Daylight factors

radiance-general
1

Dear all

I’m looking to replicate some of the functionality of IES Radiance in the full version of Radiance. This is so we will not be as restricted by licensing :wink:

The feature I’m interested in is the illuminance calc on the working plane of a space. This is frequently used to feed in to such things as BREEAM or LEED credits.

OK

In IES Radiance, working on a multi roomed model, I would:

  1. Set up some basics, i.e. CIE overcast, materials, image quality
  2. Select analysis type “illuminance - working plane” (usually 0.85 metres above floor level)
  3. Select a space
  4. Simulate!
  5. Repeat 3 until all the spaces I’m interested in are simulated.
    In terms of post processing I would
  • convert each pic to daylight factors
  • Query the PIC by tracing an “Area of Interest”. This is used to work out DF min, max, ave, uniformity, etc. With this I can exclude overlapping spaces and other features such as notional corridors from the working plane calcs.
    If I were to export a 3rd party CAD model (via, say su2rad) consisting of N number of rooms, are there alternatives to the simple IES approach?

For example, a cad model would not possess the massing data of IES.

  • Perhaps a surrogate for step 3 would be to define the view explicitly in a file?
  • How do I set step 2 up?
  • I suspect it may all be controlled in view?
  • Post processing: Is there a tool with similar “area of interest” functionality for querying PIC files?
    I’ve had a good search through “Rendering with Radiance” but it doesn’t appear to cover this.

Many thanks

Chris Yates

2

Is the desired end result a daylight factor falsecolor? or just the data
such as min, max, avg, uniformity?

Depending on what your goal is, you may want to set up a grid of
calculation points (coordinates and direction vector), process your
scene, sky, and grid points through rtrace, and manipulate the the data
as required. See 6.4, Case Study II in RwR dealing with daylight factor
generation. If you need images you could use pvalue to create an image
from the grid of results and process it through falsecolor to put a
scale on it.

-Chris

···

________________________________

In terms of post processing I would

* convert each pic to daylight factors
  
* Query the PIC by tracing an "Area of Interest". This is used to
work out DF min, max, ave, uniformity, etc. With this I can exclude
overlapping spaces and other features such as notional corridors from
the working plane calcs.

____________________________________________________________
Electronic mail messages entering and leaving Arup business
systems are scanned for acceptability of content and viruses

3

I’m looking to do both. The tutorial looks good but i can’t appear to
access the sample files at http://radsite.lbl.gov/radiance/book/index.html

IES appears to extract statistics from the PIC file by allowing the
user to trace an “Area of interest”. Perhaps they’re using their own
program for this. I gather there are other ways to do this but I’ll
have a pop at the tutorial first (if I can get hold of the sample
materials) and then ask some more informed questions (I hope!).

Many thanks

Chris

Christopher Rush wrote:

···

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

4

Try viewing the image with the ximage. I believe return allows you to
select an area of interest, check the manual page for ximage and mind
the units of your image.

···

________________________________

IES appears to extract statistics from the PIC file by allowing the user
to trace an "Area of interest". Perhaps they're using their own program
for this.
____________________________________________________________
Electronic mail messages entering and leaving Arup business
systems are scanned for acceptability of content and viruses

5

Have you checked out the "dayfact" script that's in Radiance already? If you have a model, it does most of the rest of what you're asking.

-Greg

···

From: Chris Yates <[email protected]>
Date: June 18, 2009 7:02:47 AM PDT

Dear all

I'm looking to replicate some of the functionality of IES Radiance in the full version of Radiance. This is so we will not be as restricted by licensing :wink:

The feature I'm interested in is the illuminance calc on the working plane of a space. This is frequently used to feed in to such things as BREEAM or LEED credits.

OK
In IES Radiance, working on a multi roomed model, I would:
Set up some basics, i.e. CIE overcast, materials, image quality
Select analysis type "illuminance - working plane" (usually 0.85 metres above floor level)
Select a space
Simulate!
Repeat 3 until all the spaces I'm interested in are simulated.
In terms of post processing I would
convert each pic to daylight factors
Query the PIC by tracing an "Area of Interest". This is used to work out DF min, max, ave, uniformity, etc. With this I can exclude overlapping spaces and other features such as notional corridors from the working plane calcs.
If I were to export a 3rd party CAD model (via, say su2rad) consisting of N number of rooms, are there alternatives to the simple IES approach?
For example, a cad model would not possess the massing data of IES.
Perhaps a surrogate for step 3 would be to define the view explicitly in a file?
How do I set step 2 up?
I suspect it may all be controlled in view?
Post processing: Is there a tool with similar "area of interest" functionality for querying PIC files?
I've had a good search through "Rendering with Radiance" but it doesn't appear to cover this.

Many thanks

Chris Yates

6

Dear all

I'm looking to replicate some of the functionality of IES Radiance in the
full version of Radiance. This is so we will not be as restricted by
licensing :wink:

The feature I'm interested in is the illuminance calc on the working plane
of a space. This is frequently used to feed in to such things as BREEAM or
LEED credits.

Please be prepared to defend your procedures if someone comes along
and has some doubts about the result. If you use a package that has the
official blessings (like IES) you can just push a button and point to the
certificate to make your results credible.

If you roll your own you have to know what's going on and how to validate
your results.

In IES Radiance, working on a multi roomed model, I would:

   1. Set up some basics, i.e. CIE overcast, materials, image quality
   2. Select analysis type "illuminance - working plane" (usually 0.85
   metres above floor level)
    3. Select a space
   4. Simulate!
   5. Repeat 3 until all the spaces I'm interested in are simulated.

[...]

If I were to export a 3rd party CAD model (via, say su2rad) consisting of N
number of rooms, are there alternatives to the simple IES approach?

Since you asked:

In su2rad you would

1) check that your model is suitable for daylight simulations and
patch the geometry where necessary

2) define polygons where you want your working planes. If you want
more than one you also have to define groups to keep them apart.

3) Export - this will create files with calc points suitable for rtrace
input

4a) manually feed rtrace with these files and pipe it through rcalc to get
lux or df
4b) I might have already bundled a script to do it for you ...

5) use any form of spreadsheet, data visualizer or simple command
line tool to calculate your averages or plots.

In the future it may be possible to create an image from the polygons
via John M.s stencil method and visualise rtrace results in su2rad.
But we're not there yet.

WRT the other points mentioned: You can select an area in ximage
and print out the average as well as highlight max (and min?) values.
But I think you would be better off with defining a mask image and
use a composite. It's a precise and repeatable process as oposed
to manual input of pixel coords.

The tutorial materials should be on the RwR CD. I can have a look
when I'm back home next week.

Cheers,
Thomas

···

On Thu, Jun 18, 2009 at 6:32 PM, Chris Yates <[email protected]> wrote:

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