Discrepancy between Illuminance calc and Illuminance from DA calc

hey all

I’m getting a significant and persistent discrepancy between point-in-time illuminance calcs and the illuminance from DA calc. The illuminance from DA consistently is about 3x the illuminance values of the regular point-in-time calc . I tried running the point-in-time with different sky conditions (all CIE skies) and can’t re-create the DA output. Attached are images of 6-21 at noon - we've tested other times in the year with same results.

 I changed no settings between the runs. My radiance parameters: -ab 5 -ad 1500 -as 20 -ar 300 -aa 0.1

(The color range in the images is 0-50 footcandles)

Which is 'correct'?

thanks

dan

Tags: Daylight, autonomy, illuminance

Views: 138

Attachments:

Reply to This

Replies to This Discussion

Hi Dan,

Thanks for your questions.

It looks more like the DA values are a little less than 2x the DA results. The intermediate sky is lower, but that makes sense. The overcast is a little more mysterious since it is so similar to the clear sky. It's hard to really know without the Rhino/DIVA files.

Daylight Autonomy uses the climate file and corresponding direct normal and diffuse horizontal values to generate the sky conditions, whereas the illuminance runs that you are running use the CIE skies. The weather data, for instance, takes things like cloud cover into account, where as the illuminance runs do not take local weather conditions into account. This is a major difference between annual calculations and simple point-in-time calculations. To make a more fair comparison, you can use values from the weather file, and use the Perez-Sky option - inputting the values directly from the weather file. 

Also, I can't tell from the images what geometry you are simulating, or what climate you are using. These things can affect outputs as well.

Best,

Kera

hey kera-

thanks for the reply -

yea figured the sky model had something to do with it. unfortunately we're tied to a particular illuminance level for this project (rare books reading room), and the discrepancy means the difference between significantly different glazing specs.  i also just tried running the model with same specs in 3ds Max and again got different numbers, even higher than DIVA...(both mrSky and IES sky)

here's a zip folder with all the files if you have a sec to take a look.

http://ftp.lampartners.com/_FH4FSmdvrzkUsR

thanks

dan

Hi Dan,

Just a suggestion, but what  might be a good idea for a rare book room could be to use the annual illuminance file that comes from running a climate-based calculation to calculate mean annual illuminance during working hours.

You can also use the climate-based illumiannce file to pull out the maximum values received over the entire year.

Alstan

thanks alstan-

that's a good idea.

But just comparing the 2 calcs, does that mean the climate-based illuminance calc is in fact more accurate than the point-in-time? seems like a difference between ~45fc and ~75fc is pretty large between the calculation types.

The DA calc is more accurate because it is using the weather data. To get a better handle on the comparison, you would need to run the perez sky as I mentioned above. That is not a 1:1 comparison, but it will be closer/more accurate.  3dsMax is a different software, and you would need to talk to Autodesk about that.

Kera

Dan Weissman said:

thanks alstan-

that's a good idea.

But just comparing the 2 calcs, does that mean the climate-based illuminance calc is in fact more accurate than the point-in-time? seems like a difference between ~45fc and ~75fc is pretty large between the calculation types.

Hi Dan,

I am really reiterating what Kera said about, but...

> Climate based point-in-time illuminance result is based on a sky model derived from measured data at some point in the past. So you never really know exactly what you are comparing to (could be overcast, sunny, or intermediate), but it is probably 'accurate' to whatever specific sky (solar irradiance) was measured at that point in the EPW file.

> Clear Sky / Overcast sky are ideal sky models. The overcast sky is obviously completely occluded by clouds, and the clear sky has very little atmospheric diffusion.

In the end, comparing one to the other is really an apples to oranges kind of thing.

Alstan

fair enough! will calc the mean. 

Reply to Discussion

RSS

© 2018   Created by jeff niemasz.   Powered by

Badges  |  Report an Issue  |  Terms of Service