-
1
- #1
dar654
Mechanical
- Mar 11, 2003
- 16
I have noticed there are many large companies that still use I-DEAS for FE-modeling/analysis. There are good reasons. The modeler addresses novice, intermediate, and high-end users. Users can use the functionality that meets their individual needs, whether beam, mid-surface, or solid modeling. Modern CAD tools are solid-centric, and the mesh must be solid elements associated with the geometry. I-DEAS does it all, the mesh can be associative, or non-associative. Also, the 3D modeler is extremely powerful relative to competitive products, with his design roots embellished with features useful for analysis (e.g. partition, join partition, to create multiple volumes within the same part -- extremely useful and powerful).
Multiple parts can be displayed on the screen concurrently, rather than one at a time. This has been lost in CAD modelers today, one must put away the current part, create another one separate, put away, then another. I-DEAS allows creation of a complex part, by first constructing multiple sub-component parts on the screen, each precisely located and oriented as the analyst desires. There is no need to create an assembly of parts, and define a multiplicity of relationships to orient each part to another. One can simply sketch-in-place or align the workplane to an existing part face, and then create a new part OR an extrusion to the first part. Competitive FEA tools have primitive 3D modelers, leaving that task to the CAD tools. By having a robust modeler, the analyst can better drive the design, and quickly propose and analyze conceptual changes to optimize structure shapes. More commonly, the analyst asks the designer to provide a new STEP file of all changes; whereas, I-DEAS puts tremendous power into the analyst's hands. Of course, you may often hear, "that's a design job, beneath what an analyst ought do" -- that is wrong.
I-DEAS also offers a simple to use Team Data Manager. This can be used standalone by the user, or set up as a shared database of parts and FE-models. This can simplify replicating an existing FEA model into another model file (and generally breaking TDM associativity during check-in/out for simplicity). Once one has used this function, they would find extremely powerful, and useful.
OK, so why am I rambling onwards. I-DEAS truly needs an upgrade path. I don't mean to NX, with its complexities and data structures (i.e. tree of files in folders to represent a model), limit of one part on the screen at once (requiring assemblies perhaps), or downgrading to FEMAP. Even without major upgrades since I-DEAS 12, the software is still powerful and utilitarian at the same time. There are obvious features that would be nice, such as better meshing to handle bad STEP data to overlook small features, cracks, nearly invisible adjacent vertices. Currently, the user must be given perfect geometry, or the user creates his own perfect 3D geometry (often desirable to rid of tiny design details, and conceptualize for analysis anyway). Other enhancements would include hex-dominant solid meshing, defining more types of contact interfaces between surfaces (fixed, sliding, contact, rotating), disconnecting nodes "within" surface boundaries between adjacent volumes (currently disconnects along boundary, too! not so useful), faster solvers to leverage now common dual-quad processors, etc.
If SDRC (oops, Siemens!) is listening, Siemens should provide an upgrade path for both I-DEAS and FEMAP. Let's call it I-DEAS FX. Retain the I-DEAS Master Modeler and Assembly tools, team data manager, and most the FE tools. Take the best aspects of FEMAP and integrate into I-DEAS FX. FEMAP users could quickly learn the most basic of I-DEAS modeling tools, whether line, surface, or solid primitives to get off the ground, and then learn the easy-to-learn yet powerful solid modeler to advantage. I-DEAS users would finally be able to stay with the product, rather than switch to another FEA tool (Ansys, Hypermesh, Dyna, Comsol, Femap, etc). The large set of organizations still relying on I-DEAS will most likely never switch to NX for their FEA platform. Those who could switch to meet their needs, have already switched. Those left could be recombined into a more powerful I-DEAS/FEMAP product. I-DEAS sites that stopped maintenance would restart maintenance payments. Everyone wins. I'm surprised no one has thought of this, or presented this idea to the user base.
Multiple parts can be displayed on the screen concurrently, rather than one at a time. This has been lost in CAD modelers today, one must put away the current part, create another one separate, put away, then another. I-DEAS allows creation of a complex part, by first constructing multiple sub-component parts on the screen, each precisely located and oriented as the analyst desires. There is no need to create an assembly of parts, and define a multiplicity of relationships to orient each part to another. One can simply sketch-in-place or align the workplane to an existing part face, and then create a new part OR an extrusion to the first part. Competitive FEA tools have primitive 3D modelers, leaving that task to the CAD tools. By having a robust modeler, the analyst can better drive the design, and quickly propose and analyze conceptual changes to optimize structure shapes. More commonly, the analyst asks the designer to provide a new STEP file of all changes; whereas, I-DEAS puts tremendous power into the analyst's hands. Of course, you may often hear, "that's a design job, beneath what an analyst ought do" -- that is wrong.
I-DEAS also offers a simple to use Team Data Manager. This can be used standalone by the user, or set up as a shared database of parts and FE-models. This can simplify replicating an existing FEA model into another model file (and generally breaking TDM associativity during check-in/out for simplicity). Once one has used this function, they would find extremely powerful, and useful.
OK, so why am I rambling onwards. I-DEAS truly needs an upgrade path. I don't mean to NX, with its complexities and data structures (i.e. tree of files in folders to represent a model), limit of one part on the screen at once (requiring assemblies perhaps), or downgrading to FEMAP. Even without major upgrades since I-DEAS 12, the software is still powerful and utilitarian at the same time. There are obvious features that would be nice, such as better meshing to handle bad STEP data to overlook small features, cracks, nearly invisible adjacent vertices. Currently, the user must be given perfect geometry, or the user creates his own perfect 3D geometry (often desirable to rid of tiny design details, and conceptualize for analysis anyway). Other enhancements would include hex-dominant solid meshing, defining more types of contact interfaces between surfaces (fixed, sliding, contact, rotating), disconnecting nodes "within" surface boundaries between adjacent volumes (currently disconnects along boundary, too! not so useful), faster solvers to leverage now common dual-quad processors, etc.
If SDRC (oops, Siemens!) is listening, Siemens should provide an upgrade path for both I-DEAS and FEMAP. Let's call it I-DEAS FX. Retain the I-DEAS Master Modeler and Assembly tools, team data manager, and most the FE tools. Take the best aspects of FEMAP and integrate into I-DEAS FX. FEMAP users could quickly learn the most basic of I-DEAS modeling tools, whether line, surface, or solid primitives to get off the ground, and then learn the easy-to-learn yet powerful solid modeler to advantage. I-DEAS users would finally be able to stay with the product, rather than switch to another FEA tool (Ansys, Hypermesh, Dyna, Comsol, Femap, etc). The large set of organizations still relying on I-DEAS will most likely never switch to NX for their FEA platform. Those who could switch to meet their needs, have already switched. Those left could be recombined into a more powerful I-DEAS/FEMAP product. I-DEAS sites that stopped maintenance would restart maintenance payments. Everyone wins. I'm surprised no one has thought of this, or presented this idea to the user base.