Strange results

Hi,

My name is Catarina and I am trying to perform a lightning simulation with
Desktop Radiance.

I think I followed all the necessary steps to insert my model:
- Design the geometry (its a simple cube, with the dimensions 5*5*2.8, and
a window);
- Attach materials to the surfaces;
- Insert the luminaire;
- Define a gride (I am using the reference grid simualation);
- Define the orientation.

After that, I perform the simulation but the results are very high (I
obtain values for illuminance around 14000 or higher). Additionally, the
lightning distribution around the room does not make sense. I have low
values in some points and very high value in other.

I am performing the analysis in a windows 7 (32bits) operative system. I
don't know if the operative system interfere with the results?

I read something in your site about the optics 5 software. I have already
install it and uninstall and the results remain awkward.

Can you please tell me if I am doing something wrong?

Thank you very much,
Catarina

Hi Catarina!

My name is Catarina and I am trying to perform a lightning simulation with Desktop Radiance.

As far as I know, Desktop Radiance has not been supported for several year. This list is on Radiance, which shares some underlying code with the old Desktop Radiance, but has seen continous development over the years. I suggest that you stop using Desktop Radiance and either go with "pure" Radiance (which means to get into touch with the command-line tools Radiance is made of) or one of the various frontends. Daysim, Relux, the Blender-Integration as well as several integrations in more general building simulation systems are among the options in the latter case.

I think I followed all the necessary steps to insert my model:
- Design the geometry (its a simple cube, with the dimensions 5*5*2.8, and a window);
- Attach materials to the surfaces;
- Insert the luminaire;
- Define a gride (I am using the reference grid simualation);
- Define the orientation.

After that, I perform the simulation but the results are very high (I obtain values for illuminance around 14000 or higher). Additionally, the lightning distribution around the room does not make sense. I have low values in some points and very high value in other.

It is almost impossible to give any useful comment here, as we do not know the model (geometry), the material properties or the luminaire, plus we have unsupported code. From what you wrote, you do not have any daylight in your scene - still, the values you observe could indicate that you have a sky model that you are not aware of, and measure illuminance where your sensor gets hit by direct sunlight.

Still, a very generic hint - if you observe funny results in your illuminance simulation, take a "look" using a more "natural" camera setting with a luminance calculation. This helps to identify problems in the model (missing ceilings, coordinate systems where one would not expect them to be located related to the camera, funny overlaps of surfaces, ...). Still - Desktop Radiance is for historical interest but not for simulation work any more.

I read something in your site about the optics 5 software. I have already install it and uninstall and the results remain awkward.

Optics 5 is a tool to find transmission properties of multi-layer glazings. You can import these into Radiance - but Optics does not somehow interfere with Radiance, and is a separate software.

Cheers, Lars.

Hi Lars,

Thank you very much.
Yes, after analyzing my results I also think that I have a problem with the
skylight but I can not understand where...

I have already think to use the Radiance. Daysim is not good for me because
I also want to simulate artificial light. I read somewhere that I can
insert my model in use Autodesk Ecotect Analysis software and the export it
and simulate with the Radiance. Is this right?

However I am not sure if it is free?
Where can I download it?

Best regards,
Catarina

···

2013/2/22 Lars O. Grobe <[email protected]>

Hi Catarina!

My name is Catarina and I am trying to perform a lightning simulation

with Desktop Radiance.

As far as I know, Desktop Radiance has not been supported for several
year. This list is on Radiance, which shares some underlying code with the
old Desktop Radiance, but has seen continous development over the years. I
suggest that you stop using Desktop Radiance and either go with "pure"
Radiance (which means to get into touch with the command-line tools
Radiance is made of) or one of the various frontends. Daysim, Relux, the
Blender-Integration as well as several integrations in more general
building simulation systems are among the options in the latter case.

I think I followed all the necessary steps to insert my model:

- Design the geometry (its a simple cube, with the dimensions 5*5*2.8,
and a window);
- Attach materials to the surfaces;
- Insert the luminaire;
- Define a gride (I am using the reference grid simualation);
- Define the orientation.

After that, I perform the simulation but the results are very high (I
obtain values for illuminance around 14000 or higher). Additionally, the
lightning distribution around the room does not make sense. I have low
values in some points and very high value in other.

It is almost impossible to give any useful comment here, as we do not know
the model (geometry), the material properties or the luminaire, plus we
have unsupported code. From what you wrote, you do not have any daylight in
your scene - still, the values you observe could indicate that you have a
sky model that you are not aware of, and measure illuminance where your
sensor gets hit by direct sunlight.

Still, a very generic hint - if you observe funny results in your
illuminance simulation, take a "look" using a more "natural" camera setting
with a luminance calculation. This helps to identify problems in the model
(missing ceilings, coordinate systems where one would not expect them to be
located related to the camera, funny overlaps of surfaces, ...). Still -
Desktop Radiance is for historical interest but not for simulation work any
more.

I read something in your site about the optics 5 software. I have already

install it and uninstall and the results remain awkward.

Optics 5 is a tool to find transmission properties of multi-layer
glazings. You can import these into Radiance - but Optics does not somehow
interfere with Radiance, and is a separate software.

Cheers, Lars.

______________________________**_________________
Radiance-general mailing list
Radiance-general@radiance-**online.org<[email protected]>
http://www.radiance-online.**org/mailman/listinfo/radiance-**general<http://www.radiance-online.org/mailman/listinfo/radiance-general>

--
Catarina Araújo

Hi Catarina,

Ecotect works, but is not free.

I've found the best way to install/access Radiance on Windows machines is
the OpenStudio installer - it's free and updated.
https://openstudio.nrel.gov/getting-started-developer/getting-started-radiance

Good luck!
Alex

*Alex Krickx* | LEED AP
Project Engineer
*Integral Group *| 427 13th Street | Oakland CA USA 94612
T 510.663.2070 x 247
integralgroup.com <http://www.integralgroup.com/> | *
[email protected]*

***Trust* | *Nurture** *| *Inspire*

Please consider the environment before printing this email.

···

On Fri, Feb 22, 2013 at 8:53 AM, Catarina Araújo <[email protected]>wrote:

Hi Lars,

Thank you very much.
Yes, after analyzing my results I also think that I have a problem with
the skylight but I can not understand where...

I have already think to use the Radiance. Daysim is not good for me
because I also want to simulate artificial light. I read somewhere that I
can insert my model in use Autodesk Ecotect Analysis software and the
export it and simulate with the Radiance. Is this right?

However I am not sure if it is free?
Where can I download it?

Best regards,
Catarina

2013/2/22 Lars O. Grobe <[email protected]>

Hi Catarina!

My name is Catarina and I am trying to perform a lightning simulation

with Desktop Radiance.

As far as I know, Desktop Radiance has not been supported for several
year. This list is on Radiance, which shares some underlying code with the
old Desktop Radiance, but has seen continous development over the years. I
suggest that you stop using Desktop Radiance and either go with "pure"
Radiance (which means to get into touch with the command-line tools
Radiance is made of) or one of the various frontends. Daysim, Relux, the
Blender-Integration as well as several integrations in more general
building simulation systems are among the options in the latter case.

I think I followed all the necessary steps to insert my model:

- Design the geometry (its a simple cube, with the dimensions 5*5*2.8,
and a window);
- Attach materials to the surfaces;
- Insert the luminaire;
- Define a gride (I am using the reference grid simualation);
- Define the orientation.

After that, I perform the simulation but the results are very high (I
obtain values for illuminance around 14000 or higher). Additionally, the
lightning distribution around the room does not make sense. I have low
values in some points and very high value in other.

It is almost impossible to give any useful comment here, as we do not
know the model (geometry), the material properties or the luminaire, plus
we have unsupported code. From what you wrote, you do not have any daylight
in your scene - still, the values you observe could indicate that you have
a sky model that you are not aware of, and measure illuminance where your
sensor gets hit by direct sunlight.

Still, a very generic hint - if you observe funny results in your
illuminance simulation, take a "look" using a more "natural" camera setting
with a luminance calculation. This helps to identify problems in the model
(missing ceilings, coordinate systems where one would not expect them to be
located related to the camera, funny overlaps of surfaces, ...). Still -
Desktop Radiance is for historical interest but not for simulation work any
more.

I read something in your site about the optics 5 software. I have

already install it and uninstall and the results remain awkward.

Optics 5 is a tool to find transmission properties of multi-layer
glazings. You can import these into Radiance - but Optics does not somehow
interfere with Radiance, and is a separate software.

Cheers, Lars.

______________________________**_________________
Radiance-general mailing list
Radiance-general@radiance-**online.org<[email protected]>
http://www.radiance-online.**org/mailman/listinfo/radiance-**general<http://www.radiance-online.org/mailman/listinfo/radiance-general>

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

Hi Catarina!

I would count Ecotect as one of the building simulation systems including an interface to Radiance. It is not a free one (unlike others, e.g. Openstudio, Esp-r, ...).

You may want to have a look at Livi, which is a Radiance interface for Blender. It looks pretty cool... Artificial lighting is handled by adding an IES candela distribution data block to Blender's lamp object.

http://arts.brighton.ac.uk/research/office-for-spatial-research/projects2/livi

Besides that it is a good idea to at least once "understand" Radiance without an user interface. The frontends are all nice and polished, and give the impression that things are easier than using text files and commands. But they also hide critical aspects, setting some defaults or making assumptions that one should at least be aware of. Take a look at the tutorial section at www.radiance-online.org, especially the two tutorials by Axel are a perfect start!

Cheers, Lars.

···

I have already think to use the Radiance. Daysim is not good for me because I also want to simulate artificial light. I read somewhere that I can insert my model in use Autodesk Ecotect Analysis software and the export it and simulate with the Radiance. Is this right?