The State of the Toolbox

by Richard Mandel and Gerald Jones

In the last fifteen years of product design, we've witnessed what may have been the most profound change since the advent of mass production and the idea of interchangeable parts. Engineers and designers have a singular machine on their desktop that helps draw, analyze and develop a product, whether it's an injection molded toy car or an injection mold for an automobile component. Files from the computer-aided design can then be passed to machines that produce functional models in less than a half-day, models that, in turn, can be used for everything from design or sales meetings to wind-tunnel testing.

If there is a downside to the surfeit of hardware and software platforms and their improvements over the years, it's their speed of obsolescence, almost equal to that of mil-contract aircraft. It's been a natural progression, actually -- as computers became capable of crunching more numbers, the program designers boosted the capabilities of the software. Most engineering staffs have changed design programs several times since first leaving the Rapidograph-and-plastic-template world. A typical illustration of this could be Leviton Telcom, Bothell, WA -- up to two years ago they had been using a wire-frame modeler that, when originally purchased, was "state-of-the-art." However, they required a third party to add surfaces to the file and output it in .stl format, which added $3,000 and two weeks to the design process, before a prototype could be made. The company invested in SolidDesigner 3D by CoCreate of Ft. Collins, CO, which had evolved to include such features as automatic surfacing, and options that allow the user to specify design intent within assembly modeling. After two weeks of training with the new software, engineering staffers said that their design speed had increased by 75%.

For many, it can be inconvenient to revamp the software every two to three years, in an effort to keep up. The Internet is one factor that will change that -- companies like CADMAX Corp., Baltimore, MD, are beginning to offer their software directly over the Internet. Following are comments from several other company leaders on what may be the next developments.

John Thornborrow,
President, Ashlar Inc.

Ashlar Vellum is our "classic" product, a PC-based, 2D and 3D wireframe and surface modeler. It used presumptive drafting, which allowed the user to concentrate on design rather than driving a CAD package. We now have a solid modeling product, which I think is a bit of a misnomer. I like to refer to it as an "integrated design environment," which provides wireframe surfacing capabilities plus solid modeling. It's a direction a lot of companies have moved to with similar products, into a mid-range CAD segment of the mechanical CAD market.

Robot arm designed with CADMAR

The most recent development is the continued functionality increases of the Vellum Solids product. In the last year, the company was repositioned to focus on conceptual design -- initially not usable for full-scale of the mechanical design process, at least not yet. While this addresses the front end of the design process, more and more elements will be added that will support more users in the industrial design area.

We continue to add new surface modeling functionality to the Vellum Solids product. Our strength is in the advanced surfacing capability, which gets back to our focus, conceptual design -- int egrating solids and surfaces in a very stylized product, what we call 'curvy-swoopy,' easily is a real strength. It's becoming much easier for designers to be very creative without need of a high-end design package, allowing mainstream designers to let their imaginations run wild. We use the ACIS kernel from Spatial Technologies, and we use it to its fullest, giving our users a product that maximizes their potential.

The evolution has been from a collection of lines, arcs and circles representing a three-dimensional object, to a 3D analog -- a collection of surfaces and solids representing an object -- to actual objects. There are libraries of standard objects -- say, an ANSI bolt that is a bolt object, with certain stress-load characteristics, certain material characteristics, certain finish characteristics. But I don't think we've reached a stage where the objects are functionality-defined -- say, characteristics that don't just say "This is rubber," but "This has an elasticity of these properties, and if I fill this virtual rubber bladder with water it's going to expand to this point." I think that's a very interesting evolution, an interesting problem to solve, but I don't think that's where the state of the art is today.http://www.ashlar.com
Circle 482.

Frank Perna, Chairman and CEO, MSC Software

Our acquisition of Knowledge Revolution at the beginning of this year, which is now our Working Knowledge division, has furthered our theme of "Simulating Reality." PACTRAN is very powerful for analysis and finite-element analysis, and when you have the capability of doing motion, where the loads are generated, it adds to the equation very nicely. The other, more recent acquisition has added non-linear FEA to the mix, allowing analysis of materials like rubber, plastic, cement and composites.

We have also established our Engineering-e.com division as an application and service provider, which will allow us to deliver software and services over the web. Traditional packages, such as PACTRAN and NASTRAN, and non-traditional services will be made available on an on-demand basis, either by renting, leasing or purchasing. Individuals will also be able to run their problems on our equipment, using our software. We also intend to host other, complimentary products and services on our website. On-demand tools like this reduce the need to purchase an
expensive seat.

As an example, a conveyor sales-engineer goes to a customer and takes measurements, perhaps roughs out a design. He then adjourns to an office for an hour, further develops the design, then goes on-line to our site. He transfers his CAD geometry to us, and gets back a stress analysis, even a simulation to demonstrate to the customer to confirm the performance parameters and how the system will run. And as the bandwidth of the Internet improves, you're going to see even more things like this happening. http://www.macsch.com Circle 483.

Linda Stasko, Corporate Marketing Mgr.
John Hubert, Technical Marketing Specialist
Visionary Design Systems, Inc.

What's really changed in industry over the last twenty years is that whole industries have moved from a production-push model to a customer-pull model. Instead of taking whatever gets built, customers are more likely to compare competitors and try to drive product changes. E-commerce accelerates this trend. While no one buys machine tools or semiconductor fabr ication equipment as they would books from Amazon.com, the e-world is having an impact on how machines are designed, built, sold, installed and supported.

Fusion Deposition modeler by Stratus

Our goal is "Everyone Innovates Faster" -- this goes beyond designing physical products to include things like post-sales support on the web. Paradoxically, some smaller companies -- specialized machinery, for example -- are the quickest to benefit because they're not trying to patch up twenty years of legacy systems. To gain a competitive edge in the Internet Age, companies will need to stream value from engineering to their customers, sometimes directly via the web. Four of these streams would be the proposal stage, the idea refinement stage, the product development stage and the post-production support stage. Our IronCAD product contributes to aiding these because its unique architecture and patented technologies enable all collaborators to share in a concurrent design process without regard to who built the model, how it was created, or which CAD system was used.

In less than five years, the rate of usage for these new e-capabilities (collaboration, supplier integration, outsourcing, etc.) will be at the levels mundane e-mail and web browsing usage is today. Engineering options and decisions on styling, reliability, etc., will be communicated and refined more easily and accurately with closer linkage between engineering and the end customer. This will result in a more active and timely role for the engineer of the future, and will reward those who can best understand and translate the needs and requirements of customers into products. http://www.vds.net Circle 484.

Jon Cobb, Vice
President of Marketing, Stratasys, Inc.

Users of rapid prototyping are always looking for ease-of-use and speed. A year-and-a-half ago we introduced the Quantum machine, with a linear motor drive and a larger build envelope, so we were able to increase the throughput by 5 times. We recently released WaterWorks, which provides a water-based-concentrate support system for our fusion deposition modeling. This goes a long way to putting a product like ours into an office without need for venting and hazardous waste concerns.

The rapid prototyping market is starting to be rather bifurcated -- where you have the 3D printing, and the possibility of getting near-end-use material. We'd like to see 3D printing go the way of plotters, where they went from prohibitively expensive units to one-per-three users. In this way, more people will have access to making quick, on-demand models in the same way one makes a drawing. On the other side of the fence, the higher end type of RP, we look to many more materials to satisfy the engineering or manufacturing need -- perhaps not necessarily producing mass quantity parts, but in an early prototyping stage, to produce test materials like those being used in an automotive or manufacturing industry. It's also a possibility to produce a biocompatible material that could permit the modeling of direct machine-to-patient parts. http://www.stratasys.com Circle 485.