I had to define a refinement point with very fine mesh at the origin of my model to analysis a stress field.
The simulation runs smoothly when a relatively less fine mesh size is defined. But it gets stuck at 10% when a very fine mesh size is defined.
An attachment of the problem is added with the question.
What progress are you watching? Do you think maybe the speed is related to having (0.1/1E-5)^2 = 1E8 or (0.1/1E-5)^3 = 1E12 times as many elements might have something to do with it? In other words, that is either too many elements and it gets stuck, or it takes a long time to create and/or analyze that many elements. (I am not sure how the progress is measured, but it may not be linear. It may be that 10% is "starting something", then it may jump to X% when it gets to the next stage.)
Did you try refinement point sizes of 1E-2, 1E-3, and 1E-4 to see if the stresses were converging? The stress on the aluminum block (part 2) is not going to converge because the corners are an infinite stress riser. All you can do there is ignore the "high" stress value. (You might be able to get a better stress calculation in the titanium block 1, but that is also affected by the sudden discontinuity of the bonding with part 2.)
I would recommend a radius/mesh size no larger than 10 to 20. At a ratio of 100, the mesh time increased greatly. At a ratio of 1E5 like in your original setup, I think the time to mesh the model may be approaching infinity.
Instead of creating one refinement point with a super-small mesh size inside a volume of 1 mm, I suggest that you create multiple refinements, one inside the other, with details like this:
Radius Mesh size
1 0.1
0.1 0.01
0.01 0.001
The other problem that you will encounter when trying to make the mesh too small is that you will not be able to zoom in to see the elements, and panning the model become problematic. There appears to be a limit of about 1000 to 1 on the zoom ratio. When your model is enclosed, the scale ruler was at about 12 mm. Using the mouse wheel to dynamically scroll in, the ruler would not go any smaller than 0.01 mm, and trying to pan the model was difficult. So, I predict that you will not be able to see elements that are 1E-5 mm!
The next problem in the original setup is trying to transition from a mesh size of 1E-5 to 1 mm. Even if the surface mesh could be created, the solid mesh engine is not going to like such a rapid transition. Maybe it can do it, but I did not try. When I meshed your model using the above 3 refinement points, I set the solid mesh quality to "None" just to be sure that the solid mesh could be created. ("Mesh > Mesh > 3D Mesh Settings > Options > Solid > Quality > None")
An image of the results are attached.
______________________________________________________________
If my post answers your question, please click the "Accept as Solution" button. This helps everyone find answers more quickly!