What's new

SpaceX’s Chinese rivals had a theory about Starship rocket explosion, and Elon Musk has confirmed it

Nan Yang

SENIOR MEMBER
Joined
May 1, 2010
Messages
5,252
Reaction score
1
Country
Malaysia
Location
Malaysia

SpaceX’s Chinese rivals had a theory about Starship rocket explosion, and Elon Musk has confirmed it

  • Starship’s ‘intelligent control’ appears to have gone offline when it was needed most, according to analysis by Chinese scientists developing similar rocket
  • Chinese team says it relied mostly on video for analysis because SpaceX’s telemetry appeared to be wrong

Stephen Chen
Stephen Chen in Beijing
Published: 2:00pm, 9 May, 2023

SpaceX’s next-generation Starship spacecraft atop its powerful Super Heavy rocket lifts off from the company’s launchpad on an uncrewed test flight before exploding near Brownsville, US, on April 20, 2023. Photo: EPA-EFE

SpaceX’s next-generation Starship spacecraft atop its powerful Super Heavy rocket lifts off from the company’s launchpad on an uncrewed test flight before exploding near Brownsville, US, on April 20, 2023. Photo: EPA-EFE

SpaceX’s Starship rocket suffered a catastrophic spin during its first test flight on April 20, exploding and breaking apart just 239 seconds after lift-off.

Before SpaceX released an official statement on the cause of the failure, a Chinese rival of the company claimed that the intelligent flight control system of the rocket could have failed when it was needed most.

According to the Beijing Aerospace System Engineering Institute, if the smart thrust vector control system had worked as planned, the Starship rocket could have reached space, even with multiple engine failures. The system is designed to detect and diagnose problems in real time, allowing for faster and more accurate responses to emergencies.

SpaceX Starship, world’s largest rocket, explodes minutes after lift-off on first test flight​

The Beijing institute is the overall design unit for China’s Long March rockets. An analysis of the Starship launch conducted by the institute was released by official newspaper China Space News on April 27 on its WeChat social media account.

On April 29, Elon Musk confirmed this theory in a Twitter Spaces discussion about the first Starship launch.

The Starship lost thrust vector control at T+85 seconds (85 seconds after the scheduled launch), he said.

“If we had maintained thrust vector control and throttled up, which we should have … then we would have made it to staging,” Musk added.

Where is AI?​

The Starship rocket is the largest ever built, standing at 119 metres (390 feet) tall and weighing around 5,000 tonnes at lift-off. It is designed to be fully reusable and capable of carrying up to 100 people or 100 tonnes of cargo on missions to Mars and other destinations in space.

The rocket’s engines were programmed to throttle down during flight to reduce the load on the rocket. This is a standard procedure used by SpaceX and is similar to that of the Falcon 9 rocket, according to the institute’s report.

However, the analysis by Chinese researchers said that during this flight, some of the engines failed and lost thrust, but the throttling process continued as programmed.

If the intelligent control system had been working, it could have detected the loss of thrust and adjusted the engine output accordingly to maintain the rocket’s climb, they said.

Preventable spin​

SpaceX’s Falcon 9 rocket has flight fault diagnosis and disposal capabilities.

“At least twice during flight test missions, the rocket has been able to shut down faulty engines and complete trajectory replanning, ultimately achieving accurate orbit insertion,” they said in the report.

This showed that “SpaceX has a preliminary understanding of flight fault diagnosis and online disposal technology”.

The Super Heavy Starship should therefore have a certain level of online fault disposal capability to deal with various types of faults that may occur during the first flight process, according to the report.

Frame-by-frame analysis of the launch video suggested some of the engines had failed and produced a disturbance torque during flight at around 80-90 seconds, the report said.

According to the calculations by the Chinese researchers, the remaining 11 engines with swivelling capability only needed to swivel by around 1 degree to compensate for the loss of at least eight fixed engines.

“This is because the swivelling engines have a much greater moment arm than the fixed engines that produce disturbance forces,” they said.

Based on their analysis of the video and mathematical modelling, eight engines had run into trouble, more than the number shown by SpaceX’s telemetry.

But “the rocket’s attitude control capability should be sufficient, and the rocket’s attitude should not have diverged” into a fatal spin, they added.

The Chinese team said it had to rely mostly on video for its analysis because SpaceX’s telemetry appeared to be wrong.

Musk also confirmed this issue. “At T+27 seconds, SpaceX lost communications due to some kind of energy event,” he said.

Musk disagrees with Chinese analysis​


The Starship may have encountered an unexpected fault or anomaly that the AI was not programmed to handle, leading to a shutdown or failure of the system, according to the Chinese researchers.

Starship’s servo system, for instance, uses engine drainage as a driving energy source to operate the 13 swing engines. The engine failure could lead to an energy shortage in the servo system, causing some of the swing servo mechanisms to malfunction and resulting in unstable attitude control of the spacecraft.

The abnormal engine operation could also cause control circuit failures, leading to abnormal servo operation and a loss of effective control force for stabilising the spacecraft’s attitude.

But the fate of Starship engines could have been sealed when high-speed jet streams from the Starship rocket’s engines caused damage to the launch pad during lift-off, said the report.

According to real-time video and post-launch photo analysis, the high-speed jet streams directly damaged the cement foundation of the launch pad, with some rocks flying several hundred metres away into the sea, said the researchers.

The flying debris also caused varying degrees of damage to nearby fuel tanks, cranes and broadcasting vehicles. Some of the Starship’s thermal tiles were found scattered near the launch site.

“As the rocket with the highest take-off thrust in the world, the Starship experiences extremely harsh force, heat and noise environments during lift-off due to the high-temperature and high-speed jet streams from its engines. This may be a reason why some of the engines did not function properly during ignition,” the report said.

But Musk disagreed. He said SpaceX deliberately shut down three engines at the start of launch because conditions were not ideal. SpaceX did not “see evidence that the rock tornado actually damaged engines or heat shields in a material way”, he said.

Rival’s verdict​

The Beijing-based institute is leading the development of Long March 9, a two-stage reusable rocket of a similar size and design to Starship.

The Long March 9 rocket is intended to support China’s deep-space exploration goals, including a crewed lunar landing and future crewed missions to Mars.

“This test is an example of how SpaceX is trying new things and making changes quickly,” said the institute’s research team in the report.

“The test partially assessed the performance of the Super Heavy propulsion system, but because there was no online intelligent response to faults such as thrust decline, it was unable to complete the orbital flight assessment of the Starship spacecraft,” they said.



Stephen Chen
Stephen Chen
Stephen Chen investigates major research projects in China, a new power house of scientific and technological innovation. He has worked for the Post since 2006. He is an alumnus of Shantou University, the Hong Kong University of Science and Technology, and the Semester at Sea programme which he attended with a full scholarship from the Seawise Foundation.
 
But I still think SpaceX is using too many unproven technology together.
This create too many opportunities for errors.
Any new technology is unproven till it works, that's the basis for innovation.
 
Any new technology is unproven till it works, that's the basis for innovation.
You misunderstand me.
Another method is call continuous improvement.

You make incremental improvement in technology but continuously.
 

Back
Top Bottom