Startup Phantom House needs to move into orbit towards the grain


Cantrell’s latest focus is Phantom Space, one in a sea of new launch startups seeking to take advantage of the explosion of smaller, cheaper satellite designs and build rockets that can meet the growing demand to launch these payloads into orbit. But as is par for the course with Cantrell, Phantom is trying to find success by swimming against the current. 

One of the hottest trends in rockets right now is ride-share launches, where customers buy available spots for their payloads on a midsize or large rocket with a specific departure date. This is usually a cheaper way than single launches for customers to get a payload into space—with SpaceX’s ride-share program, it costs $1 million to launch a 200-kilogram payload (its Falcon 9 rocket can take 22,800 kg total into low Earth orbit). The company launched a dedicated ride-share mission on January 21, deploying a record 143 satellites into orbit. It’s following up with a similar mission in June. In a surprising about-face in March, Rocket Lab, which has long resisted the idea of building larger rockets, unveiled the Neutron for the precise purpose of doing ride-share launches and competing with the SpaceX Falcon 9. 

Ride shares aren’t Phantom’s cup of tea. The company wants to establish its space footprint by mass-manufacturing small rockets and launching a hundred a year. “We want to be the Henry Ford of space,” says Cantrell. “We’re taking a contrarian view of how we develop this.” Just as Henry Ford didn’t reinvent the car but the way it was built, Phantom isn’t out to reinvent rockets—just their production.

How so? When SpaceX started, the supply chains for aerospace companies going into orbit were enmeshed in the financial system of the US Defense Department. To stay independent of that system, SpaceX decided to build everything itself, relying on Musk’s fortune and a ton of investment to stay afloat during years of losses. It was a long-term gamble that paid off.

But Phantom’s founders decided they didn’t need to follow suit. Even in just the last five years, aerospace supply chains have become more fluid and competitive, which means Phantom can just buy the specific parts it wants rather than build everything from scratch. It buys 3D-printed engines from Ursa Major in Colorado. The design of the flight computer was licensed from NASA, and it uses a BeagleBone Black board that some distributors sell for around $50. Other components, like batteries and telemetry systems, are bought through the missile defense supply chain. 

The Henry Ford analogy isn’t just an aspiration—it’s a model for the company. Cofounder Michael D’Angelo says the car and rocket businesses follow similar growth curves: doubling production leads to certain economies of scale also associated with greater efficiency and fewer production errors. Computers and mobile devices followed a similar path. And he argues that the supply chains nowadays are mature enough to allow for the kind of speedy manufacturing Phantom wants. 

Right now, the company is pursuing two types of rockets. There’s the 18.7-meter-tall Daytona, which should be able to lift about 450 kilograms into space. It’s perhaps on the larger end of what might be termed the small-rocket class, but according to Cantrell, the company’s analysis estimates this to be an optimum size for profitable activity. Then there’s Laguna, a 20.5-meter-tall rocket that can lift payloads of up to 1,200 kg. Phantom is developing a version of Laguna with a reusable first-stage booster, like a SpaceX Falcon 9 (with a similar vertical landing process). 

An artist’s rendering of the Daytona rocket flying into space.


Phantom is hoping to fill a gap in the market. While ride shares are cheap, customers have less control over how the mission goes. A ride-share mission, like a train, is on a fixed route. If you want your satellite to go a different orbit or trajectory, you need to install expensive thrusters that can take it there. Otherwise, you have to redesign its function for the new orbit, tolerate a less favorable orbit, or simply buy a ticket for a different mission. And you better hope your satellite can fit snugly with all the other payloads it’s going with—these flights are fully booked. 

A small-rocket launch might cost more, but it gives control back to the customer. If you have a mission with very specific requirements—like replacing a particular satellite in a constellation, launching sensitive equipment, or running an expensive tech demo—you’ll probably want a dedicated flight rather than a ride share. “There’s definitely an interest and demand for these small-rocket launches,” says Ryan Martineau, a space system engineer at the Space Dynamics Laboratory in Utah.

Cantrell thinks Phantom can meet this demand without blowing its budget. He estimates that the company’s approach can actually offer launches for a third the price of the ride-share model.

First, though, the company has to actually get to space. The aim is for Daytona to make its first spaceflight in 2023. Classically, says Cantrell, there’s a 50% reliability rate for the first four flights of a new rocket. Phantom’s plans more or less assume that at least one of its first four flights gets to orbit. It recently signed a lease from the Air Force for a launch site at Vandenberg Air Force Base in California, and it is currently looking for permission to launch from Cape Canaveral, Florida, as well—important initial steps if 100 launches a year is truly the goal.

Phantom also wants to build satellites and become something of a one-stop shop for customers. Its acquisition of Cantrell’s StratSpace this week is supposed to be a key part of this side of the business. The company is working on constellation prototypes for customers and is part of a group developing a commercially funded $1.2 billion science mission (specific details won’t be disclosed for several months). And it has been quietly working on a communications network it calls Phantom Cloud, which is essentially a mesh network other satellites can use to communicate with each other or with systems on the surface. Cantrell calls it “satellite internet for space.”

In reality, Phantom doesn’t exactly need to beat SpaceX and the other large rocket makers—it just needs to hold its own. “As the small-launcher market matures, I think you’ll see a wider variety of customers taking advantage of that capability,” says Martineau. “I think it’s unlikely that one will become dominant and edge out the other.” 

Coexistence is fine, says Cantrell: “We recognize that SpaceX has magnificently developed this large reusable space transportation system, but we think that is but one of at least two—maybe more—fundamentally different economic systems in the space transportation ecosystem.” He hopes it’s Phantom that pioneers the other.