Wang,
a cavity of 1m should definitely be modelled by a zone!
The “edge” effects shouldn’t be neglected.
Marion
Dipl.-Ing. Marion Hiller
TRANSSOLAR: KlimaEngineering - Technologien für energieeffizientes Bauen und Nutzerkomfort in Gebäude
Munich - New York - Stuttgart + 49.711.67976.0
Transsolar Energietechnik GmbH, Curiestrasse 2, 70563 Stuttgart
Amtsgericht Stuttgart - HRB 23347, Steuernummer 99073/00911, USt-IdNr.: DE152272639Geschäftsführer: Dipl.Ing. Matthias Schuler, Dipl.Ing. Thomas Auer, Dipl.Phys. Stefan Holst, Prof. Dipl.-Ing. Bleicher
Von: 王洋 [mailto:wanghongyang1767@gmail.com]
Gesendet: Donnerstag, 22. November 2012 16:25
An: Benoit Delcroix; trnsys-users
Betreff: Re: [TRNSYS-users] Fwd: Error creating the wall transfer function coeffients
Dear Benoit,
Thanks for your patient reply!
I saw your attached figure. But there is a new problem. As I said for my first email, my air layer is about 1m, however, this figure's maximal thickness of air is only 300 mm. therefore, Whether or not it is right when I still use 4.5 W/m2K your figure said as the air layer's U-value?
thanks!
Br.
yang wang2012/11/21 Benoit Delcroix <benoit.delcroix@polymtl.ca>
Hi,
The new value is including convection, radiation and conduction (and not only conduction). There is a reference in English in 2009 ASHRAE Handbook - fundamentals (chapter 26). The reference I have is in French (a Belgian regulation on energy performance of buildings). I have attached a jpg file with the table if you want. The first column is the thickness (in mm). The 3 next are depending on the heat flux direction (respectively upward, horizontal and downward). And the values in the table are resistances in m²K/W. The reference in ASHRAE is a little bit more complicated.
Best,
Benoit
Le 2012-11-21 04:27, 王洋 a écrit :Dear Benoit,
Thanks for your reply!
I did as you said, i.e. I changed the air U-value from 0.09252 W/m2K into 4.65 W/m2K. Then the min. timebase 2 hours can work. But I'd like to ask you why we should set this value 4.65 W/m2K? Could you please tell me which reference literature said? Because I checked the air conductivity as 0.09252 kJ/hmK from the literature.
Thanks!
Br.
yang wang2012/11/20 Benoit Delcroix <benoit.delcroix@polymtl.ca>
Hello,
To me, I think you should set your air layer as a resistive layer and set the air U-value to 4.55 - 4.65 W/(m²K) (if there is no ventilation in the basement). And normally, afterwards, you should be able to decrease the timebase value (normally 1 hour should be possible). If you want less, split the heaviest walls with an active layer (probably the roof and groundfloor walls).
Best,
Benoit
Le 2012-11-19 04:31, 王洋 a écrit :
Dear Leen, Jean, Marion, Benoit and others,
Many thanks for your reply!
I checked my Wall's thickness and thermal property etc. again. Why needs it 9 hours as the timebase? I think maybe it is due to my Ground floor (it is the basement's floor of the building) which includes about 0.99 m's Air Layer.
When I changed it into 0.1 m, the min. timebase 2.5 hours can work without the ERROR:Error creating the wall transfer function coeffients.
Therefore, I'd like to ask you,
1) Is it normal for TRNBuild setting how large the timebase is maximally? 5 hours or others?
2) How to solve this problem of the ground floor with about 0.99 m's Air Layer? But our building project indeed has
the ground floor with about 0.99 m's Air Layer. I can't reduce it into 0.1m? Is the only way to try the active layer approach of The group of Michael Kummert in Montreal?
Please see the Attached INF file.
Thanks in advance!
Br.
wang
2012/11/17 Benoit Delcroix <benoit.delcroix@polymtl.ca>
Hello,
I'm one of the students of Michaël Kummert working on a new method to generate the tranfer functions. I thing I can help you for your problem. As Leen said, TRNSYS is unable to generate the transfer functions coefficients for a heavy and insulated wall with a low timebase. In the TRNBuild project, you can modify that value and increase it to allow TRNSYS to generate the transfer functions. Actually, TRNSYS is unable to generate the transfer function as soon as the Fourier number of the wall (thermal diffusivity * timebase / thhickness²) is lower than 0.0005. To check that, TRNBuild make the sum of one of the coefficients series (a, b or c). The 3 sums are equal and means, according to TRNBuild (I don't personally think so...but it's like this), the Fourier number.
As Leen said, it exists a trick. You can use an active layer to split the wall in 2 and then forcing TRNSYS to generate 2 series of transfer functions coefficients for the same wall. As the 2 parts are thinner, you could use a better timebase value. When you insert the active layer in the wall, try to split the layer with the lowest Fourier number. And then to insert the active layer, use the expert mode that allows you to insert an equivalent conductivity. Insert the biggest value you can. Like that, the active layer is negligible at the thermal point of view. Actually, the active layer is acting as a mini-zone to split the wall. It is a little tricky but it works...before having better solution.
Good luck for your project and have a nice week-end,
Benoit
Le 2012-11-17 14:43, trnsys-users-request@cae.wisc.edu a écrit :Send TRNSYS-users mailing list submissions totrnsys-users@cae.wisc.eduTo subscribe or unsubscribe via the World Wide Web, visithttps://mailman.cae.wisc.edu/listinfo/trnsys-usersor, via email, send a message with subject or body 'help' totrnsys-users-request@cae.wisc.eduYou can reach the person managing the list attrnsys-users-owner@cae.wisc.eduWhen replying, please edit your Subject line so it is more specificthan "Re: Contents of TRNSYS-users digest..."
Today's Topics:1. Re: Fwd: Error creating the wall transfer function coeffients(leen peeters)
_______________________________________________TRNSYS-users mailing listTRNSYS-users@cae.wisc.eduhttps://mailman.cae.wisc.edu/listinfo/trnsys-users