totm nov 2023 SpaceX Discussion Thread

[zolotiyeruki]

6 hours ago, RyanRising said:

... I have no idea how much the heat shield will weigh, on that section of the ship, so I’m gonna call it half a ton, bringing us up to 14460 kg. ...

Why would a LES require a heat shield?  Are you expecting that it might be used all the way to near-orbit?  My impression is that LES is jettisoned well before the craft gets up to the kind of velocities that would require a heat shield for re-entry.

[Beccab]

15 minutes ago, zolotiyeruki said:

Why would a LES require a heat shield?  Are you expecting that it might be used all the way to near-orbit?  My impression is that LES is jettisoned well before the craft gets up to the kind of velocities that would require a heat shield for re-entry.

LES doesn't, Starship does. Every section of starship upper stage is covered in half with heat shield tiles

[MKI]

 

6 hours ago, RyanRising said:

The main issues with developing something like that would probably be the hybrid motors (tricky beasts, those are the mechanism to separate the launch module from the rest of the crew section cleanly, and of course the chutes. I haven’t run the numbers on this construction, though - I need to do that to work out the extra mass for hybrid motor + casing and see if we’ve developed any parachutes that could feasibly handle something more massive than Orion. 

This is all extra risk, that also directly effects other aspects of the Spacecraft making it overall more unsafe when used with its current requirements, under normal circumstances you now suddenly have more rockets that can fail/explode/leak/ignite near/in your crew compartment which will get you killed, you also have less margins due to all the weight you just added, decreasing performance across the board. 

Currently Starship is running as close to the margins as possible, because again, SpaceX needs to figure out that entire list of requirements first. Adding parts into the process will not only slow things down, and make iteration slower/worse, it will also increase complexity, which increases risk. 

Shoving in a LES in Starship just doesn't seem to be on the table. The main counterpoint to having one seems to run along the lines that "all other crewed rockets had one (except Shuttle)" yet there aren't many cases where such a system would help even if available, there are cases where such systems would get just kill people however. 

 

[wumpus]

6 hours ago, RCgothic said:

Starship won't have LES because the Mars version can't have LES.

Crew-rated Earth version with a LES? Removed the LES for Mars? That's a change and now it's not crew-rated anymore.

Mars Starship requires 10 Starship launches (9 deliver propellant).  There is no reason why humans can't be on the 9 non-Mars launches and have a LES.

Also at least one Falcon 9 had an explosion  that would require an LES to escape (so obviously things designed with the "move fast and break stuff" will break on you.  Expect similar safety on Starship, or less).  Also with the LES, they prefer to fuel the Falcon while crewed (which involved a second incident during fueling).  Including a known technology to greatly increase safety during an extremely dangerous part of the mission appears to be a no brainer.

Edited August 16, 2021 by wumpus
falcon 9 example

[Beccab]

17 minutes ago, wumpus said:

There is no reason why humans can't be on the 9 non-Mars launches and have a LES.

There really is, given that those 9 are unmanned tanker launches. Put crew and even a LES on it and it isn't a tanker anymore

[MKI]

I think we should remember that "crew rated" is a NASA/Government rating. AFAIK there is nothing preventing Elon from putting some zealots on the next Starship to Mars. 

I have no idea how or if NASA would rate Starship safety, since its so vastly different than anything before it. Nor can I see SpaceX doing much with getting it rated anytime soon with the current architecture, which is very much still a grain silo with a buncha engines, heat tiles, and a much larger grain silo with even more engines on it as the booster.

Technically the Shuttle "was human rated", and it was a well known death trap so I'm sure there is plenty of leeway. 

 

Edited August 16, 2021 by MKI

[satnet]

50 minutes ago, MKI said:

I think we should remember that "crew rated" is a NASA/Government rating. AFAIK there is nothing preventing Elon from putting some zealots on the next Starship to Mars. 

I have no idea how or if NASA would rate Starship safety, since its so vastly different than anything before it. Nor can I see SpaceX doing much with getting it rated anytime soon with the current architecture, which is very much still a grain silo with a buncha engines, heat tiles, and a much large grain silo with even more engines on it as the booster.

Technically the Shuttle "was human rated", and it was a well known death trap so I'm sure there is plenty of leeway. 

 

SpaceX could never launch humans without FAA/NASA approval. If it is from US soil there are treaties in place that make the US responsible. Even if they launch from international waters as a US company the government has oversight. If they launched without it they would probably lose every government contract overnight and every commercial contract shortly thereafter. They aren't foolish enough to do it.

Anything can be human rated if you can demonstrate that you have sufficient forethought into survivability. Looking at commercial crew the expectation will probably also include un-crewed demo flights to empirically demonstrate safety. You can never make it perfect, but you can demonstrate adequate forethought and margin in your design, which is as close as any engineering endeavor gets. Having seen something similar before makes it immensely easier, but is not a prerequisite.

[tater]

23 minutes ago, satnet said:

SpaceX could never launch humans without FAA/NASA approval. If it is from US soil there are treaties in place that make the US responsible.

This looks like it.

https://www.govinfo.gov/content/pkg/FR-2006-12-15/pdf/E6-21193.pdf

NASA certification is not mentioned at all. Mentions test flights being required. Zero specifics on ways to calculate risk for LOC, etc.

 

23 minutes ago, satnet said:

Even if they launch from international waters as a US company the government has oversight. If they launched without it they would probably lose every government contract overnight and every commercial contract shortly thereafter. They aren't foolish enough to do it.

The rules are incredibly vague, and crew loss on test flights is not at all disqualifying from anything, else we'd not have things like supersonic jet fighters, as commercial test pilots died testing them.

 

23 minutes ago, satnet said:

Anything can be human rated if you can demonstrate that you have sufficient forethought into survivability. Looking at commercial crew the expectation will probably also include un-crewed demo flights to empirically demonstrate safety. You can never make it perfect, but you can demonstrate adequate forethought and margin in your design, which is as close as any engineering endeavor gets. Having seen something similar before makes it immensely easier, but is not a prerequisite.

FAA mentions test flights, no specifics. Crew needs to be informed of risk—they are thinking largely of customers here. Ie: deadly risk is fine, but people flying these need informed consent as to risk.

Physical requirements mostly at flight crew, passengers not allowed to control vehicle.

FAA can always arbitrarily regulate, but there are certainly no laws concerning SpaceX doing a Moon mission on their own.

Edited August 16, 2021 by tater

[RyanRising]

4 hours ago, MKI said:

 This is all extra risk, that also directly effects other aspects of the Spacecraft making it overall more unsafe when used with its current requirements, under normal circumstances you now suddenly have more rockets that can fail/explode/leak/ignite near/in your crew compartment which will get you killed, you also have less margins due to all the weight you just added, decreasing performance across the board. 

Currently Starship is running as close to the margins as possible, because again, SpaceX needs to figure out that entire list of requirements first. Adding parts into the process will not only slow things down, and make iteration slower/worse, it will also increase complexity, which increases risk.

Those hybrids would have a very hard time failing, leaking, or exploding. Their fuel is solid, so it's not going anywhere, but cannot ignite without its oxidiser, which is stored in the LOX header - something that will be developed and present regardless of this system. Adding crew launch capabilities to Starship is going to be a massive amount of work whether or not you have an LES, an LES is just an added safety measure. In this guise, and in all other implementations of launch escape systems (most that do use solid propellant, mind you) this adds to crew safety, not reduces it.

You are most certainly not going to iterate as rapidly with crewed Starships as they do with uncrewed, because you need to be more careful flying meat. That means development delays due to this system would be proportionally less significant than they would on the uncrewed Starships, which they will likely still iterate with at breakneck speed while crew development is happening. You're totally right that this reduces margins and decreases performance. But this would only be implemented once they're looking at flying crew, at which point they will have margin to spare on little things like accident survivability and crew accommodations. Basically, I don't agree that adding a failsafe increases risk.

Quote

...yet there aren't many cases where such a system would help even if available, there are cases where such systems would get just kill people however. 

Here's a post that details the incidents with Soyuz where a launch abort was performed. 2/3 cases, the launch abort system helped. 1/3 it was not needed. There is indeed one case in which the launch abort system caused a disaster, killing three people. It still saved six later on its life, and I'd say that proves it was a worthwhile development. Just as a numbers game, and admittedly specific to that rocket, the launch abort system was a worthwhile development.

Edited August 16, 2021 by RyanRising

[tater]

Texted that link to my lawyer friend who is also a space nut. He immediately noticed "they seem to start by talking about flexibility and even providing an "alternative means" to satisfy the rule"" Then talk about waivers. There is no legal impediment to SpaceX flying humans.

[RyanRising]

3 hours ago, Beccab said:

There really is, given that those 9 are unmanned tanker launches. Put crew and even a LES on it and it isn't a tanker anymore

True, but that Earth launch variant should still exist and be separate from the Mars trip ship. It wouldn't replace one of the tankers, but 11 launches seems plausible if 10 does.

Edited August 16, 2021 by RyanRising

[mikegarrison]

It's just that, for years, I read in this forum that the Shuttle was fundamentally flawed as a crew vehicle because it simply could never have had a robust launch abort mode. But now, Starship fans are saying that Starship is somehow different?

I get that a launch abort mode is useless on the moon or on Mars, because there is no place to abort to anyway. But that's not true here on Earth.

One of the most-discussed problems with the Shuttle design was that it tried to do too many things. I think that we are also seeing that with the idea of Starship as a crew launch vehicle. It's trying to be a Lunar/Martian lander/launcher, and that locks in certain design choices that compromise it as a Earth lander/launcher.

However, it is still in development. So things could (and probably will) change along the way.

1 hour ago, tater said:

FAA can always arbitrarily regulate, but there are certainly no laws concerning SpaceX doing a Moon mission on their own.

Except there are laws. You can't even fly a drone above (IIRC 400 ft) without FAA approval. Experimental airplanes are regulated.

Once you get into space, there are no laws that I know of other than the Outer Space Treaties. (Could be some, but I don't know them.) However, you can't get to space without flying through the sky, and in the US the FAA is -- by law -- the regulatory agency for that.

If SpaceX built a launch platform in the open ocean, or built a launcher in some country that allowed them to fly anything they wanted to, then that would be different.

[tater]

11 minutes ago, mikegarrison said:

It's just that, for years, I read in this forum that the Shuttle was fundamentally flawed as a crew vehicle because it simply could never have had a robust launch abort mode. But now, Starship fans are saying that Starship is somehow different?

I get that a launch abort mode is useless on the moon or on Mars, because there is no place to abort to anyway. But that's not true here on Earth.

One of the most-discussed problems with the Shuttle design was that it tried to do too many things. I think that we are also seeing that with the idea of Starship as a crew launch vehicle. It's trying to be a Lunar/Martian lander/launcher, and that locks in certain design choices that compromise it as a Earth lander/launcher.

Yeah, I've never been keen on SS as a crew vehicle to/from Earth for exactly that reason.

 

11 minutes ago, mikegarrison said:

Except there are laws. You can't even fly a drone above (IIRC 400 ft) without FAA approval. Experimental airplanes are regulated.

Once you get into space, there are no laws that I know of other than the Outer Space Treaties. (Could be some, but I don't know them.) However, you can't get to space without flying through the sky, and in the US the FAA is -- by law -- the regulatory agency for that.

If SpaceX built a launch platform in the open ocean, or built a launcher in some country that allowed them to fly anything they wanted to, then that would be different.

There are laws that allow the FAA to regulate, but there are no specific laws about "airworthiness" of launch vehicles, and none about "crew rating." The FAA rules I posted are all about informed consent, really. There are clear rules regarding the pilots, as there are for aircraft. Nothing in those rules says anything about vehicle certification other than the provider must tell the passengers it is not certified.

[MKI]

2 hours ago, RyanRising said:

Basically, I don't agree that adding a failsafe increases risk.

It depends on what kinda of "failsafe" we are talking about.

Starships requirements are so inherently unsafe, risky and cover such a large range of scenarios adding a dedicate failsafe for a specific scenario will end up being redundant most of the time and not actually protect against most of the risks the platform will see.

Its one thing to add life-vests under every seat of an airliner, as that is an easy safe guard to prevent people from drowning when ditching in a body of water, it also hardly affects margins as its incredible light . Its another to try to give every single passenger an ejection seat, flight gear, and a parachute to survive a mid-air breakup or double engine failure during takeoff. Not only could a passenger just use it wrong and get them, and everyone else killed, the increased complexity could result in more issues down the line in every category of managing the system itself, from weight, margins, inspections, maintenance. If "no part is the best part", adding a bunch of parts you don't plan on using is the opposite of that. 

Even the evacuation slides on a plane aren't foolproof, people get injured using them all the time, which is why its a big decision if a pilot is to evacuate a plane or not in an emergency. Just executing a fail-safe introduces a new element of risk. Obviously that risk should be lower than the alternative, but idk if firing another rocket and executing what is essentially a full stage separation is that much safer than sticking with Starship itself.

 

When it comes to risk management you actually want to minimize the most risky part of the system. When it comes to Starship, not only is it hard to even tell where the most risky part of the system is, but there aren't many thing you can do directly to minimize them. However, when it comes to launching rockets using a Super Heavy, it actually might be the safest part of the whole system. Super heavy will have multiple redundancies, and should be the single most heavily tested spacecraft/launcher ever made by the time people are flying on it. 

 

If I was a betting man, I'd put my money on hull integrity being the single most worrying aspect of Starship. Never mind losing a heat tile, what about getting hit with a micro meteorite during Mars transfer, where you start losing propellant as your hull is the fuel tank. You might be able to fix/patch that part of the hull so you stop losing fuel to the vacuum of space, that's if you didn't instantly explode. But who's to say your hull will sustain re-entry/landing/waiting/re-fueling/takeoff/transfer/re-entry/landing??

Such a scenario would be the most insane "tire patch" ever, except instead of trying to take your patch to the nearest tire shop to get a repairs, your in the middle of freaken space on your way to your destination with Isaac Newton preventing you from making a u-turn and no gas station or roadside assistance available for months. Hell if Starship requires an aerobrake you might of not only gotten a hole in your fuel-tank but also your brakes, and might not have enough brake fluid to stop in the parking lot of your destination!

 

At that point your less of a plane and more of a classical sailing ship that starts taking on water, with multiple damaged sails, in a much bigger ocean where all the islands are flying away from you. At that point your probably just screwed, and no amount of engineering redundancy can help you. 

  

33 minutes ago, mikegarrison said:

One of the most-discussed problems with the Shuttle design was that it tried to do too many things. I think that we are also seeing that with the idea of Starship as a crew launch vehicle. It's trying to be a Lunar/Martian lander/launcher, and that locks in certain design choices that compromise it as a Earth lander/launcher.

Its not so much the Shuttle tried to do too many things, it had 1 shot to do all those things at once. IE it had a design lock that could never change or iterate due to a multitude of reasons. A Shuttle designed iteratively would of been vastly more successful because any issue found would of been fixed/changed all without forcing 7 lives to fly the damn thing. I'm not sure if it would of become what the dreams of "reusable space" would of been, but it should of been better.

 

edit I just realized the "safest redundancy" is to send multiple Starships at more or less the exact same time in "fleets" to and from Mars. Reminds me of the Pale Blue Dot book by Carl Sagan, that described this exact scenario. It also noted this is exactly what humans did when traveling across the ocean into unknown lands, with the idea that some will make the journey, and have other other vessels to lean on in-case of issues. 

Edited August 16, 2021 by MKI

[CatastrophicFailure]

15 minutes ago, mikegarrison said:

It's just that, for years, I read in this forum that the Shuttle was fundamentally flawed as a crew vehicle because it simply could never have had a robust launch abort mode. But now, Starship fans are saying that Starship is somehow different?

There’s a key difference here: Shuttle could never fly enough, especially without crew, to significant retire risks or demonstrate reliability. Starship could, and most likely will, fly hundreds of times before people are ever on board for launch. 
 

(My personal bet on DearMoon is the crew ends up getting cut to 7 and launching/returning on a Dragon.)

[mikegarrison]

5 minutes ago, MKI said:

Never mind losing a heat tile, what about getting hit with a micro meteorite during Mars transfer, where you start losing propellant as your hull is the fuel tank. You might be able to fix/patch that part of the hull so you stop losing fuel to the vacuum of space, that's if you didn't instantly explode.

Well, you won't "instantly explode" unless oxidizer and fuel are exposed to each other.

[MKI]

9 minutes ago, mikegarrison said:

Well, you won't "instantly explode" unless oxidizer and fuel are exposed to each other.

Which is possible, however even a non-explosion could be terrible. Shouldn't exposing pressurized liquid to the vacuum of space create dangerous amounts of potential force on the craft, potentially ripping it apart... and then you explode ;D

[StrandedonEarth]

23 minutes ago, MKI said:

If I was a betting man, I'd put my money on hull integrity being the single most worrying aspect of Starship. Never mind losing a heat tile, what about getting hit with a micro meteorite during Mars transfer, where you start losing propellant as your hull is the fuel tank.

This has always been one of my concerns too. And adding MMOD protection is going to mess with everything, especially mass-wise

[SpaceFace545]

46 minutes ago, MKI said:

even a non-explosion could be terrible

yeah like not having any fuel. even if starship could safely crash its still heavily reliant on it's RCS system so yeah.

Kind of a conspiracy or whatever but what if the hiccups with the FAA is just Jeff's lobbying.

[mikegarrison]

1 hour ago, tater said:

Nothing in those rules says anything about vehicle certification other than the provider must tell the passengers it is not certified.

That's basically the rule for experimental aircraft.

But it's not true that is the only rule that applies to experimental aircraft. The airspace is still regulated.

https://www.faa.gov/aircraft/air_cert/airworthiness_certification/sp_awcert/experiment/

There is also this: https://www.faa.gov/space/

Apologies if you already posted this and I missed it.

In case people don't understand this, the US has "laws" passed by Congress and "regulations" that are made by following a process approved by Congress. But "regulations" are laws in the general sense.

14 CFR is part of US law. You can get fined if you violate it. You can be sent to prison if you violate it. It's the law, despite that it is a "regulation".

[tater]

11 minutes ago, mikegarrison said:

Apologies if you already posted this and I missed it.

I didn't, thanks for the links. the relevant part seems to be part 460 (my lawyer friend said that as well skimming it).

https://www.faa.gov/space/licenses/human_spaceflight/

Unsure how they deal with airworthiness past the fact that the vehicle is experimental. The part 460 bits include saying it needs integrated testing, lfe support, etc, but no specifics I can see.

My point if you look at the initial response I made was not that there were not hoops to jumpt through at all, but arguing the specific claim that international treaties didn't allow SpaceX to do a Moon flight, or without FAA/NASA approval.

Zero NASA approval is required at all.

FAA approval is required for all launches already—but the approval for humans seems little different from uncrewed launches as long as the humans at risk know the risks, and are on the rocket, not on the ground where the rocket might crash

 

Edited August 17, 2021 by tater

[StrandedonEarth]

7 minutes ago, tater said:

as long as the humans at risk know the risks, and are on the rocket, not on the ground where the rocket might crash

Legend has it that Von Braun once stood on the bullseye of the A4 (which became the V2) testing range while trying to troubleshoot the problem of rockets breaking up mid-flight, figuring that it was the safest spot (an A4 impacted 300ft away, IIRC)

[RyanRising]

2 hours ago, MKI said:

When it comes to risk management you actually want to minimize the most risky part of the system. When it comes to Starship, not only is it hard to even tell where the most risky part of the system is, but there aren't many thing you can do directly to minimize them. However, when it comes to launching rockets using a Super Heavy, it actually might be the safest part of the whole system. Super heavy will have multiple redundancies, and should be the single most heavily tested spacecraft/launcher ever made by the time people are flying on it.

I again disagree with the first part. I think you want to minimise every risk you can. Address the riskiest bits first, of course, but that doesn't mean you should ignore the less significant risks. And this is one of those risks we know how to mitigate. The other stuff you mention later on may not have such a clear solution. But I've said before, and will say again, I agree that it's okay to eventually launch crew on a Staship with no escape system- if and only if it has practically validated some extremely low failure rate, by flying several hundred successful missions for every single launch failure. Until it reaches that point, launch is still a risky endeavour, and one we know how to reduce the risk of.

 

I want to be clear: nowhere have I said this launch escape system should come at the cost of other safety systems. I'm only saying that flying crew in a Starship, without an LES, before it has practically validated its reliability, is incredibly irresponsible and needlessly dangerous. If any of those conditions changes, so does my opinion.

 

2 hours ago, MKI said:

Just executing a fail-safe introduces a new element of risk. Obviously that risk should be lower than the alternative, but idk if firing another rocket and executing what is essentially a full stage separation is that much safer than sticking with Starship itself.

If the rocket is blowing up behind me, I'll try firing the LES. You can stay behind if you think that's safer.

[FleshJeb]

12 hours ago, Rakaydos said:

Still, that's where cheap rapid reuse WILL come into play.   If by iteration 60 they've nailed "3 flights a day per pad" reuse, that's 90 flights per month before the next iteration comes out of starbase. 3 months to get 270 flights, the commercial crew calculated LoC number.

To economically justify that kind of launch cadence you have to have demand. While I recognize that reduced launch costs help drive a "virtuous cycle", the rest of the space industry isn't going to have that much tonnage to throw into orbit within the next decade or three.

[mikegarrison]

18 minutes ago, FleshJeb said:

To economically justify that kind of launch cadence you have to have demand. While I recognize that reduced launch costs help drive a "virtuous cycle", the rest of the space industry isn't going to have that much tonnage to throw into orbit within the next decade or three.

Well ... you never know. One of the primary restrictions on the use of space is simply how expensive it is to get there. If you make it cheaper, then not only are the launch costs cheaper but also you start to lower the costs of what you send there. You don't have to make everything simultaneously as light as possible but also as robust as possible, because you can more easily afford a little extra weight or maybe a failed sat or two. Look at Starlink and how they seem to be willing to accept a certain failure rate for their sats because they are throwing so many of them up there so cheaply that they can afford to have a few fizzle.

That being said, it's something I pointed out a few days ago. A high launch cadence only works if you have paying payloads. Otherwise you have the old retail joke about "losing money on every sale but making up for it with volume". (And tanking fuel isn't really a paying payload, unless you are tanking it for someone who is paying for it.)