CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   effective size of flow boundary for aerofoil analysis (http://www.cfd-online.com/Forums/cfx/71626-effective-size-flow-boundary-aerofoil-analysis.html)

nangless January 9, 2010 15:20

effective size of flow boundary for aerofoil analysis
 
hey does anyone know what the recommended size of a flow boundary is for aerofoil analysis - I heard it was something like 5 times the size of the length of the aerofoil-any ideas (preferably with a reference)

triple_r January 14, 2010 17:35

Hi,

I've seen it being taken as small as two times (for a hydrofoil, Re~10^7) up to ~10 times (for an external flow around a car, Re~10^8). I don't remember the references right now, but if it is very important I can dig them up.

In my opinion, you should take the extension as much as you can, and the node capacity allows you to. Software usually takes the first or second derivative of the velocity as zero on the downstream boundaries, so the farther the better. Also as you don't, usually, need a lot of resolution very far from the object, you can use larger elements and make the additional computational load as small as possible.

I hope this helps.

cpinz January 16, 2010 14:56

Nangless

to have good convergence try to work ,at least, with these dimensions:

- 20c downstream (c is the chord of the airfoil)
- 10 c upstream
- 5c upper and lower

Coarse mesh near the exterior boundaries
Fine mesh around airfoil and some chords downstream for the wake

For references have a look to Aiaa, Agard, Sae papers/documents

I hope this helps
Cpinz

ghorrocks January 16, 2010 17:12

The size will depend on what you are trying to do. The further out you go the more accurate it will be. So a simulation where 10% error is acceptable will be much smaller than one where 0.1% error is required.

I know people don't like my answers on questions like this because I ask them to establish for themselves, but it is the only way to be sure. You have to do a sensitivity check of the parameter.


All times are GMT -4. The time now is 22:37.