CFD Online Discussion Forums

CFD Online Discussion Forums (
-   ANSYS Meshing & Geometry (
-   -   [ICEM] Meshing a convergent-divergent nozzle (Laval) with Boundary Layer (

Simser July 20, 2019 06:45

Meshing a convergent-divergent nozzle (Laval) with Boundary Layer
2 Attachment(s)
Hello Everyone,

I already simulated a nozzle with laminar flow and no viscosity but compressible and got good results (correlating with 1D-Flow Theory).
The geometry looks as follows and is extruded 2mm in z-direction (quasi 1D Flow). The Inlet is about 12mm high and it is 400mm long in total. The pressure drop occurs 120mm after the smallest crossection (between the green dots).

Now I want to simulate it with viscosity and friction at the nozzle wall. All other sides are symmetries.
But i dont know how to get an estimate of the Boundary Layer thickness and therefor how thick to make the boundary layer block. How do you guys proceed when meshing such a geometry?
Should I use a block with constant thickness along the whole wall or change the height of my boundary layer block along the x-axis?


Gweher August 8, 2019 14:14

Hello Simon,

I would create a biased seeding along the vertical edges for your BL. In case of turbulent flow you can estimate the first layer height using y+ calculator (

y+ is not typically mentioned in laminar flow but can also can be used if defined properly. For instance you can calculate the max displacement of the boundary layer from the Blasius solution and decide how many points you need in order to cover this. There are some interesting posts covering that topic on this forum, you should definitively have a look at it ;)

All times are GMT -4. The time now is 14:34.