Formula driven Lattices to orient on curved surface

[Garbean]

Dear PTC-Community,

I want to replace a section of a part with formula driven lattices (Gyroid in the example case).

I also want to try out different shapes for the section to be replaced and have the same surface for each version.

In the three examples below, the shape and position of the section to be replaced is slightly different.

Different versions of slots going trough the part

For each case, I am placing the first unit cell in the top left corner and orient it along the slot as shown here:

Unit Cell orientation and placement

I understand that as the Part is wound in space and does not consist of a straight surface, the unit cell gets "cut" in a different way depending on the position of the part to be replaced on the surface.

Is there is a way to get the same surface for all three different positions? To have the unit cell align itself on a curved surface so to say.

What I did try was to create a cuboid, replace it with lattices and then bend the cuboid manually to conform to the surface. The results were not statisfying.

I hope I am being clear enough with my explanation, if I missed any information please feel free to ask. I am using Creo parametric 9.

All the best and thank you for reading

Garbean

[ProFeature]

Hi @Garbean 

Have you tried creating the lattice using the Flatten Quilt method?

https://support.ptc.com/help/creo/creo_pma/r9.0/usascii/index.html#page/surfacing/surface/About_the_Flatten_Quilt_Deformation_User_Interface.html

[Garbean]

Thank you for suggesting this. I will try it out.

[ProFeature]

Hi @Garbean 

Can you share your model, or a simplified model that demonstrates your need?

[Garbean]

Yes, certainly! Sorry for taking so long. Please find attached the three different bodies side by side that I want to replace with the same lattice structure in an .asm file with the three parts as .prt.

My wish would be, to orient the lattices in such a way, that SURFACE1, SURFACE2 and SURFACE3 turn out the same, even though their geometries vary. I am unsure weather this is an impossible ask. I want the lattice to be oriented along to surface so to say.

Thank you for your replies so far and kind regards,

Garbean

[ProFeature]

Hi @Garbean 

This is what I found in the ZIP you shared, so I don't understand your requirement. Could you please share the file(s) that demonstrate the issue?

[Garbean]

I'm sorry. I seem to have not saved my files properly. I attached the hopefully working MWE.

PS: I only created Surface1, Surface2 and Surface3 to mark which surface I want the Lattice to orient on. Also I used very low resolution on the lattices to keep the file size reasonable.

[ProFeature]

Hi @Garbean 

I just checked, and Flatten-Quilt Deformation does not work with Formula-driven lattices.

Do you have to use formula-driven lattices?

Thanks

[Garbean]

Hi ProFeature,

thank you for taking a look at it. Beam based lattices can also be an option maybe. Would that be possible?

[Michael]

That right. Only certain cell types can be developed into a flat state. Here I used a 2-1/2 D hex cell lattice to create this basketball using flatten quilt. 

 

[ProFeature]

Hi @Michael 

The main problem with using Flatten-Quilt Deformation is the deformation... it distorts the lattice.

[Michael]

First make a surface that is developable. Then flatten it. Thicken the surface to solid. Perform 2d lattice. Unflatten. The order of operations matters. In my basketball example the hex cells were normal to the surfaces after unflatten, not distorted. 

[Michael]

Try using non 2d cell based lattices like stochastic beams. You can assign a shell value in surfaces too. 

[Garbean]

Thank you Michael for your suggestion. How would I go about assigning that shell value in surfaces?

[Michael]

Set the Shell Thickness in the lattice tool itself. This controls the thickness of the outer shell that surrounds the lattice structure. The shell will typically follow the outer surfaces of the model while the lattice fills the interior volume.

[Garbean]

Okay, I understand. However, I want the lattice to penetrate the outer surface as shown in my screenshots. Thank you still for trying to help! I appreciate that.

[Michael]

To avoid dangling beams at the surface is to contain them in a body and use non 2d cells like stochastic and others. These adhere better to surface ends because they are 3d.