[D.B]

Hi,
Apologies in advance for such a basic question. I tried to find an answer on other threads and online, but to no avail. So my doubt is like this :

I think CFD codes in turbulent flows ( with some turbulence model) solve the viscous terms, including molecular viscosity, but I am not sure.

My doubt is like this 1) Is my knowledge on this issue correct
Take the example of a simulation with little or no inlet turbulence value specified
2) Do commercial codes switch -On and switch-off solving of viscous terms depending on the region.
3) How does it effect the calculation of internal flows (say flow in a enclosed chamber) and external flows ( say flow over an aircraft wing). Is there some efficient way by which the codes treat viscous terms to save computational cost.

Thanks

[FMDenaro]

Originally Posted by D.B

Hi,
Apologies in advance for such a basic question. I tried to find an answer on other threads and online, but to no avail. So my doubt is like this :

I think CFD codes in turbulent flows ( with some turbulence model) solve the viscous terms, including molecular viscosity, but I am not sure.

My doubt is like this 1) Is my knowledge on this issue correct
Take the example of a simulation with little or no inlet turbulence value specified
2) Do commercial codes switch -On and switch-off solving of viscous terms depending on the region.
3) How does it effect the calculation of internal flows (say flow in a enclosed chamber) and external flows ( say flow over an aircraft wing). Is there some efficient way by which the codes treat viscous terms to save computational cost.

Thanks

First of all, you should think of what formulation you want to adopt, RANS/URANS/LES are some possibility that a commercial CFD code offers to you

[D.B]

Lets say RANS, but how would it matter if its LES(other than Sub-Grid Scale modelling ofcourse) or URANS ?

[FMDenaro]

Originally Posted by D.B

Lets say RANS, but how would it matter if its LES(other than Sub-Grid Scale modelling ofcourse) or URANS ?

LES and URANS are very different formulations and have very different computational costs

[D.B]

Originally Posted by FMDenaro

LES and URANS are very different formulations and have very different computational costs

I apologise if my answer was a bit misleading, but I know of those(cost) differences, what I meant was, is the treatment of 'molecular' viscosity term different.

My main question still stands, do codes solve the molecular viscosity term at every grid point/region in the flow ?

[FMDenaro]

The molecular diffusion is always solved for turbulent flow, implicit or explicit time integration is used. However, it is of magnitude that is disregardable almost everywhere if compared to the action of the turbulence model.
You need to resolve the boundary layer and that is depending on the grid generation not on the CFD code....