failure/bug in calculation of keff in an air cavity?

[Daniel Kehl]

Dear Therm-Team

I calculate thermal bridges of windows and constructions with Therm since 1998 in Germany and i am impressed of Therm. Thanks a lot.

Now i have a question about air cavities in constructions and I think, that there is a failure/bug in Therm or i did not understand the programm.
I have made an PDF-file, where you find a discription of the problem.

Case 1:
a) horizontal cross section
b) gravity arrow: into screen
c) two cavities in two walls (simplified - see in the PDF-file) with the same size but rotated 90° --> The heat direction in both cavities is parallel to the screen (this means: horizontal heat flow)
The result should be: both cavities should be the same conductivity (keff), but the conductivity is different. I had made a rough calculation by hand with ISO 6946. The result of cavity 1 is ok (nearly the same) but the cavity 2 is totaly different. The problem are the dimensions, they are in the opposite.

Case 2:
a) vertical cross section
b) gravity arrow: down !!! different from case 1 !!!
c) one cavity in a wall and one cavity in the roof (simplified - see in the PDF-file): both cavities have the same size but rotated 90° --> The cavity in the wall is a "lying" air cavity (heat flow direction left/horizontal), the cavity in the roof is an upright air cavity (heat flow direction upwards/verical)
The result should be: the conductivity of the air cavity of the wall should be different from conductivity of the air cavity of the roof. In this case they are the same.

In my opinion case 1 and 2 are mixed up.

Thank you for help

Daniel   

[D. Charlie Curcija/LBNL]

Dear Daniel,

Sorry about the late response. This has slipped under the radar. Is this still relevant to you?

Charlie

[Daniel Kehl]

Dear Charlie,

it is still relevant for me and for many other users!!!

Daniel

[D. Charlie Curcija/LBNL]

OK, we will look at this. Thanks for the patience.

Charlie

[Daniel Kehl]

Dear Charlie

I wish you and the whole Team a happy new year. What ist with an answer? Are you working on it?

br

Daniel

[D. Charlie Curcija/LBNL]

Daniel,

We are short staffed, so this is not yet resolved. Hopefully we will have an answer in the next couple of weeks. Could you also send me THERM file, so that we are looking at the same thing. Thank you.

[in...@bautragwerk.de]

Dear Therm team,

the solution to this problem is also interesting for us. 

Thanks for the help 

Martin Epple

[in...@bautragwerk.de]

Dear 

Therm team the solution to this problem is also interesting for us. 

Thanks for the help 

Martin Epple

[in...@bautragwerk.de]

Dear Therm team,

[Robin Mitchell/LBNL]

At this time, we are extremely limited, due to funding cuts, in our ability to look at these issues. However, I have put this on our list of bugs to look into in the next few months. But we can't promise that we will be able to look into it in any shorter time frame. 

Robin 

[Robin Mitchell/LBNL]

Daniel

Can you provide us with the actual THERM files that show this problem, rather than just a PDF? We will need to be able to reproduce the exact problem in order to debug it. 

Robin 

On Wednesday, October 29, 2014 at 7:07:58 AM UTC-7, Daniel Kehl wrote:

[in...@bautragwerk.de]

Dear Therm Team, 

Thanks and best regards

another question: is it possible to use DXF files for version 2000 or newer? 

Martin Epple

[Fabrizio Prato]

Hi.

I made some test on a case similar to Daniel Kehl's and actually something wrong does happen.
Detail regards a horizontal section of an edge between two identical walls with a thin air cavity (10mm): their thermal resistance should be the same but Therm is apparently misled to calculate them in two different ways, even flux directions are not correctly deduced. File properties are set to Jamb and gravity is set "Into the screen".
I tryed with CEN, ISO e NFRC cavities:
- CEN and NFRC cavities are miscalculated also splitting the detail in two different files
- ISO cavity is calculated correctly splitting the detail in two different files

I upload an image and .thm file for trial hoping this can help.

Fabrizio

[Kara Rosemeier]

Was this ever resolved, please? It appears that Fabrizio could reproduce the issue. An update is much appreciated!

Kara

[Robin Mitchell/LBNL]

Our software development funds have been severely cut by the US Department of Energy for the past and the coming year, so it is unclear when we will get this issue resolved. 

[Kara Rosemeier]

Sorry to hear! 

In this light: is there any consideration to go open source with THERM? It's likely that there are developers out there that would take this on. I'd be happy to coordinate an open source project like this, if desired. THERM is an excellent piece of software, and you have done a fantastic job making it available to the public. It would be a pity to have the development stalled indefinitely. Is there hope?

Kara

[Robin Mitchell/LBNL]

Part of our work in 2016 is going to be to explore setting THERM and WINDOW up as open source projects through the US Department of Energy (which has a strong mandate to try to make as much of their software OS as possible). The programs would most likely still be managed by DOE/LBNL. It is somewhat complicated by the fact that groups in the US use the software for rating and labeling products, and use of these programs is written into legislation, so there has to be careful thought about how changes can be made and still be used for certification. So even if it is open source, contributions would be carefully scrutinized before being merged into the public repository. There are several models for this type of open source development run by DOE, such as Open Studio, which is managed by NREL. So hopefully, this will get resolved in the coming year, we will be able to take contributions from other developers, and we can continue to make progress on the programs. But we will see how it all shakes out. 

I will try to keep this forum up to date on our progress in this regard. 

And by the way, it is really wonderful that a few people have taken on the role of supporting the use of THERM, by helping users via this forum, for use with the European standards, which LBNL is not really able to do. 

So the contributions you are all making to the continued use of THERM is great, and very much appreciated !!

Robin 

[Kara Rosemeier]

Great news! Looking forward to it!

[Fabrizio Prato]

Thanks to you all from Europe for your wonderful work.

Fabrizio Prato

[in...@bautragwerk.de]