×

Manufacturing

simulation speeds product design

31 Jul, 2003 By: Don LaCourse Cadalyst

---

In today's global economy, manufacturers are under tremendous pressure to reduce costs, increase quality, and deliver new products faster than their competition. Companies are focusing on the early design and development cycle, where engineers and design teams turn specifications into working prototypes, as an area where improvements can help gain a competitive edge. According to industry analysts at Daratech, more than $14 billion was spent in 2002 alone on applications that at least in part target the early design and development cycle (table 1). Recent "Modeling" columns and articles discuss many of the sectors included in the table (see box below).

Additional Reading:

Many of my recent columns and articles discuss application sectors that target the early design and development cycle:
  • “Applying Finite-Element Analysis,” March 2003
  • “Virtual Prototyping Pays Off,” May 2003
  • “Validate Designs with Today’s CAE Options,” July 2002
This month we focus on MCAD simulation, an area that tends to get merged with other topics such as virtual prototyping and analysis.

Bob Ryan, executive vice-president, products, for MSC.Software outlines the differences between virtual prototyping and simulation: "At MSC, we think of both virtual prototyping and simulation as a subset of VPD (virtual product development). Simulation, in MSC vernacular, refers to more traditional CAE-related disciplines such as linear/stress/NVH (noise, vibration, and harshness) analysis, nonlinear and crash analysis, and pre- and postprocessing. Virtual prototyping, again in MSC vernacular, refers to analysis of full-body vehicle prototypes that relates to motion and loads."

Table 1 MCAD sector sales
Sector	2002	2003 forecast
PLM	$7.49 billion (-5%)	$7.84 billion (+5%)
Mechanical CAD/CAM	$4.37 billion (-10%)	$4.51 billion (+3%)
Virtual prototyping and simulation	$1.53 billion (-1%)	$1.64 billion (+7%)
Systems performance modeling	$794 million (+8%)	$865 million (+9%)
Structural analysis	$725 million (-6%)	$766 million (+6%)
Source: Daratech

CAD industry expert Dr. Joel Orr of Cyon Research comments on the distinction between simulation and analysis in "State of the MCAD Industry," Cadalyst, March 2001, p. 24: "In analysis, an explicit understanding of the material or process is used to create a mathematical model to predict behavior. In simulation, the focus is on the ability of the model to provide the same output to a given set of inputs as the thing that is being simulated-without regard for the similarity of the internal model. . . . In other words, accurate and useful simulations do not necessarily employ a closed-form solution for a given model. That means you can simulate poorly understood and even chaotic processes."

Let's take a simulated look at a few applications in MCAD today. Range-of-motion simulation is perhaps most commonly used to validate working assemblies. CAM simulation in virtual manufacturing plans helps validate tool path operations and interactions with the machining center. High-speed simulation predicts and validates the successful deployment of safety devices.

Figure 1. Degrees of freedom refers to the six degrees in which a geometric body is free to move without limiting constraints. There are three linear and three rotational degrees each along and about the x, y, and z axes. A 3D constraint applied to the body eliminates one or more of its degrees of freedom. (Courtesy of the Solid Modeling ExChange)

RANGE-OF-MOTION
Simulation is integrated into most midrange CAD/CAM applications on the market today. It's used primarily to analyze motion. Mark Vorwaller, president and CEO of VX Corp., explains: "MCAD applications have evolved to the point where they do a reasonable job of capturing an accurate 3D representation of an engineer's design intent. This allows engineers to verify form, fit, and basic mechanical properties (for example, volume, area, and mass) before committing to a physical prototype. . . . The next step for MCAD applications is to more effectively integrate these functions with geometric modeling so that they can better influence the initial creation of design geometry."

He points to his company's VX CAD/CAM, where dynamic motion simulation, dynamic interference checking, CNC machining simulation, and a new design optimizer are the beginnings of a series of enhancements aimed at simulation-based design and virtual product testing.

One application for simulation in MCAD is the motion of geometric bodies as they react to a system of 3D constraints and multiple degrees of freedom within an assembly (figure 1). A good example is the body and linkage assembly of the heavy-equipment earth mover modeled in VX CAD/CAM (figure 2). Each body in the assembly has limited degrees of freedom and is linked so that a force applied to one is transmitted through each, resulting in a range of motion that you can simulate, check for interferences, and further evaluate (this sequence is referred to as one iteration). Based on the results, you can make adjustments and repeat the process through multiple

Figure 2. Newly released VX v8 offers dynamic interference checking during component range of motion simulations. You can check for interference among components in an assembly such as this earth mover by simulating their range of motion to verify that all components perform to specification.

iterations until you achieve the desired range of motion.

VIRTUAL MANUFACTURING PLANS
Today's CAM applications also use simulation extensively. CAM operators can create virtual manufacturing plans to prove out toolpath designs before tying up valuable machining center time. The simulation can encompass the entire CAM setup, including the part, stock, table, clamps, fixtures, tool, holders, and attachments. You can perform a rapid simulation using wire frame data or a more precise one with solid stock removal.

Toolpath simulation lets you visually inspect a virtual setup of the machining center. This includes the complete sequence of operations from rough stock removal to high-speed finishing along with the required tool changes. Automatic collision and gouge checking ensures that the tool will not collide with fixtures and clamps or gouge the part being machined.

One benefit of tool path simulation is that you can use it to accurately predict and visually inspect rest material (the amount and location of material remaining after an operation). This provides valuable decision-making feedback for operation planning.

HIGH-SPEED INTERACTIONS

Figure 3. This MSC Dytran OOP (occupant out of position) simulation was created by MSC Software engineers. MSC Dytran is an FEA (finite-element analysis) product that simulates high-speed interactions such as air bag deployment. This product is currently used by automotive manufacturers in North America, Asia, and Europe.

Tier 1 and Tier 2 automotive suppliers who manufacture airbags, steering columns, seats, and passenger restraints use simulation to predict high-speed interactions such as airbag deployment (figure 3). Simulating occupant safety not only speeds up the successful deployment of safety devices, but also allows manufacturers to test new systems and dynamics on virtual models without having to run physical crash tests with dummies, which can cost more than $500,000.

CHALLENGES
Advanced simulation poses unique challenges to application developers. The technical aspects of merging differing CAE technologies and the cultural aspects of once-independent design and analysis groups who must now share information both play a role. Humberto Roa, Centric Innovation product manager, expands on these challenges: "In the past, functional groups in automotive, such as suspension, safety, and electronics, could work independently. Each group would have its own data reference, its own postprocessing tools, and a partial connection to out-of-date geometry. Because of the relative independence of each process, isolation was not a major issue.

Figure 4. Centric Software’s Functional Prototyping solution examines and communicates the effects of design changes by overlaying acoustic and vibration CAE results with detailed chassis and suspension CAD geometry. You can also integrate multiple CAE results with CAD geometry to investigate new suspension systems. Centric software can combine rigid body analysis that describes the rocking and motion of the train with the detailed geometry of the chassis to measure collisions within the train system and with environment features such as tunnels and station platforms.

Today, the increased use of electronic networks makes these same functional groups very dependent on each other."

To design and analyze subsystems such as ABS (automatic braking systems), SRS (supplemental restraint system) airbags, and active suspensions, he notes, functional groups must have direct and continuous cross-discipline access to the network of systems and information that make up the vehicle (figure 4).

Additional simulation challenges include:

• Maintaining interoperability with both CAD and CAE applications.
• Bringing disconnected analysis information together.
• Taking advantage of the latest hardware price/ performance benefits and expanding to different operating systems and configurations.
• Data management needs to become light-years more robust, and products must all have data management interface capabilities.

CRITICAL ISSUES
Because simulation applications must integrate or interface with other MCAD and CAE applications such as solid modeling and analysis, interoperability and data management are two of the critical issues facing developers and users alike.

Bob Ryan of MSC Software says: "Data management is really the most critical issue facing big manufacturers today. For decades, they have created terabytes of critical product performance-related data, but haven't been able to manage it and use it to their advantage.

In this article

  • Centric Software (Functional Prototyping)
  • MSC.Software (MSC Dytran)
  • VX Corp. (VX CAD/CAM)
This will be a critical application area in the next few years and a large growth sector in the market for companies that can offer combined software, services, and systems packages. Interoperability is always a critical issue, and is becoming increasingly so as more simulation is done earlier in the design process. This requires that the tools used by designers and analysts be interoperable, and also better overall communication mechanisms. This is making process-related services engagements a growing opportunity at many manufacturers."

SHORT-TERM TRENDS
Both Roa and Ryan agree that users are looking for interoperable toolsets and cross-discipline integration that can perform a number of different types of analysis without making them jump between applications. Manufacturers want to do linear, nonlinear, NVH, motion, crash, and so forth, all from one environment with no data transfer or geometry cleanup.

Both commercial and proprietary tools that were developed from a single-user perspective must be more open, integrated, and interoperable. Users are also demanding the ability to integrate motion and vibration studies, states and signals of electronic controls, and other capabilities in order to optimize their products.

Developers are working to provide customers with more flexible licensing options beyond the traditional annual lease or paid-up perpetual license. These include campus licenses based on pay-per-use tokens and lease-to-own options. These license trends make it more affordable for companies to get up and running quickly.

IN THE LONG RUN
Further down the road, expect to see these developments:

• Consolidations will occur as larger vendors buy out smaller niche players.
• Large multinational public companies such as EDS, IBM, and SAP will increase their investments in simulation and design technologies, and new players such as Microsoft and Oracle will enter the market.
• The primary technological focus will be the continued evolution of interoperable applications, especially in relation to coupling motion and linear analysis with physical testing data to enhance correlation and prove model fidelity.
• A single reference of the product will contain a link to all the design, geometry, and analysis information.
• Users will be able to examine the reference with context-specific views to make key trade-off decisions about cost, weight, performance, and other factors.
• The single reference will exist across a network of enterprises, OEMs, and suppliers that are working together to bring new and innovative products to market.
• Each functional discipline will be able to share and access analysis information easily and securely while maintaining intellectual property protection.
• The ability to work in isolation will lose focus as emphasis shifts to collaborative development.