|
[Sponsors] |
November 11, 2004, 01:32 |
Confusion on 2-layer method
|
#1 |
Guest
Posts: n/a
|
Hello! I hope someone can shed a light on this.
I had forgotten to activate the 2-layer for entire surface of a wall and instead only for a small portion of a wall. But, when I plot the 2-layer result, I found that the entire wall was activated with 2-layer with a non-constant near-wall thickness. This may or may not be desirable. It seems like Star automatically assigned a 2-layer to the rest of the wall according to some computation which is not clear to me as I had forgotten to input and activate the 2-layer at portions where it should be activated. I am using version 3.15. Thank u in advance. |
|
November 18, 2004, 22:18 |
Re: Confusion on 2-layer method
|
#2 |
Guest
Posts: n/a
|
Two layer can be activated on a wall region by region basis so it sounds like your whole wall was one reqion.
(this allows huge savings on cell numbers if you only require the two layer technique locally) Thus simple add some wall boundaries of a different region number to your two layer location and only activate two-layer for this boundary region. |
|
November 20, 2004, 23:55 |
Re: Confusion on 2-layer method
|
#3 |
Guest
Posts: n/a
|
First of all thank you Ping for your reply.
You almost addressed the full issue (70%). If I can elaborate more on the problem, maybe you or somebody can shed the rest 30% that I was hoping for. Imagine ONE plate having some parallel flow over its length. Now I have refined the cells near this plate wall according to the 2-layer technique. Next I accidentally assign a small portion (the leading edge) of the wall region with active 2-layer and use the 2-layer model. The rest of the wall was not assigned as active 2-layer by mistake. Next, I ran the model and obtained a 2-layer status plot. What I saw: at the region where the 2-layer was active, the 2-layer plot was having about the right distance as assigned in the 2-layer tab. The rest of the wall did not have a consistent distance from the 2-layer plot: It grew as would a boundary layer grow! I think this was great! Or maybe not necessary! (I was expecting some error because for a very refined cells along the entire length of the wall and for non-activated 2-layer there, the k-E would have been active and hence, produce some erroneous results due to the constraint in wall function but, the 2-layer plot was really intriguing). When I assigned the entire plate with active 2-layer, The 2-layer plot after a run showed that a near-constant distance from the wall. In other words, it did not cover the growing part of the boundary layer near the trailing edge. This growing part is a lot farther from the wall for the first case above. For both cases, the mesh used was the same. So, to rephrase, for 2-layer method, does Star automatically detect boundary layer growth and automatically equate the inner region of the 2-layer to the same exact growth of the BL? Such that when the 2-layer status is plotted, the inner region represents the BL? (In the case of forgotten to activate the 2-layer for the boundary region that is). Tell me if this does not help u understand the problem still. I could be misterpreting something or worst case, this may just be a bug in 3.15...! |
|
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
some confusion about the inner product | ztdep | Main CFD Forum | 5 | January 11, 2006 06:30 |
Some confusion about editing | Harish | CFD-Wiki | 1 | September 21, 2005 00:52 |
3D Cartesian Gov. Eq. confusion | Kay | Main CFD Forum | 0 | August 19, 2004 08:21 |
Little confusion | Cfd beginner | Main CFD Forum | 2 | June 28, 2004 20:22 |
Collocation method for similar boundary layer eqs. | Stefan | Main CFD Forum | 1 | February 15, 2002 10:18 |