SpaceX Starship : First Flight : Starbase, TX : 20 April 2023 - DISCUSSION

[WindyCity]

...Assuming that the GSE is ready in July, would BN3 be fully ready, meaning that its raptors have been tested in static fires successfully and the vehicle flown on a lower altitude "hop"?

I agree that BN3 will need a static fire (at least one), but I don't think a "hop" for BN3 is necessary or likely. Its actual mission flight plan is "just" an extended "hop", not much different from the Falcon 9 booster - and a SH landing is not necessary for getting SS to orbit.

My reasoning with regard to a "hop" is that the vehicle's performance could be tested without loading the mass of propellant needed to push SS into orbit (assuming that they bypass a suborbital flight), which would run a higher risk of a catastrophic explosion if the liftoff and first minute of flight suffered a serious anomaly. A static fire wouldn't load the vehicle with as much stress as a short flight would.

[tyrred]

Procure vs obtain - in your opinion, what's the difference, Jim?

To anyone - when SpaceX applies for more complex permits to FAA... then FAA has more complexity to process per application... what is surprising about this?

[ETurner]

Procure vs obtain - in your opinion, what's the difference, Jim?

procure [ proh-kyoor, pruh- ]
1. to obtain or get by care, effort, or the use of special means: to procure evidence.

[capoman]

My understanding is that the GSE will hold enough for two complete launches. If that is the case, and my understanding is that water and nitrogen will be part of the GSE tanks there, it makes sense that the tank farm really only needs to be 2/3 complete to support a single launch. Likely 4 of 6 tanks required, and two unfinished ones could simply have the valves turned off that lead to them. If this is the case, the GSE is not as far off as it appears for a single test launch. The only caveat I can think of here is if all the tanks are required to load faster to reduce boiloff.

[novo2044]

While the GSE is a huge project it doesn't seem to involve any novel technologies?  SpaceX has proven to be pretty adept at fabricating tanks, to be fair.  Depending on what they do with the launch tower and crane/catching arms I could see that being a bit more of a sticking point.  But I'm most curious what their test program will be for the SH booster.  Will they do a 28 engine static fire?  How will the pad hold up without a blast diverter? 

I'm going to lay my chips on August.  Optimistic but ever since I saw the first visualizations of the bellyflop and thought "well there's no way they are going to use that" I've started hedging on the optimistic side.

[Robotbeat]

Agreed that getting that many Raptors to simply fire together reliably might take quite a while. Could easily see issues related to that adding a good 6 months. We don’t know at this point. No one really does.

If everything goes right, July isn’t literally impossible. But it could also be May 2022 by the time it finally launches.

[tssp_art]

...Assuming that the GSE is ready in July, would BN3 be fully ready, meaning that its raptors have been tested in static fires successfully and the vehicle flown on a lower altitude "hop"?

I agree that BN3 will need a static fire (at least one), but I don't think a "hop" for BN3 is necessary or likely. Its actual mission flight plan is "just" an extended "hop", not much different from the Falcon 9 booster - and a SH landing is not necessary for getting SS to orbit.

My reasoning with regard to a "hop" is that the vehicle's performance could be tested without loading the mass of propellant needed to push SS into orbit (assuming that they bypass a suborbital flight), which would run a higher risk of a catastrophic explosion if the liftoff and first minute of flight suffered a serious anomaly. A static fire wouldn't load the vehicle with as much stress as a short flight would.

No question that a "hop" would be useful in retiring some risk. But is that risk retirement worth the price? If the "hop" went perfectly on ascent (likely - the forces are well understood) but not so perfect on the landing (also likely - it's harder than going up) they would lose a bunch of raptors and possibly damage some of the new Ground Support Equipment - or a drone ship depending on how they try to land the booster and assuming they aren't ready to "catch" it. All of that on a flight that could have boosted the first StarShip to orbit. It just doesn't feel like a good trade and I think SpaceX will reach farther on the next flight.

[Tangilinear Interjar]

My understanding is that the GSE will hold enough for two complete launches. If that is the case, and my understanding is that water and nitrogen will be part of the GSE tanks there, it makes sense that the tank farm really only needs to be 2/3 complete to support a single launch. Likely 4 of 6 tanks required, and two unfinished ones could simply have the valves turned off that lead to them. If this is the case, the GSE is not as far off as it appears for a single test launch. The only caveat I can think of here is if all the tanks are required to load faster to reduce boiloff.

This was my point, a couple (2 maybe 3) tanks do not need to be fully operational but the tanks are really the easy part.  They may happen to be the big and obvious part but just take a look at the months of work that went into just running electrical conduit.  Every one of those conduit has wires run through them, attached to things on each side, and then tested, programmed etc.  All of the piping that needs to be fit, welded, bolted, pressure tested etc. 

Building the GSE system isn't rocket science, it's probably more like a petrochemical tank farm, but there is a lot of labor and complexity to building those things.  In this case, everything between the tanks and the rocket must be complete and functioning, that is the real part of the GSE system, the tanks are just a simple part at one end of the process.

I just had a funny realization, each end of the GSE system is a nearly identical big tank, one sits all comfy in its nice insulated jacket, the other gets shot into space.

[VaBlue]

Agreed that getting that many Raptors to simply fire together reliably might take quite a while. Could easily see issues related to that adding a good 6 months. We don’t know at this point. No one really does.

If everything goes right, July isn’t literally impossible. But it could also be May 2022 by the time it finally launches.

I don't see the firing of 28 Raptors in sync as being a large hurdle.  They have plenty of experience in lighting off 27 Merlins that are synced up pretty well (FH).  The SW base from Falcon was most likely modified for Raptor, and we've seen repeated firings of groups of 3 Raptors.  Scaling up to 28, while not physically trivial, is probably a fairly easy set of variables within the code base.  And that code is what will keep everything in sync - providing, of course, that the physical engines work as advertised.  (And there's the rub...)

Will agree, though, that launch can be pushed back by any number of issues nobody has even dreamed up, yet.  I'm not holding my breath for a July launch...

[electricdawn]

They'll probably won't even launch all Raptors at once - more like staggering them. All this within milliseconds of course. This to prevent resonances and other undesirable sh**. I think that this is not so much the problem. The problem will be more what happens when things go wrong. The software needs to be able to control 28 Raptors and handle any occurrence that might happen.

But then, SpaceX has repeatedly been flying 27 Merlins, so...

[Robotbeat]

Agreed that getting that many Raptors to simply fire together reliably might take quite a while. Could easily see issues related to that adding a good 6 months. We don’t know at this point. No one really does.

If everything goes right, July isn’t literally impossible. But it could also be May 2022 by the time it finally launches.

I don't see the firing of 28 Raptors in sync as being a large hurdle.  They have plenty of experience in lighting off 27 Merlins that are synced up pretty well (FH).  The SW base from Falcon was most likely modified for Raptor, and we've seen repeated firings of groups of 3 Raptors.  Scaling up to 28, while not physically trivial, is probably a fairly easy set of variables within the code base.  And that code is what will keep everything in sync - providing, of course, that the physical engines work as advertised.  (And there's the rub...)

Will agree, though, that launch can be pushed back by any number of issues nobody has even dreamed up, yet.  I'm not holding my breath for a July launch...

I don't mean in a showstopper sense, just in a raw numbers "this could take a while to get right" sense. Raptor is still a fairly new engine. 28 of them at once will probably cause quite a few aborted lift-offs in the early days. Just like with Falcon 9.

[Nevyn72]

They'll probably won't even launch all Raptors at once - more like staggering them. All this within milliseconds of course. This to prevent resonances and other undesirable sh**. I think that this is not so much the problem. The problem will be more what happens when things go wrong. The software needs to be able to control 28 Raptors and handle any occurrence that might happen.

But then, SpaceX has repeatedly been flying 27 Merlins, so...

Those 27 Merlins weren't clustered together and had a lot less combined thrust.

The nearest equivalence in terms of engine positioning and total thrust would be Russia's N1, so...

[punder]

They'll probably won't even launch all Raptors at once - more like staggering them. All this within milliseconds of course. This to prevent resonances and other undesirable sh**. I think that this is not so much the problem. The problem will be more what happens when things go wrong. The software needs to be able to control 28 Raptors and handle any occurrence that might happen.

But then, SpaceX has repeatedly been flying 27 Merlins, so...

Those 27 Merlins weren't clustered together and had a lot less combined thrust.

The nearest equivalence in terms of engine positioning and total thrust would be Russia's N1, so...

So, 50 years later, SpaceX can’t do better than a Soviet program that was absolutely screwed from 1965 on? Oh I see. No one will ever do hypersonic reentry of a booster. No one will ever recover or re-use a first-stage booster. No private company will ever fly a recoverable space capsule, much less re-use one. No American company will ever recapture the global launch market. And on, and on, and on... are you people even paying attention?

[Alvian@IDN]

They'll probably won't even launch all Raptors at once - more like staggering them. All this within milliseconds of course. This to prevent resonances and other undesirable sh**. I think that this is not so much the problem. The problem will be more what happens when things go wrong. The software needs to be able to control 28 Raptors and handle any occurrence that might happen.

But then, SpaceX has repeatedly been flying 27 Merlins, so...

Those 27 Merlins weren't clustered together and had a lot less combined thrust.

The nearest equivalence in terms of engine positioning and total thrust would be Russia's N1, so...

So, 50 years later, SpaceX can’t do better than a Soviet program that was absolutely screwed from 1965 on? Oh I see. No one will ever do hypersonic reentry of a booster. No one will ever recover or re-use a first-stage booster. No private company will ever fly a recoverable space capsule, much less re-use one. No American company will ever recapture the global launch market. And on, and on, and on... are you people even paying attention?

And do note that 9 engines on Falcon 9 was considered just as crazy at the early days

[Robotbeat]

I have no doubt that SpaceX can do it eventually. But until they've got a good static fire under their belt, there remains quite a bit of uncertainty about how hard that'll be. So it's rational to guess it might be, say, like 3-6 months extra time to figure that out.

BTW, I'm not too worried about supersonic flight on ascent. The main worry there on ascent is dynamic pressure, and they built a test frame for Starship for things like that, and so I think they have that handled. Separation is always kind of tricky, too, but not as crazy as a flip, which they managed first time.

Nah, I think just grinding through all the inevitable bugs of getting 28 high performance full flow staged combustion engines operating at the same time is likely to be something that could easily take 3-6 months to solve. I'm sure they'll solve it, though, and they'll probably even get to orbit soon after that.

[steveleach]

(The quoted posts were mostly removed - Chris).

I'm going to have to jump in here.

An awful lot of people are here in the NSF forums because they are specifically interested in SpaceX and Starship, and this is the best source of high quality information and discourse on that subject. These people might go and look through the old threads on other topics every now and then, but that is going to be selective at best.

And in the SpaceX/Starship forum sections, Jim tends to come across as an obnoxious troll, and so people respond to him like they do to trolls everywhere. No-one is under any obligation to research the background and posting history of the people they are responding to.

There are other posters here that display an immense depth of knowledge with almost every single post they make, and they are invariably treated with the utmost respect. If Jim wants that respect then he can adjust his own posting behaviour so that he gets it; I suspect he simply doesn't care though.

Regardless, he doesn't need anyone to defend him, and I personally believe that many of his posts (including that one) don't deserve it.

Maybe the mods should just remove Jim's post and the entire response chain (including this). None of them are adding value to the discussion of the mission.

[Chris Bergin]

Agreed. And a small trim of posts was required.

As much as you may not like what Jim has to say at times (or more so how he says it - and even he gets trimmed when he's uncivil), those of us know him from the years here's been here and has openly stated his NASA career, he knows his stuff.

[Nomadd]

 With no payload, only 3/4 of an orbit and resulting lower fuel load, maybe they'll still let it take off if a few Raptors don't light, or are a few percent off?

[Herb Schaltegger]

With no payload, only 3/4 of an orbit and resulting lower fuel load, maybe they'll still let it take off if a few Raptors don't light, or are a few percent off?

At first blush, my thinking is they will probably abort if any engine flat-out fails to ignite and ramp up to the nominal launch thrust level(*). One of the lesser-considered goals of a full stack launch will undoubtedly be gaining absolutely as much operational Raptor data they can all at once, especially in a large clustered configuration. If an engine fails to ignite and reach operating specs in the usual timeframe, the engineers will need to know why in case there’s some fault that might be shared by other engines, or an obscure issue with the vehicle.

(*) That said, I suspect the “nominal” thrust level and related parameters (chamber pressure, turbopump speeds, propellant flow rates …) might very well be more loosely constrained on the first developmental flight tests of so many engines at once as compared to later flights.

[Robotbeat]

Maybe. It’d be kind of N-1 of them to do it that way. But I think SpaceX could actually afford to do it the N-1 approach while the entirety of the Soviet Union’s lunar effort could not. SpaceX has shown ability and willingness to throw hardware at the problem.

It’s weird saying that a private company has greater space vehicle manufacturing capacity than an entire supercontinent-spanning Superpower in the late 1960s early 70s, but it’s true. That’s what half a century of progress finally gets you.

I hope they bother to do as many static fires as it takes to get things to work smoothly at lift-off, though, something the N-1 was incapable of. That, combined with actual full acceptance testing of each engine at McGregor (something that the N-1’s NK-15s couldn’t do), should help a lot.


This really, really does seem to have a lot of resemblance to N-1, doesn’t it? 28 (vs 30) staged combustion engines, a hardware-rich approach, high similarity between stages (2nd stage of Starship has 6 similar engines whereas N-1 had 8... in both cases using same fuel and similar engines as the first stage). Being used for lunar missions with intent for later Mars missions.

And I have extremely high confidence it will eventually work where N-1 failed. I’m not even worried about it. The 24/7 streamlining high definition cameras covering all aspects of Starship building and testing is the polar opposite of the extreme secrecy of the N-1.

(I gotta say that I kind of liked the N-1, and I think people crap on it too much. It nearly worked, and it probably would have on the next launch.)