4 ms·

I'm big on following Starship development, but I think people are a bit overly optimistic about what can be achieved with it in the next couple of years (while

I'm big on following Starship development, but I think people are a bit overly optimistic about what can be achieved with it in the next couple of years (while having a blind eye on what it will be able to do within a 20y timeframe). From current development and past experience at human rating their Dragon I think that it will take 7-10 years[1] for Starship to reach that level. But once it does reach it, it will make pretty much everything else obsolete and it will start a new spaceflight era, where ISRU and space mining suddenly become a thing within just a few years. [1] take Falcon 9 as a reference - it first launched in 2010 and their first manned mission was only last year in 2020. yes, SpaceX has grown significantly since then, but so have their ambitions with Starship. Fuel transfer and 2nd stage reentry/landing are significant further milestones to reach besides just getting it to orbit. And then they'll have to prove safety margins, life support systems, abort scenarios...

SEJeff 5y ago
But Falcon 9 also took a long time because it doesn't launch all that often and SpaceX had a lot to learn about how to work with NASA. The "idea" of Starship / Superheavy is that the same first and second stage can get caught by the landing tower "chopsticks", have a crane re-mate them, fuel them, and relaunch them within 24 hours or less. If they can relaunch the same entire stack more than say 2 times per day, they're going to absolutely hit the reliability margins[1] that NASA / the FAA want to see in a very short time period. [1] https://en.wikipedia.org/wiki/Human-rating_certification https://en.wikipedia.org/wiki/Human-rating_certification The NASA CCP human-rating standards require that the probability of a loss on ascent does not exceed 1 in 500, and that the probability of a loss on descent did not exceed 1 in 500. The overall mission loss risk, which includes vehicle risk from micrometeorites and orbital debris while in orbit for up to 210 days, is required to be no more than 1 in 270.[3] Maximum sustained acceleration is limited to 3 g.
- voakbasda 5y ago
  I would very much like to see the data and math used to calculate these odds.
  - cryptonector 5y ago
    According to Feynman, in the case of the shuttle there was no such calculation, just assertion.
    - SEJeff 5y ago
      The shuttle progam's motto was "beat the russians" more or less, so that checks out. There is a lot of background in the biographies of both Chuck Yaeger AND John Glenn about some of this if you like that sort of thing.
      - panick21_ 5y ago
        > The shuttle progam's motto was "beat the russians" more or less, so that checks out. And they achieved the exact opposite.
bryanlarsen 5y ago
That's a general rule of thumb in tech predictions -- people overpredict the short term and underpredict the long term. So yes, Dear Moon is unlikely to fly in 2023. But it'll definitely fly before 2028. Starship has already retired their most significant technical risks. Fuel transfer and 2nd stage reentry aren't easy, but the Shuttle and ISS did them both in the 80s. SpaceX didn't do significant work on Dragon until 2015 since Congress didn't properly fund Commercial Crew until then.
numpad0 5y ago
Am I the only one to think Starship is starting to look a lot like X-33 program...
- rbanffy 5y ago
  A bit like Salvage-1 ;-) https://www.imdb.com/title/tt0078681/ https://www.imdb.com/title/tt0078681/
- panick21_ 5y ago
  I mean, the X-33 was sub-orbital technology demonstrator. It was totally different in terms of reentry and control control surfaces. It uses different fuel. It doesn't use an areospike engine. Its a different color. Not sure what you are seeing.
Robotbeat 5y ago
I disagree. Starship HLS with requires 10 launches per mission. An extra expendable Starship/SuperHeavy launch with Orion on top is not a substantial challenge in comparison (although lots of engineering involved, but they’re good at that).

SEJeff 5y ago

But Falcon 9 also took a long time because it doesn't launch all that often and SpaceX had a lot to learn about how to work with NASA. The "idea" of Starship / Superheavy is that the same first and second stage can get caught by the landing tower "chopsticks", have a crane re-mate them, fuel them, and relaunch them within 24 hours or less. If they can relaunch the same entire stack more than say 2 times per day, they're going to absolutely hit the reliability margins[1] that NASA / the FAA want to see in a very short time period. [1] https://en.wikipedia.org/wiki/Human-rating_certification https://en.wikipedia.org/wiki/Human-rating_certification The NASA CCP human-rating standards require that the probability of a loss on ascent does not exceed 1 in 500, and that the probability of a loss on descent did not exceed 1 in 500. The overall mission loss risk, which includes vehicle risk from micrometeorites and orbital debris while in orbit for up to 210 days, is required to be no more than 1 in 270.[3] Maximum sustained acceleration is limited to 3 g.

voakbasda 5y ago
I would very much like to see the data and math used to calculate these odds.
- cryptonector 5y ago
  According to Feynman, in the case of the shuttle there was no such calculation, just assertion.
  - SEJeff 5y ago
    The shuttle progam's motto was "beat the russians" more or less, so that checks out. There is a lot of background in the biographies of both Chuck Yaeger AND John Glenn about some of this if you like that sort of thing.
    - panick21_ 5y ago
      > The shuttle progam's motto was "beat the russians" more or less, so that checks out. And they achieved the exact opposite.

bryanlarsen 5y ago

That's a general rule of thumb in tech predictions -- people overpredict the short term and underpredict the long term. So yes, Dear Moon is unlikely to fly in 2023. But it'll definitely fly before 2028. Starship has already retired their most significant technical risks. Fuel transfer and 2nd stage reentry aren't easy, but the Shuttle and ISS did them both in the 80s. SpaceX didn't do significant work on Dragon until 2015 since Congress didn't properly fund Commercial Crew until then.

numpad0 5y ago

Am I the only one to think Starship is starting to look a lot like X-33 program...

rbanffy 5y ago
A bit like Salvage-1 ;-) https://www.imdb.com/title/tt0078681/ https://www.imdb.com/title/tt0078681/
panick21_ 5y ago
I mean, the X-33 was sub-orbital technology demonstrator. It was totally different in terms of reentry and control control surfaces. It uses different fuel. It doesn't use an areospike engine. Its a different color. Not sure what you are seeing.

Robotbeat 5y ago

I disagree. Starship HLS with requires 10 launches per mission. An extra expendable Starship/SuperHeavy launch with Orion on top is not a substantial challenge in comparison (although lots of engineering involved, but they’re good at that).