When we feel the excess heat coming off of our overworked computers, most of us start hastily closing browser tabs. But where the average computer user sees danger, Lawrence Orsini sees a tremendous opportunity.
“Computers are really good heaters — that’s where they shine,” says Orsini, a former auto machinist and veteran of the energy efficiency industry. “Instead of dumping that excess heat into the atmosphere, like we do now, we should be capturing that heat and put it to good use.”
That’s the basic idea behind Orsini’s startup, Project Exergy: rather than transporting electricity across power lines to generate our heat — and wasting most of the power in the process — Exergy proposes that households and business can generate the heat they need onsite, using computation.
In thermodynamics terms, “exergy” refers to the maximum amount of useful energy in a system. It’s an appropriately compact rallying cry for wringing as much use as we can out of the heat that we’re already generating with our universe of computational devices, and even more so, the server farms that power what we misleadingly refer to as “the cloud.”
Far from being airy and untethered from our earthly concerns, Orsini points out, the cloud is a massive consumer of energy and producer of waste heat. Energy is consumed every time we perform a Web search, upload a selfie, or publish a blog post; by some estimates, moving a megabyte of data across the Web is equivalent to burning one lump of coal. So the server banks that power this ceaseless dance of data produce a tremendous amount of heat — heat that is then dumped into the atmosphere. Compounding the absurdity is the fact that these servers need to be kept cool — a process that generates even more heat.
“In the name of efficiency, companies like Amazon and Google have started moving their server farms out to colder locations, closer and closer to the Arctic Circle,” Orsini says. “But if you look at it in terms of wasted heat, it’s tremendously inefficient.”
As a proof of concept, Orsini has already built a working prototype, affectionately nicknamed “Henry,” and put it to use in his apartment in lower Manhattan. Henry is a massive PC — “a gaming rig on steroids,” Orsini says — atop a thermal storage tank, which collects and stores the excess heat generated by the computer. The computer itself performs a multitude of functions, like streaming videos to a projector and controlling the lighting and the sound system. Water blocks — metal wafers filled with water that are bolted to the circuitry — capture and transfer the heat into the storage tank below.
The beauty of the concept is that the computation itself can serve any purpose. Last winter, Orsini used the Frankensteinian rig to mine cryptocurrency; between the money he saved on electricity, and the money he mined, he came out with a little bit of money in his pocket. Henry has also been used to fold proteins, as part of a distributed computation scheme in which universities outsource their computation to idle personal computers.
Impressively — and deliberately — Orsini built his prototype using only off-the-shelf parts.
The ethos behind Project Exergy reflects Orsini’s background building racecars; the aesthetic is more steampunk than SpaceX.
“You could buy the parts build this yourself,” Orsini says. “There’s no magic here.”
Orsini hopes that Project Exergy will successfully convey the idea that combining heating and computation is a low-hanging fruit that can make a big difference in our energy future. Water and heating needs comprise 60% of residential energy consumption in the United States, according to a 2009 survey conducted by the U.S. Energy Information Administration, the results of which are prominently featured on Project Exergy’s website. And U.S. data centers, Orsini says, generate enough excess heat to warm about 80 million homes, or roughly two-thirds of the houses in the country.
Looking ahead, Orsini and his colleagues at Project Exergy envision a future where combined computation units — the space-age descendants of Henry — reduce or obviate not only the need for transporting electricity across vast distances, but also drastically reduce the number of devices we have in our homes. This, in turn, would provide even greater efficiency, reducing the need to mine the earth for the rare minerals that power our smartphones.
“Computation is computation,” says John Lilic, a computer scientist, cryptocurrency advocate, and head of business development for Project Exergy. “Why do we need separate devices to watch movies, play video games, and listen to music, when you can have one device that does all of that, while generating excess heat and reducing the load on the electrical grid?”
It’s with these high hopes , and the mission of infecting key industry players with their enthusiasm, that Project Exergy is launching a Kickstarter campaign to get the word out and fund the next prototype. Unlike Henry, Orsini says that the next version won’t be recognizable as a computer. Having proven with his first prototype that the concept works, he wants the next version to be capable of running at higher temperatures, with greater efficiency.
“This thing is 70, maybe 75 percent efficient at capturing the heat,” Orsini says. “It’s not bad for something hacked together, but with the next version, with all of the components submerged in a liquid dielectric [an insulating oil that allows for the capture and transfer of heat], we’re hoping to get that into the 90s.”
They’re also hoping that the next iteration will be modular, so that each home can customize the unit to their own energy needs. Lilic envisions the consumer version running on an operating system similar to Android, in which third party developers can create apps for, say, playing music, controlling lighting, watching movies, or anything else an app can do.
Of course, these ideas — drastically reducing the need to transport electricity, doing away with massive data centers, and replacing a universe of proprietary devices with single units — all have the potential to disrupt the business models of not only utility companies, but also this generation of established disruptors, like Google and Amazon. But Orsini and his team don’t see Project Exergy as disruptive, per se.
“I’d rather think of it as innovating, rather than disrupting,” Orsini says. “These companies have gotten where they are by innovating, and we think this is the next step in those types of innovations. We’re looking for allies.”
“Ultimately, we’re really about getting these ideas into the marketplace, for people to really start thinking about these things,” he adds. “If, out of the same amount of energy that we’re pulling out of the ground, if we could get increased use out of that energy, by several fold, why wouldn’t we do that?”
Project Exergy has generated interest, but not funding, from the U.S. National Renewable Energy Laboratory. However, not everyone is convinced it will work. Tim Jenkins, a nuclear operations instructor with over 30 years of experience in the energy industry, expressed skepticism that computing can generate sufficient heat to warm a home.
“I think I am doing exactly this right now. My laptop is generating heat and it is warming the room,” Jenkins said. “Theoretically, that is less energy that I have to supply from my furnace. I do not feel a difference right now, and I do not believe there would be an observable difference on my heat bill if I turned my laptop off for a month.”
“My initial gut feeling is it is a small amount of heat and that it is noticeable when we don’t need it, such as during the summer, for instance,” he said, “But I would need more data and think time.”
As for the energy industry, Jenkins thinks they would embrace Project Exergy “if they could see a way to make a buck. Like all industries these days, we are run by the bean counters.”
