Can three giant satellites save Viasat's business? Today on Space Stakes: after years of delays and a broken antenna, Viasat says its final ViaSat-3 satellite is now in service. Before that, in the headlines: two European firms booking SpaceX return capsules for Starship, and five launches stacked up this week including Starship's next big test. Welcome back to Space Stakes, your daily brief on the business of space. It's Tuesday, September 22, 2026. Let's get into it. Now, we've got a full board today, so let's move fast.

Two European mission integrators have booked SpaceX's Starfall reentry capsule for commercial flights on Starship, extending a vehicle first demonstrated on Falcon 9 this June into planned microgravity research and in-space manufacturing campaigns later this decade, according to NASASpaceFlight.com. Now, Starfall itself needs a quick picture for you — it's SpaceX's uncrewed, disk-shaped cargo reentry vehicle, built to carry payloads up to low Earth orbit, that's the busy zone just above us, let them sit there for research or manufacturing, then bring them home through the atmosphere. Space Cargo Unlimited plans a one-tonne in-space factory flight; Exolaunch will aggregate research and manufacturing payloads and return them to Earth. So how does it actually compare in size? FAA environmental documents described in that reporting put it at 3.1 meters in diameter, 0.75 meters tall, with an empty mass of about 2,100 kilograms and a payload capacity of up to 1,000 kilograms. Luxembourg-based Space Cargo Unlimited announced on Sept. 15 that it has contracted a dedicated Starfall mission targeting a 2028 Starship launch, intending to fly a one-tonne space factory using its BentoBox platform. Germany-headquartered Exolaunch followed on Sept. 21 with its own SpaceX contract for a Starfall mission targeted for 2029, also on Starship. And who pays here? Space Cargo says it is already talking with more than 100 prospects, about 40 percent of them in life sciences and biotechnology, while Exolaunch will aggregate experiments from pharmaceutical, biotechnology, advanced-materials, government, and research customers. Now, one more small note on status — SpaceX first flew Starfall on a Falcon 9 demonstration on June 23, 2026, from Cape Canaveral, with recovery in the Pacific days later, but neither of these new dates is an official SpaceX manifest commitment, and both depend on Starship reaching operational commercial flights. So announced deals, not funded launches yet, with Redwire's SpaceMD unit also disclosing a Starfall booking in the mix.

A total of five orbital launches are scheduled this week, according to NASASpaceFlight.com, and this is one of those traffic-jam weeks that shows you how global launch has become. Now, a caveat in our voice — this preview is built from that single roundup, which we haven't independently verified launch by launch, so take times as reported targets that can still move. First up, a Chang Zheng 8A, that's China's two-stage medium-lift rocket, is scheduled to liftoff from Commercial Launch Complex 1 at the Wenchang Space Launch Site in China on Wednesday, Sept. 23, with a planned liftoff at 13:30 UTC to deliver an unspecified payload to an unknown orbit. Then a Chang Zheng 6A is scheduled for Thursday, Sept. 24, from Taiyuan Satellite Launch Center, with a planned liftoff time of 08:45 UTC, also with an unspecified payload. Rocket Lab's Electron, that's the small 18-meter rocket that can deliver a total of 320 kilograms to low Earth orbit, is expected to launch Saturday, Sept. 26, from the Mahia Peninsula in New Zealand to deliver a 100 kilogram StriX satellite to low Earth orbit for Synspective. That would be the 97th overall Electron launch and the 18th of 2026. SpaceX will launch its Falcon 9 vehicle to deliver a payload for the United States Space Force, while the company's super heavy-lift Starship is expected to launch its first orbital mission at the beginning of next week. Specifically, the only Falcon 9 launch this week is scheduled from Vandenberg Space Force Base in California on Saturday, Sept. 26, to deliver a classified payload for the United States Space Force to low Earth orbit, with booster B1100 set for its 10th flight and recovery on the droneship Of Course I Still Love You. And the big one? Starship's 14th flight test is scheduled from Pad 2 at Starbase, Texas, on Monday, Sept. 28, with flight controllers set to decide whether to perform an orbital insertion burn at T+25 minutes. That sets up our main story nicely, because while everyone watches Starship, the old guard in high orbit just declared victory.

Our main story today I am calling The Trio Is Finally Lit — Can Viasat Sell It? According to Via Satellite, the final ViaSat-3 satellite entered service late last week, completing deployment of Viasat's high-capacity satellite trio. Now, let me level with you on where this stands — this is single-source reporting from that outlet that this show has not independently confirmed elsewhere, so we are treating their service dates as reported. This satellite, ViaSat-3 F2, was actually the second satellite to launch, and it launched in November of last year. All three satellites are now in service and the ViaSat-3 constellation is fully operational after the F3 satellite entered service about a month ago in late August. If you're hearing about ViaSat-3 for the first time, here's your one-breath bridge — Viasat bet on just three enormous satellites parked far above Earth to deliver huge amounts of internet bandwidth, instead of thousands of small ones buzzing close by. Now, for regulars, you know the baseline here — geostationary broadband, that's internet from that distant fixed ring, has to prove it can still grow while low Earth orbit constellations add capacity by the thousands. So what changed today? The F3 satellite covers the Asia-Pacific region and F2 covers the Americas. Each satellite adds more than 1 terabit per second of capacity to Viasat's network to bolster its business in aviation, maritime, government, and broadband. That is the payer list right there — airlines paying for Wi-Fi you actually use, ships, governments, and home broadband — which means Viasat finally has fresh sellable capacity in the regions where its planes and ships actually fly. More than a terabit per second per satellite is a genuine slug of throughput, the kind of number that stopped me cold when I first read it. The F1 satellite, which launched in 2023, suffered an anomaly with its antenna deployment. That satellite is operational, but only delivering a fraction of the capacity that was expected. So keep the stages straight with me — launched is not equal to operational, and operational is not equal to full design power. Beyond the capacity, the satellites are equipped with beamforming technology to allocate beams based on the location of demand, which Viasat sees as a critical differentiator. Think of beamforming as a spotlight you can steer — instead of spraying bandwidth everywhere, you point it where the planes and ships are that evening.

Now, of course, the question is whether steel in orbit turns into money on Earth, and that is where my read starts. Viasat chief executive Mark Dankberg said in a statement, quote, We believe that the VS3 satellites are the most advanced commercial satellites in the world in terms of adaptive beam forming for cost efficiencies, user performance improvements that offer higher speeds on both forward and return links, and resilience to interference, end quote. He went on, quote, We also believe VS3 technology sets new commercial standards for solar power generation and thermal dissipation – both critical technology challenges for developing economical data centers in space, end quote. That's a CEO selling both today's bandwidth and tomorrow's vision, data centers in space, which to my ear belongs firmly in the vision column until somebody shows you the customer. Here's my read — this is a years-in-the-making operational payoff built on three satellites in service after that F1 antenna failure, real hardware now doing real work. But trio in service does not equal full design capacity, and the fill-rate risk is everything now. Can Viasat migrate aviation and government customers onto F2 and F3 fast enough, at yields that repair the balance sheet, while low Earth orbit rivals sell on latency and flexibility? The reporting doesn't give us sold throughput or pricing, and I won't invent it. What would change my mind in a bullish direction? Signed capacity deals and utilization showing the trio moves revenue, not just coverage maps. What would darken it? If F1 stays impaired and F2 and F3 fill slowly while debt keeps ticking. Time for the Hype Check. On my 1-to-10 substance scale, I give this a 7 — three high-capacity satellites actually in service, with named coverage and per-satellite capacity, is major operational substance, docked three points because the economics are still unproven and F1 is still that impaired sibling. The company that wins the biggest bandwidth bets owes the public schedule honesty and cost transparency, in my view, and that applies to Viasat just as much as anyone else flying taxpayer-adjacent traffic. A geostationary trio that can steer a terabit of capacity toward demand deserves applause, and it still has to send invoices to matter.

If today's episode helped you follow the money in orbit, follow the show wherever you listen so you get tomorrow's brief on Starship's path to Sept. 28. This has been Space Stakes, an AI-voiced podcast, created and built by a real human using today's cutting-edge technology. Nothing you heard on this show is financial advice. I'm Brian Lampert, and I'll catch you all tomorrow — take care!

I also host Concrete Compute: a daily briefing on the AI buildout. The datacenters, the megawatts, and who actually pays for them. Find it wherever you get your podcasts.