| View previous topic :: View next topic |
| Author |
Message |
JesusGD
Posts: 23
|
 Posted: Mon Dec 05, 2011 8:34 am Post subject: Internal CFD Double skin façade |
 |
|
Hello everybody,
I'm trying to analyze a double skin façade with the CFD module of DB and I have always the same problem. First of all, I simulate the building for one day and afterwards I load the CFD boundary conditions for this day.
When I try to simulate the building with CFD always alerts me with unbalance flow. I adjust the flows in CFD boundary conditions and ticks ok, but when I go to simulate the CFD it alerts me again with unbalance flow. It's like a loop without solution
What can I do? Has anyone suffered the same problem?
Thank you.
|
|
| Back to top |
|
 |
Steve Potter
Posts: 615
Location: newcastle |
| Posted: Wed Dec 07, 2011 8:48 am Post subject: |
|
|
Hi,
Are you using the latest version of DB? If not, I would suggest that you download it and see if you still experience the same problem.
Regards,
Steve
|
|
| Back to top |
|
|
JesusGD
Posts: 23
|
| Posted: Thu Dec 15, 2011 12:35 pm Post subject: |
|
|
Hi,
Yes, I'm using the latest version. The problem is quite rare, I suspect could be due to the big size of the building, couldn't be?
It is a pity that the CFD module doen'st work well. I attach the file if you want to check it.
Thanks
| Description: |
|
Download |
| Filename: |
Torrelaguna_CFD.dsb |
| Filesize: |
3.13 MB |
| Downloaded: |
286 Time(s) |
|
|
| Back to top |
|
|
Steve Potter
Posts: 615
Location: newcastle |
| Posted: Thu Dec 22, 2011 10:04 am Post subject: |
|
|
Hi,
We have now fixed this problem and the fix will be available in the next update of DB (following version 3.0.0.091) which should be available within the next couple of days.
Apologies for any inconvenience caused by this bug.
Regards,
Steve
|
|
| Back to top |
|
|
JesusGD
Posts: 23
|
| Posted: Mon Dec 26, 2011 11:45 am Post subject: |
|
|
Hi Steve,
I've installed the new release 3.0.0.092 and the problem about unbalance flow looks solved. However, the software collapses trying to initialize boundary conditions. I don't know if it could be due to the large size of the building.
Thanks
|
|
| Back to top |
|
|
felipeduran
Posts: 30
|
| Posted: Mon Jan 02, 2012 2:58 am Post subject: |
|
|
Dear All,
I am trying to analyse natural ventilation on a wide double skin facade importing E+ boundary conditions to a CFD analysis and I have several questions:
1. I have modelled 6 floors one on top of the other and then I have connect them by holes. To define the CFD boundary conditions, do I have to cheek the option "merge zones connected by holes"?
2. In the natural ventilation tab, which air temperature distribution do I have to use, mixed or temperature gradient?
3. To get a reliable surface temperature distribution for the CFD, do I have to use "full exterior and interior" or "full exterior" its enough.
4. Using the option "merge zones connected by holes" I have slightly imbalance and I have corrected it. When I select the CFD tab and it gives me the following message "Flow Imbalance - Do you want to use the boundary conditions editor".
Why do I have to balance it again?
5. If I don't use the option "merge zones connected by holes" the E+ simulation gives me several warnings about "Output:Meter: Invalid name..." and the following error "Too many iterations (SLOVZP) in Airflow Network Simulation" When I reduce the absolute and relative airflow network tolerance to 0.1 and 0.001 it works fine. Is that reduction too big or its ok?
6. Finally, Is there a way to reduce the horizontal openings between zones? creating four holes instead of a big one, does it make any difference?
Thank you very much and I am sorry for this long email. I am attaching the dbs file.
Felipe Duran
| Description: |
|
Download |
| Filename: |
121201 CFD 01.dsb |
| Filesize: |
4.93 MB |
| Downloaded: |
265 Time(s) |
|
|
| Back to top |
|
|
Steve Potter
Posts: 615
Location: newcastle |
| Posted: Mon Jan 09, 2012 2:55 pm Post subject: |
|
|
Hi,
| Quote: | | I've installed the new release 3.0.0.092 and the problem about unbalance flow looks solved. However, the software collapses trying to initialize boundary conditions. I don't know if it could be due to the large size of the building. |
I think you will probably find that if you leave the boundary condition initialisation to run, it will eventually complete and the calculation dialog will appear. However the boundary condition routine is taking an exceptional amount of time to run for your model (~ 3 hours!). We have managed to identify what was taking the time and have modified the routine so that it runs a lot faster now (~ 30 min). This modification will be available in the next update of the software (3.0.0.096) which should be available within the next few days. However, as I say, if you don't want to wait for the update, the current version will run with your model albeit rather slowly.
Apologies for the inconvenience.
Regards,
Steve
|
|
| Back to top |
|
|
|
FAQ
Search
Log in to check your private messages
Usergroups
