Summary List Placement

SpaceX is on a stunning streak with Starlink, an ever-growing fleet of spacecraft that can beam internet across the planet, and with a performance greatly exceeding that of traditional satellite services.

However, company founder Elon Musk has hinted the project is not out of the woods yet in terms of its economics.

Over the past 18 months, SpaceX has launched nearly 900 Starlink spacecraft into low-Earth orbit, scooped up big contracts with the military and Microsoft, and even opened a public beta test program.

With any luck, SpaceX executives think they can bring online the floating high-speed network within a few months.

“We hope to roll it out this year,” Gwynne Shotwell, SpaceX’s president and chief operating officer, told Aviation Week space editor Irene Klotz for a May 27 episode of “Check 6,” a podcast published by the magazine.

Over time, and with expansion to perhaps tens of thousands of satellites, global revenue from the project could top $50 billion annually, Musk said last year.

The problem, though — and one which has plagued all ambitious constellations, some to the point of crippling bankruptcy — is cost.

Specifically, the issue lies with a high-tech satellite dish that Musk calls a user terminal, or the “UFO on a stick” device. Starlink user terminals rely on phased-array microelectronics to autonomously and almost instantly steer a connection beam to satellites zooming overhead.

Phased-array panels are not cheap today, though, which is why Musk recently said this is Starlink’s most significant hurdle to clear if SpaceX is establish a successful internet business.

“I think the biggest challenge will be with the user terminal and getting the user terminal cost to be … affordable,” Musk told Aviation Week in a different episode of “Check 6” posted in May. “That will take us a few years to really solve.”

Musk reminded the world of this problem in November amid buzz of the Starlink beta test, which charges users about $99 a month for service plus a $499 fee for a kit that includes a user terminal, internet router, and tripod. Estimates by industry experts suggest each phased-array panel alone (not including any other hardware) costs more than $1,000, so it’s almost certain SpaceX is taking a hit with each subscriber.

“Lowering Starlink terminal cost, which may sound rather pedestrian, is actually our most difficult technical challenge,” Musk tweeted on November 2.

A need for cheap and long-lasting ground antennas

SpaceX plans to send up to 12,000 and possibly 42,000 satellites, according to Federal Communications Commission records.

Ahead of the company’s first Starlink launch in May 2019, Musk said on a press call that it’d take about 800 satellites for “moderate” or “significant operational” coverage around the world. He also told Business Insider that making the system “economically viable” will require “on the order of 1,000 satellites.”

“If we’re putting a lot more satellites than that in orbit, that’s actually a very good thing. It means there’s a lot of demand for the system,” Musk said.

Starlink is geared primarily toward serving rural and remote regions that fiber-optic cables and mobile towers struggle to reach. To get online, subscribers will have to pay about $200 per user terminal.

“Looks like a thin, flat, round UFO on a stick. Starlink Terminal has motors to self-adjust optimal angle to view sky,” Musk tweeted in January, adding that all a user has to do is plug it in and point it upward. “These instructions work in either order. No training required.”

The devices are designed to rapidly track and communicate with in-view Starlink satellites because the spacecraft move around Earth at more than 17,000 mph. Dishes that must be physically pointed, like those used to aim at a stationary TV satellite, wouldn’t suffice for keeping an uninterrupted connection.

On each array is a platter of phase shifters, which can work together to electronically and almost instantly steer a transmission beam from one satellite to another.

SpaceX in February 2018 filed and world patents for a “distributed phase shifter array system and method.” (The US version is pending, and the world version is still under review.)

Industry analysts say phased-array devices today cost about 10 times as much as the price point Musk hopes to achieve, which is in the low four digits. The CEO is also concerned about how long such devices could stand out in the open, year after year, without requiring service or replacement, since maintenance needs would drive up cost.

“The user terminal cost is as considering … the fully installed cost — so the hardware, as well as everything required to get it set up and running, and running reliably for a decade or certainly at least five years-plus,” he told Aviation Week. “You can’t send people out to service these things because a lot of places will be in the middle of nowhere.”

‘We are focusing on making it not go bankrupt’

Musk is also aware that the path to high-speed satellite internet is littered with bankruptcies. The most recent of which is OneWeb — a competitor that hopes to launch thousands of satellites but declared Chapter 11 in March. (The FCC in late October approved OneWeb’s sale to the UK.)

More famously was Iridium: a constellation of 66 car-sized satellites whose goal was to create global wireless network for telephone and pager data. Motorola fueled the company in the 1990s with about $5 billion in investments ($6 billion when adjusted for inflation). But shortly after it deployed around Earth, new subscriptions flatlined because of the steep price of the service.

“We need to set the stage of constellations. Guess how many [low-Earth orbit] constellations didn’t go bankrupt? Zero,” Musk said at the Satellite 2020 conference in March, according to Via Satellite. “We are focusing on making it not go bankrupt.”

If SpaceX can drive down the cost of phased-array user terminals — and quickly — it may be well-positioned to avoid similar hindrances to success.

Mass production may get the company there, but it needs many subscribers. Subsidies could help the company close the gap, and to that end, SpaceX is seeking to grab millions or potentially billions of dollars in rural-internet subsidies from the US government over the next decade. (SpaceX is approved to bid for that money, though analysts think Starlink may not walk away with anything.)

Achieving cheap user terminal cost — and boosting subscribers — could also handily beat Jeff Bezos’ planned Project Kuiper satellite internet network to creating a profitable service. Though the FCC has approved that project, and Amazon has vowed to invest $10 billion to launch 3,236 satellites, a single spacecraft has yet to fly.

In his conversation with Aviation Week, Musk spoke with a mix of confidence and realism about the challenge.

“The secret to SpaceX’s success is that we’ve got … a better engineering team than I think has ever existed working on Starlink,” Musk said. “I think we’ve got a strategy where success is one of the possible outcomes.”

He added, with a laugh: “That’s high praise because I think there’s a lot of them where success was not one of the possible outcomes.”

Do you have a story or inside information to share about the space industry? Email Dave Mosher, send him a Twitter direct message, or consider more secure communication options listed here.

This story has been updated with new information and a correction. (A previous version of this story incorrectly used the word “transistors” to describe phase shifters, which represent the core technology in phased-array antennas.) It was originally published at 8:16 a.m. ET on June 5, 2020.

SEE ALSO: 2 sentences in Starlink’s terms of service show that SpaceX is serious about creating its ‘own legal regime’ on Mars

DON’T MISS: SpaceX and Amazon are making huge gambles on internet satellites that just might pay off — and transform where people live and work

Join the conversation about this story »

This content was originally published here.