|
[Sponsors] |
June 22, 2020, 12:00 |
Analysis differences with respect to size
|
#1 |
New Member
Ozgur
Join Date: Feb 2020
Posts: 7
Rep Power: 6 |
I have a wooden computer. So solving big geometries is nearly impossible. When I scale geometry like 1/10 would that matter or change the results ?
|
|
June 22, 2020, 15:18 |
|
#2 |
Senior Member
Erik
Join Date: Feb 2011
Location: Earth (Land portion)
Posts: 1,171
Rep Power: 23 |
It depends on what you are doing, but scaling the size is not a smart solution.
Geometric Size should not matter for calculation time. It is the # of mesh elements which matters. |
|
June 22, 2020, 17:47 |
|
#3 |
New Member
Ozgur
Join Date: Feb 2020
Posts: 7
Rep Power: 6 |
Yes the problem starts after meshing. I have a wooden pc and if my mesh number goes like 2m computer gives me headaches. Solution takes hours and i cant even see if its right or wrong. So I thought i could scale body, give appropriate meshing and solve it in a hour. But it is not ideal, you say.
|
|
June 23, 2020, 03:22 |
|
#4 | |
Senior Member
Filippo Maria Denaro
Join Date: Jul 2010
Posts: 6,779
Rep Power: 71 |
Quote:
Several hours of computational time are quite normal... |
||
June 23, 2020, 03:48 |
|
#5 |
New Member
Ozgur
Join Date: Feb 2020
Posts: 7
Rep Power: 6 |
So if i rearrenge re number an continue with scaled geometry it is ok. I am trying to solve for a small aircraft. But geometry is too big for pc. Always stops, program freezes etc.
|
|
June 23, 2020, 08:24 |
|
#6 | |
Super Moderator
Alex
Join Date: Jun 2012
Location: Germany
Posts: 3,400
Rep Power: 47 |
Quote:
If you decreased the geometric size of your model by a factor of 10, you would need to e.g. decrease viscosity by a factor of 10 in order to have the same Reynolds Number. Otherwise, you would change physics, and get absolutely useless results in regards to your original case. That in turn means your cell size would also have to decrease by a factor of 10 in order to achieve the same resolution. You can scale the geometric size all you want, it does nothing to reduce computational cost. |
||
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Issues with poor performance in faster CPU | gian93 | Hardware | 9 | October 29, 2018 13:34 |
Superlinear speedup in OpenFOAM 13 | msrinath80 | OpenFOAM Running, Solving & CFD | 18 | March 3, 2015 05:36 |
Bubble size analysis in Fluidized beds | Musa | FLUENT | 1 | March 18, 2013 04:01 |
[Commercial meshers] ST_Malloc: out of memory.malloc_storage: unable to malloc Velocity SA, | cfdproject | OpenFOAM Meshing & Mesh Conversion | 0 | April 14, 2009 15:45 |
Phase locked average in run time | panara | OpenFOAM | 2 | February 20, 2008 14:37 |