|
[Sponsors] |
Convergence issues and non-physical velocity fields |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
November 3, 2021, 23:24 |
Convergence issues and non-physical velocity fields
|
#1 |
New Member
Arturo Alanís
Join Date: Oct 2021
Posts: 9
Rep Power: 4 |
I'm trying to do a steady state simulation of the flow over a 3D NACA2412 airfoil at Re = 400, 000. I'm using a structured D grid mesh around the airfoil which I made in ICEM CFD. I plan on using the solution to initialize an URANS simulation and then use this other solution to initialize an hybrid URANS-LES simulation (this is the reason why I'm using a 3D domain instead of 2D, to be able to capture the 3D characteristics of turbulence). I'm using Menter's k-w SST model to account for turbulence.
The problem I'm facing is I have divergence in continuity residuals whenever I use 2nd order schemes for U, k, w and Nu. Everything blows up at around 30-100 iterations. If I use 1st order schemes it converges up to a certain point, however when I check the solution the velocity and pressure fields look VERY bad and completly non-physical. I attached a picture of the velocity field both in zoom out and around the airfoil. I've tinkered with different solvers and it doesn't seem to help. I've also tried different boundary conditions and the problem persists. I've also tried different configurations of the mesh. All my meshes pass all the checks in checkMesh command (first attempts had cells with very high Aspect Ratio due to a very small first layer around airfoil but I fixed it). I believe my mesh is very good but I still can't get a good solution. I attached a picture of the overall mesh and the boundary layer inflation (looking for y+ = 2.5, 1st layer height = 5.3 x10 ^-5). The checkMesh output is this: Code:
Mesh stats points: 6177168 faces: 18194892 internal faces: 17860782 cells: 6009279 faces per cell: 6 boundary patches: 6 point zones: 0 face zones: 1 cell zones: 1 Overall number of cells of each type: hexahedra: 6009279 prisms: 0 wedges: 0 pyramids: 0 tet wedges: 0 tetrahedra: 0 polyhedra: 0 Checking topology... Boundary definition OK. Cell to face addressing OK. Point usage OK. Upper triangular ordering OK. Face vertices OK. Number of regions: 1 (OK). Checking patch topology for multiply connected surfaces... Patch Faces Points Surface topology outlet 10246 10512 ok (non-closed singly connected) inlet 13019 13344 ok (non-closed singly connected) airfoil 13019 13296 ok (non-closed singly connected) farfield 42112 43104 ok (non-closed singly connected) periodic_1 127857 128691 ok (non-closed singly connected) periodic_2 127857 128691 ok (non-closed singly connected) Checking geometry... Overall domain bounding box (-7.84575217306 -8.00000004259 -2.975397706e-14) (32.1500015259 8.00000004482 3) Mesh has 3 geometric (non-empty/wedge) directions (1 1 1) Mesh has 3 solution (non-empty) directions (1 1 1) Boundary openness (-1.28045831943e-17 -6.02975755126e-16 1.48804105861e-16) OK. Max cell openness = 1.38576603359e-15 OK. Max aspect ratio = 604.605564814 OK. Minimum face area = 1.91070015499e-07. Maximum face area = 0.100408280885. Face area magnitudes OK. Min volume = 1.21959584361e-08. Max volume = 0.00640903919807. Total volume = 1837.26653156. Cell volumes OK. Mesh non-orthogonality Max: 34.1024120585 average: 5.48223143195 Non-orthogonality check OK. Face pyramids OK. Max skewness = 0.275168817637 OK. Coupled point location match (average 0) OK. Mesh OK. Upstream radius = 8m Downstream length = 32m z direction span = 3m airfoil chord = 1m Initial conditions are: k: Code:
dimensions [0 2 -2 0 0 0 0]; internalField uniform 0.0006; boundaryField { outlet { type inletOutlet; inletValue uniform 0.0006; value uniform 0.0006; } inlet { type fixedValue; value uniform 0.0006; } airfoil { type fixedValue; value uniform 1e-20; } periodic_1 { type cyclic; } periodic_2 { type cyclic; } farfield { type freestream; freestreamValue $internalField; } } Code:
dimensions [0 0 -1 0 0 0 0]; internalField uniform 0.024495; boundaryField { outlet { type inletOutlet; inletValue uniform 0.024495; value uniform 0.024495; } inlet { type fixedValue; value uniform 0.024495; } airfoil { type omegaWallFunction; value uniform 0.024495; } periodic_1 { type cyclic; } periodic_2 { type cyclic; } farfield { type freestream; freestreamValue $internalField; } } Code:
dimensions [0 1 -1 0 0 0 0]; internalField uniform (4 0 0); boundaryField { outlet { type inletOutlet; inletValue uniform (4 0 0); value uniform (4 0 0); } inlet { type fixedValue; value uniform (4 0 0); } airfoil { type fixedValue; value uniform (0 0 0); } periodic_1 { type cyclic; } periodic_2 { type cyclic; } farfield { type freestreamVelocity; freestreamValue $internalField; } } Code:
dimensions [0 2 -2 0 0 0 0]; internalField uniform 0; boundaryField { outlet { type fixedValue; value uniform 0; } inlet { type zeroGradient; } airfoil { type zeroGradient; } periodic_1 { type cyclic; } periodic_2 { type cyclic; } farfield { type freestreamPressure; freestreamValue $internalField; } } If anyone is able to help or give me any ideas on what I can do to fix my problem I'll be eternally grateful!! Thank you in advance. Last edited by a.aralnu; November 3, 2021 at 23:25. Reason: duplicate velocity initial BC and missing pressure |
|
November 4, 2021, 02:28 |
|
#2 |
Member
Eren
Join Date: Aug 2018
Posts: 86
Rep Power: 8 |
Your max aspect ratio is 600, that's "pretty" high.
|
|
|
|