CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   ANSYS Meshing & Geometry (http://www.cfd-online.com/Forums/ansys-meshing/)
-   -   [ICEM] For the geometry change (http://www.cfd-online.com/Forums/ansys-meshing/104720-geometry-change.html)

lnk July 13, 2012 13:04

For the geometry change
 
Hi,

I'm trying to modify my geometry to make my mesh generation easier. But how do I know how much will that modification influence my fluid behavior simulation results?

Best regards and many thanks,
lnk

PSYMN July 16, 2012 15:43

Yes, an age old question... how much can you simplify the experiment and still have meaningful results...?

For instance, i was working on submersible probes a couple years ago... These are little (less than 20cm) torpedo things than spin when you drop them in the water. They are supposed to send back sensor readings on their way down.

Our simulations were not predicting the correct behavior because we had left out a small seam (part of the manufacturing process that was not even included in our original CAD models, never mind something we intentionally removed) running from nose to tail. When spinning, this small seam was severely disrupting the boundary layer in a way that the CFD didn't predict, it actually helped the probe fall faster.

What you will need to do is validate your CFD results with physical testing... Until then, you are just guessing that you made the right compromises.

lnk July 16, 2012 16:01

Quote:

Originally Posted by PSYMN (Post 371712)
Yes, an age old question... how much can you simplify the experiment and still have meaningful results...?

For instance, i was working on submersible probes a couple years ago... These are little (less than 20cm) torpedo things than spin when you drop them in the water. They are supposed to send back sensor readings on their way down.

Our simulations were not predicting the correct behavior because we had left out a small seam (part of the manufacturing process that was not even included in our original CAD models, never mind something we intentionally removed) running from nose to tail. When spinning, this small seam was severely disrupting the boundary layer in a way that the CFD didn't predict, it actually helped the probe fall faster.

What you will need to do is validate your CFD results with physical testing... Until then, you are just guessing that you made the right compromises.

Thank you very much!

Best,
lnk


All times are GMT -4. The time now is 16:50.