[Radiance-general] specularity values

[Lars O. Grobe]

> For a research project using Ecotect and Radiance, me and my colleague
> are looking for /specularity/ values of building materials.

Dear Saskia,

I try to help out a bit, even though I will not provide you with any 
such value here. I hope that I will be able to explain why.

First, the term. I guess you are referring to specular reflectance, 
maybe also direct transmittance. These define the fraction of light 
getting reflected / transmitted directly. With a lot of simplification 
something like incident angle equals outgoing angle for reflection 
(mirror), or light passing straight through (glass). Be aware that the 
real world is not that simple, but let us assume it is.

If you want to measure the specular reflectance, you typically do it 
indirect. A very common approach is to use an integrating sphere with 
either a photocell (if you do not need color information) or a 
spectrometer connected. In a first run, you would measure the total 
reflectance. In the second, you would open a small hole in the sphere so 
that light reflected according to the "incident=outgoing angle"-rule 
would simply escape the sphere and thus be skipped in the measurement. 
This leads to the diffuse reflectance. Substracting both would give you 
the specular reflectance. For transmissive samples, the procedure works 
more or less the same. Doing a direct measurement is rather 
complicated... Measurements on an integrating sphere are a standard 
measurement.

The procedure seams to be straight forward. However it is not. First, 
you can never measure only one outgoing direction, and you never have 
only one incident direction. There is always an angular range, larger 
then zero. Imagine the device for the measurement described above - 
there will be a light beam used, and that is not perfectly parallel and 
has a diameter. And if you open a hole in the sphere, that has more then 
zero mm^2 area - thus it excludes more then one direction. In other 
terms, you are always talking about solid angles.

The reason is not only imperfection in making measurement devices. Even 
a very good mirror reflects light not only into one direction. For each 
incident direction you will always get a distribution of reflected and, 
sometimes, transmitted light. There are some labs offering measurement 
of such distributions (maybe you have seen candela distributions for 
electrical lighting - it is something like that), I have been working at 
one until just recently. Using the measured data is not simple, as the 
data isn't. That is why the observation that for some surfaces, there 
are predominant directions where most of the light is directed to, led 
to the use of terms such as specular and diffuse. We use these 
simplifications to be able to describe the characteristics and, in 
simulation, we use these to optimize algorithms.

Now, for building materials, you will find very different levels of 
support from manufacturers. Some will give you detailed information on 
reflective and transmissive properties (go to the website of Alanod, 
they let you download the bidirectional reflection distribution data of 
their frontside mirrors - the distribution I describe in the paragraphe 
above at a high angular resolution). Quite a lot of manufacturer of 
materials where lighting is involved will give you some values like 
direct/specular and diffuse (glazing, sunshades, ...) but not more. And 
then, a lot of manufacturers may be able to give you color information.

You will also find that, depending on the application, different terms 
are used when talking about reflective and transmissive characteristics. 
Gloss, shininess, application standards differ often from what is more 
established in physics.

The conclusion is that there is no way to simply collect specularity 
values of building materials. Define what you are interested in and what 
your application is. If you are talking about appearance, energy 
transports, ... Are you looking at facades? Glass? Paints? Different 
polishing techniques?

I hope that this helps to find a starting point...

Cheers, Lars.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 3740 bytes
Desc: S/MIME Cryptographic Signature
Url : http://radiance-online.org/pipermail/radiance-general/attachments/20100624/2364d092/smime.bin