totm nov 2023 SpaceX Discussion Thread

[Jacke]

First, a little personal background that is related to this flight attempt of Starship.

I am a veteran of the King's Own Calgary Regiment, which in World War 2 was known as the Calgary Regiment, informally the Calgary Tanks.  Which on 19 August 1942 participated in Operation Jubilee, the Dieppe Raid.

The only true successes of Operation Jubilee were the 2 flanking landings by Commando forces.  The main landing was a failure.  The approach and landings had problems and came under effective enemy fire.  There was insufficient naval and air bombardment to suppress that fire.  From my Regiment, among the landing casualties was the Commanding Officer's tank which sank and he and his crew were lost.  Only a few tanks got off the beach but with the other troops far insufficient to achieve any of their objectives.  All tanks that were mobile withdrew to cover the evacuation of the surviving troops.  Many Canadian soldiers died and many were taken prisoner, especially from the Calgary Tanks.  I knew a few of them personally in their later years.

At the time there was a lot of controversy over the Dieppe Raid.  There still is to some degree over the history.  There were claims that valuable lessons were learned.  Hog wash.  Nothing was learned about amphibious landings and combined arms operations that wasn't already known and demonstrated better later in the same year in other Allied assaults.  The lessons were there beforehand to be found.  Like many operations of mixed results, there was a lot wrong in the planning, preparations, and forces dedicated as well as the goals considering the challenge of the ground and opposing forces.

As for Starship....  Rocket engineering isn't the tech business.  Move fast and break things is damn more expensive.  There's nearly a century of history lessons that were learned with blood and money and any organization that glibly ignores these lessons will pay for it in more ways than one.

The flaws, let's count a few of them.

• SpaceX and the Starship design.  I'm not going to go into details on these, but they are flawed.  Just two points: SpaceX has lost a lot of competent staff who've moved on to better jobs.  Starship almost certainly can't have a launch escape system retrofitted, so it can't be crew-rated.  Because major abort -> lost LV -> lost crew.  That was one of the grave mistakes of the Shuttle and should not be repeated.
• Facility: Too small.  Wrong because it encroaches on wildlife areas, an international border, and human habitation.  No room for proper test stands and launch pads with proper safety areas for each.  It would be hard pressed to fit a pistol range with a backstop berm and its safety template in that area.
• Insufficient critical engineering design and review of what is a radically different launch vehicle.
• Lack of testing, especially no full-engine-count full-thrust full-duration burn test of Superheavy.
• A flight profile that depends on Superheavy and Starship being flung apart by spinning, which has never been done and wasn't tested with any sort of cheaper test vehicle.  And which is susceptible to being unable to stage if there are even a small number of failures in Superheavy.
• Ignorance of warning signs and problem predictions again and again (this one is so common in all engineering and military operations it's painful).
• An attitude to sweep problems under the rug and claim failures are successes.
• An attitude that the rules of engineering and the industry don't apply to them because...reasons.

Wanting to do something different and improve things doesn't mean ignore the history of engineering (centuries) and rocketry (just under a century).

[Exoscientist]

On 4/21/2023 at 11:26 AM, sevenperforce said:

Holy crap. This does not look good.

Those pilings are 150 feet deep, but still -- that's absolute devastation.

 This article link was posted on Reddit that suggest the pilings were dug 30m, 100 feet, deep:

https://www.teslarati.com/spacex-orbital-starship-launch-pad-texas/amp/

 But the article writer may have taken the 30m, 100 feet, underground idea from the length of the pilings visible above ground before they were put in place. Perhaps we can estimate how far below ground they would go  by their length as measured lying down horizontally before placement and by the length visible above ground when they are put in place, as partially dug in below ground.

  Bob Clark

 

 

 

[FleshJeb]

1 hour ago, Exoscientist said:

But the article writer may have taken the 30m, 100 feet, underground idea from the length of the pilings visible above ground before they were put in place. Perhaps we can estimate how far below ground they would go  by their length as measured lying down horizontally before placement and by the length visible above ground when they are put in place, as partially dug in below ground.

Down to local bedrock is the answer. I construction-staked a bridge a couple years ago that had similarly-sized pillars (5 ft dia, instead of 6 ft) about 75 ft long, extending about 50 ft below ground level to bedrock (This bridge was hugely overbuilt because it's designed to survive being under 20 ft of moving floodwater). It's impossible to tell from any of those photos because they're splicing the rebar into the already poured below-ground portions. 100 ft deep wouldn't be shocking, but neither would 50.

Spoiler

 

[RCgothic]

If it had only been about the vehicle, then I'm sure I'd be satisfied that objectives were met and data retrieved despite the lackluster performance in flight. It was spectacular to watch, and actually a bit impressive it kept flying as long as it did with so many failures.

But it's hard to be happy about trashing the launch site to such a degree. If it turns out they have to rebuild the OLM, that's a catastrophe. It can't be lifted down in its assembled state.

I don't necessarily buy into it being stupid to commit to launch after the pad began to disintegrate (which could only have been detected by engine failures).  If the pad is dead from shrapnel and the vehicle is damaged and can't be detanked and it's sitting on there filled with propellant, that's only going to have a worse outcome.

[JoeSchmuckatelli]

@FleshJebCan they repair the table in the condition it's in?  Enough to put some kind of diverter system in and try again in a reasonable time? 

 

Edited April 24, 2023 by JoeSchmuckatelli

[kerbiloid]

SpX previously said that they anyway were going to dig a pit below, to reinforce the launchpad basement, so optimists say that the fire show just helped with that by digging the crater where it should be.

The tower looks intact.

Mostly.

P.S.
Port Isabel, prepare.

Edited April 24, 2023 by kerbiloid

[Hannu2]

On 4/22/2023 at 2:05 AM, Meecrob said:

An explosion is an exothermic reaction that propagates at greater than the speed of sound.

Lol, I'm not sure we watched the same video. I saw two fireballs that were tamer than when I start my barbecue and have a bit too much propane under the lid.

You certainly like well done meat. Booster was almost empty but Starship had more than 1000 t fuel and oxygen on board.

[Exoscientist]

3 hours ago, FleshJeb said:

Down to local bedrock is the answer. I construction-staked a bridge a couple years ago that had similarly-sized pillars (5 ft dia, instead of 6 ft) about 75 ft long, extending about 50 ft below ground level to bedrock (This bridge was hugely overbuilt because it's designed to survive being under 20 ft of moving floodwater). It's impossible to tell from any of those photos because they're splicing the rebar into the already poured below-ground portions. 100 ft deep wouldn't be shocking, but neither would 50.

  Hide contents

 

  
 Thanks. That’s very relevant because of the ground water issue at Boca Chica. Another relevant question is how much total weight had to be supported, bridge plus cars?

 

  Robert Clark

Edited April 24, 2023 by Exoscientist

[Exoscientist]

 Good point. Has anyone done a frame by frame analysis to see if the flame front expanded at supersonic speed or not?

  Robert Clark

Edited April 24, 2023 by Exoscientist

[tater]

6 hours ago, Jacke said:

• SpaceX and the Starship design.  I'm not going to go into details on these, but they are flawed.  Just two points: SpaceX has lost a lot of competent staff who've moved on to better jobs.  Starship almost certainly can't have a launch escape system retrofitted, so it can't be crew-rated.  Because major abort -> lost LV -> lost crew.  That was one of the grave mistakes of the Shuttle and should not be repeated.

This is just wrong-headed. While I think that early versions should certainly have a LES of some form, given the goal, it simply impossible. Either human spaceflight becomes reliable enough that LES is unnecessary, or humans never leave Earth as more than a stunt.

F-22s have ejection seats. 787s don't.

Regardless, all the calculations around LOC events are just that, calculations. They make educated assumptions about the probability of events—including critical part failures because the parts on rockets are generally all "never flown" then "flown once (destroyed)"—then calculate a risk. They work to hit some arbitrary number. For Commercial Crew it's 1:270. The idea here for SS is reliability (clearly the long pole will be Raptor).

 

6 hours ago, Jacke said:

• Facility: Too small.  Wrong because it encroaches on wildlife areas, an international border, and human habitation.  No room for proper test stands and launch pads with proper safety areas for each.  It would be hard pressed to fit a pistol range with a backstop berm and its safety template in that area.

While Starbase is indeed tiny, way up the thread I think I recall posting that the entire facility—circled with pretty broad assumptions about SpaceX owning land right to the water—is small enough to fit inside the ring road around pad 39A in FL. The trouble of course is that there is not much coastal land facing East available anywhere.

 

6 hours ago, Jacke said:

• Insufficient critical engineering design and review of what is a radically different launch vehicle.

Meh. They are literally the world leader in launch vehicles, they have some idea how to design launch vehicles. They also don't want to hemorrhage money if they can avoid doing so.

 

6 hours ago, Jacke said:

• Lack of testing, especially no full-engine-count full-thrust full-duration burn test of Superheavy.

So build a factory in Alabama, ship it by barge to MI, then final product to the Cape. We'd not be having this discussion, because it would not be built yet. Some of us would likely be dead of old age before they had the thing stacked.

 

6 hours ago, Jacke said:

• A flight profile that depends on Superheavy and Starship being flung apart by spinning, which has never been done and wasn't tested with any sort of cheaper test vehicle.  And which is susceptible to being unable to stage if there are even a small number of failures in Superheavy.

Who cares? It's worth a try to test it, a mechanical stage pusher would be nontrivial on such a large vehicle. They can always end up putting stage sep motors on SH if they need to, those can be replaced on the pad (solids?).

 

6 hours ago, Jacke said:

• Ignorance of warning signs and problem predictions again and again (this one is so common in all engineering and military operations it's painful).
• An attitude to sweep problems under the rug and claim failures are successes.

Yeah, they clearly have no idea what they are doing engineering wise. Every week another Falcon 9 explodes, or careens into the wildlife area at KSC failing to land. Oh, wait.

 

6 hours ago, Jacke said:

• An attitude that the rules of engineering and the industry don't apply to them because...reasons.

I think the goal is to make decisions from a clean slate. On the other hand we have Blue Origin as the example of a new company doing literally everything you say SpaceX should be doing. We'll see if they beat Starship to orbit—and that's a false race. The real race of course is who wins the race to orbit at all, since BO was founded before SpaceX. So do things the way you suggest, and we're still waiting for Falcon 9 to see who wins, SpaceX or BO.

 

1 hour ago, Hannu2 said:

You certainly like well done meat. Booster was almost empty but Starship had more than 1000 t fuel and oxygen on board.

Yes, but the chances of a full detonation are effectively zero. A partial detonation can certainly happen, but the fuel/oxidizer mixture is then dispersed tot he point that it can no longer explode, it just burns. Spectacular, regardless. Not sure what the upper limits really is at the point where it's most concerning—at the pad, or just a few meters above it.  Enough rockets have been blown up since the 50s that I bet someone has characterized it someplace.

Edited April 24, 2023 by tater

[Hannu2]

12 minutes ago, Exoscientist said:

 Good point. Has anyone done a frame by frame analysis to see if the flame front expanded at supersonic speed or not?

This kind of "explosions" are not technically explosions. Fuel and oxygen have to mix with each other to burn which takes time. FTS has high small charge of high explosive but it only breaks tanks so that air resistance and burning fuel can rip the craft in pieces.

 

[Hannu2]

9 minutes ago, tater said:

Yes, but the chances of a detonation are effectively zero. A partial detonation can certainly happen, but the fuel/oxidizer mixture is then dispersed tot he point that it can no longer explode, it just burns. Spectacular, regardless. Not sure what the upper limits really is at the point where it's most concerning—at the pad, or just a few meters above it.  Enough rockets have been blown up since the 50s that I bet someone has characterized it someplace.

There can not be detonations with rocket fuel as far as I know. Slow explosions of mixed fuels and oxidizers (like traditional gunpowder) are called deflagration and it needs ready mixed materials which rockets do not have. But as far as I know difference between actual detonation and rapid burning is small kilometers away where supersonic shockwave of detonation has already attenuated.

I am not demolition expert but it seems that SpaceX's launch site is quite vulnerable. Tanks are on ground level very near the launch pad. If rocket fails at pad or low altitude it probably flattens and burns the whole place when several thousand tonnes of burning fuel falls on top of structures. Maybe they have thought that it is cheaper to build new simple buildings and tubings when needed than make bunkers which can handle anomalies.

[tater]

Quick Google Earth check. Production facility is ~157 acres.  The launch pad area is ~18 acres. ~175 acres total.

The SpaceX facility at 39A (counting all inside the ring road) is ~179 acres.

Spoiler

 

 

Edited April 24, 2023 by tater

[kerbiloid]

29 minutes ago, tater said:

So build a factory in Alabama, ship it by barge to MI

The 9 m diameter likely says that the four-row river barge was the origin.

4 * 2.438 + 3 * 0.05 = 9.9 m.

[tater]

April 25 Falcon 9 • Starlink 3-5
Launch time: Approx. 1302 GMT (9:02 a.m. EDT; 6:02 a.m. PDT)
Launch site: SLC-4E, Vandenberg Space Force Base, California

 

April 26 Falcon Heavy • ViaSat 3 Americas
Launch time: 2324-0018 GMT on 26th/27th (7:24-8:18 p.m. EDT on 26th)
Launch site: LC-39A, Kennedy Space Center, Florida

 

April 28 Falcon 9 • O3b mPOWER 3 & 4
Launch time: 2112 GMT (5:12 p.m. EDT)
Launch site: SLC-40, Cape Canaveral Space Force Station, Florida

 

(those sloppy people will likely blow up all 3, amiright?)

 

4 minutes ago, kerbiloid said:

The 9 m diameter likely says that the four-row river barge was the origin.

4 * 2.438 + 3 * 0.05 = 9.9 m.

Saturn V was 33 ft in dia. 10.1m, so we know that works

 

Edited April 24, 2023 by tater

[kerbiloid]

49 minutes ago, tater said:

Saturn V was 33 ft in dia. 10.1m, so we know that works

That works. Almost every rocket diameter is defined either by train standard size, or by a N-row barge payload width (ISO container = 2.438 m, spacing 5 cm).
Usually (if not always) it's also a diameter of a spherical tank (with cylindric equatorial section) of integer volume.

The engineering laws are stronger than physical ones.

Equipment to bend, transport to move. All are standard.

[JoeSchmuckatelli]

Some of the dust

Guess they did not do a FOD walk?

Edited April 24, 2023 by JoeSchmuckatelli

[magnemoe]

1 hour ago, kerbiloid said:

That works. Almost every rocket diameter is defined either by train standard size, or by a N-row barge payload width (ISO container = 2.438 m, spacing 5 cm).
Usually (if not always) it's also a diameter of a spherical tank (with cylindric equatorial section) of integer volume.

The engineering laws are stronger than physical ones.

Equipment to bend, transport to move. All are standard.

This for obvious reasons, and you can cheat, as I understand the A 380 bares could take on ballast water to float lower and pass under bridges if water level was high, 
Don't think this is done for containers but probably for oversize stuff.

12 minutes ago, JoeSchmuckatelli said:

Some of the dust

 

As an positive note lots of the dust would not cause lung issues unless it hit your chest at an ballistic trajectory. 

[sevenperforce]

6 hours ago, RCgothic said:

I don't necessarily buy into it being stupid to commit to launch after the pad began to disintegrate (which could only have been detected by engine failures).  If the pad is dead from shrapnel and the vehicle is damaged and can't be detanked and it's sitting on there filled with propellant, that's only going to have a worse outcome.

Agreed.

Not waiting to put in a diverter was obviously a mistake, in hindsight. I'm guessing that they looked at the damage to the pad after the 50% static fire, did some modeling, and accepted the model's predictions about a full-up launch. The model most likely failed to account for the failure cascade discussed upthread, and so here we are.

But if they were going to fire without a diverter, I completely agree with the decision to do it in an all-up test. Getting 5 million kg of propellant away from the pad as fast as possible is FAR better than firing and letting it sit there and go kablooey.

[sevenperforce]

4 hours ago, tater said:

It's worth a try to test it, a mechanical stage pusher would be nontrivial on such a large vehicle. They can always end up putting stage sep motors on SH if they need to, those can be replaced on the pad (solids?).

I will note that those are also nontrivial. If you put them on Superheavy, they impinge on Starship, which you may not want. If you put them on Starship, the impingement on Superheavy is less of a big deal, but that's also a lot of mass to push. 

Quick calculation -- at separation, Superheavy probably masses on the order of 650 tonnes, almost five times more than the S-IC first stage of the Saturn V. The eight retrorockets on the S-IC produced 394 kN for a period of just under a second. To get the same amount of pull-away acceleration on Starship, you'd need retrorockets producing a collective 1.9 MN of thrust...more than a Raptor 1 engine. If SpaceX used the same booster separation motors that Northrup Grumman uses for the SLS SRBs, they would need a cluster of 14 of them, which is pretty extreme.

4 hours ago, Hannu2 said:

There can not be detonations with rocket fuel as far as I know. Slow explosions of mixed fuels and oxidizers (like traditional gunpowder) are called deflagration and it needs ready mixed materials which rockets do not have. But as far as I know difference between actual detonation and rapid burning is small kilometers away where supersonic shockwave of detonation has already attenuated.

If the fuel and oxidizer are allowed to mix without igniting, and are then ignited, you absolutely end up with a detonation.

1 hour ago, JoeSchmuckatelli said:

Old strategy: use Starship to go to Mars

New strategy: use Starship to make Earth into Mars

[JoeSchmuckatelli]

1 hour ago, magnemoe said:

As an positive note lots of the dust would not cause lung issues unless it hit your chest at an ballistic trajectory. 

They could gather it together and sell it as 'Stage 0 Regolith'.  I'm certain folks would buy it.

RocketAsh?

Volcanic Ash - Etsy

4 minutes ago, sevenperforce said:

Old strategy: use Starship to go to Mars

New strategy: use Starship to make Earth into Mars

Grin!

[tg626]

3 hours ago, kerbiloid said:

That works. Almost every rocket diameter is defined either by train standard size, or by a N-row barge payload width (ISO container = 2.438 m, spacing 5 cm).
Usually (if not always) it's also a diameter of a spherical tank (with cylindric equatorial section) of integer volume.

The engineering laws are stronger than physical ones.

Equipment to bend, transport to move. All are standard.

S-IC was dictated by the size of a standing factory. In fact that helped drive the C-5 (aka Saturn V) to the top of the list of candidate launchers and in turn Lunar Rendezvous as the mission profile. Direct Ascent, the original profile, was nixed in part due to the required C-8 (aka Nova) requiring a new factory, and they had a hard deadline of 12/31/1969 to meet or beat.

[sevenperforce]

I feel like something along these lines could work well.

Big steel "dance floor" hung from the legs, so that they can still service the booster from underneath. Three giant steel triangles on hydraulics that translate inward and outward and fold up and down.  

[K^2]

5 hours ago, tater said:

Either human spaceflight becomes reliable enough that LES is unnecessary, or humans never leave Earth as more than a stunt.

F-22s have ejection seats. 787s don't.

Current generation of chemical rockets is not sustainable for the kind of mass commercial flight that gave us 787 and the like. Superheavy will never be a 787. It fundamentally can't.

I'd also point out that a 787 can glide to a survivable landing with total engine failure and is designed to climb with a single engine failure. Most critical systems are at least tripple-redundant. Starship would have to be built to this kind of standard to be even considered for passenger flight without an escape system. A Starship with dead engines is a coffin.

Edited April 24, 2023 by K^2

[tater]

13 minutes ago, K^2 said:

Current generation of chemical rockets is not sustainable for the kind of mass commercial flight that gave us 787 and the like. Superheavy will never be a 787. It fundamentally can't.

I'd also point out that a 787 can glide to a survivable landing with total engine failure and is designed to climb with a single engine failure. Most critical systems are at least tripple-redundant. Starship would have to be built to this kind of standard to be even considered for passenger flight without an escape system. A Starship with dead engines is a coffin.

I actually agree, I have said I don't see humans on Starship—even as danger-accepting astronauts—for a while.

That said, open ocean airline flights  for twin engine aircraft accept the unlikely chance of a 2 engine failure as a LOC event (a ditch in the Pacific is not likely a survivable event). If VTVL is ever to become in the same order of magnitude risk as airline travel (or even the same order of magnitude as it was in the 1950s, lol), they will need incredibly reliable engines, and ideally the ability tolerate engine(s) out situations.

My point is that humans will at some point need a reliable, safe way to get to and from space, or it's the equivalent of fighter pilot (or more) risk for space travel forever. Maybe that means wings—the trick there is that for passenger comfort, you'd then need HTHL, can't expect people to board otherwise (VTHL is a nonstarter for civlians).