Please login with a confirmed email address before reporting spam
Posted:
9 years ago
Dec 23, 2015, 4:51 a.m. EST
It all depends... in particular what features you are looking at. In some cases the difference might not affect some coarse features; in other cases the results won't just be different but will be WRONG!.
Easiest way to answer is to run simulations with both laminar and turbulent models and compare the features you are interested in.
If you want more advice it would help to describe a bit more about your geometry, what other physics are involved, what the purpose of the model is - what features you are interested in. And not least, what scale and Reynolds number you are working at.
It all depends... in particular what features you are looking at. In some cases the difference might not affect some coarse features; in other cases the results won't just be different but will be WRONG!.
Easiest way to answer is to run simulations with both laminar and turbulent models and compare the features you are interested in.
If you want more advice it would help to describe a bit more about your geometry, what other physics are involved, what the purpose of the model is - what features you are interested in. And not least, what scale and Reynolds number you are working at.
Nagi Elabbasi
Facebook Reality Labs
Please login with a confirmed email address before reporting spam
Posted:
9 years ago
Dec 23, 2015, 10:18 a.m. EST
If the flow is turbulent then using a laminar flow model should not converge. If it does then you may be using an excessively coarse mesh, too much stabilization, or loose convergence tolerances. Unsteady vortices should be triggered in the model but they cannot be resolved with any reasonable mesh resolution. That’s one of the things a turbulent model does.
Nagi Elabbasi
Veryst Engineering
If the flow is turbulent then using a laminar flow model should not converge. If it does then you may be using an excessively coarse mesh, too much stabilization, or loose convergence tolerances. Unsteady vortices should be triggered in the model but they cannot be resolved with any reasonable mesh resolution. That’s one of the things a turbulent model does.
Nagi Elabbasi
Veryst Engineering
Please login with a confirmed email address before reporting spam
Posted:
9 years ago
Dec 23, 2015, 12:45 p.m. EST
Thank you all. I have natural convection in oil with low viscosity and Rayleigh number is close to critical value of 10^6. So to be on the safe side should I use turbulent model? I have internal flow and I am using Algebraic yPlus.
Regards,
Akim
Thank you all. I have natural convection in oil with low viscosity and Rayleigh number is close to critical value of 10^6. So to be on the safe side should I use turbulent model? I have internal flow and I am using Algebraic yPlus.
Regards,
Akim
Please login with a confirmed email address before reporting spam
Posted:
9 years ago
Dec 23, 2015, 12:46 p.m. EST
Sorry, Rayleigh number is close to 10^9.
Sorry, Rayleigh number is close to 10^9.