|
[Sponsors] |
Is CFX incapable of modelling a compression ramp in 2D? |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
November 1, 2010, 12:11 |
|
#21 |
Senior Member
Michael P. Owen
Join Date: Mar 2009
Posts: 196
Rep Power: 17 |
Run it in double and see if that crazy artifact goes away.
|
|
November 1, 2010, 17:45 |
|
#22 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,854
Rep Power: 144 |
To me it looks like it is generating a shock wave off the start of the wall. I would move the inlet boundary further upstream so the flow can settle down by the time it reaches the area of interest.
|
|
November 6, 2010, 08:01 |
|
#23 |
Member
Join Date: Feb 2010
Location: Australia
Posts: 65
Rep Power: 16 |
Glen: Doing that allows the simulation to run with no problems at all, but what is causing the shock off the wall in this instance is the specification of a boundary layer profile. This occurs even if the boundary layer profile is generated from the end of a duct flow of the same height by CFX which is really odd.
The problem however is that in order for the simulation to match the experimental conditions as close as possible, the boundary layer profile must be specified. Admittedly, the boundary layer profile I specified did not deal with the viscous and buffer regions on the boundary layer accurately, but even so CFX appears to over ride the boundary layer profile I put in close to the wall anyway (maybe because of this). This I know from a graph of the velocity profile at the inlet. This is the case with the RNG k-e and SST k-w models with a mesh as refined as y+ at the wall of <1. Will try specifying a more accurate boundary layer profile. Something really odd with CFX that I've recently noticed is the lack of a 0m/s velocity at the wall even though it is no slip. The graphics of the flow (a false colours contour plot) does not indicate that this is the case, although a graph or a .csv file indicates otherwise. This is not as big a problem with a more refined (y+ at wall <1) mesh, but the velocity is still 30m/s or so which is wrong. |
|
November 7, 2010, 18:32 |
|
#24 | |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,854
Rep Power: 144 |
Quote:
|
||
Tags |
compression ramp |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
supersonic flow past compression ramp | yapalparvi | OpenFOAM Running, Solving & CFD | 1 | June 14, 2017 12:13 |
Pros and Cons for CFX, CFdesign, COMSOL | Val | Main CFD Forum | 3 | June 10, 2011 03:20 |
OpenFOAM vs. Fluent & CFX | marco | Main CFD Forum | 81 | March 31, 2009 15:22 |
VIV Modelling in FLUENT or CFX? | Sham | Main CFD Forum | 1 | July 4, 2007 05:55 |
CFD Short Course & CFX User Day | Chris Reeves | CFX | 0 | September 11, 2000 09:53 |