|
[Sponsors] |
February 11, 2021, 10:56 |
Fluid flow around a cylinder
|
#1 |
New Member
Kartikay
Join Date: Feb 2021
Location: India
Posts: 7
Rep Power: 5 |
Dear All,
I am attempting to simulate flow of fluid around a cylinder and understand its behavior with varying Re. I have created a flow field using blockMesh. I get the following error (at time 0.47) whenever I try to simulate using icoFoam. Courant Number mean: 5.54153e+024 max: 2.06512e+029 Generating stack trace... Backtrace: ZN10StackTraceC1Ev [0x705c1465+0x25] module: C:\PROGRA~1\BLUECF~1\ThirdParty-5.x\platforms\mingw_w64GccDPInt32\lib\libstack_tra ce.dll ZN4Foam5error10printStackERNS_7OstreamE [0xeb1c88+0x218] module: C:\PROGRA~1\BLUECF~1\OpenFOAM-5.x\platforms\mingw_w64GccDPInt32Opt\lib\libOpenFO AM.dll ZN4Foam6sigFpe13sigFpeHandlerEi [0xeb2af3+0x33] module: C:\PROGRA~1\BLUECF~1\OpenFOAM-5.x\platforms\mingw_w64GccDPInt32Opt\lib\libOpenFO AM.dll (No symbol) [0x403abd] module: C:\PROGRA~1\BLUECF~1\OpenFOAM-5.x\platforms\mingw_w64GccDPInt32Opt\bin\icoFOam.e xe _C_specific_handler [0x7ff9e8f48048+0x98] module: C:\WINDOWS\System32\msvcrt.dll 0_chkstk [0x7ff9e9a4184f+0x11f] module: C:\WINDOWS\SYSTEM32\ntdll.dll RtlRaiseException [0x7ff9e9a0a889+0x399] module: C:\WINDOWS\SYSTEM32\ntdll.dll KiUserExceptionDispatcher [0x7ff9e9a404be+0x2e] module: C:\WINDOWS\SYSTEM32\ntdll.dll ZNK4Foam9PBiCGStab5solveERNS_5FieldIdEERKS2_h [0xd57292+0x752] module: C:\PROGRA~1\BLUECF~1\OpenFOAM-5.x\platforms\mingw_w64GccDPInt32Opt\lib\libOpenFO AM.dll (No symbol) [0x42a59e] module: C:\PROGRA~1\BLUECF~1\OpenFOAM-5.x\platforms\mingw_w64GccDPInt32Opt\bin\icoFOam.e xe (No symbol) [0x42aca5] module: C:\PROGRA~1\BLUECF~1\OpenFOAM-5.x\platforms\mingw_w64GccDPInt32Opt\bin\icoFOam.e xe (No symbol) [0x42b004] module: C:\PROGRA~1\BLUECF~1\OpenFOAM-5.x\platforms\mingw_w64GccDPInt32Opt\bin\icoFOam.e xe (No symbol) [0x448f42] module: C:\PROGRA~1\BLUECF~1\OpenFOAM-5.x\platforms\mingw_w64GccDPInt32Opt\bin\icoFOam.e xe (No symbol) [0x4013f7] module: C:\PROGRA~1\BLUECF~1\OpenFOAM-5.x\platforms\mingw_w64GccDPInt32Opt\bin\icoFOam.e xe (No symbol) [0x40152b] module: C:\PROGRA~1\BLUECF~1\OpenFOAM-5.x\platforms\mingw_w64GccDPInt32Opt\bin\icoFOam.e xe BaseThreadInitThunk [0x7ff9e7fb7c24+0x14] module: C:\WINDOWS\System32\KERNEL32.DLL RtlUserThreadStart [0x7ff9e9a0d4d1+0x21] module: C:\WINDOWS\SYSTEM32\ntdll.dll I tweaked the solvers and smoothers/preconditioners for U and p and managed to get the solution upto 0.25 seconds (with time step of 0.025). However I am unable to solve at higher time step. I always end up getting 'sigFpe' error either for p solver or U. Also, I am consistently getting 'time step continuity error' while solving for p, intially in the range of e-06 and then suddenly going upto e21. I have checked my mesh and ensured that the geometry is free from negative volumes and parameters like skewness, cell determinant and non-orthogonality are within acceptable limits. Refer below for checkMesh results. Checking geometry... Overall domain bounding box (-25 -25 -1) (50 25 0) Mesh has 2 geometric (non-empty/wedge) directions (1 1 0) Mesh has 2 solution (non-empty) directions (1 1 0) All edges aligned with or perpendicular to non-empty directions. Boundary openness (6.05142e-019 6.08723e-020 -4.91707e-015) OK. Max cell openness = 3.8018e-015 OK. Max aspect ratio = 73.5579 OK. Minimum face area = 0.00125903. Maximum face area = 0.555892. Face area magnitudes OK. Min volume = 0.00125903. Max volume = 0.165066. Total volume = 3749.66. Cell volumes OK. Mesh non-orthogonality Max: 63.426 average: 31.3721 Non-orthogonality check OK. Face pyramids OK. Max skewness = 1.45777 OK. Coupled point location match (average 0) OK. Face tets OK. Min/max edge length = 0.003454 1 OK. All angles in faces OK. Face flatness (1 = flat, 0 = butterfly) : min = 1 average = 1 All face flatness OK. Cell determinant (wellposedness) : minimum: 0.00104199 average: 2.01206 Cell determinant check OK. Concave cell check OK. Face interpolation weight : minimum: 0.400006 average: 0.497922 Face interpolation weight check OK. Face volume ratio : minimum: 0.537888 average: 0.979276 Face volume ratio check OK. Kindly advise. Thank you. |
|
February 11, 2021, 13:58 |
|
#2 |
Member
Chris Harding
Join Date: Dec 2016
Posts: 76
Rep Power: 10 |
Hello,
Please upload your fvSchemes using the code(#) button. Add 4 to your nNonOrthogonalCorrectors because the circular cylinder is causing nonorthogonal issues. Take orthogonality issues into consideration when you complete fvSchemes. References: Harding, Chris (HappyS5). (2020, Dec.) How to calculate predominant shedding frequency in Strouhal Number. cfd-online URL: How to calculate predominant shedding frequency in Strouhal Number Qu, L. ; Norberg, C. ; Davidson, L. (2013) "Quantitative numerical analysis of flow past a circular cylinder at Reynolds number between 50 and 200". Journal of Fluids and Structures, vol. 39 pp. 347-370. DOI: https://doi.org/10.1016/j.jfluidstructs.2013.02.007 Last edited by HappyS5; February 15, 2021 at 09:47. Reason: fvSchemes |
|
February 11, 2021, 17:03 |
|
#3 |
New Member
Kartikay
Join Date: Feb 2021
Location: India
Posts: 7
Rep Power: 5 |
Please refer below for fvschemes.
Thank you for the link of the research paper. Ill share the results after changing the nNonOrthogonalCorrectors to 4 soon. Thank you again. |
|
February 11, 2021, 17:20 |
|
#4 | |
Member
Chris Harding
Join Date: Dec 2016
Posts: 76
Rep Power: 10 |
Quote:
Sorry, I complicated our communication. I should tell you why I said to add 4 to 2 to get 6. An expert, he worked at ESI OpenFOAM for three years, put on a week long seminar and gave great advice, told me that one can use a max of 20 nNonOrthogonalCorrectors safely. His name is Dr. Trushar B. Gohil. So, you could set it at 20, see if it impacts the Courant number, and move forward to minimize the value by backing off to a value that might approach 5-6. . You will likely need to reduce time step. Mine is 0.000005 at Reynolds number 100 for icoFoam. PimpleFoam can automatically adjust Courant so value does not exceed 1. Is Courant number important? Before we speak next, unless you need advice that I can help with, take a look at this VERY useful PDF: http://www.wolfdynamics.com/wiki/tipsandtricks.pdf |
||
February 14, 2021, 04:33 |
|
#5 |
New Member
Kartikay
Join Date: Feb 2021
Location: India
Posts: 7
Rep Power: 5 |
Dear Mr. HappyS5,
Thank you for that pdf document. It was quite helpful. It enabled me to select solution schemes as per the meshing conditions of my geometry. If I may, do you have something that will help me better understand the reasons behind those tips and tricks apart from OpenFoam used guide and the document itself? Thank you again for your help. |
|
February 14, 2021, 11:19 |
|
#6 | |
Member
Chris Harding
Join Date: Dec 2016
Posts: 76
Rep Power: 10 |
Quote:
Take a look at: http://www.wolfdynamics.com/wiki/fvm_crash_intro.pdf for more information on fvschemes. In the above document is an "out of box" setup that is considered the most accurate general setup. One can start from there and change settings. The key is to stay 2nd order accurate, and eliminate as much numerical diffusion, also called false diffusion, as possible. Last edited by HappyS5; February 14, 2021 at 15:10. |
||
February 16, 2021, 09:29 |
|
#7 |
New Member
Kartikay
Join Date: Feb 2021
Location: India
Posts: 7
Rep Power: 5 |
Thank you Mr. HappyS5. That document helped me to better my understanding of solution schemes. I was able to simulate the flow of fluid around a cylinder with your help.
|
|
February 16, 2021, 14:39 |
|
#8 | |
Member
Chris Harding
Join Date: Dec 2016
Posts: 76
Rep Power: 10 |
Quote:
I am happy you were successful. You might want to check your accuracy and final setup by seeing if you match Strouhal number and Coefficient of drag at specified Reynolds number. Last edited by HappyS5; February 16, 2021 at 19:19. |
||
February 16, 2021, 16:51 |
|
#9 |
New Member
Kartikay
Join Date: Feb 2021
Location: India
Posts: 7
Rep Power: 5 |
I'll do that.
Thank you again. |
|
Tags |
icofoam problem, sigfpe, time step continuity err |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Flow past a 2D cylinder - High Re (1E+05) - Cd too high | Pervispasco | OpenFOAM Running, Solving & CFD | 4 | March 14, 2022 03:19 |
Fluid Flow into Still Air from Faucet | bladder1996 | Fluent Multiphase | 2 | January 8, 2018 11:50 |
Strange flow partern (Reverse Flow) in fluid past circular cylinder problem at exit | HectorRedal | Main CFD Forum | 9 | June 9, 2016 19:14 |
Modelling flow around a Smooth Cylinder - Drag coefficient HELP | Asatorae | STAR-CCM+ | 17 | November 14, 2014 11:45 |
Turbulent steady flow around a circular cylinder | Mirek Kabacinski | FLUENT | 0 | July 23, 2003 19:40 |