|
[Sponsors] |
[snappyHexMesh] icoFoam trying to understand the process |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
May 23, 2023, 12:48 |
icoFoam trying to understand the process
|
#1 |
New Member
Andre Nel
Join Date: Sep 2022
Location: Skippack PA
Posts: 2
Rep Power: 0 |
Hi All,
I have been wandering around in the openFoam CFD space for a few months, trying to make sense of the process. At this point, I am lost and frustrated. I am trying to simulate an inline gas filter. This filter consists of a housing and a pleated fabric filter. The fabric filter has 50 pleats of 0.23-inch depth. I simulate the porosity by cutting slots in the pleats with an open area equal to 40% of a pleats surface area. Using Fusion 360, I create an stl file. I adjust the stl options until I am satisfied with the mesh. They look pretty good -- fairly uniform grid. At this point, I use the snappyHexMesh GUI add-in for Blender and export the mesh files to the openFoam case. There I run a script that creates the block mesh, extracts the edges, and finally runs snappyHexMesh with the overwrite option. What was a fairly nice mesh has become a mess of horribly small cells when examined using paraView. The model I have has 7 components. I looks like the settings of the snappyHexMesh GUI bleed across components and effect one another. A little more about what I am trying to do. I am trying to get the velocity of the flow so that I can calculate the flow rate (SCFH). I set the inlet and outlet pressures to fix values. I am using Imperial units. One of the things I noticed yesterday was the with the units set to Imperial I had to put in a scaling factor to make the measured dimensions of the filter to come out right in Blender (Blunder?). Anyhow, I am using icoFoam. What happens is that the Courant Numbers and the residuals increase rapidly and the simulation crashes with a floating point error. Not surprising since the mesh is so fine. I am shooting for a Courant number equal to around 0.7, at a velocity of 10 m/s (394 in/s) with a cell length of around 0.006 m (0.25 in). This means the delta T should be 0.0004 seconds. Unfortunately, my cell length, after snappyHexMesh is way too small. Finally, here are the questions: 1. Are there any recommendations for the settings in snappyHexMesh GUI. Most of the model is pretty course except for the pleated filter fabric. I have tried to reduce if to a mesh surface but could not get is to Castelate. 2. Can I create a mesh in the CAD software by subtracting my filter assembly from a bar, taking the interior portion and using that in the icoFoam simulation? |
|
Tags |
blender, icofoam, snappyhexmesh gui |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Not isotropic phenomenon during natural convection process | 4513645ygq | Main CFD Forum | 0 | July 28, 2020 10:59 |
Process Killed icoFoam | Akshay_11235 | OpenFOAM Programming & Development | 1 | April 3, 2020 06:17 |
Problem running in parralel | Val | OpenFOAM Running, Solving & CFD | 1 | June 12, 2014 03:47 |
error 262144 and unexpected disconnect when run on 2 nodes | Emily1412 | SU2 Installation | 1 | April 10, 2014 18:08 |
How to setup a simple OpenFOAM cluster? | TommiPLaiho | OpenFOAM Installation | 3 | October 27, 2013 16:15 |