| View previous topic :: View next topic |
| Author |
Message |
thanhtruc
Posts: 11
|
Posted: Thu May 19, 2011 8:38 am Post subject: DB anoucement |
|
|
Hi,
I am using the DB trial version 236005. it has happened to me quite often that DB stop working suddenly (i attached the note from DB) and i lost my all settings for the model. If my company purchases DB, this problem will be solved or not?
Look forward to your feedback.
Thank you.
Regards,
Truc Pham
| Description: |
|
| Filesize: |
16.29 KB |
| Viewed: |
23764 Time(s) |

|
|
|
| Back to top |
|
 |
Andy Tindale

Posts: 2432 Location: Stroud |
Posted: Thu May 19, 2011 9:32 am Post subject: |
|
|
Truc,
Could you tell us more about the problems you have been having? It is difficult for us to comment based on your post. Perhaps you could send a model you have created with instructions to help us to reproduce the crash?
We have produced a new release v2.4.2 which has many fixes for CFD which you may like to try.
There is a diagnostics file which may provide useful information to tell us what DB was doing just before it crashed.
Lastly are you using ATI Radeon graphics cards? Some of these can cause instabilities in DB due an incomplete implementation of OpenGL.
I hope this helps to answer your question.
Andy
|
|
| Back to top |
|
 |
thanhtruc
Posts: 11
|
Posted: Fri May 20, 2011 8:04 am Post subject: complex CFD model |
|
|
HI Andy,
Thank you for your feedback. I have tried to upload my model but the size is too big. However, i tried to reduce the amount of CFD boundaries and this error message has stopped to appear.
by the way, i would like to ask you regarding to the quantity of CFD boundaries. It seems like if i have some many CFD boundaries in my model. The file size is really big (>600MB) and the DB process is really slow (e.g. saving, click to simulation...). Could you give me some advice to build a model with complex CFD boundaries. please see the attached model.
Thank you for your support.
Regards,
Truc Pham
|
|
| Back to top |
|
 |
thanhtruc
Posts: 11
|
Posted: Fri May 20, 2011 8:32 am Post subject: |
|
|
i can not attach my model to the forum. I got this message 'no post mode specified'. the model file is 157MB.
Regards,
Truc
|
|
| Back to top |
|
 |
Andy Tindale

Posts: 2432 Location: Stroud |
Posted: Fri May 20, 2011 10:01 am Post subject: |
|
|
The file will be too big to upload to the forum at 157MB. From your message it sounds like you may have solved the problem with the crash by reducing the number of boundary conditions. Do you still need us to look at the model? If so I suggest that you save the file without the CFD simulation data sets which should make the file small enough to post on the forum.
Andy
|
|
| Back to top |
|
 |
thanhtruc
Posts: 11
|
Posted: Mon May 23, 2011 4:02 am Post subject: |
|
|
Hi,
here is my simplified model. If i put the CFD supply and exhaust boundaries to each block (according to the real set-up of our project), then the crashing problem will occur and file getting biger and bigger. Does DB prefer to deal with the simple CFD model, not too complicated set up of HVAC?
Look forward to your feedback.
best regards,
Truc Pham
| Description: |
|
 Download |
| Filename: |
110513_GC_Atrium.rar |
| Filesize: |
3.12 MB |
| Downloaded: |
4965 Time(s) |
|
|
| Back to top |
|
 |
Steve Potter

Posts: 572 Location: newcastle |
Posted: Thu May 26, 2011 3:24 pm Post subject: |
|
|
Hi Truc,
Please could you upload the version of your model that actually causes a crash? You will need to remove the CFD projects first to keep the file size to a minimum.
Generally speaking, if you can make the arrangement of supply boundaries as simple as possible, you will reduce the complexity of the grid and increase the stability of the solution.
One suggestion I can make concerning your arrangement of supply boundaries is that if you set the minimum discharge velocity for the supply diffusers at a very small value (~ 0.01m/s), then the supply diffusers will be created as a single element which will significantly reduce the complexity of the grid. Supply diffusers are normally created using a number of small sub-elements to maintain whatever minimum discharge velocity has been specified and each of these sub-elements requires a grid cell. If you specify a very small velocity, these elements are not required.
Regards,
Steve
|
|
| Back to top |
|
 |
thanhtruc
Posts: 11
|
Posted: Fri May 27, 2011 1:57 am Post subject: |
|
|
Hi Steve,
I have solved the corrupted problem by using a strong processing computer. Yes, but your suggestion is good to consider when my model is getting more complex.
by the way, when i install the trial version of new DB release 242016 (after using the old version 236005), i only can build them model, but i can not run the simulation. in fact 30days is not enough to get used to the software and comparing the strength and weakness of DB to other CFD simulation software. i hope i can spend a bit more time with DB before my company decides to buy DB for our consultant work.
Look forward to your feedback.
Thank you for your support.
Best regards,
Truc Pham
|
|
| Back to top |
|
 |
Steve Potter

Posts: 572 Location: newcastle |
Posted: Fri May 27, 2011 8:08 am Post subject: |
|
|
Hi Truc,
If you need an extension to the evaluation period for the CFD module, that should not be a problem. Please send an application email to our administration manager (gina@designbuilder.co.uk).
Regards,
Steve
|
|
| Back to top |
|
 |
|