Activity
Mon
Wed
Fri
Sun
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan
What is this?
Less
More

Memberships

CFD Dose

1k members • $99/m

4 contributions to CFD Dose
IDDES simulation in OpenFOAM diverging due to CFL instability
Hi everyone, I’m working on an IDDES simulation in OpenFOAM for a Formula car. My RANS simulation runs fine, and I’d like to use mapFields to take that solution as the initial condition for the IDDES run. From my understanding this should make sense, but please correct me if I’m wrong. I’ve also read that meshes designed for RANS are usually not suitable for LES. Would the same be true for IDDES? If so, what kind of modifications would be necessary? At this point I’m not really sure how to proceed. I’ve tried different setups, but the simulation always blows up. Below are the parameters I’ve modified in case someone can give me some guidance. The main issue I’m facing is with the Courant number. Should it always be < 1 for IDDES? Here’s how I tried to control it in controlDict: runTimeModifiable true; // allow controlDict to be read at each timestep adjustTimeStep true; // allow timestep to change based on maxCo deltaT 2e-6; // initial timestep value, adjustable at each iteration maxCo 1.0; // maximum CFL, deltaT changes based on this maxDeltaT 0.05; // upper limit for deltaT This gives acceptable values at the beginning, but after a while the CFL starts growing uncontrollably and the simulation diverges. Other parameters I’ve set are: LESModel: kOmegaSSTIDDES delta: IDDESDelta In fvSchemes: div(phi,U) Gauss DEShybrid linear // scheme 1 linearUpwind grad(U) // scheme 2 hmax 0.65 // DES coefficient, typically = 0.65 1 // Reference velocity scale 0.028 // Reference length scale 0 // Minimum sigma limit (0-1) 1 // Maximum sigma limit (0-1) 1; // B function limiter, usually 1e-03 div(phi,k) bounded Gauss upwind; div(phi,omega) bounded Gauss upwind; div((nuEff*dev2(T(grad(U))))) Gauss linear; Solver: PIMPLE In fvSolutions (under-relaxation factors): fields { p 0.3; } equations { U 0.7; k 0.7; omega 0.7; "(U|k|omega)Final" 1.0; } Any advice on how to stabilize the simulation or set this up properly would be really appreciated. Thanks in advance!
0 likes • Sep '25
Thank you for this interesting question. If the mesh is problem, you can use face or cell limited grad schemes and add more non orthogonal correctors. If your simulation becomes stable, you may want to change your k and omega to more accurate schemes like limitedLinear. From my experience, if your mesh is too coarse, DES will run in RANS mode rather than diverge.
First Episode of the CFD Dose Podcast!
💻I'm excited to share with you the very first episode of my CFD Dose podcast! I was very pleased to have Jeff Franklin as my first guest to talk about the control volume approach and its connection to co-located grids. 🕸️🌊 If that rings a bell for you, feel free to check this one out ;)
0 likes • Jan '25
Great news, Mo!
Sarah's new job
I am happy to announce that I have started a new job position as Lead Generation Executive at AsRa Tech. As part of this job, I’ll be: ✔️ Researching and identifying new opportunities in the market ✔️ Building connections with potential clients ✔️ Collaborating with the team to drive growth for the company. If you are in the Aerospace, Automotive, or other industry and want to collaborate, feel free to contact me. Here's a link to my LinkedIn ID: www.linkedin.com/in/sarah-mirza123
Sarah's new job
0 likes • Dec '24
Congratulations 🎉 🎉
Have You Confirmed Your Cloud HPC Registration Yet?
If you’re like me, managing hardware upgrades and resource limitations has probably been a constant battle while running high-resolution simulations. That’s where Cloud HPC comes in. It’s designed to handle all the heavy lifting for you—so you can focus on the physics instead of the hardware. 🎁 Here’s what you'll immediately get: - 300 vCPU-Hours for Free to test your cases. - Seamless OpenFOAM integration (and other software too). - Blazing speeds—no joke, my simulations saw up to a 10x speedup! - The new HyperCore Nodes (just launched!) that make even complex cases run like a breeze. 💻 How to Get Started: 1. Register: Create a free account on the platform and claim your free vCPU-Hours. 2. Install CloudHPCexec: Easy download and setup. 3. Upload Your Case: Drag and drop—it’s as simple as that. 4. Run Your Simulation: Watch it fly through the queue and get notified when it’s done. 5. Download Results: Pull the results back to your machine and dive into your analysis. I’ve been using Cloud HPC for a while now, and honestly, it’s hard to imagine going back. Whether you’re running quick studies or tackling complex setups, it’s a total game-changer. If you haven’t registered yet, go give it a shot, and let me know what you think! 👉 Register here and start simulating! Let me know in the comments how your experience goes—looking forward to hearing your thoughts.
Have You Confirmed Your Cloud HPC Registration Yet?
0 likes • Nov '24
@Mohamed Sayed Thank you!
1-4 of 4
Thaw Tar
1
5points to level up
@thaw-tar-2934
Building tools for CFD

Active 108d ago
Joined Nov 19, 2024