Need to calculate more than 10 layers

[ali hatuki]

Hi everyone,

I am trying to use WINDOW to calculate SHGC for the product.

The product consists of 14 layers, and I notice that WIDNOW is limited to 10 layers.

What can I do to overcome this?

Best regards,

[Robin Mitchell]

What are you trying to model? 

--
You can view the WINDOW forum online at:
https://groups.google.com/g/berkeley-lab-window
---
You received this message because you are subscribed to the Google Groups "Berkeley Lab WINDOW" group.
To unsubscribe from this group and stop receiving emails from it, send an email to berkeley-lab-wi...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/berkeley-lab-window/24686eff-a180-4294-8a49-2c9ce62f9ab7n%40googlegroups.com.

--

Robin Mitchell

Building Technology and Urban Systems Division
Energy Technologies Area
Lawrence Berkeley National Laboratory

[ali hatuki]

Hi,

I am trying to understand the SHGC of a multiwall sheet made of polycarbonate.

The sheet has 14 layers.

Best regards,

[Collin Robinson]

Ali, I have worked with products like this before, something like Kingspan Quadwall. Some layers are coplanar with the glazing, some are cellular. It's complex. To me, a good case would be a "laminate" material whose optical properties are tested by manufacturer and found in IGDB, visible in the Optics 6 program. There are generic polycarbonate materials in both Optics and Window programs. Are some of the various layers of polycarbonate so different they can't be combined into one thicker layer? That doesn't capture the angular performance of this material, which would really be best described by measured BSDF. That would capture the unusual scattering of cellular polycarbonate. When I was studying this, I was only concerned with visible performance, so I just made a radiance trans material with transmittance based on manufacturer data. 

Another path in WINDOW would be

1. combine some of the polycarbonate layers into one thicker layer

2. simulate the cells by making a Shade layer of Venetian blinds at a spacing similar to your product

3. by using a transparent Shading Material such as 31037 C82 Transparent Plastic. 

Please report back with what you decide. I am curious. 

Best, Collin

To view this discussion visit https://groups.google.com/d/msgid/berkeley-lab-window/3342a88e-3fc0-4cb7-a688-34568678a528n%40googlegroups.com.

[D. Charlie Curcija]

You can use our open-source calculation engine WinCalc and its Python script API implementation PyWinCalc (https://github.com/LBNL-ETA/pyWinCalc), which does not have a limit on the number of glazing layers. BTW, WinCalc is the engine in our newly released WINDOW 8.1 Beta (https://windows-downloads.lbl.gov/software/window/Window8_1_30_SetupSmall.exe). However, WINDOW 8.1 GUI still supports only 10 glazing layers.

Charlie

To view this discussion visit https://groups.google.com/d/msgid/berkeley-lab-window/3342a88e-3fc0-4cb7-a688-34568678a528n%40googlegroups.com.

--

-------------------------------------------------------------------------------------------

D. Charlie Curcija, Ph.D.                              Tel: (510) 495-2602

Lawrence Berkeley National Laboratory         Fax: (510) 486-4089

Windows & Envelope Materials Group           Cell:(510) 604-8668

1 Cyclotron Rd., 90R3147                           Email: dccu...@lbl.gov

Berkeley, CA 94720                                    Web: http://windows.lbl.gov/

-------------------------------------------------------------------------------------------

[ali hatuki]

Dears,

Collin,

I thought about the option of combining layers into a single layer. For example, taking the first 7 layers and simulating them as a glazing, this gives me new Tsol, Rsol, and Keff.

These parameters I can add into a new "Glass" with the new thickness and parameters. 

This way, I can simulate the whole as a double-glazing system. Do you think this is correct?

Charlie, 

Is there a manual or a guide that I can read to understand how this software works?

Best regards,

[D. Charlie Curcija]

This approach has an inaccuracy in calculating convective heat transfer in cavities, which depends on the temperature difference across each cavity. When you model 7 cavities separately, their delta T will be higher than if they were part of 14 layers. I am unsure of the magnitude of inaccuracy, but it is likely not very high and mostly reflected in U-factor calculations. You can get a sense of the error by modeling 10 layers and comparing that to modeling 2 x 5 layers.

Charlie 

To view this discussion visit https://groups.google.com/d/msgid/berkeley-lab-window/e23b432c-9914-449e-8aaf-15124347d6d4n%40googlegroups.com.

[Robin Mitchell]

There is technical documentation on the WINDOW webpage under the documentation link

https://windows.lbl.gov/window-documentation

Here is a link to the PDF from that page

https://windows.lbl.gov/sites/default/files/software/WINDOW/WINDOW%20Technical%20Documentation.pdf

Robin

To view this discussion visit https://groups.google.com/d/msgid/berkeley-lab-window/CAPCvEOFyNR2KME8v31An-LGD-SB-ycVsTy4_yUeC9_YXTOQ8fw%40mail.gmail.com.

[ali hatuki]

Dear  Charlie,

I performed a similar test to what you suggested, but at a lower scale. 4 layers instead of 10.

I did two tests:

1 - Regular test of 4 layers according to the required structure. 

2. "parted" test combined of a glass consisting of the first 2 layers and a glass consisting of the last 2 layers.

In test number 2, I took the Tsol, Rsol, and Keff of each of the combinations.

The results I got were nowhere near the results of the initial test. Only when I adjusted the Keff to an unreasonable value was it close.

Final results were :

* Test number 1: 4 layers -  SHGC 0.407.

* Test number 2: 2 layer x 2 -  SHGC 0.438 

*Test number 3: 2 layers x 2, but I kept conductivity to 0.2 like PC and not 0.06 -> U factor 1.414 and SHGC - 0.417

Also, changing the environmental conditions didn't change the results because they do not affect the Tsol Rsol Keff, which matters. 

I keep testing to see if I can get around this. 

Best regards,