presentation and explanation of all these methods are
out of our scope. Hence selected methods are discussed as
follows:
6.1 The sky component (SC)
The sky component is usually the greatest in magnitude
among the three. Its magnitude depends on the area of sky
visible from the point considered, and in the case of CIE
sky, also on the position of this area at the sky dome.
(A) BRE protractors
The Building Research Establishment BRE (formerly the
Building Research Station BRS) developed a set of
protractors which give direct reading of the sky component
in percentages. There are ten nos. of such protractors, of
which five are for the uniform sky and five for the CIE sky:
 Uniform sky CIE sky

(Protractor No.) (Protractor No.)

Vertical windows 1 2

Horizontal glazing 3 4

30oC glazing 5 6

60oC glazing 7 8

Unglazed apertures 9 10

The protractors No. 2 and No. 4 are given in Figures 7 & 8 respectively.

The procedures to use BRE Protractor are as follows:

Determine the initial sky component (infinite length of window)

(a) Refer to the elevation view of a room in Figure 9.

(b) Mark the work plane and the point considered (A).

(c) Connect this point to the window sill (AS) and the window head (AH).

(d) Make a tracing of protractor (Figure 7).

(e) Place the tracing with its centre on point (A) and its base along the working plane,
with scale 1 upright.

(f) Read the values where lines AS & AH intersect the protractor scale.

(g) The difference between the two values gives the initial sky component.
Figure 7 BRS Sky Component Protractor for Vertical Glazing

(CIE Overcast Sky)

 Figure 8 BRS Sky Component Protractor for Horizontal Glazing

(CIE Overcast Sky)

Figure 9 Determining Initial Sky Component

Find the average altitude

(a) Read the angles of elevation of lines AH and AS, on the inner degree-scale.

(b) Add the two angles and divide the sum by 2.

Determine the correction factor (defined length of window)

(a) Refer to the plan view of the same room in Figure 10.

(b) Mark on the plan the point considered (A).

(c) Connect this point to the two sides of the window opening (AL & AR).

(d) Place the protractor with its centre on point (A), with the base parallel to the window
and scale 2 towards the window.
(e) Interpolate an imaginary semicircle between those designated 0o, 30o, 60o and 90o,
corresponding to the average altitude previously obtained.

(f) Read the values along the curves on the inner scale. Interpolate if necessary.

(g) Find the sum of the two readings, if they were taken on either side of the centre line.

(h) Find the difference of the two readings, if they were both taken on the same side of
the centre line.

The sky component is the product of the initial sky component and the correction factor

(B) Tables
Table 1 is used for determining the sky components for windows with clear, vertical,
rectangular glazing in conjunction with a CIE standard overcast sky. Other tables are
available for other forms of glazing, e.g. roof lights.

Figure 10 Determining Correction Factor

 Table 1 Sky Component (CIE) Standard Overcast Sky) for Vertical Glazed
Rectangular Windows

The procedure for a measurement position on the line of window and at sill level (Figure
11) is as follows:

(a) Establish H, the window head height above the work plane level;

(b) Establish D, the distance from the window to the point considered,

(c) Express D as a multiple of H (i.e., the H/D ratio) and locate this at the head of the
table;
(d) Establish W1 and W2, i.e., the width of window to either side of the perpendicular
(W1 + W2 = total width);

(e) Express both widths as a multiple of D (i.e., ratios W1/D and W2/D) and locate these
in the first column of the table;

(f) Read the two values in the table. The sky component for the point considered is the
sum of the two values.

Figure 11 Sky Component for a Measurement Position on the Line of the Window
and at sill Level

For a position off the line of the window, the sky component can be obtained by
evaluating sky components for a series of windows and obtaining the required sky
component by a process of subtraction and addition. (Figure 12)

6.2 The externally reflected component (ERC)

(A) Charts
If there is an obstruction opposite the window, then the lower limit for the sky component
will be a line drawn from point (A) to the top of this obstruction with reference to Figures
9 and 10. The sector between this line and the sill (AS) line will give the ERC. Follow the
same procedure as for the sky component. Multiply the result by the reflectance of the
obstruction (if known) or by 0.2 if the CIE-sky protractors were used. Using the uniform
sky protractors, take half of the reflectance value or 0.1 as the multiplying factor.

Figure 12 Sky Component for a Measurement Point that is off from the Window
and below Sill Level

(B) Tables
ERC can also be done using Table 2. The procedure is to treat the external obstruction
visible from the reference point as a patch of sky whose luminance is some fraction of the
unobstructed sky luminance. In other words, the SC for the obstructed area is first
calculated as described above and is then converted to the ERC by multiplying the ratio
of the luminance of the obstructed area to the sky luminance.
 Table 2 The Minimum IRC of DF (percent)

6.3 The internally reflected component (IRC)

(A) Formula
The average IRC can be determined quite precisely from the BRS inter-reflection
formula. The simplified form of this is:

(5)

where,

0.85 = transmittance of window glazing assumed,

W = window area (m2),

A = total area (ceiling + floor + walls including windows),

r = average reflectance of area A,

r fw = average reflectance of floor and the three walls below the plane at the level mid-
height of the window (excluding the window wall),

r cw = average reflectance of ceiling and the upper (remaining) part of the above three
walls,

C = a coefficient depending on external obstructions, as given below:

Angle of 0o 10o 20o 30o 40o 50o 60o 70o 80o

obstructions

Coefficient C 39 35 31 25 20 14 10 7 5

The average IRC can be converted to minimum IRC by the following factors:

Average ARC (%) 30 40 50 60

Conversion factor 0.54 0.67 0.78 0.85

Table 2 was formulated by the BRE to give minimum internally reflected components of
daylight factor, assuming a CIE standard overcast sky and a known scheme of decoration.
The table was primarily for rooms of the following conditions:

(a) Dimensions : 6 m length x 6 m width x 3 m ceiling height. Floor area = 36 m2

(b) A window (glazed with ordinary glass) on one side extending from a 1 m sill to the
ceiling.

Minimum IRC for rooms of 10-100 m2 floor area, with ceiling heights ranging from 2.5
to 4 m can be obtained by multiplying the following conversion factors.

Floor Wall reflection factor

Area 20% 40% 60% 80%

10 m2 0.6 0.7 0.8 0.9

 100 m2 1.4 1.2 1.0 0.9

6.4 Additional correction factors

The SC, ERC and IRC will be reduced by the following factors:

(a) Maintenance factor, M, allowing for dirt and other causes of deterioration of glazing
(see Table 3);

(b) Glass factor, G, allowing for the type of glazing if other than clear glass is used (see
Table 4);

(c) Bars, or Framing Factor, B, which may reduce the effective area of the window (see
Table 5).

Location of Inclination of Maintenance Factor

Building Glazing Non-industrial or Dirty Industrial

Clean Industrial Work
Work

Non-industrial or Vertical 0.9 0.8

Clean Industrial Sloping 0.8 0.7

Area Horizontal 0.7 0.6

Dirty Industrial Vertical 0.8 0.7

Area Sloping 0.7 0.6

Horizontal 0.6 0.5

Table 3 Maintenance Factors to be Applied to the DF or to Each of its Three

Components to Allow for Dirt on the Glazing

The IRC for dirt will also be reduced by the maintenance factor, D, allowing on room
surfaces (See Table 6)
7. Daylight factor contours
If the overall distribution of daylight in a room is to be
predicted, the best method is to set out a grid and calculate
the SC and ERC for each of the grid-points. The average
IRC can then be found, corrected by factor D, and added to
each value, after which the factors M, G and B can be
applied. By interpolating between grid-point values, a set of
daylight factor contours can be drawn, indicating the
distribution of daylight.

Figure 15 The Effect of Window Shape on DF

8. Models of artificial sky
Model studies can be used to predict daylight penetration into any building.
This is the only reliable prediction method in the case of unusual situations,
complex geometries or heavily obstructed windows. Model studies can be
carried out under outdoor conditions, but the limitations imposed by weather
variations can be quite severe.

The artificial sky simulates the standard overcast sky conditions, giving either
uniform luminance or the CIE luminance distribution.

There are two basic forms of artificial sky: the hemispherical and the
rectangular (Figure 13). The former may be a diffusely reflective opaque dome
surface, illuminated from below, or a translucent dome (hemispherical), with
lighting mounted behind. The rectangular sky has a luminous ceiling and four
strictly vertical walls lined with mirrors. The multiple reflections between
accurately parallel opposing mirrors give an infinite horizon effect. The mirror
glass absorption through multiple reflections ensures a luminance distribution
similar to that of the CIE sky.

Figure 13 Artificial Sky

9. Design Consideration
9.1 Environmental factors

Daylight design is closely related to a number of environmental factors because

the use of glass windows to let in light also allows the penetration of solar heat
and noise and increases the rate of heat gain or loss of the building fabric.
Therefore daylight design can never be considered alone. Fortunately by the
proper choice of window orientation and suitable sun shading device or
constructions, the solar heat gain can be greatly reduced. The use of double
glazed window units is able to cope with the noise problem and reduce
significantly the heat gain or loss through the window. The remaining problem
would be the proper design of windows or rooflights to provide a visually
comfortable and efficient environment for the type of tasks and activities to be
performed.

9.2 Fenestration

The lighting conditions in a room depend primarily on the fenestration. As

Figure 14 shows, the DF and consequently the illumination near the window
decreases as the height of the window above the floor is increased. However,
the illumination increases away from the window, giving greater uniformity to
the lighting .

Figure 15 shows how the DF varies across a room using two types of windows,
both having the some total area:

(a) is for three tall windows, and

(b) is for one long, high-level window.

Window (a) will give a good open view, but (b) will provide more even
daylight illumination.

The following points should be considered in the design of fenestration:

(a) The window head should be as high as possible, say at least 2 m above floor
level, to enable one can see out when standing.
(b) The window sill should not be higher than 1 m from floor level to enable
one can see out when sitting.

(c) The window surface area should be evenly distributed over the outside wall
and the window heights and widths should not be too small in relating to the
window wall because this reduces the uniformity of lighting and produces
undesirable shadows.

Figure 14 The Effect of Window Height on DF

 Figure 15 The Effect of Window Shape on DF

(d) Better lighting can be achieved with windows on opposite walls since the
illumination produced by the individual windows are superimposed. The region
with the minimum daylight factor (see Figure 16) is then displaced towards the
centre of the room. The usable depth of the room thus increase.

For rooflights, as shown in Figure 17, the daylighting is evenly spread over the
working area and much higher DF can be obtained. Yet DF are often drastically
reduced by overhead obstructions and poor window maintenance.
 Figure 16 The Effect of Windows on Two Sides of a Room on DF

Figure 17 The Distribution of Daylight Across the Floor for a Roof-Lit Interior

9.3 Obstructions

Any obstruction (buildings opposite the window, trees, etc.) decreases the
illuminance in rooms. Obstructions which subtend elevation angles of more
than 25-30o significantly reduce the daylight penetration. This can be improved
by increasing the height and width of the window. If possible, the angle
subtended by obstructions should not be greater than 30o. The minimum
distances between adjacent building and the window as laid down in the
building regulations should be complied with.

9.4 Room Dimensions

With room heights of 2.5-3 m which are usual in residential building, room
depths, from the lighting point of view, should ideally be limited to 4-6 m,
depending on the width of the room and the nature of obstructions if any. To
achieve good illumination, greater room depths are only possible with windows
on two sides, with rooflights or with the installation of supplementary artificial
lighting.

9.5 Uniformity of lighting

Uniformity is essentially dependent on:

(a) the dimensions of the room, the windows and all the obstructions to light,

(b) the reflectances of the surfaces enclosing the room, the obstructions to light
and the room furnishings, as well as the type of glazing.

The uniformity of illumination is expressed in terms of the ratio of the

minimum illumunance in the room to the average illuminance measured on a
horizontal reference plane. Uniform distribution of the luminance in the room
should be the objective.

9.6 Glare

One will experience glare if the difference in the luminances of adjacent

surfaces become excessive. This refers to impairment of the visual efficiency of
the eye by excessive relative and absolute luminances. In addition to direct
glare from the sun, glare may also arise in interior rooms through excessive
contrast between the high luminance of the patch of sky visible through the
windows and the substantially lower luminances of the adjacent surfaces
enclosing the room. The degree of glare thus depends on :

(a) the luminance and the size of the light-emitting surface seen by the eye,
(b) the ratio of this luminance to the luminance of this environment or
background,

(c) the distance of the glare-producing surface from the eye and its position in
the field of vision.

All types of glare, both direct glare from the sun and glare from reflections, e.g.
from polished floors, are to be avoided.

9.7 Shadow intensity and direction of incident light

For the perception of the solidity and surface texture of objects a reasonable
shadow intensity is required. Adequate shadow intensity is generally provided
by the lateral incidence of daylight into rooms with side windows. Working
positions should be arranged so that hand and body shadows do not fall on the
working surface.

9.8 Permanent supplementary artificial lighting of the interior (PSALI )

By considering windows solely as admitters of daylight it may seem that the

achievement of a certain minimum DF is the main design criterion. Electric
lighting is today often used during daylight hours and it may be the case that in
trying to achieve a minimum DF, say 2 per cent, the window becomes too large
causing more serious glare and heating /cooling problems. It may be preferable
to design to a lower DF and integrate the daylight with the electric light.

The method known as 'permanent supplementary artificial lighting of the

interior' (PSALI) (See Figure 18) is based on three principles:

(a) Utilisation of daylight as far as practicable,

(b) Use of electric lighting to supplement the daylight in the interior parts of the
room,

(c) Installation of the electric lighting in such a way that the daylight character
of the room is retained.
 Figure 18 PSALI Scheme

The required quantity of artificial light input is to achieve an illumance at the

rear of the room comparable to, but slightly less than, the daylight illuminance
near the window. For example:

Daylight factor Artificial light (lux)

0.5% 250

1.0% 500

1.5% 750

2.0% 1,000