Spacex Starship IFT-2 launch today

[coppice]

Speculation is that the relight of the booster engines happened without stable acceleration to keep the fuel from floating/sloshing around in the tanks and many engines inhaled air and exploded. Resulting in an inability to continue the booster return and water landing mission, which triggered FTS.

     That shouldn't have been a problem.  Ullage engines have been used for MANY years to provide a slight amount of acceleration so that the fuel and oxidizer are pushed to the bottom of the tanks so that they will be properly picked up for engine re-ignitions.

The booster was twisting around at the time, which makes it pretty hard to get the fuel in the right part of the tanks. Most rockets which start twisting around quickly trigger their self destruct. This is not uncharted territory for Space X, though. This is an issue they also face with the Falcon, where they seem to have it sorted out very well. You might have expected them to have carried over what has proven reliable there, and nailed this on the first try.

It's very different from Falcon though. Different fuel, very different engines with different ignition mechanism, autogenous pressurization instead of helium, pretty much uncharted territory in every corner.

Sure, but they are much better positioned to get this right than anyone else. They are the only ones with experience. Sloshing is going to be the big issue, and that should be a fairly similar issue between the two designs.

[CatalinaWOW]

Truly amazing how much failure analysis can be done with zero telemetry.  Hopefully the guys with actual data can narrow it down.  Could be lots of different things. But lots of good data.

[Psi]

On a side note, I recently heard the Nasa versus SpaceX dynamic being described as, "it's rocket scientists versus space cowboys." Yee har...

This year "space cowboys" put 4x more mass to the orbit (1000 tons) than the rest of the world combined. Also they are the only US launch provider that can currently deliver astronauts to ISS (for NASA you just mentioned).

I more prefer the NASA vs SpaceX analogy of Scientists vs Engineers.

A scientist takes their time to do lots of theoretical testing and report writing until everything is proved to be perfect before building anything.
The engineer just tries shit until it works.

It's an over simplification, but is mostly accurate.

[CatalinaWOW]

On a side note, I recently heard the Nasa versus SpaceX dynamic being described as, "it's rocket scientists versus space cowboys." Yee har...

This year "space cowboys" put 4x more mass to the orbit (1000 tons) than the rest of the world combined. Also they are the only US launch provider that can currently deliver astronauts to ISS (for NASA you just mentioned).

I more prefer the NASA vs SpaceX analogy of Scientists vs Engineers.

A scientist takes their time to do lots of theoretical testing and report writing until everything is proved to be perfect before building anything.
The engineer just tries shit until it works.

It's an over simplification, but is mostly accurate.

Neither end of that binary is optimum for most things.  The real art is figuring out where in the middle of that spectrum best fits the job at hand.  Close to the science extreme is appropriate for a deep space probe that will cost billions of dollars and take advantage of a once in a century planetary configuration.  Close to the other end is appropriate for an RC toy that will be built in the millions and where consequences of failure are nil. 

More interesting question is the best approach for problems like a WiFi light switch that will be built in huge volume but could kill many people if screwed up.   What is novel about SpaceX is their vision that the application volume will be large enough to justify enormous spending on trial and error.   It worked on Falcon9.  The jury is still out on Starship, but from my point of view it looks like they might well end up making much more than they spend.

[wraper]

What is novel about SpaceX is their vision that the application volume will be large enough to justify enormous spending on trial and error.   It worked on Falcon9.  The jury is still out on Starship, but from my point of view it looks like they might well end up making much more than they spend.

The thing is that SpaceX's trial and error is cheaper than current "traditional" rocket development which in the last decades was more like industry rotting from inside due to complacency and government bailouts. Look no further than obscene spending on SLS with all old tech, including 40 years old engines. https://www.space.com/nasa-sls-rocket-artemis-moon-plans-unaffordable-gao-report

In late 2021, a report by NASA's Office of Inspector General showed that NASA will likely spend a total of $93 billion on the Artemis program between 2012 and 2025, and that each SLS launch will cost about $4.1 billion. A large chunk of the budget was attributed to hiring contractors in every U.S. state and more than 20 similar partners across Europe.

[coppice]

On a side note, I recently heard the Nasa versus SpaceX dynamic being described as, "it's rocket scientists versus space cowboys." Yee har...

This year "space cowboys" put 4x more mass to the orbit (1000 tons) than the rest of the world combined. Also they are the only US launch provider that can currently deliver astronauts to ISS (for NASA you just mentioned).

I more prefer the NASA vs SpaceX analogy of Scientists vs Engineers.

A scientist takes their time to do lots of theoretical testing and report writing until everything is proved to be perfect before building anything.
The engineer just tries shit until it works.

It's an over simplification, but is mostly accurate.

Its mostly BS. Science is mostly stumbling around until you hit on something new and interesting. Good engineering is mostly using the understanding that comes from science, and applying it in a methodical way, with a lot of testing to demonstrate your thinking didn't go wrong. NASA started life with scientists, and a LOT of stuff blowing up on the launch pad or soon after. The engineering people in the air force used to mock the NASA people's lack of faith in their own work with famous comments like "When a rocket goes up, the NASA people cheer". The big problem with NASA has been that bureaucracy inhibited progress. A real engineering organisation would never have let the Challenger disaster happen, as the seeds of that disaster were well know. The design would have been improved to avoid a warranty claims disaster. NASA just flew the poor design until people died.

[tom66]

What is novel about SpaceX is their vision that the application volume will be large enough to justify enormous spending on trial and error.   It worked on Falcon9.  The jury is still out on Starship, but from my point of view it looks like they might well end up making much more than they spend.

The thing is that SpaceX's trial and error is cheaper than current "traditional" rocket development which in the last decades was more like industry rotting from inside due to complacency and government bailouts. Look no further than obscene spending on SLS with all old tech, including 40 years old engines. https://www.space.com/nasa-sls-rocket-artemis-moon-plans-unaffordable-gao-report

In late 2021, a report by NASA's Office of Inspector General showed that NASA will likely spend a total of $93 billion on the Artemis program between 2012 and 2025, and that each SLS launch will cost about $4.1 billion. A large chunk of the budget was attributed to hiring contractors in every U.S. state and more than 20 similar partners across Europe.

ULA is still using rocket engines dating back to the days of the USSR, the RD-180.  A major headache is being able to re-engineer that product so that it can be built in the west, given the, err... somewhat less cordial relationship the US and Russia now have following their actions in Ukraine.  Apparently they have under 20 engines left.   Given they throw away the rocket every time, that's not much cargo.  Meanwhile SpaceX is reusing Falcon 9 at least 10 times over and trying to discover what the limit actually is.

Pork barrel politics is very toxic in the US, which is why NASA finds companies like SpaceX so compelling to work with.  It's up to them how they spend the billions they get, and once the contract is signed the fed is on the hook for the funds either way.   Privatisation in government contracting often has negative effects but SpaceX shows great exceptions happen.

[wraper]

ULA is still using rocket engines dating back to the days of the USSR, the RD-180.  A major headache is being able to re-engineer that product so that it can be built in the west, given the, err... somewhat less cordial relationship the US and Russia now have following their actions in Ukraine.  Apparently they have under 20 engines left.   Given they throw away the rocket every time, that's not much cargo.  Meanwhile SpaceX is reusing Falcon 9 at least 10 times over and trying to discover what the limit actually is.

Pork barrel politics is very toxic in the US, which is why NASA finds companies like SpaceX so compelling to work with.  It's up to them how they spend the billions they get, and once the contract is signed the fed is on the hook for the funds either way.   Privatisation in government contracting often has negative effects but SpaceX shows great exceptions happen.

RD-180 first flight happened in 2000. It's based on earlier RD-170 which still is not nearly as old as RS-25. The main difference is that RS-25 is a very expensive reusable engine, but in SLS they become single use  . Not to say SLS will be not only reusing old RS-25 engines in stock, their production is restarted basically from scratch while RD-180 had continuous production. Also RD-180 is 2x more powerful and about 7 times cheaper than newly built RS-25 and still much cheaper than modernized used engines.

[wraper]

Meanwhile SpaceX is reusing Falcon 9 at least 10 times over and trying to discover what the limit actually is.

So far one booster has flown 18 times, one 17 times, two 16 times, one 15 times, one 14 times and 3 other more than 10. Also there were launches for multiple customers on 10+ times flown boosters, not only their own Starlink as initially. So SpaceX should be quite confident about their reliability.

[coppice]

RD-180 first flight happened in 2000. It's based on earlier RD-170 which still is not nearly as old as RS-25.

The RD-180 is just a stripped down RD-170, so in terms of technology its basically the same. The RS-25 is early 70s. The RD-170 is late 70s. Not a lot of difference.

[Dan123456]

@Wrapper said: “The thing is that SpaceX's trial and error is cheaper than current "traditional" rocket development which in  last decades was more like industry rotting from inside due to complacency and government bailouts.”

We do not know if that is true. Musk says it is cheaper… but he also says we will have full self driving cars “next year”… every year for the last decade 

He could easily open the books to the world and show us that is cheaper, but till then I don’t trust a word that comes out of his mouth as 99% of what he says is bullshit!

Also Musk takes soooooo much government money for his companies it isn’t even funny. He is just a massive hypocrite as hold his hand out for hundreds of millions of dollars in government assistance while at the same time saying “companies getting government assistance is bad”.

Edit: I messed up the quote thingy so was trying to fix that… but this will have to do for now 

[wraper]

we do not know if that is true. Musk says it is cheaper…

We know that very well actually. If you actually look into it, before SpaceX it was all about cost+ contracts (like paying $1B annually to ULA for basically existing (worded like for providing launch availability or something of sorts). And SLS being all about politics and contractors spread over all US states to make senators happy. Just before SpaceX came into the game, ULA lost all sense of shame and on average was charging $400+ million per launch of Atlas V, and that was besides receiving $1B/year as mentioned above. Basically no commercial customer was launching from US soil due to utterly ridiculous cost that only US government was ready to pay. SpaceX had to sue US government to get any air force contracts at all. Since then ULA halved their prices but still are not competitive. As of SpaceX, there is plenty of info how much money they receive from the government, and for the services they provided it's not that much, especially compared to others (2-3 times less for the same job).

[CatalinaWOW]

@Wrapper said: “The thing is that SpaceX's trial and error is cheaper than current "traditional" rocket development which in  last decades was more like industry rotting from inside due to complacency and government bailouts.”

We do not know if that is true. Musk says it is cheaper… but he also says we will have full self driving cars “next year”… every year for the last decade 

He could easily open the books to the world and show us that is cheaper, but till then I don’t trust a word that comes out of his mouth as 99% of what he says is bullshit!

Also Musk takes soooooo much government money for his companies it isn’t even funny. He is just a massive hypocrite as hold his hand out for hundreds of millions of dollars in government assistance while at the same time saying “companies getting government assistance is bad”.

Edit: I messed up the quote thingy so was trying to fix that… but this will have to do for now 

I think I detect a little animosity towards Musk here

You are right.  We don't have a detail breakdown of SpaceX costs for Starship or anything else.   But we do know that NASA spent $92 Billion for 1 SLS launch to date.  And to their credit for $92 billion we got a successful launch.  And we have some sort of idea of the total cash flows into the Musk companies.  Which indicates expenditures of single digit billions per year.  If they get the Starship worked out in less than a decade his method will be cheaper.  It seems likely to be far less than a decade, and the annual costs are best guessed in the few billion a year meaning that his method will have turned out far cheaper.  For a far better product. 

Don't let your anger over what Musk did to Twitter or the oversold claims for underground transportation and solar rooftops blind you to the things that have turned out pretty well.  Teslas are among the best electric cars out there, Falcon9 is a solid space launch system and the real cost for launch is less than half of the competitors, possibly far less than that.  Users of Starlink are pretty darned happy with that product and even the Russian astronauts like the Dragon better as a way to get to the ISS.

[Stray Electron]

Don't let your anger over what Musk did to Twitter or the oversold claims for underground transportation and solar rooftops blind you to the things that have turned out pretty well.  Teslas are among the best electric cars out there, Falcon9 is a solid space launch system and the real cost for launch is less than half of the competitors, possibly far less than that.  Users of Starlink are pretty darned happy with that product and even the Russian astronauts like the Dragon better as a way to get to the ISS.

   I'm not a fan of the Musk and I don't want this thread to go further askew but you DO have to admit that he's accomplished a lot of things that others thought were impossible and he's clearly developed several entirely new technologies.

[coppice]

Here's a fun video if you like rockets self-destructing:



He seems to have only included the ones that self-destructed, but the ones that went wild and had to be destroyed are some of the more amusing failures. There have been a LOT of those, especially in the early days. In the 70s I watched a hour of launches going wild just due to EMC problems, which was part of a training course on the importance of taking EMC seriously in design, I think produced by NATO. It might have been the US military.