merge operation was unsuccessful 2

[wallison.silva]

Hi Guys... Well, I am trying to run a simulation in Abaqus importing iges/step files from another software. However, when I import the files first says that cointain imprecise geometry and I do not know how to fix this, since the geometry is generated in another program (digimat, by the way). When I try to merge the parts abaqus says "merge operation was unsuccessful" and because of that I cannot proceed in my analysis. I am also facing a problem in the mesh part and I think is related to this. I am not an expert in abaqus and I really apreciate it if you could help me to find a solution for this.

Attached, I am sending you a file, maybe you could take a look and help me to find a solution for this. I am a bit desesperated, to be honest.

 

[FEA way]

Try using Geometry Edit --> Part --> Convert to precise. This geometry is quite complex, be careful when meshing it.

 

[wallison.silva]

I have already tried this option in all possible combinations, haha. However the Matrix and Fibers parts does not seem to merge in a way I can generate the mesh. I understand that the geometry is complex, but I am trying to reproduce a result of an author in the same conditions. I really do not know what try anymore.

But thank you very much for the suggestion.

 

[FEA way]

Step format might be better for import, give it a try if you can. Also, be careful because these parts seem to be recognized as, at least partially, hollow.

 

[IceBreakerSours]

I managed to import one of the geometries into a solver agnostic pre-processor (ANSA) without any hiccups. However, as warned previously, meshing this beast is going to be a massive pain - unless you choose of tetrahedrals (lots of them).

*********************************************************
Are you new to this forum? If so, please read these FAQs:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083

 

[wallison.silva]

Hey Man! Thank you very much for your reply!

So, are you suggesting to use a pre-processor before import to ABAQUS?

And yes! I've noticed this in a painful way (haha) and besides, I don't have much experience in the software, but I could learn a lot in the past two weeks. I've used the tetrahedral elements, because they are better to complex geometries, but the problem seems to be in the interaction of surfaces of the fibers and matrix.

When I try to merge intances containing the Fiber and the Matrix, for example, occur this problem. So, Abaqus says that it cannot finish the job because of the error "ErrElemVolSmallNegZero" and "ErrElemDistorted". So, Now, I am trying to look for options to correct these elements, but as you said it is a painful work.

I've looked for options and I found the GMSH, which is a finite element generator, but I don't know if it is possible to import only the mesh. I am up to any suggestions. I understand that is a complex geometry and if there is another smart way do to this I would apreciate it. If you still interested I could give you much more details

Sorry, If I seem stupid, but I've been reading a lot and maybe I am too blind to see the solution.

Thank you again!

################################################

P.S.: Yes, I am new in this forum. Did I do something wrong or is this a normal procedure? haha

 

[FEA way]

External mesher is usually needed only when a hexahedral mesh is absolutely necessary for a very complex shape that can't be handled by mesher in Abaqus. Otherwise, it's typically sufficient to perform meshing directly in Abaqus. This software also offers some tools for geometry preparation and correction. There are virtual topology features, for instance. But you can also remove and replace faces, extract middle surfaces, convert parts to solid and so on. It's often enough to handle even imported geometries that are broken but sometimes corrections have to be done in CAD software, before import.

When it comes to commercial meshers, there's mainly Ansa and Hypermesh. Hypermesh is a preprocessor very often used in conjunction with Abaqus. Open-source meshers include Gmsh and Salome_Meca. Both are very good. Gmsh is used mainly for tetrahedral meshes and is very powerful but has a simple GUI and some operations require editing script files. Salome_Meca offers a complex but unintuitive GUI. Both programs will let you import geometry in STEP format. From Gmsh you can export directly to .inp file while Salome uses .unv format which requires some conversions.

 

[IceBreakerSours]

Here is one thought:

Create a surface triangular mesh on the external surface of the fibers (and then later fill the volume of the fibers with that surface mesh as the starting point for generating tetrahedrals). Use the same external triangular surface mesh on the surface of the fibers as the starting mesh for the matrix and fill it up with tetrahedrals.

Whether you can accomplish this recipe with Abaqus/CAE, I do not know. I haven't used it many years. I think I can do it with one of the more powerful pre-processors but, even with tetrahedrals, meshing this baby is not trivial. And, there will be LOTS of tetrahedrals so the analysis time is going to be excruciating which means iterating through versions of the model will be painful as well. If you had to run some variations, then that adds another layer of complexity. Finally, add the fact that you are new to Abaqus into the mix and the combination starts to look like a brutal battle ahead of you. I would talk to my manager and get the support I need or, more generally, have a proper strategy in place.

("Are you new to .." is simply my signature.)

*********************************************************
Are you new to this forum? If so, please read these FAQs:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083