[AntonZ]

Hi,
I have read lots of things about y+, including ANSYS help, this forum and other CFD-dedicated sites.
But I still have a mess up in my head regarding that value.

"Officially" y+ is a non-dimensional (regarding velocity and viscosity) distance to the wall.
y+ tells me, in which region of the near wall layer I put my first node, if I choose a certain cell size.

So why in Ansys Turbogrid do I set y+ value?

I should calculate an y+ value and then choose a cell size, depending on what I want to model, isn't it?

Or the question here

"Today we will consider three critical questions that are often asked by CFD engineers when developing or refining a CFD simulation:
Do I have an appropriate Y+ value and a sufficient number of inflation layers?"

must be reformulated to "Do I have an appropriate 1st cell size with respect to my y+ value?"
Because for the flow with fixed viscosity, Reynolds, velocity, etc.
y+ is also fixed. Is that right?

It looks like some people define y+ as a dimensionless distance, while others (incl. ansys) treat y+ as a 1st cell size.

[Antanas]

Originally Posted by AntonZ

Hi,
I have read lots of things about y+, including ANSYS help, this forum and other CFD-dedicated sites.
But I still have a mess up in my head regarding that value.

"Officially" y+ is a non-dimensional (regarding velocity and viscosity) distance to the wall.
y+ tells me, in which region of the near wall layer I put my first node, if I choose a certain cell size.

So why in Ansys Turbogrid do I set y+ value?

I should calculate an y+ value and then choose a cell size, depending on what I want to model, isn't it?

Or the question here

must be reformulated to "Do I have an appropriate 1st cell size with respect to my y+ value?"
Because for the flow with fixed viscosity, Reynolds, velocity, etc.
y+ is also fixed. Is that right?

It looks like some people define y+ as a dimensionless distance, while others (incl. ansys) treat y+ as a 1st cell size.

Your choice of y+ value usually depends on turbulence model choosen. Turbulence models have recommended y+ ranges. Next, y+ is related to y (dimensional distance of first mesh node from wall) and flow properties (I think you can easily find formulas). So after you choosed y+ and determined characteristic flow properties you can evaluate y to create your mesh properly.

[AntonZ]

OK, lets be more specific.
I have a centrifugal compressor, RANS + SST.
The initial meshing was done with y+ 200 (Re 1.4e+06).
In real calculation in post processing CFX Post shows me y+ = 830.
Which mesh size should I set next (in a refined grid), if I don't need viscous sublayer.
Should I refine my mesh at all?

[ghorrocks]

You need to do a mesh refinement study and find out. All simulations are different. Some will require a finer mesh, and some will be just fine with large y+ like that.

[Antanas]

Originally Posted by AntonZ

OK, lets be more specific.
I have a centrifugal compressor, RANS + SST.
The initial meshing was done with y+ 200 (Re 1.4e+06).
In real calculation in post processing CFX Post shows me y+ = 830.
Which mesh size should I set next (in a refined grid), if I don't need viscous sublayer.
Should I refine my mesh at all?

I think that if you don't need viscous sublayer your y+ shouldn't be less than 11. Better >30.