Basic HTML Version
NORDIC LIGHT & COLOUR
91
Such an eventuality would probably hinder rather than encour-
age general progression towards better metrics. Such eventu-
alities should be avoided, or at least their detrimental effects
mitigated as much as possible.
There are several approaches that can be employed to avoid
these undesirable outcomes. Firstly, all architecture students
should, I believe, be taught climate-based daylight modelling –
at least in outline. Hands-on use, where appropriate, should be
encouraged.
The Diva4Rhino package (development led by Christoph
Reinhart) was designed for use by architecture students, and
largely hides the complexities of CBDM from the user. With
increasingly popularity Diva4Rhino could become an estab-
lished tool for use by designers. For those not carrying out
hands-on simulation, examples from CBDM could be used to
help understand the daylighting properties of a space. Because
CBDM predicts daylight as it is experienced – sun and sky
acting together – it can reveal performance insights that were
previously accessible only to those very experienced daylight
designers who could intuit the spatio-temporal dynamics of
natural illumination for a design from their, hard-won, practi-
cal knowledge. Designers have often remarked to the author
how much
easier
it is to understand the daylighting perfor-
mance of a space from, say, UDI plots than trying to guess how
a DF relates to
actual
daylight.
Architects often commission consulting engineers to carry
out performance analyses of their designs. Education should
prepare architectural students to become knowledgeable pur-
chasers of specialist services, e.g. daylight modelling. For this
the architect does not necessarily need hands-on simulation
experience, instead a sound understanding of the principles
and what to expect in a competently carried out study should
be sufficient. In any negotiations with a specialist consultant,
the person commissioning the work should demonstrate that
they know what constitutes ‘best practice’, and so are clearly
in a position to distinguish between a well executed perfor-
mance analysis and a poor or mediocre one. The following two
examples give illustrations of climate-based daylight modelling
results from actual studies.
Evaluation of a classroom using UDI
The space used to demonstrate the application of UDI is a
classroom with high-level clerestory windows on the north
and south facades, and view windows looking north, Figure
11. The evaluation was carried out for the period 09h00 and
17h00, giving a total of 2,920 hours throughout the year. Overall
there is a general good provision of daylight on the desks with
illuminances in the UDI-a range (i.e. 300–3000 lux) typically oc-
curring around 2,000 hours of the (occupied) year. Illuminances
in the UDI-s (100–300 lux) range were achieved for about 500
hours, with a similar occurrence for illuminances less than 100
lux which mostly result from the hours of darkness in winter
that are included in the evaluation. Particularly revealing is the
plot for UDI-e showing the occurrence of illuminances greater
than 3000 lux. The very low incidence of these (i.e. typically
about 40 hours or less) suggests that the design is effective in
shading the desks from the sun for much of the year. Thus one
would expect that the occupied space should achieve some-
thing close to its predicted daylighting potential because the
incidence of the high illuminances that are associated with us-
ers deploying blinds is low. Note, the 300 lux daylight autonomy
value for each desk would be approximately equal to the UDI-a
plus UDI-e.
Daylight exposure study for conservation
Lighting simulation has been used to assess the the long-
term exposure of art works to daylight for management and
conservation purposes. The graphic in Figure 12 is a false-
colour image showing the predicted cumulative annual daylight
illumination incident on surfaces around a daylit stairwell. The
approximate location of the painting under study is marked by
the rectangle in the centre of the image, note that the scale is
logarithmic (see accompanying legend). The annual exposure
across the painting was predicted to be
≈
2.8 Mlux-hrs at the
bottom and
≈
4.4 Mlux-hrs at the top. The simulations were
further used to investigate amelioration strategies (e.g. lower-
ing cupola reflectance, reducing rooflight glazing transmission)
to reduce the annual exposure to preferred limits. The cumu-
lative annual illumination can be calculated directly using a
cumulative annual luminance sky (including sun component),
or by taking the sum of all the individual illuminance values
at each time-step. Either method should arrive at the same
result, but only the time-series approach will also permit an
evaluation of the occurrence of instantaneous illuminance
values for the purpose of determining the levels of daylight
for viewing of the art works, e.g. illuminances in the range X
to Y lux occurred for N hours of the year – preferred ranges
often depend on the type and sometimes age of work, e.g. oil,
watercolour.