|
[Sponsors] |
April 26, 2011, 19:23 |
y+ confusion
|
#1 |
Senior Member
Join Date: Nov 2010
Location: USA
Posts: 1,232
Rep Power: 24 |
Alright, I'm officially confused on y+ values. Here we go:
I understand what y+ is (I think so, at least), and I understand how and why it works for the transitional and turbulence models. Things get messed up in my head though when I work on real meshes instead of just reading papers. I've got an external automotive car simulation, with ke model, so y+ should be between 30 and 500, but close to 30 is better. When I plot y+ to check the grid after running a simulation for a while, should I plot only the prism layer and interior layer cap from the car itself? Are these the only y+ values I care about since they're the viscous calculation? I don't have any layers grown on the ground. When I look at my y+ plot of the prism layer and interior cap, FLUENT shows all of the prism layer to be y+ of zero - what does that mean? It shows the interior cap to have nonzero y+, but the values are pretty small, generally around 11. Is my layer too small? I understand how to calculate the first layer size, but calculating to total layer size is a bit of a guess - is there a more succinct way to calculate a total layer size estimate based on flow conditions and/or geometry without iterating? Should I also be looking at the y+ of the entire domain? Any help is appreciated - my y+ plot is also attached. |
|
April 26, 2011, 19:24 |
|
#2 |
Senior Member
Join Date: Nov 2010
Location: USA
Posts: 1,232
Rep Power: 24 |
Apparently the image insert doesn't work. Here's the link:
http://s82.photobucket.com/albums/j2...¤t=y.jpg |
|
April 26, 2011, 22:06 |
|
#3 |
Senior Member
Travis Carrigan
Join Date: Jul 2010
Location: Arlington, TX
Posts: 161
Rep Power: 15 |
Y+ is just the non-dimensional FIRST cell height in the boundary layer. A y+ of 11 puts the first cell center inside the buffer layer, not where you want to be when using wall functions...or resolving the flow for that matter.
I've seen it mentioned that the boundary layer region should be resolved by at least 10 cells with a growth rate no greater than 1.3. This rule of thumb should help you to estimate the total prism layer height for your grid. |
|
April 26, 2011, 22:11 |
|
#4 |
Senior Member
Join Date: Nov 2010
Location: USA
Posts: 1,232
Rep Power: 24 |
Yeah, I know the 11 was wrong, I've adapted the grid to fix it. I was just wondering why the prism layer showed a y+ of 0, and why the interior-10 shows all the y+, especially if y+ is only for the first cell.
Does anyone know of a more concrete method of generating the layer height estimate? I'm using settings near that - 15 layers at 1.2, I think. |
|
April 27, 2011, 07:49 |
|
#5 | |
Senior Member
Raashid Baig
Join Date: Mar 2010
Location: Bangalore, India
Posts: 135
Rep Power: 16 |
Quote:
You can try http://geolab.larc.nasa.gov/APPS/YPlus/ for calculating the height of the first cell. |
||
April 27, 2011, 14:00 |
|
#6 |
Senior Member
Join Date: Nov 2010
Location: USA
Posts: 1,232
Rep Power: 24 |
Thanks cfd_newbie - but I know about that particular grid calculator. It's not so much the first cell height I'm worried about rather the prism layer total height.
|
|
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Sub-domain Confusion | Flaky | CFX | 6 | October 7, 2010 08:40 |
Confusion about the wall oundary conditions. | windhair | CFX | 0 | August 13, 2006 15:18 |
reference pressure confusion | hiren | CFX | 1 | November 23, 2005 10:27 |
Confusion about pressures in CFX5.6 | Forrest | CFX | 3 | January 20, 2004 13:11 |
Avoiding Code Confusion | Jonas Larsson | CFX | 3 | April 18, 2001 04:44 |