Inconsistency in results from Winkelman and equivalent layer blinds models
I am trying to model vertical blinds and I am getting inconsistent results when modelling this system with the WindowMaterial:Blind (Winkelman, blue in graphs below) and WindowMaterial:Blind:EquivalentLayer(Ashwat, red in graphs below)
I mainly have doubts about the results I get with the ASHWAT model. I tested vertical blinds, 127 cm wide and positioned at a static 45 degrees using both modelling approaches. To be able to simulate 127 cm wide blinds in the equivalent layer model I had to overwrite the maximum width in the IDD file (set to 2.5 cm by default).
Transmitted solar radiation is consistent for both models but the equivalent layer approach gives extremely low indoor surface temperatures (lower than indoor and outdoor temp), no radiative exchange and very high convective exchange.
These links contain the two IDF's I used for this comparison:
The glazing properties are defined differently in each model. I aligned these properties using LBNL-window and I do get consistent and plausible results when I model glazing system without blinds using these models:
I am using EnergyPlus 8.8 for reasons outlined in the answer to this post.
Addition following MJWitte's suggestion. The same comparison but than using a blind width/slat separation of 2 cm and the original unedited IDD file: This gives the same discrepancies. IDF's: Winkelman, ASHWAT.
I also tried the 2 cm blind width but than using horizontal blinds in both models to see if something was wrong in the vertical blind implementation. This also gives the same discrepancies.
Perhaps there is a good reason for the 2.5 cm limit in the IDD. How does this comparison look with 2.5 cm blinds?
This gives the same discrepancies. Tried with 2 cm blinds and the uneditted IDD file. Also tried horizontal blinds. This gives the same conclusion. See the comparison in the question.
New issue posted. 7367
There are two defects identified in ASHWAT that are being addressed. #7367 being one of them. Your test case has the field "Slat Infrared Transmittance" in the WindowMaterial:Blind:EquivalentLayer blank. This currently sets 0K mean radiant temperature for the room, as the results the room acts a huge radiant heat sink and in turn results very low window inside surface temperature. For your test case change input field "Slat Infrared Transmittance" value to 0 instead of blank.This will improved the inside surface temperature prediction. Full fix, will be included in version 9.2.