search

LBNL-ETA / pyWinCalc

Other
19 stars 10 forks source link

Values not matching in WINDOW #2

Closed joshtate closed 5 years ago

joshtate commented 5 years ago

I'm having an issue on certain glazing systems where it does not match the values in WINDOW. Here are the versions I'm using:

For both examples, width and height are set to 1.0m with standard W5_NFRC_2003.std

Also, here are the exported Optics files that were used in the script, but for someone testing this you may want to export them yourself to see if there is any discrepancy there.

Glazing System 1 (closely matches WINDOW):

Glass 1: Activ6.lof Gap 1: AIR @ .0127m Glass 2: Clear_6.VTA

pyWinCalc WINDOW Difference
U-Value 2.690 2.689 0.001
SHGC 0.679 0.679 -
Tvis 0.740 0.740 -
Tsol 0.5873 0.5873 -
Tuv 0.3483 0.3475 0.0008

Glazing System 2 (not matching WINDOW):

Glass 1: SB60 Clear_6.VTA Gap 1: AIR @ .0127m Glass 2: Clear_6.VTA

pyWinCalc WINDOW Difference
U-Value 2.690 1.651 1.039
SHGC 0.412 0.391 0.021
Tvis 0.704 0.704 -
Tsol 0.3356 0.3355 0.0001
Tuv 0.1736 0.1812 (0.0076)

Could anyone explain why these values are not matching?

StephenCzarnecki commented 5 years ago

Hi

The values should match WINDOW. We found a couple calculation bugs that we’ve been fixing and are currently testing the fixes. Thanks for the comparison and sending the files that you were using, I’ll add your system those to the ones we are testing.

I’ll let you know if the current round of bug fixes resolve the differences in results you are seeing. If they do not then we’ll try to find whatever is causing the differences.

StephenCzarnecki commented 5 years ago

Thanks for submitting this issue, it was indeed another bug. I think I've patched it in the latest code. TUV should also be much closer to WINDOW now.

I should say I'm doing the unwise thing of making a patch immediately before going on vacation. The values you were seeing were bad enough that I wanted to correct the issue sooner rather than later.

If you find any more inconsistencies with WINDOW or have any other problems please let us know. Hopefully someone else may be able to help you before I am back. Otherwise I will be back on July 1st.

joshtate commented 5 years ago

Glad to hear you were able to reproduce my issue! I'll give this a try later today or early next week and I'll report back with my findings.

StephenCzarnecki commented 5 years ago

Ok thanks. As I mentioned before I'll be away today and next week with not much internet access. So apologies in advance if something goes wrong. I'll be back July 1st.

On Friday, June 21, 2019, Josh Tate notifications@github.com wrote:

Glad to hear you were able to reproduce my issue! I'll give this a try later today or early next week and I'll report back with my findings.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/LBNL-ETA/pyWinCalc/issues/2?email_source=notifications&email_token=AB2AZH6DBJXAPNYGF6UXQPLP3TH2LA5CNFSM4HZ2W34KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODYINU3Y#issuecomment-504420975, or mute the thread https://github.com/notifications/unsubscribe-auth/AB2AZHZWMLLM7HU5GA5RF6LP3TH2LANCNFSM4HZ2W34A .

joshtate commented 5 years ago

The new version is producing consistent results now. Thank you!