Hi! I'm a beginner CFD user, and I've encountered a strange bug with my mesh. When I reduce the firstLayerThickness to a smaller value, the mesh stops working. The layer doesn't generate properly, and I get an error saying my refineBox isn't inside my blockMesh, which is incorrect.

Hi everyone,

I woud like to export the relative velocity URel magnitude, instead of the absolute velocity U magnitude, in the interfaces files created by TCAE in ../stationnary/postProcessing/interfaceName/0/interfaces.csv. I tried to change the file ./stationnary/system/interfaces. fun by replacing magU by magURel in the list of fields per interface but it did'nt work... How can I do it ?

Thanks in advance !

Zacharie Prigent

Hi Foamers, 

Based on the given instructions underneath, I have been able to come up with this design. Could you help confirm if the attached image to this topic looks promising or requires some modifications? Thanks 😊 

“Introduction

Finishing a small meshing task is the first possibility multi-block mesh using the blockMesh utility or Salome,

e. The task, creation of a small

Geometry

wall

outlet

cut

inlet

Z

Parameters d=50.8 mm

I=508 mm a=12.7 mm

= a/2

Description

The geometry represents half of the 2" pipe inner volume. It is composed of four building blocks. The first block (No. 1) represents the core of the pipe. The other blocks (No. 2, 3, and 4) fill the outside pipe volume.

The mesh blocks must be split by more than one cell in every direction. You can select how many cells you want to use. Apply the mesh grading, if possible.

The outside volume surfaces are called inlet, outlet, wall, and cut. Select an appropriate patch type for each surface.

The blockMesh utility writes out a log. Ensure no errors are reported.

Hint: The local coordinate system in each block should be aligned with the global coordinate system!”

I ma trying to create a thin boundary layer, to achieve a yplus <1 and I need the first cell to be positioned at 10^-5 from the wall of my 3D airfoil. I created the mesh using blockmesh and now I want to use snappyhexmesh only the add layer function.However when I do this snappyhexmesh does not generate the boundary layer. Thank you

Nicolas

I would like to report an issues with SnappyHexMesh when running with OpenFoam v2306. I'm currently try to run SnappyHexMesh with high mesh number structure in HPC cloud environment. Unfortunately, the process of "snappyhexmesh -overwrite" stuck and hanging at refinement phase with message "Found point (-0.001 0.0015 0.0015) in cell 40402 on processor 0". I did run this on serial processing method with VM specification 44vCPU and 315GiB of memory. I already try to run over parallels processing but it's same. Below are sequences of steps I did run and produce the message. Mesh number are define by "blockMeshDict". From my observation, this happens when I'm running with higher mesh number but if I'm lowering the number the process works fine. 

Running steps:

foamCleanPolyMesh

blockMesh

topoSet

createPatch -overwrite

surfaceFeatureExtract

# Run snappyHexMesh before decomposition

snappyHexMesh -overwrite << this steps produce the hanging process.

This is more like an open question than a fully formed topic with both question and answer.

I am running an incompressible laminar flow simulation with the pimpleFoam solver. It has two openings on one side (n1 and n2) and one on the other (n3). I have set up the pressure boundary conditions so that the flow is driven primarily by the pressure applied at n3. The flow should be changing direction halfway. Realistically, the total volume passing through an opening should be the same for the two flow direction. Yet, when I sum over the patch flow rates (postProcessing function) of the two flow directions and compare them, they are different. What could be the potential reason(s) for this phenomenon?

Following is the pressure boundary condition I set up:

/*--------------------------------*- C++ -*----------------------------------*\
  =========                 |
  \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox
   \\    /   O peration     | Website:  https://openfoam.org
    \\  /    A nd           | Version:  10
     \\/     M anipulation  |
\*---------------------------------------------------------------------------*/

FoamFile
{
    version     2.0;
    format      ascii;
    class       volScalarField;
    object      p;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

dimensions      [0 2 -2 0 0 0 0];

internalField   uniform 0;

boundaryField
{
    n1
	{
		type            totalPressure;
		p0              uniform 0;  // Reference total pressure (e.g., atmospheric pressure in Pascals)
		U               U;               // Velocity field name
		phi             phi;             // Flux field name
		rho             none;            // Density, use 'none' for incompressible flow
		psi             none;            // Compressibility, use 'none' for incompressible flow
		gamma           1;               // For incompressible flow, use 1
	}

	n2
	{
		type            totalPressure;
		p0              uniform 0;  // Reference total pressure
		U               U;
		phi             phi;
		rho             none;
		psi             none;
		gamma           1;
	}


    n3
    {
		type		uniformTotalPressure;
	p0
	{
		type		sine;
		frequency	0.3333333333333333;
		amplitude	-50.0;
		level		0;
	}


    }
    walls
    {
        type            zeroGradient;
    }

    defaultFaces
    {
        type            empty;
    }

}

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



1. Public CFD processor benchmark! 💻🚀

   
   

   
   CFDSUPPORT has initiated a benchmark that tests and compares the speed of various processors, providing crucial insights for engineering simulations. 

📊🖥️ Not just that, we're inviting academicians, professionals, and enthusiasts like YOU to contribute! Yes, ALL participants are welcome! 🙌
Your contribution will help us better understand the performance of different CPUs and their role in streamlining engineering simulations. This is a unique opportunity to be a part of a global research initiative, right from the comfort of your home or office. 🌍🏢
For a comprehensive understanding, do check out our full case study 📄 (download option available) at https://www.cfdsupport.com/motorbike-cpu-processor-benchmark.html
Help us create a more informed academic and professional community. Let's benchmark for a better tomorrow! 💪🌐
#OpenFOAM #CFDBenchmark #CPUSpeedTest #EngineeringSimulations #GlobalResearchInitiative 📚🌍🔬🖥️🚀

Hi there

I don't know if tihs would help but I was wondering if anyone can help me to apply a "opening" condition in OpenFoam, in the picture you can see almost a square, and what I have in my problem is that from a little part of the left side I have an inlet and a part of the top part I have an outlet. Also I know that my outlet takes twice the flow that goes in from the inlet, and in the right side its like a room, we have more space, so what I wanted to do is to stablish an opening boundary to make my system have air outside of that, I'm currently using a pressure of 0 and the velocity I don't know what to set, I have set "zeroGradient' "opening" "fixedGradient" "calculatedValue" and nothing work, my simulation always diverges.

I hope someone could help me, 

thanks in advance

Hello everyone,


I have questions about the calculation of circumferential angles circA by TCAE.


I have understood that, if a surface is oriented in the radial direction, circA on this surface will be the angle between the tangent and the velocity in the radial/tangential plane. According to graphical representation of the manual, circA shoud be positiv when the projection of the velocity in the radial/tangential plane points out (of the cylinder in the picture), and negativ when it points in.


Let's imagine that we are on a surface radially oriented where some back flow occurs and that the absolute velocity has a circA > 90° when there is no back-flow. In most of the cells of the surface, circA points out of the cylinder (90° < circA < 180°), but when back-flow occurs, circA points in the cylinder (-180° < circA < -90°). Thus, there are positiv and negativ values of circA at the same time, isn't it ? Therefore, because the circA are either > 90° or < -90°, in the in the interfaces-final.csv file in simulationRun1/TCFD/stationary/postProcessing/surfaceName/0/ folder, it seems to me that surface averaged value of circA is distorted.

I had this reflection because I find particularly high values of circA on some surfaces of my simulations.

Could someone confirm my interpretation ? Or explain me why it isn't true ?

Besides, does someone know how to visualize circA angles on Paraview after a calculation ?


Thanks in advance !


Zacharie

Hi,

I installed OF4Win-20.09v2-win64.exe on my Windows PC. I am interested in using swak4Foam to set specific boundary conditions in my case. Do I need to compile it or what would be the necessary steps?

Thank you.

Hey, I was wondering if anyone could help me to understand the different solvers OpenFoam has for different purposes, I thing I have understood icoFoam and simpleFoam but if anyone could help me to know when and how to use the potentialFoam and all of them I would appreciate it a lot.

Thanks in advance for the help 

Hi, I'm a newbie to OpenFoam and I was trying to do the tutorials, I've been doing great but then I got to the motorbike one and tried to run in parallel the meshing and simulation but didn't work.

As you can see I set my meshing for parallel and gives me errors no matter if I use mpirun or mpiexec (I used both of them because I saw in other forums one of those should work), however if I delete the "-parallel" part it does everything normal.

My computer has 16 cores and the decomposePar works perfectly, it has for subDomains 6 and when i run it it gives me the processor0-processor5 folder

I would appreciate if anyone could help me

Thanks,

Armando