China got the booster back.
On July 10, 2026, a Long March 10B rocket launched from Hainan, placed a payload into orbit, and brought its first stage back to an offshore platform.
Instead of landing on legs, the booster descended into a large net and was caught by hooks attached to the stage.[1]
So is China now operating a reusable rocket?
Not yet in the operational sense. Recovery has been proved. Reflight of the same hardware is the next test.
A landing makes the video. The economics are decided later, inside the inspection hangar, on the next launch, and across many repeated flights.
Recovery is the first milestone. Reflight and repetition determine whether reuse becomes an operating system.
What Did Long March 10B Actually Prove?
It proved much more than a low-altitude landing test.
The rocket completed a real orbital mission while attempting recovery on the same flight. CASC says the flight validated high-precision navigation and control, engine restart at altitude, propellant-management technologies, and the offshore net-capture system.[1]
| Question | Status after July 10, 2026 |
|---|---|
| Can it deliver a payload to orbit? | Yes — demonstrated |
| Can the first stage return under control? | Yes — demonstrated |
| Can the offshore net capture it? | Yes — demonstrated once |
| Can the same stage fly again? | Not yet publicly verified as of September 20, 2026 |
| Can it repeat the cycle reliably? | Not yet demonstrated |
| Does reuse lower operating cost? | Not yet demonstrated publicly |
CASC says it expects a first-stage reuse flight before the end of 2026.[1]
One Technical Detail Is Easy to Get Wrong: What Fuel Does It Use?
Long March 10B uses different propellants in its two stages.
- First stage: liquid oxygen + kerosene
- Second stage: liquid oxygen + methane
The first stage is the part that returned and was caught by the net.[1]
Why Catch a Rocket in a Net Instead of Landing on Legs?
The main attraction is mass.
Landing legs have to travel with the booster. A net-capture system moves more recovery hardware off the rocket and onto the recovery vessel. Reuters reported that the concept is intended to reduce vehicle mass and preserve payload capacity.[2]
But the complexity does not disappear. It moves to the ship, net, cables, hooks, sea operations, and post-capture handling.
Long March 10B approaches the offshore net-capture platform on July 10, 2026. Video still: CCTV, via Space.com.
| Recovery choice | What stays on the rocket | What the system must solve |
|---|---|---|
| Landing legs | Deployable legs and support structure | Stable touchdown, leg loads, pad/deck operations |
| Net capture | Hooks and capture interfaces | Ship position, sea state, cable loads, capture accuracy, handling |
Is the Net Better Than Falcon 9’s Landing Legs?
We do not have enough operational evidence to say.
The useful comparison is not which landing looks smarter. It is which architecture can recover, inspect, refly, and repeat with the least time and cost.
What Changed After Zhuque-3 Landed in August?
July answered one question: Can China recover an orbital-class first stage?
Then the picture changed again.
On August 18, 2026, Chinese commercial launch company LandSpace landed the first stage of its Zhuque-3 rocket on deployable legs after an orbital mission.[3] Reuters reported that LandSpace planned to reuse the recovered stage within six months.
China therefore ended the summer with two different orbital-booster recovery architectures:
- Long March 10B: state-developed, offshore net capture
- Zhuque-3: commercial, leg-based land recovery
The question is no longer whether China can bring a booster back. It is whether one of these systems can refly the same hardware repeatedly.
How Far Is China from Operational Reuse?
The cleanest benchmark is not rocket size. It is repeated useful flights by the same hardware.
On August 25, 2026, SpaceX launched Falcon 9 booster B1067 for its 37th flight. Spaceflight Now counted the mission as the 654th Falcon booster landing at that point.[4]
By mid-September, reporting counted more than 650 Falcon booster landings and six boosters with more than 30 flights each.[5]
Long March 10B has proved that its recovery architecture can work once. Falcon 9 has proved that recovery can become a repeated operating process.
This does not make the Chinese milestone small. It tells us what the next benchmark should be.
Why Does Reflight Matter More Than Another Landing?
A recovered booster is finally available for the test that simulation cannot fully replace: inspection of real flight hardware.
Engineers can examine engine wear, tanks, structure, valves, plumbing, avionics, thermal damage, grid fins, and the loads created by the recovery hooks.
The next chain is:
inspect → service → certify → launch again
The key question is whether that can happen without rebuilding so much of the vehicle that reuse loses its value.
What Actually Makes Reuse Cheaper?
Getting the rocket back is not enough.
Effective booster cost per flight ≈ build cost spread across useful flights + inspection + refurbishment + recovery operations + replacement parts
Reuse becomes more attractive when one booster flies many missions, inspection is fast, few parts need replacement, recovery is reliable, and enough payloads exist to keep the fleet busy.
CASC says Long March 10B’s reusable configuration can carry about 16 metric tons to low Earth orbit and is intended for missions including low-orbit satellite-constellation deployment and large commercial satellites.[1]
A reusable rocket needs something to launch often enough for reuse to matter.
The Reuse Maturity Ladder
| Step | What it proves | Long March 10B — Sep. 20, 2026 |
|---|---|---|
| 1. Recover | Guidance, relight, descent and terminal recovery work | Proved |
| 2. Inspect | Real flight wear can be measured | Public detail limited |
| 3. Refly | Same hardware returns to mission service | Not yet publicly verified |
| 4. Repeat | Reflight is not a one-off | Not yet |
| 5. Shorten turnaround | Inspection and servicing become efficient | Not yet demonstrated |
| 6. Build cadence | Launch, recovery and processing become routine | Not yet demonstrated |
| 7. Prove economics | Reuse lowers cost or raises availability enough to matter | Not yet demonstrated publicly |
Landing is a milestone. Reflight is a capability. Cadence is a business system.
What Should You Watch Next?
- Same-booster reflight: Does the recovered Long March 10B stage actually launch again?
- Turnaround time: How long does inspection and servicing take?
- Refurbishment: How much hardware must be repaired or replaced?
- Recovery reliability: Can the net work repeatedly in real sea conditions?
- Flights per booster: Does one stage fly two times, five times, ten times, or more?
- Launch cadence: Are there enough missions to use the reusable fleet frequently?
How Should an Ordinary Reader Read the Next Headline?
If a headline says “reusable rocket successfully landed,” ask one more question:
Did the same hardware fly again?
If yes, ask how long the turnaround took. Then ask how many times the stage repeated the cycle.
Only after that does the economic question become clearer.
The Main Idea
China’s July 2026 Long March 10B recovery was a real engineering milestone.
Then LandSpace added a second Chinese orbital-booster recovery in August.
That means the question changed quickly.
In July, the question was: “Can China recover an orbital booster?”
In September, the better question is: “Can China refly one repeatedly and turn recovery into an operating system?”
The next important video may not be another landing. It may be the same booster leaving the launch pad again.
Continue Reading
Reusable Rockets Explained: Technology, Economics, and the Global Race — Compare recovery, reflight, launch cadence, and economics across reusable launch systems.
Key English Words
- recovery: bringing a rocket stage back after flight
- reflight: flying hardware that has already completed a previous mission
- turnaround time: the time required to prepare recovered hardware for another mission
- refurbishment: inspection, repair, replacement, and servicing performed before reuse
- cadence: how frequently a launch system can operate
- dry mass: vehicle mass excluding usable propellant and payload
Sources
- China Aerospace Science and Technology Corporation — Long March 10B maiden flight and controlled recovery — mission, vehicle specifications, stage propellants, and planned reuse flight.
- Reuters — China successfully tests sea-based rocket booster recovery system — independent reporting on the July 10 recovery and net-capture concept.
- Reuters — LandSpace lands Zhuque-3 booster — China’s August 2026 leg-based land recovery and planned reuse timeline.
- Spaceflight Now — Falcon 9 booster B1067’s 37th flight — operational reuse benchmark and Falcon booster landing count.
- Space.com — Falcon 9 booster landing and reuse record — September 2026 operational-reuse context.
- OrbitalIntel — China’s reusable rocket programmes, updated September 2026 — programme-status cross-check.
Status checked September 20, 2026. Recovery and reflight plans can change quickly. “Reusable” can describe a vehicle designed for reuse; this article uses “operational reuse” for hardware that has actually been recovered and flown again.