| View previous topic :: View next topic |
| Author |
Message |
adrien26
Posts: 20
|
 Posted: Tue May 06, 2008 4:12 pm Post subject: can not produce EPC |
 |
|
Hi,
I usually do not have any problems to calculate and produce EPC certificates with design builder.
However, I have been struggling a lot on a University Hall Building. I have tryed to keep the model as simple as possible but at the end of the calculation the EPC certificate is not displayed.
Also, during the calcs I can see on MSdos: "error loading EPCgen.dll library"
Best regards,
ADrien
| Description: |
|
Download |
| Filename: |
St Andrews uni final.dsb |
| Filesize: |
1.14 MB |
| Downloaded: |
839 Time(s) |
|
|
| Back to top |
|
 |
adrien26
Posts: 20
|
| Posted: Tue May 06, 2008 4:17 pm Post subject: Re: can not produce EPC |
|
|
| adrien26 wrote: | Hi,
I usually do not have any problems to calculate and produce EPC certificates with design builder.
However, I have been struggling a lot on a University Hall Building. I have tryed to keep the model as simple as possible but at the end of the calculation the EPC certificate is not displayed.
Also, during the calcs I can see on MSdos: "error loading EPCgen.dll library"
Best regards,
ADrien |
The last updated file is the following:
| Description: |
|
Download |
| Filename: |
St Andrews uni last calcs.dsb |
| Filesize: |
2.47 MB |
| Downloaded: |
954 Time(s) |
|
|
| Back to top |
|
|
Andy Tindale
Posts: 1198
Location: Stroud |
| Posted: Tue May 06, 2008 4:23 pm Post subject: |
|
|
Hi Adrian,
Your model seems to run fine using the latest version of DesignBuilder (version 1.5.0.061 beta). I would recommend downloading that to see if it helps at your end too.
We expect to release a full version of 1.5 soon.
Andy
| Description: |
| EPC generated from your file |
|
Download |
| Filename: |
EPC.pdf |
| Filesize: |
8.75 KB |
| Downloaded: |
103 Time(s) |
|
|
| Back to top |
|
|
Andy Tindale
Posts: 1198
Location: Stroud |
| Posted: Tue May 06, 2008 4:48 pm Post subject: |
|
|
OK I've reproduced the problem with the second (much larger) file you posted. I'll send the file to the SBEM developers at BRE to find out why SBEM is failing with this input data. The most likely cause is the quantity of data requiring processing.
DB also crashes attempting to process the large htm data reflection reports and I will add some code to avoid this happening in future.
I'll let you know when I hear back from BRE.
Andy
|
|
| Back to top |
|
|
adrien26
Posts: 20
|
| Posted: Tue May 06, 2008 6:15 pm Post subject: |
|
|
It seems to be because I have cloned the first floors of my building to elevate the 2nd, 3rd and 4th.
If I run the calcs without these floors the EPC is produced with no problems.
P.S. I have cloned the building including activities, lighting, HVAC and constructions data. I hope the cloning tool will be operational. This will save us a lot of time...
Thank you,
Adrien
|
|
| Back to top |
|
|
Andy Tindale
Posts: 1198
Location: Stroud |
| Posted: Tue May 06, 2008 7:45 pm Post subject: |
|
|
Well the cloning tool should be 'operational' and I suspect that the problem is that SBEM is not able to cope with the volume of data. We'll see when BRE respond.
Andy
|
|
| Back to top |
|
|
adrien26
Posts: 20
|
| Posted: Wed May 07, 2008 12:12 pm Post subject: |
|
|
Another one which did not work. It is quite a simple building.
Adrien
| Description: |
|
Download |
| Filename: |
building.dsb |
| Filesize: |
1.35 MB |
| Downloaded: |
855 Time(s) |
|
|
| Back to top |
|
|
Andy Tindale
Posts: 1198
Location: Stroud |
| Posted: Wed May 07, 2008 1:39 pm Post subject: |
|
|
I've also passed this model onto BRE for comment.
Andy
|
|
| Back to top |
|
|
Andy Tindale
Posts: 1198
Location: Stroud |
| Posted: Wed May 07, 2008 3:15 pm Post subject: |
|
|
A quick update:
It looks like the problems you found in building.dsb are at least partly caused by the use of the obsolete 'Hospital cellular office' activity. This activity should not be in the DB activity database. I got a simple test version of your models to run by replacing these activities with the open plan equivalent.
Hopefully more info on how to get round the problem to follow...
|
|
| Back to top |
|
|
Andy Tindale
Posts: 1198
Location: Stroud |
| Posted: Wed May 07, 2008 3:34 pm Post subject: |
|
|
Try changing the activity in Zones 2, 23, 2 and 15 in the Ground floor eye surgery block to Hosp_OpenOff - this allows EPC to be generated on my computer.
Still looking into the problems with the University building.
Andy
|
|
| Back to top |
|
|
adrien26
Posts: 20
|
| Posted: Thu May 08, 2008 12:42 pm Post subject: |
|
|
Thanks for that Andy.
Looking forward to hearing from the university one.
Adrien
|
|
| Back to top |
|
|
adrien26
Posts: 20
|
| Posted: Thu May 08, 2008 1:30 pm Post subject: |
|
|
| well, we have just tried to run the calcs after changing the cellular offices to hosp_openoff and it still does not work. It is definitely coming from the eye surgery floor. We have run the calcs without it and produced the EPC with problems...
|
|
| Back to top |
|
|
adrien26
Posts: 20
|
| Posted: Thu May 08, 2008 1:33 pm Post subject: |
|
|
Could you please send me the file you have run?
I could have a look at the changes you have made.
Many thanks,
Adrien
|
|
| Back to top |
|
|
Andy Tindale
Posts: 1198
Location: Stroud |
| Posted: Thu May 08, 2008 1:36 pm Post subject: |
|
|
The response from BRE on the University model is:
Azza Nabil, On behalf of the SBEM team
wrote: | Indeed, the calculation seems to be crashing due to the large number of objects being processed. Although there seems to be 17 zones in the input file, there are 5584 envelopes and 857 windows, which is in itself unusual ( i.e., average 328 envelopes per zone). For building regulations calculations purposes, we recommend that users generally avoid generating more than 100-150 zones in iSBEM. However, the processing time will depend on the total number of objects (not just zones), i.e., zones, envelopes, windows, etc.
Note that for building regulations compliance checking, the calculation has to generate 2 buildings only: the actual and notional, so the number of objects (all the zones, envelopes, windows, etc.) that the calculation has to process is multiplied by 2, while for the EPC to be generated, 4 buildings need to be generated: actual, notional, typical, and reference, i.e., all the objects in the input are multiplied by 4 to give the total number of objects being processed by SBEM.
Are you sure your client needs this many objects? For example, have they merged all contiguous similar areas (according to the zoning rules)? Have they made full use of the multiplier function to describe multiple occurrences of the same zone (e.g. on many floors)? This is a much more efficient use of computational time because it does not simply create repeated zones but multiplies up the results. And there is also the fact that having lots of zones increases the likelihood of errors and the difficulty of applying QA procedures.
If there are still too many zones, then we suggest merging zones which are different but sufficiently similar. This might reduce 'accuracy' but is a short-term 'fix'. Alternatively, they can choose to use a dynamic simulation software package in order to model such large complex buildings. A list of CLG-accredited software can be accessed from http://www.ncm.bre.co.uk/software.jsp.
However, if funding allows, we will look into memory use optimisation options in the next releases of SBEM. In the meantime, we generally suggest your client seek alternatives to the current zoning structure in their building in order to reduce the number of objects that need to be processed within SBEM. |
The issue is certainly not number of zones but is perhaps related more to number of walls and windows. I'll take another look to see if there is a way to get it running. The model is right on the edge in that deleting just the roof and part of the top floor gets it running so it may be possible just by some simplification.
My view is that SBEM should be able to deal with buildings of this size without problems and that this is a bug.
|
|
| Back to top |
|
|
Andy Tindale
Posts: 1198
Location: Stroud |
| Posted: Thu May 08, 2008 4:29 pm Post subject: |
|
|
Adrien,
Just to let you know I've made a modification in DB that reduces the number of surfaces generated by lumping similar surfaces together reducing the calculation load on SBEM. The change allows your models to generate EPCs correctly. Results are very simular to calculations without lumping and providing it passes our internal testing this option will be included in the next release.
This change significantly speeds up SBEM calcs in all models.
Andy
|
|
| Back to top |
|
|
|
Home
FAQ
Search
Log in to check your private messages
Usergroups
