What did Orbital Matter print in space?
On 3 November 2026, TechCrunch reported that a satellite built by Orbital Matter had printed a half-metre-long cylindrical beam in the vacuum of space. The company's executives called it a first for a free-flying spacecraft. The beam was made from a custom plastic that starts as a liquid resin and hardens under ultraviolet light.
The beam was not a demonstration for its own sake. Replicator 2 carries a flexible solar array folded against the spacecraft. Two of its printers are designed to build one-metre booms that hold that array open. On this attempt, one of those two failed after an anomaly, so a single half-metre boom did the work. The array still deployed, which the company treats as proof that printing and deployment can be joined in orbit.
Orbital Matter describes the mission as a sequence rather than a single test. First the satellite is brought online and each system is commissioned. Then a printer proves it can extrude and cure material in the space environment. Then the settings are refined and the print is scaled to full-length booms, which finally deploy the array folded on board. The November beam sits at the third step, not the last.
The word “first” needs unpacking. Engineers have printed parts in space for years, but almost always inside the International Space Station. That cabin has air, pressure and temperature control. Orbital Matter says one of its printers was the first to extrude and cure material in open vacuum. The company reported that result on 12 August 2026, with footage from the printer's own camera.
An independent assessment came from Advenit Makaya, an advanced manufacturing engineer at the European Space Agency. He told Aerospace America in September that a successful flight would be the first publicly known case. The structure would have been printed on orbit, outside a space station or capsule. The distinction matters because a station is a workshop. Open space is the environment where large structures would actually have to be built.
Size is the point of the exercise. A half-metre cylinder is small. The structures Orbital Matter wants to build are not. Its design targets booms of 100 metres or more, long enough to support solar arrays measured in tens or hundreds of kilowatts.
How does a 3D printer work in a vacuum?
Replicator 2 is an 8U CubeSat weighing about 13 kilograms. It launched on 7 July 2026 on a SpaceX Falcon 9 rideshare flight, Transporter-17, into an orbit roughly 584 by 597 kilometres above Earth. Four independent printer modules sit inside the small body, each about the size of a one-unit CubeSat.
Each printer pushes out photopolymer resin in a coil and cures it with ultraviolet light. Layer by layer, the material stacks into a rigid cylinder. The finished boom has no hinges, latches or motors, which is the company's main argument against conventional deployable structures. Those mechanisms add mass, cost and failure points as an array grows.
Printing outside a spacecraft changes the physics. There is no gravity to settle the resin, no air to carry heat away, and temperatures swing sharply between sunlight and Earth's shadow. Orbital Matter's first extrusion in August came out over-extruded: the printer pushed out more material than the design called for.
Engineers revised the settings on a ground model in Warsaw and uploaded them to the satellite, with printing due to resume in mid-September. The company's chief technology officer, Robert Ihnatisin, said the team expected to tune the process in orbit. "We preferred to learn in orbit and adapt the software as needed rather than waste years testing every possibility on the ground," he told TechCrunch.
Each printer has its own inspection camera, and the company built the image-recognition software that watches every print. That vision system is how a printer checks its own work without a person in the loop. Orbital Matter has also patented the vacuum-printing method, according to Payload. The company calls the material SPACRETE and the deployment hardware PADS, for printer-assisted deployment system.
Why do orbital data centres need 100-metre solar arrays?
Satellites are built around the rocket that carries them. Solar arrays, antennas and telescope mirrors are folded like origami, then unfolded by hinges, motors and tensioning cables. That approach has worked for decades, but it scales badly. "Deployables will hit a certain ceiling," Makaya said. "The limitation you have when you want to make large structures in space is the size of your rocket fairing."
The demand side is changing. Google's Project Suncatcher is testing whether AI chips can run in orbit. Google says satellites in low Earth orbit see near-constant sunlight and can generate up to eight times more solar power than on Earth. Future designs would carry dozens of TPUs per satellite, flying in clusters. Cooling is the catch: in a vacuum, heat can only leave through radiators. The first prototype was scheduled to fly on SpaceX's Transporter-18 rideshare with Planet as a partner, and Google says its next milestone is 2027.
That combination of more power and more heat pushes arrays and radiators beyond what folding allows. SpaceX's Starlink V2 satellites, among the most powerful mass-produced spacecraft in orbit, use solar arrays about 20 metres long. Orbital Matter argues that the same compact printer module could make booms over 100 metres, and that the cost per metre falls as the structure grows. The company says next-generation constellations and orbital data centres need arrays in the tens of kilowatts, far larger than anything flying today.
Orbital Matter's commercial case rests on that gap. Jakub Stojek, the chief executive, has said its potential customers are operators that need large structures, especially developers of orbital data centres. Ihnatisin told Aerospace America that the company had a number of prospective customers but declined to name them.
The cooling problem follows the same logic. Google's engineers are testing heat pipes and radiators in a thermal vacuum chamber because a data centre in orbit cannot dump heat by warming air. The radiator becomes as important as the array, and both want more area than a fairing allows.
Government money has gone into the same idea before. TechCrunch notes that NASA funded a Redwire effort that printed structural beams in a vacuum chamber in 2023, but the mission never reached orbit. A chamber can show that a printer works. It cannot show that a boom survives launch, thermal cycling and radiation at the same time.
How close is Orbital Matter to a product?
The honest answer is early. The company was founded in 2022 and has flown two demonstration missions, with Replicator 2 the second. Payload reports that a third, larger mission is in preparation.
TechCrunch reports that the next demonstration is scheduled for 2028. It will test a larger version of the printer on a bigger satellite, with the aim of deploying an array that generates 5 kilowatts. The company describes that as a stepping stone to 100 kilowatts or more per wing by 2030. Payload's October profile says Orbital Matter expects construction missions in the 2030s that deploy radiators and kilometre-long solar arrays.
Each step is roughly an order of magnitude. A half-metre boom holding a small array is not a 100-metre structure carrying kilowatts. Stiffness, thermal expansion and pointing all become harder as length grows, and a solar array has to survive years of cycling between sunlight and shadow.
Earlier reporting pointed to a faster schedule. In September, Ihnatisin described a follow-up mission that would print two three-metre beams to deploy a one-kilowatt array. Its launch was targeted for before the end of 2027. The TechCrunch account in November describes a 2028 mission and a five-kilowatt array. The two reports differ on both date and power, and Orbital Matter has not publicly reconciled the two accounts.
TechCrunch reports a $3 million seed round. In July 2026 the European Innovation Council awarded the company €500,000 through its Pre-Accelerator programme. Orbital Matter is headquartered in Berlin, with its main office and research work in Warsaw.
The partner list is notable. The European Space Agency and Thales Alenia Space worked with Orbital Matter on Replicator 2, according to TechCrunch. Thales Alenia is also developing a deployable VHF antenna that would use a printed boom as its support. A mockup of that antenna flew on the same satellite. Orbital Matter was accelerated by Thales Alenia Space's Space Business Catalyst.
What went wrong on the flight, and what still limits the design?
The mission did not match its own plan. In September, Ihnatisin described a primary demonstration in which two printers would build one-metre beams in unison by mid-October. The result reported on 3 November was half a metre from a single printer, after the other failed.
That is not a failure of the concept. Four independent printers were flown for exactly this reason: one working unit can still prove the process. But it does show how much of the engineering is still being learned in flight rather than in a laboratory.
The most basic limit is material. A boom can only be as long as the resin supply allows. TechCrunch reports that the company's engineers believe 100-metre booms are feasible, but that belief is not yet backed by flight data.
Then there is the environment. Over the satellite's expected three-year life, engineers plan to watch how the beams expand and contract between sunlight and shadow. They will also measure how much the beams degrade under radiation. The array's power output over time will be measured too, using off-the-shelf solar cells, so the experiment tests the structure rather than a new photovoltaic design. Ihnatisin framed the question simply: "It's basically how well the beam holds together."
Makaya expects the industry to need more than printing. He told Aerospace America that building large structures in space will likely require a mix of additive manufacturing and robotic assembly. Printing a boom is one capability. Building a 100-metre array on orbit is a system of capabilities.
There is also an economic question that no demonstration answers. Printing in orbit only makes sense if the saved mass and complexity outweigh the cost of carrying resin, printers and power. That trade will vary by mission, and those numbers will decide whether printing beats folding for a given satellite.
The competing technology is not standing still. Conventional deployable arrays are mature and well understood after decades of use. A printer adds resin mass and a new failure mode, while folding adds hinges, latches and motors that grow heavier as an array grows. Both sides can point to real engineering problems, and the market will decide which set is cheaper to solve.
Who else is trying to manufacture in space?
Orbital Matter is not alone. SpaceNews reported in September that Dcubed, a German company, is also running an on-orbit 3D-printing demonstration in a CubeSat. Both companies are chasing solar arrays with additively manufactured booms.
The better-established route runs through a pressurised module. Previous in-space manufacturing, including parts printed for astronauts, happened inside the station's pressurised cabin, in air rather than vacuum. Those experiments proved that extrusion can work in microgravity. They did not test vacuum, which is the harder problem for large structures.
What separates the current round is that the hardware is flying. A vacuum chamber can simulate pressure and temperature. It cannot reproduce launch vibration, orbital thermal cycling and radiation together. Every flight test, successful or partial, adds data that ground tests cannot produce.
What to watch next
The next year should produce the data that matters. Orbital Matter says it will track the printed boom through thermal cycles and radiation, and measure the array's output over time. Those results will show whether a printed structure behaves like an engineered boom or merely looks like one.
One test of the company's confidence will be how much of that data it publishes. Another will be whether satellite builders adopt its deployment hardware rather than waiting for a full solar array product.
- •Whether Orbital Matter publishes power output for the array deployed by the printed boom.
- •Flight data on how the beam expands, contracts and degrades over the satellite's three-year life.
- •Progress on the 2028 demonstration, which aims at a 5-kilowatt array.
- •Adoption of the printer-assisted deployment system by satellite builders, which TechCrunch reports is under discussion.
- •Google's Project Suncatcher results, which will show how much power and cooling orbital AI hardware needs.