[Radiance-general] Exterior light levels and daylight factors

[Caroline Prochazka]

Hello,
I am a graduate student in architecture at the
University of Waterloo, Canada. My thesis centres on
the modelling of two Canadian 'sustainable' buildings
to investigate the relationship between daylighting
and energy consumption for heating and cooling. I am
modelling each building in Desktop Radiance in order
to evaluate the performance of both Ecotect and
Energy-10 in daylighting analysis.

I had hoped to use Desktop Radiance to evaluate
daylight factors in key areas of the buildings. While
I am able to model fairly realistic interior
illumination results and analyse these using the False
Colour plot I have encountered several problems:

1. When I opt for a daylight factor plot, the scale
remains at a range of 0 to 1.0 and is unaffected by
any changes I try to make through the analysis
dialogue box.  The result in Radiance is a washed out
image all in the 'maximum' colour (my other software
suggests DF% between 2 and 4%) confirming DF%>1.0
How can I adjust the scale for DF% readings in Desktop
Radiance?

2. I am trying to assess the effect of changing
glazing types. For daylight studies in the late winter
afternoon (based on sun charts and other software, I'm
using 4pm Dec21, where the sun is not yet set, but
very low in the sky), where I have not modelled any
surrounding obstructions such as trees or buildings,
my internal illuminance levels are consistently about
200 lux, even on the east side of the building. Common
sense tells me there should be MUCH MUCH less light
than this...probably no more than 100 lux. Also, when
changing glazing types (ie, from clear to reflective)
a significant change is noticed for noon hour
lighting, but the 4pm Dec21 lighting remains at 200
lux. What could I be mis-assuming?

I have recently run a few simplified test models and
compared my results to true exterior light conditions
using a hand-held light meter. The results are
somewhat frustrating, but interesting nonetheless.

My questions here are particularly:
3. How does Radiance calculate the external light
levels?
4. On a perfectly clear morning in a low pollution
area (assumed turbidity=2) my light meter reading read
approx. 55600 lux. A Radiance test (modelling an
unobstructed horizontal plane outdoors) for the same
date and time, and lat. long settings, with
turbidity=2 produced an exterior light level of 47000
lux (15% off the true reading!). While there may have
been some diffuse reflection off nearby buildings,
there was certainly no direct reflection or shadow on
the light meter. This seems to be much too large a
difference. Any suggestions?

Apologies for the long email. I often read that these
software applications are tested and confirmed but my
data isn't showing an appropriate correlation of the
simulation to reality. Any remarks would be greatly
appreciated.

Caroline Prochazka
M.Arch candidate
University of Waterloo


__________________________________
Do you Yahoo!?
Yahoo! Small Business $15K Web Design Giveaway 
http://promotions.yahoo.com/design_giveaway/