# How to create perfect mesh transition from coarse Hexagonal elements to fine mesh?

Member

Hi everyone,

I am testing a structure in ANSYS APDL that includes several cubes attached to each other. Two of these cubes are much thinner (in the middle) and I would like to apply fine mesh on them. However, for the beginning and ending cubes, I would like to have coarse mesh. Please see the attached picture and also you can copy and paste the code to generate this geometry.

/prep7

ET,1,SOLID186

MP,EX,1,2e11

MP,PRXY,1,0.3

BLC4,0,0,1,0.5,0.2

VOFFST,2,0.2, ,

VOFFST,7,0.2, ,

VOFFST,12,0.1, ,

VOFFST,17,0.02, ,

VOFFST,22,0.1, ,

VOFFST,27,0.02, ,

VOFFST,32,0.1, ,

VOFFST,37,0.2, ,

VOFFST,42,0.2, ,

VOFFST,47,0.2, ,

If I want to use hexagonal elements, my ideal mesh transition method is like the following picture:

So, first of all, I apply the fine mesh to my thin cubes

vsel,s,,,5,7,2

esize,0.01

vsweep,all

allsel,all

vplot

vsel,s,,,1,2

vsel,a,,,10,11

vplot

esize,0.1

vsweep,all

allsel,all

So, the final mesh for the selected volumes becomes like the following picture:

So now, here is my question. Does anyone know how I can apply that type of mesh transition (2nd pic) to my model? Because now, the only transition I could apply to this model was using tet elements.

Best,

Ali

• Member

“Perfection is the enemy of progress.” – Winston Churchill

• Member

@peteroznewman Thank you for sending this quote! No doubt it is a great quote. I think I didn't have to use the term "perfect". What actually I meant is how to create this mesh transition without using tet elements, like the second picture.

• Forum Coordinator

@peteroznewman: :-)

@Ali_Safian, The image that you have shared is nice to transition from 4 elements to 2, but it would be possible in some tools such as ICEM CFD, but would be a lot of tedious work as far as I understand,

what is your intent and purpose of creating such fine mesh? why the tet would not work? Why the bonded contacts won't work?

-Aniket

How to access Ansys help links

Guidelines for Posting on Ansys Learning Forum

• Member

@Aniket Thank you for your response! This geometry is continuous structure and I prefer not to consider bonded contact. For the fine mesh, I would like to apply area pressure on those small volumes. What I mean actually is that, consider a roller passing over this structure with a high radial load. Based on Hertz contact theory a pressure distribution and an area is generated at the contact location. But now, I would like to remove that roller from my structure and instead apply area pressure at those small volumes where roller was placed. The thickness of those small volumes is equal to contact width by Hertz contact theory. I know Hertz contact does not have uniform pressure, but for simplification, I want to apply a uniform pressure on my model. This is the concept behind this model.

For tet elements, they do not create a symmetry pressure distribution as the hex elements. Also, I read they are stiffer elements and less accurate compared with Hertz (my knowledge in this sentence is limited).