×

Manufacturing

virtual prototyping pays off

30 Apr, 2003 By: Don LaCourse

---

Additional Reading:

Virtual prototyping encompasses many of the CAD/CAM/CAE disciplines that I often discuss in this column and in articles, including 3D parametric feature-based solid and surface modeling and analysis. Refer to the following articles:
  • “When Surfaces Meet Solids,” January 2003
  • “Applying Finite-Element Analysis,” March 2003
  • “Validate Designs with Today’s CAE Options,” July 2002
Virtual prototyping (also called systems performance modeling) refers to the design, simulation, and testing of new ideas, concepts, products, schemes, or processes in a synthetic but interactive computer environment. Systems performance modeling will enjoy vigorous growth of 14% annually over the next five years, topping $1.4 billion in 2007, according to a forecast (figure 1) by market research firm Daratech.

Robert R. Ryan, president and CEO of Mechanical Dynamics (recently acquired by MSC.Software), says we may be reaching levels of diminishing return in applying CAD/CAM/CAE technologies to part design. The big opportunity to increase quality and reduce time and cost has now shifted to the system level.

But what about the three Fs of product design?

• Form. Make it look and feel good.
  
• Fit. Make it easy to assemble within acceptable variances in the manufacturing process.
  
• Function. Make it perform as specified under normal conditions.
  

Aren't the traditional CAD/CAM/CAE disciplines getting the job done?

More significant returns on investment, Ryan says, today can come through the effective use of simulation-based design processes and virtual prototyping applied to system-level design (figure 2). Manufacturers need a means to quickly assess form and fit of entire assemblies of 3D models that define a digital product mockup. They need to assess the operating function of the entire assembled product (functional virtual prototyping), not just component parts.

Virtual prototyping may entail different analysis solutions, depending on the application. The Procter & Gamble Company, for example, uses the ABAQUS FEA application, along with customized programs and

Figure 2. Traditional component-focused CAD/CAM/CAE compared with system-focused virtual prototyping. (Courtesy of MSC Software.)

interfaces, for what they refer to as their VPS (Virtual Packing System) for the structural design of plastic containers. VPS automates all required operations, including preprocessing, job submission, and postprocessing. David Henning of Procter & Gamble says that the automation of analysis routines for designing packages led to the development of the VPS. Through computer automation, Procter & Gamble greatly reduced the time required to do virtual testing of product design and analysis of bottle geometry.

Researchers with Siemens in Munich developed a digital model using Centric Innovation's Functional Prototyping solution. Siemens' engineers use Functional Prototyping to create a complete cross-functional product definition and system-level simulation environment to digitally validate total product functionality during the crucial concept phase. The benefit: planners significantly reduce time-to-market for new subway systems (figure 3).

The implementation of durability analysis at John Deere Welland Works, Welland, Ontario, was a key factor in shortening development time for its rotary cutter systems. On the last two product generations, it cut the cycle from 4-5 years to 1-2 years.

According to Terry Ewanochko, product engineer for John Deere Welland Works, the company used the ADAMS/Durability product to integrate key virtual prototyping techniques such as FEA (finite element analysis), multibody simulation, and fatigue life prediction. Designers generated the initial ADAMS model of the cutter and test fixture in Pro/ENGINEER using MECHANISM/Pro, an embedded product from Mechanical Dynamics. The virtual prototyping procedure mimics a rotating drum test used in physical durability testing. The virtual tests produce stress time-histories that the durability analysis software then uses to predict fatigue life. These predictions have proven very accurate when compared to physical testing.

PROS AND CONS
Like any technology, virtual prototyping has its strengths and weaknesses. As Robert Ryan notes, virtual prototyping can cut costs and increase product quality and manufacturing efficiency. However, these benefits occur only if the process of virtual prototyping works efficiently. Design and analysis are by tradition separate disciplines performed by different professionals. Virtual prototyping must bring these disciplines together (figure 4).

Design geometry must be fed to visualization and analysis applications, and they in turn must provide feedback so that the design can be tweaked.

The key to the iteration process is speed. Designers must make critical what-if decisions quickly. Some vendors provide a suite of integrated applications. Others are entering strategic development agreements, such as the recent Dassault Systemes and ESI Group announcement. However, people and departments by tradition are not integrated.

Also, virtual prototyping models are always simplified representations of the real world in terms of geometry and conditions because an increase in accuracy directly increases time and cost. As such, the results will only be approximate until CPU speeds advance to the point where the software can analyze exact replicas. On the other hand, a quick and approximate analysis can often bring to light design flaws that might otherwise be overlooked.

In this article

  • ABAQUS FEA application
  • Dassault Systemes
  • ESI Group
  • MSC.Software
Virtual prototypes are productive because they can test many variations of a design, including geometry, materials, and conditions, thus providing valuable feedback to the design team. The new role of physical prototypes is to validate the final design of the virtual prototype.

Efficiency also comes into play when you can apply a single validation to a family of parts or when a history of validations leads to a knowledge base of understanding and trust in the virtual prototyping process. Virtual prototyping is also more cost effective for products that require millions of dollars to bring to market and where the risk of failures are high, such as automotive, aerospace, industrial equipment, and weapons systems design.

BOTTOM-LINE BENEFITS
Virtual prototyping can encompass a myriad of disciplines, depending on the manufacturer, the products, and the software vendors involved. They use many different front-end CAD systems for geometry definition and a host of applications to handle model analysis. Companies realize the maximum benefits of virtual prototyping when:

• Various disciplines are tightly integrated.
• Designers can move seamlessly among modeling, meshing, finite-element analysis, process-oriented performance analysis, and design optimization.
• They share design and data models and the results of one analysis are fed directly as inputs to another and then back again.
• Relatively quick what-if decision making occurs within the same user interface.

A big "thank you" to the software developers and manufacturers cited for sharing their insights and first-hand experiences with virtual prototyping.

Glossary of Virtual Prototyping Terms In describing the applications and benefits of virtual prototyping, vendors often use buzzwords and phrases that may not be readily understood even by people well-versed in the world of MCAE (mechanical computer-aided engineering). This glossary should prove helpful. (Courtesy of MSC Software.) Automatic loads calculation When virtual prototyping software simulates the motion behavior of a mechanical system, it automatically computes the reaction forces at each joint for each split second during a simulation. The software can automatically reformat this data as accurate dynamic loading conditions for input into FEA (finite element analysis) software to study stress and strain of individual parts. CAD/CAM Computer-aided design and manufacturing. Software systems enable designers to capture their design intent for mechanical parts or assemblies by graphically describing the precise 2D or 3D geometry of the parts. Synonym: mechanical design automation. CAD mechanism module Optional add-on (or embedded) module that lets you model not just parts, and not just static assemblies of parts, but moving assemblies (mechanisms or mechanical systems) as well. Typically, these modules let you connect parts using joints from an online library, and then articulate, or move, the mechanism in 2D or 3D. With such a mechanism module, you can detect collisions, locate lockup positions, define motion envelopes, and calculate loads at the joints, all without leaving your familiar CAD environment. Design optimization Capability, in virtual prototyping software, to define design objectives, constraints, and variables, and then have the software automatically iterate to the optimal configuration. Design study Virtual prototyping software's ability to select a design variable (say, the length of a link in a mechanical system design), sweep that variable through a range of values, and then simulate the motion behavior of the various designs to understand the sensitivity of the overall system to these design variations. Synonym: parametric design simulation. DOE Design of experiments. A technique complementary to design optimization, DOE is a method for running a statistically significant battery of tests or computer simulations on a design to determine its sensitivity to design and manufacturing variations. Applied with virtual prototyping, DOE automatically produces the arrays of simulation permutations required for a complete experiment. By submitting new design variations to a consistent, standard battery of virtual tests, you can develop sufficiently rigorous knowledge of and confidence in the performance of your system designs to avoid all but a single, perfunctory physical prototype. Dynamics Study of multibody systems (mechanisms, linkages, and other mechanical systems or subsystems) as they undergo large-displacement motion, taking into account the effects of applied and inertial forces. Synonyms: multibody dynamics; mechanical system simulation. FEA Finite-element analysis. Software technique used to study stresses and strains on mechanical parts or components. Virtual prototyping software automatically generates input loads for FEA. In addition, you can import FEA-computed component flexibility ("stiffnesses") into virtual prototyping software to accurately simulate the motion behavior of systems with flexible parts. Flexible-body dynamics Study of the motion behavior of mechanical systems in which some of the components are flexible (that is, not infinitely rigid). Kinematics Study of multibody systems (mechanisms, linkages, or other mechanical systems or subsystems) as they undergo large-displacement motion, not taking into account the effects of applied or inertial forces. Mechanical system Collection of parts or components that are connected to each other by joints and that move or articulate in two or three dimensions. Mechanical system simulation Software-based engineering discipline that entails modeling a mechanical system, simulating and visualizing its 3D motion behavior under real-world operating conditions, and refining and optimizing the design through iterative design studies before building the first physical prototype. Synonym: virtual prototyping. Parametric design simulation See Design study. Virtual prototyping See Mechanical system simulation.