The Hefei company is betting that electric motors driving propellant pumps, an approach Rocket Lab used on its small Electron rocket and then dropped for its larger follow on, can carry 1,500 kilograms to orbit by 2027.
Spark Space, a Chinese startup founded in 2024 in Hefei, has hot-fired a rocket engine that uses electric motors to drive its propellant pumps and this week closed a fresh Pre-A+ funding round, positioning itself as the first credible non-Rocket Lab entrant to ask whether the electric-pump architecture can scale beyond a single orbital vehicle.
The Lieyan-2 engine burned in a 20-second test announced in early March, according to SpaceNews, and the Cathay Capital-led Pre-A+ announced June 18 follows a Pre-A of about 100 million yuan ($14.8 million) on June 1 led by Yunze Capital and Orbital Chenguang. Spark Space has now run an angel round, an angel+ round, and the two Pre-A rounds inside roughly twelve months, and has staked its scale-up on an engine cycle that, on paper, should be simpler and lighter than a turbopump or gas generator.
That simplicity is the whole point. An electric-pump engine uses battery- or motor-driven propellant pumps, which strips out the gas generator and most of the plumbing that a traditional staged-combustion or gas-generator cycle needs. The tradeoff is real: as the engine gets bigger, the motors and batteries start to weigh more than the turbomachinery they replace, and the specific impulse (a measure of fuel efficiency) tends to drop. Rocket Lab reached orbit with the architecture on Electron, which carries about 300 kilograms to low Earth orbit, and then walked away from the cycle for Neutron, its planned medium-lift follow-on, after concluding the cycle does not scale to a more powerful vehicle. SpaceNews reported in February that Neutron's debut has slipped to late 2026 at the earliest, after a composite-tank failure on January 21.
Spark Space's pitch is that the cycle does scale; it just has not been tried yet by anyone other than Rocket Lab. Lieyan-2, burning kerosene and liquid oxygen, is rated by the company at roughly 10 tonnes of thrust per engine, against Electron's Rutherford at about 2.4 tonnes at sea level. The proposed Jinhua-1, also called Evolution-1, would cluster nine Lieyan-2s on the first stage and a vacuum-optimized Lieyan-2 on the second, for a 90-tonne liftoff thrust and a 27.5-meter, 2.25-meter-diameter airframe. The 1,500-kilogram-to-LEO target is roughly five times Electron's payload to the same orbit.
The "world's largest electric-pump-fed rocket" framing is Spark Space's own claim, and the company has not disclosed chamber pressure, specific impulse, throttling range, or the full duration of the March test. The only public artifact is a 20-second video clip, and independent benchmark data is not yet available.
The competitive context is crowded. China already has three solid-fueled small launchers flying regularly: Galactic Energy's Ceres-1, CAS Space's Kinetica-1, and Expace's Kuaizhou-1A. Solid-fueled rockets trade payload for simplicity and quick turn-around, and they have flight heritage Spark Space does not yet have. The Hefei startup is building a production and test footprint of more than 10,000 square meters in the Hefei National High-Tech Industry Development Zone, with a 20,000-square-meter expansion underway, and has drawn staff from CASC, CASIC, and the Chinese Academy of Sciences, per Chinese tech outlet 36Kr as cited by SpaceNews.
The investor map is the second story. Cathay Capital, the lead on this week's Pre-A+, also appears in Orbital Chenguang's Pre-A1. Orbital Chenguang, a Beijing-based space-computing startup incubated by the Beijing Astro-future Institute of Space Technology, is building what it calls a dawn-dusk sun-synchronous orbit constellation to host orbital data centers, and on April 20 disclosed $8.4 billion in credit lines from 12 Chinese banks including Bank of China, Agricultural Bank of China, Bank of Communications, SPDB, and CITIC Bank, per SpaceNews. The alignment is a hypothesis, not a fact: the same investor backing a launch startup and a constellation buyer could be threading the supply and demand for orbital compute into one portfolio. Spark Space has not named Orbital Chenguang as a customer.
What to watch next is specific. Spark Space's 2027 debut is a company target, not a milestone achieved, and the company still has to extend Lieyan-2 beyond a 20-second firing, prove throttling and restart, and qualify the first-stage cluster before any flight. Rocket Lab's own abandonment of the cycle for Neutron is the closest available proxy for the engineering ceiling Spark Space is betting against, and Neutron's first flight is now, after the January tank failure, slipped to late 2026 at the earliest. If Lieyan-2 stays at a 20-second firing and a 10-tonne thrust claim, the scaling question will answer itself.