CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Software User Forums > OpenFOAM > OpenFOAM Bugs

ChannelOodles

Register Blogs Community New Posts Updated Threads Search

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   May 2, 2007, 13:03
Default nuSgs boundary condition: c
  #1
Senior Member
 
Maka Mohu
Join Date: Mar 2009
Posts: 305
Rep Power: 18
maka is on a distinguished road
nuSgs boundary condition:

case details: SUSE linux 10.2, OpenFOAM1.3, channelOodles - tutorial case (channel395).

how to reproduce it:

cd channel395;
rm -r 0;
cp -r 0.org 0;
change nuSgs b.c. on the bottom wall to:
bottomWall
{
type nuSgsWallFunctionWrongLetter;
value uniform 1;
}
assume that you had a new boundary condition defined but while typing the name you misspelled. The solver will run and seems to calculate nuSgs base k b.c.; It would be hard to know whether your new b.c. was applied without checking the fields and doing some hand calculation. This is because the code does not check for "type" to be a valid boundary condition, as long as you provide a value.

Best regards,
Maka.
maka is offline   Reply With Quote

Old   May 24, 2007, 06:57
Default correction: ... calculate nuSg
  #2
Senior Member
 
Maka Mohu
Join Date: Mar 2009
Posts: 305
Rep Power: 18
maka is on a distinguished road
correction: ... calculate nuSgs based on k value; ...

Best regards,
Maka
maka is offline   Reply With Quote

Old   May 24, 2007, 07:34
Default There is an option in .OpenFoa
  #3
Senior Member
 
Join Date: Mar 2009
Posts: 854
Rep Power: 22
henry is on a distinguished road
There is an option in .OpenFoam-1.4/controlDict

disallowDefaultFvPatchField 0;

which if set to 1 disallows undefined boundary conditions which is useful for testing.

Henry
henry is offline   Reply With Quote

Old   May 25, 2007, 18:59
Default Thanks Henry.
  #4
Senior Member
 
Maka Mohu
Join Date: Mar 2009
Posts: 305
Rep Power: 18
maka is on a distinguished road
Thanks Henry.
maka is offline   Reply With Quote

Old   July 6, 2007, 13:14
Default paraFoam when disallowDefaul
  #5
Senior Member
 
Maka Mohu
Join Date: Mar 2009
Posts: 305
Rep Power: 18
maka is on a distinguished road
paraFoam
when disallowDefaultFvPatchField is set to 1,

paraFoam will refuse to process any case that uses a user defined wallFunction; This was tested in channelOodles. It only allows built-in boundary conditions. On the other hand OpenFOAM works well after setting such switch to 1. Thanks.

Best regards,
Maka.
maka is offline   Reply With Quote

Old   July 6, 2007, 13:18
Default Yes that is the point of the s
  #6
Senior Member
 
Join Date: Mar 2009
Posts: 854
Rep Power: 22
henry is on a distinguished road
Yes that is the point of the switch, this is not a bug but an important feature.
henry is offline   Reply With Quote

Old   July 7, 2007, 15:48
Default nuSgsWallFunction broken link
  #7
Senior Member
 
Maka Mohu
Join Date: Mar 2009
Posts: 305
Rep Power: 18
maka is on a distinguished road
nuSgsWallFunction broken link with paraFoam:

I got that point from your first message but here is what I mean after I understood it better:

paraFoam does not accept nuSgsWallFunction which is an LES wall function that is distributed with the standard release. May be this is because it is compiled into libincompressibleLESmodels.so together with other LES sub-filter models, which is not felt by paraFoam.

Check the following error message. You can reproduce it by using the wall function with standard channelOodles tutorial. My guess is that paraFoam needs to know about the fact that some b.c. are compiled into libincompressibleLESmodels.so

--> FOAM FATAL IO ERROR : Unknown patchField type nuSgsWallFunction for patch type wall

Valid patchField types are :

28
(
fixedGradient
symmetryPlane
syringePressure
oscillatingFixedValue
freestreamPressure
turbulentInlet
freestream
wallBuoyantPressure
default
empty
fixedFluxPressure
fixedValue
uniformFixedValue
cyclic
mixed
processor
calculated
slip
timeVaryingUniformFixedValue
directionMixed
sliced
partialSlip
outletInlet
inletOutlet
wedge
zeroGradient
totalPressure
pressureTransmissive
)


file: /data/maka/OpenFOAM/run-1.3/channelOodles.01/delete/0/nuSgs::bottomWall from line 34 to line 35.

From function fvPatchField<type>::New(const fvPatch&, const Field<type>&, const dictionary&)
in file lnInclude/newFvPatchField.C at line 115.

FOAM exiting

Best regards,
Maka.
maka is offline   Reply With Quote

Old   July 7, 2007, 16:42
Default I was thinking about a solutio
  #8
Senior Member
 
Maka Mohu
Join Date: Mar 2009
Posts: 305
Rep Power: 18
maka is on a distinguished road
I was thinking about a solution. Not only paraFoam need to know about the names of b.c. (wall functions) defined in libincompressibelLESmodels.so but also any user defined ones. If you read in this posted link in message: Maka Mohu on Thursday, June 21, 2007 - 11:43 am. You will find that LES wall function are special in such a way that one can not use foamUser library to add them, since they will be exposed to solvers which does not need nuSgs (icoFoam, ...). What I came up with is to make a foamUserLESmodelsIncompressible which is linked with libincompressibelLESmodels.so, which worked. As a result if paraFoam is need to be aware of LES wall functions in libincompressibelLESmodels.so, where that latter is linked to foamUserLESmodelsIncompressible.so. This will provide a solution where the precompiled version of paraFoam will work not only with LES standard-release wall functions but with user defined ones also. This solution may reflect my lack of knowledge of openFOAM, but it is the best I came up with.

Best regards,
Maka
maka is offline   Reply With Quote

Old   July 7, 2007, 17:50
Default Why not simply set disallo
  #9
Senior Member
 
Join Date: Mar 2009
Posts: 854
Rep Power: 22
henry is on a distinguished road
Why not simply set

disallowDefaultFvPatchField 0;

in .OpenFoam-1.4/controlDict then paraFoam will post-process BCs it does not know the details of.
henry is offline   Reply With Quote

Old   January 24, 2009, 18:51
Default I recently got a strange error
  #10
Senior Member
 
Maka Mohu
Join Date: Mar 2009
Posts: 305
Rep Power: 18
maka is on a distinguished road
I recently got a strange error from channelOodles tutorial (channel395) in its standard setup. I did not find similar error posted on the forum. I found this error before and when I run the tutorial again I do not get the error in the same time step.


Time = 440
Courant Number mean: 0.0632141 max: 0.336486
BICCG: Solving for k, Initial residual = 0.00724241, Final residual = 4.61516e-06, No Iterations 2
BICCG: Solving for Ux, Initial residual = 0.00863137, Final residual = 7.95612e-06, No Iterations 2
BICCG: Solving for Uy, Initial residual = 0.0497597, Final residual = 7.79092e-07, No Iterations 3
BICCG: Solving for Uz, Initial residual = 0.0526165, Final residual = 8.45029e-07, No Iterations 3
ICCG: Solving for p, Initial residual = 0.140782, Final residual = 9.45507e-07, No Iterations 94
time step continuity errors : sum local = 2.98458e-10, global = -1.88345e-18, cumulative = -3.82092e-15
ICCG: Solving for p, Initial residual = 0.00933368, Final residual = 9.56856e-07, No Iterations 76
time step continuity errors : sum local = 3.02209e-10, global = -1.70725e-18, cumulative = -3.82263e-15
Uncorrected Ubar = 0.1335 pressure gradient = 4.95291e-05
ExecutionTime = 6807.97 s ClockTime = 22454 s


--> FOAM FATAL ERROR : NO_READ specified for read-constructor of object Bmean of class IOobject

From function regIOobject::readStream(const word&)
in file db/regIOobject/regIOobjectRead.C at line 53.

FOAM aborting

Foam::error::printStack(Foam:stream&)
Foam::error::abort()
Foam::regIOobject::readStream()
Foam::regIOobject::readStream(Foam::word const&)
Foam::regIOobject::read()
Foam::regIOobject::readIfModified()
Foam::objectRegistry::readModifiedObjects()
Foam::objectRegistry::readIfModified()
Foam::objectRegistry::readModifiedObjects()
Foam::Time::operator++()
channelOodles [0x419f9c]
__libc_start_main
__gxx_personality_v0



Best regards,
Maka.
maka is offline   Reply With Quote

Old   January 24, 2009, 18:54
Default I noticed that the time step w
  #11
Senior Member
 
Maka Mohu
Join Date: Mar 2009
Posts: 305
Rep Power: 18
maka is on a distinguished road
I noticed that the time step when the solver stops with such error is when the solver are about to write output to the disk.
maka is offline   Reply With Quote

Old   January 24, 2009, 20:00
Default This is a complex file time-st
  #12
Senior Member
 
Join Date: Mar 2009
Posts: 854
Rep Power: 22
henry is on a distinguished road
This is a complex file time-stamp interaction problem that took sometime to fix. Try version OpenFOAM-1.5.x, I am sure that will solve the problem for you.

H
henry is offline   Reply With Quote

Old   February 5, 2009, 19:17
Default I will be very grateful if you
  #13
Senior Member
 
Maka Mohu
Join Date: Mar 2009
Posts: 305
Rep Power: 18
maka is on a distinguished road
I will be very grateful if you could give me a small hint on which files to update between V1.3 and 1.5.x; I know I may be asking too much but I can not go through an upgrade now since, I'm facing a close deadline. Many Thanks.
best regards,
Maka
maka is offline   Reply With Quote

Reply


Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
[Gmsh] ChannelOodles with gmsh lofty OpenFOAM Meshing & Mesh Conversion 7 April 16, 2008 17:44
GradP in channeloodles nikos_fb16 OpenFOAM Running, Solving & CFD 0 September 11, 2007 05:28
GradP in channeloodles nikos_fb16 OpenFOAM Running, Solving & CFD 0 September 10, 2007 10:46
GradP in channeloodles nikos_fb16 OpenFOAM Running, Solving & CFD 1 September 4, 2007 11:52
ChannelOodles in parallel maka OpenFOAM Bugs 3 August 21, 2007 18:30


All times are GMT -4. The time now is 21:32.