Crane Aerospace & Electronics used Flomerics' Flotherm thermal simulation software to develop an unusual heatsink that works in the vertical or horizontal orientation. The heatsink is for a DC power subsystem for a commercial airliner. Developing an entirely new heatsink geometry that would have met the performance requirements would have taken an enormous amount of time using conventional build and test methods. Instead Crane mechanical engineer Mark Resler simulated a range of different heatsink configurations using thermal simulation software. He determined that an unusual design with a plate and standoffs provided the best performance.
Normally, this type of system would use fin type heatsinks on the exterior sidewalls of the enclosure. In this case, however, the default design would work only in the horizontal or vertical orientation but not in both. Resler brainstormed with other Crane engineers and developed six different alternative heatsink configurations. They included: 1) Plate and standoffs; 2) Short fins; 3) Long fins; 4) 45 degree angle fins; 5) Short flanged fins with outer flange; 6) 45 degree angled fins with outer flange. The conventional approach to evaluating these concepts would be to build prototypes and test their performance. But this approach is expensive and time-consuming and would not have met the customer schedule.
Thermal simulation with Flotherm made it possible for Resler to evaluate a wide range of designs quickly and at a relatively low cost. "Flotherm quickly generates a model and makes it easy to change boundary conditions, environmental conditions and gravitational direction," Resler said. "I used these capabilities to model all six geometries at various altitudes and ambient temperatures." Resler used the pattern generator in Flotherm to quickly produce the fins.
Resler ran the simulation results twice for each design, in the horizontal and vertical orientations. The simulation results showed that the plate and standoff design provides the best overall performance for both horizontal and vertical mounting. Resler then proceeded to perform a trade study that involved simulating the performance of the plate and standoff design while varying the length of the standoff.
"Building and testing several different prototype configurations would have had a significant cost and schedule impact," Resler concluded. "Thermal simulation allowed us to optimize the heatsink thermal performance and weight in a very short time period. Thermal simulation also enabled us to optimize the design to improve its performance to a level that would have been impossible to achieve with the build and test method within our schedule constraints."