Question-and-Answer Resource for the Building Energy Modeling Community
Get started with the Help page

# U-Value for Simple Glazing in OpenStudio Results

I am entering Simple Glazing with U-1.22, SHGC-0.25 and VT-0.275 for my window construction. The OpenStudio Results popping out in OpenStudio Application 1.2.1 report this with U-1.03, SHGC-0.25, VT-0.28, so the U-value is a bit off - and the VT-value rounded, I suppose. I reviewed https://bigladdersoftware.com/epx/doc... trying to understand why, but the approach discussed there (Rw = 1/U - Ri - Re) gives a Uw-value (1 over Rw-value) MUCH lower than even 1.03, so this 1.03 is probably not even the Uw-value. Any thoughts?

edit retag close merge delete

Sort by » oldest newest most voted

The linked screenshot compares OSApp v1.2.1 (SDK v321, E+ v9.5) vs v1.3.0 (SDK v330, E+ v9.6) input vs output (Benchmark Warehouse skylights, in SI units). I think this is simply linked to the following fix. From the E+ v9.6 Input Output Reference: "Field: U-Factor. ... In versions up till 9.6.0, the maximum allowable input is U-7.0 W/m2·K (~1.22), and the effective upper limit of the glazing generated by the underlying model is around U-5.8 W/m2·K (~1.03). In later versions, such upper bound of the input U value is removed. So is the mis-match between the user input U and the effective U is resolved."

more

The Simple Glazing model uses a much more complex calculation to derive the window U-Value then Rw = 1/U - Ri - Re. It has to convert the U/SHGC/VT summary metrics into detailed window parameters representing an equivalent, feasible, single-layer window, so the derived summary metrics tend not to match inputs. The Simple Window documentation provides the details of the calculation.

more

Thank you. I am wondering then how the U-value from the OpenStudio Results report can be used to prove to USGBC/LEED that the window has indeed the U-1.22 it is supposed to have. I'd have to tell them that yes, it is entered as U-1.22, but due to internal calculations, EnergyPlus actually uses a different number?

( 2022-01-01 22:55:47 -0500 )edit

No, you'll have to find the detailed inputs for the actual window construction you're representing and build up a layer by layer detailed window assembly. So, assuming your getting the summary metrics from some window manafacturer, you just need to ask them for what the actual glass emittance/transmittance, conductivity etc. are. Then the calculated U-Value should match your target.

( 2022-01-02 01:23:37 -0500 )edit

I must admit I am confused. Are you referring to executing all the steps in the Simple Window documentation link you sent earlier? I can only see Step 1 effecting a re-computation of the U-Value, from U-value given (e.g. U-1.22) to glass resistance Rlw = 1/U - Riw - Row, which in turn may become its own U-value, namely Ulw = 1/Rlw? Or are you referring to using something like LBNL's WINDOW?

( 2022-01-02 13:47:53 -0500 )edit

I apologize, yes you're right. I thought there was a regression being done at some point but I must be confusing this with something. I think Denis has the right answer below.

( 2022-01-02 18:18:41 -0500 )edit

## Stats

Seen: 291 times

Last updated: Jan 03 '22