[esp-r] Re: again about Radiance

[Achim.geissler]

Dear Gian Luca

I have added some comments, below.

Best
Achim

--------- Original-Nachricht --------
Von: Gian Luca Brunetti <gianluca.brunetti at polimi.it>
An: esp-r at lists.strath.ac.uk <esp-r at lists.strath.ac.uk>,
gianluca.brunetti at polimi.it <gianluca.brunetti at polimi.it>
Betreff: [esp-r]  again about Radiance
Datum: 13/09/11 16:58

> Dear list,
> I am writing again about the ESP-r Radiance interaction.
> 
> I August while away from my computer I optimistically posted a couple of 
> messages here in which I reasoned about the possibility to modify the 
> insolation values in the .shda files taking into account the illuminance 
> values produced by Radiance at each time step. (This by using a ratio of 
> the illuminance levels calculated with solar obstruction as black bodies 
> and the illuminance levels calculated with their true reflectances.)
> 
> Now inspecting the interface to Radiance in ESP-r, which I only 
> previously used for daylight factors calculations, I have seen that in 
> the interactive set-up is a menu entry for the ESP-r - Radiance coupling 
> and another for the calculation of daylight coefficient.
> 
> I also saw the 1998 article  by Clarke and Janak &quot;Simulating the
Thermal 
> Effects of Daylight-controlled Lighting&quot; (Building Performance,
1998).
> 
> About the coupling mode, I think I have understood that it comes close 
> to my needs to take into account the effect of reflections by external 
> obstructions; but also that this operation only obviously regards just 
> illuminance levels, without effect on solar gains.

I would agree, here. If you give your obstructions a material, this material
is used by the generated Radiance model. However, as you write, only
daylight performance is calculated by the coupling.

> And about the daylight coefficient calculations, I think I have 
> understood that each set of them belongs to a specific position in time.

The radiance coupling is generally on a (building) simulation time step
basis.

> It doesn't help my understanding that I am not able to run the two 
> operations, nor the one, nor the other. (Incidenlty, I also have not 
> been able to run the operations to obtain glare evaluations, which I 
> presently am not longing for.)

Which operation does not work? The setup of the coupling is done via "casual
gains control". Have you set this up? It is control type 4.

> I imagine that there must be some Radiance files to write or modify to 
> place sensors at surfaces or else, but I have no clue about where to 
> start; and I also most fundamentally suspect that I am not on the right 
> track at a more general level.

Yes, in the .cgc file you enter the coordinates of the sensors used. The
Radiance results are compared against a set point (desired lux level) and
then "the rest" is added as internal gain due to artificial lighting.

> That said, running the two above described operations has just 
> indirectly to do with present aim, and might be not essential for me. My 
> priority is to find a way to extract grids of illuminance levels for 
> each opaque internal surface from the  rendering operations, to obtain 
> the mean illuminance levels for each of them. (Morevoer, relative 
> differences of illuminance from surface to surface would be enough for 
> my needs - I woun't necessarily need absolute values; and relative 
> values might even be inferred from to-be-rendered colors.)  But the 
> Radiance rendering options I have seen produce graphical results, not 
> textual.

If you do not really need the energy level, but the illuminance levels,
only, maybe "daysim" is the better tool? This also uses Radiance but is
geared toward whole year daylighting analysis and gives illuminance levels
for a grid which you can define.

> About all this, I poke in the source of the Radiance's file
&quot;dayfact&quot;, 
> which is there presented as being able to calculate daylight factors and 
> illuminance levels, but I haven't found no clue about the calculation or 
> set up needed to request illuminance levels.
> 
> Does anybody using Radiance here have a suggestion about how to obtain 
> the mean illuminance level for a certain surface in a certain position 
> in time?

Another possible approach is via more basic Radiance tools. There is a very
good tutorial on "whole year simulations", however, I do not have it to hand
right now (search the net for Axel Jacobs tutorial and you should find quite
a lot information).

> I fear that if I can't perform this most basic operation I'll have to 
> throw in the towel.
> 
> Thank you in advance.
> 
> Best regards
> 
> Gian Luca Brunetti
> 
> Politecnico di Milano
> Dipartimento BEST
> via Bonardi 11
> 20133 Milano
> tel./fax 02 2399 5750
> 
> 
> 
> 
> _______________________________________________
> esp-r mailing list
> esp-r at lists.strath.ac.uk
> http://lists.strath.ac.uk/mailman/listinfo/esp-r
> 
>