Best procedure for setting mesh density for imported 3D mold assemblies

[piotr.wozniacki]

Here is the best procedure for differentiating mesh densities on the inner and outer surfaces, which works well with a Mold Wizard - generated, simple mold block:

Changing mold internal and external edge lengths for CAD mold models

The part, feed system and cooling channels should all be meshed before you mesh the mold.

1. Make sure that the layers representing the part, feed system, cooling channels and mold are all visible.
2. Using the Ctrl key and the Select tool, rotate the mold and select all 6 faces.
3. Click (Mesh tab > Mesh panel > Density) to open the Define Mesh Density dialog.
4. Set the Global edge length to reflect the mold internal edge lengths, or accept the default value.
5. In the left-hand panel, select the first mold face, then deselect Use global edge length in the Mesh density of selected entity section. This will activate the Target edge length entry box.
6. In the Target edge length box, enter a suitable edge length for the mold external edges. Mold external edges may be as much as 3-4 times as long as the internal edges.
7. Click Apply to accept the new external edge length for that mold face.
8. Repeat steps 5 - 7 for all the remaining mold faces, then click Close.

Now, with complete mold assembiels imported from SAT files, selecting all outside surfaces may not be that easy / obvious. Therefore, I'm trying to use another procedure:

- generate mesh for the mold while stopping after surface mesh is ready (allows me to set a coarse mesh for outer surfaces)

- generate mesh again to create tetras: this time, I'm presented with a couple of parameters, one of them being "Mesh size gradation ratio from internal to external surfaces ". The Help system suggests its value to be of 1.5, but this doesn't work well for me...

Question: which of the above 2 approaches is better suited for 3D molds, imported into AMI for 3D (FEM) analysis purposes?

BTW, the Help system's "show me" video shows yet another method, whereby the mold mesh is generated in a single sstep withpout differentiating its inside/ouside densities; while the most straight-forward, this method tends to produce enormous number of tetras: for a 4-cavity mold with some 1.5 million tetras (in parts + cooling system + runner system), over 10 million mold tetras can be generated this way!

[piotr.wozniacki]

OK - after some more experiments, I find that setting the Mesh size gradation ratio from internal to external surfaces parameter to 2 works well, with the inner / outer min. edge length equal to global edge / 2x global edge length, respectively.

This gives fine mesh on the cavity interface, while keeping an acceptably low number of mold tetras.

Piotr

[piotr.wozniacki]

I still have some doubts about the optimum mesh edge sizes for internal vs. external surfaces of the mold, imported from a CAD system...

Yannick Moret in his presentation on 2012 new features says:

§External mold surface versus internal mold surface

          §A ratio of 3 should be a good compromise for mesh quality and number of elements

- and I'd agree that 3 would be nice, and result in a much more manageable number of mold tetras. However, the  size gradation ratio from internal to external surfaces parameter in the Mold (3D) mesher only allows values up to 2.

Why such discrepancy, Yannick?

Thanks,

Piotr