Mapped meshes and spheres

[346gnu]

Hello,

A quarter cylinder is easy

But what about a quarter sphere with the top chopped off?

The model really is as simple as it looks.

I can't be looking at the meshing problem in the right way.

Any thoughts

Thanks

[skunks]

example (asm).

regards

paul

[346gnu]

Thanks Paul,

I thought along similar lines but with volume regions in a part rather than an assy of parts. I couldn't get it to work.

It is a reasonable amount of prep.

Consider the following. just spent 10mins including geometry creation and manual shell meshing in Independent and 5 mins to remember my log in and post.

(I'm not suggesting for a minute that the modern user would want to do this, I just thought that since my last foray into this the methods may have moved on)

Thanks

[skunks · Accepted solution]

OK,

example 2 (volume regions only, but sphere with the top chopped off).

regards

paul

[346gnu]

Paul,

That's what I tried !!

I have absolutely no idea what I was doing wrong.

I have taken your model and refined in different ways with volume regions as I thought it mattered whether these overlapped or were individual. It seems I can't break your model.

I will go back and look at my model again.

Thanks for you help

[346gnu]

Paul,

OK, I broke my model straight away.

Have a look at the attached.

A  6deg wedge of sphere rather than a 90deg portion. (6deg because the minimum edge/face angle=5eg)

Only 1 angular subdivision.

Can you spot the deliberate mistake?

Thanks

[346gnu]

Creo 3.0m110

[346gnu]

I think I can repeat the issue.

IF the 2 central wedges are selected such that the

FIRST wedge - pick triangular surfaces

SECOND wedge - does not permit you to pick faces so you have to use datum points

It then refuses to mesh.

Changing the following makes no difference :

wedge angle

angular subdivisions

degenerate form (taking care to change both central wedges)

It is as if the software has got stuck.

Redefine the FIRST wedge by removing vertices and replacing them with datum points

It is still stuck.

But interestingly, the FIRST wedge has forgotten which degenerate it is; a default is not selected

Selecting a degenerate for the FIRST wedge (and making sure this is the same as the SCOND wedge) seems to 'unstick' the software.

Thoughts?

[skunks]

test in creo simulate 3.0 m110

6 deg

regards

paul

[skunks]

Charles,

test 2 as asm (creo simulate 3.0 m110)

regards

paul

[SweetPeasHub]

See my entry for mapping spheres.

https://community.ptc.com/t5/Simulation/Did-you-know-Mapped-Meshing-of-Spheres/m-p/555970#M6812

[skunks]

yes,

creo simulate stand alone (mechanica)  ist fine,

until creo 2.

user lost

regards

paul

[346gnu]

Paul,

it's not a bad thing Independent is now history.

The 'revolve' of shell elements was easy just so long as the user made sure it was only element edges were along the axis of revolution.

Using the Creo functionality we have to approach it differently and are forced to have the cylindrical central volume region; which isn't a problem.

bfn

Charles

[skunks]

Charles,

wedges and tetras are compatible, not bricks.

regards

paul

[sdensberger]

You'll need to slice your volume up, but use the butterfly meshing scheme. The mesh below is all brick elements.