Aspect Ratio
166 views
Skip to first unread message
Ghaleb Hamdan
Jan 4, 2012, 5:36:36 PM1/4/12
to FDS and Smokeview Discussions
Hi,
I read in the manual that numerical instability could be caused by an
aspect ratio larger than 2:1.
I am running a 2D case, with aspect ratio of 3:1 and so far the
calculations have not failed, and so far it looks reasonable.
I am not sure if I should trust my results or not, other than
numerical instability that causes FDS to end the run, would an aspect
ration larger than 2:1 cause any other issues?
I read in the manual that numerical instability could be caused by an
aspect ratio larger than 2:1.
I am running a 2D case, with aspect ratio of 3:1 and so far the
calculations have not failed, and so far it looks reasonable.
I am not sure if I should trust my results or not, other than
numerical instability that causes FDS to end the run, would an aspect
ration larger than 2:1 cause any other issues?
Kevin
Jan 4, 2012, 5:58:16 PM1/4/12
to FDS and Smokeview Discussions
The 2:1 ratio is a suggestion, not a formal stability constraint. Just
remember that accuracy is only as good as the longest dimension of a
grid cell, so you are not buying anything by stretching the cells.
remember that accuracy is only as good as the longest dimension of a
grid cell, so you are not buying anything by stretching the cells.
Craig Weinschenk
Jan 4, 2012, 6:05:37 PM1/4/12
to FDS and Smokeview Discussions
A potential way to check if your aspect ratio is causing significant
issues with your particular simulation would be run cases with aspect
ratios of 1:1 and 2:1. Then look to see if the parameter of interest
in your model is changing as a function of aspect ratio. I suspect the
results change based on what Kevin mentioned with respect to
accuracy.
issues with your particular simulation would be run cases with aspect
ratios of 1:1 and 2:1. Then look to see if the parameter of interest
in your model is changing as a function of aspect ratio. I suspect the
results change based on what Kevin mentioned with respect to
accuracy.
Ghaleb Hamdan
Jan 4, 2012, 6:20:59 PM1/4/12
to FDS and Smokeview Discussions
2D - My cell number in X is 3000 while in Z it's 70, which makes the
cell size aspect ratio 1:3, my flow reaches steady state and is as
expected
Going with X 2000 and Z 70, I get flow buckling, and my flow never
reaches steady state
going for an aspect ratio of 2:1, means my cell count is 3000 for X
and 210 for Z, my computer can't handle that, so aspect ratio of 1:1
won't work either
Question is, if i stick to 3000X70, which means the cell sizing aspect
ratio is 1:3, if FDS doesn't fail, and in the absense of any other
methods of calculating with a finer grid size, do I trust my results?
( I know it's not a black and white kind of question, i apologize in
advance)
> > ration larger than 2:1 cause any other issues?- Hide quoted text -
>
> - Show quoted text -
cell size aspect ratio 1:3, my flow reaches steady state and is as
expected
Going with X 2000 and Z 70, I get flow buckling, and my flow never
reaches steady state
going for an aspect ratio of 2:1, means my cell count is 3000 for X
and 210 for Z, my computer can't handle that, so aspect ratio of 1:1
won't work either
Question is, if i stick to 3000X70, which means the cell sizing aspect
ratio is 1:3, if FDS doesn't fail, and in the absense of any other
methods of calculating with a finer grid size, do I trust my results?
( I know it's not a black and white kind of question, i apologize in
advance)
>
> - Show quoted text -
Ghaleb Hamdan
Jan 4, 2012, 6:25:27 PM1/4/12
to FDS and Smokeview Discussions
Correction:
3000:210 = 1:1 aspect ratio
3000:210 = 1:1 aspect ratio
dr_jfloyd
Jan 4, 2012, 8:27:58 PM1/4/12
to fds...@googlegroups.com
If your results look OK. That is you don't see anything that looks like a stability problem, then you probably don't have a stability issue. Typically when there is a stability problem, it is noticeable. Now that doesn't mean there isn't an accuracy issue due to the stretching. Unfortunately, the only way to really evaluate that would be to run at different resolutions / stretching. If you can't do 1:1, what about 2:1? Or 1:1 but at a slightly lower resolution.
Ghaleb Hamdan
Jan 4, 2012, 11:00:39 PM1/4/12
to fds...@googlegroups.com
I am trying on a different computer that would run a better resolution! I hope it works out! thank you
--
On Wed, Jan 4, 2012 at 5:27 PM, dr_jfloyd <drjf...@gmail.com> wrote:
If your results look OK. That is you don't see anything that looks like a stability problem, then you probably don't have a stability issue. Typically when there is a stability problem, it is noticeable. Now that doesn't mean there isn't an accuracy issue due to the stretching. Unfortunately, the only way to really evaluate that would be to run at different resolutions / stretching. If you can't do 1:1, what about 2:1? Or 1:1 but at a slightly lower resolution.
To view this discussion on the web visit https://groups.google.com/d/msg/fds-smv/-/h094goghW9AJ.
--
You received this message because you are subscribed to the Google Groups "FDS and Smokeview Discussions" group.
To post to this group, send email to fds...@googlegroups.com.
To unsubscribe from this group, send email to fds-smv+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/fds-smv?hl=en.
Ghaleb Hamdan
Graduate Student
San Diego State University Solar and Combustion Laboratory
5500 Campanile Drive
San Diego, Ca 92182
Phone: 708-890-0530
San Diego, Ca 92182
Phone: 708-890-0530
Reply all
Reply to author
Forward
0 new messages