modeling glazings

Dear colleagues,

I know, I know. The topic of modeling glazings in Radiance using Optics 5
data has been extensively discussed in January 2004. I just reread the
emails with great interest in the archives. BUT I am still undecided whether
to go with Optics 5 is worth the effort.

in January, everybody agreed that creating a specific glazing in Optics 5
creates an output as follows:

···

#
# header info
void glass TestSystem_glass
0
0
3 0.638 0.647 0.568

void BRTDfunc TestSystem_front
10
      0.230 0.263 0.326
      0.585 0.594 0.521
      0 0 0
      .
0
9 0 0 0 0 0 0 0 0 0

void BRTDfunc TestSystem_back
10
      0.289 0.336 0.371
      0.585 0.594 0.521
      0 0 0
      .
0
9 0 0 0 0 0 0 0 0 0

Philip mentioned that to use this material description, one would need "two
glass polygons for each window, one for the front and one for the back". I
assume this means:

TestSystem_front polygon FrontOfWindow
0 0
12 0 0 0
         1 0 0
        1 1 0
        0 1 0

TestSystem_back polygon BackOfWindow
0 0
12 0 0 -0.01
        0 1 -0.01
        1 1 -0.01
        1 0 -0.01
         
Is this correct? How does the glazing know its transmittance, i.e. how does
TestSystem_glass come into play? Also, does anybody have any references or
experience how much more accurate a simulation gets if one uses the full
optics 5 file instead of a stripped down version such as:

void glass TestSystem_glass
0
0
3 0.638 0.647 0.568

Any thoughts are elcome,

Christoph
Christoph Reinhart, Ph.D. tel: (613) 993-9703
Research Officer fax: (613) 954-3733
Institute for Research in Construction e-mail:
[email protected]
National Research Council Canada 1200 Montreal Road M-24, Ottawa
Ontario K1A 0R6 Canada

Daylighting Analysis Software:
The Lightswitch Wizard (initial design) www.buildwiz.com
<http://www.buildwiz.com/lightswitch/index.cfm>
DAYSIM (expert software) www.daysim.com
<outbind://26/www.daysim.com> <outbind://26/www.daysim.com>

IEA Task 31
http://www.iea-shc.org/task31/

Hi Christoph,

Have you investigated the new glaze script in the HEAD release? I just added a feature whereby you can enter your own data for layer transmittance & reflectances, and I expect it is quite a bit more accurate than Optics5 for multiple glazing systems, especially when frits are involved.

-Greg

From: "Reinhart, Christoph" <[email protected]>
Date: September 8, 2004 6:16:31 AM PDT

Dear colleagues,

I know, I know. The topic of modeling glazings in Radiance using Optics 5 data has been extensively discussed in January 2004. I just reread the emails with great interest in the archives. BUT I am still undecided whether to go with Optics 5 is worth the effort.

  
...

Hi Christoph,

I think that there are a lot of things that need to be considered here. Based on some work that Greg has consulted with us on, the basic radiance material definitions output from Optics 5 are perhaps too simplistic to fully account for a variety of glazing related behaviors. Nevertheless let me make a couple of points that may be obvious to many:

   1. There are many types of glazing that will have front and back side
      reflectances that are different and the "glass" material type is
      probably not suitable for these materials.
   2. If you want, it is possible to combine the front and back side
      definitions to one material definition and thus surface geometry
      (one polygon) with uniform normal orientation into our out of the
      space as follows

        void BRTDfunc TestSystem_front_and_back
        10
            if(Rdot, 0.230, 0.289) if(Rdot, 0.263, 0.336) if(Rdot,
        0.326, 0.371)
            0.585 0.594 0.521
            0 0 0
            .
        0
        9 0 0 0 0 0 0 0 0 0

    The "if(Rdot, ReflFront, ReflBack)" will apply the relevant
    reflectance depend upon which side of the polygon is being
    evaluated. This may be suitable depending on the needs (and
    information available) at hand. However this material does not
    suitably account for angular dependencies that may impact
    reflectance and transmittance.

    Note also that if you were to use two pieces of geometry, then you
    would need to be very careful of what you think you are applying
    material wise. For example if you developed an insulating glass in
    Optics 5 made up of 2 glass layers and export this then this needs
    to be applied to "one" polygon not "two."

There is a shell script called "glaze" that is in the current release of Radiance (probably need to get the latest HEAD release for some updates). This script works by enabling the user to specify the makeup of the glazing (currently allowing for single and double pane glazing as well as frit coverage). The makeup is based on a small database that includes front and back reflectances as well as transmittances for a few glass materials. Some of these have been taken directly from the Radiance output provided by Optics 5. We have also added a feature recently that allows a separate file to be specified with additional glazing types (eg build your own). The material definitions (BRTDfunc) that the script outputs have a more robust model for single and double pane glazing systems. Probably, the best thing to do is try it out to see how it works.

One last note/thought. I think that as far as things go, Optics 5, is an excellent resource. The issue is really how to effectively use the data in a useful Radiance material definition.

Regards,

-Jack de Valpine

Reinhart, Christoph wrote:

···

Dear colleagues,
I know, I know. The topic of modeling glazings in Radiance using Optics 5 data has been extensively discussed in January 2004. I just reread the emails with great interest in the archives. BUT I am still undecided whether to go with Optics 5 is worth the effort.
in January, everybody agreed that creating a specific glazing in Optics 5 creates an output as follows:
#
# header info
void glass TestSystem_glass
0
3 0.638 0.647 0.568
void BRTDfunc TestSystem_front
10
      0.230 0.263 0.326
      0.585 0.594 0.521
      0 0 0
      .
0
9 0 0 0 0 0 0 0 0 0
void BRTDfunc TestSystem_back
10
      0.289 0.336 0.371
      0.585 0.594 0.521
      0 0 0
      .
0
9 0 0 0 0 0 0 0 0 0
Philip mentioned that to use this material description, one would need "two glass polygons for each window, one for the front and one for the back". I assume this means:
TestSystem_front polygon FrontOfWindow
0 0
12 0 0 0
         1 0 0
        1 1 0
        0 1 0
TestSystem_back polygon BackOfWindow
0 0
12 0 0 -0.01
        0 1 -0.01
        1 1 -0.01
        1 0 -0.01
         Is this correct? How does the glazing know its transmittance, i.e. how does TestSystem_glass come into play? Also, does anybody have any references or experience how much more accurate a simulation gets if one uses the full optics 5 file instead of a stripped down version such as:
void glass TestSystem_glass
0
3 0.638 0.647 0.568
Any thoughts are elcome,
Christoph
Christoph Reinhart, Ph.D. tel: (613) 993-9703
Research Officer fax: (613) 954-3733
Institute for Research in Construction e-mail: [email protected]
National Research Council Canada 1200 Montreal Road M-24, Ottawa
Ontario K1A 0R6 Canada

Daylighting Analysis Software:
The Lightswitch Wizard (initial design) www.buildwiz.com <http://www.buildwiz.com/lightswitch/index.cfm>
DAYSIM (expert software) www.daysim.com <outbind://26/www.daysim.com>

IEA Task 31 http://www.iea-shc.org/task31/

------------------------------------------------------------------------

_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general

--
# John E. de Valpine
# president
#
# visarc incorporated
# http://www.visarc.com
#
# channeling technology for superior design and construction

I have been looking for fritted and translucent glazing data and had to resort to trans to model them. They seem not to be part of Optics5, unless I did not search carefully. Therefore, I just measure glazing properties with luminance and illuminance meters and approximate them with trans. I am not quite sure when optics5 is needed.

Martin

···

-----Original Message-----
  From: Reinhart, Christoph [mailto:[email protected]]
  Sent: Wed 9/8/2004 9:16 AM
  To: [email protected]
  Cc:
  Subject: [Radiance-general] modeling glazings
  
  Dear colleagues,
   
  I know, I know. The topic of modeling glazings in Radiance using Optics 5 data has been extensively discussed in January 2004. I just reread the emails with great interest in the archives. BUT I am still undecided whether to go with Optics 5 is worth the effort.
   
  in January, everybody agreed that creating a specific glazing in Optics 5 creates an output as follows:
   
  #
  # header info
  
  void glass TestSystem_glass
  0
  0
  3 0.638 0.647 0.568
   
  void BRTDfunc TestSystem_front
  10
        0.230 0.263 0.326
        0.585 0.594 0.521
        0 0 0
        .
  0
  9 0 0 0 0 0 0 0 0 0
   
  void BRTDfunc TestSystem_back
  10
        0.289 0.336 0.371
        0.585 0.594 0.521
        0 0 0
        .
  0
  9 0 0 0 0 0 0 0 0 0
   
  Philip mentioned that to use this material description, one would need "two glass polygons for each window, one for the front and one for the back". I assume this means:
   
  TestSystem_front polygon FrontOfWindow
  0 0
  12 0 0 0
           1 0 0
          1 1 0
          0 1 0
   
  TestSystem_back polygon BackOfWindow
  0 0
  12 0 0 -0.01
          0 1 -0.01
          1 1 -0.01
          1 0 -0.01
           
  Is this correct? How does the glazing know its transmittance, i.e. how does TestSystem_glass come into play? Also, does anybody have any references or experience how much more accurate a simulation gets if one uses the full optics 5 file instead of a stripped down version such as:
  
  void glass TestSystem_glass
  0
  0
  3 0.638 0.647 0.568
   
  Any thoughts are elcome,
   
  Christoph
  Christoph Reinhart, Ph.D. tel: (613) 993-9703
  Research Officer fax: (613) 954-3733
  Institute for Research in Construction e-mail: [email protected]
  National Research Council Canada 1200 Montreal Road M-24, Ottawa
  Ontario K1A 0R6 Canada

  Daylighting Analysis Software:
  The Lightswitch Wizard (initial design) www.buildwiz.com <http://www.buildwiz.com/lightswitch/index.cfm>
  DAYSIM (expert software) www.daysim.com <outbind://26/www.daysim.com> <outbind://26/www.daysim.com>

  IEA Task 31 http://www.iea-shc.org/task31/

I should have given proper credit regarding the glaze script in my last e-mail. Indeed, it was Jack who had the idea to create this script, and did most of the design on it and a lot of the debugging. It would be nice to develop an interface between glaze.csh and Optics 5, but I know very little about the latter.

-Greg

Mmm, Greg is perhaps being overly generous in his credit. Without Greg's input and expertise, we would be at square one in this conversation.

-Jack

Greg Ward wrote:

···

I should have given proper credit regarding the glaze script in my last e-mail. Indeed, it was Jack who had the idea to create this script, and did most of the design on it and a lot of the debugging. It would be nice to develop an interface between glaze.csh and Optics 5, but I know very little about the latter.

-Greg

_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general

--
# John E. de Valpine
# president
#
# visarc incorporated
# http://www.visarc.com
#
# channeling technology for superior design and construction

Jack de Valpine wrote:

Mmm, Greg is perhaps being overly generous in his credit. Without Greg's input and expertise, we would be at square one in this conversation.

...and as I recall, you two started that glaze script conversation in Fribourg almost two years ago! Jack, are you going again?

···

----

      Rob Guglielmetti

e. [email protected]
w. www.rumblestrip.org

Hey Rob,

Unfortunately, I am not going to be able to make it to Fribourg this year. The dates fall too near my son's birthday (three years old!) to make it possible for me to attend. I will look forward to hearing about it and at least obtaining the cdrom image if it is made available at radiance-online.org

-Jack

Rob Guglielmetti wrote:

···

Jack de Valpine wrote:

Mmm, Greg is perhaps being overly generous in his credit. Without Greg's input and expertise, we would be at square one in this conversation.

...and as I recall, you two started that glaze script conversation in Fribourg almost two years ago! Jack, are you going again?

----

     Rob Guglielmetti

e. [email protected]
w. www.rumblestrip.org

_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general

--
# John E. de Valpine
# president
#
# visarc incorporated
# http://www.visarc.com
#
# channeling technology for superior design and construction

Jack de Valpine wrote:

Hey Rob,

Unfortunately, I am not going to be able to make it to Fribourg this year. The dates fall too near my son's birthday (three years old!) to make it possible for me to attend. I will look forward to hearing about it and at least obtaining the cdrom image if it is made available at radiance-online.org

Yeah, I'll be looking for that iso image too!

Three years old? Wow, you know what that means? You're OLD. =8-) Yesterday, my office took our intern out for a sendoff lunch, as he goes back to architecture school tomorrow. He was talking about dorm rooms, cafeteria food, professors, and I got to thinking "my oh my, that seems like such a long time ago". Then I realized it *was* a long time ago! Sheesh. Now where did I put my glasses...

···

----

      Rob Guglielmetti

e. [email protected]
w. www.rumblestrip.org