[CFD Germany]

Hello everyone,

I do a student research project at the moment. The topic is to calculate the lift and drag-coefficients of a NACA4415 Airfoil with/without winglet and to create the polar-diagramms. I use the k-w model, the freestream velocity is 95 km/h the length of the airfoil 1.5m.

I asked a prof of mine in which way the Y+ Value is important to calculate the coefficients. He meant its ok to have a Y+ between 30-70 or 30-100. We have had a training course where i was said to use a Y+ less than 1. This would be very important.

On the internet I found out that a Y+ less than 1 would be important at heat transfer and the drag-coefficient. Is it really necessary to have such a low Y+ for right results?

Thank You

[colinda1]

Indeed for a correct drag prediction it is important to have low y+ values IF you use a turbulence model without wall functions. But be careful; it is not the only thing that is important. Very often the focus is so much on Y+ that the rest is forgotten. E.g. the expansion ratio should not be too high either.
Also it would be important to have a sufficiently fine mesh in the wake. It would be wise to perform a mesh dependency study.

[CFD Germany]

Thank you for the fast answer. I'm using a turbulence model with wall condition.

I will try to use a Wall function 1 or 2 mm (or should it be less for example 0,5?) thick with 5 Layer. I would choose a lowest thickness round about 0.05mm. That gives me a Y+ < 1 on the Wing Surface.

I have big Expansion ratios from the Wall. I made 3 areas. The inner with 50mm, middle 250 and outher 1250mm, for reasons of better Performance. I could also use more areas to get a lower ratio.

The Flow room is 10 times the lenght of the Wing. Thats important for the Lift Report.

Sent from my D5803 using CFD Online Forum mobile app

[colinda1]

When you write "wall condition" do you mean wall boundary condition?

A wall function is different: in the viscous sublayer it allows to approximate the boundary layer by a function instead of solving the Navier-Stokes equations inside it. It allows to work with higher Y+ values at the wall outside of the viscous sub-layer. Expansion ratios should be really small in the boundary layer (approx. 1.2 to 1.5 would be good)

[CFD Germany]

Yes thats what i meant. I understood it wrong because I was so focused on the mesh.

I will look it up if a wall function is used i'm not sure yet.

So to get a lower ratio i should create a continous change in the mesh size, getting fine at the Wing Surface?

Sent from my D5803 using CFD Online Forum mobile app

[CFD Germany]

Ok i looked it up, i'm using a wall function. Also the graphs of Lift and drag follow the right pattern, so i think its correct, thank you now i understand it better!

Sent from my D5803 using CFD Online Forum mobile app