The automation of engineering tools has generally not kept up with designers’ needs. As a result, engineers typically start with a conceptual design, then simulate the design, refine it to evaluate a change, and re-simulate it. This tedious iterative process is time consuming and inefficient, and while it may eventually converge on an adequate design, it rarely results in an optimal one.
Diabatix has solved this issue by developing a new methodology that integrates analysis methods and refines the design autonomously. The new methodology integrates heat, flow, and form optimization into an automatic algorithm that designs and optimizes the cooling component. The user enters thermal requirements into a web-based interface and initiates the analysis, and a supercomputer performs the analysis remotely, returning results to the web browser. The web browser shows the progression of the design graphically, enabling the user to watch as the design develops. A video demonstrates the design of a liquid cooling heat sink to illustrate the process.
|
The process begins with an empty cooling platform (courtesy Diabatix.com)
|
Liquid pathways are automatically defined (courtesy Diabatix.com)
|
|
The pathways are optimized by simulation (courtesy Diabatix.com)
|
The final result focuses cooling under the devices where it is most needed (courtesy Diabatix.com)
|
The platform evaluates product manufacturability as well as optimal thermal performance. Each manufacturing method has different constraints that must be taken into account during the design process. For example, machined metal parts have minimum radii for corners due to practical size limitations on the tooling, while casting requires a design with no undercuts so that the part can be extracted from the mold, and 3D printing imposes yet another set of constraints. The platform designs the cooling component to be compatible with the fabrication technology to produce a manufactural design.