There is an increasing trend toward use of simulation early in the product development cycle by mechanical design engineers. Yet today only about 30,000 out of over 1 million mechanical design engineers worldwide use computational fluid dynamics (CFD) to simulate the flow of fluids inside and around their products. Engineers give reasons such as “not aware of benefits,” “too difficult,” “not well integrated,” and “too expensive” for why CFD is not more widely used. These limitations are gradually being overcome as flow simulation becomes more embedded in the mechanical design process. The result is that simulation results are becoming available nearly as fast as changes occur in the design cycle so that CFD can play a central role in design decision-making and optimization.
Value of Simulating Early
A recent report by the Aberdeen Group highlighted the value of using simulation early in the product development cycle. The study stated that a broad range of manufacturers are looking to early simulation to arrive at a good design earlier and minimize the time spent in the verification and testing phase of product development. Every single one of the best-in-class performers surveyed by Aberdeen uses simulation in the design phase while only about 75% of laggards do so. This early use of simulation enables those leading manufacturers to reduce the number of prototypes necessary to pass quality tests and avoid unnecessary change orders after design release.
Aberdeen says that simulation is typically embedded within the computer-aided design (CAD) application and transferred from CAD to preprocessor applications. Embedding simulation within a CAD application keeps the engineer in a familiar environment and removes the additional step of transferring geometry to a second application. Best-in-class manufacturers surveyed by Aberdeen are 63% more likely than all other manufacturers to access simulation capabilities directly within CAD applications (36% vs 22%).
Fluid Flow Analysis Challenges
Mechanical and structural analysis are widely employed as simulation tools early in the design process but CFD lags behind. Meanwhile, fluid flow analysis is becoming increasingly important to the performance and manufacturing of a wide range of products. While CFD has long been used to design high value-added products that depend heavily on fluid flow, such as airplanes and automobiles, it has the potential to substantially improve performance of every product that relies upon fluid flow and heat transfer phenomena. CFD uses computers to solve the fundamental nonlinear differential equations that describe fluid flow (the Navier-Stokes and allied equations), for predefined geometries and a set of initial boundary conditions, process flow physics, and chemistry. The result is a wealth of predictions of flow velocity, temperature, and species concentrations for almost any piece of chemical and process equipment. These predictions can be used to improve the design of products and manufacturing processes that involve liquids and gases.
But the Navier-Stokes equations governing fluid flow and heat transfer processes are inherently more complex and nonlinear and hence harder to solve mathematically than the equations governing mechanical stress and solid deformation. Partly for this reason, companies developing CFD software have spent more time and effort in speeding up their mathematical algorithms than embedding their software in the mainstream mechanical design environment. Progress has been made but even most CFD codes that are connected to CAD software take a copy of the 3D geometry, translate it via a neutral format such as Parasolid or ACIS, and add boundary conditions to create a model for fluid analysis that loses intelligence such as assembly hierarchy, constraints, and features.
Because the empty flow space which forms the basis of the fluid model does not exist as a discrete object in the original CAD design, most integrated CFD codes extract all of the cavities from the CAD model, add them to the feature-tree as new objects, and then grid them separately using grid generation software. The data consistency and links to the original CAD model are lost, which makes it impossible to retain model history and parameterization. These traditional approaches are an improvement over traditional stand-alone analysis but still require the user of the software to get involved with the computational side of the analysis to a considerable degree.
This helps explain the results of a recent survey by Flomerics in which engineers were asked why CFD is currently used by so few of them. The largest percentage of respondents, 58.1%, said that most mechanical design engineers don’t have the necessary expertise and knowledge to use a CFD code. The next greatest number, 53.5%, said that most mechanical design engineers are not aware of what CFD can do for them. Then 41.1% of the engineers said that today’s CFD codes are too expensive. Finally, 36.8% of the engineers said that most mechanical designers are too busy doing basic mechanical design and that they don’t have time for CFD analysis. The percentages add up to more than 100% because engineers were allowed to select as many reasons as they wished.
Overcoming the Difficulty Obstacle
The most critical obstacle toward greater use of CFD simulation early in the design process -- high level of expertise required -- is being overcome through the proliferation of a new generation of CFD tools that are fully embedded in mechanical design. This new generation of software operates as an extension of the CAD software, automates routine simulation operations, reduces activities that require specialist knowledge, and enables direct interaction between the CFD software and the native CAD data to keep pace with ongoing design changes.
In this new approach, native 3D CAD data is used directly for fluid flow simulations without the need for translations or copies. The fluid flow software has the look and feel of the mainstream CAD tool and uses the same feature tree and geometry model. All design changes based on simulation results are carried out directly in the CAD system using familiar solid modeling functions. All ancillary data required for flow simulation, such as material properties and boundary conditions, are associatively linked to the CAD model and carried along with all design changes.
Flow conditions are defined directly on the CAD model in the graphics window of the CAD system and organized similarly to other design data in the feature tree. The new generation of CFD software analyzes the CAD model, automatically identifies fluid and solid regions and allows the entire flow space to be defined and gridded without user interaction and without adding extra objects to the CAD model.
Fluid flow simulation in the early stages of the design process provides most benefit when it involves a systematic search for the optimal solution to a design problem. This normally requires simulation of a large number of design iterations that involve changes to geometrical parameters as well as input variables, temperatures, and flow conditions. A modern product lifecycle management (PLM) system provides an ideal platform by parameterizing parts and assemblies so that design changes can be easily made. The newest generation of embedded CFD software stores fluid flow parameters, temperatures, flow rates, etc. as object-based features, managed in the feature tree like other object-based data, and used directly to update the simulation software. This approach enables a large number of model variants to be simulated automatically.
Focus On the Design Rather Than the Software
The new generation of CFD software addresses the major reasons that engineers give for the relative lack of use of CFD software. Its use of native 3D CAD data, automatic gridding of the flow space, and managing the flow parameters as object-based features eliminates the need for engineers to understand the computational part of CFD and instead enables them to focus on the fluid dynamics of the product which is already their responsibility to master. The skills required to operate the CFD software are simply knowledge of the CAD system and of the physics of the product, both of which the design engineer already possesses. The engineer is thus able to focus his or her time and attention on optimizing the performance of the product as opposed to operating the software.
The new generation of embedded CFD software also helps address the concern mentioned by engineers of lack of awareness of the benefits of CFD by making it easy to provide reliable answers and tangible results that lead to improvements in product performance and reductions in time to market.
Embedded CFD software also provides faster and more certain results that make it much easier to justify the cost of CFD to management. For example, consider the case where the customer requires that the flow rate be increased from the current 8 liters per second to 10 liters per second. Without CFD, this may require many iterations of building and testing hardware prototypes to achieve the goal. With CFD, it may be possible to evaluate a dozen software prototypes in a day, making it possible to achieve the required performance at lower cost.
Finally, by integrating CFD into the mechanical design process that engineers are already performing, the new generation of software helps address the engineers who feel they are too busy to take on another analysis tool. All in all, embedding CFD into the design environment enables simulation results to be incorporated into the design process by the right person in the right place at the right time.