Racing technology has the reputation of being a rich source of innovations that find their way into commercial passenger vehicles. Fuel injection, turbochargers, and carbon-fiber materials are among the racing-originated developments that make today’s passenger cars more efficient and agile. But, surprisingly, mainstream automotive OEMs might be able to teach racing teams something about vehicle design.
Today’s consumer automotive platforms are engineered with software-based digital tools that automate steps ranging from suspension design to harness configuration while integrating the disciplines of design, manufacturing, and management. In contrast, many racing teams still turn to old-school methods when designing the harnesses and electrical systems of their 360-km/h (224-mph) “products.”
A passenger vehicle may appear in thousands of configurations varying in minute ways, each requiring its own slightly different harness. Options and variants (O&V) dramatically complicate the consumer automotive designer’s challenge. In the racing environment these problems don’t exist. A racecar is essentially a custom, “one-off” product with just one configuration and harness layout.
However, the racing calendar forces teams to complete an entire vehicle in the span of a few offseason months. So O&V complexity is not an issue for racecar harness designers, but the tremendous pressure to meet whirlwind design schedules, combined with weeks spent waiting for physical prototypes to be fabricated, tends to compress the final integration steps into an uncomfortably brief time span.
This same situation is equally applicable to other low-volume harness design realms, from giant mining shovels to seagoing yachts. Designers need tools that give them a competitive edge, whether the challenge is one of complexity or time urgency.
Today, racecar designers are taking a long look at solutions that can expedite their electrical system work. Some teams have begun to adopt software-based automation and modeling tools that improve productivity and enable “scenario-based” design. These same tools also create drawings, track changes, maintain documentation, and improve communication with suppliers. While a harness design application will never circle a racetrack, its contributions might enable timely completion of next season’s championship-winning racecar.
Columns and rows, wires and nodes
Automotive wire harness designers were quick to adopt electronic spreadsheets, particularly Microsoft Excel, to help them keep orderly lists and track every wire and connection in a harness.
But a spreadsheet functions as a purely administrative record-keeping tool in harness design applications. It lacks an integrated way to interact with external applications that can transform stored data points into usable harness drawings. Therefore, racing designers historically have used spreadsheets as a database from which to read out numbers and transcribe wiring details into a drawing. As always, the manual transcription from list to drawing is time-consuming and error-prone.
Integrated tools such as Visual Basic can supplement the basic spreadsheet tools, automating some functions and optimizing the spreadsheet for specific needs. Using these enhancements, engineers can verify point-to-point connections and search for wire gauge errors, providing a limited degree of data validation.
But there remains a barrier between the spreadsheet’s stored data and the harness drawing applications that use it. This makes it difficult to predict costs and gives engineers very little time to create and refine a harness design before creating a physical prototype.
Entering the virtual domain
An entirely different and more modern approach based on simulation with software models—virtual devices—is available, and mainstream automotive OEMs have embraced it almost universally. In this respect, they are ahead of their racing team colleagues who still work from wire lists on spreadsheets.
As delivered in electrical CAD (ECAD) solutions such as Mentor Graphics VeSys 2.0 and CHS, simulation embodies a design environment that resides entirely within a computer and lets engineers quickly create complex system designs and test them nondestructively, without physical prototypes. Instead, virtual—but visual—images allow manipulation and observation of electrical activities.
Functional models are at the heart of the simulation process. A model’s symbol—whether it is a motor, connector, lamp, or wire—is more than just a two-dimensional representation; its electrical and physical attributes are implicit in the symbol. Modeling and simulation are often integrated into a larger process known as model-driven development (MDD), in which high-level system models contain subsystem models. These, in turn, contain component models and so on.
Manipulating models is simple. Moving a symbol such as a light-emitting diode (LED) into an existing wire causes that wire to part automatically and make room for the component. Once the LED is connected, its function as part of the larger system can be tested. Because the design environment knows how the LED works, it is possible to throw a “switch” (yet another model) elsewhere in the circuit and see the LED symbol light up, confirming the correctness of the design. One can imagine designing and validating an entire harness without ever touching a single hardware component.
Models reside in a database accessible throughout an enterprise. Reusable models can be developed entirely in-house, acquired from suppliers or ported from one vehicle platform to another. A motor model, for example, can be version-controlled to ensure that the latest proven version will always be specified for emerging platforms.
Integration and optimization
Not to be overlooked is the prospect of integration with the MCAD tools used by the team’s mechanical designers. ECAD is designed to communicate with MCAD as a design evolves. Ultimately it is possible for the mechanical and electrical engineers to work concurrently on their respective subsystems, tracking each others’ progress and making informed compromises where necessary. Interdisciplinary modeling, as it is called, keeps disparate processes in step and ensures that harnesses and chassis elements will literally fit and work together.
Modeling and simulation also make it possible to try out more new ideas. Until recently, most teams had to go with their first educated guess in hopes of completing the whole vehicle design on time. Design schedules allowed time for one, perhaps two, electrical concepts to be developed before the car launch and the first race of the season.
But successes in the consumer auto sector have proven that the greatest simulation and MDD benefit often comes from trying many alternatives and homing in on an ever-improving design rather than completing one proposal in record time. Automotive designers have discovered that simulation tools can do in days what might take weeks using the traditional methods.
Those newly available weeks offer an opportunity to seek ways to save weight or cost, or even eliminate some electrical elements altogether; the process is called “optimization.”
Breaking and keeping records
Issues such as cost estimation, product documentation, and communication with suppliers can make or break a vehicle design project. In the ECAD domain, documentation is innate and automated. The tools track every component and wire in a harness and know much more than just the electrical performance.
The design application also knows each element’s dimensions and vital statistics. It accrues a detailed database of the quantity of, say, coaxial connectors used in the harness; similarly it captures the length, gauge, and color of every wire in the emerging design.
All this stored data is ready in the form of a bill of materials, a key deliverable of a full-featured electronic modeling toolset. In addition, the toolset creates engineering drawings of the harness, formboard layouts, reports and cost analyses, and information for manufacturing.
Toward the checkered flag
Can electrical simulation and optimization help a racecar win championships? That would be a grand claim indeed. In truth, no amount of design automation can take the place of human creativity, diligence, and insight.
But imagine a world in which engineers have much more time to try out new ideas while still meeting stringent design schedules—more time to validate and test their most ambitious concepts. This is the world that opens up to those who take full advantage of simulation and MDD.
Phil Davies, Product Manager, Mentor Graphics Corp., wrote this article for Automotive Engineering.