Enabling “cooler” electronics
I’ve written in the past that there are a pile of GE products that have thermal challenges, and our research teams have no shortage of ideas for new thermal technologies to solve these problems. Heat pipes are one such example. A heat pipe is a device that, on the outside, looks like a rod or bar of copper, but appears to have a thermal conductivity that is several times higher than that of copper. But the heat pipe is hollow and on the inside it passively creates a fluid recirculation loop. The fluid is evaporated at the hot end of the heat pipe, travels along its length, re-condenses and the cold end, and then the liquid travels back to the hot side to start the process over again. The liquid carries the heat from one end to the other, and can do so much more efficiently than mere conduction through the solid copper walls. Heat pipes are very common in today’s electronics. In fact, practically every laptop has one or more heat pipes to distribute heat from CPU’s and GPU’s to the heat sinks elsewhere in the laptop.
Meanwhile, the cooling needs of electronics continues to escalate, and existing heat pipes have some limits. In response to these trends, DARPA, the Defense Advanced Research Projects Agency, put out a request for teams to develop an advanced Thermal Ground Plane, which in essence is a high performance planar heat pipe. GE was one of the teams selected to attempt to develop such a device.
So here’s what we are going to build. First of interest is the form factor. Most heat pipes are literally pipes, say 6 mm in diameter and a few inches long. But DARPA wanted something that looks more like a circuit board in size and scale. So we are attempting to build a heat pipe that is only 1 mm thick, but is up to 20 cm long. This is very thin! Maintaining structural integrity will be very challenging.
The other big requirement is that this device needs to be able to operate at up to 20 g’s. Depending on orientation, the g-forces can impede and even halt the flow of the liquid in a regular heat pipe, thus stopping the operation of the heat pipe and driving the temperature of the electronics through the roof. There are ways to make heat pipes work at high g’s, but then one must severely de-rate the amount of heat the heat pipe can carry. A major innovation was required.
One of the key points of innovation for this project is to leverage some of our recent advancements in nanotechnology. By carefully inventing and constructing special nano-sized features in various regions of the TGP, we believe we are going to set records for heat fluxes at high g’s.
The other thing that makes this project very daunting, but very fun, is the wide range of disciplines needed to successfully create the TGP device. A great thing about the GE Global Research Center is that we have just about every type of technologist available. So it is true that some of my thermal experts are working on this project, but they constitute only a fraction of the technologists. We’ve got a team of experts on computational heat transfer methodologies building a new suite of models to predict the performance of our TGP devices. We have chemists who are experts at fabricating new material technologies, and engineers who have devoted their research over the last several years to nano-scale multi-phase heat transfer. And we have packaging experts who are extremely knowledgeable at selecting substrate materials, bonding the TGP packages together, even how to interface the electronics to these devices in the future. Plus we have the pleasure of teaming with the University of Cincinnati and the Air Force Research Lab. The result is a diverse, world-class team of scientists who are tackling a truly hard problem. But when we succeed, you will see our TGP in a wide range of GE’s electronics products!
Glad to learn how cool cooling technologies are. Are you forecasting that such technologies, when commercially available, will cause another Moore’s law acceleration?