View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0004222 | OpenFOAM | Bug | public | 2025-03-06 09:55 | 2025-03-06 14:07 |
| Reporter | rcarneiro | Assigned To | will | ||
| Priority | normal | Severity | major | Reproducibility | always |
| Status | resolved | Resolution | fixed | ||
| Product Version | 12 | ||||
| Fixed in Version | dev | ||||
| Summary | 0004222: stochasticDispersionRAS | ||||
| Description | I am running a transient isothermal 2D test case (in the xy plane) that resembles air flowing in a ventilated room. In the Lagrangian model, I assume that the particles behave close to a tracer by defining their diameter to be 1e-16; the drag model is the SchillerNaumannDrag, and the dispersion model is the stochasticDispersionRAS. So far, everything seems to be working fine. However, when I checked the tTurbLoc calculation, I noticed something that didn't look right. For a tracer particle, this function (min(k/epsilon, cps*pow(k, 1.5)/epsilon/(UrelMag + small));) should return k/epsilon, but in some regions, I have verified that this doesn't hold. After further analysing the results, I concluded that in the vector dir, defined as: const vector dir(a*cos(theta), a*sin(theta), u); the z-component is present, resulting in a slip velocity. As a result, tTurbLoc retrieves cps*pow(k, 1.5)/epsilon/(UrelMag + small) instead of k/epsilon. I hope it is clear. Thank you | ||||
| Additional Information | As a test I have recompiled the stochasticDispersionRAS by defining the z-component of the vector dir to 0; the results look in better agreement with my expectations. Note that this is just a test case; no validation data is available for particle concentration. | ||||
| Tags | No tags attached. | ||||
|
|
I agree with your analysis. I have changed these models so that UTurb is constrained to the mesh dimensions. https://github.com/OpenFOAM/OpenFOAM-dev/commit/122c68c5dc20a642ef50552df5b3a77eaffef926 As you have not provided a case, however, this involves a bit of guesswork at my end. If the problem turns out not to be fully resolved then you will need to open another report. |
