CAD User Online Article Archive

Case Study

Features and History

From CAD User AEC Magazine Vol 15 No 01 - JANUARY 2002

Sending Features and History along with your model to a design collaborator might imperil your companies knowledge assets - but for some solutions, it is absolutely imperative.

Features and History - everybody wants it, but it could be a double-edged sword - according to John Wedrychowski of Theorem Solutions, the company that supplies data translation software for customers of the major CAD software suppliers - "everybody wants it, but if you pass both your CAD model and its' Features and History to one of your sub-contractors, you could be presenting him with more than you intended - basically, part of your company's intellectual property!" The problem is, however, that models created using the latest solid modelling software cannot be used effectively unless they are accompanied by the history of their creation.

This is discouraging companies within, for instance, the aerospace industry who obviously do not want to pass out data to outside organisations. But even they may have their own design and manufacturing divisions working on different systems, and if they wish to work effectively together, will need to be able to manipulate the geometry of the model as if it was created on their own system, and will need Features and History to accompany shared models.Similarly, within the same company, a model constructed under a dedicated surface modeller may need to be moved to a different system for final assembly, where it also may require the use of Features and History to make changes to the original model.

Talking about exporting sensitive data to external organisations - currently decisions on what should be sent, and how, are made at the sharp end. One designer phones his counterpart, asking for a particular file, which is duly zipped up and transported - along with, quite possibly, some rather unique homegrown expertise. Using Features and History might, at least, serve to remind senior directors of the potential risks involved in arbitrary transmission of company information.

Getting the right tool to tackle a particular job rightly saves considerable time and expense. Taking a sledgehammer to crack a nut is rather wasteful of resources, and this applies to the common practice of sending too much data to the sub-contractor, when only a small element is required.

In many cases, CAD models are exchanged so that companies working within a supply chain can be sure that the component that they design fits in the appropriate space. They do not need to modify the models they receive, and so it is best if they receive an explicit solid or even a set of surfaces which define the space envelope. Some software has been developed specifically for this purpose. The EAI visualisation product for instance, is designed to solve problems of fit and clashes, merely using triangulated geometry with its' visualisation software, rather than the complete geometry. The models, represented in this way are ideally suited for the purpose of checking space and fit. A practical example is fitting out overhead lockers in aeroplane fuselages, where the outside shape has already been strictly defined, and all that remains is for the locker to be designed to fit the shape. CAD models represented in this way are "light" and as a result very large assemblies can be viewed interactively, without processing delays and heavy demands on graphics.

Another common requirement is for the exchange of CAD data for manufacturing, and here again, the needs of the application should dictate the kind of CAD model which is exported or imported. Where a model is being translated so that a machine tool can be used to cut a mold or a die, the controlling element is the shape of the surface. The tool paths are laid down on the surface, the solid geometry, the assemblies and linkages are simply not necessary. More than this, the unwanted geometry can actually inhibit the process. Fully defined CAD model files including solids, surfaces, parameters, assemblies and other attributes are far larger than those which comprise just the surfaces. The bigger the file the more risk that something will not translate, and what a waste of time and effort if it is a solid that will not translate when the application only depends on surfaces.

The Origins of the Problem.

Features and History have only become an issue since the wider use of solid modelling systems which work almost entirely with boolean operations. Original hybrid solid modelling software from Catia and UGS was able to work interactively with wireframes, solids and surfaces. Even now, they work on 'explicit geometry' as well as with solid boolean operations, having access to all the data available within the model. 'Explicit modification' is only really possible with hybrid software solutions such as Catia, UGS and VX. Interestingly, in this day where the solid model is King it is perhaps a reflection of the value of hybrid CAD systems when we see that one of the biggest current aerospace projects (Airbus) is heavily dependent on CADDS, the Daddy of hybrid modellers.

CAD models created in hybrid CAD systems have structures that are well served by current direct translation technology and so with the right software can be simply translated from one system to another with sufficient data to enable the receiving party to complete their aspect of the design project, whether it be further design or manufacturing. The advent of systems almost totally dependant on Boolean operations changed all that, as models were subsequently developed with a history of such operations, and the design couldn't be easily re-used unless the history accompanied the model. Similarly with parametric solid modelling, originating with Pro/Engineer, and continuing with other Parasolid and ACIS software developments, which developed model histories as they built it up from basic blocks.

Theorem Solution's Features and History has been announced as a landmark in the development of CAD interoperability. It stems from the companies long experience in the data translation business, and from a demand from customers to provide a tool that they could use to send Features-based translations to their suppliers - if they required. This new technology is not for everybody, but it will bring big benefits where it is appropriate, particularly where the CAD models being exchanged are not just being used for reference, but need to be modified in the different systems. Being able to turn it 'on or off' provides originators with the ability to protect their intellectual property where they deem it necessary.

Features and History is going to be available first for the 'big four' CAD systems - where most of Theorem Solutions data translaters are currently being used. According to Stuart Thurlby, Theorem's MD, 'although our Features and History translators are new products, they won't be replacing the current CADverter range'. Theorem sees it, instead, as an entirely new line of business. Instead, as market research shows, there is a continual demand for geometry translators and 'our Features and History translators will complement our range, and will all come with the option of direct geometric model output'.

Another problem that has been directly addressed by Theorem Solutions is sending complete and large models to suppliers where smaller and more detailed assemblies would have been quite sufficient.

It is easy to imagine the situation that many designers find themselves in. Pressing ahead with their own work, they get a call from one of their sub-contractors wanting a copy of some part of the model, so that they can design their bits to fit. With no available time to mess about looking for the relevant part of the model the whole assembly is sent instead - all 100Mb of it! The unfortunate designer at the other end has to plough through the whole model to find the relevant section.

Another practical example? Car manufacturers have been known to send complete body shells to manufacturers of door handles or boot lid hinges.

Theorem Solutions has now introduced Data eXchange Navigator (DXN) to enable recipients of data to search complete assemblies for only that particular sub-assembly that they need to translate. The software was used during the design process of a major programme for leading aircraft engine designer and manufacturer Pratt & Whitney.

Data eXchange Navigator software allows engineers to access and integrate product information from different CAD systems and STEP formats, as well as digitally inspect, evaluate and verify the data to determine precisely what needs converting. Allowing access to any associated attribute information, plus, presumably some of the Features and History that we have already covered, DXN integrates incoming information and highlights a hierarchical product breakdown.

Pratt & Whitney, a division of United Technologies Corporation, encountered assembly constraints that highlighted one of the other problems that occurred when exchanging geographical data for a digital mock-up on one of its major engine programs. During the product development process, a Pratt & Whitney customer used a version of CADDS5 and add-on system EPD Connect for integrating and assembling components, while Pratt & Whitney engineers used UNIGRAPHICS software.

“Components were sent individually and due to the incompatibility between the two software systems, so the entire mock-up had to be manually re-assembled, one part at a time into a single large file,” said Pratt & Whitney CAD-System Analyst, David Yamarik. “This data exchange problem required some of our engineers to spend additional hours attempting to re-create the assembly. A faster translation was necessary.”

“We requested that our customer send the EPD Connect file, which outlined component placement, and then utilized Theorem’s Data eXchange Navigator to recreate the entire assembly. We also isolated components that failed to translate, and that resulted in faster problem resolution. With the automated translation process and improved problem interrogation on this particular program, we reduced overall translation time by 50%.”

Pratt & Whitney began working with Theorem’s United Kingdom office two years ago during the development of DXN. The company’s designers evaluated a prototype of the software for four months and provided Theorem with a final analysis. Their input helped Theorem’s engineers enhance the product before its official release.

Theorem has developed Data eXchange Navigator to support the reading and writing of structure, non-graphical attributes and geometry data for CADDS5, CATIA, I-DEAS, UNIGRAPHICS, and STEP AP203

Other Developments.
Theorem Solutions is one step ahead of the game, again, bringing out the only available translation of Catia 5 into UGS. This Catia 5 translator enables bidirectional translation of 3D models between Catia 5 and Unigraphics, and we expect Theorem to announce Catia 5 to I-DEAS and Pro/E translators early (maybe January) in the new year. Don't hold your breath waiting for other Catia 5 translators to appear, as we explained in our last issue that Catia's jump from Version 4 to Version 5 involved considerable rewriting of the Catia base code. CU
www.theorem.co.uk

Case Study

Click here for a Print Friendly Version

The products referenced in this site are provided by parties other than BTC. BTC makes no representations regarding either the products or any information about the products. Any questions, complaints, or claims regarding the products must be directed to the appropriate manufacturer or vendor. Click here for usage terms and conditions.
For Comments towards this website please contact the webmaster