[musahossein]

Dear all:

I am trying to use probesDict to capture pressure at 0.1m from the bottom of 2D tank (sloshingtank2D, interDymFoam). The tank is 1mX1mX0.1m. Fluid depth is 0.4m. The axis origin (0,0,0) is at midpoint of the tank 0.4m above the bottom (at the water surface). My probesDict file is as follows:

/*--------------------------------*- C++ -*----------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  2.1.1                                 |
|   \\  /    A nd           | Web:      www.OpenFOAM.org                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     2.0;
    format      ascii;
    class       dictionary;
    object      probesDict;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

// Fields to be probed. runTime modifiable!
// To run this file, type "probeLocations" in the solver directory

fields
(
    p
);

// Locations to be probed. runTime modifiable!
probeLocations
(
    (0.0 0.5 -0.3)

);

// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

However, when I run the file, I get the following error:

Time = 0.55

Time = 0.6

Time = 0.65
--> FOAM Warning : 
    From function findElements::findElements(const fvMesh&)
    in file probes/probes.C at line 105
    Did not find location (0 0.5 -0.3) in any cell. Skipping location.

Does anyone have any suggestions as to why this is occuring?

Thanks.

[nimasam]

Did not find location (0 0.5 -0.3) in any cell. Skipping location.

warning is completely obvious, the given point is not in computational domain, omit the point or give the appropriate point position

[musahossein]

Originally Posted by nimasam

warning is completely obvious, the given point is not in computational domain, omit the point or give the appropriate point position

I want to probe at the tank wall, which is at 0 0.5. Even if I put 0.49, I get the same error. Only when I use 0.45 is the output error free.

Any suggestions?

[wyldckat]

Greetings to all!

For reference, I'm coming from Musa's other thread: http://www.cfd-online.com/Forums/ope...tml#post455342

OK, the situation is this: the "probesDict" specifies points in their absolute position in the reference world. This means that they do not move with the mesh itself.

Therefore, there are two possible solutions here:

1. To implement a modified version of the probes utility, which is able to also transform the point positions provided at the beginning, or to use a time dependent point list.
2. Or think about this the other way around: if you can calculate the position of the point you want to track, for every time snapshot, then you can define all of the positions that the point will take during the simulation. This way, you take the final array of values and process it accordingly to the time snapshot it refers to.

For example, if the list of points is something like this:

probeLocations
(
    (0.0 0.5 -0.3) //0s
    (0.0 0.52 -0.31) //0.1s
    (0.0 0.53 -0.32) //0.2s
    (0.0 0.54 -0.33) //0.3s
    (0.0 0.55 -0.36) //0.4s
    (0.0 0.56 -0.37) //0.5s
);
//

Then as the comments are indicating, you use only the values for each probed point for each respective time instance.

Best regards,
Bruno

[musahossein]

Thankyou for your response. However, if I am running a simulation with 4000 time steps, then it would be impractical to try and create a list of coordinates - would you not agree, or is there an easy way of obtaining the probe coordinates as the mesh moves?

Musa

[wyldckat]

Hi Musa,

Then all that is left is for you to take care of the possibility #1:

Originally Posted by wyldckat

1. To implement a modified version of the probes utility, which is able to also transform the point positions provided at the beginning, or to use a time dependent point list.

I don't have the time necessary to implement this myself, so I suggest that you either:

1. Try to do this yourself;
   • edit: Associated to this hypothesis, you can try asking in the Programming sub-forum: OpenFOAM Programming & Development
2. Request this feature at the bug tracker: http://www.openfoam.org/bugs/
3. Request a small support contract to get this implemented in OpenFOAM. Have a look at the section "8.3 Commercial Companies" in the http://openfoamwiki.net website.

Best regards,
Bruno

[musahossein]

Originally Posted by wyldckat

Hi Musa,

Then all that is left is for you to take care of the possibility #1:
I don't have the time necessary to implement this myself, so I suggest that you either:

1. Try to do this yourself;
   • edit: Associated to this hypothesis, you can try asking in the Programming sub-forum: OpenFOAM Programming & Development
2. Request this feature at the bug tracker: http://www.openfoam.org/bugs/
3. Request a small support contract to get this implemented in OpenFOAM. Have a look at the section "8.3 Commercial Companies" in the http://openfoamwiki.net website.

Best regards,
Bruno

I submitted this to the bug tracker:

http://www.openfoam.org/mantisbt/view.php?id=1039

hopefully we will see something soon.

[wyldckat]

Quick update - As I've just reported on the bug tracker:

If I'm not mistaken, this has been fixed in OpenFOAM 2.3: http://www.openfoam.org/version2.3.0...processing.php - and I quote:

probes include appropriate treatment for moving mesh cases in which, by default, probes positions are fixed; for moving mesh cases, it can be useful to fix the probed cell instead by setting the optional fixedLocations entry to false.

[musahossein]

Thankyou very much for your response. I will take a look. I am using the older version of OpenFOAM (2.2.1) not the 2.3 which is now available from the OpenFoam site.

[musahossein]

Originally Posted by wyldckat

Quick update - As I've just reported on the bug tracker:

Is there a "patch" for older versions, such as 2.2.1 which I am running? Just curious. Thanks.

[wyldckat]

I took a better look into this and found out that this was fixed here: http://www.openfoam.org/mantisbt/view.php?id=1090

The respective commit: https://github.com/OpenFOAM/OpenFOAM...f27b5d573c2a9b

To apply this fix, try to see if this works:

cd $FOAM_SRC/sampling/probes/

wget "https://raw.githubusercontent.com/OpenFOAM/OpenFOAM-2.2.x/541b8b0e6498e56728743128a3f27b5d573c2a9b/src/sampling/probes/probes.C" -O probes.C
wget "https://raw.githubusercontent.com/OpenFOAM/OpenFOAM-2.2.x/541b8b0e6498e56728743128a3f27b5d573c2a9b/src/sampling/probes/probes.H"  -O probes.H
wget "https://raw.githubusercontent.com/OpenFOAM/OpenFOAM-2.2.x/541b8b0e6498e56728743128a3f27b5d573c2a9b/src/sampling/probes/probesTemplates.C"  -O probesTemplates.C

cd ..
wmake libso

Best regards,
Bruno

[alia]

Originally Posted by wyldckat

Quick update - As I've just reported on the bug tracker:

Dear Bruno,

is this also the case for foam-extend 3.1 or other versions of foam-extend?

Thank you in advance!