The First Look At Polaris - Helion's Latest Nuclear Fusion Machine
It's extraordinary. Trust me
Around a dozen nuclear fusion start-ups in the U.S. have raised more than $100 million each in recent years. The investors backing these companies have clearly bet that fusion reactors can become real energy sources for humans in the relatively near future, even if history has not been kind to such hopes and dreams.
No fusion company is more ambitious or more controversial than Helion Energy, which is backed by $1 billion from Sam Altman and others. Helion has plans to bring a fusion system online by 2028 and a contract in place with Microsoft to purchase the resulting electricity.
Plenty of people doubt that Helion can meet this timeline. More to the point, plenty of people doubt that Helion’s technology works. Unlike most of its rivals, Helion does not seek to create electricity by using fusion to heat up water, create steam and spin a turbine. Instead, Helion plans to suck electricity straight from its fusion machine via a process that involves extreme heat, the creation of plasma and magnetic fields.
Helion, which was founded in 2013, has irked scientists and competitors by refusing to publish much data about its work. It also has not let many outsiders into its facilities. This air of secrecy has made it relatively easy for rivals to suggest that Helion’s technology might not be so crash hot. In particular, there are questions as to whether or not Helion can actually produce more electricity than it consumes with its fusion machines.
As you’ll see in our episode, David Kirtley, Helion’s co-founder and CEO, appears rather unfazed by the skepticism.
During our recent visit to Helion’s headquarters in Everett, Washington, Kirtley took me inside Ursa, which is the company’s fusion facility that houses Polaris, which is its latest and largest fusion machine. I was, in fact, the first reporter to see the machine.
Polaris is extraordinary. It’s an hourglass-shaped machine connected to a shit-ton of cables, which then connect to a shit-ton of capacitors.
This is how it looks in my mind’s eye.
I could prattle on for longer, but you should just watch the video for, like Polaris, it is glorious to behold.
I spell “shit ton” without the hyphen, but reasonable people can disagree. Regardless, it’s a wonderful volumetric descriptor and nice to see it deployed in my favorite substack. Take that, Bloomberg style guide!
Wishing you happy holidays and a shit ton of success with Core Memory in the new year.
Love the direct electricity approach here. Most fusion projects stick with the thermal route because it's familiar, but Helion's magnet-field capture method is way more elegant if it can actually work at scale. I've always wondered if their secrecy is strategic or a sign they're not hitting the energy gain numbers, but getting access to teh facility shows they're at least confident enough to let people see it. That 2028 timeline with the Microsoft contract seems aggresive as hell tho.