Layers and Layer Factor

[BradSawyer]

I am currenty working on simulating vortex style centrifugal pumps. My question has a few parts. I will try to ask them in a coherent manner.

Firstly, in laymans terms, what do the layers and the layer factor effect? I'm assuming it has something to do with boundry later formation, but I want to make sure I have a grasp on what I am controlling when I play with these numbers.

Second, I have been slowly increasing my mesh and have run into problems with the model diverging. I have found that I can make the models converge by playing with the number of layers and the layer factor. My question is, as you increase mesh density is there any general rule with number of layers and layer factor?? Does the layer factor have to increase or decrease? 

What seems to be happening with my model is a small high velocity pocket forms on the wall of my discharge (I have a long discharge, a few pipe diameters long, as reccomended) and then propogates throuout the whole model and causes it to blow up.

I appreciate any feedback in advance.

Thanks!

Brad

[apolo_vanderberg · Accepted solution]

Layers controls the number of Enhancement layers we generate along the wall

Layer Factor controls the relative thickness of the total number of layers off the wall

We would typically alter these to ensure that we have sufficient elements in the boundary layer to be able to capture the wall effects present.

There is no specific general rule that you should alter these paramaters as you alter the mesh settings globally.

Typically (in general) we want to have 3 layers as a minimum and then ensure that the mesh settings within the domain are appropriately refined to capture the needed gradients.

There is a minimum quality of mesh that will be required to avoid divergence regardless of Layer settings. If you happen to be on the edge of this, then yes adjusting Layer parameters may allow you to converge, or cause divergence.

Go through the Basic Meshing video on  SimTV as this will outline the minimum requirements for the mesh. From here you might still need further refinement to capture the specific physics at hand

[BradSawyer]

Apolo,

Thank you very much for the explaination. I attended an Autodesk webinar yesterday about meshing which also helped clear things up as well.

Brad