Stirling Engine Cools Chips
Using a Stirling engine (which converts heat into useful work) for electronics cooling seems like a perfect fit - at least according to Micro-Star International (MSI). In retrospect you wonder why no one thought of the concept sooner.
Electronics (and specifically chip) cooling has become a limiting factor in the drive for ever more capable (and smaller) computers. With each increase in performance, incrementally more heat is generated, causing chip manufacturers to resort to ever more elaborate thermal management methods. Specially shaped (often finned) heat sinks, made of high conductance materials such as aluminum and copper, are combined with electric fans in an effort to keep temperature sensitive chips working reliably. These methods rely on transferring heat to the surrounding air using a combination of conductive heat transfer (heat sink in contact with the chip) and convective heat transfer (air blown across the heat sink). In extreme conditions some chips are liquid cooled using a system similar to that found in a typical car engine (internal combustion engine). I've also discussed using an Ionic wind for chip cooling in a previous post.
As an aside, Computational Fluid Dynamics (CFD) is playing an ever increasing role in analyzing the effectiveness of electronics cooling methods. A large CFD industry dedicated to electronics cooling applications has evolved to address the specific concerns of heat generation in electronic devices.
A Stirling engine alternately heats and cools a fixed (constant mass) amount of gas (usually air) to extract useful work.
The air within a closed (to the outside atmosphere) cylinder is first brought into contact with a heat source (hot), which causes the gas to expand, forcing a power piston to ascend and transfer power to a flywheel. Once the expansion is complete a secondary loose fitting piston below the power piston descends into the region near the heat source. This secondary motion causes the air to move into the heat sink (cold) region and lose heat, causing the air to contract in combination with the descent of the power piston. The secondary piston ascends to allow air back into contact with the heat source and the cycle repeats.
MSI's novel concept combines a heat-pipe/heat-sink combination with a Stirling engine. The heat pipe and Stirling engine are mounted on the hot chip surface and a fan, driven by the Stirling engine, directs air through the heat sink. This combination makes use of the abundant heat produced by modern chips to power the Stirling engine's fan without using any additional power for electric fans. While adding an electric fan can locally reduce heat in the vicinity of a chip, its addition will also raise the temperature of the overall system due to friction and electrical losses.
A consequence of using the Stirling engine for cooling is that it forms a stable (negative) feedback control loop - where the hotter the chip gets the faster the fan runs.
So finally there might be justification for fitting a window on your desktop computer case, not to see flashing LEDs attached to your cooling fans, but instead to see an energy saving Stirling engine whirling away.
Recent blog posts
- Meshing in a Single, Integrated Simulation Environment
- CFD Geometry Creation
- CFD Prefers NURBS Over STL
- DIY CFD
- How to Control Your CFD Simulations in Caedium
- How to Control Your CFD Mesh in Caedium
- Walkalong Tumblewing Meets CFD
- Tumblewing Meets CFD
- Spinning Paper Sheet Meets CFD
- Get Started With CFD