Using specific .cal file for sky.rad

radiance-daysim
#1

Hi,

I would like to use a specific file for sky definiton. However the sky.rad file does not take into account the .cal file I am trying to use. It always uses the default skybright.cal file.

Here is my definition (trial_sky.cal file that I have generated with gencumulativesky.exe is not taken into account):

Sky Definition

void brightfunc skyfunc
2 skybr trial_sky.cal
0
0
skyfunc glow sky_glow
0
0
4 1 1 1 0
sky_glow source sky
0
0
4 0 0 1 180

#2

It’s not clear what your issue is. Are you saying that the quoted sky definition is still somehow calling “skybright.cal”, or do you have two sky definitions, one calling “trial_sky.cal” and the other calling “skybright.cal”? Is the issue with what you have put in “trial_sky.cal”, which you have not shared? What commands are you using, and what is the output versus the expected output?

You need to give us more to work with.

1 Like
#3

Hi,

Thank you for your reply. The problem was with the syntax. As you said “the quoted sky definition was still somehow calling “skybright.cal”.

I have changed the corresponding line like below:

4 skybr “.\trial_sky.cal”

The “trial_sky.cal” is in a different folder than the sky.rad (one folder up - main folder). Now it is working. Btw, there was also a problem with running rtrace with multiple CPUs. I have removed ambient file generation part in the syntax to avoid bad ambient file error. In brief, I have managed to make it work. Thank you again.

#4

I don’t think ambient file synchronization works under Windows. That is a Unix-only feature, as are many of Radiance’s multi-processing methods.

Glad you were able to sort it out!
-Greg

1 Like
#5

Thank you. I have another question, too:) I am trying to create CIE sky definitions in a sky.rad file by referencing skybright.cal like below:

void brightfunc skyfunc
2 skybr skybright.cal
0
7 xxxxxxxxx

Gendaymtx or Gencumulativesky are both using weather files. It may be ridiculous but I am really wondering if it is possible to create cumulative clear, cloudy or uniform sky for a location to run an annual analysis without weather file? As I suppose, the last option is gensky which helps to create a sky definition for point in time analysis. But it is not suitable for an annual radiation analysis.

#6

Well, it’s a bit of a pain, but you could always run gensky many times, send the output into genskyvec, then average together the vector output values. You would probably need to write a short program or script to do that, though.

1 Like
#7

I think that means running gensky for 8760 hours to achieve annual sky model. As you suggest, I may try to write a script to run gensky for every hour in a year. I have never user genskyvec but will take a look at it.

Btw, do you think this is a good practice? I mean running annual radiation analysis without weather file. Thank you again.

Edit: I will try the syntax below for every hour and than merge sky.dat files in a single file. Last I will try to take the average from the combined file to run it with rtrace.

gensky 9 21 12 +s -a 41 -0 -29 -m -30 | genskyvec -m 2 > sky09_21_12.dat

#8

Nope. Because, without normalising the sun and sky to plausible values at each time-step (i.e. weather file data), you’ve no assurance that the gensky output has validity for whatever location you happen to be using.

Some years ago, un-normalised single-sky conditions featured in LEED and ASHRAE guidance. This was not a good idea for the reasons discussed here.

Cheers
John

Daylight Experts Ltd.
Expert Witness | Simulation | Measurement | Conservation

Associate Editor Lighting Research & Technology

2 Likes
#9

Thank you for the comprehensive information.