China Loses Two Rockets Within Hours as Launch Ambitions Expand

Two losses interrupt a record launch push

China lost two rockets and their satellite payloads less than 12 hours apart on January 16 and 17, 2026, interrupting an expanding launch campaign after a record year. A Long March 3B carrying the classified experimental satellite Shijian-32 failed first. The debut flight of Galactic Energy's Ceres-2 followed, with approximately six commercial satellites believed to be aboard.

The Long March 3B lifted off from the Xichang Satellite Launch Center in southwest China at 1655 UTC on January 16, or 11:55 a.m. Eastern time. The China Aerospace Science and Technology Corporation (CASC) later confirmed an anomaly in its third stage. Ceres-2 launched from the Jiuquan Satellite Launch Center in northwest China at 0408 UTC on January 17, or 11:08 p.m. Eastern time on January 16. Galactic Energy also confirmed that its mission had failed.

The launches were separated by 11 hours and 13 minutes. Both ended in the complete loss of their payloads, affecting a state space program with decades of flight experience and a private company introducing a larger rocket.

China conducted 92 orbital launch attempts in 2025, with two failures. It expects to exceed 100 launches in 2026 for the first time, while pursuing satellite networks involving thousands of spacecraft. Recent paperwork submitted to the International Telecommunications Union (ITU) covers ambitions totaling more than 200,000 satellites.

As of January 18, both failures were under investigation. The immediate task is to identify what went wrong, while launch planners and satellite customers assess how the losses could affect an already crowded schedule.

What happened during the two launches?

The Long March 3B appeared to launch on schedule. Amateur footage showed its departure from Xichang at the time indicated by airspace closure notices, which warn aircraft away from areas affected by a rocket flight. Public confirmation of the outcome did not follow immediately.

CASC announced the failure approximately 12 hours after liftoff, identifying a third stage anomaly and confirming the loss of Shijian-32. The third stage is the upper part of the rocket that supplies propulsion during the later portion of the flight, helping place a satellite on its intended path.

The official Xinhua News Agency carried the announcement that the malfunction was still being examined:

the specific cause is under further analysis and investigation

Galactic Energy's Ceres-2 launched before that difficult period had ended. The company said an anomaly had occurred during its first flight and that the specific cause was under further investigation.

The sequence also included a success just before the two losses:

• April 2020: A Long March 3B failure involving its third stage destroyed Indonesia's Palapa-N1/Nusantara-2 satellite.
• November 2025: Galactic Energy's smaller Ceres-1 suffered a failure.
• January 15, 2026: Ceres-1 returned to flight with a successful launch from a barge off Shandong province, placing four Tianqi constellation satellites into orbit.
• January 16, 1655 UTC: Long March 3B launched from Xichang carrying Shijian-32 and subsequently failed.
• January 17, 0408 UTC: Ceres-2 launched from Jiuquan on its debut mission and also failed.

That sequence left Galactic Energy moving from a successful recovery flight to the loss of a new vehicle within roughly a day.

Why the Long March 3B loss matters

The Long March 3B has been a workhorse of China's space program for decades. Developed by the China Academy of Launch Vehicle Technology, it is a principal launcher for communications, weather, remote sensing and technology demonstration satellites headed toward geostationary orbit.

A satellite in geostationary orbit travels around Earth at the same rate that Earth rotates, allowing it to remain above roughly the same location. The Long March 3B usually delivers spacecraft to a geostationary transfer orbit, an intermediate route from which they move toward that operating position.

The rocket uses hypergolic propellants, fuels and oxidizers that ignite when they come into contact. Its upper stage uses hydrolox, a combination of liquid hydrogen and liquid oxygen. That upper stage is central to the current investigation because CASC identified the anomaly in the third stage.

The Shijian-32 loss was the first complete loss of a Long March 3B mission since April 2020, when Indonesia's Palapa-N1/Nusantara-2 satellite was destroyed following another third stage problem. The January failure also appears to have been the first outright launch failure across the Long March family in approximately 300 launches.

Upcoming flights could include satellites in the TJS experimental series and the Tianlian relay network. Tianlian satellites pass communications between spacecraft and ground stations, supporting human spaceflight and other missions. Delays to their deployment would therefore reach beyond the loss of a single payload.

The newer Long March 7A, which launches from the coastal Wenchang spaceport, provides another option for these missions. It also inherits its hydrolox upper stage from the Long March 3 series. Investigators must determine whether the malfunction was specific to the failed rocket or involved technology shared by other vehicles.

A grounding affecting that upper stage family could disrupt both Long March 3B and Long March 7A schedules. China's alternatives for deploying geostationary satellites would become harder to arrange if two launchers were affected at once.

What was aboard Shijian-32?

CASC released little information about Shijian-32 or its intended work. The Shijian name covers a varied family of spacecraft used to conduct experiments, test technology and verify ways of operating in orbit.

Official descriptions commonly identify Shijian satellites as experimental or capable of several roles. Western defense analysts have frequently associated the series with advanced signals intelligence and surveillance from space. Signals intelligence involves collecting and studying electronic transmissions.

Shijian-32's precise mission remained undisclosed. Its loss comes as China expands space surveillance and reconnaissance capabilities within its broader military modernization effort. Classified spacecraft can contain sensitive equipment that takes years to develop and replace, making a launch failure costly in both money and lost capability.

Other Shijian missions show the range of technology China is testing. Shijian-21 and Shijian-25 separated in geosynchronous orbit late in 2025 after spending months apparently docked together for refueling tests in orbit.

Geosynchronous satellites take about one Earth rotation to complete an orbit. Refueling requires spacecraft to approach one another, connect and transfer fuel in space. The Shijian-21 and Shijian-25 activities highlighted growing Chinese experience with advanced operations and autonomous rendezvous, in which spacecraft manage their approach using onboard systems.

Ceres-2's debut carried commercial expectations

Beijing based Galactic Energy designed Ceres-2 as a larger successor to its Ceres-1 rocket. Its long delayed first flight was intended to demonstrate a substantial increase in the company's ability to carry satellites into low Earth orbit (LEO), the region relatively close to Earth used by many commercial satellite networks.

Ceres-2 was designed to deliver 1,600 kilograms to an orbit at an altitude of 500 kilometers. The smaller Ceres-1 can carry 400 kilograms to LEO. Those advertised capacities make Ceres-2's payload figure four times the smaller rocket's figure, although the stated destination conditions are more specific for Ceres-2.

The new vehicle weighed 100 tonnes and had four stages. Three core stages used solid propellant, followed by an upper stage using liquid fuel. Solid propellant contains the fuel and oxidizer together, while liquid propulsion feeds propellants into an engine. Combining stages allows a rocket to discard spent sections as it climbs.

Approximately six satellites were thought to be aboard the failed flight. They included Lilac-3, an ultra flat satellite shaped like a disk, associated with the Harbin Institute of Technology and Beijing University of Aeronautics and Astronautics.

Galactic Energy acknowledged the impact on the customers and other participants in the mission, extending what it called:

sincerest apologies to all parties involved in the mission

The company had positioned Ceres-2 to win a larger share of commercial launch demand across the Asia-Pacific region and compete with international providers carrying small satellites. Industry observers viewed its loss as a strategic setback for China's growing private aerospace sector, alongside the technical setback to the established Long March fleet.

Financial pressure could extend to investors and insurers supporting Chinese space startups. Development costs for comparable vehicles typically exceed several hundred million yuan, with approximately US$42 million cited as a scale of expenditure. The loss of the satellites adds to the burden, although the mission's actual financial loss was not disclosed.

Galactic Energy faces several demands at once

Galactic Energy's successful Ceres-1S launch from waters off Shandong province had demonstrated its ability to operate from both land and sea. The company promoted that flexibility as a way to serve regional customers and arrange missions from different locations.

The launch from a barge delivered four Tianqi constellation satellites into orbit and marked the smaller Ceres-1's return after its November failure. The subsequent Ceres-2 loss showed how a company can restore one launcher while still facing difficulties introducing another.

Galactic Energy is also preparing the first flight of Pallas-1, its first launcher using liquid propellant. Pallas-1 is designed to carry approximately 8,000 kilograms to LEO, allowing the company to compete for larger satellite deployment contracts.

The company has filed initial paperwork toward an initial public offering (IPO). An IPO allows a company to sell shares to public investors, providing access to capital for development and expansion. Ceres-2's investigation now sits alongside preparations for Pallas-1 and those financing plans.

India's Indian Space Research Organisation (ISRO), together with emerging private launch providers in Japan and Australia, forms part of the regional competition as China works to maintain its dominant position in the launch schedule. India also suffered a failure earlier in the same week: an anomaly resulted in the loss of the PSLV-C62 rocket and 16 satellites.

The Indian loss and China's two failures illustrate the technical difficulty of delivering spacecraft reliably, even as governments and businesses seek more frequent launches.

A record year sets a demanding pace

China's 92 orbital launch attempts in 2025 carried more than 300 spacecraft into orbit and included successful debuts of new launchers. Government, military and commercial missions all contributed to a program that had grown more varied as well as more active.

Private companies attempted 16 commercial launches, most of them successful. CASC conducted 73 state launches. The growing commercial contribution showed a sector taking on a larger role alongside established state contractors.

China recorded only two launch failures across the whole of 2025. By January 17, 2026, it had suffered two losses in less than 12 hours. That contrast places reliability at the center of preparations for a much busier year.

Launch activity has risen every year since 2019. Rockets departed from Wenchang, Jiuquan, Taiyuan and Xichang in 2025, often with only days between missions. The pace was described as "nothing short of industrial," requiring trained personnel, coordinated supplies and equipment, and rapid preparation of launch facilities.

China expects more than 100 launches in 2026 for the first time. CASC alone aims to increase its activity from 73 state launches in 2025 to more than 100 in 2026, with commercial flights adding to the national total.

Planned activity includes new launchers with potential for reuse: Landspace's Zhuque-3, Space Pioneer's Tianlong-3, iSpace's Hyperbola-3, Deep Blue Aerospace's Nebula-1, Galactic Energy's Pallas-1 and Orienspace's Gravity-2. Reuse means recovering rocket hardware so that it can fly again rather than being discarded after one mission.

The two January failures bring renewed attention to whether quality control and oversight can keep pace with expanding operations. Investigations into the individual vehicles will guide the practical decisions about repairs, future flights and launch scheduling.

Satellite ambitions stretch far beyond current capacity

China's recent ITU paperwork outlines ambitions totaling more than 200,000 satellites. Two proposed constellations, CTC-1 and CTC-2, account for nearly 200,000 of that number: each would contain 96,714 satellites distributed across 3,660 orbits.

The two networks, backed by the Chinese government, would together contain 193,428 satellites. A constellation is a group of spacecraft operating as a network, often to provide communications across a wide area. ITU filings concern the radio frequencies and orbital arrangements needed for those services.

For comparison, approximately 14,300 active satellites were in orbit at the time, including about 9,400 SpaceX Starlink satellites operating across a handful of orbits. SpaceX has filed with the ITU for as many as 42,000 satellites.

Victoria Samson of the Secure World Foundation suggested that China's huge filing might represent a:

land grab

Her assessment concerned securing orbital spectrum rights. Experts considered deployment on the proposed scale almost impossible with current launch capabilities. Placing 200,000 satellites in orbit over seven years would require more than 500 satellites a week and hundreds, possibly thousands, of launches annually.

That demand would far exceed even China's ambitious projected growth in launches. The filings nevertheless show the scale of its interest in satellite internet services serving tens or hundreds of millions of users, and in the ability to influence the global flow of information.

Guowang advances while orbital tensions grow

China had already completed other missions in January before the two failures. A Long March 6A launched Yaogan-50 (01) into an unusual highly retrograde orbit, meaning it traveled strongly against the direction of Earth's rotation.

A subsequent Long March 8A carried nine satellites for Guowang, China's planned constellation of nearly 13,000 spacecraft in low Earth orbit. Guowang is a commercial and strategic response to Starlink. It had 145 operational satellites in orbit, with its operator planning to reach 400 by 2027.

Guowang satellites are estimated to have relatively high mass, suggesting possible roles beyond communications, including navigation, remote sensing and space situational awareness. Remote sensing gathers information about Earth from a distance. Space situational awareness involves tracking objects and activity in orbit.

China is also expanding Qianfan, known as G60. Both Qianfan and Guowang are projects backed by the state, making satellite deployment a matter of national policy as well as commercial competition.

Earlier in January, Chinese representatives warned the United Nations Security Council about the rapid, unregulated expansion of commercial satellite constellations, specifically Starlink. They described the risks to global orbital stability as:

pronounced safety and security

The statement cited incidents in 2021 when China's Tiangong space station had to perform what it called:

preventive collision avoidance control

Those maneuvers were undertaken to avoid Starlink satellites. China's representative argued that inadequate international regulation was a major driver of collision risks, drawing particular attention to spacecraft with limited ability to change their paths:

for spacecraft operated by developing countries that lack orbit-control capability... this undoubtedly constitutes a major risk.

China's warnings accompany the expansion of its own competing networks. Reliable launches, the ability to manufacture satellites in large numbers and safe operations in increasingly crowded orbits are all part of the same effort. The January losses interrupt that effort at the point where spacecraft must first reach space.

The Bottom Line

• Long March 3B and Ceres-2 failed less than 12 hours apart on January 16 and 17, destroying their payloads.
• CASC identified a third stage anomaly in the Long March 3B mission carrying Shijian-32. Both failures remained under investigation as of January 18.
• The Long March 3B loss was its first complete mission loss since April 2020 and appears to have ended approximately 300 Long March launches without an outright failure.
• Shared upper stage technology means the investigation could affect Long March 7A schedules as well as Long March 3B flights.
• Ceres-2's failed debut was believed to carry approximately six satellites, including Lilac-3, while Galactic Energy also prepares Pallas-1 and pursues an IPO.
• China made 92 orbital launch attempts in 2025, with two failures, and expects to exceed 100 launches in 2026.
• China's proposed satellite networks require deployment capacity far beyond its current launch rate, while Guowang aims to grow from 145 operational satellites to 400 by 2027.