Hi,

I'm trying to run a LEED v4  sDA + ASE simulation with some of the smart new glare controls in DIVA4.

In the example files i could in the 02_AnnualDaylight_C_WindowsAndShades file select two different ways;

Manual Control

- Which I read looks at the points in the grid and see how often they are hit by light

Automatic Control

- Which I read mimics an automatic system based on a sensor in the glass pane.

However, this example file does not work when I select more than a few breps for windows.

I've tried to tweak it with my own test file, and it seems like it runs both manual and automatic fine, but only when running on less than 3-7 windows. Is this really a limitation, or how am I drawing it wrong?

I've attached my test file.

Best, Mathias

My output is:

{0}
0. Write failed.
18 manual shades
0 auto shades
manual sensor ray count = 9495
The pipe has been ended.

The pipe has been ended.

The pipe has been ended.

The pipe has been ended.

The pipe has been ended.

The pipe has been ended.

raytrace loop encountered a break at 1
Run complete.
Results directory: C:\Users\mathias.nielsen\Documents\DIVA\GH_Data\_diva_Ann\AnnualDaylight

Tags: ASE, LEED, sDA

Views: 164

Attachments:

Reply to This

Replies to This Discussion

Hi Mathias,

There's no inherent limitation on the number of windows.  And your file runs fine on my end.

I'm guessing your issue is the same as Luciano's, so try switching your decimal char to a period...

Jon

Thanks Jon,

This worked. and wow - it finds the schedules really fast!

Jon Sargent said:

Hi Mathias,

There's no inherent limitation on the number of windows.  And your file runs fine on my end.

I'm guessing your issue is the same as Luciano's, so try switching your decimal char to a period...

Jon

Oh and by the way - when it creates manual blinds schedules by checking if points in the simulation grid are hit by direct sun - is this the "real" sun vector or is it a gensky? and if its a gensky - how many patches is it made with? I've seen on the Ladybug forum that they have made a ASE calculation from pure vectors -> ASE(250h) is much lower, as the sky patches may get too large and illuminate too many points.

See more here:

http://hydrashare.github.io/hydra/viewer?owner=chriswmackey&for...



Jon Sargent said:

Hi Mathias,

There's no inherent limitation on the number of windows.  And your file runs fine on my end.

I'm guessing your issue is the same as Luciano's, so try switching your decimal char to a period...

Jon

Hi Mathias,

Schedules are created using real sun vectors, not the Daysim ASE method.  For speed's sake, the vectors are actually binned every few days, but the radiance is still calculated for each hour of the year (using weather data).  The geometric binning is quite reasonable given that sun positions shift a maximum of ~0.5 degrees from one day to the next (vs. 15 degrees from hour to hour).

At some point, we intend to add options for sub-hourly sampling and specular reflections (which are missed by the current method).

Jon

ps. A minor technicality, but the Daysim ASE method does not make use of sky patches, since it is direct only.  The illuminance at a particular hour is interpolated using up to 4 nearby sun positions, which are modeled as small (0.5 degree) light sources.  This will both over- and underestimate illuminance values, so the error vis-a-vis any particular threshold may be scene and sensor dependent.

Reply to Discussion

RSS

© 2017   Created by jeff niemasz.   Powered by

Badges  |  Report an Issue  |  Terms of Service