totm nov 2023 SpaceX Discussion Thread

[kerbiloid]

2 hours ago, tater said:

They can already build a rocket larger than the whole Shuttle stack every month or so. The next testing regime is orbital reentry, and first stage. Both require expending vehicles until they nail it.

35 engines per test.

Every ~2 days is about the min they need to test at a reasonable pace of a few times a year.

The only thing I can't get in this chain.

Why bother with reusable rockets when you can just build produce the cheap rockets like sausages.

Especially since they are made of cheap iron.

[Beccab]

7 minutes ago, kerbiloid said:

The only thing I can't get in this chain.

Why bother with reusable rockets when you can just build produce the cheap rockets like sausages.

Especially since they are made of cheap iron.

Launch cadence is the main factor into this. Falcon 9 have been seen having a turnaround time of <30 days, while a falcon 9 first stage (according to reddit) takes 4-6 months to produce. Assuming they can make a starship+superheavy in a month and a half and not taking into account parallel construction, that's very far from the aspirational goal of relaunch in a day or even a more realistic goal of reducing the turnaround time to a week-ish.

That, plus the constant growth of the fleet would help both Starlink launches and SpaceX's mars ambitions 

[RCgothic]

An engine every two days doesn't mean the manufacturing process takes just two days. It means they're working on many at once.

Assuming they eventually reach their goal of $250000 per engine and half of that is labour, then that's roughly a worker-year of labour per engine.

If the engine's still costing ~$1m each then that's ~4 worker years per engine.

Even with multiple people working on the engine at once it'll take months to make each engine from start to finish.

[Beccab]

Do we know if they are implementing some kind of automatization for the Raptor production? Admittedly I know very little of how engine construction is done, but seeing how much Tesla factories are automated i'd assume they want to make the assembly as automated as possible

[RCgothic]

Almost certainly the production will be both highly serialised, with a new engine immediately stepping into a manufacturing phase the moment the previous moves to the next, and massively parallel, with multiple engines being worked on at once at each phase of manufacturing.

Edited May 30, 2021 by RCgothic

[RCgothic]

36 minutes ago, kerbiloid said:

The only thing I can't get in this chain.

Why bother with reusable rockets when you can just build produce the cheap rockets like sausages.

Especially since they are made of cheap iron.

This may be the cheapest rocket ever produced at this scale, but it'll still be cheaper to reuse it than not. Methane and LOX are inexpensive fuels.

As long as the recovery cost is not more than the manufacturing cost, and the manufacturing cost is more than a small fraction of the fuel cost, then disposable rockets won't make sense from a cost perspective.

Alternatively, if desired launch cadence exceeds manufacturing capability, then reuse still makes sense even if it's more expensive than building new. Not that I think it would be more expensive, but demand will certainly exceed supply for what spacex has planned.

[RCgothic]

And right on cue:

1 raptor every 2.x days.

 

[kerbiloid]

50 minutes ago, Beccab said:

Falcon 9 have been seen having a turnaround time of <30 days, while a falcon 9 first stage (according to reddit) takes 4-6 months to produce.

Just produce more Falcons at once, and have a hundred of them in the storehouse.

47 minutes ago, RCgothic said:

Assuming they eventually reach their goal of $250000 per engine and half of that is labour, then that's roughly a worker-year of labour per engine.

They should use the additive technologies more actively.

They anyway need this for Mars. No rolled steel can be bought there.

21 minutes ago, RCgothic said:

This may be the cheapest rocket ever produced at this scale, but it'll still be cheaper to reuse it than not. Methane and LOX are inexpensive fuels.

A reusable rocket needs more expensive parts, more exclusively chosen from the manufactured ones.

[YNM]

6 hours ago, SpaceFace545 said:

Either they have magically reduced engine mania factoring time from years to hours or they have a sweat shop pumping these things out.

Or they have enough production lines with enough workers in each of them, and combined they finish one rocket engine after the other one every 48 hrs if averaged out.

Before the pandemic and all, airplane manufacturers (as in, commercial airliners) produce one more aircraft every week or something too. This just shows their commitment to the scale of what they're working, not necessarily that they're making one airliner within one week from scratch to final furnish. Same goes for Raptors.

 

What a difference in production scale makes, everyone.

Edited May 30, 2021 by YNM

[tater]

6 hours ago, RCgothic said:

And right on cue:

We were having the engine cost discussion because of that tweet last night

 

[tater]

 

[RealKerbal3x]

(Probably Elon time, but exciting to see nonetheless)

[SunlitZelkova]

On 5/29/2021 at 6:13 PM, YNM said:

If you can tell which payload is the "hot" one, and which is just a stack of Starlinks...

They would just hit all of them. In fact, even without a missile variant in development, every aerospace site in the US- including SpaceX Boca Chica- is probably a nuclear target.

5 hours ago, RealKerbal3x said:

(Probably Elon time, but exciting to see nonetheless)

You never know. It took only 18 months to convert the Japanese battleship Hyuga into a hybrid aircraft carrier, with 1940s technology and in an empire with dwindling resources. It seems like building a launch pad/ship would be something much easier to figure out the time for and actually do, compared to determining the date of the first crewed flight of Starship or something else space related.

[CollectingSP]

20 hours ago, kerbiloid said:

The only thing I can't get in this chain.

Why bother with reusable rockets when you can just build produce the cheap rockets like sausages.

Especially since they are made of cheap iron.

Coming from an *admittedly*  EELV fan, I must accept defeat. falcon is much cheaper to Re-fly than any (new) Legacy EELV vehicle. that's why they are re-usable.  

Edited May 31, 2021 by CollectingSP

[SpaceFace545]

35 minutes ago, CollectingSP said:

Coming from an *admittedly*  EELV fan, I must accept defeat. falcon is much cheaper to Re-fly than any (new) Legacy EELV vehicle. that's why they are re-usable.  

I think EELV's (I am a ULA fanboy) really allow for the specialization that reusable launchers can't achieve. The Atlas V can launch a GPS satellite one day, a spy satellite the size of a school bus another, then it can send a probe into deep space a week later. All by changing the number of SRBs.

[YNM]

2 hours ago, SunlitZelkova said:

In fact, even without a missile variant in development, every aerospace site in the US- including SpaceX Boca Chica- is probably a nuclear target.

2 miles away from another country at that... hope they have good CEP if they do...

What about the oil-rig launchpads ?

Edited May 31, 2021 by YNM

[SpaceFace545]

27 minutes ago, YNM said:

2 miles away from another country at that... hope they have good CEP if they do...

What about the oil-rig launchpads ?

I honestly doubt that anyone will nuke an offsea launchpad, that just seems impractical. Spacex also isn’t affiliated with any country, it’s private. Starship and their future endeavors don’t really have a nationality.

[Spaceman.Spiff]

8 minutes ago, SpaceFace545 said:

Spacex also isn’t affiliated with any country, it’s private.

I mean….

They’re based in the US and receive national security contracts. Spacex is about as affiliated as a company can get. 

[tater]

I think the problem with comparing EELVs to SpaceX is that while the EELVs can certainly work for specific missions—and ULA has what up to this point is the best upper stage (arguably best stage, period), Centaur—they are built for customers. To make money. So there are missions where the C3 is such that they really shine, and their customers like that.

But from a broader commercial standpoint cost and cadence matter, and the market is just not that big.

Most importantly, ULA (Boeing/LockMart) is in the business of making money.

ULA makes launch vehicles to make money.

SpaceX makes money to make launch vehicles.

 

When they started, they had pretty much no money. They needed customers so they could try and achieve their goals. They now have a reasonable amount of money. They are building Starlink to make money—so they can make giant rockets. Failing to realize this difference will really confuse the way you read the situation. I constantly say "I'm not a colonize Mars guy." I'm not. They are. They are serious about it. Making money is a means to that end, nothing more.

Edited May 31, 2021 by tater

[kerbiloid]

2 minutes ago, tater said:

When they started, they had pretty much no money.

They probably had good friends in NASA.

And as every child knows, friends are better than money, lol.

2 minutes ago, tater said:

They are building Starlink to make money

They claim a Martian colony.
So, Starship 3d-print manufacturing is not a luxury but an inevitable phase of R&D for that.

Edited May 31, 2021 by kerbiloid

[Beccab]

Looks like the 32 engines configuration will be 3-9-20, not 4-8-20 as people expected given the current configuration. Curious

[tater]

2 minutes ago, kerbiloid said:

They probably had good friends in NASA.

And as every child knows, friends are better than money, lol.

They managed to get the COTS contract. $360N they used to develop F9 and Dragon. The idea that they were somehow the lucky kid with rich parents is pretty wrong-headed, the people writing the checks prefer to write them to the companies that then donate to their campaigns, etc. The large defense contractors are rather better at this game than SpaceX. NASA's budget while huge for a civilian space agency is noise level money in the US compared to DoD.

 

2 minutes ago, kerbiloid said:

They claim a Martian colony.
So, Starship 3d-print manufacturing is not a luxury but an inevitable phase of R&D for that.

No idea about the Mars stuff, but they already 3D print engine components, so sure.

[kerbiloid]

6 minutes ago, tater said:

The idea that they were somehow the lucky kid with rich parents is pretty wrong-headed,

The idea was not that. The idea is that somebody in NASA wants to have a NASA Corp, so SpaceX is a good protege.

As well as Axiom.

[tater]

3 minutes ago, kerbiloid said:

The idea was not that. The idea is that somebody in NASA wants to have a NASA Corp, so SpaceX is a good protege.

As well as Axiom.

Nope. To borrow from Pauli, "It is not even wrong."

 

[CollectingSP]

1 hour ago, SpaceFace545 said:

I think EELV's (I am a ULA fanboy) really allow for the specialization that reusable launchers can't achieve. The Atlas V can launch a GPS satellite one day, a spy satellite the size of a school bus another, then it can send a probe into deep space a week later. All by changing the number of SRBs.

 

*EXTREME BIAS WARNING*

I agree completely. Spacex just doesn’t have the capability to do that at the moment. They can’t even get many contracts from the AF because of vertical integration requirements (which they do not fit at this time.)

as for the specialization, that’s what I think as well. Falcon is “one size fits all,” and ULA has many different sizes of delta/atlas for every mission you could possibly think of.

Spacex’s Reusable rockets are good for cheap and quick launches.  Things like GPS, for example.

if you want something in a wonky orbit, or you want to send a mission far out into the solar system, pick ULA’s expendables, as they are better for that. Just look at how many missions spacex has sent outside of LEO. One? Two?

 

not saying that they can’t, but the government probably wants to pick the company that has more experience doing that kind of thing.

Edited May 31, 2021 by CollectingSP